Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
listlengths 1
91
| keywords
listlengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
Pseudomonas aeruginosa is a common environmental bacterium that is also a significant opportunistic pathogen , particularly of the human lung . We must understand how P . aeruginosa responds to the lung environment in order to identify the regulatory changes that bacteria use to establish and maintain infections . The P . aeruginosa response to pulmonary surfactant was used as a model to identify transcripts likely induced during lung infection . The most highly induced transcript in pulmonary surfactant , PA5325 ( sphA ) , is regulated by an AraC-family transcription factor , PA5324 ( SphR ) . We found that sphA was specifically induced by sphingosine in an SphR-dependent manner , and also via metabolism of sphingomyelin , ceramide , or sphingoshine-1-phosphate to sphingosine . These sphingolipids not only play a structural role in lipid membranes , but some are also intracellular and intercellular signaling molecules important in normal eukaryotic cell functions as well as orchestrating immune responses . The members of the SphR transcriptome were identified by microarray analyses , and DNA binding assays showed specific interaction of these promoters with SphR , which enabled us to determine the consensus SphR binding site . SphR binding to DNA was modified by sphingosine and we used labeled sphingosine to demonstrate direct binding of sphingosine by SphR . Deletion of sphR resulted in reduced bacterial survival during mouse lung infection . In vitro experiments show that deletion of sphR increases sensitivity to the antimicrobial effects of sphingosine which could , in part , explain the in vivo phenotype . This is the first identification of a sphingosine-responsive transcription factor in bacteria . We predict that SphR transcriptional regulation may be important in response to many sites of infection in eukaryotes and the presence of homologous transcription factors in other pathogens suggests that sphingosine detection is not limited to P . aeruginosa .
Pseudomonas aeruginosa is a common , Gram negative , environmental bacterium that can cause significant disease as an opportunistic pathogen , particularly in the lung . P . aeruginosa lung infections are prevalent in people with cystic fibrosis ( CF ) and chronic obstructive pulmonary disease ( COPD ) , as well as individuals undergoing mechanical ventilation [1]–[4] . These infections cause significant morbidity and mortality and continue to be a major health care burden [5]–[8] . P . aeruginosa has a large genome by bacterial standards ( ∼6 Mbp ) containing a high proportion of regulatory genes ( ∼8% are predicted to be transcriptional regulators ) [9] . This large regulatory capacity is likely important for P . aeruginosa success as an opportunist , enabling it to rapidly alter gene expression in response to host-derived factors and environmental conditions . Understanding mechanisms by which P . aeruginosa detects and responds to the host could present new avenues to combat these devastating , and often antibiotic resistant [10] , opportunistic infections . Our current understanding of P . aeruginosa response to the host come from transcriptional profiling using epithelial cells , mucus , or CF sputum [11] , [12] . We were interested in the response of P . aeruginosa to the environment of the distal airway , particularly the response to mammalian pulmonary surfactant , the lipid rich mixture that coats the airway surface liquid of the lungs and participates in both respiratory physiology and host defense ( reviewed in [13] , [14] ) . This mixture is rich in phosphatidylcholine ( ∼75% by mass ) but also has a substantial fraction of other phospholipids , cholesterol and its esters , and sphingolipids [15] . Sphingolipids constitute a class of molecules that are critical components of eukaryotic cell membranes . In addition to this structural role in membranes and their biophysical role in pulmonary surfactant , many sphingolipids have been shown to act as signaling molecules that play critical roles in regulation of diverse physiological processes . The broad importance of sphingolipid signaling in eukaryotic hosts has only recently been appreciated , and the rapidly expanding field has many recent reviews [16]–[19] . Sphingosine serves as a backbone component for all sphingolipids , which include the signaling molecules sphingosine-1-phosphate ( S1P ) and ceramide , as well as the structural lipid sphingomyelin . S1P , in particular , has been intensely studied in the past decade as a potent immune signaling molecule that plays a critical role in diverse immune functions such as lymphocyte trafficking , myeloid cell activation , and epithelial and endothelial barrier function , mediated by five G-protein coupled receptors [20]–[24] . Importantly , S1P is released by endothelial cells and platelets during the acute phase response and therefore plays an important role in the initial response to infection [25]–[28] . A specific transcriptional response to host derived sphingolipids and S1P has never been previously shown in bacteria . Here we have identified P . aeruginosa genes induced in response to mammalian pulmonary surfactant and subsequently characterized a subset of genes that are specifically and directly regulated by sphingosine or via metabolism of S1P , sphingomyelin , or ceramide to sphingosine . This response to sphingosine and its precursors is dependent on an AraC-family transcription factor in response to physiological levels of sphingosine and its precursors . This transcription factor binds sphingosine , which alters its association with DNA . A bacterial system to detect and respond to sphingosine may have broad implications in the modulation of host immune function and aid P . aeruginosa in altering host immune response in the human lung . In support of this prediction , deletion of the sphingosine-responsive transcription factor confers a survival defect during mouse lung infections .
Microarray studies were used to identify a group of P . aeruginosa transcripts that were induced when the bacteria were grown in minimal media supplemented with lung surfactant ( Survanta ) . When wild type PAO1 was exposed to minimal media containing lung surfactant compared to minimal medium with pyruvate alone , 125 transcripts ( both predicted open reading frames ( ORFs ) and intergenic regions ) were changed more than 3-fold ( p<0 . 05 ) , with 96 being induced ( Table S1 ) and 29 being reduced ( Table S2 ) . Of the induced transcripts , 56 were characterized and 40 were predicted or hypothetical , while in the reduced transcript group , 11 were characterized and 18 were predicted or hypothetical . The induced class was dominated by genes from the Anr-regulon ( 29 genes , 16 of which were recently demonstrated as induced in surfactant [29] ) and the choline catabolic pathway ( 15 genes ) [30]–[32] . One observation of note in the induced group is the preponderance of transcripts encoding stress-related proteins including the chaperones hslU , groEL , dnaK , and dnaJ , and the universal stress protein family members sspK , PA1789 , PA4352 , and PA5027 . We were interested in using the response to lung surfactant to identify gene function and novel biology in P . aeruginosa , and thus we have focused on highly induced genes of unknown function . The PA5325 transcript was induced ∼18 fold in the presence of lung surfactant and was the most highly induced transcript in these experiments ( Table S1 ) . The PA5325 gene is divergently transcribed from PA5324 , which encodes a probable AraC-family transcription factor that we hypothesized could be the transcriptional regulator of PA5325 ( Fig . 1A ) . The robust induction of PA5325 in the presence of lung surfactant ( Fig . 1B ) suggested a possible role of this gene in the early stages of lung infection by P . aeruginosa . For the following studies , we generated two reporter plasmids; pAL5 contained both sphR and the PA5325-lacZYA reporter and pAL4 contained only the PA5325-lacZYA reporter . Unless specified , the reporter used was pAL5 as it resulted in more robust induction . In addition to verifying the microarray results with surfactant , the PA5325-lacZ reporter was also induced in response to mouse fibroblasts ( L-cells ) and defibrinated sheep's blood ( Fig . 1B ) . To determine which component of these eukaryotic-derived mixtures was a specific inducer of PA5325 , we extracted mouse fibroblasts into aqueous and organic fractions and tested induction of PA5325-lacZ . The organic fraction contained the inducing activity , suggesting a lipid or other hydrophobic compound ( Fig . 1C ) . Mouse fibroblasts and sheep's blood both contain high percentages of sphingomyelin [33] , [34] , and sphingomyelin makes up ∼4% of lung surfactant . Therefore , we tested induction of PA5325 by sphingomyelin and related sphingolipids including ceramide , S1P , and sphingosine . PA5325 was induced by sphingomyelin , ceramide , S1P , and sphingosine , but not the likely degradation products of sphingosine: palmitate and glycine ( Fig . 1D ) . Other common lipid components of surfactant such as phosphatidylcholine and cholesterol did not induce transcription of PA5325-lacZ , and neither did unsaturated fatty acids ( Supplemental Fig . S1 ) . The strong induction by sphingosine compared to the other sphingolipids led us to hypothesize that the specific inducer of PA5325 is sphingosine . We tested the sensitivity of our PA5325-lacZ reporter to sphingosine ( Fig . 1E ) , which demonstrated PA5325-lacZ induction in a dose-dependent manner and showed response to physiological levels of sphingosine , which range from 200 nM ( as S1P ) in the serum and lymph up to 2–13 mM of free sphingosine in the skin and some epithelial surfaces [35] , [36] . We did not reach saturation in this assay due to a combination of sphingosine insolubility , plastic binding , and bactericidal effects ( discussed below ) . We hypothesized that the reduced induction of PA5325-lacZ in response to sphingomyelin , S1P , and ceramide compared to sphingosine was due to a processing step required by P . aeruginosa to yield sphingosine . To test this hypothesis we generated a clean deletion in the neutral ceramidase encoded by PA0845 and measured enzyme activity from the PA5325-lacZ reporter construct pAL4 . Ceramide fails to induce PA5325-lacZ in the absence of the neutral ceramidase , whereas the response to sphingosine was unaffected ( Fig . 1F ) . This finding strongly supports our hypothesis that induction of PA5325 occurs in response to sphingosine . In addition , PA5325-lacZ is induced in response to S1P in P . aeruginosa ( Fig . 1D ) , but not significantly induced by S1P in E . coli ( Fig . 1G ) , although the reporter in E . coli could still be induced in the presence of sphingosine ( Fig . 1G ) . This suggested that P . aeruginosa may be processing S1P and that E . coli does not possess an orthologous activity under these conditions . When S1P was pretreated with shrimp alkaline phosphatase , induction of PA5325-lacZ was partially restored in E . coli ( Fig . 1G ) . Given the transcriptional control of PA5325 in response to sphingosine , we have renamed it sphingosine regulated gene A , sphA . PA5324 encodes a predicted AraC-family transcription factor divergently transcribed from sphA ( Fig . 1A ) . This arrangement led us to suspect that PA5324 was the transcriptional regulator of sphA . To confirm the requirement of PA5324 for induction of sphA we generated an in-frame deletion of PA5324 . The PA5324 deletion strain carrying our sphA-lacZ reporter construct ( pAL4 ) showed no induction in the presence of sphingosine ( Fig . 2A ) . Insertion of PA5324 onto the chromosome at the attTn7 site restored induction in the deletion strain ( Fig . 2A ) . Furthermore , PA5324 was necessary to induce sphA-lacZ in a heterologous E . coli system in response to sphingosine ( Fig . 2B ) . Our data suggest that the sphingosine responsiveness via sphA transcription is dependent on PA5324 , and PA5324 was sufficient to confer sphingosine responsiveness in an E . coli system , therefore we have renamed PA5324 as the Sphingosine-responsive Regulator , SphR . Our a priori prediction was that SphR , controlling expression of a strongly induced gene by pulmonary surfactant , would be important for colonization and/or survival in the mammalian lung . To test this hypothesis , we examined bacterial survival 24 hours after infection in the mouse lung . The sphR deletion strain had significantly lower survival than wild type ( 7 . 7-fold decrease , Dunnett's multiple comparisons p<0 . 001 ) and the survival defect was complemented by addition of sphR at the attTn7 site ( Fig . 3 ) . In this comparison , wild type and ΔsphR both contained the empty attTn7 insertion cassette on the chromosome . The contribution of sphR to survival in the mouse lung led us to a more in-depth study of SphR and its target genes . Deletion of sphR resulted in reduced P . aeruginosa survival in the mouse lung ( Fig . 3 ) , leading us to hypothesize that one or more of the genes in the SphR regulon were likely candidates for this phenotype . To identify SphR-regulated genes in addition to sphA , we conducted microarray transcriptome analyses to compare wild type and the sphR deletion mutant in the presence and absence of pulmonary surfactant . Using a two-fold change cutoff and a p-value <0 . 05 , there are six genes that differ between wild type and ΔsphR in the presence of surfactant ( Table 1 ) . Transcripts that are induced in wild type but not in the sphR deletion mutant include sphA , the neutral ceramidase ( PA0845 ) , and a three gene operon convergently transcribed toward sphA , PA5328-PA5326 . The argB gene ( PA5323 ) was induced more strongly in the sphR deletion than in wild type , which we think is likely due to a cis effect of the sphR ( PA5324 ) deletion , as these genes are convergently transcribed ( Fig . 1A ) . To denote their placement in the SphR regulon , we have renamed the genes in the predicted PA5328-PA5326 operon as sphBCD . The sphB gene encodes a predicted periplasmic cytochrome and sphC and sphD encode a predicted flavin-dependent oxidoreductase and a predicted pyridoxalphosphate-containing threonine aldolase-like enzyme , respectively . The predicted functions of SphC and SphD suggest a potential two-step pathway for sphingosine degradation to glycine and a long chain aldehyde by oxidation to an aldol and subsequent cleavage by the aldolase , a prediction we are currently exploring . The neutral ceramidase ( PA0845 ) was previously designated PaCD [37] , which does not conform to standard bacterial nomenclature . We propose that PA0845 be renamed cerN for ceramidase , neutral . The induction of sphA by surfactant in wild type ( 17 . 8-fold ) versus the difference of sphA induction between wild type and ΔsphR ( 5 . 9-fold ) suggested altered regulation of sphA in the absence of sphR ( Table 1 ) . The relative induction of sphA in the sphR mutant compared to wt under pyruvate ( non-inducing ) conditions supports a de-repression of sphA transcription in the absence of sphR at baseline . The remaining genes in the operon appear solely regulated by SphR under these conditions , as their induction levels in wild type compared to the difference between wild type and ΔsphR are not different . We used promoter mapping to identify the promoter proximal regions of the sphA , sphBCD , and cerN promoters that were important for sphingosine and sphR-dependent regulation . Using lacZ reporter fusions to each upstream region , we identified a portion of each promoter-proximal region required for responsiveness to sphingosine ( Fig . 4A ) . The regions required for sphingosine responsiveness were aligned using KALIGN [38] , which produced an alignment that highlights the general format of an AraC-family binding site ( Fig . 4B ) . The MEME consensus for a single half-site is shown below the alignment ( Fig . 4B ) . Bioinformatic search of the P . aeruginosa genome ( DNA Motif Search [39] ) turned up only one additional predicted binding site ( two direct repeats of the consensus ( TGNCCSNNRNNSNCC ) separated by 6–8 bp ) in the genome apart from those present in the three identified promoters . The additional binding site is in the intergenic region between PA0428 and PA0429 , upstream of the PA0428 gene . We did not detect any change in the PA0428 transcript for wild type or ΔsphR in the presence of surfactant or in either strain in the absence of surfactant . Therefore , based on our microarray data and bioinformatic analysis , we predict that sphA , sphBCD , and cerN likely comprise the core SphR regulon . The upstream sequences for the SphR regulon members showing the predicted SphR binding sites , promoter elements , and ribosome binding sites are shown in Supplemental Figure S2 . To test both specificity and the importance of conserved consensus sequences we mutated the first two residues in the consensus sequence TG to AA in half-site 1 ( sphA** ) ( Fig . 4B ) , and tested the ability of the mutant sequence to permit induction of the reporter gene in response to sphingosine . The mutant reporter was unable to support reporter induction in response to sphingosine ( Fig . 4C ) , demonstrating the importance of these conserved binding site residues . We conducted electrophoretic mobility shift assays ( EMSAs ) with purified MBP-SphR fusion protein to test if SphR directly bound the sphA , sphBCD , and cerN promoters . The binding of MBP-SphR to the sphA promoter probe was greatly enhanced by the addition of sphingosine to the binding reaction in a concentration-dependent manner , providing evidence that sphingosine was a direct ligand of SphR ( Fig . 5A ) . In the presence of sphingosine , MBP-SphR specifically shifted the sphA , sphBCD , and cerN promoters in a protein concentration-dependent manner and the binding could be competed with unlabeled sphA promoter probe , which gives a sense of the relative affinities for each binding site ( Fig . 5B ) . MBP-SphR did not shift the non-specific plcH probe ( Fig . 5B ) . The plcH probe is a useful negative control and demonstrates the specificity of SphR binding , as it has a known binding site for the AraC-family transcription factor GbdR in P . aeruginosa and its regulation is well described [31] , [40]–[43] . To test the predicted SphR binding site , 59-mer oligonucleotides containing the proposed SphR binding site from the sphA promoter were annealed and the resultant probe was used in binding reactions . MBP-SphR was able to shift the 59-bp sphA probe ( Fig . 5C , left ) , but only in the presence of sphingosine . Based on the inability of the mutated consensus sequence ( sphA** ) to support sphingosine-dependent reporter expression ( Fig . 4C ) , we predicted that an oligonucleotide carrying these mutations would also be unable to bind SphR . As shown in the right side of Figure 5C , MBP-SphR was unable to bind this mutated probe . Together with the reporter fusions , these data support both the specificity of SphR binding and the importance of the conserved residues in the consensus . Based on the enhancement of SphR DNA binding in the presence of sphingosine and our genetic evidence , we predicted that SphR would directly bind sphingosine . We used 3H-sphingosine to test the ability of SphR to bind sphingosine ( Fig . 6 ) . The binding assay conditions were similar to those used for EMSA studies with MBP-SphR in the presence of 3H-sphingosine . Amylose resin beads were used to pull down the MBP-SphR , and bead-associated sphingosine was assayed by liquid scintillation counting . 3H-sphingosine was substantially enriched in the fraction containing amylose-bound MBP-SphR , while relatively little remained associated with the amylose beads alone , or beads bound to a non-specific MBP-tagged P . aeruginosa AraC-family transcription factor , CdhR ( MBP-CdhR ) [44] . These data , in combination with the EMSAs ( Fig . 5 ) , demonstrate direct interaction between sphingosine and SphR . Because deletion of sphR led to reduced survival in the mouse lung , we were interested in determining which of the SphR regulon members contributed to survival in the lung . We generated deletions in cerN , sphA , and sphC and compared to wild type in our 24 hour lung infection model . Deletion of sphA led to a significant reduction in bacterial survival in the mouse lung ( 9-fold decrease , Dunnett's multiple comparisons p<0 . 001 ) , while deletion of cerN or sphC had no impact on bacterial survival in vivo ( Fig . 7 ) . The sphA mutant phenotype could be complemented by supplying the sphA under its native promoter control at the attTn7 site ( Supplemental Fig . S3 ) . These data suggest an important role for sphA in survival during infection . We did not test deletions of sphB and sphD in the animal model , given their predicted coordinate role with sphC in sphingosine metabolism and their similar phenotype to an sphC deletion during in vitro sphingosine killing ( Fig . 8 and Figure S4 ) . Sphingosine has previously been shown to have antimicrobial properties and is able to inhibit growth and kill many Gram positive and Gram negative bacteria [19] . Previous studies suggest that P . aeruginosa is not sensitive to killing by sphingosine [45] . We hypothesized that SphR might play a role in the response of P . aeruginosa to sphingosine and could regulate sphingosine resistance . Using a modified sphingosine killing assay , we show that the ΔsphR deletion strain is more sensitive to sphingosine compared to wild type ( Fig . 8 ) , an effect that could be complemented by sphR on a plasmid ( Supplemental Fig . S4 ) . Most of the sensitivity of the ΔsphR strain appears to be due to loss of sphA induction , as the ΔsphA strain is also more sensitive to sphingosine than wild type and is nearly as sensitive as ΔsphR ( Fig . 8 ) . The deletion phenotype of sphA could be complemented by sphA on a plasmid ( Supplemental Fig . S4 ) . Deletion of sphC and transposon insertions into sphD and sphB also led to small but reproducible decreases in survival on sphingosine , suggesting a minor role for this operon in the response to sphingosine ( Fig . 8 and Supplemental Fig . S4 ) . Deletion of cerN , befitting its known function as an extracellular ceramidase , had no effect on survival in sphingosine ( data not shown ) .
The induction of ceramidase activity in response to sphingosine has been demonstrated in a few bacteria [37] , but the mechanism of sphingosine detection and conversion into a response had not previously been elucidated . In this study we show that sphingosine is directly detected by the AraC-family transcription factor SphR ( PA5324 ) leading to the induction of sphA , sphBCD , and cerN transcripts . Deletion of sphR or sphA resulted in survival defects in a mouse model of acute pneumonia , suggesting that the ability to detect and respond to host-derived sphingolipids is important for survival in the lung . Sphingolipids are abundant in mammals , plants , and fungi , constituting a diverse family of molecules that serve as essential structural components of eukaryotic cell membranes and as dynamic signaling molecules that mediate diverse cellular functions [16]–[19] . In particular , S1P has been implicated as a critical component of mammalian innate and adaptive immune function , particularly in the acute phase response to pathogens [25]–[28] . Interestingly , orthologs of SphR and some of the SphR-regulon members are present in other opportunistic pathogens including Acinetobacter haemolyticus and Burkholderia pseudomallei , as well as the professional pathogen Mycobacterium tuberculosis . Sphingolipids play important roles in host-pathogen interactions , particularly S1P and ceramide signaling [46]–[48] . In addition to host modulation of sphingolipid pathways to combat infection , pathogens can modulate host sphingolipids . M . tuberculosis alters sphingolipid signaling in macrophages by undetermined mechanisms [49] , and S1P levels in the lungs of patients infected with M . tuberculosis are significantly decreased [50] . Interestingly , M . tuberculosis has an AraC-family transcription factor that is 47% similar along the whole length to SphR ( RV1395 ) that was identified though signature-tagged mutagenesis where the RV1395 transposon mutant strain had an ∼1 . 5 log reduced survival in a mouse lung infection model [51] . Similarity between RV1395 and SphR is not restricted to the helix-turn-helix DNA-binding domain , as the two proteins are 44% similar when the DNA-binding domain is removed from the alignment analysis . RV1395 was characterized and found to be an activator of a divergently transcribed cytochrome gene , however the signals that govern RV1395 activation and its direct contribution to virulence have yet to be determined [52] . Based on the similarity of RV1395 to SphR we predict that a sphingolipid , perhaps sphingosine , may be the inducing ligand of RV1395 . The AraC-family transcription regulators are one of the largest groups of regulatory proteins in bacteria , and are often involved in the regulation of catabolism , stress response , and virulence [53] . Many members of the AraC family have been shown to respond to host-derived chemical signals present at the site of infection , but relatively few inducing ligands have been demonstrated to bind directly to their cognate regulator [54] . We found that addition of sphingosine altered the binding of SphR to the sphA promoter in EMSA studies and observed a dose response curve of SphR DNA binding at physiologically relevant concentrations of sphingosine . Bioinformatic analysis suggest similarity of SphR to ToxT ( 44% similarity and 20% identity ) , which directly regulates the major virulence factors in Vibrio cholerae . ToxT activation is inhibited by unsaturated fatty acids found in bile [55] . Subsequently , the crystal structure of ToxT was solved revealing a bound 16-carbon fatty acid that alters the structure of ToxT to prevent DNA binding in the presence of these bile associated fatty acids [56] . The similar size and hydrophobic nature of the regulatory ligands ( palmitate vs . sphingosine ) coupled with the sequence similarity allows us to speculate that SphR may bind sphingosine in a manner analogous to ToxT binding of palmitate . Ito et al . identified a neutral ceramidase encoded by PA0845 ( renamed cerN in this study ) that was induced in the presence of sphingomyelin , ceramide and sphingosine , however the regulatory mechanism was not reported [37] . The discovery of SphR control of neutral ceramidase allows us to expand a model of bacterial utilization of sphingomyelin by linking it to our previous work on regulation of the phospholipase C/sphingomyelinase PlcH . We previously characterized the AraC-family regulator GbdR that is integral to a positive feedback loop controlling PlcH expression in response to a metabolite of the choline headgroup of sphingomyelin [31] . Sphingomyelin hydrolysis by PlcH yields ceramide [57] , which P . aeruginosa can further metabolize through the action of ceramidases [54] . Here we show that CerN is produced as part of an SphR-dependent positive feedback loop in response to the ceramide metabolite sphingosine , in a manner analogous to GbdR control of PlcH . Both of these positive feedback loops link induction of secreted catabolic enzymes not to the availability of the substrate itself , but to metabolic products derived from the substrate . In each case , this ensures that the positive feedback loop will robustly operate only if the substrate is being metabolized at sufficient rates . Sphingolipids such as sphingosine have long been known to have antimicrobial properties and sphingosine is found in high concentration in the skin where it is thought to be part of the barrier function against microbial infections [58]–[61] . A variety of Gram positive and Gram negative bacteria are sensitive to sphingosine , including Staphylococcus aureus and Escherichia coli [62] . The precise bactericidal mechanism of sphingosine remains unknown . However , recent evidence suggests that sphingosine may directly damage bacterial membranes [63] . P . aeruginosa has recently been reported to be resistant to the bactericidal effects of sphingosine [45] . While none of the deletion strains generated in this study showed growth defects under normal conditions , we found that both the sphR and sphA deletion strains were susceptible to the antimicrobial effects of sphingosine compared to wild type in vitro . Strains with deletions in sphR and sphA were also shown to have reduced survival in the mouse lung . We hypothesize that the sensitivity of sphA and sphR mutants to sphingosine contributes to their observed reduced survival in vivo . It is interesting to note that the double deletion ΔcerNΔsphA strain did not survive better or worse than ΔsphA , minimally suggesting that if the defect is due to sphingosine sensitivity , it is not sphingosine derived from P . aeruginosa hydrolysis of host-derived ceramide; in other words , they are not causing their own death by sphingosine derived from sphingomyelin and ceramide hydrolysis . Therefore , while the in vitro sphingosine killing correlates well with the in vivo phenotypes , we currently do not know the mechanism governing reduced survival of the sphR and sphA mutants in the lung . We speculate that SphR responds to sphingosine to induce transcripts encoding proteins that protect P . aeruginosa from the bactericidal effects of sphingosine by induction of membrane stabilizing factors and/or catabolism of sphingosine to non-bactericidal metabolites . Here we show that SphR binds to sphingosine to initiate transcription of sphA , sphBCD and cerN . sphA encodes a hypothetical protein with some homology to proteins involved in meta-pathway phenol degradation . Protein localization predictions for SphA using the structure similarity-based prediction of Phrye2 [64] suggests that SphA is an outer membrane porin . Perhaps P . aeruginosa responds to sphingosine by providing a porin for sphingosine import and subsequent degradation that could aid in protecting the outer membrane from the damaging effects of free sphingosine . Okino and Ito demonstrated sphingosine utilization by P . aeruginosa by measuring removal of sphingosine from the culture supernatants and cell fractions [54] . Based on bioinformatic predictions , SphB , SphC and SphD are most likely involved in the metabolism of sphingosine . The sphB gene encodes a predicted cytochrome , while sphC encodes an FMN-linked oxidoreductase , and sphD encodes a pyridoxalphosphate serine-threonine aldolase . The latter two activities could work in concert to oxidize carbon 1 , generating an aldol , which SphD could hypothetically act upon , rendering a long chain aldehyde and glycine . Transposon insertion into the sphC coding sequence ( PA5327 ) resulted in reduced bacterial survival in a chronic rat lung infection model [65] , suggesting that while our sphC deletion strain did not show a phenotype in the acute mouse lung infection ( Fig . 7 ) , it nonetheless impacts survival in the mammalian lung . The microarray data comparing wild type in the presence and absence of pulmonary surfactant suggests some interesting biology in the presence of surfactant . The first observation has been covered by Jackson et al . , who recently analyzed the changes in transcript levels of P . aeruginosa exposed to pulmonary surfactant , and compared wild type to both plcH and gbdR mutants [29] , but did not publish results of these strains in the absence of pulmonary surfactant . They noted a reduction in transcript levels for Anr-controlled genes in both the gbdR and plcH mutants grown in surfactant , as do we ( Table S1 ) . Given the high levels of phosphatidylcholine and sphingomyelin in pulmonary surfactant , it was not surprising that the transcripts encoding proteins from the choline catabolic pathway were also highly induced in the presence of surfactant ( Table S1 ) . In addition to the high proportion of transcripts encoding stress-related proteins ( mentioned in the Results section ) , there are also a high proportion ( ∼8% ) of transcriptional regulators: NalC , BetI , NirG , PsrA , NarL , CgrA , PA3458 , and PA4596 . It is possible that the effects of induction of these transcription factors is contained in our regulation data , however our transcriptome analyses were a snapshot of transcripts at four hours post-induction and effects from changes in these transcription factors may not have sufficiently accumulated in the transcriptome . Of the reduced transcripts ( Table S2 ) , we note that three of the pyrroquinoline quinine biosynthesis genes are down , suggesting a change in requirement for this cofactor between surfactant and pyruvate conditions . The demonstration of sphingosine detection by P . aeruginosa also opens up the possibility that this bacterium , and others with similar detection systems , could alter sphingosine and related sphingolipid signals , including S1P in the host . We have not yet examined the contribution of host immune signaling effected by the SphR regulon , but the impact of altering such an important and tightly controlled signaling network by bacterial factors has not been elucidated and may be an important contributing factor to the survival of P . aeruginosa in vivo .
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 for animal infection was approved by the University of Vermont Institutional Animal Care and Use Committee ( Permit number A3301-01 ) . All procedures were performed under pentobarbital anesthesia and all efforts were made to minimize animal suffering . P . aeruginosa PAO1 , isogenic mutant strains , and E . coli ( Table 2 ) were maintained in LB-Lennox ( LB ) medium . Morpholinepropanesulfonic acid ( MOPS ) medium [66] supplemented with 25 mM sodium pyruvate , 5 mM glucose and 50 µg/ml gentamicin ( for P . aeruginosa ) or MOPS with 10% LB ( v/v ) , 5 mM glucose , and 10 µg/ml gentamicin ( for E . coli ) was used to grow strains prior to transcriptional induction studies . See LaBauve and Wargo ( 2012 ) for further details on P . aeruginosa growth methods [67] . For bactericidal assays , 1% neopeptone was supplemented with varying sphingosine concentrations in ethanol to reach a final concentration of 6 . 25% ( w/v ) ethanol in the assay . All lipids were purchased from Avanti Polar Lipids and other chemicals were purchased from Sigma-Aldrich or Fisher . We used the oropharyngeal route of mouse lung infection previously described [68] , [69] . Briefly , P . aeruginosa PAO1 and isogenic strains were streaked onto LB plates from −80°C stocks . Colonies from the first plate were restreaked onto a new LB plate after 24 hours and incubated at 37°C for 24 hours . Cells from the second plate were used to start 3 ml cultures in LB that were grown for 16–18 hours at 37°C on a roller drum . From these overnight cultures , cells were collected by centrifugation , washed in Dulbecco's PBS ( DPBS ) , and resuspended to give ∼1×107 viable P . aeruginosa in 40 µL , with actual inoculum determined by serial dilution and plate counting . Eight to twelve week old male C57Bl/6J mice ( Jackson Labs ) were inoculated with 40 µL of the bacterial suspension via oropharyngeal aspiration . Anesthesia , surgery , bronchoalveolar lavage fluid ( BALF ) collection , organ harvest , and organ homogenization were done as previously described [68] , [69] at 24 hours post-infection . Viable bacterial counts in organs were determined by serial dilution plating onto Pseudomonas Isolation Agar ( PIA ) ( BD-Difco ) followed by incubation at 37°C for 24 hours . Mouse experiments ( Fig . 3 and 7 and Supplemental Fig . S3 ) show CFU counts from all animals from duplicate experiments with each replicate having 4–6 animals per experimental group . All informative comparisons: mutants versus wild type ( both Figures ) and mutant versus complementation strain ( Fig . 3 and Supplemental Fig . S3 ) were conducted in at least one additional experiment , included with comparator strains from other studies . Therefore , all informative comparisons were assessed three times . All experiments met the same statistical criteria , i . e . all replicates were consistent with regards to effect size and significance of changes . Inoculation order and harvest order alternated between experiments to eliminate potential issues related to the difference between the duration of inoculation ( ∼20–30 min ) and the duration of harvest ( ∼1 . 5 h ) . For group comparisons , data ( log10 transformed CFU counts ) were analyzed by ANOVA followed by Tukey's ( Fig . 3 and Supplemental Fig . S3 ) or Dunnett's ( Fig . 7 ) Multiple Comparisons tests . All calculations were done using GraphPad Prism . P . aeruginosa PAO1 wild type and ΔsphR were grown overnight in MOPS media supplemented with 20 mM pyruvate and 5 mM glucose . Overnight cultures were collected by centrifugation and resuspended in either MOPS supplemented with 20 mM pyruvate alone or 20 mM pyruvate and a 1∶50 dilution of the bovine surfactant preparation Survanta ( Abbott ) and induced for 4 hours at 37°C . Bacteria were collected by centrifugation , resuspended in MOPS and RNA Protect Bacterial Reagent ( Qiagen ) , and the resultant pellets stored overnight at −20°C . RNA was extracted using an RNeasy kit ( Qiagen ) , and eluted samples were treated with DNase I followed by a second round of RNeasy purification including an on-column DNase I treatment . Purified RNA samples were checked for DNA contamination by PCR and RNA integrity scores based on Agilent Bioanalyzer analysis were indicative of little to no DNA contamination . Microarray analysis was performed on a Pseudomonas aeruginosa PAO1 gene chip using raw oligonucleotide probes generated from each condition using the NuGen Pico system . Each condition was analyzed in duplicate ( N = 2 ) , and summarized in one probe intensity by the Vermont Genetics Network Microarray Facility using Affymetrix GCOS software . Information from multiple probes was combined to obtain a single measure of expression for each probe set and sample . Probe-level intensities were background-corrected , normalized , and summarized , and Robust Multichip Average ( RMA ) statistics were calculated for each probe set and sample as is implemented in Partek Genomic Suites , version 6 . 6 ( Copyright 2009 , Partek Inc . , St . Louis , MO , USA ) . Sample quality was assessed based on relative log expression ( RLE ) , and normalized unscaled standard error ( NUSE ) . To identify differentially expressed genes , linear modeling of sample groups was performed using ANOVA as implemented in Partek Genomic Suites . The magnitude of the response ( fold change calculated using the least square mean ) and the p-value associated with each probe set and binary comparison were calculated . The data have been submitted to NCBI GEO with accession number GSE48982 . Deletion mutants were generated using the pMQ30 plasmid [70] carrying the flanking regions of each of the four genes , sphR , sphA , sphC , and cerN , using conjugation-mediated deletion as described previously [30] , [69] . Primers for these constructs are listed in Table S3 . Single cross-over mutants were selected on PIA with gentamicin and selection of double crossover deletion mutants were carried out on LB 5% sucrose plates prepared without NaCl . Unmarked deletion mutants were verified using PCR . Complementation was done by integration of the sphR or sphA coding sequence under control of their native promoter at the attTn7 locus using the pUC18-miniTn7T-Gm vector as we described previously [68] , [69] using the method of Choi and Schweizer [71] . This allowed stable complementation in the absence of antibiotic . For complementation where reporter plasmids were used , the gentamicin resistance cassette was excised by FLP-mediated recombination [71] . All sphR::attTn7 and sphA::attTn7 complementation strains were compared with wild type or mutant strains carrying the empty attTn7 integration region from the pUC18-miniTn7T-Gm vector . Two reporter constructs were generated in this study using yeast homologous recombination [70] to generate translational fusions to lacZYA . A target lacZYA-containing vector suitable for yeast cloning ( pMW42 ) was generated by excising the lacZYA region from pMW5 [31] with HindIII and EcoRI and cloning into the similarly cut pMQ80 backbone [70] , which removes egfp-mut3 . Either the sphA promoter ( pAL4 ) , or the entire sphR gene and the sphA promoter ( pAL5 ) were recombined with pMW42 linearized with KpnI and HindIII . P . aeruginosa strains were electrotransformed with the reporter constructs and grown overnight in MOPS media supplemented with 20 mM pyruvate , 5 mM glucose , and 50 µg/ml gentamicin prior to induction . Inductions were carried out in MOPS media supplemented with 20 mM pyruvate and the inducing compound and incubated at 37°C for 6 hours . β-galactosidase assays were done as previously described [31] , [72] , using the method of Miller [73] . Studies of heterologous sphA induction in E . coli were carried by transforming pAL4 and pAL5 into E . coli NEB5α . Resulting E . coli strains were grown overnight in MOPS media supplemented with 10% LB ( v/v ) , 5 mM glucose and 10 µg/ml gentamicin . For induction assays with S1P in E . coli , 2 . 4 µg of S1P or sphingosine were pre-treated with or without 5 U shrimp alkaline phosphatase ( SAP ) in 100 µL of water with 1× SAP buffer ( USB ) , and incubated at 37°C for 60 minutes . Induction assays were carried out in MOPS supplemented with 10% LB ( v/v ) , treated inducing compounds , and 10 µg/ml gentamicin . All E . coli strains were induced for 8 hours prior to ß-galactosidase assays . Full-length reporter constructs and truncations of sphA , sphB , and cerN promoters were cloned into pMW5 [31] . The resultant lacZYA reporter constructs were transformed into wild type P . aeruginosa and used to identify the region required for response to sphingosine . Inductions were carried out in MOPS media supplemented with 20 mM pyruvate and 150 µM sphingosine and incubated at 37°C for 6 hours followed by ß-galactosidase assays . We constructed a maltose binding protein ( MBP ) fusion to SphR by using the pMALc2 vector system ( NEB ) . The sphR gene was amplified from genomic DNA . The PCR product was gel purified and ligated into the pCR Blunt vector ( Invitrogen ) . The insert was excised with KpnI and HindIII , gel purified , and ligated into a similarly digested pMALc2 vector to generate pAL11 . E . coli NEB5α ( New England Biolabs ) carrying the pAL11 plasmid were grown overnight in LB supplemented with 120 µg/ml carbenicillin . The overnight culture was transferred to two 500 ml flasks containing 100 ml of LB-carbenicillin and shaken at 220 rpm for 5 hours . Isopropyl-β-D-thiogalactopyranoside ( IPTG ) was added to a final concentration of 1 mM , and the cells were induced for 3 hours . Cells were collected by centrifugation , lysed in column buffer ( 20 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 1 mM EDTA ) supplemented with 3 mg/ml lysozyme and Halt protease inhibitor 1× cocktail ( Thermo Scientific ) . Lysates were clarified by centrifugation , and the soluble fraction was applied to a column containing amylose resin ( NEB ) . The column was washed with ten volumes of column wash buffer ( 20 mM Tris-HCl , 150 mM NaCl 1 mM EDTA pH 7 . 4 ) , followed by elution with column wash buffer supplemented with 10 mM maltose . Elution fractions were run on 10% SDS-PAGE gels and visualized by Coomassie staining . Fractions containing the MBP-SphR were pooled and dialyzed against 20 mM Tris-HCl , pH 7 . 5 at 4°C in a 20 , 000 kDa cutoff Slide-A-lyzer cassette ( Pierce ) . The full length MBP-SphR fusion protein was used in electrophoretic mobility shift assays , as the MBP tag did not prevent sequence specific DNA binding ( Fig . 5 ) or binding to sphingosine ( Fig . 6 ) . EMSA DNA probes were generated using PCR ( Primers in Table S3 ) and were spot dialyzed against 2 . 5 mM Tris-HCl , 0 . 25 mM EDTA , pH 8 . 0 . Labeled probes , generated using a primer with a covalently linked 5′ biotin tag ( IDT ) , were used at 0 . 5 fmol/µl , and unlabeled competitor probes were used at a final concentration of 0 . 5 pmol/µl . EMSA was carried out using a Thermo Scientific Thermoshift kit . The final binding buffer was modified to contain 1× binding buffer ( 10 mM Tris-HCl , pH 7 . 5 , 50 mM KCl , 1 mM dithiothreitol ) , 0 . 1 mM glycine betaine , and 2 µg/ml poly-dI-dC . Various concentrations of sphingosine dissolved in ethanol were added to reaction tubes and allowed to dry to eliminate ethanol prior to binding reactions . Binding reactions were carried out at 37°C for 15 minutes and electrophoresed on a 5% non-denaturing polyacrylamide gel then transferred to a BioDyne B membrane ( Thermo Scientific ) . Detection was carried out using streptavidin-linked horseradish peroxidase according to the supplied protocol ( Thermo Scientific ) . Sphingosine association with SphR was measured by conducting binding reactions using 3H-D-erytho-sphingosine ( Perkin-Elmer ) . Binding reactions were carried out as described for EMSA except 3H-D-erytho-sphingosine was used at a final concentration of 50 nM . Samples were incubated with and without either 10 µM MBP-SphR or 10 µM MBP-CdhR for 30 minutes then added to amylose resin . The amylose beads were collected by centrifugation and washed 3 times with amylose column wash buffer . After washes , amylose beads were resuspended in 200 µl of amylose wash buffer and transferred to a glass vial containing 10 ml of Biosafe II scintillation cocktail ( RPI ) . Samples were quantified using a Tri-Carb 2910 TR liquid scintillation analyzer ( Perkin-Elmer ) . Killing assays were carried out as previously described [61] . Briefly , overnight P . aeruginosa strains were grown in trypticase soy broth ( TSB ) and diluted 1∶40 . Diluted cultures ( 100 µl ) were added to glass tubes containing 250 µl of 1% neopeptone supplemented with 50 µl of the appropriate sphingosine stock in ethanol or ethanol alone as the vehicle control . The cultures were shaken at 170 rpm for one hour . Survival was determined by serial dilution plating on PIA . Colonies were counted after 24 hour incubation and survival calculated by comparison to vehicle only controls . | Many opportunistic pathogens transition from an environmental niche into the host . To establish an infection , these bacteria must rapidly adapt their transcriptional profile to the conditions at the site of infection . We used the response of Pseudomonas aeruginosa to lung surfactant as a model to discover genes important for bacterial survival during mouse lung infection . Using this model we identified transcripts induced in response to host-derived sphingolipids , accomplished by detection of the core component sphingosine by a sphingosine-binding transcription factor , SphR . Deletion of this transcription factor in P . aeruginosa reduced bacterial survival , highlighting the importance of a proper response to host-derived sphingosine . We present evidence that impaired survival against the antimicrobial effects of sphingosine may explain part of the in vivo survival defect of mutants in this response system . This is the first description of a specific bacterial response to sphingosine and its precursors , some of which are important immune signaling molecules . Thus , P . aeruginosa is capable of intercepting and responding to host immune modulatory signals . The importance of this response during infection and the presence of similar systems in other pathogens opens up a new avenue for investigation and expands our understanding of bacterial metabolic interactions with the host . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"microbial",
"metabolism",
"lipids",
"gram",
"negative",
"sphingolipids",
"gene",
"expression",
"genetics",
"gene",
"regulation",
"molecular",
"genetics",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"bacterial",
"pathogens",
"dna",
"transcription"
]
| 2014 | Detection of Host-Derived Sphingosine by Pseudomonas aeruginosa Is Important for Survival in the Murine Lung |
Entamoeba histolytica is an obligate protozoan parasite of humans , and amebiasis , an infectious disease which targets the intestine and/or liver , is the second most common cause of human death due to a protozoan after malaria . Although amebiasis is usually asymptomatic , E . histolytica has potent pathogenic potential . During host infection , the parasite is exposed to reactive oxygen species that are produced and released by cells of the innate immune system at the site of infection . The ability of the parasite to survive oxidative stress ( OS ) is essential for a successful invasion of the host . Although the effects of OS on the regulation of gene expression in E . histolytica and the characterization of some proteins whose function in the parasite's defense against OS have been previously studied , our knowledge of oxidized proteins in E . histolytica is lacking . In order to fill this knowledge gap , we performed a large-scale identification and quantification of the oxidized proteins in oxidatively stressed E . histolytica trophozoites using resin-assisted capture coupled to mass spectrometry . We detected 154 oxidized proteins ( OXs ) and the functions of some of these proteins were associated with antioxidant activity , maintaining the parasite's cytoskeleton , translation , catalysis , and transport . We also found that oxidation of the Gal/GalNAc impairs its function and contributes to the inhibition of E . histolytica adherence to host cells . We also provide evidence that arginase , an enzyme which converts L-arginine into L-ornithine and urea , is involved in the protection of the parasite against OS . Collectively , these results emphasize the importance of OS as a critical regulator of E . histolytica's functions and indicate a new role for arginase in E . histolytica's resistance to OS .
Amebiasis is a parasitic infection of the intestines and is mainly caused by fecal contamination [1] . Although 90% of infected individuals are asymptomatic , amebic dysentery affects 50 million people in India , Southeast Asia , Africa , and Latin America and amebiasis is the cause of at least 100 , 000 deaths each year [2 , 3] . Following excystation within the small intestinal lumen , trophozoites colonize the large intestine and they usually reside in the colon as a non-pathogenic commensal in most infected individuals . Due to as yet unidentified causes , these trophozoites can cause amebic dysentery , become virulent and invasive , and migrate to the liver , via the portal veins , where they cause hepatocellular damage . Following host invasion , invading E . histolytica trophozoites are challenged by oxidative stress ( OS ) and nitrosative stress ( NS ) , which originate from fluctuations in ambient oxygen tension in the intestinal lumen and the generation of reactive oxygen species ( ROS ) and reactive nitrogen species ( RNS ) by cells of the immune system . Once formed , these reactive species can oxidatively damage proteins and change their structural conformation and functional activity [4] , [5] , [6] . The parasite's complex response to OS involves modulation of a large number of genes which encode proteins that are associated with signaling/regulatory and repair/metabolic pathways and proteins whose exact functions are still unknown [7] . It has been recently reported that the expression of these genes is regulated by a recently identified transcription factor that binds to a specific promoter motif of hydrogen peroxide ( H2O2 ) -responsive genes [8] . It has also been reported that those genes in E . histolytica which confer resistance to OS also contribute to its virulence [9] . Since antioxidant enzymes , such as catalase , glutathione reductase , and γ-glutamyl transpeptidase , are missing from E . histolytica's enzyme resource [10] , one of the functions of proteins , such as the 29-kDa peroxiredoxin [11] and the iron-containing peroxide dismutase [12] , is to protect the parasite against OS . Since OS glycolysis is inhibited and metabolic flux is redirected towards glycerol production in oxidatively stressed E . histolytica trophozoites , these findings suggest that the glycerol synthesis pathway is a component of the parasite's metabolic antioxidative defense system [13] . Despite these informative data on the parasite's response to OS , our knowledge on the identity of oxidized proteins in E . histolytica is still incomplete . Here , we report the results of a study whose aim was to identify and to determine the biological relevance of oxidized proteins ( OX ) in E . histolytica using resin-assisted capture ( RAC ) coupled with mass spectrometry ( MS ) [14] . The results of this analysis revealed 154 OXs which include antioxidant proteins , cytoskeleton proteins , protein involved in translation , and transport proteins . We also found that oxidation of cysteine residues in the carbohydrate recognition domain ( CRD ) of the 260-kD heterodimer and multifunctional virulence factor of E . histolytica , Gal/GalNAc lectin ( gl ) , impairs its ability to adhere to host cells . We also found that arginase , the enzyme which converts L-arginine into L-ornithine and urea , confers resistance to OS in E . histolytica .
E . histolytica trophozoites strain HM-1:IMSS were grown under axenic conditions in Diamond's TYI S-33 medium at 37°C . Trophozoites in the exponential phase of growth were used in all experiments . For the construction of the pJST4-arginase expression vector , arginase was amplified by polymerase chain reaction ( PCR ) using the primers , Arginase KpnI and Arginase BglII ( table 1 ) . The PCR product was subcloned using the pGEM-T Easy vector system ( Promega ) and then digested with the restriction enzymes KpnI and Bgl II . The digested DNA insert was cloned into the E . histolytica expression vector pJST4 which had been previously linearized with KpnI and Bgl II . The pJST4 expression vector ( pcontrol ) contains a tandem affinity purification tag for use in protein purification and identification [15] . This CHH-tag contains the calmodulin binding protein , hemagglutinin ( HA ) , and histidine ( His ) residues and its expression is driven by an actin promoter . This vector was used as control in our experiment in order to exclude the possibility that the CHH tag is responsible for the phenotypes of the arginase-overexpressing strain . A previously described protocol was used to transfect E . histolytica trophozoites [16] . E . histolytica trophozoites ( 1x106 ) were exposed to 1 mM , 2 . 5 mM , 5 mM , 7 mM , or 10 mM H2O2 for 60 minutes at 37°C . At the end of the exposure , a 10-μl aliquot of each culture was stained with eosin ( 0 . 1% final concentration ) , and the number of living trophozoites was counted in a counting chamber under a light microscope . Resistance of the control , trophozoites overexpressing arginase and pcontrol trophozoites to OS was measured by calculating the median lethal dose ( LD50 ) of hydrogen peroxide ( H2O2 ) by linear regression analysis using Microsoft Excel . The assay was repeated three times with two replicates in each assay . Control and oxidatively stressed E . histolytica trophozoites were incubated with 0 . 4 mM ( final concentration ) 2 , 7-dichlorofluorescein diacetate ( H2DCFDA; Sigma ) for 15 minutes in the dark . The cells were washed twice in phosphate buffered saline ( PBS; pH 7 . 4 ) and immediately examined under a Zeiss Axio Scope . A1 fluorescence microscope . Intracellular ROS levels were determined by measuring fluorescence intensity using the ImageJ software [17] . E . histolytica trophozoites ( 5x107 ) were incubated with 2 . 5 mM H2O2 for 60 minutes at 37°C . At the end of the incubation , a total protein extract was prepared by lysing the oxidatively stressed trophozoites with 1% Igepal ( Sigma ) in PBS . OXs in the extract were detected by OX-RAC using a previously described protocol [14] with minor modifications . Briefly , the total protein extract ( 9 mg ) was incubated in mixture of 50 mM N-ethylmaleimide and 2 . 5% sodium dodecyl sulfate ( SDS ) for one hour at 50°C with frequent vortexing in order to block the free thiols . The proteins were then precipitated with three volumes of cold 100% acetone and incubated at -20°C for 20 minutes . The mixture was centrifuged at 1820g for five minutes , and the pellet was then washed three times with 70% acetone ( 3 volumes ) and then resuspended in HENS buffer which contains 100 mM HEPES , 1 mM EDTA , 0 . 1 mM neocuproine , and 1% SDS ) . The resuspended samples were added to 80 μl thiopropyl sepharose 6B resin ( GE Healthcare ) in the presence or absence of dithiothreitol ( DTT , final concentration 10 mM ) . DTT is a reducing agent , which enables the oxidized thiol group of cysteine to bind to the resin by forming disulfide bonds between the reduced thiol groups of the proteins and the thiol group of the resin . The samples were rotated in the dark at room temperature for 1–2 hours , and then overnight at 4°C . The resin was washed four times with 1 ml HENS buffer , and then twice with 1 ml HENS/10 buffer ( 1:10 HENS buffer ) . Captured proteins were eluted with 30 μl HENS/10 buffer which contained 100 mM 2-mercaptoethanol for 20 minutes at room temperature , and the proteins in each eluent were resolved on a 12 . 5% SDS-PAGE gel . Each gel was then stained with silver ( Pierce Silver Stain ) and each gel slice was independently analyzed by MS . The proteins in each gel slice were reduced with 2 . 8 mM DTT ( 60°C for 30 minutes ) , modified with 8 . 8 mM iodoacetamide in 100mM ammonium bicarbonate in the dark at room temperature for 30 minutes , and digested overnight in 10% acetonitrile and 10 mM ammonium bicarbonate with modified trypsin ( Promega-Biological Industries , Israel ) at 37°C . The resulting peptides were resolved by reverse-phase chromatography on 0 . 075 x 200-mm fused silica capillaries ( J&W Scientific , Agilent Technologies , Israel ) packed with Reprosil reversed phase material ( Dr . Maisch GmbH , Germany ) . The peptides were eluted at flow rates of 0 . 25 μl/min on linear gradients of 7–40% acetonitrile in 0 . 1% formic acid for 95 minutes followed by eight minutes at 95% acetonitrile in 0 . 1% formic acid . MS was done by an ion-trap mass spectrometer ( Orbitrap , Thermo ) in a positive mode using a repetitively full MS scan followed by collision-induced dissociation ( CID ) of the seven most dominant ions selected from the first MS scan . The MS data was analyzed using the Proteome Discoverer software version 1 . 3 which searches the Ameba section of the NCBI-NR database and the decoy databases ( in order to determine the false discovery rate ( FDR ) ) using the Sequest and the Mascot search engines . The OXs were classified according to their protein class using the PANTHER software ( Protein ANalysis THrough Evolutionary Relationships ) Classification System ( http://www . pantherdb . org/ ) [18] . The adhesion of oxidatively stressed trophozoites to HeLa cell monolayers was measured using a previously described protocol [19] . Briefly , trophozoites ( 2×105 ) were exposed to 2 . 5 mM H2O2 for 20 minutes at 37°C , washed twice with Dulbecco's modified Eagle's medium ( DMEM ) without serum , added to wells that contained fixed HeLa monolayers in 1 ml of DMEM without serum , and incubated for 30 minutes at 37°C . The number of adherent trophozoites was determined by counting the number of trophozoites that remained attached to the HeLa cells after gentle decanting ( twice ) of the non-adherent trophozoites with warm ( 37°C ) DMEM under a light microscope . The Costar Transwell System ( 8-μm pore size polycarbonate membrane , 6 . 5-mm diameter , Corning Inc , Corning , NY , USA ) was used to determine trophozoite motility [20] . Briefly , 24-well culture plate was filled with serum-free Diamond’s TYI-S-33 medium ( 500-μl per well ) . A transwell insert was then inserted into each well . Control and oxidatively stressed ( 2 . 5 mM and 1 mM for one hour at 37°C ) trophozoites were washed three times in serum-free Diamond’s TYI-S-33 medium , and then suspended in serum-free Diamond’s TYI-S-33 medium . A 500-μl aliquot of the suspension ( 26x105 trophozoites/ml ) was then loaded into the transwell inserts . The 24-well culture plate containing the transwell inserts was then placed in anaerobic bags ( Mitsubishi Gas Chemical Company , Inc . , Tokyo , Japan ) , and incubated for three hours at 37°C . At the end of the incubation , the inserts and culture medium were removed from the 24-well culture plate , and trophozoite migration was determined by counting the number of trophozoites that were attached to the bottom of the wells of the 24-well culture plate . Gal/GalNAc lectin was purified using a previously described protocol [21] Aliquots ( 5 μg ) of purified Gal/GalNAc lectin were incubated with either 0 . 1 mM or 2 . 5 mM H2O2 for ten minutes at 37°C . The Gal/GalNAc lectin was then incubated with 10 μl D-galactose-coated agarose beads ( Thermo Scientific-Pierce ) overnight at 4°C . At the end of the incubation , the beads were washed in 20 volumes of PBS and then boiled in Laemmli sample buffer . The amount of Gal/GalNAc lectin that was released from the beads was determined using SDS-PAGE gel electrophoresis and silver staining ( Pierce ) . Aliquots ( 5 μg ) of purified Gal/GalNac lectin [21] were treated with 1 mM H2O2 for ten minutes at room temperature in order to introduce carbonyl groups into protein side chains . Using the OxyBlot Protein Oxidation Detection Kit ( Millipore , Israel ) [22] , the carbonyl groups are derivatized with 2 , 4-dinitrophenylhydrazine ( DNPH ) . The DNPH-treated Gal/GalNac lectin was separated by SDS-PAGE , transferred onto a nitrocellulose membrane and then detected by a specific antibody against the dinitrophenyl ( DNP ) moiety of the OXs . The nitrocellulose membrane has been stripped and probed with a polyclonal Gal/GalNAc lectin antibody ( a kind gift from of N . Guillen , Pasteur Institute , Paris , France ) to confirm that equal amounts of purified Gal/GalNac lectin were loaded on the gel . SUnSET was performed using a previously described protocol [23] . Briefly , trophozoites ( 2x106/ml ) that were treated with 2 . 5 mM H2O2 for 15 minutes at 37°C and untreated control trophozoites were incubated with 10 μg/ml puromycin ( Sigma ) , a structural analog of tyrosyltRNA , for 20 minutes . For pretreatment of the trophozoites with cycloheximide ( Sigma ) , the trophozoites were incubated with 100 μg/ml cycloheximide for five minutes before adding puromycin . The trophozoites were lysed using 1% Igepal ( Sigma ) in PBS . Puromycin was detected by immunoblotting using a monoclonal puromycin antibody ( 12D10 clone , Millipore ) . Protein quantification was measured by band intensity ( densitometry ) using ImageJ software [17] . Arginase activity in E . histolytica crude lysate was spectrophotometrically measured by quantifying the amount of urea that is generated when L-arginine is hydrolyzed by arginase using a previously described protocol [24] . Briefly , 104 trophozoites were dissolved in 100 μl of 0 . 1% Triton X-100 ( Sigma ) in the presence of 50 μM L-3-carboxy-2 , 3-trans-epoxypropionyl-leucylamido ( 4-guanidino ) -butane ( E-64 ) ( Sigma ) , a cysteine protease inhibitor . The lysate ( 50 μl ) was mixed with 50 μl of Tris-HCl ( 50 μM; pH 7 . 5 ) which contained 10 mM MnCl2 , and then activated by heating for ten minutes at 55°C . The hydrolysis of L-arginine by arginase was initiated by adding 25 μl L-arginine ( 0 . 5 M; pH 9 . 7 ) to a 25 μl aliquot of activated lysate . After a 30-minutes incubation at 37°C , the reaction was stopped by adding 400 μl of an acid solution mixture ( H2SO4: H3PO4: H2O = 1: 3: 7 ) . The urea concentration in the mixture was measured at 570 nm after adding α-isonitrosopropiophenone ( 25 μl , 9% in absolute ethanol ) to the mixture , heating the mixture for 45 minutes at 100°C , and incubating the mixture in the dark for ten minutes at room temperature . The amino acid levels in culture supernatants were analyzed by high-performance liquid chromatography ( HPLC ) using a previously described protocol [24] in which the proteins in the culture supernatant are first precipitated with methanol , and the amino acids are derivertized using o-phthalaldehyde ( OPA ) in an alkaline medium . Briefly , 200 μl of culture supernatants are added to an 800-μl mixture of methanol and internal standard ( homocysteic acid ) . After centrifugation , the samples are loaded into the HPLC autosampler , which converts the samples to fluorescent derivatives ( by mixing them with OPA ) before their injection into the columns ( C-18 ) . A JASCO FP 1520 fluorescence detector at an excitation wavelength of 360 nm with emission detection at 455 nm was used to separate , detect , and quantify the fluorescent derivatives . For quantification of the intracellular amino acids , trophozoites ( 107 ) were lysed in 1 ml of trichloroacetic acid ( TCA ) 10% for 30 minutes at 4°C , and centrifuged , and the pH of the supernatants was adjusted to 12 using 10N NaOH . The amino acid concentration in the supernatants was then measured by HPLC on two biological replicates . The quantification of putrescine in the trophozoite lysates was performed using 1H- nuclear magnetic resonance ( NMR ) spectroscopy as previously described [25 , 26] . Two biological replicates were used for each measure . Briefly , 400 μl H2O was used to dissolve the lysate and mixed with 200 μl buffer solution containing the internal standard TSP ( 3-trimethylsilyl-[2 , 2 , 3 , 3-D4]-1-propionic acid ) ( Sigma-Aldrich ) . All NMR spectra were obtained at 600 . 27 MHz at a temperature of 310 K , using a Bruker Avance-II 600 NMR spectrometer operated by TOPSPIN 3 . 2 software ( Bruker Biospin GmbH ) . Spectral referencing was done relative to the TSP signal ( final concentration 0 . 33mM ) . Data analysis ( identification ) was done using AMIX v3 . 9 . 14 software ( Bruker Biospin GmbH ) as previously described [25 , 26] . The quantification was performed using the software package CHENOMX ( version 8 . 1 ) .
When E . histolytica trophozoites strain HM-1:IMSS were incubated with 1 mM , 2 . 5 mM , 5 mM , 7 mM , or 10 mM H2O2 for 60 minutes at 37°C , the calculated LD50 of H2O2 is 5 . 5 ± 0 . 1 mM ( Table 2 ) and the intracellular ROS levels in living trophozoites are high ( Fig 1A ) . Based on these results , we selected 2 . 5 mM as the H2O2 concentration to oxidatively stress trophozoites in our various assays because this concentration is not lethal ( 85% of the trophozoites are viable; S1 Fig ) , the intracellular ROS levels are relatively low ( Fig 1A ) , and OXs are formed ( this work ) . We then used OX-RAC coupled to label-free quantification LC-MS for the detection and quantification of OXs in the lysate of oxidatively stressed trophozoites ( Fig 1B ) . A protein was considered to be oxidized when its relative amount in the DTT-treated lysates was at least two times greater than that in the untreated lysates ( Fig 1C ) . We identified 154 proteins that met this condition ( S1–S3 Tables ) . These 154 proteins were then classified ( Fig 1D ) using PANTHER sequence classification tool [27 , 28] . The protein classes were phosphatases ( exemplified by phosphoinositide phosphatase ( EHI_141860 ) ; transporters ( exemplified by plasma membrane calcium-transporting ATPase , EHI_030830 ) ; membrane traffic proteins ( exemplified by putative vacuolar sorting protein , EHI_025270 ) ; chaperones ( exemplified by Hsc70-interacting protein , EHI_158050 ) ; hydrolases ( exemplified by arginase , EHI_152330 ) ; oxidoreductases ( exemplified by superoxide dismutase , EHI_159160 ) ; enzyme modulators ( exemplified by Ras family GTPase , EHI_058090 ) ; lyases ( exemplified by tRNA pseudouridine synthase , EHI_151650 ) ; transferases ( exemplified by histone acetyltransferase , EHI_152010 ) ; nucleic acid binding proteins ( exemplified by 13 kDa ribonucleoprotein-associated protein , EHI_104600 ) ; ligases ( exemplified by ubiquitin-conjugating enzyme family protein , EHI_070750 ) ; kinases ( exemplified by galactokinase , putative , EHI_094100 ) ; isomerases ( exemplified by cysteine synthase A , EHI_024230 ) ; cytoskeletal proteins ( exemplified by actin-binding protein , cofilin/tropomyosin family , EHI_168340 ) and proteases ( exemplified by methionine aminopeptidase , EHI_126880 ) . In order to evaluate the consistency of MS-based identification of OXs , the purified heavy subunit of Gal/GalNac lectin ( Hgl ) was exposed to 1 mM H2O2 for ten minutes and its oxidation was confirmed independently by using the OxyBlot kit . The presence of carbonyl groups was detected using a specific antibody which recognizes the DNP moiety in the purified lectin that has been exposed to H2O2 and treated with DNPH ( Fig 2A ) . As expected , carbonyl groups were not detected in purified lectin that was not exposed to H2O2 and treated with DNPH . According to the PANTHER statistical overrepresentation test which compares classifications of multiple clusters of lists to a reference list , very significant enrichment ( fold enrichment > 5 and P > 2 . 42x10-6 ) was found for proteins involved in the process of translation , such as the 60S ribosomal protein L9-like protein ( EHI_193080 ) and the 13 kDa ribonucleoprotein-associated protein ( EHI_104600 ) . The results of a previous study showed that ( a ) gene expression of oxidatively stressed E . histolytica trophozoites triggers a stress response in which 185 genes are upregulated and 102 genes are downregulated and ( b ) these genes are involved in signaling/regulatory processes , metabolic/repair processes , energy metabolism , the stress response , and transport [7] . We found that some of the OXs in the oxidatively stressed trophozoites are involved in the stress response , transport , and metabolism . Of all the proteins that were found to be oxidized and all the genes whose expression had changed in the oxidatively stressed trophozoites , EHI_179080 ( sulfate adenylyltransferase ) was the only common gene . Cysteine residues are the predominant targets of oxidation or S-nitrosylation in redox-sensitive proteins . We have recently identified 142 S-nitrosylated ( SNO ) -proteins in E . histolytica after its exposure to NO [21] . 21 proteins were shared in our OX-RAC and SNO-RAC analysis ( Table 3 ) . The shared proteins include rubrerythrin , protein disulfide isomerase and iron-containing superoxide dismutase which have been associated with resistance to OS [12 , 27–29] and the Gal/GalNac lectin , a cell surface protein which is involved in binding of E . histolytica troohozoites to host cells [17 , 18] . In order to gain information on the consequence of oxidation on the activity of some of the proteins that were identified in the OX-RAC analysis , we decided to focus our analysis on the function of the oxidized Gal/GalNac lectin . The Gal/GalNac lectin consists of Hgl ( 170 kDa ) and a light subunit ( Lgl ) ( 35/ 31 kDa ) . Hgl mediates E . histolytica adherence , and indirect evidence suggests that Lgl plays a role in E . histolytica virulence [13 , 19] . The occurrence of Hgl among the OXs ( see S2 Table ) and SNO proteins ( [21] and Table 3 ) suggests that OS and NS affect the parasite's adherence . In order to test this hypothesis , we compared the ability of untreated and oxidatively stressed trophozoites to adhere to HeLa cells [13] . We found that the binding of the oxidatively stressed trophozoites to the HeLa cell monolayer was significantly less than that of the untreated trophozoites ( 65% and 85%; Fig 2B ) . Previous research results show that the cysteine-rich region ( CRR ) of the Hgl and the CRD are important for the binding activity of the lectin [19 , 20] . Our MS analysis of OXs indicates that many carbamidomethylated cysteine residues , that possibly represent oxidized cysteines , are located in the CRR and in the CRD of Hgl ( Fig 2C ) . When we investigated the binding ability of oxidized Gal/GalNAc lectin , we observed that H2O2 prevents the binding of the Gal/GalNAc lectin to the galactose beads ( Fig 2D ) . Motility is important for E . histolytica survival and pathogenicity , and requires a dynamic actin cytoskeleton [30] . The presence of cytoskeleton-associated proteins , such as the ARP2/complex 20kDa subunit , among the OXs ( S2 Table ) and the SNO- proteins ( [21] and Table 3 ) suggests that the parasite's motility is redox-regulated . We tested this hypothesis by comparing the migration of control and oxidatively stressed trophozoites: the migration of the oxidatively stressed trophozoites was significantly less than that of the control trophozoites ( Fig 3 ) . Inhibition of translation is a typical response of cells exposed to stress conditions [31 , 32] , and such inhibition may avoid constant gene expression during error-prone environments . We recently reported that NS inhibits protein synthesis in E . histolytica [33] . The presence of the 60S ribosomal protein L7 , the 60S ribosomal protein L18a and the 40S ribosomal protein S13 among the OXs ( S2 Table ) and SNO proteins ( [21] and Table 3 ) suggest that oxidation is similar to S-nitrosylation [21] , in that it regulates the translation of proteins in the parasite . In order to test this hypothesis , we used SUnSET [23][33 , 34] to determine protein synthesis . We found that protein synthesis is strongly inhibited in oxidatively stressed trophozoites and that the level of inhibition is comparable to that found in cyclohexamide-treated trophozoites ( Fig 4 ) [35] . We previously reported that the enzymatic conversion of L-arginine to L-ornithine is an significant source of L-ornithine for E . histolytica and that arginase activity is essential for the resistance of the parasite to NS [24] . The presence of L-arginase among the OXs raises questions about the regulation of its activity by OS and its involvement in the resistance of the parasite to OS . In order to obtain information about the effect of OS on arginase activity , we measured arginase activity in control and oxidatively stressed trophozoites using a previously described assay [24] . We found that arginase activity is markedly inhibited ( 90% inhibition ) in the oxidatively stressed trophozoites ( Fig 5 ) . In order to establish whether arginase gene expression is involved in the resistance of the parasite to OS , we decided to upregulate its expression ( Fig 5 ) . We detected modest arginase activity in arginase-overexpressing trophozoites and no arginase activity in control trophozoites that were exposed to 2 . 5 mM H2O2 for 15 minutes ( Fig 5 ) . We found that the arginase-overexpressing trophozoites are more resistant to OS ( LD50 6 . 2±0 . 08 mM ) than the control trophozoites ( IC50 LD50 5 . 1±0 . 1 mM ) ( table 2 ) . We also found that the intracellular arginine concentration in the arginase-overexpressing trophozoites ( 53 ± 5 μM ) was significantly lower than that in the control trophozoites ( 136±9 μM ) . No significant difference in the intracellular ornithine concentration in the arginase-overexpressing trophozoites ( 616±7 μM ) and the control trophozoites ( 577±15 μM ) was detected . In contrast , we found that the intracellular concentration of putrescine in the arginase-overexpressing trophozoites ( 293±20 nmol/mg protein ) was significantly higher than that of the control trophozoites ( 123±3 nmol/mg protein ) .
The current understanding of the antiamebic effect of OS [36] has greatly beneficiated from previous omics studies [13] [7] on the response of E . histolytica to OS . However , these studies did not address the nature of OXs in oxidatively stressed E . histolytica trophozoites . We decided to use OX-RAC coupled to MS [14] to generate new data about OXs in E . histolytica . Some of the proteins that we identified in our OX-RAC analysis of OXs are of particular interest because ( i ) they have an important function in the parasite's physiology and/or virulence and ( ii ) their activity is regulated by OS in other organisms . The first notable OX that we identified in our OX-RAC analysis is the cysteine-rich ( C-rich ) region ( amino acids 356–1143 ) and a CRD ( amino acids 895–998 ) of Hgl [20] . We recently showed that S-nitrosylation of cysteine residues in the CRD inhibits the galactose binding activity of the Gal/GalNAc lectin and contributes to the reduced binding of NO-treated trophozoites to their target cells [21] . In this study , we found that carbamidomethylated cysteine residues are also located in the CRR and the CRD of the lectin . This result suggests that oxidation of these cysteines is responsible for the loss of galactose-binding activity of the Gal/GalNAc lectin . We found that cytoskeletal proteins are oxidized in oxidatively stressed trophozoites . For its invasion into the host's tissues , E . histolytica relies on its dynamic actin cytoskeleton [37] , [38] . Oxidation of the actin cytoskeleton can modulate its cellular functions in mammalian cells [39] [40] [41] and the results of proteomics studies in human peripheral blood mononuclear cells have previously identified actin as an oxidation target [42] . Actin oxidation inhibits its polymerization [43] [44] and leads to cytoskeletal rearrangements [45] . According to previous reports , the cysteines in actin are some of the most susceptible targets of oxidation [45] and their oxidative modification is the likely cause of cytoskeletal rearrangements in oxidatively stressed mammalian cells [46] [47] [48] . Actin is a highly conserved protein among species [49] . The actin of rats and E . histolytica share a number of oxidized amino acid residues ( Met44 , Met47 ) which are located in actin−actin contact regions [50] suggesting that actin oxidation is part of the process that results in the inhibition of the transwell migration that we observed in the oxidatively stressed trophozoites . We detected strong enrichment of OXs which are components of the parasite's translational machinery , such as ribosomal proteins and elongation factors . This finding is in agreement with the results of a recent study in yeast cells , in which it was reported increases in ribosomal proteins and elongation factors due to oxidative thiol modifications following a short-term exposure to H2O2 [51] . At first glance , this finding is also in agreement with the inhibition of protein synthesis in oxidatively stressed E . histolytica trophozoites ( this work ) and mammalian cells [52] . However , a recent report suggests that the oxidation of components of the translational machinery is not the direct cause of the inhibition of protein synthesis , but rather a global , enzymatic downregulation of almost all tRNA species in OS [53] . It will be interesting to determine whether the same enzymatic downregulation of tRNA species also occurs in oxidatively stressed E . histolytica . An additional group of OXs which we identified are the oxidoreductases , which includes the iron-containing superoxide dismutase . OS induces the expression of this protein [12] and the protein can form adducts with metronidazole metabolites [54] . In E . coli , iron-containing superoxide dismutase is inactivated by H2O2 via a reaction of H2O2 with the iron at the active site that generates a potent oxidant which attacks tryptophan residues [55] . It has been recently showed for Trypanosoma superoxide dismutase ( Fe-SODB ) that Cys ( 83 ) in Fe-SODB acts as an electron donor that repairs the tyrosyl radical ( Tyr35-O• ) via intramolecular electron transfer in order to prevent inactivation of Fe-SODB by peroxynitrite , which is produced by immunostimulated macrophages [56] . Interestingly , this Cys ( 83 ) is conserved in E . histolytica's iron-containing superoxide dismutase and Cys ( 83 ) was detected by MS as a carbamidomethylated cysteine residue which strongly suggest that it has been oxidized . Based on this data , it is tempting to speculate that Cys ( 83 ) in E . histolytica's iron-containing superoxide dismutase is crucial for protecting the protein against inactivation by H2O2 . The phosphatases are another group of OXs which we identified in our OX- RAC analysis . This group includes the protein-tyrosine phosphatases ( PTP ) . Two PTPs have been cloned from E . histolytica , EhPTPA and EhPTPB . EhPTPA but not EhPTB is strongly up-regulated in trophozoites that have been recovered from amebic liver abscesses suggesting that EhPTPA is involved in the parasite’s virulence . [57] . The regulation of phosphatase function by OS is well documented [58–60] . PTPs are characterized by an 11-residue signature motif ( I/V ) HCXAGXXR ( S/T/G ) in their active site [61] . The oxidation of the catalytic cysteine in this signature sequence leads to their reversible inactivation . Remarkably , the cysteine at the catalytic site of E . histolytica's EhPTPA ( IKGIKLNGPPIIHCSAGLGRSGTFI ) was detected as a carbamidomethylated cysteine residue by MS , and this finding strongly suggests that it has been oxidized . Accordingly , we surmise that E . histolytica EhPTPA activity will be inhibited by OS and in the future , it will be interesting to study the consequence of this inhibition on the parasite’s virulence . Another group of OXs which we identified in our OX- RAC analysis are the transport proteins and this group includes plasma membrane calcium-transporting ATPase . Ca2+-ATPases regulate intracellular calcium levels in eukaryotic cells and are thus essential to the correct functioning of the cell machinery . Calcium is also important in numerous cellular processes in E . histolytica , such as development and virulence [62 , 63] . Five putative Ca2+-ATPases , which could be important in the regulation of the cytoplasmic calcium concentration , have been recently identified in E . histolytica [64] . Ca2+-ATPases are very sensitive to OS and undergo functional and conformational changes when exposed to oxidants [65] . It is possible that the same observation applies to the E . histolytica Ca2+-ATPases identified in our Ox-RAC analysis of oxidized proteins . Arginase ( EHI_152330 ) is a hydrolase which we identified as being oxidized in our OX-RAC analysis . Whereas arginase activity has been associated with resistance to NS in various unicellular parasites including E . histolytica [24 , 66 , 67] , its involvement in OS resistance has never been investigated . In this study , we found that overexpression of arginase protects E . histolytica against OS . The inhibitory effect of OS on arginase activity has also been found in Helicobacter pillory [68] . Only three cysteine groups are present in Eharginase and their role in the enzyme's activity is unknown . A clue about their role may be deduced from the results of a previous study which found that arginase activity was inhibited in erythroleukemic K562 cells that were exposed to aurothiomalate , a gold analog that can specifically react with a protein sulphydryl group to form a thiol-gold adduct [69] . Since this finding suggests that one or more of the cysteine residues in arginase are essential for its activity , we surmise that Eharginase activity is also dependent on the presence of cysteine residues . A possible mechanism to explain why arginase overexpression protects E . histolytica against OS involves the strong reduction of intracellular arginine that we detected in the arginase overexpressing trophozoite . We presume that this reduction might be due to the conversion of arginine into ornithine by the excess of arginase and from ornithine into putrescine by ornithine decarboxylase ( ODC ) [70–72] . This presumption is supported by the higher intracellular concentration of putrescine found in the arginase-overexpressing trophozoites compared with that of the control trophozoites . ODC is the only enzyme of polyamine biosynthetic pathway that has been reported to exist in E . histolytica , [73 , 74] [70] . Putrescine has been linked to OS resistance and one of the proposed mechanism of OS resistance is based on its polycationic nature that enables it to couple with nucleic acids and membrane phospholipids . Putrescine is also free radical scavenger and an antioxidant [75] . Putrescine probably plays the same antioxidant role in E . histolytica but in absence of an efficient inhibitor of E . histolytica's ODC ( α-difluoromethylornithine , which is a potent irreversible ODC inhibitor in many organisms ) is not effective against E . histolytica ODC [70] ) , this hypothesis cannot be directly tested . We found a very weak overlap between the results of this OX-RAC analysis of OXs and the results of an transcriptomics analysis of oxidatively stressed trophozoites [7] . This weak overlap indicates that the parasite's response to OS occurs at two different levels . One level is protein oxidation and the second level is a global change in gene expression which is characterized by the expression of general stress response-related proteins , such as heat shock proteins ( HSPs ) [8] . Another explanation is that oxidized proteins are not immediately expressed to replenish the parasite's cellular needs but are recycled through reduction processes . Such recycling may be done by EhPDI , an oxidoreductase that catalyzes oxidation , reduction and isomerization of disulfide bonds in polypeptide substrates [76] . A third explanation is that proteins which are encoded by OS response genes are resistant to inactivation by oxidation and for this reason they were not detected by OX-RAC . This explanation appears not to apply to chaperone-like heat-shock proteins or ubiquitin-conjugating enzymes because their expression is upregulated by OS [7] and oxidation inhibits their activity [77 , 78] . We identified 21 common OX and SNO proteins ( Table 3 ) . With the exception of the Gal/GalNac lectin ( this work ) , the effect of oxidation or S-nitrosylation on their activity/function has yet to be determined . This effect may be complex and often antagonist . For example , the activity of mammalian protein disulfide isomerase [79] and prokaryotic iron-containing superoxide dismutase [80] is inhibited when they are S-nitrosylated . In contrast , S-nitrosylation of superoxide dismutase and protein phosphatase 1B prevents their inactivation by OS [81 , 82] . Thioredoxin must be S-nitrosylated for it to be an efficient antioxidant in plants [83] . It has been suggested that OXs and SNO proteins are mediators between stress pathways that are induced by OS and NS [84] . A good candidate for such mediator function is a member of the Ras-family GTPase , one of the common OX and SNO proteins which we identified in this study . Ras-family GTPases are involved in cell proliferation and their activity in mammalian is regulated by both S-nitrosylation [85] and oxidation [86] . To conclude , we inform on the presence of many novel OXs in oxidatively stressed E . histolytica trophozoites . Of these oxidized proteins , we discovered that ( a ) protein oxidation can regulate the activity of an important virulence factor of E . histolytica , namely Gal/GalNAc lectin , and ( b ) a protective role for arginase against OS . OX-RAC detects , enriches , and identifies OXsby detecting their oxidized cysteine residues , and it is possible that some of the OXs may not have been detected because of oxidation of other amino acid residues , such as methionine and tyrosine [87] . Finally , we envisage that these results will pave the way for further studies on the activity of OXs in E . histolytica . | Reactive oxygen species are the most studied of environmental stresses generated by the host immune defense against pathogens . Although most of the studies that have investigated the effect of oxidative stress on an organism have focused on changes which occur at the protein level , only a few studies have investigated the oxidation status of these proteins . Infection with Entamoeba histolytica is known as amebiasis . This condition occurs worldwide , but is most associated with crowded living conditions and poor sanitation . The parasite is exposed inside the host to oxidative stress generated by cells of the host immune system . The nature of oxidized proteins in oxidatively stressed E . histolytica has never been studied . In this report , the authors present their quantitative results of a proteome-wide analysis of oxidized proteins in the oxidatively stressed parasite . They identified crucial redox-regulated proteins that are linked to the virulence of the parasite , such as the Gal/GalNAc lectin . They also discovered that arginase , a protein involved in ornithine synthesis , is also involved in the parasite's resistance to oxidative stress . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| []
| 2016 | Proteomic Identification of Oxidized Proteins in Entamoeba histolytica by Resin-Assisted Capture: Insights into the Role of Arginase in Resistance to Oxidative Stress |
Parasitological cure for Chagas disease is considered extremely difficult to achieve because of the lack of effective chemotherapeutic agents against Trypanosoma cruzi at different stages of infection . There are currently only two drugs available . These have several limitations and can produce serious side effects . Thus , new chemotherapeutic targets are much sought after . Among T . cruzi components involved in key processes such as parasite proliferation and host cell invasion , Ca2+-dependent molecules play an important role . Calcineurin ( CaN ) is one such molecule . In this study , we cloned a new isoform of the gene coding for CL strain catalytic subunit CaNA ( TcCaNA2 ) and characterized it molecularly and functionally . There is one copy of the TcCaNA2 gene per haploid genome . It is constitutively transcribed in all T . cruzi developmental forms and is localized predominantly in the cytosol . In the parasite , TcCaNA2 is associated with CaNB . The recombinant protein TcCaNA2 has phosphatase activity that is enhanced by Mn2+/Ni2+ . The participation of TcCaNA2 in target cell invasion by metacyclic trypomastigotes was also demonstrated . Metacyclic forms with reduced TcCaNA2 expression following treatment with morpholino antisense oligonucleotides targeted to TcCaNA2 invaded HeLa cells at a lower rate than control parasites treated with morpholino sense oligonucleotides . Similarly , the decreased expression of TcCaNA2 following treatment with antisense morpholino oligonucleotides partially affected the replication of epimastigotes , although to a lesser extent than the decrease in expression following treatment with calcineurin inhibitors . Our findings suggest that the calcineurin activities of TcCaNA2/CaNB and TcCaNA/CaNB , which have distinct cellular localizations ( the cytoplasm and the nucleus , respectively ) , may play a critical role at different stages of T . cruzi development , the former in host cell invasion and the latter in parasite multiplication .
Chagas disease , whose etiological agent is Trypanosoma cruzi , is a neglected tropical parasitic infection . An estimated 10 million people are infected worldwide , predominantly in Latin America , where it is endemic , and more than 25 million people are at risk of acquiring the disease [1] . The two chemical therapeutic agents used to treat the disease ( Nifurtimox and Benznidazole ) may cause side effects , and parasitological cure is not achieved in all cases [2] , [3] . Identification of key factors in the life cycle of the parasite that could be targets for new chemotherapeutic strategies is therefore very important . In the life cycle of T . cruzi , epimastigote forms replicate in the insect vector and then differentiate into metacyclic trypomastigotes , which are infective to the mammalian host . Cell invasion by metacyclic forms is crucial for the establishment of T . cruzi infection . Inside host cells , the parasite replicates as amastigotes , which subsequently transform into trypomastigotes . When the host cell ruptures , these are released to the circulation . There is evidence that Ca2+-dependent events are implicated in various processes that are critical for the maintenance of the T . cruzi life cycle . It has been shown that the Ca2+ chelator EGTA decreases epimastigote multiplication and that intracellular Ca2+-concentration increases about six-fold during differentiation of epimastigotes into metacyclic trypomastigotes , an event that is blocked by calmodulin inhibitors [4] . Induction of Ca2+ signaling in insect-stage and bloodstream trypomastigotes is an important requirement for target cell invasion [5] , [6] . Further , it has been suggested that the Ca2+ signal induced in metacyclic forms is associated with the activation of a protein tyrosine kinase [7] . Protein kinases and phosphatases , which control the phosphorylation state of tyrosine , serine and threonine residues , play a pivotal role in cell signal regulation and integration in all living organisms , including trypanosomatids [8] , [9] . T . cruzi protein phosphatase 2A ( PP2A ) , for instance , has been implicated in the transformation of trypomastigotes into amastigotes [10] . In this scenario , a homolog of mammalian calcineurin has emerged as an important factor for T . cruzi infection . In cells of different tissues , the Ca2+-dependent phosphatase calcineurin , also known as PP2B or CaN , is involved in a number of different signaling pathways . An evolutionarily conserved protein in all eukaryotes , it appears to be ubiquitously expressed [11] , [12] , [13] . It is heterodimeric and consists of calcineurin A ( CaNA ) , the catalytic subunit , and calcineurin B ( CaNB ) , the Ca2+-binding subunit [12] . In T . cruzi clone CL Brener , Moreno et al . [14] identified a protein homologous to CaNA , which is predominantly localized in the nucleus and , unlike its mammalian counterpart , has a catalytic domain and a CaNB-binding domain but lacks the binding domain to calmodulin and the autoinhibitory domain ( AID ) . A protein phosphatase with the same characteristics was also detected in T . cruzi CL and G strains , and the sequence of its regulatory subunit ( TcCaNB ) was determined , revealing the presence of three Ca2+-binding domains , known as EF-hand motifs [15] . Treatment of CL strain metacyclic or tissue culture trypomastigotes with CaN inhibitors , such as cyclosporin and cypermethrin , or with antisense phosphorothioate oligonucleotides directed to TcCaNB was shown to inhibit parasite entry into host cells [15] . Whether TcCaN plays other biological functions essential for T . cruzi development had not been investigated prior to the present study . We addressed this question and found that TcCaN is also involved in parasite multiplication . In addition , we identified a new isoform of TcCaNA , TcCaNA2 ( HM854297 ) , which is localized in the cytoplasm and is implicated in a number of important events , including trypomastigote entry into target cells .
All animal handling protocols were performed according to the “Guide for the Care and Use of Laboratory Animals” from the National Institutes of Health , USA [16] and approved by the Institutional Ethics Committee at the Faculty of Health Sciences , University of Antofagasta , Chile ( CEIC REV/200 ) under FONDECYT-Chile grant number 1051045 . T . cruzi CL strain [17] , used throughout this study , was maintained cyclically in Balb/c mice and in axenic liver infusion tryptose ( LIT ) medium containing 5 . 0 g liver infusion , 5 . 0 g tryptose , 4 . 0 g NaCl , 0 . 4 g KCl , 8 . 0 g Na2HPO4 , 2 . 0 g glucose and 10 . 0 mg hemin per liter and supplemented with 5% fetal bovine serum . Epimastigote forms were grown at 28°C in LIT medium , and Grace's medium was used to obtain cultures enriched in metacyclic trypomastigote forms , which were purified by chromatography using a diethylaminoethyl ( DEAE ) cellulose column ( Sigma Chemical Co ) , as described by Teixeira et al . [18] . Cell invasion assays were performed as described by Ramirez et al . [19] . Briefly , 2×105 HeLa cells were cultured in 4-well Lab-Tek Chamber Slides ( Nunc , Thermo Scientific ) . After adhesion and growth at 37°C in a humidified 5% CO2 atmosphere , the cells were incubated with 1×106 metacyclic trypomastigotes ( MT ) previously treated or not with different calcineurin inhibitors . After 3 h incubation , the cells were washed with PBS and fixed with methanol followed by Giemsa staining . The number of intracellular parasites was counted in 100 cells . Assays were conducted in triplicate . The viability of MT was evaluated by Trypan blue exclusion and by parasite migration assay through gastric mucin layer . Briefly , polycarbonate transwell filters ( 3 mm pores , 6 . 5 mm diameter , Costar ) were coated with 50 µL of a preparation containing 10 mg/ml gastric mucin . T . cruzi metacyclic trypomastigotes , in 600 µL PBS were added to the bottom of 24-well plates ( 1×107 parasites/well ) and incubated for 1 h at 37°C . Thereafter , the mucin-coated transwell filters were placed onto parasite-containing wells , and 100 µL PBS were added to the filter chamber . After 1 h of incubation at 37°C , 10 µL were collected from the filter chamber for determination of parasite number and the volume in this chamber was corrected by adding 10 µL PBS [20] . Epimastigotes maintained in LIT medium were used for proliferation assays . Parasites in growth phase were incubated with different concentrations of calcineurin inhibitor cyclosporin A ( CsA ) . Untreated or treated parasites with ethanol were used as controls . Assays were performed with three different cultures , each containing 5×105 epimastigotes per sample in a volume of 1 mL of LIT medium . Parasite cultures were analyzed daily for one week by taking samples to measure the number of parasites using a Neubauer hemocytometer , and the viability of epimastigotes was determined under light microscopy using Trypan blue exclusion and CFDA-SE assays [21] . Results were expressed as mean ± standard error of three independent experiments . Similarly , epimastigotes were incubated with different concentrations of calcineurin inhibitors [CsA , tacrolimus ( FK-506 ) , INCA-6 and kaempferol ( Kmp ) ] at 0 , 10 , 20 and 40 µM . Crystal violet was used as a positive control because of its trypanocidal effect . Assays were performed in triplicate in 96-well microplates containing 5×105 epimastigotes per well in 200 µL of LIT medium . After 72 h incubation , the number of parasites was counted as described above . Untreated epimastigote cultures were used as a negative control . Results were expressed as a percentage of proliferation ( number of cells ) in the control group . Isolation of genomic DNA and total RNA from T . cruzi CL strain , Southern and Northern blot hybridizations and separation of chromosomal bands by pulsed-field gel electrophoresis were performed as described previously [10] . To clone the CaNA2 gene of T . cruzi CL strain , sense and antisense primers were designed based on the sequence of the genome of CL Brener clone available in GenBank ( http://www . ncbi . nlm . nih . gov ) under accession number XM_816360 . 1 . Sense primer 5′-ATG TTG TCT ACA TCA GAT TCT-3′ and antisense primer 5′-TCA TTT GCA TCC CTT ATT TAG-3′ were used . After amplification of the TcCaNA2 gene by RT-PCR with 1–20 ng of T . cruzi cDNA synthesized using poly A+ mRNA ( using oligo dT ) obtained from epimastigotes , PCR products were analyzed by agarose gel electrophoresis . The amplification product was cloned in the vector pCR 2 . 1-TOPO ( TOPO Cloning Vector Kit for sequencing , Invitrogen by Life Technologies ) according to the manufacturer's instructions . The CaNB gene was amplified by RT-PCR using a pair of primers: sense 5′-CGG AAT TCA TGG GCG AGG GGG T-3′ and antisense 5′-CGG AAT TCC TAA ATG GAG AGG C-3′ , which were based on a cDNA sequence from the CL strain ( accession number AY570505 ) . A cloning protocol similar to that described for the TcCaNA2 gene was used . Sequences were analyzed using DNASTAR and GeneDoc software and National Center for Biotechnology Information ( NCBI ) programs ( http://www . ncbi . nlm . nih . gov ) . The coding sequences of TcCaNA2 and TcCaNB genes were subcloned into the expression vector pGEX-1λT ( GE Healthcare ) in-frame with glutathione S-transferase ( GST ) gene . After sequencing the construct to check that the open reading frame was in the correct orientation , expression of the recombinant protein was induced in E . coli BL21 ( DE3 ) after addition of 1 mM isopropyl thio-β-D-galactoside ( IPTG ) . The recombinant protein was purified by the cleared lysate method using gluthathione-Sepharose 4B ( Amersham Biosciences ) . After washing with PBS pH 7 . 3 , the proteins of interest were eluted with 200 mM Tris-HCl pH 8 . 0 , 40 mM reduced glutathione , 150 mM NaCl , 5 mM DTT and 0 . 1% Triton X-100 . Analysis of the purified protein was carried out by SDS-PAGE on 10% gels stained with Coomassie blue . Additionally , the coding sequence of TcCaNA2 was cloned in-frame into the expression vector pET-SUMO ( Champion pET Expression System , Invitrogen by Life Technologies ) . The recombinant protein was expressed in E . coli BL21 ( DE3 ) after induction with IPTG , and the protein ( 6×His-SUMO-TcCaNA2 ) was purified from cleared lysates by affinity chromatography on Ni-NTA agarose ( Invitrogen by Life Technologies ) . After elution in the presence of imidazole ( SIGMA ) , the purity of the protein was determined as above . To measure TcCaN activity , parasites were washed three times with TBS ( 150 mM NaCl , 20 mM Tris , pH 7 . 2 ) and then lysed in lysis buffer ( 50 mM Tris pH 7 . 5 , 1 mM DTT , 100 µM EDTA , 100 µM EGTA , 0 . 2% NP-40 ) and centrifuged at 100 , 000× g for 45 min at 4°C . The high-speed post-lysis supernatants were used to evaluate calcineurin-type phosphatase activity with a Calcineurin Cellular Activity Assay Kit , Colorimetric ( Calbiochem , USA ) , according to the manufacturer's instructions . To determine the phosphatase activity of TcCaNA2 , the assay was performed using 1 µg of the recombinant protein in the presence of one of the following metal ions at 1 mM: CaCl2 , MgCl2 , MnCl2 or NiCl2 . The recombinant TcCaNA2 was incubated for 15 min at 30°C with 80 mM p-nitrophenyl phosphate ( p-NPP ) ( Calbiochem , USA ) as substrate . The reaction was stopped by adding 950 µL of 1 M NaOH , and enzyme activity was measured by the change in absorbance at 405 nm , as described by Sagoo et al . [22] . Similarly , the activity of 1 µg of TcCaNA2 combined with 1 µg of TcCaNB was measured in the presence of 1 mM of MnCl2 with or without EGTA . Two protocols were developed to generate antibodies . First , BALB/c mice and/or New Zealand white rabbits ( after obtaining pre-immune serum ) were immunized with GST-TcCaNA2 and GST-TcCaNB recombinant proteins . Each animal received 4 doses of 10 µg antigen plus 0 . 5 mg Al ( OH ) 3 as adjuvant at 7-day intervals . After the last immunization dose , blood was obtained by cardiac puncture . The polyclonal antiserum was divided into aliquots and stored at −20°C in the presence of 0 . 1% sodium azide as preservative . Monospecific polyclonal antibodies against the TcCaNA2 isoform were raised in New Zealand white rabbits by immunization with the synthetic peptides 246–264 ( CGSKSDYYTPSAGPSYGSKP-amide ) and 206–224 ( Ac-IKLNHIDLIHRFRE PPSRGC-amide ) conjugated to KLH ( Keyhole limpet hemocyanin ) or ovalbumin using MBS ( m-Maleimidobenzoyl-N-hydroxysuccinimide ester ) . The monospecific antibodies were purified by affinity chromatography using the peptide 246–264 , which is only present in the TcCaNA2 isoform , as ligand . To evaluate the interaction between TcCaNA2 and TcCaNB , a Far-Western blotting was performed as described by Wu et al . [23] using the recombinant proteins TcCaNB ( tagged with GST ) and TcCaNA2 ( tagged with 6×His ) . BSA and the unrelated fusion protein 6×His-SUMO-CAT were used as negative controls . One microgram each of BSA , CAT and 6×His-SUMO-TcCaNA2 ( target protein ) were resolved in 10% SDS-PAGE and transferred onto PVDF membranes , which were then incubated with decreasing concentrations of guanidine-HCl ( 6 , 3 , 1 , 0 . 1 , and 0 M ) to denature and renature the target protein . The membrane was then blocked with PBS containing 0 . 5% Tween 20 and 5% skim milk and incubated with 10 µg of the bait protein GST-TcCaNB . A rabbit polyclonal antibody directed to GST-TcCaNB and an anti-rabbit IgG antibody conjugated to horseradish peroxidase ( Sigma-Aldrich ) were used to detect TcCaNA2-TcCaNB interaction . The immunocomplexes were revealed using diaminobenzidine ( DAB ) as substrate . Epimastigotes ( 1×107 ) were lysed in Laemmli buffer ( 62 . 5 mM Tris-HCl pH 6 . 8 , 25% glycerol , 2% SDS , 0 . 01% bromophenol blue ) and exposed to different concentrations of urea ( 0 , 2 , 4 and 6 M ) overnight at 4°C to stabilize the solutions , which were then separated on 10% SDS-PAGE gels , electrotransferred to a PVDF membrane ( Amersham Hybond-P ) and visualized by Western blot . The antisense oligonucleotide ( AS-ON ) used was 5′- GAG AAT ATG CTG TAG ACA ACA TTA T-3′ , and the AS-ON control was a standard unrelated oligonucleotide: 5′-CCT CTT ACC TCA GTT ACA ATT ATT-3′ . The assays were performed in the presence of antisense oligonucleotides directed to TcCaNA2 ( third generation morpholino , Gene Tools , LLC ) , and the effects were evaluated as described above for the assays used to detect inhibition of cell proliferation and invasion . Each AS-ON was added at a concentration of 10 µM with 6 µM of Endo-Porter , according to the Gene Tools “Endo-Porter delivery of morpholino oligos” protocol . Epimastigotes in exponential growth phase ( CL strain ) were centrifuged at 6 , 000× g for 5 min , washed twice in PBS pH 7 . 2 and then resuspended at 2×106 parasites/mL . Parasites were decanted on poly-L-lysine-treated cover slips and fixed in PBS containing 4% paraformaldehyde . Next they were permeabilized with PBS-0 . 5% Triton X-100 for 5 min at room temperature and blocked with a solution of 2% glycine , 2% BSA , 5% FCS and 50 mM NH4Cl in PBS pH 7 . 2 . The cover slips were then incubated with the primary antibodies , washed three times for 5 min in PBS pH 7 . 2 and incubated with the secondary antibodies ( anti-mouse Alexa Fluor 488 and anti-rabbit Alexa Fluor 568 ) . The samples were visualized with a reflected light microscope ( Carl Zeiss , model Axiovert 10 ) , photographed and analyzed using QCapture Pro 6 . Statistical analysis was carried out with GraphPad Prism v . 5 . 0 ( GraphPad Software , Inc . , San Diego , CA ) . Values of p<0 . 05 were considered significant .
It has previously been shown that the treatment of CL strain metacyclic trypomastigotes or tissue culture-derived trypomastigotes with calcineurin inhibitors CsA or cypermethrin resulted in strong inhibition ( 62–64% ) of parasite entry into HeLa cells [15] . To determine the role of TcCaN in parasite growth , epimastigotes were incubated in the absence or presence of varying concentrations of the calcineurin inhibitor CsA . Starting on the day CsA was added , parasite density was measured daily for one week , and it was observed that parasite growth was inhibited by CsA . At 20 µM and 40 µM , CsA inhibited parasite growth completely whereas at 10 µM inhibition was partial ( Figure 1A ) . Epimastigotes treated with CsA for 24 h remained viable and were indistinguishable from the control untreated parasites ( morphology and motility ) , although their growth was impaired . Loss of viability was observed when epimastigotes were treated with CsA for 48 h or longer ( data not shown ) . Other calcineurin inhibitors , such as FK-506 and INCA-6 , were also tested , with similar results ( Figure 1B ) . Among the inhibitors tested , CsA and FK-506 were more effective than INCA-6 to inhibit parasite proliferation . No inhibition was observed after treatment with kaempferol ( Figure 1B ) . These data indicate that TcCaN plays an important role at different stages in T . cruzi development . The nucleotide sequence corresponding to a gene coding for the catalytic subunit of the CL strain CaN , which is distinct from that previously reported [15] and was therefore called TcCaNA2 , was obtained by RT-PCR amplification using primers based on a genomic sequence of clone CL Brener and deposited in GenBank under accession number HM854297 . In silico analysis using BLASTp showed 44% identity between TcCaNA2 and CL strain TcCaNA ( accession number EU195113 ) , described by Araya et al . [15] , and between TcCaNA2 and the catalytic subunit A ( CnA-like: accession number AJ878872 ) , described by Moreno et al . [14] for CL Brener clone . The TcCaNA2 nucleotide sequence showed an open reading frame ( ORF ) of 1179 bp , which codes for a polypeptide of 392 amino acids with an estimated molecular mass of 44 . 5 kDa . The deduced amino acid sequence of TcCaNA2 possesses the domain that interacts with CaNB , but the calmodulin-binding domain and AID are absent ( Figure 2A ) . The conserved GDXHG , GDXVDRG and GNHE ( phosphoesterase ) motifs , which are characteristic of serine-threonine protein phosphatases , are present ( Figure 2A ) . A strong homology between TcCaNA2 and CaNA from other species was found: 44% identity with Homo sapiens and 45% with Drosophila melanogaster and Neurospora crassa . Among trypanosomatids , TcCaNA2 showed 46% and 67% identity with CaNA from Leishmania major and Trypanosoma brucei , respectively . The neighbor-joining cladogram revealed two major clusters , one of which contains CaNA from three trypanosomatids: T . brucei TREU927 ( XP_822888 . 1 ) , Leishmania infantum ( XP_001469789 . 1 ) and T . cruzi ( TcCaNA2-CL , HM854297 ) . This shows the ancestral relation of these organisms and the closer relationship between the two Trypanosoma species ( Figure 2B ) ; the CaNA2 gene of T . cruzi is the most conserved among the species analyzed . The second cluster included Schizosaccharomyces pombe 972 h- ( NP_596178 . 1 ) , D . melanogaster ( Q27889 . 2 ) and the alpha isoform of H . sapiens ( EAX06123 . 1 ) , showing an evolutionary variation among species ( Figure 2B ) . A BLASTn search [http://www . genedb . org; Gish , W . ( 1996–2006 ) ; http://blast . wustl . edu] retrieved the 21178 bp contig 8328 , which contains one copy of the TcCaNA2 gene; a query in the kinetoplastid genome database ( 2012 The EuPathDB Project Team; http://tritrypdb . org/tritrypdb/ ) showed that the TcCaNA2 gene is located on chromosome TcChr37P , which belongs to the non-Esmeraldo haplotype of CL Brener clone . However , no sequence similarity was found within chromosome TcChr37S from Esmeraldo haplotype ( data not shown ) . Southern blot analysis of CL strain genomic DNA showed a simple pattern of hybridization with a probe derived from TcCaNA2 gene , with two internal BamHI restriction sites ( Figure 3A ) . The same probe revealed only one chromosomal band ( chromosomal band XX ∼3 . 27 Mb ) in clone CL-Brener and CL strain chromosomes resolved in pulsed field gel electrophoresis ( Figure 3B ) . This hybridization profile suggests that a single copy of TcCaNA2 gene is present , although the presence of two co-migrating chromosomal bands cannot be ruled out . Northern blot hybridizations showed a transcript of approximately 1 . 5 kb in all developmental stages of the parasite ( Figure 3C ) , indicating that the transcription of TcCaNA2 is constitutive . Densitometric analysis on the same autoradiogram hybridized with a tubulin-derived probe confirmed equal loading of the samples ( data not shown ) . Similarly , using a pair of primers flanking the ORF of the TcCaNA2 gene , an RT-PCR on cDNA freshly synthesized from parasite mRNA revealed amplification in all developmental forms ( Figure 3D ) , confirming what was observed in the Northern blot ( Figure 3C ) . Correspondingly , TcCaNA2 protein was also detected in all developmental stages ( TCT , MT , A and E ) , being the protein levels slightly greater in MT and TCT , as demonstrated by immunoblotting ( Figure 3D ) . The activity of the purified recombinant TcCaNA2 was determined after its purity and specific recognition by polyclonal monospecific anti-TcCaN2 antibody had been confirmed . Recombinant TcCaNA2 was incubated in the absence or presence of 1 mM CaCl2 , MgCl2 , MnCl2 or NiCl2 , and the reaction using p-NPP as substrate proceeded for 15 min at 30°C . Absorbance reading at 405 nm revealed higher TcCaNA2 activity in the presence of Mn2+ or Ni2+ , indicating that these metal ions are more effective cofactors than Ca2+ or Mg2+ ( Figure 4A ) . Our data showing that the recombinant TcCaNA2 is activated by Mn2+/Ni2+ ( Figure 4A ) are compatible with the report suggesting that enzyme activation by Mn2+/Ni2+ is mainly mediated via the catalytic domain , if it is assumed that the non-catalytic domains of subunit A negatively regulate the activity of calcineurin by acting as intra-molecular inhibitors [24] . As TcCaNA2 interacts with TcCaNB in vivo , the activity of TcCaNA2 combined with TcCaNB was measured , by adding 1 µg of each protein recombinant . TcCaNA2 activity did not depend on its association with TcCaNB ( Figure 4B ) . Taking into account that the specific activity is expressed as nmol per min per µg protein , the reduced activity of TcCaNA2 , when combined with TcCaNB , is only apparent and is due to the presence of 2 µg protein . Activity of TcCaNA2/TcCaNB was significantly inhibited by EGTA ( Figure 4B ) . To confirm the molecular interaction between TcCaNA2 and TcCaNB , we performed a Far-Western blot assay as described by Wu et al . [23] . The recombinant proteins GST-TcCaNB and 6×His-SUMO-TcCaNA2 were used as bait and target , respectively . As negative controls , BSA and the unrelated fusion protein 6×His-SUMO-CAT were used . Binding of GST-TcCaNB to 6×His-SUMO-TcCaNA2 was shown using the mouse polyclonal anti-TcCaNB antibody and anti-IgG tagged with horseradish peroxidase . As shown in Figure 5A , TcCaNB interacts with TcCaNA . We ascertained that anti-TcCaNB antibody specifically reacts with TcCaNB and does not recognize 6×His-SUMO-TcCaNA2 ( Figure 5B ) . In addition , to show that native TcCaNA2 and TcCaNB are associated in the parasites , we performed an assay to detect the dissociation of the two subunits . Total extracts from epimastigotes ( 1×107 ) were exposed to different concentrations of urea at 4°C , subjected to SDS-PAGE and electroblotted onto PVDF , and TcCaNA2 was revealed using anti-TcCaNA2 antibody and peroxidase-conjugated IgG , followed by ECL . In the absence of urea , a band of approximately 63 . 5 kDa was detected ( Figure 5C ) that is compatible with the size of TcCaNA2/TcCaNB assuming that the molecular mass of TcCaNA2 is 44 . 5 kDa and that of TcCaNB 19 kDa , as determined by Western blot using antibodies specific to these subunits ( Figure 5D ) . After treatment of parasite extracts with urea , there was a change in this profile . A component of about 45 kDa appeared as a weak band in the sample treated with 2 M of urea , had its intensity increased with 4 M of urea and was the sole band detected by anti-TcCaNA2 antibody ( Figure 5C ) . We inferred from this observation that TcCaNA2 and TcCaNB are closely associated in T . cruzi . It has been shown by Moreno et al . [14] in clone CL Brener that CaNA is present predominantly in the parasite nucleus . We examined the cellular distribution of TcCaNA2 in epimastigotes of CL strain . To this end , parasites were fixed in paraformaldehyde , permeabilized with 0 . 5% Triton X-100 , incubated with specific antibodies to TcCaNA2 and then processed for immunofluorescence . Figure 6A shows that TcCaNA2 has a predominantly cytoplasmic localization , with a diffuse and/or slightly speckled pattern . To show that CaN activity was detectable in the parasite cytosol , we performed in vitro assays with high-speed cytosolic extracts from metacyclic trypomastigotes and epimastigotes depleted of phosphates and nucleotides . The cytosolic extracts were assayed in the presence or absence of EGTA to determine the contribution of calcium-dependent activity ( CaN/PP2B ) to total phosphatase activity ( PP1+PP2A+PP2B+PP2C ) . In both parasite forms CaN activity was detected in the cytosol , amounting to 34 . 6% and 40 . 0% of the total phosphatase activity in metacyclic forms and epimastigotes , respectively . Figure 6B shows the result on CaN activity in cytosol of epimastigotes . We had previously found that CaN is implicated in host cell invasion [15] . To determine if TcCaNA2 was involved in that process , we performed an inhibition assay using third-generation AS-ONs ( morpholino oligonucleotides ) associated with the Endo-Porter carrier ( Gene Tools ) . This strategy has been used successfully in many studies [15] , [25]–[28] . Metacyclic trypomastigotes ( 4×107 ) were incubated with 10 µM of sense or antisense morpholino oligonucleotides directed to TcCaNA2 for 24 h and then incubated with HeLa cells for 3 h to analyze their invasive capacity . After washing with PBS , cells were fixed and stained with Giemsa , and the number of intracellular parasites was counted . Parasites treated with antisense TcCaNA2 oligonucleotides showed a significant decrease in invasivity ( 70% ) while no inhibitory effect of sense oligonucleotides on parasite infectivity was observed ( Figure 7A ) . The decrease in TcCaNA2 expression in parasites treated with antisense oligonucleotide but not with their sense counterparts was ascertained by immunoblotting using anti-TcCaNA2 antibodies , 45% according to densitometric analysis ( Figure 7A , lower panel ) . To demonstrate that the reduced cell invasion capacity of metacyclic forms treated with antisense oligonucleotides was not due to diminished parasite viability , we performed an additional assay that consisted of determining the parasite's capacity to migrate through a polycarbonate transwell filter coated with gastric mucin . Viable parasites cross the gastric mucin-coated filter propelled by ATP-driven flagellar movement [20] . Metacyclic forms were treated as above with 10 µM of sense or antisense morpholino oligonucleotides directed to TcCaNA2 with Endo-Porter or with the carrier alone . After 16 h at 28°C , the parasites were used in a parasite migration assay through a mucin layer and in a cell invasion assay . Gastric mucin-coated transwell filters were placed onto the wells of 24-well plates containing metacyclic forms in PBS ( 107/mL ) , and 100 µL PBS were added to the filter chamber . After 1 h at 37°C , 10 µL were collected from the filter chamber to determine the number of parasites . We found that a comparable number of parasites had crossed the gastric mucin-coated filter regardless of whether they had been pretreated with sense or antisense oligonucleotides and Endo-Porter or with Endo-Porter alone ( Figure S1 ) , similar results were observed by Trypan blue exclusión . Although all parasites exhibited high motility , only those treated with antisense oligonucleotide and Endo-Porter had their capacity to invade HeLa cells diminished ( by about 75% ) . Untreated parasites or those treated with sense oligonucleotides and Endo-Porter or with Endo-Porter alone did not show a reduction in their ability to invade these cells . As epimastigote proliferation was reduced by CaN inhibitors ( Figure 1 ) , we examined the contribution of TcCaNA2 in the process using the antisense strategy as above . Epimastigotes in exponential growth phase were treated or not with sense or antisense oligonucleotides against TcCaNA2 for 72 h . Parasites treated with antisense TcCaNA2 oligonucleotides showed a slight decrease in their proliferative capacity ( Figure 7B ) . The decrease in TcCaNA2 expression in epimastigotes treated with antisense oligonucleotides but not with their sense counterparts was ascertained by immunoblotting using anti-TcCaNA2 antibodies , 55% according to densitometric analysis . As shown in Figure 7B ( lower panel ) , there was a marked decrease in TcCaNA2 expression when epimastigotes were treated with antisense oligonucleotides . If TcCaNA2 played a critical role in epimastigote replication , such a decrease would have had a greater effect on proliferation .
T . cruzi calcineurin ( TcCaN ) has been shown to play a role in host cell invasion [15] . Recently , Kulkarni et al . [29] demonstrated that parasites exposed to cyclophilin-trialysin exhibit enhanced binding to and invasion of host cells , leading to higher infectivity via calcineurin activation . In this study we found that TcCaN is also implicated in the process of epimastigote replication . Furthermore , existing knowledge about Ca2+-dependent phosphatase TcCaN was further enhanced by characterization of the gene coding for a new isoform of TcCaNA , the catalytic subunit A , which exerts its activity through its association with the Ca2+-binding regulatory subunit B ( CaNB ) . Unlike the CaNA homologous protein described by Moreno et al . [14] , which is predominantly localized in the parasite nucleus , TcCaNA2 is cytoplasmic . TcCaNA2 contains two functional domains: a catalytic domain homologous to the protein phosphatase 2A and the domain that interacts with TcCaNB , both domains characteristic of previously reported CaNA homologs [14] , [15] . The lack of a calmodulin-binding domain in TcCaNA2 indicates that its activity is independent of calmodulin . Compatible with this was the finding that TcCaNA2 activity is enhanced by Mn2+ rather than by Ca2+ . The catalytic domain of PPP ( phosphoprotein phosphatase ) has the phosphoesterase consensus motif , with three conserved motifs in separate regions showing the configuration DXH ( X ) n GDXXDR ( X ) m GNHD/E [30] , [31] . Although there are mutations in the catalytic site of TcCaNA2 , suggesting that it may behave as a pseudophosphatase [32] , arguments against this idea include the existence in TcCaNA2 of histidine ( H ) in the GNHE domain , which acts as a proton donor in the catalysis [33] , in addition to the four amino acids involved in metal coordination , of which only one is non-conserved . The recombinant protein TcCaNA2 showed enzymatic activity using p-NPP as substrate . Also , near the C-terminal portion of the TcCaNA2 sequence there is the highly conserved SAPNY motif , which is “conventional” in eukaryotic PPP , tyrosine ( Y ) being implicated in the interaction with regulators and inhibitors [34]–[37] . This is in contrast to what is found in other catalytic subunits of CaN [for instance the α isoform of rat CaNA , in which a leucine ( L ) residue is present in the SAPNYL motif making it more susceptible to okadaic acid , a characteristic of phosphatases PP1 and PP2A] [38] . Together with the configuration of the invariant PPP motifs in TcCnA-like , which adjusts to the configuration -GDXHG- , -GDXVXRG- , -GNH- [14] , the differences between TcCnA-like and TcCaNA2 in the hydrophobic profiles of the domain that interacts with TcCaNB suggest that these two calcineurin-type protein phosphatases may play distinct functional roles ( data not shown ) . This is supported by their different subcellular localization . Multiple sequence alignment of TcCaNA2 with catalytic subunits of calcineurin from other organisms , revealed a 20 residues long stretch that is found only in trypanosomatids , whose composition in T . cruzi is 243-VSGGSGSDYYTPSAGPSYGS-262 ( Figure 2A ) . The functional relevance of this sequence is not known . Bioinformatics tools associated the sequence with two motifs: one the NURR type present in orphan nuclear receptors , and the other associated with calpain-type cysteine proteases ( data not shown ) . Between one third and one half of all enzymes described to date must associate with metals to perform their function [39] . In the case of calcineurin , which belongs to the hydrolase class ( EC Number 3 . 1 . 3 . 16 ) and whose systematic name is phosphoprotein phosphohydrolase , Fe3+ and Zn2+ have been described as cofactors in its active site [40]–[42] . Our results showed that , under the conditions assayed , the recombinant TcCaNA2 is activated by Mn2+ and Ni2+ , with no substantial activation by Mg2+ or Ca2+ . By sequentially removing the non-catalytic domains of CaN , such as the calmodulin binding domain and autoinhibitory domain , Liu et al . [24] observed increases in phosphatase activity , clearly demonstrating that non-catalytic domains negatively regulate the activity of the enzyme and act as intra-molecular inhibitors . This sequential domain deletion favors CaN activation by Mn2+/Ni2+ but not by Mg2+ , suggesting that enzyme activation by Mn2+/Ni2+ is mainly mediated via the catalytic domain [24] . Our finding that TcCaNA2 lacks the calmodulin-binding domain and the autoinhibitory domain is consistent with its activation by Mn2+/Ni2+ . Assays using the recombinant TcCaNA2 showed that the enzyme can function when dissociated from the regulatory subunit TcCaNB . However , it should be borne in mind that inside the parasite TcCaNA2 is bound to the Ca2+-binding regulatory subunit TcCaNB ( Figure 5 ) and that the enzyme is activated during T . cruzi invasion of host cells , a process associated with an increase in cytosolic Ca2+ concentration . Because of the presence of a highly hydrophobic CaNB domain , which is located between the catalytic and the calmodulin-binding domains , CaNA and CaNB subunits form a heterodimer [43] . This interaction can only be dissociated by strong denaturing agents such as urea at 6 M [44] . We found that 6 M urea dissociates TcCaNA2 from TcCaNB . Calcineurin contributes in a variety of cellular signaling events and activation processes [45] . In eukaryotic pathogens , it has been associated in the regulation of specific steps of the cell cycle; in intracellular Toxoplasma gondii , the loss of host cell potassium , activates a phospholipase C that , in turn , causes an increase in cytoplasmic [Ca2+] causing the parasite output from host cell , by the activation of at least two signaling pathways: the protein kinase and calcineurin [46] . Similarly , in T cruzi , entry into the host cell is a process dependent [Ca2+] , recent studies show that calcineurin , specifically , the regulatory subunit CaNB is present in this parasite ( TcCaNB ) and is involved in the process of invasion of target cells . Treatment of parasites with antisense phosphorothioate oligonucleotides directed to TcCaNB , which reduced the expression of TcCaNB and affected TcCaN activity , resulted in ∼50% inhibition of HeLa cell entry by MT or TCT [15] . Kumar et al . , [47] , show that CsA inhibits the intraerythrocytic replication of P . falciparum and that both , Cyp19A and Cyp19B , are potent effectors of CsA-mediated inhibition of recombinant P . falciparum CaNA in vitro . CsA-resistant parasites , isolated from erythrocytic cultures , contained mutations in the CaNA and CaNB subunits and in Cyp19A and Cyp19B . Using Geldanamycin ( GA ) , an inhibitor of plasmodial Hsp90 [48] , [49] , they also show that parasitic Hsp90 is associated with CaN , strongly suggesting that Hsp90 regulates CaNA folding and hence regulates all cellular events that require the phosphatase activity of calcineurin , so the antimalarial activities of CsA and GA would be synergistic . Potenza et al . , [50] , studied cDNA clones encoding cyclophilin isoforms in epimastigotes of T . cruzi . These genes were also detected in amastigotes and trypomastigotes . Four cyclosporin A-binding proteins were isolated in epimastigote extracts , which were identified by mass spectrometry as TcCyP19 , TcCyP22 , TcCyP28 and TcCyP40 , these cyclophilins of T . cruzi would be of importance to the mechanism of action of CsA . In the present study , CsA and others calcineurin inhibitors , inhibit the invasion and proliferation processes . On the other hand , calcineurin is involved in the morphogenesis and virulence of multiple pathogenic fungi: in Candida spp . , calcineurin participates in antifungal drug resistance/tolerance , survival in serum , and virulence [51]–[54]; in Paracoccidioides brasiliensis , calcineurin plays a role in morphogenesis [55]; and in Aspergillus fumigatus , calcineurin regulates morphogenesis and thereby pathogenesis [56] . In addition , calcineurin is essential for growth at elevated temperatures in the human fungal pathogen Cryptococcus neoformans [57] . Studies conducted in the protozoan parasite Leishmania major suggest that Ca2+ influx and activation of calcineurin signaling is required for parasite differentiation and adaptation to cellular stress encountered ( elevated temperatures ) during infection of the mammalian host [58] . In this study , the disruption of calcineurin function , achieved by deletion of the gene encoding the CnB subunit , had no effect on promastigote growth at 27°C or the development of infectious metacyclic promastigotes in stationary-phase cultures . However , disruption of calcineurin function was associated with a marked increase in the sensitivity of promastigotes to elevated temperature and perturbations in membrane lipid composition . In our study , the temperature changes do not affect the infective form ( MT ) in motility or viability , both ATP-dependent processes [20] , as demonstrated by the parasite migration assay through gastric mucin layer . The decrease of the invasive capacity of metacyclic forms , treated and not treated with antisense oligonucleotides at 28°C , was not due to diminished viability , because the parasites that were incubated at 37°C crossed through a polycarbonate transwell filter coated with gastric mucin , demonstrating that are not affected by the temperature stress . TcCaNA2 is expressed in all T . cruzi developmental forms ( Figure 3D ) . In metacyclic trypomastigotes , the decrease in TcCaNA2 expression brought about by the use of a TcCaNA2-targeted anti-sense strategy resulted in reduced capacity to invade host cells . In previous studies , the anti-sense approach directed to inhibit TcCaNB , the regulatory subunit with EF-Hands Ca2+-linking motifs , rendered metacyclic forms as well as tissue culture-derived trypomastigotes less infective toward target cells [15] . It was also demonstrated that T . cruzi protein dephosphorylation by TcCaN is in fact associated with a decrease in parasite internalization assuming that treatment of metacyclic forms with TcCaN inhibitor CsA , which diminishes the phosphorylation levels of serine/threonine residues of high-molecular-weight proteins , inhibits host cell invasion [15] . An expansion of the serine/threonine phosphatase family and a low proportion of tyrosine phosphatases have been found in T . cruzi compared with other eukaryotic genomes [32] . Another role of TcCaN is its involvement in parasite multiplication . TcCaN inhibitor CsA affected epimastigote multiplication , confirming the data reported by Búa et al . [59] . CsA probably binds to the cytosolic protein cyclophilin ( CyP ) , forming a complex that , through its association with the invariant regulatory subunit TcCaNB , would inhibit TcCaN in a manner similar to that of its mammalian counterpart [60] . Members of the CyP family identified in T . cruzi , called peptidyl-prolyl cis/trans isomerases , have their activity inhibited by CsA and its analogs [59] , and the affinity of CyP for CsA has been documented [61] . It appears that the involvement of TcCaNA2 in epimastigote proliferation is partial , as judged by the weak inhibitory effect on epimastigote replication following treatment with antisense oligonucleotides against TcCaNA2 . It is possible therefore that the major contribution to epimastigote replication comes from TcCaNA bound to TcCaNB rather than from TcCaNA2/CaNB . Assuming that T . cruzi expresses TcCaNA and TcCaNA2 , both of which can associate with TcCaNB , one interesting possibility is that the two isoforms are engaged in distinct events during the parasite life cycle . TcCaNA2/TcCaNB present in T . cruzi cytosol would be predominantly activated in the infective trypomastigote forms , leading to dephosphorylation of serine/threonine residues of proteins implicated in cell invasion . In favor of this view is the fact that a partial decrease in TcCaNA2 expression resulted in a significant reduction in metacyclic trypomastigote internalization , a short process involving signaling events in the cytosol , whereas TcCaNA/TcCaNB may be activated in replicative epimastigote forms to promote parasite proliferation , which implicates transcription in the nucleus [62] . In addition to being inhibited by CsA , epimastigote multiplication is inhibited by FK506 and INCA-6 , which are also CaN inhibitors . In a mechanism similar to that associated with CsA , FK506 exerts its inhibitory effect by forming a drug-immunophilin complex with CaN [63] , whereas the mechanism of action of INCA-6 is linked to blocking of the substrate recognition site by a covalent union to CaN , inhibiting dephosphorylation of nuclear factor of activated T cells ( NFAT ) and interrupting the formation of the CaN-NFAT complex [64] . Similarly , kaempferol , binds directly to the catalytic site in CaNA interacting with Leu 312 [65] . Our results showed no inhibition in proliferation and cell invasion in parasites treated with kaempferol . These findings are consistent with the absence of Leu 312 in TcCaNA2 ( Figures 1B and 2A ) . The mechanism involving phosphorylation/dephosphorylation events that play an important role in cell cycle progression , may operate in T . cruzi epimastigotes . It is of note that T . cruzi TcCnA-like protein has been detected predominantly in the nucleus of this parasite [14] . TcCaNA2/CaNB and TcCaNA/CaNB , which have distinct cellular localizations , may play a critical role at different stages of T . cruzi development . Many authors suggest calcineurin as a potential chemotherapeutic target against pathogenic; fungi , helminths and protozoa [46] , [47] , [66]–[70] . Our revision strongly supports that TcCaNA2 is a good candidate for chemotherapeutic target given the differences with its human counterpart . In fact , the T . cruzi calcineurin does not possess the calmodulin binding domain and the autoinhibitory domain which are present in the human enzyme . Besides this , TcCaNA2 presents a 20 amino acids long stretch ( 243-VSGGSGSDYYTPSAGPSYGS-262 ) in the catalytic domain that is absent in the human calcineurin and is conserved in all trypanosomatids with minimal differences . Also , comparative sequence analysis shows only 44% of identity between human calcineurin and TcCaNA . Taking into account that TcCaNA2 differs considerably in its primary structure from human CaNA and that it may play a key role in host cell invasion by T . cruzi , it should be considered a potential target for chemotherapeutic intervention in Chagas disease . | Chagas disease is a neglected tropical parasitic infection . An estimated 10 million people are infected worldwide , and more than 25 million people are at risk of acquiring the disease . The therapeutic agents used to treat the disease may not be effective in all cases and also produce considerable side effects . Therefore , it is important to identify the key factors in the life cycle of the parasite that could be targets for new chemotherapeutic strategies . This paper provides evidence that a new cytoplasmic catalytic subunit of T . cruzi calcineurin ( TcCaNA2 ) may play a critical role in host cell invasion by metacyclic trypomastigotes . Metacyclic forms with reduced TcCaNA2 expression following treatment with antisense morpholino oligonucleotides had significantly decreased capacity to invade HeLa cells . Epimastigote proliferation was inhibited to some extent by treatment with an antisense morpholino oligonucleotide targeted to TcCaNA2 , but to a lesser degree than by calcineurin inhibitors ( CsA , FK506 and INCA-6 ) . The structural differences between TcCaNA2 and its human ortholog CaNA were analyzed to determine the potential of this newly identified calcineurin subunit as a chemotherapeutic target . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"biochemistry",
"molecular",
"cell",
"biology",
"protozoology",
"biology",
"microbiology",
"pathogenesis",
"parasitology"
]
| 2014 | A Cytoplasmic New Catalytic Subunit of Calcineurin in Trypanosoma cruzi and Its Molecular and Functional Characterization |
Electrical synapses between neurons , also known as gap junctions , are direct cell membrane channels between adjacent neurons . Gap junctions play a role in the synchronization of neuronal network activity; however , their involvement in cognition has not been well characterized . Three-hour olfactory associative memory in Drosophila has two components: consolidated anesthesia-resistant memory ( ARM ) and labile anesthesia-sensitive memory ( ASM ) . Here , we show that knockdown of the gap junction gene innexin5 ( inx5 ) in mushroom body ( MB ) neurons disrupted ARM , while leaving ASM intact . Whole-mount brain immunohistochemistry indicated that INX5 protein was preferentially expressed in the somas , calyxes , and lobes regions of the MB neurons . Adult-stage-specific knockdown of inx5 in αβ neurons disrupted ARM , suggesting a specific requirement of INX5 in αβ neurons for ARM formation . Hyperpolarization of αβ neurons during memory retrieval by expressing an engineered halorhodopsin ( eNpHR ) also disrupted ARM . Administration of the gap junction blocker carbenoxolone ( CBX ) reduced the proportion of odor responsive αβ neurons to the training odor 3 hours after training . Finally , the α-branch-specific 3-hour ARM-specific memory trace was also diminished with CBX treatment and in inx5 knockdown flies . Altogether , our results suggest INX5 gap junction channels in αβ neurons for ARM retrieval and also provide a more detailed neuronal mechanism for consolidated memory in Drosophila .
Pavlovian olfactory learning in Drosophila melanogaster , the fruit fly , is a well-characterized behavioral paradigm in which flies are subjected to a training session of sequential exposures to two distinct odors ( conditioned stimulus , CS ) with or without electric foot shock ( unconditioned stimulus , US ) [1] . The assay involves CS-US coincidence detection in the mushroom bodies ( MBs ) , the olfactory learning and memory centers of the fly brain . The MBs are a pair of neuropils composed of around 2 , 000 major intrinsic neurons , called the Kenyon cells ( KCs ) , in each brain hemisphere[2] . The dendrites of the MB neurons form a calyx , and their axons project anteriorly through the peduncle to give rise to the αβ , α'β' , and γ lobes in the middle brain . Three hours after a single training session consists of two genetically distinct forms of memory , anesthesia-sensitive memory ( ASM ) and anesthesia-resistant memory ( ARM ) , with each accounting for about half of the retention level[3–5] . Synaptic transmission in the brain has two different modalities , chemical and electrical synapses . Neurons mainly use neurotransmitters or neuropeptides to communicate and regulate one another’s functions , which is mediated by chemical synapses . In contrast , electrical synaptic transmission depends on clusters of intercellular channels called gap junctions , which form the pores approximately 1 . 2 nm in diameter between neurons[6] . These pore structures allow diffusion of small molecules and ions , thus enabling bidirectional electronic signal transmission between neurons . The synchronization of neuronal activity in the hippocampus is mediated by gap junctions in the mammalian brain , which is critical for memory consolidation[7 , 8] . In the human and mouse genomes , 21 and 20 gap junction genes have been identified , respectively[9] . The connexin and pannexin gap junction gene families are found in vertebrates , whereas the innexin ( inx ) gene family is found in invertebrates[10 , 11] . Drosophila melanogaster has 8 gap junction genes , named inx1-inx8 . Our previous study showed that two MB modulatory neurons in the Drosophila brain , the anterior paired lateral ( APL ) and dorsal paired medial ( DPM ) neurons , formed heterotypic gap junction channels via INX6 and INX7 , and that disrupting communication through these gap junctions impaired 3-hour ASM[12] . Moreover , a recent study indicated that gap junctions in αβ , α'β' , and MB output neurons ( MBON-β'2mp ) were involved in Drosophila visual learning[13] . To determine whether gap junctions in MB neurons are essential for olfactory memory formation , we knocked down each innexin gene in MB neurons and found that only the downregulation of inx5 specifically disrupted 3-hour ARM . Consistent with this result , whole-mount brain immunostaining showed INX5-positive signals in the somas , calyxes , and lobes of the MBs , suggesting the existence of gap junction channels between MB neurons . Knockdown of inx5 in αβ , but not α'β' or γ , neurons disrupted ARM , indicating that INX5 in αβ neurons was involved in ARM formation . Furthermore , adult-stage-specific knockdown of inx5 in αβ neurons disrupted ARM , demonstrating that the ARM deficiency was not caused by defects in MB development . We performed a transient inhibition of the action potential in αβ neurons by expressing an engineered halorhodopsin protein ( eNpHR ) [14] , which acts as a light-driven chloride pump , specifically during memory retrieval , but not during acquisition or consolidation . This also led to the disruption of ARM , suggesting that INX5 was involved in ARM retrieval in αβ neurons . We observed a training-induced increase in the proportion of odor-responsive αβ neurons to the training odor ( CS+ odor ) 3 hours after conditioning , and this phenomenon was disrupted by treatment with the gap junction blocker carbenoxolone ( CBX ) . Finally , we found increased calcium responses to the training odor in the MB α-lobe branch region 3 hours after conditioning , and this increased calcium response was diminished by both gap junction blocker CBX treatment and in inx5 knockdown flies . These data suggest that INX5 channels coordinate the MB neuronal activity changes to training odor at 3-hour after odor/shock association . Together , our results show that ARM retrieval in Drosophila is mediated by gap junction channels composed of INX5 in αβ neurons .
In our previous study , we found that heterotypic gap junctions in two MB modulatory neurons , the APL and DPM neurons , are required for 3-hour ASM formation[12] . In addition , the existence of gap junctions in MB neurons was reported in a recent study[13] . We therefore sought to examine whether gap junctions in MB neurons were involved in olfactory memory formation . We used a Drosophila RNAi library to express UAS-inxRNAi transgenes under the control of OK107-GAL4 to individually silence the inx genes in the entire MBs[15 , 16] . We found that only inx5 knockdown disrupted 3-hour memory ( Fig 1A and S1A Fig ) . In the fly , olfactory 3-hour memory consists of the labile ASM and consolidated ARM . We applied 2-min cold-shock anesthetization at 2-hour after training to abolish ASM , and tested the 3-hour ARM retention[5 , 17–20] . Interestingly , the memory defect persisted after the cold-shock treatment in inx5 knockdown flies , suggesting that the memory loss was attributable to the disruption of ARM rather than ASM ( Fig 1B–1D; S1B–S1E Fig ) . It has been shown that inx5 is expressed 50 hours after pupal formation[21] . To further characterize INX5 protein expression in the fly brain , a rabbit polyclonal antibody recognizing Drosophila INX5 was generated . Whole-mount brain immunohistochemistry with this antibody showed that INX5 was expressed in the MB calyxes , somas , and lobes ( Fig 2A–2C and S2 Fig ) . Western blotting confirmed that the INX5 protein levels were dramatically decreased in head extracts from two independent UAS-inx5RNAi flies ( v6950 and JF02877 ) in which the RNA transgene was under the control of the pan-neuronal GAL4 driver , elav-GAL4 ( Fig 2D ) . Furthermore , quantitative brain immunohistochemistry indicated that two independent UAS-inx5RNAi flies with the transgene under the control of OK107-GAL4 had greatly decreased INX5 levels in the MBs ( Fig 2E ) . Brain immunostaining showed INX5-positive signals in most if not all MB calyxes and somas . To identify the subset of MB neurons in which INX5 expression is required for ARM formation , we performed behavioral screening using RNAi-mediated inx5 knockdown with GAL4 drivers specific for different MB neurons . VT30604-GAL4 , VT44966-GAL4 , and C739-GAL4 drive expression of UAS-inx5RNAi in α′β′ , γ , and αβ neurons respectively ( Fig 3A ) . Genetic knockdown of INX5 in αβ neurons , but not in α′β′ or γ neurons , disrupted ARM , suggesting that INX5 in αβ neurons regulates the ARM process ( Fig 3B ) . It is important to consider that C739-GAL4 is not exclusively expressed in αβ neurons ( Fig 3C ) . We therefore combined the MB-GAL80 transgene to reduce GAL4 expression in the MB neurons . The presence of the MB-GAL80 transgene specifically abolished GAL4 activity in MB neurons , but left its expression unchanged in non-MB neurons ( Fig 3D ) . Three-hour ARM of C739-GAL4/MB-GAL80; UAS-inx5RNAi/+ flies were statistically indistinguishable from both wild-type and MB-GAL80/+ flies and were also statistically different from that of C739-GAL4/+; UAS-inx5RNAi/+ flies ( Fig 3E ) . Moreover , this behavioral result was further confirmed in an additional GAL4 line ( VT49246-GAL4 ) with specifically labeled αβ neurons in the fly brain ( Fig 3F and 3G ) . Furthermore , our previous study found that glutamate release from MB αβ output neurons ( MBON-β2β′2a ) was required for ARM[20] . To test the possibility that gap junctions between αβ neurons and MBON-β2β′2a are involved in ARM formation , we genetically knocked down each inx gene in MBON-β2β′2a and found that all the modified flies displayed normal ARM ( S3 Fig ) . In addition , using a dye-coupling approach , Liu and colleagues found no gap junction connectivity between αβ neurons and their target MBONs[13] . Furthermore , our pervious study found that knockdown of each inx gene in the projection neurons and APL neurons via expressing individual UAS-inxRNAi by GH146-GAL4 did not impair ARM[12] . Based on these results taken together , we conclude that ARM formation requires gap junctions between αβ neurons . In order to rule out the possibility that the chronic RNA-mediated knockdown of inx5 causes developmental defects in the MBs , we examined the gross morphologies of the MBs in the inx5-manipulated flies and found no significant differences as compared to control flies , suggesting that the MB structure was unaffected by the chronic inx5 knockdown ( S4 Fig ) . Furthermore , we used an inducible knockdown strategy to silence INX5 expression specifically in the adult stage utilizing a temperature-sensitive GAL80 repressor ( tubP-GAL80ts ) . Inducible knockdown of INX5 still induced significant impairment of 3-hour ARM , but not initial learning ( Fig 4 ) , suggesting that INX5 in αβ neurons is required post-developmentally for ARM formation . Consistent with previous studies , blocking chemical synaptic transmission by a temperature sensitive shibire ( shits ) in αβ neurons during memory retrieval[20 , 22] but not during acquisition or consolidation ( S5A Fig ) , disrupted ARM . The shibire gene encodes the Drosophila homologue of dynamin , which is required for the fission of endocytic vesicles from the presynaptic membrane . A dominant-negative mutant form of shibire interferes with the recycling of neurotransmitters but cannot block the gap junction-mediated propagation of action potential between adjacent neurons[23–27] . We therefore applied optogenetic tools to transiently block action potential in αβ neurons during memory acquisition , consolidation , or retrieval which subsequently allowed us to examine the specific memory phase of ARM that gap junctions were involved in . We overexpressed eNpHR to transiently hyperpolarize αβ neurons during different phases of the ARM process . The eNpHR-mediated hyperpolarization of αβ neurons during memory retrieval impaired ARM in C739-GAL4 > UAS-eNpHR and VT49246-GAL4 > UAS-eNpHR flies . In contrast , eNpHR-mediated hyperpolarization of αβ neurons during memory acquisition or consolidation did not impair ARM , suggesting that the activity in αβ neurons is only required for ARM retrieval ( Fig 5 and S5B Fig ) . Although the eNpHR-mediated hyperpolarization in αβ neurons also affects the secretion of neuropeptides , blocking these secretions ( e . g . amnesiac ) in αβ neurons does not affect ARM [28] . Knockdown of inx5 gap junction gene in αβ neurons disrupted ARM and highlights the role of gap junctions between αβ neurons[23–27] ( Figs 2 , 3 , and 4 ) . ARM deficiency was only observed in all-trans-retinal-fed flies indicates that the ARM defect was caused by the transient blockade of αβ neuronal activity through eNpHR-mediated neuronal silencing ( Fig 5C and S5B Fig ) . Taken together , these data suggest that INX5 gap junctions are involved in ARM retrieval from αβ neurons in the fly brain . To further test the hypothesis that INX5 is involved in ARM retrieval , the calcium indicator GCaMP6 was applied to monitor αβ neuronal activity during ARM retrieval[29] . The preferential expression of INX5 in the somas of MB neurons led us to examine the αβ neuronal activity changes in this region . We first observed that odor/shock association increases the proportion of odor responsive αβ neurons to the conditioned odor ( CS+ odor ) at 3-hour after a 2-min cold shock given at 2-hour after training ( Fig 6A ) . The increased CS+ odor responsive αβ neurons were diminished by treatment with the gap junction blocker CBX , 10 min before image recording ( Fig 6A ) . To further analyze whether this increased CS+ odor responsive αβ neurons correlate to the memory trace[30 , 31] , we recorded the neuronal activity in the MB α- and β- lobes respectively . The MB α-lobe branch has been shown to produce training-induced modifications in odor-evoked cellular calcium responses[30] . We visualized the functional responses of naïve flies to different odors , 3-octanol ( OCT ) or 4-methyl-cyclohexanol ( MCH ) , in the αβ lobes by expressing UAS-GCaMP6m under the control of C739-GAL4 . However , only the MB α-lobe branch exhibited a significantly elevated GCaMP6 intensity 3 hours after shock/odor association as compared to naïve flies ( Fig 6B–6C and S6 Fig ) . These results indicated that the MB α-lobe branch exhibits a training-induced increase in calcium responses 3 hours after training . The increased calcium response was abrogated by treatment with the gap junction blocker CBX 10 min before image recording , and was recovered after CBX washout ( Fig 6B–6C ) . Although a significant 3-hour memory trace was also observed in α'β' neurons , this phenomenon still occurred with CBX treatment suggesting that the memory trace is independent of the gap junction ( S7 Fig ) . Finally , genetic knockdown of inx5 in αβ neurons also abolished the increased calcium responses in the MB α-lobe to the training odor at 3-hour after conditioning , which suggests that INX5 gap junctions are required for the branch-specific modification of neuronal responses to the conditioned odor during memory retrieval ( Figs 6 and 7 ) .
In fruit flies , two parallel MB circuits , containing αβ and α'β' neurons , are involved in ARM formation . Radish expression in αβ neurons is required for partial ARM , whereas octβ2R expression in MB α'β' neurons is required for the rest part of ARM , suggesting that two distinct cellular mechanisms regulate ARM in different MB neurons[17 , 19 , 20 , 32] . The radish gene encodes a protein with a predicted cAMP-dependent protein kinase phosphorylation site , which can bind Rac1 to regulate the rearrangement of the cytoskeleton and affect synaptic structural morphology[32] . The interaction of RADISH and BRUCHPILOT at the synaptic active zone has been proposed to regulate neurotransmitter release[17] , and genetic knockdown of radish or bruchpilot in αβ neurons disrupts ARM[17 , 20] . A recent study indicated that Drk–Drok signaling is essential for ARM formation in αβ neurons , and related to dynamic cytoskeletal changes[33] . In addition , the dopamine type 2 ( D2R ) and serotonin ( 5HT1A ) receptors in αβ neurons are also critical for ARM formation[34 , 35] . The key finding of our study is that the gap junction protein INX5 in αβ neurons is critical for 3-hour ARM retrieval . This conclusion is supported by four independent lines of evidence . First , immunohistochemistry data indicated that INX5 was preferentially expressed in the MB calyxes and somas , and these INX5-positive signals were reduced in OK107-GAL4 > UAS-inx5RNAi flies ( Fig 2 ) . Second , adult-stage-specific knockdown of inx5 in αβ neurons impaired ARM ( Fig 4 ) . Third , eNpHR-mediated inhibition of action potential in αβ neurons during retrieval also impaired ARM ( Fig 5 ) . Forth , knockdown of inx5 in αβ neurons inhibited the training-induced cellular calcium responses in the MB α-lobe region 3 hours after odor/shock association ( Fig 6 ) . Previous studies have concluded that αβ neuronal activity is involved in 3-hour memory retrieval using shibirets to transiently block chemical synaptic transmissions via inhibiting neurotransmitter recycling[22 , 36] . Three-hour memory is composed of ASM and ARM , each accounting for about half of the memory retention level[3–5] . In our recent study , we further showed that the inhibition of neurotransmitter recycling in αβ neurons during memory retrieval disrupted 3-hour ARM[20] . However , blocking neurotransmitter recycling in αβ neurons during memory acquisition and consolidation did not affect 3-hour ARM ( S5A Fig ) . The function of gap junctions in the electrical synapses is to coordinate the propagation of action potential in neuronal networks[23–27] , and shibirets cannot block gap junction-mediated electrical synapses . We therefore used eNpHR to transiently silence action potential in αβ neurons to confirm the requirement of αβ neuronal activity during the ARM formation process . Our data showed an eNpHR-mediated hyperpolarization of αβ neurons during memory retrieval but not acquisition or consolidation , impaired ARM , suggesting that action potential in αβ neurons is required only for ARM retrieval ( Fig 5 and S5B Fig ) . Brain immunostaining data showed that INX5 gap junction proteins are strongly expressed in the calyxes and somas of αβ neurons ( Fig 2 and S2 Fig ) , and knockdown of inx5 gap junction gene in αβ neurons disrupted ARM ( Figs 3 and 4 ) . The expression of gap junction is critical for neuronal functions since it plays a role in the propagation of action potential between adjacent neurons[23–27] . We therefore conclude that the gap junction channels composed of INX5 in αβ neurons are critical for ARM retrieval ( Fig 7 ) . A recent study showed that the gap junction protein INX2 regulates calcium transmission across the follicle cells during Drosophila oogenesis[37] . In addition , INX1/INX2 induces calcium oscillations in the glial cells of the blood-brain barrier ( BBB ) , enabling signal amplification and synchronization across the BBB in fruit flies[26] . Furthermore , the gap junction protein INX6 is important for promoting synchronous neuronal activity in the dorsal fan-shaped body ( dFB ) in the fly brain that is critical for the sleep switch[38] . In mammals , most neuronal gap junctions in the brain are composed of Connexin-36 ( Cx36 ) and are involved in synchronizing the hippocampal neuronal oscillatory patterns[39] , which is required for emotional memories[8] . Therefore , it is possible that gap junction channels composed of INX5 mediate neuronal activity amplification and synchronization across αβ neurons , boosting the synaptic output strength during ARM retrieval ( Fig 7 ) . By using the newly developed calcium indicator GCaMP6[29] , we observed the increased proportion of training odor-responsive αβ neurons 3 hours after odor/shock association , and this phenomenon was abolished after treatment with gap junction blocker CBX ( Fig 6A ) . Furthermore , we observed significant enhancement of the training-induced cellular calcium response to the training odor in the MB α-lobe branch 3 hours after odor/shock association ( Fig 6B–6C ) . According to the broad consensus of the field , the memory trace is supposed to be formed in the vertical lobe of the MBs by the activity contingency of MBs and dopaminergic Protocerebral Posterior Lateral 1 ( PPL1 ) neurons , which represent odor and punitive shock , respectively[40–42] . Therefore , it is possible that 3-hour memory trace back propagation of somas’ activity occurs from the MB lobes during memory retrieval . In addition , the branch specific modifications via MB input neurons ( e . g . , Protocerebral Anterior Medial , PAM ) may occur during memory retrieval[43 , 44] , hence the memory trace was only observed in α-lobe branch but not the β-lobe of MBs ( Fig 6B–6C & S6 Fig ) . This training-induced 3-hour ARM-specific memory trace was eliminated by treatment with the gap junction blocker , CBX , during memory retrieval or by genetic knockdown of inx5 in αβ neurons . Although a significant 3-hour ARM-specific memory trace was also observed in α'β' neurons , this phenomenon was independent of the gap junction ( S7 Fig ) . From this , we propose that an unknown dynamic mechanism regulates the permeability of gap junction channels composed of INX5 in αβ neurons after training . Recently , cryoelectron microscopy revealed that the structure of C . elegans INX6 was highly similar to that of the vertebrate gap junction protein Connexin-26 ( Cx26 ) [45] . Connexin properties , such as gating and assembly , can be regulated by phosphorylation[46 , 47] . Additionally , the functions of Innexins or Connexins can also be regulated by changes in the intracellular pH and calcium levels[48–50] . Establishing whether the properties of INX5 in the MBs are modified following conditioned training will provide insights into the neuronal mechanisms of ARM .
Flies were raised on standard cornmeal food media at 25°C and 70% relative humidity under a 12:12-hour light: dark cycle . The “Cantonized” w1118 w ( CS10 ) strain was used as a wild-type control . The MB-GAL80 , UAS-GCaMP6m , and OK107-GAL4 flies were obtained from Bloomington Drosophila stock center . The UAS-eNpHR-YFP; UAS-eNpHR-YFP fly line was obtained from Ann-Shyn Chiang . The RNAi lines were obtained from the Vienna Drosophila RNAi Center or TRiP RNAi fly stocks . All RNAi lines from the Vienna Drosophila RNAi Center have been described previously[12] . The VT30604-GAL4 , VT44966-GAL4 , C739-GAL4 , VT49246-GAL4 , VT0765-GAL4 , UAS-shits , and tubP-GAL80ts flies have been described[20] . Brain samples were stained with the mouse 4F3 anti-discs large ( DLG ) monoclonal antibody ( Hybridoma Bank ) to label all neuronal synapses , or with a rabbit polyclonal anti-INX5 antibody . The rabbit INX5 antibody was generated by Antibody International , Inc . , with an HPLC-purified synthetic peptide , NH2– PHFRSSLRRIGEYNEAYAR–COOH , selected from the INX5 sequence . Fixed brain samples were incubated in PBS containing 1% Triton X-100 and 0 . 25% normal goat serum ( PBS-T ) with mouse 4F3 anti-DLG antiserum ( 1:10 ) or rabbit anti-INX5 ( 1:1 , 000 ) as primary antibodies at 25°C for 1 day . After three washes in PBS-T , the samples were incubated in biotinylated goat anti-mouse or rabbit IgG ( 1:200; Invitrogen ) at 25°C for 1 day . Next , the brain samples were washed and incubated in Alexa Fluor 633 streptavidin ( 1:500; Invitrogen ) at 25°C overnight . After extensive washing , the brain samples were cleared and mounted in FocusClearT ( CelExplorer ) for confocal imaging . Sample brains were imaged under a Zeiss LSM 700 confocal microscope with either a 40× C-Apochromat water-immersion objective lens for whole-brain images ( N . A . value , 1 . 2; working distance , 220 μm ) or a 63× glycerol-immersion objective lens for horizontal , sagittal , and frontal cross sections ( N . A . value , 1 . 4; working distance , 170 μm ) . To overcome the limited field of view , some samples were imaged twice , one for each hemisphere , with overlap in between . We then stitched the two parallel image stacks into a single dataset online with the ZEN software , using the overlapping region to align the two stacks . Groups of approximately 100 flies were exposed first to one odor ( CS+; OCT or MCH ) paired with 12 1 . 5-s pulses of 75-V DC electric shock presented at 5-s interpulse intervals . This was followed by the presentation of a second odor ( CS–; MCH or OCT ) without electric shock . In the testing phase , the flies were presented with a choice between the CS+ and CS–odors in a T-maze for 2-min . At the end of the 2-min period , the flies in each T-maze arm were trapped , anesthetized , and counted . From the distribution of flies between the 2 arms , the performance index ( PI ) was calculated as the number of flies avoiding the shock-associated odor ( CS+ ) minus the number avoiding the non-shock-associated odor ( CS– ) , divided by the total number of flies and multiplied by 100 . If the flies did not learn , they were distributed equally between the 2 arms; hence , the calculated PI was 0 . If all flies avoided the shock-paired odor and were distributed 0:100 between the CS+ and CS–arms in the T-maze , the PI was 100 . To assess learning , performance was measured immediately after training . To evaluate intermediate-term memory , testing was performed 3 hours after training . ARM was defined as 3-hour memory after a 2-min cold shock presented at 2-hour post-training ( i . e . , 1 hour before testing ) by placing a plastic vial containing the trained flies in ice water . A brief cold shock completely erases short-term memory and the labile ASM , preserving only ARM . For the adult-stage-specific RNAi-mediated knockdown of inx5 with tubP-GAL80ts , flies were kept at 18°C until eclosion and then shifted to 30°C for 7 days before training . The 3-hour ARM assay was also performed at 30°C . Control flies were kept at 18°C throughout the experiment . For eNpHR-mediated light-inactivation during memory acquisition , a group of approximately 100 flies were put into a custom-made light-delivering electrical shock tube and received electrical shock alternately paired with either OCT or MCH . For eNpHR-mediated light-inactivation during memory consolidation , the conditioned flies were put into the LED-embedded tube for 1 . 5 hours immediately after training . For eNpHR-mediated light-inactivation during memory retrieval , the conditioned flies were tested for approach to OCT or MCH in LED-embedded tubes . The light intensity was approximately 9 . 35 mW/cm2 , and the wavelength was 590 nm . Flies were fed a standard food medium with or without 100 μM all-trans-retinal ( Sigma-Aldrich ) for at least 5 days before the experiments . Sample preparation for in vivo calcium imaging was modified from a previous study[31] . The fly was fixed in a 250-μl pipette tip , a small window was opened on the head capsule using fine tweezers and fixed in place with dental glue . Next , a drop of adult hemolymph-like ( AHL ) saline ( 108 mM NaCl , 5 mM KCl , 2 mM CaCl2 , 8 . 2 mM MgCl2 , 4 mM NaHCO3 , 1 mM NaH2PO4 , 5 mM trehalose , 10 mM sucrose and 5 mM HEPES [pH 7 . 5 , 265 mOsm] ) was added to the window to prevent dehydration . The fly and pipette tip were fixed to a coverslip by tape and time-lapse recording of changes in GCaMP6m intensity before and after odor delivery was performed on a Zeiss LSM 700 confocal microscope with a 40X water-immersion objective ( W Plan-Apochromat 40× /1 . 0 DIC M27 ) , a 488-nm excitation laser , and a detector for emissions passing through a 555 nm short-pass filter . An optical slice with a resolution of 512 × 512 pixels was continuously monitored for 60 s at 2 frames per second . Odorants were delivered at 11 s and 29 s in each 60 s trial . To correct the motion artifacts , frames were aligned using a lightweight SIFT-implementation[51] . Response amplitudes were calculated as the mean change in fluorescence ( dF/F ) in the 0 . 1–5 s window after stimulus onset . To quantify the numbers of the MB neurons in odor stimulus , the response was judged to be significant if the peak was > 0 . 2 ( dF/F ) . For the lobe specific memory trace assay , regions of interest ( ROI ) were manually assigned to anatomically different regions of the MB lobe . To evaluate responses to different odors in flies , we calculated the change in GCaMP6 fluorescence as ΔF ( Ft–F0 ) /F0 . Changes in GCaMP6 fluorescent intensity for the CS+ vs . CS− odors were calculated as log10 ( ΔFCS+/ΔFCS- ) . For the CBX treatment experiments , flies were dissected and immediately placed in a drop of adult AHL saline containing 1 mM CBX for 10 min before image recording . The CBX solution was washed out using standard AHL solution . ΔF/F0 intensity maps were generated using ImageJ . For the quantification of INX5 protein , fly brains were immunostained with rabbit INX5 antibody ( 1:1 , 000 ) at 25°C for 1 day . After three washes in PBS-T , the samples were incubated in biotinylated goat anti-mouse or rabbit IgG ( 1:200; Invitrogen ) at 25°C for 1 day . The brain samples were then washed and incubated in Alexa Fluor 635 streptavidin ( 1:500; Invitrogen ) at 25°C overnight . After extensive washing , the brain samples were cleared and mounted in FocusClear ( CelExplorer ) . Brain images were obtained using a Zeiss LSM 700 confocal microscope under the same confocal settings for each sample , and the images were further analyzed using ImageJ . Single optical sections were used to calculate the average intensity values per voxel of the INX5 immunopositive signals in the MB calyx , α lobe , α' lobe , β lobe , β' lobe , γ lobe , and protocerebrum bridge ( PB ) . The fluorescent intensity in the PB was used as an adjacent-region control . The efficiency of gene inactivation in each inxRNAi line from TRiP collections was verified with qPCR . Flies for qPCR were generated by crossing elav-GAL4 virgin flies to either wild-type males or the various UAS-inxRNAi males . RNA from the isolated heads of adult flies was extracted with TRIzol Reagent ( Invitrogen ) . The extracted RNA was used to synthesize first-strand cDNA with RevertAid First Strand cDNA Synthesis Kit ( Thermo Fisher Scientific ) . RNA expression levels were quantified with SYBR Green PCR Master Mix on a StepOnePlus System ( Thermo Fisher Scientific ) . For western blotting , the heads of adult flies were homogenized in lysis buffer ( 25 mM HEPES [pH 7 . 5] , 100 mM NaCl , 1 mM MgCl2 , 1 mM CaCl2 , 0 . 1% SDS , 0 . 2% TritonX-100 , 0 . 2% NP-40 , 1 mM EDTA , 1 mM EGTA , and protease inhibitor cocktail [Roche , CH] ) , the lysates were centrifuged at 14 , 000 rpm for 30 min at 4°C , and the supernatants were collected . Lysate proteins were electrophoresed on an SDS-PAGE and electroblotted onto PVDF membranes . The immobilized proteins were probed with rabbit anti-INX5 ( 1:20 , 000 ) and anti-β-actin ( 1:10 , 000 ) antibodies . The membrane was then incubated with horse radish peroxidase ( HRP ) -conjugated goat-anti-rabbit IgG secondary antibody ( 1:10 , 000 ) . The positive signal was detected with SuperSignal West Pico PLUS Chemiluminescent Substrate ( Thermo Fisher Scientific ) . Raw data were analyzed parametrically using the Prism 7 . 0 software ( GraphPad ) . Because of the nature of their mathematical derivation , PIs were distributed normally . Hence , data from more than two groups were evaluated by one-way analysis of variance and Tukey’s multiple-comparisons tests . Data from only two groups were evaluated by the paired t-test . A statistically significant difference was defined as P < 0 . 05 . The data in the bar graphs are presented as means ± SEM . | One of the most important questions in the neuroscience is how the brain process memory . Memory formation requires neuronal communication in the brain via synaptic transmissions , which include chemical and electrical synapses . Unlike the chemical synapses , the biological functions of electrical synapses in memory formation remain poorly understood . Here , we revealed that electrical synapses between mushroom body ( MB ) αβ neurons in Drosophila are critical for consolidated memory retrieval . We also showed that the electrical synapses are important for the branch-specific modification of calcium influx into the αβ neurons during memory retrieval . Our results provide novel insights into the molecular mechanisms and synaptic networks underlying memory retrieval . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"cell",
"physiology",
"invertebrates",
"plant",
"anatomy",
"medicine",
"and",
"health",
"sciences",
"nervous",
"system",
"membrane",
"potential",
"junctional",
"complexes",
"electrophysiology",
"neuroscience",
"learning",
"and",
"memory",
"animals",
"gap",
"junctions",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"plant",
"science",
"cognition",
"experimental",
"organism",
"systems",
"memory",
"drosophila",
"flower",
"anatomy",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"animal",
"cells",
"animal",
"studies",
"memory",
"consolidation",
"insects",
"hyperpolarization",
"arthropoda",
"calyx",
"cellular",
"neuroscience",
"immunostaining",
"eukaryota",
"cell",
"biology",
"anatomy",
"synapses",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cognitive",
"science",
"neurophysiology",
"organisms"
]
| 2019 | Electrical synapses between mushroom body neurons are critical for consolidated memory retrieval in Drosophila |
One of the three most frequently documented copy number variations associated with autism spectrum disorder ( ASD ) is a 1q21 . 1 duplication that encompasses sequences encoding DUF1220 protein domains , the dosage of which we previously implicated in increased human brain size . Further , individuals with ASD frequently display accelerated brain growth and a larger brain size that is also associated with increased symptom severity . Given these findings , we investigated the relationship between DUF1220 copy number and ASD severity , and here show that in individuals with ASD ( n = 170 ) , the copy number ( dosage ) of DUF1220 subtype CON1 is highly variable , ranging from 56 to 88 copies following a Gaussian distribution . More remarkably , in individuals with ASD CON1 copy number is also linearly associated , in a dose-response manner , with increased severity of each of the three primary symptoms of ASD: social deficits ( p = 0 . 021 ) , communicative impairments ( p = 0 . 030 ) , and repetitive behaviors ( p = 0 . 047 ) . These data indicate that DUF1220 protein domain ( CON1 ) dosage has an ASD-wide effect and , as such , is likely to be a key component of a major pathway underlying ASD severity . Finally , these findings , by implicating the dosage of a previously unexamined , copy number polymorphic and brain evolution-related gene coding sequence in ASD severity , provide an important new direction for further research into the genetic factors underlying ASD .
Autism Spectrum Disorder ( ASD ) is a common neurodevelopmental condition characterized by impaired social reciprocity and communicative skills , as well as increased repetitive behaviors and stereotyped interests [1] . ASD has been frequently linked to an accelerated postnatal brain growth [2] that likely involves excessive neuron number and increased neuron density [3] which may affect symptom presentation through gray matter and total volumetric increases [4]–[6] . To date , despite the existence of a strong genetic component for ASD etiology [7] , only rare- and minor-affect genetic loci have been identified [8] , raising the possibility that major genetic contributors to ASD reside in previously unexplored parts of the genome . One such genomic candidate is DUF1220 , a protein domain with an unusually broad spectrum of allelic copy number variation within the human population [9] , [10] . Found within the NBPF gene family and primarily in the 1q21 . 1 region , DUF1220 sequences have undergone a rapid , recent and extreme increase in copy number specifically in the human lineage [11] , [12] . Humans have approximately 290 haploid copies of DUF1220 that can be subdivided into 6 clades defined by sequence similarity ( CON1-3 and HLS1-3 ) [12] . Further , DUF1220 copy number ( dosage ) has been implicated in normal and pathological variation in human brain size and in neuron number across primate lineages [10] . These findings , together with our recent research implicating DUF1220 domains as drivers of neuronal stem cell proliferation ( J . Keeney , submitted ) , make DUF1220 an attractive candidate for modifying ASD symptoms through brain growth mechanisms . Finally , many DUF1220 domain paralogs reside in or adjacent to a widely documented 1q21 . 1 duplication that is one of the three most prevalent copy number variations ( CNVs ) significantly enriched in individuals with autism [13]–[15] , lending further support to the link between DUF1220 copy number and ASD . The association between DUF1220 copy number and the evolutionary expansion of the human brain [10] , [15] , [16] , and the rapidity with which DUF1220 copy number increased in the human genome suggests there were strong selection pressures acting on these sequences [9] . We have suggested that this has also resulted in a deleterious genomic side effect: increased 1q21 instability that predisposes the region to deletions and duplications that in turn contribute to a large number of neurodevelopmental diseases including ASD [15] . This association of DUF1220 copy number increase with evolutionary adaptation may also help explain why ASD , which is genetic but maladaptive , has persisted at such a high frequency across human populations . Given these insights and the link between the copy number of the CON1 subtype ( clade ) of DUF1220 domain and gray matter volume [10] , along with the known associations between gray matter volume irregularities and ASD symptomology [6] , we investigated the association between CON1 copy number and both parent-reported and clinically evaluated ASD-related symptoms . Phenotypic characteristics of children with ASD were determined by clinically robust metrics and CON1 copy numbers were determined using droplet digital PCR ( ddPCR ) , a third-generation PCR technique designed for accurate assay of copy number measurement .
Notably , the CON1 copy number profile in individuals with ASD followed a Gaussian distribution ( Figure 1 ) . In ASD samples CON1 had a mean of 70 copies and extended from 56 to 88 , a range that was similar to that found in otherwise healthy individuals ( ASD mean = 70 , SD = 5 . 5 , healthy mean = 70 , SD = 6 . 9 , unequal variance ttest p = 0 . 98 ) . However , multivariate linear regression detected a linear increase in CON1 dosage that was progressively associated with increasing severity of each of the three primary symptoms associated with ASD as measured by the ADI-R ( Table 1 ) . With each additional copy of CON1 , Social Diagnostic Score increased on average 0 . 25 points ( SE 0 . 11 p = 0 . 021 ) , Communicative Diagnostic Score increased 0 . 18 points ( SE 0 . 08 p = 0 . 030 ) and Repetitive Behavior Diagnostic Score increased 0 . 10 points ( SE = 0 . 05 p = 0 . 047 ) . Further , the association between CON1 copy number and Vineland Adaptive Behavior Scale ( VABS ) -measured Standardized Social Score was nearly significant ( p = 0 . 057 ) , also indicating a progressively worsening condition with increasing dosage of CON1 . CON1 copy number was not associated with cognitive outcomes measured from the Stanford Binet or Raven Matrices . Diagnostic scores were moderately correlated with CON1 copy number , exhibiting a Pearson's r of 0 . 49 and 0 . 67 in social and communicative domains , respectively . Repetitive behavior score demonstrated a more modest correlation with CON1 copy number , with a Pearson's r of 0 . 26 .
These findings represent the first evidence indicating that , in individuals with ASD , increasing DUF1220 CON1 dosage is associated with increasing severity of the primary symptoms of ASD . Further , the apparent dosage effect detected here suggests a causal role for DUF1220 in ASD symptoms , as previous variants in the 1q21 region detected in ASD are exceedingly rare and do not exhibit the broad normal distribution displayed by DUF1220 CON1 copy number . While the precise manner by which DUF1220 dosage affects ASD symptom severity is not yet known , the evidence presented here indicates that DUF1220 protein domains ( specifically clade CON1 ) have an ASD-wide effect and , as such , are likely to be part of a key pathway underlying ASD severity . Given our recent data linking DUF1220 with neural stem cell proliferation ( J . Keeney , submitted ) , this effect could be related to the timing and rate of neurogenesis , such that too many neurons produced too quickly may result in an overabundance of poorly connected neurons . This initial overabundance would in turn inhibit the formation of long distance projection neurons . This process , resulting from ( or exacerbated by ) CON1 dosage increase , could in turn lead to the excess of localized versus long-distance connectivity seen in individuals with ASD [17] . The correlation of the dosage of a highly repeated DNA sequence with symptom severity , while new to ASD , has been seen in other cognitive diseases such as Fragile X and Huntington's disease [18]–[20] . However , in contrast to the small size of the repeating unit in those diseases ( i . e . 3 nucleotides ) , the example presented here is the first to link copy number increase of an entire protein domain ( approximately 1 . 7 kb ) to disease severity . Also , it is particularly striking that the data presented here , together with our previous findings relating DUF1220 copy number to human brain evolution [10] , [15] , [16] , imply that both expansion of the human brain and increase in autism severity appear to involve increasing dosage of sequences within the same gene family . This intriguing observation may help explain the fact that autism , though maladaptive and heritable , nevertheless persists at a high frequency worldwide . Our finding that the DUF1220 CON1 copy number spectrum is not demonstrably different between ASD and otherwise healthy individuals suggests that , while DUF1220 CON1 dosage increase contributes to symptom severity in individuals with ASD , an additional contributing factor is needed for disease manifestation . Such factors could include epigenetic effects or other types of previously unexamined genetic variations such as a copy number imbalance among the six DUF1220 clades , both of which represent testable hypotheses for future research . The study also provides evidence that genetic variants that exert significant effects on complex disease phenotypes , such as described here for ASD , can be found in previously unexamined parts of the human genome . Finally , these findings , by implicating the dosage of a previously unexamined , highly copy number polymorphic and brain evolution-related protein domain in ASD severity , provide a major new direction for further research into the genetic factors underlying ASD .
All participants utilized in this study participated in the Autism Genetic Research Exchange ( AGRE ) and all data was de-identified . The Colorado Multiple institutional Review Board approved this research . Using the AGRE database , we selected 170 well-characterized non-Hispanic white unrelated individuals with idiopathic autism as subjects for this study ( Table 2 ) . AGRE is an academic genetic repository containing genetic material and extensive phenotype information from individuals with autism and unaffected family members [21] . Individuals utilized from the AGRE database were clinically identified utilizing the Autism Diagnostic Interview–Revised ( ADI-R ) and the Autism Diagnostic Observation Schedule ( ADOS ) . All non-idiopathic forms of autism such as fragile X were excluded from this study . Simplex and multiplex status was also collected due to previous reports suggesting different symptoms and different etiologies depending on familial status [22] . Simplex families are defined in AGRE as those with either a single affected child with an unaffected sibling , or one set of affected identical ( monozygotic ) twins with an unaffected sibling . Multiplex families are defined as those with more than one affected child ( except for one set of monozygotic twins , as noted ) . Additionally , raw head circumference was collected as a potential confound due to the link between head circumference and autism-like symptoms [5] and the link between CON1 copy number and head circumference [10] . Sex and age were also collected for adjustment purposes . Finally , a control population of 25 healthy non-Hispanic white male individuals was utilized to explore DUF1220 copy number differences between individuals with ASD and otherwise healthy individuals . All DNA samples , including those from unaffected individuals , were collected and prepared from cell lines by the Rutgers branch of the AGRE repository . Characteristics related to ASD were measured by common diagnostic and assessment tools including the ADOS , ADI-R , Vineland Adaptive Behavior Scales ( VABS ) , Raven Progressive Matrixes ( RM ) , and the Stanford-Binet Intelligence Scales ( SB ) . The ADOS is a clinician administered , structured-play diagnostic exam designed to evaluate the core symptoms of autism . The ADOS has 5 versions that are administered to the child's developmental ability regardless of age . Due to the age independence of this assessment , deriving severity from the ADOS is non-trivial . Therefore , this study used the ADOS only as an enrollment mechanism , dropping children with a negative autism ADOS indication . The ADI-R is a 2–3 hour parent interview administered by a trained clinician focused on a thorough developmental history and specific behaviors associated with the core symptoms of ASD . ADI-R Social Diagnostic Score , Communicative Diagnostic Score , and Repetitive Behavior Diagnostic score were used as outcomes in this analysis . Importantly , sub-domain scores of the ADI-R have been used quantitatively [5] , [23] and higher scores on a diagnostic algorithm indicate greater symptom manifestation . The VABS is a parent questionnaire that addresses the child's personal skills . It is widely used in children with various neurodevelopmental conditions to assess adaptive functioning in social , communication , daily living , and motor skills . The VABS Social Score , Daily Living Score , and Motor Skills Score were used in this study , with lower scores indicating a greater impairment . The RM are multiple-choice tests of abstract reasoning that rely primarily on pattern recognition and are considered good measures of non-verbal abstract abilities . The SB is a commonly used , psychometrically validated measure of intellectual functioning . Verbal ( VIQ ) and Non-Verbal IQ ( NVIQ ) measures were used in this analysis . Droplet digital polymerase chain reaction ( ddPCR ) , a third-generation PCR protocol was utilized following the manufacturer's protocol to assess CON1 copy number in each individual . Primer sequences were as follows: CON1: Left – ‘AATGTGCCATCACTTGTTCAAATAG’ , Right – ‘GACTTTGTCTTCCTCAAATGTGATTTT’ , Hyb – ‘CATGGCCCTTATGACTCCAACCAGCC’; RPP30 ( reference sequence ) : Left – ‘GATTTGGACCTGCGAGCG’ , Right – ‘GCGGCTGTCTCCACAAGT’ , Hyb – ‘TTCTGACCTGAAGGCTCTGCGC’ . Each sample was run in triplicate to confirm results and the copy number estimates were then merged to produce a final copy number for each sample . The ddPCR assay was found to be highly reproducible ( Pearson's r = 0 . 87–0 . 97 , and ICC>0 . 75 ) . Importantly , all samples were assayed in a blinded and randomized order . Blinding and randomization of samples guarded against biases by eliminating differential misclassification and as such the results presented are likely underestimates . Randomization is a critical step in this study because it ensures the error due to imperfect measurement is not disproportionately distributed among individuals . Multivariate linear regression was then utilized to test associations of CON1 with the behavioral phenotypes described . Linear regression was utilized due to the normal distributions of the psychometric outcomes described and due to the normal distribution of CON1 ( Figure 1 ) . Diagnostic analyses did not identify outlying or highly leveraged residuals . In all models covariates were explored because of their known or suspected association with autism-like symptoms and/or potential association with CON1 copy number . These included: sex , age , SB IQ ( in the case of autism symptoms measured from the ADI-R and VABS ) , head circumference , multiplex/simplex status and the interaction of CON1 copy number with multiplex/simplex status . We hypothesized that the interaction of CON1 by multiplex/simplex status could be important due to reports suggesting different symptoms , and potentially different etiologies based on this classification [22] . Interactions of CON1 by sex were similarly explored due to increased prevalence of ASD identified in males [1] . A p-value of less than 0 . 05 was used for definition of significance for main effects . While interactions of CON1 by sex were not significant , the interaction of CON1 by multiplex/simplex approached significance ( p = 0 . 088 ) in the ADI-R Social Diagnostic Score analysis . Given this finding , subsequent ADI-R Social Diagnostic Score analyses were stratified and results are presented from multiplex individuals . Prior to stratification CON1 copy number was associated with ADI-R Social Diagnostic Score ( p = 0 . 020 ) . | Autism Spectrum Disorder ( ASD ) is a common behaviorally defined condition noted by impairments in social reciprocity and communicative abilities and exaggerated repetitive behaviors and stereotyped interests . Individuals with ASD frequently have a larger and more rapidly growing brain than their typically developing peers . Given the widely documented heritability suggesting that ASD is predominantly a genetic condition and the well-established link between ASD and abnormal brain growth patterns , genes involved in brain growth would be excellent candidates to study regarding ASD . One such candidate is DUF1220 , a highly copy number polymorphic protein domain that we have previously linked to brain evolution and brain size . However , due to the extreme copy number variability of DUF1220 , it has not been directly investigated in previous genome wide polymorphism studies searching for genes important in ASD . Here we show that , in individuals with ASD , 1 ) DUF1220 subtype CON1 is highly variable , ranging from 56 to 88 copies , and 2 ) the copy number of CON1 is associated , in a linear dose-response manner , with increased severity of each of the three primary symptoms of ASD: as CON1 copy number increases each of the three primary symptoms of ASD ( impaired social reciprocity , impaired communicative ability and increased repetitive behaviors ) become incrementally worse . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"behavioral",
"neuroscience",
"functional",
"genomics",
"developmental",
"and",
"pediatric",
"neurology",
"biology",
"genomics",
"neuroscience",
"pediatrics"
]
| 2014 | DUF1220 Dosage Is Linearly Associated with Increasing Severity of the Three Primary Symptoms of Autism |
Bistable epigenetic switches are fundamental for cell fate determination in unicellular and multicellular organisms . Regulatory proteins associated with bistable switches are often present in low numbers and subject to molecular noise . It is becoming clear that noise in gene expression can influence cell fate . Although the origins and consequences of noise have been studied , the stochastic and transient nature of RNA errors during transcription has not been considered in the origin or modeling of noise nor has the capacity for such transient errors in information transfer to generate heritable phenotypic change been discussed . We used a classic bistable memory module to monitor and capture transient RNA errors: the lac operon of Escherichia coli comprises an autocatalytic positive feedback loop producing a heritable all-or-none epigenetic switch that is sensitive to molecular noise . Using single-cell analysis , we show that the frequency of epigenetic switching from one expression state to the other is increased when the fidelity of RNA transcription is decreased due to error-prone RNA polymerases or to the absence of auxiliary RNA fidelity factors GreA and GreB ( functional analogues of eukaryotic TFIIS ) . Therefore , transcription infidelity contributes to molecular noise and can effect heritable phenotypic change in genetically identical cells in the same environment . Whereas DNA errors allow genetic space to be explored , RNA errors may allow epigenetic or expression space to be sampled . Thus , RNA infidelity should also be considered in the heritable origin of altered or aberrant cell behaviour .
Altered proteins can result from errors incurred at any step during information transfer from DNA to protein . Errors in DNA , RNA , and protein synthesis occur at rates of , very roughly , 10−9 , 10−5 , and 10−4 errors per residue , respectively [1] . Although rare , errors in DNA synthesis can be fixed as permanent errors—mutations—which can generate heritable change in cellular phenotype . Transcription and translation errors occur more frequently , but are considered transient and their effects fleeting , since the altered molecules are present for a limited time . It has been shown that transcription over a damaged DNA template can generate altered proteins in nondividing DNA repair–deficient cells [2] , and it has been suggested that transient errors can produce transient mutators , thereby generating phenotypic change by introducing mutations [3 , 4] . However , the capacity for transient errors to generate heritable epigenetic phenotypic change has not been considered . The stochastic nature of gene expression results in random fluctuations in protein numbers per cell [5 , 6] . Theoretical and experimental studies have culminated in stochastic chemical kinetic models that describe the statistics of molecular noise [7–9] . Many aspects of gene expression have been considered , including rates of transcription and translation and rates of destruction of the corresponding mRNA and protein products . These models address protein quantity; the quality of the protein produced is not considered with transcription and translation deemed error-free processes . However , due to RNA transcription errors , approximately 1% of all mRNAs encoding polypeptides of 300 amino acids will encode erroneous messages [3] . It has been shown in bacteria , yeast , and mammalian cells that gene expression , and the accompanying noise , occurs in stochastic bursts dominated by the production of mRNAs [10–12] . Since one mRNA is translated many times , RNA errors become amplified , challenging the cell with erroneous proteins that may exhibit partial function , loss-of-function , gain-of-function , or dominant-negative properties . Therefore , any cell at any time may be transiently impaired for a function encoded in a rarely made transcript [3] . As first suggested by Delbrück [13] , epigenetic differences can be understood in terms of multistability: a given cell can persist in one of many stable steady states , which differ from each other by the genes that are ON and those that are OFF . This multistable nature of biological switches is fundamental for the determination of cell fate in unicellular and multicellular organisms [14–21] . Bistability can arise in gene networks that contain a positive-feedback loop [15] . Such gene networks are often regulated by transcription factors that are present in low abundance and therefore subject to noise [22–26] . The lac operon , a set of coordinately expressed genes under the negative control of the lac repressor , is a classic bistable gene network with stable ON and OFF states [14 , 27] . We determined the contribution of RNA errors to molecular noise using a biologically relevant context to monitor noise , namely , heritable stochastic switching in the bistable lac gene network .
To monitor the proportion of cells that are ON or OFF , we have replaced the lacA gene in the wild-type E . coli MG1655 chromosome ( Table S1 ) with a gfp cassette , so that when the lacZYA::gfp transcript is expressed , β-galactosidase , galactoside permease , and green fluorescent protein are produced from the lacZ , lacY , and gfp genes , respectively ( Figure 1A and Figure S1 ) . The galactoside permease promotes the accumulation of the nonmetabolizable inducer thio-methylgalactoside ( TMG ) . This permease induction and inducer accumulation provides the positive-feedback loop that throws the ON switch: permease transports the inducer of permease synthesis , TMGin [14 , 27] . Our system exhibits bistability and hysteresis ( history dependence ) , as was shown by Novick and Weiner [14] and Ozbudak et al . [27] for the lac system ( Figure 1B and Figure S1 ) . Moreover , Novick and Weiner [14] introduced the concept of a “maintenance” concentration of inducer , able to maintain the pre-existing state of cells , either maximally induced or uninduced . Cells already containing permease can accumulate enough inducer ( TMGin ) to maintain induction and so remain ON; uninduced cells , which lack permease , cannot accumulate TMG and so remain OFF ( Figure 1A ) . lac repressor is rare ( ∼5–10 tetramers per cell ) with the lacI gene transcribed about once per cell generation [28] and , due to intrinsic noise , repressor levels differ among isogenic cells and fluctuate with time within each cell [5] . A transient depletion of repressor within a cell will lead to a transient derepression of the operon and to a burst of lacZYA::gfp gene expression [10] . We reasoned that during growth of uninduced cells in a maintenance concentration of TMG , some cells would undergo stochastic events leading to permanent ( DNA replication errors becoming mutations ) or transient ( transcription errors producing erroneous proteins ) derepression of the lac operon . In the case of mutations , constitutive lac expression will ensue; in the case of transcription errors , the lac operon will be transiently derepressed but the subsequent appearance of permease will trigger the autocatalytic positive-feedback response so that the induced state will be heritably maintained , mimicking mutation [14 , 27] ( Figure S6 ) . For our analysis , the MG1655 reporter strain was grown in a maintenance level of 6 μM TMG ( Figure 1B; see also Figure 2B in Ozbudak et al . [27] ) . To determine the genetic ( lacI or lacOc ) mutation frequency , we selected for colony-forming ability on agar plates containing phenyl-β-D-galactoside ( Pgal ) as the sole carbon source ( Text S1 ) . To determine the epigenetic-switch frequency , we measured the appearance of green cells by fluorescence microscopy ( Figure S1 ) . For wild-type MG1655 , the frequency of epigenetically ON cells ( 0 . 3% ) is ∼1 , 000-fold higher than the frequency of genetic lac constitutive mutations ( Table 1 and Figure 1C ) . This result directly compares the frequency of permanent and transient errors in information transfer that lead to the same phenotype in the same system . Bistability in lac operon expression at 6 μM TMG ( the maintenance effect [14] ) , depends upon the low native lac repressor concentration; when we introduced the lacIq up-promoter mutation into the reporter strain , causing a 10-fold increase in lacI transcription [28] and ∼50–100 repressor molecules per cell , bistability and the maintenance effect are abolished , regardless of genetic background ( Table 1 ) . To determine whether the epigenetic-switch frequency is influenced by transcriptional fidelity , we deleted the greA and greB genes that encode auxiliary fidelity factors known from in vitro studies to facilitate the proofreading of misincorporations that arise in nascent mRNAs during transcription [29–31] . The ΔgreA ΔgreB double mutant increased the epigenetic-switch frequency 55-fold over the wild-type level , demonstrating a strong in vivo phenotype , but had no effect on the genetic-mutation frequency ( Figure 1B and 1C , Table 1 , and Figure S1 ) . The ΔgreA and ΔgreB single mutants showed no large effect on the epigenetic-switch frequency , which is consistent with their common function in transcription fidelity ( Table 1 ) . To confirm that transcription infidelity is causing the increase in the epigenetic-switch frequency , we replaced the wild-type rpoB gene ( encoding the β subunit of RNA polymerase , RNAP ) with the ack-1 allele that produces a RNAP known to increase ribonucleotide misincorporation both in vitro and in vivo [32] . The ack-1 allele increased the epigenetic-switch frequency almost 6-fold compared with the wild-type level but had no effect on the genetic-mutation frequency ( Table 1 ) . Another rpoB allele—rpoB8 , which shares many characteristics with ack-1 , including rifampicin resistance [33] , but is not known to affect transcription fidelity—had no significant effect on the epigenetic-switch frequency ( Table 1 ) . Finally , a ΔmutS deletion , which abolishes post-replicative mismatch repair and confers a DNA-mutator phenotype , had the complementary effect to the “RNA-mutators”: the genetic-mutation frequency was increased almost 12-fold over the wild-type level , but the epigenetic-switch frequency was unchanged . Therefore factors involved in the fidelity of RNA transcription affect the frequency of epigenetic heritable stochastic switching , while factors involved in the fidelity of DNA synthesis affect the frequency of heritable genetic stochastic events . While a decrease in transcriptional fidelity leads to an increase in intracellular noise that manifests itself in an increased switch frequency , it would be of equal interest to determine if an increase in transcriptional fidelity would lead to a decrease in intracellular noise and a decreased switch frequency . Until such an “anti-mutator” RNA polymerase is obtained , the role of transcription fidelity in the wild-type noise level remains suggestive; however , the analogous situation for DNA synthesis errors and their role in spontaneous mutation levels also exists . Our results , presented in Table 1 and Figure 1C , show that a decrease in transcriptional fidelity increases heritable stochastic switching in a model bistable gene network . This is not because transcriptional infidelity makes the cell more responsive to 6-μM TMG , in that the ΔgreA ΔgreB strain , which exhibited the greatest increase in epigenetic-switch frequency , behaves the same as the wild-type strain with respect to sensing extracellular TMG: when ON cells from each strain are grown in the presence of decreasing concentrations of TMG , at 6-μM TMG , both wild-type and ΔgreA ΔgreB populations are fully ON ( maintenance ) , and when the TMG concentration decreases beneath this point , the proportion of ON cells in the two strains falls in concert ( Figure 1B ) . If ΔgreA ΔgreB cells were more responsive to TMG , then at lower concentrations than 6-μM TMG , a greater proportion of cells in this strain would remain fully ON compared with wild-type , which is not what we observe . We have also determined that these strains respond in an identical manner to isopropyl-thiogalactoside ( IPTG ) , a similar gratuitous inducer to TMG but one that can more readily penetrate the cell ( Figure S4 ) . These results suggest that compromising the fidelity of transcription transiently reduces the concentration of functional proteins in a stochastic manner in rare individual cells , and that in the great majority of cells for the great majority of the time , there is no difference in the quantity or quality of the proteins being made in wild-type and ΔgreA ΔgreB cells . However , although Western blot analysis of lac repressor levels supports this contention that there is no general reduction in lac repressor levels in the ΔgreA ΔgreB background compared with wild-type ( Figure S3 ) until repressor levels are quantified in individual cells in the wild-type and ΔgreA ΔgreB backgrounds , the possibility of slight differences in protein levels between the strains remains . Finally , β-galactosidase induction curves show that the ΔgreA ΔgreB strain and the ack-1 strain are slightly dampened in their induction rates compared with wild-type , which would not be expected if transcriptional infidelity made the cell more responsive to TMG ( Figure S2 ) . Therefore , transcriptional infidelity does not make a cell more responsive to maintenance-level TMG . It should also be noted that although the ack-1 strain exhibits a slower doubling time than does the wild-type strain , a reduced growth rate , by itself , does not increase the stochastic switching frequency . Indeed , while the rpoB8 control strain has a much longer doubling time than the RNA fidelity strains ( Table S2 ) [33] , it does not exhibit any increase in stochastic switching above the wild-type level ( Table 1 ) . The mechanism ensuring fidelity of transcription by RNAP and the biological consequences of transcription errors are not well understood . The consequences of increasing transcription errors have been obscured by the transient nature of mRNAs and a lack of appropriate experimental systems [34] . Our demonstration that a bistable switch can capture transient errors in transcription provides a ready approach to study and dissect in vivo the nature and the specificity of transcription fidelity . Structures from prokaryotic and eukaryotic RNAPs have revealed a funnel-shaped pore , called the secondary channel , leading from the surface of the enzyme directly to the active site ( Figure S5 ) . This secondary channel is thought to be the major point of entry of ribonucleoside triphosphate substrates to the active site . We have now sequenced the DNA of the ack-1 allele and reveal it to be a P564L substitution in the RNAP β subunit ( Text S1 ) that is positioned at the strategic β turn , where the secondary channel opens directly onto the main channel and the active site . The only other RNAP fidelity mutation characterized in E . coli , an RpoB D675Y substitution [35] , is also positioned at a β turn in the secondary channel but at the opposite end , on the surface or entry region of the secondary channel ( Figure S5 ) . Moreover , it is through this secondary channel that auxiliary fidelity factors ( GreA , GreB , and TFIIS , the eukaryotic analogue of the Gre factors ) access the active site to accomplish their proofreading activity . Misincorporations at the 3′ end of a nascent transcript have been proposed to arise by a misalignment mechanism common to all multisubunit RNAPs [36] . It has been proposed that a “product-assisted catalysis” occurs such that the 3′ misincorporated residue participates in its own excision via an intrinsic cleavage property of the RNAP that provides a proofreading fidelity mechanism [31] . This intrinsic proofreading is stimulated by auxiliary fidelity factors such as GreA and GreB [29–31] and TFIIS [31 , 37] . Therefore , residues composing the secondary channel , and auxiliary factors that reach the active site via the secondary channel , affect both transcriptional fidelity and the frequency of heritable stochastic switching . Additional mutational analysis of the secondary channel should provide insight into the mechanism of transcription fidelity . In yeast , trans-acting mutations have been identified that contribute to noise and include the deletion of components of the SWI/SNF , INO80 , and SAGA chromatin-remodeling complexes [38] . Recently , another study in yeast [39] has shown that mutations that impair uracil metabolism ( ura3Δ ) , transcriptional elongation ( dst1Δ or spt4Δ strains ) or inactivate the Leo1p or Cdc73p subunits of the Paf1 complex can increase the level of noise in gene expression . Dst1 ( TFIIS ) and Spt4 are considered putative elongation factors , but it has recently been shown that neither factor affects the polymerase elongation rate [40] . The dst1Δ mutant does lack TFIIS activity , which is the analogous activity lost in ΔgreA ΔgreB E . coli; however , an altered transcriptional fidelity leading to increased noise was not considered for the dst1Δ mutant [39] . Since TFIIS has been demonstrated to be an auxiliary transcription fidelity factor [31 , 37 , 41 , 42] , we propose that an increase in molecular noise resulting from a decrease in transcriptional fidelity may be a universal phenomenon , because it is observed in dst1Δ yeast and ΔgreA ΔgreB E . coli . Since GreA is itself regulated , and GreA levels change in response to environmental cues [43] , it is possible that the level of noise is under cellular control , and therefore subject to natural selection . Similar to a decrease in the fidelity of DNA polymerases under stressful conditions , which is a possible means to generate genetic diversity [44] , a decrease in the fidelity of RNA transcription may also be viewed as a means to generate phenotypic diversity . Indeed , individuality in bacterial cells is now being considered as an evolutionary solution to the challenges of life in stochastic environments [45] . RNA errors occur frequently but are short-lived . Such transient errors can produce heritable epigenetic consequences if , for example , a positive-feedback loop is initiated . The idea that a transient transcription error can trigger a heritable epigenetic phenotypic change may also be considered in the origin of prion-like particles , where the altered protein acquires a dominant ability to alter its normal counterpart . Population heterogeneity and cell fate are being increasingly viewed as stochastic noise-driven processes [6 , 9 , 24 , 46] . Gene regulatory circuits are thought to control , exploit , or tolerate noise and so maintain order in the behavior and development of cells . Dysregulation in cell behavior is often due to mutation of control genes . Transient errors , by heritably altering epigenetic switch dynamics , may mimic mutation , and occur at much higher frequency than permanent errors . The resulting phenotypic heterogeneity may provide the raw material upon which selection can act with the subsequent evolution of novel cell characteristics during , for example , tumourigenesis: commitment to proliferation in mammalian cells is controlled by a positive-feedback loop , the “restriction point . ” This critical bistable switch [19] may be triggered by transcription infidelity , and proliferation may become independent of growth stimulation as found in cancer cells . Thus “RNA mutators , ” by introducing noise , should also be considered in the origin of altered or aberrant cell behavior .
All strains are derived from the wild-type sequenced E . coli MG1655 strain ( Table S1 ) . The gfp-cat cassette was amplified from plasmid p1G ( gift of Ariel Lindner , Université René Descartes , Paris , France ) and integrated into the native lac operon by homologous recombination producing the lacZYA::gfp reporter . The ack-l allele ( gift of Jonathan Gallant , University of Washington , Seattle , Washington , United States of America ) and the ΔgreA , ΔgreB and ΔmutS deletion mutations ( Keio Collection , Japan ) were moved into MG1655 by P1 transduction . A small number of uninduced OFF cells ( ∼200 ) were grown for 42 h at 37 °C in minimal A salts plus MgSO4 with succinate as the sole carbon source , supplemented with varying amounts of TMG ( to determine the genetic-mutation and epigenetic-switch frequencies , cells were grown at a maintenance level of 6-μM TMG ) ( Text S1 ) . To determine the genetic-mutation frequency , we selected for colony forming ability on agar plates containing Pgal ( 75 μg/ml ) as the sole carbon source . Only cells constitutively expressing β-galactosidase ( lacI– and lacOc mutants ) can form colonies on Pgal plates . To determine epigenetic-switch frequency , 1 . 0 ml of cells from the same subcultures used to determine genetic-mutation frequency was washed and concentrated 20-fold in minimal A salts buffer and 4 μl was used to prepare a microscope slide . Green fluorescence values of single cells were measured using a Zeiss HAL 100 inverted fluorescence microscope . Fields were acquired at 100× magnification with an EM-CCD camera ( Hammamatsu ) . For each field , phase contrast and fluorescence ( EGFP cube = Chroma , #41017 ) images were acquired . Image analysis was performed using AxioVision Rel . 4 . 6 ( Zeiss ) . The program identified individual cells on phase contrast images and extracted pixel values from corresponding regions of fluorescence images to obtain the average fluorescence of each cell above the fluorescence background . The fluorescence levels of uninduced ( OFF ) cells are very similar to the fluorescence background . | Errors in information transfer from DNA to RNA to protein are inevitable and ubiquitous . When errors that occur in DNA are not repaired and become fixed as permanent mutations , they can have heritable phenotypic consequences for cells . In contrast , errors that occur during RNA transcription are considered transient , because the life span of mRNAs and their encoded proteins is thought to be too short to have heritable consequences . Here , we show that transient errors that arise during transcription can cause a heritable phenotypic change within a population of genetically identical Escherichia coli cells growing in the same environment . We used single-cell observation of the bistable lac operon ( which allows a cell to have two alternate states ) in a stable ON or OFF state . We show that the epigenetic-switch frequency from the OFF to ON state is increased when the fidelity of RNA transcription is altered: bacterial strains that contain error-prone RNA polymerases , “RNA mutators , ” and strains deficient in auxiliary RNA fidelity factors exhibit an increased epigenetic-switch frequency compared with wild-type strains . Therefore , like DNA mutation , transient stochastic events can also have long-lived heritable consequences for the cell . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"biology",
"microbiology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
]
| 2009 | Transcriptional Infidelity Promotes Heritable Phenotypic Change in a Bistable Gene Network |
Immunologically intact BALB/c mice infected with Leishmania mexicana develop non-healing progressively growing lesions associated with a biased Th2 response while similarly infected IL-4Rα-deficient mice fail to develop lesions and develop a robust Th1 response . In order to determine the functional target ( s ) for IL-4/IL-13 inducing non-healing disease , the course of L . mexicana infection was monitored in mice lacking IL-4Rα expression in specific cellular compartments . A deficiency of IL-4Rα expression on macrophages/neutrophils ( in LysMcreIL-4Rα−/lox animals ) had minimal effect on the outcome of L . mexicana infection compared with control ( IL-4Rα−/flox ) mice . In contrast , CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice infected with L . mexicana developed small lesions , which subsequently healed in female mice , but persisted in adult male mice . While a strong Th1 response was manifest in both male and female CD4+ T cell specific IL-4Rα−/− mice infected with L . mexicana , induction of IL-4 was manifest in males but not females , independently of CD4+ T cell IL-4 responsiveness . Similar results were obtained using pan-T cell specific ( iLckcreIL-4Rα−/lox ) IL-4Rα−/− mice . Collectively these data demonstrate that upon infection with L . mexicana , initial lesion growth in BALB/c mice is dependent on non-T cell population ( s ) responsive to IL-4/IL-13 while progressive infection is dependent on CD4+ T cells responsive to IL-4 .
New world cutaneous leishmaniasis resulting from infection with Leishmania mexicana is under different genetic and immunoregulatory controls to those controlling L . major infection [1] . Also , unlike L . major , the majority of mouse strains are susceptible to L . mexicana infection [2]–[3] . As with the other Leishmania species , protective immunity against L . mexicana is the result of a STAT-4 dependent type-1 immune response , although this can be generated independently of IL-12 [4] . While the immunological pathways resulting in non-healing L . major infections in susceptible BALB/c mice remain somewhat controversial , IL-4 plays the major role in promoting non-healing L . mexicana infections in this mouse strain [5]–[7] . Thus , mice lacking IL-4 develop small lesions that heal while those lacking IL-4Rα fail to develop lesions [6] . This also indicates some input from IL-13 in the non-healing response to L . mexicana infection as IL-4 and IL-13 receptors share the IL-4Rα sub unit [6] . However , IL-4 and IL-13 are pleiotropic cytokines and numerous cell types of both the innate and adaptive immune responses produce these cytokines as well as express their receptors . In order to better differentiate both the cellular sources and targets of IL-4/IL-13 initiating lesion growth and facilitating progressive non-healing disease , we have previously examined parasite growth in SCID mice reconstituted with IL-4−/− , IL-4Rα−/− , or wild type splenocytes [5]–[6] . These studies indicated that non-lymphocyte sources of IL-4/IL-13 may contribute to early lesion growth during L . mexicana infection . However , the non-healing disease phenotype was dependent on a lymphocyte source of IL-4 and , in its absence , IL-4-deficient splenocyte-reconstituted SCID mice generated a healing response [5] . In addition , SCID mice reconstituted with IL-4Rα−/− splenocytes demonstrated that initial lesion development was also dependent upon cells from this source responding to IL-4/IL-13 [6] . In order to better differentiate the specific role of IL-4/IL-13 responding cells from global effects in vivo , tissue specific IL-4Rα−/− mice have been produced . So far macrophage/neutrophil specific ( LysMcreIL-4Rα−/lox ) [8] and CD4+ T cell specific ( LckcreIL-4Rα−/lox ) [9] IL-4Rα−/− mice have been generated and the consequences for L . major infection studied . In contrast to susceptible BALB/c mice , BALB/c LysMcreIL-4Rα−/lox mice showed a significantly delayed disease progression after infection with L . major , concomitant with normal Th2 and type 2 antibody immune responses but with improved macrophage leishmanicidal activities [8] . These results suggest that alternatively activated macrophages were contributing to the susceptible phenotype in non-healer BALB/c mice . Furthermore T cell-specific LckcreIL-4Rα−/lox BALB/c mice infected with L . major were significantly more resistant than global IL-4Rα−/− mice and developed a disease phenotype and clinical immunity similar to genetically resistant C57BL/6 mice [9]; not only showing the importance of IL-4Rα signaling via CD4+ T cells in the non-healing BALB/c phenotype but paradoxically indicating a protective role for IL-4Rα signaling in a non-CD4+ T cell population . In the present study we demonstrate that in contrast to L . major infection [8] , macrophage/neutrophil signaling via IL-4Rα has minimal effect on the outcome of L . mexicana infection in BALB/c mice . In addition , unlike global IL-4Rα−/− mice infected with L . mexicana that display no lesion growth , infected CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice initially develop lesions indicating that early susceptibility to L . mexicana is dependent on an IL-4 responsive non-CD4+ T cell population . However , subsequent lesion growth is significantly curtailed in infected CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice compared with IL-4Rα intact mice , and a strong Th1 response generated in the presence of significant elements of Th2 activity . Despite reduced susceptibility in all CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice , a dichotomy between the sexes was identified during L . mexicana infection and while lesions in female CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice healed they persisted in male mice associated with elevated IL-4 production in this sex compared with females . Together , our results suggest that initial development of the L . mexicana lesion is dependent on an IL-4/13-responsive non-T cell population , whilst progressive infection is dependent on CD4+ T cells responsive to IL-4 .
LysMcreIL-4Rα−/− , LckcreIL-4Rα−/− , IL-4Rαlox/lox mice were generated and maintained as previously described [9]–[10] . Cell-specific gene disruption in macrophages/neutrophils or T cells was achieved through an intercross between either LysMcreIL-4Rα−/− or LckcreIL-4Rα−/− and IL-4Rαlox/lox mice . Transgene-bearing LysMCreIL-4Rα−/lox and LckcreIL-4Rα−/lox , were identified by PCR genotyping as described [9]–[10] . The mice were maintained under specific pathogen free conditions . Animal experiments were performed in strict accordance with the UK Home Office Animal [Scientific Procedures] Act 1986 ( licence number 60/3929 ) with approval by the University of Strathclyde Ethical Review Panel . L . mexicana ( MYNC/BZ/62/M379 ) was maintained by serial passage of amastigotes inoculated into the shaven rumps of BALB/c mice . Amastigotes for use in infections were isolated from lesions and enumerated as described below . Two sites of infection were examined and either 5×106 L . mexicana amastigotes in a final volume of 50µl were inoculated subcutaneously into the shaven base of the tail , or 2×105 L . mexicana amastigotes in a final volume of 25µl were inoculated subcutaneously into the hind footpad . 6–8 week old male or female LysMCreIL-4Rα−/lox and LckcreIL-4Rα−/lox mice were used in each experiment , with age and sex matched cre negative IL-4Rα−/lox littermates used as controls . The lesion diameter was measured using a sliding gauge micrometer at weekly intervals . Lesions were excised from L . mexicana infected LysMCreIL-4Rα−/lox , LckcreIL-4Rα−/lox and IL-4Rα−/lox mice and disrupted through a metal mesh with 5mL of RPMI 1640 ( Cambrex Bio Science Verviers , Belgium ) . The parasites were washed twice at 350g in RPMI and then enumerated using an improved Naubauer haemocytometer . Alternatively parasite numbers were quantified by limiting dilution , as previously described [11] . Splenocytes were isolated from infected mice and cultured for 72 hours in 96-well plates ( Corning-Costar , NY , USA ) in the presence or absence of L . mexicana antigenic lysate , as previously described [6] . IFN-γ and IL-4 levels were detected in the supernatants by capture ELISA . Briefly the wells of Immulon 1B flat-bottomed microtitre plates ( ThermoLabsystems , MA , USA ) were coated with 50µL of 1µg ml−1 purified anti-mouse IFN-γ capture antibody R4-6A2 ( BD Biosciences , Oxford , UK ) or 500ng ml−1 IL-4 capture antibody 11B11 ( BD Biosciences ) in PBS ( pH 9 . 0 ) overnight at 4°C . Supernatants were then added to the individual wells and either 30µL recombinant mouse IFN-γ ( R&D Systems , Abingdon , UK ) or recombinant mouse IL-4 ( Genzyme , Cambridge , UK ) added to individual wells in duplicate in a doubling dilution with a solution of pH 7 . 4 PBS supplemented with 10% v/v FCS ( Harlan Sera-Lab Ltd . , Crawley , UK ) , ranging form 20ng mL−1 to 0 . 01ng mL−1 ( IFN-γ ) or 2ng mL−1 to 0 . 977pg mL−1 ( IL-4 ) . The plates were then incubated for 2 hours at 37°C . The bound cytokines were incubated with either biotinylated rat anti-mouse IFN-γ monoclonal antibody XMG1 . 2 or biotinylated rat anti-mouse IL-4 antibody BVD6-24G2 ( both BD Biosciences ) and detected with either conjugated streptavidin-alkaline phosphatase or conjugated streptavidin-horseradish peroxidase ( BD Biosciences ) . The appropriate substrate was then added to the wells , p-nitrophenyl-phosphate ( Sigma-Aldrich , Poole , UK ) or tetramethylbenzidine in pH 5 . 5 sodium acetate buffer , containing 0 . 0003% hydrogen peroxide ( BDH , Poole , UK ) . Finally the plates were read at an absorbance of 405nm for IFN-γ or at 450nm for IL-4 . L . mexicana specific-IgG1 and -IgG2a were detected in the plasma of infected mice by ELISA , as previously described [12] . Briefly , Immulon 1B flat-bottomed microtitre plates were coated with 100µL of 10µg ml−1 Leishmania mexicana lysate ( lysate preparation previously described [13] in PBS ( pH 9 . 0 ) overnight at 4°C . Plasma samples were serially diluted in duplicate , followed by a 1 hour incubation at 37°C . Bound Leishmania specific antibodies were detected with a 1 hour incubation with horseradish peroxidase conjugated goat anti-mouse IgG1 or goat anti-mouse IgG2a ( Southern Biotechnology Associates Inc . , AL , USA ) . The substrate tetramethylbenzidine in pH 5 . 5 sodium acetate buffer , containing 0 . 0003% hydrogen peroxide , was then added to the wells and , following colour development , the reaction stopped by the addition of 10% sulphuric acid , absorbance measured at 450nm using a SOFTmax PRO ( Molecular Devices , CA , USA ) and the endpoint dilution was determined . Total IgE was detected in the plasma of infected mice by capture ELISA as previously described [12] , using R35–72 capture IgE mAb ( BD Biosciences ) and biotinylated rat anti-mouse IgE ( Southern Biotechnology Associates Inc . ) . Draining lymph node cells were activated for 4 hours with 50 ng ml−1 PMA and 500 ng ml−1 Ionomycin ( both Sigma-Aldrich ) along with GolgiPlug ( BD Biosciences ) . Following stimulation , cells were harvested and washed , resuspended in FACS Buffer containing Fc Block ( 2 . 4G2 hybridoma supernatant ) together with the appropriate combinations of the following antibodies: CD4-APC , CD8-PerCP or B220-FITC ( all from BD Biosciences ) . Intracellular cytokine staining was carried out using PE-conjugated anti-mouse IL-4 or IFN-γ with Cytofix/Cytoperm solution ( all from BD Biosciences ) . Data was obtained using FACSCanto ( BD Bioscience ) and analysed using FlowJO ( Tree Star Inc . , CA , USA ) . Antibody analysis was performed using the Mann-Whitney U test and all other analysis used an unpaired Student's t test .
To compare the progression of L . mexicana lesion growth in LysMCreIL-4Rα−/lox with IL-4Rα−/lox littermate control and global IL-4Rα−/− mice , animals were infected with 5×106 amastigotes into the shaven base of the tail . While no discernible lesions were identified in infected global IL-4Rα−/− mice , as previously demonstrated [6] , rapidly growing non-healing lesions were observed in both LysMCreIL-4Rα−/lox and IL-4Rα−/lox mice ( Figure 1A ) . Parasite burdens were also similar in LysMCreIL-4Rα−/lox , and IL-4Rα−/lox mice and significantly higher ( p<0 . 0001 ) than those recorded from global IL-4Rα−/− animals ( Figure 1B ) . In line with the non-healing progressive disease phenotype displayed by macrophage/neutrophil IL-4Rα−/− mice , parasite antigen induced spleen cell IFN-γ production was similar to IL-4Rα−/lox mice and significantly less ( p<0 . 01 ) than that of global IL-4Rα−/− mice ( Figure 1C ) . Antigen induced splenocyte IL-4 production was similar in all 3 strains ( Figure 1D ) , demonstrating once again that IL-4 induction can be independent of IL-4Rα signaling [6] , [14]–[16] . Further studies demonstrated that the close similarities in the disease phenotypes of LysMCreIL-4Rα−/lox and IL-4Rα−/lox mice were independent of site , dose of inoculum , and life cycle stage initiating infection ( data not shown ) . These data suggest that the expression of IL-4Rα by a macrophage/neutrophil population is not important in determining susceptibility to L . mexicana infection . In subsequent studies , infection of CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice with 5×106 L . mexicana amastigotes into the shaven base of the tail resulted in control of the lesion growth observed in control IL-4Rα−/lox mice ( wild-type equivalent ) . Interestingly , in one experiment using male mice , lesions in LckcreIL-4Rα−/lox mice did not heal completely , while experiments utilizing female mice fully resolved ( Figure S1 ) . Consequently , to confirm and further investigate this apparent gender-dependent difference in control of L . mexicana infection male and female LckcreIL-4Rα−/lox , control IL-4Rα−/lox , and global IL-4Rα−/− mice were infected in parallel with 5×106 L . mexicana amastigotes into the shaven base of the tail ( Figure 2A–D ) . While infected global IL-4Rα−/− mice displayed a non-lesion growth phenotype , lesion growth was progressive in control IL-4Rα−/lox mice ( Figure 2A and B ) . These disease phenotypes were independent of gender . By comparison L . mexicana infected female CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice developed lesions which completely healed after 4–5 weeks ( Figure 2A ) , while male CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− developed lesions which failed to fully heal ( Figure 2B ) . Indeed in agreement with lesion size parasite burdens up until week 6 were similar in both male LckcreIL-4Rα−/lox and IL-4Rα−/lox mice ( Figure S2 ) . Parasite numbers at the termination of the study at week 12 were of a similar order of magnitude in both female and male control IL-4Rα−/lox mice while global IL-4Rα−/− mice of both sexes were equally able to control infection with L . mexicana ( Figure 2C and D ) . By contrast while infected CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice of both sexes were able to significantly control parasite growth ( p<0 . 0001 and p<0 . 01 respectively for female and male mice ) , male mice ( Figure 2D ) were significantly limited in this ability and had significantly higher parasite burdens ( p<0 . 001 ) than female mice ( Figure 2C ) . L . mexicana infection of CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice resulted in an enhanced Th1 response in both male and female LckcreIL-4Rα−/lox mice compared with control IL-4Rα−/lox mice , as demonstrated by significantly enhanced antigen induced splenocyte IFN-γ production ( Figure 3A and B; p<0 . 025 for females; p<0 . 01 for males ) . As previously described , antigen stimulated splenocyte IFN-γ production from infected global IL-4Rα−/− mice was significantly higher than for wild-type equivalent ( IL-4Rα−/lox ) mice and this was true whether examining female or male mice ( p<0 . 02 ) . Antigen stimulated splenocytes from infected female but not male LckcreIL-4Rα−/lox mice produced significantly more IFN-γ than antigen stimulated splenocytes from infected global IL-4Rα−/− animals . An expanded Th1 response was also indicated by enhanced antigen specific IgG2a production compared with control IL-4Rα−/lox mice , and in magnitude similar to that generated by global IL-4Rα−/− mice ( Figure 3C and D ) . However , while specific IgG2a production in the absence of IL-4Rα signaling via CD4+ T cells in female LckcreIL-4Rα−/lox mice was significantly greater ( p<0 . 025 ) than in infected control IL-4Rα−/lox mice as early as week 6 post-infection ( Figure 3C ) it was not until week 12 post-infection that male LckcreIL-4Rα−/lox mice were producing significantly more IgG2a ( p<0 . 025 ) than their infected control IL-4Rα−/lox counterparts ( Figure 3D ) . Similar results were recorded in 3 separate experiments . A clear dichotomy in antigen-induced splenocyte IL-4 production was identified between infected female and male LckcreIL-4Rα−/lox mice both compared with each other and compared with their wild-type equivalent counterparts ( Figure 4A and B ) . Splenocyte IL-4 production was barely detectable in female LckcreIL-4Rα−/lox mice , and significantly less than IL-4 production by male LckcreIL-4Rα−/lox mice ( p<0 . 01 ) . On the other hand , antigen induced splenocyte IL-4 production was similar in all infected male mice , independent of IL-4Rα expression ( Figure 4B ) . Similar results were obtained with another Th2 cytokine , IL-5 ( Figure 4C and D ) , as well as IL-10 ( Figure 4E and F ) , with production of both cytokines significantly lower in female but not male LckcreIL-4Rα−/lox compared with similarly infected sex matched control IL-4Rα−/lox mice . Similarly , Th2 associated antigen specific IgG1 production was significantly less in infected female ( p<0 . 05 week 12 ) but not infected male LckcreIL-4Rα−/lox mice compared with their respective control IL-4Rα−/lox counterparts ( Figure S2 ) . Minimal IgG1 production was detected in the serum of infected global IL-4Rα−/− mice . Total IgE levels in infected female and male LckcreIL-4Rα−/lox mice were similar to each other and intermediate between those induced in control IL-4Rα−/lox and the absence of IgE in infected IL-4Rα−/− mice ( Figure S3 ) . Anti-CD3 and ConA stimulation ( Figure 5A and B ) of spleen cells from 12 week infected mice demonstrated quite clearly that not only was IL-4 production from female LckcreIL-4Rα−/lox mice significantly less than that of female control IL-4Rα−/lox mice ( p<0 . 003 for ConA and p<0 . 01 for anti-CD3 respectively ) , but also significantly less than similarly treated male LckcreIL-4Rα−/lox mice ( p<0 . 02 for ConA and p<0 . 05 for anti-CD3 respectively ) . Conversely male LckcreIL-4Rα−/lox mice splenocytes produced similar quantities of IL-4 to control male IL-4Rα−/lox mice with either treatment . Examination of draining inguinal lymph node cells indicated a significantly lower ( p<0 . 05 ) percentage of IL-4 and greater ( p<0 . 05 ) percentage of IFN-γ secreting CD4+ T cells in infected female but not male LckcreIL-4Rα−/lox mice compared with control sex-matched IL-4Rα−/lox mice ( Figure 6 A–D ) .
We have previously demonstrated that signaling via IL-4Rα plays the major role in the non-healing response of BALB/c mice following infection with L . mexicana [6] and that IL-4Rα−/− mice , unlike their wild-type counterparts that produce progressively growing non-healing lesions , display a non-lesion growth disease phenotype associated with an enhanced type-1 response . In the course of the present study using macrophage/neutrophil specific IL-4Rα−/− mice ( LysMcreIL-4Rα−/lox ) we failed to identify any significant role for IL-4Rα signaling via macrophages/neutrophils in the normal non-healing response of BALB/c mice infected with L . mexicana . In contrast , following early lesion growth , CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice were able to inhibit disease progression . However , while lesions in female CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice healed those in male CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice persisted . Furthermore , although both female and male CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice had significantly enhanced type-1 responses compared with IL-4Rα intact ( IL-4Rα−/lox ) mice , male CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice maintained strong type-2 responses compared with their female counterparts . Although signaling via IL-4Rα plays a significant role in the outcome of infection with L . mexicana as well as L . major [6] , [8]–[9] , [15] the cell targets for IL-4/IL-13 activity and whether they promote or inhibit the disease process differ significantly between species . Thus , while IL-4Rα signaling via macrophages/neutrophils promotes early lesion growth in L . major infected BALB/c mice and macrophage/neutrophil specific ( LysMcreIL-4Rα−/lox ) IL-4Rα−/− mice display delayed lesion growth [8] , we have failed to identify any contributory role for IL-4Rα signaling via macrophages/neutrophils during L . mexicana infection . The control of L . major early in infection in LysMcreIL-4Rα−/lox mice has been identified as being due to enhanced macrophage microbicidal NO mediated activity in the absence of alternative macrophage activation . What may be critical in this regard is that L . amazonensis parasites , which belong to the “mexicana” complex of parasites , have been shown to be more resistant to macrophage-mediated control than L . major requiring higher levels of NO to induce killing [16]–[17] . Furthermore , recent evidence indicates that , unlike L . major , there is in fact enhanced replication of the amastigote stage of L . amazonensis in IFN-γ-stimulated murine macrophages [18] , reportedly due to the induction of a novel L-arginine pathway independent of iNOS or host arginase [19] . In addition it has been demonstrated that arginase null-mutant L . mexicana have attenuated virulence in vitro and in vivo with the indication that the parasite arginase has a potential role in depleting host L-arginine available for iNOS activity [20]–[21] . Furthermore , the authors suggest that there could be different roles of arginase between L . mexicana and L . major as the Th2 response is blunted in animals infected with arginase null mutant L . mexicana parasites while pharmacological inhibition of arginase during L . major infection did not inhibit the Th2 immune response [22] . CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice are more resistant than global IL-4Rα−/− mice to infection with L . major , indicating that in the absence of a polarized Th2 response , there is a role for an IL-4/IL-13 responsive non-CD4+ T cell in early resistance to infection [9] . Conversely CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice are more susceptible than global IL-4Rα−/− mice to infection with L . mexicana , indicating a role for an IL-4/IL-13 responsive non CD4+ T cell population in early susceptibility . We have now studied the course of L . mexicana infection in newly generated iLckcreIL-4Rα−/lox female and male mice that have IL-4Rα deleted on all T cell populations [23] . These produce the same disease and immunological phenotypes as CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice ( data not shown ) . Consequently IL-4 responsive CD8+ T cells do not play a role in early susceptibility or the non-healing response following infection with L . mexicana . Studies utilizing macrophage specific BALB/c IL-4Rα−/− mice have demonstrated that IL-4/IL-13 operates through this population to enhance L . major parasite growth via alternative macrophage activation [8] and consequently these are unlikely to be the population driving a Th1 response in CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− mice . However , IL-4 treatment of BALB/c mice prior to T cell priming has previously been demonstrated to instruct dendritic cells to produce IL-12 and facilitate a protective Th1 response against L . major [24] and is required for protective type-1 responses to Candida [25] . In addition IL-13 is able to prime monocytes for IL-12 production[26] , which is also observed in listeriosis [27] . Furthermore both IL-4 and IL-13 promote CD40L-induced IL-12 production by macrophages and dendritic cells [28] . This would indicate that dendritic cells may be the IL-4/IL-13 responsive cells facilitating protection against L . major in the absence of IL-4Rα responsive CD4+ T cells in BALB/c mice . As no distinct disease phenotype could be discerned in macrophage/neutrophil specific ( LysMcreIL-4Rα−/lox ) IL-4Rα−/− mice compared with IL-4Rα intact animals infected with L . mexicana , no IL-4/IL13 responsive non-T cell population can easily be ruled out in promoting early infection against this parasite . While a role for B cells and antibody production in the non-healing response to L . mexicana is well established ( as reviewed in [29]–[30] ) , the fact that IL-4Rα−/− BALB/c mice are more resistant to this parasite than IL-4−/− mice suggests that IL-13 responsive cells and consequently non-lymphoid cells via IL-4Rα signaling play a role in disease susceptibility during L . mexicana infection [6] . Unlike the epidemiological and experimental reports on L . major and L . tropica which identify females as more susceptible , females are more resistant than males to cutaneous infection with L . mexicana ( humans and mice ) [5] , [12] , [31]–[33] and visceral leishmaniasis caused by L . donovani ( humans ) [34]–[35] or L . infantum ( dogs ) [36] . Female DBA/2 mice infected with L . mexicana develop much stronger Th1 responses , as measured by IFN-γ production , delayed–type hypersensitivity and IgG2a antibody levels , than similarly infected male mice [12] , [32] . Similarly , in humans infected with L . mexicana , females generally have increased Th1 responses as measured by DTH reactions and decreased Th2 responses as measured by IgE production than males [33] . The present studies using CD4+ T cell specific IL-4Rα−/− BALB/c mice have revealed a previously undetected , underlying male susceptibility to L . mexicana involving T cells . Thus , unlike female mice , male mice were unable to resolve infection and overall had a less polarized Th1 response and more polarized Th2 than their female counterparts . This was associated with IL-4 production independently of IL-4Rα signaling in male but not female CD4+ T cell specific ( LckcreIL-4Rα−/lox ) IL-4Rα−/− BALB/c mice . The L . mexicana induced IL-4 producing Th2 phenotype in male LckcreIL-4Rα−/lox BALB/c mice was not the result of differential Cre-mediated deletion efficiency of IL-4Rα in male mice as compared with female mice , as no CD4+ T cell population expressing IL-4Rα was detected from either gender ( data not shown ) . While IL-4 production independently of IL-4Rα signaling has been observed in a number of immunological models previously [6] , [14]–[15] this is the first time a sex associated influence on this ability has been identified . How significant this could be with regard to the numerous gender related differences observed in inflammatory and infectious diseases is an intriguing question but outside the scope of the present study . To conclude in this study utilising tissue specific IL-4Rα−/− mice we demonstrate that upon infection with L . mexicana initial lesion growth is dependent on a non-CD4+ T cell population responsive to IL-4/IL-13 , while progressive infection is dependent on CD4+ T cells responsive to IL-4 . Furthermore whether lesions heal or not is gender determined , suggesting a subtle but significant effect of sex hormones on CD4+ T cell function whereby infected male but not female CD4+ T cell specific IL-4Rα−/− mice can drive IL-4 production independently of IL-4Rα signaling . | Leishmania species are parasites , transmitted by sandflies which are of extensive public health importance in the tropical and subtropical regions of the world . A large number of distinct Leishmania species cause cutaneous disease and the vast majority of studies utilize the caustive agent of Old World cutaneous leishmaniasis , L . major . Other species , for example , L . mexicana , are associated with quite different patterns of disease following infection of mice when compared with L . major . Thus , while susceptible BALB/c mice deficient in the ability to respond to the cytokines IL-4/IL-13 are not protected against development of cutaneous leishmaniasis caused by L . major they are totally resistant to infection with L . mexicana . Here we describe the outcome of L . mexicana infection in BALB/c mice with cell-specific deletion of the receptor for IL-4/IL-13 on macrophages/neutrophils or T helper cells . Infections develop in both mutants but lesion growth is controlled only in T cell specific knockouts and female but not male mice heal . Male but not female T cell specific knockouts maintain a strong IL-4/IL-13 response . This highlights the role of IL4/IL-13 in driving a non-healing response and may in part explain why human males are more susceptible to this infection than females . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"immunology/immunity",
"to",
"infections"
]
| 2011 | BALB/c Mice Deficient in CD4+ T Cell IL-4Rα Expression Control Leishmania mexicana Load although Female but Not Male Mice Develop a Healer Phenotype |
A well-known mechanism through which new protein-coding genes originate is by modification of pre-existing genes , e . g . by duplication or horizontal transfer . In contrast , many viruses generate protein-coding genes de novo , via the overprinting of a new reading frame onto an existing ( “ancestral” ) frame . This mechanism is thought to play an important role in viral pathogenicity , but has been poorly explored , perhaps because identifying the de novo frames is very challenging . Therefore , a new approach to detect them was needed . We assembled a reference set of overlapping genes for which we could reliably determine the ancestral frames , and found that their codon usage was significantly closer to that of the rest of the viral genome than the codon usage of de novo frames . Based on this observation , we designed a method that allowed the identification of de novo frames based on their codon usage with a very good specificity , but intermediate sensitivity . Using our method , we predicted that the Rex gene of deltaretroviruses has originated de novo by overprinting the Tax gene . Intriguingly , several genes in the same genomic region have also originated de novo and encode proteins that regulate the functions of Tax . Such “gene nurseries” may be common in viral genomes . Finally , our results confirm that the genomic GC content is not the only determinant of codon usage in viruses and suggest that a constraint linked to translation must influence codon usage .
Modification of existing genes , such as by duplication or fusion , is a common and well-understood mechanism by which protein-coding genes originate [1] , [2] . In contrast , we have shown that viruses generate many proteins de novo ( hereafter called “de novo proteins” ) [3] , [4] . Preliminary observations indicate that these proteins play an important role in the pathogenicity of viruses [3] , [5] , for instance by neutralizing the host interferon response [6] or antagonizing the host RNA interference [7] . Strikingly , p19 , the only de novo protein characterised both structurally and functionally , has both a previously unknown structural fold and a previously unknown mechanism of action [7] . Thus , protein innovation seems to be a significant , but poorly understood part of the evolutionary arms race between hosts and their pathogens [5] , 8 , 9 . Studying de novo proteins should thus greatly enhance our understanding of host-pathogen co-evolution and our knowledge of the function and structure of viral proteins [3] , [10]–[14] . However , a major bottleneck that prevents the study of such proteins is their identification , which is very challenging . Finding that a viral protein has no detectable sequence homolog does not reliably indicate that it has originated de novo , because viral proteins evolve so fast that they can easily diverge in sequence beyond recognition . To circumvent this problem , in our previous work [3] , [4] and in the current study , we focused on a special case of de novo proteins: those generated by overprinting . Overprinting is a process in which mutations in a protein-coding reading frame allow the expression of a second reading frame while preserving the expression of the first one ( Figure 1 ) , leading to an overlapping gene arrangement [10] . It is thought that most overlapping genes evolve by this mechanism , and that consequently each gene overlap contains one ancestral frame and one originated de novo [10] . Because overlapping genes are particularly abundant in viruses [15]–[17] , they constitute a rich source of de novo proteins . Identifying which frame is ancestral and which one is de novo ( the “genealogy” of the overlap ) can be done , in principle , by examining their phylogenetic distribution ( the frame with the most restricted distribution is assumed to be the de novo one ) . One can exclude the possibility that the phylogenetically restricted frame is in fact present in other genomes but has diverged beyond recognition , by checking that outside of its clade , the ancestral frame is not overlapped by any reading frame [4] . This approach is simple and reliable [3] , [4] , but is not applicable to cases where the homologs of both frames have an identical phylogenetic distribution . For instance , it could identify the de novo frame in only a minority ( 40% ) of overlaps in our previous study [3] . Therefore , a new method is needed to identify the de novo proteins in most overlapping genes . The approach we investigated is based on the hypothesis that the ancestral frame should have a pattern of codon usage ( i . e . which synonymous codon ( s ) is preferred to encode each amino acid [18] ) closer to that of the rest of the viral genome than the de novo frame [10] . Indeed , analyses of plant RNA viruses and animal DNA viruses [19] , [20] have shown that , within a given viral genome , genes generally have a similar pattern of codon usage , which is thought to depend on the overall GC content of the genome [19]–[21] . In overlapping genes , the ancestral frame , which has co-evolved over a long period with the other viral genes , is expected to have a codon usage similar to that of the rest of the genome ( Figure 1 ) . On the other hand , the de novo frame , at birth , will have a codon usage in effect randomized by the shift and thus unlikely to be close to that of the genome . In addition , constraints imposed by the ancestral frame might prevent the de novo frame from adopting , later , the typical genomic codon usage . Consequently , the de novo frame is expected to have a codon usage less similar to that of the viral genome than the ancestral frame ( Figure 1 ) . This approach has been empirically used to try and identify the de novo frame in a number of cases , as have been related methods which rely on the frequency on nucleotides at some or all codon positions [10] , [22]–[29] . However , the reliability or accuracy of these methods has never been tested . Here we gathered a reference ( “benchmark” ) dataset composed of overlaps with known genealogy , and used it to answer the following questions: do de novo frames have a codon usage distinguishable from ancestral frames ? If yes , can codon usage be used to identify the de novo frame ? What is the reliability of the method and its sensitivity ? Finally , we applied this method to overlapping genes whose genealogy was undeterminable by the phylogenetic method .
As described in Material and Methods , we assembled a dataset of 27 independent , experimentally proven overlapping genes longer than 140 nt ( Table 1 ) . 16 of them have been described previously [3] , as indicated by an asterisk in Table 1 , and 11 additional overlaps were collected for this study . The 27 overlaps come from 25 genera , distributed in 16 viral families covering a wide range of viruses ( Table 1 ) . 18 overlaps involve one gene being completely overlapped by the other , while in 9 the overlap is partial ( e . g . Figure 2 ) . All overlapping genes are in the same orientation , i . e . there are no antiparallel overlapping genes in the dataset . To identify the genealogy of the overlaps , we used the same stringent criterion as in our previous study [3] , selecting only cases in which one frame , predicted ancestral , had a much wider taxonomic distribution than the other frame , predicted de novo . To be confident about the taxonomic distribution of each frame , we carried out extensive searches involving the most up to date similarity search tools , supplemented by in-depth manual searches using contextual information ( see Material and Methods ) . The taxonomic distribution of each frame , and the corresponding evidence , are presented in Supplementary Table S1 . Our predictions of ancestry are supported by functional data: almost all proteins encoded by a frame identified as ancestral have a function central to the viral cycle ( such as capsid or replication ) , while most proteins identified as de novo have a “secondary” function related to pathogenicity ( such as silencing suppressor or apoptotic factor ) ( Table 1 ) . Thus , the predicted genealogy of the overlapping genes of the dataset is highly reliable . We needed to exclude from the dataset ancestral frames that have entered their genome by distant horizontal transfer since these frames are not expected to have the same codon usage as that of their new viral genome , and are thus not suitable for codon usage analysis . Performing a detailed recombination analysis on all ancestral frames of the dataset was out of the scope of this article , and thus we simply detected cases of taxonomic incongruence ( see Material and Methods ) . We detected two cases in which the ancestral frame had originated from another viral genome by distant horizontal transfer . The ancestral protein p104 of Providence virus ( genus alphacarmotetravirus , family Carmotetraviridae ) has statistically significant similarity with the replicase of viruses from a different family , Tombusviridae . Also , the capsid protein of Maize chlorotic virus ( genus machlomovirus , family Tombusviridae ) has significant similarity to that of sobemoviruses , an unassigned genus unrelated to Tombusviridae [30] . We established that horizontal transfer took place towards alphacarmotetravirus and machlomovirus from the other families by analysing the phylogenetic distribution of homologs of the ancestral proteins ( not shown ) . Our results agree with previously reported findings that Providence virus has originated through recombination between a Tombusviridae-like and a Tetraviridae-like virus [31] , and that the machlomovirus capsid protein is taxonomically incongruent [32] . We excluded these two cases from our analyses , and the final benchmark dataset is thus composed of 25 overlaps ( Table 1 ) . As a measure of codon usage similarity between a given frame and the rest of the viral genome , we used the Spearman's rank correlation coefficient ( rs ) between the number of occurrences of each codon in that frame and in the viral genome ( see Materials and Methods ) . Accordingly , the higher the rs of a frame , the more similar its codon usage is to that of the genome . For all gene overlaps of the benchmark dataset , we evaluated rsA , rsN ( the rs of the ancestral and the de novo frame , respectively ) , and the difference ( d21 ) between rsA and rsN ( d21 = rsA−rsN ) . They are listed in the left moiety of Table 2 , ranked by decreasing value of t-Hotelling . rsA is higher than rsN in 21 cases ( i . e . d21>0 ) and lower ( i . e . d21<0 ) in only 4 cases . This distribution is not random ( P<0 . 001 , in accordance to the binomial proportion test ) , suggesting that ancestral frames have a codon usage closer to their genome than de novo frames . This conclusion is supported quantitatively , since the median rsA ( 0 . 42 ) is significantly ( P<0 . 01 ) higher than the median rsN ( 0 . 19 ) according to the Wilcoxon signed rank test [33] . These findings support the hypothesis that codon usage can , in principle , be used to determine the ancestral frame . We now needed a method to infer , given any pair of overlapping frames , whether one frame had a codon usage significantly closer to the rest of the viral genome than does the other frame . In principle , a suitable method to assess the significance of the difference between the rs coefficients of each frame is Hotelling's t-test [34] , [35] . However , Hotelling's t-test is applicable to correlation coefficients estimated from independent data , whereas our data are clearly not independent ( see Material and Methods ) . Therefore , we assessed whether Hotelling's t-test was robust to the violation of the non-independence of data by comparing the results of Hotelling's t-test with simulated codon usage data ( see Material and Methods ) . Values of rsA , rsN and d21 for simulated frames corresponding to each overlap are presented in the right moiety of Table 2 . We performed a McNemar test [33] , which indicated that both methods provide equivalent results ( McNemar chi-square = 0 . 6; P = 0 . 50 ) . Therefore , Hotelling's t-test is reasonably robust to violation of independence and is applicable to our problem . Having established the validity of Hotelling's t-test , we used it to predict the ancestral frame ( and consequently the de novo frame ) in our dataset . Given two overlapping frames 1 and 2 , a frame ( for instance frame 2 ) was predicted ancestral only if it matched the following two criteria: The first criterion corresponds to our main biological hypothesis , whereas the second criterion avoids a scenario in which the first criterion gives results that are mathematically significant but not biologically meaningful . For instance , if one frame had an rs of −0 . 7 and the overlapping frame had an rs of −0 . 1 , the difference would be significant . However , it would be unjustified to return a prediction that the second frame is ancestral , because the negative value of its rs contrasts with our central hypothesis that the ancestral frame has conserved traces of the genome's codon usage . The overlaps are listed in Table 3 by decreasing value of t-Hotelling . We found that both criterions were fulfilled for 13 of our 25 overlaps , and in all these cases the ancestral frame prediction was correct , i . e . concordant with that established by phylogeny ( Table 3 ) . Consequently , the specificity of the codon usage approach is high ( specificity = 1 . 0 , 95% confidence interval [CI] 0 . 77–1 . 00 ) , but its sensitivity is moderate ( sensitivity = 0 . 52 , 95% CI 0 . 31–0 . 72 ) . We examined several factors that could influence the ability to predict the de novo frame by its codon usage . A first factor is genome segmentation: five overlaps of the dataset belong to viruses with segmented genomes ( Aquabirnavirus , Begomovirus , Hordeivirus , Omegatetravirus , Orthobunyavirus ) . The calculations above were done by considering all genomic segments of such viruses as their “genome” . However , considering only the segment encoding the overlap under study yielded the same predictions , suggesting that genome segmentation is not a confounding factor . Second , an extreme GC content could also , in principle , confound codon usage analysis . However , the GC contents of the genomes we analysed here are in a moderate range ( 35–57% ) , and thus are probably not a source of bias . Third , in principle , the relative frame ( +1 or +2 ) of the de novo region with respect to the ancestral region could influence the power of codon usage analysis to distinguish their genealogy . As can be seen in Supplementary Table S2 , 16 de novo coding regions are in the +1 frame relative to the ancestral region they overlap , while the remaining 9 de novo regions are in the +2 frame . Among the 13 overlaps for which there was a significant difference in codon usage between the two overlapping regions , in 9 cases the de novo region was in the +1 frame relative to the ancestral region , while in 4 cases it was in the +2 frame ( Supplementary Table S2 ) . A chi-square test ( chi-square = 0 . 023; P = 0 . 90 ) indicates that the sensitivity of our method does not change depending on the relative frame of the de novo region with respect to the ancestral region , and thus that the relative frame is not a confounding factor . A fourth factor is the age of overlaps: as de novo proteins age , they may progressively impose increased constraints on the ancestral frames , which may change their codon usage , and make it difficult or impossible to distinguish them from de novo frames [4] . Precisely estimating the age of overlaps is not possible given the state of our knowledge of viruses . However , one can use the taxonomical distribution of de novo frames as a heuristic to obtain a very approximate idea of their relative ages . For instance , a de novo frame found in a single species of viral family A has almost certainly appeared more recently than a de novo frame found in a whole family B ( provided there is a good sequencing coverage in both families ) . We have applied this idea to infer the age of overlaps of the benchmark dataset . De novo frames found only in one species were considered “young” ( provided there are several species in the genus considered ) ; overlaps found in more than one species but less than one genus were considered of “Intermediate” age , and overlaps found in more than one genus were considered “old” . We have indicated these estimated relative “ages” in Supplementary Table S2 ( the exact taxonomic distribution of de novo frames is in Supplementary Table S1 ) . There is insufficient taxonomic coverage to estimate the age of overlaps in two genera , for which only a single species is known ( betatetravirus and mandarivirus ) . The remaining 23 overlaps cluster in the following way: 3 young , 13 medium , and 7 old ( supplementary Table S2 ) . By codon usage analysis we have ( correctly ) predicted the genealogy of 3 young , 6 medium and 2 old overlaps ( supplementary Table S2 ) . We have analysed these data by the chi-square contingency table test . The Chi-square value was 1 . 95 ( P = 0 . 30 ) . Therefore , the predictive power of codon usage to identify the de novo frames does not seem to be dependent on their taxonomic distribution , and by extension , on their relative ages . A fifth potential confounding factor is that some de novo frames have a biased amino acid ( aa ) composition . This raises the possibility that the aa composition of de novo frames could be the major explanatory factor of our results , and that differences in codon usage would be a consequence of this biased aa composition . To empirically determine whether aa composition contains more information about frame ancestry than codon usage , we carried out a correlation analysis of the aa composition of overlapping frames with the same statistical test as for codon usage , and compared the predictive power of both methods . We performed the same analysis as on codon usage data but on the frequency of the 18aas that have a degree of codon-degeneracy >1 . The median value of the Spearman correlation between the aa composition of the ancestral frame and that of non-overlapping regions was 0 . 62 , while the median value of the Spearman correlation between the aa . composition of the de novo frame and that of non-overlapping regions was 0 . 50 . Unlike for codon usage ( see above ) , the difference was not significant ( P = 0 . 35 in accordance to the Wilcoxon signed rank test ) . Therefore , aa composition does not have as much predictive power regarding the genealogy of overlaps as codon usage , and our results are unlikely to be explained by the difference in aa composition between ancestral and de novo frames . Finally , to study whether recombination could be a confounding factor , we examined codon usage in the two cases in which the ancestral frame had arisen by recombination ( see above ) , excluded from the above statistics . For machlomovirus , the difference between rsN and rsA was not significant ( Table 3 , bottom , t-Hotelling = 1 . 01 , P<0 . 20 ) . On the other hand , in the case of Providence virus ( Alphacarmotetravirus ) , rsN ( 0 . 51 ) was significantly higher than rsA ( 0 . 00 ) ( t-Hotelling = 2 . 94; P<0 . 005 ) , and positive . Thus , ignoring the recombination event would lead to the erroneous prediction that the replicase is the de novo frame . It would be interesting to determine whether the codon usage of the Providence virus replicase gene is similar to that of its original genome . However , we could not find the species from which the recombination had occurred , since a similarity search based on the nucleotide sequence of the replicase found no similar viral ( or cellular ) sequence . We applied the codon usage method defined above to seven pairs of overlapping genes ( gathered from the literature ) , in which both frames have the same phylogenetic distribution . Table 4 presents the codon usage values for these overlaps by decreasing value of t-Hotelling , and the corresponding predictions of ancestry . The codon usage of overlapping frames was significantly different in only two cases ( deltaretrovirus Tax/Rex and alphanodavirus replicase/B2 ) . Deltaretrovirus Tax and the betanodavirus replicase , respectively , had a codon usage significantly closer to that of the viral genome than the other frames , suggesting that they are the ancestral frames and that the de novo frames are Rex and B2 . We discuss these two overlaps in more depth below ( case studies number 1 and 2 ) . In the five other overlaps analyzed in Table 4 , both frames had a comparable codon usage , preventing prediction of the de novo frame . We examined in more detail the deltaretrovirus genome , which contains a complex pattern of overlapping genes at its 3′ end , in the pX region [36]–[39] . In addition to Tax and Rex , the pX region of Human T-lymphotropic virus 1 ( HTLV1 ) encodes two other proteins in the sense strand , p12 and p30 , and a fifth protein , HBZ , from the antisense strand [36] , [37] , [40] ( Figure 3 ) . The resulting arrangement has two long ( >80 aa ) triple overlaps: the N-terminus of p30 overlaps both p12 and the N-terminus of HBZ , while the C-terminus of p30 overlaps the N-termini of both Tax and Rex ( Figure 3 ) . The phylogenetic distribution of the overlapping genes in deltaretroviruses is summarized in Figure 4 . P30 is expressed only in HTLV1 [36] . p12 has only been reported in HTLV1 [36] , and its coding sequence is interrupted by a stop codon in HTLV2 , while it has no equivalent in bovine leukemia virus . HBZ is present in HTLV1 but also in HTLV2 , 3 and 4 , where it is called respectively APH2 , APH3 and APH4 ( these proteins have statistically significant similarity with HBZ , indicative of homology ) . In the bovine leukemia virus genome , no equivalent of HBZ is expressed from the antisense strand in the region between the Env and Tax genes ( Luc Willems , personal communication ) ; instead microRNAs are expressed from the sense strand [41] , [42] . Taking into account this phylogenetic distribution , and our codon usage predictions , the most likely evolutionary scenario ( Figure 4 ) is that HBZ has originated in the common ancestor of HTLV1 to 4 , after its divergence from bovine leukemia virus; p12 has originated de novo in HTLV1 by overprinting HBZ; and p30 has originated de novo in HTLV1 by overprinting both HBZ ( in the N-terminus of p30 ) and Tax and Rex ( in the C-terminus of p30 ) . It is not possible to conclude whether p30 or p12 originated first , nor how Tax or HBZ originated ( de novo or by horizontal gene transfer ) . We made two additional observations regarding codon usage . First , the fact that Tax and Rex are involved in a triple overlap with a short region of p30 ( Figure 3 ) constitutes a potential confusing factor in our prediction of ancestry by codon usage above . Nevertheless , the region of p30 overlapping Tax and Rex has a codon usage significantly more distant to that of the genome than that of Tax ( t-Hotelling = 2 . 16; P<0 . 025 ) . Therefore , the codon usage of Tax is close to that of the genome over the entire length of its overlapping region , which further suggests that Tax is the ancestral gene . Second , genes expressed from an antisense strand are not expected to have a similar codon usage to genes from the sense strand . Accordingly , the codon usage of HBZ is not correlated to that of the rest of the genome ( rs = 0 . 00 for the entire HBZ gene , rs = 0 . 06 for the region of HBZ overlapping p30 ) . The existence of triple overlaps poses severe constraints on the sequence of the proteins encoded by the pX region , and we thus examined whether they had an unusual sequence composition , or were predicted to be structurally disordered [3] ( see Material and Methods ) . We found that all proteins encoded by the pX region , with the exception of Tax , contained long regions with low sequence complexity ( as defined by SEG [43] ) over most of their length ( dashed lines in Figure 3; see Supplementary Table S3 ) , indicating that they were unlikely to adopt a typical globular structure [43] , [44] . Tax has no specific region of low sequence complexity , but both its N-terminus , overlapping Rex and p30 , and its non-overlapping C-terminus have a highly biased composition , being enriched in hydrophobic residues ( P<0 . 005 ) and depleted in polar and charged residues ( P<0 . 005 ) . In addition , HBZ and Rex were predicted to be mostly disordered , at least in the absence of binding partners , while p30 contained several long regions predicted disordered ( see Supplementary Table S3 ) . Only p12 and Tax were predicted to be mostly ordered . These results suggest that sequence constraints imposed by triple overlaps may lead to proteins with a highly biased sequence composition and/or structurally disordered [3] , and further highlight the fact that Tax is different from the other proteins encoded by the pX region . Finally , it may seem extraordinary that triple overlaps exist at all , given the sequence constraints they impose; in that light , we note that the relative frame arrangement that would impose the highest constraint , called “−2” [45] , is not used for the overlap involving HBZ . ( In this arrangement , codon positions 1 and 2 of a frame overlap respectively codon positions 2 and 1 of the antisense frame , with the result that the sequences of each frame are almost fixed by each other ) . As can be seen in Figure 3 , the frame that is in the −2 arrangement relative to HBZ is the non-coding frame 0 , rather than the p12 or p30 frames . In the second case , the codon usage of alphanodavirus B2 ( a suppressor of RNA silencing [46] ) suggests that it has originated de novo by overprinting the disordered C-terminal extension of the polymerase domain ( Table 4 ) . However , several observations cast a doubt on the reliability of this prediction . A similar genomic arrangement occurs in a closely related genus , betanodavirus ( though there is no detectable sequence similarity between either the C-terminal extensions of the replicases or the B2 proteins of both genera ) ( Figure 5 ) . A parsimonious scenario would demand that the overlaps of both genera have the same origin and thus presumably a similar codon usage pattern . Yet this is not the case: in betanodavirus it is B2 that has a codon usage closer to that of the genome ( though not significantly so ) . This discrepancy might be due to horizontal transfer ( see below ) . Intriguingly , a very similar arrangement occurs in two genera ( ilarviruses and cucumoviruses ) of another family of positive-strand RNA viruses , Bromoviridae , in which a silencing suppressor called 2b overlaps a C-terminal extension of the polymerase ( Figure 5 ) . Like in Nodaviridae , neither the overlapping regions of the replicases nor the 2b proteins of the two genera have any similarity . The codon usage of the 2b proteins of ilarviruses and cucumoviruses is indistinguishable from that of the region of the replicase they overlap ( Table 4 ) , making a prediction of ancestry impossible . In fact , the 2b proteins of ilarviruses might have a different origin from those of cucumoviruses , since these genera do not form a monophyletic clade [47] . Despite their similar genomic location , size and function , alphanodavirus B2 and cucumovirus 2b have different structural folds and different modes of binding to RNA , both previously unknown [46] , [48]–[50] , clearly indicating an independent origin . We think that together , these observations indicate that the overlaps have a complex evolutionary origin; the ancestral protein could differ in each genus ( for instance it might be the C-terminal extension of the replicase in alphanodaviruses and the B2 protein in betanodaviruses ) , and in some genera the ancestral proteins might have entered their genome by horizontal transfer ( see below ) .
We have shown that de novo frames originated by overprinting have a pattern of codon usage distinguishable from ancestral frames , which can be used to predict the de novo frame with a good specificity but intermediate sensitivity ( working in around half the cases ) . How do our results compare with previous empirical studies of codon usage ? The codon usage of six of the overlaps presented here has been studied previously using a different method , the “codon similarity index” [4] . The qualitative trends reported were similar to the ones we observe . Four of the overlaps presented here were also analysed previously , by Pavesi et al [24] who studied their information content and their codon usage . Again , the numerical values they reported for codon usage are in very good agreement with those obtained here , as are their general conclusions . However , our improved statistical analysis allowed us to draw more powerful conclusions . For instance , Pavesi et al reported that both the tymovirus replicase and movement genes had a codon usage correlated with that of their genome , and concluded that it was consequently not possible to determine the ancestral gene [24] . In the present article , the use of Hotelling's t-test to compare two dependent correlation coefficients [51] allowed us to determine that the replicase gene had a codon usage significantly closer to its genome than the movement gene ( Table 3 ) , indicating ( correctly ) that it was the ancestral frame . Another study , on the VP2/VP5 overlap of avibirnavirus ( homologous to the aquabirnavirus overlap studied herein ) , showed that VP5 had an unusual nucleotide usage and predicted that it was the de novo frame [27] . Our quantitative analysis is in agreement with these findings: VP2 has a codon usage significantly closer to the viral genome than VP5 ( Table 3 ) . Finally , a previous analysis of the cucumovirus replicase/2b overlap predicted that 2b was the de novo frame , based on its uridine content at the third codon position [22] . In contrast , our analysis detects no statistically significant difference between the codon usage of the overlapping region of the replicase and that of 2b ( Table 4 ) . Why are ancestral and de novo frames distinguishable by their codon usage in only half of the overlaps ? We examined in the Results several confounding factors , such as the relative frame of the overlapping regions , their sequence composition , and the estimated age of overlaps . They did not appear to have a significant impact on the predictive power of codon usage analysis . One note of caution is that we used a very crude estimate of the relative ages of overlaps ( i . e . their taxonomic distribution ) . We could not use a more precise estimate , unlike a previous study [4] , because our dataset contains both RNA and DNA viruses , which have no protein in common that could be used as a molecular clock , and because the proteins we studied often have more than 50% sequence divergence , preventing the determination of reliable phylogenies . During the revision of this manuscript , following the suggestion of a reviewer , we tested a distance measure based on information theory approaches: the modified Kullback-Leibler ( KL ) distance [52] . Since dinucleotide frequency is an important genome signature [53] , we have re-analysed our dataset by calculating the KL distance ( based on the frequency of the 16 dinucleotides at codon positions 1-2 , 2-3 , and 3-1 ) between the ancestral frame and the non-overlapping coding regions of the genome ( KLA ) , and between the de novo frame and the non-overlapping coding regions of the genome ( KLN ) . The ancestral frame had a KL distance to non-overlapping regions lower than that of the de novo frame in 23 out of 25 overlaps . In contrast , in our approach , the rs of the de novo frame was lower than the rs of the ancestral frame in 21 out of 25 overlaps . We could not evaluate by analytical methods whether the KL distance between the ancestral frame and the non-overlapping regions ( KLA ) was significantly smaller than that of the KL distance between the novel frame and the non-overlapping regions ( KLN ) , because KL distances are gamma-distributed [52] and there is no generic analytical solution for the distribution of the difference in gamma distributed variables . Therefore , instead , we performed , on each pair of overlapping genes from our dataset , a permutation test to estimate whether the observed ( KLN−KLA ) was significantly higher than the null distribution of ( KLN−KLA ) derived from 10 , 000 random permutations of the nucleotide sequence of the ancestral and the novel frame . We found that , on our dataset , this permutation test on the KL distance has the same specificity as the t-Hotelling test ( 1 . 00 ) and a slightly better sensitivity ( 0 . 60 ) than the t-Hotelling test ( 0 . 52 ) , although the performance of the two methods is not significantly different ( McNemar chi-square = 0 . 16; P = 0 . 70 ) . We hope that our publicly available dataset of overlapping genes with known genealogy ( expected to grow ) will encourage others to continue exploring these methods and others . Our new method allowed us to make predictions of ancestry for two overlaps in which both frames have the same phylogenetic distribution ( Table 4 ) . In the alphanodavirus replicase/B2 overlap ( case study 2 ) , several elements suggest that horizontal transfer might have taken place and thus that the codon usage prediction is not reliable . In the deltaretrovirus Tax/Rex overlap ( case study 1 ) , our prediction that Rex has originated de novo by overprinting Tax is consistent with the function of Tax , which occurs upstream of that of Rex in the viral cycle [37] , [39] , [54] . It is also coherent with the fact that Tax has a higher sequence complexity than Rex or p30 , and is under stronger selection pressure than Rex [55] , [56] , which is generally the case of ancestral frames [3] , [4] . Our prediction is in agreement with that of a previous work , reached by analyzing the substitution rates of Tax and Rex , their nucleotide composition and their amino acid composition [55] . Tax and Rex are encoded by the same mRNA but have different start codons [57] and thus Rex presumably originated by the appearance of a new ATG upstream of Tax . Both Tax and Rex are present in all deltaretroviruses and only in those viruses , which suggests that Tax originated first in the common ancestor of deltaretroviruses , and that Rex originated by overprinting it “rapidly” afterwards ( by biological timescales ) , before the divergence of deltaretroviruses . Rex must have then undergone a rapid functionalization , since it is indispensable for the viral cycle , like Tax [37] , [39] , [54] . An alternative scenario is possible but appears much less parsimonious: Rex might have originated in another organism with a different codon usage , and entered the genome of the ancestor of deltaretroviruses by horizontal transfer . It would then have diverged in sequence beyond recognition , and have been overprinted by Tax ( which would have a codon usage similar to that of the genome by coincidence ) . The pX region encodes five genes unique to deltaretroviruses [36]–[39] , at least three of which ( p12 , p30 and Rex ) have originated de novo , while the two others ( Tax and HBZ ) have either also originated de novo too ( although earlier ) , or by horizontal transfer ( Figure 3 ) . The pX region thus constitutes a hotspot of gene origination , or gene “nursery” [58] . Strikingly , the two genes that have overprinted Tax , Rex and p30 , play roles that are respectively complementary and antagonistic to Tax [38] , [39] , [59] , while HBZ plays a role antagonistic to that of Tax [60] , [61] . This suggests that the function of Tax was gradually controlled and refined by the appearance of new genes encoded in the same genomic location . Interestingly , other gene nurseries are found in a similar genomic position in other Retroviridae , such as lentiviruses or spumaviruses [62] . As seen above , the 3′ end of the replicase gene of some positive-strand viruses may also favour the origination of gene encoding silencing suppressors ( Figure 5 ) . Such hotspots of origination of genes coding for proteins involved in the same pathways , and combining horizontal transfer and de novo origin , may be common in viruses . For instance , the movement proteins of Alphaflexiviridae and Betaflexiviridae are encoded in the same genomic position ( downstream of the replicase gene ) but belong either to the Triple Gene Block type [63] , [64] or to the 30K type [65] , indicating that at least one or possibly both types of movement proteins have entered these families by horizontal transfer [66] . The mechanisms that presumably favour the appearance of gene nurseries are unknown , but obviously of great interest . In the case of the deltaretrovirus pX region , we note that the high constraints imposed by the triple overlaps severely restrict the evolution of p12 , p30 and Rex , and that consequently their present-day sequence composition is probably rather similar to the one they had when they first originated . We speculate that the pattern of origination seen in the pX region , in which de novo genes regulate the function of an ancestral protein , may be facilitated by the fact that low sequence complexity ( and/or structural disorder ) is compatible with a range of regulatory functions [67]–[69] . Thus , at birth , despite having a very “simple” sequence not honed by natural selection , these proteins may have had , by chance , a regulatory function and provided the virus with a fitness advantage that lead to their fixation . Retroviridae encode numerous short , accessory genes , often accessed by alternative splicing or complex mechanisms leading to partially overlapping genes , and no doubt many remain to be discovered [62] . Yet at the time this article was submitted , almost none of these genes were annotated in the NCBI reference genomes [70] . This poor annotation is prejudicial to the study of these viruses . It is important that researchers who discover , or have discovered new genes , contact the NCBI viral genomes team to ensure that they are annotated properly . Another , more general implication for genome annotation is that long , triple overlaps may have the potential to yield functional proteins relatively easily . Therefore , triple overlaps might be more abundant than previously thought ( we know only two triple overlaps outside of deltaretroviruses , involving the P , V , and D or W proteins in Paramyxovirinae [71]–[74] ) . We thus recommend re-investigating known overlapping gene pairs to find whether in some cases a third overlapping frame might be expressed . It has been proposed that the GC content of a genome was the main , though not the only , determinant of codon usage [19]–[21] . Our results confirm that it cannot be the unique determinant , otherwise the de novo and ancestral frames ( which have the same GC content ) would necessarily have a similar codon usage . Therefore , a significant evolutionary constraint ( s ) on codon usage must operate in addition to the GC content , and this constraint must be greater on ancestral frames than on de novo frames . Belalov et al . recently reported that the frequency of the dinucleotide CpG was an important factor of viral codon usage , and that CpG was significantly rarer at codon positions 2-3 than at positions 3-1 [75] . CpG is known to be underrepresented in RNA viruses [76] , perhaps to avoid recognition from an antiviral CpG sensor [77] . However , the difference in frequency of CpG at different codon positions suggests that a second type of pressure exists , imposed by the translational apparatus . The authors thus suggested the existence of an evolutionary constraint on the genome deriving from a hypothetical cellular CpG sensor coupled ( by an unknown mechanism ) to the translational machinery . The existence of such a constraint would be coherent with our results , and we speculate that it might cause the difference in codon usage between ancestral and de novo frames . Very little is known about de novo protein origination , although it is by now clear that this mechanism plays an important role in viral pathogenicity . Our method should allow the identification of more de novo proteins , and thus enhance our understanding of host-pathogen co-evolution . It will be of particular interest to apply it to gene “nurseries” such as the ones we have identified here , and to elucidate the pressures that shape them . Finally , we note that recent experimental and computational reports suggest that de novo origination of genes by overprinting may not be confined to viruses but on the contrary , be a much wider phenomenon than previously thought , both in eukaryotic [78]–[82] and bacterial genomes [83] .
We retrieved all sequences from the NCBI viral genome database [84] . We used MAFFT [85] for multiple sequence alignment , HHpred [86] and HHblits [87] for remote homology detection , Phylogeny . fr [88] for phylogenetic analyses , and METAPRDOS [89] for prediction of protein structural disorder , respecting the guidelines of [44] . We used Composition Profiler [90] for analyses of protein global compositional bias with respect to Swiss-Prot ( release 51 ) , and SEG for analyses of protein local compositional bias [43] . SEG analyses were obtained from the web server ANNIE [91] with parameters 45/3 . 75/3 . 4 in order to identify long regions with a composition bias indicative of non-globular proteins [44] . We searched the NCBI genome database [84] for viruses that infected eukaryotes , with a genome shorter than 30 , 000 nucleotides , and which contained overlapping genes longer than 120 nucleotides . The cut-off of 30 , 000 nucleotides was chosen because curation of larger genomes is impractical [3] . We included an overlapping gene into the benchmark dataset only when two criteria were fulfilled: 1 ) the expression of both overlapping reading frames was experimentally verified; 2 ) the genealogy of the overlapping reading frames could be determined with good support by using the very stringent criterion described previously [3] , regarding the taxonomic distribution of both overlapping frames . According to this criterion , one reading frame can be considered ancestral only if it has homologs in at least two viral families whereas the other , overlapping frame had in at most one viral family . Since viral proteins diverge very fast , identifying viral proteins conserved in at least two families requires powerful similarity search techniques , which are described below . The final dataset , presented in Table 1 , contains 27 independent ( non-homologous ) overlapping genes whose genealogy is reliably established . The dataset contains no antiparallel overlapping genes because we could not find any whose existence had been convincingly proven experimentally in the genomes of short or medium size considered ( <30 kb ) . We used the following conventions to define the precise boundaries of the overlapping regions on which we performed calculations of codon usage . There are two types of overlaps: in internal overlaps , one overlapping gene is contained entirely within the other gene whereas terminal overlaps involve only the 3′ end of one gene and the 5′ end of another [92] . In the case of internal overlaps , for the longest frame , the first codon counted as overlapping was the most upstream codon that overlaps the start codon of the internal frame , and the last codon counted as overlapping was the most downstream codon that overlaps the stop codon of the internal frame . In the case of terminal overlaps , for the upstream frame , the first codon counted as overlapping was the most upstream codon that overlaps the other frame , and for the downstream frame the last codon counted as overlapping was the most downstream codon that overlaps the stop codon of the other frame . In order to obtain a highly reliable genealogy of the overlaps , we needed to identify as distant homologs as possible for each protein of the dataset . However , not all homologs of a protein can be detected by conventional sequence similarity searches even if they have retained some sequence identity with the query , for a number of reasons [93] , including the fact that databases of protein domains are underrepresented for viruses ( our observations ) . We thus exploited “contextual” information available for viral proteins , such as taxonomy and genome organisation , to identify distant homologs overlooked by conventional searches [94] . We proceeded in the following way ( the procedure is the same as in our previous article [3] but had not been described in detail ) . We first identified “straightforward” homologs of the query protein in the NCBI nr database ( release 1st April 2012 ) , by using HHpred [95] and HHblits [87] and selecting hits whose E-value was below the standard cut-off of 10−3 . We then examined subsignificant hits ( i . e . hits with an E-value superior to 10−3 ) up to E-values of 1000 , looking for viral proteins that came from a taxonomically related virus , and/or occurred in the same position of the genome . Such subsignificant hits , which have weak similarity with the query protein and occur in a similar genomic context , constitute potential homologs . In order to test whether they were actually homologous with the query , we gathered homologs of these subsignificant hits ( with E≤10−3 ) , and used HHalign [96] to compare homologs of the query protein ( obtained above ) with homologs of the subsignificant hits . We considered that an HHalign E-value inferior to 10−3 indicated homology between the subsignificant hit and the query , but performed additional checks , such as verifying that the secondary structure and function ( when available ) of the hits were compatible with that of the query . Whenever the structure of a protein from the dataset was available , we also performed structural similarity searches to identify structural homologs , using DALI [97] and FATCAT [98] . Because overlapping genes are not systematically recognised [16] , [99] there is a theoretical possibility that some homologs of an overlapping frame might exist in related genomes but not be annotated , and therefore missed by similarity searches . For each overlap , we thus systematically checked that the genomes of other taxonomically related viruses did not contain conserved , unannotated open reading frames , as in [4] . We present in Supplementary Table S1 the taxonomic distribution of the homologs detected by our searches , together with the corresponding PFAM family ( or clan ) identified in the process . Genes that have entered their genome by horizontal transfer can be identified by the fact that their phylogeny is discordant with the rest of the genome . A robust measure of this discordance is taxonomic incongruence , e . g . the existence of close homologs in a distant taxon . To detect taxonomic incongruence , we collected homologs of the protein products of each ancestral reading frame using blastp [100] on the Refseq database [101] with a cutoff E-value of 10−3 . Hits to proteins from a different viral family than that of the query indicated taxonomic incongruence . To infer the direction of horizontal transfer , we analysed the phylogenetic distribution of homologs of the ancestral protein , both from the same family and from the distant taxon detected , and applied a parsimony criterion: the clade that has the wider phylogenetic distribution of the gene was most likely to be the clade of origin . In the genetic code , 18 amino acids ( aas ) are degenerate , e . g . encoded by more than one codon , and they are encoded by 59 “synonymous” codons in total . For each viral genome sequence , we measured the number of occurrences of the 59 synonyms in the non-overlapping coding regions and in each of the two overlapping reading frames ( Figure 2 ) . For clarity we will refer to the ensemble of the numbers of occurrences of the 59 synonymous codons of a given reading frame as its “codon usage” . The codon usage of non-overlapping regions will be called the “codon usage of the genome” . In some overlapping reading frames ( generally short , i . e . less than 400 nucleotides ) , the number of occurrences of the synonymous codons for a given aa was smaller than the degree of degeneracy of this aa ( for instance only 3 synonyms for arginine , a 6-fold degenerate aa ) . In these cases , we restricted the analysis to synonymous codons whose number of occurrences was at least equal to the degree of degeneracy of the encoded aa . We indicated in Table 2 the number of synonymous codons on which the analysis was carried out . We wanted to utilize codon usage as a method to predict the genealogy of overlapping genes , and not simply to characterise the behaviour of overlapping genes . Therefore , we needed a method to assess whether the codon usage of ancestral frames was closer to the rest of the genome that the codon usage of de novo frames , and to assess whether this difference was statistically significant . We have examined various canonical methods to evaluate codon usage bias: the Effective number of codons ( ENC ) , [102] Codon Adaptation Index ( CAI ) [103] , and Dmean index [104] . We found that ENC and Dmean had poor predictive power on the genealogy of overlaps ( not shown ) . Initial tests suggested that CAI may have been more sensitive , but we could not easily test the statistical significance of the difference between the observed CAI distances . Therefore , we developed a new method , that had a good predictive power and that could yield estimates of statistical significance . Our hypothesis was that , in overlapping reading frames , the ancestral frame could be identified by having a codon usage that was more similar to the codon usage of the genome than that of the other frame . Thus we designed a measure of the similarity of codon usage of each frame with that of the genome , and a method to assess whether one frame had a codon usage significantly closer to that of the genome than the other frame . In order to quantify the similarity between the codon usages of two given reading frames , we used as a measure the Spearman's rank correlation coefficient ( rs ) [33] between the number of occurrences of the 59 synonymous codons of these two frames ( i . e . between their “codon usages” , see above ) . Each viral genome was divided into three sets: a ) the overlapping region of the reading frame 1; b ) the overlapping region of the reading frame 2 , and c ) non-overlapping regions of the genome , composed of the sequences of non-overlapping genes , and , in cases where some genes of the genome partially overlapped , of their non-overlapping regions ( Figure 2 ) . For viruses with segmented genomes , all segments were included in the calculations . For simplicity , the codon usage of the third set , i . e . non-overlapping regions , will be referred to as the “codon usage of the genome” . In all viral genomes , we calculated the rs between the codon usage of the genome and that of each of the two overlapping frames under consideration ( rs1 and rs2 ) . The reason we collected the non-overlapping coding regions of a virus genome into an integrated set ( as opposed to studying individual non-overlapping genes and analyzing their variance ) is because the individual non-overlapping genes ( or their non-overlapping regions , in cases of genes that partially overlap ) are often short , which would have restricted correlation analysis to 2 or 3 dozens of synonyms . Determining if a given frame “1” has a codon usage closer to that of the genome than the other frame “2” is equivalent to determining whether rs1 is significantly greater than rs2 , i . e . whether the correlation between the codon usage of the first frame and that of the genome is significantly greater than the correlation between the codon usage of the second frame and that of the genome . This comparison involves two correlations coefficients that refer to a common variable ( the codon usage of the genome ) , a situation categorized as “dependent correlation” [51] or as the study of “correlated correlation coefficients” , which can be addressed by the Hotelling t-test [34] , [35] . The conventional Hotelling formula involves comparing Pearson correlation coefficients rp , but can be used with Spearman's correlation coefficients rs by converting them into their Pearson equivalents: [105] . The Hotelling t-value was calculated as follows:where n is the number of the compared codon frequencies , rp1 and rp2 are respectively the Pearson equivalents of rs1 and rs2 , and rp12 is the Pearson equivalent of rs12 ( codon usage correlation between the overlapping frames ) . We assess the Hotelling t-value according to the one-tailed Student's t-test . The Hotelling's t-test is designed for correlation coefficients estimated from independent data . However , the data we examine in this study ( the number of occurrences of synonymous codons ) are clearly not independent , since the sum of the numbers of synonymous codons encoding a given aa is fixed . Consider , for example , a reading frame containing 28 Glutamine codons ( an aa encoded by two synonyms , CAA and CAG ) . If the number of occurrences of CAA is 11 , that of CAG will inevitably be 17 ( i . e . 28−11 ) , i . e . the number of occurrences of CAA and CAG are not independent . Therefore , we assessed whether Hotelling's t-test was robust to non-independence of data by comparing it with a simulation-based exact test . For each pair of overlapping frames of the dataset , we generated two simulated overlapping frames with an aa composition identical to that of the two original frames , and used the actual non-overlapping regions of the genome as a reference set . One round of simulation was performed as follows: we randomly generated a number ( n ) of occurrences for each of the 59 codons encoding the 18 degenerate aas , keeping the sum of the occurrences of codons encoding each aa equal to that of the original frame ( e . g . if there were 28 Glutamine codons in the original frame , the simulated frame could have any number of CAA and CAG totalling 28 ) . We calculated the correlation coefficients rs1 and rs2 between the number of occurrences of all synonyms in both simulated frames and that of the actual genome . We repeated the same process 10 , 000 times , thus simulating the distribution of d21 expected assuming that the reading frames are randomly generated and that codon usage is not related to ancestry ( i . e . the null distribution ) . We then tested whether the observed d21 ( Table 2 ) was significantly larger than this null distribution . Finally , we used the McNemar's non-parametric test [33] to determine whether the Hotelling's t-test and the simulation provide equivalent results ( which would indicate that the Hotelling's t-test is robust to non-independence of data ) . | How does novelty originate in nature ? It is commonly thought that new genes are generated mainly by modifications of existing genes ( the “tinkering” model ) . In contrast , we have shown recently that in viruses , numerous genes are generated entirely de novo ( “from scratch” ) . The role of these genes remains underexplored , however , because they are difficult to identify . We have therefore developed a new method to detect genes originated de novo in viral genomes , based on the observation that each viral genome has a unique “signature” , which genes originated de novo do not share . We applied this method to analyze the genes of Human T-Lymphotropic Virus 1 ( HTLV1 ) , a relative of the HIV virus and also a major human pathogen that infects about twenty million people worldwide . The life cycle of HTLV1 is finely regulated – it can stay dormant for long periods and can provoke blood cancers ( leukemias ) after a very long incubation . We discovered that several of the genes of HTLV1 have originated de novo . These novel genes play a key role in regulating the life cycle of HTLV1 , and presumably its pathogenicity . Our investigations suggest that such “gene nurseries” may be common in viruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genome",
"evolution",
"virulence",
"factors",
"and",
"mechanisms",
"virology",
"molecular",
"genetics",
"biology",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"microbiology",
"computational",
"biology",
"viral",
"evolution"
]
| 2013 | Viral Proteins Originated De Novo by Overprinting Can Be Identified by Codon Usage: Application to the “Gene Nursery” of Deltaretroviruses |
Mechanical stretch-induced tyrosine phosphorylation in the proline-rich 306-residue substrate domain ( CasSD ) of p130Cas ( or BCAR1 ) has eluded an experimentally validated structural understanding . Cellular p130Cas tyrosine phosphorylation is shown to function in areas without internal actomyosin contractility , sensing force at the leading edge of cell migration . Circular dichroism shows CasSD is intrinsically disordered with dominant polyproline type II conformations . Strongly conserved in placental mammals , the proline-rich sequence exhibits a pseudo-repeat unit with variation hotspots 2–9 residues before substrate tyrosine residues . Atomic-force microscopy pulling experiments show CasSD requires minimal extension force and exhibits infrequent , random regions of weak stability . Proteolysis , light scattering and ultracentrifugation results show that a monomeric intrinsically disordered form persists for CasSD in solution with an expanded hydrodynamic radius . All-atom 3D conformer sampling with the TraDES package yields ensembles in agreement with experiment when coil-biased sampling is used , matching the experimental radius of gyration . Increasing β-sampling propensities increases the number of prolate conformers . Combining the results , we conclude that CasSD has no stable compact structure and is unlikely to efficiently autoinhibit phosphorylation . Taking into consideration the structural propensity of CasSD and the fact that it is known to bind to LIM domains , we propose a model of how CasSD and LIM domain family of transcription factor proteins may function together to regulate phosphorylation of CasSD and effect machanosensing .
p130Cas ( mouse: NP_001185768; rat: NP_037063 ) is a proline-rich scaffold protein that plays an essential role in various cell functions , including motility [1] , survival [2] , apoptosis [3] and transformation [4] . The substrate domain , CasSD , is centrally located and contains 15 repeats of YxxP motifs that can be a substrate of Src family kinases [5] . Tyrosine phosphorylation of the CasSD YxxP motifs creates binding sites for the SH2 and PTB domains of effector signaling proteins , such as Crk and Nck . The presence of other domains in p130Cas , namely the N-terminal SH3 domain , the serine-rich domain and the C-terminal Src-binding domain , also allow p130Cas to interact with various other signaling molecules , including focal adhesion kinase ( FAK ) , 14-3-3 proteins and Src family kinases . The ability of p130Cas to associate with a large array of signaling proteins appears to facilitate the formation of multi-protein complexes that allow protein–protein interactions among the bound molecules to promote effective transduction of cellular signals [6] . Various growth factors , hormones , and integrin-mediated adhesion have been reported to regulate tyrosine phosphorylation of CasSD . For example , activation of receptor protein tyrosine kinases by growth factors [7] , activation of estrogen receptor via estrogen binding [8] , or direct interaction between integrin and FAK [9] result in activation of Src and FAK , leading to phosphorylation of tyrosine residues within CasSD . Of those , the most intriguing function that is assigned to p130Cas is its ability to act as a force sensor . We previously demonstrated that physical stretching of CasSD renders it susceptible to phosphorylation of its tyrosine residues by Src family kinases [10] . The multiply phosphorylated CasSD can then act as a docking site for a variety of signaling molecules as described earlier . Evidence from a variety of methods also exists that shows that the LIM domain proteins zyxin [11] and TRIP6 [12] , [13] bind to unphosphorylated p130Cas , localized to sequence within CasSD , and requiring at least 2 LIM domain repeats for binding . In cells , p130Cas can localize to focal adhesions by interacting with FAK through its N-terminal SH3 domain [14] . Since focal adhesions are where FAK associates with actin cytoskeletons via talin [15] , we postulated that extension of p130Cas depends on the tensile forces generated between actin cytoskeletons and cell–extracellular matrix ( ECM ) contacts ( Figure 1A and B ) [10] . By transforming a mechanical event that occurs at a cell-stretching site into a tyrosine phosphorylation signal , p130Cas can act effectively as a cellular mechanosensor . However , the details of the strength of the type of cell-generated forces that stretch CasSD and facilitate its phosphorylation have remained poorly defined . In addition , structural mechanism underlying the responsiveness of CasSD to mechanical stretching is yet to be determined . Since structural information would be critical in understanding how the conformational change of CasSD can occur in response to a tensile force , we set out to determine the biophysical and structural properties of CasSD using a combination of various in vivo , in vitro and in silico characterization techniques . Proline accounts for 19 . 9% ( 61 out of 306 residues ) and 20 . 9% ( 64 out of 306 residues ) of the composition of mouse and rat CasSD , respectively . We therefore anticipated , and demonstrate herein , that CasSD is an intrinsically disordered domain ( IDD ) . There are many intrinsically disordered proteins ( IDP ) found in nature [16] , but only a few IDDs have undergone intense structural scrutiny . Several of these IDD-containing proteins are known scaffold proteins [17]–[20] . Just as CasSD has been known to interact with several different protein partners , it has been noted that IDDs also associate with promiscuous interacting partners and often form hubs of interactions networks [21]–[25] . While there is clearly no apparent single low-energy folded structure in uncomplexed IDDs [26] , advancements in experimental and computational approaches have allowed better characterization of ensemble states and insight into local polypeptide backbone conformational preferences . A growing consensus suggests that the normal peptide backbone angle distribution of IDDs contains a large number of PPII conformations [27]–[29] , except in those instances where there is some evolutionary conservation of protein-fold sequence as in the SH3-like DRK IDD [30] , or where there are local regions of strong α-helical propensity , such as those found in the NTAIL protein of measles and related viruses [31] . In NMR studies of IDDs , the consensus approach to working with ensemble information has been to generate large numbers of candidate structures using various software systems that sample conformational space [32]–[34] , and then remove those structures that are excluded by a variety of measurable constraints [30] , [35] . The NMR fitting of several IDPs has provided a general knowledge that IDD sequences have a natural propensity to sample from PPII conformations [31] , while chemical or thermal denaturation alters this propensity more toward unpaired β-strand type dihedral angle conformations [29] . The Trajectory Directed Ensemble Sampling package ( TraDES-2 , http://trades . blueprint . org ) [32] has been used for generating conformational space samples in some of these studies . It employs brute-force sampling of protein conformations to search for fully folded structures and for creating ensembles of conformations for disordered proteins . In principle , the method could constrain the conformations according to given experimental data . However , no such constraints were used in this study . Taking this new information about PPII conformational sampling propensity into account , we set out to create large ensembles of plausible all-atom 3D structures of CasSD with varying amounts of PPII and β bias , and compare the polymer properties of this in silico ensemble with similar ensembles made with increasing amounts of unpaired β-strand dihedral conformations . We then compared the computed polymer properties of three separate ensembles to those measured by a variety of biophysical techniques to determine whether a coil ( PPII biased ) ensemble can recapitulate the experimental parameters we have determined .
p130Cas is phosphorylated at cell–matrix contact sites ( focal adhesions ) where cytoskeletal tensile force is transmitted to ECM ( Figure 1A and B ) [36] . Since cell stretching is thought to increase the tensile force exerted on the molecules localized at the sites of ECM–cytoskeleton linkage [10] , we initially speculated that phosphorylation of p130Cas molecules at adhesion sites would depend on the contractility of actin cytoskeletons derived from myosin motor activity [37] . Cells exert centripetal traction forces on substrate to which they adhere , even while stretching forces are not externally applied [36] . We therefore expected that inhibition of myosin II would decrease stretching forces on p130Cas at focal adhesions and thereby affect its phosphorylation . Contrary to this notion , p130Cas exhibited distinct phosphorylation at the leading edge of migrating cells even when cells were treated with a myosin II inhibitor , blebbistatin ( Figure 1C ) . Furthermore , when we treated spread NIH3T3 cells with blebbistatin , we found that p130Cas phosphorylation was not significantly decreased ( Figure 1D ) . These results indicated that p130Cas phosphorylation does not depend upon actomyosin contractility . In contrast , inhibition of actin polymerization by cytochalasin D or latrunculin B greatly attenuated the phosphorylation of p130Cas ( Figure 1D ) . Since p130Cas is phosphorylated at the leading edge of migrating cells where actin is actively polymerized independently of myosin II activity ( Figure 1A and B ) [38] , these results suggested that CasSD may be stretched for phosphorylation by the force generated by actin polymerization ( ∼5 pN ) , which would be significantly weaker than the actomyosin-generated force ( ∼30 pN per integrin bond ) [39] . To analyze the mechanical stability of CasSD , single-molecule force measurements by atomic-force microscopy ( AFM ) in a constant-velocity mode have been carried out on a protein construct CasSD–I27–CasSD–I27 , where two I27 domains are introduced as referenced unfolding signature . I27 domain was used , because the elastic property of this domain has been well characterized [40] , and its good mechanical strength makes it easy to be identified from other proteins [41] . Though a hexahistidine ( His6 ) -tag is introduced at the N-terminal of the construct and nickel-nitrilotriacetic acid ( Ni-NTA ) on substrate surfaces to promote the binding of protein molecules to the substrate surface at its N-terminal ( Figure 2A ) , it is still possible to pull a molecule from any two points along its length in the actual experiment . However , the two I27 domains will always have one CasSD domain in between them as shown by the scheme in Figure 2B . Once an I27 domain unfolds ( Figure 2A parts b and c ) , a signature force peak will be recorded on the force–extension trajectory ( peaks labeled b in the bottom panel of Figure 2A ) . Thus , we can be certain that at least one CasSD domain is stretched in trails that show two force peaks for I27 domains in the force-extension trajectories . Any peaks other than the two I27 peaks in those trajectories would be considered as the signal from stretching any mechanically stable structure associated with CasSD . Out of 73 curves obtained with two identified force peaks for I27 , there were 42 curves ( type-1 ) that showed no other distinct feature as shown in Figure 2B , curve i . This indicates that the CasSD domains stretched in these trials consist of only floppy structures with limited mechanical strength that cannot be detected by AFM ( <15 pN ) . Because a fast pulling speed of 600 nm/s was used in these AFM measurements , the unfolding force of CasSD in vivo is expected to be even smaller . The rest of the trajectories ( type-2 ) did show some features ( Figure 2B , curves ii–iv ) other than I27's . Both Funfold ( Figure 2C , top side panel ) and ΔL ( Figure 2C , right side panel ) were broadly distributed , ranging from 30 to 120 pN and from 5 to 120 nm , respectively . The unfolding peak force Funfold and contour length change ΔL showed no correlations since no dominant region can be found in Figure 2C . The relationship between Funfold and ΔL as well as their distributions indicate that within those CasSDs showing type-2 curves , only random structures with random mechanical strength are found . Therefore , results from single-molecule force measurements suggest that the structure of CasSD is predominantly random and flexible . The variability of the pull distance of CasSD indicates significant variation among structures of CasSD , which may be related to its fundamental function as a reporter of subtle mechanical transformations in its environment . Importantly , most structures of CasSD possess little mechanical stability , implying that CasSD can be stretched readily with the weak force generated by actin polymerization . This unexpected mechanical flexibility of CasSD requires modification to the stretch-sensor model illustrated in Figure 1B that involves stretching of CasSD with much stronger tensile forces derived from actomyosin contractility . To gain better understanding of the structural basis of this intrinsic mechanical flexibility of CasSD , further biophysical analyses and simulations were undertaken . To obtain a large-scale preparation of a purified protein for further structural characterizations , CasSD was expressed as a tobacco etch virus ( TEV ) protease-cleavable C-terminal His6-tagged protein in E . coli BL21 ( DE3 ) . We also employed recombinant rat CasSD for some of our experiments , because a method for faster and higher yielding CasSD production became available . The purified CasSD was analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis ( SDS–PAGE ) to confirm that the sample was at least 95% pure ( Figures S1A and S3A for mouse and rat CasSD , respectively ) . While we expected that mouse CasSD and rat CasSD would behave in a virtually identical manner because of their high overall amino acid sequence homology ( 96 . 4% identity and 97 . 4% similarity ) , we confirmed this by comparing their profiles during the SDS–PAGE ( Figure S1A vs . S3A ) , SEC ( Figure S3B ) and circular dichroism ( Figure S3C ) analyses . The first indication of CasSD being an IDD was observed during SDS–PAGE analysis , where the protein anomalously migrated at 15–20% larger than the molecular weight determined by mass spectrometry ( Figure S1B ) . This is a typical behavior of polar-than-normal IDPs , which bind less SDS and hence migrate more slowly than typical protein molecules [21] . Another hallmark characteristics of IDP is their elevated susceptibility to degradation by proteases [42] . CasSD was readily degraded by limited proteolysis using trypsin at 1∶2000 mass ratio to CasSD at a low reaction temperature ( i . e . , on ice ) ( Figure S2 ) . Proteolytic degradation crudely indicates that , like other IDPs , CasSD does not assume a tightly folded structure . Those initial observations indicated that CasSD is likely an IDD . To gain better understanding of the unique structural property of CasSD , we applied various analytical techniques to the purified recombinant CasSD . When we performed an analytical SEC experiment to examine the hydrodynamic property of CasSD , we found that CasSD clearly behaved as a single , homogeneous species ( Figure 3 ) but with a broader peak width than standards . When compared to the standard reference proteins , CasSD was eluted from the column much earlier than a typical monomeric globular 35-kDa protein . Based on the chromatograms obtained for the reference proteins , the apparent molecular weight of CasSD based on the elution volume can be estimated to be close to a bovine catalase tetramer , which has a molecular weight of 250 kDa and Stokes radius of 51 . 2 Å [43] . The peak breadth may have arisen from conformational heterogeneity in the sample . Dynamic light scattering ( DLS ) was also measured to obtain additional information on the hydrodynamic property of CasSD obtained based on a different physical principle employed in SEC , where the outcome can be biased by ionic interactions between the sample and the matrix . DLS indicated that CasSD exhibits a monomodal , reasonably monodispersed distribution in a neutral potassium phosphate buffer with an apparent molecular weight of around 200 kDa ( Figure 4 ) , a result that is in agreement with the results from the SEC experiment . Those results suggest that CasSD assumes a shape that deviates from a typical globular protein to give an apparent molecular weight that is significantly larger than its calculated monomeric molecular weight . However , neither technique could directly distinguish whether the observed large molecular weight was due to an oligomer or a non-globular structure . In order to address this , we employed the sedimentation velocity analytical ultracentrifugation ( SV-AUC ) technique to characterize CasSD ( Figure 5 ) . SV-AUC on the purified CasSD allowed determination of its experimental molecular weight to be 34 . 2 kDa . With the calculated molecular weight of 34 . 9 kDa , this result confirms that CasSD exists as a monomer in solution . SV-AUC also allows calculation of the Stokes radius of the sample , which represents the hydrodynamic radius ( RH ) of a protein molecule . From the SV-AUC data , the RH of CasSD was calculated to be 48 . 1 Å , which is in agreement with the SEC result . Since the minimal RH of an ideal protein sphere with a molecular weight of 34 . 9 kDa is calculated to be 21 . 6 Å [44] , the friction ratio of CasSD is 2 . 23 . Friction ratio is an indicator of size and shape of a protein . Empirically , it has been shown that a nearly globular protein exhibits a friction ratio of around 1 . 2 to 1 . 3 , whereas an elongated or branched protein has a ratio of 2 . 0 to 3 . 0 [44] . Accordingly , CasSD was thought to assume a non-globular and elongated shape , behaving closely to the previously defined native coil-like protein [45] . Combining these results , we can begin to formulate that CasSD is a coil-like intrinsically disordered monomeric 35 kDa protein that persists in a heterogeneous ensemble of predominantly elongated prolate forms . It has an expanded RH compared to folded proteins of the same length but smaller than the calculated value for the chemically denatured form , which would have an RH value of approximately 61 Å [46] . The circular dichroism ( CD ) spectrum of CasSD was collected to determine what type of secondary structure is present . The result shows that CasSD lacks α-helices or β-sheets as its predominant secondary structure constituents ( Figure 6A ) . Negative ellipticity at around 215 nm and the strong the negative peak at 200 nm suggests the presence of PPII-type dihedral angle conformations in residues including proline and other amino acids [47] , [48] . An increasing concentration of urea , up to 6 M , does not effect a large change in the spectra , confirming lack of α-helices or β-sheets ( Figure 6B ) . The CD spectrum of CasSD appears nearly identical to those of other intrinsically disordered or unstructured proteins including ActA [49] , β-casein [50] , bovine viral diarrhea virus core [51] and a synthetic hydrophilic recombinant gelatin [52] . Compositional bias varies in these four examples from 4 . 9–22 . 4% proline , 5 . 8–33 . 7% glycine and 3 . 0–20 . 6% lysine . The similarity in these CD spectra indicates that unstructured proteins exhibit similar subsets of backbone conformational space that are tolerant to a wide range of amino acid compositional biases . Slightly negative ellipticity in the 222 nm region has been interpreted in the past to possibly indicate the minor presence of α-helix or β-sheet secondary structure . However , a new interpretation arises from recent results from a comprehensive library of 400 blocked dipeptide CD spectra [29] which shows that this spectral feature at 222 nm is a general property of amino acid pairs in two dominant conformations , PPII and β , where the β conformers are not stabilized by strand-paired hydrogen bonds . The negative ellipticity feature at 222 nm in blocked dipeptide CD spectra is also temperature dependent as is the 222 nm feature of an IPD , ActA [49] . The dipeptide library results show that these full-length protein CD spectra are consistent with a population of dominant PPII and unpaired β conformations , with β conformations increasing with temperature . General decrease of the ellipticity at 222 nm over increasing temperature was in fact observed with CasSD ( Figure S4A ) , and similar decrease of the ellipticity with increase in the buffer acidity was also observed with CasSD ( Figure S4B ) . This observation follows precisely the known behavior of IDPs termed “turned out” response to heat and changes in pH [53] . This partial folding of IDPs under elevated temperature and low pH is thought to be induced by increased hydrophobic interaction and dampened electrostatic repulsion among the protein backbones , leading to the shift of the conformational states of CasSD toward β . In addition , there is a urea-induced increase of ellipticity in CasSD at around 222 nm ( Figure 6B ) , which suggests that urea changes the conformational states of CasSD , possibly altering the mixed populations of PPII and β conformations towards β as suggested by recent NMR results [54] . Such shift of conformations toward β would enlarge the ensemble RH , as observed in the case of chemically denatured ActA by SEC [49] . When the amino acid sequence of CasSD is analyzed using various disorder prediction programs listed in the Materials and Methods section , all programs indicate that the predominant portion of the protein is disordered . High propensity for disorder is predicted for residues 115–189 and 265–394 ( Figure 7 , thick-lined segments with orange and red letters ) , with highest probability predicted for residues 124–174 and 294–394 ( red letters ) . On the other hand , the central and C-terminal regions ( residues 190–264 and 386–410 , respectively ) are predicted to be least disordered ( black letters ) within the domain . In line with those disorder predictions , most sequence-based secondary structure prediction algorithms also assign CasSD to be comprised of random coil for its entire length , a typical result obtained for IDPs . NetTurn P1 . 0 , a program for sequence-based prediction for occurrence of β-turn motifs [55] , suggests that turn-prone positions ( Figure 7 , Ω-shaped pink bars ) exist in between most of the YxxP motifs ( Figure 7 , yellow circled Ps with green bars ) . We note that to be stable turns , they would require flanking β-strands forming antiparallel hydrogen bonds , which is not supported by the CD data . Thus , we speculate that those predicted “β-turn motifs” may represent positions that introduce discontinuity into the CasSD structure . A multiple alignment of CasSD sequences from 11 placental mammalian species is shown in Figure S5 . Those sequences have 71% identity across the domain , but the spacing of the YxxP motifs are absolutely conserved in all species . Occurrence of highly variable positions relative to the YxxP motifs is also conserved well and coincides with the turn-prone regions suggested in Figure 7 . This highly conserved motif organization found among different CasSDs hints toward functional importance of the spatial arrangement of the YxxP motifs . The GOR [56] 3-state secondary structure prediction of CasSD is shown in Table S1 , which represents the weights applied internally by the TraDES-2 package to the three basis sets of dictionary , α- , β- and coil subsets of φ , ψ dihedral angles , for the conformational sampling . TraDES samples Ramachandran space using these frequencies as a cumulative distribution function . The input dictionary ψ and φ distributions are obtained from non-secondary-structure regions of 7 , 030 representative non-redundant X-ray and NMR structures . There are significant differences in the three ensembles of backbone conformational space-sampling that are caused by the different sampling weights . The distribution of radius of gyration ( RGyr ) values extracted from the three different simulations show distinct differences ( Figures 8A , B and C ) . The mean RGyr value of coil-biased ensemble , 50 . 0 Å , matches the experimentally determined structure the best . The GOR 3-state biased conformations ( mean RGyr of 53 . 8 Å ) are similar to the coil-biased conformations as the predicted secondary structures were almost completely coil . The β-sampled conformations show a significantly enhanced average radius of 70 . 7 Å . A sampling of structures extracted from the three ensembles shows different proportions of PPII regions in the structures ( Figure 9A , B and C ) . Clearly , the PPII regions are more abundant in the coil-sampled and the 3-state conformational ensembles , especially around the region of the experimentally measured value . However , it is only in the coil-sampled ensemble that we observe an enrichment of the PPII conformation at the expense of β structure . This is in good agreement with the results of the CD experiments . While the simulated structures do not provide accurate predictions at the level of single amino acid residues , it does provide a qualitative picture of the general behavior of protein conformational space . As the coil-biased conformations match the experimentally measured determined value of RH the best and apparently reproduce the secondary structure content better than the other ensembles , it is reasonable to conclude that the average conformation sampled in this ensemble is a good approximation to the reality .
The structural and sequence properties of CasSD underlie an unknown stretch-based force detection mechanism . The experimental results obtained during the current study show that CasSD is devoid of α-helix and β-sheet structures and contains significant local PPII-type structure throughout its entire length . While it was earlier suggested that a stable compact structure of CasSD might hide tyrosine residues from phosphorylation that initiates downstream signaling events , our current results are inconsistent with this earlier hypothesis [10] . Instead , CasSD appears to contain conserved short blocks of sequence whose elongated structure is most likely comprised of local PPII-type left-handed helices on the C-terminal side of each tyrosine substrate ( green bars next to yellow circled Ps in Figure 7 ) . These short PPII blocks seem to be often flanked by sequence regions that are both variable in sequence and prone to form turn-like elements ( Ω-shaped pink bars in Figure 7 ) , possibly introducing structural breaks in the PPII-rich domain . We suspect that this closely interspersed structure–sequence organization prevents CasSD from forming regular secondary structure and packing tightly into a globular state . The computed RGyr and relative secondary structure content for the coil-sampled ensemble best reproduces the corresponding values deduced from the experimentally measured Stokes radii ( RH ) and CD spectra , respectively . The RGyr distribution of the coil-sampled ensemble , which has a fewer proportion of β structure , is also narrower ( Figures 8A ) with full width at half-maximum ( FWHM ) of 26 . 2 Å as compared to 38 . 9 Å ( Figures 8B ) and 28 . 8 Å ( Figures 8C ) for the β-sampled and 3-state ensembles , respectively . This can also be attributed to the increase in β-to-PPII ratio in the 3-state- and β-sampled ensembles . Thus , we conclude that the coil-sampled prolate state having low β and high PPII secondary structural content approximates the observed solution structure of CasSD , whereas the most elongated of the β structures likely approach the mechanically stretched forms of CasSD . Results of our experimental and computational analyses suggest that stretching of CasSD is likely to elongate without resistance by undergoing a transformation from non-proline PPII and isolated α dihedral angle-based random coil structures into an elongated configuration with mixed β dihedral angles that appear wherever there are no local proline ring constraints . The conformational propensities of the CasSD ensemble do not appear sufficiently compact to maintain the overall ensemble in such a fashion that the YxxP phosphorylation motifs would be all simultaneously protected from phosphorylation by Src family kinases when the molecule is not stretched . The possibility arises that , instead , the unstretched CasSD may be blocked by several LIM domains present in LIM domain-containing proteins , such as zyxin and TRIP6 that are in fact known to bind p130Cas [11] . In addition , we note that zyxin itself shows stretch sensing properties [57] , [58] where zyxin , upon stretching , reinforces actin stress fibers [58] and accumulates in the nucleus where it may be involved in gene regulation [57] as are other LIM domain-containing transcription factors [59] , including homeodomain proteins [60] . If there is indeed a complex between p130Cas and zyxin or TRIP6 in the unstretched state of a focal adhesion , it can be speculated that the release of the LIM domains from p130Cas for phosphorylation of its substrate domain might be accomplished by physical stretching of such a complex . This raises the mechanistic question of how a mechanical force can disrupt a pre-existing LIM-domain–CasSD complex . Currently , very little is known about the relationship between stretch-sensing and gene expression despite its known clinical relevance in hypertension [61] . Our biophysical and computational experiments have clearly shown that there are significant PPII regions in CasSD . This ties up with the observation that CasSD is known to bind LIM domains . Analysis of LIM domain structures [62]–[64] reveals that LIM domains bind their substrate peptides in PPII conformation . This suggests that LIM domains likely bind to p130Cas at the PPII-rich CasSD . Any change to the backbone PPII conformation of CasSD , for instance by the application of a mechanical force that elongates the peptide and converts the PPII region to a β-stranded region , can lead to misalignment of hydrogen bonding partners between LIM domains and LIM-binding motifs in CasSD ( Figure 10 , moving from top to bottom panel ) . This would result in weakening of the LIM domain–CasSD interactions , allowing LIM domains to dissociate from CasSD and expose CasSD to Src family kinases for subsequent phosphorylation . Furthermore , LIM domain-containing proteins frequently carry two to three copies of LIM domain in tandem repeats . Thus , segments of LIM-binding substrate peptides that directly interact with LIM domains also occur in a relatively regular interval . As pointed out earlier ( Figure S5 ) , CasSDs across different mammalian species show an absolute conservation of the spacing of the YxxP motifs . This may be a reflection of the sensitivity of the stable LIM domain–CasSD complex formation toward spacing of the LIM-binding motifs present in CasSD that would allow formation of optimal hydrogen bonding and other favorable interactions between the two binding partners . Breaking of a single hydrogen bond requires a weak force that is approximately 5 pN in magnitude . We believe that the application of forces of this magnitude or slightly higher would rupture the hydrogen bonds between CasSD and the LIM domains bound to it . While we have no direct experimental evidence for such a mechanism currently , this model is proposed here to account for the good agreement observed in the experimental and computational analyses of the biological , biophysical and structural characteristics of CasSD .
p130Cas-deficient mouse embryonic fibroblasts expressing p130Cas tagged with GFP ( GFP–p130Cas ) were grown overnight in DMEM containing 10% FBS on a 50 µg/ml collagen-coated μ-Dish ( ibidi , Martinsried , Germany ) to form a monolayer . The cells were then treated with DMSO ( 0 . 1% ) or 10 µM blebbistatin for 1 hour and scratched by a pipette tip 1 . 5 hours before fixation . This scratching of the cells simulates wounding of the monolayer . Cells were fixed with cold methanol for 20 minutes at −20°C , permeabilized with 0 . 1% Triton X-100 in PBS for 5 minutes at room temperature , blocked with 1% BSA in PBS for 1 hour at room temperature , incubated with a polyclonal antibody against phospho-p130Cas-Y165 ( pCas-165 ) ( Cell Signal Technology , Danvers , MA ) as a primary antibody in PBS containing 1% BSA for overnight at 4°C , Lastly , the cells were incubated with an Alexa546-conjugated goat anti-rabbit IgG antibody ( Invitrogen , Carlsbad , CA ) as a secondary antibody for 1 hour at room temperature to fluorescently label pCas-165 . Image acquisitions were performed on an IX81 inverted microscope ( Olympus , Tokyo , Japan ) equipped with an Olympus Total internal reflection fluorescence ( TIRF ) illumination arm , fiber-coupled 488 and 559 nm lasers to excite GFP and Alexa546 , respectively , 60× 1 . 45 numerical aperture oil immersion objective lens , and an electron multiplying charge-coupled device camera with a 512-by-512 pixel chip ( Evolve 512 , Photometrics , Tucson , AZ ) . 1 . 5×106 NIH3T3 cells were allowed to adhere to collagen-coated substrates overnight in DMEM containing 10% FBS . Subsequently , the cells were exposed to DMSO ( 0 . 1% ) , blebbistatin ( 50 µM ) , cytochalasin D ( 0 . 5 µM ) or latrunculin B ( 0 . 5 µM ) for 30 minutes , solubilized with SDS sample buffer , and subjected to SDS–PAGE . The gel was subjected to immunoblotting using anti-pCas-165 and anti-p130Cas ( αCas3 ) antibody to visualize phospho-p130Cas and total p130Cas , respectively . Single-molecule stretching experiments were performed on a commercial AFM ( DI Multimode AFM with Picoforce system , Veeco , Plainview , NY ) in a buffer comprised of 25 mM HEPES and 125 mM sodium chloride at pH 7 . 4 . CasSD–I27 ( titin immunoglobulin domain 27 ) –CasSD–I27 was labeled with an N-terminal hexahistidine ( His6 ) -tag for later binding to Ni-NTA-coated substrates . Before measurements , purified proteins were incubated on a Ni-NTA-coated slide [65] for 15 min . In AFM experiments , a gold-coated cantilever ( HYDRA2R-100NGG , Appnano , Santa Clara , CA ) with a spring constant around 15 pN/nm was repeatedly moved toward the slide surface 1 µm above , held at the surface with a contact force of 800 pN for 2 seconds , and then retracted from the surface at a constant velocity of 600 nm/s . When a single protein molecule [40] was absorbed to the cantilever , a force vs . extension curve was recorded . In the force-extension curves , each unfolding event was fitted by a worm-like-chain ( WLC ) model [40] to get the contour length . The difference in the contour length between consequent force peaks was treated as ΔL for the unfolding event associated with the former peak . Trajectories showing two unfolding force peaks of I27 domains ( ΔL = 28±2 nm , F>100 pN ) were chosen for final data processing , because any other ( or none ) feathers other than the two I27 peaks in such trajectories would come from CasSD . Mouse CasSD was produced as a tobacco etch virus ( TEV ) protease-cleavable C-terminal His6-tagged protein in the E . coli BL21 ( DE3 ) Rosetta2 strain ( Merck Biosciences , Darmstadt , Germany ) . Induction of the gene expression was achieved by 37°C incubation for three hours after addition of 400 µM isopropyl-β-d-thiogalactopyranoside ( IPTG ) to LB culture . Cell suspension in a lysis buffer ( 50 mM potassium phosphate pH 7 . 8 , 300 mM potassium chloride , protease inhibitor cocktail VII ( Merck Biosciences , Darmstadt , Germany ) was sonicated and centrifuged to obtain a cleared cell lysate . This lysate was subjected to cobalt-affinity chromatography using HisPur cobalt resin ( Thermo Scientific Pierce Protein Research Products , Rockford , IL ) . CasSD was eluted with 50 mM imidazole . The eluate was exchanged into a buffer composed of 10 mM potassium phosphate pH 7 . 5 , 100 mM potassium chloride , 1 mM EDTA and 5% ( v/v ) glycerol using PD-10 desalting column ( GE Healthcare , Waukesha , WI ) and concentrated to approximately 1 . 5 mg/mL prior to being subjected to preparative SEC using a Superdex 10/300GL column on an ÅKTA purifier liquid chromatography system ( GE Healthcare , Waukesha , WI ) . Purity of the protein was judged by SDS–PAGE . Rat CasSD was also produced as a N-terminal His12-tagged , C-terminal Avi-tagged protein using the E . coli BL21-CodonPlus ( DE3 ) -RP strain ( Agilent Technologies , Santa Clara , CA ) . Protein production was induced with 1 mM IPTG at 37°C for three hours in the M9 media supplemented with 3 µM thiamine . Cells were harvested and lysed in a denaturing lysis buffer containing 8 M urea . Cleared lysate was supplemented with sodium chloride to a final concentration of 50 mM before being subjected to nickel-affinity chromatography using Ni-NTA resin ( QIAGEN , Hilden , Germany ) . Eluted rat CasSD was concentrated to approximately 1 mg/mL prior to being subjected to reversed-phase high-performance liquid chromatography using a semi-preparative Luna 10 micron C18 ( 2 ) column ( Phenomenex , Torrance , CA ) on a Shimadzu LC-6AD semi-preparative system ( Shimadzu Corporation , Kyoto , Japan ) . Samples were separated on a 0–80% acetonitrile linear gradient in water supplemented with 0 . 1% ( v/v ) trifluoroacetic acid . CasSD was eluted with 38–40% acetonitrile . The fractions containing CasSD were pooled and lyophilized . The lyophilized CasSD was kept at −80°C and used in subsequent experiments after reconstituting it in a suitable buffer . Purified mouse CasSD was subjected to analytical SEC using the same condition for preparative SEC described earlier . Purified CasSD was injected at 1 . 5 mg/mL concentration and eluted from the column at a flow rate of 0 . 5 mL/min in a buffer comprised of 10 mM potassium phosphate pH 7 . 5 , 100 mM potassium chloride , 1 mM EDTA and 5% ( v/v ) glycerol . As a reference , proteins used as standard molecular weight references , namely horse spleen ferritin ( type 1 ) , bovine liver catalase , bovine serum albumin and bovine pancreatic ribonuclease A , were also subjected to gel-filtration chromatography using the same condition . DLS measurements were taken on DynaPro Titan ( Wyatt Technology Corporation , Santa Barbara , CA ) using the purified mouse CasSD . Measurements were collected at 1 to 3 mg/mL of purified CasSD in 10 mM potassium phosphate at pH 7 . 5 , 100 mM potassium chloride , 1 mM EDTA and 5% glycerol at room temperature . Data analysis was performed using DYNAMICS V6 software to calculate the diffusion coefficient , hydrodynamic radius , molecular weight and polydispersity of CasSD . Rat CasSD was subjected to SV-AUC experiment in 10 mM potassium phosphate at pH 7 . 5 , 100 mM potassium chloride , 1 mM EDTA and 5% glycerol at a concentration of 1 mg/mL using XL-I analytical ultracentrifuge ( Beckman-Coulter , Brea , CA ) . Samples were centrifuged at 40 , 000 rpm at 4°C over 7 . 3 hours with continuous scan from 5 . 8 to 7 . 2 cm at 0 . 003 cm interval . Data was fit using the program SednTerp ( Alliance Protein Laboratories , Thousand Oaks , CA ) and SedFit [66] with a continuous distribution model to obtain the experimental molecular weight , Stokes radius and frictional ratio of CasSD . Limited proteolysis was performed on the purified mouse CasSD using trypsin as follows . A 2 , 000-fold excess of CasSD to protease was mixed in the reaction buffer ( 10 mM potassium phosphate and 20 mM calcium chloride at pH 7 . 8 ) . Reaction was allowed to proceed on ice for 5 , 10 , 30 , 60 and 120 minutes . At each time point , an aliquot is taken out and mixed with suitable protease inhibitor to quench the reaction . Aliquots were analyzed by SDS–PAGE . CD measurements are taken on JASCO J-715 spectropolarimeter ( JASCO Corporation . Tokyo , Japan ) using the purified mouse CasSD at 0 . 2 mg/mL concentration in 10 mM potassium phosphate at pH 7 . 8 . Measurements were also taken on the rat CasSD in the presence of increasing concentrations of denaturing agents ( 0–6 M urea ) in 10 mM potassium phosphate pH 7 . 8 at the same protein concentration to study the change in the secondary structure content of the protein upon denaturation . For the pH measurements , 10 mM potassium phosphate buffer was used for pH 6 . 6 and 7 . 5 , while 100 mM citrate/phosphate buffer was used for pH 2 . 6 , 3 . 6 , 4 . 6 and 5 . 6 . The mouse CasSD amino acid sequence isoform 1 ( Accession NP_001185768 . 1 ) was subjected to disorderliness prediction by web-based algorithms DRIP-PRED [67] , DISOPRED [68] , IUPred [69] , DisProt VL3H and VSL2P [70] , Scratch [71] , FoldUnfold [72] , RONN [73] , CSpritz [74] , and FoldIndex [75] . Because each algorithm is based on a different theoretical framework , results were compared and combined to obtain a crude consensus of predicted disorderliness of this protein . Predicted degrees of disorderliness from different programs were normalized to a scale of 0 to 9 with 9 being the most disordered . Then , residues predicted as disordered by less than 50% of the programs were labeled as “less disordered ( black ) , ” 50–60% as “intermediately disordered ( purple ) , ” 70–80% as “disordered ( orange ) , ” and 90–100% as “extensively disordered ( red ) . ” The classification is arbitrary and hence only meant to illustrate a crude trend of the predicted disorderliness of the protein . Prediction for the occurrence of a β-turn was performed using the web-based program NetTurn P1 . 0 [55] . BLAST of the CasSD domain without SEG masking , using Blosum80 on the RefSeq database returned a number of vertebrate sequences , however only the subset of placental mammals showed conservation over the CasSD domain sequence . ClustalX [76] was used to illustrate the multiple sequence alignment of the 306-residue domain . Mouse CasSD composition included excessive proline residues ( 19 . 9% ) , yet the aligned sequences exhibited 71% sequence identity across the domain . A conserved deletion is observed in the common ancestor of horse and cow , corresponding to exactly one pseudo-repeat unit . As IDPs are not known to conform to any given 3D shape , an ensemble of possible representations of the 3D-shapes of the mouse CasSD sequence was generated . The new TraDES-2 seq2trj program available at http://trades . blueprint . org was used for this purpose [33] . The working of the TraDES software has been described in detail elsewhere [32] , [33] . Briefly , given the secondary structure preferences of amino acids of a sequence , an ensemble of non-clashing 3D structures of the sequence is generated by assigning backbone Ramachandran angles ( φ , ψ ) according to the predicted ( or assigned ) secondary structure . In the new version of the TraDES-2 seq2trj program , the ( φ , ψ ) frequency information was derived from an updated non-redundant set of 7 , 030 structures including NMR single model structures for which no corresponding X-ray structure is available . The structures and chains used are listed in the TraDES-2 package data file filtmmdblist . The output of seq2trj is a sampling trajectory file containing sequence-weighted frequency φ , ψ with a 400×400 Ramachandran grid square resolution , representing the propensity for backbone conformational space that can be explored at each step in chain construction . Three sets of 300 , 000 structures each were constructed using the following biases to the φ , ψ-sampling frequencies: In generating an ensemble of 3D structures , TraDES computes the values of the following parameters for each structure: the radius of gyration ( RGyr ) , hydrodynamic radius , N-to-C-terminal distance , accessible surface area , hydrophobic accessible surface area , secondary structure content , and three statistical energy scoring functions . Of these , this study only concerns itself with the values of RGyr . These values are computed during structure generation and are output in log files . 30 , 000 ( 10% ) structures were randomly chosen from each of the 3 sampled sets of 300 , 000 structures . RGyr values of these samples were computed . The sampled structures are available at http://www . iiserpune . ac . in/~madhusudhan/pCas130_mechanosensing in VAL format . Each of the VAL format files could be converted to PDB format using the str2pdb package of the TraDES software . Input instructions are also provided to reproduce similar ensembles . To compare the radius of gyration of sampled CasSD to experimental hydrodynamic radius , 17 additional tag residues with MG at the N-terminus and ENLYFQSLEHHHHHH at the C-terminus had to be accounted for . At this size range the Flory polymer ratio term corresponding to RGyr2/Nl2 ( length l = 3 . 81 Å ) has a constant distribution with peak to mean values in the range of 0 . 407–0 . 595 . From this the RGyr correction for the additional N = 17 residues can be calculated to contribute an additional 1 . 3 ( +/−0 . 1 ) Å to the peak , median or mean values . RH as measured experimentally and the computed RGyr parameter are related [46] by the approximate RGyr/RH ratio of 1 . 06 , based on measurements of urea denatured proteins . Estimates of urea-denatured protein RGyr estimates may be computed from protein length by the relation 1 . 927N0 . 598 [78] , which yields 61 . 0 Å for the tagged CasSD length of N = 306+17 . | Mechanical stretching of cells causes the substrate domain of p130Cas ( CasSD ) to be phosphorylated on 15 tyrosine residues embedded along its length . CasSD is rich in proline and surprisingly well conserved in placental mammals . Stretching of CasSD by atomic force microscopy has identified that it requires far less force than normal folded proteins . Classical biophysical analyses have determined that CasSD is a typical intrinsically disordered protein , a difficult-to-study group of molecules covering about 30% of human proteins . The average size of CasSD is larger and elongated than folded globular proteins but smaller than chemically denatured proteins . We have simulated a large number of all-atom protein structures using a fast all-atom sampling method . The result is in good agreement with the experimental observation . As it is already known that stretching somehow exposes the tyrosine residues to phosphorylation , a mechanism is proposed where straightening of the p130Cas substrate domain backbone conformation through mechanical stretching can lead to dissociation of p130Cas-binding LIM domain proteins and exposure of CasSD tyrosine residues for phosphorylation . This study has led to a new model of a protein-based mechanism of force sensing at the leading edge of cells that allows the cells to feel their way as they move . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"molecular",
"cell",
"biology",
"biophysics"
]
| 2014 | Biophysical Properties of Intrinsically Disordered p130Cas Substrate Domain — Implication in Mechanosensing |
Copper toxicity and copper limitation can both be effective host defense mechanisms against pathogens . Tolerance of high copper by fungi makes toxicity as a defense mechanism largely ineffective against fungal pathogens . A forward genetic screen for Histoplasma capsulatum mutant yeasts unable to replicate within macrophages showed the Ctr3 copper transporter is required for intramacrophage proliferation . Ctr3 mediates copper uptake and is required for growth in low copper . Transcription of the CTR3 gene is induced by differentiation of H . capsulatum into pathogenic yeasts and by low available copper , but not decreased iron . Low expression of a CTR3 transcriptional reporter by intracellular yeasts implies that phagosomes of non-activated macrophages have moderate copper levels . This is further supported by the replication of Ctr3-deficient yeasts within the phagosome of non-activated macrophages . However , IFN-γ activation of phagocytes causes restriction of phagosomal copper as shown by upregulation of the CTR3 transcriptional reporter and by the failure of Ctr3-deficient yeasts , but not Ctr3 expressing yeasts , to proliferate within these macrophages . Accordingly , in a respiratory model of histoplasmosis , Ctr3-deficient yeasts are fully virulent during phases of the innate immune response but are attenuated after the onset of adaptive immunity . Thus , while technical limitations prevent direct measurement of phagosomal copper concentrations and copper-independent factors can influence gene expression , both the CTR3 promoter induction and the attenuation of Ctr3-deficient yeasts indicate activation of macrophages switches the phagosome from a copper-replete to a copper-depleted environment , forcing H . capsulatum reliance on Ctr3 for copper acquisition .
To successfully infect and colonize a host , pathogens must acquire sufficient nutrients from the host to enable microbe growth and proliferation . These metabolic resources include , but are not limited to , essential metals . The nutrient-limited phagosome represents a particularly challenging environment for intracellular pathogens as mammalian hosts can sequester essential elements such as iron and zinc from pathogens . This has been termed “nutritional immunity” [1 , 2] . For example , host molecules such as heme , ferritin , transferrin , and lactoferrin make iron largely inaccessible to microbes [1] . However , successful pathogens have developed sophisticated strategies to combat iron limitation . For example , Mycobacterium tuberculosis and the fungal pathogen Histoplasma capsulatum secrete iron-chelating siderophores [3–5] . Accordingly , inability to synthesize siderophores severely impairs intracellular growth [5 , 6] . In addition , H . capsulatum maintains a slightly acidic intra-phagosomal pH which is sufficient to release iron from host transferrin [7] . Mammalian hosts also restrict available zinc by production of zinc chelating proteins such as S100 family proteins and calprotectin [8 , 9] . In addition , host zinc transporters ( ZIPs ) are employed to tightly control zinc levels in different cellular compartments [10] . Host zinc limitation mechanisms are an important aspect of activation of cellular immunity [11] . However , as with iron limitation , some pathogens have evolved efficient mechanisms to counteract zinc sequestration . High affinity transporters expressed by Salmonella species and H . capsulatum ( ZnuABC and Zrt2 , respectively ) enable these pathogens to import zinc in environments with low zinc concentrations [12–15] . Without these zinc transporters Salmonella and H . capsulatum intracellular proliferation is significantly attenuated . Employing an alternative strategy , the fungal pathogen Candida albicans expresses zincophore ( Pra1 ) , a zinc-chelating molecule , to scavenge zinc during endothelial invasion [16] . Like iron and zinc , pathogen acquisition of copper during infection is essential , but high levels of copper are toxic . Copper killing mechanisms involve reactive oxygen-generating fenton-type reactions , nitrosative stress , or iron-sulfur cluster attack [1] . Recent evidence has shown that immune cells can utilize excessive copper as a powerful weapon to kill pathogens during innate immunity [17–19] . For pathogens , an inability to decrease cellular copper can impair pathogen virulence . For example , M . tuberculosis survival in host cells depends on copper exporter proteins [19] and Salmonella systemic infection requires detoxification of excess copper by a multi-copper-ion oxidase ( CueO ) [20] . The fungus Cryptococcus neoformans ( grubii ) utilizes copper-sequestering metallothionein ( Cmt ) proteins for full virulence during pulmonary infection [17] . On the other hand , there is evidence that host defenses also use copper limitation in some tissue environments . During kidney infection , C . albicans switches from copper-dependent superoxide dismutase 1 ( Sod1 ) to expression of the copper-independent Sod3 [21] . Proliferation of C . neoformans in murine brains requires two copper transporters ( Ctr1 and Ctr4 ) indicating that copper is limited in the mouse CNS [22 , 23] . Thus , maintenance of copper homeostasis in host environments with high or low copper environments is essential for pathogens to establish successful infections . H . capsulatum is a primary fungal pathogen that is not efficiently controlled by innate immunity alone since clearance requires activation of cell-mediated immunity [24 , 25] . H . capsulatum resides within the phagosome of host phagocytes , an environment that is initially permissive for fungal proliferation . Through a forward genetic screen , we identified a homolog of copper transporters ( Ctr3 ) which was required for growth of H . capsulatum in low copper and within the phagosome of host macrophages . We determined that Ctr3 enhances H . capsulatum survival in vivo specifically during the peak of the adaptive immune response to pulmonary infection . Consistent with this , expression of CTR3 increases in low copper concentrations in vitro and in activated , but not in unactivated macrophages . These findings show that copper is sufficiently available to intramacrophage H . capsulatum during innate immunity , but that activation of macrophages induces copper limitation to enact fungal control .
To identify genes required for intramacrophage growth , a genetic screen was designed to identify mutants unable to proliferate within macrophages . Insertion mutants were created using Agrobacterium tumefaciens-mediated transformation of a T-DNA element previously shown to provide relatively random and trackable mutations [26] . To facilitate efficient identification of mutants with reduced intramacrophage proliferation , two indirect assays of H . capsulatum yeast replication within macrophages were used . First , increasing fluorescence of red-fluorescence protein ( RFP ) expressing yeasts was used to indicate intramacrophage yeast replication [27] . Second , a lacZ-expressing macrophage cell line was used to rapidly quantify the ability of mutant yeast to lyse infected macrophages as a result of yeast replication [28] . Individual H . capsulatum mutants were added to macrophage populations to initiate infections . RFP-fluorescence was monitored daily over 7–8 days after which remaining macrophages were quantified by the remaining β-galactosidase activity . Mutants were selected that showed less than 30% increase in RFP fluorescence and/or at least 30% reduction in macrophage lysis . Of 40 , 000 insertion mutants , 178 had reduced intramacrophage growth and/or attenuated virulence in macrophages . The insertion mutations were mapped to the genome by sequencing the regions flanking the T-DNA insertion , and two mutants ( 27H11 and 84D11 ) were identified which had T-DNA insertions in the promoter region of a gene encoding a putative copper transporter ( 193 bp and 215 bp upstream of the CDS initiation codon for 27H11 and 84D11 , respectively ) . We designated the gene CTR3 based on phylogenetic analysis that showed the gene product was similar to copper transporters , including the high affinity Ctr3 copper transporter of Saccharomyces cerevisiae ( S1 Fig ) . Each mutant had approximately 50% reduced RFP-fluorescence ( intramacrophage fungal growth ) and 60% reduced macrophage lysis compared to wild type ( Fig 1A and 1B ) . Consistent with the RFP-based measurement of intracellular growth , CFU-based measurement of viable yeasts within macrophages in culture confirmed that the Ctr3-deficient mutant proliferated only 30% to 50% as well as wild type at 48 hours and 72 hours post-infection ( Fig 1C ) . However , the ctr3 mutant grew as well as the wild-type CTR3 parent in liquid culture ( Fig 1D ) . Complementation of the ctr3 mutants with the wild-type CTR3 gene restored intramacrophage proliferation ( Fig 1A ) and virulence in macrophages ( Fig 1B ) linking intramacrophage growth to Ctr3 function . We also found that Ctr3 is required for intracellular proliferation in another phylogenetically distinct H . capsulatum strain G186A , but the virulence defect observed in G186A yeasts was not as significant as that in G217B background strain ( S2A Fig ) . Consistent with the function of Ctr3 homologs in other fungi , Ctr3 enables H . capsulatum acquisition of copper when copper is limited . Restriction of copper showed Ctr3-deficient yeasts are more sensitive to reduced copper availability; the IC50 of the copper chelator bathocuproine disulfonate ( BCS ) for the ctr3 mutant is over 150-fold lower than that of the CTR3 parent and ctr3/CTR3 complemented strains ( Fig 2A ) . A similar pattern was observed in the G186A strain background; lack of Ctr3 rendered yeasts approximately 10-fold more sensitive to BCS ( S2B Fig ) . Supplementation of BCS-chelated media with excess copper , but neither zinc nor iron , restored the growth of the ctr3 mutant demonstrating the specificity of the phenotype for copper ( S3 Fig ) . Consistent with this , loss of Ctr3 function did not affect growth in iron restricted conditions ( using the iron chelator bathophenanthroline disulfonate ( BPS ) ; Fig 2B ) , suggesting that Ctr3 plays no role in iron uptake ( Fig 2B ) . In contrast to copper-chelation , loss of Ctr3 function does not affect H . capsulatum growth in high copper as both the ctr3 mutant and CTR3 parent strain grow equally well in media with millimolar concentrations of copper ( Fig 2C ) . To directly show Ctr3-dependent copper acquisition , we measured intracellular copper levels in yeasts by inductively-coupled plasma mass spectrometry ( ICP-MS ) . Ctr3-deficient yeasts have lower overall copper levels compared to Ctr3-expressing yeasts during exponential growth in medium containing 10 nM copper ( Fig 3A ) , and starvation of yeasts for copper reduces intracellular copper to baseline levels . Upon replenishment of copper , Ctr3-expressing but not Ctr3-deficient yeasts accumulate intracellular copper ( Fig 3A ) . Both Ctr3-expressing and Ctr3-deficient yeasts have equivalent iron levels in exponential growth and after iron starvation and accumulate iron upon supplementation ( Fig 3B ) , showing the specificity of Ctr3 for copper , but not iron acquisition . Together these data indicate that Ctr3 functions as a copper-specific importer to facilitate growth of yeasts when available copper is low . The Ctr3 requirement for yeast growth in low copper suggests CTR3 expression may be regulated by copper concentrations . Bioinformatic analysis of the H . capsulatum genome identified two additional putative copper transporters which were designated Ctr1 and Ctr2 ( S1 Fig ) . Examination of CTR1 , CTR2 , and CTR3 gene expression by qRT-PCR showed that low copper ( 10 nM ) significantly increased mRNA levels of all three CTR genes compared to high copper ( 10 μM ) conditions; CTR1 , CTR2 , and CTR3 were all induced in low copper media compared to high copper media , regardless of whether cells were grown as yeasts or mycelia ( Fig 4A ) . Interestingly , CTR3 had the highest overall expression of the CTR genes . In mycelia , CTR3 expression was induced by low copper and was expressed at similar levels to that of CTR1 and CTR2 . However , in yeast cells , the expression of CTR3 in high copper was 2 . 5-fold higher than that of CTR1 and CTR2 ( Fig 4A ) and CTR3 expression was further induced nearly 10-fold when yeasts were grown in low copper ( Fig 4A ) . These data indicate that while expression of CTR1 , CTR2 , and CTR3 are all induced by low available copper , differentiation of H . capsulatum cells into pathogenic yeasts establishes an overall higher baseline of expression . Since copper regulates CTR3 expression , we created a green-fluorescent protein ( gfp ) transcriptional fusion to the H . capsulatum CTR3 promoter as a fluorescent indicator of copper availability . CTR3 promoter activity , as indicated by GFP fluorescence of yeast cells , was measured by microscopy after growth in liquid culture and normalized to the fluorescence of an isogenic strain in which gfp expression was controlled by the constitutive H . capsulatum TEF1 promoter . Consistent with the transcriptional analysis , decreasing copper concentrations increased the CTR3 promoter activity , and addition of BCS further increased the reporter GFP-fluorescence to levels at least 5-fold greater than expression in high copper ( Fig 4B ) . Conversely , addition of copper greatly decreases but does not eliminate CTR3 promoter activity . The CTR3 promoter responds to changes in copper concentrations but is not affected by changes in iron or zinc ( S4 Fig ) . In addition , the CTR3 promoter activity is not affected by reactive oxygen stress ( S5A Fig ) or changes in pH ( S5B Fig ) , two physiologically-relevant environmental changes encountered by Histoplasma in the phagosome . These data indicate the specificity of the CTR3 promoter regulation for available copper . To provide quantitative estimates of phagosomal copper levels , the fluorescence of the gfp reporter strains was measured in a gradient of copper concentrations . Analysis of media by inductively coupled plasma mass spectrometry ( ICP-MS ) showed that media without any metal addition had 60 nM trace copper . To reduce copper concentrations below 60 nM necessitated culture of cells in increasing concentrations of BCS . The dose response-data for increasing copper was used to generate a curve of the CTR3-promoter ( PCTR3 ) -controlled GFP fluorescence in 60 nM to 10 μM copper , which showed that CTR3 promoter activity decreased to baseline levels at concentrations above 240 nM copper ( Fig 4B ) . Maximal CTR3 promoter activity reached a plateau of approximately 5-fold higher relative expression in media containing at least 8 μM of BCS ( Fig 4B ) . These data show that the CTR3 promoter is regulated by copper and the CTR3 promoter-dependent fluorescence of gfp reporter yeasts provides an estimate of the available copper in H . capsulatum’s environment . To determine the role of Ctr3 in H . capsulatum virulence , Ctr3-producing and Ctr3-deficient yeasts were tested in a murine model of pulmonary histoplasmosis . Respiratory infections of mice were established using a mixed inoculum of the wild-type and ctr3 mutant strains and the fungal burden in lungs determined over time . To enable measurement of the relative fitness of the Ctr3-deficient ctr3 strain , the ctr3 mutant strain was marked with constitutive GFP-fluorescence , and the viable colony forming units ( cfu ) of wild-type versus mutant strains were differentiated by colony fluorescence . At day 6 post-infection , a time point before significant adaptive immune responses , the ctr3 mutant showed equivalent lung infection as wild type ( Fig 5A ) . However , after the peak of the adaptive immune response ( day 9–21 post-infection ) , the ctr3 mutant was less fit compared to the co-infecting wild-type strain . Complementation of the ctr3 mutant with the CTR3 gene restored the virulence of the mutant ( S6 Fig ) indicating the loss of fitness was due to loss of Ctr3 function . These data show that Ctr3 is required for full virulence , specifically at time points following activation of cell-mediated immunity . The requirement for Ctr3 function in H . capsulatum growth both in limited copper ( Fig 2A ) and for pathogenesis during adaptive immune response stages suggests that copper becomes limiting in the phagosome of phagocytes during adaptive immunity . To probe the intraphagosomal copper concentration during H . capsulatum infection , we used the copper-regulated CTR3 promoter-gfp fusion to measure the CTR3 promoter activity in vivo . Following respiratory infections in mice , H . capsulatum yeasts were collected from lung tissue and the fluorescence of the CTR3 promoter-gfp yeasts measured . Consistent with the equivalent fitness of the wild type and the Ctr3-deficient strain during innate immunity ( Fig 5A ) , the CTR3 promoter activity remained low at 6 days post-infection ( Fig 5B ) . Comparing the yeast GFP fluorescence to the copper concentration dose-response curve for the CTR3 promoter ( Fig 4B ) estimates the copper concentration H . capsulatum encounters at day 6 post-infection is approximately 100 nM , a concentration sufficient to allow yeast growth without Ctr3 function ( Fig 2A ) . However , the CTR3 promoter activity at day 10 and day 14 post-infection was 3- to 5-fold higher than that at day 6 post-infection indicating less available copper in the H . capsulatum-containing phagosome at these time points ( Fig 5B ) . At day 14 post-infection , the fluorescence distribution appears bimodal . The low fluorescent yeasts may reflect a sub-population that is inhibited for growth due to macrophage activation or that not all phagocytes have equivalent changes in phagosomal copper . The average CTR3 promoter activity measured in yeast in vivo at day 10 and 14 ( including the low-fluorescent population ) was similar to growth in liquid medium containing at least 64 μM BCS , a concentration which induces the CTR3 promoter ( Fig 4B ) and at which the ctr3 mutant cannot grow ( Fig 2A ) . Together these data suggest that phagosomal copper becomes significantly limited in phagocytes during the adaptive immune response . As one of the central features of the adaptive immune response involves cytokine activation of phagocytes , we tested which cytokines induce phagosomal copper restriction in H . capsulatum-infected macrophages . For these experiments , the H . capsulatum CTR3 promoter-regulated GFP fluorescence was used to indicate the levels of available copper within the macrophage phagosome . H . capsulatum yeasts within the phagosome of non-activated bone marrow-derived macrophages ( BMDMs ) expressed GFP at moderate levels ( Fig 6A ) . Quantification of the CTR3-driven GFP fluorescence and correlation with the in vitro-derived dose-response data ( Fig 4B ) estimates the phagosomes of unactivated macrophages is around 80 nM . Treatment of H . capsulatum-infected BMDMs with IFN-γ increased CTR3 promoter activity in a dose-dependent manner indicating phagocyte activation with IFN-γ stimulates restriction of phagosomal copper availability ( Fig 6A and 6B ) . However , treatment with TNF-α or GM-CSF did not significantly increase CTR3 promoter activity , suggesting these cytokines do not significantly influence phagocyte phagosomal copper concentrations ( Fig 6B ) . IFN-γ-induced phagosomal copper restriction also occurs in vivo . At day 6 post-infection , the CTR3 promoter activity is normally low ( Fig 5B ) . However , administration of IFN-γ to mice increased the CTR3 promoter activity of H . capsulatum yeasts ( Fig 6C ) ; the average fluorescence was at least two-fold higher with just two IFN-γ treatments indicating that IFN-γ is sufficient to induce copper restriction in vivo . Copper concentration estimation using the dose-response curve ( Fig 4B ) indicates that this IFN-γ treatment causes copper concentration to decrease from 100 nM to well below 60 nM . We used the growth of Ctr3-deficient yeasts as an alternate indicator of cytokine-induced changes to phagosomal copper availability . Intramacrophage growth of Ctr3-deficient yeasts , which are sensitive to low copper , was compared to that of wild-type H . capsulatum yeasts . Without cytokine treatment , Ctr3-deficient yeasts proliferate equally as well in BMDMs as Ctr3-expressing yeasts ( Fig 6D ) . This indicates copper is not limited in the phagosomes of these macrophages and is consistent with the CTR3 promoter activity measurements ( Fig 6A and 6B ) . Treatment of BMDMs with IFN-γ , but not TNF-α , restricted the growth of Ctr3-deficient H . capsulatum yeasts ( Fig 6D ) , demonstrating that IFN-γ triggers restriction of available copper in the phagosome to levels which impair the growth of Ctr3-deficient yeasts . These results are consistent with the CTR3 promoter activity data in IFN-γ-treated macrophages ( Fig 6A and 6B ) indicating that IFN-γ activation of macrophages changes the available phagosomal copper from high to low concentrations ( significantly less than 60 nM ) . For quantifying CTR3 promoter activity , GFP fluorescence driven by the CTR3 promoter was normalized to fluorescence of intracellular yeasts with a GFP-promoter fusion to the TEF1 promoter to control for any changes in global gene expression due to the state of intracellular yeast cells . The activity of the TEF1 promoter is not affected by copper levels ( S7A Fig ) or by residence within unactivated or activated macrophages in culture ( S7A Fig ) . Furthermore , normalization of the GFP fluorescence driven by the CTR3 promoter to a different housekeeping gene ( H2B ) promoter fusion showed a similar increase of the CTR3 promoter in intramacrophage yeasts before and after activation ( S7B Fig ) . Finally , GFP-fluorescence driven by the CTR3 promoter was normalized to RFP-fluorescence driven by the TEF1 promoter within the same yeast cells . This also showed the same induction of the CTR3 promoter in activated BMDMs ( S7C Fig ) . These data indicate that TEF1 promoter activity serves as an accurate normalization factor to account for global transcription variation due to intracellular residence of yeasts . Surveying primary murine phagocytes as well as common macrophage cell lines showed that CTR3 promoter activity of intracellular H . capsulatum yeasts was high in cultured macrophage cell lines , indicating significantly restricted phagosomal copper even without cytokine treatment in these cells ( Fig 7 ) . A similar pattern among cell lines was also observed when the H2B promoter was used for normalization ( S8A Fig ) or when the CTR3-driven GFP fluorescence was normalized to the TEF1 promoter activity within the same cells ( S8B Fig ) . Among primary cells , resident peritoneal macrophages and alveolar macrophages had high phagosomal copper concentrations ( Fig 7 ) estimated at 280 nM and 320 nM , based on the copper dose-response curve for the CTR3 promoter ( Fig 4B ) . This is consistent with the equivalent in vivo proliferation of the Ctr3-deficient and Ctr3-expressing yeasts when H . capsulatum yeasts are primarily present in alveolar macrophages before phagocytes are activated .
In order to establish infections and proliferate in macrophages , H . capsulatum yeasts must acquire essential metals within the phagosomal environment . H . capsulatum secretes siderophores and expresses zinc transporters to combat host limitation of iron and zinc , respectively . In this study , we demonstrate that growth in macrophages also imposes challenges on yeasts to maintain copper homeostasis . Specifically , H . capsulatum yeasts rely on the Ctr3 copper transporter to acquire sufficient copper when copper becomes limiting , both in liquid culture and within macrophages . Besides Ctr3 , the H . capsulatum genome encodes two additional putative copper transporters ( Ctr1 and Ctr2 ) . However , Ctr1 and Ctr2 are not simply redundant with Ctr3; Ctr1 and Ctr2 are not as highly expressed as Ctr3 , and they are not sufficient for copper acquisition when phagosomal copper levels become severely limited . Thus , Ctr3 is the primary transporter involved in copper acquisition as part of H . capsulatum’s pathogenesis program . Two aspects related to H . capsulatum pathogenesis contribute to Ctr3 expression . First , differentiation of H . capsulatum into pathogenic yeasts induces Ctr3 expression independent of copper levels , consistent with a virulence role facilitating H . capsulatum infection of macrophages and persisting within a copper-limited environment . Second , restriction of available copper further increases Ctr3 expression above the level set by yeast-phase differentiation . In support of this dual regulation of Ctr3 transcript levels , the Ctr3 promoter contains putative binding sites for two transcription factors , Ryp1 and Mac1 , which have been implicated in yeast-phase gene regulation and fungal transcriptional responses to copper concentration , respectively [17 , 29 , 30] . Furthermore , ChIP-chip analysis showed that two yeast-phase transcriptional regulators , Ryp1 and Ryp4 , preferentially interact with the CTR3 locus in the yeast phase compared to mycelia [29] . While we can not rule out the possibility that other copper-independent features of macrophage infection do not impact the CTR3 promoter activity , the data suggest that yeast-phase expression of CTR3 is regulated primarily by available copper . Other physiologically-relevant conditions encountered by yeasts within the phagosome ( i . e . , iron and zinc concentrations , reactive oxygen , and pH changes ) do not influence the CTR3 promoter . Consistent with our results , CTR3 is part of a copper-responsive regulon in a microarray-based study of copper-regulated genes [31] , although in this study only yeast responses were examined . H . capsulatum strain differences in CTR3 expression are due to trans-acting factors [32] , likely from variations in either Ryp or Mac1 production or activity among strains . Together , these data are consistent with the model that yeast phase differentiation primes H . capsulatum cells for pathogenesis by inducing basal Ctr3 expression and the level of Ctr3 production is further tuned to the precise level of copper availability in the phagosome . Technical challenges in direct measurement of phagosomal copper required the use of surrogate indicators of copper availability for intracellular H . capsulatum . Using a transcriptional gfp fusion as a semi-quantitative reporter of available copper , we determined that during innate immune stages , H . capsulatum resides within a phagosomal environment with copper concentrations above 100 nM . Consistent with this , CTR3 transcription is lowest for yeast within alveolar macrophages ( Fig 7 ) . At 10 days post-infection , which coincides with the onset of adaptive immune responses , copper becomes restricted . Both in vitro and in vivo , IFN-γ is sufficient to trigger copper restriction . Comparison of in vivo CTR3 transcription to a standard curve generated in vitro estimates phagosomal copper concentrations become significantly lower than 60 nM . As H . capsulatum yeasts are found almost exclusively within phagocytes during mammalian infection [33] , these levels reflect copper concentration within the phagosomal environment at these two points of infection . We note that these copper concentrations are inferred using regulation of the CTR3 promoter , which assumes differential expression is not influenced by copper-independent changes during infection . However , our conclusions about CTR3 transcription reflecting copper dynamics are completely supported by the differential growth of Ctr3-deficient and Ctr3-expressing H . capsulatum strains as a second indicator of copper availability within the phagosome; Ctr3-mediated copper transport is required for H . capsulatum yeast proliferation in vivo at 9 days post-infection but not before . While higher intraphagosomal copper concentration can be microbicidal to some intracellular pathogens , H . capsulatum can tolerate high copper ( up to mM levels ) and thus copper toxicity mechanisms of immune defense are ineffective . As a nearly exclusive intracellular pathogen , H . capsulatum responses to copper levels provide unique insights into the dynamics of the phagosomal environment . Like H . capsulatum , C . neoformans ( grubii ) yeasts up-regulate transcription of copper transporters ( Ctr4 and Ctr1 which are homologs of H . capsulatum Ctr3 and Ctr1 , respectively ) in response to copper restriction [30 , 34 , 35] . For C . neoformans , these transcriptional responses vary by tissue with CTR4 promoter activity increasing in the CNS environment but not in the lung environment , suggesting limited copper in the CNS , but not the lung [17 , 22 , 35] . These results are corroborated by the reciprocal expression profile of C . neoformans copper-binding metallothioneins ( Cmt1 and Cmt2 ) which are induced by high copper concentrations; C . neoformans yeasts in the lung have elevated Cmt1 and Cmt2 expression but not in the brain [17 , 22] . Multiple functional studies with mutants of C . neoformans support the transcriptional profiles since Ctr4- and Ctr1-deficient C . neoformans yeasts , which are unable to grow in limited copper [23 , 30] , are impaired in CNS but not lung infection [22 , 23 , 30 , 36] . In contrast , C . neoformans mutants lacking the Cmt1 and Cmt2 metallothioneins are attenuated in lung infection . Together these studies indicate copper concentrations in the lungs during C . neoformans infection are sufficiently high to not require the Ctr1 and Ctr4 transporters . Initially , these findings appear to contrast with those we observe with H . capsulatum yeasts . However , while H . capsulatum yeasts are nearly exclusively intracellular during infection [33] , C . neoformans has multiple mechanisms to avoid long-term residence within macrophages ( e . g . , formation of phagocytosis-resistant titan cells [37] , production of the anti-phagocytic capsule [38] , secretion of anti-phagocytic protein App1 [39] , and vomocytosis [40] ) . Thus , C . neoformans infection studies indicating the lung is not copper limiting likely include the general extracellular environment , whereas H . capsulatum yeasts indicate phagosome-specific copper concentrations . Indeed , C . neoformans infection of RAW264 . 7 or J774 . 1 macrophages in vitro show up-regulation of CTR4 but not CMT1 expression in intracellular yeasts consistent with our data showing the phagosome in these macrophage cell lines have low available copper [23 , 35] . In addition , Ding et al . found that expression of the mammalian phagosomal copper transporter ATP7A decreased in bronchoalveolar lavage cells at day 14 following Cryptococcus infection consistent with reduced transport of copper into the phagosome during adaptive immunity [17] . These findings and our results with Histoplasma establish a general model that while the extracellular lung environment has ample copper , the phagosome of lung phagocytes becomes copper limiting , particularly following IFN-γ activation . Copper restriction as a mechanism to control fungal pathogens contrasts with copper toxicity as a means to control bacterial pathogens . Phagosomes of macrophages infected with the intracellular bacterial pathogen M . tuberculosis have approximately 400 μM Cu after 1 hour which decreases to 20 μM after 24 hours [41] . Despite this decrease in copper , 20 μM is still a considerably high amount of copper . Consistent with elevated copper levels in the M . tuberculosis-containing phagosome , M . tuberculosis bacteria which have lost the outer membrane copper export protein have reduced tolerance to copper and reduced virulence compared to wild type [19] . Similarly , M . tuberculosis mutants in the RicR regulon are inhibited by high copper ( > 60 μM ) in vitro and are attenuated in vivo [42] . Loss of the CsoR regulon improves copper resistance of M . tuberculosis enabling full virulence during early stages of infection . These data indicate that the M . tuberculosis-containing phagosome contains relatively high concentrations of copper . Supporting this , treatment of macrophages with LPS or IFN-γ increases the host copper transporting protein ATP7A on the phagosome membrane [43] . Even though these studies lack direct measurement of intraphagosomal copper concentrations , they are consistent with findings of elevated copper within latex bead-containing phagosomes and a requirement for ATP7A for phagocyte killing of Escherichia coli [43] . In contrast , 14 days following C . neoformans pulmonary infection ( a time point consistent with IFN-γ production ) alveolar macrophages have decreased ATP7A levels [17] . These differences and our data with intramacrophage H . capsulatum yeasts suggest that macrophages may differentiate between bacterial and fungal pathogens and employ copper toxicity or copper limitation , respectively , in their attempts to limit replication of these two classes of pathogens . With the involvement of adaptive immunity , host utilization of copper for control of fungi switches from copper toxicity to copper restriction . Aspergillus fumigatus conidia infection of alveolar macrophages increases ATP7A expression consistent with elevation of phagosomal copper [18] as an initial phagocyte response . A . fumigatus conidia lacking the AceA transcription factor are less tolerant of high copper and accordingly are less virulent in vivo [18] . Aspergillus cells lacking two Ctr transporters homologous to the fungal Ctr3/Ctr4 and Ctr1 proteins are unable to grow in low copper but lack any virulence defects in vivo [44] . These data indicate that copper toxicity is the primary host defense initially employed against fungal cells ( i . e . , during the innate immune response ) . In contrast to H . capsulatum yeasts , innate immune mechanisms are sufficient for control of A . fumigatus infections and cells that escape initial clearance by phagocytes grow as extracellular hyphae . H . capsulatum yeasts , on the other hand , are not controlled by innate immunity and are primarily intracellular . Instead restriction of H . capsulatum , as well as C . neoformans , requires activation of phagocytes by the adaptive immune system . Our data shows that IFN-γ is key to the switch of macrophage phagosomes from a high copper environment to a copper-limited environment , and it explains , in part , how adaptive immunity contributes to the control of intracellular primary pathogens .
H . capsulatum strains used in this study are listed in the S1 Table and were derived from the G217B and G186A clinical isolates . H . capsulatum yeasts were grown in H . capsulatum-macrophage medium ( HMM , which contains 10 nM CuSO4 ) or in 3M media [45] without added copper for metal supplementation tests with FeSO4 , ZnSO4 , or CuSO4 as appropriate . For growth of uracil auxotrophs , HMM was supplemented with 100 μg/ml uracil . Yeasts were grown with continuous shaking ( 200 rpm ) at 37°C and mycelia cultures at 25°C . Cultures were grown to exponential phase for use in infection studies . For dose-response tests with chelators and metals , yeasts were grown at 37°C in microtiter plates with twice-daily agitation [46] . For growth on solid medium , HMM was solidified with 0 . 6% agarose and supplemented with 25 μM FeSO4 . Growth of G217B derived strains in liquid culture was quantified by measurement of culture turbidity ( optical density at 595 nm ) or enumerating viable CFU by plating dilutions on solid HMM . Growth of G186A derived strains was determined by resazurin-based yeast metabolic assay [46] . Briefly , 100 μM resazurin was added to the yeast culture at 37°C and resorufin fluorescence ( 530 nm excitation , 590 nm emission ) was measured over 90 minutes . LacZ-expressing P388D1 cell line was created from mouse cell line P388D1 ( ATCC CCL-46 , [29] ) . LacZ-expressing P388D1 RAW264 . 7 ( ATCC TIB-71 ) and J774 . 1 ( ATCC TIB-67 ) macrophage cell lines were maintained in Ham’s F-12 medium supplemented with 10% fetal bovine serum ( FBS , Atlanta Biologicals ) . L929 cells ( ATCC CCL-1 ) were maintained in Dulbecco’s modified Eagle medium ( DMEM ) supplemented with 10% FBS . THP-1 cells ( ATCC TIB-202 ) were maintained in RPMI-1640 medium supplemented with 10% ( FBS ) and were differentiated in 10 ng/ml phorbol 12-myristate 13-acetate ( PMA ) for 48 h before use . All cell lines were cultured at 37°C in 5% CO2/95% air . For infection experiments , macrophage cell lines were co-cultured with yeasts in Ham’s F-12 medium supplemented with 10% FBS . Peritoneal macrophages were obtained from wild-type C57BL/6 mice by peritoneal lavage with phosphate-buffered saline ( PBS ) . For elicitation of macrophages , peritoneal injection of 3% protease peptone was performed 4 days prior to lavage . Bone marrow cells were isolated from femurs of C57BL/6 mice ( Charles River ) and differentiated by culturing in Dulbecco’s modified Eagle medium ( DMEM ) supplemented with 30% L929 cell culture supernatant for 7 days to obtain bone marrow derived macrophages ( BMDMs ) . Non-adherent cells were removed from plastic dishes by washing with PBS . Alveolar macrophages were obtained from C57BL/6 mice by bronchoalveolar lavage ( BAL ) with PBS . All primary cells were cultured in DMEM at 37°C in 5% CO2/95% air . H . capsulatum strain OSU233 was used as the genetic background for insertional mutagenesis . OSU233 was constructed by A . tumefaciens-mediated transformation of H . capsulatum yeasts [47] with plasmid pQS01 which contains the apt3 gene for selection ( G418-resistance ) and the tdTomato red-fluorescent protein transgene under control of the H . capsulatum TEF1 constitutive promoter . OSU233 yeasts were mutagenized by A . tumefaciens-mediated transformation [48] using strain LBA1100 harboring plasmid pBHt2 [26] . Bacteria and yeasts were co-cultured for 40 hours on solid Agrobacterium induction medium containing 0 . 1 mM acetosyringone at 25°C . Cells were then transferred to HMM medium containing 100 μg/ml uracil , 100 μg/ml hygromycin to select for H . capsulatum transformants , and 10 μg/ml tetracycline to counter select A . tumefaciens . Plates were incubated at 37°C for 10–12 days until transformants appeared . Individual transformants were picked into liquid HMM with 100 μg/ml uracil in wells of a 96-well microtiter plate and incubated at 37°C for 5 days . Monolayers of P388D1 lacZ-expressing macrophage cells in 96-well microtiter plates [28] were then inoculated with mutant yeasts at a multiplicity of infection ( MOI ) of 1:1 ( yeasts:macrophages ) . Intramacrophage growth of yeasts was monitored daily by measuring RFP fluorescence ( 530 nm excitation , 590 nm emission ) with a Synergy 2 microplate reader ( Biotek ) . After 7 days , surviving macrophages were quantified by removal of culture media from the infected macrophages , lysis of the remaining macrophages with 0 . 1% Triton X-100 , addition of 1 mg/mL o-nitrophenyl-β-D-galactopyranoside ( ONPG ) , and determination of the β-galactosidase activity ( absorbance at 420 nm with correction at 600 nm ) . Mutants with at least 30% reduction in intramacrophage growth or in lysis of the macrophages were retained as candidate attenuated strains . The location of the T-DNA insertion in individual mutants was determined by thermal asymmetric interlaced PCR ( TAIL-PCR; [49] ) . 100 ng of genomic DNA was used as the template for primary PCR , with a T-DNA left or right border-specific primer ( LB11 or RB9 ) and one of four semi-random primers ( LAD1-4 ) . The primary PCR reaction was diluted 500-fold and used as the template for the secondary PCR with nested left- or right-border primers ( LB12 or RB10 ) and the AC1 primer . PCR products were sequenced and aligned to the H . capsulatum genome sequence . T-DNA insertion at the CTR3 locus was confirmed by PCR and sequencing using CTR3-specific primers in conjunction with LB11 and RB9 . Primer sequences are listed in the Key Resource Table . A 1 . 7 kb fragment consisting of the wild-type CTR3 gene and 825 bp of upstream sequence was amplified by PCR from H . capsulatum G217B genomic DNA using CTR3-specific primers ( ORF9-3 and CTR-13 ) and cloned into a URA5-based T-DNA plasmid fusing the CTR3 gene with sequence encoding a C-terminal FLAG epitope . Either the CTR3 complementation vector ( pDT06 ) or a control gfp-expression vector ( pCR628 ) were transformed by A . tumefaciens-mediated transformation into the ctr3 mutants and Ura+ transformants were selected by plating on solid HMM . Reciprocal BLAST searches of fungal genomes using the H . capsulatum Ctr3 protein sequence and other known fungal copper transporters ( e . g . , Saccharomyces Ctr proteins ) were used to identify copper transporter homologs in Saccharomyces cerevisiae , Schizosaccharomyces pombe , Candida albicans , Neurospora crassa , Aspergillus nidulans , Aspergillus fumigatus , Blastomyces dermatitidis , Paracoccidioides braziliensis , Magnaporthe oryzae , Trichophyton rubrum , Ustilago maydis , and Cryptococcus neoformans var grubii . Proteins with E-values less than 10−4 with and at least 50% coverage were aligned and used for construction of a phylogenetic tree ( Clustal Omega ) . H . capsulatum sensitivity to bathocuproine disulfonate ( BCS ) , bathophenanthroline disulfonate ( BPS ) , or excess CuSO4 was assayed by addition of two-fold dilutions of chelators or CuSO4 to HMM in 48-well plates containing 4 × 104 H . capsulatum yeasts/ml . Plates were incubated at 37°C with continuous shaking ( 200 rpm ) for 5 days . Yeast proliferation was quantified by measurement of culture turbidity ( optical density at 595nm ) with a Synergy 2 microplate reader ( Biotek ) . Relative growth in the presence of chelators or CuSO4 was determined by normalization of growth to wells lacking BCS , BPS or CuSO4 . Dose-response curves were determined by non-linear regression of the data and the 50% inhibitory concentrations calculated from the regression curve . Histoplasma yeasts were pre-grown in HMM to late exponential phase and subsequently treated with 2 mM BCS or 16 μM BPS for 24 h to deplete residual copper or iron carried over from HMM . Thereafter , yeast cells were washed 3 times in PBS to remove BCS or BPS and resuspended in HMM for 3 h or 24 h . Yeast cell concentrations were determined by plating serial dilutions on solid HMM . To measure the intracellular total iron or copper , the yeasts were heated at 95°C to remove all the water . To each yeast sample , 0 . 1 ml of concentrated ultrapure nitric acid ( Fisher Scientific Tracemetal grade distilled at TERL using a Savillex DST-1000 ) was added . Samples were digested by floating in a 100°C water bath using a floating tube rack for 15–30 min . Samples were visually inspected to ensure complete digestion until no solids were seen . Samples were cooled and diluted by 10-fold in deionized water spiked with 11 . 11ppb Indium . Final Indium concentration was 10 ppb which was used as an internal standard . All samples were analyzed on a Thermo Finnigan Element 2 Inductively Coupled Plasma Sector Field Mass Spectrometer . Samples were introduced to the ICP at 100 μl/min using a PFA-100 Microflow self aspirating nebulizer ( Elemental Scientific ) pumped at 100 μl/min using 0 . 25 mm I . D . pvc pump tubing and a Gilson 3 peristaltic pump . Iron was measured at m/z 54 , 56 and 57 in medium resolution ( R = 4000 ) and copper was measured at m/z 63 and 65 in medium resolution ( R = 4000 ) . Calibration standards were prepared by dilution from commercially available single element 1000 μg/ml Fe and 1000 μg/ml Cu ( Inorganic Ventures ) into 10% v/v nitric acid to match the sample solvent . Calibration standards also included 10 ppb Indium ( Inorganic Ventures ) as an internal standard . Intracellular total copper or iron were calculated based on 108 yeast cells . CTR gene transcriptional analyses were determined by quantitative RT-PCR ( qRT-PCR ) or by CTR3 promoter-gfp reporter fusions . For qRT-PCR , wild-type yeasts or mycelia were grown in HMM containing low ( 10 nM ) or high ( 10 μM ) concentrations of CuSO4 . RNA was isolated from fungal cells by mechanical disruption with 0 . 5 mm glass beads , extraction with RiboZol ( Amresco ) and alcohol precipitation of nucleic acids . Following DNA removal with DNase , RNA was reversed transcribed with Maxima reverse transcriptase ( Thermo Scientific ) primed with random pentadecamers . Quantitative PCR was carried out using CTR gene specific primer pairs with SYBR green-based visualization of product amplification ( Bioline ) . Changes in CTR transcript levels relative to actin ( ACT1 ) were determined using the ΔΔCt method [50] after normalization of cycle thresholds to that of the TEF1 gene . For construction of the CTR3 promoter-gfp fusion , 1453 bp of sequence upstream of the CTR3 coding sequence was used to drive gfp transcription ( PCTR3-gfp; pMK32 , [32] ) . For normalization purposes , a similar gfp reporter fusion was created using 661 bp of sequence upstream of the TEF1 gene ( PTEF1-gfp ) . Constructs were cloned into a Ura5+ T-DNA vector and transformed by A . tumefaciens-mediated transformation into the WU15 ura5 auxotroph . For relative quantification of CTR3 expression , GFP-fluorescence of PCTR3-gfp transformed yeasts was normalized to GFP-fluorescence of PTEF1-gfp transformed yeasts grown in identical conditions ( in vitro and in vivo ) . For population-based measurements in multi-well plates , GFP fluorescence ( 485/20 nm excitation , 528/20 nm emission ) was measured and the TEF1 or CTR3 promoter activity was calculated by normalization to the number of yeasts present ( OD595 ) . CTR3 promoter activity was then normalized to that of the constitutively TEF1 promoter . For determination of the GFP-fluorescence of individual yeasts , 0 . 1% Uvitex 3BSA was added to yeast suspensions to label the cell wall and yeasts were examined by microscopy ( Nikon E400 ) . GFP-fluorescent ( 480/30 nm excitation , 535/40 nm emission ) and Uvitex-fluorescent images ( 350/50 nm excitation , 460/50 nm emission ) were captured with identical exposure settings and the average GFP fluorescence contained within the Uvitex cell outline was measured ( Micro-manager Studio v1 . 4 . 5 ) and ImageJ [51] . Relative CTR3 promoter activity was determined as the ratio of PCTR3 GFP fluorescence of individual yeasts to the average PTEF1 GFP fluorescence . A standard curve of CTR3 promoter activity at different copper concentrations was generated by incubation of H . capsulatum yeasts with TEF1 or CTR3 promoter-gfp fusions in 3M medium with a gradient of copper concentrations . After 48 hours , GFP fluorescence and culture turbidity ( OD595 ) were measured . TEF1 and CTR3 promoter activities were determined per OD595 unit , and the CTR3 promoter activity was normalized to that of the constitutively expressed TEF1 gene . The actual available copper in the 3M-based media was determined using inductively coupled plasma mass spectrometry ( ICP-MS ) . Media samples were introduced into a Perkin Elmer Nexion 350D ICP-UCT mass spectrometer at the speed of 400 μl/min after being spiked with 10 ppb indium as an internal standard . Copper was measured in DRC ( dynamic reaction cell ) mode using ammonia gas ( 0 . 35 ml/min ) to reduce polyatomic and molecular overlaps . Copper concentration analysis showed that 3M medium without any copper addition contained 60 nM copper . Copper concentrations lower than 60 nM were achieved by adding increasing amounts of BCS . For copper concentrations above 60 nm , a curve was fit to the data by four-variable non-linear regression . For determination of CTR3 promoter activity in macrophages , macrophages were infected with PTEF1-gfp or PCTR3-gfp yeasts at an MOI of 1:2 in 6-well microtiter plates . Phagocytes were previously seeded into 6-well plates at 5 × 105 ( P388D1 , RAW 264 . 7 , J774 . 1 , and THP-1 ) or 3 × 106 ( peritoneal macrophages , and BMDMs ) . Due to the limited yield of alveolar macrophages , 3 × 104 alveolar macrophages were seeded into 96-well plates . For activated macrophages , BMDMs were treated with IFN-γ ( 200 U/ml and 1000 U/ml ) , TNF-α ( 20 U/ml and 100 U/ml ) or GM-CSF ( 2 ng/ml or 10 ng/ml ) for 24 hours before infection . 48 hours following macrophage infection , extracellular yeasts were removed by washing macrophages with PBS and intracellular yeasts released by lysis of macrophages with 1% Triton-X100 . Yeasts in the macrophage lysates were stained by addition of 0 . 1% Uvitex and the GFP fluorescence of individual yeasts determined by microscopy and fluorescence quantification . Phagosomal copper concentrations were estimated by comparison of the CTR3 promoter activity of yeasts recovered from macrophages to the in vitro fluorescence versus copper concentration standard curve . For determination of the phagosome copper concentration in vivo , C57BL/6 mice were intranasally infected with 2 × 104 PTEF1-gfp or PCTR3-gfp yeasts . At day 6 , 10 and 14 post-infection , intracellular yeasts were recovered by euthanizing mice , collection and homogenization of lungs in water , and filtration of the homogenate through a 70 μM cell strainer to remove large debris . The filtrate was treated with collagenase and DNAse for 60 minutes . 0 . 1% Uvitex was added to the filtrate to label yeasts and the GFP-fluorescence of individual yeast was imaged and measured by microscopy . For IFN-γ treatment of mice , infected mice received 2000 U of IFN-γ intranasally at days 4 and 5 post-infection , with control mice receiving PBS in parallel . At day 6 post-infection , intracellular yeasts were recovered from lung homogenates and the GFP-fluorescence measured by microscopy as above . Macrophage monolayers were established in 96-well plates by seeding with 3 × 104 ( P388D1 ) , 2 × 105 ( peritoneal macrophages ) or 1 × 105 ( BMDMs ) cells . For experiments with activated macrophages , BMDMs were treated with IFN-γ ( 1000 U/ml ) , TNF-α ( 100 U/ml ) or both for 24 hours before infection . Macrophages were then infected with Ctr3-expressing or Ctr3-deficient H . capsulatum yeasts at an MOI of 1:2 . Yeast-infected macrophages were incubated at 37°C for up to 72 hours . At different time points , intracellular yeasts were quantified by removal of any extracellular yeasts with the culture supernatant followed by lysis of the macrophages with sterile H2O and plating of the macrophage lysate on solid HMM to enumerate H . capsulatum CFU . To measure in vivo fitness of Ctr3-deficient yeasts compared to Ctr3-expressing yeasts , wild type C57BL/6 mice were infected with H . capsulatum by intranasal delivery of approximately 2 × 104 yeast cells consisting of equal numbers of Ctr3-expressing ( GFP-negative ) and Ctr3-deficient yeasts ( GFP-fluorescent ) . Actual numbers and ratios of yeasts delivered were determined by plating serial dilutions of the inocula on solid media for enumeration of CFU . For fungal burden determination at 6 , 9 , 12 , 15 , 18 and 21 days post-infection , mice were euthanized , lungs were collected and homogenized in HMM , and serial dilutions of the homogenates were plated on solid HMM to determine the fungal burden ( CFU ) and fluorescence of recovered colonies . Colony fluorescence was determined with a modified transilluminator and image capture system [52] . The competitive index was calculated as the number of fluorescent colonies ( ctr3 ) divided by the number of non-fluorescent colonies ( CTR3 ) . For testing complementation of the ctr3 mutant in vivo , the competition assay was repeated using the ctr3/CTR3 complemented strain and wild type at day 6 , 10 and 14 post-infection . Data were tabulated and analyzed by Student’s t-test ( Prism v5 , GraphPad Software ) for determination of statistically significant differences which are indicated in graphs with asterisk symbols ( * , P < 0 . 05; ** , P < 0 . 01; *** P < 0 . 001 ) . Dose-response curves were generated by four-variable non-linear regression . The number of mice and the number of biological replicates used in experiments is specified in the relevant figure legends . H . capsulatum G217B ( ATCC 26032 ) and H . capsulatum G186A ( ATCC 26029 ) were obtained from American Type Culture Collection ( ATCC ) . All animals were housed in Ohio State University’s AAALAC- and OLAW-accredited animal research facilities ( OLAW assurance # A3261-01 ) . All experiments involving mice followed standards in the Public Health Service ( PHS ) “Guide for the Care and Use of Laboratory Animals” and were approved by The Institutional Animal Care and Use Committee ( IACUC ) at Ohio State University ( protocol # 2007A0241 ) . Mice were anesthetized by inhalation of isoflurane and were euthanized by CO2 inhalation as described in the AVMA "Guidelines for the Euthanasia of Animals” | Control of primary pathogens that infect phagocytes often requires adaptive immunity , but the mechanisms that convert host cells from permissive to antimicrobial states are only partially understood . The intracellular fungal pathogen Histoplasma capsulatum resides and proliferates within the macrophage phagosome . During innate immunity , macrophages which normally control fungi prove ineffective against H . capsulatum yeasts . At this stage , the phagosome of unactivated macrophages has ample copper that facilitates intracellular growth of Histoplasma but does not cause copper toxicity . However , the onset of adaptive immunity and the subsequent activation of macrophages decreases phagosomal copper and macrophages become less permissive to Histoplasma proliferation . IFN-γ acts as a key cytokine for switching the macrophage strategy by changing phagosomes from a copper-sufficient to a copper-depleted state in order to control intracellular pathogens . In such activated macrophages , H . capsulatum yeasts upregulate expression of the Ctr3 copper transporter to enable continued acquisition of essential copper . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"yeast",
"infections",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"vesicles",
"markov",
"models",
"pathogens",
"immunology",
"microbiology",
"organisms",
"fungi",
"mathematics",
"phagosomes",
"fungal",
"diseases",
"cellular",
"structures",
"and",
"organelles",
"fungal",
"pathogens",
"infectious",
"diseases",
"mycology",
"white",
"blood",
"cells",
"animal",
"cells",
"medical",
"microbiology",
"hidden",
"markov",
"models",
"alveolar",
"macrophages",
"microbial",
"pathogens",
"probability",
"theory",
"yeast",
"eukaryota",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"macrophages",
"histoplasma",
"capsulatum"
]
| 2018 | Macrophage activation by IFN-γ triggers restriction of phagosomal copper from intracellular pathogens |
Gene expression dynamics have provided foundational insight into almost all biological processes . Here , we analyze expression of environmentally responsive genes and transcription factor genes to infer signals and pathways that drive pathogen gene regulation during invasive Candida albicans infection of a mammalian host . Environmentally responsive gene expression shows that there are early and late phases of infection . The early phase includes induction of zinc and iron limitation genes , genes that respond to transcription factor Rim101 , and genes characteristic of invasive hyphal cells . The late phase includes responses related to phagocytosis by macrophages . Transcription factor gene expression also reflects early and late phases . Transcription factor genes that are required for virulence or proliferation in vivo are enriched among highly expressed transcription factor genes . Mutants defective in six transcription factor genes , three previously studied in detail ( Rim101 , Efg1 , Zap1 ) and three less extensively studied ( Rob1 , Rpn4 , Sut1 ) , are profiled during infection . Most of these mutants have distinct gene expression profiles during infection as compared to in vitro growth . Infection profiles suggest that Sut1 acts in the same pathway as Zap1 , and we verify that functional relationship with the finding that overexpression of either ZAP1 or the Zap1-dependent zinc transporter gene ZRT2 restores pathogenicity to a sut1 mutant . Perturbation with the cell wall inhibitor caspofungin also has distinct gene expression impact in vivo and in vitro . Unexpectedly , caspofungin induces many of the same genes that are repressed early during infection , a phenomenon that we suggest may contribute to drug efficacy . The pathogen response circuitry is tailored uniquely during infection , with many relevant regulatory relationships that are not evident during growth in vitro . Our findings support the principle that virulence is a property that is manifested only in the distinct environment in which host–pathogen interaction occurs .
Which genes does a pathogen express during infection ? Which regulatory pathways govern expression of those genes in vivo ? These questions are central to the study of microbial pathogenesis , and they have been addressed by diverse approaches [1–4] . Despite those efforts , we have a limited understanding of gene expression dynamics during infection of humans or animals by most pathogens . Even new genome-wide technologies on the horizon face several well-acknowledged technical hurdles before they can be implemented in tissues of infected animals [5] . Here , we have used an exquisitely sensitive technology to elucidate gene regulation during tissue invasion by the fungal pathogen Candida albicans . C . albicans is a human commensal that lives on mucosal surfaces of the gastrointestinal and genitourinary tracts [6] . Deep tissue infection begins when the organism gains access to the bloodstream due to disruption of mucosal surfaces or biofilm growth on an implanted device . C . albicans disseminates via the bloodstream and can infect almost any tissue [6] . A mouse hematogenously disseminated candidiasis ( HDC ) infection model , in which C . albicans yeast cells are inoculated into the lateral tail vein , has been widely used to study invasive candidiasis [7] . Although C . albicans invades and infects virtually all tissues , the kidney is the principal target organ . In the kidney , C . albicans proliferates as hyphae [7] , which are long tubular cells attached end to end . During the first 12 hr postinfection , relatively few fungal cells are present in the kidney . Pro-inflammatory cytokines , including TNFα and IL-1β , are detected in the kidney and in circulation by this time [7] . By 24 hr , the fungal burden increases by a factor of 100 , and leukocyte infiltration begins . By 48 hr , the fungal burden increases by another factor of 10 , and leukocyte infiltration is extensive [7] . Prior studies have profiled gene expression in the kidney during invasive C . albicans infection using microarray technology [4 , 8 , 9] . These pioneering studies established several basic principles that have shaped the thinking in Candida infection biology . Specifically , examination of C . albicans gene expression revealed the induction of stress response genes , adhesins , and fatty acid utilization genes during infection [4 , 8] . One study , which used yeast-form cell RNA for comparison , detected induction of hyphal genes , as expected from the extensive hyphal growth observed in infected kidney [8] . These findings argued that adaptation ability is central for proliferation in a novel niche like the kidney , and that hyphal morphogenesis during infection is accompanied by the hyphal gene expression program that has been characterized during growth in vitro . Host gene expression in the kidney is also broadly affected at 24 hr postinfection [9] . There is extensive induction of pro-inflammatory cytokine genes , including IL-6 , Kc/Cxcl1 , Cxcl10 , Cxcl11 , Cxcl13 , IL-1β , and Tnfα [9] . In addition , there is substantial induction of host pattern recognition receptor genes and their signaling pathway components , including TLR2 , Dectin-2 , DC-Sign , and Myd88 [9] . These profiling studies paint a picture of kidney invasion in which C . albicans proliferates in hyphal form , adapting to such stresses as the need to use alternative carbon sources , and induces a pro-inflammatory response in the host that ultimately leads to leukocyte trafficking into the kidney . While prior profiling studies have provided foundational insight into mammalian infection , the dynamic range and sensitivity of microarray signals [5 , 10 , 11] may have limited detection of key regulatory genes that govern the infection process . In addition , while reliable gene expression signals were recorded for infection with wild-type C . albicans strains , it was unclear whether those profiling approaches could be applied under conditions in which organ fungal burden was diminished , such as with attenuated mutants or after drug therapy . We have recently implemented nanoString technology for analysis of pathogen gene expression during infection [12 , 13] . For oral and abdominal C . albicans infection models , in which pathogen cells are relatively numerous , we have been able to collect a snapshot of infection samples and elucidate the roles of two well-established virulence regulators , Bcr1 and Rim101 [12 , 13] . Here we turn our attention to the most widely used C . albicans infection model to capture the first time-course analysis , to our knowledge , of large-scale gene expression during invasive Candida infection . Our findings elucidate the unique functional interactions among virulence regulators that are manifested during infection and document an unexpected relationship between genes that respond to the infection environment and to antifungal therapy .
We used a nanoString n-counter [10] to quantify selected C . albicans gene transcripts in whole kidney lysates at 12 , 24 , and 48 hr postinfection . We first tested our methodology using a probe set that contained 248 environmentally responsive genes chosen from previously described genome-wide datasets ( S1 Data ) . Expression levels of these genes are known to respond to a range of signals , such as nutrient limitation and cell morphogenesis . Also , many of the response pathways that govern their expression are known to impact virulence , as inferred from mutant analysis and comparisons to other pathogens [14] . NanoString probe background signals were assessed by comparing uninfected and infected tissue samples . Signal-to-noise ratios for the majority of probes in all samples fell into the range of 101 to 106 ( S1 Fig . ) . Agreement among independent samples was excellent ( S1 Fig . ; R2 values of 0 . 94 to >0 . 99 ) , and induction ratios were confirmed by quantitative reverse transcription polymerase chain reaction ( QRT-PCR ) for all eight genes examined ( S2 Fig . ) . Probe signals from infected tissue samples varied over a range of three orders of magnitude , and comparison with pure C . albicans cultures indicated that pathogen RNA makes up 0 . 2% or less of the total RNA in the infected kidney over the 48 hr time frame examined . These results show that this approach can reliably detect C . albicans RNAs of low abundance in invasive infection samples . To assess changes in gene expression during infection , we compared the infected tissue samples to each other and to inoculum samples ( stationary phase in yeast extract peptone dextrose medium [YPD] ) . The results revealed that C . albicans undergoes both early and late infection responses ( Fig . 1; S1 Data ) . The early gene expression response comprises genes with RNA levels significantly different from the inoculum at 12 hr postinfection ( p < 0 . 05 and ≥2-fold change in expression; Fig . 1 ) . We found 65 up-regulated early genes and 74 down-regulated early genes . The late gene expression response comprises genes with RNA levels that are significantly different from the 12 hr time point at 48 hr postinfection ( Fig . 1; S1 Data ) . We found 79 up-regulated late genes and 13 down-regulated late genes . These results indicate that C . albicans gene expression is dynamically regulated during invasive infection of a mammalian host . To interpret the early and late gene expression changes , we compared our data to 166 published gene expression studies ( S2 Data ) . We identified significant correlations among genes with expression changes of 2-fold or greater between datasets , as assessed with Fisher's exact test ( FET ) . The most significant correlations were with genes up-regulated during invasive infection ( Fig . 2A ) . Early up-regulated genes correlated well with genes expressed during the yeast-to-hyphal transition , including growth at 37° in Roswell Park Memorial Institute medium 1640 ( RPMI ) , Lee's medium , Spider medium , or serum , all compared to growth at 30° in YPD . This correlation extended to the profiles of mutants defective in regulation of hyphal formation . For example , early up-regulated genes correlated with those up-regulated in a hyperfilamentous tup1Δ/Δ mutant ( “tup1/wild type [WT]” dataset in Fig . 2A ) and with those down-regulated in a nonfilamentous tec1Δ/Δ mutant ( “WT/tec1” dataset in Fig . 2A ) . Early up-regulated genes also correlated well with genes that are repressed by iron acquisition gene repressor Sfu1 ( “sfu1/WT” dataset ) or activated by zinc acquisition gene activator Zap1/Csr1 ( “WT/zap1” dataset ) . There was also a strong correlation with genes that depend upon transcription factor Rim101 for expression ( "WT/rim101" dataset ) . Although Rim101-dependent genes overlap with yeast-to-hyphal genes , the correlation was significant with Rim101-dependent genes from an nrg1Δ/Δ background , in which hyphal gene expression changes were minimized [15] . There was also significant similarity to genes that were up-regulated mid-way through a zebrafish infection ( “8 hr postinfection [PI]” dataset [16] ) , a point we return to in the Discussion . These correlations suggest the hypothesis that hyphal formation , iron limitation , zinc limitation , and Rim101 activation may be driving forces that govern early gene expression responses . We tested several of those inferences through the profiling of mutant strains presented below . The late up-regulated genes correlated significantly with a distinct spectrum of gene expression responses ( Fig . 2B ) . The late up-regulated genes correlated with genes up-regulated at late times during zebrafish infection . In addition , they correlated with genes induced by growth on acetate as carbon source , compared to glucose . The acetate response was assayed by Lorenz et al . [17] to model the metabolic state induced after phagocytosis by macrophages , and indeed the late infection response we describe here correlates with the response to co-culture with macrophages and to peroxide stress ( Fig . 2B ) . Interestingly , we note that the zinc- and iron-limitation responses that were prominent early in infection begin to subside . This point is illustrated by the fact that early genes correlated with the "WT/zap1" dataset , while late genes correlate with its inverse , designated "zap1/WT . " This reversal of iron and zinc responses may reflect a release of nutrients from damaged tissue , a phenomenon documented by Potrykus et al . with respect to iron levels [18] . These results suggest that the early and late infection environments are substantially different . The late gene expression responses may be driven by tissue damage and an inflammatory cell influx . Two phases of host gene expression were also discernible ( Fig . 3A; S3 Data ) . RNA levels for 46 mouse genes associated with fungal infection responses were assayed in each kidney sample . One set of genes was induced early in infection , reaching peak levels by 24 hr postinfection . This set included genes for pro-inflammatory cytokines ( such as Tnfα and IL-6 ) and pattern recognition receptors ( such as Mincle and Dectin-2 ) , in keeping with the study by MacCallum at 24 hr postinfection [9] . These rapid responses are likely to be mediated by cells that are resident in the kidney . A second set of genes was induced late in infection , with significant expression increases between 24 and 48 hr ( Fig . 3A ) . This second set included the neutrophil marker NOX ( NADPH oxidase 1 ) and the inflammatory macrophage marker CCR2 . The set also included genes whose expression is associated with T cells , such as interferon γ and IL-17a . These late responses probably reflect leukocyte migration into the kidney . Thus , both pathogen and host display distinct early and late temporal gene expression responses during invasive infection . In order to identify regulators that govern gene expression during infection , we employed a probe set that included all 231 known or predicted C . albicans transcription factor genes . Early and late gene expression phases were evident among these genes ( Fig . 4 , S4 Data ) . Early up-regulated transcription factor genes included regulators of hyphal formation ( Ume6 , Tec1 ) , zinc acquisition ( Zap1 ) , and iron acquisition ( Hap43 , Sef1 ) . The late response included many transcription factors whose function has not been analyzed in detail previously . Functions of the early class of transcription factor genes that are known are consistent with the properties of environmentally responsive genes presented above . Transcription factors that are expressed at high levels or significantly up-regulated during infection may govern pathogen proliferation and virulence . An analysis of previously studied transcription factors supports this idea ( Table 1 ) . Expression level was an excellent predictor of function . Specifically , among the 30 most highly expressed transcription factor genes at 48 hr postinfection , 18 have been found to be required for proliferation or virulence among 24 tested previously ( 75% ) . Among the 30 most weakly expressed transcription factor genes at this time point , one is required for proliferation or virulence among 12 tested previously ( 8% ) . This difference is highly significant ( p < 0 . 001 by FET ) . The up- or down-regulation of expression was a weaker predictor of function . Among the 30 most highly up-regulated transcription factor genes ( comparing 48 hr postinfection to the inoculum samples ) , 15 are required for proliferation or virulence among 20 tested previously ( 75% ) . Among the 30 most down-regulated transcription factor genes , seven are required for proliferation or virulence among 16 tested previously ( 44% ) . There is no significant difference between these expression classes ( p > 0 . 08 by FET ) . Therefore , the transcription factor gene expression level , rather than its regulatory response , is the most discerning predictor of function in this infection model . However , transcription factors that are most highly up-regulated or most highly expressed are equally likely to have clear roles in proliferation and virulence ( p = 1 . 0 by FET ) . We sought to understand regulatory relationships during infection , and we began this analysis with the transcription factor Rim101 . We chose RIM101 because it is the most highly expressed transcription factor gene at 48 hr postinfection; its RNA levels comprise over 3% of the total RNA levels for all 231 transcription factor genes ( S4 Data ) . Rim101 clearly functions during infection: early up-regulated genes correlate with known Rim101-dependent genes , and prior studies show that it is required for virulence in the mouse disseminated infection model , where it promotes both hyphal formation and proliferation [19] . In order to define the gene expression basis for the rim101Δ/Δ mutant virulence defect , we assayed expression of 148 environmentally responsive genes by the rim101Δ/Δ mutant and rim101Δ/Δ+pRIM101 complemented strain at 24 hr postinfection ( S5 Data ) . The rim101Δ/Δ mutant presented significantly reduced accumulation of 33 RNAs ( ≥2-fold change and p < 0 . 05 , including RIM101 itself ) , and increased expression of 14 RNAs , compared to the wild-type strain ( S5 Data ) . RNA accumulation in the complemented strain closely mirrored that of the wild type ( Fig . 5A ) . Specifically , RNA accumulation levels were restored by the complementing RIM101 allele for all genes except GAP2 , PHR1 , and HYR1 ( p < 0 . 05 ) . In the complemented strain , expression of GAP2 and PHR1 trended toward the wild-type level , thus suggesting that these few gene expression differences from the wild type are the result of reduced gene dosage of RIM101 alleles in the complemented strain compared to the wild type , as documented previously for these strains [19] . Genes that were down-regulated in the rim101Δ/Δ mutant included many that are induced during hyphal growth ( such as HWP1 , ECE1 , SOD5 , and SAP5 ) . Their reduced expression in the mutant is consistent with its deficiency in hyphal growth in vivo [19] . Two transcription factor genes that are required for hyphal growth , EFG1 and TEC1 , were also down-regulated in the mutant and may account for its failure to form hyphae and express hyphal genes . We note that the few direct Rim101 target genes assayed have altered RNA levels in the mutant , including Rim101-activated genes PHR1 and PRA1 , as well as Rim101-repressed gene PHR2 [20 , 21] . Overall , these in vivo profiling data are consistent with known features of the rim101Δ/Δ pathogenicity defect and the molecular data about the Rim101 protein . The gene expression impact of the rim101Δ/Δ mutation has been characterized extensively in vitro , and we noted several differences between published in vitro profiles and our in vivo data . Such differences can arise from the assay platform , and we thus examined the rim101Δ/Δ mutant and rim101Δ/Δ+pRIM101 complemented strain under hypha-inducing conditions in vitro ( Spider medium at 37°C ) with the nanoString platform . We examined expression of 144 environmentally responsive genes ( S5 Data ) . In this set , 37 genes responded significantly ( ≥2-fold change , p < 0 . 05 ) to Rim101 in vivo , 11 genes responded significantly to Rim101 in vitro , and only eight responded significantly to Rim101 under both conditions ( compare blue and green bars in Fig . 5B and Fig . 5C ) . The genes that responded significantly to Rim101 only in vivo include several with roles in pathogenicity ( EFG1 , HWP1 , SAP6 , and TEC1 ) . These results emphasize that the invasive infection environment can alter the spectrum of genes that respond to a transcription factor , and they suggest that such alterations have the potential to influence virulence . We have recently analyzed the rim101Δ/Δ mutant and rim101Δ/Δ+pRIM101 complemented strain on the nanoString platform during abdominal infection [12] . Among 144 genes assayed in both infection models , we found 46 genes that responded significantly to Rim101 during abdominal infection , and only 15 responded similarly in both infection models ( Fig . 5B and Fig . 5C , blue and red bars ) . Notably , the hyphal regulatory genes EFG1 and TEC1 did not require Rim101 for expression during abdominal infection . Therefore , the site of infection can affect the relationship between a transcription factor and its target genes . We also examined the gene expression impact of a defect in the transcription factor Efg1 . An efg1Δ/Δ mutant was among the first attenuated C . albicans strains characterized [22 , 23] . Efg1 governs numerous pathogenicity-related phenotypes , including adherence to diverse cells and substrates , formation of hyphae at 37°C , colonization of the gastrointestinal tract , and antifungal drug susceptibility [14 , 24] . The impact of efg1Δ/Δ mutations has been assessed on the C . albicans transcriptome during growth in vitro [25] , but never in vivo . We confirmed that the proliferation of an efg1Δ/Δ mutant was severely defective in the mouse disseminated infection model , yielding overall RNA levels too low to detect expression of many C . albicans genes at 24 hr postinfection ( Fig . 6B ) and a greatly diminished host response ( Fig . 3B ) . Therefore , we assayed C . albicans gene expression in an efg1Δ/Δ mutant and efg1Δ/Δ+pEFG1 complemented strain at 48 hr postinfection . Among 148 environmentally responsive genes assayed , we found 26 genes that were significantly down-regulated ( including EFG1 itself ) and 63 genes that were significantly up-regulated ( ≥2-fold and p < 0 . 05 ) , compared to the wild-type strain ( Fig . 6A; S5 Data ) . Presence of one EFG1 allele in the complemented strain increased EFG1 expression to 26% of the wild-type level , and permitted partial or complete restoration of environmentally responsive gene expression levels ( S3 Fig . ; S5 Data ) . This dosage effect is consistent with prior studies that document haploinsufficiency for EFG1 [26] . The 26 down-regulated genes included many hyphal genes ( for example , ALS3 , ECE1 , HWP1 , IHD1 , and SOD5 ) , as expected from profiling of an in vitro–grown efg1Δ/Δ mutant [25] . Among the 63 up-regulated genes , only four were up-regulated in the in vitro efg1Δ/Δ characterization [25] , and their fold-change was much greater in vivo than in vitro . These results indicate that the efg1Δ/Δ mutant is defective in expression of hyphal genes during proliferation in vivo , as expected from its previous phenotypic characterization . In addition , the efg1Δ/Δ mutant , like the rim101Δ/Δ mutant , has impact on gene expression during proliferation in vivo that is distinct from what is observed in vitro . In order to visualize gene expression alterations associated with virulence defects in a broader context , we profiled a panel of mutants that are defective in proliferation in vivo . We selected a zap1Δ/Δ mutant because many known Zap1-dependent genes are up-regulated during infection , as discussed above . The zap1Δ/Δ mutant has been shown to be defective in proliferation previously [27] , a phenotype that was confirmed by its reduced RNA levels in infected tissue ( Fig . 6B ) and a diminished host response ( Fig . 3B ) . We also made mutations in three highly expressed or up-regulated transcription factor genes that had not been studied previously in a disseminated infection model: ROB1 ( orf19 . 4998 ) , RPN4 ( orf19 . 1069 ) , and SUT1 ( orf19 . 4342 ) . Each mutation caused reduced proliferation in vivo , as indicated by a decreased fungal RNA yield ( Fig . 6B ) and host response ( Fig . 3B ) . Histopathology examination of kidney sections at 24 hr postinfection showed that the attenuated mutants were able to invade the kidney parenchyma , though fungal cells were much less abundant than observed in mice infected with the wild-type strain ( Fig . 6C ) . Complementation with a wild-type copy of each gene led to increased fungal RNA yield and host response gene expression ( Fig . 6B and Fig . 3B ) . Therefore , defects in ROB1 , RPN4 , and SUT1 cause defects in proliferation in vivo . In order to identify the virulence pathways that are governed by these transcription factors , we assayed expression of 148 C . albicans environmentally responsive genes at 24 hr postinfection in the wild type and the attenuated transcription factor mutant strains ( Fig . 6A; S5 Data ) , and compared the results to the rim101Δ/Δ and efg1Δ/Δ data presented above . Gene expression alterations in each mutant were largely restored in the respective complemented strains ( S3 Fig . ; S5 Data ) , thus indicating that the mutation introduced into each strain was the cause of the gene expression alteration . Only three genes had significantly altered expression ( ≥2-fold , p < 0 . 05 ) in all mutants , the up-regulated cell wall or secreted protein genes GCA2 , RBR1 , and RBR2 . Although no firm conclusion can be drawn from so few genes , the results suggest that attenuated mutants may undergo a common cell surface alteration . Prior and current studies indicate that Rim101 , Rob1 , and Efg1 are required for hyphal formation , so we looked for common gene expression alterations in those three strains . Common down-regulated genes included CRH11 , ECE1 , FAV2 , HWP1 , IFD6 , SAP5 , and SAP6 , most of which are associated with hyphal morphogenesis . Therefore , this feature of the in vivo gene expression profile is consistent with prior in vitro analysis of the hyphal morphogenesis program and the mutant phenotypes observed in vivo . Three of the mutants , rpn4Δ/Δ , sut1Δ/Δ , and zap1Δ/Δ , had extremely similar gene expression profiles ( Fig . 6A top right; S5 Data ) . These results were unexpected because these three C . albicans mutants have distinct phenotypes in vitro . For example , on low-zinc medium , the zap1Δ/Δ mutant fails to grow , while the rpn4Δ/Δ and sut1Δ/Δ mutants grow well ( S4 Fig . ) . Also , during growth in vitro , the gene expression alterations of the three mutants have little similarity ( S4 Fig . ) . We considered two explanations for our observations . One possibility is that Rpn4 , Sut1 , and Zap1 act in a pathway that operates in the invasive infection environment to govern proliferation; in vitro growth conditions may alter their spectra of target genes . A second possibility is that Rpn4 , Sut1 , and Zap1 are functionally unrelated , and the in vivo gene expression profiles of the three mutants represent general proliferation-defective responses . We focused on Sut1 and Zap1 to determine whether there may be a functional relationship in vivo . We chose those two transcription factors because , during proliferation in vivo , the sut1Δ/Δ mutant had reduced RNA levels for ZAP1 and the Zap1-dependent genes PRA1 , ZRT1 , and ZRT2 ( Fig . 7A ) . Therefore , we considered the specific hypothesis that the sut1Δ/Δ phenotypic defect during infection arises from its ZAP1 expression defect . This model predicts that overexpression of ZAP1 will suppress the defects of a sut1Δ/Δ mutant . We created a ZAP1 overexpression allele by fusing the ZAP1 coding region to a strong TDH3 promoter in a sut1Δ/Δ strain . The sut1Δ/Δ TDH3-ZAP1 strain expressed zinc acquisition genes PRA1 , ZRT1 , and ZRT2 during infection at higher levels than the sut1Δ/Δ mutant ( Fig . 7A ) . In order to assess the biological significance of the Sut1-Zap1 relationship , we made use of the finding that the sut1Δ/Δ mutant is defective in virulence ( Fig . 7B ) . Specifically , mice infected with the sut1Δ/Δ mutant survived substantially longer than those infected with the wild type or complemented strain . The sut1Δ/Δ TDH3-ZAP1 strain displayed much greater virulence than the sut1Δ/Δ mutant , causing host lethality with nearly wild-type kinetics ( Fig . 7B ) . We considered the possibility that ZAP1 overexpression may cause a nonspecific increase in virulence . However , an otherwise wild-type strain that carried the TDH3-ZAP1 allele had reduced virulence rather than increased virulence ( Fig . 7B ) . These findings are consistent with the model that Sut1 is necessary for pathogenicity because it is required for ZAP1 expression during infection . To test the specific hypothesis that the sut1Δ/Δ mutant is attenuated because it is defective in zinc uptake , we assayed virulence of strains that express the zinc transporter gene ZRT2 from the TDH3 promoter . The TDH3-ZRT2 allele did not alter virulence in a wild-type background , but fully restored virulence of a sut1Δ/Δ mutant ( Fig . 7C ) . This finding proves that the zinc transporter expression defect of the sut1Δ/Δ mutant is the cause of its virulence defect . Therefore , Sut1 is required for zinc acquisition in the invasive infection environment . Caspofungin , a cell wall inhibitor , is an extremely effective antifungal drug [6] . The gene expression response to caspofungin treatment has been assayed during growth in vitro [28–30] , but not under infection conditions . We reasoned that the response to caspofungin may be different in vivo . Therefore , we assayed expression of the 248 environmentally responsive genes 2 hr after caspofungin administration in mice that had already been infected for 24 hr . There was no detectable decline in fungal cell number after this brief treatment time , but an extensive gene expression response was manifested ( S6 Data ) . A much broader set of genes was induced by caspofungin in vivo than had been detected through previous in vitro studies ( Fig . 8A ) . Specifically , 44 of the genes assayed showed significantly increased RNA accumulation ( ≥2-fold , p < 0 . 05 ) after caspofungin treatment in vivo compared to untreated infection samples . Several induced genes specify cell wall or secreted proteins ( ALS4 , ALS9 , GCA2 , PGA13 , PGA26 , PGA31 , PGA37 , PHR2 , PIR1 , RBR2 , SAP1 , and SAP9 ) , and others specify enzymes that function in glucose generation ( PCK1 , GPM2 , and DAK2 ) . Thus , the in vivo response suggests that the cell wall is restructured in response to the drug through an alteration of cell wall protein composition and an increase in β-glucan synthesis . The gene expression response to caspofungin during infection showed little similarity to the previously characterized response during growth in vitro ( Fig . 8A ) . Among significantly regulated genes from one previous in vitro study [30] , 32 could be detected by our nanoString probes , and only four were regulated in parallel ( up-regulated: PHR2 , RTA4 , SAP9; down-regulated: FGR41 ) . Previous studies had been carried out with yeast-form cells and with a microarray platform , both of which may contribute to the divergence of results . Therefore , we assayed the gene expression response to caspofungin during hyphal growth conditions ( RPMI , 37°C ) with the same 248 nanoString probes used for in vivo profiling . We found a broad response to caspofungin under these in vitro conditions , with 19 down-regulated genes and 32 up-regulated genes ( Fig . 8A; S6 Data ) . The up-regulated genes included numerous cell wall or secreted protein genes ( ALS1 , CHT2 , CRH11 , DFG5 , ECM331 , KRE1 , PGA17 , PGA26 , PHR1 , RBR1 , and SOD4 ) as well as glucose generation genes ( FBP1 , ICL1 , and PCK1 ) . Surprisingly , though , there was still limited similarity to the in vivo response to caspofungin: seven genes were regulated in parallel ( up-regulated: PGA26 , PCK1 , MET3 , and YHB1; down-regulated: ATO2 , CIP1 , and LAP3 ) . These results indicate that the infection environment has considerable impact on the gene expression response to caspofungin antifungal treatment . In order to determine whether any previously characterized gene expression responses resemble the in vivo caspofungin response , we compared in vivo caspofungin-responsive genes to our database of gene expression data . We found striking overlap between caspofungin up-regulated genes and those genes that are repressed at 12 hr postinfection ( Fig . 8A; p < 0 . 0001 by FET ) . Specifically , among 44 genes that were up-regulated in vivo in response to caspofungin , 35 genes had been repressed within 12 hr postinfection ( Fig . 8A ) . These 35 genes include several with roles in cell wall biogenesis and integrity , such as PGA31 , PHR2 , PIR1 , and SAP9 . Down-regulation of these genes at an early time during infection may render infecting cells more vulnerable to cell wall inhibitors than in vitro-grown cells . The gene expression responses to caspofungin may differ in vivo and in vitro because of a difference in the transcription factors that mediate the response under each condition . Therefore , we examined the transcription factor genes that are induced by caspofungin under the two conditions ( Fig . 8B; S6 Data ) . Under infection conditions , we detected induction of 18 transcription factor genes ( ≥2-fold induction , p < 0 . 05 ) . Under in vitro conditions ( RPMI , 37°C ) , we detected induction of 13 transcription factor genes . The two sets of induced genes had minimal overlap; only iron regulator SFU1 was induced under both conditions . However , the caspofungin up-regulated transcription factor genes included 11 genes that were down-regulated within 12 hr postinfection . It seems reasonable that the different gene expression responses to caspofungin treatment in vivo and in vitro reflect the difference in transcription factor gene responses that we have discovered here .
The expression profile reported here is , to our knowledge , the largest time-course analysis of C . albicans infection of a mammalian host . The results indicate that there are both early and late gene expression changes during infection . Early genes underscore the importance of many driving forces in infection that have been deduced from previous studies of C . albicans and other invasive pathogens: hyphal formation and limitation for iron and zinc [31 , 32] . Late genes reflect responses that have been logically inferred from interactions with innate immune cells: oxidative stress and the consequences of phagocytosis by macrophages [32 , 33] . In addition , late responses suggest that iron and zinc limitation become less severe as infection proceeds , as reported for iron availability by Potrykus et al . [18] . Therefore , the overall features of gene expression during infection that we describe here fit well with the current understanding of C . albicans infection biology . The changes in carbon metabolic gene expression we observe during infection suggest that invasive cells shift from glycolysis to gluconeogenesis . Early up-regulated genes included four genes that specify subunits of pyruvate dehydrogenase , which is needed for carbon flux from glycolysis into the tricarboxylic acid cycle . Early down-regulated genes included the glycerol biosynthetic genes GPD2 and RHR2 , a likely reflection of metabolic feedback from high levels of the kidney osmoprotectant glycero-phosphocholine [34] , which is utilized by C . albicans during infection [35] . Also down-regulated early was the PCK1 gene , which specifies the gluconeogenic enzyme phosphoenolpyruvate carboxykinase . This observation is consistent with the idea that carbon is metabolized through glycolysis , perhaps augmented by catabolism of glycerol , at early times in infection . Late gene expression changes suggest that carbon metabolism shifts toward gluconeogenesis as infection progresses; the glyoxylate cycle genes ICL1 and MLS2 were up-regulated , as was the gluconeogenic gene PCK1 . These changes may reflect increased lipid catabolism for carbon because the β-oxidation gene FOX2 is up-regulated as well . Interestingly , up-regulation of ICL1 , MLS2 , PCK1 , and FOX2 occurs upon internalization by macrophages [17] . Our results are consistent with an analysis of green fluorescent protein ( GFP ) fusion gene expression during infection by Barelle et al . [36] , who detected higher expression of PCK1 and ICL1 in infecting cells than in glucose-grown cells , which are arguably equivalent to our inoculum . Our interpretation that cells shift from glycoloysis to gluconeogenesis during infection is consistent with the finding by Barelle et al . that a defect in glycolysis blocks establishment of infection , whereas defects in the glyoxylate cycle or gluconeogenesis impair later events in infection [36] . Therefore , the gene expression data reported here are consistent with the observation that glycolysis is required early in infection and that invasive C . albicans cells respond to phagocytosis at later times during infection . Our gene expression data correlate with a recent gene expression profile by Chen et al . of whole zebrafish through a time-course of C . albicans infection [16] . The study normalized samples to the mean expression level over the time-course for each gene . Our early up-regulated genes correlated with genes that Chen et al . detected as up-regulated at 8 hr postinfection ( roughly the midpoint of the time-course ) and as down-regulated at 0 . 5 hr postinfection . Because of the normalization procedure , it is reasonable that both of these zebrafish samples would correlate with our early class . Our late up-regulated genes correlate with genes that Chen et al . detected as up-regulated at 16 hr postinfection , close to the end of the time-course . These correlations suggest that C . albicans carries out similar biological processes in both infection models . Many matches among both early and late genes map to metal ion homeostasis GO terms . This correlation is consistent with the idea that nutritional immunity is used broadly as a defense mechanism [31] . Does C . albicans express different sets of genes at different infection sites ? We can begin to address that question by comparison of the kidney infection profile reported here with infection profiles of other mouse infection models: oropharyngeal candidiasis [13] and abdominal infection [12] . Expression levels for 115 genes were assayed in all three models , and we converted the data to expression ratios relative to the inoculum samples used here ( Fig . 9 ) . Thirty-two genes are up-regulated in all infection models , and they include many genes related to hyphal formation ( ALS3 , ECE1 , HWP1 , SOD5 , TEC1 , UME6 ) , zinc limitation ( PRA1 , ZRT1 , ZRT2 ) and iron limitation ( PGA7 , RBT5 ) . Therefore , with the qualification that the dataset is small , there seem to be common driving forces in all three infection models . There are some examples of genes up-regulated only during oral infection , though too few to allow an inference about relevant regulatory or functional relationships . There are also genes up-regulated only during abdominal infection , which include the adhesin genes ALS5 , ALS6 , and ALS9 . These adhesins may be involved in the abscess formation that characterizes this model [12] . The most striking conclusion , though , is that we see no examples of genes that are expressed only during kidney infection . We have sampled only 2% of the C . albicans genome , and it seems likely that kidney-specific genes will be found . The genes we have sampled are enriched for stress response genes , though , and we speculate that the kidney may be a common target organ for disseminated infection because it is a relatively hospitable environment . The C . albicans transcription factors that govern infection have been extensively studied through both single gene analyses and large scale competitive growth assays [27 , 37] . We find that strong in vivo expression or up-regulation are excellent predictors of function during invasive infection . Expression level seems to be the most discerning predictor; highly expressed transcription factor genes are much more likely to have measurable impact on infection than the weakly expressed . Such considerations may offer useful prioritization of prospective virulence regulators in pathogens that lack the wealth of prior analysis of C . albicans . The infection environment modifies the gene expression impact of the majority of transcription factors that we examined , compared to what is observed under in vitro conditions . The gene expression impact of a mutation in vivo may reflect an amalgam of the direct effects of the transcription factor bound to its target genes , along with input from a spectrum of stress response pathways activated in each mutant when it is unable to proliferate . This explanation seems especially compelling for the efg1Δ/Δ mutant . The profile of this strain in vivo displayed a good correlation among down-regulated hyphal genes but little correlation among up-regulated genes , when compared with in vitro efg1Δ/Δ expression data [25] . The genes up-regulated in vivo were diverse , with many annotated to the GO term Response to Stress ( e . g . , CAP1 , CAT1 , HSP104 , HSP70 , and YHB1 ) . Our data suggest that the kidney environment encountered by each mutant is similar , in that host gene expression response is essentially proportional to the fungal burden assessed by fungal RNA level; we see no evidence for qualitative alteration of the host response by individual mutant strains . Therefore , we would expect to see a common set of genes with altered expression in many different attenuated mutants . We see some evidence of a shared gene expression response among all of the mutants assayed . We believe that genome-wide analysis combined with additional replicates , given the low mutant titers , will document a more extensive shared gene expression response among attenuated strains . Are in vivo mutant profiles informative ? In the case of Rim101 , comparison of the mutant profiles from two different infection sites offers a simple explanation for niche-specific rim101Δ/Δ mutant phenotypes . During invasive growth in the kidney , the rim101Δ/Δ mutant is defective in hyphal formation [19] . Our results here provide a simple explanation that was not understood previously ( see [14 , 38] for recent reviews ) : the rim101Δ/Δ mutant has reduced expression of two hyphal regulatory genes , EFG1 and TEC1 , during kidney invasion . In contrast , during growth in an abdominal infection model , the rim101Δ/Δ mutant is not defective in hyphal formation , nor in EFG1 or TEC1 expression [12] . Many mutant strains are attenuated in multiple infection models , and we often infer that the same pathways or functions are thus required for infection in multiple contexts . Interestingly , though , if the gene expression impact of a mutant defect is tailored differently in distinct infection environments , then the same genetic lesion may lead to different causes for attenuation . If regulatory relationships are modified during infection , then it seems possible that some regulatory pathways may be more prominent during growth in vivo than in vitro . Here , we have identified such a pathway relationship , in which transcription factor Sut1 is required for Zap1 expression , and Zap1 in turn activates zinc acquisition genes that are necessary for proliferation in vivo . The sut1Δ/Δ and zap1Δ/Δ mutants have very similar gene expression alterations in vivo , an observation that led us to hypothesize that they function in the same pathway . Interestingly , the two mutants have little similarity between their gene expression alterations during growth in vitro , and only the zap1Δ/Δ mutant has a zinc acquisition deficiency in vitro . In addition , there is no indication of a functional interrelationship between the orthologous genes in Saccharomyces cerevisiae . In other words , the traditionally sought indications that two genes are functionally related were negative . We demonstrated that Sut1 and Zap1 are functionally related in vivo with the demonstration that ZAP1 overexpression restores zinc acquisition gene expression and virulence to a sut1Δ/Δ mutant . This conclusion was strengthened by the finding that overexpression of the zinc transporter gene ZRT2 also restores virulence to a sut1Δ/Δ mutant . Therefore , in vivo profiling data can define biologically relevant functional relationships that are not evident from in vitro analysis . Our hypothesis is that Rpn4 is functionally related to Sut1 and Zap1 , given the similarity of the three mutant gene expression profiles during infection . Rpn4 is not required for zinc acquisition gene expression , so Rpn4 may govern a response to zinc limitation . There are many alternative hypotheses , though , and the issue may be resolved by future functional studies . It seems logical that pharmacological perturbations , like genetic perturbations , may have distinct effects in vivo and in vitro . The gene expression response to the cell wall inhibitor caspofungin is a clear example . The functional spectrum of caspofungin-induced genes seems similar under all conditions and includes many cell wall–modification genes . However , the particular genes that are induced in infecting cells are quite different from those induced in cells grown in vitro . Our finding that a different selection of transcription factor genes is induced by caspofungin in vivo and in vitro helps to explain the overall difference in transcriptional responses . An unanticipated finding from our caspofungin response profiling is that caspofungin induces many of the same genes that are repressed early in infection . It is not clear why caspofungin reverses the repression of many genes that occurs early in infection; perhaps the high osmolarity of the kidney relieves the need for cell wall reinforcement by eliminating turgor , thus repressing cell wall reinforcing functions early in infection . Alternatively , the sequestration of nutrients in tissue may cause repression of genes early in infection , and perhaps lysis of the first drug-exposed C . albicans cells relieves nutrient limitation and thus repression . What we find exciting is that the correlation may mean that infecting cells , by virtue of repressing genes that help survive drug treatment , may be in a more drug-susceptible state than in vitro–grown cells . This inference fits well with the study of Wheeler et al . [39] , who observed that cell wall β-glucan is more exposed on the C . albicans cell surface of infecting cells than of in vitro–grown cells . Thus the Wheeler study indicates that cell wall structure is different in vitro and in vivo; our data argue that one source of this difference is the transcriptional regulation of cell wall biogenesis genes . Many of the key regulatory and environmental signals that have been deduced to govern C . albicans behavior during infection are indeed manifested at the level of gene expression . Regulatory relationships are modified during infection though , with two main consequences that we have reported . First , some regulatory relationships are more evident in vivo than in vitro , and the in vivo relationships can reveal pathways that are relevant to pathogenicity . Our analysis of SUT1 , ZAP1 , and ZRT2 stands as an example . Second , the response to drug treatment manifested during infection can be distinct from that detected in vitro , as revealed by our analysis of caspofungin treatment . The drug response during infection includes induction of many infection-repressed genes , a relationship that may contribute to drug efficacy . Casadevall and colleagues [40] have argued that virulence is an emergent property , a unique state manifested only when host and pathogen interact . Our results provide support for this concept by illustrating that unique regulatory relationships emerge in the environment of invasive infection .
All animal procedures were approved by the Institutional Animal Care and Use Committee at the Los Angeles Biomedical Research Institute ( protocol 011000 ) and carried out according to the National Institutes of Health ( NIH ) guidelines for the ethical treatment of animals . The mice were caged in an AAALAC-accredited facility located on the campus of Harbor-UCLA Research and Education Institute . A full-time veterinarian who specializes in laboratory animal medicine oversaw their care . Caging and husbandry was provided according to the guidelines in the United States Public Health Service publication Guide for the Care and Use of Laboratory Animals . Every attempt was made to treat the mice humanely . The survival and health of the mice was monitored three times daily . Obviously sick , lethargic mice were segregated from the group and euthanized to minimize suffering . The mice were euthanized by pentobarbital overdose ( 210 mg/kg ) , as recommended by the Panel on Euthanasia of the American Veterinary Medical Association . C . albicans strains were grown at 30°C in YPD ( 2% Bacto peptone , 2% dextrose , 1% yeast extract ) or 37°C in RPMI-1640 ( with L-glutamine and 0 . 165M MOPS , without sodium bicarbonate ) , both with shaking at 200 rpm . Transformants were selected on synthetic medium ( 2% dextrose , 1 . 7% Difco yeast nitrogen base with ammonium sulfate and auxotrophic supplements ) or on YPD+clonNAT400 ( 2% Bacto peptone , 2% dextrose , 1% yeast extract , and 400 μg/ml nourseothricin [clonNAT , WERNER BioAgents] ) for nourseothricin-resistant isolates . Growth on low-zinc medium [41] was assayed with synthetic medium lacking added zinc ( 2% dextrose , 1 . 7% yeast nitrogen base without ammonium sulfate and without zinc sulfate , 0 . 2% ammonium sulfate , 2 . 5 μM EDTA , and auxotrophic supplements ) . The C . albicans IRO1 gene was polymerase chain reaction ( PCR ) amplified from strain SC5314 genomic DNA using primers pSG1 Nde1 IRO1 18 F and pSG1 Nde1 IRO1 257 , which have flanking homologous sequences to plasmid pSG1 , a derivative of pDDB78 [42] from which the S . cerevisiae TRP1 marker had been deleted . The PCR product was co-transformed with NdeI digested pSG1 into S . cerevisiae for homologous recombination . The resultant plasmid was isolated from S . cerevisiae and transformed into Escherichia coli for amplification . The plasmid pSG1 IRO1 can be linearized by Afe1 digestion and then transformed into a His- C . albicans strain to direct integration at the IRO1 locus , making the strain His+ IRO1 . C . albicans deletion mutant strains were constructed in the BWP17 strain background through homologous recombination [43] . Briefly , gene disruption PCR products were synthesized using the plasmids pRS-ARG4 or pGEM-URA3 as templates . The primers were designed to include homology to the sequence immediately upstream of the start codon or the sequence immediately following the stop codon of the target gene . Arg+ Ura+ homozygous deletion strains were then made His+ along with restoring IRO1 by integration of linearized plasmid pSG1-IRO1 at the HIS1 locus . For complementation of deletion strains , PCR primers were designed to amplify genomic DNA of strain SC5314 from 1 kb upstream to 0 . 5 kb downstream of the open reading frame of specific genes . Shorter distances were used when there were additional genes located within this region . These primers have 5′ flanking sequences with homology to pSG1 IRO1 . The resulting PCR product was co-transformed into S . cerevisiae with Not1 and Sac1 digested pSG1 IRO1 . Plasmid DNA was isolated , transformed into E . coli , and isolated plasmid DNA was digested with AfeI and transformed into the respective C . albicans mutant strains . The complementing cassette was targeted to the IRO1 locus and these complemented strains were then also His+ IRO1 . Where indicated , we used the prototrophic reference strain CW696 as a control . It is derived from the SC5314 background [44] strain DAY286 [45] , which in turn derives from strain BWP17 [43] , as do all of the deletion mutants studied here . Strain DAY286 was transformed with Afe1-digested plasmid pSG1 IRO1 to restore both HIS1 and IRO1 . His+ transformants were selected on CSM-his . We verified that CW696 expresses IRO1 through nanoString assays , and confirmed that its virulence is comparable to that of strain SC5314 , based on median survival assays in the mouse hematogenously disseminated infection model . For ZAP1 and ZRT2 overexpression strains , PCR primers were designed to amplify the NAT1-pTDH3 cassette from plasmid CJN542 [15] . These primers were designed to include homology to the sequence immediately upstream of the start codon or the sequence immediately following the stop codon of the ZAP1/ZRT2 gene . The PCR products were transformed into the respective C . albicans mutant strains , replacing one allele of ZAP1/ZRT2 promoter with the TDH3 promoter by homologous recombination . C . albicans strains used in this study are listed in S1 Table . PCR primers used in this study are listed in S2 Table . Male Balb/c mice weighing 20–22g ( Taconic Farms ) were used for all studies . In the survival studies , five mice per strain of C . albicans were inoculated intravenously with 5 × 105 yeast-phase cells that had been grown in YPD at 30°C to saturation as described previously [46] . A portion of each inoculum was plated on a YPD plate , and CFU was measured after 2 d of growth to check viability of cells in the inoculum . The animals were monitored three times daily for 21 d , and moribund mice were humanely euthanized . The results of the survival experiments were analyzed with the Log-Rank Test . For isolation of host and fungal RNA and histopathology , three mice per strain were inoculated intravenously with 1 × 106 yeast-phase organisms , and the kidneys were harvested at specific time points . The right kidney was snap frozen in liquid nitrogen and stored at −80°C for later RNA extraction . The left kidney ways fixed in zinc-buffered formalin , embedded in paraffin , sectioned , and stained with Periodic acid-Schiff . All mouse experiments were approved by the Animal Care and Use Committee at the Los Angeles Biomedical Research Institute ( protocol 011000 ) and carried out according to the National Institutes of Health ( NIH ) guidelines for the ethical treatment of animals . RNA isolations were performed using Qiagen RNeasy mini kit ( Cat#74104 ) with modifications described here . Approximately 1 . 5 ml of buffer RLT with 1% β-ME was added to each kidney immediately before homogenization using gentelMACS dissociator ( Miltenyi Biotec ) on pre-loaded setting RNA_02 . 01 . The M tube ( Miltenyi Biotec ) was centrifuged at 2 , 000 rpm for 1 min in Eppendorf tabletop centrifuge 5810R at room temp . 600 μl homogenate was transferred to a fresh 2 ml screw-cap tube and mixed with 600 μl Phenol:Chloroform:Isoamyl Alcohol 25:24:1 and 300 μl of Zirconia beads ( Ambion ) , vortexed on a mini-beadbeater ( Biospec Products ) for 3 min , and centrifuged at 14 , 000 rpm for 5 min in a 4°C cold room . The aqueous phase was transferred to a new 1 . 5 ml microfuge tube , mixed well with equal volume of 70% ethanol , and loaded onto the RNeasy spin column . The washing and RNA elution steps were carried out following Qiagen Quick-Start Protocol for RNeasy Mini Kit ( Cat#74104 ) . For caspofungin treatment in vivo , caspofungin was administrated at a dose of 1 mg/kg ( approximately 20 ug per mouse ) intraperitoneally at 24 hr postinfection . The control mice received a similar volume of PBS intraperitoneally . The mice were killed 2 hr after drug administration and the kidneys were collected for RNA isolation . For caspofungin treatment in vitro , an overnight YPD 30°C culture of wild-type SC5314 was diluted 1:1 , 000 into 50 ml RPMI cultures and grown at 37°C with 200 rpm shaking for 4 hr . Caspofungin was then added to a final concentration of 100 ng/ml , and the control cultures received equal amount of water . Cells were collected by filtration 2 hr after drug administration . Because nanoString technology is not genome-wide , we developed strategies to select high-priority fungal and host genes for investigation . First , we selected 248 C . albicans genes that play important roles in environmental responses , such as iron/zinc acquisition , oxidative/nitrosative stress , pH response , hyphal growth , and general stress responses . We reasoned that these environmental response gene probes ( ER codeset ) should give us clues about what environmental challenges Candida cells face during infection and which pathways are employed to cope with such challenges . Secondly , we designed a codeset to include 231 Candida genes that specify known and predicted transcription factors ( TF codeset ) . Finally , we selected 46 mouse genes that are critical in host responses ( HR codeset ) to microbial infection , including genes encoding for fungal pattern recognition receptors , chemokines and chemokine receptors , interleukins and interleukin receptors , interferons and antimicrobial peptides . All the Candida gene codesets also include control genes TDH3 and YRA1 , and the mouse gene codeset includes control genes ACTB , GAPDH , and PPIA . The complete list of genes for all codesets is shown in S1 , S3 , and S4 Data . For each nanoString assay , 10 μg of total tissue RNA isolated from a mouse kidney or 100 ng of pure Candida RNA isolated from an in vitro culture was mixed with a nanoString codeset mix and incubated at 65°C overnight ( 16–18 hr ) . The reaction mixes were loaded on the nanoString nCounter Prep Station for binding and washing , and the resultant cartridge was transferred to the nanoString nCounter digital analyzer for scanning and data collection . A total of 600 fields were captured per sample . The raw data were first adjusted for binding efficiency and background subtraction , following nCounter data analysis guidelines . The total adjusted counts ( i . e . , before normalization using the internal control genes ) for genes on the same codeset were used to estimate relative fungal RNA abundance in each sample . For example , 10 ug RNA isolated from kidney 24 hr postinfection by wild-type strain SC5314 generated 324 , 648 total counts for 150 genes ( mean of six determinations ) , while that from rim101Δ/Δ infected kidney generated 49 , 083 counts ( mean of three determinations ) . We estimated that at 24 hr postinfection , the relative fungal RNA abundance for rim101Δ/Δ infected kidney is roughly 15% ( 49 , 083/324 , 648 ) of that of the wild type . Following the same logic , we estimated that fungal RNA consists approximately 0 . 1% of total RNA isolated from kidney infected by the wild-type strain SC5314 at the 24 hr time point ( by comparing to total counts from in vitro samples ) . To calculate gene expression ratios among different samples , we normalized adjusted raw counts using internal control genes . For the ER codeset , data were normalized using the control gene TDH3 . For the TF codeset , data were normalized by total counts of 231 genes . We tested a number of methods for normalization , including using another control gene YRA1 , using total counts for all genes , and using geometric mean of robustly expressed genes . We found that the normalization factors resulting from these methods are similar and will not affect our main conclusions should we use a different method for normalization . For the HR codeset , data were normalized using the geometric mean of three internal control genes: ACTB , GAPDH , and PPIA . All expression ratios were calculated using mean values of three independent biological samples , and statistical significance was determined by two-tailed Student’s t-test ( n = 3 , p < 0 . 05 , unless specified otherwise ) . The heat maps were generated using Multiexperimental Viewer 4 . 9 . 0 . Dataset comparisons were carried out with one-sided Fisher's Exact Test , using query gene sets from in vivo nanoString data described here ( applying cut-off at 2X , 4X or 10X changes ) and the entire set of nanoString probes as a background set . Queries were matched to a database we assembled of 166 published expression datasets , as well as nanoString datasets we generated during this study . Quantitative reverse transcription PCR reactions were carried out as previously described [47] . Briefly , 10 μg total RNA was treated with the DNA-free kit ( Ambion ) followed by first-strand cDNA synthesis from half of the DNA-free RNA using the AffinityScript multiple temperature cDNA synthesis kit ( Stratagene ) . Absence of DNA contamination was confirmed using control sets for which reverse transcriptase was omitted from the cDNA reaction . Primers were designed to amplify a 150–200 bp region for target genes including TDH3 , which was used as a reference gene for normalization . 2X iQ SYBR Green Supermix ( Bio-Rad ) , 1 μl of first-strand cDNA reaction mixture , and 0 . 1 μM of primers were mixed in a total volume of 50μl per reaction . Real-time PCR was performed in triplicate using an iCycler iQ real-time PCR detection system ( Bio-Rad ) . The program for amplification had an initial denaturation step at 95°C for 5 min , followed by 40 cycles of 95°C for 45 s and 58°C for 30 s . Product amplification was detected using SYBR Green fluorescence during the 58°C step , and specificity of the reaction was monitored by melt-curve analysis following the real-time program . Gene expression was determined using Bio-Rad iQ5 software ( ΔΔCT method ) . | We have a limited understanding of how the expression of pathogens’ genes changes during infection of humans or other animal hosts , in contrast to in vitro models of infection . Here we profile the alteration in gene expression over time as a predictor of functional consequences during invasive growth of Candida in the kidney; a situation in which the limited number of pathogen cells makes gene expression challenging to assay . Our findings reveal that there are distinct early and late phases of infection , and identify new genes that govern the early zinc acquisition response necessary for proliferation in vivo—and thus required for infection . We also find that the response to drug treatment that manifests during infection can be distinct from that detected in vitro . We show that a well-known gene expression response to the antifungal drug caspofungin is naturally down-regulated in infecting cells , suggesting that the efficacy of the drug may be enhanced by a susceptible state of the pathogen during invasive proliferation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Activation and Alliance of Regulatory Pathways in C. albicans during Mammalian Infection |
Shuni virus ( SHUV ) is an orthobunyavirus that belongs to the Simbu serogroup . SHUV was isolated from diverse species of domesticated animals and wildlife , and is associated with neurological disease , abortions , and congenital malformations . Recently , SHUV caused outbreaks among ruminants in Israel , representing the first incursions outside the African continent . The isolation of SHUV from a febrile child in Nigeria and seroprevalence among veterinarians in South Africa suggests that the virus may have zoonotic potential as well . The high pathogenicity , extremely broad tropism , potential transmission via both biting midges and mosquitoes , and zoonotic features warrants prioritization of SHUV for further research . Additional knowledge is essential to accurately determine the risk for animal and human health , and to assess the risk of future epizootics and epidemics . To gain first insights into the potential involvement of arthropod vectors in SHUV transmission , we have investigated the ability of SHUV to infect and disseminate in laboratory-reared biting midges and mosquitoes . Culicoides nubeculosus , C . sonorensis , Culex pipiens pipiens , and Aedes aegypti were orally exposed to SHUV by providing an infectious blood meal . Biting midges showed high infection rates of approximately 40–60% , whereas infection rates of mosquitoes were lower than 2% . SHUV successfully disseminated in both species of biting midges , but no evidence of transmission in orally exposed mosquitoes was found . The results of this study show that different species of Culicoides biting midges are susceptible to infection and dissemination of SHUV , whereas the two mosquito species tested were found not to be susceptible .
Arthropod-borne ( arbo ) viruses continue to pose a threat to human and animal health [1 , 2] . In particular the order Bunyavirales comprises emerging pathogens such as Crimean-Congo haemorrhagic fever virus ( CCHFV ) and Rift Valley fever virus ( RVFV ) [3 , 4] . The World Health Organization ( WHO ) has included both CCHFV and RVFV to the “Blueprint” list of ten prioritized viruses likely to cause future epidemics and for which insufficient countermeasures are available [5] . In the veterinary field , prioritized viral diseases of animals , including RVFV , are notifiable to the World Organization for Animal Health ( Office International des Epizooties , OIE ) . Apart from pathogens that are recognised as major threats by WHO and OIE , many have remained largely neglected . Before the turn of the century , West Nile virus , chikungunya virus , and Zika virus were among these neglected viruses until they reminded us how fast arboviruses can spread in immunologically naïve populations [2] . Although these outbreaks came as a surprise , in hindsight , smaller outbreaks in previously unaffected areas could have been recognised as warning signs . Shuni virus ( SHUV; family Peribunyaviridae , genus Orthobunyavirus , Simbu serogroup ) recently emerged in two very distant areas of the world [6] . SHUV was isolated for the first time from a slaughtered cow in the 1960s in Nigeria [7] . During subsequent years , the virus was isolated on several occasions from domestic animals including cattle , sheep , goats , and horses [7–10] , from wild animals including crocodiles and rhinoceros [10] , and from field-collected Culicoides biting midges and mosquitoes [8 , 11 , 12] . More recently , SHUV was associated with malformed ruminants in Israel [13 , 14] . Emergence of SHUV in areas outside Sub-Saharan Africa shows the potential of this virus to spread to new areas , and increases the risk for SHUV outbreaks in bordering territories such as Europe . Isolation of SHUV from a febrile child and detection of antibodies in 3 . 9% of serum samples from veterinarians in South Africa shows that SHUV can infect humans as well , although its ability to cause human disease is still uncertain [7 , 15 , 16] . Proper risk assessments rely on accurate knowledge of disease transmission cycles . Arbovirus transmission cycles can only become established when competent vectors and susceptible hosts encounter under suitable climatic conditions . Although SHUV has been isolated from pools of field-collected Culicoides biting midges and mosquitoes [7 , 11 , 12] , the role of both insect groups as actual vectors remains to be confirmed . Detection of virus in field-collected insects is not sufficient to prove their ability to transmit the virus . Arboviruses need to overcome several barriers ( i . e . midgut and salivary gland barriers ) inside their vector , before they can be transmitted [17 , 18] . In addition to virus isolation from field-collected vectors , laboratory studies are therefore needed to experimentally test the ability of blood-feeding insects to become infected with , maintain , and successfully transmit arboviruses ( i . e . , vector competence ) [19] . To gain insights into the potential of Culicoides biting midges and mosquitoes to function as vectors of SHUV , we studied the susceptibility of four main arbovirus vector species ( Culicoides nubeculosus and C . sonorensis biting midges , and Culex pipiens biotype pipiens and Aedes aegypti mosquitoes ) for SHUV .
African green monkey kidney cells ( Vero E6; ATCC CRL-1586 ) were cultured in Eagle’s minimum essential medium ( Gibco , Carlsbad , CA , United States ) supplemented with 5% fetal bovine serum ( FBS; Gibco ) , 1% non-essential amino acids ( Gibco ) , 1% L-glutamine ( Gibco ) , and 1% antibiotic/antimycotic ( Gibco ) . Cells were cultured as monolayers and maintained at 37°C with 5% CO2 . Vero E6 cells that were used in biting midge and mosquito infection experiments in the biosafety level 3 ( BSL3 ) facility were cultured in Dulbecco's modified Eagle medium ( Gibco ) supplemented with 10% FBS , penicillin ( 100 U/ml; Sigma-Aldrich , Saint Louis , MO , United States ) , and streptomycin ( 100 μg/ml; Sigma-Aldrich ) . Prior to infections in the BSL3 facility , Vero E6 cells were seeded in 4- ( 2-hydroxyethyl ) -1-piperazineethanesulfonic acid-buffered DMEM medium ( HEPES-DMEM; Gibco ) supplemented with 10% FBS , penicillin ( 100 U/ml ) , and streptomycin ( 100 μg/ml ) , fungizone ( 50 μg/ml; Invitrogen , Carlsbad , United States ) , and gentamycin ( 50 μg/ml; Gibco ) . C6/36 cells ( ATCC CRL-1660 ) , derived from Ae . albopictus mosquitoes , were cultured in Leibovitz-15 ( L-15 ) growth medium ( Sigma-Aldrich ) supplemented with 10% FBS , 2% Tryptose Phosphate Broth ( Gibco ) , 1% non-essential amino acids solution , and 1% antibiotic/antimycotic . Cells were cultured as monolayers and incubated at 28°C in absence of CO2 . KC cells , derived from embryos of colonized C . sonorensis biting midges [20] , were cultured as monolayers in modified Schneider’s Drosophila medium ( Lonza , Basel , Switzerland ) with 15% FBS , and 1% antibiotic/antimycotic at 28°C in absence of CO2 . SHUV ( strain An10107 , P2 Vero , 1980 ) was kindly provided by the World Reference Center for Emerging Viruses and Arboviruses ( WRCEVA ) . The virus was originally isolated from the blood of a slaughtered cow in 1966 in Nigeria by inoculation of neonatal mice , and passaged twice in Vero cells [21] . The passage 3 ( P3 ) stock was generated by inoculation of Vero E6 cells with the P2 stock at a multiplicity of infection ( MOI ) of 0 . 001 . The supernatant was harvested at 6 days post inoculation , centrifuged , and stored in aliquots at -80°C . The P4 stock was generated by inoculating Vero E6 cells at MOI 0 . 01 using the P3 stock . At this MOI , full cytopathic effect ( CPE ) was present at 3 days post infection . Virus titers were determined using endpoint dilution assays ( EPDA ) on Vero E6 cells [22] . Titers were calculated using the Spearman-Kärber algorithm and expressed as 50% tissue culture infective dose ( TCID50 ) [23 , 24] . The virus detection and titration procedure was validated using a SHUV-specific reverse transcriptase quantitative PCR ( RT-qPCR; S1 Supporting Information ) . Cells were seeded in T25 cell culture flasks at densities of 7 . 5 × 105 ( Vero E6 ) , 1 . 5 × 106 ( C6/36 ) , or 2 . 5 × 106 ( KC cells ) per flask in 10 ml complete medium . After overnight incubation , the flasks were inoculated with SHUV at an MOI of 0 . 01 ( P4 stock ) . The MOI calculation for each cell line was based on the virus titer that was determined on Vero E6 cells . One hour after inoculation , the medium was removed and replaced with fresh medium . At time points 0 ( sample taken directly after medium replacement ) , 24 , 48 , and 72 h post infection , 200 μl samples were taken and stored at -80°C for later analysis . For each cell line , virus titers were determined in triplicate per time point by EPDA using Vero E6 cells , which showed distinct CPE [22] . Culicoides nubeculosus were kindly provided by The Pirbright Institute , Pirbright laboratories , United Kingdom , in 2012 [25] , and were maintained at 23°C with 16:8 light:dark cycle and 60% relative humidity . Culicoides sonorensis were kindly provided by the Arthropod-Borne Animal Diseases Research Laboratory , USDA-ARS ( courtesy of Dr . Barbara Drolet ) in 2017 [26] , and were maintained at 25°C with 16:8 light:dark cycle and 70% relative humidity . Similar rearing protocols were used for both biting midge species . Eggs were transferred to square larval holding trays ( C . nubeculosus: 25 x 25 x 8 cm , Kartell , Noviglio , Italy; C . sonorensis: 19 x 19 x 20 cm , Jokey , Wipperfürth , Germany ) with filter wool ( Europet Bernina International , Gemert-Bakel , The Netherlands ) attached with double-sided tape to the bottom . Trays were filled with tap water , a few millilitres of rearing water in which larvae had completed their life cycle , and two drops of Liquifry No . 1 ( Interpet , Dorking , United Kingdom ) . Larvae were fed with a 1:1:1 mixture of bovine liver powder ( MP biomedicals , Irvine , CA , US ) , ground rabbit food ( Pets Place , Ede , The Netherlands ) , and ground koi food ( Tetra , Melle , Germany ) . Culicoides nubeculosus larvae were additionally fed with nutrient broth No . 2 ( Oxoid , Hampshire , UK ) . Pupae were transferred to plastic buckets ( diameter: 12 . 2 cm , height: 12 . 2 cm; Jokey ) and closed with netting on the top through which the biting midges could feed . Emerged adults were provided with 6% glucose solution ad libitum . Cow blood ( Carus , Wageningen , The Netherlands ) was provided through a Parafilm M membrane using the Hemotek PS5 feeding system ( Discovery Workshops , Lancashire , United Kingdom ) for egg production . The Cx . pipiens pipiens colony was established in the laboratory from egg rafts collected in the field in The Netherlands during August 2016 . Egg rafts were individually hatched in tubes . Pools of approximately 10 first instar larvae were identified to the biotype level using real-time PCR [27] . The colony was started by grouping larvae from 93 egg rafts identified as the pipiens biotype . Mosquitoes were maintained at 23°C with 16:8 light:dark cycle and 60% relative humidity [28 , 29] . Adult mosquitoes were kept in Bugdorm-1 rearing cages and maintained on 6% glucose solution ad libitum . Cow blood or chicken blood ( Kemperkip , Uden , The Netherlands ) was collected in BC Vacutainer lithium heparin-coated blood collection tubes ( Becton Dickinson , Breda , The Netherlands ) , and stored at 4°C . Blood was provided through a Parafilm M membrane using the Hemotek PS5 feeding system for egg production . Egg rafts were transferred to square larval holding trays ( 25 x 25 x 8 cm , Kartell ) filled with tap water and two drops of Liquifry No . 1 . Hatched larvae were fed with a 1:1:1 mixture of bovine liver powder , ground rabbit food , and ground koi food . Pupae were collected every 2 days and placed in Bugdorm-1 insect rearing cages . Aedes aegypti mosquitoes from the Rockefeller strain ( Bayer AG , Monheim , Germany ) were used in all experiments . The mosquito colony was maintained as described before [30] . In short , mosquitoes were maintained at 27°C with 12:12 light:dark cycle and 70% relative humidity . Adult mosquitoes were kept in Bugdorm-1 rearing cages and maintained on 6% glucose solution ad libitum . Human blood ( Sanquin Blood Supply Foundation , Nijmegen , The Netherlands ) was provided through a Parafilm M membrane using the Hemotek PS5 feeding system for egg production . Eggs were transferred to transparent square larval holding trays ( 19 x 19 x 20 cm , Jokey ) , filled for approximately one-third with tap water and three drops of Liquifry No . 1 . Hatched larvae were fed with Tetramin Baby fish food ( Tetra ) . Larval trays were closed with fine-meshed netting , to allow adult mosquitoes to emerge inside larval trays . Twice a week , adults were aspirated from the larval trays and collected in Bugdorm-1 insect rearing cages . Groups of adult C . nubeculosus ( 1–7 days old ) , C . sonorensis ( 1–11 days old ) , Cx . p . pipiens ( 4–20 days old ) , and Ae . aegypti ( 4–7 days old ) were transferred to plastic buckets ( diameter: 12 . 2 cm , height: 12 . 2 cm; Jokey ) and closed with netting before being taken to the BSL3 facility . Culex p . pipiens mosquitoes were kept on water for 3 days , whereas the other species were maintained on 6% glucose solution until being offered an infectious blood meal . SHUV P3 stock with a mean titer of 3 . 0 x 106 TCID50/ml was mixed 1:1 with cow blood . The used cow blood was tested negative for Schmallenberg virus ( SBV ) antibodies , to prevent cross-neutralisation with SHUV . The infectious blood meal was provided through a Parafilm M membrane using the Hemotek PS5 feeding system , under dark conditions at 24°C and 70% relative humidity . After 1 h , insects were anesthetized with 100% CO2 and kept on a CO2-pad to select fully engorged females . For each species , five fully engorged females were directly stored at -80°C for each replicate . These samples were used to determine the ingested amounts of SHUV for each species . All remaining and fully engorged females were placed back into buckets with a maximum group size of 110 individuals per species per bucket . All insects were provided with 6% glucose solution via a soaked ball of cotton wool on top of the netting ad libitum . Culicoides sonorensis and Ae . aegypti were kept at 28°C for 10 days , whereas C . nubeculosus and Cx . p . pipiens were kept at 25°C for 10 days . These temperatures were selected for optimal replication of the virus , and to reflect differences in the rearing temperature for each species . Three replicate experiments of C . nubeculosus ( N1 = 84 , N2 = 82 , N3 = 77 , Ntotal = 243 ) , C . sonorensis ( N1 = 9 , N2 = 9 , N3 = 30 , Ntotal = 48 ) , and Cx . p . pipiens ( N1 = 89 , N2 = 57 , N3 = 65 , Ntotal = 211 ) were carried out , and two replicate experiments of Ae . aegypti ( N1 = 72 , N2 = 77 , Ntotal = 149 ) . During each replicate , biting midges and mosquitoes were fed in parallel with the same infectious blood meal . Adult female Cx . p . pipiens ( 3–9 days old ) and Ae . aegypti ( 4–6 days old ) mosquitoes were injected with SHUV into the thorax to investigate the role of mosquito barriers on dissemination of SHUV . Mosquitoes were anesthetized with 100% CO2 and positioned on the CO2-pad . Female mosquitoes were intrathoracically injected with 69 nl of SHUV ( P3 stock with a titer of 3 . 0 x 106 TCID50/ml ) using a Drummond Nanoject II Auto-Nanoliter injector ( Drummond Scientific , Broomall , Unites States ) . Injected Cx . p . pipiens were maintained at 25°C and injected Ae . aegypti were maintained at 28°C . Mosquitoes were incubated for 10 days at the respective temperatures , and had access to 6% glucose solution ad libitum . Injections were done during a single replicate experiment for Cx . p . pipiens ( N = 50 ) and Ae . aegypti ( N = 50 ) . After 10 days of incubation at the respective incubation temperatures , samples from surviving biting midges and mosquitoes were collected . Biting midges were anesthetized with 100% CO2 and transferred individually to 1 . 5 ml Safe-Seal micro tubes ( Sarstedt , Nümbrecht , Germany ) containing 0 . 5 mm zirconium beads ( Next Advance , Averill Park , NY , United States ) . For a selection of C . nubeculosus ( N = 77 ) and C . sonorensis ( N = 30 ) from one replicate experiment , heads were removed from bodies and separately stored in tubes . All samples were stored at -80°C until further processing . Mosquitoes were anesthetized with 100% CO2 to remove legs and wings . Mosquito saliva was then collected by inserting the proboscis into a 200 μl yellow pipette tip ( Greiner Bio-One ) containing 5 μl of a 1:1 solution of 50% glucose solution and FBS . The saliva sample was transferred to a 1 . 5 ml micro tube containing 55 μl of fully supplemented HEPES-DMEM medium . Mosquito bodies were individually stored in 1 . 5 ml Safe-Seal micro tubes containing 0 . 5 mm zirconium beads . Frozen biting midge and mosquito tissues were homogenized for 2 min at maximum speed ( setting 10 ) in the Bullet Blender Storm ( Next advance ) , centrifuged for 30 seconds at 14 , 500 rpm in the Eppendorf minispin plus ( Eppendorf , Hamburg , Germany ) , and suspended in 100 μl of fully supplemented HEPES-DMEM medium . After addition of the medium , samples were blended again for 2 min at maximum speed , and centrifuged for 2 min at 14 , 500 rpm . Mosquito saliva samples were thawed at RT and vortexed before further use . In total 30 μl of each body or saliva sample was inoculated on a monolayer of Vero E6 cells in a 96 wells plate . SHUV stock or infectious blood mixture was included as positive control and wells to which no sample was added were included as negative controls . After 2–3 h the inoculum was removed and replaced by 100 μl of fully supplemented HEPES-DMEM medium . Wells were scored for virus induced CPE at 3 and 7 days post inoculation , with full CPE being observed at the latter time point . Afterwards , virus titers for positive samples of biting midge bodies and heads , as well as mosquito bodies and saliva were determined with single EPDA on Vero E6 cells [30] . Virus titers were determined using the Reed & Muench algorithm [31] . A subset of samples was validated by RT-qPCR , to confirm that observed CPE was induced by SHUV ( S1 Supporting Information ) . Infection rate ( virus-infected whole body ) and dissemination efficiency ( virus-infected head ) were determined for biting midges , whereas infection rate ( virus-infected whole body ) and transmission efficiency ( virus-infected saliva ) were determined for mosquitoes . Infection rate , dissemination efficiency , and transmission efficiency were calculated , respectively , by dividing the number of females with virus-infected bodies ( infection ) , virus-infected heads ( dissemination ) , or virus-infected saliva ( transmission ) by the total number of females tested in the respective treatment and that survived the incubation period . The values were subsequently expressed as percentages by multiplying with 100 . Two biting midge samples of which only the head was virus-positive , but not the body , were considered to be uninfected .
Mammalian , mosquito , and midge cells were inoculated with SHUV to gain insight into the replicative fitness of this virus and strain in different host cell types . The results show that SHUV is capable to produce progeny in all three cell types ( Fig 1 and S1 Data ) . Of note , a strong CPE was observed in the Vero E6 cells upon infection whereas no CPE was observed in the insect cell lines . Therefore , Vero E6 cells were used to determine titers by EPDA . To evaluate the susceptibility of two species of biting midges ( C . nubeculosus and C . sonorensis ) for SHUV , groups of individuals of both species were orally exposed to an infectious blood meal with a mean SHUV titer of 3 . 0 x 106 TCID50/ml . SHUV titers of ingested blood were determined for a selection of 10 fully engorged females for each species , that were directly stored at -80°C after feeding . Both species ingested low amounts of SHUV that were below the detection limit of the endpoint dilution assay of 103 TCID50/ml . Infection rates were also determined after 10 days of incubation at temperatures of 25°C ( C . nubeculosus and Cx . p . pipiens ) or 28°C ( C . sonorensis and Ae . aegypti; Fig 2 and S2 Data ) . Both biting midge species showed high infection rates of 44% for C . nubeculosus ( N = 243 ) , and 60% for C . sonorensis ( N = 48; Fig 2A ) . SHUV replicated to median titers of 2 . 4 x 103 TCID50/ml in body samples of C . nubeculosus and 1 . 1 x 104 TCID50/ml in body samples of C . sonorensis ( Fig 2E ) . For one replicate experiment , heads were separated from the bodies and tested for presence of SHUV to assess whether the virus successfully passed from the midgut to the haemocoel , indicative of dissemination throughout the body . Dissemination efficiencies were 18% ( N = 77 ) for C . nubeculosus and 10% ( N = 30 ) for C . sonorensis ( Fig 2C ) . In all virus-positive heads that induced CPE , SHUV titers were lower than 103 TCID50/ml . Because only very low amounts of SHUV were detected in biting midge heads , the actual percentage of disseminated infections might be higher . A subset of the samples was additionally tested by RT-qPCR to confirm that CPE was induced by SHUV ( S1 Supporting Information ) . The relatively high infection rates and dissemination efficiencies observed in this study and the absence of a salivary glands barrier in biting midges as shown in previous studies [17 , 32] , suggests that both C . nubeculosus and C . sonorensis have the potential to transmit SHUV . SHUV was previously isolated from field-collected mosquitoes [8] . Therefore , we determined vector competence for two mosquito species ( Cx . p . pipiens and Ae . aegypti ) which are important vectors for several arboviruses [22 , 28 , 30] . SHUV titers of ingested blood were determined for a selection of 10 fully engorged female mosquitoes that were directly stored at -80°C after feeding on an infectious blood meal with a SHUV titer of 3 . 0 x 106 TCID50/ml . Similar to results obtained with the biting midges , the amounts of SHUV ingested by both mosquito species was less than 103 TCID50/ml . No SHUV infection was observed in the Cx . p . pipiens mosquitoes ( N = 211 ) following oral exposure , whereas infection rates of 2% were found for orally exposed Ae . aegypti mosquitoes ( N = 149; Fig 2B ) . SHUV replicated to median titers of 6 . 3 x 103 TCID50/ml in body samples of Ae . aegypti ( Fig 2F ) , which was comparable to titers found in biting midges . No SHUV was detected in any of the saliva samples taken from either Cx . p . pipiens or Ae . aegypti ( Fig 2D ) . Thus , SHUV was able to successfully infect a small proportion of Ae . aegypti mosquitoes but not Cx . p . pipiens , and no evidence was found for transmission of SHUV by mosquitoes . The very low infection rates of mosquitoes triggered further investigation into potential mosquito barriers against SHUV infection . To this end , Cx . p . pipiens and Ae . aegypti mosquitoes were intrathoracically injected with SHUV , to bypass the potential midgut barrier . Direct injection of SHUV into the thorax resulted in high infection rates of 70% for Cx . p . pipiens ( N = 50 ) , and 100% for Ae . aegypti ( N = 50; Fig 3A ) . Transmission efficiency of 32% ( N = 50 ) was found for Cx . p . pipiens and 8% ( N = 50 ) for Ae . aegypti ( Fig 3B ) . Interestingly , although infection rates of Cx . p . pipiens were below 100% , we found a relatively high transmission efficiency . This may indicate a relatively weaker salivary gland barrier in Cx . p . pipiens compared to Ae . aegypti mosquitoes that had 100% infection rate , but relatively low transmission efficiency . To gain more insight in replication of SHUV , virus titers were determined for virus-infected mosquito body and saliva samples . Titers of virus-infected Cx . p . pipiens body samples were almost all below the detection limit of 103 TCID50/ml of the endpoint dilution assay ( Fig 3C ) . This indicates that even when SHUV is injected into the thorax , there is no productive virus replication . In contrast , we found median titers of 7 . 1 x 104 TCID50/ml for virus-infected Ae . aegypti body samples . This shows that SHUV is able to successfully replicate in Ae . aegypti when the midgut barrier is bypassed . In the majority of mosquito saliva samples , SHUV titers were less than 103 TCID50/ml ( Fig 3D ) . Taken together , SHUV is able to disseminate in mosquitoes , but both the midgut and salivary glands form a barrier for SHUV .
SHUV was previously isolated from field-collected pools of Culicoides biting midges and from mosquitoes , but their involvement in SHUV transmission remained to be confirmed [8 , 11 , 12] . Here , we show for the first time that SHUV is able to infect and replicate in biting midges as well as in mosquitoes , but only the biting midge species evaluated in the present study can be considered highly susceptible to infection . Both C . nubeculosus and C . sonorensis showed high infection rates of 44% and 60% when incubated for 10 days at 25°C and 28°C , respectively . It has been demonstrated that a salivary gland barrier is absent for Orbiviruses and Schmallenberg virus in biting midges [17 , 32] . This knowledge , in combination with evidence of successful dissemination of SHUV to the heads indicates that the biting midge species evaluated in the present study are likely competent vectors of SHUV . Importantly , the finding that SHUV replicates efficiently in two biting midge species from a different geographic background suggests that various species of Culicoides may function as vectors of SHUV . SHUV infection and replication in biting midges seems more efficient compared to other biting midge-borne viruses such as SBV and bluetongue virus ( BTV ) , which generally show infection rates up to 30% [32–36] . Both SBV and BTV have caused sudden and large-scale epizootics in Europe , with devastating consequences for the livestock sector [37 , 38] . The relatively high susceptibility and efficiency of replication in biting midges , and recent spread of SHUV to areas outside Sub-Saharan Africa [13] , should therefore be interpreted as a warning for its epizootic potential . In contrast to the high infection rates in biting midges , only few orally exposed Ae . aegypti mosquitoes became infected with SHUV during 10 days of incubation at 28°C . In addition , no evidence of successful dissemination to the salivary glands of the two mosquito species was found . SHUV replication and transmission ( 8% ) was observed when the virus was directly injected into the thorax of Ae . aegypti mosquitoes . This indicates that both the midgut infection barrier and the salivary gland barrier prevent infection and subsequent transmission of SHUV by Ae . aegypti mosquitoes . Of the Cx . p . pipiens mosquitoes that were orally exposed to SHUV , none became infected during 10 days of incubation at 25°C . Moreover , replication of SHUV was low in Cx . p . pipiens , as evidenced by low titers when it was directly injected into the thorax . However , a relatively high percentage of mosquito saliva samples contained SHUV . We therefore conclude that the midgut barrier is the main barrier that prevents infection of Cx . p . pipiens with SHUV . Our findings are in line with an earlier study on the closely-related SBV , which showed no evidence for involvement of Cx . pipiens in virus transmission , although SBV was able to infect Cx . pipiens mosquitoes [39] . However , as Cx . theileri has been identified as a vector of several other bunyaviruses , this mosquito may also be a possible vector of SHUV [40 , 41] . Thus , vector competence studies with additional mosquito species collected from the field are needed to fully understand the possible role of mosquitoes in natural transmission cycles of SHUV . In this study , we determined infection , dissemination , and transmission of SHUV by infectivity assays and virus titers by EPDA ( i . e . assays based on inoculation of samples on Vero cells which are then screened for CPE ) . Such infectivity assays and EPDAs have the advantage of detecting infectious virus particles , whereas other methods like qPCR that quantify genome equivalents , may include defective virus particles and thereby not accurately represent infectious virus . Of note , observed CPE in the infectivity assays and EPDAs was found to invariably correspond with SHUV RNA as determined by RT-qPCR ( S1 Supporting Information ) . Recent outbreaks of SBV and BTV exemplified the tremendous impact of midge-borne viruses on animal health [37 , 38] . Our study demonstrates highly efficient infection , replication , and dissemination of SHUV in two biting midge species ( C . nubeculosus and C . sonorensis ) . However , conclusive evidence for SHUV transmission by biting midges should be provided by experiments with infected biting midges and susceptible mammals , although these kind of experiments are costly and complex . We cannot exclude that results obtained with laboratory-reared vectors are different from those obtained with field-collected vectors . Therefore , future studies should test vector competence of field-collected Culicoides biting midge and mosquito species exposed to different quantities of SHUV , to more accurately predict the risk of SHUV transmission in specific areas . These experiments in combination with behavioural and ecological research will contribute to our understanding of the transmission cycle of SHUV . | Arthropod-borne ( arbo ) viruses are notorious for causing unpredictable and large-scale epidemics and epizootics . Apart from viruses such as West Nile virus and Rift Valley fever virus that are well known to have a significant impact on human and animal health , many arboviruses remain neglected . Shuni virus ( SHUV ) is a neglected virus with zoonotic potential that was recently associated with severe disease in livestock and wildlife . Isolations of SHUV from field-collected biting midges and mosquitoes suggests that SHUV may be transmitted by these insects . Laboratory-reared biting midge species ( Culicoides nubeculosus and C . sonorensis ) and mosquito species ( Culex pipiens pipiens and Aedes aegypti ) , that are known to transmit other arboviruses , were exposed to SHUV via an infectious blood meal . SHUV was able to successfully disseminate in both biting midge species , whereas no evidence of infection or transmission in both mosquito species was found . Our results show that SHUV infects and disseminates in two different Culicoides species , suggesting that these insects could play an important role in the disease transmission cycle . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"rift",
"valley",
"fever",
"virus",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"culicoides",
"microbiology",
"saliva",
"animals",
"viruses",
"rna",
"viruses",
"insect",
"vectors",
"bunyaviruses",
"infectious",
"diseases",
"aedes",
"aegypti",
"medical",
"microbiology",
"microbial",
"pathogens",
"viral",
"replication",
"disease",
"vectors",
"insects",
"arthropoda",
"mosquitoes",
"eukaryota",
"blood",
"anatomy",
"arboviruses",
"virology",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"organisms"
]
| 2019 | Vector competence of biting midges and mosquitoes for Shuni virus |
The regulation of intracellular levels of reactive oxygen species ( ROS ) is critical for developmental differentiation and virulence of many pathogenic fungi . In this report we demonstrate that a novel transmembrane protein , TmpL , is necessary for regulation of intracellular ROS levels and tolerance to external ROS , and is required for infection of plants by the necrotroph Alternaria brassicicola and for infection of mammals by the human pathogen Aspergillus fumigatus . In both fungi , tmpL encodes a predicted hybrid membrane protein containing an AMP-binding domain , six putative transmembrane domains , and an experimentally-validated FAD/NAD ( P ) -binding domain . Localization and gene expression analyses in A . brassicicola indicated that TmpL is associated with the Woronin body , a specialized peroxisome , and strongly expressed during conidiation and initial invasive growth in planta . A . brassicicola and A . fumigatus ΔtmpL strains exhibited abnormal conidiogenesis , accelerated aging , enhanced oxidative burst during conidiation , and hypersensitivity to oxidative stress when compared to wild-type or reconstituted strains . Moreover , A . brassicicola ΔtmpL strains , although capable of initial penetration , exhibited dramatically reduced invasive growth on Brassicas and Arabidopsis . Similarly , an A . fumigatus ΔtmpL mutant was dramatically less virulent than the wild-type and reconstituted strains in a murine model of invasive aspergillosis . Constitutive expression of the A . brassicicola yap1 ortholog in an A . brassicicola ΔtmpL strain resulted in high expression levels of genes associated with oxidative stress tolerance . Overexpression of yap1 in the ΔtmpL background complemented the majority of observed developmental phenotypic changes and partially restored virulence on plants . Yap1-GFP fusion strains utilizing the native yap1 promoter exhibited constitutive nuclear localization in the A . brassicicola ΔtmpL background . Collectively , we have discovered a novel protein involved in the virulence of both plant and animal fungal pathogens . Our results strongly suggest that dysregulation of oxidative stress homeostasis in the absence of TmpL is the underpinning cause of the developmental and virulence defects observed in these studies .
Oxidative stress arises from a significant increase in the concentration of reactive oxygen species ( ROS ) inside the cell , and is primarily caused by either an imbalance of the cellular antioxidant capacity or a deficiency in the antioxidant system controlling ROS levels [1] . The damaging effects of ROS on DNA , proteins , lipids and other cell components and their role in pathological and aging processes is well established [2] , [3] , [4] . Numerous studies of pathogenic fungi have documented the crucial role of ROS produced by either fungal pathogens or their hosts in pathogenesis and defense-related activities [5] , [6] , [7] . There is also increasing evidence supporting an alternative view that ROS play important physiological roles as signaling molecules . ROS have been shown to be critical in immunity , cell proliferation , cell differentiation , and cell signaling pathways . However , the mechanisms by which ROS and their associated enzymes regulate development in microbial eukaryotes remain to be defined [8] , [9] . Taken together , all the deleterious , pathological , and regulatory roles of ROS have generated great interest in defining the mechanisms by which ROS are produced , sensed , and managed in eukaryotes . Because ROS readily lead to oxidative injuries , it is extremely important that the cellular ROS level be tightly controlled by complex and sophisticated redox homeostasis mechanisms . In the yeast Saccharomyces cerevisiae , the transcription factors Yap1 and Skn7 and a pair of related factors , Msn2 and Msn4 ( Msn2/4 ) , are implicated in controlling intracellular ROS levels [10] , [11] , [12] . Yap1 and Skn7 activate the expression of proteins that intercept and scavenge ROS . Yap1 is primarily controlled by a redox-sensitive nuclear export mechanism that regulates its nuclear accumulation when activated [13] . The Msn2/4 regulon contains only a small number of antioxidants but also includes heat shock proteins ( HSPs ) , metabolic enzymes , and components of the ubiquitin-proteasome degradation pathway [14] . Recently , a heat shock transcription factor , Hsf1 , has been added to the list of oxidative stress-responsive activators [15] . In addition to those found in S . cerevisiae , hybrid histidine kinase Mak1 and response regulator Prr1 ( a Skn7 homolog ) , and bZIP transcription factors Atf1 and Pap1 ( a Yap1 homolog ) in Schizosaccharomyces pombe are also involved in transducing hydrogen peroxide ( H2O2 ) signals . These proteins are required to induce catalase gene ctt1+ and other genes in response to H2O2 [16] , [17] . Although several similar proteins have been found and characterized in filamentous fungi , little is known about other transcriptional regulators or the defined regulatory mechanisms implicated in oxidative stress responses in filamentous fungi [7] , [18] , . However , orthologs of most components of the oxidative stress-sensing pathway described in yeasts are also known to be conserved in filamentous fungi such as Aspergillus nidulans and Neurospora crassa [20] , [21] . Pathogenic fungi need specialized , multi-faceted mechanisms to deal with the oxidative stress encountered in vivo during infection . Therefore , adaptive mechanisms that confer resistance to the oxidative stress from intra- or extracellular sources may contribute to the efficient colonization and persistence of fungal pathogens in their hosts . One of the most rapid plant defense reactions encountered by plant pathogens is the so-called oxidative burst , which constitutes the production of ROS , primarily superoxide and its dismutation product , H2O2 , at the site of attempted invasion [22] , [23] . The ROS produced by the oxidative burst either activate plant defense responses , including programmed cell death , or function as secondary messengers in the induction of various pathogenesis-related ( PR ) genes encoding different kinds of cell wall-degrading enzymes [24] , [25] , [26] . Furthermore , the presence of H2O2 is essential for the formation of lignin polymer precursors via peroxidase activity , which provide additional plant barriers against pathogen attack [27] . Similarly , animal phagocytic cells produce ROS to combat invading fungal pathogens . For example , following inhalation of airborne Aspergillus fumigatus conidia , the normal host is protected by pulmonary innate immunity , including phagocytosis by macrophages , where the killing of the engulfed conidia is known to be directly associated with ROS production [28] , [29] . In vitro studies of neutrophil function have shown that H2O2 effectively kills fungal hyphae [30] and that neutrophil-mediated damage is blocked by the addition of a commercial catalase [31] . Consequently , to counteract the potentially dangerous accumulation of ROS surrounding infection sites , fungal pathogens have developed diverse strategies . These include physically fortified or specialized fungal infection structures and various antioxidant defense systems through transporter-mediated effluxing , non-enzymatic antioxidants , and enzymatic scavenging systems , generally using NAD ( P ) H as reducing equivalents [32] , [33] , [34] , [35] . Through a combination of computational and functional genomics approaches a novel gene tmpL , encoding a transmembrane protein with a N-terminal AMP-binding domain and C-terminal NAD ( P ) /FAD-binding domain , was characterized in this study . Previously , a protein with approximately 50% identity but lacking the AMP-binding domain present in TmpL was discovered in A . nidulans to be important for regulation of conidiation [36] . TmpL was initially identified during this study and referred to as the large TmpA homolog but was not functionally characterized [36] . In the present study , we characterize TmpL in both a plant and an animal fungal pathogen and provide cytochemical and genetic evidence that demonstrate a filamentous fungi-specific mechanism for control of intracellular ROS levels during conidiation and pathogenesis .
Previously , seven putative nonribosomal peptide synthetase ( NPS ) genes designated as AbNPS1 to AbNPS7 , for Alternaria brassicicola nonribosomal peptide synthetase , were identified in the A . brassicicola genome via HMMER and BLAST analyses in our lab [37] . During this study , a NPS-like gene was identified with only a putative AMP-binding domain similar to an adenylation domain , followed by six transmembrane domains . There were no sequences in the adjacent region similar to thiolation and condensation domains which are typical components in the multi-modular organization of NPS genes . We designated this AMP-binding domain containing gene as tmpL , referring to the previous nomenclature but designating it as tmpL in lieu of large tmpA homolog [36] . The entire sequence of the tmpL gene was determined and confirmed by several sequencing events using genomic DNA and cDNA as templates for PCR based amplification and sequencing with primers based on information derived from the A . brassicicola genome sequence ( http://www . alternaria . org ) . The open reading frame ( ORF ) of the tmpL is 3450 bp long and predicted to encode a protein of 1025 amino acids . The predicted TmpL hybrid protein contains an AMP-binding domain , six putative transmembrane domains , and a FAD/NAD ( P ) -binding domain ( Figure 1A ) . The A . brassiciola TmpL protein sequence was used to search for an A . fumigatus ortholog via BLASTP analysis in the genome sequence of strain CEA10 . The highest sequence similarity was found for a protein encoded by a gene with the locus ID AFUB_085390 . The protein sequences are 41% identical and use of protein domain prediction tools suggested that the A . fumigatus protein , like the A . brassicicola protein , has a putative N-terminal AMP-binding domain , followed by six transmembrane domains and a FAD/NAD ( P ) -binding domain at the C-terminus . Based on the high sequence and structural similarities to the A . brassicicola tmpL gene , we named this gene A . fumigatus tmpL as well . The ORF of the A . fumigatus tmpL is 3357 bp long , contains 8 predicted introns and encodes for a protein of 994 predicted amino acids . Phylogenetic analysis indicated that TmpL and its putative orthologs are present only in filamentous fungi ( Figure S1 ) . The majority of fungal genomes shown in the phylogenetic tree contained a single putative TmpL ortholog , including A . nidulans that has TmpA [36] . Notable exceptions included the Basidiomycete , Coprinus cinerea , which contained 3 , and the Sordariomycetes Fusarium graminearum ( Gibberella zeae ) ( 3 ) , F . oxysporum ( 2 ) , and F . verticillioides ( 2 ) . A . brassicicola did not contain a putative TmpA homolog , while A . fumigatus contained one ( EAL91362 ) . The AMP-binding domain of the TmpL protein showed high similarity to adenylation domains of the NPS proteins [38] , which are generally involved in the activation of an amino acid substrate in the nonribosomal synthesis of polypeptides . One of the most similar sequences in the GenBank NR database was Cochliobolus heterostrophus NPS12 ( score = 2901 , ID = 54% ) , which was reported as a putative NPS gene [39] . However protein functional domain searches conducted against NCBI conserved domains and the InterPro database did not detect any thiolation and condensation domains in the predicted TmpL protein . This indicates that the TmpL is indeed lacking both thiolation and condensation domains that are conserved among NPSs , and thus is not a true NPS protein . Given that TmpL does not appear to be a true NPS , we next sought to determine the function of this protein in A . brassicicola . The transmembrane and FAD/NAD ( P ) -binding domains demonstrated a high sequence similarity and predicted structure to the previously identified plasma membrane flavoprotein , TmpA , in Aspergillus nidulans ( Figure S2 ) [36] . As with TmpA , the sequence analysis of the FAD/NAD ( P ) -binding domain showed that TmpL contains two important consensus sequences which are highly conserved in flavoproteins that bind both FAD and NAD ( P ) . They are hypothetical FAD ( RLHFD ) and NAD ( P ) ( GSGIGP ) phosphate-binding domains ( Figure S1 ) , and correspond to the RXYS ( T ) motif for the FAD-binding domain and the GXGXXG or GT ( S ) G ( A ) IXP consensus sequences for the NAD ( P ) -binding domain , respectively [40] , [41] , [42] . In addition , protein structure homology modeling with TmpL C-terminal 247 amino acids using Azotobacter vinelandii NADPH:ferredoxin reductase as a template [42] via SWISS-MODEL at ExPASy ( http://swissmodel . expasy . org/ ) showed a possible cleft formed by the two domains where the FAD and NAD ( P ) -binding sites were juxtaposed ( data not shown ) . This finding was also reported in the TmpA study [36] . To support this in silico data , we generated a partial TmpL recombinant protein containing the FAD/NAD ( P ) -binding domain via E . coli expression . The UV-visible spectra of the partial protein observed were characteristic of a flavoprotein ( Figure 1B ) . The absorbance peaks at 367 and 444 nm indicated that the enzyme contained bound flavin . All of these analyses suggest that TmpL possesses an enzymatic function using its FAD/NAD ( P ) -binding domain like other NAD ( P ) H-dependent flavoenzymes containing FAD or FMN cofactors such as the ferric reductase ( FRE ) protein group . Fungal proteins belonging to the FRE group include metalloreductase [43] , NADPH-cytochrome P450 reductase [44] , ferric-chelate reductase [45] , and NADPH oxidases ( NOX ) [9] . Next , we examined the putative subcellular localization of TmpL to gain possible insights into its cellular functions . First , in silico analyses were performed using wolf psort , sherloc , targetp , tmhmm , pred-tmr and signalp [46] , [47] , [48] , [49] , [50] , [51] . sherloc predicted a possible subcellular localization of the TmpL protein to the peroxisomal membrane with a high probability score ( 0 . 94 ) , while wolf psort and targetp assigned no definitive subcellular location . tmhmm and pred-tmr analyses predicted six possible transmembrane helices in TmpL similar to the results of initial protein conserved domain searches . There was no predictable N-terminal signal peptide sequence for co-translational insertion into a specific subcellular component by signalp . Taken together , these predictions indicated that TmpL might be a peroxisomal integral membrane protein with six transmembrane helices . To experimentally determine the localization of TmpL within the various cell types and intracellular compartments and organelles in A . brassicicola , a strain expressing a TmpL-GFP fusion protein was generated . Two transformants carrying a single copy of the tmpL:gfp allele tagged at the genomic locus were identified by PCR analysis and further confirmed by Southern blot analysis ( data not shown ) . Compared with the wild-type strain , neither of the two transformants exhibited differences in growth or pathogenesis except for expression of green fluorescence in conidia , suggesting that TmpL-GFP is fully functional . One of the transformants , A1G4 , was used to analyze the localization of TmpL-GFP using confocal laser scanning fluorescence microscopy . The GFP signal was detected in conidia , but no GFP signal was detected in the vegetative mycelia of the A1G4 strain grown in complete media ( CM ) ( Figure 2A ) . The GFP signals were localized in a punctate pattern in the cytoplasm as one or two tiny spots in each conidial cell , either near septae or associated with the cortical membrane . Given the previous in silico analyses , we hypothesized that the GFP signal might come from a specialized peroxisomal structure , the Woronin body ( WB ) . In order to perform a co-localization test , we selected the known WB core protein HEX1 in N . crassa , and searched for the orthologous abhex1 gene in A . brassicicola . Using the same strategy with the TmpL-GFP fusion constructs , we produced a DsRed-AbHex1 fusion protein-expressing transformant in the TmpL-GFP strain A1G4 background . DsRed-AbHex1 showed a similar punctate distribution in the cytoplasm , mostly near septal pores , but a few distant from septal pores . Figure 2A shows only DsRed-AbHex1 that are distant from septal pores co-localized with the TmpL-GFP . A separate analysis by confocal microscopy of strains that expressed either TmpL-GFP or DsRed-AbHex1 ruled out any possible cross talk between the two fluorescence signals . Although there is no literature indicating two distinct types of WBs in fungal conidia , this might suggest that TmpL is associated with a specific WB that is not associated with septal pores . Using transmission electron microscopy ( TEM ) of A . brassicicola conidia , we confirmed several WBs located distantly from septal pores ( Figure S3A ) . As mentioned , there was no TmpL-GFP detected in vegetative hyphae , while the DsRed-AbHex1 was distributed near septal pores ( Figure 2A ) as reported in other studies [52] , [53] . The WB has been described as evolving or being formed from peroxisome . The HEX1 assemblies emerge from the peroxisome by fission ( budding off ) and the nascent WB is subsequently associated with the cell cortex [54] , [55] . To observe the peroxisomes and their relationship to TmpL , we co-expressed TmpL-GFP and peroxisome matrix-targeted DsRed which has a C-terminal SKL tripeptide , a peroxisome targeting signal 1 ( PTS1 ) . The TmpL-GFP was mostly associated with relatively large peroxisomes ( Figure 2B ) . Interestingly , depending on whether conidia were harvested from the center or edge of the colony ( old to young ) prior to microscopic examination , three different types of association between TmpL-GFP and DsRed-PTS1 were observed . The TmpL-GFP signals in young conidia most often showed complete association with peroxisomes . Some TmpL-GFP signals mainly in older conidia were detected in a partial association with or complete dissociation from DsRed-PTS1 ( Figure 2B ) . Together with TmpL-GFP localization with DsRed-AbHex1 , these sequential associations might indicate a sequential process of WB biogenesis in A . brassicicola: AbHex1 assemblies in large peroxisomes ( Figure 2B , a green circle ) , a budding event of nascent WB out of the peroxisome ( Figure 2B , white circles ) , and a mature WB that is completely separated from the peroxisome ( Figure 2B , red circles ) . This result was also supported by the observation of aged conidia from 21-day-old colonies , which rarely showed co-localization between TmpL-GFP and DsRed-PTS1 fusion proteins ( data not shown ) . It has been recently shown that PEX14 , an essential component of the peroxisomal import machinery , is essential for the biogenesis of both peroxisome and WB . The deletion of pex14 leads to complete mis-localization of peroxisomal matrix proteins containing PTS1 signal and HEX1 to the cytosol [53] . To determine whether deletion of the A . brassicicola homolog of pex14 affects TmpL localization , we generated Δpex14 mutant strains in a TmpL-GFP strain background using a linear minimal element ( LME ) gene disruption construct [56] and examined the mutants with confocal microscopy . In most of the TmpL-GFP:Δpex14 mutant conidia , disruption of pex14 resulted in an uneven distribution of the TmpL-GFP in the cytoplasm ( Figure 2C ) . The DsRed-AbHex1:Δpex14 mutants used as control also showed cytoplasmic distribution of the DsRed-AbHex1 as reported in the study mentioned earlier [53] . Therefore , pex14 is related to the proper localization of TmpL protein in association with WB and peroxisome proteins governed by pex14-related peroxisomal import machinery , further suggesting that TmpL is associated with a specific type of WB that is not associated with septal pores . HEX1 and its orthologs in filamentous fungi possess a PTS1 at their C-terminal end that target it to the peroxisomal matrix [57] . However , as other known peroxisomal membrane proteins , the predicted TmpL sequences do not carry any defined localization signal peptides or PTS peptides . To identify the organelle targeting information in TmpL , we produced three transformants by appending GFP marker protein at three locations of TmpL: the AMP-binding domain , transmembrane domain , and FAD and NAD ( P ) -binding domain . This produced truncated TmpL-GFP fusion proteins under the control of the wild-type tmpL promoter ( Figure 3 ) . Using each construct , we generated three different GFP-tagged strains and examined their localization pattern . The AMP-binding-GFP fusion protein resulted in cytoplasmic distribution of the GFP signal , while the transmembrane- and FAD and NAD ( P ) -binding-GFP fusion proteins were concentrated in a punctate pattern in the cytoplasm ( Figure 3 ) . This suggests that the transmembrane domain carries the targeting signal to the organelle membrane . To gain further insights into the possible function of TmpL , we next examined tmpL mRNA abundance in diverse fungal developmental stages . Relative abundance of tmpL mRNA transcripts during vegetative growth , conidiation , and plant colonization were estimated by quantitative real-time polymerase chain reaction ( QRT-PCR ) ( Figure 4A ) . The abundance of tmpL mRNA during vegetative growth in liquid CM was extremely low compared with the internal reference gene , A . brassicicola glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) . Interestingly , the mRNA abundance of tmpL increased almost six fold at 12 hr post-inoculation ( hpi ) on plant leaves ( i . e . , approximately at the time when penetration and infection hyphae develop from appressoria ) , compared with that of conidia ( 0 hpi ) . This result was also supported by in planta observation of the TmpL-GFP strain using epifluorescence microscopy ( Figure 4B ) . At 24 and 48 hpi , however , the mRNA abundance was significantly decreased from the 12 hpi level . From 48 hpi , the mRNA abundance gradually increased until 120 hpi . To examine tmpL mRNA abundance during conidiation , vegetative mycelia grown in liquid CM were exposed to ambient air to stimulate conidiophore formation and subsequent conidia production . tmpL mRNA abundance was gradually elevated up to six-fold during conidiation compared with vegetative growth in liquid CM . Epifluorescence microscopy with the TmpL-GFP strain confirmed the increased expression of tmpL in young conidia ( Figure 2A ) and conidiophores ( Figure S3B ) . Overall , these data indicate that tmpL transcript is strongly accumulated during conidiation and during infection in planta . To further characterize the role of TmpL in fungal development and pathogenesis , a targeted gene replacement strategy was adopted to produce tmpL deletion mutants in A . brassicicola ( Figure S4 ) and A . fumigatus ( Figure S5 ) . For the complementation of the A . brassicicola ΔtmpL ( AbΔtmpL ) strain we introduced both the full-length tmpL gene and nourseothricin resistance gene ( NAT ) fragments into the AbΔtmpL strain . Re-introduction of full-length tmpL gene in A . fumigatus ΔtmpL ( AfΔtmpL ) strain was conducted as well by introducing full length tmpL with hph gene for hygromycin resistance ectopically into the AfΔtmpL strain . The resulting complemented strains were named AbtmpL rec and AftmpL rec for A . brassicicola and A . fumigatus mutant strains , respectively . All strains were rigorously confirmed with Southern blot and PCR analyses ( Figure S4 and S5 ) . Analysis of developmental characteristics , including germination , growth , and conidiation on CM and in planta , of A . brassicicola tmpL deletion mutants indicated that they were indistinguishable from wild-type and an ectopic mutant A1E1 . The mutant strains also showed no defects related to osmotic stress , cell wall perturbation , or responses to antifungal drugs ( data not shown ) . However , it was noted that the AbΔtmpL strains displayed less pigmentation in culture ( Figure 5A ) . Light microscopy showed that the conidia of the mutants were less pigmented and were narrower than the wild-type . Few multicellular conidia with longitudinal septa were detected among the mutants , which may explain the larger minor axis in wild-type conidia . In addition , increased conidial chain branching was observed in AbΔtmpL strains compared with the wild-type ( Figure 5A ) . Further investigation of the abnormal mutant conidia using TEM revealed that the conidial cell wall was significantly more electron-dense and thicker in the wild-type than the AbΔtmpL strain ( wild-type , 746±116 nm , n = 53; AbΔtmpL , 504±83 nm , n = 64; p<0 . 01 ) . The reconstituted strain AbtmpL rec showed the rescue of the less pigmented conidia and abnormal conidiogenesis seen in the AbΔtmpL strains ( data not shown ) . Another interesting difference between A . brassicicola wild-type and ΔtmpL strains was noticed in older fungal colonies . The conidial suspension of a 21-day-old AbΔtmpL strain appeared more yellow in color than a comparable wild-type suspension ( Figure 5B ) . We analyzed the conidial suspensions to obtain a secondary metabolite profile using high performance liquid chromatography but the profiles were comparable ( data not shown ) . A protein quantification assay , however , detected large differences in the amount of protein . The 21-day-old AbΔtmpL strain released more cytoplasm than the wild-type as judged by the amount of total protein quantified in the conidial suspensions ( Figure 5B ) . This result was further supported by our finding that the 21-day-old AbΔtmpL conidia showed frequent cell bursts in water under light microscopy , which resulted in exuding large amounts of cytoplasm ( Figure 5C , LM ) . Ultrastructural analysis revealed more frequent cell necrosis-like phenotypes in cells of the AbΔtmpL conidia compared with seemingly intact wild-type conidia ( Figure 5C , TEM ) . In order to clarify the TEM observation , we determined the percentage of old conidia that stained positive with annexin V-FITC , a compound that specifically stains apoptotic or dead cells by binding to phosphatidylserine present on the outer leaflet [58] , [59] . The annexin V-stained conidia from 21-day-old AbΔtmpL strain were increased significantly to 30% , whereas the annexin V-positive wild-type conidia had increased less than 10% after 21 days of growth on CM ( Figure S6 ) . These phenotypic abnormalities suggest that the membrane protein TmpL is required for proper fungal conidiation and maintenance of fungal cell integrity with aging in A . brassicicola . A . fumigatus ΔtmpL strains displayed no noticeable phenotypic change when grown on glucose minimal media ( GMM ) plates compared with the wild-type strain CEA10 . Unlike A . brassicicola ΔtmpL strains , A . fumigatus ΔtmpL strains displayed normal pigmentation and cell wall thickness in conidia compared with CEA10 ( data not shown ) . However , when we examined aged conidia using TEM , obvious differences were observed in the AfΔtmpL strain conidia ( Figure 5D ) . The 10-day-old A . fumigatus wild-type conidia featured cells with normal structure and clearly identifiable organelles , nuclei surrounded by a nuclear membrane , and mitochondria with well-preserved outer and inner membranes ( Figure 5D , CEA10 ) . TEM of the reconstituted strain AftmpL rec conidia were comparable to the wild-type conidia ( data not shown ) . However , AfΔtmpL conidia had an abnormal subcellular morphology ( Figure 5D , AfΔtmpL ) . The mitochondria were less well defined and often displayed discontinuous or missing outer membranes ( Figure 5D , a ) . Chromatin condensation and margination was observed in many nuclei ( Figure 5D , a and b ) and amorphous electron-dense fragments were frequently aggregated in the cytoplasm ( Figure 5D , c ) . Signs of cell death , such as distorted organelles and numerous small vacuoles , were also observed in some conidia ( Figure 5D , d ) . These features appeared frequently , but not all were observed in every cell . Given the peroxisomal association of TmpL and the dramatic phenotype during conidiation observed in ΔtmpL strains , we suspected a possible involvement of TmpL in oxidative stress responses . To investigate this hypothesis , wild-type and ΔtmpL mutants of A . brassicicola were examined for sensitivity to two different sources of oxidative stress , the superoxide generator KO2 and H2O2 . The AbΔtmpL strain showed increased sensitivity to oxidative stress compared with the wild-type ( Figure 6A ) . The minimal inhibitory concentration ( MIC ) of KO2 for A . brassicicola wild-type was 12 . 5 mM and for the AbΔtmpL strain , 7 . 5 mM; the MIC of H2O2 for wild-type , 7 . 5 mM and for AbΔtmpL , 5 mM . The reconstituted strain AbtmpL rec showed comparable sensitivity to oxidative stress with the wild-type , indicating deletion of tmpL caused the hypersensitivity to oxidative stress . In order to investigate the functional conservation of the A . fumigatus tmpL , we also examined A . fumigatus ΔtmpL strains for sensitivity to oxidative stress . We tested germling sensitivity to H2O2 for the A . fumigatus strains ( Figure 6B ) . The germlings of the AfΔtmpL strain were more sensitive to H2O2 than the wild-type ( p = 0 . 0018 ) . The reconstituted strain AftmpL rec showed comparable sensitivity to H2O2 as the wild-type , and a slight , but statistically not significant , increase in tolerance to oxidative stress created by H2O2 in the germling test ( Figure 6B ) . Visualization of the accumulation of reactive oxygen species ( ROS ) was examined to investigate oxygen metabolism during conidiation and plant infection in A . brassicicola wild-type and ΔtmpL strains . We first investigated the production of ROS by using nitroblue tetrazolium ( NBT ) , which forms a dark-blue water-insoluble formazan precipitate upon reduction by superoxide radicals . Using this technique , it appeared that the AbΔtmpL strain conidia accumulated higher amounts of superoxide than the wild-type ( Figure 7A ) . Such increased accumulation of superoxide was also detected in the AbΔtmpL strain inoculated on onion epidermis . Formazan precipitates were typically more intense in the mature appressoria and emerging infection hyphae of the AbΔtmpL strain , normally after 12 hpi ( Figure 7B ) . However , wild-type appressoria and infection hyphae had less formazan precipitate than the AbΔtmpL strain . To investigate production of other ROS in conidia of A . brassicicola strains , we used 2′ , 7′-dichlorodihydrofluorescein diacetate ( H2DCFDA ) . This cell-permeable ROS indicator remains nonfluorescent until it is deacetylated by intracellular esterases and oxidized to yield DCF . The H2DCF can be oxidized by several ROS generated by intracellular peroxidases , but not directly by H2O2 [60] , [61] . Conidia released from 7-day-old colonies were subject to the H2DCFDA staining . More than half of the AbΔtmpL strain conidia examined were stained by H2DCFDA while only few wild-type conidia showed green fluorescence ( Figure 7C ) . Staining with 3 , 3′-diaminobenzidine tetrahydrochloride ( DAB ) visualized that mature appressoria of the AbΔtmpL strain on green cabbage cotyledons also accumulated more H2O2 than wild-type appressoria at 12 hpi ( Figure 7D ) . Together these data indicate that deletion of tmpL in A . brassicicola caused an intracellular burst of ROS in conidia and infection structures . This accumulation of ROS was also visualized in A . fumigatus wild-type and the ΔtmpL strain conidia using H2DCFDA ( Figure 7E ) . H2DCFDA staining of conidia from 3-day-old colonies showed a greater intensity of fluorescence in the AfΔtmpL conidia than in the wild-type CEA10 conidia . This brighter fluorescence was detected mainly in the smaller , younger AfΔtmpL conidia ( Figure 7E , inset ) . ROS production appeared to be greater in the conidiophores of AfΔtmpL than wild-type conidiophores , especially in the phialides and not the inflated vesicle of the conidiophore . This indicates that the oxidative burst first takes place mostly within phialides and then young conidia that are formed on the phialides in the absence of the tmpL gene in A . fumigatus . Taken together , these data indicate that deletion of tmpL in A . fumigatus resulted in the same phenotype as the A . brassicicola ΔtmpL strains: a burst of ROS in conidia and conidiophores . Given the increased ROS accumulation in the absence of TmpL , we next sought to determine whether the ROS scavenging system was impaired in the ΔtmpL strains of A . brassicicola . We compared the expression of general antioxidant and redox control gene orthologs: ctt1 ( catalase T ) , sod1 ( Cu , Zn superoxide dismutase ) , gsh1 ( gamma glutamylcysteine synthetase ) , gsh2 ( glutathione synthetase ) , trx2 ( thioredoxin ) , gpx1 ( glutathione peroxidase 1 ) , and two redox-regulating genes yap1 and skn7 in A . brassicicola wild-type and ΔtmpL strains ( Figure 8A ) . In the wild-type strain , the relative transcript levels of all genes increased up to nine-fold during conidiation ( 36 hr air-exposed mycelia ) compared with the transcript levels in vegetative mycelia . During conidiation all stress-associated genes examined showed up to a two-fold increase in mRNA abundance in the AbΔtmpL strain compared with the wild-type , while there was a very slight difference observed between the two strains during vegetative growth . Based on the fact that increased ROS levels typically result in higher expression of the enzymes that neutralize them [10] , [62] , these observations indicate a higher ROS level in the AbΔtmpL conidia . When combined with excess ROS accumulation observed in AbΔtmpL conidia ( Figure 7 ) , these results also indicate a fundamental inability of the mutant to reduce cellular ROS levels . This may be because it's beyond the cellular capability to neutralize them , even with increased activity of antioxidants . These results also strongly suggest that the Yap1 and Skn7 regulators are not downstream of TmpL activity . It has been demonstrated in multiple yeast and fungal systems that during oxidative stress , the transcription factor Yap1 facilitates targeted gene expression by migrating into the nucleus from its location in the cytosol [13] . This cellular movement of Yap1 might provide additional information about the state of oxidative stress in the AbΔtmpL strain . Wild-type and AbΔtmpL strains were transformed with a GFP-Yap1 construct under the control of the A . brassicicola yap1 promoter . Cellular localization of the GFP-Yap1 strains was examined by confocal microscopy ( Figure 8B ) . During normal conidiation on solid CM , fluorescence of GFP-Yap1 was distributed evenly throughout the cytoplasm of wild-type conidia ( Figure 8B , 0 mM H2O2 ) . In contrast , the AbΔtmpL:pYap1-GFP-Yap1 strains showed a focal , condensed GFP signal typical of nuclear localization , suggesting the mutant is in a state of constitutive oxidative stress during conidiation . By constrast , there was cytoplasmic distribution of the GFP signals observed in mycelia of the AbΔtmpL:pYap1-GFP-Yap1 strains ( data not shown ) . This observation not only indicates excess ROS accumulation only in conidial cells , but also excludes any possible involvement of environmental factors generating ROS in fungal cells , such as UV radiation , temperature shift , mechanical damage , etc [63] . In a parallel experiment , treatment of WT:pYap1-GFP-Yap1 and AbΔtmpL:pYap1-GFP-Yap1 strains with 1 mM H2O2 for 1 hr resulted in substantial nuclear localization of GFP-Yap1 in both strains ( Figure 8B , 1 mM H2O2 ) . This indicates that the GFP-Yap1 proteins in both strains are functional . Staining with DAPI confirmed our observations that GFP-Yap1 was indeed localized to the nucleus in these experiments ( data not shown ) . Given the above phenotypes of the ΔtmpL strains , we hypothesized that TmpL may play a key role in fungal virulence . To investigate the role of TmpL in A . brassicicola virulence , susceptible green cabbage ( Brassica oleracea ) were inoculated with two different concentrations of young , 7 day old conidia ( 2×105 and 5×104 conidia ml−1 ) ( Figure 9A ) . Plants inoculated with either wild-type or ectopic mutant ( A1E1 ) developed extensive , typical black spots on leaves at both concentrations of conidia tested . However , the black necrotic spots resulting from inoculation with AbΔtmpL strains ( A1–3 and A1–4 ) at both conidial concentrations was significantly smaller than those produced by the wild-type or ectopic mutant inoculations ( p<0 . 01 ) . The reconstituted strain AbtmpL rec ( A1C2 ) was found to be just as virulent as the wild-type at both concentrations of conidia . The average reduction in disease severity caused by the mutants compared with the wild-type was more than 62% and 80% when using the higher and lower conidial concentrations , respectively . Similar results were obtained in virulence assays with Arabidopsis . We next asked the question whether tmpL is also involved in fungal virulence in the human fungal pathogen Aspergillus fumigatus . Deletion of tmpL in A . fumigatus led to a statistically significant reduction ( p<0 . 01 ) in virulence in a chemotherapeutic murine model of invasive pulmonary aspergillosis ( Figure 9B ) . Mice infected with the AfΔtmpL strain did not display normal symptoms associated with invasive aspergillosis ( IA ) in contrast to wild-type and reconstituted strain infected mice which displayed well described symptoms of IA including ruffled fur , hunched posture , weight loss , and increased respiration . Consequently , like the ΔtmpL mutant in A . brassicicola that has reduced virulence on plants , TmpL is also required for fungal virulence in mammalian hosts . To understand the reasons for the reduced virulence of A . brassicicola ΔtmpL strains on green cabbage , we performed microscopic analyses of the infection processes . Examination of green cabbage cotyledons using light microscopy at 12 hpi revealed that the mutants formed appressoria on the plant surface similar to those formed by wild-type ( Figure S7A ) . Intracellular infection hyphae formed directly under the appressoria of the AbΔtmpL strain , however , rarely developed inside of plant epidermal cells , while development of infection hyphae from wild-type appressoria was consistently observed . An onion epidermis assay also showed similar results as the cotyledon assay ( Figure 10A ) . Only 7% of AbΔtmpL appressoria produced visible intracellular infection hyphae at 12 hpi ( Figure 10B ) , but initial penetration hyphae from most individual appressoria were frequently visible ( Figure 10A , inset ) . At 24 hpi , ∼11% of the AbΔtmpL appressoria developed intracellular infection hyphae . The remaining AbΔtmpL appressoria did not develop infection hyphae , but in some cases , produced one or several germ tubes that formed additional appressoria ( Figure 10A , 24 hpi ) . In contrast , more than half of the wild-type appressoria successfully produced intracellular infection hyphae at 12 hpi ( Figure 10B ) , which usually penetrated cross-walls and spread within 24 hr ( Figure 10A , 24 hpi ) . To characterize the host-pathogen interface , inoculated green cabbage leaves were examined by light and electron microscopy . In vertical leaf sections inoculated with the compatible wild-type , fungal appressoria successfully penetrated , formed intracellular infection hyphae , and killed most plant tissue below the infection sites within 24 hr ( Figure S7B and 10C ) . In contrast , leaf sections inoculated with the less virulent AbΔtmpL strain appeared undamaged , though it was noted that necrosis similar to a hypersensitive response or papillae formation ( callose deposition ) developed below the infection site ( Figure S7B ) . Transmission electron microscopy revealed penetration hyphae and appressoria of the AbΔtmpL strain showing typical cell death-like phenotypes ( cytoplasmic fragmentation , enlarged vacuoles , and distorted organelles ) and the penetration hyphae were completely arrested by papillae formation in plant epidermal cells ( Figure 10C ) . Callose deposition was also detected by cytological staining using aniline blue ( Figure 10D ) . The wild-type induced small , scattered deposits in close proximity to the sites of penetration and tissue necrosis was extensive . In contrast , callose deposits observed following AbΔtmpL inoculation were much more pronounced and often localized at the site of penetration . In order to investigate whether the AbΔtmpL strains can colonize the host plant when the first physical barrier , the plant cell wall , is removed , wounded leaf assays were performed ( Figure S7C ) . Symptoms produced by inoculation of the wild-type on wounded tissue were more severe than on intact ( non-wounded ) tissue . The AbΔtmpL strain formed larger lesions on wounded leaves than on intact leaves , but were still smaller than wild-type lesions on wounded leaves . Together , these results indicate that A . brassicicola ΔtmpL strains have defects in pathogenicity associated primarily with very early stages of plant infection , resulting in the failure of appressoria penetrating into epidermal cells and an induction of callose deposition . To further understand the potential mechanism behind the virulence defect of the A . fumigatus ΔtmpL mutant , we examined lung histopathology from mice on days +2 and +4 of the infection . On day 2 , AfΔtmpL mice generally displayed less necrotic lesions and less fungal burden as observed by H&E and GMS stains ( Figure 11 ) . However , the differences with regard to inflammation were subtle between wild-type and mutant infected animals and it is clear that both fungal strains were able to germinate and colonize the lung tissue ( Figure 11 ) . QRT-PCR analysis of fungal burden based on amplification of fungal 18S rRNA revealed an approximate 10 fold decrease in fungal burden in mice infected with the AfΔtmpL mutant ( data not shown ) . However , by day 4 , both wild-type and AfΔtmpL mutant mice displayed significant histopathological findings associated with Aspergillus infections including the development of granulomatous like lesions , massive influx of inflammatory cells ( primarily neutrophils ) to sites of infection , subsequent peribronchiolar and alveolar inflammation , and substantial fungal growth in silver stained tissue ( Figure 11 ) . In general , the inflammation and necrosis observed was much more significant in wild-type infected animals than AfΔtmpL infected animals ( Figure 11 ) . However , it was clear that the AfΔtmpL mutant was still persistent and causing pathology at this time point . These results partially mimic findings with regard to the virulence of the A . brassicicola ΔtmpL mutant during infection of wounded plants that displayed a slower colonization and disease progression than the wild-type strain . With regard to these animal experiments , it is unclear if the slower colonization of the mouse lung tissue by the AfΔtmpL strain observed on day 2 and day 4 of the infection is due to lack of growth by the fungus in the in vivo environment , or improved clearance by the host immune response . Additional studies are ongoing to further characterize the mechanism behind the virulence defect of the AfΔtmpL mutant strain . Given the excess oxidative burst phenotypes of the ΔtmpL strains , we hypothesized that overexpression of yap1 may rescue the ΔtmpL mutant phenotypes . To determine whether overexpression of the Yap1 transcriptional regulator can enhance the cellular scavenging ability of fungal cells and consequently restore the abnormal phenotype and reduced virulence in ΔtmpL strains , we generated a ToxA promoter-driven yap1 overexpression cassette using fusion PCR methods . Subsequently , we introduced the overexpression cassette into both A . brassicicola wild-type and ΔtmpL backgrounds and examined its effect on each strain . As shown in Figure 12A , the mRNA abundance of yap1 significantly increased at least 25-fold compared with each recipient strain: wild-type and AbΔtmpL , indicating that yap1 overexpression cassettes were successfully integrated in the genome and expressed under the control of the ToxA promoter . To evaluate whether Yap1 overproduction affected the induction of the antioxidant defense system , we monitored the transcriptional activation of ctt1 and sod1 orthologs as representative downstream genes regulated by Yap1 . During vegetative growth , there was no induction of the ctt1 and sod1 transcripts . During conidiation in 36 hr air-exposed mycelia , however , the yap1 overexpression mutant in the AbΔtmpL background ( AbΔtmpL:pToxA-Yap1 ) showed significantly increased expression ( almost two-fold ) of antioxidant genes . Yet , yap1 overexpression in the wild-type ( WT:pToxA-Yap1 ) resulted only in a slight increase of these antioxidant genes , possibly because of the mechanism of Yap1 activation; Yap1 is post-translationally activated only in the presence of cellular ROS [13] , [64] . Overexpression of yap1 restored oxidative stress tolerance of the AbΔtmpL strain , resulting in comparable sensitivity to H2O2 as the wild-type ( Figure 12B ) . Furthermore , the AbΔtmpL:pToxA-Yap1 strain produced wild-type-like conidia ( Figure 12C ) , indicating that yap1 overexpression complemented , at least to a substantial degree , the ΔtmpL phenotypes . There was no distinguishable phenotypic difference between the WT:pToxA-Yap1 strain and the wild-type recipient strain . In addition to the conidial phenotype , green cabbage infection assays showed that the AbΔtmpL:pToxA-Yap1 strain partially restored its virulence compared with the AbΔtmpL recipient strain , but was still not comparable to the wild-type ( AbΔtmpL , 4 . 1±2 . 83 nm , n = 26; AbΔtmpL:pToxA-Yap1 , 12 . 9±4 . 52 mm , n = 26; p<0 . 01 ) ( Figure 12D ) . Interestingly yap1 overexpression in the wild-type caused slightly decreased lesion size compared with its wild-type recipient strain ( wild-type , 17 . 2±2 . 5 mm , n = 22; WT:pToxA-Yap1 , 15 . 7±3 . 8 mm , n = 22; p<0 . 05 ) , indicating that excess antioxidant activity resulting from yap1 overexpression did indeed negatively affect the pathogenesis of the A . brassicicola wild-type . Overall , yap1 overexpression in the AbΔtmpL strain strongly suggested that the phenotypic defects and reduced virulence were attributable to failure in the regulation of intracellular ROS levels , particularly in conidia and infection-related structures during the conidiation process and during plant infection , respectively . However , the residual virulence defect in the presence of yap1 overexpression may suggest additional roles of tmpL in fungal virulence .
Mechanisms for adapting to stress either from intracellular or extracellular sources are among the most relevant and timely topics in fungal biology . During normal developmental processes , a fungal organism encounters various stresses from toxic by-products of its metabolism or oxidative stress generated mainly through aerobic respiration [33] , [65] . The cellular environment within a host , whether plant or animal , also represents a major source of stress to an invading fungal pathogen [26] , [66] , [67] . In order to evade or circumvent stress , the fungus must possess special adaptation mechanisms . In this study we provide the first evidence that a novel , pathogenicity-related gene from a plant and animal fungal pathogen , tmpL , is critical for proper conidiogenesis and infection of healthy host tissues . Furthermore , tmpL appears to be associated with a filamentous fungi-specific stress defense system that particularly responds to oxidative stress . TmpL is a novel hybrid protein consisting of an AMP-binding domain , six putative transmembrane domains , and a FAD/NAD ( P ) -binding domain . Based on our phylogenetic analysis , TmpL and its putative orthologs are present only in filamentous fungi ( Figure S1 ) and not highly related to proteins with known functions . Although portions of the predicted TmpL amino acid sequence showed high similarity to putative NPS protein sequences in the GenBank NR database , its sequence lacked thiolation and condensation domains necessary to create a minimal module in typical NPS proteins . The AMP-binding domain is very similar to an adenylation domain . The latter is most often associated with modular NPS enzymes , where it activates amino acids prior to their incorporation into nonribosomal peptides ( NRP ) [68] . Interestingly , all fungi that contained a TmpL homolog also contain numerous NPS genes . Though the exact function of TmpL remains to be determined , it may modify or activate specific amino acids associated with certain nonribosomal peptides acting as a signal molecule for oxidative stress responses in filamentous fungi . It is also proposed that based on the similarity of the C-terminal sequences of TmpL to a previously identified , although smaller , plasma membrane flavoprotein in A . nidulans , TmpA , TmpL might be involved in production of a regulatory signal , which eventually leads to fungal differentiation . As predicted in TmpA [36] , we suspected that the C-terminal region of TmpL had enzymatic activity . Bioinformatic analysis also showed TmpL and its orthologs contain proposed sites for FAD and NAD ( P ) -binding , based on protein modeling and the existence of two important consensus sequences , suggesting that the protein is specifically reduced by NAD ( P ) H with a reduction potential . Indeed in our study , a partial recombinant protein of TmpL , which includes FAD/NAD ( P ) -binding domain , supports this hypothesis by showing that the partial protein is capable of binding flavin . In addition , NCBI conserved domain BLAST searches identified a ferric reductase ( FRE ) domain with low similarity ( E-value 0 . 004 ) in the FAD/NAD ( P ) -binding domain of the TmpL protein , suggesting that TmpL might be distantly related to the FRE group of proteins . Indeed several FRE proteins are known to be involved in the response to oxidative stress in various organisms [69] , [70] , as part of a system that activates a number of different enzymes involved in redox control . When considered together , it is likely that TmpL uses electrons from NAD ( P ) H , transferred via FAD , to activate or modify unknown substrates or possibly downstream proteins in a redox-related signal transduction pathway . Our localization assays indicated that TmpL is associated with the Woronin body ( WB ) in filamentous fungi . WBs are known to plug septal pores in response to fungal cell injury , preventing excess cytoplasmic leaking [57] , [71] . Early TEM studies indicated a peroxisomal origin for WBs [72] . More recently , genetics and cell biology research confirmed that the WB is first assembled in large peroxisomes [54] , [55] . Our confocal microscopy analysis showing a sequential association between TmpL and peroxisomes suggests that TmpL is first targeted into peroxisomes by an unknown peroxisomal targeting signal and then goes through WB biogenesis , eventually becoming part of a mature WB . However , WB in A . brassicicola conidia appeared to be divided into two groups based on their location and the localization of TmpL . It is generally accepted that depending on the organism , cell type , and metabolic requirements , distinct sets of proteins could be housed within certain multipurpose organelles or microbodies [53] , [73] . Confocal analyses with TmpL-GFP and DsRed-AbHex1 double-labeled strain and TEM analysis of A . brassicicola conidia showing existence of one or two WB located in the cytoplasm near the cell cortex support this hypothesis . In addition , cytoplasmic redistribution of the TmpL-GFP fluorescence in a Δpex14 strain indirectly , albeit strongly , supports the idea that TmpL is associated with a specific WB where AbHex1 is localized . Several reports on WB from other fungi have established the presence of WB in non-septal regions , such as the tips of the germlings and secondary infectious hyphae , or at the cell periphery [74] , [75] , [76] . These WB showed no association with the hyphal septum , suggesting other possible functions than plugging septal pores in response to cell injury . For example , loss of WB in Magnaporthe grisea Δhex1 strains led to increased cell death in response to nitrogen starvation . This suggests that WB may function in response to environmental stress [76] . PRO40 , associated with WB in Sordaria macrospora , was pivotal in triggering the developmental switch from protoperithecia to perithecia [52] . Together , these findings indicate other possible functions of the WB associated with development or the multicellular growth characteristic of filamentous fungi . On the other hand , it is also true that very little is known about the WB function in other fungal structures such as conidia and specialized infection structures . Although we cannot rule out the possibility that DsRed-AbHex1 was targeted incorrectly to the peroxisome-like organelles where TmpL-GFP was localized because of its ectopic expression , it is more likely that these observations reflect the existence of a specific WB which is associated with TmpL . To confirm the association between TmpL and WB in the future , more detailed biochemical analyses are needed . These include either immunodetection assays using TmpL- and Hex1-specific antibodies following differential and density gradient centrifugation , or immunofluorescence microscopy . It has been well documented that regulation of ROS level is important during fungal development [21] , [63] . In this study , we also highlighted the significance of intracellular ROS concentration in relation to fungal development . Given the observations that tmpL was highly expressed during conidiation and the loss-of-function mutation resulted in abnormal conidiogenesis and excess ROS accumulation in conidia , we can speculate that TmpL is involved in important mechanisms for balancing ROS level during conidiation . Deletion of a catalase gene ( CATB ) in M . grisea caused similar phenotypic changes as was observed in the ΔtmpL strains , such as less pigmentation , fragile conidia , and reduced virulence [77] , indicating a possible common effect of excess intracellular ROS in filamentous fungi . In many fungi , inhibition of ROS generation or excess intracellular ROS levels affected various fungal developmental processes [6] , [35] , [63] , [78] . Even a fungus-plant mutualistic symbiosis requires a sophisticated regulation of the ROS production [79] , [80] . Consistent with the involvement of ROS in cell-wall biosynthesis [81] , it seems probable that the excess ROS levels in ΔtmpL strains resulted in lighter pigmentation in the A . brassicicola conidia . Several studies also reported that accumulation of ROS within the cytoplasm played a central role in apoptosis-like cell death [82] , [83] , as shown in our observations of apoptosis-like cell death phenomena in aged conidia of both A . brassicicola and A . fumigatus ΔtmpL strains . Increased expression of antioxidant genes in A . brassicicola ΔtmpL strains is another indicator of increased ROS levels in the cell . Indeed , several reports in different microorganisms have shown a correlation between the up-regulation of specific antioxidant enzymes and increased cellular ROS levels [21] , [84] , [85] , suggesting that increased ROS levels result in higher expression of the enzymes that neutralize them . On the other hand , it could be questioned why the increased antioxidant expression in the ΔtmpL strains did not result in reducing cellular ROS levels in the mutant cells . The possible reason for that would be excess ROS levels in the ΔtmpL strains were far beyond the cellular capability ( or threshold ) to neutralize them . Our results from experiments of yap1 overexpression in ΔtmpL mutant background provide major evidence for this hypothesis . Upon oxidative stress , Yap1 is involved in activating genes involved in a cellular antioxidant system , such as GSH1 ( γ-glutamylcysteine synthetase ) , TRX2 ( thioredoxin ) , GLR1 ( glutathione reductase ) , and TRR1 ( thioredoxin reductase ) [86] . Therefore we can speculate that Yap1 overproduction led to the increase of the cellular antioxidant defense capability in the ΔtmpL strain that produces excess intracellular ROS in conidia . Indeed , yap1 overexpression suppressed most of the phenotypic defects shown in the ΔtmpL strain , indicating excess intracellular ROS was most likely the primary reason for the phenotypic changes observed in the ΔtmpL mutants . Interestingly yap1 overexpression in the wild-type strain did not affect the expression levels of downstream antioxidant genes ctt1 and sod1 , consistent with the post-translational activation model of the Yap1 protein by intracellular ROS . When considered together , these results demonstrate that TmpL may be associated with a filamentous fungi-specific oxidative stress defense system . However , we cannot rule out another possibility that TmpL is involved in cellular ROS production . As a consequence of the loss of TmpL-operated ROS production , an additional means of ROS generation may be up-regulated during conidiation , resulting in excess production of ROS . Indeed M . grisea Δnox1Δnox2 mutant displayed increased ROS generation during hyphal growth compared with wild-type strain [6] , indicating that there is an alternative ROS source that is activated upon loss of the Nox enzymes . Similarly , in Podospora anserina inactivation of panox1 led to an enhanced ROS production in mycelia [87] . However there was no difference observed in the expression levels of A . brassicicola nox homologs , AbnoxA and AbnoxB between wild-type and ΔtmpL stains during conidiation process ( data not shown ) , suggesting the NADPH oxidase-mediated ROS production is not the cause of excess oxidative stress in the ΔtmpL stains . A major question from our work is the role of TmpL in fungal virulence . We observed that loss of TmpL function resulted in avirulence in both plant and animal fungal pathogens . With regard to plant pathogenesis , A . brassicicola ΔtmpL conidia successfully germinated and formed normal appearing appressoria on plant surfaces at similar rates as wild-type . Thus , a defect in germination or appressoria development cannot explain the mutant phenotype during plant pathogenesis . However , only 7% of the total appressoria were capable of penetrating the host and growth was rapidly arrested in the epidermal cells . Additionally , the mutant appressoria and penetration hyphae observed by TEM showed a cell-death-like phenotype that we speculate may be due to excess oxidative stress , as indicated by NBT and DAB staining . To understand whether the infection failure in ΔtmpL strains was related to the excess buildup of ROS therein , we tried to reduce the levels of ROS during in planta appressoria development and penetration using a NADPH oxidase inhibitor diphenylene iodonium or antioxidant ascorbic acid . However , none of the treatments were successful in restoring the infection failure of the ΔtmpL strains . Even the infection of wild-type strains treated with these agents was seriously suppressed and resulted in tiny lesions on host leaves ( data not shown ) . The latter result seems to be explained by the same reasoning with the observation that yap1 overexpression in the wild-type strain caused reduced lesion size compared with its wild-type recipient strain . All of these results suggest that an excess reduction in intracellular but not extracellular oxidative stress also leads to a significant suppression of fungal infection . In other words , a sophisticated balancing of ROS levels is critical in fungal pathogenesis of plants . As an alternative method of reducing excess ROS in appressoria and/or penetration hyphae of the ΔtmpL mutants , we chose to manipulate the existing antioxidant system present in filamentous fungi by overexpressing yap1 . NBT staining showed less superoxide accumulation in the appressoria of the AbΔtmpL:pToxA-Yap1 overexpression strain compared with the AbΔtmpL recipient strain ( data not shown ) . Although the overexpression strain exhibited significantly restored virulence , it still was not comparable to the wild-type . Thus , our yap1 overexpression analyses clearly demonstrated that the infection failure in ΔtmpL strains was related to the intracellular accumulation of excess ROS in fungal infection structures . Regulation of ROS level during pathogenesis has been a critical factor that governs success or failure of the infection process . For example , M . grisea showed considerable amount of oxidative burst in appressoria during its pathogenesis , and inhibition of the ROS production by some inhibitors resulted in abnormal appressoria and further failure of plant infection [6] . Deletion of the Yap1 oxidative stress response protein in Ustilago maydis caused avirulence on corn , resulting from an excess oxidative stress on infection structures [19] . In addition , numerous fungal pathogens of animals have been reported to possess a defined genetic program to respond to oxidative killing by the host [10] , [88] , [89] , [90] . However , yap1 deletion mutants in the human fungal pathogen A . fumigatus are still virulent in chemotherapeutic models of invasive aspergillosis [91] . This observation , coupled with the lack of full virulence restoration in the A . brassicicola ΔtmpL mutant strains overexpressing yap1 may suggest that the virulence defect of tmpL deficient strains is due to additional unknown causes . Indeed , in our studies the virulence of the A . fumigatus ΔtmpL mutant was also attenuated in gp91phox−/− mice , which are deficient in generating a respiratory burst and highly susceptible to A . fumigatus infection ( data not shown ) . Collectively , these studies and our observations suggest that production and accumulation of excess intracellular ROS , and not increased sensitivity to extracellular ROS , in both ΔtmpL mutants of plant and animal pathogenic fungi is the primary cause for reduced virulence . Thus increased sensitivity to and detoxification of host derived , extracellular ROS , is most likely not the reason for the avirulence observed in ΔtmpL mutants in both pathosystems . Recent discoveries of functional ROS-generating enzymes within filamentous fungi have elucidated some possible roles of the fungus-derived ROS in pathogenic species [6] , [92] . Fungal contributions to ROS production have been obtained from fungi showing such activity without any contact of host cells . For example , spores of M . grisea germinating in water generated H2O2 , O2− , and OH+ extracellularly [93] and ROS production was associated with the development of infection structures on glass coverslips [6] . Previous studies have also speculated the possible involvement of fungus-derived ROS production in the rapid growth and spread of the pathogens inside their hosts [94] , [95] . Together with the excess ROS accumulation in the ΔtmpL conidia it is more plausible to speculate that the failure of regulating intracellularly produced ROS caused the penetration failure of the ΔtmpL strains , and thus the reduced virulence . While A . fumigatus is not known to produce penetration structures like appressoria to invade mammalian hosts , it may be possible that failure to handle fungal ROS accumulation during the initial stages of host infection result in the avirulence of the ΔtmpL mutants . In conclusion , we have identified a novel transmembrane protein , TmpL , involved in plant and animal fungal virulence . Our results suggest that TmpL is involved in a complex redox homeostasis mechanism in A . brassicicola and A . fumigatus during fungal development and pathogenesis . Although the biochemical function of TmpL needs to be further investigated , it is plausible that the AMP-binding domain may activate signaling molecules and , together with the enzymatic activity generated by the FAD/NAD ( P ) -binding domain , regulate intracellular redox homeostasis . Since WBs have a peroxisomal origin [54] , we can speculate that the TmpL protein , associated with WB , might also have a peroxisomal origin . Considering that peroxisomes play a key role in both the production and scavenging of ROS in the cell , H2O2 in particular [96] , the peroxisome-originated TmpL may act as a detoxifier of ROS in the same way as many enzymatic peroxisomal membrane proteins and previously identified peroxisomal antioxidant regulators [97] , [98] . Another recent finding to support this connection between fungal WB and oxidative stress is that disruption of abhex1 in A . brasscicola resulted in mutant strains lacking WBs and were more sensitive to oxidative stress ( H2O2 ) than wild-type ( Kim et al . , unpublished data ) . This result was meaningful because the deletion of hex1 in other fungi also causes the complete loss of WB in the resulting mutants [76] , [99] . Of further interest is that the Δabhex1 mutants are not as hypersensitive as ΔtmpL strains , suggesting a possible , complicated relationship between the antioxidant involvement of the TmpL protein and its association with WB . Future studies will focus on identification of the specific substrate ( s ) directly or indirectly interacting with TmpL , and definitively determining the role of this interesting protein in plant and animal fungal virulence .
Alternaria brassicicola strain ATCC 96866 was used in this study ( American Type Culture Collection , Manassas , VA ) . The growth and maintenance of A . brassicicola and media composition were performed as described by Kim [37] except for a minimal medium ( MM ) ( 1% glucose , 0 . 5% ( NH4 ) 2SO4 , 0 . 2% KH2PO4 , 0 . 06% MgSO4 , 0 . 06% CaCl2 , 0 . 0005% FeSO4 . 7H2O , 0 . 00016% MnSO4 . H2O , 0 . 00014% ZnSO4 . 7H2O , 0 . 00037% CoCl2 . 6H2O ) . Aspergillus fumigatus strain CEA10 was used as the wild-type , stored as frozen stock in 20% glycerol at −80°C , and grown at 37°C , on glucose minimal medium ( GMM ) with appropriate supplements as previously described [100] . A . fumigatus strain CEA17 , a uracil auxotroph derived from CEA10 , was used as the recipient strain for generation of the ΔtmpL mutant . In our study , solid complete medium ( CM ) refer to potato-dextrose agar and liquid CM to glucose-yeast extract broth ( 1% glucose , 0 . 5% yeast extract ) . The virulence test on green cabbage ( Brassica oleracea ) was performed as described by Kim [37] . Briefly , A . brassicicola was inoculated with a 10 µl drop of conidial suspension ( 5×104 or 20×104 conidia ml−1 ) on each leaf of 5-week-old plants . Inoculated plants were kept in a plastic box at ambient temperatures and incubated at 100% humidity for 24 hr in the dark , followed by 16 hr fluorescent lights per day for 4–6 days . Lesion diameters were measured for all virulence tests . Statistical analyses were performed to test the differences in lesion diameters among the tested strains by a pairwise t-test using JMP software ( SAS Institute Inc . ) . P-values≤0 . 01 were considered statistically significant . To test the ability of AbΔtmpL strain to colonize on wounded plants , the same conidial suspensions were applied to needle scratches on host plant leaves . A tmpL replacement construct was made by a double-joint PCR method from three PCR fragments , with slight modifications [101] . Using A . brassicicola genomic DNA as a template , a 993 bp tmpL 5′ flanking region was amplified with primers TMPLR1 and TMPLR2 ( Table S1 ) , and a 992 bp tmpL 3′ flanking region was generated with primers TMPLR5 and TMPLR6 . Using pCB1636 [102] as a template , a ∼1 . 4 kb hygromycin B phosphotransferase ( hph ) gene cassette was amplified with primers TMPLR3 and TMPLR4 . The reverse primer TMPLR2 that amplifies the 5′ flanking region and the forward primer ATMR5 that amplifies the 3′ flanking region , contained 20 bp tail sequences that overlapped the 5′ and 3′ ends of the hph cassette . Likewise , the forward and reverse primer TMPLR3 and TMPLR4 that amplified the hph cassette also contained a 20 bp tail sequences that overlapped the 5′ and 3′ flanking regions . The three PCR fragmentswere purified with the QIAquick PCR purification kit ( Qiagen , Valencia , CA ) , were then diluted 10-fold , and subjected to fusion PCR with primers TMPLR1 and TMPLR6 . The final 3 . 4 kb tmpL replacement construct was purified again with the QIAquick PCR purification kit and reduced to 1 µg/µl under vacuum before transformation . Fungal transformation was based on protocol described previously [56] . Transformants with expected genetic integrations were identified by PCR and Southern blot analysis . In order to reintroduce wild-type tmpL into the ΔtmpL mutant , we amplified the wild-type tmpL allele from A . brassicicola genomic DNA using primer set TMPLcomF and TMPLcomR . The resulting PCR product covers 5 . 2 kb between the 953 bp upstream in relation to the start codon and the 1132 bp downstream in relation to the stop codon . A 1449 bp long nourseothricin resistance gene ( NAT ) cassette was amplified with primer set PNRcomF and PNRcomR from pNR1 plasmid [103] . The final two PCR products were used simultaneously to transform ΔtmpL mutant A1–3 , and the transformants were selected using a nourseothricin antibiotic . PCR and Southern blot analyses were used to identify transformants with expected genetic integrations . Generation of a tmpL null mutant in A . fumigatus strain CEA17 was accomplished by replacing an ∼1 . 9 kb internal fragment of the tmpL coding region ( ∼3 . 36 kb; GenBank accession no . EDP49089 ) with A . parasiticus pyrG . The disruption construct was generated by cloning a sequence homologous to the tmpL locus into plasmid pJW24 ( donated by Nancy Keller , University of Wisconsin—Madison ) . The 5′ and 3′ tmpL homologous sequences , each ∼1 kb in length , were cloned to flank A . parasiticus pyrG in pJW24 . The resulting plasmid , pTMPLKO , was used as a template to amplify the ∼5 . 2 kb disruption construct ( primer RAC39 and RAC41 ) for use in fungal transformation . To complement the ΔtmpL strain , a plasmid with the tmpL gene connected to the hph gene was constructed . Therefore , the tmpL gene was amplified using genomic DNA of CEA10 as template and the primers RAC357 and RAC110 . The ∼5 . 3 kb PCR product and the plasmid pBC-Hygro were digested with NotI and SpeI . The PCR product was then ligated into the vector . The resulting plasmid , pTMPLREC , was used as a template to amplify the ∼9 . 5 kb reconstitution construct ( primer RAC325 and RAC326 ) for use in fungal transformation . Generation of fungal protoplasts and polyethylene glycol-mediated transformation of A . fumigatus were performed as previously described [104] . Briefly , 10 µg of the tmpLKO PCR-generated disruption construct was incubated on ice for 50 min with 1×107 fungal protoplasts in a total volume of 100 µl . Gene disruption transformants were initially screened by PCR to identify potential homologous recombination events at the tmpL locus . PCR was performed with primers designed to amplify only the disrupted tmpL locus - RAC109 and RAC22 ( PCR product: 2077 bp ) ; RAC21 and RAC110 ( PCR product: 1595 bp ) . For the reconstituted strain , 10 µg of the tmpLREC PCR-generated reconstitution construct was used in the protoplast transformation . Colonies were selected for growth on hygromycin containing media . Reconstitution events were then screened by PCR by amplifying a part of the tmpL that was replaced by pyrG in the mutant [RAC351 and RAC352 ( PCR product: 778 bp ) ] . Homologous recombination of the disruption cassette and random integration of the reconstituition construct was confirmed by Southern analysis with the digoxigenin labeling system ( Roche Molecular Biochemicals , Mannheim , Germany ) as previously described [105] . To eliminate the chance of heterokaryons , each transformant was streaked with sterile toothpicks a minimum of two times to obtain colonies from single conidia . DNA isolation and Southern blot analysis were performed as described by Kim [37] . The tmpL 3′ fragment was used as an tmpL specific probe and a 500 bp hph fragment from the pCB1636 plasmid was used as a hph specific probe , and a ∼1 kb NAT fragment from pNR1 plasmid as a NAT specific probe . All sequencing was done using the ABI Prism 310 automated sequencer ( Applied Biosystems , Forster City , CA ) . Total RNA was extracted from fungal samples using the RNeasy Plant Kit according to the manufacturer's protocol ( Qiagen , Valencia , CA ) . For the expression analysis with QRT-PCR , leaves of green cabbage were inoculated with 10 µl drops of wild-type conidial suspension ( 1×107 conidia ml−1 ) , and infected samples were collected at 12 , 24 , 48 , 60 , 72 , 96 , and 120 hr after inoculation . Total RNA was also extracted from mycelia grown in liquid CM for 72 hr . In order to maintain vegetative growth with no stress , the liquid CM was changed every 24 hr . About 20 mycelial balls collected from the above 72 hr-liquid culture were spread onto sterilized filter paper , incubated for conidiation , and collected at 24 and 48 hr for total RNA extraction . First-strand cDNA was generated from the total RNA of 48 hr air-exposed mycelial balls with random primers using SuperScript™ First-Strand Synthesis System ( Invitrogen™ Life Technologies , Carlsbad , CA , USA ) . A 635 bp tmpL partial coding sequence containing the FAD/NAD ( P ) -binding region was amplified from the cDNA using primers A1fn_ExpKpnFor and A1fn_ExpHndRev , and cloned between the KpnI and HindIII sites in plasmid pKLD66 [106] to obtain plasmid pA1FN . E . coli BL21 ( DE3 ) was transformed with pA1FN . E . coli BL21 ( DE3 ) ( pA1FN ) was grown to an optical density of 0 . 6–0 . 8 , followed by induction with 0 . 2 mM IPTG . After 3 hr of induction , cells were harvested by centrifugation at 7000×g for 10 min at 4°C . The resulting 2 g cell pellet was resuspended in 2 . 5 ml nickel-nitrilotriacetic acid ( Ni-NTA ) 50 mM sodium phosphate buffer , pH 7 . 5 . The cell suspension was passed three times through a French pressure cell at a pressure of 1 . 28×108 Pa . The resulting cell lysate was centrifuged at 8000×g for 25 min at 4°C to remove cell debris . The resulting supernatant was mixed with l ml Ni-NTA His Bind Resin ( Novagen ) and incubated for 1 hr at 4°C with constant agitation . The incubated solution was loaded onto a column bed and the column was washed with 10 ml Ni-NTA washing buffer ( 50 mM sodium phosphate ( pH 7 . 5 ) and 20 mM imidazole ) . The column was sequentially eluted with 50–500 mM imidazole containing 50 mM sodium phosphate buffer ( pH 7 . 5 ) . Fractions at about 250 mM imidazole were pooled and concentrated on an YM-30 membrane ( Amicon ) . The 1 mg protein concentrate was incubated with 0 . 2 mM FAD at 4°C for 5 hr . Free flavin was removed by filtration and three 1 mL washes with 50 mM sodium phosphate buffer ( pH7 . 5 ) , on the membrane of a YM-3 concentrator ( Amicon ) . The product was recovered in 50 mM sodium phosphate buffer ( pH7 . 5 ) and assayed for protein content . A UV-visible spectrum of the protein was analysed with 200–800 nm wavelength range . A tmpL C-terminal gfp fusion construct was generated by fusion PCR . Using A . brassicicola genomic DNA as a template , an 1 kb tmpL 3′ region was amplified with primers TMPLGFP1 and TMPLGFP2-GA . Another set of primers , TMPLGFP3-GA and TMPLGFP4 , were used to amplify a 2 . 4 kb gfp and hph cassette from template plasmid pCB16G6-Nac [56] . Two resulting fragments , the 1 kb tmpL 3′ fragment and the 2 . 4 kb gfp and hph cassette , were mixed and subjected to second fusion PCR with primers TMPLGFP1 and TMPLGFP4 . The resulting 3 . 4 kb PCR products were transformed in the A . brassicicola wild-type to make TmpL-GFP fusion transformants . Transformants with expected genetic integration events were identified by PCR and Southern blot analyses . The same fusion PCR strategy was applied to generate a series of fusion proteins in which different portions of tmpL , an AMP-binding and transmembrane domain , were appended to the N terminus of the gfp . For the tmpL AMP-binding-gfp fusion construct , two primers , A1AdeGFP1 and A1AdeGFP2-GA , were used to amplify an 881 bp tmpL AMP-binding domain region . Another set of primers , A1AdeGFP3-GA and A1AdeGFP4 , were used to amplify a 2 . 4 kb gfp and hph cassette from template plasmid pCB16G6-Nac . The two resulting PCR fragments were subjected to second fusion PCR with primers A1AdeGFP1 and A1AdeGFP4 . The resulting 3 . 4 kb PCR products were transformed in the wild-type . In the same way , four primers were designed to generate the tmpL transmembrane-gfp fusion construct as follows: A1TmGFP1 and A1TmGFP2-GA for a 756 bp tmpL transmembrane region; A1TmGFP3-GA and A1TmGFP4 for a gfp and hph cassette . To generate the DsRed-abhex1 fusion construct by fusion PCR , three PCR fragments were amplified as follows: a 573 bp Pyrenophora tritici- repentis ToxA promoter fragment using primers ToxAFor and ToxA-DsRedRev from template plasmid pCB16G6-Nac; a 728 bp DsRed ORF fragment using primers DsRed-ToxAFor and DsRed-AbHEX1Rev from template plasmid pCAG-DsRed [107]; a 969 bp abhex1 fragment using primers AbHEX1-DsRedFor and AbHEX1Rev from A . brassicicola genomic DNA . These final three PCR fragments were subjected to second fusion PCR with primers ToxAFor and AbHEX1Rev . The final construct was transformed into the TmpL-GFP fusion strain to generate TmpL-GFP:DsRed-AbHex1 dual fluorescence-labeled strains . To construct the DsRed-PTS1 construct that serves as marker of peroxisomal matrix , the DsRed fragment was amplified from pCAG-DsRed plasmid using primers DsRedPTS1For and DsRedPTS1Rev , which append the PTS1 tripeptide SRL to the C terminus of DsRed . Using pNR1 as template , a 1 . 4 kb nourseothricin resistance gene ( NAT ) cassette was amplified with primers DsRedPTS1NATFor and DsRedPTS1NATRev . These final two PCR fragments were subjected to second fusion PCR with primers DsRedPTS1For and DsRedPTS1NATRev . The final construct was transformed into the TmpL-GFP strain to generate TmpL-GFP:DsRed-PTS1 dual fluorescence-labeled strains . To disrupt pex14 in TmpL-GFP and DsRed-AbHex1 strains , a linear minimal element ( LME ) construct was generated as previously described [56] . Primers pex14KOFor and pex14KORev were used to amplify a 415 bp pex14 partial fragment from the A . brassicicola genomic DNA and another set of two primers , pex14HygFor and pex14HygRev , were used to amplify an 1 . 4 kb NAT cassettes from the plasmid pNR1 . The two fragments were subjected to second fusion PCR with primers pex14KOFor and pex14HygRev . The final construct was transformed into the TmpL-GFP and DsRed-AbHex1 strains to generate TmpL-GFP:Δpex14 and DsRed-AbHex1:Δpex14 mutant strains , respectively . To generate gfp-yap1 construct under the control of the yap1 promoter , four PCR fragments were amplified by fusion PCR . A 500 bp fragment of the yap1 promoter region was produced from A . brassicicola genomic DNA using primers PromoYap1For and PromoYap1Rev , a 570 bp fragment of the gfp ORF region from pCB16G6-Nac plasmid using primers GFPYap1For and GFPYap1Rev , an 1 kb yap1 ORF from the genomic DNA using primers Yap1For and Yap1Rev , and an 1 . 4 kb NAT cassette from plasmid pNR1 using primers Yap1NATFor and Yap1NATRev . These four fragments were subjected to second fusion PCR with primers PromoYap1For and Yap1NATRev . The final construct was transformed into the wild-type and A . brassicicola ΔtmpL mutant . To generate ToxA-yap1 overexpression construct , a 400 bp fragment of the ToxA promoter region from pCB16G6-Nac plasmid using primers ToxAFor and ToxAYap1Rev , and a 3 . 4 kb yap1 and NAT cassette from the above gfp-yap1 construct using primers Yap1overFor and Yap1NATRev were subjected to second fusion PCR with primers ToxAFor and Yap1NATRev . The final construct was transformed into wild-type and the A . brassicicola ΔtmpL mutant . All construct were subject to sequence verification with the ABI Prism 310 automated sequencer ( Applied Biosystems , Forster City , CA ) . All transformants with expected genetic integration events were identified by PCR and Southern blot analysis . To analyze the mRNA abundance of tmpL by quantitative real-time ( QRT ) PCR , 1 µg of total RNA was used for first-strand cDNA with random primers using SuperScript™ First-Strand Synthesis System ( Invitrogen™ Life Technologies , Carlsbad , CA , USA ) according to the manufacturer's instruction and diluted 1∶3 with nuclease-free water . Reactions were performed in a 25 µl volume containing 100 nM of each primer , 2 µl of cDNA ( 25 ng of input RNA ) and 12 . 5 µl of 2X iQ™ SYBR® Green Supermix ( Bio-Rad , Hercules , CA , USA ) . QRT-PCR was run on the iCycler iQ Real-Time PCR Detection System ( Bio-Rad , Hercules , CA , USA ) . After a 3 min denaturation at 95°C , samples were run for 40 cycles of 15 s at 95°C , 30 s at 60°C and 30 s at 72°C . After each run , amplification specificity was checked with a dissociation curve acquired by heating the samples from 60 to 95°C . To compare relative abundance of tmpL transcripts , average threshold cycle ( Ct ) was normalized to that of Glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) for each condition as 2−ΔCt , where −ΔCt = ( Ct , tmpL−Ct , GAPDH ) . Fold changes during conidial development and during infectious growth compared with growing fungus in liquid CM were calculated as 2−ΔΔCt , where −ΔΔCt = ( Ct , tmpL−Ct , GAPDH ) test condition− ( Ct , tmpL−Ct , GAPDH ) liquid [108] . The same real-time PCR strategy was used to analyze the expression of yap1 and other antioxidant-related genes in A . brassicicola wild-type , WT:pToxAYap1 mutant , AbΔtmpL , and AbΔtmpL:pToxAYap1 strains , except for the method of calculating relative fold change . It was determined by comparing each expression level with the one of vegetatively growing wild-type in liquid CM , where −ΔΔCt = ( Ct , gene of interest−Ct , GAPDH ) rest conditions− ( Ct , gene of interest−Ct , GAPDH ) WT , vegetative mycelia . Each QRT-PCR was conducted twice with two replicates and all the data is presented . The primer pairs for the transcript amplification of each gene were as follows: For the tmpL gene , TMPL-expFor and TMPL-expRev; yap1 , Yap1-expFor and Yap1-expRev; skn7 , SKN7-expFor and SKN7-expRev; ctt1 , CTT1-expFor and CTT1-expRev; sod1 , SOD1-expFor and SOD1-expRev; gsh1 , GSH1-expFor and GSH1-expRev; gsh2 , GSH2-expFor and GSH2-expRev; trx2 , TRX2-expFor and TRX2-expRev; gpx1 , GPX1-expFor and GPX1-expRev . For amplification of the internal control GAPDH gene , AbGAPDH-For and AbGAPDH-Rev were used . For the oxidative stress tests , A . brassicicola and A . fumigatus were grown on solid MM with or without the stress agents KO2 and H2O2 . Sensitivity to each stressor was determined by comparing the colony radius of 5-day-old A . brassicicola cultures on media containing each stressor . The tests were repeated at least three times for each condition . For the germling susceptibility assay in A . fumigatus , a protocol from the laboratory of Judith Rhodes University of Cincinnati was followed . Briefly , conidia from CEA10 , AfΔtmpL and AftmpL rec were harvested after growth on GMM plates for 3 days and incubation at 37°C . The conidia were diluted and counted in a hemocytometer . The strains were adjusted to 200 colonies per plate when 100 µl was plated . The strains were challenged in triplicate on GMM plates with 1 . 25mM H2O2 , plus the control . The plates were incubated at 30°C until microscopic germlings appeared on the plates ( about 16 hrs ) . Then the plates were overlaid with 10 ml of 1 . 25mM H2O2 or 10 ml distilled water as a control and incubated at 37°C for 10 minutes . After aspirating off the H2O2 and washing the plate twice with 10 ml of sterile distilled water the plates were returned to the 30°C incubator and incubated until colonies were large enough to count . In this study , an outbred CD1 ( Charles River Laboratory , Raleigh , NC ) strain was used . All animals were kept in specific pathogen-free housing , and all animals were handled in strict accordance with good animal practice as defined by the relevant national and/or local animal welfare bodies , and all animal work was approved by the appropriate institutional internal review board ( IACUC ) committee . Male mice ( 26 to 28 g in size , 6–8 weeks old ) , were housed five per cage and had access to food and water ad libitum . Mice were immunosuppressed with intraperitoneal ( i . p . ) injections of cyclophosphamide at 150 mg/kg 3 days prior to infection and with Kenalog injected subcutaneously ( s . c . ) at 40 mg/kg 2 days prior to infection . On day 3 post-infection ( p . i . ) , repeat injections were given with cyclophosphamide ( 150 mg/kg i . p . ) and on day 6 p . i . with Kenalog ( 40 mg/kg s . c . ) . Ten mice per A . fumigatus strains ( CEA10 , tmpL-deficient mutant , or the reconstituted strain AftmpL rec ) were infected intranasally . The mice were inoculated intranasally following brief isoflurane inhalation , returned to their cages , and monitored at least twice daily . Infection inoculum was prepared by growing the A . fumigatus isolates on GMM agar plates at 37°C for 3 days . Conidia were harvested by washing the plate surface with sterile phosphate-buffered saline-0 . 01% Tween 80 . The resulting conidial suspension was adjusted to the desired concentration of 1×106 conidia/25 µl by hemacytometer count . Mice were observed for survival for 14 days after A . fumigatus challenge . Any animals showing distress were immediately sacrificed and recorded as deaths within 24 hr . Mock mice were included in all experiments and inoculated with sterile 0 . 01% Tween 80 . Survival was plotted on a Kaplan-Meier curve and a log-rank test used to determine significance of pair-wise survival ( two-tailed P<0 . 01 ) . The animal experiments were repeated on two separate occasions with similar results . For histopathology studies , additional CD1 mice were infected with the wild-type CEA10 , tmpL mutant , or tween/saline control as described for virulence studies , and 3 mice were sacrificed at set time points of day 2 and day 4 after A . fumigatus challenge . When mice were sacrificed , lungs were immediately removed on that day . Lung tissue was fixed in 10% phosphate-buffered formalin , embedded in paraffin , sectioned at 5 µm , and stained with hematoxylin and eosin ( H&E ) or Grocott methenamine silver ( GMS ) by using standard histological techniques . Microscopic examinations were performed on a Zeiss Axioscope 2-plus microscope and imaging system using Zeiss Axiovision version 4 . 4 software . For confocal microscopy , an inverted confocal laser scanning microscope ( LSM-510 , Carl Zeiss , Göttingen , Germany ) and an argon ion laser for excitation at 488 nm wavelength and GFP filters for emission at 515–530 nm were used . Transformants carrying each fluorescent protein fusion construct were grown on solid and liquid CM . Newly formed conidia and conidiophores from solid CM plates and vegetative mycelia from liquid CM were collected for viewing . For in planta expression analysis , the lower epidermis of green cabbage cotyledons was peeled off at 4 and 12 hpi and observed . For the DsRed fusion strains , a He-Ne laser ( 543 nm excitation , 560–615 nm emission ) was used . The imaging parameters used produced no detectable background signal from any source other than from each fluorescent protein . Confocal images were captured with LSM-510 software ( version 3 . 5; Carl Zeiss ) and recorded simultaneously by phase contrast microscopy and fluorescence confocal microscopy . Brightfield and DIC images were captured with a photomultiplier for transmitted light using the same laser illumination for fluorescence . For the electron microscopy , conidia from each strain were released in sterile water and processed as described previously for transmission electron microscopy [37] . Examination was conducted with a JEM-1010 transmission electron microscope ( JEOL , Tokyo , Japan ) operating at 60 kV . For cross-sections of the green cabbage leaves inoculated with A . brassicicola wild-type and ΔtmpL strains , leaf samples were collected , embedded in Epon resin , cut thick sections with an ultramicrotome ( MT-X , RMC , USA ) , and stained with 1% toluidine blue O . The thick sections were observed using a light microscope ( Eclipse E600; Nikon , Tokyo , Japan ) . For cytological analysis , the lower epidermis of green cabbage cotyledons was peeled off 12 hpi , stained with lactophenol-cotton blue [109] , and observed by light microscopy . For the onion epidermis assay , the epidermis was peeled off , carefully washed with distilled water , then inoculated with the conidia on the adaxial surface . After 12 , 24 , 48 , and 72 hr incubation in a closed Petri dish at 100% RH , the epidermis was stained with lactophenol-cotton blue and observed by light microscopy . For the detection of callose papillae , green cabbage cotyledons inoculated with A . brassicicola were fixed and decolorized in boiling 95% ethanol , then stained in aniline blue ( 0 . 005% ( w/v ) in 0 . 07 M K2HPO4 ) . Callose was observed by mounting stained tissue in 70% glycerin and water viewing on an Axioplan Universal microscope ( Carl Zeiss Microscope Division , Oberkochen , Germany ) with a fluorescein filter set with excitation at 365 nm and emission at 420 nm . ROS was detected by staining with following solutions . For superoxide detection , nitroblue tetrazolium ( Sigma-Aldrich ) was used at 5 mg·ml−1 and the staining performed for 1 hr at room temperature prior to observation . A . brassicicola conidia collected from a nutrient-rich medium ( 5 g yeast extract , 5 g casamino acid , 340 g sucrose , 15 g agar in 1 L deionized water ) and fungus-inoculated leaves at each time point were subjected to the staining . For detection of ROS other than superoxide , A . brassicicola and A . fumigatus conidia were collected from PDA and GMM media , respectively , and stained with 5 µg·ml−1 5- ( and 6 ) -carboxy-2′ , 7′-dichlorodihydrofluorescein diacetate ( carboxy-H2DCFDA; Molecular Probes , Eugene , OR ) . The intracellular distribution of ROS in appressoria was visualized after staining with 2 mg/ml DAB 2 hr , followed by a short rinse PBS . Sequence data for tmpL can be found in the GenBank data libraries under accession number EU223383 for A . brassicicola and EDP49089 for A . fumigatus . | The critical roles of reactive oxygen species ( ROS ) in fungal development and virulence have been well established over the past half a century since the first experimental detection of hydrogen peroxide in fungal cells by Bach ( 1950 ) . In the cell , ROS act as signaling molecules regulating physiological responses and developmental processes and are also involved in sophisticated virulence processes for many pathogenic fungi . Therefore , uncovering the biological roles of cellular ROS appears to be very important in understanding fungal development and virulence . Currently we have limited knowledge of how intracellular ROS are generated by fungal cells and which cellular ROS regulatory mechanisms are involved in establishing homeostasis . In this study we describe a novel protein , TmpL , involved in development and virulence in both plant and animal pathogenic fungi . In the absence of TmpL , dysregulation of oxidative stress homeostasis in both fungi caused developmental and virulence defects . Therefore , elucidating the role of TmpL presents an opportunity to uncover a common pathogenicity mechanism employed by both plant and animal pathogens and to develop efficient and novel therapeutics for both plant and animal fungal disease . Our findings provide new insights into mechanisms underlying the complex web of interactions between ROS and cell differentiation and the involvement of ROS for both plant and animal fungal pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"biology/microbial",
"physiology",
"and",
"metabolism",
"plant",
"biology/plant-biotic",
"interactions",
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics",
"respiratory",
"medicine/respiratory",
"infections",
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses",
"genetics",
"and",
"genomics/functional",
"genomics",
"cell",
"biology/microbial",
"growth",
"and",
"development",
"biochemistry/bioinformatics",
"cell",
"biology/gene",
"expression"
]
| 2009 | TmpL, a Transmembrane Protein Required for Intracellular Redox Homeostasis and Virulence in a Plant and an Animal Fungal Pathogen |
Ivermectin is the only drug currently recommended for the treatment of onchocerciasis , the second leading infectious cause of blindness in the world . This drug kills only the first stage larvae—microfilariae ( mf ) of Onchocerca volvulus and is to be used cautiously in areas where Loa loa is prevalent because of severe adverse events observed with coinfected patients . This study investigated the anti-filarial activities of two Cameroonian medicinal plants , Lantana camara and Tamarindus indica locally used to treat onchocerciasis . Twelve ( 12 ) extracts were prepared and tested in vitro on the bovine model parasite , O . ochengi as well as L . loa mf . Both mf and adult male worm viabilities were assessed by motility scoring , while adult female worm viability was determined biochemically by standard MTT/formazan colorimetry . Cytotoxicity and acute toxicity were determined respectively , in monkey kidney epithelial cells and in BALB/c mice . Pure compounds were isolated by LC/MS using a bio-assay guided strategy . All the extracts showed 100% activity at 500 μg/mL against O . ochengi adult worms and mf . The highest activity against O . ochengi was observed with the hexane extract of L . camara leaves ( LCLhex ) , with IC50 of 35 . 1 μg/mL for adult females and 3 . 8 μg/mL for the mf . Interestingly , this extract was more active against O . ochengi mf than L . loa mf . Further studies on the extracts led to the isolation of lantadene A from the methylene chloride extract of L . camara leaves , with IC50s of 7 . 85 μg/mL for adult males , 10 . 38 μg/mL for adult females , 10 . 84 μg/mL for O . ochengi mf and 20 . 13 μg/mL for L . loa mf . We report for the first time the anti-onchocercal activities of these locally consumed medicinal plants and lantadene A , a potential lead for further development as an onchocerciasis cure .
Onchocerciasis ( river blindness ) is a blinding and debilitating disease caused by the parasitic nematode , Onchocerca volvulus . According to estimates of the World Health Organization ( WHO ) [1] , 37 million people are infected , 800 , 000 visually impaired and 270 , 000 blinded . Adult worms of O . volvulus can live for up to 15 years in subcutaneous nodules ( onchocercoma ) and produce millions of microfilariae ( mf ) which parasitize skin and eye tissues , resulting in major pathologies such as intense and often unbearable itching , disfiguring dermatitis , atrophy , visual impairment and blindness [2] . The microfilaricide , ivermectin was shown to be safe and effective in the treatment of onchocerciasis and is currently the only recommended drug for control of the disease by a mass drug administration ( MDA ) strategy [3] . The emergence of animal parasite strains resistant to ivermectin and an abundance of reports of resistance or low response rates of O . volvulus mf to the drug are worrisome . Additionally , the use of ivermectin in MDA in areas of high Loa loa co-endemicity is limited due to severe adverse events ( including encephalopathy and death ) observed with some coinfected patients [4] . Since ivermectin is only effective against the mf , prolonged annual therapy for at least 10 to 15 years is required to interrupt transmission and clear onchocerciasis from a human population [5] . Therefore , there is the need for a safe and more effective macrofilaricidal drug for the cure of onchocerciasis or an alternative microfilaricide , preferably one that does not kill L . loa mf . Since onchocerciasis is a neglected tropical disease , such a drug has been difficult to find with the conventional for-profit pharmaceutical company approach , requiring alternative strategies to aid its discovery and development . One strategy employed has been the exploitation of medicinal plants and other natural materials as alternative medicines or for the identification of novel potential drug leads . It has been shown that medicinal plants play a very important role in the health care needs of rural populations in Africa because they are cheap and readily available locally [6 , 7] . The majority of drugs active against infectious agents are in fact derived from natural products [8] , including ivermectin derived from Streptomyces avermitilis [9] and artemisinin from the medicinal plant , Artemisia annua [10] . Previous studies have revealed the filaricidal properties of several Cameroonian medicinal plants [11–13] . Tamarindus indica lotions and extracts are widely used by indigenes to treat conjunctivitis , dysentery , jaundice , hemorrhoids and onchocerciasis . Lantana camara contains principles active against Mycobacterium tuberculosis and has been used in the traditional treatment of onchocerciasis in parts of Cameroon [14] . This study thus sought to investigate the filaricidal properties of extracts from Lantana camara ( Verbenaceae ) and Tamarindus indica ( Leguminosae ) against cattle derived O . ochengi , the closest known relative of O . volvulus [15] , and against L . loa mf , in order to assess their acclaimed activities and their possible use as sources of new drug leads for onchocerciasis .
Ethical clearance ( No . 2013/11/371/L/CNERSH/SP ) and administrative clearance ( No . 631–06 . 14 ) for blood collection from L . loa infected humans were obtained from the Cameroon National Ethics Committee and the Ministry of Public Health , respectively . All subjects of age 20–55 granted written and informed consent before any blood for diagnosis or worm preparation was collected . Both L . camara and T . indica were collected in January , 2013 based on ethno pharmacological information from Bafoussam and Oshei communities in the West and North West Regions of Cameroon , respectively . Voucher specimens were taken to the Yaoundé herbarium where they were authenticated by Mr . Onana Jean Marie and voucher numbers were assigned to them ( L . camara: 25900 SRF CAM; T . indica: HNC/42429 ) . The leaves , stem bark and roots of each plant were air dried and ground to fine powder using a grinding mill . Each powder was macerated for 48 hours , sequentially , in hexane , methylene chloride , and methanol . The filtrate was concentrated using a rotary evaporator ( BUCHI Rotavapor R-200 , Switzerland ) and crude extracts weighed and preserved at -20°C for further use . The percentage yield in extract was calculated using the following formula: %yield= ( Weightofcrudeextractx100 ) /Weightofdrygroundedplantmaterial A stock solution of 25 mg/mL was prepared in >99 . 8% DMSO ( Sigma , Germany ) and kept at -20°C until tested in biological assays . Worms were isolated from umbilical areas of infected cattle skin as previously described by Cho-Ngwa et al . [11] . Briefly , cattle skin containing palpable nodules obtained from the butchery in Douala Cameroon were washed with soap and rinsed with distilled water . The skin was then sterilized with 70% ethanol after which the nodules were carefully opened and the entire nodular content removed and submerged in 2 mL of complete culture medium ( CCM ) comprising of RPMI-1640 ( SIGMA cat: R0883 ) , supplemented with 25 mM HEPES , 2 g/L sodium bicarbonate , 2 mM L-glutamine , 5% heat inactivated new born calf serum ( SIGMA Cat: N4762 ) , 200 units/mL penicillin/ 200 μg/mL streptomycin and 0 . 25 μg/mL amphotericin B , pH 7 . 4 in 12-well culture plates . The worms were left in cultures in a HERACELL-CO2 incubator ( Thermo Fisher , UK ) overnight and checked for any contamination before drugs were added . Monkey kidney epithelial cells ( LLC-MK2 ) obtained from the American Type Culture Collection ( ATCC ) were proliferated in 96-well microtitre plates in CCM medium at 37°C in 5% CO2 humidified air . At confluency , the cells served as feeder layers for the mf cultures . This was prepared as described by Bianco et al . , [16] with slight modifications . Briefly , fresh pieces of umbilical cattle skin were obtained from the butchery and washed thoroughly . Few skin snips from different locations of the skin were obtained and incubated in small amounts of culture medium for 15 minutes , after which the emergent mf were qualified and quantified using an inverted microscope and standard atlases for reference [17] . The remainder of a selected piece of skin was shaved , rinsed and sterilized with 70% ethanol and sliced into thin slivers . The slivers were incubated in CCM for 2 hours , and the emergent highly motile O . ochengi mf were concentrated by centrifugation . The mf were transferred into 96 -well microtitre plates ( 15 mf /100 μL/ well ) already containing fully confluent LLC-MK2 cell layer in 100 μL of CCM and monitored for viability and sterility for 24 hours before addition of test and control compounds . Identification of L . loa mfs was done using standard atlases after staining with giemsa and observing the slide under the microscope [18] . Whole blood was collected in an EDTA tube from patients not receiving treatment and transported immediately to the Laboratory . The mf load was determined with the aid of an inverted microscope after diluting a portion of the blood in RPMI-1640 medium . After this the blood was diluted according to the number of mf present at initial count so as to obtain a total of 15 mf/100 μL/well . After dilution , the mf were distributed in 96-well plate and monitored for 24 hours before addition of test and control compounds . Extracts were tested at 500 μg/mL in triplicates in CCM . Auranofin at 10 μM , which had previously shown activity against O . ochengi adult worms and mf [19] was used as positive control , while negative control wells received the diluent , 2% DMSO only , previously shown to have no effect on parasite viability . The worm cultures with drug were incubated for 168 hours ( 7 days ) , at 37°C in 5% CO2 atmosphere . On the last day of incubation , the female worms were removed and incubated in 500 μL of 0 . 5 mg/mL MTT for 30 minutes . Inhibition of formazan formation from MTT directly correlates with worm death . The worms were blotted on absorbent paper and observed visually for blue coloration against a white background . Scores based on activity were assigned , ranging from 100% inhibition of formazan formationcompletely pale yellow worm , 90% inhibition; only one or few spots of blue color seen on worm , 75% inhibition; about 75% of worm remained pale yellow , 50% inhibition; about 50% of worm remained pale yellow , to 25% inhibition; near total blue coloration , to 0% inhibition; for total blue color on worm for inactive compound . Adult male worm motility was evaluated using an inverted microscope . Scores were attributed to the worms using the following code: Vigorous or normal movement of whole worm , corresponding to 0% inhibition of worm motility; near normal movement of whole worm or 25% inhibition of worm motility; whole body of worm motile but sluggish i . e . 50% inhibition of worm motility; only head or tail of worm moving i . e . 75% inhibition of worm motility; completely immotile worm i . e . 100% inhibition of worm motility . Extracts with 100% activity at primary screens were re-tested as described under primary screens and at serial dilutions of seven concentrations ( from 500 to 7 . 8125 μg/mL ) , in order to determine the IC50 values . The IC50 assays were done in triplicates and each experiment repeated for confirmation . The means of all activities at a concentration were calculated and used in the statistical analyses . GraphPad prism version 6 . 0 ( GraphPad Software , CA , USA ) was used to generate dose response curves from which the IC50 values were obtained . Assays were conducted at 500 μg/mL in duplicates in the 96-well microtitre plates . The positive control drug was amorcazine at 30 μM and negative control was the diluent ( DMSO ) . The mf were incubated with drug for 120 hours in a total of 200 μL of medium . Mf viability was assessed by microscopy once every day and motility inhibition scores were recorded as: 100% ( immotile ) , 75% ( only head or tail shaking ) , 50% ( sluggish ) , 25% ( almost vigorous motility ) , 0% ( vigorous motility as with negative control ) . The day 5 data were used in determining drug activity . Motility inhibition correlates to drug activity . Extracts showing 100% activity in the primary screens were re-tested as described under primary screens at 8 serial dilutions ( from 500–3 . 9 μg/mL ) to determine the IC50 values . All assays were repeated at least once . The selectivity index ( SI ) of each extract was calculated as the ratio of the IC50 of the extract on mammalian cell ( termed CC50 ) to the IC50 on parasites . All the extracts were screened against L . loa mf and the IC50 values determined . This was done at serial dilutions of eight concentrations ( 500–3 . 9 μg/mL ) , and according to the protocol used to screen extracts against O . ochengi mf . All assays were repeated at least once for confirmation of results . Data were analyzed using GraphPad prism 6 . The statistical significance of differences in means between the effects of extracts at various concentrations on parasites were determined by one-way analysis of variance ( ANOVA ) , followed by Newman-Keuls multiple comparison tests . A value of p < 0 . 05 was considered significant . Active extracts were chromatographed on a Sephadex LH-20 column using 4:1 MeOH/CH2Cl2 as eluent to give fractions A-E which were screened in quadruplicates at 50 μg/mL on all the developmental stages of O . ochengi and on L . loa mf . Fractions C and D were combined and further fractionated on a silica gel column with a gradient from 100% hexane to 100% EtOAc to give sub-fractions A-J and screened at the same concentration . Fraction D was recrystallized twice in MeOH to obtain needle-shaped crystals . The crystals were identified as lantadene A by analysis of their NMR and MS data ( Fig 1 ) . Lantadene A: Colorless crystal ( 4 . 0 mg ) ; 1H NMR ( 600 MHz , CDCl3 ) δ 5 . 97 ( m , 1H ) , 5 . 35 ( m , 1H ) , 5 . 06 ( m , 1H ) , 3 . 02 ( dm , J = 13 . 9 Hz , 1H ) , 2 . 53 ( m , 1H ) , 2 . 35 ( dm , J = 14 . 2 Hz , 1H ) , 1 . 99 ( m , 1H ) , 1 . 95 ( dm , J = 6 . 5 Hz , 3H ) , 1 . 90 ( m , 1H ) , 1 . 87 ( dm , J = 13 . 5 Hz , 2H ) , 1 . 79 ( br . d , J = 14 . 1 Hz , 2H ) , 1 . 74 ( m , 1H ) , 1 . 73 ( br . d , J = 5 . 6 Hz , 3H ) , 1 . 65 ( m , 1H ) , 1 . 49 ( m , 1H ) , 1 . 46 ( m , 2H ) , 1 . 39 ( dm , J = 6 . 7 Hz , 2H ) , 1 . 28 ( m , 1H ) , 1 . 27 ( m , 2H ) , 1 . 26 ( m , 2H ) , 1 . 15 ( br . s , 3H ) , 1 . 07 ( br . s , 3H ) , 1 . 03 ( br . s , 3H ) , 1 . 02 ( br . s , 3H ) , 0 . 98 ( br . s , 3H ) , 0 . 87 ( br . s , 3H ) , 0 . 80 ( br . d , J = 5 . 7 Hz , 3H ) ; 13C NMR ( 150 MHz , CDCl3 ) δ217 . 9 , 179 . 7 , 166 . 4 , 143 . 3 , 139 . 3 , 127 . 8 , 122 . 7 , 76 . 0 , 55 . 5 , 50 . 8 , 47 . 6 , 47 . 1 , 46 . 1 , 42 . 2 , 39 . 4 , 39 . 3 , 38 . 6 , 37 . 9 , 37 . 0 , 34 . 3 , 33 . 9 , 32 . 4 , 30 . 2 , 27 . 8 , 26 . 7 , 26 . 3 , 26 . 0 , 24 . 4 , 23 . 7 , 21 . 7 , 20 . 8 , 19 . 7 , 17 . 1 , 15 . 9 , 15 . 3; HRESIMS [M+Na]+m/z 575 . 3708 ( calcd for C35H52O5Na , 575 . 3712 ) . Lantadene A was first tested at 50 μg/mL in the primary screens against the males , females and mf of O . ochengi and L . loa mf . Thereafter , it was tested in secondary screens from 40 μg/mL—0 . 31 μg/mL for adult worms , 40 μg/mL—0 . 16 μg/mL for O . ochengi mf and 40 μg/mL– 10 μg/mL for Loa loa mf in order to determine its IC50 values , using the same assays described for the extracts . Cytotoxicity of extracts with anti-Onchocerca activities was assessed on LLC-MK2 cells , microscopically , on day 5 of the mf assay . Living cells were flattened out and attached to the culture plate , while dead cells were rounded up and detached from the plate . The IC50 values for these cells were estimated from the morphological deformation data . This test was conducted in accordance with the OECD guideline for testing of chemicals [20] and the animal protocol IACUC No UBAP2014-001 was approved by the Animal Care and Use Committee of the Faculty of Science , University of Buea . The three most active extracts ( LCLmc , TILmc and LCLhex ) were tested for acute toxicity in BALB/c mice . Thirty-two animals of approximately 21 . 67g body weight each were divided into 4 groups ( each group consisting of 1 subgroup of 4 males and 1 subgroup of 4 females ) . Each of the 3 treatment groups received one of the extracts at a limit dose of 2000 mg/kg body weight , administered orally in a maximum volume of 280 μL of corn oil per animal; while the control group received the diluent only . The animals were observed for any changes in physical activity , food intake , water intake , stool sample , loss of fur , sensitivity to sound , sensitivity to pain , motility and mortality , every day for 14 days . The phytochemical properties of the active extracts were determined using standard procedures: Mayer and freshly prepared Dragendoff’s reagents for alkaloïds , Liebermann-Buchard test for triterpenoids and sterols , FeCl3 and K3Fe[ ( CN ) 6] for phenols , Shinoda’s test for flavonoids , frothing test for saponins [21] .
The twelve ( 12 ) crude extracts obtained from the 2 plants , L . camara and T . indica , were first tested at 500 μg/mL in primary screens against O . ochengi worm stages . All of the extracts showed 100% activity against adult worms and mf . The extracts were further screened at various concentrations on adult worms and mf in order to determine their IC50s . The hexane and methylene chloride extracts of L . camara leaves ( LCLhex and LCLmc , respectively ) were the most active against adult male worms with IC50s of 7 . 3 and 7 . 8 μg/mL , and O . ochengi mf with IC50s of 3 . 8 and 3 . 9 μg/mL , respectively . Moreover , LCLhex and methylene chloride extract of T . indica leaves ( TILmc ) were the most active extracts against female worms with IC50s of 35 . 1 μg/mL and 62 . 5 μg/mL , respectively . Seven of the twelve extracts had lower activities ( higher IC50s ) against L . loa mf than O . ochengi mf ( Table 1 ) . LCLhex , LCLmc , and TILmc had IC50s of 62 . 5 μg/mL , 55 . 6 μg/mL and 64 . 5 μg/mL , for L . loa respectively . LCLhex and LCLmc had IC50 values for L . loa 16 . 4 and 14 . 3 times higher than that for O . ochengi mf , respectively ( Fig 2 ) . Comparing the mean activities of the different extracts tested against males , females and microfilariae of O . ochengi and L . loa , we observed a dose dependent effect from 500–3 . 90625 μg/mL . At a fixed extract concentration ( 250–7 . 8125 μg/mL ) there were significant differences ( p < 0 . 05 ) between the different types of extracts . For the same extract type at a fixed concentration there was a difference in the response of males , females and microfilariae of O . ochengi and L . loa ( Table 2 , 3 , 4 and 5 ) . The drug concentrations inducing cytotoxicity in 50% of cells ( CC50 ) were 46 . 4 μg/mL , 7 . 8 μg/mL and 7 . 8 μg/mL for TILmc , LCLhex and LCLmc , respectively . Thus , the selectivity index ( SI ) values of the extracts for adult worms and mf ranged from 0 . 12–2 . 07 ( Table 6 ) . In general , the cytotoxicity assay demonstrated that about 57% of the 12 extracts had SI values below 1 , a clear indication of cytotoxic tendencies for the crude preparations . LCLhex , TILmc and LCLmc were selected for acute toxicity studies in BALB/c mice at a limit dose of 2 , 000 mg/kg body weight . No sign of acute toxicity was noticed in BALB/c mice . The average weights of the mice increased from 21 . 67g pre-treatment to 26 . 67g post-treatment . No change was observed in the physical appearance of the animals throughout the 14-day study period . Phytochemical screening revealed different classes of secondary metabolites present in the three most active extracts ( Table 7 ) . Further fractionation of the active extracts , LCLhex , LCLmc and TILmc yielded sub-fractions that were each screened on O . ochengi mf and adults at 50 μg/mL . The 6 fractions from LCLhex were inactive against the adult worms and mf , while 5 fractions from TILmc showed moderate activity against adult male and female worms , although inactive against them . Five fractions were obtained from LCLmc and marked activity was observed with two of the fractions ( C and D ) at 50 μg/mL against adult worms and mf . Combining these two fractions and fractionating further , 11 sub-fractions were obtained . When tested at 50 μg/mL against the male , female and mf of O . ochengi , 6 sub-fractions [C+D ( C , D , D2 , F , G , H ) ] were very active , 2 sub-fractions [C+D ( E , I ) ] moderately active , while the others [C+D ( A , B , J ) ] were inactive . From the above active fractions , the sub-fraction C+D-D4 was obtained , which showed 100% activity at 50 μg/mL ( Table 8 ) . Further analysis by liquid chromatography/mass spectrometry ( LC/MS ) and recrystallisation of combined fractions C+D-D4 enabled isolation of the major compound , lantadene A ( Fig 1 ) . This pure compound showed 100% activity against O . ochengi and L . loa mf at 50 μg/mL in the primary screens; and for the secondary screens , IC100 of 20 μg/mL for the adult worms and IC50s of as low as 7 . 85 μg/mL for the adult male worms; while the IC100 and IC50 for L . loa mf were 30 μg/mL and 20 . 13 μg/mL , respectively ( Fig 3 ) .
In this study , we investigated the in vitro filaricidal activities of extracts of L . camara and T . indica; carried out a bioassay guided fractionation for identification of new drug leads for onchocerciasis and then isolated and determined the filaricidal properties of lantadene A from L . camara for the first time . Dose-dependent activity relationships were observed with the twelve extracts with IC50s ranging from 385 . 2 down to 3 . 8 μg/mL ( Table 1 ) . This indicates high anti-onchocerca properties of the plant extracts . The anti-filarial properties of lantadene A , which completely killed the parasites at 20 μg/mL were deemed encouraging , necessitating further studies on the compound . In the secondary screens , the L . camara hexane extract ( LCLhex ) and the methylene chloride extract ( LCLmc ) showed significant differences against male worms and O . ochengi mf when compared to all the extracts . For the adult female worms , significant differences were observed with LCLhex and TILmc as compared to the other extracts . These differences of a particular extract tested at a single concentration acting differently on the different parasites might be due to the differences in proteins being expressed at the different stages or species . Also the differences observed when testing different extracts on one parasite shows the difference in composition of the extracts . Overall , higher activities were observed with the non-polar extracts than with the polar ones , corroborating previous findings that showed non-polar compounds , including essential oils to be nematicidal [22 , 23] . It is therefore suggested that , traditional healers find a way of reducing the polarity of their usual aqueous solvents in preparing the corresponding herbal medicines . But this must be confirmed in any clinical trials . The reduction in polarity could be by way of addition of suitable edible oils to the extracting media . After confirming the filaricidal activity of the extracts , it was deemed necessary to investigate and obtain preliminary data on their safety . About 57% of the extracts tested were more toxic to the monkey kidney epithelial cells than to the worms as reflected in their SI < 1 , probably due to complexity of the extracts , although none of the mice died after administration of the selected extracts at limit dose . These results underscore the importance of carrying out full toxicity and dosage studies on traditional medicines , which may generate new problems after the patient might have been cured of the original problem . Serious adverse events ( SAE ) associated with incidental killing of L . loa mf in blood during treatment of patients coinfected with O . volvulus and L . loa have been reported [24] , suggesting a need for drugs that will selectively kill O . volvulus without affecting or only moderately affecting L . loa mf . Interestingly , the most active extracts against O . ochengi adult males and mf ( LCLhex and LCLmc ) were less active for L . loa mf ( Fig 2 ) . This indicates that these extracts could be potential sources of such selective anti-Onchocerca drugs . To identify novel leads for the development of new drugs for onchocerciasis , bio-assay guided fractionations of the extracts were carried out . Fractions from LCLhex and TILmc each showed no activity and moderate activity against the parasites respectively , although the whole extract itself was highly active . This implies that active principles from these extracts may be unstable , unable to withstand the fractionation process; or may be acting in synergy to provide the anti-parasitic activity , or even might have been over retarded in the column or missed out in the chromatographic process . Two fractions ( C+D ) from LCLmc showed marked activity against all the developmental stages of O . ochengi and L . loa mf when tested at 50 μg/mL . These yielded the compound lantadene A . The compound had previously been isolated from L . camara [25] . It is a pentacyclic triperpenoid , with molecular weight 552 . 78 , is only sparingly soluble in water and crystallizes in methanol . It was shown to be active against tumors , Leishmania and soil nematodes [26 , 27] . At least 12 triterpenoids have been isolated and reported from L . camara , with some of their analogues being less toxic on cells and others having CC50s which go as low as <1 μM [26] . Like most or all drugs , lantadene A is toxic at higher doses , especially to livestock . Liver injury occurred after sheep were injected intravenously with lantadene A . A single dose of 1–3 mg/kg of the compound caused mild hepatocellular injury in sheep . Higher doses resulted in hepatic necrosis . It did not require metabolism in the alimentary tract for toxicity in sheep [25] . Reports of human toxicity by L . camara are rare . Most children with exploratory exposures to the plant remain asymptomatic . In the minority who develop mild effects , gastrointestinal irritation was most common . It is not known what substance produces these mild toxic effects in humans , but it does not appear that lantadene A or phototoxins contained therein are responsible [23] . Additional studies will therefore be needed to further investigate the anti-filarial effects of the other numerous compounds reported to be present in the plant by exploiting the structure–activity relationships to design better analogues that could be more active and non-toxic lead compounds for onchocerciasis treatment . The apparent low level of toxicity of lantadene A to humans could be supported by the current use of L . camara in traditional medicine of the present study population . The IC50s of 7 . 85 μg/mL for adult males , 10 . 38 μg/mL for adult females and 10 μg/mL for monkey kidney epithelial cells obtained for lantadene A were deemed encouraging suggesting that it could be a potential lead for the development of an onchocerciasis cure . Also the IC50 of 20 . 13 μg/mL of lantedene A for L . loa mf as compared to the IC50 of 10 . 84 μg/mL for O . ochengi mf is an indication that lantadene A is less active against L . loa and so could be somewhat safe to administer in areas of Loa loa co-endemicity . Further bioassay-guided fractional studies on the extracts are also recommended as we believe that we have not yet isolated all the main anti-Onchocerca principles from them . This study has thus , reported for the first time the anti-filarial activity of these medicinal plants and of lantadene A . | Onchocerciasis is a chronic disease of humans that affects mainly the skin and eyes . It is an insect-borne disease , caused by a nematode worm , Onchocerca volvulus . It is a public health problem and an obstacle to socio-economic development in affected communities . There is currently no vaccine , and no adult worm drug to cure the infection . The only recommended drug , ivermectin can give short-term relief by killing the microfilariae of the parasite but cannot always be safely administered in mass drug administration ( MDA ) in areas where another filarial worm , Loa loa is co-endemic . To prevent infection and completely eradicate the disease there is an urgent need for alternative drugs that can kill the adult worms but to a lesser extent Loa loa . In the past , medicinal plants have served as good starting points for the development of drugs . In the present study , we determined the filaricidal properties of extracts , chromatographic fractions of Lantana camara and Tamarindus indica and lantadene A against Onchocerca ochengi , the bovine model of the parasite and L . loa . Interestingly , most of the active extracts were more active against O . ochengi than on L . loa mf . No death was recorded following oral administration of the active extracts to BALB/c mice . Lantadene A is a potential lead molecule for the development of a cure for onchocerciasis . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"onchocerca",
"volvulus",
"chemical",
"compounds",
"pathology",
"and",
"laboratory",
"medicine",
"helminths",
"methylenes",
"hexanes",
"tropical",
"diseases",
"parasitic",
"diseases",
"animals",
"onchocerca",
"toxicology",
"toxicity",
"medicinal",
"plants",
"neglected",
"tropical",
"diseases",
"onchocerciasis",
"plants",
"hydrocarbons",
"chemistry",
"loa",
"loa",
"chlorides",
"helminth",
"infections",
"eukaryota",
"nematoda",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
]
| 2018 | Filaricidal properties of Lantana camara and Tamarindus indica extracts, and Lantadene A from L. camara against Onchocerca ochengi and Loa loa |
Hearing and vestibular function depend on mechanosensory staircase collections of hair cell stereocilia , which are produced from microvillus-like precursors as their parallel actin bundle scaffolds increase in diameter and elongate or shorten . Hair cell stereocilia contain multiple classes of actin-bundling protein , but little is known about what each class contributes . To investigate the roles of the espin class of actin-bundling protein , we used a genetic approach that benefited from a judicious selection of mouse background strain and an examination of the effects of heterozygosity . A congenic jerker mouse line was prepared by repeated backcrossing into the inbred CBA/CaJ strain , which is known for excellent hearing and minimal age-related hearing loss . We compared stereocilia in wild-type CBA/CaJ mice , jerker homozygotes that lack espin proteins owing to a frameshift mutation in the espin gene , and jerker heterozygotes that contain reduced espin levels . The lack of espins radically impaired stereociliary morphogenesis , resulting in stereocilia that were abnormally thin and short , with reduced differential elongation to form a staircase . Mean stereociliary diameter did not increase beyond ∼0 . 10–0 . 14 µm , making stereocilia ∼30%–60% thinner than wild type and suggesting that they contained ∼50%–85% fewer actin filaments . These characteristics indicate a requirement for espins in the appositional growth and differential elongation of the stereociliary parallel actin bundle and fit the known biological activities of espins in vitro and in transfected cells . The stereocilia of jerker heterozygotes showed a transient proximal-distal tapering suggestive of haploinsufficiency and a slowing of morphogenesis that revealed previously unrecognized assembly steps and intermediates . The lack of espins also led to a region-dependent degeneration of stereocilia involving shortening and collapse . We conclude that the espin actin-bundling proteins are required for the assembly and stabilization of the stereociliary parallel actin bundle .
A stunning example of cytoskeleton-mediated morphogenesis is the formation of hair cell stereocilia , which act as primary mechanosensory detectors in the auditory and vestibular systems [1] , [2] . Stereocilia are fingerlike projections that contain a specialized cytoskeletal element , the parallel actin bundle [3] , aligned axially at their core . The parallel actin bundle , which consists of hexagonally packed unidirectional actin filaments cross-linked by actin-bundling proteins to produce a regular ∼12–13 nm ( center-to-center ) interfilament spacing , exhibits the properties of a molecular scaffold that sets the dimensions of stereocilia and influences their mechanical properties [3]–[5] . During development , highly precise staircase collections of stereocilia are produced from microvillus-like precursors as their parallel actin bundle scaffolds selectively undergo an increase in diameter , through the addition of more actin filaments to the parallel actin bundle , and their constituent actin filaments elongate or shorten [4] , [5] . The plasma membrane of the stereocilium remains in close proximity to the parallel actin bundle throughout morphogenesis , so that increases in stereociliary diameter reflect increases in the number of actin filaments in the parallel actin bundle and changes in stereociliary length correspond to changes in the length of the actin filaments in the parallel actin bundle [6] . The dimensions of stereocilia vary in a remarkably regular way , not only within a given collection , but also according to hair cell type and position in the cochlea or vestibular system [7]–[10] . This attests to an impressive degree of spatial precision in actin-cytoskeletal regulation . A growing list of deaf mutant mice with malformed stereocilia demonstrates the importance of stereociliary morphogenesis to hair cell mechanoelectrical signal transduction [11] , [12] . The modifications in parallel actin bundle dimensions that underlie stereociliary morphogenesis are presently thought to involve actin-bundling proteins [13]–[17] , actin-capping proteins [18]–[20] , unconventional myosin motors and their cargoes [21]–[23] . Although multiple classes of actin-bundling protein have been identified in hair cell stereocilia , relatively little is known about what each class contributes [13]–[17] . One class of actin-bundling protein implicated in hair cell stereocilia is the espin family [24] . Discovered originally in Sertoli cell junctional plaques [25] , espins are encoded by a single gene , but are produced in multiple isoforms [24]–[27] . Espins bind to and cross-link actin filaments into parallel actin bundles in vitro with high affinity and in a Ca2+-insensitive manner [28]–[30] , exert a potent cooperative effect on the twist of actin filaments in parallel actin bundles [31] and elicit a dramatic , concentration-dependent elongation of parallel actin bundles in cells [27] , [32]–[34] . In hair cell stereocilia , espin antibody staining is detected along the length of the parallel actin bundle in the body of the stereocilium , both in adulthood and during morphogenesis , but not in the rootlet [14] , [16] , [32] , [35] , [36] . Espin expression and accumulation in stereocilia are hallmarks of hair cell differentiation in situ and by stem cells in culture [35] , [37] , [38] . In addition , the espin gene is the target of mutations associated with deafness and vestibular dysfunction , including the jerker mutation in mice [14] , [39] , [40] . The jerker mutation is a spontaneous mutation noted in the collection of a mouse fancier and first described in 1941 [41]–[43] . Homozygous jerker mice exhibit the stereotyped shaker-waltzer behavior indicative of hair cell defects , including deafness , circling , head tossing and hyperactivity . In 2000 , the jerker mutation was shown to be a frameshift mutation in the espin gene ( c . 2426delG; Espnje ) on mouse chromosome 4 [14] , and this was verified by independent physical mapping studies [44] . Because homozygous jerker mice lack espin proteins and jerker heterozygotes contain approximately half-normal espin levels [14] , the examination of jerker mice promises to reveal a great deal about the functions of espins . Earlier studies examining stereociliary ultrastructure in inbred jerker mice with uncharacterized genetic backgrounds detected the degeneration of stereocilia and loss of hair cells in jerker homozygotes [33] , [45]–[47] . Although the results were encouraging , these earlier studies did not compare wild-type mice of the same genetic background and either were not systematic or examined only a single hair cell type and inner ear location . In addition , the effects on the vestibular system [45] were not investigated in detail . Importantly , the detection of a related group of degenerative changes in jerker heterozygotes with later onset [45]–[48] was difficult to reconcile with the presumed recessive nature of the jerker mutation . This naturally raised concerns about possible complications owing to strain-specific genetic modifier effects or age-related hearing loss , which have been detected in a number of mouse strains [49] . A recent proteomic analysis of stereocilia detected espins at lower levels than some other actin-bundling proteins [17] , raising additional questions about the roles of espins . To help elucidate the roles of espins in hair cell stereocilia , we have carried out a systematic scanning electron microscopic study of hair cell stereocilia examining a congenic jerker mouse line we prepared using the CBA/CaJ inbred strain . The CBA/CaJ strain was chosen for the genetic background because CBA/CaJ mice exhibit excellent hearing and minimal age-related hearing loss [50] . Unlike earlier studies , we compared jerker homozygotes and heterozygotes to wild-type mice of the same genetic background , analyzed hair cells from multiple inner ear locations in the cochlea and vestibular system , examined specimens without metal coating , measured stereociliary width and length and focused on the critical period of early postnatal development . We determined that the absence of espin proteins drastically alters stereociliary morphogenesis , resulting in marked decreases in stereocilium diameter , length and stability . In addition , we uncovered an informative group of transient developmental defects in jerker heterozygotes , which contain reduced espin levels .
Examination of the stereocilia of vestibular hair cells in +/je mice revealed an unexpected developmental defect: transient tapering . As shown in Figure 11C , the width of stereocilia on the extrastriolar hair cells of +/je mice ( dashed line ) was intermediate to those in je/je and +/+ mice at P0 , but eventually increased to become highly similar to that in +/+ mice by P20 . What is remarkable is that this increase in stereociliary width was gradual and took place first in the proximal part of stereocilia and later in the distal part , resulting in stereocilia that were transiently tapered from P0 through P10 ( Figure 11C and Figure 12A ) . A similar gradual tapering of stereocilia was observed in the peripheral zone of the cristae ampullares in +/je mice at P5 ( Figure 12D ) . Stereociliary tapering , but of a more extreme and abrupt nature , was also seen in the central zone of the cristae ampullares in +/je mice at P0 , P5 and P10 ( P5 shown in Figure 12B and 12C ) . The distal segment of these stereocilia was often dramatically thinner than the proximal segment , giving the appearance of a candle with a wick ( Figure 12B and 12C ) . Notably , up to a length of ∼4 µm , the width of the proximal segment of these stereocilia was similar to that in +/+ mice ( compare Figure 12B and Figure 7A ) . By P20 , few of these extremely thin distal segments remained ( Figure 12F ) , suggesting that most had grown longer and widened sufficiently to match their proximal segments . Also evident in the central zone of P5 +/je mice were stereocilia with eccentric protruding distal tips , which were suggestive of intermediates caught in a relatively early stage of additional elongation ( Figure 12E , arrowheads ) . A close scrutiny of inner hair cells also revealed the transient tapering of stereocilia in +/je mice . This tapering was especially evident for stereocilia in the tallest row in the apical region of the cochlea at P5 ( Figure 13D ) , but was also detected in the middle region ( Figure 13B ) . The tapering was not detected in +/+ mice ( Figure 13A , 13C , 13E and 13G ) . In the basal and middle regions of the cochlea , the tapering was no longer detected at P10 ( Figure 13F ) , but it was still partially evident in the apical region ( Figure 13H ) . By P20 , the tapering of the tallest stereocilia was only observed in the extreme apical region of the cochlea ( >95% from cochlear base ) . In this region , a partial tapering was still observed at P20 ( Figure 13J and 13K ) and even at 8 months of age ( Figure 13L ) , but not in +/+ mice ( Figure 13I ) .
The most consistent morphogenesis defect we observed in je/je mice was the failure of stereocilia to increase in mean diameter beyond ∼0 . 10–0 . 14 µm . Because stereocilia grow to different diameters in +/+ mice , this made the stereocilia of je/je mice ∼50–60% thinner than wild type for inner hair cells and utricular hair cells and ∼30–40% thinner than wild type for outer hair cells . The increase in stereocilium diameter during morphogenesis is presumed to reflect the appositional growth of the parallel actin bundle scaffold at the core , in which additional actin filaments are added at the periphery of the existing parallel actin bundle [4]–[6] . A 50–60% decrease in stereocilium diameter translates into a 75–84% decrease in cross-sectional area . Assuming that the actin filaments in the abnormally thin projections maintain the standard ∼12–13 nm parallel actin bundle interfilament spacing [30] , [31] , the parallel actin bundle scaffold of je/je mouse hair cell stereocilia could contain as little as 16–25% of the number of actin filaments found in the stereocilia of +/+ mice . The abnormally thin utricular stereocilia in P20 je/je mice labeled with fluorescent phalloidin ( Figure 1G ) , suggesting that they contain actin filaments , a conclusion that has been confirmed by transmission electron microscopy ( GS and JRB , unpublished results ) . Definitive assessments of the numbers , continuity and packing of these actin filaments will come from systematic serial-section analyses . The present examination of stereociliary dimensions suggests that espin proteins are required to increase the diameter of the stereociliary parallel actin bundle beyond a limiting value . A role in the appositional growth of the stereociliary parallel actin bundle would be entirely consistent with the espins' activity as actin-bundling proteins that can efficiently cross-link actin filaments into parallel actin bundles in vitro [28]–[30] and with the immunocytochemical localization of espins to hair cell stereocilia throughout the process of stereociliary morphogenesis [35] , [37] . Defects in the appositional growth of the parallel actin bundle could also explain the transient stereociliary tapering we discovered in young +/je mice . This novel phenotype is most likely a sign of haploinsufficiency , in which espin protein levels are limiting . The transient tapering is consistent with a slowing in the appositional growth of the stereociliary parallel actin bundle , and we propose that this slowing revealed some exclusive views of assembly intermediates and steps that are difficult to resolve in +/+ mice . For example , the direction of the taper and its subsequent filling suggest that the appositional growth of the parallel actin bundle proceeds in a proximal-to-distal direction and , thus , likely involves the barbed-end elongation of shorter actin filaments positioned in the peripheral layers of a wider , more proximal segment of the core bundle . How could reducing espin levels by approximately one-half slow the appositional growth of the parallel actin bundle ? Beyond cross-linking actin filaments into parallel actin bundles [28]–[30] , espins cause a concentration-dependent , barbed-end elongation of microvillus-type parallel actin bundles in transfected epithelial cells [32] , [34] . Like espin-mediated actin filament bundling [28] , the 116-amino acid espin carboxy-terminal actin-bundling module is necessary and sufficient for this barbed-end elongation activity , and putative F-actin-binding sites located at either end of the module are required [32] . Thus , one possibility is that wild-type levels of espin cross-links are needed both to cause the barbed-end elongation of shorter actin filaments situated at the periphery of the parallel actin bundle and to attach the newly elongated filament segments to the bundle . In addition to these activities that emphasize the role of the espin actin-bundling module , espins can bind monomeric actin via their WH2 domain [27] , [32] , [57] and can elicit WH2 domain-dependent parallel actin bundle formation when targeted to specific locations – centrosomes [57] , nucleoli [57] or filopodial tips [23] – in transfected cells . Thus , it is possible that wild-type espin levels are also needed to deliver polymerizable actin monomer into the stereocilium and to sustain the actin polymerization reactions needed to increase the number and length of actin filaments at the periphery of the parallel actin bundle . A slowing of stereociliary morphogenesis in +/je mice might also account for the eccentric protruding distal tips of stereocilia we observed in the central zone of cristae ampullares at P5 . These structures , which may represent pioneering elongation intermediates of reduced diameter , are a potential source of the abruptly tapered distal segments of stereocilia ( “wicks” ) we observed as assembly intermediates in the central zone of early postnatal +/je mice . Thus , with the slowing brought about by reduced espin levels , the morphogenesis of these long vestibular stereocilia beyond the immature stage ( Figure 5A ) appears resolvable into two additional phases of elongation , each requiring an increase in stereociliary diameter: a phase A that produces stereocilia of relatively similar intermediate length and a phase B involving differential elongation to final length . In the vestibular hair cells of je/je mice , stereociliary morphogenesis appears to stall when the increase in diameter associated with phase A does not proceed to completion . Unlike the situation in je/je mice , stereocilia in +/je mice can largely recover from having reduced espin levels . In fact , we found aged +/je mice to be remarkably similar to +/+ mice in stereociliary morphology , hair cell abundance and auditory brainstem response thresholds . Thus , we conclude that the jerker mutation is indeed recessive and that the stereociliary degeneration , extensive hair cell loss and deafness observed previously by Sjöström and Anniko [45]–[48] in aged jerker heterozygotes of an uncharacterized genetic background are attributable to another influence , e . g . , a genetic modifier , age-related hearing loss or disease . Although multiple studies have suggested a connection between espins and the elongation of stereocilia [23] , [32]–[35] , our results indicate that the relationship between espins and stereociliary length is complicated . It is true that , in general , we found the stereocilia of je/je mice to be significantly shorter than their counterparts in +/+ mice . However , we determined that this seemingly generic response to a lack of espins actually reflects a complicated mixture of defects in stereociliary morphogenesis , affecting width and length , together with defects in stereociliary stability , which vary according to inner ear region . For example , the early-stage graded elongation of stereociliary precursors appears remarkably similar in the presence and absence of espins . Cochlear stereocilia are shorter in je/je mice primarily because they subsequently shorten and disappear . Our examination of extrastriolar hair cells in the utricular macula suggests that , for long vestibular stereocilia , it is elongation phase B , involving the final differential elongation , that is markedly attenuated in je/je mice . Although espins could contribute directly to this differential elongation through the parallel actin bundle elongation activity mentioned above [32] , it is also possible that they contribute in a more indirect manner . For example , a certain threshold in the number of espin cross-links , in actin filament twist or in parallel actin bundle diameter might need to be attained before additional stereocilium elongation can proceed via mechanisms involving other proteins . Construction of a taller stereocilium might simply require a broader base with suitable cross-links . Importantly , stereocilia in the vestibular system of je/je mice showed pronounced elongation beyond the precursor stage , e . g . , up to lengths of ∼6 µm in utricular maculae and ∼8 µm in cristae . Thus , clearly substantial stereociliary elongation can take place in the absence of espins . Superimposed on an inability of stereocilia to widen and elongate fully , we observed major defects in stereocilium stability in je/je mice . This was especially noticeable in the cochlea , where stereocilia rapidly shortened and disappeared , often so fast as to obscure defects in morphogenesis . A qualitatively different and slower form of stereociliary collapse and resorption was evident in the striolar/central regions of the vestibular system in je/je mice . Thus , espins are required to avoid these types of degenerative change , which are suggestive of mechanical weakness , fragmentation and/or depolymerization of the stereocilium's parallel actin bundle scaffold . A likely possibility is that the espins' high-affinity , Ca2+-resistant cross-links are needed to stabilize the parallel actin bundle against depolymerization , fragmentation and mechanical insult . Actin-bundling proteins are known to retard actin depolymerization in vitro [58] , [59] , and an espin-mediated increase in the number of actin filaments in the parallel actin bundle would be expected to make the bundle more sturdy [60] . In a related way , the presence or absence of espin cross-links could determine why some parallel actin bundle-containing projections ( stereocilia ) are spared while others ( microvilli ) are cleared from the apical surface of the hair cell during stereociliary morphogenesis in +/+ mice . The occasional loss of a short-row stereocilium from outer hair cells that we observed in +/je mice could reflect a similar , yet localized parallel actin bundle disassembly process initiated when espin levels drop below a critical threshold . Higher levels of immunolabeling for espins have been detected at suspected sites of stereociliary damage [61] , raising the intriguing possibility that espins may also play important roles in parallel actin bundle repair . Thus , it is conceivable that the various types of stereociliary degeneration we observed in je/je mice are extreme manifestations of faulty parallel actin bundle repair . Remarkably , stereocilia in the extrastriolar/peripheral regions of the je/je mouse vestibular system , although abnormally thin and short , resisted shortening and collapse for much longer periods , even though we showed that these stereocilia normally contain espins . Since the striolar/central and extrastriolar/peripheral regions contain similar numbers of type I and type II hair cells [55] , [62] , we conclude that the pattern of stereociliary degeneration in the vestibular system of je/je mice varies primarily according to region instead of vestibular hair cell type . These differences could be tied to known regional differences in parameters such as hair cell birth date [63] , afferent response characteristics [64] , hair cell physiology [65] or interstereociliary links [66] , but could also reflect regional differences in stereociliary actin dynamics or actin-cytoskeletal proteins . Parallel actin bundles in cells typically contain multiple classes of actin-bundling protein [3] . Accordingly , espins are not the only actin-bundling proteins in hair cell stereocilia . The other actin-bundling proteins believed to be present include plastin 1 ( I-fimbrin ) , plastin 3 ( T-fimbrin ) , fascin-2 and TRIOBP , and all except fascin-2 are believed to be present in hair cells in early postnatal development [15]–[17] , [35] , [67] . Given the multiplicity of actin-bundling proteins , it is truly remarkable that these other actin-bundling proteins and the espin-like protein , which has also been detected in stereocilia [17] , are insufficient to compensate for the lack of espins in je/je mice . Like fimbrins/plastins and fascin , espins are relatively small monomeric globular proteins that preferentially cross-link actin filaments in parallel fashion [30] , [31] , [68] , [69] . The fact that espins show no obvious sequence homology with fimbrins/plastins and fascins raises the possibility that they supply cross-links of a qualitatively different nature . Accordingly , espin cross-links are much more potent than those of fascin at over-twisting the actin filaments in parallel actin bundles [31] . This over-twisting , which likely reflects a high degree of conformational rigidity in the espins , is predicted to allow for an optimum number of interfilament cross-links to form and could lead to enhanced stability for the parallel actin bundle [31] . Actin filament bundling and the cooperative effect on actin filament twist are both realized even at relatively low espin stoichiometry ( espin-actin ratio , ∼1/50 ) [29] , [31] . This may be important because , despite the intense espin antibody labeling we observed along vestibular stereocilia ( Figure 1C and 1E ) , one group recently reported that they recovered espin tryptic peptides at lower yield than those from other actin-bundling proteins in ripped-off preparations of chicken and rat vestibular hair cell stereocilia [17] . Irrespective of their actual stoichiometry , however , we conclude that espins fulfill indispensable and relatively early roles in the multistep assembly of the stereociliary parallel actin bundle . A precedent for multistep parallel actin bundle assembly can be seen in the developing neurosensory bristles of Drosophila , which form through the sequential actions of a putative espin ortholog , forked , followed by a fascin ortholog , singed [70] . Shin et al . [17] recently showed that fascin-2 appeared in stereociliary parallel actin bundles relatively late during the differential elongation of stereocilia and tended to concentrate near the distal end of the longest stereocilia . Thus , the sequential actions of espin and fascin-2 in hair cell stereocilia may be orthologous to the sequential actions of forked and singed in Drosophila bristle parallel actin bundles . A lack of espins may not only impede later-acting actin-bundling proteins , but might also irretrievably impair the parallel actin bundle substrate on which the other proteins involved in length regulation and mechanoelectrical signal transduction depend .
This study was carried out on 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 protocol was approved by the Northwestern University Animal Care and Use Committee ( Protocols 2004-0427 , 2007-0427 and 2008-1321 ) . Perfusion fixation and measurements of auditory brainstem response were performed under sodium pentobarbital anesthesia . Organ removal for western blotting was performed following euthanasia under CO2 gas-induced narcosis by decapitation with a rodent guillotine . All efforts were made to minimize animal suffering . Inbred CBA/CaJ mice ( stock number 000654 ) and jerker mice of the standard commercially available strain ( JE/LeJ; stock number 000259 ) were purchased from the Jackson Laboratory and bred and housed in the barrier-level mouse vivarium in the Center for Comparative Medicine at Northwestern University Feinberg School of Medicine . Homozygous jerker males were bred with heterozygous jerker or wild-type females . Homozygous animals older than P10 could be identified by their distinctive shaker-waltzer behavior . Genotypes were confirmed by DNA sequence analysis of PCR products obtained from tail genomic DNA [14] . We produced the congenic jerker mouse line used in this study ( CBA/CaJ . JE/LeJ-Espnje ) by repeated backcrossing into the CBA/CaJ inbred strain for 13–15 generations , according to the following scheme: male jerker homozygotes of generation n were mated with wild-type CBA/CaJ females , and the resulting heterozygous progeny from two different breeder pairs were mated to produce male jerker homozygotes of generation n+1 . Approximately every 3 generations , the wild-type CBA/CaJ female breeder stock was refreshed with mice newly purchased from the Jackson Laboratory . Wild-type mice of the JE/LeJ strain were produced by mating heterozygotes and identified by genotyping . Mice were anesthetized by intraperitoneal injection with sodium pentobarbital ( 60 mg/kg ) and briefly perfused through the ascending aorta with 0 . 9% ( w/v ) NaCl followed by fixative solution: 4% ( w/v ) formaldehyde ( freshly prepared from paraformaldehyde ) in 0 . 1 M sodium phosphate buffer , pH 7 . 4 . The inner ear was removed by dissection . A small hole was made at the top of the cochlea with the tip of a fine forceps , and the semicircular canals were broken open . Through these openings the inner ear was gently flushed with ∼0 . 3 ml of 2% ( w/v ) paraformaldehyde in 0 . 1 M sodium phosphate buffer , pH 7 . 4 , and then postfixed for an additional 1 h . Utricular maculae were dissected away from bony labyrinths . To carefully expose the epithelium , the overlying membranous labyrinth was removed , and the otoconial membrane was gently removed using a single strand from a brush . Specimens were treated with 3% ( v/v ) normal goat serum , 1% ( w/v ) bovine serum albumin , 0 . 2% ( v/v ) Triton X-100 in TBS ( 100 mM Tris , 150 mM NaCl , pH 7 . 4 ) , and incubated overnight with affinity purified rabbit polyclonal espin antibody at a concentration of 1 µg/ml . The espin antibody , which we raised against purified recombinant rat espin 2B and affinity purified on columns of rat espin 2B-Sepharose 4B , is known to react with all espin isoforms , including epitopes that are amino-terminal to site of the frameshift mutation in jerker espins [27] , [51] . The bound antibody was detected by Alexa594-goat anti-rabbit IgG ( Invitrogen ) . F-actin was visualized using Alexa488-phalloidin ( Invitrogen ) . Specimens were mounted with Vectashield ( Vector Laboratories ) and examined using the Nikon PCM2000 system confocal microscope and Simple PCI Program . Dissected cerebella were cryoprotected in 30% ( w/v ) sucrose dissolved in phosphate-buffered saline . Frozen sections , 30 µm thick , were cut in the sagittal plane on a freezing-stage sliding microtome . For bright-field microscopy , sections were processed for immunohistochemistry according to an avidin-biotin amplification protocol . Briefly , the endogenous peroxidase activity was blocked with 0 . 3% ( v/v ) H2O2 and 10% ( v/v ) methanol in TBS . Sections were treated with 3% ( v/v ) normal goat serum , 1% ( w/v ) bovine serum albumin , 0 . 2% ( v/v ) Triton X-100 in TBS and then incubated with either mouse anti-calbindin monoclonal antibody ( 1:5 , 000; Sigma ) or affinity purified rabbit polyclonal espin antibody ( see above ) . Bound antibody was detected using biotinylated donkey anti-mouse or anti-rabbit IgG ( GE Healthcare ) , the ABC Elite kit ( Vector Laboratories ) and diaminobenzidine ( Sigma ) . Images were captured with the Spot RT CCD video camera ( Diagnostics Instruments ) mounted on the Nikon Eclipse 800 microscope using the Spot RT Software 3 . 5 . 8 . Whole-mount images of eyes were captured with the Nikon digital DN100 camera mounted on the Olympus SZH10 stereomicroscope . All images were stored and processed in Adobe Photoshop CS2 . Brightness and contrast were adjusted . Mice ( ∼6 months old ) were euthanized by decapitation while under CO2 gas-induced narcosis . Testes and kidneys were removed by dissection , weighed and homogenized in 9 ( kidney ) or 18 ( testis ) volumes ( ml/g ) of ice-cold 0 . 25 M sucrose , 3 mM imidazole-HCl , pH 7 . 4 , containing 2 mM phenylmethylsulfonyl fluoride and 1% ( v/v ) Protease Inhibitor Cocktail ( Sigma P 8849 ) using 8 up-and-down strokes of a motor-driven 10-ml Teflon-glass Potter-Elvehjem homogenizer spinning at 3000 rpm . SDS gel buffer concentrate containing dithiothreitol was added , and the samples were heated at 95-100°C for 3 min with intermittent agitation on a vortex mixer . Gel samples derived from 1 . 8 mg ( testis ) or 3 . 6 mg ( kidney ) of wet tissue mass were resolved in SDS gels and transferred to nitrocellulose membrane . The blots were labeled with affinity purified rabbit polyclonal espin antibody ( see above ) at a concentration of 0 . 1 µg/ml using the ECL system ( GE Healthcare ) . Apparent molecular mass was estimated using the BenchMark Prestained Protein Ladder ( Invitrogen ) . Mice of the designated age and of either sex were anesthetized by intraperitoneal injection with sodium pentobarbital ( 60 mg/kg ) and briefly perfused through the ascending aorta with 0 . 9% ( w/v ) NaCl followed by 5-20 ml of fixative solution: 2 . 5% glutaraldehyde and 2 mM CaCl2 in 0 . 1 M sodium cacodylate buffer , pH 7 . 4 . The inner ear was removed by dissection . A small hole was made at the top of the cochlea with the tip of a fine forceps , and the semicircular canals were broken open . Through these openings the inner ear was gently flushed with ∼0 . 3 ml of 2 . 5% glutaraldehyde fixative solution and then postfixed overnight at 4°C . The membranous labyrinth , containing the cochlea and vestibular end organs , was removed by dissection . Cochlear specimens were prepared by removing the stria vascularis , Reissner's membrane and tectorial membrane . The cochlear spiral was cut into basal , middle and apical segments for processing . The utricular maculae and two adjacent , horizontal and anterior , cristae ampullares were also removed by dissection . The otolithic membrane was gently removed from the macular surface using a single strand from a brush . Specimens were processed using an osmium-thiocarbohydrazide method adapted from Hunter-Duvar [71] , which included three 1-h incubations with 1% ( w/v ) OsO4 , with 20-min incubations with saturated thiocarbohydrazide inserted after the first and second OsO4 treatments . The specimens were then dehydrated using a graded series of ethanol solutions and critical-point dried using liquid CO2 as the transitional fluid . Uncoated specimens were mounted on a Hitachi specimen stub using silver electroconductive paint and viewed using a Hitachi S-4800 field emission scanning electron microscope operated at 5 kV . Stereociliary dimensions were measured using NIH ImageJ ( Wayne Rasband , National Institutes of Health , Bethesda , MD; http://rsb . info . nih . gov/ij/ ) , and analyzed using Instat 3 . 0 ( GraphPad Software ) . To minimize foreshortening , specimens were rotated and tilted in the scanning electron microscope so that , for the purposes of measurement , the stereocilia under examination were approximately perpendicular to the direction of view . The length and width of cochlear hair cell stereocilia were measured at P0 ( n = 3 mice of each genotype ) and P5 ( n = 2 mice of each genotype ) . For each mouse , images of 10 outer and 10 inner hair cells were collected at 20 , 000X magnification from each of three different cochlear locations: base ( ∼20% from base ) , mid ( ∼50% from base ) and apex ( ∼80% from base ) . We measured the lengths of the 4 tallest stereocilia and the widths of 5 randomly selected stereocilia on each cochlear hair cell . The perpendicular alignment of stereocilia was difficult to accomplish for cochlear specimens from je/je mice , because these stereocilia were often bent in different directions ( e . g . , Figure 8G ) . We measured the lengths of the 4 tallest stereocilia on each hair cell , regardless of their location in the collection . In +/+ mice , the tallest stereocilia were located in the tallest row , near the kinocilium , whereas in je/je mice the tallest stereocilia could be found at other locations in the collection ( e . g . , Figure 8H ) . Widths were measured at 2 different positions , at the midpoint and near the top , and then averaged . The number of total surface projections ( stereocilia and microvillus-like structures ) on 10 outer hair cells from each row ( 30 outer hair cells total ) at each of the three cochlear locations were counted at P5 and P10 for mice of all three genotypes . Likewise , the numbers of missing stereocilia on outer hair cells , as evidenced by a characteristic gap in the shortest row of stereocilia ( Figure 4D and 4E ) , were counted for +/+ and +/je mice at P10 . The width of stereocilia on extrastriolar hair cells was measured at 1-µm intervals along the length . The analysis included 3 or 4 mice of each of the three genotypes at P0 , P5 , P10 and P20 , 2 je/je mice at P40 and 1 je/je mouse each at P60 and P90 . To avoid immature hair cells ( e . g . , Figure 5A and 5B ) , only hair cells with stereocilia of lengths >7 µm ( +/+ or +/je ) or >3 µm ( je/je ) were included . Images of the stereociliary collection ( n = 20–28 hair cells for each genotype and each time point ) were transferred to Adobe Photoshop , where a grid of lines ( spaced at 1-µm intervals ) was placed over the image ( Figure 11A ) . Widths were measured for clearly resolved stereocilia that crossed a line or , in the case of the distal tip , ended close to a line . In addition , we measured widths at the extreme base , where the stereocilia emerged from the apical surface of the hair cell . The width measurements at each distance were averaged for individual hair cells ( see graph in Figure 11A ) . These data were combined ( Figure 11B ) to calculate mean ± SD for each genotype and postnatal age ( Figure 11C ) . In most instances , we obtained large numbers of measurements ( >100 ) , which did not show a Gaussian distribution . Therefore , such measurements were analyzed using the nonparametric Kruskal-Wallis test followed by Dunn's multiple comparisons test . In two instances , sample sizes were , by necessity , considerably smaller and showed a Gaussian distribution: width measurements at 11 and 12 µm from the base of extrastriolar stereocilia ( Figure 11C ) . These measurements were analyzed by one-way analysis of variance followed by the Bonferroni multiple comparisons test . The average height of the collection of stereocilia on the extrastriolar hair cells of je/je mice ( Figure 5E ) was estimated from measurements of 100 hair cells at each postnatal age . Mice were anesthetized by intraperitoneal injection with sodium pentobarbital ( initial , 80 mg/kg; maintenance , 17 mg/kg ) . Auditory brainstem responses were obtained by subtracting ipsilateral mastoid potentials from vertex potentials measured relative to a ground electrode placed in the neck . The electrodes were connected to a differential amplifier ( ISO-80 , World Precision Instruments ) with a high input-impedance ( >1012 Ω ) set to 80 dB . Further filtering of the signal ( 300–3000 Hz ) was obtained through an IP90 filter ( Frequency Devices ) . The sampling rate was 200 kHz , and responses to 100 stimulus presentations were averaged . Auditory brainstem response thresholds were defined as sound levels required for a visible response to acoustic stimuli . The noise floor in an average recording was typically 1 µV . In particular , the appearance of wave II was monitored . Voltage commands for acoustical stimuli were generated using a computer KPCI 3110 I/O board ( Keithley Instruments , Inc . ) inserted into a personal computer and were used to drive a DT 770Pro headphone ( Beyerdynamic ) . For acoustically evoked auditory brainstem responses , tone bursts ( 12 ms duration , 1 ms rise/fall ) with different carrier frequencies were presented at a rate of 4 Hz . The sound pressure was measured with a real head coupler [72] . Means and standard errors were calculated for the electrophysiological thresholds . Measurements were analyzed by one-way analysis of variance followed by the Tukey-Kramer multiple comparisons test . | Stereocilia are the fingerlike projections of inner ear hair cells that detect sound and motion . Stereocilia grow to specific lengths and diameters and form staircase-like arrays . The changes in size appear to be driven by matching alterations in the dimensions of an underlying molecular scaffold consisting of a bundle of actin filaments cross-linked by actin-bundling proteins . To elucidate the roles of the espin actin-bundling proteins in hair cell stereocilia , we carry out an in-depth accounting of stereociliary size and shape in the jerker mutant mouse , which lacks the espin proteins because of a mutation in the espin gene . We examine a new and improved jerker mouse with a genetic background known for high-quality lifelong hearing . We find that , in the absence of espins , stereocilia do not increase in diameter or complete their elongation , but instead bend , shorten , and disappear . Although the specifics vary according to inner ear region , the stereociliary defects are profound and can readily account for the deafness and balance problems of jerker mice and humans with certain espin gene mutations . Even reducing espin levels by one-half leads to temporary defects in stereociliary diameter . Thus , espins play crucial roles in the formation and maintenance of hair cell stereocilia . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"cellular",
"structures",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"animal",
"genetics",
"neuroscience",
"cell",
"differentiation",
"animal",
"models",
"developmental",
"biology",
"otology",
"model",
"organisms",
"hearing",
"disorders",
"cytoskeleton",
"proteins",
"biology",
"otorhinolaryngology",
"otolaryngology",
"mouse",
"biochemistry",
"cytoskeletal",
"proteins",
"auditory",
"system",
"cell",
"biology",
"genetics",
"sensory",
"systems",
"molecular",
"cell",
"biology",
"genetics",
"of",
"disease",
"genetics",
"and",
"genomics"
]
| 2011 | Roles of the Espin Actin-Bundling Proteins in the Morphogenesis and Stabilization of Hair Cell Stereocilia Revealed in CBA/CaJ Congenic Jerker Mice |
Vertebrate dentitions originated in the posterior pharynx of jawless fishes more than half a billion years ago . As gnathostomes ( jawed vertebrates ) evolved , teeth developed on oral jaws and helped to establish the dominance of this lineage on land and in the sea . The advent of oral jaws was facilitated , in part , by absence of hox gene expression in the first , most anterior , pharyngeal arch . Much later in evolutionary time , teleost fishes evolved a novel toothed jaw in the pharynx , the location of the first vertebrate teeth . To examine the evolutionary modularity of dentitions , we asked whether oral and pharyngeal teeth develop using common or independent gene regulatory pathways . First , we showed that tooth number is correlated on oral and pharyngeal jaws across species of cichlid fishes from Lake Malawi ( East Africa ) , suggestive of common regulatory mechanisms for tooth initiation . Surprisingly , we found that cichlid pharyngeal dentitions develop in a region of dense hox gene expression . Thus , regulation of tooth number is conserved , despite distinct developmental environments of oral and pharyngeal jaws; pharyngeal jaws occupy hox-positive , endodermal sites , and oral jaws develop in hox-negative regions with ectodermal cell contributions . Next , we studied the expression of a dental gene network for tooth initiation , most genes of which are similarly deployed across the two disparate jaw sites . This collection of genes includes members of the ectodysplasin pathway , eda and edar , expressed identically during the patterning of oral and pharyngeal teeth . Taken together , these data suggest that pharyngeal teeth of jawless vertebrates utilized an ancient gene network before the origin of oral jaws , oral teeth , and ectodermal appendages . The first vertebrate dentition likely appeared in a hox-positive , endodermal environment and expressed a genetic program including ectodysplasin pathway genes . This ancient regulatory circuit was co-opted and modified for teeth in oral jaws of the first jawed vertebrate , and subsequently deployed as jaws enveloped teeth on novel pharyngeal jaws . Our data highlight an amazing modularity of jaws and teeth as they coevolved during the history of vertebrates . We exploit this diversity to infer a core dental gene network , common to the first tooth and all of its descendants .
Teeth are ancient vertebrate structures . During the early evolution of vertebrates , the appearance of a pharyngeal dentition greatly enhanced the capacity for processing food . Tooth-like structures located on elements of the pharyngeal series or skeleton were present in extinct jawless fishes ( agnathans ) , for example members of the conodonts and later the thelodonts , which both possessed intricate , well-organized replacing dental systems [1–4] . Although tooth-like elements ( denticles ) were also present on the dermal surface of some agnathans ( including thelodonts ) and chondrichthyans , it was the occurrence of uniquely patterned pharyngeal teeth in agnathans that likely foreshadowed all other vertebrate oropharyngeal teeth [1 , 3–5] . Intriguingly , some extant fish still retain this ancient population of teeth in the posterior pharyngeal skeleton . More advanced groups of teleosts have adapted their posterior pharyngeal skeleton with teeth housed in discrete functional jaws , as in the cichlids and other groups [6–14] ( Figure 1 ) . Teeth arise from a collaboration of different cell types that coalesce during the formation of the pharyngeal arches . Pharyngeal arches develop as a set of bulges on the ventrolateral side of the embryonic vertebrate head [15–17] ( Figure 1 ) . The formation of the pharyngeal arches involves the combination of all “germ” tissue layers: ectoderm covering each arch externally , endoderm lining the arches , and between these layers , the neural crest–derived mesenchyme surround a core of mesoderm [16] . Numerous common key developmental genes are required to regulate both arch patterning and development of the dentition ( e . g . , bmp4 and dlx2; Figure 1C–1F ) . During the evolution of vertebrates , a general reduction in the number of pharyngeal arches is observed , from fossil agnathans ( e . g . , “ostracoderms” [18] ) that possessed tens of arches and multiple ( up to 45 ) gill openings [15 , 18] , to amniotes that have five arches [15] . Teleost fish have seven pharyngeal arches [19] . The first pharyngeal arch ( PA1 ) in the series forms the oral jaws ( Figure 1 ) . The second arch ( PA2 ) forms the hyoid and the jaw support; the remaining posterior arches either contribute to the formation of the gills and gill-related skeletal structures ( branchial ) in fish or become incorporated into the throat of tetrapods [15] . The most posterior arch ( PA7 ) in teleost fish houses a pharyngeal dentition , and in some groups , PA7 forms a second set of jaws , the pharyngeal jaws ( Figure 1 ) . The numbering of the arches PA1–7 in teleosts generally reflects the order of metameric development from anterior to posterior . However , the most terminal posterior arch ( PA7 in teleost fish ) develops out of series , ahead of most of the anterior arches [20] . The evolutionary origin of toothed oral jaws galvanized the dominance of gnathostomes and may have been prompted by the loss of Hox gene regulation in PA1 [21 , 22] . This notion has been supported by a report [21] of Hox gene expression during first arch formation in the lamprey ( Lampetra fluviatilis ) , a jawless fish , although this observation is controversial ( see Takio et al . [23] ) . All extant , jawed vertebrates do not express Hox genes in developing PA1 . Numerous studies conclude that for correct first arch ( PA1 ) fate , Hox genes must be absent , and consequently , for posterior arch fate , Hox genes must be present [24–28] . A branchial Hox code maintains the identity of more posterior pharyngeal arches , including the seventh pharyngeal arch ( PA7 ) in teleosts [29] that house the terminal pharyngeal jaws . Osteichthyan fish have retained the potential to form teeth throughout the oropharyngeal cavity , which includes the most posterior arch , PA7 ( Figure 1 ) . The ancestral condition for osteichthyans is teeth located throughout the oropharynx ( e . g . , Amia calva , the bowfin ) [1 , 5] . However , an evolutionary trend toward a reduction in the sites that house teeth throughout the oropharynx is observed , similar to that of the arches themselves . For example , tetrapods have a dentition reduced to the oral jaws; further reductions are observed in tetrapods with complete ( birds and turtles ) or partial ( mammals ) loss of teeth within the single oral row . Even the developmental teleost model , the zebrafish ( Danio rerio ) , has a reduced dentition with few teeth present on the lower PA7 and a complete loss of oral teeth [30 , 31] . Oral and pharyngeal teeth are assumed to be serially homologous [5 , 32 , 33] . This is thought to be the case despite the likelihood that tissue origins are not equivalent , with teleost oral teeth having the potential for ectodermal cell participation and pharyngeal teeth born out of endodermal epithelial tissue [1 , 5 , 34] . Tissue origin identification of the oral epithelium that contributes to tooth development has been consistently elusive . The break down of the stomodeum or the oropharyngeal/buccopharyngeal membrane leads to mixing of both anterior ectodermal and posterior endodermal cells within the oropharyngeal cavity , therefore a definite ectoderm/endoderm boundary may be unidentifiable . The mixed interface between the endoderm and ectoderm within the oropharyngeal cavity may be variable among vertebrate groups [35 , 36] . Reports of both histological and cell labeling evidence have suggested that some vertebrates develop oral teeth in close proximity to endodermal cells , even mammalian incisors [37] and molars ( P . Sharpe , personal correspondence ) . Recently , Soukup et al . [36] observed that oral teeth of the Mexican axolotl form from epithelium either born of ectoderm , endoderm , or a mixture of the two , and teeth that form as a result of these specific cell types or their collaboration are indistinguishable . This therefore suggests that at least in the oral region , the origin of the epithelium may vary; the important combination for odontogenesis is some source of epithelium plus the underlying neural crest–derived ectomesenchyme [36] . The data of Soukup et al . [36] lead to the interpretation that most anterior oral teeth are likely ectodermal , posterior oral teeth develop from a mixed population of ectodermal and endodermal epithelia , and the most posterior teeth , such as those on PA7 in teleost fishes , are likely formed from strictly endodermal cells [1 , 5 , 34] . Isolated reports have concluded that the teeth on the oral and pharyngeal elements of teleost fish share expression of a small set of genes , with notable differences [31 , 38–41] . In addition , certain genetic factors , key to the developmental programming of the mammalian oral dentition , are similarly expressed in equivalent regions of the developing teleost pharyngeal dentition [31 , 38–44] . Despite the coordination of tooth and arch development ( above ) , oral and pharyngeal odontogenesis is partly decoupled from associated bones and/or cartilage [20 , 22 , 30] . Mutations affecting the pharyngeal cartilages , including PA7 , do not necessarily disrupt the development of pharyngeal teeth [20 , 30 , 45] , and mutations affecting pharyngeal teeth do not necessarily disrupt cartilage development [46] . Interestingly , other zebrafish mutations that affect pharyngeal/branchial cartilage formation in most arches do not always affect the most posterior tooth bearing PA7 [20] . This suggests that PA7 has unique properties separating it from more anterior arches . The involvement of Hox genes during the development and organization of the pharyngeal skeleton [29] implies that pharyngeal teeth develop and fuse to skeletal elements in a Hox-positive environment , unlike those of the oral jaws that develop consistently in a Hox-negative region , unless the appropriate conditions for jaw formation regardless of location require a loss ( albeit temporary , in the case of PA7 ) of Hox regulation . The available data are thus equivocal on the molecular regulation of oral versus pharyngeal dentitions . These dentitions are evolutionarily decoupled; teeth arose first in the pharynx prior to the origin of jaws . They are functionally decoupled; many vertebrates possess pharyngeal teeth and not oral teeth ( e . g . , zebrafish ) , and many others possess oral teeth and not pharyngeal teeth ( e . g . , mammals ) . They are developmentally decoupled in space ( PA1 vs . PA7 ) , tissue distribution ( contribution of ectoderm in PA1 vs . endoderm in PA7 ) , and possibly by the influence of the Hox code . One of the major difficulties in interpreting available data is that they are drawn from organisms , often with only a single dentition ( zebrafish or mouse ) , separated by vast evolutionary distances , or sampled species are taken from lineages exhibiting reduced dental diversity ( e . g . , medaka , trout ) among close relatives . Our aim is to understand the relationships and constraints between evolutionarily , developmentally , and functionally decoupled oral and pharyngeal dentitions . Our models for this project are cichlid fishes from Lake Malawi , for three primary reasons . First , Malawi cichlids exhibit a tremendous diversity in oral and pharyngeal jaw dentitions , and this variation has evolved in a short evolutionary window [47] . Second , all cichlids possess modified posterior pharyngeal arches , which act as a functional jaw ( Figure 1 and 2; [6 , 7] ) . Cichlids , and a few other teleost lineages [6–9 , 12–14 , 48] , feature fusion of the bilateral units of the lower pharyngeal jaw cartilages ( LPJ ) and a novel muscular sling that pulls the LPJ upward to contact the hinged upper pharyngeal jaw units . This formidable pharyngeal machinery for food processing can produce enough force in some species to crush hard prey such as shelled molluscs [6 , 7] . Cichlid oral and pharyngeal jaws are evolutionarily and functionally decoupled [7 , 8] . Third , we have recently characterized a gene network ( including bmp2 , bmp4 , dlx2 , eda , edar , pax9 , pitx2 , runx2 , shh , and wnt7b ) associated with variation in oral jaw tooth row number , tooth number within rows , and the spacing of teeth [49]; we can therefore ask how this network of genes is expressed in dentitions on oral and pharyngeal jaws of the same organism . Here we use “gene network” in the sense of coordinated expression . Exact similarities between network topologies ( e . g . , interactions between nodes ) , while implicated , remain to be determined in each evolutionary lineage . We integrate these molecular data with comparative morphology and paleontology to ( 1 ) infer the ancient dental network used to pattern the first teeth and ( 2 ) suggest a core regulatory circuit common to all dentitions .
We took advantage of oral and pharyngeal dental diversity among Lake Malawi cichlids to ask whether tooth number was controlled similarly on each jaw . We therefore estimated the number of teeth on both oral and pharyngeal jaws of adult fishes for a range of Malawi cichlid species spanning the major evolutionary lineages and the extremes of dental diversity ( Figure 2 ) . For instance , the large ( 0 . 5 m ) pelagic predator Rhamphochromis esox possesses an average of 65 oral and 110 pharyngeal teeth , whereas the rock-dwelling algal brusher Petrotilapia nigra has on average 1 , 170 oral and 722 pharyngeal teeth ( Figure 2; Table S1 ) . There was a positive and highly significant correlation between the numbers of teeth on oral versus pharyngeal jaws ( r = 0 . 53 without P . nigra and r = 0 . 66 including P . nigra; p < 0 . 00001; Figure 2J ) . This correlation is independent of evolutionary history as demonstrated by removing phylogenetic effects with independent contrasts ( r = 0 . 39; p < 0 . 019; Figure 2K ) . These data indicate that regulators of tooth initiation ( tooth number ) are similar across the two dentitions , a surprise given functional independence , developmental differences , and evolutionary separation . Following ( 1 ) the correlation described above and ( 2 ) the idea that lack of Hox expression is permissive for toothed jaw development on PA1 [21] , we predicted that Hox genes would be down-regulated during the development of the cichlid toothed pharyngeal jaw on PA7 . We therefore examined the expression of seven Hox genes ( hoxA2b , hoxA5a , hoxB2a , hoxB5b , hoxB6b , hoxC6a , and hoxD4a; Figure 3 ) in two cichlid species , Metriaclima zebra ( MZ ) and Copadichromis conophorus ( CC ) representing the two major Malawi evolutionary lineages [50] , during a critical period when pharyngeal jaws and dentitions develop . Notably , all seven Hox genes were strongly expressed within the mesenchyme enveloping the pharyngeal jaw cartilages of PA7 with six of the seven genes examined ( hoxA2b , hoxB5b , hoxB6b , and hoxD4a , Figure 3A–3C and 3G–3O; hoxB2a and hoxC6a; unpublished data ) expressed in the dental mesenchymal cells directly surrounding the tooth germs . Furthermore , hoxB5b ( Figure 3G–3I ) and hoxB6b ( Figure 3J–3L ) are expressed in the basal dental mesenchyme within individual tooth germs ( dental papilla ) at this stage . hoxA5a ( Figure 3D–3F ) is the only Hox gene we examined not expressed in close proximity to the developing teeth , but is strongly expressed around the future regions of tooth attachment and cartilage maturation of both upper and lower elements of PA7 ( Figure 3E and 3F ) . These data demonstrate , contrary to our prediction , that cichlid pharyngeal jaws and their dentitions develop in a Hox-positive environment . Tooth number is correlated on cichlid oral versus pharyngeal jaws ( Figure 2J and 2K ) , but these jaws represent distinct cellular and developmental ( Hox-negative vs . Hox-positive ) environments ( Figure 3 ) . We therefore hypothesized that conservation in adult tooth pattern was due to conservation in a genetic network establishing tooth initiation on both jaws . We have recently described a dental gene network for cichlid oral jaws , a “periodic pattern generator” for interspecific variation in tooth row number , tooth number within rows , and tooth spacing [49] . Genes involved in this dental regulatory circuit include bmp2 , bmp4 , dlx2 , eda , edar , pax9 , pitx2 , runx2 , shh , and wnt7b; specific roles in odontogenesis have also been documented in the mouse ( Mus musculus ) . A noted corollary of this hypothesis is that it might be surprising to observe the expression of ectodysplasin pathway genes eda and edar in the pharyngeal dentition ( derived from endoderm ) , because these molecules are seemingly specified to ectodermal epithelial organs ( see below ) , although expression has been observed in murine endoderm [51] . Most of the genes analyzed ( six of eight; bmp2 , bmp4 , dlx2 , pitx2 , runx2 , and shh ) have equivalent expression patterns in dental epithelium and/or mesenchyme during cichlid oral and pharyngeal tooth development ( Figure 4 ) . The two exceptions are provided by pax9 and barx1 . pax9 is expressed within the developing dentition of the oral jaws ( Figure 4A; [49] ) but not in close proximity to developing teeth in the pharynx , although expression is noted in cells of the pharyngeal mesenchyme lateral to the teeth ( similarly described in zebrafish [D . rerio] , medaka [Oryzias latipes] , and the Mexican tetra [Astyanax mexicanus] by Stock and colleagues [31] ) but not associated with cells of the dental mesenchyme ( Figure 4B ) . Conversely , barx1 is not localized to the oral dentition . It is expressed in the flanks of the oral jaw outside of the tooth-forming region ( Figure 4C ) . However , barx1 is expressed in the pharyngeal mesenchyme underlying the dental epithelial thickenings of the pharyngeal teeth on CB5 ( PA7 ) ( Figure 4D ) . We observed the expression of ectodysplasin pathway genes , eda and its receptor edar , in conserved dental cell types on both oral and pharyngeal jaws . The ectodysplasin receptor , edar , is expressed in the epithelial thickenings and within the oral epithelial odontogenic band ( OB ) , similar to shh and pitx2 [49] . During morphogenesis , expression of edar remains confined to the epithelial tooth germ ( Figure 5A–5C ) . eda is similarly expressed in both oral [49] and pharyngeal teeth , restricted to the mesenchymal cells directly surrounding the developing initial epithelial thickening ( the mesenchymal “zone of inhibition” [ZOI]; [49]; Figure 5D–5F ) and later during morphogenesis around the maturing tooth germs . This is in contrast to its expression in mammalian teeth where it is exclusively expressed in the epithelial cells of the intergerm space ( ZOI ) before its expression is recruited into the tooth during morphogenesis of the outer dental epithelium [52 , 53] . This is the first documentation of ectodysplasin pathway genes expressed in teeth likely derived from endoderm , deep within the pharyngeal/branchial arches . Our data complement a recent report that mutations in eda and edar result in loss of zebrafish pharyngeal teeth [46] . This result might have been expected due the expression of Tabby-A ( eda ) localized to the visceral and definitive endoderm in the mouse [51] , although not related to an epithelial appendage per se ( e . g . , hair , tooth , or scale ) . eda and edar are members of the tumor necrosis factor ( TNF ) superfamily and are imperative for the correct formation and patterning of ectodermal appendages in vertebrates such as feathers , hair , teeth , scales , and glands [52–56] . Human mutations in these and other members of the ectodysplasin pathway cause various forms of hypohidrotic ectodermal dysplasia ( HED ) , which manifests by specifically affecting ectodermal appendages [52] . Both eda and edar are involved in “gill raker” patterning along the mesiodistal axis of each gill bar . Gill rakers are skeletal elements of the oropharyngeal cavity that line the dorsal region of the cartilaginous gill arches from PA3 to PA6 ( Figure 5A , 5C , 5D , and 5F ) . Each gill/branchial arch ( PA3–6 ) is defined by a band of eda/edar expression , from which edar is up-regulated at the site of initiation for gill raker primordia ( Figure 5A–5C ) . eda is expressed during gill raker initiation similar to its expression in teeth ( Figure 5E and 5F; [49] ) , labeling the interraker mesenchyme region for each gill raker primordium ( Figure 5D and 5F ) . Thus , cichlid eda and edar are expressed in all seven arches , from the teeth and jaw of PA1 ( oral jaw ) throughout the series PA3–6 during gill raker placode formation , and in PA7 , where they mark the pharyngeal dentition ( Figures 5 and 6 ) . Additionally , the non-dental , non–gill-bearing arch PA2 exhibits expression of the two ectodysplasin pathway genes in both internal pharyngeal endoderm and external arch ectoderm ( Figure 5A and 5D ) , presumably recruited for the extension of PA2 to form the opercular flap gill cover in teleosts . The role of ectodysplasin pathway genes in the development of tetrapod PA2 remains unclear; in tetrapods PA2 skeletal elements support the jaw ( hyoid ) and contribute to the neck . A collection of other dental markers ( β-catenin , bmp2 , bmp4 , dlx2 , pitx2 , and shh; unpublished data ) is also expressed in a similar manner during the patterning of the gill rakers . Gill rakers are iteratively initiated from a band of competence similar to the odontogenic band on the jaws , expressing these genes in a mesiodistal pattern , from which “raker buds” show localized expression . Furthermore , later in development , these elements house an additional set of teeth/denticles ( unpublished data ) [39 , 57 , 58] . Our data suggest that a conserved dental gene network periodically patterns distinct gill arch structures on PA3–6 .
Our data demonstrate that Hox genes are expressed in cichlid pharyngeal jaws as the pharyngeal dentition initiates . Moreover , expression of a subset of these genes is observed within dental mesenchyme ( hoxA2b , hoxB5b , hoxB6b , and hoxD4a , Figure 3; and hoxB2a and hoxC6a; unpublished data ) . Hox expression in teeth has not been noted before . This is not surprising; the majority of vertebrate developmental models do not possess a pharyngeal dentition ( Hox genes are not expressed in PA1 ) , and zebrafish Hox expression [27 , 61] has not been assessed during pharyngeal odontogenesis . This observation prompts a series of related questions: what role , if any , do Hox genes play in the pharyngeal dentition ? Did ancient pharyngeal teeth express Hox genes ? Hox genes ( or their absence ) are neither necessary nor sufficient for tooth initiation . All vertebrate oral dentitions develop in a Hox-negative environment . Initiation of the mouse dentition is unaffected when Hoxa2 is overexpressed in the first arch [22] . Additionally , tooth development proceeds normally when Hox-positive mesenchyme from the second arch is recombined with first arch epithelium [22] . These latter results may have more to say about the independence of teeth and the jaws that house them ( also [24–26] ) than about the role of Hox genes in a developing tooth . Notably , in other organ ( e . g . , limb ) regulatory networks [62] , Hox genes are upstream of a number of dental markers ( barx1 , bmp2 , bmp4 , dlx2 , pitx2 , and shh ) , and as such , Hox regulation might affect later aspects of pharyngeal tooth morphogenesis , replacement , or shape . One putative Hox target , barx1 [62] , is expressed in cichlid pharyngeal , but not oral , dentitions ( Figures 4 and 6 ) . Barx1 is downstream of Fgf8 in mouse molars [63 , 64]; fgf8 is absent from the oral and pharyngeal dentitions of all fishes examined to date [43 , 49] . Therefore , it is intriguing to speculate that Hox and barx1 expression were components of an ancient dental program deep in the pharynx of jawless fishes , retained in the pharyngeal teeth of extant fishes . Hox genes may have played a patterning role for these first teeth , which lined an endoderm-rich pharyngeal cavity devoid of bony jaw elements . We speculate that as gnathostomes evolved a Hox-negative oral jaw , barx1 expression was initially absent in oral teeth , and later recaptured by Fgf8 in mammalian molars . pax9 , a paired domain ( not homeodomain ) transcription factor , may have replaced Hox expression in gnathostome oral dental mesenchyme . pax9 is not essential for vertebrate tooth development ( it is not expressed in all teeth ) , although it is necessary for mammalian odontogenesis [65 , 66] . We propose that an ancient dental gene network constructed the first tooth-like structures deep within the pharyngeal arches of jawless fishes , more than half a billion years ago [1–4 , 59 , 67 , 68] . This ancient dental regulatory circuit has been conserved in modern fishes as those markers expressed in pharyngeal dentitions . This dental network is comprised of genes present during pharyngeal tooth initiation: barx1 , bmp2 , bmp4 , dlx2 , pitx2 , runx2 , and shh , including the ectodysplasin pathway genes , eda and edar , with a contribution from Hox molecules . In addition to genes described here , a number of others expressed in the pharyngeal dentition of teleosts are part of the ancient dental network , including eve1 [41 , 69] , lhx6 , and lhx7 [43] ( Figure 7; Table 1 ) . β-catenin , fgf3 , fgf10 , and notch2 , a set of stem cell markers recruited during cichlid oral jaw tooth replacement ( G . J . Fraser and J . T . Streelman , unpublished data ) are also assigned to the ancient dental network , based on expression in pharyngeal teeth ( Figure 7; Table 1 ) . We hypothesize that this ancient dental network has patterned all pharyngeal teeth , from the first dentitions in now-extinct jawless vertebrates to modern osteichthyan and chondrichthyan fishes . Although an ambiguous relationship exists between the homology of the elements of the dermal skeleton and teeth/denticles of the oropharynx [68] , we envisage a plausible scenario that follows the general “inside-out” model of odontode evolution [1 , 3 , 4 , 59] in which pharyngeal endoderm in collaboration with neural crest–derived ectomesenchyme permitted the development of the first discrete , patterned dental units in jawless vertebrates . In contrast , the “outside-in” notion of vertebrate odontode evolution [1 , 3 , 4 , 59 , 70] , that dermal denticle units like those of modern elasmobranchs ( sharks and rays ) “migrated” into the mouth cavity coinciding with the appearance of oral jaws , is confidently contested as the earliest “toothed” vertebrates ( i . e . , conodonts ) lacked a dermal skeleton . Thus , it seems that pharyngeal teeth were the progenitor population for all vertebrate dentitions . We therefore propose that this ancient dental network was established close to the origin of vertebrates and was adopted for the formation of the first teeth . This regulatory network was later co-opted and modified ( Figure 7 ) to form teeth on the first jaws of gnathostomes ( the oral jaws , PA1 ) , providing the prerequisite for extreme predatory feeding , in the absence of Hox gene expression [21 , 71] . We infer from these data that the transition from agnathans to the first gnathostomes coincided with further modifications of the dental network governing the development of the early oral dentition . We observe a number of genes with variable dental expression patterns between vertebrates . pax9 ( Pax9 ) is imperative for mammalian tooth development , expressed during cichlid ( Figures 4 and 6 and Table 1; [49] ) and Mexican tetra [31] oral tooth initiation , but not expressed in relation to the pharyngeal teeth of cichlids ( Figure 4 ) and zebrafish [43] . This suggests that pax9 was not part of the ancient dental network but became a player in Hox-negative oral tooth evolution . Fgf8 is an important regulator of murine odontogenesis , retained during the establishment of a potential avian odontogenic cascade [72] , but is not present during the development of any teleost dentition [43 , 49] . Thus , we can assume Fgf8 was recruited in collaboration with tooth development during tetrapod dental evolution . In addition to these modifications , a member of the ancient dental network ( present in pharyngeal teeth ) , barx1 , not involved in oral tooth development in fish ( Figure 4 ) was subsequently adopted for mammalian molar formation . Some genetic participants of tooth development that made the transition from the ancient gene network to the evolution of oral teeth , e . g . , eve1 [41 , 69] , are not involved in tetrapod odontogenesis . Taken together , these reports of variable dental genes ( pax9 , fgf8 ) and members of the ancient gene network that have been lost ( and regained ) in oral dentitions ( barx1 , eve1 ) across vertebrates suggest that they are not “evolutionarily essential” for tooth development . The first oral dentitions during the advent of gnathostomes likely developed in a transitional genetic and cellular environment as the consequences of major changes in cell signaling ( e . g . , loss of Hox ) sorted . Thus it is intriguing to note lack or the simplified peripheral oral dentition in the first gnathostomes ( e . g . , derived placoderms [73–75] ) , an experimental dental transition . Subsequently in some groups of advanced teleost fishes [6–9] , including cichlids , the ancient dental network located on PA7 , in coordination with a recent adaptation of the pharyngeal skeleton , led to the evolution of a new functional toothed jaw , the pharyngeal jaw . Thus , the ancient dental gene network , once used for the first teeth in the pharynx of extinct jawless vertebrates , has been deployed on an entirely novel set of jaws ( Figures 1 , 2 , and 7 ) . Our analysis identifies a number of genes expressed commonly on cichlid oral and pharyngeal jaws ( Figures 4 and 5 ) . Many of these are similarly employed in the pharyngeal dentition of zebrafish , across oral and pharyngeal dentitions in the rainbow trout , Japanese medaka , and Mexican tetra , and in the oral jaws of tetrapods ( Table 1 and references therein ) . These common patterns define a core dental regulatory network likely expressed in the first tooth and all of its evolutionary descendants , regardless of anatomical location within the oropharynx . By definition , core genes are part of the ancient dental network , but not necessarily vice versa . shh is a core marker of dental epithelial initiation , as is pitx2 , bmp2 , edar , and to some degree , bmp4 , dlx2 , and eda . In response to initial epithelial signals [76] , molecules within the neural crest-derived ectomesenchyme activate the collaboration between these cell layers toward morphogenesis of the unit tooth; mesenchymal instigators of tooth development include bmp2 , bmp4 , dlx2 , runx2 , and eda ( with eda deployment variable between vertebrates [49] although its role is potentially equivalent; Table 1 ) . β-catenin , fgf3 , fgf10 , and notch2 are active during the initiation of dentitions and are recruited similarly in the dental stem cell niche of cichlid replacement teeth ( G . J . Fraser and J . T . Streelman . unpublished data ) and in continuously growing mouse incisors [77–79] . The core dental network represents a conserved set of molecules for tooth development that provides the molecular machinery and developmental constraints for all teeth , regardless of cellular origin ( endodermal or ectodermal ) or Hox gene contribution . We suggest that this core set is evolutionarily essential; no known examples of correctly patterned dentitions occur without the involvement of core genes . It is likely that nature has never made a tooth without this core genetic network . It is notable that members of the core dental network are coexpressed in the development of other vertebrate organs such as scales , feathers , and hairs [54 , 56 , 80–86] and that the origin of these gene families predate vertebrates altogether . Regulatory interactions among the core genes are themselves likely to be ancient , and therefore evolutionarily successful . Such ancient developmental regulatory networks may be particularly robust to failure ( for instance , null mutations in human , dog , and cow ectodysplasin pathway genes affect morphogenesis but usually do not lead to loss of all teeth [87–89] ) while retaining the capacity for evolvability [46 , 49 , 52 , 90 , 91] . It is impossible to study the developmental programs that controlled morphologies of extinct organisms . It is possible , however , to infer evolutionary transitions from modern phenotypic diversity through to origins [92–98] . Here , we have combined paleontology , molecular developmental biology , and comparative morphology to infer the developmental basis of ancient dental structures close to the origin of vertebrates and their evolutionary progression through time to recent diversity .
To generate a phylogenetic tree for the species examined , we assembled published ND2 data from a total of 37 species of Lake Malawi cichlids and several outgroup species . Modeltest 3 . 06 [99] was used to identify the best model of molecular evolution for each codon site . With the ND2 gene partitioned into its codon sites , Bayesian analyses were executed to find approximations of the maximum likelihood tree using MrBayes 3 . 0 [100] with methods similar to those described in Hulsey et al . [101] . Pharyngeal tooth counts were performed on high-magnification images of Malawi cichlid lower pharyngeal jaws , and each tooth was counted ( see Figure 2 ) . We counted all lower oral teeth for a number of specimens and devised a system to estimate oral tooth number that replicated the counts . Oral tooth number was estimated based on the exact numbers of teeth on one half of the lower jaw first row , multiplied by the exact number of rows , multiplied by 2 ( for the two halves of the oral jaw process; Figure 2 ) . To assess the putative relationship among oral jaw and pharyngeal lower jaw tooth number , we first examined the correlation between Malawi species values . Between one and four individuals ( 70 specimens total ) were documented per 37 species ( Table S1 ) ; the correlation was then based on the mean values . However , because species are not evolutionarily independent [102] , we also performed an independent contrast analysis . For the phylogenetic backbone of this analysis , we used the single best ND2 phylogeny recovered above . Many of the species relationships were recovered only as polytomies due to a lack of base pair changes among these recently diverged species [103] . Phylogenetic independent contrast analyses were then performed in order to assess the independent contrast correlations among the number of teeth on each jaw . The “crunch” algorithm was used in CAIC because it treats all variables as continuous . Analyses were run with untransformed tooth number because the correlation of species values suggested the variance was equal for species possessing both large and small numbers of teeth . Because tooth number may increase over ontogeny ( but see [104] ) , we also examined the correlation of tooth number on the two jaws with log10 specimen standard length ( SL ) used as a phylogenetic covariate . P . nigra is a species with extremely large numbers of teeth on both the oral and pharyngeal jaws; thus the inclusion of P . nigra may have disproportionate affects on the regression trend . We therefore examined , and report the species level correlation with and without P . nigra . Embryos and fry of multiple species of Lake Malawi cichlids ( Copadichromis conophorus [CC] , Dimidiochromis compressiceps [DC] , Metriaclima zebra [MZ] , and Labeotropheus fuelleborni [LF] ) were raised to the required stage in a recirculating aquarium system ( GIT ) at 28 °C . Embryo ages ( in days postfertilization [dpf] ) were set after the identification of mouth brooding females ( day 0 ) . Embryos were then removed from the mouths of brooding females and , if required , were maintained for further development in separate culture tanks at 28 °C . All animals were handled in strict accordance with good animal practice as defined by the relevant national and/or local animal welfare bodies , and all animal work was approved by the appropriate committee at Georgia Institute of Technology . Cloned sequences used to generate digoxigenin-labeled antisense riboprobes from Malawi cichlid species have been published [49] , additional sequences have been deposited in GenBank ( http://www . ncbi . nih . nlm . gov; accession numbers FJ594754–FJ594761 and FJ597647 ) . Many genes were identified through partial genome assemblies of L . fuelleborni and M . zebra [50] and cloned from M . zebra and L . fuelleborni cDNA libraries , including all of the Hox genes present in this study ( cichlid Hox sequences from genomic contigs are also published in [105] ) . Overall , Malawi cichlids exhibit almost no sequence divergence; the average nucleotide diversity for comparisons across the Malawi assemblage is 0 . 26% , less than among laboratory strains of the zebrafish [50] . In situ hybridization experiments were based on a protocol from [49] and references therein . Specimens for in situ hybridization were anesthetized in tricaine methanesulfonate ( MS222; Argent ) and fixed overnight in 4% paraformaldehyde ( PFA ) in 0 . 1% phosphate-buffered saline ( PBS ) at 4 °C . Specimens were stage-matched based on external features , including pectoral and caudal fin development and eye development and maturity . All in situ hybridization experiments were performed with multiple specimens ( multiple individuals were fixed at regular intervals , within single broods , then experiments were repeated at least twice with alternative broods ) to fully characterize the expression patterns . After color reaction ( NBT/BCIP; Roche ) embryos were washed in PBS and fixed again in 4% PFA , before whole-mount imaging using a Leica Microsystems stereo microscope ( MZ16 ) . Embryos for sectioning were embedded in gelatin and chick albumin with 2 . 5% gluteraldehyde . The gelatin-albumin blocks were postfixed in 4% PFA before sectioning . Thin sections were cut at 15–25 μm using a Leica Microsystems VT1000 vibratome . | During evolution , teeth originated deep in the pharynx of ancient and extinct jawless fishes . Later , with the evolution of bony fish , teeth appeared in the mouth , as in most current vertebrates , although some living fishes retain teeth in the posterior pharynx . We integrate comparative morphology , paleontology , and molecular biology to infer the genetic control of the first dentition . We identify Hox genes as important components of an ancient dental gene-regulatory circuit and pinpoint subsequent modifications to this gene network that accompanied the evolution of toothed oral jaws . Furthermore , we highlight a set of genes conserved in the construction of all teeth , regardless of location and lineage . This core dental gene network is evolutionarily essential: nature appears never to have made a dentition without it . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"evolutionary",
"biology",
"developmental",
"biology"
]
| 2009 | An Ancient Gene Network Is Co-opted for Teeth on Old and New Jaws |
Mutations in the retinoblastoma tumor suppressor gene ( rb1 ) cause both sporadic and familial forms of childhood retinoblastoma . Despite its clinical relevance , the roles of rb1 during normal retinotectal development and function are not well understood . We have identified mutations in the zebrafish space cadet locus that lead to a premature truncation of the rb1 gene , identical to known mutations in sporadic and familial forms of retinoblastoma . In wild-type embryos , axons of early born retinal ganglion cells ( RGC ) pioneer the retinotectal tract to guide later born RGC axons . In rb1 deficient embryos , these early born RGCs show a delay in cell cycle exit , causing a transient deficit of differentiated RGCs . As a result , later born mutant RGC axons initially fail to exit the retina , resulting in optic nerve hypoplasia . A significant fraction of mutant RGC axons eventually exit the retina , but then frequently project to the incorrect optic tectum . Although rb1 mutants eventually establish basic retinotectal connectivity , behavioral analysis reveals that mutants exhibit deficits in distinct , visually guided behaviors . Thus , our analysis of zebrafish rb1 mutants reveals a previously unknown yet critical role for rb1 during retinotectal tract development and visual function .
Biallelic mutations in the retinoblastoma susceptibility gene rb1 are causal for intraocular childhood retinoblastomas . Rb1 is a member of a gene family that consists of three members , p105/Rb1 , p107/Rb-like1 , and p130/Rb-like2 , collectively known as “pocket proteins” [1] . The activity of these proteins is controlled , in part , by cyclin/cyclin-dependant kinase complexes . Upon activation , Rb proteins bind to an array of proteins , including members of the E2F family of transcription factors to execute a range of cellular functions , including cell cycle exit , terminal differentiation , and cortical cell migration [2] . In humans , germline or somatic mutations occur throughout the 180 kb genomic region spanning the rb1 gene , including its promoter region , exons , and intronic essential splice sites , resulting in bilateral or unilateral retinoblastomas within the first 2 years of life [3] , [4] . Given its clinical relevance , the role of Rb1 during embryonic development and during tumor suppression has been studied intensely , mostly using mouse models [5] . Rb1 is expressed ubiquitously during murine development , postnatally , and continues to be expressed in adults . Embryos harboring non-conditional Rb1 knockout alleles exhibit ectopic proliferation and apoptosis throughout the nervous system and die prenatally at embryonic day 14 . 5 [6] , [7] , [8] . Embryos with conditional loss of Rb1 in the retina display ectopic division and considerable apoptosis of retinal transition cells starting at E10 [9] , [10] , [11] , [12] . Retinas in these animals contain reduced numbers of rods , bipolar cells , and RGCs , yielding a retina with a thin outer nuclear layer and a hypoplastic optic nerve . However , the etiology of optic nerve hypoplasia and if/how Rb1 functions in RGC axonal guidance has not been examined . Similarly , electroretinogram recordings from Rb1 deficient mouse retinas have revealed reduced photoreceptor to bipolar to amacrine signal transmission [10] , yet the behavioral consequences have not been examined . Here , we report that zebrafish space cadet mutants carry a rb1 mutation found in cases of sporadic and familial human retinoblastoma [13] , [14] , [15] , [16] . In zebrafish rb1 mutants , RGC precursors show delayed exit from the cell cycle and hence a delay in the generation of early-born , postmitotic RGCs , whose axons are critical for pioneering the retinotectal tract . This delay leads to RGC intrinsic axon guidance defects , aberrant retinotectal connectivity , and deficits in phototactic behaviors . Together , this work describes a novel model for understanding the developmental role of rb1 and reveals a previously unknown role for rb1 in the formation of the retinotectal tract .
We previously identified two mutant space cadet alleles , based on abnormal startle response behavior to acoustic or tactile stimuli [17] , [18] . Using recombination mapping , we mapped the space cadette226a allele to a 1 . 1 cM interval on chromosome 21 between single nucleotide polymorphic markers in the myo5b locus ( 20 recombinants/2688 meioses ) , and in the ncam1 locus ( 12 recombinants/2688 meioses ) , respectively ( Figure 1A ) . This genomic interval contains several annotated genes , including rb1 , lpar6 , and cystlr2 , which have retained syntenic positional conservation between humans , mice and zebrafish ( Figure 1B ) . Sequencing of rb1 cDNAs isolated from spcte226a larvae revealed the presence of 4 nucleotides inserted between exon 19 and exon 20 . Subsequent sequencing of genomic DNA isolated from spcte226a larvae confirmed a single nucleotide change in the splice donor sequence of intron 19 ( nt1912+1: G to A; Figure 1C ) . This generates a cryptic splice site donor , resulting in the 4 base pair insertion into the rb1 mRNA . This 4 base pair insertion causes a premature stop codon in exon 20 , predicted to truncate the protein at amino acid 677 , thereby severely truncating the B domain and cyclin domain essential for Rb1 function ( Figure 1D ) [15] . Interestingly , identical mutations have been reported in human patients with familial and sporadic forms of retinoblastoma [13] , [14] , [15] , [16] . The zebrafish rb1 gene is 67% similar ( 52% identical-based on amino acid sequence ) to the mouse and human rb1 . The Rb1 protein consists of an A and B domain forming Rb1's binding “pocket” , and a cyclin binding domain ( Figure 1D ) , and shows 81% amino acid similarity ( 66% identical ) between zebrafish and mammalian rb1 in these critical domains . Thus , sequence homology , syntenic conservation , and cDNA sequence analysis provide compelling evidence that space cadet phenotypes are caused by an rb1 gene mutation known to cause retinoblastoma . Sequence analysis of the second mutant space cadetty85d allele did not reveal any changes in the coding sequence or in any of the splice donor and acceptor sites , suggesting that this allele is caused by a regulatory mutation in the rb1 locus . Importantly , analyses of spcte226a and spcty85d mutants revealed no significant differences with regards to the strength of the phenotypes examined below ( Table 1 ) . From here on , we will refer to the space cadet gene as rb1 and describe anatomical and behavioral defects in the rb1te226a allele . During zebrafish embryogenesis , the earliest born RGCs begin extending axons at 32 hpf , cross the ventral midline of the diencephalon to form the optic chiasm at 36 hpf , and project dorsally to the contralateral optic tectum by 48 h to form a retinotectal pathway critical for mediating visually guided behaviors by 120 hpf [19] . We previously reported that 120 hpf stage rb1 mutants display wild type like retinal lamination and expression of terminal RGC cell differentiation markers , but exhibit various RGC axonal pathfinding defects [18] . To determine the temporal onset and spatial site of RGC pathfinding errors in rb1 mutant embryos , we used the ath5:gfp transgenic line expressed in RGCs to examine the development of the retinotectal trajectory [20] . Importantly , between 28–96 hpf we did not detect a difference in the intensity of GFP fluorescence in the retinas of rb1 mutant compared to wild type retinas ( Figure S1 and data not shown ) . At 36 hpf wild type RGCs have exited the retina and pioneered across the ventral midline to form an optic chiasm ( n = 40 , Figure 2A ) . In contrast , only 13% ( n = 30 ) of rb1 mutant retinas had RGC axons that exited the retina ( Figure 2B , 2C ) , suggesting that the loss of rb1 function causes a delay in the initial outgrowth of RGC axons from the retina . At 48 hpf the optic nerve in rb1 mutants was significantly thinner , with a mean diameter of 3 . 04 µm ( n = 18 ) , compared to the thicker optic nerves in wild type siblings ( 13 . 76 µm , n = 22; Figure 2D–2F ) . At 72 and 96 hpf , when wild type RGC axons have reached and innervated the optic tectum , rb1 mutant tecta show a significant reduction in RGC axon tectal innervation ( Figure 2G–2L; see Methods for quantification details ) . Together , these results reveal that rb1 mutants exhibit a delay in RGC axonal outgrowth , leading to a delay in optic nerve development , and reduced innervation of the optic tectum . To determine whether the identified mutation in rb1te226a is causative of the delay in retinotectal development , we injected wild type rb1 mRNA into one-cell stage rb1te226a mutants and examined optic nerve diameter at 48 hpf . Microinjection of wild type rb1 mRNA restored optic nerve diameter in rb1 deficient mutants in a dose dependent manner , demonstrating that mutations in zebrafish rb1 cause RGC outgrowth defects ( Figure 3A–3C , 3E ) . To determine if and to which extent the mutant rb1te226a allele has retained biological activity , we examined the ability of rb1te226a mRNA to rescue retinotectal development in rb1te226a mutants ( Figure 3D–3E ) . Injection of rb1te226a mRNA failed to significantly increase optic nerve diameter in rb1te226a mutants , suggesting that the rb1te226a protein product has very limited , if any , functionality . However , we cannot exclude the possibility that the mutant phenotype is ameliorated by maternal rb1 mRNA and/or protein deposition . Finally , we asked whether Rb1 functions within RGCs for their axons to exit from the retina and enter the retinotectal path . Because zebrafish rb1 is expressed ubiquitously throughout development ( Figure 4A–4B ) , we generated chimeric embryos by transplanting cells at the blastula stage between rb1 mutant and wild type embryos , and then examined their ability to exit from the retina ( Figure 4C ) . A significant fraction of axons from genotypically mutant rb1 RGCs transplanted into wild type hosts failed to exit from the retina ( 19% of retinas showed failure of transplanted rb1 mutant RGC axons to exit , n = 31 , Figure 4E–4F ) , consistent with the low but significant frequency of rb1 mutant retinas in which we observed a complete failure of RGCs to exit from the eye ( 11%; see below ) . Conversely , 100% of rb1 mutant retinas showed exit of axons from transplanted wild type RGCs ( n = 69 , Figure 4D ) . Thus , during zebrafish development rb1 acts RGC autonomously for axons to exit the retina and to form the optic nerve . Given the RGC intrinsic defects observed in rb1 mutants , we next wanted to determine the primary defect leading to the delay of RGC axons to exit from the retina . Rb1 canonically functions to regulate cell cycle checkpoints , promoting cell cycle exit and differentiation of progenitors and suppressing cell cycle re-entry of differentiated cells [2] . In the retina , rb1 has been shown to promote the exit of retinal progenitor cells from the cell cycle into the various postmitotic cell types that populate the retinal lamina [9] , [10] , [11] , [12] . To examine rb1 deficient retinas for cell cycle defects , we labeled wild type and rb1 mutant retinas for M-phase positive nuclei with an anti-phosphohistone-H3 antibody ( anti-pH3 ) during the initial phase of RGC birth and axon outgrowth , between 28 and 36 hpf . During this time window , premitotic ath5 positive retinal progenitors divide , with one daughter becoming a postmitotic RGC and the other maintaining its progenitor potency to give rise to other retinal cell types that become postmitotic at later stages of development [21] . Although the total number of M-phase positive increased with time between 28 and 36 hpf in rb1 mutant and wild type retinas , we observed fewer M-phase positive nuclei in rb1 deficient retinas , compared to wild type retinas , at each time point examined ( Figure 5 ) . One possibility is that the reduction of M-phase retinal precursors in rb1 deficient retinas is due to increased cell death . Indeed , compared to wild type retinas , rb1 deficient retinas showed a slight , but significant increase of TUNEL positive nuclei between 28 and 36 hpf ( Figure S1 ) . Importantly though , comparing the increased number of TUNEL positive nuclei to the decreased number of pH3 positive nuclei in rb1 mutants at 28 , 32 , and 36 hpf revealed that apoptosis accounts for only 18–26% of the observed reduction in M-phase positive retinal precursors in rb1 mutant retinas at each time point examined . This suggests that in rb1 mutant retinas cell death contributes only partially to the deficiency of M-phase positive nuclei ( Figure S1G ) . Thus , the reduction in M-phase retinal precursors in rb1 mutant retinas suggests a prolonged terminal cell cycle for the retinal precursors , which need to exit their final cycle to become the earliest population of postmitotic RGCs . To determine if loss of rb1 function indeed causes an initial delay in the presence of postmitoic , differentiated RGCs , we examined expression of isl2b-gfp , one of the earliest transgenic markers indicative for postmitotic RGCs [22] . We found that in wild type retinas postmitotic , differentiating RGCs marked by isl2b-gfp expression emerged first at 32 hpf , increased significantly in their abundance by 36 hpf , and by 48 hpf isl2b-gfp positive RGCs were densely packed throughout the ganglion cell layer ( Figure 6A , 6C , 6E , n = 25 , 16 , and 29 , respectively ) . In contrast , isl2b-gfp positive RGCs were present in only 13% of rb1 deficient retinas at 32 hpf ( n = 23 ) . Because of the cytoplasmic localization of the GFP signal and the density at which RGCs normally populate the ganglion cell layer , it is difficult to determine the total number of isl2b-gfp positive RGCs . Nonetheless , semi-quantitative analysis revealed that by 36 hpf , isl2b-gfp positive RGCs were present in 90% of rb1 mutant retinas ( n = 29 ) ; however , their distribution with the retina was more similar to that of younger wild type retinas at 32 hpf ( Figure 6B , 6D ) . By 48 hpf , all rb1 mutant retinas harbored isl2b-gfp neurons ( n = 32 ) , although differentiation still appeared to lag in 81% ( n = 32 ) of the rb1 mutant retinas compared to the more densely packed ganglion cell layer in wild type retinas ( n = 29 , Figure 6E–6F ) . Despite the reduced number of RGCs present at 48 hpf , rb1 mutant isl2b-gfp positive RGCs express DM-GRASP , a late marker of RGC differentiation , demonstrating that mutant RGCs were fully differentiated once becoming postmitotic ( Figure S2 ) . Importantly , the number of ath5-gfp positive RGC precursors was unaffected in rb1 mutants ( Figure S1 and data not shown ) . Thus , the rb1 deficiency causes a delay in the transition of RGC precursors to postmitotic RGCs , but not in the specification of RGC precursors . Aside from the reduced population of early born RGCs , rb1 mutant retinas appear grossly normal , and at 120 hpf , show proper lamination by each retinal cell type [18] . Although we did not determine whether birth dating of other retinal cell types is affected in rb1 mutant retinas , netrin-positive exit glial cells and Muller glia cells are present in appropriate numbers and location , indistinguishable from wild-type retinas ( Figure S2 ) . Taken together , these results suggest that a delay in cell cycle exit by rb1 deficient RGC precursors leads to a transient reduction in the early born postmitotic RGCs without consequence to the gross morphology and overall cellular landscape of the rb1 mutant retina . The early born RGCs are located within the central retina and pioneer the retinotectal tract to the contralateral optic tectum [22] . In the absence of the early pioneering RGC axons , the axons of later born , more peripherally located RGCs fail to exit the eye and project aberrantly within the retina [22] . Given the reduced number of these early born , central RGCs in rb1 mutants , we sought to determine whether peripheral RGC axon trajectories were affected . For this , we labeled small groups of RGCs in the anterior peripheral retina with DiO ( green ) and in the posterior peripheral retina with DiI ( red , Figure 7A–7F ) . In 120 hpf larvae wild type larvae , all labeled axons from anterior and posterior RGCs fasciculated shortly after sprouting from their soma and extended as an axon bundle , forming a path directly toward the retinal exit point ( n = 83 , Figure 7A–7B , 7D–7E ) . In contrast , 91% ( n = 65 ) of rb1 deficient retinas harbored a significant subset of axons that had extended aberrantly throughout the retina and failed to exit ( Figure 7C , 7F ) . These results demonstrate that the delayed differentiation of the early born RGCs in rb1 mutants impairs the ability of later born RGC axons to exit the retina . The delayed cell cycle exit and differentiation of pioneering RGCs lacking rb1 may also affect axon navigation by later born RGC axons at key choice points: the ventral midline of the diencephalon and/or the optic tectum . To examine these possibilities , we filled the RGC layer of the left and right eyes of wild type and rb1 mutant larvae with either DiI or DiO , respectively ( Figure 7G ) . In wild type siblings , 99% of dye filled optic nerves projected to their appropriate contralateral tectum ( n = 946 , Figure 7H ) . In contrast , rb1 mutant optic nerves displayed a variety of phenotypes . The majority of rb1 deficient optic nerves were significantly thinner than their wild type counterparts ( 37% , n = 663 , Figure 7I–7J , 7L ) , consistent with what we observed in with ath5:gfp ( Figure 2 ) . In a significant portion of rb1 deficient optic nerves , 17% , RGCs projected to both the contralateral but also to the ipsilateral tectum , indicative of midline pathfinding defects ( n = 663 , Figure 7K–7L ) . Focal DiI/DiO labeling of RGC axons arising from the anterior and posterior retina revealed that retinotopic mapping , a function of retinal cell body location [23] , remains intact in rb1 mutants despite the aberrant pathfinding en route to the optic tectum ( Figure S3 ) . Finally , in 11% of rb1 mutant retinas , there was a complete failure of RGCs to exit from the eye , even at 120 hpf ( Figure 7J ) . Taken together , these results suggest that rb1 deficient RGC axons make intraretinal and midline pathfinding errors , leading to reduced and incorrect tectal innervation . By 120 hpf , zebrafish larvae perform an array of sensorimotor behaviors , including responses to visual stimulation . For example , changes in visual field illumination , such as the sudden absence of light or a shift from uniform to focal illumination , elicit specific , stereotyped turning behaviors [24] , [25] . We first examined the ability of rb1 mutant larvae to perform positive phototaxis , defined as navigating toward a target light source that is presented after extinguishing the pre-adapted uniform light field [25] . Positive phototactic navigation is characterized by larvae first turning towards the target light source and then swimming forward towards the target . As previously reported , when presented with a target light source wild type larvae facing away from the light target show significant initiation of turns , which are preferentially biased towards the light target ( Figure 8A–8B ) . Once facing the target , wild type larvae initiate forward scoot swims ( Figure 8C ) . In contrast , turn initiation in rb1 mutant larvae facing away from the light target was dramatically reduced ( Figure 8A ) . On the few occasions when they initiated a turn , turning direction was unbiased with respect to the light target ( Figure 8B ) . Moreover , rb1 mutants facing the light target did not show an increase in forward scoot swim initiation above baseline ( Figure 8C ) . To further determine whether rb1 mutants respond to more extreme changes in illumination , we examined their ability to perform an O-bend response to a visual dark flash stimulus , a sudden extinction of light [24] . Again , compared to their wild type siblings , rb1 mutants displayed a minimal O-bend response to dark flash stimulation ( Figure 8D ) . Despite their impaired visual responses , rb1 mutants showed no difference in the spontaneous initiation of turning or swimming behaviors compared to wild type siblings ( Figure 8D ) . Importantly , the kinematic parameters of spontaneously occurring turning and swimming movements were indistinguishable between rb1 mutants and their wild type siblings , demonstrating that the neural circuits required for initiation and execution of turning behaviors are largely intact in rb1 mutants . Together , these results demonstrate that rb1 mutants exhibit visual deficits .
Children with biallelic germline or sporadic inactivation of rb1 are likely to form ocular tumors during early childhood . Initially , the retinas of affected individual show an otherwise grossly normal morphology . In contrast , even conditional rb1 knockout mouse models exhibit ectopic proliferation and cell death leading to significant morphological defects throughout affected retinas . We find that inactivation of the zebrafish rb1 gene through a rb1 causing mutation results in mutant retinas that display very limited signs of cell death , with differentiated retinal cell types that are properly laminated , similarly to childhood retinas lacking rb1 . Thus , the fairly ‘normal’ retinal landscape of zebrafish rb1te226a mutants provided us with a unique opportunity to investigate if and how rb1 is required to establish the retinotectal projection . Our analysis reveals a RGC autonomous requirement for rb1 in regulating RGC axon pathfinding within the retina and at presumptive choice points en route to the optic tectum . Moreover , we demonstrate that zebrafish rb1te226a mutants exhibit deficits in visually guided behaviors , suggesting that the retinotectal path defects in rb1 mutants may be sufficient to impair vision . Together , this work reveals a novel role for rb1 in the establishment of RGC axon projections during development and establishes a unique model for understanding the developmental and tumor suppressor roles of the rb1 gene . Zebrafish rb1te226a mutants harbor a human retinoblastoma causing rb1 gene mutation . The mutant protein is truncated in the B-domain and lacks the cyclin-binding domain , reducing Rb1's capacity to form a ‘pocket’ , and reducing its capacity for phosphorylation by cyclin dependent kinases [2] , [26] . Consistent with the notion that the rb1te226a mutant allele is largely non-functional , mRNA over-expression in rb1 mutants does not ameliorate the rb1 mutant phenotype . Despite the absence of biological activity of the truncated rb1te226a protein , mutant zebrafish show a significantly milder retinal phenotype compared to conditional or even germline rb1 mouse knockouts [6] , [7] , [8] , [9] , [10] , [11] , [12] . One possible explanation is the strong maternal contribution of rb1 in zebrafish ( Figure 4A ) , which may suppress phenotypic expressivity at early stages of development . Consistent with this idea , formation of the initial scaffold of axon tracts during the first day of development appears unaffected in rb1te226a mutants , yet visual and hindbrain pathways that develop after the first day of development show defects [18] . In humans the rb1 nt1960+1 mutation , which is identical to the zebrafish rb1te226a mutation , causes ocular tumors [13] , [14] , [15] , [16] , raising the possibility that zebrafish rb1 mutants might also develop tumors as juveniles . However , rb1 mutants fail to inflate a functional swim bladder , and die ∼7 days post fertilization , precluding the analysis of ocular tumors in juveniles . Although wild type rb1 mRNA injection rescues the early RGC and retinotectal defects through 48 hpf , these transiently rescued rb1te226a mutants do not survive beyond 7 days of development , indicating that rb1 plays a critical role after the injected mRNA has been degraded . Establishing stable , inducible rb1 transgenic lines to rescue developmental deficits will therefore be required to monitor juvenile and adult zebrafish for retinal tumors . In rb1te226a mutants , a significant subset of RGC axons fail to exit the retina , and many of the exiting axons then project incorrectly to the ipsilateral tectum , revealing a previously unrecognized requirement for rb1 in regulating axon pathfinding . One possible explanation for the RGC guidance defects is that in rb1te226a mutant RGCs the expression of guidance factors might be disrupted . Interestingly , cortical cell migration was shown to be dependent on rb1 regulated neogenin expression [27] , suggesting that rb1 deficient RGC axons might lack guidance factors required to navigate towards the retinal exit point and properly cross the ventral midline . To investigate this possibility , we performed microarray gene expression analysis of the retina and brains of 32 hpf rb1te226a mutants ( MAW and MG , unpublished ) . However , this approach did not reveal a significant change in the expression levels of neogenin or other known axon guidance genes . Although it remains possible that rb1 regulates expression of un-identified guidance factors , it is more likely that rb1 regulates axon pathfinding indirectly by ensuring the timely exit of RGC precursors from the cell cycle and hence the appropriate temporal appearance of differentiated RGCs . In fact , genetic ablation of the earliest born RGCs prevents the formation of the retinotectal tract [22] . This suggests that RGC birth order imprints a critical hierarchical pathfinding role on RGC axons , such that axons from the earliest born RGCs pioneer the retinotectal tract that later born RGC axons will follow [22] . The delayed onset of RGC birth in rb1te226a mutants may therefore reduce the population of pioneering RGCs present during a restricted window of environmentally expressed guidance factors . Zebrafish rb1te226a mutants display deficits in the acoustic startle response [17] , [18] and in visually guided behaviors , reflecting the importance of rb1 function for the development of neural circuitry underlying behavior . The deficits in startle behavior are due to defects in a small subset of hindbrain neurons , the spiral fiber neurons [18] , and giving the results presented here , it is tempting to speculate that rb1 plays a similar role for the transition of these neurons from precursors to postmitotic neurons . Unfortunately , markers that follow the development of spiral fiber neurons are not available , precluding such analysis . Therefore , we focused on the well-characterized development of RGCs . Although we demonstrate a defect in the early development of these cells and their axonal connectivity , we cannot exclude the possibility that zebrafish rb1 mutants exhibit defects in the development and/or function of other retinal cell types , and that these defects contribute to the deficits in visual behaviors we observe . Future analysis of transgenic lines expressing the wild type Rb1 gene in individual retinal cell types will reveal which cell type ( s ) and connections are causative of the visual deficit . In summary , we report a zebrafish mutant carrying a human disease causing rb1 mutation , which reveals novel roles of rb1 in regulating RGC axon pathfinding and visually guided motor behavior . Furthermore , these mutants provide a non-murine vertebrate model of rb1 and offer new potential for identifying the elusive retinoblastoma cell of origin and further insight into the developmental role of rb1 .
All experiments were conducted according to an Animal Protocol fully approved by the University of Pennsylvania Institutional Animal Care and Use Committee ( IACUC ) on January 27 , 2011 , protocol number 803446 . Veterinary care is under the supervision of the University Laboratory Animal Resources ( ULAR ) of the University of Pennsylvania . The zebrafish ( Danio rerio ) strain used in this study was the spcte226a allele ( now referred to as rb1te226a ) of space cadet [17] , [18] , maintained on a mixed TLF and Tubingen background . The rb1te226a allele was also crossed into the ath5:gfp and isl2b:gfp transgenic backgrounds for RGC analysis [20] , [22] . rb1te226a+/−;ath5:gfp+/− or rb1te226a+/−;isl2b:gfp+/− adults were always crossed with rb1te226a+/−;TLF adults to ensure that rb1te226a embryos analyzed for GFP-expressing RGCs were hemizygous for GFP . Throughout the manuscript , rb1−/− , “rb1 deficient” , and rb1 mutant refers to rb1te226a homozygotes . The other space cadet allele spcty85d [17] was only used where mentioned . Embryos were collected in the morning , maintained on a 14/10 hour light/dark cycle at 28°C , and staged as described previously [28] . Larvae were raised in 6 cm plastic Petri dishes at a density of 20–30 per 7 mL in E3 medium ( 5 mM NaCl , 0 . 17 m mM KCl , 0 . 33 mM CaCl2 , 0 . 33 mM MgSO4 ) with medium changes at 48 hpf ( hours post fertilization ) and 96 hpf . Behavioral experiments were conducted on 120 hpf larvae . A three generation mapping cross between rb1te226a heterozygous and WIK fish was generated , and pools of 25 F2 mutant and F2 sibling 5 dpf larvae were collected in the F2 generation and used for bulk segregant mapping ( see Table 2 for simple sequence length and single nucleotide polymorphic markers ) . Mutant larvae were identified by performing successive , unilateral C-bends to acoustic or tactile stimulation [17] , [18] . To identify the mutation , cDNA was prepared following total mRNA extraction from 5 dpf larvae as previously described [29] . rb1 cDNA was amplified with primers ( rb1:1–6 , Table 2 ) designed against overlapping regions of the rb1 reference sequence ( Ensembl ) with the following RT-PCR conditions: 94°C for 3 min and then 40 cycles of 94°C for 45 sec , 57°C for 1 min , and 70°C for 1 min . Products were gel purified and cloned into the pCR2 . 1-TOPO-TA vector for sequencing . After detecting a frameshift and 4 nucleotide addition to the end of exon 19 in rb1te226a cDNA clones , gDNA was isolated from 5 dpf larvae , and intron 19 was amplified with the rb1:8 primers , using identical PCR conditions to those described above . For rb1 RNA injection , cDNA was prepared from genotyped homozygous wild type or rb1te226a mutant 5 dpf larvae ( dCAPS protocol , see below ) and amplified with the rb1:FL primers ( similar conditions as above , except extension time increased to 3 min ) , which includes the coding region of rb1 , and cloned into the pCS2+ vector . Wild type rb1 and rb1te226a mRNA was prepared using the mMessage mMachine kit ( Ambion , NY ) and injected at the 1-cell stage at doses ranging from 1–100 picograms . Embryos injected with 20 or greater picograms of rb1 mRNA showed gross morphological abnormalities and necrosis , whereas embryos injected with 10 picograms or less appeared morphologically normal . To genotype rb1te226a embryos , we developed a dCAPS assay [30] using the dCAPS program ( http://helix . wustl . edu/dcaps/dcaps . html ) to design appropriate primers ( Table 2 ) . After gDNA isolation , PCR was performed as described above . The PCR product is then digested with SspI ( New England Biolabs , Ipswich , MA ) , cleaving the rb1te226a allele and producing a 120 bp fragment that can be distinguished from the 150 bp wild type allele on a 3% agarose gel containing 1 . 5% Metaphor agarose ( Lonza , Rockland , ME ) . All genotyping , except for BrdU labeled embryos , was performed following immunolabeling experiments . For immunostaining , embryos were fixed in 4% paraformaldehyde ( PFA ) overnight at 4°C , permeabilized with 1 mg/mL collagenase , and blocked for 1 hour with 5% normal goat serum in 0 . 1 M phosphate buffer . Embryos were then incubated in the primary antibodies anti-GFP ( 1∶200 mouse JL8 , Clontech , Mountain View , CA or 1∶500 rabbit , Invitrogen , Carlsbad , CA ) , anti-phosphohistone-H3 ( Millipore , Charlottesville , VA ) , 1∶100 anti-BrdU ( Roche , Branchburg , NJ ) , and/or 1∶50 A2-J-22 polyclonal antisera ( recognizes carbonic anhydrase II , kindly provided by Dr . P . Linser ) overnight at 4°C in blocking solution , washed out , and then detected by the addition of AlexaFluor488 or AlexaFluor594 conjugated secondary antibodies ( 1∶500 , Invitrogen , Carlsbad , CA ) . TUNEL assay was performed as previously described [31] using Apoptag Peroxidase In Situ Apoptosis Detection Kit ( Chemicon , Temecula , CA ) . After staining , samples were mounted in DAPI containing Vectashield ( Vector Labs , Burlingame , CA ) . Images were acquired with a Zeiss 710 confocal laser scanning microscope ( LSM 710 ) using ZEN2010 software . For in situ hybridization , digoxygenin-UTP labeled antisense riboprobes for rb1 were synthesized and hydrolyzed from the full length rb1 cDNA construct [32] . Whole-mount in situ hybridization was performed as described previously [33] . Images were acquired with a Zeiss Axioskop compound microscope . For RT-PCR based expression analysis , the rb1:FL and B-actin primers ( Table 2 ) were run against cDNA prepared from total mRNA extracted from 25 embryos/larva at each stage . 120 hpf larvae were anesthetized ( 0 . 01% Tricaine ) and fixed in 4% paraformaldehyde at 4°C overnight . Larvae were removed from fix , washed briefly in phosphate buffered saline ( PBS ) , and mounted dorsal side up for whole retinal injection or laterally for discreet RGC labeling on glass microscope slides in a bed of 1 . 5% agarose . To label all RGCs , the vitreal space of each eye was filled with either of the fluorescent lipophilic dyes DiI ( red ) or DiO ( green ) ( Molecular Probes , Eugene , OR ) dissolved in 1% chloroform , using a WPI PV820 picopump injector fitted with a glass micropipette . For discreet labeling , a small region of the exposed eye was labeled with pulses of DiI/DiO dissolved in 0 . 5% dimethylformamide . Injected larvae were kept moist with PBS and incubated overnight at room temperature in a humidity chamber in darkness . Larvae were then examined for phenotype analysis using a Zeiss Axioplan compound fluorescent microscope . Eyes were carefully removed from selected representative larvae , which were then remounted on coverslips in agarose for imaging . Images were recorded using a Zeiss 510 confocal laser scanning microscope ( LSM510 ) and Zeiss LSM510 analytic software . For transplant direction wild type donor into space cadet host , wild type transgenic Tg ( ath5:gfp ) and rb1te226a heterozygous fish were used to generate wild type GFP expressing donor embryos and non-GFP expressing rb1te226a mutant embryos , respectively . For transplant direction space cadet donor into wild type host , rb1te226a; Tg ( ath5:gfp ) double heterozygotes and either TU or TLF strain wild type mating pairs were used to generate rb1te226 GFP expressing donor embryos and non-GFP expressing wild type embryos , respectively . Once the appropriate donor-host embryos were collected , embryos were immediately placed in E3 medium and kept at room temperature . Donor embryos were pressure injected into the yolk sac at the 1–2 cell stage with the lineage tracer tetramethylrhodamine dextran , 3 Kd , 5% w/v ( Molecular Probes , Eugene , OR ) dissolved in 0 . 2 M KCL and filter sterilized . Donor and host embryos were then incubated at 28 . 5°C in E3 medium in darkness to grow synchronously to the 1000 cell stage . Embryos were then transferred into room temperature complete E2 medium ( E2 ) to retard growth , and dechorionated using Pronase ( 1∶50 in E2 of 30 mg/ml stock , Roche ) in glass 60 mm petri dishes . Dechorionated embryos were washed extensively with E2 , transferred using a fire polished glass Pasteur pipette into individual wells in a transplantation dish containing E2 , and properly oriented . Transplantation needles were made using #1BBL No Fil borosilicate glass pipettes ( WPI ) , pulled to produce fine tips in a P87 pipette puller ( Sutter Instruments , Novato , CA ) , broken at various diameter openings , and polished using a microforge . Needles were then inserted into a standard pipette holder connected to a modified manual injection apparatus , and mounted in a micromanipulator arm for precision control . Thirty to fifty blastomeres were carefully removed from the donor embryo using the transplantation pipette/manual injector apparatus , and transferred into the adjacent host embryo at the apex of the animal pole ( eye/nose region ) . Operated embryos were maintained in the transplantation dish wells in E2 at 28 . 5°C in darkness following transplantation , and were allowed to develop undisturbed until epiboly completed . Embryos were then transferred from the transplantation wells/dish into either separate 1 . 5% agarose coated 60 mm plastic Petri dishes for donors and hosts , or 1 . 5% agarose coated wells in 12-well tissue culture plates as host-donor pairs , depending on the direction of the transplant , and incubated at 28 . 5°C for five days . The later was necessary in order to correctly identify rb1te226a donors from each donor-host pair , as the motility phenotype does not manifest itself until 120 hpf . 120 hpf larvae were screened for the presence of GFP expressing RGC clones using a Leica MZFIII fluorescence stereomicroscope , and further analyzed for misprojecting RGC axons using a Zeiss axioplan compound fluorescence microscope . Host larvae suspected of containing misprojecting RGC axons ( ie , not exiting the eye , or midline defects ) were then fixed and stained with anti-GFP antibody as described above , and imaged using a Zeiss LSM510 microscope and Zeiss LSM510 analytic software . Confocal z-stacks were of sufficient depth ( 150–220 µm ) to insure optic nerves were not inadvertently missed . Confocal stacks were processed into maximum and/or summation intensity projections using ImageJ for quantification . We used the full width half maximum algorithm to calculate optic nerve diameter from maximum intensity projections of GFP-labeled retinal ganglion cell axons . Tectal innervation was determined by making 20 µm summation projections of GFP labeled tecta , tracing the area of the labeled tectum to determine the Raw Integrated Density ( RID ) per µm2 , and subtracting the RID/µm2 of an unlabeled , background region . TUNEL and anti-pH3 labeled nuclei were counted from 30 µm stacks using Volocity ( PerkinElmer , Waltham , MA ) , with individual cells distinguished by fluorescent intensity and size . Statistical analysis was performed on all data using the Graphpad prism software ( www . graphpad . com ) . Behavioral experiments were performed on 120 hpf larvae and analyzed with the FLOTE software package as previously described [25] , [34] , [35] . rb1te226a and wild type siblings were identified based on acoustic startle behavior [17] , [18] and then grouped by phenotype for visual behavior testing in 6 cM petri dishes at a density of 12 fish per dish . For all behavioral experiments , N = 48 rb1te226a and 48 wild type sibling larvae . For phototaxis experiments , video recordings were triggered every 500 msec , with each recording covering a 400 msec time window , for a total duration of 4 sec of recorded behavior . Each group of 12 larva were subjected to 3 rounds of phototaxis testing , with 3 min between trials . Orientation of larvae to target light was determined at the beginning of each 400 ms recording as previously described [25] , such that the behavior of each larva was tested multiple times and in different orientations with respect to the target light . Therefore , the N for Figure 8A and 8C ranged from 75 to 547 for wild type siblings and 136 to 311 for rb1te226a larvae . In Figure 8B , the N ranged from 39 to 106 for wild type siblings and 13–33 for rb1te226a larvae . For dark flash response experiments , N = 4 groups of 12 larvae . Spontaneous behavior was analyzed on individually housed larvae on a 4×4 grid array . | Before an organism can execute necessary behavioral responses to environmental stimuli , the underlying neural circuits that regulate these behaviors must be precisely wired during embryonic development . A properly wired neural circuit is the product of a sophisticated collaboration of multiple genetic pathways that orchestrate cell type specification , the extension and growth of the cell processes that connect each circuit component , and the refinement of these connections . In an unbiased genetic screen designed to identify the genes required for proper circuit formation in developing zebrafish embryos , we identified a human disease causing mutation in the retinoblastoma-1 ( rb1 ) gene that disrupts the formation of the zebrafish visual circuit . rb1 canonically functions to regulate the cell cycle , and when mutated the loss of rb1-mediated cell cycle control elicits childhood ocular tumor formation . Genetic models of rb1 have been developed to study the developmental role of rb1 in the retina; however , ectopic cell proliferation and death within the retina have largely precluded the ability to evaluate the formation and integrity of neural circuits connecting the retina with the brain . In this study , through genetic and cellular analysis of a zebrafish rb1 mutant , we reveal a novel role for rb1 in regulating the establishment and functionality of the visual circuitry . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biology"
]
| 2012 | The Tumor Suppressor Gene Retinoblastoma-1 Is Required for Retinotectal Development and Visual Function in Zebrafish |
Differential equation models that describe the dynamic changes of biochemical signaling states are important tools to understand cellular behavior . An essential task in building such representations is to infer the affinities , rate constants , and other parameters of a model from actual measurement data . However , intuitive measurement protocols often fail to generate data that restrict the range of possible parameter values . Here we utilized a numerical method to iteratively design optimal live-cell fluorescence microscopy experiments in order to reveal pharmacological and kinetic parameters of a phosphatidylinositol 3 , 4 , 5-trisphosphate ( PIP3 ) second messenger signaling process that is deregulated in many tumors . The experimental approach included the activation of endogenous phosphoinositide 3-kinase ( PI3K ) by chemically induced recruitment of a regulatory peptide , reversible inhibition of PI3K using a kinase inhibitor , and monitoring of the PI3K-mediated production of PIP3 lipids using the pleckstrin homology ( PH ) domain of Akt . We found that an intuitively planned and established experimental protocol did not yield data from which relevant parameters could be inferred . Starting from a set of poorly defined model parameters derived from the intuitively planned experiment , we calculated concentration-time profiles for both the inducing and the inhibitory compound that would minimize the predicted uncertainty of parameter estimates . Two cycles of optimization and experimentation were sufficient to narrowly confine the model parameters , with the mean variance of estimates dropping more than sixty-fold . Thus , optimal experimental design proved to be a powerful strategy to minimize the number of experiments needed to infer biological parameters from a cell signaling assay .
Biological cells continuously process stimuli that they receive from their inside and outside , using interconnected signaling pathways . Since strength , timing , and combination of inputs typically determine the output behavior in a non-linear manner , dynamic models of such signaling networks in the form of differential equations are becoming widely used to complement experimental studies [1] , [2] . A major challenge to systems biology is therefore , to infer the rate constants , binding affinities and other parameters of such models from actual measurement data [3] , [4] . Parameter estimation is a widely applied method for model calibration in mechanical and chemical engineering , and — if transferred to the field of cell signaling — may facilitate the development of predictive models of disease and therapy [5] . However , because measurements are always afflicted with error , parameter estimates which minimize the discrepancy between model and data often come along with high uncertainty . Gutenkunst et al . [6] studied parameter sensitivities in 17 published models in the field of systems biology , and concluded that parameter estimation in biology is difficult not only because of often large measurement error , but also because model behavior is often insensitive to combined changes of parameter values . The resulting working models thus remain limited in mechanistic insight and in their capability to predict system dynamics in unforeseen conditions . For a given set of differential equations that describes a dynamic process , not every experimental protocol is equally suited to produce data from which parameters can be inferred . However , it is often possible to control a dynamic process in such a way that predicted measurements are very sensitive to the parameter values that underlie the dynamic system . In that case , the data would be better suited for parameter estimation because it restricts the range of parameter values that could explain the data . To efficiently determine experimental protocols that sensitize model predictions to parameter values , optimal control problems need to be solved . Körkel et al . [7] , [8] developed and implemented an efficient scheme for solving such optimal control problems in chemical engineering , given deterministic models in the form of ordinary differential equations ( ODE ) or differential algebraic equations ( DAE ) . These solutions propose how to best feed a reactor , control its temperature , and when to take measurements in order to estimate parameters with highest accuracy , while obeying constraints imposed by feasibility or cost . Although the computed experimental protocols and sampling schemes are unavoidably based only on approximate models and parameter values , experience shows that data evaluated from optimized experiments are typically more informative than data from intuitively planned experiments[8] . By designing experiments sequentially , the iterative process of performing experiments can be paralleled by an increasing predictive power to plan the next experiment . Several computational studies on models of the MAP kinase cascade , apoptosis , STAT5 activation , and the pathway [9]–[14] suggested that this strategy could also be valuable for the experiment-driven modeling of signal transduction networks . These studies showed that given a specific model of a biological process , several numerical methods can be used to identify stimulus and sampling patterns that would reveal parameter values with higher accuracy . The practical value of such an approach could be immense , because experiments are sensitized to reveal otherwise hidden biological parameters . However , despite a wealth of diverse numerical approaches [8] , [9] , [11] , [13] , [15] ( reviewed in [16] ) , those methods have not been adopted by experimentalists . The theoretical nature of previous work left it unclear , whether optimized experiments would drive a true biological system within the predictive range of a model [15] , how experimental uncertainty and imperfection affect the attainable accuracy of estimates , and how design performance is affected by the possibly large discrepancy between true biology and a differential equations model . Moreover , the increased complexity of proposed stimulus designs has raised the question whether numerically optimized experiments are even feasible enough to return net savings of experimental effort in a real-world cell biology lab . Here , we addressed these questions explicitly by employing a numerical approach to enrich an existing single-cell assay for pharmacologically , biochemically , and clinically relevant parameters by an optimized protocol . Our study focused on the accumulation of the phosphatidylinositol 3 , 4 , 5-trisphosphate ( ) second messenger lipid and the subsequent recruitment of downstream signaling elements through pleckstrin homology ( PH ) domains . is produced by the catalytic subunit p110 of phosphoinositide 3-kinase ( PI3K ) in response to chemokine or other receptor stimuli in the plasma membrane of eukaryotic cells . Slow diffusion and rapid degradation of result in gradients that are steep enough to mediate cell polarity as in migration and differentiation , but elevated synthesis of is associated with cancer . Signaling elements downstream of PI3K , like the oncogenic Akt protein , engage by binding to -enriched membranes through pleckstrin homology ( PH ) domains . Previously , Suh et al . [17] described a chemical method for activating endogenous p110 at the plasma membrane . This technique makes use of the rapamycin-dependent heterodimerization of the FK506-binding protein ( FKBP ) with the mammalian target of rapamycin ( mTOR ) . A genetic fusion of cyan fluorescent protein ( CFP ) , FKBP12 , and a peptide from the regulatory subunit p85 of PI3K was constructed ( CF-p85 ) that resides in the cytosol and does not intrinsically stimulate p110 activity . However , upon addition of the rapamycin derivative iRap to cells co-transfected with a construct made from the N-terminal plasma membrane-targeting sequence of Lyn and the FKBP-rapamycin binding domain of mTOR ( Lyn-FRB ) , CF-p85 translocates to the plasma membrane and induces the production of ( Figure 1A ) . Elevated concentrations of can be monitored through translocation events of a construct in which yellow fluorescent protein is fused to the PH-domain of Akt ( Y-PH ) . Our experimental setup observed translocation dynamics of CF-p85 and Y-PH in NIH 3T3 fibroblasts by live-cell confocal fluorescence microscopy in order to infer the kinetic parameters of the underlying dynamic system . If such parameters were inferable with high statistical confidence , it would be possible , for example , to compare the inhibitory effects of different cancer drug candidates on PI3K directly and in situ . To shed light on nonlinear signal transmission , it would be desirable also to reveal differential affinities of the more than 20 PH domains that were shown to be responsive [18] . Moreover , the degradation of is compromised in many tumors , whereas in others , sustained growth factor inputs could be responsible for elevated downstream signaling . We asked if the established single-cell assay would yield dynamic data from which such relevant properties could be inferred by fitting a differential equations model . However , we found that our intuitively planned and often used experimental protocol was not informative about these parameters . Even for a simplified model , the parameter estimates were largely undefined and covaried strongly . Since the traditional experimental protocol involved two drugs , iRap and the PI3K inhibitor LY294002 ( LY29 ) , we utilized a numerical optimization method [7] , [8] to design better concentration-time profiles for these drugs in order to reveal model parameters with higher accuracy . We performed these numerically optimized experiments sequentially , and found that the uncertainty of parameter estimates could be reduced dramatically and covariance be largely eliminated . Given the potential implications on systems biology methodology , the paper derives meaningful constraints of the optimization problem as it pertains to live-cell experimentation , and investigates the implications of experimental uncertainty and imperfection .
In order to infer biological parameters from the observed translocation dynamics , we set up a differential equations model ( Figure 1C , Text S1 ) that reflects this system . It is a system of ordinary differential equations ( ODE ) of the form ( 1 ) It comprises differential states for the amounts of ( ) , of CF-p85 bound to iRap ( ) , of CF-p85 in a ternary complex with Lyn-FRB via iRap ( ) and of Y-PH bound to the plasma membrane ( ) . All other molecular species were derived from total amounts via mass balance . The parameters of this model are the rate constant for the degradation of , the inhibitor constant for the effect of LY29 on p110 , the rate constants and for the stepwise formation of the ternary complex , for the attachment of Y-PH to -enriched membranes , and for its dissociation ( Figure 1C , blue ) . This simplified model assumes that the formation of the complex of CF-p85 , iRap , and Lyn-FRB begins with the binding of iRap to CF-p85 [19] , that both this reaction and the subsequent binding of Lyn-FRB are irreversible , and that Y-PH does not shield from degradation . Because the amounts of the exogenously expressed CF-p85 , Lyn-FRB , and Y-PH vary from experiment to experiment , the total amount of each of those species , , and were additional model parameters that solely served the purpose of calibration ( Figure 1C , gold ) . The time-dependent concentrations of iRap and LY29 were represented as piecewise constant functions stepping from at 5 min for iRap , and from for LY29 10 min later: ( 2 ) ( 3 ) where denotes the concentration of the respective drug in experiment at time ( Figure 1C , green ) . The cytosolic fluorescence intensities of CF-p85 and Y-PH were measured for a single cell from 25 frames captured every minute . In order to relate the measured values from frame to a cytosolic concentration in the model at time , the observation functions ( 4 ) ( 5 ) were defined . Using these observation functions , the model can be fitted to the two fluorescence trajectories by finding a parameter set that minimizes a weighted least squares functional: ( 6 ) The measurement errors were estimated from individual frames ( see Methods ) . For collective fits from independent experiments , the least squares functional is a sum over of the above expression ( 6 ) . Figure 1D shows a schematic of the parameter estimation concept . For the experimental treatment described by , the model predicts observations that should coincide — within the range of measurement error — with data . If the predicted observations depend on the model parameters , it is possible to infer the parameter set that minimizes the discrepancy between model and data . If this discrepancy falls below the confidence level which is defined by the measurement errors , the model is considered a “fit” . To infer parameter values from microscopy experiments , i . e . to minimize expression ( 6 ) , a multiple shooting Gauss-Newton type algorithm was employed ( see Methods ) . The resulting parameter set yielded good agreement between model and data ( Figure 1E ) . However , the behavior of biological systems is often robust to changes in rate constants , binding affinities , and concentrations . Thus , large regions in parameter space often correspond to similarly good model predictions , and therefore , it is often not possible to infer parameter values from experimental data with satisfying precision [6] . Figure 1D depicts this uncertainty region as the area where the least squares functional falls below the confidence level hyperplane ( blue shaded region ) . In other words , the uncertainty of parameter estimates is inversely related to the sensitivity of testable model predictions to changes in parameter values : ( 7 ) where is the matrix of diagonalized and are the vectorized predictions of the ODE model . From these sensitivities , the parameter covariance matrix can be obtained . Note that the covariance matrix depends on the experimental protocol . describes the uncertainty region in parameter space in proximity of the solution of the estimation problem . In particular , the diagonal element of this matrix approximates the variance of a specific parameter estimate . For the initial fit shown in Figure 1E , the relative uncertainties indicate that the intuitive experimental protocol is not very suitable to define the parameters of interest ( Figure 1F , Table 1 ) . In order to maximize the information gained about kinetic and pharmacological properties , we asked if we could find a second experimental protocol with given that would minimize the predicted parameter variances from a collective fit . Such a minimum would correspond to a smaller confidence region ( shaded blue in Figure 1D ) . Figure 1G shows the ellipsoid approximation of this confidence region for a case with two parameters . Different geometric properties of this confidence ellipsoid correspond to properties of the covariance matrix . We chose to minimize the mean of the diagonal elements ( the trace of divided by the number of parameters ) to balance the experimental effort between parameters . We would like to note that several other properties of have been proposed as design criteria . For example , minimizing the determinant corresponds to minimizing the volume of the uncertainty region , whereas minimizing the maximum eigenvalue ( ) corresponds to minimizing the longest half-axis ( Figure 1G ) . Using the trace criterion , the second experimental design was optimized with the previous experiment included in the prediction of and based on the current parameter estimates . were piecewise constant functions that were discretized into three intervals . This means that the composition of extracellular buffer was allowed to change twice at 5 and 15 min with measurements taken every minute for 24 min . The maximum concentrations of drug allowed were for iRap or for LY29 . The traditional experimental protocol served as an initial design that was optimized by a numerical method that employs in its core a sequential quadratic programming ( SQP ) method [20] ( see Methods ) . SQP is an efficient derivative-based local optimization method that approaches the Karush-Kuhn-Tucker conditions ( local optimality conditions ) iteratively . It is particularly suited for large optimization problems with many design variables such as drug concentrations . How good a local minimum is approximated by an SQP run depends on termination critera . Typically , good improvements of the initial design can be observed in practice ( see for example [8] ) . Here , we obtained from a single SQP run an experimental protocol ( Figure 2A ) for which the mean of variances was predicted to be more than 50-fold reduced when compared to a repetition of the established protocol . The numerically optimized protocol suggested that we add a low dose of iRap for the first 5 min . Then , the concentration of iRap should be increased to in the presence of LY29 for the next 10 min . Finally , LY29 should be washed out in the presence of iRap ( Figure 2B ) . While in general , the predicted improvement of parameter estimates is subject to parameter and model uncertainty , the predicted gain of information appeared promising ( Figure 2C , blue arrows ) . Figure 2D shows confocal microscopy images from the realization of this numerically optimized experiment . Translocation of CF-p85 was induced submaximally using 30 nM iRap for 5 min , but was then reinforced by the addition of iRap in the presence of LY29 . The addition of iRap induced rapid translocation of CF-p85 while Y-PH remained cytosolic due to the inhibitory effect of LY29 on PI3K . Ten minutes later , LY29 was washed out with extracellular buffer containing iRap . The buffer exchange of most of the solution occurred in less than one minute by making use of combined addition of buffer and vacuum suction ( Figure 2B ) . Immediately after the washout of LY29 , Y-PH started to translocate to the plasma membrane , indicating that the production of resumed ( see also Video S2 ) . The experiment illustrates an important problem that arises often from numerical optimization of biological experiments , in that these experiments are often challenging and introduce additional uncertainty . For example , the withdrawal of a growth factor stimulus may only partially reduce signaling due to tight receptor binding . Similarly , the knock-down of a protein by siRNA is never complete , and the magnitude of reduction in protein expression may vary from cell to cell . In this particular case , it was conceivable that some LY29 remained in the imaging chamber after the washout . While such experimental uncertainties are often irrelevant to traditional biology , the implications on parameter estimation from biological data can be significant . Thus , the implications of experimental imperfection on the performance of numerically optimized and idealized experiments need to be addressed . In this specific case , we accounted for experimental uncertainty by estimating the residual concentration from the time course data as part of the parameter estimation problem . To this end , the concentration of LY29 ( eq . 3 ) was redefined posteriorly: ( 8 ) The parameter estimates obtained from a combined fit to the traditional and the optimized experiment ( Table 1 ) calibrated the model in excellent agreement with data ( Figure 2E ) . While the estimate for clearly suggested that residual LY29 was present , it did not significantly affect the large improvement of the resulting parameter estimates ( Figure 2C , green arrows; Table S1 ) . A practical limitation in live-cell imaging experiments is that only a limited number of buffer changes can be performed without exerting cell stress . Instead of employing a fixed time-point model for adding drugs , it could also be beneficial to optimize at which time points drugs are added . We therefore derived a constraint formulation that extends the previous approach by optimizing the time points of two buffer changes in addition to their drug composition . To this end , the system of ODEs was subjected to a time transformation , by which the differentials were multiplied by a piecewise constant control function that was discretized into the three segments of incubation . In this parameterization , large values of correspond to longer intervals between buffer changes with more measurements , which was reflected by transforming the error model accordingly to . In addition , a dynamic constraint formulation was employed to ensure that the duration of the optimized experiment would equal to the duration of the established protocol: ( 9 ) Further , to facilitate the experiment , we limited the protocol to drug additions ( see Methods ) . This also reflects a protocol constraint in high-throughput microscopy where wash-out procedures are difficult to implement . Under these constraints , experimental designs were optimized for different design criteria ( Figure 3A , 3B ) based on the diagonal elements , the maximum eigenvalue , and the determinant of . While for any experiment only small overall improvements of the parameter estimates were predicted when compared to a repetition of the established protocol , the three optimized designs mostly aimed at the uncertainty of . Such an opportunity for narrowing down selectively a single parameter would have been difficult to predict by intuition . In this specific situation , we used the determinant criterion to optimize particularly for the inhibitor constant of LY29 . According to this second optimized protocol ( Figure 3C ) , cells were treated with iRap for 10 min , eliciting fast translocation of CF-p85 . In the course of 10 min , membrane-targeted CF-p85 stimulated the production of , as was indicated by a decline of cytosolic fluorescence from Y-PH . The concentration of iRap was then doubled with a concomitant addition of 10 nM LY29 . PI3K activity was largely insensitive to such a low dose of inhibitor , thus establishing a lower bound for . Ten minutes later , LY29 was added , triggering rapid dissociation of Y-PH ( Figure 3D , Video S3 ) . Finally , we formulated the parameter estimation problem on the data obtained from the traditional and both optimized experiments . To obtain a satisfying collective fit for all three experiments , we considered background fluorescence from CF-p85 and Y-PH recruited to the plasma membrane and to internal membranes . For simplicity , a fraction of bound CF-p85 and Y-PH was included into the observation functions: ( 10 ) ( 11 ) Here the index denotes the experiment number , the time point in experiment . With this assumption , parameter estimation established good agreement between the model and the measurements from all three experiments combined ( Figure 3E ) . Note that for , the revised observation functions revert to equations ( 4 ) and ( 5 ) . To assess if optimal experimental design yielded data that reasonably defined the parameters of the proposed signaling model , we approximated the uncertainties of parameter estimates after each of the three experiments ( Table 1 ) . Overall , the mean of variances dropped from 193 to 2 . 9 ( Figure 4A ) , much faster than one could expect from triplicate measurements . Most of this decline reflects improvements of the parameter estimates with the highest initial uncertainty , , , and . Moreover , the parameters controlling downstream signaling , and could be inferred with much improved accuracy . From experiment to experiment , however , the parameter estimates varied strongly , in agreement with most relative uncertainties remaining above 100% . Nevertheless , we identified a protocol that renders all endogenous parameters accessible in few single-cell experiments . This protocol of three designs could be refined further by taking advantage of the improved previous knowledge . On the other hand , the uncertainty of the exogenous parameter increased from the initial experiment to the final model because it describes the association of CF-p85 to Lyn-FRB . Lyn-FRB is not fluorescently tagged and is virtually unidentifiable with free ( Table S1 ) . Consistently , is the least defined system parameter in the final model ( Figure 4B ) . Nevertheless , this uncertainty does not propagate to endogenous parameters because the localization of CF-p85 is observable . This is a favorable property because it suggests that the exact transfection efficiency for Lyn-FRB is not critical as long as it expresses much better than CF-p85 . In their study , Gutenkunst et al . used published intuitively designed experimental protocols to characterize parameter identifiability in different models [6] . Independently of the size of these models , they found that a substantial fraction of eigenvectors of the uncertainty region ran along multiple axes in parameter space . The eigenvectors of computed after the intuitive experiment from Figure 1 suggested a similar co-dependence also for this model ( Figure 4C ) . However , the data from a single numerically optimized experiment largely eliminated the correlation between estimates . The corresponding eigenvalue spectrum of confirms that also the extent of the uncertainty region was consistently reduced ( Figure 4D ) .
Our study focused on how an optimal control method can be employed to interrogate a cell signaling system by live-cell microscopy . The goal of this approach was to reveal biologically relevant parameters of a model that describes the dynamic behavior of the lipid second messenger , the recruitment of a cytosolic pleckstrin homology domain to the plasma membrane , and the inhibition and activation of PI3K by means of pharmacological and synthetic biology techniques . We found that the data obtained from the established experimental protocol resulted in large parameter variances and , in effect , inconclusive estimates . However , the model was useful to inform an optimal control method that proposed modifications of the existing protocol to markedly reduce parameter uncertainty . While we found that the appropriate parameterization of the design space is important for the attainability of optimized concentration-time profiles , experimental imperfection led only to a small reduction in information gain ( Figure 2C ) . On the other hand , the uncertainty of parameter estimates appeared to be more important ( Figure 2C ) , suggesting that methods for robust optimal experimental design will be of much practical value [15] . Importantly , not only robustness to uncertainty of parameter estimates , but also to uncertainty in model structure and experimentation will need to be addressed more formally by future optimization methods , e . g . along the developments described in [21] . Considering the specific model structure of the translocation assay , the first optimized experiment at a first look did not seem intuitive . However , it can be rationalized as an intelligent maneuver to separate the translocation kinetics of CF-p85 from the production of . To this end , a high concentration of iRap was added in the presence of the inhibitor LY29 , and then LY29 was washed out after CF-p85 was brought into position . By employing a submaximal dose of LY29 , the identifiability of the inhibitor constant could be improved simultaneously . This suggests that the first optimized experiment clearly exploited features of the experimental system that were independent of details in model structure and parameter values . Moreover , it is interesting to note that for the last segment of the experiment , it was not necessary to wash out LY29 completely . Our posterior analysis that accounted for experimental imperfection suggests that significant residual amounts of LY29 did not significantly affect the realized design performance ( Figure 2C ) . The second optimized experiment employed novel dynamic constraint formulations to optimize a limited number of buffer changes and time points . These formulations could be of general value for reflecting actual experimental constraints . With the constrained design spaces used for optimizing the second experiment , we were limited to three additions of drug . While this design was clearly informative , its performance with respect to parameter identifiability was sensitive to uncertainty of . This exemplifies a drawback of minimizing the determinant of the covariance matrix . Since the determinant criterion aims at the volume of the confidence ellipsoid ( Figure 1G ) , an experimental design that defines a single parameter value such as very tightly , would correspond to large improvements in the objective function . In other words , minimizing in some cases focuses experimental effort at eliminating a full dimension of the uncertainty region ( Figure 3B ) . Thus , despite the significant improvement predicted for , it should be noted that focusing effort on a single parameter can be sensitive to overall uncertainty . The practical non-identifiability of many parameters is an important challenge for the development and use of signaling models . Intuitively , this means that there are too many parameters for the number and the type of experiments that can be practically performed . Moreover , the structure of such models was suggested to reflect natural robustness to changes in many parameter values [3] , [6] . For example , signaling pathways often translate graded stimuli into all-or-none responses [22] , such that downstream observations become insensitive to upstream kinetic parameters . Strong covariance between parameters of reaction cascades , fast reversible reactions , and binding events makes it difficult to constrain single parameter values [3] , [6] , [10] , [23] . Therefore , pharmacological and synthetic tools that drive a biological system internally will become important for eliciting informative dynamics . The pair of Lyn-FRB and CF-p85 is one example of such tools , because endogenous PI3K is activated directly and downstream of complex receptor relays . Likewise , the small molecule inhibitor of PI3K that we employed exemplifies the value of pharmacological tools in the context of such perturbation studies . In essence , our work suggests that chemical tools may be useful for dividing complex pathways into tractable signaling domains in order to conquer non-identifiabilities in situ . However , it was not clear intuitively how to combine these perturbations in order to generate data from which parameters could be inferred . Our work exemplifies that in such a situation , model-based experimental design can dramatically reduce the experimental effort to establish a useful protocol . To summarize , several issues should be considered when a signaling system is explored using a dynamic model . First , it is often difficult to develop comprehensive cell signaling models that are simple enough to be useful but that do retain enough detail to represent the essential features of a system quantitatively . Finding a trade-off between overparameterized model formulations and formulations that neglect important interactions is made difficult by our limited knowledge of the mammalian signaling network and by often large uncertainties in measurements . Second , our study suggests that pharmacological perturbations are powerful tools to enhance the quality of parameter estimates . We show that when such pharmacological agents are not available , synthetic biology approaches can be employed . In all those cases , numerical methods for optimal experimental design can propose protocols that take advantage of these experimental techniques . Third , the experimental design space should be constrained during optimization to result in feasible protocols . This ensures that the experiment can be realized according to the proposed plan . At the same time , our results suggest that some experimental protocols can be robust to imperfection ( Figure 2C ) . Our approach to reveal parameter values of biological models complements other approaches of optimal experimental design in systems biology , such as designing experiments for model discrimination [24] or for minimizing prediction uncertainty [25] . Ideally , the parameter values of a biochemical model correspond to biologically meaningful properties . In our example , reflects the degradation rate of , and describe the affinity of a PH domain , and reflects the inhibitor constant of LY29 . Because any model can only be an approximation of the true process in nature , the parameter values we determine are model-specific . Nevertheless , if such parameters could be inferred with sufficient accuracy , they may reflect molecular differences that underlie biochemical mechanisms . We could envision that in a pharmacological context , drug candidates for targeting PI3K could be compared in situ by inferring from few single-cell experiments . Likewise , in a biochemical context , differences in PH domain affinities could be revealed by estimating . Finally , reflects the activity of phosphatases like phosphatase and tensin homolog ( PTEN ) which is lost in many cancers [26] . Importantly , inferring in cell lines established from cancer patients could reveal changes in PTEN activity that originate from causes beyond gene expression . To achieve this , instead of minimizing a criterion on the full covariance matrix ( e . g . the trace or determinant ) , experimental effort could focus only on the subset of diagonal elements of interest . Applications of such an approach may include a comprehensive estimation of signaling parameters from cancer samples , possibly leading to individualized therapy . Also , by treating drug action as a systems property , this approach may also be useful to evaluate possible off-target effects of drugs . We showed here for the first time that optimal experimental design for parameter estimation yields rich data for rapid model development in systems biology . This approach aims to restrict the range of possible parameter values , and is therefore a suitable way to challenge the validity of a model by quantitative experimentation . In particular , the results exemplify the use of this optimal control method to interrogate individual cells for pharmacological and biological parameters that underlie a growth-related second messenger signaling system .
Cell culture , constructs , and drugs . 4-well LabTek chambered coverslips ( nunc , Rochester , NY ) were coated with 0 . 1 mg/ml poly-L-lysine ( Sigma-Aldrich , St . Louis , MO ) in hydroborate buffer . NIH 3T3 cells were seeded per well , and grown close to confluency . Per well ( ) , of each construct ( Lyn-FRB , CF-p85 , Y-PH ) were transfected using Lipofectapine 2000 in total transfection mix on DMEM supplemented with 10% FBS for 6 hrs ( all reagents from Invitrogen , Carlsbad , CA ) . 4 to 12 hours prior to imaging , cells were serum-starved in 0 . 1% BSA ( Sigma-Aldrich ) in DMEM . CF-p85 is described in [17] , and Lyn-FRB ( as ) in [27] . Stock solutions of LY294002 ( Tocris Bioscience , Ellisville , MO ) and iRap were prepared in DMSO . For the organic synthesis of iRap from rapamycin and 3-methylindole , see [27] . Numerical methods . The numerical methods used are incorporated in the software package VPLAN [7] , [8] . They include a variable–step variable–order BDF method for initial value problems in differential equations , which is used for the simulation of the translocation model . This solver is based on [28] , [29] and computes first and second derivatives of the solution of the initial value problem with respect to initial values and parameters , utilizing a sophisticated combination of internal numerical differentiation [30] and automatic differentiation [31] . Parameters in the translocation model were estimated using a multiple shooting method with a generalized Gauss-Newton algorithm that is implemented in PARFIT ( see [30] , [32] , reviewed in [33] ) . Multiple shooting methods consider the parameter estimation problem as a least squares problem that is constrained by a differential equation model . The differential model constraint is discretized like a boundary value problem . This enhances stability of the method and allows for initialization of the discretized output trajectories close to measurement data , which results in fast convergence to statistically stable minima . In contrast , single shooting methods only solve the initial value problem from the beginning of an experiment , and integrate with possibly poor initial guesses for parameter values along the full time axis . If at all feasible , a solution is found only with much more effort than by multiple shooting . Recently , Balsa-Canto et al . proposed a hybrid approach that switches from a global search to a multiple shooting Newton-type method [34] . Optimal experimental designs are determined by minimizing a function of the covariance matrix of the parameter estimation problem ( e . g . the sum of the diagonal elements ) with respect to the concentration-time profile of drugs and possibly the sampling scheme . The method is applicable to constrained parameter estimation problems in systems of differential algebraic equations ( DAE ) . In the general case of constrained problems , the covariance matrix can be represented in terms of ( see eq . 7 ) and the sensitivities of the constraints: ( 12 ) To exclude washout procedures in the second optimized experiment , we defined a differential state variable as a “short-term memory” for each drug , and defined diffusion-like equilibration with the current concentration : ( 13 ) By introducing the dynamic inequality constraint which needs to be evaluated only once after each buffer change , we can approximate a constraint of the form for . The constraint is stricter the larger one choses . In our specific case , was sufficient . These constrained optimal control problems were solved by a direct approach for optimal control , the core of which is solving a structured nonlinear program by SQP-type methods [20] . All calculations were performed on a Dell Optiplex GX620 workstation . Live cell microscopy . Cells were imaged at using a UApo/340 40x/1 . 35 oil objective ( Olympus , Center Valley , PA ) mounted on an Olympus IX70 stage customly equipped with an UltraVIEW spinning disk confocal scanner ( PerkinElmer , Waltham , MA ) . CF-p85 and Y-PH were excited with a He-Cd laser ( IK series from Kimmon , Centennial , CO ) or an Ar laser ( model 60B from ALC , Salt Lake City , UT ) , using 442/10 or 515/10 excitation filters , and 480/40 or 530LP emmission filters , respectively . The variable power supply of the Ar laser was taped at the beginning of this study , and the power of each laser was measured on stage through an Olympus UPlanFl 10x/0 . 30 objective before every experiment using an optical power meter ( model 835 from Newport , Irvine , CA ) to safeguard quantitative analysis . Images were captured from a Hamamatsu C4742 Orca ( Hamamatsu Photonics , Shizuoka , JP ) at 12 bit/pixel and 300 msec exposure time . Shutters were emulated using a closed position of the excitation filter wheel controlled by a Sutter Lambda 10–2 ( Sutter Instrument Company , Novato , CA ) . Extracellular buffer contained 5 mM KCl , 125 mM NaCl , 1 . 5 mM , 1 . 5 mM , 10 mM D-glucose ( Sigma-Aldrich ) , and 20 mM HEPES ( Invotrogen ) . Image analysis . For each experiment , an empty region of the coverslip was captured in both fluorescence channels , and used for background subtraction . Image stacks were aligned with subpixel resolution to compensate for planar drift where needed . Fluorescence was quantified using the median of a region of 2525 pixels , that was chosen to minimize interference from passing organelles , and that is marked by green squares in Figures 1B , 2D , and 3D . The standard deviation of measurements was obtained from the standard deviation of intensities in the region of 2525 pixels for each frame . For optimizing designs , we assumed a constant error model , however with larger errors for CF-p85 ( 0 . 2 AU ) than for Y-PH ( 0 . 06 AU ) . | Differential equation models of signaling processes are useful to gain a molecular and quantitative understanding of cellular information flow . Although these models are typically based on simple kinetic rules , they can often qualitatively describe the behavior of biological systems . However , in the quest to transform biomedical research into an engineering discipline , biologists face the challenge of estimating important parameters of such models from laboratory data . Measurement noise as well as the robust architecture of biological circuits are causes for large uncertainty of parameter estimates . This makes it difficult to plan informative experiments . Here , we used a computational method to predict and minimize the uncertainty of parameter estimates we would obtain from prospective experiments given a cancer-relevant signaling model . This was achieved by optimizing the concentrations and time points for adding drugs in a live-cell microscopy experiment . Our experimental results demonstrated that the advice given by this algorithm resulted in many-fold more informative data than we would obtain by repeating an intuitively planned experiment . Our study shows that significant experimental effort and time can be saved by adopting an optimal experimental design strategy for inferring relevant parameters from biomedical experiments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"oncology",
"computational",
"biology/synthetic",
"biology",
"cell",
"biology/cell",
"signaling",
"mathematics",
"computer",
"science/systems",
"and",
"control",
"theory",
"computer",
"science/numerical",
"analysis",
"and",
"theoretical",
"computing",
"diabetes",
"and",
"endocrinology",
"computational",
"biology/signaling",
"networks",
"computational",
"biology/systems",
"biology"
]
| 2009 | Optimal Experimental Design for Parameter Estimation of a Cell Signaling Model |
The steady states of cells affect their response to perturbation . Indeed , diagnostic markers for predicting the response to therapeutic perturbation are often based on steady state measurements . In spite of this , no method exists to systematically characterize the relationship between steady state and response . Mathematical models are established tools for studying cellular responses , but characterizing their relationship to the steady state requires that it have a parametric , or analytical , expression . For some models , this expression can be derived by the King-Altman method . However , King-Altman requires that no substrate act as an enzyme , and is therefore not applicable to most models of signal transduction . For this reason we developed py-substitution , a simple but general method for deriving analytical expressions for the steady states of mass action models . Where the King-Altman method is applicable , we show that py-substitution yields an equivalent expression , and at comparable efficiency . We use py-substitution to study the relationship between steady state and sensitivity to the anti-cancer drug candidate , dulanermin ( recombinant human TRAIL ) . First , we use py-substitution to derive an analytical expression for the steady state of a published model of TRAIL-induced apoptosis . Next , we show that the amount of TRAIL required for cell death is sensitive to the steady state concentrations of procaspase 8 and its negative regulator , Bar , but not the other procaspase molecules . This suggests that activation of caspase 8 is a critical point in the death decision process . Finally , we show that changes in the threshold at which TRAIL results in cell death is not always equivalent to changes in the time of death , as is commonly assumed . Our work demonstrates that an analytical expression is a powerful tool for identifying steady state determinants of the cellular response to perturbation . All code is available at http://signalingsystems . ucsd . edu/models-and-code/ or as supplementary material accompanying this paper .
Transient activation of signaling molecules is a hallmark of the cellular response to perturbation . Far from acting as a simple relay , however , the dynamics of signaling molecules can encode information about the instigating stimulus [1]–[3] . Interestingly , these dynamics are affected by the steady state prior to perturbation [4] , [5] . Non-genetic variation in the proteome , for example , is sufficient to explain variability in the sensitivity of HeLa cells to the pro-apoptotic ligand TRAIL [6] . Like other TNF superfamily members , TRAIL is a promising anti-cancer therapeutic [7] . Recombinant human TRAIL , or dulanermin , as well as antibodies raised against the TRAIL receptors DR4 and DR5 , are currently in clinical trials [8] . To improve the efficacy of these and other drugs , understanding how sensitivity is affected by the cellular resting state is of great importance [9] . Mathematical models are powerful tools for characterizing the behavior of signaling systems in response to perturbation [10]–[13] . Assuming conservation of mass , these models equate the change in concentration of a molecular species with the sum of reaction velocities that produce the species , minus the sum of those that consume it . The reactions themselves are often modeled by the Law of Mass Action . This law assumes that the velocity of a reaction is proportional to the product of the concentrations of its reactants . Since many signaling reactions are bimolecular , the resulting mass balance equations are non-linear in the concentrations . A system is at steady state if no species is consumed faster than it is produced , nor produced faster than it is consumed . By this formalism , the steady state of a signaling system is equivalent to the root of a non-linear system of equations . Because of this , no universal method has been developed to identify the steady states of mass action models , despite their importance to basic and clinical research . As a result , even with the help of mathematical models , investigating the relationship between steady state and stimulus-responsiveness remains cumbersome . Of course with any model , simulating the response to perturbation often requires the system to be at steady state prior to perturbation . To achieve this , one of several techniques is currently used . The most common technique is to assume a “trivial” steady state where every reaction velocity is zero [2] , [14] . While straightforward , this approach may not reflect biological reality , where tonic signaling is common [15] , [16] and can strongly influence the response to perturbation [17]–[19] . A second technique is to approach the steady state asymptotically via numerical integration of the mass balance equations [1] , [13] , [20] . While this approach can yield non-trivial steady states , the number of integration steps required to reach the steady state may dominate the number of steps required to simulate the perturbation . Also , identifying the parameter values that result in a desired steady state is an inverse problem that requires non-linear optimization . For these reasons , numerical derivation of the steady state is impractical when characterizing its effect on the response to perturbation , and an analytical expression is required instead . The best-known method for deriving analytical expressions for the steady states of mass action models was developed by King and Altman in 1956 [21] . This method assumes that all molecular species can be divided into enzymes and substrates , that no enzyme is itself a substrate , and that all substrates remain constant over the time-scale of steady state formation [22] . A number of improvements have been made to the King-Altman method over the years [23]–[25] . Many of these are now implemented in the Matlab application , KAPattern [26] . The King-Altman methodology was also recently formalized using concepts from algebraic geometry [22] , [27] , and extended to layered signaling cascades [28] and post-translational modification networks [29] . Despite these improvements , however , these methods do not extend to mass action models with arbitrary reaction structure , as is common in contemporary models of signaling systems . Furthermore , only the King-Altman method has been reduced to practice . For these reasons we developed py-substitution , a simple , algebraic method for deriving steady state expressions for mass action models with arbitrary structure . Our method can be explained using concepts from linear algebra , and full code has been provided for all examples in this manuscript , implemented in either Matlab or Maple . A particular benefit of py-substitution is that it affords considerable flexibility when selecting independent quantities for the steady state expression . Often , this permits explicit derivation of kinetic rate constants from steady state concentration measurements . More generally , it allows independent quantities to be chosen that maximize incorporation of known or measured parameter values . This not only simplifies model fitting , but typically reduces the total number of parameters required as well . We compare py-substitution to the King-Altman method and show that , where King-Altman is applicable , the two methods yield equivalent results . Computationally , however , we find that our method is more efficient , and , because py-substitution does not require a particular reaction structure , more general than King-Altman . Finally , we use py-substitution to derive a steady state expression for a recent model of apoptosis induced by the death-receptor ligand TRAIL [14] . We find that incorporation of a non-trivial steady state changes the qualitative behavior of the model . Specifically , tonic signaling desensitizes the system to low doses of TRAIL , while high doses of TRAIL still result in the “snap-action” signaling dynamics indicative of cell death . We then systematically alter the steady state and show that changes in steady state affect the threshold at which TRAIL results in death . We find that the threshold is highly sensitive to the steady state abundances of procaspase 8 and its negative regulator , Bar , but not the other procaspase molecules . This suggests that the activation of caspase 8 is a critical point in the cell death decision . Finally , without recourse to a model that is tolerant to low doses of TRAIL , a common practice is to approximate the sensitivity to TRAIL by the time at which death occurs . Using our tonic signaling model , we show that these two metrics are not universally equivalent . Caution should therefore be taken when equating the dynamics of cell death with the probability that death occurs .
Let be the set of non-negative natural numbers and be the set of non-negative real numbers . Let be a set of species and be a set of reactions . Each reaction follows the normal definition , where is the stoichiometric coefficient of the reactant and is the stoichiometric coefficient of the product [30] . We define to be the concentration of species and to be the velocity at which converts reactants into products . By the Law of Mass Action , ( 1 ) The quantity is often , but not necessarily , equal to . The coefficient is called the rate constant . Assuming conservation of mass , the concentration changes according to ( 2 ) where is the first derivative of with respect to time . Any collection where the concentration of obeys Equation 2 and the velocity of obeys Equation 1 is called a mass action model . In what follows , we assume , , and are indices over the interval and is an index over . When are such that all ( 3 ) the model is said to be at steady state . If all we call the steady state trivial . In this manuscript we are concerned with symbolic , non-trivial solutions to Equation 3 . A solution is symbolic if all and are left as uninterpreted variables , rather than being assigned numerical values . For a complete list of symbols and their meanings , see Table S1 . Let and be the vectors with elements = , and . Throughout this manuscript , we use to denote the element of vector and to denote the element at row , column of matrix . Let be the stoichiometric matrix , i . e . , the matrix whose elements are . Using this notation , Equation 2 becomes ( 4 ) and the steady state equation becomes ( 5 ) By convention we use the overline to denote vectors that satisfy steady state . Equation 5 often takes this form in flux balance analysis [31]–[34] . Here is a real-valued vector and is calculated numerically . However , prior work has shown that Equation 5 can also be used to calculate a vector of rate constants from a vector of steady state concentrations [35] . Let be the vector with elements . Let be the diagonal matrix with elements . The vector can then be expressed as ( 6 ) Substituting Equation 6 into Equation 5 and solving for yields the -cone [35] — equivalently , the left null space of the matrix product . Given a basis for this null space and a vector of steady state concentrations , a vector of rate constants can be calculated that satisfies Equation 5 . While this approach is useful for deriving kinetic parameters from metabolomic measurements , it is less well suited to signaling systems where transient and low-abundance species confound accurate measurement of the concentrations . If the velocity of every is homogeneous of degree in , then an analogous approach allows to be expressed in terms of . We call models that satisfy this condition linear models . An alternative , stoichiometric definition for a linear model is given by the following , ( 7 ) Equation 7 requires that every reaction defines a transition from exactly one time-varying species to another . Let be the matrix with elements . If is a vector of linear reaction velocities , it can likewise be expressed as ( 8 ) Substituting Equation 8 into Equation 5 results in the matrix product , also called the Jacobian matrix [36] . Given a basis for the null space of the Jacobian , a vector of steady state concentrations can be calculated from a vector rate constants . For linear models , an alternative , graphical method for deriving expressions for the steady state species concentrations was introduced by King and Altman in 1956 [21] . Notice that Equation 7 permits a two-dimensional indexing of the rate constants , ( 9 ) We call a transition rate constant since the product defines the rate of transition from species to . Substituting Equation 9 into Equations 1 and 2 gives ( 10 ) By defining the matrix with elements ( 11 ) the steady state equation becomes ( 12 ) Note that is simply the Jacobian matrix for a linear model , . The general solution to Equation 12 was found in [21] to be the vector with elements ( 13 ) where is the minor of , formed by removing its row and column and computing its determinant . For sufficiently small systems , Equation 13 can be solved directly using modern mathematical computing software [37] . Prior to the advent of modern computers , King and Altman realized that the minors can also be derived by graph theoretic means . Note that for a linear model , and imply a directed graph , ( 14 ) where each defines a vertex and each defines an edge between vertices and ( provided and are such that ) . The King-Altman method enumerates for each species the set of simple connected subgraphswhere vertex has out-degree and all other vertices have out-degree [23] , [24] . These are the directed spanning trees of , with all edges directed towards root . A subgraph is called a King-Altman pattern . The minor can then be expressed as ( 15 ) where is the transition rate constant between species and . For a more thorough derivation of the King-Altman method , see [38] . Of course , many biochemical reactions are bimolecular . By Equation 1 , the velocity of a bimolecular reaction is degree in . To preserve linearity , one can assume the concentration of one reactant is so high as to be effectively constant . This concentration is incorporated into the kinetic rate constant , and the techniques described above can still be used to solve Equation 3 . If this assumption fails , then Equation 2 describes a polynomial in with coefficients in . In this case the solutions to Equation 3 form an algebraic variety . Deriving an expression for the steady state of a non-linear model thus requires finding a parameterization of the variety [39] . One way to achieve this is to calculate a Gröbner basis for the ideal generated by and eliminate variables [40] , [41] . Alternatively , if the model displays certain structural properties , variables can be eliminated by identifying conservation relationships . The best-known example of this is when defines a cascade of post-translational modifications . In this case , enzyme-substrate intermediates can be eliminated and the variety can be parameterized by rational functions of the free enzyme concentrations with coefficients in [22] , [28] . Although these methods do not require linearity , calculating a Gröbner basis can be computationally intractable , while identifying conservation relationships can be difficult for models of arbitrary reaction structure . Py-substitution allows mass action models — a particular class of non-linear model — to be solved using simple linear algebra . We make use of the following observations: ( a ) is always homogeneous of degree in , and ( b ) is often no greater than degree in . If a subset of elements in can be found on which every has only linear dependence , then Equation 5 can be solved using linear methods . To begin , we define sets of symbolic variables and such that and . We then relabel , or map , every element in to a unique element in so that every is linear in . By Equations 1 and 2 this requires that all are linear in . Variables that we want to remain independent , as well as variables on which has non-linear dependence , should be mapped to . As we shall see , there is considerable flexibility in choosing this map . Let and be partitioned into disjoint ( but possibly empty ) subsets and . We define to be a bijective map ( with a restriction given below ) and extend it homomorphically over . Our linearity restriction is to consider maps of this form such that ( 16 ) for some . For , the exponent is . For , the exponent . In words , defines a change of variables such that is homogeneous of degree in . By Equation 2 , becomes a homogeneous polynomial of degree in with coefficients in . We can now write ( 17 ) where is the Jacobian matrix with elements . Here and elsewhere we use the notation to mean that is the vector formed by applying the function element-wise to . Note that the trivial partition and recovers the -cone procedure described above . For the remainder of this section , we treat as an index over . Substituting Equation 17 into Equation 5 gives ( 18 ) where is called the coefficient matrix . The solution to Equation 18 is precisely the null space of . Let be a matrix whose columns form a basis for this null space . Let be the number of columns in . By the rank-nullity theorem , we have ( 19 ) where is the number of columns in . Furthermore , because is linear in and exists , the matrix must be full rank . By the properties of the rank , we can write ( 20 ) Together , Equations 19 and 20 give ( 21 ) thus calling for the constraint . This , in conjunction with Equation 16 , are the only constraints on . If we now let be some linear combination of the basis vectors , ( 22 ) then satisfies Equation 18 and steady state is achieved . In general , Equation 22 is underdetermined . Equation 22 therefore implies a partition of into independent variables ( denoted ) and dependent variables ( denoted ) . We will now describe this partition by a second mapping function , . Recall that a basis for the null space of can be constructed from , the reduced row echelon form of . Let be the column of . If contains a pivot position , then is a dependent variable . If does not contain a pivot , then is free , or independent . Let ( 23 ) Let be the cardinality of . Enumerate these variables as , with . For every not containing a pivot , there is a basis vector ( related by ) whose element equals and whose elements in positions are . By Equation 22 , this gives an independent parameter , . Equation 22 thus defines a function . Let be the set of independent parameters . If column does contain a pivot , then depends on variables in , giving where is the specific function resulting from the row operations used to reduce to . Equation 22 can now be described in its entirety by the mapping function , ( 24 ) The notation indicates that for every . Note that we define as the set of all linear combinations , where and are distinct elements of . is the set of all polynomials in variables with rational numbers as coefficients . is the field of fractions of : any can be expressed as , where , . As with , there is some flexibility in choosing how is partitioned into free variables , , and dependent variables , . A different indexing of the variables in simultaneously permutes the vector and the columns of . This leads to different reduced row echelon forms , with different partitions into free and dependent variables . The null space basis obtained by reducing to greedily classifies low-numbered columns as dependent columns when possible , or free columns when not possible . Quantities in for which good numerical estimates exist should therefore be assigned to higher indices . These quantities are favored , but not guaranteed , to be mapped to independent parameters . Quantities for which good numerical estimates do not exist should be assigned to low indices in . Finer control over the partition of into dependent and independent parameters is possible by working directly with or . Let be the set of elements in that we want mapped to . Let be the square matrix formed by rows of . To map to requires that we find a vector such thatwhere is the vector with elements . Solving for gives ( 25 ) Thus , for a given map , not all partitions of into and are possible , but only those for which . An example of this can be seen in the file “fum2 . m” in Supporting Protocol S1 , discussed below . Next let , and . Let and be defined analogously . The composition captures the entire process of linearizing with the function , solving the linear system , and taking an arbitrary combination of solution space basis vectors: Applying to the sets and results in a parametric description of the steady state that is typically the most useful: every element in or is mapped to an element in or , or a function in . Assigning numerical values to elements in and results in elements in taking values that satisfy the steady state equation . In some cases we may wish to reverse the substitution so that functions of variables are mapped back to functions of . To do so , let and . Let be the inverse of restricted to the independent parameters , . The composition of and now defines a map from the set of independent parameters to their counterparts in and , If we extend to homomorphically , we can compose with , The function then defines a map for whichwhere steady state velocities in are in terms of elements in and . A visual overview of the py-substitution method is given in Figure 1 .
An important goal in developing py-substitution was that it be generally applicable to any model whose reaction rates obey mass action kinetics . This requires that the independent quantities be chosen freely among the species concentrations and reaction rate constants , and that non-linear rate equations do not confound the derivation of a steady state expression . To demonstrate these capabilities we consider an open-system analog of the classical Michaelis-Menten model of enzyme kinetics ( OMM , see also Figure 2 ) . Substrate synthesis and product degradation allow this system to achieve a non-trivial steady state , which we derive here using four different substitution strategies . The set of reactions for this model is given by The symbol Ø represents a source or sink for mass and is not modeled by a time-varying species . From the set we derive the stoichiometric matrix and reaction velocity vector , By Equation 4 this results in the following system of equations , for which we now derive functions such that . Some chemical reaction systems are linear in the species concentration vector , or can be rendered linear by assuming that the concentrations of certain species don't change over time . The classical model for malate synthesis is an example of the latter [42] . Here , the enzyme fumarase binds reversibly to fumarate and hydrogen in either order , followed by reversible binding of hydroxyl and reversible formation of malate ( Figure 3 ) . The reactions for this model are By Equation 1 , the corresponding reaction velocities are ( 32 ) The stoichiometric matrix is Notice that the submatrix formed by the first five rows of satisfies the definition for a linear model given in Equation 7 . Call this submatrix and let be the vector of enzyme concentrations . If we assume that the substrate concentrations are time-invariant , the steady state equation for this model becomes ( 33 ) Because satisfies Equation 7 , we may define the following transition rate constants ( 34 ) Substituting Equations 34 into 32 results in a velocity vector that is linear in . Let as before , where and . This gives ( 35 ) If we now define a matrix ( 36 ) Equation 33 becomes ( 37 ) where the elements of are given in Equation 11 . The solution to Equation 37 is given by Equation 13 , which we saw may be evaluated using the King-Altman method . Alternatively , we may solve Equation 33 directly using py-substitution . Given that py-substitution applies to a more general class of mass action models then King-Altman , we wondered whether this flexibility came at the cost of computational efficiency . Here we show that , for models that can be treated using the King-Altman method , py-substitution yields an equivalent result , and at no loss of efficiency . Finally , we sought to use py-substitution to characterize the relationship between steady state and the response to the cancer drug , dulanermin . Dulanermin is a recombinant human form of the endogenous ligand TRAIL , whose mechanism for triggering cell death is modeled in version 1 . 0 of the extrinsic apoptosis reaction model , or EARM [14] . This model considers the biochemical events following engagement of the death receptors 4 and 5 ( DR4/5 ) , including receptor-induced cleavage of initiator caspases , positive-feedback by effector caspases , and feed-forward amplification by the mitochondrial pathway following outer membrane permeabilization , or MOMP ( Figure 5 ) . The EARM model was trained on data derived from HeLa cells co-treated with cyclohexamide , an inhibitor of protein synthesis that results in hypersensitivity to TRAIL [43] . Accordingly , any amount of ligand in the EARM model results in cell death . The abundance of ligand still affects the time of death , defined for example by the time at which half of the caspase 3 target protein PARP has been cleaved ( Figure 6A , left ) [44] . Note in this section we refer to the abundance of a species rather than its concentration , as these are the units chosen by the original authors . In the absence of cyclohexamide , however , HeLa cells do not all die following exposure to TRAIL . Rather , a fraction of cells persist , and this resistance is a function of the proteomic state prior to stimulation [6] . To capture this phenomenon , we extended the EARM model so that proteins continued to be synthesized and degraded following exposure to TRAIL . Specifically , we introduced 43 new synthesis and degradation fluxes as well as 2 protein inactivation reactions ( see “xearm . mpl” in Protocol S1 ) . These reactions were chosen so that every species is subject to at least one efflux . We refer to our extended model as xEARM . Because xEARM satisfies our definition of a mass action model , we use py-substitution to identify an analytical expression for its steady state . To derive this expression , a mapping function was chosen so that every non-zero parameter in EARM was mapped to an independent parameter in . As a result , we were able to preserve the snap-action dynamics of MOMP that is central to the original model ( Figure 6A , right ) . Honoring the published parameters required that we introduce two pseudospecies , one for each the di- and tetrameric forms of Bax ( variables and , respectively ) , ( 43 ) ( 44 ) The coefficient matrix and null space basis matrix were calculated as before , with the latter calculation requiring less than a minute on our benchmark PC . The null space of has 17-dimensions , resulting in a matrix of basis vectors of the form Basis vectors to preserve the steady state ratios of paired synthesis and degradation reactions . Vector , for example , ensures that a change in results in a change in , where is the abundance of Cytochrome C in the mitochondria and and are its rates of synthesis and degradation , respectively . The vector scales the steady state abundances of mitochondrial Bax and Bcl2 complexes with respect to changes in the rate of Bcl2 synthesis . Vectors and are algebraically intractable and thus defy simple biochemical interpretation . Two of these vectors , and , are constrained by the pseudospecies and . To resolve these constraints , note that Equations 43 and 44 require that ( 45 ) ( 46 ) By our mapping function ( see “xearm . mpl . trace . pdf” in Protocol S1 , pp . 120–121 ) , Equations 45 and 46 become ( 47 ) ( 48 ) where . Solving Equation 48 for gives ( 49 ) where . Substituting Equation 49 into Equation 47 and solving for gives ( 50 ) where ( see “xearm . mpl . trace . pdf” in Protocol S1 , pp . 121–126 ) . Obviously , Equation 50 identifies an explicit bistability in the xEARM model . Basis vector coefficient — and by Equation 49 , — can take either of two values for any numerical realization of the model . By examination of , we find that these two coefficients affect all modified and compound species , as well as synthesis rates for proteins within and upstream of the mitochondria . Using the parameter values supplied in [14] , however , we find that one of the solutions to Equation 50 is negative . The corresponding steady state is therefore infeasible and the solution was discarded . In addition to parameters in [14] , a full numerical realization of the xEARM model requires values for parameters and . All but three of these elements represent first-order degradation rate constants , to which we assigned values equivalent to a half-life of one hour . This value was based on global quantifications of protein turnover in mammalian cells , which revealed that signaling proteins tend to be short-lived [45] . Two of the elements , and , represent first-order inactivation fluxes , which we assumed to be ten times faster than protein degradation . The final element is the steady state abundance of the mitochondrial Bax2∶Bcl2 complex , which we set to 20 molecules . Six of the elements were then modified from their initial values to better match the dynamics of caspase activation and PARP cleavage , as reported in [14] . The complete table of parameter values required to initialize and numerically integrate the xEARM model is given in Table S2 . For comparison , Table S3 lists the steady state abundances of species in the original and extended EARM models , sorted in order of decreasing difference . As expected , every species in EARM with a non-zero abundance has precisely the same abundance in xEARM , since these are independent parameters in the steady state solution . Among species with zero abundance in EARM , the mitochondrial Bax∶Bcl2 complex exhibits the greatest disparity , with the steady state abundance in xEARM being in the low thousands of molecules . Ubiquitinated , cleaved caspase 3 and cleaved PARP are also in the low hundreds of molecules , but this represents only a small fraction of their total cellular abundance . A full 25 species with zero abundance in the EARM model have an abundance of less than 1 molecule in xEARM . This indicates that , even though the steady state reaction velocities are markedly different between EARM and xEARM , by using py-substitution we were able to engineer a steady state where the species abundances are appreciably similar between the two models . Next we asked whether the xEARM model remained viable in the presence of low doses of TRAIL , but still exhibited MOMP when stimulated with high doses of TRAIL . To do so we created a numerical realization of the model using the parameters from Table S2 , then perturbed the model from its steady state using a step increase in the abundance of TRAIL ( variable ) . The magnitude of the step ranged from to -fold and was followed by numerical integration of the mass balance equations out to 48 hours . As shown in Figure 6A , MOMP is only observed in xEARM when TRAIL is increased by -fold or more . We label this minimum dose of TRAIL required for MOMP . Increments less than result in a small and transient change in cleaved PARP abundance , followed by a return to the pre-stimulated steady state . By comparison , any magnitude dose of TRAIL causes MOMP in the original EARM model . This ability of xEARM to distinguish between low and high doses of TRAIL , in conjunction with an analytical expression for its steady state , allowed us to systematically perturb the steady state and ask how these perturbations affect the sensitivity to TRAIL . To illustrate this capability we varied the steady state abundance of each major xEARM species over a 100-fold range , centered about each species' wildtype value as reported in Table S2 . For each variation , we performed a binary search to identify . The results from this procedure are plotted in Figure 6B . As expected , increases in XIAP , Bcl2 , FLIP , and Bar result in reduced sensitivity to TRAIL stimulation , while increases in Procaspase 8 , TRAIL receptor DR4/5 , Bax , and Bid result in increased sensitivity [46] . What is interesting , however , is the following . First , TRAIL sensitivity is most affected by changes in the abundance of Procaspase 8 and Bar , an inhibitor of active caspase 8 [47] . The ability to activate caspase 8 , then , appears to be a critical determinant of TRAIL sensitivity , as previously suggested [48] , [49] . Second , the abundances of Procaspase 3 , 6 , and 9 have little effect on the sensitivity to TRAIL . This observation is in good agreement with the model-based prediction that induction of MOMP does not require positive-feedback via this caspase loop [14] . A common metric for describing how model parameters affect the sensitivity to TRAIL is to calculate the change in time at which death occurs in response to a small change in each parameter [6] , [44] , [50] . It is conceivable , however , that changes in the time of death do not accurately reflect changes in the threshold of TRAIL at which death occurs . Therefore , to test this assumption we calculated parameter sensitivity coefficients for the ligand threshold , , and the time at which death occurs , , using the xEARM and EARM models , respectively . The numerators and were calculated by backward finite difference approximation and all sensitivities were normalized to the maximum observed sensitivity for each metric ( Figure 6C ) . The data show good agreement for positive regulators of TRAIL sensitivity , but some disparity in the negative regulators . Specifically , while is particularly sensitive to changes in XIAP and Bcl2 , is most sensitive to changes in Bar . This result argues that some caution should be taken when equating changes in the time of death with changes in TRAIL sensitivity .
We have described a simple but flexible method for deriving analytical expressions for the steady states of mass action models . Central to our method is the observation that mass action models are systems of polynomial equations that are generally no greater than degree 2 . This permits a partitioning of rate constants and species concentrations into disjoint sets of quantities , and , where the reaction velocity vector is linear with respect to the variables in . If the cardinality of is greater than the rank of the stoichiometric matrix , then the steady state equation can be solved analytically using simple linear methods . There is considerable benefit to deriving an analytical expression for the steady state of a model . An analytical expression can be used to identify network ultrasensitivity [51] , robustness [52] , multistationarity [53] , and invariants [54] . For enzyme catalytic models that have no true steady state but nevertheless satisfy the assumptions for quasi-steady state , an analytical expression can relate the rate of product formation to the initial concentrations of the substrates and enzyme [55] . Critically , these properties do not depend on the numerical values of the parameters , which may be difficult to measure [56] . In our companion manuscript , we show that analytical steady state expressions can be used to identify changes in the kinetic rate constants that do not alter the species concentrations . These isostatic perturbations can be used to characterize the dynamic plasticity of a system , and also how changes in the rates of protein turnover can affect the response to perturbation , independently of changes to steady state concentrations . Even if numerical interrogation is ultimately intended and all parameters must be assigned values , deriving an analytical expression for the steady state still confers a number of benefits . First , including steady state constraints can facilitate the construction of a model [57] . As illustrated by our treatment of the Open Michaelis-Menten model , py-substitution affords considerable flexibility in selecting which quantities are independent — thus requiring numerical values prior to simulation — and which quantities can be derived from the independent quantities . This partly transforms the problem of parameterizing a model from one of numerically fitting the rate constants to available data [58] , to one of identifying the steady state expression that maximizes incorporation of known quantities into the independent set of parameters . Second , incorporating steady state concentration measurements can reduce the total number of parameters required . In the traditional approach to parameterization , every rate constant is assigned a value prior to simulation , as well as the abundance of any species not subject to synthesis and degradation . Using py-substitution , only independent quantities must be assigned a value . This number is equal to the total number of species and reactions , minus the rank of the stoichiometric matrix . As the stoichiometric matrix approaches full rank , this number converges to the number of species . Since most systems have more reactions than species , py-substitution often requires fewer parameters than the traditional approach . This can be observed in the xEARM model , where 119 parameters are required for simulation after deriving a steady state expression using py-substitution ( 100 rate constants , 18 species , and the mitochondrial volume ) , versus 133 parameters required for traditional parameterization ( 115 rate constants , 17 species , and the mitochondrial volume ) . Further , in the case of the xEARM model , we have demonstrated that an analytical expression of the steady state allows systematic characterization of its effect on the response to perturbation . This was made possible in two ways . First , it allowed the model to operate at a non-trivial steady state . In the original EARM model , infinite sensitivity to TRAIL is caused by unbalanced reactions . Once the receptor is engaged , caspase cleavage and pore formation proceed deterministically to completion . As a result , for cells to be “alive” prior to stimulation , the model must assume a trivial steady state in which the abundance of TRAIL and all reaction velocities are zero . Using py-substitution , we were able to engineer a non-trivial steady state that is viable at low doses of TRAIL . Second , we were able to apply systematic changes to the steady state concentrations . By virtue of the mapping function , these resulted in compensating changes to the kinetic rate constants such that steady state was preserved . For each modification , we were then able to calculate the number of TRAIL molecules required to induce cell death , as well as the sensitivity of this threshold to changes in the steady state concentrations of different species . Previous studies with models operating at trivial steady states employed sensitivity metrics that were with respect to the time at which death occurs , and not whether it occurs [6] , [44] . These studies suggested that the dynamics of TRAIL-induced cell death depend critically on Bcl-2 [44] . Also , whether cell death proceeds to completion depends on XIAP [44] , and whether the mitochondrial feed-forward loop is required depends on the ratio of XIAP to Procaspase 3 [59] . In contrast , our analysis indicates that whether cell death occurs is primarily determined by the ratio of Procaspase 8 to its negative regulator , Bar . Our sensitivity analysis with respect to the threshold at which death occurs is therefore related to but distinct from analyses that consider only the timing of death , and may relate better to clinical applications since we don't assume co-treatment with cyclohexamide . For all these reasons , an analytical expression for the steady state of a model can be of general benefit to cell systems modeling . Indeed , other methods have previously addressed the challenge of deriving analytical steady state expressions , most notably the King-Altman method . Prior to the advent of modern computers , the authors realized that for a particular class of mass action models , the laborious calculation of steady state enzyme ratios could be achieved by a conceptually simpler graphical method . As we have shown , however , this simpler approach is no longer more efficient . More significantly , the King-Altman method requires that all reactions be first- or pseudo-first order in the time-varying species . Without this stipulation , Equation 7 no longer holds and the reaction network can no longer be described by a graph . This requirement is often stated as a pair of assumptions: 1 ) that no enzyme is itself a substrate and 2 ) that all substrates remain constant over the time scale of steady state formation [22] . The second of these can be considered common to any method that treats time-varying species as constants when solving the steady state equation . The first of these , however , is violated by any cascade of post-translational modifications , for example the well-known MAP kinase cascade [60] . Although recent methods relax these assumptions [28] , [29] , in the contemporary systems biology literature , analytical derivation of the steady state rarely , if ever , precedes numerical interrogation of a model . Since this derivation is of considerable value , we sought to develop a method that was simple , scalable , and general to mass action models . First , we described our method using only concepts from linear algebra , and we have provided complete code for all seven examples described in this manuscript , with implementations in either Matlab or Maple . Second , we show that py-substitution scales well . The xEARM model has 58 species and 115 reactions , and we were able to derive a steady state expression in less than a minute on a conventional desktop computer . Finally , we demonstrated that py-substitution can be generally applied to chemical reaction networks whose reaction velocities are modeled by mass action kinetics . This is a considerably broader class of models than can be addressed using the King-Altman and other methods , which require that the reaction network exhibit specific structural properties . This does , however , open up an interesting avenue for further research: precisely what properties must a mass action model exhibit for its steady state to be derived using py-substitution ? How many different steady state expressions are possible , and which of these is the “best” ? As we have shown with the fumarase model , even after the rate constants and species concentrations were partitioned into sets and , 72 different steady state expressions were possible . These different expressions arose from flexibility in selecting the pivot columns in the coefficient matrix , since the pivot vs . free columns partition the linear variables into dependent vs . independent variables . Equivalently , these different expressions arise from flexibility in ordering the linear variables , since different orderings permute the columns of the coefficient matrix and result in a different reduced row echelon form . Since the number of possible steady state expressions is large but finite , a combinatorial optimization strategy ought to be able to identify the best steady state expression , where the difference between any two expressions could take into account measurement uncertainty in the independent quantities , as well as computational complexity in deriving the final steady state expression . Finally , we consider that the steady state may not be the only state of interest , but perhaps specified dynamic states as well . Essentially , this replaces the zero vector in Equation 5 with a vector of non-zero values . From linear algebra , we know that the solution to this dynamic equation can be expressed as the sum of a particular solution to the dynamic equation and an arbitrary point in the null space of the coefficient matrix . The solution is thus straightforward , raising the possibility of incorporating specific dynamic states into the parameterization of a model as well . | Diagnostic markers are derived from steady state measurements , but are used to predict the cellular response to therapy . To develop new and better diagnostics , we would like to systematically characterize the relationship between steady state and the response to a given therapeutic . Mathematical models have powerfully complemented empirical studies in this regard , but it remains challenging to employ these models to characterize the effects of steady state . To do so requires a mathematical expression for the steady state , for which no universal method has been developed . Here , we present a method for deriving a mathematical expression for the steady state of a common class of models , those that obey the Law of Mass Action . We show that our method is easy to use and scales well to large models . We then use our method to characterize the relationship between steady state and the sensitivity to the anti-cancer drug candidate , dulanermin . We find that sensitivity to the drug is strongly affected by the concentration of the signaling molecule , procaspase 8 , and its inhibitor , Bar . Our work thus demonstrates the utility of analytical studies of the steady state and its relationship to drug sensitivity . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"biochemistry",
"biochemical",
"simulations",
"mathematics",
"theoretical",
"biology",
"biochemistry",
"simulations",
"algebra",
"linear",
"algebra",
"biology",
"computational",
"biology",
"algebraic",
"equations"
]
| 2013 | Characterizing the Relationship between Steady State and Response Using Analytical Expressions for the Steady States of Mass Action Models |
Unrepaired or inaccurately repaired DNA damage can lead to a range of cell fates , such as apoptosis , cellular senescence or cancer , depending on the efficiency and accuracy of DNA damage repair and on the downstream DNA damage signalling . DNA damage repair and signalling have been studied and modelled in detail separately , but it is not yet clear how they integrate with one another to control cell fate . In this study , we have created an integrated stochastic model of DNA damage repair by non-homologous end joining and of gamma irradiation-induced cellular senescence in human cells that are not apoptosis-prone . The integrated model successfully explains the changes that occur in the dynamics of DNA damage repair after irradiation . Simulations of p53/p21 dynamics after irradiation agree well with previously published experimental studies , further validating the model . Additionally , the model predicts , and we offer some experimental support , that low-dose fractionated irradiation of cells leads to temporal patterns in p53/p21 that lead to significant cellular senescence . The integrated model is valuable for studying the processes of DNA damage induced cell fate and predicting the effectiveness of DNA damage related medical interventions at the cellular level .
Multiple DNA lesions arise in each cell within an organism every day , caused by errors in DNA replication , by exposure to external factors such as UV light and by a variety of hydrolytic and oxidation reactions [1] . Most simple lesions are repaired quickly and accurately by the cellular DNA-damage response ( DDR ) . The more complex double-strand breaks ( DSBs ) , however , are often left either unrepaired or are repaired incorrectly . Accumulation of persistent DNA lesions leads to apoptosis , cellular senescence or cancer [2 , 3] . The outcome for a cell after a DNA-damaging insult depends largely on the cell type ( or state ) and on its DDR capacity: e . g . , while irradiation of human fibroblasts in culture leads to cellular senescence [4] , irradiating cancer cells leads to apoptosis or mitotic catastrophe [5] . Therefore , clear understanding of control of DDR is important when seeking to identify novel targets for interventions in cancer and ageing [6–9] . Although DNA damage drives cell fate decisions , the actual outcome depends on effects that play out through downstream DNA-damage signalling pathways such as those involving ATM , p53 and p16 [10 , 11] . Recent studies in ATM/p53 signalling have shown that although the amplitude of the signal is affected by the level of damage , it is the temporal pattern of ATM/p53 activity that more strongly affects cell fate [12 , 13] . UV-induced damage causes a sustained response of p53 and strong induction of its target p21 , leading to senescence , whereas γ-irradiation generates pulses of p53 activity that must endure over time if they are to induce p21 signalling and senescence . Interestingly , regardless of the type of damage insult and the temporal pattern of p53 , induction of p21 occurs only in the presence of DNA damage , and not after spontaneous pulses of p53 that occur without damage [14] . Thus , it seems that studying DNA damage signalling without DNA damage occurrence/repair , or vice versa , can explain only part of the cell fate story . A complete explanation requires an integrative , systems-biology approach . While many separate mathematical models of DNA damage repair and DNA damage signalling exist [15–21] , including some from our group , there have been few integrative efforts . Some work has focused on the onset of senescence as a result of damage with varying levels of mechanistic detail [22] [23]; apoptosis has also been included as an alternative cell fate [23] . To date the majority of models are deterministic and DNA damage is considered as a constant input rather than integral part of the system that can change . Two notable integrative studies were done by Passos et al and Ma et al [4 , 24] . Ma et al . built a model of random DNA damage induction and stochastic repair , ATM signalling and p53/MDM2 negative feedback to explain undamped oscillations in p53 after irradiation . Passos et al added p21-based early senescence signalling downstream of p53 but did not include details of DNA damage repair; their model and accompanying in vitro experiments demonstrated that irradiation-induced senescence requires a positive feedback between reactive oxygen species and DNA damage [4] . Crucially , since their model did not include mechanistic details of DNA damage repair or feedback between p53/p21 signalling and DNA damage factors , it fell short of predicting the amount of senescence after different amplitudes/time courses of irradiation or explaining the long-term decrease in DNA damage repair following irradiation observed in experiments [4 , 16] . In parallel , a mechanistic stochastic model of DNA damage repair by NHEJ , was developed to try to explain this decrease [16] . This model agreed with the measured repair dynamics for the first eight hours after irradiation ( short-term ) , but overestimated the speed of repair after the eight hour mark ( long-term ) . It was hypothesized that the disparity might be explained by events downstream of the DDR , e . g . cellular senescence . In the present work we undertake the important task of integrating the model of early senescence with the mechanistic model of NHEJ and examine whether the combined model can both explain the long-term DNA damage repair dynamics and predict senescence for different temporal patterns of irradiation [4 , 16] . The integrated model was built in rule-based format in BioNetGen [25–27] and for simulation we used the network-free simulator NFsim [28] , which lead to a considerable reduction in computation time . We show that the model is able to explain the long-term decrease in DNA damage repair and also to predict experimental results of two p53 signalling studies not used in the model calibration . Additionally , we demonstrate that the model generates a novel prediction that significant cellular senescence can occur after repetitive low-dose irradiation , for which we provide some experimental support .
We built a single-cell stochastic model of non-homologous end joining ( NHEJ ) , DNA damage signalling and early senescence signalling by developing and integrating two recent stochastic models: one of DNA damage repair by NHEJ [16] and one of irradiation-induced senescence focusing on the DDR [4] ( Fig 1 ) . The model describes the formation of DNA DSBs by irradiation and background ROS and their repair via the fast DNA-PKcs NHEJ ( D-NHEJ ) and the slower backup NHEJ pathway ( B-NHEJ ) , also known as microhomology-mediated end joining [29] . In D-NHEJ , Ku70/80 and DNA-PKcs are recruited to the DSB and form a complex . Ku80 is redox sensitive , which causes an increase in the rate of Ku70/80 dissociation from a DSB with high ROS levels [30] . DNA-PKcs then undergoes autophosphorylation and ligase IV ( LiIV ) ligates the DSB . In B-NHEJ , Parp-1 is recruited to the break site followed by ligase III ( LiIII ) which ligates the DSB [31 , 32] . Since tracking the dynamics of individual DSB breaks was one of the aims of the study , each DNA break site was modelled as an independent species . In addition to binding of Ku70/80 and Parp-1 , the DSB causes activation of ATM which phosphorylates H2AX leading to formation of a DNA damage focus [33] . Each focus resolves independently of the DSB repair ( i . e . , a focus can be resolved even if the DSB is still present and in that case a new focus often forms , but it is also possible for the focus to still exist after the DSB has been repaired ) , but the repair of the DSB requires the presence of the damage focus [34] . The phosphorylation of H2AX at the site of a DSB can also be carried out by the other Phosphatidylinositol 3-kinase-related kinases ( PIKK ) present in the model DNA-PKcs [35] once it has undergone autphosphorylation . Activated ATM phosphorylates p53 , which increases its stability by reducing its capacity to bind to MDM2 for targeted degradation . Activated p53 increases transcription of MDM2 and CDKN1A ( p21 ) . p21 in turn activates early senescent signalling ( phenotype Sen in Fig 1 ) , increasing production of ROS via GADD45 and phosphorylation of p38 . Once early senescent signalling is activated , the repair factors Ku70/80and Parp-1 are depleted to the levels observed in senescent cells [36 , 37] ( S2 Table ) . Our model of DNA damage repair is limited to NHEJ as this is the dominant mechanism of repair during the G1 phase of the cell cycle [38] which also corresponds to the state of cell cycle arrest for early senescent cells . G1 lasts approximately 24 hours in MRC5 human fibroblasts ( unpublished observation ) , and we assume the validity of the model for early senescent cells for at least this period of time . As the senescent state deepens other processes take over [39 , 40] and the validity of the model prediction much beyond 48 hours has to be regarded with some caution . The original , separate models were created in the SBML format; however , due to their size , SBML was unsuitable for their integration . Instead , the original reactions were translated to a BioNetGen rule-based format [25 , 26] with appropriate recalibration where the separate models overlapped ( Fig 1 and Methods ) . The integrated model , consisting of 4 , 583 species and 22 , 341 reactions , is available to download as a bngl file in Supporting Information . A full description of the model reactions and reaction rates can be found in the Supplementary Information ( S1 Text and S1–S3 Tables ) . In all key readouts , the integrated model matched qualitatively the output of the original published models . To ensure the consistency of the integrated model , and to constrain further its parameters , we used three additional published datasets that were not used in constructing the original models [13 , 16 , 41] . For each dataset a subset of the model parameters was recalibrated , while ensuring that the results matched the dataset used in the construction of the original models . These recalibrated parameters were: 1 ) the rate at which p21 activity leads to early senescence , established by Passos et al and based on measurement of senescence after a single level of irradiation ( selected because the model is very sensitive to its value ) [4]; and 2 ) the parameters of the feedback between senescence and abundance of Ku70/80 and Parp-1 , as literature only provided approximate steady-state levels of both molecules after senescence and not the dynamics of their change [36 , 42] ( S2 and S3 Tables ) . A further parameter that was altered from the original models was the rate of background ROS production , which was determined experimentally ( S3 Table ) . The first dataset used was the full dynamics of DNA damage foci observed in MRC5 fibroblasts after irradiation [16] . Whereas only short-term data was used in parameterization of the original NHEJ model , we here included the long-term data . As noted earlier , the original NHEJ model was able to explain short-term but not the long-term dynamics of DNA damage foci [16] . The early-senescence feedback , which increases the production of ROS and depletes the repair factors Ku70/80and Parp-1 , enabled the model to match both short-term and long-term dynamics of DNA damage foci ( Fig 2 and S1 Fig ) . Resolution of foci is faster in non-irradiated cells compared to irradiated ones; additionally almost all foci in non-irradiated cells resolve in 24 hours , whereas only around half of the foci in irradiated cells do so . All in all , the decrease in DNA damage repair after irradiation can be explained by a reduction in the efficiency of D-NHEJ and a shift from the D-NHEJ in low ROS normal cells to the slower B-NHEJ in high ROS/senescent cells , as has been demonstrated in the original model [16] , and by a stabilization of high ROS levels and further reduction of both D-NHEJ and B-NHEJ due to depletion of Ku70/80 and Parp-1 after early-senescence signalling has become activated , which was added in the integrated model . The proposed early senescent signalling feedback is therefore a plausible mechanism behind the long-term decrease in DNA damage repair . The second dataset used was from a study that measured the dynamics of p53 after exposing MCF7 breast cancer cells to different levels of irradiation [41] . The same dataset was also used for validation purposes by Ma et al [24] . The study showed that the number of p53 pulses depends strongly on the level of irradiation , but the average amplitude of the pulses and time between consecutive pulses have only a weak dependence , a phenomenon they named “digital behaviour of p53 oscillations” . The population data derived from a minimum of 100 cell simulations agreed very well with the measured p53 pulse properties ( Fig 3A , 3B , and 3C ) ( t-test; p > 0 . 05 for all experiment vs simulation comparisons , except pulse amplitude at 2 . 5 Gy , where p < 0 . 001 ) . The discrepancy in p53 pulse amplitude at 2 . 5 Gy ( Fig 3C ) may be attributed partly to the stochasticity of the system , but may also arise from the lack of a precise definition of a pulse in Lahav et al ( for the definition of pulse used in this study see S2 Text ) . In accordance to previous observations , the p53 amplitude was much more variable ( coefficient of variation of about 60% ) than the period between pulses ( coefficient of variation about 25% ) [43] . The width of each p53 pulse was 350 ± 160 min in Lahav et al . vs 385 ± 95 min predicted by the model . The time to the first pulse peak was 360 ± 240 min vs 355 ± 165 min , and the time between consecutive pulses was 440 ± 100 min vs 495 ± 130 min . These results computationally confirm that the link between DNA damage repair and p53 signalling is well represented in the integrated model . The final dataset used was from a recent study where MCF7 cells were exposed to different levels of irradiation whilst inhibiting the p53-MDM2 interaction with Nutlin-3 [13] . The p53 abundance was measured immediately after irradiation and the number of senescent cells several days later . Using the same time course concentration of Nutlin-3 , the p53 response in simulations of the integrated model switched from pulsed to a more sustained p53 signalling ( Fig 3D ) . Although single simulations of the p53 response to irradiation with Nutlin-3 inhibition did not produce a consistently high signal , the levels of p53 did not fluctuate nearly as much as without Nutlin-3 inhibition and were on average much higher . At the same time , after recalibrating the rate at which p21 leads to early senescence ( MCF7 cells were more robust than the MRC5 cells used for calibrating the original models ) the fraction of cells entering early senescence in the simulations matched well with the fraction of Sen-β gal stained cells in the experiments [13] ( Fig 3E; Chi-square , p > 0 . 05 for all experiment vs simulation comparisons ) , again suggesting that the abstracted presentation of senescence in our model is adequate . It has been shown that different patterns of p21 after irradiation can explain why some cells progress to senescence while others escape [4] . However , the contributions of DNA damage repair and signalling to senescence have not yet been analyzed . We reasoned that looking into individual stochastic runs of the recalibrated model could provide valuable insight . In non-irradiated cells under normal conditions the abundance of ROS is approximately constant . The actual number of DNA damage foci in a cell at any time varies stochastically , a higher number of ROS molecules usually but not always producing more DSBs . The induction of DNA damage subsequently increases p53 , which , due to the feedback between p53 and MDM2 , becomes a transient pulse [44] . Higher levels of p53 lead to an increase of p21 , which in turn leads to additional production of ROS , providing the p21 levels are sufficiently high ( compare Fig 4A and 4B ) . Fig 4B shows how a relatively high maintained number of DNA damage foci results in continuous activation of p53 , which eventually increases p21 in the cell , ultimately increasing the level of ROS . In non-irradiated cells the repair machinery is able to keep the levels of damage under control; none of the simulated cells progressed into early senescence under normal background levels of ROS . However , if the background ROS production was increased 10-fold , the number of induced DNA damage foci increased and 32% of cells entered into early senescence after 30 hours . Simulations showed that cells have much more difficulty coping with high-damage events; in our case this was the 20 Gy X-ray typically used to generate irradiation-induced senescence . A pulse of irradiation induces a large number of DNA damage foci which are rapidly repaired through D-NHEJ and B-NHEJ ( Fig 5 ) . The p53 levels peak about 6 hours after irradiation ( Fig 5 and S2 Fig ) . If repair is rapid the level of DNA damage decreases enough to prevent a next high pulse of p53 ( Fig 5B ) , which in turn prevents p21 from causing early senescence and from feeding strongly back into ROS production . In the simulations , over three and a half days approximately 85% of all cells progressed into early senescence , while 15% did not . Of these 85% , almost all had a second large ( but generally smaller than the first ) peak or a sustained high level of p53 shortly after the first peak ( Fig 5A ) , consistent with previous studies [12 , 41] . In the early-senescent cells the ROS levels and the number of DNA damage foci rose , and there were repeated pulses of high p53 and an increasing p21 signal . The average signals show an increase with time for ROS , DNA damage foci , p53 and p21 ( S2 Fig ) , as more cells enter early senescence . If the irradiation pulse is decreased , the feedbacks are weaker and fewer cells enter early senescence over three and a half days ( 38% for 5 Gy , 74% for 10 Gy , 85% for 20 Gy ) . Ultimately , the difference between non-irradiated and irradiated cells is the repair machinery’s capacity to handle the level of damage induced . While the low level of damage in the non-irradiated cells poses little problem , the higher level in cells that have undergone significant amounts of stress is much more likely to overwhelm the repair system . If the damage is not repaired fast enough the feedback reduces the cell’s repair capability , increasing the longevity of DNA damage foci and leading to senescent cells with several permanent DNA damage foci [4] . The standard protocol for irradiation-induced senescence involves exposing cells to a single 20 Gy pulse of irradiation [4 , 45] . However , since the temporal pattern of DNA damage seems to have as great an effect on cell fate as the amplitude of the DNA damage signal , we wished to see whether a fractionated exposure to low-dose irradiation was also likely to produce senescence ( S3 Fig ) . Simulations predicted that a single 5 Gy pulse would cause 38% of the cells to undergo early-senescence; however this would be increased to 55% if 5 x 1 Gy pulses were used instead ( 1 pulse/1 hour or 1 pulse/2 hours ) ( Fig 6A ) . The simulations also suggested that each additional 1 Gy pulse would push more cells into senescence than the preceding one , but that senescence would plateau after five pulses . A single 1 Gy pulse would push 3% of cells into senescence , 2 x 1 Gy pulses 10% , 3 x 1Gy pulses 22% and 4 x 1 Gy pulses 39% ( Fig 6A ) . The model does not predict that 5 x 1 Gy irradiation pulses produce more DNA damage breaks than a single 5 Gy pulse , however it predicts a delay in the decrease of p53 levels and thus widening of the first p53 pulse ( pulse width 585 ± 170 min ) ( S4 Fig ) . This delay is enough to induce senescence in an additional 17% of cells . While both intervals between pulses were predicted to cause about the same increase in senescence , their p53 dynamics were different: 2 hours between pulses caused lower p53 amplitudes but wider pulses , whereas 1 hour between pulses caused higher p53 amplitudes , but a narrower pulse . The model’s predictions were tested by irradiating MRC5 cells with the same time-course of irradiation as used in the model , with Sen-β gal and cell-cycle arrest staining being used as proxies for senescence ( Fig 6C ) . We opted to investigate the temporal pattern with 1 hour between irradiation pulses as it produced a clear distinction in the production of a senescent population within the model simulations and would take much less time than the 2 hour pattern . The experiments confirmed the model predictions that a series of low-dose pulses produced significant senescence and that the higher the number of pulses the more senescence occurred ( Fig 6A ) . However , the experimental data suggested a plateau in the senescent population already after 4 x 1 Gy pulses , just above the values seen from a 1 x 5 Gy dose , suggesting that the delay in the decrease in p53 level achieved by repeated pulsing did not produce the entire predicted effect . When trying to adjust the model parameters to match this data , we were not able to achieve an improvement by changing any single model parameter , suggesting that either more experimentally driven parameter estimation is necessary or that more details should be included in the model to fully capture the process of DNA damage-induced early senescence signalling .
In this study we constructed an integrated model combining two previous stochastic models , one of DNA damage repair by non-homologous end joining , the other of p53 signalling-induced cellular senescence ( Fig 1 ) . We simulated the production and resolution of DSBs in the G1 phase of the cell cycle with and without induction by irradiation . Damage by ROS is the most common source of DSBs even in unirradiated cells although it should be noted that DSBs may also be generated in S-phase which we do not consider here . The integrated model is , to our best knowledge , the first stochastic mathematical model that takes into account both the mechanistic details of DNA damage repair , the downstream DNA damage signalling which leads to cell fate choice and the feedback in-between . The integrated model captures much more of the relevant biology and should therefore provide significantly enhanced explanatory and predictive power . Recent experimental data shows that irradiation not only induces DSBs , but also decreases the rate of DNA damage repair [16] . The shift towards slower repair occurs during the first minutes/hours after irradiation and does not reverse for cells entering senescence ( Fig 2 and S1 Fig ) . Our integrated model shows that the long-term slowdown can be attributed to the redox sensitivity of Ku70/80 , increased production of ROS and decreased transcription of Ku70/80 and Parp-1 during early cellular senescence ( Fig 2 ) . Although the slower repair in the model is due primarily to the redox sensitivity of Ku70/80 , the higher number of complex DSBs being formed by increased levels of stress , from irradiation or ROS generated as a result of the DDR , also contribute to slowing the rate of observed repair . This result significantly extends the capability of one of the earlier models , which showed that the short-term part of the shift can be explained by redox sensitivity of Ku70/80 , but which could not explain the persistence of DNA damage in the long-term [16] . Recently , it has been suggested that a high proportion of permanent damage foci are telomere associated , and thus resistant to NHEJ repair [46 , 47] , and this will need to be included in the model for longer term predictions . The model simulations also matched the p53 dynamics determined experimentally by single-cell imaging after irradiation . While Ma et al’s model was also able to explain this digital behaviour , it could not match the cell-cell variability found in the experiments [24] . Although other mathematical models , which did not explicitly account for DNA damage or the dynamics of its repair , have been used to explain cell-cell variability of the p53 system after irradiation [43] , the deterministic approach used in those models required low-frequency noise to be added to match the experiments . In contrast , our integrated stochastic model was able to match the observed cell to cell variability in p53 signalling without adding additional noise . Slow fluctuations in p53 occurred in cells with little DNA damage , a smaller number of pulses in cells where DNA damage was successfully repaired and continuing pulsing in cases where DNA damage was not repaired ( eventually leading to senescence ) ( Fig 5 ) . Furthermore , our predictions on the entrance of cells into cellular senescence , which greatly depends on the p53/p21 signalling , also matched the experimental data ( Fig 3 ) [13] . Following the proposal of Lahav et al that the temporal pattern of p53 signalling plays a decisive role in determining cell fate , we investigated how DNA damage repair and the p53 signal affect the cell fate outcome [41] . In a low ROS environment , few DSBs occur and the number of p53 molecules fluctuates around a steady-state . In rare cases , when a few DSBs are detected by ATM there is a rapid increase in p53 followed by a fall back to the steady-state . As this single pulse of p53 is not enough to induce much p21 transcription , and as a second pulse does not occur because the DSBs have been repaired by then , the downstream early-senescence signalling is not activated . When irradiation is used , many more DSBs occur and the DNA repair machinery is not able to repair all the damage as quickly . If DSBs are present , many sequential pulses of p53 ensue , causing enough p21 to accumulate in the system to induce early senescence signalling . The accumulation of p21 then causes the production of greater levels of ROS via the activation of p38 and GADD45 which decreases DNA repair further and causes more DSBs to form , further increasing p21 and finally pushing the cells into senescence . As our model ( calibrated to MCF7 cells , which have been shown to be more resistant to senescence[48] ) shows , exposure to 20 Gy irradiation does not lead to complete senescence , with a few cells managing to escape due to the inherent stochasticity in the system . The ability of some cells to evade senescence may be an important step in the initiation of carcinogenesis , since these cells are highly likely to have incorrectly repaired DNA , a feature that often leads to cancer [44 , 45] . However , further mutations are probably necessary , as we have never witnessed any cancerous outgrowth after 20 Gy irradiation of human fibroblasts . Although p53 dynamics predicted by our model is very similar to the dynamics measured by Lahav et al [41] , there are a few important discrepancies . One is that our model predicts an almost 4-fold higher initial pulse of p53 after irradiation than the following pulses ( while there is almost no difference between the pulses in [41] ) . The other is that the delay in subsequent pulses in our model has more variability than the measured one . Interestingly , when we increased the steady-state abundance of p53 in our simulations to values closer to the measured abundance in MCF7 cells ( used in [41] ) , the difference between the amplitude of the first p53 pulse to other pulses dropped below 2-fold and the time delay between consecutive pulses became more uniform ( S5 Fig; 495 ± 130 ( low p53 ) and 475 ± 95 ( high p53 ) ) . These values are also closer to those measured in another study of p53 oscillations [49] . It is worth noting that to achieve a new higher p53 steady-state , but at the same time keep the senescence outcomes at the same level , as in the original integrated model , we had to change the rates of production and degradation of p53 molecules , thus somewhat changing the dynamics of the original model . Nevertheless , we believe our results indicate that the model is very sensitive to the number of molecules used and that the low number of p53 used in our model might not be adequate for simulating p53 dynamics in cells with higher p53 abundances . Because sustained p53 signalling was shown to be more efficient in causing senescence than p53 pulsing , we investigated whether repetitive induction of DNA damage would induce senescence differently than single damage events , as has been recently suggested [13] . The model predicted that fractionated low-dose irradiation with 1 Gy would cause a significant number of cells to undergo senescence , although almost no cells undergo senescence when only a single 1 Gy pulse is applied , and that each further irradiation event ( up to about 5 ) would send more cells towards senescence . This is consistent with a study that observed accumulation of DSBs after fractionated therapy [50] . When we sought to confirm this experimentally , the results matched the model’s predictions in broad terms: fractionated irradiation lead to a significant amount of senescence and more pulses ( but the same cumulative dose ) lead to more senescence; however , the difference between 5x1 and 1x5 Gy pulses was smaller than predicted by modelling and not statistically significant . We currently do not have an explanation for this , but it could be due to different DNA sites , e . g . telomeres , having different repair dynamics [47] . Although others before us have found senescent cells after fractionated irradiation , the irradiation events were further apart ( e . g . 5 doses over 2 days ) , of higher dose ( 2 Gy or more ) and generally achieved much lower senescence ( below 5% ) [51 , 52] . Our study suggests that the temporal pattern of irradiation events could play a very important role in radiation therapy and that an integrative modelling approach might therefore be useful for its optimization . In general , the system is controlled by the dynamics of just two components , the damage foci and p53 . Most damage foci are permanently resolved within 30 minutes of occurring in MRC5 . However the response of p53 is much slower , a pulse triggered by damage may last over 300 minutes . The system , therefore has a rapid trigger for DDR buts its effects endure for ten times longer than the duration of a typical break . As a result , a single burst of damage only triggers a single pulse of p53 unless the damage is not resolved—which is the case for the overwhelming damage of a 20Gy pulse of radiation . However by introducing more bursts of damage ( even relatively small ones ) the decline in p53 levels is prevented and levels remain elevated driving the production of a more ROS to make more breaks and thereby inducing senescence . During the writing of this paper we have become aware of a new study that looked at the effects of Parp-1 inhibition on radiation sensitivity in cancer cells [53] . The study found that the same cumulative dose of irradiation spread over more time induces more cellular senescence than a quick high dose , and that Parp-1 inhibition can make cells more sensitive to irradiation induced senescence . This partly confirms our modelling predictions and suggests that our modelling approach could also have value in investigating molecular interventions that affect the DNA damage repair or DNA damage signalling pathways . In summary , the integration of two previously constructed mathematical models has allowed us to make more complete and powerful predictions than with either of the two original models: the senescence feedback was shown to be a plausible explanation for the decreased DNA damage repair that eluded the NHEJ model , while the dynamic DNA damage input from the NHEJ model enabled the senescent model to predict senescence after fractionated irradiation . The size of the original model forced us to abandon SBML and opt for a rule-based approach , which proved very effective both in the ease of integration and in the speed of simulation [25] . Altogether , the model provides a suitable tool for investigating DNA damage induced senescence and can easily be expanded to add more details of the DNA repair pathways [15 , 21 , 54] , or adapted for modelling other cell fates or cancer therapies that intervene at the level of DNA damage repair or signalling .
The conversion of the two original SBML models to a single BioNetGen rule-based model was performed by first translating all distinct molecular species within the SBML models into molecules with components ( e . g . binding and phosphorylation sites ) and states in the BioNetGen language . The following list ( see also Fig 1 ) provides a summary of the molecules , their components and their states in the BioNetGen model ( a more detailed explanation can be found in the Supplementary model files ) : Source_of_ROS ( ) ; ROS ( ) ; DNA ( id~1~2~3…~50 , site~ok~sdsb~cdsb , h2ax~u~p~foci ) ; Ku ( dna , cs , cys~red~ox ) ; DNAPKcs ( ku , liIV , psite~u~p ) ; LiIV ( cs ) ; PARP ( dna , liIII ) ; LiIII ( PARP ) ; ATM ( state~0~1 , h2ax ) ; P ( ) ; p53_mRNA ( ) ; p21_mRNA ( ) ; MDM2_mRNA ( ) ; GADD45 ( ) ; p53 ( psite~u~p ) ; MDM2 ( psite~u~p ) ; p38 ( psite~u~p ) . In BioNetGen each molecule can have different components , which can either bind to components of other molecules or be in different states . For example , Ku ( dna , cs , cys~red~ox ) has three components: component dna for binding a DNA molecule , component cs for binding DNA-PKcs and component cys that can be either in the reduced or oxidized state . As the original models partially overlapped , some reactions needed to be reformatted in the integrated model . For example , in the NHEJ model , ATM and the MRN complex ( MRE11A-RAD50-NBN ) were single molecules located at a specific site of damage to facilitate phosphorylation of H2AX , which eventually form part of the damage focus . However , in the senescence model , a pool of ATM molecules was activated by DNA damage and in turn caused the phosphorylation of p53 and MDM2 . To account for both functions of ATM we modelled it with an active and inactive state ( state~0~1 ) and gave it a binding domain ( h2ax ) so that it could also become part of the DNA damage foci which is built up and around the phosphorylated H2AX . Within the model the damage foci and the repair complexes form independently of one another but the ligation of a dsb does not occur until both the damage foci and one of the repair complexes have been formed around the site of DNA damage . Each of the reactions from the original SBML models was then re-created in the rule-based format using the molecules defined above . Rate constants were derived from the original models , although some reactions , such as the recruitment and binding of repair proteins , were combined to change from a two-step process to a one-step process for quicker simulation ( Ku70/80 , DNA-PKcs , XRCC4 , LIG4 , Parp-1 , XRCC3 and LIG3 ) . We also recreated a timed irradiation event used in the senescent model to simulate the treatment of a cell with irradiation [4] . All the reactions of the integrated model and the reaction rates can be found in S1–S3 Tables . A typical ( reaction ) rule in BioNetGen is composed of multiple species that interact and change as a result of their interaction , for example the rule DNA ( site ! ? ~sdsb , h2ax~u ) + ATM ( state~1 , h2ax ) -> DNA ( site ! ? ~sdsb , h2ax~p ) + ATM ( state~1 , h2ax ) kh2axp1 can be interpreted as a reaction that occurs between a DNA molecule with a simple DSB ( ~sdsb ) and unphosphorylated h2ax and an ATM molecule in the active state ( state~1 ) with the h2ax binding site free . The reaction leads to phosphorylation of h2ax ( h2ax~p ) at a rate of kh2axp1 . While the DNA molecule has one more component—id , it does not feature in this reaction . This is therefore interpreted as irrelevant for the reaction , in other words the reaction will take place regardless of the state of the id component , therefore this single rule actually codes for 50 different reactions ( DNA ( id~1 , … ) , DNA ( id~2 , … ) , … , DNA ( id~50 , … ) ) . Another useful syntax is site ! ? , which is interpreted in the following way: only those DNA molecules , which have something bound ( ! ) to the site can participate in the reaction , however what exactly is bound is not important ( ? ) . In our case , Ku70/80or Parp-1 can be bound to the DNA and the reaction takes place with the same reaction rate in both cases . In human fibroblasts p21 is the major driving factor behind the transition into senescence due to DNA damage [55–57] . When p21 increases , the pro-senescence pathway becomes active; however senescence can be avoided if the levels are not kept consistently high . To recreate this behaviour in our model we created a molecule called Sen: Sen ( int~1~10~PLUS~MINUS , State~norm~sen ) Sen has a component int made up of integers between 1 and 10 which increase/decrease depending on the level of p21: high p21 causes the int to increase ( PLUS function ) , while in low levels int decreases ( MINUS function ) in the rule ( S2 Table ) . If the int component reaches 10 the cell has become senescent and a reaction takes place that changes the second component called State from state norm to sen . The presence of the molecule Sen with the component state sen triggers a number of functions in the model that gradually reduce the levels of Ku70/80 and Parp-1 to levels observed in senescent states and reduce the ability of the Ku complex to bind to a DSB [36] . All simulations were carried out using the network-free stochastic simulator NFsim [28] running on a desktop PC . Non-irradiated cells were simulated for 30 hours . Irradiated cells were simulated for 3 hours prior to irradiation , and for 78 hours after irradiation , which matched the corresponding experimental data and also allowed the simulation of the early senescent feedback . The parameters chosen for recalibration were recalibrated using relevant experimental data and the fmincon optimization function in Matlab ( Matlab 2010b , MathWorks , Massachusetts , United States ) using the default settings ( the model was run > 500 times for evaluation ) . Among the parameters varied in the simulations were the levels of environmental ROS for the non-irradiated cells , the level of the irradiation exposure for the irradiated cells , and finally the number and frequency of irradiation events for the study of senescence after fractionated irradiation . Human MRC5 fibroblasts ( obtained from ECACC ) were cultured in Dulbecco’s modified Eagle’s medium ( DMEM; Sigma , Dorset , UK ) enriched with 10% heat inactivated foetal bovine serum ( FBS; BioSera , Ringmer , UK ) , 2mM L-Glutamine and 1% penicillin/streptomycin . Fibroblasts were grown on 150 cm2 flasks ( Corning Incorporated , Corning , NY , USA ) in a humified atmosphere of 5% CO2 , 20% O2 and 95% air at 37°C . Cells were split at 90% confluence into identical fresh medium . Cells were seeded onto 18 mm Ø No 1 . 5 glass coverslips 48 hours before irradiation and grown to 80% confluence over 48 hours as described above . Coverslips to be irradiated were subjected to a 1 or 5 Gy irradiation pulse in a XRAD225 Biological Irradiator ( Precision X-ray Inc , N Branford , CT , USA ) . Media for each slide was then replaced and slides left for an hour before a repeat pulse was delivered ( where applicable ) . Slides were washed in phosphate buffer solution ( PBS ) and fixed 96 hours after the start of irradiation protocol in 2% paraformaldehyde in PBS for 5 minutes . Slides were then washed and stored in PBS . Before staining slides were washed in 5 mM MgCl2 in PBS ( PBS-Mg ) . Cells were stained for senescence overnight using Sen β gal staining solution ( 5 mM MgCl2 , 50 mM K4Fe ( CN ) 6·3H2O , 50 mM K3[Fe ( CN ) 6] , 1mg/ml X-gal ( Sigma ) in PBS at pH 5 . 7 ) . The next day slides were washed with PBS-Mg and mounted onto microscope slides using Vectashield Mounting Medium with 4' , 6-diamidino-2-phenylindole ( DAPI ) nuclear staining ( Vector Laboratories , Peterborough , UK ) . Cells were imaged with a Leica DFC420 camera ( Leica Microsystems UK , Milton Keynes , UK ) mounted to a Nikon Eclipse E800 microscope with a Nikon Plan Fluor 40x/0 . 75 air objective ( Nikon UK , Kingston Upon Thames , UK ) . Areas were selected for counting using only the DAPI stain ( excitation at 340–380 nm , emission 425nm ) so as not to bias the results . Both DAPI fluorescence and Brightfield images were taken for senescence analysis . The two sets of images were then overlaid in Image J so that the presence of DAPI around the nucleus of a cell could be detected . Presence of the stain within the cell indicated a transition to a senescent state . To better calibrate the levels of ROS production during senescence in the model we measured the rates of the release of hydrogen peroxide ( H2O2 ) from cultured cells using Amplex Red reagent ( Invitrogen , A12222 ) . H2O2 is the most common source of damage to DNA as it has a long half-life and is mobile including being able to diffuse across cellular membranes . Amplex Red ( 10-acetyl-3 , 7-dihydroxyphenoxazine ) reacts with H2O2 in a 1:1 stoichiometry to produce the red-fluorescent oxidation product , resorufin , in the presence of horseradish peroxidase . The cells were trypsinised and resuspended in culture media , and the time courses of resorufin appearance were followed fluorometrically at 37°C at an excitation 544 nm and an emission 590 nm in a black bottom 96 well plate using a FLUOstar Omega ( BMG Labtech ) . Each well consisted of 150 , 000cells , 50μM Amplex Red and 2U/ml horseradish peroxidase . Known amounts of H2O2 were added to blank wells to construct the H2O2 standard curve in order to convert the fluorescence arbitrary units to moles of H2O2 . | All cells are subject to damage and DNA is the most important molecule to protect . Cells communicate DNA damage through p53—‘the guardian of the genome’—and the dynamics of p53 signalling is one the main mechanisms that determine the outcome for the cell . On detection of DNA damage , p53 is activated and cell cycle arrest is induced: if the DNA damage is repaired quickly then the signalling ends and the cell returns to normal function; if the DNA damage persists then the signalling continues and cells may undergo senescence or apoptosis . Here , we develop a computational model that can simulate the whole process of DNA damage occurrence , DNA damage repair , p53 signalling and cell fate and successfully predict how persistent DNA damage can lead to cellular senescence . The model predicts that using repeating low dose irradiation as a source of damage is as effective as a single large dose , which could have important implications for radiation therapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| []
| 2015 | Integrated Stochastic Model of DNA Damage Repair by Non-homologous End Joining and p53/p21- Mediated Early Senescence Signalling |
Influenza virus has the ability to evade host immune surveillance through rapid viral genetic drift and reassortment; therefore , it remains a continuous public health threat . The development of vaccines producing broadly reactive antibodies , as well as therapeutic strategies using human neutralizing monoclonal antibodies ( HuMAbs ) with global reactivity , has been gathering great interest recently . Here , three hybridoma clones producing HuMAbs against influenza B virus , designated 5A7 , 3A2 and 10C4 , were prepared using peripheral lymphocytes from vaccinated volunteers , and were investigated for broad cross-reactive neutralizing activity . Of these HuMAbs , 3A2 and 10C4 , which recognize the readily mutable 190-helix region near the receptor binding site in the hemagglutinin ( HA ) protein , react only with the Yamagata lineage of influenza B virus . By contrast , HuMAb 5A7 broadly neutralizes influenza B strains that were isolated from 1985 to 2006 , belonging to both Yamagata and Victoria lineages . Epitope mapping revealed that 5A7 recognizes 316G , 318C and 321W near the C terminal of HA1 , a highly conserved region in influenza B virus . Indeed , no mutations in the amino acid residues of the epitope region were induced , even after the virus was passaged ten times in the presence of HuMAb 5A7 . Moreover , 5A7 showed significant therapeutic efficacy in mice , even when it was administered 72 hours post-infection . These results indicate that 5A7 is a promising candidate for developing therapeutics , and provide insight for the development of a universal vaccine against influenza B virus .
Influenza virus remains a constant public health threat . During annual epidemics 5–15% of the worldwide population are typically infected , resulting in 3 to 5 million cases of severe illness and between 250 , 000 to 500 , 000 deaths per year [1] , [2] . While all age groups are affected by the disease , most influenza-related hospitalizations in industrialized countries occur in young children , the elderly , and the immunocompromised [3] . Like H1 and H3 subtypes of influenza A virus , influenza B virus also causes epidemics in humans [2] . In contrast to influenza A virus , influenza B virus is found almost exclusively in humans and has a much slower mutational rate [3]–[5] . However , cocirculation of two phylogenetically and antigenically distinct lineages , represented by B/Yamagata/16/1988 and B/Victoria/2/1987 , has caused antigenic variation through genetic reassortment and antigenic drift from cumulative mutations , leading to annual epidemics [6] , [7] . Currently , two main countermeasures , vaccines and antiviral drugs , are used against influenza virus [8] . Vaccination has been the mainstay of infection control . However , the protection afforded by vaccination varies widely , depending on the antigenic match between the viral strains in the vaccine and those that are circulating during a given influenza season , as well as on the recipient's age and health status [1] , [3] , [9] . Neuraminidase inhibitors , such as oseltamivir ( Tamiflu ) and zanamivir ( Relenza ) , and matrix 2 ( M2 ) ion channel inhibitors like amantadine , have been widely used for the treatment of influenza viral infection and are proven to be quite effective against susceptible strains [10] . However , they have limited efficacy in delayed administration after the onset of illness [11] , and widespread use has resulted in the emergence of resistant viral strains , as seen in H1N1 and H3N2 [12]–[15] . Of note , M2 blockers are not active against influenza B virus as it has no equivalent to the M2 ion channel protein of influenza A virus [16] . Thus , the development of therapeutic approaches and vaccine design that provide potent and broadly cross-protective host immunity are a global public health priority [17] . Human monoclonal antibodies ( HuMAbs ) prepared from vaccinated donors and patients with viral infections could have potential therapeutic application , and provide significant information on human epitopes that could be important for developing the next generation of universal vaccines [18] , [19] . In addition to classical hybridoma methods [20] , recent advances in technology , such as transgenic mice [21] and yeast or phage display [22] , [23] , have renewed interest in the development of HuMAbs [17] , [24] , [25] . Thus , using phage display or single cell culture methods , several HuMAbs with broad neutralizing activities have been identified against the hemagglutinin ( HA ) protein in influenza A viruses , including C6261 and F10 which react with group 1 [26] , [27] , and CR8020 which reacts with group 2 viruses [28] . Another HuMAb , FI6v3 , that neutralizes both group 1 and group 2 influenza A viruses , has also recently been described [29] . For influenza B virus , by contrast , broadly neutralizing HuMAbs , CR8033 , CR8071 and CR9114 , have firstly reported on September 2012 [30] . A hybridoma method for establishing HuMAbs was developed previously in this laboratory by fusion of the peripheral blood mononuclear cells ( PBMCs ) of influenza-vaccinated healthy volunteers with the fusion partner cell line SPYMEG , which has been optimized for higher reliability of cell fusion by overcoming chromosome deletion problem [31] , [32] . In this study , three HuMAbs that neutralize influenza B virus were prepared using this method . One of the three HuMAbs , 5A7 , reacted broadly with influenza B virus isolates from 1985 to 2006 that belong to both the Yamagata and Victoria lineages , and recognized a highly conserved region in the HA protein . Moreover , 5A7 showed therapeutic efficacy even in mice treated with the HuMAb 72 hours post-infection . These results indicate that 5A7 is a promising candidate for developing therapeutics and will provide insight for the development of the next generation of vaccines universally effective against influenza B virus .
Healthy volunteers were vaccinated with the trivalent HA split vaccine including A/Brisbane/59/2007 ( H1N1 ) , A/Uruguay/716/2007 ( H3N2 ) , and B/Florida/4/2006 strains . Then , 1–2 weeks later , the vaccine-derived PBMCs were fused with SPYMEG cells . After screening for MAb specificity to influenza viruses , the cells in the specific MAb-positive wells were cloned by limiting dilution . Ultimately , three hybridoma clones producing HuMAbs , designated 5A7 , 3A2 and 10C4 , were established against influenza B virus . These HuMAbs did not react with influenza A virus . HuMAb reactivity was tested by immunofluorescence assay ( IFA ) and western blotting using Madin-Darby canine kidney ( MDCK ) cells infected with B/Florida/4/2006 , homologous with the vaccine antigen . All three HuMAbs reacted with HA protein expressed by transfection into 293T cells ( Table S1 ) . IgG isotyping was performed by ELISA and revealed that HuMAbs 5A7 and 10C4 were IgG1 , and 3A2 was IgG3 ( Table S1 ) . Sequences of the VH and VL region of the three HuMAbs were compared and analyzed to the closest germline sequences using IgBlast software in NCBI database . These three HuMAbs were derived from different germ lines except D region in VH of 3A2 and 10C4 ( Figure 1 ) . Next , the three HuMAbs were evaluated for their ability to neutralize influenza B viruses by in vitro virus neutralization ( VN ) assay ( Figure 2 ) . The VN test was carried out on MDCK cells infected with influenza B virus under the treatment with serial four-fold dilutions of HuMAbs . The viruses used were B/Florida/4/2006 , B/Shanghai/361/2002 , B/Johannesburg/5/1999 , B/Yamanashi/166/1998 and B/Mie/1/1993 for the Yamagata lineage , and B/Malaysia/2506/2004 , B/Shandong/7/1997 and B/Victoria/2/1987 for the Victoria lineage . The mouse-adapted B/Ibaraki/2/1985 in the Victoria lineage , used in the passive transfer experiment described below , was also subjected to VN assay . HuMAb 5A7 showed a lower neutralizing activity compared with 3A2 and 10C4 against the Yamagata lineage; however , 5A7 neutralized all strains in the Yamagata and Victoria lineages that were isolated during 1985 to 2006 . HuMAbs 3A2 and 10C4 neutralized the Yamagata lineage effectively , whereas they had little neutralization effect on all the Victoria lineage viruses . An anti-dengue virus HuMAb ( D23-1B3B9 ) derived from PBMCs of a patient infected with dengue virus serotype 2 [33] was used as a control IgG . It did not neutralize any influenza B viral strains . To clarify the mechanism of neutralization by the three HuMAbs , hemagglutinin inhibition ( HI ) and fusion inhibition assays were performed . All three HuMAbs had HI activity; 3A2 and 10C4 showed markedly higher HI titers ( 0 . 39 µg/ml ) than 5A7 ( 25 . 0 µg/ml ) . Fusion inhibition assay showed that they all had the ability to inhibit cell-cell fusion; the concentration of HuMAb necessary for complete inhibition was lower for 3A2 and 10C4 ( 25 µg/ml ) than 5A7 ( 100 µg/ml ) at pH 5 . 5 ( Figure S1 ) . By contrast , the control IgG did not show any HI or fusion inhibition activity , even at 100 µg/ml . In addition , HuMAbs 5A7 , 3A2 and 10C4 were subjected to surface plasmon resonance analysis to examine their binding affinities . Each HuMAb was immobilized on the surface of the sensor chip . The vaccine antigen , HA protein of B/Florida/4/2006 , at concentrations 12 . 5 , 25 , 50 , 100 and 200 nM was consecutively injected on the chip surface and the association and dissociation phases were monitored . KD value could not be calculated for 5A7 precisely as it was difficult to dissociate from HA ( Table 1 and Figure S2 ) . However , the estimated KD value of 5A7 ( less than 5 . 6×10−9 M ) was similar to that of 10C4 ( 1 . 8×10−9 M ) . By contrast , the control IgG did not associate with HA at all ( data not shown ) . Next , the epitope regions recognized by the three HuMAbs were determined . At first , escape mutants were selected by culturing B/Florida/4/2006 in the presence of serial ten-fold diluted HuMAb . MDCK cells were infected with the mixture in a 24-well plate and 72 hours later the supernatants were subjected to VN and HI assays , and direct sequencing analysis of the HA gene ( Figure S3 ) . Escape mutants were not established for 5A7 , even after the virus was serially passaged 10 times in the presence of this HuMAb . 5A7 reacted with the HA protein by western blotting under reducing conditions . Therefore , the region of 5A7 involved in recognition was refined using HA truncation vectors containing HA segments of varying length ( Figure 3A ) . Western blotting with 5A7 was carried out on 293T cells transfected with the truncated HA expression vectors . HuMAb 5A7 reacted with truncated HA segments that included amino acid residues 1–324 but not with those with residues 1–314 ( Figure 3A ) : amino acid numbering was started after the signal peptide [34] . These results indicate that 5A7 recognizes amino acid residues between 315 to 324 ( IGNCPIWVKT ) in the HA protein , which locates near the C terminal of the HA1 protein . The epitope region of 5A7 was determined by mutating each of the 315–324 amino acid residues singly . Each residue was replaced by alanine using a site-directed mutagenesis method . Mutant HAs expressed in 293T cells were tested for reactivity with 5A7 by IFA . Mutants G316A , C318A and W321A did not react with 5A7 ( Figure 3B ) , indicating that 316G , 318C and 321W amino acids critically affected the structure of the epitope of 5A7 . To estimate the conservation of the amino acid sequences in the epitope of 5A7 , HA sequences of influenza B virus were extracted from the NCBI database . Notably , 2 , 851 among 2 , 853 viral sequences ( 99 . 93% ) showed an identical amino acid sequence in the epitope region ( Table S2 ) . Among the Yamagata and Victoria lineages , there was only 1 divergent strain , respectively . By contrast , escape mutants were obtained in the presence of 3A2 and 10C4 . They showed four-fold reduced VN and HI activities compared with the parent virus after just one passage of the virus ( Figure S3 ) . Each escape mutant obtained in the presence of 3A2 and 10C4 had amino acid substitutions at identical positions , 194D and 196T ( Figure S4 ) , both located in the readily mutable 190-helix antigenic site near the receptor binding site [34] . In addition , the different amino acid residues in VH and VL of 3A2 and 10C4 ( Figure 1 ) indicated that these two HuMAbs were derived from different germ lines . These results are consistent with 3A2 and 10C4 reactivity with only the Yamagata strains , since Yamagata and Victoria strains differ in amino acid sequence at this position . HuMAb 3A2 showed low reactivity against B/Shanghai/361/2002 ( Figure 2A ) and was therefore examined for an additional distinct epitope region . To do this , various chimeric sequences of HA were constructed from B/Florida/4/2006 and B/Shanghai/361/2002 , which differ at seven residues ( positions 37 , 40 , 88 , 131 , 227 , 249 and 456 ) , expressed in plasmids , and transfected into 293T cells . IFA of the chimeric HA proteins expressed in 293T cells showed that 131P and 227S were essential for reaction with 3A2 ( Figure S5 ) . These results indicate that the epitope of 3A2 is dependent on residues at positions 131 , 194 , 196 and 227 , and the epitope of 10C4 is dependent on residues at positions 194 and 196 . The epitope regions to which the three HuMAbs map are shown in an HA monomer and trimer three-dimensional models in Figure 3C . HuMAbs 3A2 and 10C4 recognized the top of the globular head including the 190-helix antigenic site , whereas 5A7 reacted with the stalk region distant from the viral membrane . The evaluation of 5A7 as a passive transfer therapy for influenza B viral infection was examined in mice . Six-week-old mice were treated intraperitoneally with 5A7 at 1 , 5 , 10 or 15 mg/kg or with control IgG at 10 mg/kg , 4 hours after an intranasal injection with a lethal dose ( 1 . 47×103 50% mouse lethal dose ( MLD50 ) /mouse ) of mouse-adapted B/Ibaraki/2/1985 . Survival rate and body weight change were checked daily . When body weight decreased to less than 60% of starting weight , mice were sacrificed . With respect to survival rate , complete therapeutic efficacy against the virus challenge was seen with 5 , 10 and 15 mg/kg of 5A7 examined ( Figure 4A , Upper panel ) . The weight change was mild in the groups , especially in those treated with 5A7 at 10 or 15 mg/kg ( Figure 4A , lower panel ) . Next , viral load in the lungs was titrated . Mice infected with mouse-adapted B/Ibaraki/2/1985 ( 1 . 47×103 MLD50/mouse ) , or B/Florida/4/2006 that had been passaged eight times in mice lungs ( 5 . 0×103 focus-forming units ( FFU ) /mouse ) , were treated with 5A7 or control IgG at 10 mg/kg 4 hours post-infection , and then sacrificed on day 3 and day 6 post-infection . The lungs were homogenized and tested in focus-forming assays with MDCK cells . The viral titers were significantly lower in 5A7-treated mice compared to the control IgG-treated group for both viral infections ( Figure 4B ) . Finally , mice were treated with 5A7 or control IgG at 10 mg/kg intraperitoneally at 4 , 24 , 48 or 72 hours after an intranasal injection with a lethal dose ( 1 . 47×103 MLD50/mouse ) of mouse-adapted B/Ibaraki/2/1985 . Survival rate and body weight change were monitored ( Figure 4C ) . Two independent experiments were similarly performed for 5A7 treatment with five mice/group/experiment ( Experiments 1 and 2 in Figure 4C ) . When administered 4 hours post-infection , 5A7 treatment showed complete therapeutic efficacy , as was observed above ( shown in Figure 4A ) . Notably , 80% of mice were alive in the group treated with 5A7 24 hours post-infection . Moreover , several mice survived after treatment with 5A7 48 hours post-infection and , surprisingly , more mice survived when treated 72 hours post-infection . By contrast , 80% of mice treated with control IgG at 4 hours post-infection died and the surviving 20% did not show recovery of body weight during the experiment . All of mice died by 8 days post-infection in the groups treated with control IgG after 24 hours post-infection ( control IgG in Figure 4C ) .
A broadly neutralizing HuMAb , 5A7 , was established using PBMC from a vaccinated healthy volunteer . This antibody neutralizes influenza B virus strains isolated between 1985 and 2006 that belong to the Yamagata or Victoria lineage . Moreover , therapeutic efficacy was shown in mice even when the HuMAb was administered 48 or 72 hours after viral challenge . As previously reported for almost all MAbs broadly neutralizing influenza virus , 5A7 recognized the stalk region of the HA protein [26]–[29] . Importantly , the epitope region recognized by 5A7 is highly conserved in influenza B virus and the divergent strains occur only sparsely ( Table S2 ) . Although influenza B virus has a much slower mutational rate than that observed for influenza A subtypes like H1 and H3 , cocirculation of two phylogenetically and antigenically distinct lineages of influenza B virus leads to annual epidemics in humans [6] , [7] . Dreyfus et al . first reported broadly neutralizing HuMAbs against influenza B virus in September 2012 [30] . The epitope regions they identified were distinct from that of our HuMAb , 5A7 . Characterization of the epitope region recognized by such a HuMAb could therefore provide insight for the development of a universal vaccine . The high degree of conservation of amino acid residues in the epitope region implies that influenza B virus would not easily induce mutation in this region . Indeed , amino acid residues in the epitope region did not mutate even when the virus was passaged ten times under 5A7-treatment conditions . It could be considered that the poor inhibitory activity of 5A7 ( weak HI , fusion inhibition and VN50 ) is the cause of failure to establish escape mutants . However , a previous report showed that escape mutants could be prepared even from an MAb with weak fusion inhibition and complete neutralizing activities ( 25 and 50 µg/ml , respectively ) , and without HI activity , in order to map its epitope [35] , [36] . Since 5A7 shows similarly weak activity , we would also expect to have been able to prepare 5A7 escape mutants and determine the epitope region . Failure to establish escape mutants in the presence of 5A7 could be an advantage for its development as a therapeutic tool , and also for designing the next generation of globally effective vaccines . Generally , MAbs recognizing the globular head show strong HI activity , whereas those against the stalk region usually show none [35] . Thus , it is considered that MAbs against the globular head inhibit the receptor binding step , while MAbs against the stalk region inhibit the fusion step , in viral replication [26] , [37] . In fact , HuMAbs , 3A2 and 10C4 , that recognize the 190-helix in the globular head near the receptor binding site , showed strong HI activity , which suggests that they can inhibit viral binding to the receptor on the host cells . Their ability to inhibit the fusion process ( Figure S1 ) implies that they could secondarily disturb the low pH-dependent structural change in HA by binding to the globular head . Surprisingly , although 5A7 reacted to the stalk region it also showed specific and weak HI activity . This suggests that MAb recognizing the stalk region distal from the viral membrane could affect the ability of the virus to bind to the receptor . As reported for MAbs reacting to the stalk region , 5A7 also showed specific fusion inhibition activity [28] , [38] . These results suggest that HuMAb 5A7 could inhibit viral entry by preventing receptor binding , and the subsequent fusion process . Previous reports show that the concentration of MAbs necessary for viral neutralization was much higher for those recognizing the stalk region of the envelope protein than for those reacting with the globular head in influenza viruses [28] , [36] , [39] , as well as other viruses [40] . In agreement with these reports , the concentration required for viral neutralization in this study was higher for 5A7 than 3A2 and 10C4 ( Figure 2 ) . Such results can be explained by either a difference in binding affinity or in physical accessibility of HuMAbs to the epitope region . 5A7 showed similar KD with 10C4 in binding kinetics analysis ( Table 1 and Figure S2 ) , indicating that 5A7 has more difficulty physically accessing the epitope region of the HA protein . Modifying the HuMAb structure to enable easier access to the epitope region and improving its binding affinity , as described [41] , [42] , could lead to the development of better therapeutic HuMAbs for influenza . HuMAb 5A7 had specific therapeutic efficacy in mice even when administered after viral challenge ( Figure 4C ) . Two independent passive transfer experiments were performed . Surprisingly , in both experiments , mice treated with 5A7 HuMAb 72 hours post-infection had a better survival rate than those treated 48 hours post-infection . It is reported that in the first 24 hours after infection , the levels of lymphocyte apoptosis increase transiently in both nasal-associated lymphoid tissue and spleen , and cellular immune suppression occurs [43] . Temporal cellular immune suppression or infection-mediated endogenous signals could therefore interfere with the efficacy of exogenous HuMAb . These results imply that the timing of HuMAb treatment could be critical for efficacy , and therefore , injection at several time points may be necessary . Mouse-adapted B/Ibaraki/2/1985 was used to examine the kinetics of survival rate and body weight change in passive transfer experiments because other viral strains are not lethal to mice , even if passaged several times in vivo . HuMAb 5A7 protected mice against mouse-adapted B/Ibaraki/2/1985 viral challenge although 5A7 was obtained from a volunteer vaccinated with B/Florida/4/2006 and had shown the lowest sensitivity to this viral strain in vitro ( Figure 2 ) . These results suggested that 5A7 would have therapeutic efficacy against a wide spectrum of influenza B viruses and , in fact , lung viral titers of both mouse-adapted B/Ibaraki/2/1985 and B/Florida/4/2006 were reduced significantly under 5A7-treatment conditions ( Figure 4B ) . Further study using both mice and ferrets with several viral strains is needed to confirm the wide-ranging therapeutic potential of 5A7 in vivo .
Human materials were collected using protocols approved by the Institutional Review Boards of the Research Institute for Microbial Diseases , Osaka University ( #19-8-6 ) . Written informed consent was obtained from the participants . Animal studies were conducted under the applicable laws and guidelines for the care and use of laboratory animals in the Research Institute for Microbial Diseases , Osaka University . They were approved by the Animal Experiment Committee of the Research Institute for Microbial Diseases , Osaka University ( #H21-24-0 ) , as specified in the Fundamental Guidelines for the 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 , Japan , 2006 . HuMAbs were prepared as described previously [31] . Briefly , 10 ml blood was drawn from a healthy volunteer vaccinated in the 2008/2009 winter season with trivalent HA split vaccine , which included A/Brisbane/59/2007 , A/Uruguay/716/2007 , and B/Florida/4/2006 ( The Research Foundation for Microbial Diseases of Osaka University , Osaka , Japan ) , and PBMCs were collected by density gradient centrifugation through Ficoll-Paque Plus ( GE Healthcare ) . SPYMEG cells were used as fusion partner cells . SPYMEG cells , which are non-secretors of human and murine immunoglobulins , were established by fusion between mouse myeloma cell line SP2/0-Ag14 and human megakaryoblastic cell line MEG-01 [31] . The PBMCs were fused with SPYMEG cells using polyethylene glycol #1500 ( Roche ) . The fused cells were cultured in Dulbecco's modified Eagle medium ( DMEM; Invitrogen ) supplemented with 15% fetal bovine serum and hypoxanthine-aminopterin-thymidine . The first screening for antibody specificity to influenza virus was performed by IFA . Cells in the specific MAb-positive wells were then cloned by limiting dilution , followed by a second screening by IFA . Hybridoma cells taken from IFA-positive wells that had a single colony per well were cultured and expanded in Hybridoma-SFM ( Invitrogen ) . MAb was purified from 100 ml hybridoma culture supernatant by affinity chromatography using HiTrap Protein G HP Columns ( GE Healthcare ) and then dialyzed into phosphate buffered saline ( PBS ) using Slide-A-Lyzer Dialysis Cassettes ( Thermo Scientific ) . Control IgG ( D23-1B3B9 ) used in this study was derived from the PBMCs of a dengue patient infected with dengue virus serotype 2 [33] . Eight influenza B vaccine strains ( B/Florida/4/2006 , B/Malaysia/2506/2004 , B/Shanghai/361/2002 , B/Johannesburg/5/1999 , B/Yamanashi/166/1998 , B/Shandong/7/1997 , B/Mie/1/1993 , and B/Victoria/2/1987 ) and a mouse-adapted influenza B virus strain ( B/Ibaraki/2/1985 ) were used . The B/Malaysia/2506/2004 and B/Florida/4/2006 strains were kindly provided by the National Institute of Infectious Diseases , Tokyo , Japan . The mouse-adapted B/Ibaraki/2/1985 strain was provided by Dr . S . Tamura , National Institute of Infectious Diseases [44] . Viruses were propagated either in MDCK cells or in 9-day-old embryonated chicken eggs . Infectivity was titrated by focus-forming assay: MDCK cells in a 96-well plate were adsorbed with viruses serially ten-fold diluted at 37°C for 1 hour . The cells were then washed with PBS and incubated at 37°C for 12 hours . Cells were fixed and subjected to IFA . For IFA , the infected cells were fixed with absolute ethanol for 2 minutes at room temperature and then reacted with hybridoma supernatant for 30 minutes at 37°C , followed by incubation with FITC-conjugated anti-human IgG for 45 minutes at 37°C . For western blotting , the infected samples in a loading buffer containing 2-mercaptoethanol were used for electrophoresis and then blotted to PVDF membrane . They were probed with hybridoma supernatant for 1 hour at 37°C , followed by incubation with HRP-conjugated anti-human IgG for 1 hour at 37°C . The VN test was carried out as described previously [45] , with minor modifications . HuMAb at a concentration of 100 µg/ml was serially four-fold diluted with Minimum Essential Medium ( MEM; Invitrogen ) and incubated with 200 FFU of viruses at 37°C for 1 hour . Then , MDCK cells were adsorbed with the mixtures at 37°C for 1 hour . After incubation for 12 hours , the cells were fixed and subjected to IFA . First , viral titers were determined with a hemagglutination assay . Briefly , the viruses were serially diluted two-fold with PBS and mixed with 0 . 7% ( v/v ) human O-type red blood cells . After incubation at room temperature for 1 hour , hemagglutination units ( HAUs ) were estimated . Next , HI titration was performed as follows: MAb at a concentration of 100 µg/ml was serially two-fold diluted and mixed with 8 HAU per 50 µl of viral sample . After incubation at 37°C for 1 hour , the mixtures were further incubated with 0 . 7% ( v/v ) human red blood cells for 1 hour at room temperature . The lowest concentration of HuMAb that completely inhibited hemagglutination was designated the HI titer . Cell-cell fusion was accomplished as described previously [35] . Briefly , monkey kidney cell line CV-1 cells were infected with B/Florida/4/2006 at an MOI of 0 . 3 . After incubation for 24 hours , the cells were washed with MEM and then incubated for 15 minutes at 37°C in MEM supplemented with 2 . 5 µg/ml of acetylated trypsin ( Sigma ) . After washing , the cells were incubated for 30 minutes with diluted HuMAbs . Thereafter , the cells were treated for 2 minutes at 37°C with MEM supplemented with 10 mM MES and 10 mM HEPES ( pH 5 . 5 ) . After the medium was completely removed by washing , the cells were incubated for 3 hours . Then they were fixed with absolute methanol and stained with Giemsa ( Wako ) . Surface plasmon resonance analysis was performed [46] with a Biacore T200 ( GE Healthcare ) . Interaction was measured in running buffer ( 10 mM HEPES-Na pH 7 . 5 , 150 mM NaCl , 3 mM EDTA , 0 . 005% Tween 20 ) at 25°C . The surface of a CM5 sensor chip was coated with goat anti-human IgG Fcγ antibody ( Jackson ImmunoResearch ) at a density of 10 , 000 resonance units ( RU ) by an amine coupling technique using 1-ethyl-3-[3-dimethylaminopropyl] carbodiimide hydrochloride and Sulfo-NHS for activation , and ethanolamine for blocking . As the ligand , each of the anti-HA HuMAbs was captured on the chip surface via anti-human IgG Fcγ to a density of 10–80 RU . The HA protein of B/Florida/4/2006 vaccine antigen ( The Research Foundation for Microbial Diseases of Osaka University ) was diluted in running buffer at concentrations of 12 . 5 , 25 , 50 , 100 and 200 nM ( as monomers ) . Each concentration was injected sequentially on to the chip at a flow rate of 60 µl/min for a contact time of 60 seconds , and then washed with the running buffer for a dissociation time of 30 minutes . Regeneration of the chip surface was performed with 10 mM glycine-HCl pH 1 . 5 . Binding kinetics were evaluated using Biacore T200 evaluation software version 1 . 0 ( GE healthcare ) using the single kinetics analysis method with a 1∶1 binding model . Escape mutants were selected by culturing B/Florida/4/2006 in the presence of HuMAb as described previously [47] , with minor modifications . Viruses ( to give final concentrations of 100 to 1 , 000 FFU/ml ) were incubated with HuMAb serially ten-fold diluted ( to give final concentrations of 0 . 0025 , 0 . 025 , 0 . 25 and 2 . 5 µg/ml ) , at 37°C for 1 hour . Then , MDCK cells in a 24-well plate were inoculated with the mixtures ( n = 6 for each ) and cultured in DMEM/F-12+GlutaMAX-I supplemented with 0 . 4% bovine serum albumin ( BSA ) , antibiotics and 2 µg/ml acetylated trypsin . After culturing for 72 hours , the supernatants in each well were collected separately and subjected to VN and HI assays . The entire HA gene was directly sequenced from the mixed population in the supernatants of those viral samples that showed a reduced neutralization and HI titer of more than four-fold . Viral RNA extracted with QIAamp Viral RNA Mini Kit ( Qiagen ) was subjected to one step RT-PCR ( Superscript III One-Step RT-PCR System with Platinum Taq High Fidelity; Invitrogen ) with the following HA primer set: 5′-CAGAATTCATGAAGGCAATAATTGTACTAC-3′ forward and 5′-CTCfCGCGGCCGCTTATAGACAGATGGAGCATGAAACG-3′ reverse . The PCR products were purified with Qiaquick PCR Purification Kit ( Qiagen ) . After electrophoresis , the discrete band was extracted using the Qiaquick Gel Extraction Kit ( Qiagen ) and sequenced . HA gene of B/Florida/4/2006 and B/Shanghai/361/2002 was amplified by one step RT-PCR , as described above , and inserted into the pGEM-T Easy Vector ( Promega ) . Mutant and truncated HA genes were generated by site-directed mutagenic PCR ( GeneTailor Site-Directed Mutagenesis System; Invitrogen ) and conventional PCR ( Expand High FidelityPLUS PCR System; Roche ) , respectively , using the HA plasmid inserted into pGEM-T easy vector . Each of the plasmids was subcloned into the expression vector pCAGGS/MCSII [48] . The expression plasmids were transfected into human embryonic kidney 293T cells with lipofectamine 2000 ( Invitrogen ) according to the manufacturer's instructions . Amino acid sequence and sequence information were downloaded from the FTP site of Influenza Virus Resource ( ftp://ftp . ncbi . nih . gov/genomes . INFLUENZA/ ) on January 31 , 2012 . HA sequences of influenza B virus were extracted from the database , and then aligned using the MAFFT program [49] , [50] . The number of variants in the target sequence ( IGNCPIWVKT ) was counted . ELISA microplates ( MaxiSorp; Nunc ) were coated overnight at 4°C with goat anti-human IgG ( Jackson ImmunoResearch Laboratories ) in 0 . 05 M sodium bicarbonate buffer ( pH 8 . 6 ) . After washing with PBS including 0 . 1% Tween-20 , the wells were blocked with 0 . 5% BSA in PBS for 1 hour at 37°C . After washing again , the wells were incubated with hybridoma supernatants or control serum for 2 hours at 37°C . Following further washing , the wells were incubated with HRP-conjugated anti-human IgG1 , IgG2 , IgG3 or IgG4 ( SouthernBiotech ) for 1 hour at 37°C . The wells were washed five times followed by incubation with TMB peroxidase substrate ( KPL ) at room temperature in the dark . After 20 minutes , the reaction was stopped with 2N H2SO4 solution . The color development was read at 450 nm in an ELISA Photometer ( Biotek ELISA Reader; Biotek ) . All samples were run in triplicate . Total RNA extracted from the hybridoma using an RNeasy Mini Kit ( Qiagen ) was subjected to RT-PCR using a PrimeScript RT reagent Kit ( Takara ) with an oligo ( dT ) primer . The coding region of the H- and L-chains of HuMAb was amplified by PCR with the following primers: 5′-ATGGAGTTTGGGCTGAGCTGGGTT-3′ ( H-chain-forward ) and 5′-CTCCCGCGGCTTTGTCTTGGCATTA-3′ ( H-chain-reverse ) ; and 5′-ATGGCCTGGRYCYCMYTCYWCCTM-3′ ( L-chain-forward ) and 5′-TGGCAGCTGTAGCTTCTGTGGGACT-3′ ( L-chain-reverse ) . PCR products were ligated into pGEM-T Easy Vector ( Promega ) and their sequences were analyzed using a BigDye Terminator v3 . 1 Cycle Sequencing Kit and an ABI Prism 3100 Genetic Analyzer ( Applied Biosystems ) . Determined sequences were analyzed and compared to the NCBI database using the IgBlast softwares ( http://www . ncbi . nlm . nih . gov/igblast/ ) . The HA structure was constructed using Molecular Operating Environment software ( Chemical Computing Group Inc . ) based on the crystal structure of B/Hong Kong/8/1973 ( PDB ID: 2RFU ) [51] . All mice used were 6-week-old female BALB/c mice from Japan SLC Inc . Before infection , mice were anesthetized by intraperitoneal administration of pentobarbital sodium ( Somnopentyl; Kyoritsu Seiyaku Corporation ) . MLD50 was determined by inoculating intranasally with serial 10-fold dilutions of virus and calculating with the Reed-Muench method [52] . For passive transfer experiments , mice were treated with HuMAb at 1 , 5 , 10 or 15 mg/kg intraperitoneally at 4 , 24 , 48 or 72 hours post-infection with 25 µl mouse-adapted B/Ibaraki/2/1985 virus at a lethal dose ( 1 . 47×103 MLD50/mouse ) . Mice were weighed daily and sacrificed if they fell to 60% of starting weight . To titrate the viruses in the infected lungs , mouse-adapted B/Ibaraki/2/1985 or B/Florida/4/2006 passaged eight times in mice lungs were infected at 1 . 47×103 MLD50/mouse or 5 . 0×103 FFU/mouse , respectively . The lungs were harvested on day 3 and day 6 post-infection and virus titers in lung homogenates were determined by focus-forming assay . Data are expressed as the means ± standard errors of the means ( SEM ) . Statistical analysis was performed by Student's t test . A P value of <0 . 05 was considered significant . The GenBank ( http://www . ncbi . nlm . nih . gov/GenBank/ ) accession numbers for the genes discussed in this paper are as follows: IgG-VH in clone 5A7 ( AB729122 ) , IgG-VL in clone 5A7 ( AB729123 ) , IgG-VH in clone 3A2 ( AB729120 ) , IgG-VL in clone 3A2 ( AB729121 ) , IgG-VH in clone 10C4 ( AB729124 ) , and IgG-VL in clone 10C4 ( AB729125 ) . | Influenza virus is classified into types A , B and C . Influenza A virus is further divided into many subtypes , all of which exist in animals , indicating pandemic potential . By contrast , influenza B virus circulates almost exclusively in humans and , as there is no evidence for reassortment with influenza A virus , there is no indication of pandemic potential . Hence , there is far less accumulated research information regarding influenza B virus than influenza A virus . Influenza B virus , which is classified into two phylogenetic lineages , does , however , cause annual epidemics in humans and is therefore as essential to control as influenza A virus . Recently , the development of a universal vaccine and therapeutic strategies using human monoclonal antibodies ( HuMAbs ) has been gathering great interest . The present study reports a HuMAb neutralizing a wide range of influenza B viruses of both lineages . This HuMAb recognizes the conserved region of hemagglutinin . Moreover , therapeutic efficacy of this HuMAb was also confirmed by in vivo animal experiments . Thus , this study provides insight for the development of broad-spectrum therapeutics and a universal prophylactic vaccine against influenza B virus . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"rna",
"viruses",
"virology",
"viral",
"classification",
"biology",
"microbiology"
]
| 2013 | Human Monoclonal Antibodies Broadly Neutralizing against Influenza B Virus |
DNA demethylation mediated by the DNA glycosylase ROS1 helps determine genomic DNA methylation patterns and protects active genes from being silenced . However , little is known about the mechanism of regulation of ROS1 enzymatic activity . Using a forward genetic screen , we identified an anti-silencing ( ASI ) factor , ASI3 , the dysfunction of which causes transgene promoter hyper-methylation and silencing . Map-based cloning identified ASI3 as MET18 , a component of the cytosolic iron-sulfur cluster assembly ( CIA ) pathway . Mutation in MET18 leads to hyper-methylation at thousands of genomic loci , the majority of which overlap with hypermethylated loci identified in ros1 and ros1dml2dml3 mutants . Affinity purification followed by mass spectrometry indicated that ROS1 physically associates with MET18 and other CIA components . Yeast two-hybrid and split luciferase assays showed that ROS1 can directly interact with MET18 and another CIA component , AE7 . Site-directed mutagenesis of ROS1 indicated that the conserved iron-sulfur motif is indispensable for ROS1 enzymatic activity . Our results suggest that ROS1-mediated active DNA demethylation requires MET18-dependent transfer of the iron-sulfur cluster , highlighting an important role of the CIA pathway in epigenetic regulation .
DNA 5-cytosine methylation ( 5mC ) is an epigenetic mark that is critical for maintaining genome integrity , regulating gene expression and responding to environmental stress in many higher eukaryotes [1–6] . A proper DNA methylation pattern is important for plant development and plant responses to various stress conditions [1 , 7] . In Arabidopsis , the DNA methylation pattern is established and maintained through different pathways by several DNA methyltransferases such as MET1 , CMT3 , DRM2 and CMT2 [1 , 8] . Active removal of 5mC also contributes to the steady state of DNA methylation levels [9] . Defects in active DNA demethylation can result in abnormal gene silencing due to increased DNA methylation in gene regulatory sequences [10 , 11] . In Arabidopsis , the ROS1/DME family of DNA glycosylases/lyases initiates active DNA demethylation by excising the 5mC and then cutting the phophodiester backbone through either β- or β , δ-elimination [2 , 6 , 10 , 12–14] . The β , δ-elimination generates a 3’ phosphate at the gap , which is then hydrolyzed by phosphatase ZDP [15] , while β-elimination generates a 3’-phosphor-α , β- unsaturated aldehyde ( 3’-PUA ) , which is processed to a hydroxyl group by APE1L [16] . DNA polymerase and ligase fill the gap with an unmethylated cytosine through the base-excision repair pathway [2] . DNA demethylation by DME in the central cell contributes to parental specific expression of some imprinted genes in the endosperm , whereas ROS1 and other family members mainly affect the DNA methylation status in vegetative tissues [2] . The recruitment of ROS1 to the chromatin requires histone acetylation , which is catalyzed by IDM1 [11] . The in vivo activity of IDM1 and its targeting require other anti-silencing factors such as IDM2 , IDM3 and MBD7 [17 , 18] . In addition , ROS3 , which is an RNA-binding protein , has been suggested to mediate ROS1 recruitment at some loci because its mutation disrupts the subnuclear localization pattern of ROS1 and increases DNA methylation [19] . It is unclear , however , how the enzymatic activity of ROS1 may be regulated . Many DNA-repair and -replication proteins contain iron-sulfur ( Fe-S ) clusters as cofactors [20 , 21] , and such clusters are important for the enzymatic function and/or stability of the proteins . ROS1/DME DNA glycosylases are predicted Fe-S cluster-binding proteins since they have a conserved Fe-S-binding motif . A previous study has shown that the predicted Fe-S-binding motif is essential for DME enzymatic activity [22] . Eukaryotes have the ISC ( iron–sulfur cluster assembly ) apparatus in mitochondria , the SUF ( sulfur mobilization ) pathway in mitochondria and the CIA ( cytosolic Fe-S assembly ) pathway in cytoplasm for the assembly of Fe-S proteins in respective cellular compartments [23 , 24] . The CIA pathway is mainly responsible for the maturation of cytosolic and nuclear Fe-S proteins and depends on parts of the ISC pathway for a sulfur-containing component [23 , 24] . In the CIA pathway , two P loop NTPases , Cfd1 and Nbp35 , form a protein complex which serves as a scaffold for Fe-S cluster assembly [25 , 26] . The Fe-S cluster is assembled to its binding protein by the CIA targeting complex , which consists of NAR1 , CIA1 , AE7 , and MET18 in Arabidopsis [27] . In the complex , NAR1 shares sequence homology with Fe hydrogenases and binds to the Fe-S cluster [28] . CIA1 is a WD40 protein that functions as a scaffold for protein interactions , and AE7 is a DUF59 family protein [27 , 29] . MET18 is an ARM repeat-containing protein , which is a conserved component in the CIA pathway from yeast to plant to animal [21 , 23] . In yeast and animal cells , MET18 is the delivering protein that directly interacts with various Fe-S cluster-binding proteins [21 , 30 , 31] . Here , we identified MET18 as a component required for preventing transgene silencing as well as for normal DNA methylation patterns in Arabidopsis . We show that mutations in MET18 result in genome-wide DNA hypermethylation patterns similar to those in the ros1 mutant . We found a direct interaction between MET18 and ROS1 , which suggests that MET18 is likely the end of the CIA pathway that transfers the Fe-S cluster to its apoprotein . These results also reveal for the first time a direct connection between the CIA pathway and DNA demethylation , and thus make a significant contribution to understanding how iron-sulfur cluster assembly affects epigenetic regulation .
To search for anti-silencing factors in Arabidopsis , we previously developed a forward genetic screen system in Arabidopsis ( Wang et al . , 2013; Lei et al . , 2014 ) . In this system , the SUCROSE TRANSPORTER 2 ( SUC2 ) gene driven by the cauliflower mosaic virus 35S promoter is overexpressed in Col-0 transgenic plants ( referred to as 35S::SUC2 ) . When grown on a medium containing sucrose , 35S::SUC2 seedlings produce short roots because the roots over-accumulate sucrose ( Fig 1A ) . Both 35S::SUC2 and mutant plants grow normally on glucose-containing medium ( Fig 1A ) . We generated an ethylmethane sulfonate ( EMS ) -mutagenized population and screened for mutants with long-root phenotype on sucrose-containing medium . With this strategy , many known components associated with DNA methylation were identified , including the loss-of-function of DNA demethylase ROS1 mutants ( Fig 1A ) and also RNA-directed DNA methylation ( RdDM ) mutants [32–34] . In this study , we identified a new recessive mutant named anti-silencing 3–1 ( asi3-1 ) with a long-root phenotype ( Fig 1A ) . We first examined the expression of the SUC2 transgene by real-time quantitative RT-PCR ( qRT-PCR ) . Compared to its expression in 35S::SUC2 plants , SUC2 expression in asi3-1 plants was significantly reduced ( Fig 1B ) . 35S::SUC2 plants also express two marker genes , neomycin phosphotransferase II ( NPTII ) and hygromycin phosphotransferase II ( HPTII ) , driven by shorter forms of 35S promoter for transgenic plant selection . In contrast to 35S::SUC2 plants , which grew well on a kanamycin-containing medium , asi3-1 and ros1-13 were sensitive to kanamycin ( Fig 1C ) . qPCR results indicated that the sensitivity was due to a reduction in NPTII transgene expression ( Fig 1B ) . Moreover , HPTII expression was also down-regulated in the asi3-1 mutant ( Fig 1B ) . The observation that ASI3 dysfunction leads to down-regulation of transgenes suggested that ASI3 may function as an anti-silencing factor that prevents transgene silencing in Arabidopsis . In addition to transgene silencing , the asi3-1 mutant also displayed morphological phenotypes such as reduced plant and leaf sizes ( S1A Fig ) . Through genetic mapping , the asi3-1 mutation was mapped to a narrow region between two simple sequence length polymorphism ( SSLP ) markers located on chromosome 5: 19 , 050 , 000 and 20 , 070 , 000 ( Fig 1D ) . By whole genome re-sequencing , we detected in this region a “CAA” to “TAA” mutation in the first exon of AT5G48120 gene , which encodes the Armadillo repeat motif ( ARM ) -containing protein MET18 . This mutation causes an amino acid change from Q to a stop codon , likely leading to premature termination of translation ( Fig 1E ) . To further confirm that the mutation in AT5G48120 caused the silencing phenotype , we cloned the wild-type genomic sequence of AT5G48120 and transformed it into the asi3-1 mutant . Like the 35S::SUC2 , the transgenic plants exhibited a short-root phenotype ( Fig 1F ) . We selected the transgenic lines with resistance to Basta and assessed the expression of the recombinant protein MET18-3FLAG-3HA ( MET18-3FH ) in the T1 generation by Western blotting ( S1B Fig ) . The SUC2 transgene expression was recovered to the 35S::SUC2 level , indicating that the AT5G48120 gene complemented the transgene-silencing phenotype of the asi3-1 mutant ( Fig 1G ) . Moreover , the reduced-plant size phenotype was rescued by the introduction of MET18 genomic DNA , indicating that this developmental defect was caused by MET18 dysfunction ( S1B Fig ) . All of these results confirmed that AT5G48120 is the gene corresponding to the asi3-1 mutation . AT5G48120 ( MET18 ) encodes an ARM repeat superfamily protein , which was previously identified as a homolog of yeast Met18 [27] . Because two T-DNA insertion alleles ( met18-1 and met18-2 ) have been previously identified ( Fig 1E ) [27] , we named the asi3-1 mutant met18-3 ( Fig 1E ) . To investigate whether silencing of transgenes was associated with DNA methylation , we performed whole-genome bisulfite sequencing ( WGBS ) of 35S::SUC2 and met18-3 mutant seedlings . We examined the DNA methylation level in the 35S::SUC2 promoter region , which carries ~800 bp full-length 35S promoter different from the 35S promoters of NPTII and HPTII transgenes ( ~400bp containing two tandem repeats of 35S 3’ terminal sequence ) ( Fig 2A ) . The whole-genome methylation data showed clear increases in methylation in all cytosine contexts ( CG , CHG , and CHH ) in the met18-3 mutant compared to the 35S::SUC2 , and especially in the upstream region ( Region B ) of the 35S promoter ( Fig 2A and 2B ) . These results indicated that SUC2 transgene silencing was associated with DNA hypermethylation-mediated transcriptional repression . To confirm the DNA hypermethylation phenotype in met18-3 , DNA methylation-sensitive PCR ( Chop PCR ) was performed . The results indicated that met18-3 displays a hypermethylation phenotype in all five examined loci ( Fig 2C ) that were previously identified as ROS1-target loci [11 , 33 , 34] . As a positive control , ros1-13 also showed hypermethylation at the examined loci ( Fig 2C ) . As expected , introduction of MET18 genomic DNA rescued the DNA methylation phenotype in the met18-3 mutant , indicating that MET18 dysfunction is responsible for the hypermethylation phenotype in met18-3 . Moreover , unlike RdDM mutants in which the transcript level of ROS1 gene is reduced [33 , 34] , MET18 dysfunction did not affect ROS1 mRNA level ( S2 Fig ) , suggesting that MET18 may be a new factor critical for DNA demethylation . To further investigate the effects of MET18 dysfunction on whole genome DNA methylation level , we subjected two T-DNA insertion alleles , met18-1 and met18-2 [35] ( Fig 1E ) , for whole-genome bisulfite sequencing . Two biological replicates were sequenced for each allele ( S3 Fig ) . Using the published Col-0 WGBS data as the control [11] , we identified thousands of differentially methylated regions ( DMRs ) in the mutants . There are 4653/729 , 3884/732 , 4826/509 and 3687/730 hyper/hypo-DMRs in met18-1 replicate 1 ( rep . 1 ) , met18-1 rep . 2 , met18-2 rep . 1 and met18-2 rep . 2 , respectively ( S1 Table and Fig 3A and 3C ) . The predominance of hyper-DMRs in all 4 WGBS datasets strongly supports our hypothesis that MET18 has a negative effect on DNA methylation . We then compared the DMRs between different replicates and different mutant alleles . There are 2584 and 2406 overlapping hyper-DMRs between two technical replicates of met18-1 and met18-2 , respectively ( S4 Fig ) . The percentages of overlapping hyper-DMRs between replicates are about 56% ( met18-1 rep . 1 ) / 67% ( met18-1 rep . 2 ) and 50% ( met18-2 rep . 1 ) / 65% ( met18-2 rep . 2 ) ( S4 Fig and S1 Table ) , respectively , indicating the majority of identified hyper-DMRs are reliable . Moreover , even for hyper-DMRs that appeared to be unique to one replicate , the DNA methylation level was also increased in the other replicate , compared to Col-0 control ( S4B and S4D Fig ) , even though the increases in DNA methylation level were not enough to be considered as hyper-DMRs due to our stringent criteria . We then compared the hyper-DMRs between met18-1 and met18-2 and obtained 1254 common hyper-DMRs , which account for about 49% and 52% of total hyper-DMRs in met18-1 and met18-2 , respectively ( S4E and S5 Figs ) . Similarly , even for one allele-unique hyper-DMRs , DNA methylation level was also increased in the other allele , suggesting that more loci are overlapping between the two alleles than the Venn diagram showed ( S4F Fig ) . Therefore , we used met18-1 ( overlapping hyper-DMRs of met18-1 two replicates ) , met18-2 ( overlapping hyper-DMRs of met18-2 two replicates ) and met18 ( overlapping hyper-DMRs of met18-1 and met18-2 ) hyper-DMRs for subsequent analysis . ROS1 , DML2 and DML3 are the main 5mC DNA glycosylases/demethylases functioning in active DNA demethylation in vegetative tissues in Arabidopsis [2 , 10] . To investigate whether MET18 and ROS1 control a common set of target loci , we compared met18 DNA methylation data with two published DNA methylomes from the ROS1 mutant ros1-4 and a triple mutant of ROS1 and its two paralogs DML2 , DML3 , rdd [11 , 36] . We identified 6151 ( 5400 hyper- and 751 hypo-DMRs ) and 9319 ( 8295 hyper- and 1024 hypo-DMRs ) DMRs in ros1-4 and rdd mutants , respectively ( S1 Table and Figs 3A and S5 ) . Based on TAIR10 annotation , we analyzed the distribution of genomic features for the hyper-DMRs in met18 , ros1-4 and rdd mutants . The results show that the distribution patterns are similar among met18 , ros1-4 and rdd mutants ( Fig 3B and 3C ) . We then compared the hyper-DMRs of met18 with those of ros1-4 and rdd mutants . As expected , the majority of the hyper-DMRs in ros1-4 ( 64% ) overlapped with those in the rdd mutant ( S6 Fig ) . For the met18-1 allele , there are 55% ( 1419/2584 ) and 59% ( 1535/2984 ) of hyper-DMRs overlapping with those of ros1-4 and rdd , respectively ( Fig 4A and 4B , left panels ) . For the met18-2 allele , the percentages of overlapping hyper-DMRs with those of ros1-4 and rdd are 56% ( 1339/2412 ) and 58% ( 1400/2412 ) , respectively ( Fig 4A and 4B , left panels ) . For the overlapping hyper-DMRs between met18-1 and met18-2 ( here refer to as met18 ) , about 65% ( 814/1254 ) and 67% ( 837/1254 ) are overlapping with those of ros1-4 and rdd mutants , respectively ( S6 Fig ) . These results indicate that MET18 and ROS1 control a common set of genomic loci to prevent their DNA hypermethylation . Moreover , the box plots based on hyper-DMR overlap of met18-ros1 and met18-rdd also indicated that even at ros1 , rdd or ros1/rdd-unique loci , the average DNA methylation level was higher in met18 than in Col-0 ( Fig 4A and4B , right panels ) , implying that there are more loci affected by MET18 than the Venn diagram showed ( Fig 4A and 4B ) . Nevertheless , the partial-overlapping patterns also indicate a role of MET18 in controlling ROS1-independent , in addition to ROS1-dependent , DNA demethylation . The heatmaps based on hyper-DMR loci identified in met18 , ros1 and rdd also suggested that hyper-DMR loci in met18 were also hypermethylated in ros1-4 and rdd mutants ( S7 Fig ) . Collectively , these results indicate that hyper-DMRs in met18 highly overlap with those in ros1-4 and rdd , demonstrating that MET18 facilitates DNA demthylation mainly at ROS1-dependent loci . Proteins in the ROS1 DNA demethylase family contain a conserved Fe-S motif ( Fig 5A ) which was reported to be required for DME enzymatic activity [22] . To determine whether this motif is important for ROS1 enzymatic activity , we performed a DNA nicking assay as previously described [13] . We generated mutated forms of MBP-ROS1 fusion proteins by changing one of the four conserved cysteine to serine ( C to S ) and compared their enzymatic activities with that of the wild-type ROS1 protein . The MBP protein was used as a negative control . Coomassie Brilliant Blue ( CBB ) staining of purified proteins indicated that the mutated forms of ROS1 were properly expressed as wild-type ROS1 ( Fig 5B lower panel ) . We incubated the purified recombinant proteins with either unmethylated or methylated DNA substrates . Any significant DNA nicking activity would result in the disappearance of the fast-migrating supercoiled DNA plasmid . Neither wild-type nor any of the mutated forms of ROS1 could cut the unmethylated substrate . However , wild-type but not the mutated ROS1 could cut the CG- or CHG-methylated substrates ( Fig 5B , upper panel ) . The results show that mutating any of the four conserved cysteines reduces ROS1 activity , suggesting that all four cysteines in the Fe-S-binding motif are critical for ROS1 enzymatic activity ( Fig 5B ) . Consistently , mutations of all 4 Fe-S-binding cysteines in DME were shown to impair DME activity [22] . We performed affinity purification of the MET18 complex using inflorescence tissues of Flag-tagged MET18 transgenic plants . The co-purified proteins were identified through mass spectrometric ( MS ) analyses ( Fig 6A and S2 Table ) . Mass spectrometry showed that AE7 and CIA1 , another two components functioning together with MET18 in the CIA pathway , were co-purified with MET18 ( Fig 6A and S2 Table ) . Interestingly , ROS1 was also co-purified from MET18 transgenic plants but not from 35S::SUC2 control plant purification . These results suggest that CIA pathway components may interact with ROS1 in vivo . To test for direct interactions between ROS1 and CIA components , yeast two-hybrid ( Y2H ) assays were performed . Yeast bearing MET18 in the AD vector and AE7 in the BD vector grew well in SD media lacking leucine , tryptophan , or histidine amino acids ( SD-L/T/H ) ( Fig 6B ) , which is consistent with a previous report that MET18 directly interacts with AE7 [27] . The direct interaction of AE7 and MET18 was further confirmed by split luciferase assay in Arabidopsis protoplasts ( Fig 6C ) . Yeast bearing ROS1 in the BD vector and MET18 in AD the vector also grew in the SD-L/T/H media , demonstrating that MET18 directly interacts with ROS1 . Yeast bearing AE7 in the AD vector and ROS1 in the BD vector also grew in the SD-L/T/H media ( Fig 6C ) , suggesting AE7 can also directly bind to ROS1 . Moreover , the interactions of ROS1 with MET18 and AE7 were further confirmed by a split luciferase assay ( Fig 6C ) . These results demonstrate that ROS1 physically interacts with the CIA components MET18 and AE7 . To gain insights into the interaction domain of ROS1 with MET18 , we generated truncated forms of ROS1 , including the N terminal lysine-rich domain ( NTD ) , the HhH-GPD motif-containing GPD domain and the C terminal domain ( CTD ) [37] ( Fig 6D ) . The iron-sulfur motif is located in the overlapping region ( 100 amino acids ) of GPD and CTD domain ( Fig 6D ) . Split luciferase assay was performed to test the interactions between MET18 and the different forms of ROS1 . The results showed that , besides full length ROS1 , MET18 can also interact with ROS1GPD and ROS1CTD , but not with ROS1NTD ( Fig 6D ) . This result suggests that the iron-sulfur motif may be the critical region for MET18-ROS1 interaction .
DNA methylation and histone modifications are major epigenetic marks determining transcriptional activation or silencing of the associated genes [1 , 2] . In Arabidopsis , DNA demethylation mediated by the 5mC DNA glycosylase ROS1 prevents thousands of genomic loci from being hyper-methylated [11] and prevents transcriptional silencing at some loci [6 , 10 , 38 , 39] . In our previous searches for anti-silencing factors in a genetic screen , we identified many ros1 mutant alleles , illustrating that ROS1 is required to prevent the silencing of the 35S::SUC2 transgene [32–34] . Using the same system , we identified the asi3-1 mutant and mapped the mutation to the MET18 locus . Whole-genome bisulfite sequencing identified thousands of hyper-methylated loci in met18 mutants ( Fig 3 and S1 Table ) , indicating that MET18 is a new factor controlling DNA demethylation . The comparison of DNA methylomes illustrated that hyper-DMRs in the met18 mutant highly overlap with those identified in ros1-4 and rdd mutants ( Fig 4 ) , supporting that MET18 functions in the ROS1/DML2/DML3-mediated DNA demethylation . AtMET18 is a homolog of the yeast MET18 protein , which was identified as a component of the cytoplasmic Fe-S assembly machinery [27] . In Arabidopsis , AtMET18 forms a protein complex with other CIA components including AE7 , CIA1 , and NAR [27] . About 25 years ago , a DNA glycosylase functioning in base excision repair was found to contain an Fe-S cluster that was essential for its activity [40] . A recent study showed that mutation in the conserved Fe-S motif of DEMETER , a ROS1 paralogue in Arabidopsis , impaired the protein’s enzymatic activity in base excision repair [22] . Our site-directed mutagenesis analysis of ROS1 in a nicking assay showed that the iron-sulfur motif is also critical for ROS1 enzymatic activity ( Fig 5 ) . This result suggests that the enzymatic activity of ROS1 may require MET18-mediated Fe-S assembly . Interestingly , ROS1 was detected with other CIA components in the MET18-containing protein complex ( Fig 6A ) . Our Y2H and split luciferase assays further confirmed that MET18 directly interacts with ROS1 and AE7 ( Fig 6B and 6C ) . AE7 also directly interacted with ROS1 . Thus , MET18 and AE7 may have partially redundant functions in delivering Fe-S clusters to target proteins . It is unclear how CIA components select target apoproteins because no conserved amino acid sequence was identified in the Fe-S proteins [41] . Our results are consistent with previous reports that CIA components target specific apoproteins via physical interaction [31 , 42] . Therefore , it is important that future studies determine the structural features that enable recognition between CIA components and the target apoproteins . Biogenesis of Fe-S clusters is a multistep process that takes place in mitochondria and the cytoplasm , but how it is linked to nuclear Fe-S proteins is not known [30] . ROS1-mediated DNA demethylation is a nuclear process [10] . Our localization assay in tobacco leaves indicated that MET18 is mainly localized to the cytoplasm , although occasional localization in the nucleus was also observed ( S8 Fig ) . ROS1 functions in the nucleus [10] . Luo et al . ( 2012 ) reported that the MET18-C terminus interacts with AE7 in both the nucleus and cytoplasm in bimolecular fluorescence complementation ( BiFC ) assays [27] . It is possible that the CIA machinery may facilitate ROS1 protein maturation in the cytoplasm before ROS1 is translocated into the nucleus . Based on our results , a working model for the anti-silencing and DNA demethylation roles of MET18 can be proposed ( S9 Fig ) . In this model , MET18 and other CIA components transfer an Fe-S cluster onto one of their apoproteins , ROS1 , in the nucleus or cytoplasm . The mature ROS1 protein is then enzymatically active and recruited to the 35S::SUC2 promoter or other methylated genomic loci to remove methylated cytosines . The promoter is then demethylated , allowing for active transcription of the SUC2 transgene . The model helps to explain an earlier observation that a defect in AE7 causes DNA hypermethylation at two loci known to be ROS1 targets [27] . CIA proteins are important for the replication of DNA and the maintenance of genomic integrity [20 , 24 , 31 , 43–45] . Consistent with the importance of the CIA pathway in plants , both EMS and T-DNA insertion mutants of MET18 displayed a stunted phenotype ( S1 Fig ) , indicating that MET18 is also indispensable for normal development . Unlike AE7 , NAR1 , and CIA1 , whose homozygous mutants are lethal , the met18 mutant can grow and propagate normally , indicating that MET18 is less important than the other CIA components and there may be other proteins that are functionally redundant with MET18 . Although there are no other MET18 homologs in Arabidopsis , other proteins in the CIA pathway such as AE7 may also be capable of delivering the Fe-S cluster to some apoproteins . This notion is consistent with our finding that not only MET18 but also AE7 could interact with ROS1 ( Fig 6 ) . Therefore , it can be speculated that , without MET18 , some ROS1 protein probably can still retain Fe-S clusters and confer demethylation activity at some loci . This hypothesis can explain the observation that in some of the hyper-DMR loci met18 hypermethylation is not as strong as in the ros1 mutant . The met18ros1 double mutant displayed the same phenotype of reduced size as the met18 single mutant but the ros1 single mutant or the rdd mutant displayed a normal growth phenotype ( S1 Fig ) , indicating that MET18 participates in additional processes besides the ROS1-mediated DNA demethylation . Collectively , our study reveals that MET18 affects transgene silencing and DNA methylation by interacting with the DNA demethylase ROS1 for Fe-S delivery . Our study thereby demonstrates a direct linkage between the CIA pathway and ROS1-mediated active DNA demethylation . Through metabolites such as acetyl-CoA , SAM and SAH , which are important in histone and DNA modifications , the state of cellular metabolism is intimately connected to epigenetic modifications [46] . Knowledge of such connections is essential for our understanding of epigenetic regulation and how nutrition and metabolism affect development and diseases through epigenetic regulation [47] . In this regard , the linkage between the CIA pathway and active DNA demethylation represents an important but previously underappreciated connection between epigenetics and nutrition and metabolism .
All plants were grown under a long-day photoperiod ( 16-h light/8-h dark ) . 1/2-strength Murashige and Skoog ( MS ) medium containing 1% sucrose was used for studying root phenotype . All other experiments used growth medium with 1% glucose . Kanamycin of 100 mg/L was used for the kanamycin-sensitivity assay . Seedlings were photographed 14 days after germination . Two other alleles of the met18 mutant , met18-1 ( SALK_121963 ) and met18-2 ( SALK_147068 ) , were obtained from the Arabidopsis Biological Resource Center ( ABRC ) and confirmed by PCR-based genotyping [35] . MET18 genomic DNA including the upstream 2-kb promoter region was amplified by PCR and cloned into a modified pEarleyGate 302 vector [48] , which bears three copies of FLAG and HA tags . The construct was introduced into the met18-3 mutant by Agrobacterium-mediated transformation via the floral dip method [49] All primers used in the study are listed in S3 Table . The met18-3 mutant was obtained by a screen method described in our previous report [33 , 34] . In brief , EMS-treated M2 seeds were grown on ½ MS medium containing 1% sucrose . Seedlings with roots longer than the wild type plants were selected as potential transgene silencing mutants . To clone the ASI3 gene , the asi3-1 mutant was crossed with Landsberg erecta , and F1 plants were self-pollinated to obtain the F2 population . F2 seeds were planted on 1/2 MS medium containing 1% sucrose and 25 μg/mL hygromycin . Seedlings with normal , long roots were selected for genetic mapping in order to calculate asi3-1 linkage . The asi3-1 mutant genomic DNA was then re-sequenced to determine the location of the mutation in the mapping region . DNA methylation-sensitive PCR was performed according to a previous report [34] . In brief , genomic DNA was extracted from 2-week-old seedlings using the DNeasy Plant Minikit ( QIAGEN ) and was quantified with NanoDrop 2000c UV-Vis Spectrophotometer ( Thermo SCIENTIFIC ) . About 50 ng of DNA was subjected to DNA methylation-sensitive restriction enzyme digestion in a 50-μL reaction . The digested DNA was used as template for PCR , and the PCR products were subjected to agarose gel electrophoresis . The non-digested genomic DNA was used as control template for PCR . For whole-genome bisulfite sequencing , DNA was extracted from 2-week-old seedlings . The DNA samples were submitted to BGI ( Shenzhen , China ) for bisulfite treatment and DNA sequencing . The bisulfite treatment and DNA methylation analysis were performed as previously described [50] . When counting the number of overlapping DMRs , if aregion in one mutant overlaps with several regions in another mutant , we calculated the number of overlapping regions from the perspective of met18 , ros1-4 , and rdd . For TE annotation , if a region overlaps with a TE or a transposable elemental gene , the region is classified as a TE region . Regions having no overlap with TE or transposable element gene and overlapping with protein-coding genes are classified as Gene regions . For real-time quantitative RT-PCR , total RNAs were extracted from 2-week-old seedlings using the RNeasy Plant Minikit ( QIAGEN ) . After TURBO DNase I treatment ( Ambion ) , 2 μg of RNA was subjected to RT reaction using the SuperScript III First-Strand Kit according to the manufacturer’s instructions ( Invitrogen ) . The 1st-strand cDNAs were then amplified using IQ SYBR green supermix ( BIO-RAD ) with the CFX96 real-time PCR detection system ( BIO-RAD ) . The MBP-ROS1 in the pMAL-c2x vector was constructed as previously reported [6 , 51] . Site-directed mutagenesis was performed via PCR to obtain the expression constructs of ROS1C1038S- , ROS1C1045S- , ROS1C1048S- , and ROS1C1054S-mutant proteins . All constructs were transformed into dcm- Codon Plus cells of E . coli strain BL21 . The induction and purification of ROS1 and mutant proteins were performed as previously described [15] . The ROS1 enzymatic activity assay ( the nicking assay ) was performed as previously described [13] . In brief , plasmid pBluescript KS purified from the dcm- strain of E . coli BL21 ( DE3 ) was methylated in vitro with MspI ( mCHG ) or SssI ( mCG ) methylases ( NEB ) . The methylation status was confirmed by digestion with MspI and HpaII restriction endonucleases . The unmethylated plasmid was used as a control . The purified , closed-circular plasmids ( 250 ng ) were then incubated with purified MBP-ROS1 protein or its mutated forms in nicking buffer containing 40 mM Hepes-KOH ( pH 8 . 0 ) , 0 . 1 M KCl , 0 . 5 mM EDTA , 0 . 5 mM DTT , and 0 . 2 mg/ml BSA at 37°C for 2 h . After adding stop solution ( 0 . 4 M EDTA and 1% SDS ) , the reaction mixtures were heated at 70°C for 5 min and subjected to 1% agarose gel electrophoresis . The full-length coding regions of MET18 , ROS1 , and AE7 were first individually cloned into the pENTR/D-TOPO directional cloning vector ( Invitrogen ) and then transferred into the destination vectors pDEST 22 ( AD ) and pDEST 32 ( BD ) ( Invitrogen ) via LR recombination using Gateway Clonase II Enzyme ( Invitrogen ) . For the split luciferase assay , constructs carrying full-length coding regions of MET18 , AE7 , and full-length/truncated forms of ROS1 fused with split luciferase in the pEarleyGate vector were co-transfected into Arabidopsis protoplast or Nicotiana Benthamiana for overnight incubation or 3-day-growth , respectively [48] . The empty vectors was used as a negative control . The luciferase activity was determined using CCD camera equipped with Winview software ( Princeton instruments ) . Protoplast was prepared as previously reported [52] . For MET18-GFP cellular localization , Agrobacterium containing MET18-GFP construct driven by 35S promoter was infiltrated into Nicotiana benthamiana leaves . The cellular localization was examined by Zeiss Microscope at 3 day-post-infiltration . For affinity purification of MET18 and its associated proteins , 5 g of flower tissues were collected from MET18-3FLAG-3HA transgenic plants , and tissue from 35S::SUC2 plants was used as a negative control . Extraction of total proteins and affinity purification were performed as described previously [53] with minor modifications . Briefly , flower tissues were ground to fine powders in liquid nitrogen and suspended in 30 ml of lysis buffer ( 50 mM Tris pH 7 . 6 , 150 mM NaCL , 5 mM MgCl2 , 10% glycerol , 0 . 1% NP-40 , 0 . 5 mM DTT , 1 mM PMSF , and 1 protease inhibitor cocktail tablet ( Roche ) ) . After further homogenization by douncing and centrifugation at 4°C , the supernatants were incubated with 120 μL of anti-FLAG M2 agarose beads ( Sigma ) , which had been pre-equilibrated with lysis buffer . After incubation at 4°C with rotation for 2–3 hours , the agarose beads were washed three times for 5 minutes each time with 40 ml of lysis buffer , three times for 5 minutes each time with 1 mL of lysis buffer , and three times for 5 minutes each time with 1 ml of PBS buffer . The agarose beads were finally resuspended in 120 μL of PBS buffer for mass spectrometry according to a previous report [54] . The raw WGBS sequencing dataset of met18-1 rep . 1 , met18-1 rep . 2 , met18-2 rep . 1 and met18-2 rep . 2 was deposited to the public database NCBI GEO ( accession number GSE69281 ) . The Col-0 , ros1-4 and rdd whole genome bisulfite sequencing data were from GEO accession GSE33071 [11] . | DNA cytosine methylation is a major epigenetic mark that confers transcriptional regulation . Active removal of DNA methylation is important for plants and mammals during development and in responses to various stress conditions . In the model plant species Arabidopsis thaliana , active DNA demethylation depends on a family of 5-methylcytosine DNA glycosylases/demethylases including ROS1 , DME , and others . While the epigenetic function of this demethylase family is well-known , little is known about how their enzymatic activities may be regulated . In this report , we carried out a forward genetic screen for anti-silencing factors and identified MET18 , a conserved component of cytosolic iron-sulfur cluster assembly ( CIA ) pathway in eukaryotes , as being required for the ROS1-dependent active DNA demethylation . Dysfunction of MET18 causes DNA hyper-methylation at thousands of genomic loci where DNA methylation is pruned by ROS1 . In addition , ROS1 physically interacts with MET18 and other CIA pathway components; while a conserved iron-sulfur-binding motif is indispensable for ROS1 enzyme activity . Our results suggested that MET18 affects DNA demethylation by influencing ROS1 enzymatic activity via direct interaction with the iron-sulfur-binding motif of ROS1 , highlighting a direct connection between iron-sulfur cluster assembly and active DNA demethylation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | MET18 Connects the Cytosolic Iron-Sulfur Cluster Assembly Pathway to Active DNA Demethylation in Arabidopsis |
The mouse organ of Corti , housed inside the cochlea , contains hair cells and supporting cells that transduce sound into electrical signals . These cells develop in two main steps: progenitor specification followed by differentiation . Fibroblast Growth Factor ( FGF ) signaling is important in this developmental pathway , as deletion of FGF receptor 1 ( Fgfr1 ) or its ligand , Fgf20 , leads to the loss of hair cells and supporting cells from the organ of Corti . However , whether FGF20-FGFR1 signaling is required during specification or differentiation , and how it interacts with the transcription factor Sox2 , also important for hair cell and supporting cell development , has been a topic of debate . Here , we show that while FGF20-FGFR1 signaling functions during progenitor differentiation , FGFR1 has an FGF20-independent , Sox2-dependent role in specification . We also show that a combination of reduction in Sox2 expression and Fgf20 deletion recapitulates the Fgfr1-deletion phenotype . Furthermore , we uncovered a strong genetic interaction between Sox2 and Fgf20 , especially in regulating the development of hair cells and supporting cells towards the basal end and the outer compartment of the cochlea . To explain this genetic interaction and its effects on the basal end of the cochlea , we provide evidence that decreased Sox2 expression delays specification , which begins at the apex of the cochlea and progresses towards the base , while Fgf20-deletion results in premature onset of differentiation , which begins near the base of the cochlea and progresses towards the apex . Thereby , Sox2 and Fgf20 interact to ensure that specification occurs before differentiation towards the cochlear base . These findings reveal an intricate developmental program regulating organ of Corti development along the basal-apical axis of the cochlea .
The inner ear contains six sensory organs required for the senses of hearing and balance . The cochlea , a snail-like coiled duct , is the auditory organ . It contains specialized sensory epithelia , called the organ of Corti , composed of hair cells ( HCs ) and supporting cells ( SCs ) . In mammals , this sensory epithelium is elegantly patterned , with one row of inner hair cells ( IHCs ) and three rows of outer hair cells ( OHCs ) , separated by two rows of pillar cells forming the tunnel of Corti . Each row of OHCs is associated with a row of supporting cells called Deiters’ cells . Here , we refer to pillar cells and Deiters’ cells collectively as SCs . Organ of Corti development has been described as occurring in two main steps: prosensory specification and differentiation [1] . During prosensory specification , proliferative progenitors at the floor of the developing cochlear duct are specified and then exit the cell cycle to form the postmitotic prosensory domain . Here , we define specification to be a process that makes progenitors competent to differentiate . We also use cell cycle exit as a marker for specified cells in the prosensory domain ( prosensory cells ) . During differentiation , prosensory cells differentiate into both HCs and SCs [2] . Interestingly , cell cycle exit , marking the completion of specification , and initiation of differentiation occur in waves that travel in opposite directions along the length of the cochlear duct . At around embryonic day 12 . 5 ( E12 . 5 ) in the mouse , progenitors begin to exit the cell cycle and express the cyclin-dependent kinase inhibitor CDKN1B ( p27Kip1 ) in a wave that begins at the apex of the cochlea ( the cochlear tip ) and reaches the base of the cochlea by around E14 . 5 [3 , 4] . Afterwards , the specified prosensory cells begin differentiating into HCs and SCs in a second wave that begins at the mid-base at around E13 . 5 , and spreads quickly to the rest of the base and to the apex over the next few days [1] . Thus , while prosensory specification occurs in an apical-to-basal gradient , differentiation occurs in a basal-to-apical gradient . Notably , while the basal end of the cochlear duct differentiates immediately after prosensory specification , the apical end has a longer time between specification and differentiation , providing a larger “temporal buffer” for apical development . The spiral ganglion , containing neurons that synapse with HCs , has been shown to be important for this delay in apical differentiation , via inhibitory Sonic Hedgehog ( SHH ) signaling [5–8] . The transcription factor Sox2 is one of the earliest markers of prosensory cells [9 , 10] . Mice with specific Sox2 hypomorphic mutations that affect inner ear expression have hearing impairment due to decreased HC and SC number , while mice with inner ear-specific Sox2 null mutations are completely deaf and have no HCs or SCs [11 , 12] . Genetic experiments show that Sox2 is both necessary and sufficient for prosensory specification . Absence of Sox2 expression leads to the loss of Cdkn1b expression at E14 , a marker for the prosensory domain [12] , while ectopic Sox2 expression in cochlear nonsensory epithelium can induce ectopic sensory patches [13–15] . The Fibroblast Growth Factor ( FGF ) signaling pathway also plays vital roles in organ of Corti development [16] . Studies utilizing cochlear explants showed that inhibition of FGF signaling prior to and during stages of HC and SC differentiation results in decreased HC and SC number [17] . Signaling through FGF receptor 1 ( FGFR1 ) , in particular , is essential during this process . Conditional deletion of Fgfr1 ( Fgfr1-CKO ) in the developing cochlear epithelium resulted in dramatically reduced HC and SC number [18–20] . This has been attributed to decreased Sox2 expression in the prosensory domain of Fgfr1-CKO mice , leading to a defect in prosensory specification [19] . FGF20 has been hypothesized to be the FGFR1 ligand during organ of Corti development . Both in vitro inhibition of FGF20 with an anti-FGF20 antibody [17] and in vivo knockout of Fgf20 ( Fgf20-KO ) [21] led to decreased HC and SC number , similar to the Fgfr1-CKO phenotype . However , the Fgf20-KO phenotype is clearly not as severe as that of Fgfr1-CKO . Almost all OHCs and some IHCs are missing in Fgfr1-CKO mice [19] , while only 2/3 of OHCs are missing in Fgf20-KO mice , without any loss of IHCs [21] . This suggests that another FGF ligand may be redundant with and compensating for the loss of FGF20 , the identity of which is currently unknown . Another difference between Fgfr1-CKO and Fgf20-KO mice is the proposed mechanism accounting for the decrease in HCs and SCs . Interestingly , unlike in Fgfr1-CKO mice , Sox2 expression in the prosensory domain is not disrupted in Fgf20-KO mice [19 , 21] . Rather , FGF20 seems to function during HC and SC differentiation . These differences between the Fgfr1-CKO and Fgf20-KO phenotypes and their relationship with Sox2 suggest that FGF20/FGFR1 signaling has a more complex and as yet unexplained role during organ of Corti development . Here , we hypothesize that FGFR1 signaling has functions in both steps of organ of Corti development: an earlier role in prosensory specification that involves Sox2 , and a later role in the initiation of differentiation . We provide evidence that FGF20 regulates differentiation but not specification . Moreover , while Fgfr1 functions upstream of Sox2 , Fgf20 is downstream of Sox2 . We further show that Sox2 and Fgf20 genetically interact during organ of Corti development . Interestingly , downregulation of both genes leads to the loss of HCs and SCs preferentially towards the outer compartment and the basal end of the cochlear duct . To explain the more severe basal phenotype , we provide evidence that Sox2 regulates the timing of prosensory specification , while Fgf20 regulates the timing of differentiation . As these two steps occur along a developmental pathway , we hypothesize that prosensory specification must occur prior to differentiation . In Sox2 hypomorphic mice , prosensory specification is delayed , while in Fgf20-KO mice , the onset of differentiation occurs prematurely . When combined , these two defects led to differentiation attempting to initiate prior to the completion of specification towards the basal end of the cochlear duct . These results define unique functions of and complex interactions among FGF20 , FGFR1 , and Sox2 during organ of Corti development and highlight the potential importance of the timing of specification and differentiation along different regions of the cochlear duct .
Previous studies showed that deletion of Fgf20 leads to a loss of two thirds of OHCs in the mouse organ of Corti [21] , while conditional deletion of Fgfr1 from the cochlear epithelium leads to a loss of almost all OHCs and some IHCs [19 , 20] . To rule out the effect of genetic background accounting for these differences , we generated Fgf20 knockout ( Fgf20-KO: Fgf20-/- ) and Fgfr1 conditional knockout ( Fgfr1-CKO: Foxg1Cre/+; Fgfr1flox/- ) mice along with littermate controls ( Fgf20+/- for Fgf20-KO and Fgfr1flox/+ , Fgfr1flox/- , and Foxg1Cre/+;Fgfr1flox/+ for Fgfr1-CKO ) on a mixed C57BL/6J and 129X1/SvJ genetic background . Fgf20-KO and Fgfr1-CKO mice were generated in separate matings; therefore , some genetic background differences could persist . Foxg1Cre targets most of the otic vesicle as early as E9 . 5 [22] and has been used in other studies to conditionally delete Fgfr1 [18–20] . In the Fgf20- allele , exon 1 of Fgf20 is replaced by a sequence encoding a GFP-Cre fusion protein [18] . We also refer to this null allele as Fgf20Cre . We examined the cochleae at P0 ( Fig 1A and 1B ) and quantified the length of the cochlear duct and the total number of IHCs , OHCs , and SCs ( Fig 1C–1F ) , as well as the number of cells along the basal , middle , and apical turns of the cochlear duct ( S1A–S1C Fig ) . Refer to Fig 1G for the positions of basal , middle , and apical turns along the cochlear duct . We identified HCs based on Phalloidin labeling and SCs based on Prox1/Sox2 labeling . IHCs and OHCs were distinguished based on location relative to p75NTR-labeled inner pillar cells ( IHCs are neural , or towards the center of the coiled duct; OHCs are abneural ) . In both Fgf20-KO and Fgfr1-CKO cochleae , there were gaps in the sensory epithelium that lacked HCs and SCs along the entire cochlear duct . Quantitatively , Fgf20-KO cochleae had a 6% reduction in cochlear length compared to control ( Fgf20+/- ) cochleae , while Fgfr1-CKO cochleae had a 28% reduction compared to control ( Fgfr1Cre/+;Fgfr1flox/+ ) . Fgf20-KO did not have a significant reduction in the number of IHCs , while Fgfr1-CKO cochleae had a 40% reduction . Fgf20-KO cochleae only had a 76% reduction in the number of OHCs , while Fgfr1-CKO cochleae had almost a complete lack of OHCs , a 97% reduction . For SCs , Fgf20-KO cochleae had a 59% reduction , while Fgfr1-CKO cochleae had an 84% reduction . These patterns persisted when HC and SC numbers were normalized to cochlear length . These results were all consistent with previous studies [19 , 21] and showed that the Fgfr1-CKO phenotype is more severe than the Fgf20-KO phenotype in cochlear length and in the number of HCs and SCs . We hypothesize that during organ of Corti development , there is an additional FGFR1 ligand that is partially redundant with FGF20 . Notably , while the total number of IHCs was decreased in Fgfr1-CKO cochleae , the decrease was only observed in the basal and middle turns of the cochlea , not in the apical turn ( S1A Fig ) . In addition , the number of IHCs normalized to cochlear length was slightly increased in Fgf20-KO cochleae ( Fig 1D ) , and this increase was only prominent in the middle and apical turns of the cochlea , but not in the basal turn ( S1A Fig ) . The increase in IHCs could be explained by the shortened cochlear duct length in Fgf20-KO mice . No such basal/middle/apical turn discrepancies existed in the number of OHCs or SCs in either genotype ( S1B and S1C Fig ) . Our previous studies also noted that the apical tip of Fgf20-KO cochleae has delayed differentiation relative to control at E16 . 5 and P0 , but catches up by P7 [21] . We confirmed this result , finding that at P0 in control cochleae , sensory epithelium at the apical tip has begun to differentiate , based on phalloidin and p75NTR expression , while in Fgf20-KO cochleae , there was no sign of differentiation at the apical tip . There was a similar delay in differentiation at the apical tip of Fgfr1-CKO cochleae relative to control ( S1E Fig ) . Refer to S1D Fig for the location of the apical tip . Next , we examined Sox2 expression in Fgf20-KO and Fgfr1-CKO cochleae at E14 . 5 by RNA in situ hybridization and immunofluorescence . In control cochleae , Sox2 mRNA and protein were highly expressed in the prosensory domain ( Fig 2A , refer to Fig 2C ) . The expression of Sox2 was not changed in Fgf20-KO cochleae compared to control; however , it was noticeably decreased in Fgfr1-CKO cochleae ( Fig 2A ) , in agreement with previous findings [19–21] . This indicates that FGFR1 has an additional role , independent of FGF20 , in regulating Sox2 , which is required for prosensory specification [12] . Similar to Sox2 , CDKN1B expression in the prosensory domain is also regulated by FGFR1 , but not by FGF20 [18 , 19 , 21] . We confirmed these results , finding that while CDKN1B expression was not changed in Fgf20-KO cochleae at E14 . 5 relative to control , it was dramatically downregulated in Fgfr1-CKO cochleae ( Fig 2B ) . This is consistent with the role of Sox2 in regulating CDKN1B expression [12] . We hypothesize that a yet unidentified FGF ligand ( in addition to or independent of FGF20 ) signaling via FGFR1 regulates Sox2 expression ( and therefore CDKN1B expression ) during prosensory specification , while FGF20 signaling via FGFR1 regulates differentiation ( Fig 2D ) . We also wanted to confirm that FGF20 signals to epithelial FGFR1 at around the initiation of differentiation . To do so , we examined the expression of Etv4 ( also known as Pea3 ) and Etv5 ( also known as Erm ) , two well-established downstream effectors of FGF signaling [23] , by in situ hybridization . The expression of these two genes are downregulated with FGF signaling inhibition in E14 cochlear explants [17] . At E14 . 5 , there were two domains of Etv4 and Etv5 expression in control cochleae: the prosensory domain and the outer sulcus ( S2A Fig , brackets ) . The outer sulcus is the region of the cochlear epithelium abneural to the prosensory domain at E14 . 5 . In Fgf20-KO cochleae , expression of both genes was not detected in the prosensory domain . In Fgfr1-CKO cochleae , expression of both genes was similarly not detected in the prosensory domain . Expression of Etv4 and Etv5 in the outer sulcus was not affected in Fgf20-KO and Fgfr1-CKO cochleae ( S2A Fig ) . These results confirm that FGF20 signals through epithelial FGFR1 in the prosensory domain . Previous studies have also reported a decrease in proliferation in Kölliker’s organ ( neural to the prosensory domain , S2B Fig ) in Fgfr1-CKO cochleae [20] . We replicated this result by examining EdU ( 5-ethynyl-2’-deoxyuridine ) incorporation at E14 . 5 . Fgfr1-CKO mice had a complete lack of EdU-incorporating Kölliker’s organ cells , while Fgf20-KO mice did not show a decrease in EdU incorporation ( S2B Fig ) . This finding is also consistent with an additional FGF ligand signaling via FGFR1 , likely at an earlier stage . We do not know whether the proliferation defect in Kölliker’s organ contributes to the reduction in HC and SC number in Fgfr1-CKO mice . We have previously shown that recombinant FGF9 , which is biochemically similar to FGF20 with similar receptor binding specificity [23 , 24] , is able to rescue the loss of HCs and SCs in Fgf20-KO explant cochleae [21] . Interestingly , while treatment with FGF9 at E13 . 5 and E14 . 5 was able to rescue the Fgf20-KO phenotype , treatment at E15 . 5 was not . This temporal rescue specificity suggests that FGF20 signaling is required for the initiation of HC and SC differentiation . To confirm the hypothesis that FGF20 is involved in differentiation and not specification ( Fig 2D ) , we sought to more accurately determine the temporal requirement of FGF20 signaling . To achieve this , we developed an in vivo genetic rescue model of the Fgf20-KO phenotype by ectopically expressing FGF9 . We decided to use FGF9 again as we have already developed a system for in vivo FGF9 expression . We combined Fgf20Cre with the Fgf20βgal [21] , ROSArtTA [25] and TRE-Fgf9-IRES-eGfp [26] alleles to generate Fgf9-rescue ( Fgf20Cre/βgal;ROSArtTA/+;TRE-Fgf9-IRES-eGfp ) mice along with littermate controls: Fgf20-het ( Fgf20Cre/+;ROSArtTA/+ ) , Fgf9-OA ( Fgf20Cre/+;ROSArtTA/+;TRE-Fgf9-IRES-eGfp ) , and Fgf20-null ( Fgf20Cre/βgal;ROSArtTA/+ ) . These mice express the reverse tetracycline transactivator ( rtTA ) in the Fgf20Cre lineage , which contains the prosensory domain and Kölliker’s organ at E13 . 5 to E15 . 5 [18] . In mice expressing TRE-Fgf9-IRES-eGfp , rtTA drives the expression of FGF9 upon doxycycline ( Dox ) induction . The Fgf20βgal allele is another Fgf20-null allele , in which exon 1 of Fgf20 is replaced by a sequence encoding β-galactosidase . We combined Fgf20Cre with Fgf20βgal to generate homozygous mutant mice while maintaining a constant dosage of Fgf20Cre in control and knockout mice . Initially , pregnant dams were fed a Dox diet from E13 . 5 to E15 . 5 and pups were harvested at P0 to examine HC and SC development . As expected , Dox treatment itself did not appear to affect HC or SC development in Fgf20-het and Fgf20-null cochleae , both of which showed the expected phenotypes ( Fig 3A and 3B ) . Ectopic expression of FGF9 during these stages also did not affect HC or SC development in Fgf9-OA cochleae , showing that excess FGF20/FGF9 was not sufficient to produce ectopic HCs and SCs . Importantly , ectopic expression of FGF9 resulted in a full rescue of the number and patterning of HCs and SCs in Fgf9-rescue pups . The organ of Corti in these rescue pups had one row of IHCs , three rows of OHCs , and five rows of SCs throughout the entire length of the cochlear duct , without any gaps ( Fig 3A and 3B ) . This shows that FGF20/FGF9 signaling at E13 . 5-E15 . 5 is sufficient for HC and SC differentiation . The quantified results from all of the rescue experiments are summarized in Fig 3C , where the number of OHCs and SCs are represented as a percentage of that of Fgf20-het mice treated with the same Dox regimen . All of the quantified data are presented in S3 Fig . To more precisely determine the timing of rescue sufficiency , we fed pregnant dams Dox for a period of 24 hours starting at E13 . 5 , E14 . 5 , or E15 . 5 ( see S3 Fig for schematic of Dox regimens ) . With E13 . 5 Dox , patterning and OHC number in the basal turn of the cochlea were completely rescued in Fgf9-rescue mice ( Fig 3A ) . However , OHC number in the middle and particularly the apical turns were only partially rescued , resulting in regions with two rows of OHCs instead of three . For instance , in the apical turn , OHC number was restored to 81% of Fgf20-het mice , which is statistically significantly increased compared to Fgf20-null , but also statistically significantly decreased compared to Fgf20-het , indicating partial rescue ( Fig 3C ) . With E14 . 5 Dox , patterning and OHC number in the middle and apical turns were completely rescued . However , OHC number in the basal turn was not completely rescued , with regions of one or two rows of OHCs , instead of three . With E15 . 5 Dox , patterning and OHC number was not rescued in the basal and middle turns , as gaps still formed between islands of HCs ( Fig 3A ) . However , OHC number in the apical turn was partially rescued , with two or three rows of OHCs not separated by gaps towards the tip of the apex . In all of these experiments , the rescue of SCs followed the same pattern as that of OHCs ( Fig 3B ) . These rescue results show that FGF20/FGF9 is sufficient for OHC and SC differentiation in the basal turn of the cochlea at E13 . 5 , in the middle and apical turns at E14 . 5-E15 . 5 , and in the tip of the apical turn at E15 . 5 . Since the initiation of HC and SC differentiation occurs in the base/mid-base of the cochlea at E13 . 5 and progresses apically over the next few days , these results strongly imply that FGF20 functions during the initiation of differentiation , rather than prosensory specification , consistent with our model ( Fig 2D ) . Our results and previous findings suggest that FGFR1 regulates prosensory specification via Sox2 [19] . Mice with an inner ear-specific Sox2 hypomorphic mutation ( Sox2Ysb/Ysb , see below ) have defects in prosensory specification , accounting for a small loss of HCs and SCs , whereas mice with inner-ear specific Sox2 null mutations have a complete lack of prosensory specification and a complete absence of sensory epithelium [12] . To examine how much the reduction in Sox2 expression in Fgfr1-CKO cochlea contributes to the phenotype at P0 , we combined the Sox2- ( Sox2 constitutive null ) and Sox2Ysb alleles to closely examine the effects of reduction in Sox2 expression on organ of Corti development , on a similar genetic background as our Fgf20-KO and Fgfr1-CKO mice . We hypothesized that if Fgfr1 acts upstream of Sox2 , then reducing Sox2 expression should at least partially recapitulate the Fgfr1-CKO cochlea phenotype . The Sox2Ysb allele is a regulatory mutant in which transgene insertion in chromosome 3 disrupts some otic enhancers , resulting in hypomorphic Sox2 expression in the inner ear [11 , 12] . We generated a Sox2 allelic series of mice with the following genotypes , in order of highest to lowest levels of Sox2 expression: Sox2+/+ ( wildtype ) , Sox2Ysb/+ , Sox2Ysb/Ysb , and Sox2Ysb/- . In this allelic series , decrease in Sox2 expression had a dose-dependent effect on cochlea length at P0 ( Fig 4A–4C ) . Sox2Ysb/+ cochleae had a 6% reduction in length compared to wildtype ( although not statistically significant ) , Sox2Ysb/Ysb cochleae had a 24% reduction , and Sox2Ysb/- had a 46% reduction . Sox2Ysb/+ organ of Corti developed relatively normally , with three rows of OHCs and one row of IHCs ( Fig 4A ) . Interestingly , there were occasional ectopic IHCs neural ( inner ) to the normal row of IHCs , especially in the middle and apical turns of the Sox2Ysb/+ cochlea ( Fig 4A , arrowheads ) . However , there was no significant increase in IHC number ( total or normalized to length ) compared to wildtype cochleae ( Fig 4D ) . The Sox2Ysb/Ysb cochlea appeared much more abnormal , with gaps in the sensory epithelium that lacked HCs and SCs in the basal turn ( Fig 4A and 4B ) , similar to what was observed previously [12] . Moreover , at the base , in the sensory islands between the gaps , there were often four rows of OHCs and six rows of SCs . In the middle and apical turns , there were the normal three rows of OHCs and five rows of SCs . There were also numerous ectopic IHCs throughout the middle and apical turns , sometimes forming an entire second row of cells ( Fig 4A ) , resulting in increased number of IHCs in the middle turn compared to wildtype ( S4A Fig ) . However , the total and length-normalized number of IHCs in Sox2Ysb/Ysb cochleae did not significantly differ from that of wildtype cochleae ( Fig 4D ) . In terms of OHCs , Sox2Ysb/Ysb cochleae exhibited a 40% decrease in total number compared to wildtype cochleae ( Fig 4E ) . This decrease was not quite as severe when normalized to cochlear length ( 21% decrease ) . Strikingly , Sox2Ysb/- cochleae lacked almost all HCs and SCs , except in the apical turn ( Fig 4A and 4B ) . The decrease in OHC number ( 93% ) in Sox2Ysb/- cochleae compared to wildtype was more severe than the decrease in IHC number ( 75% ) . Notably , IHC number was significantly decreased in the basal and middle turns , but not in the apical turn ( S4A Fig ) . OHC number was significantly decreased throughout all three turns ( S4B Fig ) . In all of these genotypes , the number of SCs followed the pattern of loss of OHCs ( Fig 4F and S4C Fig ) . Interestingly , while Sox2Ysb/- cochleae almost completely lacked HCs and SCs in the basal and middle turns , in 7 of 11 Sox2Ysb/- cochleae examined , one or two small islands of HCs or SCs were found at the basal tip ( S4D Fig ) . Overall , these results showed that the basal end of the cochlea is more sensitive to the loss of Sox2 expression than the apical end . Furthermore , while both IHCs and OHCs were affected , OHCs were more sensitive to decrease in Sox2 expression than IHCs . Importantly , both of these features were found in Fgfr1-CKO cochleae , where the decrease in IHCs was only found in the basal and middle turns and there were almost no OHCs along the entire cochlear duct ( S1A and S1B Fig ) . Therefore , we conclude that decrease in Sox2 expression , leading to defects in prosensory specification , could account for the Fgfr1-CKO phenotype . Furthermore , the decrease in Sox2 expression could also account for the difference in severity between the Fgf20-KO and Fgfr1-CKO phenotypes , since Fgf20-KO cochleae , which had normal Sox2 expression , did not have a decrease in the number of IHCs , unlike Fgfr1-CKO and Sox2Ysb/- cochleae . We sought to determine why a decrease in Sox2 expression more severely affected the basal end of the cochlear duct . Initially , we examined Sox2 expression at E14 . 5 . As expected , Sox2 expression was almost completely absent in Sox2Ysb/- cochleae ( S5A Fig ) . This decrease in expression was not more severe at the basal turn of the cochlear duct , relative to the middle and apical turns , suggesting that the more severe basal phenotype in Sox2Ysb/- cochleae cannot be explained by differential Sox2 expression . Similarly , CDKN1B expression was downregulated in the prosensory domain of Sox2Ysb/- cochleae , consistent with previous studies [12] . Interestingly , the decrease in expression was also not more severe at the basal turn relative to the middle and apical turns ( S5B Fig ) . Using CDKN1B as a marker of prosensory specification , this suggests that the more severe basal phenotype also cannot be explained by differential regulation of prosensory specification along the length of the cochlea . As described in the introduction , the wave of cell cycle exit ( marking the completion of prosensory specification ) and the wave of differentiation travel in opposite directions along the cochlear duct during development , resulting in the basal end of the cochlear duct differentiating immediately after specification . The apical end , meanwhile , exhibits a delay in differentiation , resulting in a longer temporal buffer between specification and differentiation . In this developmental pathway , specification must be completed prior to the initiation of differentiation . We reasoned , therefore , that disruptions to the timing of prosensory specification will preferentially interfere with basal sensory epithelia development , potentially accounting for the more severe basal phenotype in Sox2 hypomorphs . Notably , Sox2 expression in the prosensory domain has recently been shown to follow an apical-to-basal pattern , suggesting that Sox2 may play a role in the apical-to-basal progression of cell cycle exit and the completion of prosensory specification [27] . To test this hypothesis , we examined cell cycle exit in the prosensory domain via Ki67 expression , as a marker of the status of prosensory specification . Ki67 is expressed by cycling cells , but not cells in the G0 phase of the cell cycle [28] . In the developing cochlea at around E12 . 5 to E15 . 5 , cells of the prosensory domain , sometimes referred to as the zone of non-proliferation , have turned off or are beginning to turn off Ki67 expression as they exit the cell cycle [3] . At E14 . 5 in Sox2Ysb/+ cochleae , the prosensory domain along most of the cochlear duct ( serial sections 2–6 ) has turned off Ki67 expression , except at the very base ( serial section 1; Fig 5A , brackets ) . See graphical summary below Fig 5A; also see S5C Fig for serial “mid-modiolar” sections through the cochlea . This indicates that the wave of cell cycle exit , which starts at the apex , has reached the very base of the cochlear duct . However , in Sox2Ysb/- cochleae , only the prosensory domain at the apical turn of the cochlear duct ( serial section 6 ) has turned off Ki67 , not at the mid-basal or basal turns ( serial sections 1–3 ) ; the middle turns ( serial sections 4 and 5 ) , meanwhile , were just starting to turn off Ki67 ( Fig 5A , brackets ) . In all , in the 5 Sox2Ysb/+ cochleae examined , the most basal section that has not yet turned off Ki67 are 1 , 1 , 1 , 1 , and 2; in the 6 Sox2Ysb/- cochleae examined , they are 4 , 4 , 4 , 5 , 5 , and 5 ( p = 0 . 008 , Mann-Whitney U test ) . This indicates that in Sox2Ysb/- cochleae , cell cycle exit is delayed relative to Sox2Ysb/+ cochleae , suggesting a delay in the completion of prosensory specification . In addition , the nuclei of prosensory domain cells shift away from the luminal surface of the cochlear epithelium upon specification [29] . This basal shift of nuclei localization within the cell leaves a blank space between DAPI-stained nuclei and the luminal surface of the cochlear duct , which can be visualized in all six serial sections in Sox2Ysb/+ cochleae at E14 . 5 ( Fig 5A , asterisks ) . However , in Sox2Ysb/- cochleae , cells of the prosensory domain mostly did not exhibit this nuclei shift at E14 . 5 . At E15 . 5 , the prosensory domain along the entire length of the cochlear duct has turned off Ki67 expression in both Sox2Ysb/+ and Sox2Ysb/- cochleae , in all samples examined ( Fig 5A , brackets ) . This suggests that cell cycle exit and prosensory specification in Sox2Ysb/- cochleae has caught up by this stage . Prosensory nuclei localization has also begun to catch up at E15 . 5 in Sox2Ysb/- cochleae ( Fig 5A , asterisks ) . Overall , these results suggest that prosensory specification is delayed in Sox2Ysb/- cochleae , but not permanently disrupted . Prosensory specification must occur prior to differentiation to generate HCs and SCs . Therefore , the period of time in between cell cycle exit and the initiation of differentiation represents a temporal buffer ( Fig 5B , green shading ) preventing differentiation from initiating prior to specification . As differentiation begins in the basal/mid-basal cochlear turns shortly after specification , the delay in specification in Sox2Ysb/- cochleae leads to progenitors not having been specified in time for differentiation at the basal end of the cochlear duct ( Fig 5B , crosshatch pattern ) . We propose that this at least partially explains why the basal end of the cochlea is more sensitive to decreases in the level of Sox2 expression . Moreover , since differentiation begins in the mid-base and spreads to the rest of the base , progenitors at the basal tip in Sox2Ysb/- cochleae may still have time to undergo specification prior to differentiation . This may explain why small islands of HCs and SCs are sometimes seen in the basal tip of Sox2Ysb/- cochleae ( S4D Fig ) . Notably , while we detected a difference in Ki67 expression between Sox2Ysb/+ and Sox2Ysb/- cochleae at E14 . 5 ( Fig 5A ) , we did not detect a difference in EdU incorporation after 1 hour of EdU injection . In the basal turn of Sox2Ysb/- cochleae at E14 . 5 , cells of the presumptive prosensory domain still expressed Ki67 , but did not incorporate EdU ( S6A Fig , brackets ) . This indicates that while these presumptive prosensory cells have not exited the cell cycle , they are no longer actively undergoing DNA synthesis . We predict that these cells are either cycling slowly or are transiently “stuck” at a particular stage of the cell cycle . While the delay in prosensory specification can explain the preferential loss of sensory epithelium from the basal end of Sox2 hypomorph cochleae , it does not readily explain the preferential loss of OHCs , relative to IHCs . Since this preference for OHC loss is reminiscent of the Fgf20/Fgfr1 deletion phenotypes , we investigated the possibility that Sox2 may be upstream of FGF20-FGFR1 signaling . Interestingly , both Etv4 and Etv5 were dramatically downregulated in the prosensory domain of Sox2Ysb/- cochleae compared to control ( Fig 6A ) . This shows that FGF20-FGFR1 signaling was disrupted in the Sox2 hypomorph cochleae . Examination of Fgfr1 and Fgf20 expression by in situ hybridization revealed that while Fgfr1 expression did not appear to be affected in Sox2Ysb/- cochleae at E14 . 5 , Fgf20 expression was absent ( Fig 6B ) . This suggests that while Fgfr1 functions upstream of Sox2 ( Fig 2A ) , Fgf20 is downstream of Sox2 . This model predicts that Fgf20 expression would be downregulated in Fgfr1-CKO cochleae , which was confirmed by in situ hybridization ( Fig 6C ) . The above results indicate that the loss of Fgf20 could partially account for the Sox2Ysb/- phenotype . Therefore , to determine whether loss of Fgf20 also causes delayed prosensory specification , we examined Ki67 expression in Fgf20-KO cochleae . At E14 . 5 , there was no detectable delay in cell cycle exit in Fgf20-KO cochleae , as loss of Ki67 expression reached the base ( serial section 1 ) in all 6 control and all 6 Fgf20-KO cochleae examined ( S6B Fig , brackets ) . There was also no detectable delay in prosensory basal nuclei shift in Fgf20-KO cochleae ( S6B Fig , asterisks ) . These results were expected as the Fgf20-KO phenotype is not more severe at the basal end of the cochlear duct . This is also consistent with Fgf20 being required during differentiation rather than prosensory specification ( Fig 2D ) . However , these results do not answer whether and how the loss of Fgf20 contributes to the Sox2 hypomorph phenotype . Since Fgfr1 is upstream of Sox2 , we next asked whether Fgfr1 deletion also results in a delay in prosensory cell cycle exit via decrease in Sox2 expression . Interestingly , similar to the Fgf20-KO , there was no detectable delay in cell cycle exit or basal nuclei shift in Fgfr1-CKO cochleae compared to control at E14 . 5 ( S6B Fig , brackets , asterisks ) . We hypothesized that this lack of a detectable difference may be due to the relatively small reduction in Sox2 expression in Fgfr1-CKO cochleae compared to Sox2Ysb/- cochleae . To examine cell cycle exit more closely for subtle changes , we used Sox2 as a marker of prosensory cells and quantified the number of Sox2-expressing cells that also expressed Ki67 in serial sections through the cochlear duct . Because at E14 . 5 Sox2 also labels the Kölliker’s organ , which is neural ( inner ) to the prosensory domain , we quantified cells at the abneural ( outer ) border of the prosensory domain . In E14 . 5 Fgf20-KO cochleae , a few Ki67+/Sox2+ cells could be found at the abneural border , mostly towards the basal end of the cochlear duct , similar to control cochleae ( Fig 6D and 6E ) . However , there were significantly more Ki67+/Sox2+ cells at the abneural border in Fgfr1-CKO cochleae compared to control ( Fig 6D and 6E , arrowheads ) . This shows that , as expected , deletion of Fgfr1 , but not Fgf20 , does lead to a quantifiable defect in prosensory cell cycle exit . We also asked whether a decrease in Sox2 expression can account for the absence of proliferation in Kölliker’s organ in Fgfr1-CKO cochleae . Interestingly , EdU-incorporation was decreased in Kölliker’s organ in Sox2Ysb/- cochleae at E14 . 5 , especially in the region adjacent to the prosensory domain ( S6C Fig , bracket ) . However , EdU-incorporation was not completely absent from Kölliker’s organ , unlike in Fgfr1-CKO cochleae . This suggests that loss of Sox2 in combination with other factors contributes to the Kölliker’s organ phenotype in Fgfr1-CKO cochleae . To explore how the loss of Fgf20 contributes to the Sox2 hypomorph phenotype , we combined the Fgf20- and Sox2Ysb alleles to generate Fgf20 and Sox2 compound mutants . We also hypothesized that reducing Sox2 expression in Fgf20-KO mice would recapitulate ( or phenocopy ) the more severe Fgfr1-CKO phenotype . We interbred F1 mice from the same parents to generate nine different F2 genotypes encompassing all possible combinations of the Fgf20- and Sox2Ysb alleles: Fgf20+/+;Sox2+/+ , Fgf20+/+;Sox2Ysb/+ , Fgf20+/-;Sox2+/+ , Fgf20+/-;Sox2Ysb/+ , Fgf20+/+;Sox2Ysb/Ysb , Fgf20+/-;Sox2Ysb/Ysb , Fgf20-/-;Sox2+/+ , Fgf20-/-;Sox2Ysb/+ , and Fgf20-/-;Sox2Ysb/Ysb ( Figs 7 and 8 ) . At P0 , an overview of HCs and SCs showed that the Fgf20+/-;Sox2Ysb/+ phenotype mostly resembled that of Fgf20+/+;Sox2+/+ , Fgf20+/+;Sox2Ysb/+ , and Fgf20+/-;Sox2+/+ cochleae , except for the prevalence of ectopic IHCs ( Fig 7A , arrowheads ) . The Fgf20+/-;Sox2Ysb/Ysb phenotype mostly resembled that of Fgf20+/+;Sox2Ysb/Ysb cochleae , but with more gaps in the basal cochlear turn and two rows of IHCs throughout the length of the cochlear duct , except where there were gaps . The Fgf20-/-;Sox2Ysb/+ phenotype mostly resembled that of Fgf20-/-;Sox2+/+ cochleae , but with smaller sensory islands in between gaps . The Fgf20-/-;Sox2Ysb/Ysb phenotype appeared by far the most severe , with almost a complete absence of IHCs , OHCs , and SCs from the basal turn , and tiny sensory islands in the middle turn; however , the apical turn appeared similar to that of Fgf20-/-;Sox2Ysb/+ and Fgf20-/-;Sox2+/+ cochleae ( Fig 7A and 7B ) . Quantification of the phenotypes are presented in Fig 8B–8E and S7B–S7D Fig . We analyzed the quantified P0 phenotype via two-way ANOVA with the two factors being gene dosage of Fgf20 ( levels: Fgf20+/+ , Fgf20+/- , Fgf20-/- ) and Sox2 ( levels: Sox2+/+ , Sox2Ysb/+ , Sox2Ysb/Ysb ) . Results from the two-way ANOVA and post-hoc Tukey’s HSD are presented in Fig 8A and 8F and S7A and S8 Figs . Cochlear length and the total number of IHCs , OHCs , and SCs were all significantly affected by both the Fgf20 dosage and the Sox2 dosage , as well as an interaction between the two factors ( Fig 8A–8E ) . The statistically significant interaction between Fgf20 and Sox2 dosages suggests that Fgf20 and Sox2 have a genetic interaction in regulating cochlear length as well as the number of IHCs , OHCs , and SCs ( Fig 8A ) . Notably , Fgf20+/-;Sox2Ysb/Ysb cochleae had significantly fewer OHCs and SCs than Fgf20+/+;Sox2Ysb/Ysb cochleae , and Fgf20-/-;Sox2Ysb/+ cochleae had significantly fewer OHCs than Fgf20-/-;Sox2+/+ cochleae ( Fig 8F ) . Importantly , Fgf20-/-;Sox2Ysb/Ysb cochleae had decreased total and length-normalized number of IHCs , which was not observed in any of the other genotypes , strongly supporting a genetic interaction between Fgf20 and Sox2 ( Fgf20+/+;Sox2Ysb/Ysb cochleae did have a slight decrease in the total number of IHCs , but not in the length-normalized number of IHCs ) . Interestingly , while the total number of IHCs was decreased in Fgf20-/-;Sox2Ysb/Ysb cochleae relative to all other genotypes , this decrease was only found in the basal and middle turns , but not the apical turn ( S7B and S8 Figs ) . No such basal/middle/apical turn discrepancies existed in the number of OHCs or SCs ( S7C , S7D and S8 Figs ) . This is reminiscent of the Fgfr1-CKO and Sox2Ysb/- phenotypes . To ensure that the Fgf20- and Sox2Ysb interaction is not purely an artifact of the Sox2Ysb allele , we generated Fgf20+/+;Sox2+/+ ( wildtype ) , Fgf20+/-;Sox2+/+ ( Fgf20-het ) , Fgf20+/+;Sox2+/- ( Sox2-het ) , and Fgf20+/-;Sox2+/- ( double het ) mice to look for an interaction between the Fgf20- and Sox2- alleles ( S9A Fig ) . At P0 , cochlear length did not significantly differ among the four genotypes ( S9B Fig ) . HC quantification showed that neither Fgf20 nor Sox2 exhibited haploinsufficiency for total or length-normalized number of IHCs or OHCs ( S9C and S9D Fig ) . However , in Fgf20-het and much more so in Sox2-het cochleae , occasional ectopic IHCs can be found in the middle and apical turns of the cochlear duct ( S9A Fig , arrowheads ) . Interestingly , in double het cochleae , many more ectopic IHCs were found , even in the basal turn . These ectopic IHCs led to an increase in the total and length-normalized number of IHCs in double het cochleae , compared to wildtype ( S9C Fig ) . Notably , a significant increase in IHCs was only found in the basal turn , not the middle or apical turns ( S9E Fig ) . In the basal turn , IHC number was significantly increased in double het cochleae compared to wildtype , Fgf20-het , and Sox2-het cochleae . Double het cochleae also had a significant decrease in total and length-normalized number of OHCs compared to wildtype ( S9D Fig ) . Again , a significant decrease in OHCs was only found in the basal turn , not the middle or apical turns ( S9F Fig ) . These results confirm a genetic interaction between Fgf20 and Sox2 . We propose that the Fgf20-/-;Sox2Ysb/Ysb phenotype lies in between that of Fgfr1-CKO and Sox2Ysb/- in terms of severity of reductions in cochlear length and in the number of HCs and SCs . We further hypothesize that these three phenotypes form a continuum with the Fgf20-KO phenotype ( Fig 9A ) . Along this continuum , all four genotypes lack FGF20 signaling , but vary in the level of Sox2 expression and phenotype severity in the basal end of the cochlear duct and the outer compartment ( outer rows of OHCs and SCs ) . From this , and from the Fgf20- and Sox2Ysb series of alleles , we conclude that the basal end of the cochlear duct and the outer compartment are more sensitive to the loss of Fgf20 and Sox2 , relative to the apical end and inner compartment , respectively . To determine the mechanism underlying the Sox2 and Fgf20 interaction , we asked whether the loss of Fgf20 further reduces Sox2 expression on a Sox2 hypomorphic background . In other words , we asked whether Fgf20 has a role in regulating Sox2 expression on a sensitized background . Examination of prosensory domain Sox2 expression at E14 . 5 revealed , as expected , that Fgf20-/-;Sox2Ysb/+ cochleae did not have a decrease in Sox2 expression compared to Fgf20+/-;Sox2Ysb/+ ( S10A Fig ) . Fgf20-/-;Sox2Ysb/Ysb cochleae also did not have a further decrease in Sox2 expression compared to Fgf20+/-;Sox2Ysb/Ysb cochleae . Moreover , despite the loss of sensory epithelium in most of the basal turn , Sox2 expression was not further decreased in the basal turn at E14 . 5 relative to the rest of the Fgf20-/-;Sox2Ysb/Ysb cochlea ( S10A Fig ) . A similar pattern of expression was observed for CDKN1B across the different genotypes ( S10B Fig ) . Loss of Fgf20 did not contribute to a further decrease in CDKN1B expression on a Sox2Ysb/Ysb background , nor was there a basal-apical difference in CDKN1B expression in Fgf20-/-;Sox2Ysb/Ysb cochleae at E14 . 5 . These results are consistent with our hypothesis that Fgf20 does not regulate Sox2 expression or prosensory specification . Next , we asked whether Sox2 and Fgf20 interact to delay prosensory specification . We showed that Fgf20-KO cochleae do not exhibit a delay in prosensory specification ( Fig 6D and 6E ) . However , this does not rule out the possibility that the loss of Fgf20 may contribute to a delay on a Sox2 hypomorphic background . We examined Ki67 expression at E14 . 5 and found that in Fgf20+/-;Sox2Ysb/+ cochleae , prosensory domain cell cycle exit has reached the end of the base ( serial section 1; S10D Fig , brackets ) . Similarly , cell cycle exit in Fgf20-/-;Sox2Ysb/+ cochleae also reached the end of the base . As expected , Fgf20+/-;Sox2Ysb/Ysb cochleae exhibited a slight delay in prosensory specification; cell cycle exit has reached the base ( serial section 2 ) , but has not yet reached the end of the base ( serial section 1 ) . Importantly , Fgf20-/-;Sox2Ysb/Ysb cochleae did not show a further delay relative to Fgf20+/-;Sox2Ysb/Ysb . There was also no detectable delay in nuclei shift in Fgf20-/-;Sox2Ysb/+ or Fgf20-/-;Sox2Ysb/Ysb cochleae ( S10D Fig , asterisks ) . We then quantified the number of Sox2-expressing prosensory cells that still express Ki67 in serial sections through the cochlear duct at E14 . 5 . As expected , both Fgf20+/-;Sox2Ysb/Ysb and Fgf20-/-;Sox2Ysb/Ysb cochleae had significantly more Ki67+/Sox2+ cells than Fgf20+/-;Sox2Ysb/+ cochleae , confirming that decrease in Sox2 expression delays specification ( Fig 9B and 9C , arrowheads ) . Importantly , loss of Fgf20 alone had no detectable effect on the number of Ki67+/Sox2+ cells: Fgf20-/-;Sox2Ysb/+ and Fgf20-/-;Sox2Ysb/Ysb cochleae did not have significantly more Ki67+/Sox2+ cells than Fgf20+/-;Sox2Ysb/+ and Fgf20+/-;Sox2Ysb/Ysb cochleae , respectively ( Fig 9B and 9C ) . These results suggest that the loss of Fgf20 does not contribute to delayed specification , even on a Sox2 hypomorphic background . They also show that the severity of the Fgf20-/-;Sox2Ysb/Ysb basal phenotype cannot be completely attributed to delayed specification . Lastly , we examined proliferation in the Kölliker’s organ of Fgf20- and Sox2Ysb E14 . 5 cochleae . Interestingly , there was a noticeable decrease in the number of EdU-incorporating cells in Kölliker’s organ in Fgf20-/-;Sox2Ysb/Ysb cochleae , compared to Fgf20+/-;Sox2Ysb/+ , Fgf20-/-;Sox2Ysb/+ , and Fgf20+/-;Sox2Ysb/Ysb cochleae ( S10C Fig ) . This phenotype is similar to that of Sox2Ysb/- cochleae and is less severe than that of Fgfr1-CKO cochleae . This suggests that Fgf20 and Sox2 interact to regulate proliferation in Kölliker’s organ , although other factors downstream of Fgfr1 also contribute . One such factor could be Fgf10 , which has been shown to be downregulated in the Kölliker’s organ in Fgfr1-mutant mice [19] . We showed that based on the timing of FGF9 rescue , Fgf20 likely plays a role during the initiation of differentiation . Previous studies showed that deletion of both transcription factors Hey1 and Hey2 results in premature differentiation in the organ of Corti [30] . Furthermore , it has been suggested that FGF signaling , in particular FGF20 , regulates Hey1 and Hey2 expression during this process [8 , 30] . To test whether Fgf20 is upstream of Hey1 and Hey2 , we looked at the expression of the two transcription factors via in situ hybridization . In Fgf20-KO cochleae at E14 . 5 , Hey1 expression is downregulated while Hey2 is almost completely absent compared to control ( Fig 10A ) . To test whether FGF20 loss leads to premature differentiation , we examined myosin VI ( Myo6 ) expression , a marker of differentiated HCs [30] . At E14 . 5 , the cochleae of 3 of 12 control embryos examined contained Myo6-expressing HCs , while the cochleae of 18 of 19 littermate Fgf20-KO embryos contained Myo6-expressing HCs ( p < 0 . 001 , Fisher’s exact test; Fig 10B ) . If present , the Myo6-expressing HCs at this stage were always found in the basal and mid-basal turns of the cochlea . These results show that there is premature onset of differentiation in Fgf20-KO cochleae , which begins in the basal/mid-basal turns . This result is surprising given our previous finding of delayed differentiation in the apical end of Fgf20-KO cochleae at later stages , which we confirm here ( S1E Fig ) . These findings suggest that while initiation of differentiation occurs earlier in Fgf20-KO cochleae , apical progression of differentiation may be slower . Next , we asked whether ectopic activation of FGF signaling via overexpression of FGF9 will delay the onset of differentiation . We generated Fgf20-het ( Fgf20Cre/+;ROSArtTA/+ ) , Fgf20-null ( Fgf20Cre/βgal;ROSArtTA/+ ) , Fgf9-OA ( Fgf20Cre/+;ROSArtTA/+;TRE-Fgf9-IRES-eGfp ) , Fgf9-rescue ( Fgf20Cre/βgal;ROSArtTA/+;TRE-Fgf9-IRES-eGfp ) mice as before and started Dox induction at E13 . 5 until E15 . 0 ( Fig 3 ) . At E15 . 0 , all of the Fgf20-het ( 4/4 ) and Fgf20-null ( 4/4 ) cochleae contained Myo6-expressing HCs , while none of the Fgf9-OA ( 0/4 ) and Fgf9-rescue ( 0/4 ) cochleae contained Myo6-expressing HCs ( Fig 10C ) . This suggests that ectopic expression of FGF9 was able to delay the onset of differentiation , even with the lack of endogenous FGF20 . Despite this delay in onset of differentiation , by P0 , differentiation has apparently caught up in both Fgf9-OA and Fgf9-rescue cochleae ( Fig 3A ) . Similar to a delay in prosensory specification , premature onset of differentiation narrows the temporal buffer between the completion of specification and initiation of differentiation . In the context of a slight delay in specification due to decreased Sox2 levels , premature differentiation from the loss of Fgf20 can lead to an attempt at differentiation before specification in the basal end of the cochlea . We propose that Sox2 and Fgf20 interact to regulate the boundaries of the temporal buffer , helping to ensure that differentiation begins after the completion of specification ( Fig 11 ) .
Fgf20 and Fgfr1 are required for HC and SC development . Based on similarities in the phenotype caused by the loss of FGF20 and loss of FGFR1 signaling , FGF20 has been hypothesized as the FGFR1 ligand during organ of Corti development [17–21] . However , the exact role of FGF20/FGFR1 during organ of Corti development has been a topic of debate . We previously reported that Fgf20-KO mice do not have defects in prosensory specification , and have a normally formed prosensory domain [21] . We further showed that FGF20 signaling is important during the initiation stage of differentiation , and that Fgf20-KO cochleae have gaps in the differentiated sensory epithelium filled with undifferentiated prosensory progenitors . However , other studies have shown in vitro that FGF20 regulates prosensory specification via Sox2 [31] and in vivo that FGFR1 is required for prosensory specification via Sox2 [19] . Here , we show , using an in vivo rescue model , that ectopic FGF9 signaling is sufficient to rescue the Fgf20-KO phenotype in a spatiotemporal pattern that matched the timing of initiation of differentiation along the length of the cochlear duct . We conclude , therefore , that FGF20 is involved in differentiation and is not necessary for prosensory specification . Notably , the Fgf20-KO phenotype , in which two-thirds of OHCs fail to develop , is not as severe as the Fgfr1-CKO phenotype , which lacks almost all OHCs as well as half of IHCs . Potential explanations for this include differences in mouse genetic background , and the existence of a redundant FGF ligand ( s ) . To rule out the former , we examine here Fgf20-KO and Fgfr1-CKO mice on a similar genetic background , and replicated the difference in phenotype severity . We also replicated the decrease in Sox2 expression in the prosensory domain previously reported in Fgfr1-CKO mice [19] . We further reaffirmed that Sox2 expression in the prosensory domain is not affected by the loss of Fgf20 . This suggests that another FGF ligand signaling through FGFR1 is required to maintain Sox2 expression during prosensory specification . The identity of this ligand is currently unknown . We hypothesized that the severity of the Fgfr1-CKO phenotype is due to decreased Sox2 expression causing disrupted prosensory specification and the loss of FGF20 signaling during differentiation . Consistent with this hypothesis , the combination of Fgf20-/- and Sox2Ysb/Ysb mutations phenocopied Fgfr1-CKO cochleae . The similarities in phenotype include approximately a 30% reduction in cochlear length and almost a complete loss of OHCs and SCs and approximately a 50% loss of IHCs . Interestingly , the Fgf20-/-;Sox2Ysb/Ysb phenotype is also similar to the Sox2Ysb/- phenotype . We conclude that the Fgfr1-CKO , Fgf20-/-;Sox2Ysb/Ysb , and Sox2Ysb/- phenotypes likely lie along the same continuum , as these three genotypes all exhibited a lack of Fgf20 expression or signaling and varying levels of Sox2 expression ( Fig 9A ) . Fgf20-KO cochleae , in which Sox2 expression was not affected , lies at the mild end of this continuum . Interestingly , this continuum shows that in the absence of Fgf20 expression or signaling , reductions in the level of Sox2 most severely affected sensory epithelium development of the cochlear base and the outer compartment . Moving from the Fgf20-KO ( mild ) end of the spectrum towards the Sox2Ysb/- ( severe ) end , increasing numbers of HCs and SCs are lost , preferentially form the cochlear base and the outer compartment . Importantly , while these results seem to suggest that the main function of FGFR1 signaling during early stages of organ of Corti development is to regulate Sox2 expression , we have not ruled out the potential for other functions of FGFR1 signaling . There are also notable differences between the Fgfr1-CKO and Fgf20-/-;Sox2Ysb/Ysb phenotypes at P0 . For instance , Fgfr1-CKO cochleae have a slightly less severe OHC phenotype at the cochlear base relative to the apex , while Fgf20-/-;Sox2Ysb/Ysb cochleae have a slightly more severe OHC phenotype at the base . We attribute these phenotype differences to differences in the level of Sox2 expression and the timing of decrease in Sox2 expression . However , they may be instead attributable to additional functions of FGFR1 signaling not captured by the combination of Fgf20-/- and Sox2Ysb/Ysb mutations . Foxg1Cre has been used in several studies to target the otic epithelium , including to conditionally delete Fgfr1 [18–20] . One concern with Foxg1Cre is that it is a null allele [22] . Foxg1-null mice have shortened cochlear length , although HC and SC differentiation did not appear to be directly affected [32] . Previous work [33] and our results here showed that Foxg1 is not haploinsufficient during cochlea development , as Foxg1Cre/+;Fgfr1flox/+ cochleae had very similar phenotypes to Fgfr1flox/- cochleae . Moreover , the use of the Six1enh21-Cre transgene , which targets the otic epithelium in a similar spatiotemporal pattern as Foxg1Cre , to conditionally delete Fgfr1 resulted in the same phenotype as Foxg1Cre/+;Fgfr1flox/- cochleae [19] . This included the loss of almost all OHCs , loss of IHCs , and decreased prosensory Sox2 expression . Therefore , the increased severity of Foxg1Cre/+;Fgfr1flox/- cochleae relative to Fgf20-/- cochleae is likely not attributable Foxg1 haploinsufficiency . We show here conclusive evidence that Sox2 and Fgf20 genetically interact during cochlea development . Interestingly , HC and SC development towards the basal end of the cochlea is more severely affected by the loss of Sox2 and Fgf20 and their interaction . While we hypothesize that Sox2 and Fgf20 are involved in distinct steps during organ of Corti development ( prosensory specification and differentiation , respectively ) , there is nevertheless potential for a strong interaction . We propose that the timing of specification and differentiation define a temporal buffer that normally prevents differentiation from initiating prior to the completion of specification , and that Sox2 and Fgf20 modulate the borders of this buffer . In a developmental pathway , the upstream event ( specification ) must occur prior to the downstream event ( differentiation ) . Therefore , loss of Sox2 and Fgf20 leading to delayed specification and premature differentiation onset , respectively , disrupts the temporal buffer , especially towards the cochlear base ( Fig 11 ) . Here , we use cell cycle exit in the prosensory domain ( also known as the zone of non-proliferation ) as a marker for the completion of specification [3] . We hypothesize that prosensory cells become specified and primed for differentiation upon withdrawal from the cell cycle . Previous studies showed that prosensory cells are indeed capable of differentiating into HCs and SCs directly after cell cycle exit , even in the apex . When Shh was deleted from the spiral ganglion , differentiation began in the apex shortly after cell cycle exit and progressed towards the base [5] . This suggests that specification occurs in an apex-to-base direction . It also suggests that normally , SHH signaling prevents the apex from differentiating immediately after specification . We cannot rule out , however , that specification occurs in the same direction as differentiation ( base-to-apex ) , independently of cell cycle exit . Such a scenario would still be consistent with our model that a combination of delayed specification and premature onset of differentiation accounts for the more severe basal phenotype in Fgf20/Sox2 mutants . The effect of loss of Fgf20 on the timing of differentiation is small . We estimate that the onset of differentiation in Fgf20-KO cochleae is advanced by only around 0 . 5 days . By itself , this effect does not lead to a more severe mid-basal or basal phenotype in Fgf20-KO cochleae . However , we present evidence that on a sensitized genetic background causing delayed specification , this small change in the timing of differentiation leads to a large defect in HC and SC production towards the basal end of the cochleae . We propose that this at least partially explains the interaction between Sox2 and Fgf20 . Furthermore , the relative sparing of development towards the apical end of Sox2Ysb/Ysb;Fgf20-/- cochleae , especially of IHCs , can be further explained by a delay in differentiation at the apical end due to the loss of Fgf20 . We do not know why an apical-basal difference in timing of differentiation exists in Fgf20-KO cochleae . Perhaps there is a delay in the apical progression of differentiation , or perhaps other factors contribute to the differentiation of the apical end of the cochlea . Consistent with the latter , by P7 in Fgf20-KO cochleae , the apical tip contains a full complement of IHCs and OHCs , unlike the rest of the cochlea [21] . Importantly , the model that Sox2 and Fgf20 regulate the distinct processes of specification and differentiation , respectively , is a simplified take on a complex developmental pathway for the sake of addressing our specific question . While we show the potential for a Sox2 and Fgf20 interaction in modulating the temporal buffer between specification and differentiation , Sox2 also has known roles during HC and SC differentiation [13 , 15 , 34 , 35] . Therefore , the genetic interaction may occur during differentiation as well . While interaction at this stage may explain the preferential loss of outer compartment cells in Sox2 and Fgf20 mutants , it does not explain the selective loss of basal cochlear HCs and SCs . Therefore , we conclude that the Sox2 and Fgf20 interaction regulates the temporal buffer , with potential further interactions during differentiation . The Notch ligand Jagged1 ( Jag1 ) is thought to be important for cochlear prosensory specification via lateral induction [36–42] . Interestingly , Notch signaling has also been shown to be upstream of both Fgf20 and Sox2 in the developing cochlea [31] . Conditional deletion of Jag1 or Rbpj , the major transcriptional effector of canonical Notch signaling , resulted in the loss of HCs and SCs , particularly from the basal end of the cochlear duct , similar to Fgf20/Sox2 mutants . Unlike Fgf20/Sox2 mutants , however , deletion of Jag1 or Rbpj led to preferential loss of Sox2 and CDKN1B expression from the prosensory domain at the basal end of the cochlea [37 , 39 , 43] . This suggests that Jag1-Notch signaling is required for prosensory specification , especially towards the cochlear base . This likely accounts for the more severe basal phenotype of Jag1 or Rbpj mutants . This same mechanism likely does not explain the more severe basal phenotype of Fgf20/Sox2 mutants , as Sox2 and CDKN1B expression was not more severely reduced or absent in the cochlear base in these mice . Notably , not all studies agree that Jag1 or Rbpj is required for Sox2 and CDKN1B expression or for prosensory specification [44] . More studies are required to further elucidate the functional relationship between Jag1/Notch , Fgf20 , and Sox2 during cochlea development . Other genes that potentially interact with Fgf20 and Sox2 during cochlea development include Mycn ( N-Myc ) and Mycl ( L-Myc ) . Interestingly , deletion of Mycn and Mycl from the cochlear epithelium results in accelerated cell cycle exit and delayed initiation of differentiation [45] , opposite to the effects of loss of Sox2 and Fgf20 . Addressing potential interactions between Sox2 , Fgf20 , Mycn , and Mycl is another topic for future studies . In all of the genotypes we observed in this study , the loss of outer compartment cells ( i . e . OHCs ) was predominant . Only in the most severe cases in which almost all OHCs were missing , as seen in Fgfr1-CKO , Fgf20-/-;Sox2Ysb/Ysb , and Sox2Ysb/- cochleae , were IHCs also lost . Similarly , reduction in SC number always preferentially affected the outermost cells . This suggests that the organ of Corti outer compartment is more sensitive to the loss of Fgfr1 , Fgf20 , and Sox2 than the inner compartment . The combination of Fgf20- and Sox2Ysb alleles elegantly demonstrates this: as the number of Fgf20- and Sox2Ysb alleles increased , the number of OHCs progressively decreased . In the double homozygous mutants , the number of IHCs decreased as well . We also show here that in Fgfr1-CKO and Fgf20-/-;Sox2Ysb/Ysb cochleae , there is a delay in cell cycle exit and completion of specification , especially towards the abneural or outer side of the prosensory domain , which may contribute to the more severe outer compartment phenotype . However , Fgf20-KO cochleae , which also exhibit a more severe outer compartment phenotype , did not show a delay in specification of the outer side of the prosensory domain , at least at the stage we examined . Further studies are needed to elucidate differences in the timing of cell cycle exit and specification in the outer and inner compartments , and how they may affect differentiation . Previous studies noted that the dosage of Fgfr1 affects the degree of organ of Corti outer compartment loss . In Fgfr1 hypomorphs with 80% reduction in transcription , only the third row of OHCs were missing , while hypomorphs with 90% reduction had a slightly more severe phenotype [20] . Therefore , Fgfr1 loss preferentially affects the outermost HCs . Other studies suggested that the timing of Fgfr1 deletion is important in determining the degree of outer compartment loss and level of Sox2 expression . When an earlier-expressed Cre driver ( Six1enh21-Cre ) was used to conditionally delete Fgfr1 , almost all OHCs and some IHCs were lost , with a 66% reduction in Sox2 expression at E14 . 5 [19] . When a later-expressed Cre driver ( Emx2Cre ) was used , many more OHCs and IHCs remained , with only a 12% reduction in Sox2 expression . Our results are consistent with both of these studies . We show that FGF20-independent FGFR1 signaling and Sox2 are required early , affecting both IHC and OHC development , while FGF20-FGFR1 signaling is important during later stages , affecting only OHC development . Differentiation in the organ of Corti not only occurs in a basal-to-apical gradient , but also in an orthogonal inner-to-outer gradient . That is , IHCs differentiate first , followed by each sequential row of OHCs [46] . This wave of differentiation suggests that perhaps outer compartment HCs and SCs require a longer temporal buffer between specification and differentiation . The genetic interaction between Sox2 and Fgf20 in modulating this temporal buffer , therefore , could also account for the preferential loss of outer compartment HCs and SCs . We hypothesize that the requirement for a longer temporal buffer may also be involved in determining OHC fate . In Fgf20+/-;Sox2+/- cochleae , there was a slight decrease in OHCs that was compensated for by ectopic IHCs , suggesting a fate switch from OHCs into IHCs . Here , we confirmed previous suggestions that Fgf20 regulates Hey1 and Hey2 to prevent premature differentiation in the developing organ of Corti [8 , 30] . Interestingly , in Hey1/Hey2 double knockout cochleae , there was a similar slight decrease in OHCs compensated for by ectopic IHCs [30] . Furthermore , inner ear-specific deletion of either Smoothened or Neurod1 , which led to premature differentiation in the apical cochlear turn , also led to loss of OHCs and the presence of ectopic IHCs at the apex [6 , 8] . These findings further support a model where timing of specification and differentiation affect IHC versus OHC fate , an interesting and important topic for future studies . Previously , we hypothesized that Fgf20 is strictly required for the differentiation of an outer compartment progenitor [21] . However , data we present here show that Fgf20 , on a sensitized , Sox2 hypomorphic background , is also required for inner compartment differentiation . We conclude that inner and outer compartment progenitors likely are not distinct populations . Rather , all prosensory progenitors giving rise to the organ of Corti exist on an inner-to-outer continuum . FGF20 signaling , in combination with other factors including Sox2 , are required for the proper development of all of these cells , though with varying sensitivities . We show in vivo that Fgf20 is upstream of Hey1 and Hey2 . Supporting this result , Fgfr1 has also been shown in vivo to be upstream of Hey2 [19] . Interestingly , in explant studies , inhibition of FGF signaling alone did not result in decreased Hey1/Hey2 expression or premature differentiation [30] . However , FGF inhibition has been shown to rescue the overexpression of Hey1/Hey2 and the delay in differentiation induced by SHH signaling overactivation [8 , 30] . Notably , these studies suggest that SHH signaling from the spiral ganglion regulates Fgf20 expression , which in turn regulates Hey1 and Hey2 expression to prevent premature differentiation in the organ of Corti [5 , 8 , 30] . Our results here showing that Fgf20 regulates Hey1 and Hey2 expression and timing of differentiation are mostly consistent with these studies . However , Hey1/Hey2 double knockout cochleae do not exhibit a loss of OHCs to the extent of Fgf20-KO cochleae , suggesting that other genes downstream of Fgf20 are important in prosensory cell differentiation ( Fig 11 ) . Moreover , deletion of Fgf20 only led to premature differentiation at the basal and mid-basal turns . Fgf20 deletion actually delayed differentiation in the apical end of the cochlea . Deletion of Hey1/Hey2 , contrarily , led to premature differentiation along the entire length of the cochlear duct , although it is unclear how Hey1/Hey2 loss affects the timing of apical differentiation beyond E15 . 0 [30] . This suggests that other factors downstream of Fgf20 interact with Hey1/Hey2 to regulate the timing of differentiation . Perhaps these same genes contribute to the loss of OHCs in Fgf20-KO cochleae . Mekk4 , which has been shown to be downstream of Fgf20 and necessary for OHC differentiation [47] could be one of these genes . Identifying other factors downstream of Fgf20 will be a topic of future studies .
All studies performed were in accordance with the Institutional Animal Care and Use Committee at Washington University in St . Louis ( protocol #20160113 ) and University of Nebraska Medical Center ( protocol #16-004-02 and 16-005-02 ) . Mice were group housed with littermates , in breeding pairs , or in a breeding harem ( 2 females to 1 male ) , with food and water provided ad libitum . For timed-pregnancy experiments , embryonic day 0 . 5 ( E0 . 5 ) was assigned as noon of the day the vaginal plug was found . For postnatal experiments , postnatal day 0 ( P0 ) was determined as the day of birth . Mice were of mixed sexes and maintained on a mixed C57BL/6J x 129X1/SvJ genetic background . All mice were backcrossed at least three generations onto this background . The following mouse lines were used: Pregnant dams were starved overnight the night before initiation of Dox induction and fed Dox Diet , Grain-Based Doxycycline , 200 mg/kg ( S3888 , Bio-Serv , Flemington , NJ ) ad libitum starting at noon on the start date of Dox induction . On the stop date of Dox induction , Dox Diet was replaced with regular mouse chow at noon . For whole mount cochleae , inner ears were dissected out of P0 pups and fixed in 4% PFA in PBS overnight at 4°C with gentle agitation . Samples were then washed x3 in PBS . Cochleae were dissected away from the vestibule , otic capsule , and periotic mesenchyme with Dumont #55 Forceps ( RS-5010 , Roboz , Gaithersburg , MD ) . The roof of the cochlear duct was opened up by dissecting away the stria vascularis and Reissner’s membrane; tectorial membrane was removed to expose hair and supporting cells . For sectioning , heads from E14 . 5 embryos were fixed in 4% PFA in PBS overnight at 4°C with gentle agitation . Samples were then washed x3 in PBS and cryoprotected in 15% sucrose in PBS overnight and then in 30% sucrose in PBS overnight . Samples were embedded in Tissue-Tek O . C . T . compound ( 4583 , VWR International , Radnor , PA ) and frozen on dry ice . Serial horizontal sections through base of the head were cut at 12 μm with a cryostat , dried at room temperature , and stored at -80°C until use . Probe preparation: mouse cDNA plasmids containing the following inserts were used to make RNA in situ probes , and were cut and transcribed with the indicated restriction enzyme ( New England Biolabs , Ipswich , MA ) and RNA polymerase ( New England Biolabs , Ipswich , MA ) : Fgfr1 transmembrane domain ( pBluescriptKS-Fgfr1TM , 325 bp , HincII , T7 , gift of K . Peters ) , Fgf20 ( pGEMT-Fgf20 , 653 bp , NcoI , Sp6 ) , Sox2 ( pBluescriptSK-Sox2 , 750 bp , AccI , T3 , gift of A . Kiernan ) , Etv4 ( pGEM-Etv4 , ~2300 bp , ApaI , Sp6 , gift of G . Martin ) , Etv5 ( pBluescriptSK-Etv5 , ~4000 bp , HindIII , T3 , gift of G . Martin ) , Hey1 ( pT7T3D-Hey1 [IMAGE clone #478014] , 343 bp , EcoRI , T3 , gift of S . Rentschler ) , Hey2 ( pCMVSPORT6-Hey2 [IMAGE clone #5374813] , 819 bp , EcoRI , T7 , gift of S . Rentschler ) . Restriction digest and in vitro transcription were done according to manufacturer’s instructions , with DIG RNA Labeling Mix ( 11277073910 , Sigma-Aldrich , St . Louis , MO ) . After treatment with RNase-free DNase I ( 04716728001 , Sigma-Aldrich , St . Louis , MO ) for 15 min at 37°C , probes were hydrolyzed in hydrolysis buffer ( 40 mM NaHCO3 , 60 mM Na2CO3 ) at 60°C for up to 30 min , depending on probe size . Frozen section in situ hybridization: frozen slides were warmed for 20 min at room temperature and then 5 min at 50°C on a slide warmer . Sections were fixed in 4% PFA in PBS for 20 min at room temperature , washed x2 in PBS and treated with pre-warmed 10 μg/ml Proteinase K ( 03115828001 , Sigma-Aldrich , St . Louis , MO ) in PBS for 7 min at 37°C . Sections were then fixed in 4% PFA in PBS for 15 min at room temperature , washed x2 in PBS , acetylated in 0 . 25% acetic anhydrate in 0 . 1M Triethanolamine , pH 8 . 0 , for 10 min , and washed again in PBS . Sections were then placed in pre-warmed hybridization buffer ( 50% formamide , 5x SSC buffer , 5 mM EDTA , 50 μg/ml yeast tRNA ) for 3 h at 60°C in humidified chamber for prehybridization . Sections were then hybridized in 10 μg/ml probe/hybridization buffer overnight ( 12–16 h ) at 60°C . The next day , sections were washed in 1x SSC for 10 min at 60°C , followed by 1 . 5x SSC for 10 min at 60°C , 2x SSC for 20 min at 37°C x2 , and 0 . 2x SSC for 30 min at 60°C x2 . Sections were then washed in KTBT ( 0 . 1 M Tris , pH 7 . 5 , 0 . 15 M NaCl , 5 mM KCl , 0 . 1% Triton X-100 ) at room temperature and blocked in KTBT + 20% sheep serum + 2% Blocking Reagent ( 11096176001 , Sigma-Aldrich , St . Louis , MO ) for 4 h . Blocking Reagent was dissolved in 100 mM Maleic acid , 150 mM NaCl , pH 7 . 5 . Sections were then incubated in sheep anti-Digoxigenin-AP , Fab fragments ( 1:1000 , 11093274910 , Sigma-Aldrich , St . Louis , MO ) in KTBT + 20% sheep serum + 2% Blocking Reagent overnight at 4°C . Sections were then washed x3 in KTBT for 30 min at room temperature , and then washed x2 in NTMT ( 0 . 1 M Tris , pH 9 . 5 , 0 . 1 M NaCl , 50 mM MgCl2 , 0 . 1% Tween 20 ) for 15 min . Sections were next incubated in NTMT + 1:200 NBT/BCIP Stock Solution ( 11681451001 , Sigma-Aldrich , St . Louis , MO ) in the dark at room temperature until color appeared . Sections were then washed in PBS , post-fixed in 4% PFA in PBS for 15 min and washed x2 in PBS . Finally , sections were dehydrated in 30% and then 70% methanol , 5 min each , followed by 100% methanol for 15 min . Sections were then rehydrated in 70% and 30% methanol and then PBS , 5 min each , and mounted in 95% glycerol . Whole mount: cochleae were incubated in PBS + 0 . 5% Tween-20 ( PBSTw ) for 1 h to permeabilize . Cochleae were then blocked using PBSTw + 5% donkey serum for 1 h and then incubated in PBSTw + 1% donkey serum with the primary antibody overnight at 4°C . Cochleae were then washed x3 in PBS and incubated in PBS + 1% Tween-20 with the secondary antibody . After wash in PBS x3 , cochleae were mounted in 95% glycerol with the sensory epithelium facing up . Frozen slides were warmed for 30 min at room temperature and washed in PBS before incubating in PBS + 0 . 5% Triton X-100 ( PBST ) for 1 h to permeabilize the tissue . Sections were then blocked using in PBST + 5% donkey serum for 1 h and then incubated in PBST + 1% donkey serum with the primary antibody overnight at 4°C in a humidified chamber . Sections were then washed x3 in PBS and incubated in PBS + 1% Triton X-100 with the secondary antibody . After wash in PBS x3 , slides were mounted in VectaShield antifade mounting medium with DAPI ( H-1200 , Vector Labs , Burlingame , CA ) . Primary antibodies: rabbit polyclonal anti-P75NTR ( 1:300 , AB1554 , EMD Millipore ) , rabbit polyclonal anti-Prox1 ( 1:1000 , ABN278 , EMD Millipore ) , goat polyclonal anti-Sox2 ( 1:200 , sc-17320 , Santa Cruz Biotechnology ) , rabbit polyclonal anti-p27Kip1 ( 1:50 , RB-9019-P , Neomarkers ) , rabbit polyclonal anti-Ki67 ( 1:200 , ab15580 , Abcam ) , rabbit polyclonal anti-Myo6 ( 1:100 , sc-50461 , Santa Cruz Biotechnology ) . Secondary antibodies: donkey polyclonal anti-Rabbit IgG , Alexa Fluor 488 ( 1:500 , A-21206 , Thermo-Fisher Scientific ) , donkey polyclonal anti-Goat IgG , Alexa Fluor 555 ( 1:500 , A-21432 , Thermo-Fisher Scientific ) , goat polyclonal anti-Rabbit IgG , Alexa Fluor 555 ( 1:500 , A-21428 , Thermo-Fisher Scientific ) , donkey polyclonal anti-Rabbit IgG , Alexa Fluor 594 ( 1:500 , A-21207 , Thermo-Fisher Scientific ) . Other compounds: Alexa Fluor 488-conjugated Phalloidin ( 1:50 , A12379 , Invitrogen ) . EdU ( E10187 , Thermo-Fisher Scientific , Waltham , MA ) was injected i . p . into pregnant dams at 100 μg per gram body weight . Embryos were harvested at 1 h after injection . EdU was detected using the Click-iT EdU Alexa Fluor 488 kit ( C10337 , Thermo-Fisher Scientific , Waltham , MA ) according to manufacturer’s instructions . Brightfield microscopy was done using a Hamamatsu NanoZoomer slide scanning system with a 20x objective . Images were processed with the NanoZoomer Digital Pathology ( NDP . view2 ) software . Fluorescent microscopy was done using a Zeiss LSM 700 confocal or Zeiss Axio Imager Z1 with Apotome 2 , with z-stack step-size determined based on objective lens type ( 10x or 20x ) , as recommended by the ZEN software ( around 1 μm ) . Fluorescent images shown are maximum projections . Low magnification fluorescent images shown of the whole cochlear duct required stitching together , by hand , several images . Images were processed with ImageJ ( imagej . nih . gov ) . Measurements and cell quantification ( using the Cell Counter plugin by Kurt De Vos ) were done using ImageJ . Total cochlear duct length was defined as the length from the very base of the cochlea to the very tip of the apex , along the tunnel of Corti . Hair cells were identified via Phalloidin , which binds to F-actin [51] . Supporting cells ( SCs , including pillar cells and Deiters’ cells ) were identified based on positive labeling with both Prox1 [52] and Sox2 [10] . Inner hair cells ( IHCs ) were differentiated from outer hair cells ( OHCs ) based on their neural/abneural location , respectively , relative to p75NTR-expressing inner pillar cells [53] . For total cell counts , IHCs , OHCs , and SCs were counted along the entire length of the cochlea . Total cell counts were also normalized to cochlear length and presented as cell count per 100 μm of cochlea ( e . g . IHCs/100 μm ) . For cell quantification at the basal , middle , and apical turns of the cochlea , the cochlear duct was evenly divided into thirds , and total IHCs , OHCs , and SCs were quantified for each third and normalized to length . For the Fgf9-rescue experiments in Fig 3 , IHCs , OHCs , and SCs from at least 300 μm regions of the basal ( 10% ) , middle ( 40% ) , and apical ( 70% ) turns of the cochleae were counted and normalized to 100 μm along the length of the cochlear duct . In Sox2Ysb/- cochleae , p75NTR expression was mostly absent , resulting in sensory islands without p75NTR-expressing inner pillar cells . In these cochleae , HCs not associated with inner pillar cells were presumed to be IHCs during quantification . When a curved line was drawn connecting the p75NTR islands along the organ of Corti , these presumed IHCs were always neural ( inner ) to that line . All figures were made in Canvas X ( ACD systems ) . Data analysis was performed using the Python programming language ( python . org ) in Jupyter Notebook ( jupyter . org ) with the following libraries: Pandas ( pandas . pydata . org ) , NumPy ( numpy . org ) and SciPy ( scipy . org ) . Plotting was done using the Matplotlib library ( matplotlib . org ) . Statistics ( t-test , Mann-Whitney U test , one-way ANOVA , two-way ANOVA , and Fisher’s exact test ) were performed using the SciPy module Stats; Tukey’s HSD was performed using the Statsmodels package ( statsmodels . org ) . All comparisons of two means were performed using two-tailed , unpaired Student’s t-test . For comparisons of more than two means , one-way ANOVA was used , except in Fig 8 and S7 Fig , where two-way ANOVA was used , with the factors being Fgf20 ( levels: Fgf20+/+ , Fgf20+/- , Fgf20-/- ) and Sox2 ( levels: Sox2+/+ , Sox2Ysb/+ , Sox2Ysb/Ysb ) gene dosage . For significant ANOVA results at α = 0 . 05 , Tukey’s HSD was performed for post-hoc pair-wise analysis . In all cases , p < 0 . 05 was considered statistically significant . All statistical details can be found in the figures and figure legends . In all cases , each sample ( each data point in graphs ) represents one animal . Based on similar previous studies , a sample size of 3–5 was determined to be appropriate for our experiments . Error bars represent mean ± standard deviation ( SD ) . All numerical data underlying graphs can be found in S1 File . For qualitative comparisons ( comparing expression via immunofluorescence or RNA in situ hybridization ) , at least three samples were examined per genotype . All images shown are representative . Evaluation of onset of Myo6-expressing cells ( Fig 10B and 10C ) : 3 or 4 serial sections through the entire cochleae were immunostained for Myo6 and evaluated , blinded to genotype , for the presence of Myo6-expressing cells . E14 . 5 embryos were further stage-matched based on interdigital webbing of the hindlimb ( at E14 . 5 , roughly half of the hindlimb interdigital webbing is still present ) . Of the 34 embryos at E14 . 5 , 3 were removed from analysis due to lack of or minimal hindlimb interdigital webbing ( too old relative to the other embryos ) . | The mammalian cochlea contains the organ of Corti , a specialized sensory epithelium populated by hair cells and supporting cells that detect sound . Hair cells are susceptible to injury by noise , toxins , and other insults . In mammals , hair cells cannot be regenerated after injury , resulting in permanent hearing loss . Understanding genetic pathways that regulate hair cell development in the mammalian organ of Corti will help in developing methods to regenerate hair cells to treat hearing loss . Many genes are essential for hair cell and supporting cell development in the mouse organ of Corti . Among these are Sox2 , Fgfr1 , and Fgf20 . Here , we investigate the relationship between these three genes to further define their roles in development . Interestingly , we found that Sox2 and Fgf20 interact to affect hair cell and supporting cell development in a spatially-graded manner . We found that cells toward the outer compartment and the base of the cochlea are more strongly affected by the loss of Sox2 and Fgf20 . We provide evidence that this spatially-graded effect can be partially explained by the roles of the two genes in the precise timing of two sequential stages of organ of Corti development , specification and differentiation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"ears",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"cell",
"differentiation",
"endocrine",
"physiology",
"developmental",
"biology",
"organ",
"of",
"corti",
"inner",
"ear",
"growth",
"factors",
"fibroblast",
"growth",
"factor",
"endocrinology",
"biological",
"tissue",
"head",
"anatomy",
"cell",
"biology",
"phenotypes",
"cochlea",
"physiology",
"genetics",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cochlear",
"ducts"
]
| 2019 | Sox2 and FGF20 interact to regulate organ of Corti hair cell and supporting cell development in a spatially-graded manner |
Campylobacter jejuni is a leading cause of food-borne illness . Although a natural reservoir of the pathogen is domestic poultry , the degree of genomic diversity exhibited by the species limits the application of epidemiological methods to trace specific infection sources . Bacteriophage predation is a common burden placed upon C . jejuni populations in the avian gut , and we show that amongst C . jejuni that survive bacteriophage predation in broiler chickens are bacteriophage-resistant types that display clear evidence of genomic rearrangements . These rearrangements were identified as intra-genomic inversions between Mu-like prophage DNA sequences to invert genomic segments up to 590 kb in size , the equivalent of one-third of the genome . The resulting strains exhibit three clear phenotypes: resistance to infection by virulent bacteriophage , inefficient colonisation of the broiler chicken intestine , and the production of infectious bacteriophage CampMu . These genotypes were recovered from chickens in the presence of virulent bacteriophage but not in vitro . Reintroduction of these strains into chickens in the absence of bacteriophage results in further genomic rearrangements at the same locations , leading to reversion to bacteriophage sensitivity and colonisation proficiency . These findings indicate a previously unsuspected method by which C . jejuni can generate genomic diversity associated with selective phenotypes . Genomic instability of C . jejuni in the avian gut has been adopted as a mechanism to temporarily survive bacteriophage predation and subsequent competition for resources , and would suggest that C . jejuni exists in vivo as families of related meta-genomes generated to survive local environmental pressures .
The Gram-negative bacterium Campylobacter jejuni is now recognised as a major cause of human gastroenteritis worldwide [1] and has been linked to serious neurological sequelae such as Guillain–Barré syndrome and Miller–Fisher syndrome [2] . Poultry are considered a major source of C . jejuni infections in humans , though numerous other risk factors have been proposed , including the consumption of pork , barbequing , living or working on farms , working in slaughterhouses , seasonal changes in flying insect populations , travel abroad , and the consumption of raw milk [3–8] . Targeted control of food-borne pathogens generally requires identification of the major route of transmission and thereby the most effective place to control infection . For C . jejuni , however , the ubiquitous presence of the organism in the environment , and the sporadic nature of the disease , coupled with the inherent genetic heterogeneity , make the task of tracing of individual strains , and thereby the source of infection , extremely difficult [9 , 10] . The molecular mechanisms behind this extensive diversity are not fully understood . However , C . jejuni exhibits slip-strand mutation within homopolymeric tracts , which is thought to alter the expression of a significant number of genes [11] . The majority of these genes have been identified as being involved in the production of surface structures , including key fitness determinants such as motility [12–14] and lipo-oligosaccharide synthesis [15] . C . jejuni is also known to be naturally competent under environmental conditions [16 , 17] , though analyses of multi-locus sequence typing ( MLST ) profiles indicate that short lengths of DNA ( less than 3 kb ) are involved [18 , 19] . Intra-genomic recombination has been observed in C . jejuni [20 , 21] and C . fetus [22–24] but these events are reported to be highly localised and limited in size . Larger scale intra-genomic recombination events leading to genome diversity have , however , been reported for a wide range of other bacterial species [25–31] . As part of a study to investigate bacteriophage therapy and its impact on Campylobacter populations in poultry , we report that chromosomal inversions of up to 590 kb that include the origin of replication of C . jejuni arise in response to exposure to virulent bacteriophage . These inversions are associated with bacteriophage resistance , an inability to colonise chickens without reversion to bacteriophage sensitivity , and the production of a functional Mu-like bacteriophage . These data have profound implications with respect to the evolution of pathogen genomes under the strong and widespread pressure of bacteriophage predation in the environment , and the propagation of prophage under conditions in which host populations are falling .
Following bacteriophage CP34 treatment of chickens colonised by C . jejuni HPC5 , a series of CP34-insensitive isolates were recovered and examined by Loc Carrillo et al . [32] . The frequency of resistance was found to be 4% from the intestinal contents of these birds . Pulsed-field gel electrophoresis ( PFGE ) analysis of these isolates following SmaI digestion indicated that a number of these strains had novel PFGE macro-restriction profiles ( MRPs ) compared to the parent strain HPC5 , though some bands were clearly related . Two novel PFGE-MRPs were observed to be common , and an example of each type was selected for further analysis . These two isolates , R14 and R20 , shared five of seven SmaI bands with HPC5 , but contained two novel bands of approximately 420 kb and 240 kb in R14 and 170 kb and 125 kb in R20 ( Figures 1 and S1 ) . The combined size of the novel bands in R14 and R20 was approximately equal to those of the missing bands from HPC5 , indicating that gross loss or gain of genetic material was unlikely . To rule out the possibility of contamination , strains R14 and R20 were analysed by MLST . They were found to have identical MLST profiles to HPC5 ( type 356 ) , thus confirming their origin . R14 and R20 had indistinguishable growth characteristics in vitro compared to HPC5 , where the resistant phenotypes were stable for five passages representing at least 100 generations . To examine the stability of R14 and R20 in vivo , the strains were administered to chickens in the absence of bacteriophage CP34 . The colonisation potentials of R14 and R20 after 5 d following administration of log10 8 colony-forming units ( CFU ) of each strain were determined to be log10 7 . 1 ( ±0 . 5 ) CFU g−1 for R14 and log10 6 . 7 ( ±0 . 3 ) CFU g−1 for R20 , not significantly different from that of HPC5 ( log10 6 . 9 [±0 . 2] CFU g−1 ) . However , of the recovered isolates tested , almost all ( 98% ) had reverted to bacteriophage sensitivity ( n = 100 ) . Clearly , the resistance phenotype has a large fitness cost associated with it such that it is rapidly out-competed in chickens by a relatively low initial number of sensitive revertants . PFGE of revertant strain DNAs demonstrated all to possess MRPs different from those of their respective parent strains , where the SmaI fragments involved were those that discriminated R14 or R20 from HPC5 to link the phenotype of bacteriophage sensitivity to the observed genomic changes ( Figure S2 ) . The MRPs of the revertant strains fell into several distinct classes , termed R14-A , R14-B , R20-A , R20-B , and R20-C , with reference to their parent strains . There was an unequal distribution of the MRPs , with a preponderance of R14-B ( 75% of R14 revertants ) and R20-A isolates ( 80% of R20 revertants ) . To determine whether similar genomic alterations could be observed in vitro , a number of growth curves were performed using C . jejuni HPC5 and bacteriophage CP34 . After 24 h of growth , 91% of the isolates were resistant to CP34 ( n = 148 ) . However , genomic alterations were not observed in any of these , and it is assumed that resistance arose through a different mechanism , such as point mutation of the receptor . Motility assays indicated that these strains were essentially non-motile , identifying the flagella or motility as being involved in resistance . In contrast , strains R14 and R20 derived in vivo were as motile as the parent strain HPC5 . Binding assays were performed to determine whether the resistance observed in R14 and R20 was receptor mediated or an abortive infection . Bacteriophage CP34 was capable of binding to HPC5 , exhibiting a 98 . 7% drop in titre following a 90-min period of co-incubation . When incubated with either R14 or R20 , however , CP34 showed no reduction in titre at the end of the 90-min period ( Figure 2A ) . Initial colonisation experiments indicated that the bacteriophage-resistant strains R14 and R20 were compromised in their ability to colonise broiler chickens because they were found to revert to bacteriophage sensitivity . This was characterised further by examining the colonisation response of broiler chickens to a range of Campylobacter doses for each strain ( Figure 3 ) . These data show that when administered at higher doses , all of the strains tested achieved similar colonisation values at 48 h . However , at the lowest doses of log10 1 . 9 CFU and log10 2 . 8 CFU , the mean cecal colonisation values of R14 and R20 were determined to be log10 3 . 9 ( ±0 . 5 ) CFU g−1 and log10 3 . 7 ( ±0 . 6 ) CFU g−1 , respectively . These were significantly different ( p = 0 . 003 and p = 0 . 002 , respectively ) from those of the parent strain HPC5 ( log10 5 . 7 [±0 . 7] CFU g−1 ) and the revertant strains R14-A ( log10 6 . 3 [±0 . 5] CFU g−1 ) , R14-B ( log10 6 . 3 [±0 . 3] CFU g−1 ) , R20-A ( log10 5 . 8 [±0 . 7] CFU g−1 ) , R20-B ( log10 5 . 7 [±0 . 5] CFU g−1 ) , and R20-C ( log10 5 . 4 [±0 . 6] CFU g−1 ) . To identify the elements responsible for the MRP changes observed in the genomes of the HPC5 derivatives , a series of restriction maps were created for these strains . SmaI sites were located by the PCR amplification of genomic DNAs using primers designed on the basis of the SmaI sites present in the genome of C . jejuni NCTC11168 , followed by digestion of the PCR product with SmaI . The presence of SmaI sites within specific gene sequences were found not to vary between strains , and thus point mutation within the SmaI sites was discounted as the reason for the MRP changes . Once identified , digoxigenin ( DIG ) -labelled probes for the genes immediately adjacent to the SmaI sites were created along with probes spanning the SmaI sites and used for Southern hybridisation against transferred SmaI-PFGE DNA . This essentially created a range of SmaI restriction maps . Comparison of these maps for the various strains indicated that the genomes were essentially co-linear , but could be divided into sections . The polarity of these sections with respect to each other varied between the strains , indicating that genomic rearrangements involving considerable regions of the genome had occurred ( up to 590 kb in R14 and 220 kb in R20 ) . The polarities of the 540-kb and 100-kb SmaI fragments present in the R14 change and the 190-kb and 100-kb SmaI fragments in the R20 change were reversed ( Figure 1 ) . Since the SmaI sites are asymmetrically distributed , this affected the observed MRP . It is noticeable that the generation of R14 involves a rearrangement about the origin of replication ( located within the 130-kb SmaI fragment ) , whilst the rearrangement to generate R20 does not . This procedure was similarly carried out on the R14-A , R14-B , R20-A , R20-B , and R20-C strains to demonstrate that these strains had undergone further genomic rearrangements , all utilising the SmaI-PFGE bands that were observed to change in the R14 and R20 genomic profiles ( Figures 4 and S2 ) . However , the rearrangements observed were not all simple reversions to the HPC5 MRP; rather , the most common isolates ( R14-B and R20-A ) were the result of two separate events . These data indicate that the HPC5 lineage contains three genomic locations capable of recombining with each other . Free recombination between these locations would result in eight genome configurations derived from HPC5 . The strains described here represent four of eight of these potential arrangements . All of the rearrangements involve a central location within the 100-kb SmaI fragment of HPC5 , which limits the permutations possible to four , all of which are observed . This central location appears key to the generation of the counter-selective phenotypes of bacteriophage resistance and inefficient chicken colonisation that are selected upon exposure to virulent bacteriophage . To identify the sites of recombination , a system of chromosome walking using long-range PCR , Southern hybridisation , and direct sequencing from genomic DNA was developed . Using the genes adjacent to the SmaI sites as an anchor , long-range PCR was performed with the C . jejuni NCTC11168 genome as a guide . Once linked by PCR , DIG-labelled probes were created for individual genes to determine in which SmaI-PFGE band the gene was located . In places where the gene order in the HPC5 lineage diverged from that of NCTC11168 , a system of sequencing directly from genomic DNA and inverse PCR was employed to determine the identity of the adjacent sequences . These DNA sequences appear in Figures S3–S8 . The rearrangement end points were determined to be within copies of Mu-like prophages , similar to the prophage identified in C . jejuni RM1221 [33] . However , whereas RM1221 contained a single copy of the prophage , HPC5 and its daughter strains contained two complete copies and a partial copy ( ORFs CjE0227 to CjE0241 ) at distinct genomic locations . In HPC5 , the complete copies are between the 3′-end of Cj1470 ( CtsF ) and the 5′-end of Cj0167c in the 540-kb SmaI fragment ( designated CampMu-I ) , and between the 3′-end of Cj0167 and the 3′-end of Cj1470c ( CtsF ) in the 100-kb SmaI fragment ( CampMu-II ) , resulting in the disruption of both CtsF and Cj0167c . The partial copy is between an unknown gene and a paralog of CjE0225 in the 190-kb SmaI fragment ( CampMu-III ) . The R14 rearrangement involved recombination between the two complete copies of the CampMu prophage , whilst the R20 rearrangement involved recombination between CampMu-II and CampMu-III . Similarly , the rearrangements to create the R14 and R20 derivative strains took place within these CampMu prophage DNA sequences ( Figure 5A ) . The discovery that the inversion sites featured CampMu prophage sequences led to studies of whether the CampMu lysogens could be induced to liberate bacteriophage particles . However , it was determined that both R14 and R20 were capable of producing a CampMu bacteriophage without the need for induction . Bacteriophages were produced at a rate of approximately one particle per 50 cells ( R14 = 49 , R20 = 61 ) . These bacteriophages were examined by transmission electron microscopy ( Figure 5B ) and identified as corresponding to the CampMu prophage by PCR amplification of DNA extractions from the bacteriophage using CampMu primer pairs , but not from control C . jejuni 16s rDNA primers . Infectious CampMu bacteriophage particles could not be detected in supernatants from HPC5 or from the R14 and R20 revertant strains exhibiting phage sensitivity as tested by titration of the supernatants on all of the strains used in this study and a further panel of 139 independent C . jejuni isolates from broiler chickens , chicken meat , and humans . The strains capable of supporting the replication of the virulent bacteriophage CP34 ( HPC5 and the R14/R20-derived revertants ) were also capable of supporting replication of bacteriophage R14-CampMu and R20-CampMu whilst R14 and R20 were resistant to CP34 and the bacteriophage they produce . To compare whether the resistance observed with R14-CampMu and R20-CampMu was due to the failure of the phage to bind the bacterial host in a similar way to CP34 or to abortive infection , binding assays of the CampMu phage were performed under similar conditions ( Figure 2B and 2C ) . Bacteriophage R14-CampMu and R20-CampMu were capable of binding the progenitor host strain HPC5 , exhibiting approximately 90% reductions in phage titre after a 90-min incubation . R14-CampMu showed no reduction in phage titre when incubated with the R14 strain producing it , but incubation with R20 produced a 95% fall in phage titre . R20-CampMu showed no evidence of binding to either R14 or R20 . Further evidence for differences between R14-CampMu and R20-CampMu became apparent upon testing the susceptibility of a variety of C . jejuni strains , which revealed that the CampMu bacteriophage exhibited different host ranges ( Figure 5C ) , and that these were maintained following growth of the CampMu bacteriophage on susceptible strains not of the HPC5 lineage . It was also notable that R14-CampMu and R20-CampMu could replicate on independent strains carrying CampMu prophage genes ( Figures S9 and S10 ) .
One of the major fears concerning bacteriophage therapy is the potential for bacteriophage-induced genome evolution . Numerous examples exist where temperate bacteriophages are associated with virulence determinants , for example , the genes encoding the toxins of cholera , diphtheria , and verotoxigenic Escherichia coli [34–36] . However , it is generally assumed that using virulent bacteriophage will avoid this problem . This report indicates that virulent bacteriophage have the potential to activate dormant prophage , leading to rapid pathogen evolution; and via host recombination the evolution of temperate bacteriophage leading to the production of chimeric phage with novel phenotypes . However , we also show that whilst pathogen evolution can be rapid , resistance to the therapeutic bacteriophage is associated with a draconian fitness cost that renders the resistant strains non-competitive in the absence of the bacteriophage . Clearly though , the primary benefit of bacteriophage therapy in this instance is to temporarily reduce the carriage of C . jejuni rather than to eliminate it . Indeed , the ability of C . jejuni to enter what is effectively a transient survival state is evidence of the unusual measures Campylobacter can employ to survive environmental pressures . Recent evidence suggests that Campylobacter is limited in stress response mechanisms [37] and can use genome alterations such as localised frame-shift mutations and slip-strand phase variation to modify gene expression as a substitute for the maintenance of structured regulatory mechanisms [11–15] . The evidence presented here indicates that C . jejuni can use specific genome inversions to survive adverse ecological conditions . Under these conditions , any given Campylobacter recovered is actually but a single representative of a larger family of related meta-genomes under continual flux , the relative proportions of which are dictated by local environmental pressures . Amongst derivate genomes are those in which the origin of replication has been inverted , which could give rise to yet wider changes in gene regulation . This form of chaotic genome regulation is a striking example of the extraordinary strategies adopted by C . jejuni to survive . This type of genomic scale regulation would also suggest that complementary typing methods are required to adequately differentiate C . jejuni strains; methods should be selected that sample the whole genome in parallel with those that are highly discriminatory for smaller sections of the genome . In this example , strains R14 and R20 cannot be differentiated from HPC5 by MLST alone despite the large phenotypic differences observed . A combination of MLST and PFGE methods are required to distinguish these strains and identify them as being different but closely related . Intra-genomic rearrangements have been reported previously for the flagellin locus of C . jejuni [20 , 21] and the sap locus of C . fetus [22–24] . However , these rearrangements are relatively short ( <5 kb ) and highly localised , utilising areas of sequence homology ( flagellin ) or specific recombination pathways ( sap locus ) . The genome sequence data available for C . jejuni are notable for their lack of repeated sequences , and the completed genomes of NCTC11168 and RM1221 are essentially co-linear; therefore , it would not be unreasonable to suggest that genome rearrangements of C . jejuni are either limited or , given the idea of Campylobacter as a meta-genomic organism , that the observed genome organisations are optimal for in vitro cultures . However , changes to PFGE-MRPs have been noted elsewhere [17 , 38 , 39] , indicating that chromosomal rearrangements are possible for strains carrying repeated sequences as substrates for homologous recombination such as the prophage sequences documented here . The observation that the R14 and R20 rearrangements occur in vivo rather than the generation of resistance through mutation of the receptor or a specific binding component is likely a consequence of the essential nature of these components . The frequency at which bacteriophage-resistant mutants are generated in vitro ( 91% ) with HPC5 suggests that there are easier paths to escape bacteriophage predation . However , all the mutants selected in vitro were impaired in motility . Flagella components have been demonstrated to be dominant colonisation factors [40–42] , and thus it is not surprising that resistant isolates lacking motility do not survive long in chickens . Bacteriophage CP34 appears to have selected an essential component of the bacteria's intestinal lifecycle , where dense host populations are likely to be most abundant . C . jejuni flagellin is known to be polymorphic and variably glycosylated , leading to differences in sero-specificity [43–46] . Bacteriophage predation may be the direct driving force behind the development of such antigenically variable flagellins rather than host immune evasion , as considered previously [47] . The spontaneous production of CampMu bacteriophages following bacteriophage therapy is of concern because Mu bacteriophages are potential agents of mutation . However , the influence of potential mutator phage needs to be considered against the mutation-driven lifestyle of C . jejuni , which does not carry a full complement of DNA repair mechanisms in the expectation that genomic variation will modify gene expression to overcome adverse conditions . Moreover , evidence suggests that Campylobacter populations are already exposed to CampMu bacteriophage through the mechanism outlined here . Virulent bacteriophages of the family Myoviridae , like CP34 , are common in chickens harboring campylobacters . Isolation rates of around 20% in United Kingdom conventional broiler flocks , and more frequently in environmentally exposed free-range and organic flocks , have been reported [48–50] . A recent survey of C . jejuni and C . coli isolates found that 19 of 67 and two of 12 of the respective isolates contained at least one prophage gene [51] . This corresponds well with the four of 12 positive C . jejuni strains reported here . If these figures are representative of general Campylobacter populations , then the likelihood is that these processes are quite common . The recombination events leading to the strain variants reported here are centred on a 9-kb region of DNA sequence that is shared between prophages CampMu-I , -II , and -III ( genes CjE0227 to CjE0241 ) . Recombination between CampMu-I and CampMu-II gives rise to the R14 genome that produces bacteriophage R14-CampMu , and recombination between CampMu-III and CampMu-II gives rise to the R20 genome that produces R20-CampMu . These events lead to the generation of chimeric CampMu prophage in which the genes CjE0242 to CjE0273 adjacent to the recombination resolution of R14 are exchanged , and genes CjE0215 to CjE0226 adjacent to the recombination resolution of R20 are exchanged . These exchanges enable functional excision of CampMu bacteriophages with different gene contents that are themselves distinguishable by their Campylobacter host range ( Figure 5C ) . What these events have in common is that they lead to resistance to the virulent phage CP34 that is unable to bind the host bacterium . The non-binding of CP34 may arise through two potential mechanisms: 1 ) changes in host surface structures that are required for phage adsorption; or 2 ) receptor saturation , if CP34 shares a receptor recognition site with the CampMu phages , and these sites are saturated in the R14 and R20 cultures that produce them . Changes in the surface structures of host bacteria leading to bacteriophage immunity often accompany the state of lysogeny and are mediated through the acquisition of additional genes , commonly known as morons , the control of which are generally independent of the regulation of prophage within which they are sited [52] . In the case of R14 and R20 , the change in surface structure expression would have to be associated with the activation of the prophage , for although a wider set of genes other than those affected by the recombination resolution site could be influenced by the gross inversions , the second site reversion events that reinstate phage sensitivity as a consequence of the chromosome rearrangements would militate against the inversions themselves being responsible for the change in phenotype . A consequence of the prophage control of surface structures is that campylobacters carrying CampMu may be biased for certain phage types through the regulated expression of their receptors , or indeed for specific environments according to the need of the organism to express these surface structures . In the latter case , there would be strong selective pressure to inactivate the prophage even at the expense of inverting significant parts of the genome to reassert the control necessary to respond to alternative environments . Bacteriophage R14-CampMu can bind strain R20 , suggesting the receptor site for the phage is still available on this strain , but despite this the R14-CampMu phage does not form plaques on R20 , indicating there is likely an underlying resistance mechanism that results in abortive infections . However , neither bacteriophage R14-CampMu nor R20-CampMu are able to bind the respective C . jejuni strains that produced them , and therefore receptor saturation remains a plausible mechanism by which R14 and R20 prevent super-infection . Considering the above , it is of interest to contemplate how multiple prophage copies have been fixed within the HPC5 lineage . It is a general tenet that the state of lysogeny renders the bacteria immune to infection by homologous bacteriophage . Therefore , it is unlikely that the naïve HPC5 precursor was lysogenised by multiple copies of the same bacteriophage , though not of course impossible . Indeed , it is more likely that a single prophage was present and replicated itself by transposition during the first stages of prophage lytic multiplication . It is known that the position of DNA replication forks influences the location of transposition [53] , and the equidistant spacing of the CampMu-I and CampMu-II copies about the origin of replication may indicate that these prophage inserted here as a result of the presence of replication forks symmetrically arranged around the origin . This does lead to the question as to why the replication of the putative CampMu phage was not carried through to completion , namely , the lysis of the host cell . A potential answer to this is a recombination event . The R14 genome structure is similar to that of C . jejuni NCTC11168 and RM1221 , whereas HPC5 has a section of genome of reversed polarity . It is possible the R14 configuration represents the original bacteriophage-negative progenitor that became lysogenised by CampMu . When this prophage began to replicate by transposition , the sudden presence of extensive regions of homology allowed recombination within the genome of this strain . Presumably , this recombination led to strain HPC5 , where the CampMu prophage was inactivated , and the cell survived . This is supported by the fact that the CampMu cannot be recovered from HPC5 , and yet frequently exits the cell in R14 and R20 , suggesting that in HPC5 , the Mu is inactivated . If true , this is another example of how flexibility within the C . jejuni genome has enabled it to survive the induction of a lysogenic bacteriophage that should have resulted in cell death , and to capitalise on the outcome through evasion of virulent bacteriophage . The genes present in the partial copy of the CampMu prophage ( CampMu-III ) have previously been identified as being present en masse in a variety of Campylobacter [51] that lack the other prophage genes ( CjE0215 to CjE0226 and CjE0242 to CjE0273 ) . Analysis of the unique sequences adjacent to CampMu-III in HPC5 indicates that these have similarity to bacteriophage genes from other sources , most notably to a phage major tail tube protein from C . jejuni 260 . 94 . It would appear that genes CjE0227 to CjE0241 represent a module of a CampMu genome ( CampMu-III ) comprising a central region similar to that of the RM1221 CampMu , but flanked by novel prophage genes . Recombination between prophage genomes leads to exchange of these modules and the evolution of the prophage genome . Intra-chromosomal recombinations between the prophage in HPC5 are a direct example of such events , producing chimeric bacteriophage that can exploit differing host ranges .
Campylobacters were cultured on blood agar plates ( blood agar base No . 2 with 5% defibrinated horse blood; Oxoid , http://www . oxoid . com/ ) in gas jars under microaerobic conditions ( 5% O2 , 85% N2 , 10% CO2 ) at 42 °C for 24 to 48 h . Growth curves were conducted by inoculating log10 7 CFU of the C . jejuni into 100 ml of nutrient broth No . 2 ( Oxoid ) and incubating at 42 °C under microaerobic conditions with 100 rpm orbital rotation . Bacteriophage CP34 was propagated on C . jejuni HPC5 and recovered using a plate lysis method and stored at 4 °C in SM buffer [32] . Bacteriophage R14-CampMu and R20-CampMu were recovered from blood agar plate cultures of either C . jejuni R14 or R20 by swabbing into SM buffer and passage through a 0 . 2-μm filter to remove bacteria . Testing of Campylobacter strain susceptibility to bacteriophage was performed as described previously [48] . The susceptibility of Campylobacter strains to bacteriophage R14-CampMu and R20-CampMu was tested by growth of the appropriate Campylobacter strain in 100 ml of nutrient broth No . 2 in the presence of log10 3 plaque-forming units ( PFU ) ml−1 . Samples were recovered and bacteriophage enumerated before and after growth for 24 h at 42 °C on HPC5 . Bacteriophage binding assays were performed to determine whether insensitivity to bacteriophage was due to surface or intracellular factors . Overnight Campylobacter growth from blood agar plates was swabbed into nutrient broth No . 2 , centrifuged at 13 , 000g for 1 min , and the cell pellet resuspended in nutrient broth No . 2 . The cells were washed in this manner twice more and , upon final resuspension , were adjusted to contain log10 10 CFU ml−1 as estimated from OD600 . Bacteriophage was added at concentrations of log10 4–5 PFU ml−1 to the Campylobacter suspension and incubated at 42 °C with 100 rpm shaking under aerobic conditions for 90 min . Samples were taken at 0 and 90 min , filtered through a 0 . 2-μm filter , and stored at 4 °C until enumeration of the bacteriophage . Campylobacter-free Ross broiler chickens were used to determine the colonisation of different Campylobacter strains in the presence and absence of bacteriophage . To ensure that the experimental birds remained free of naturally occurring infection , faeces and cloacal swabs were taken each day from hatch and tested for Campylobacter by direct plating on CCDA agar and for Salmonella by enrichment in Rappaport–Vassiliadis soya peptone broth ( Oxoid ) , then plating on xylose-lysine desoxycholoate agar ( Oxoid ) . Birds were dosed with Campylobacter at 21 d of age and with bacteriophage where applicable at 25 d of age . Following sacrifice , the ceca , upper ( proximal small intestine ) and lower intestines of the birds were first separated by ligature and then removed by sterile dissection . The lumenal contents were collected for Campylobacter and bacteriophage isolation as described previously [32] . MLST was performed as described previously [54] with reference to the C . jejuni MLST database ( http://pubmlst . org/campylobacter/ ) to determine the sequence alleles . PFGE was carried out on SmaI-digested genomic DNA and compared to the known profiles of the test strains [55] . Campylobacter DNA isolation was carried out by using GenElute Bacterial Genomic DNA purification kit ( Sigma-Aldrich , http://www . sigmaaldrich . com/ ) or Wizard Genomic DNA purification kit ( Promega , http://www . promega . com/ ) . Bacteriophage genomic DNA isolation was performed according to standard procedure [32] using proteinase K digestion followed by phenol-chloroform extraction and precipitation . Oligonucleotide primers were designed using the NCTC11168 and RM1221 sequences ( Sigma-Genosys , http://www . sigmaaldrich . com/Brands/Sigma_Genosys . html ) . A list of primers used in this study can be found in Table S1 . PCRs were performed in 50-μl volumes using a Techne Progene thermal cycler . Reactions consisted of 2 . 5 U AccuTaq DNA polymerase ( Sigma-Aldrich ) , each dNTP at 500 μM ( Promega ) , forward and reverse primers at 400 nM each , 2% v/v dimethyl sulphoxide , and 100–500 ng of genomic DNA as template in AccuTaq DNA polymerase buffer . DIG-labelled probes for Southern hybridisation were synthesised by PCR with the replacement of the 400 μM dTTP with dTTP at 368 μM and DIG-11-dUTP at 32 μM ( Roche , http://www . roche . com/ ) . Sequencing of PCR products was carried out by MWG Biotech AG ( http://www . mwg-biotech . com/ ) using the ValueRead system . Direct sequencing of genomic DNA was achieved using the same system but with 20 μg of genomic DNA prepared using the Wizard Genomic DNA purification kit . DNA fragments separated in PFGE gels were transferred to Hybond N+ nylon membranes ( Amersham Biosciences , http://www . gelifesciences . com/ ) using the capillary method . Hybridisation probes were synthesised by PCR as described above . Hybridisations were performed overnight at 42 °C using DIG Easy Hyb . Buffer ( Roche ) . The membranes were blocked using 1% blocking reagent ( Roche ) before antibody binding with 150 mU ml−1 anti-DIG-AP in 1% blocking reagent . Colour development was performed by incubation in 100 mM tris-HCl ( pH 9 . 5 ) , 100 mM sodium chloride , 0 . 45 mg ml−1 nitro-blue tetrazolium chloride , and 0 . 175 mg ml−1 5-bromo-4-chloro-3-indolyl-phosphate , 4 toluidine salt . Bacteriophage particles at log10 8 PFU ml−1 were absorbed onto a glow-discharged carbon-coated Pioloform grid and stained with uranyl acetate . These were examined using a JEOL 100CX transmission electron microscope ( http://www . jeol . com/ ) operating at an acceleration voltage of 80 kV . C . jejuni strains were grown on blood agar overnight . A loop of bacteria was inoculated into the centre of a motility plate ( Mueller–Hinton broth with 0 . 4% agar ) and grown micro-aerobically for 24 h . Motility was assessed as a function of the radius of the motility halo .
DNA sequences associated with this manuscript appear in the following supplementary figures with the corresponding GenBank ( http://www . ncbi . nlm . nih . gov/Genbank/index . html ) accession numbers: Figure S3 , Cj1468-CjE0213 ( EF581842 ) ; Figure S4 , ORF0656-Cj0167 ( EF581846 ) ; Figure S5 , miaA-CjE0213 ( EF581841 ) ; Figure S6 , ORF0656-Cj1470 ( EF581845 ) ; Figure S7 , Unk-CjE0241 ( EF581844 ) ; Figure S8 , Unk-CjE0227 ( EF581843 ) . | Campylobacter jejuni is the major cause of bacterial food-borne illness worldwide . Predation of C . jejuni by virulent bacteriophage offers the prospect of controlling bacterial populations at source in poultry . We report that in chickens , bacteriophage resistance is infrequent because the mutants that escape bacteriophage are not proficient in poultry colonisation but readily revert back to colonisation-proficient phage-sensitive types . Bacteriophage resistance is generated by reversible genomic scale inversions , leading to the activation of an unrelated bacteriophage integrated into the bacterial genome . These data not only suggest that bacteriophage therapy of C . jejuni would remain a sustainable measure to reduce poultry contamination but also demonstrate how bacterial genomes can evolve under the strong and widespread pressure of bacteriophage predation in the environment . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
]
| [
"viruses",
"infectious",
"diseases",
"chicken",
"in",
"vitro",
"microbiology",
"evolutionary",
"biology",
"genetics",
"and",
"genomics",
"eubacteria"
]
| 2007 | Genome Dynamics of Campylobacter jejuni in Response to Bacteriophage Predation |
The functional interpretation of high throughput metabolomics by mass spectrometry is hindered by the identification of metabolites , a tedious and challenging task . We present a set of computational algorithms which , by leveraging the collective power of metabolic pathways and networks , predict functional activity directly from spectral feature tables without a priori identification of metabolites . The algorithms were experimentally validated on the activation of innate immune cells .
Knowledge of many metabolic pathways has accumulated over the past century . For instance , glycolysis , citric acid cycle and oxidative phosphorylation fuel cellular processes through the generation of adenosine triphosphate; glycans and cholesterols not only serve as structural blocks but also mediate intercellular communication . In fact , metabolites pervade every aspect of life [1] , [2] . Their roles are increasingly appreciated , as advancing tools allow deeper scientific investigations . The most notable progresses in recent years come from metabolomics and genome-scale metabolic models . Metabolomics is the emerging field of comprehensive profiling of metabolites . As metabolites are the direct readout of functional activity , metabolomics fills in a critical gap in the realm of systems biology , complementing genomics and proteomics [3]–[6] . The technical platforms of metabolomics are mainly based on mass spectromety and nuclear magnetic resonance [4] , [7] . Among them , untargeted LC/MS ( liquid chromatography coupled mass spectrometry ) , especially on high resolution spectrometers , produces unparalleled throughput , measuring thousands of metabolite features simultaneously [5] , [8]–[10] . On the other hand , genome-scale metabolic models have been largely driven by genomics , as the total list of metabolic enzymes of a species can be derived from its genome sequence [11] , [12] . The reconstruction of microbial metabolic network models is an established process [13] , [14] . Intense manual curation , however , was required in the building of two high-quality human models [15] , [16] , which were followed by a number of derivatives [17]–[20] . The coverage of these metabolic models greatly exceeds the conventional pathways . Even though they are a perfect match in theory , metabolomics and genome-scale metabolic models have had little overlap so far . The use of metabolomics in building metabolic models is rare [21] , due to the scarcity of well annotated metabolomics data . The application of genome-scale metabolic models to metabolomics data is not common either [22] . The limited throughput of targeted metabolomics usually does not motivate large scale network analysis . Untargeted metabolomics cannot move onto pathway and network analysis without knowing the identity of metabolites . A typical work flow of untargeted metabolomics is illustrated in Figure 1A . After ionized molecules are scanned in the spectrometer , the spectral peaks are extracted , quantified and aligned into a feature table . At this point , each feature is identified by a mass-to-charge ratio ( m/z ) and retention time in chromatography , but its chemical identity is not known . To assign a spectral feature to a bona fide metabolite , it usually involves tandem mass spectrometry to examine the fragmentation pattern of a specific feature , or coelution of isotopically labeled known references - both are inherently low throughput . Considerable effort is needed to build a spectral library , which is often of limited size and interoperability . Thus , metabolite identification forms the bottleneck of untargeted metabolomics [23] . A number of informatics tools have been developed for LC/MS metabolomics , ranging from feature extraction [24]–[26] , pathway analysis and visualization [27]–[29] to work flow automation [30]–[32] . Yet , whereas pathway and network analysis is concerned , the existing tools require identified metabolites to start with . Computational prediction of metabolite identity , based on m/z alone , is deemed inadequate as a single m/z feature can match multiple metabolites even with high instrumental accuracy [33] , [34] , and multiple forms of the same metabolite often exist in the mass spectra [35] . Although automated MS/MS ( tandem mass spectrometry ) search in databases is improving the efficiency of metabolite identification [36] , [37] , this requires additional targeted experiments and relies on extensive databases , where data from different platforms often do not match . How to bring untargeted metabolomics data to biological interpretation remains a great challenge . In this paper , we report a novel approach of predicting network activity from untargeted metabolomics without upfront identification of metabolites , thus greatly accelerating the work flow . This is possible because the collective power in metabolic networks helps resolve the ambiguity in metabolite prediction . We will describe the computational algorithms , and demonstrate their application to the activation of innate immune cells .
The genome-scale human metabolic network in mummichog is based on KEGG [38] , UCSD Recon1 [15] and Edinburgh human metabolic network [16] . The integration process was described previously [39] . The organization of metabolic networks has been described as hierarchical and modular [40] . When we perform a hierarchical clustering on the metabolic reactions in our network , its modular structure is clear ( Figure 2A ) . This modular organization , as reported previously [41] , often but not always correlates with conventional pathways ( Figure 2B ) . The module definition in this work is adopted from Newman and Girvan [42] , [43] , where a module is a subnetwork that shows more internal connections than expected randomly in the whole network . Modules are less biased than pathways , which are defined by human knowledge and conventions , and outgrown by genome-scale metabolic networks . Activity of modules may exist within and in between pathways . Deo et al [22] convincingly demonstrated the advantage of unbiased module analysis over pathways . On the other hand , pathways have built-in human knowledge , which may be more sensitive under certain scenarios . Pathway analysis and module analysis are rather complementary , and both are included in mummichog . The reference metabolic network model contains both metabolites and enzymes . Since metabolomics only measures metabolites , the model is converted to a metabolite centric network for statistical analysis . Enzymes are only added later in the visualization step to aid user interpretation . Within the predefined reference metabolic network model , mummichog searches for all modules that can be built on user input data , and compute their activity scores . This process is repeated many times for the permutation data to estimate the background null distribution . Finally , the statistical significance of modules based on user data is calculated on the null distribution . The specific steps are as follows: The basic test for pathway enrichment here is Fisher's exact test ( FET ) , which is widely used in transcriptomic analysis . The concept of FET is , when we select features ( ) from a total of features ( ) , and find of the features present on a pathway of size , the chance of getting in theory can be estimated by enumerating the combinations of , and . To apply FET to an enrichment test of metabolic features on pathways , we need to understand the additional layer of complexity . Our metabolic features can be enumerated either in the m/z feature space or in the metabolite ( true compound ) space . Since metabolic pathways are defined in the metabolite space , either way needs to factor in the many-to-many mapping between m/z features and metabolites ( Figure S1 ) . This mapping is effectively covered in our permutation procedure , which starts from the m/z features and reruns the mapping every time . The overall significance of a pathway enrichment is estimated based on a method by Berriz et al [44] , which ranks the p-value from real data among the p-values from permutation data to adjust for type I error . Finally , a more conservative version of FET , EASE , is adopted to increase the robustness [45] . The key idea of EASE is to take out one hit from each pathway , thus preferentially penalize pathways with fewer hits . The specific steps are as follows: Both the module analysis and pathway analysis above serve as a framework to estimate the significance of functional activities . In return , the predicted metabolites in significant activities are more likely to be real . Mummichog collects these metabolites , and look up all their isotopic derivatives and adducts in . A confidence rating system is applied to filter for qualified metabolites . For instance , if both the single-charged form M+H[1+] and the form M ( C13 ) +H[1+] are present , this metabolite prediction carries a high confidence . All the qualified metabolites carry over their connections in the reference metabolic network , and form the “activity network” for this specific experiment ( e . g . Figure 3 ) . The activity network gears towards a quality and clear view of user data , as modules and pathways can be redundant and fragmented . It also accentuates the activity in a specific experimental context , in contrast to the generic nature of the reference metabolic network . We next illustrate the application of these algorithms to a novel set of immune cell activation data , and two published data sets on human urinary samples and yeast mutants . The innate immunity plays a critical role in regulating the adaptive immunity , and the field was recognized by the 2011 Nobel Prize in Physiology or Medicine [46] . According to the nature of stimuli , innate immune cells direct different downstream molecular programs , which are still under intense scientific investigation [47] , [48] . In this study , we examine the metabolome of human monocyte-derived dendritic cells ( moDC ) under the stimulation of yellow fever virus ( YF17D , a vaccine strain ) . We have shown previously that yellow fever virus activates multiple toll-like receptors , and induces cellular stress responses [49]–[51] . This data set is , to our knowledge , the first high throughput metabolomics on any immune cells ( macrophages were previously studied by limited throughput ) . The cell extracts from our activation experiment were analyzed by LC/MS metabolomics , and yielded 7 , 995 spectral features ( denoted as ) after quality control . Among them , 601 features were significantly different between the infected samples and both the baseline and time-matched mock controls ( , student t-test; denoted as ) . Using and , mummichog computes significant pathways and modules and the activity network . Viral infection induced a massive shift of metabolic programs in moDCs ( pathways in Table S1 , modules in Figure S2 ) . The predicted activity network is shown in Figure 3A , and we will focus our investigation on a small subnetwork ( Figure 3B ) , which includes the metabolisms of nucleotides , glutathione/glutathione disulfide and arginine/citrulline . Nucleotides are required for viral replication , and the hijacking of host nucleotide metabolism by virus has been well described [52]–[54] . Glutathione is best known as intracellular antioxidant , where it is oxidized to glutathione disulfide ( GSSG ) . However , our data show that both glutathione and GSSG are depleted in activated moDCs , departing from this conventional wisdom . The across-the-board depletion is consistent with the down-regulation of genes for glutathione synthesis ( Figure 4B ) . Our data support the notion that glutathione is released by dendritic cells and conditions the extracellular microenvironment during their interaction with T cells [55]–[57] . Arginine is known to be an important regulator in innate immune response [58] , [59] . Arginine metabolism can lead to two pathways: to ornithine ( catalyzed by arginase ) or to citrulline ( catalyzed by nitric oxide synthase ) . The decrease of arginine and increase of citrulline suggests the latter pathway , which is the main reaction of producing intracellular nitric oxide . We indeed detected the inhibition of eNOS and iNOS expression later ( Figure 4C ) , a well documented feedback effect by nitric oxide [60] . We also performed tandem mass spectrometry on the metabolites in Figure 3B , using authentic chemicals as references . All the metabolites , except glutamylcysteine and thyroxine , were confirmed ( Figure 5 , Figure S3 ) . The depletion of arginine and accumulation of citrulline in moDC was also triggered by dengue virus but not by lipopolysaccharide ( LPS , Figure S4 ) . It is known that iNOS is induced in dendritic cells by LPS but not by virus [47] , [61] . Our data suggest a different nitric oxide pathway in viral infection , driven by constitutive nitric oxide synthases . The intracellular nitric oxide has a fast turnover and we did not detect its accumulation by fluoremetric assays ( data not shown ) . We previously demonstrated that the phosphorylation of EIF2A was induced by YF17D [50] . An upstream mechanism is now suggested by this metabolomic experiment , as both the production of nitric oxide and depletion of arginine induce the activity of EIF2A kinases [62] . The nature of metabolomics data often varies by platforms and sample types . We thus extend our mummichog approach to two published data sets on human urinary samples [63] and on yeast cell extracts [64] . Both data sets carry metabolite annotation by the original authors , which can be used to evaluate the prediction by mummichog . The human urinary data contained both formal identification by matching to local library of chemical references and putative identification by combining multiple public resources [63] . We used mummichog to investigate the gender difference in this data set , and predicted an activity network of 45 metabolites . Among them , 13 were not found in the original annotation . For the remaining metabolites , 97% ( 31/32 ) were agreed between mummichog and the original annotation ( Figure 6 ) . There is an option in mummichog to enforce the presence of M+H[+] form ( for positive mode , M−H[−] for negative mode ) in metabolite prediction . With this option , 3 out of 44 metabolites were not in the original annotation , and the remaining 41 metabolites were in 100% agreement . The mummichog algorithms are not tied to a specific metabolic model . We adopted the yeast metabolic model from BioCyc database [11] for the yeast data [64] , to predict an activity network contrasting mutant and wild type strains . This data set was only annotated for 101 metabolites through the authors' local library . As a result , the majority of metabolites in the predicted network by mummichog were not found in the original annotation . Out of the remaining 28 metabolites , 24 ( 86% ) were agreed between mummichog and the original annotation ( Figure 6 ) . Enforcing the presence of primary ion M−H[−] ( data collected in negative ion mode ) had little effect to the result , since the original annotation was already biased to metabolites that are ionized easily . These results show that the prediction by mummichog is robust cross platforms and sample types .
Critical to the success of mummichog is the integration of genome-scale metabolic models . We have used in this study a recent human metabolic model . An alternative human model from BioCyc [11] produced comparable results ( Figure S6 ) . The coverage of the models in all three case studies is shown in Table 1 . These genome-scale metabolic models are more extensive than conventional pathways , and were shown to capture activities in between pathways [22] . The pathway organizations differ between the two human models , as the BioCyc model tends to use smaller pathways . This creates some model dependency in the pathway analysis , but little effect to the “activity network” , as mummichog is more network centric . The two test cases in Figure 6 also indicate that these models tend to capture more information than conventional annotations . However , as mentioned earlier , the new data from metabolomic studies are yet to be integrated into these genome-scale metabolic models . For example , a number of metabolites in metabolomics databases [36] , [65] , [66] are not in any of these metabolic models . In general , the features from a high resolution profiling experiment by far exceed the current annotations in metabolite databases . This leads to a de facto filtering when data are run on mummichog ( similiar situation in database searches ) . Meanwhile , the features that can be mapped to the current metabolic model are more likely to be biologically relevant . This “filtering” is pertinent to the metabolic model , not to mummichog algorithms - mummichog still has to choose the true metabolites from multiple possible candidates ( Figure S1B ) . It will be an important future direction to advance metabolic modeling with the chemical data . We also expect the metabolic models to improve on lipid annotation , physiological context and tissue specificity . As lessons learned from transcriptomics , pathway and network analysis not only provides functional context , but also the robustness to counteract noises at individual feature level , which are commonly seen in omics experiments . Similarly , the prediction on activity by mummichog is tolerant to errors at individual feature level . In the moDC data , we chose by a cutoff value . When we vary this cutoff from to , the program returned networks of a stable set of metabolites ( Figure S7 ) . The module finding procedure in the program was designed to extensively sample subnetwork structures . Among the modules will be many variations , but the subsequenct “activity network” will collapse on stable results . In deed , we tested an alternative algorithm of modularization [67] , and it returned almost identical predicted networks , in spite of moderately different intermediate modules ( Figure S8 ) . In theory , there are merits to incorporate a statistical matrix from the feature selection step prior to mummichog's analysis and mass flow balance of metabolic reactions [22] , [68] . While these are appealing directions for future research , the current version of mummichog confers some practical robustness , such as tolerance to technological noise and biological sampling limitation . For example , mass balance is almost impossible within serum or urine samples , because the reactions producing these metabolites are likely to occur in other tissues . The number of overlap metabolites is used in the enrichment calculation in both module analysis and pathway analysis . Sometimes , a single m/z feature may match to several metabolites in the same module/pathway , inflating the overlap number . Thus , mummichog always compares the number of overlap metabolites and the number of corresponding m/z features , and uses the smaller number for enrichment calculation , since the smaller number is more likely to be true . The size of each metabolic pathway is defined by the number of metabolites in the pathway . mummichog uses only the metabolites that can be matched in to define a pathway size , because this reflects the analytical coverage of the experiment and is confined by the same coverage . Overall , mummichog uses the whole feature list from an experiment for resampling , therefore the computation of statistical significances effectively circumvents analytical biases . In spite of the fantastic progress in mass spectrometry , these are the early days of metabolomics . Effective computational integration of resources , the combination of cheminformatics and bioinformatics , will greatly benefit the field [69] , [70] . As data accumulate , further method refinement will become possible . Mummichog presents a practical solution of one-step functional analysis , bypassing the bottleneck of upfront metabolite identification . It trades off some sensitivity in the conventional approach for the much accelerated work flow of high throughput LC/MS metabolomics . Mummichog is not designed to replace tandem mass spectrometry in metabolite identification . It is the biological activity not metabolites per se that mummichog predicts . Even with some errors on individual metabolites , as long as the biology is pinpointed to a subnetwork structure , investigators can focus on a handful of validations , steering away from the lengthy conventional work flow . In conclusion , we have demonstrated that mummichog can successfully predict functional activity directly from a spectral feature table . This benefits from the convergence of genome-scale metabolic models and metabolomics . Mummichog will continue to improve as the metabolic network models evolve . We expect this method to greatly accelerate the application of high throughput metabolomics . The mummichog software is available at http://atcg . googlecode . com .
Human peripheral blood mononuclear cells ( PBMCs ) were isolated from Buffy coats by separation over a Lymphoprep gradient . CD14+ monocytes were isolated from the PBMCs with MACS beads ( Miltenyi Biotec , Auburn , CA ) and cultured for 7 days with 20 ng/ml GM-CSF and 40 ng/ml IL-4 ( Peprotech , Rocky Hill , NJ ) . MoDCs were then harvested , washed twice and resuspended in serum-free medium . MoDCs ( ) were stimulated in triplicate in 48-well plates in a 200 µL volume with Yellow Fever virus ( M . O . I . of 1 ) , or mock infected . After 2 hrs , 800 µL of 10% FBS-RPMI was added to all wells . MoDCs were harvested at 6 hr or 24 hr after infection and centrifuged . Supernatants were aspirated , and dry cell pellets were frozen at −80°C . Supernatants of moDC cultures were assayed for the concentration of IL-6 and TNF using ELISA kits ( BD , San Diego , CA ) . Three biological replicates were used for LC/MS and QPCR . Full scan LC/MS ( m/z range 85–2000 ) was performed essentially as previously described [8] . Cell extracts or supernatants were treated with acetonitrile ( 2∶1 , v/v ) and centrifuged at 14 , 000× g for 5 min at 4°C to remove proteins . Samples were maintained at 4°C in an autosampler until injection . A Thermo Orbitrap-Velos mass spectrometer ( Thermo Fisher , San Diego , CA ) coupled with anion exchange chromatography was used for data collection , via positive-ion electrospray ionization ( ESI ) . Metabolites of interest were identified by tandem mass spectrometry on a LTQ-FTMS , where the biological sample , biological sample spiked in with authentic chemical and authentic chemical reference were run sequentially . The and were done in the ion trap of the LTQ-FTMS , with an isolation width of 1 amu and a normalized collision energy of 35 eV . The LC/MS data were processed with apLCMS program [25] for feature extraction and quantification . Significant features were also verified by inspecting the raw data ( Figure S5 ) . Features were removed if their intensity is below 10 , 000 in every sample class . Missing intensity values were imputed to 1000 . The intensities were log2 transformed . Low quality features were further filtered out if their averaged in-class coefficient of variation was greater than 0 . 2 . Averaged ion intensity over three machine replicates was used for subsequent analysis . These 7 , 995 features constituted the reference list . No normalization was used because total ion counts in all samples were very similar . Student's t-test was used to compare infected samples ( at 6 hr ) versus mock infections ( at 6 hr ) , and infected samples ( at 6 hr ) versus baseline controls ( at 0 hr ) . Features with in both tests were included in the significant list . The feature table , , and predictions are given in Dataset S1 . For gene expression quantification , mRNA was extracted by RNeasy Mini Kit ( Qiagen ) according to manufacturer's protocol , where the cell lysate was homogenized by QIAshredder spin columns . Reverse transcription was performed with SuperScript III reverse transcriptase and oligo-dT ( Invitrogen ) according to manufacturer's recommendation . Real-time PCR was performed on a MyiQ Icycler ( BioRad ) , using SYBR Green SuperMix ( Quanta Biosciences ) . The PCR protocol used 95°C 3 mins; 40 cycles of 95°C 30 seconds and 60°C for 60 seconds . The amplicons were verified by melting curves . Quantafication was performed by the method , normalized by microglobulin levels . The primer sequences are given in Table S2 . Data on human urinary samples [63] were retrieved from MetaboLights server [71] . The positive ion feature table for study “439020” contained 14 , 720 features . A feature is only included if its ion intensity is above 100 , 000 in 5 or more samples . This leaves 11 , 086 features , which consist for this study . Data were normalized by total ion counts . We next compared the metabolite difference between females ( 8 samples of low testosterone glucuronide level ) and males ( 11 samples of high testosterone glucuronide level ) . is consisted of 524 features ( by student t-test ) . The original authors annotated 3 , 689 metabolite features , and their annotation was used to compare with the prediction by mummichog . The yeast data [64] were downloaded from MAVEN website [32] in mzXML format . Feature extraction was performed in MAVEN through two approaches: targeted approach and untargeted approach . The targeted approach used chemical library from the same lab and produced 177 annotated features , which corresponded to 101 metabolites . The untargeted approach extracted 6318 features without annotation . After the same processing procedure as in our moDC data , contained 5707 features . We thus used mummichog to predict on the untargeted data , and compared the result to MAVEN annotation . The consisted of 426 features that were significantly different between the prototrophic wild type and the auxotrophic mutant ( by student t-test ) . The yeast metabolic model was compiled from BioCyc data [11] . | Mass spectrometry based untargeted metabolomics can now profile several thousand of metabolites simultaneously . However , these metabolites have to be identified before any biological meaning can be drawn from the data . Metabolite identification is a challenging and low throughput process , therefore becomes the bottleneck of the filed . We report here a novel approach to predict biological activity directly from mass spectrometry data without a priori identification of metabolites . By unifying network analysis and metabolite prediction under the same computational framework , the organization of metabolic networks and pathways helps resolve the ambiguity in metabolite prediction to a large extent . We validated our algorithms on a set of activation experiment of innate immune cells . The predicted activities were confirmed by both gene expression and metabolite identification . This method shall greatly accelerate the application of high throughput metabolomics , as the tedious task of identifying hundreds of metabolites upfront can be shifted to a handful of validation experiments after our computational prediction . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"systems",
"biology",
"metabolic",
"networks",
"biology",
"computational",
"biology"
]
| 2013 | Predicting Network Activity from High Throughput Metabolomics |
Tubular protrusions are a common feature of living cells , arising from polymerization of stiff protein filaments against a comparably soft membrane . Although this process involves many accessory proteins in cells , in vitro experiments indicate that similar tube-like structures can emerge without them , through spontaneous bundling of filaments mediated by the membrane . Using theory and simulation of physical models , we have elaborated how nonequilibrium fluctuations in growth kinetics and membrane shape can yield such protrusions . Enabled by a new grand canonical Monte Carlo method for membrane simulation , our work reveals a cascade of dynamical transitions from individually polymerizing filaments to highly cooperatively growing bundles as a dynamical bottleneck to tube formation . Filament network organization as well as adhesion points to the membrane , which bias filament bending and constrain membrane height fluctuations , screen the effective attractive interactions between filaments , significantly delaying bundling and tube formation .
Individual cells generate tubular membrane protrusions in order to sense and interact with their environment [1] . The necessary work for their formation is performed by the directed polymerization of a tightly aligned parallel actin filament bundle against the load of the cell membrane [2] . Although the core of the underlying molecular machinery required for actin driven membrane tube formation is well known , a key role in the process has been attributed to different accessory proteins in different experimental scenarios [3] . Recently , in-vitro reconstituted branched actin networks , containing only a minimum set of three purified proteins ( i . e . actin , Arp2/3 , and N-WASP ) and growing from outside against the membrane of a giant unilamellar vesicle , were shown to yield filopodia-like protrusions [4] . This finding highlighted the importance of subtle physical interactions between a reduced set of molecular ingredients in bundling filaments and forming membrane tubes . It suggests a much less elaborate mechanism for bundling and protrusion , principally involving effective attractive interactions between neighboring filaments that are mediated by nearby small-amplitude deformations of the membrane due to the individual filaments’ stochastic polymerization ( cf . Fig 1 ( a ) ) . When this bundling process eventually accumulates a sufficient number of filaments to overcome membrane resistance , filopodia-like structures emerge . In recent years , several studies explored the nature of filopodia protrusion . Specifically the impact of the stochastic nature of capping and polymerization reactions [5–7] , of actin network crosslinkers and reorganization [8–10] , and of mechanical ( in- ) stability [11] on filopodia formation and growth have been analyzed . Here , we develop a novel grand canonical membrane simulation method to establish a comprehensive statistical dynamics framework for such phenomena . By relaxing specific rather limiting model assumptions that were made for simplification in previous theoretical approaches [4 , 12–15] , we arrive at a physically highly plausible , yet relatively simple , computational model system . In detail , our model explicitly accounts for thermal fluctuations of membrane and filaments , stochastic and quantized polymerization dynamics at the filament tip , cooperativity of multiple filaments , and steric interactions between all model constituents . Within this framework we are able to reproduce previously established results for the force-extension curves of membrane tubes [16] and the growth rate of a single filament against an obstacle [12 , 17] . Much more importantly , our simulation method allows a detailed analysis of the dynamics of the tubulation process in filament-driven membrane protrusion . It had been shown before , by computing static zero-temperature minimum elastic energy shapes of two nearby membrane deformations as a function of their distance , that the membrane can induce an effective lateral attraction between the two [4 , 14] . Within our dynamical approach , we find that filament tips that are attracted to a remote primary membrane protrusion are initially arrested in a metastable state in which net polymerization ceases . A rare combination of bending and polymerization fluctuations is necessary to overcome the energy barrier to joining the protruding bundle . The typical escape time from this arrested state strongly depends on subtle polymer network parameters , like filament length and orientation , as well as possible adhesion points to the membrane , which constrain height fluctuations . As the waiting time for this bundling process competes with other network related processes like filament capping and network turnover , it is expected to be an important parameter in the biological system . By analyzing many individual trajectories of semiflexible filaments polymerizing against a fluctuating membrane patch , we establish a comprehensive statistical understanding of the dynamical transition from independently growing individual filaments to cooperatively polymerizing filament bundles that efficiently drive membrane tube extension .
In typical in-vitro experiments , e . g . in [4 , 18] , micrometer-length membrane tubes protrude from the surface of a large vesicle , which provides a fixed reference frame supporting the deformation . The vesicle’s membrane also serves as a reservoir of lipid molecules , from which the tube draws material as it elongates . Because a tube constitutes a very small fraction of the total lipid population , its area can change many-fold without influencing the surface tension imposed by the much larger lipid reservoir . In order to avoid the great computational expense of representing the lipid reservoir in numerical simulations explicitly , we extended a widely used dynamically triangulated surface model [19 , 20] to allow for fluctuations in lipid population within a grand canonical ensemble , where surface tension is held fixed . Specifically , we examine a periodically replicated square membrane patch , fluctuating around its flat ground state according to Metropolis-Hastings grand canonical Monte Carlo ( GCMC ) dynamics . We account for the membrane energy with a discretized form of the standard Helfrich Hamiltonian [21–23] , H = ∫ dS κ 2 2 H 2 + γ , ( 1 ) with mean curvature H , bending rigidity κ , and surface tension γ . Fluctuations in lipid population are achieved by Monte Carlo moves that attempt to change the number of vertices in the triangulated surface . Appropriate acceptance criteria for these moves are derived in S1 Text . Addition and removal of membrane area is regulated by a fugacity z characterizing the implicit lipid reservoir . For the case of an incompressible fluid membrane , z can be related directly to the membrane tension γ , z = C exp - γ ρ k B T , ( 2 ) where ρ is the constant lateral density of the fluid surface and kBT is the thermal energy scale . Other system specific parameters are combined into the constant C ( see S1 Text for details ) . A membrane patch simulated in this way can be manipulated as if it were part of a much larger vesicle ( Fig 2 ( a ) ) . For instance , membrane tubes can be pulled from the initially flat patch by applying an additional external potential to the triangulated surface ( Fig 2 ( b ) and 2 ( c ) ) , much as in optical tweezer experiments [18] . The equilibrium radii of such tubes are simply determined ( at zero temperature ) by the membrane’s rigidity and tension , R 0 = κ / ( 2 γ ) [16] . We exploit this relationship to determine values of the constants in Eq ( 2 ) for various z ( Fig 2 ( b ) ) . As a quantitative test for our simulation methodology , we calculated force-extension relations f ( L ) for membrane tube formation from GCMC sampling . As shown in Fig 2 ( c ) , our results are consistent with numerical zero-temperature calculations [24]: The computed pulling force initially increases with extension L , then decreases towards an asymptotic plateau value , f 0 = 2 π 2 κ γ , in the limit of long tubes [16] . As a further test of our methods , we examined the irreversible polymerization kinetics of a single rigid filament ( i . e . , depolymerization rate koff = 0 , persistence length Lp → ∞ ) growing against a simulated membrane patch . The position and orientation of the filament’s base is held fixed throughout the simulation . Stochastic polymerization attempts , occurring at rate kon , 0 and resulting in an increase δfil = 2 . 7nm in filament length , were accepted whenever permitted by constraints of excluded volume , i . e . , whenever the membrane’s fluctuating shape could accommodate monomer addition ( see S1 Text and Materials and Methods Sec . A for additional details of the simulations ) . The normalized rate of successful polymerization , kon ( n ) /kon , 0 , decays rapidly with the polymerization event number n in membrane simulations ( see Fig 2 ( d ) ) . This is consistent with the decrease as expected from near-equilibrium theory , kon ( n ) /kon , 0 ≃ exp[− ( f ( L ) δfil ) / ( kBT ) ] , with retraction force f ( L ) of the membrane at extension L [12 , 25] . Under typical in-vitro conditions ( i . e . globular actin concentration ∼ 10μM and nonzero depolymerization rate koff ) , net polymerization of a single actin filament ceases early in protrusion once the steady state condition kon ( n ) /kon , 0 = koff/kon , 0 ≃ 10−2 is met . Forming extended membrane tubes thus typically requires cooperative polymerization of multiple neighboring filaments , which share the load of the membrane . To establish the necessary physical conditions for cooperative filament growth , we analyzed 50 independent trajectories of collections of Nfil = 10 growing , structurally fluctuating ( persistence length Lp = 15μm ) , and additionally depolymerizing ( koff/kon , 0 = 10−2 ) filaments anchored at uniformly random lateral positions below the membrane . To fix the membrane’s base relative to the filaments’ anchor points , we constrain the positions of three boundary nodes of the membrane to be constant throughout the simulation runs . The inset of Fig 3 ( a ) shows in black the location of these frozen nodes ( together with their periodically replicated copies ) within the membrane patch . Despite the lack of additional accessory proteins for filament bundling and crosslinking in our model , all simulated trajectories eventually yielded tight cooperative filament bundles and subsequently a single polymerization-driven membrane tube of at least 260nm in height . For sufficiently large filament density , bundle formation is thus only a matter of time . It is initiated by the interaction of individual filaments with the membrane , which leads to bending and directed growth towards a single point of protrusion as sketched in Fig 1 ( b ) . Roughly half of the tubes formed before a simulation time of 800/kon , 0 , while the slowest tube formation process required about 1800/kon , 0 to reach this height ( cf . solid line in Fig 3 ( c ) ) . This range in tube formation times reflects a distribution of waiting times for filament bundling by the membrane . In simulation , the accumulation of filaments into a tight bundle occurs through a cascade of dynamical transitions . In the early stages of a representative trajectory Fig 4 ( a ) , significant net polymerization and membrane protrusion take place only where the lateral density of filaments happens to be high ( red in Fig 4 ( a ) ) . The filaments’ growing ( “barbed” ) ends are pushed together by forces from the membrane , as described in [4 , 14] . For growth conditions we have studied , the resulting premature bundle is not sufficient to overcome the membrane’s restoring force . Net polymerization therefore stalls , and the interfacial deformation fluctuates around a steady state height . The recruitment of additional filaments to the premature bundle occurs sequentially , as each one traverses a dynamical bottleneck evidenced by sudden changes in its growth rate ( Fig 4 ( a ) , colors other than red ) . Remote filaments experience a membrane-induced attraction to the bundle , but it is offset by the elastic cost of their own bending . Their tendency to join the bundle is thus not a direct consequence of mechanical forces on the barbed end . It is instead a product of the nonequilibrium nature of these dynamics ( which generate net filament growth ) , together with the inevitability of fluctuations that create sufficient space for polymerization and the bias provided by the bundle’s deformation of the membrane . Once the bundle amasses enough filaments to generate polymerization forces greater than f0 , it grows steadily at a rate that increases with each additional recruitment event . These elongation kinetics suggest a high degree of cooperativity , with nearly equal sharing of the membrane’s load . Perfect load sharing , in which each of N filaments grows against a force f0/N , would yield a net polymerization rate kon for each filament: k on k on , 0 ≃ exp - f 0 δ fil N k B T - k off k on , 0 . ( 3 ) In simulations we observe bundle elongation rates of nearly this magnitude , with an effective value of N that is about 90% of the true bundled filament population . The empirical effective polymerization rate of a complete bundle ( i . e . , Eq ( 3 ) with N set to 0 . 9Nfil ) is indicated by the black dashed line in Fig 4 ( a ) . Based on the measured radius , the membrane tension of the fully grown tube , γ = κ/ ( 2R2 ) , is consistent with the corresponding tube pulling simulation in Fig 2 ( b ) at the same fugacity z = 21 , i . e . γ ≃ 0 . 01kBTnm−2 . To analyze in detail the kinetics of recruiting filaments to a premature bundle , we focus on the simplified scenario sketched in Fig 4 ( c ) . Here , the bundle’s influence on a distant tagged filament is represented by a static average tilt angle θ and the vertical distance L0 between the filament anchor point and the membrane due to the bundle’s protrusion ( cf . S1 Text , Fig S3 for a sketch indicating these parameter definitions ) . Growing beneath such a tilted patch of membrane , the tagged filament adds monomers at rate kon , 0—koff until approaching the membrane within a distance δfil . Further polymerization awaits a sufficient fluctuation in the shape of the membrane and/or the filament . The most likely of these fluctuations involve bending of the filament toward the bundle ( cf . inset sketch Fig 4 ( c ) ) . Monomer addition locks in the resulting filament curvature , reducing the cost of further filament deformation and thus enhancing subsequent growth . Once the ground-state filament orientation at its barbed end is parallel to the tilted membrane , polymerization is again unimpeded and very rapid . Through this facile growth , the tagged filament quickly joins the nascent bundle . First passage time statistics for reaching the bent state can be understood thoroughly in the context of a simplified model for growth kinetics . As equilibration of membrane and filament shape fluctuations is fast relative to the timescale of polymerization , we assume that structural fluctuations which create sufficient space for monomer addition occur with probabilities corresponding to thermal equilibrium , independent of growth history . The rate kon ( n ) for polymerization of a tagged filament already comprising n monomers is then given by k on ( n ) = k on , 0 P bend ( n → n + 1 ) , ( 4 ) where Pbend ( n → n + 1 ) is the equilibrium probability of a filament bending fluctuation that creates a gap of sufficient size δfil between membrane and filament . The S1 Text presents an analytical approximation for Pbend ( n → n + 1 ) that considers only the softest modes of deforming the membrane and filament . This result , which we use below , is plotted in Fig 4 ( b ) inset . The set kon ( n ) of effective polymerization rates informs a master equation for stochastic growth dynamics , ∂ t P ( n , t ) = k off P ( n + 1 , t ) - P ( n , t ) + k on ( n - 1 ) P ( n - 1 , t ) - k on ( n ) P ( n , t ) , ( 5 ) where P ( n , t ) denotes the probability that the tagged filament is composed of exactly n monomers at time t . We computed probability distributions p ( τFP ) of the first passage time by numerically solving Eq ( 5 ) . The results of this approximate treatment , plotted in Fig 4 ( b ) , agree well with detailed simulations of a single fluctuating filament growing against a membrane that fluctuates about a uniformly tilted state ( see S1 Text details and the S4 Movie of the simulation ) . The average waiting time 〈τFP〉 to reach the bundle can be directly calculated from Eq ( 5 ) [26] . As shown in Fig 4 ( c ) , it varies strongly with the membrane tilt angle and filament length . Modest changes in these two key parameters alter the mean first passage time by orders of magnitude . Hence , our results suggest that the efficiency of dynamic bundling and , as a direct consequence , the rate of membrane tube formation is highly sensitive to even subtle structural changes in the underlying actin network architecture . To test the impact of suppressed membrane height fluctuations on protrusion formation , we performed 50 additional simulations , using identical filament base positions as before , but with an additional frozen membrane node at the center of the patch ( cf . Fig 3 ( a ) and 3 ( b ) insets ) . In a laboratory context , these constrained vertices could represent transient binding of filaments within a growing network to membrane-bound nucleation promoting factor N-WASP [27] . Due to this additional constraint on the membrane , the waiting time for filament bundle formation indeed increases substantially . Furthermore , the mature tubes that do form include fewer filaments in the growing bundle . The filaments that fail to join the bundle in this case are separated by an immobile node from the primary protrusion , as illustrated by representative simulation snapshots in Fig 3 ( a ) and 3 ( b ) ( cf . S1 and S2 Movies of these simulations ) . This depletion of the bundle reduces its effective polymerization force , strongly delaying the emergence of protrusions and slowing their subsequent growth . The simulations and analysis described so far considered all filaments in the actin network to be growing in the same direction , normal to the initial plane of the membrane . Real networks , however , include filaments with a substantial range of orientations . Previous work has shown that this diversity has important consequences for reconstituted actin networks and migrating cells [28 , 29] . Given the sensitivity of filament recruitment kinetics to the angle between filament and membrane , we expect orientational diversity within a network to also significantly impact dynamics of bundling and protrusion . To explore the roles of filament alignment in tube formation , we advanced growth trajectories from initial conditions in which filament orientations were assigned randomly . In detail , we selected each filament’s initial polar angle from a Gaussian distribution centered at zero ( with the polar axis pointing along the membrane’s normal vector ) and with standard deviation std; its azimuth was selected uniformly without bias . We then monitor which filaments join a tube bundle at various stages of its development , as assessed by the tube’s height L . Different trajectories require different amounts of time to achieve the same value of L . The time at which the bundling transition occurs also varies from trajectory to trajectory , but the corresponding value of L is consistent . We performed 50 independent simulations for the case std = 20° , and another 50 for the case std = 40° . The sequence of events in these trajectories is very similar to that previously described for filaments that all point in the same direction . A cascade of filament bundling events again accumulates filaments into a bundle ( colored red in Fig 5 ) to form the tube . And all filaments that join the bundle subsequently contribute in load sharing and accelerate tube protrusion . The principal distinction is that some poorly aligned filaments fail to join the tube even at very long times . A well-developed tube ( L = 260 nm ) typically includes filaments whose orientations lie within ∼ 40° of the surface normal . Filaments initially pointing outside this cone have either stalled at this stage ( still awaiting recruitment to the bundle ) or have failed mechanically and grown away from the membrane ( as indicated by green and black coloring in Fig 5 ( a ) –5 ( c ) ) . For std = 20° this discrimination excludes only a small fraction of filaments . Many more filaments fail to join the mature bundle in the case std = 40° . Among filaments included in the tube , the distribution of initial orientations is very similar for the two values of std ( see Fig 5 ( d ) ) . Robust tube formation thus primarily requires a sufficient density of filaments at orientations that are compatible with bundling . In addition to selecting filaments with sufficiently aligned initial orientations , bundling and protrusion significantly bias filaments’ orientation at the growing barbed end . Fig 5 ( e ) shows the degree of alignment among barbed end orientations for filaments that eventually join the bundle ( see Materials and Methods Sec . A for details ) , plotted as a function of tube height . These barbed ends are highly aligned already at low tube extension , reflecting the orientation selection process discussed above . At small membrane deformation , 0 ≤ L ≤ 50nm ( i . e . before a major filament bundle has formed ) , alignment of these filaments transiently declines due to dynamic bending in a direction tangential to the membrane and subsequent bundling . Strong barbed end alignment is restored within the mature membrane tube . In long tubes , ≥ 150nm , the degree of alignment even exceeds its initial value , which is indicated by the black dashed horizontal line in Fig 5 ( e ) . Once filaments are recruited to protrude the membrane tube , they are corralled into a highly parallel bundle that efficiently drives tube extension .
In this work , we introduced a novel grand canonical simulation method for a fluctuating biological fluid membrane , based on a randomly triangulated surface , within a statistical ensemble of constant surface tension . Coupling the system to an implicit lipid reservoir , while at the same time explicitly accounting for thermal fluctuations and excluded volume constraints , our computational method is very well suited to efficiently simulate typical experimental vesicle assays and , more generally , biological processes in which the relevant membrane area changes over time . By additionally including quantized and stochastic polymerization kinetics of fluctuating filaments , we identified and quantified key necessary microscopic conditions for filament polymerization-driven membrane tubes , which are extremely difficult to access experimentally . Our simulations revealed a cascade of single filament bending transitions as an important dynamical bottleneck to filament bundling and subsequent tubular protrusion formation . Even subtle changes in structural filament network parameters , such as filament orientation and length or additional constraints on the membrane’s height fluctuations have a profound impact on the waiting time for filament bundling . Inside the cell , this waiting time competes not only with the timescale of other polymer network related processes , like capping and actin turnover , but also with the viscous response time of the treadmilling actin gel . In this biological context , the waiting time for filament bundling will directly impact the rate of membrane tube formation and filopodia emergence in biological systems , or might even prevent their occurrence altogether . These results suggest a possible explanation for unexpected experimental observations in reconstituted assays of actin-driven membrane tube formation [4] . Specifically , the number of filopodia-like tubes on vesicles was found to decline upon increasing Arp2/3 or N-WASP concentration , despite the concomitant increase in filament density expected by the biochemical perturbation . While higher filament density should enhance the collective force of polymerization , the denser filament network likely also features more numerous adhesions to the membrane . In our analysis , the resulting suppression of height fluctuations could indeed strongly diminish filament bundling and subsequent formation of linear protrusions . Our model could be straightforwardly extended to account for the mechanics and dynamics of an explicitly crosslinked actin network , and of various actin associated proteins . Other parameter regimes of the model will also be worthwhile to explore . At even larger actin filament density , packing effects and the resulting thickness of the tubular actin bundle will become important as has been highlighted before [30–32] . Our results could be directly tested in biochemical experiments by manipulating the filament orientation distribution of the polymerizing actin network [33] or by incorporating artificial adhesion points into a reconstituted membrane . Together with such extensions and related measurements , our approach promises to enable a thorough microscopic understanding of the combined biochemical and biophysical requirements for the formation of actin-driven membrane protrusions .
Each of the grand canonical Monte Carlo simulations feature a square membrane patch with linear size in the range of 200–300nm . The size of the membrane patch mostly determines the cutoff for long wavelength fluctuations . In the biological system we expect such long wavelength fluctuations to be efficiently damped by tethers between the membrane and the relatively rigid growing actin network . In simulations we are choosing the patch size to be able to resolve the smallest important wavelength on the size of the membrane’s thickness . To constrain the center of mass motion of the membrane patch vertically , three equidistant boundary nodes ( and their periodic images ) were immobilized ( except where explicitly stated differently in the main text ) . Fluctuating actin filaments are included into the Metropolis MC scheme as worm-like chains , using the standard discretized Hamiltonian in combination with free rotation and crankshaft MC moves . The base segment of each discretized filament is fixed in position and orientation throughout the entire simulation . The membrane and filament MC quasi-dynamics can be mapped to physical dynamics for small step size . In this spirit , we related the timescale of filament and membrane MC dynamics by comparing the equilibration time of relevant normal modes for the process under consideration ( see S1 Text for details ) . Additionally , in the dynamical simulation we account explicitly for stochastic , quantized ( de- ) polymerization events in which filaments change their contour length by 2 . 7nm with a kinetic MC approach that draws a random reaction time and filament that is subject to the corresponding reaction [34] . The effects of volume exclusion due to the membrane ( with nodes assigned a diameter of 5nm ) and other filaments ( node diameter 10nm ) on these biochemical kinetics are also included explicitly . Unless otherwise noted in the main text , we used typical in-vitro conditions for actin on- and off-rates , kon , 0/koff = 102 . Minimum energy shapes of the membrane at zero-temperature were calculated using Surface Evolver [24] . Filament segment alignment a in Fig 5 ( e ) was measured by averaging barbed end orientations for filaments that eventually joined the mature bundle , a ( L ) = | ∑ i = 1 N o ^ i | / N , where o ^ i are the unit orientation vectors of filament segments whose height lies between L and L + ΔL , with ΔL = 10nm . Averages included results from all simulations that eventually formed a tube of minimum height , 260nm , before a given maximum time , 2500/kon , 0 . | The necessary biophysical conditions for the formation of tubular membrane protrusions by polymerizing actin filament bundles have not yet been fully understood . For this reason we introduce a novel grand canonical simulation model that describes stochastic polymerization of filaments against a fluctuating fluid membrane , while only considering a minimum set of biological proteins . Although still relatively simple and highly tractable , our model explicitly accounts for thermal fluctuations of membrane and filaments , stochastic and quantized polymerization dynamics at the filament tip , cooperativity of multiple filaments , and steric interactions between all model constituents in a physically realistic way . This approach enables us to go well beyond previous static zero-temperature theoretical considerations to filament bundling and explore the physical origins of membrane tube formation dynamics on length and time scales that are currently inaccessible to both experiments and atomistically detailed simulations . Our results suggest a membrane mediated dynamical transition from single filaments to cooperatively growing bundles as an important dynamical bottleneck to tubular protrusion . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"classical",
"mechanics",
"simulation",
"and",
"modeling",
"damage",
"mechanics",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"bending",
"polymer",
"chemistry",
"contractile",
"proteins",
"actins",
"lipids",
"proteins",
"deformation",
"chemistry",
"biophysics",
"polymerization",
"cell",
"membranes",
"physics",
"biochemistry",
"biochemical",
"simulations",
"cytoskeletal",
"proteins",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"computational",
"biology",
"biophysical",
"simulations"
]
| 2016 | The More the Tubular: Dynamic Bundling of Actin Filaments for Membrane Tube Formation |
Among human RNA viruses , hepatitis C virus ( HCV ) is unusual in that it causes persistent infection in the majority of infected people . To establish persistence , HCV evades host innate and adaptive immune responses by multiple mechanisms . Recent studies identified virus genome-derived small RNAs ( vsRNAs ) in HCV-infected cells; however , their biological significance during human HCV infection is unknown . One such vsRNA arising from the hepatitis C virus ( HCV ) E2 coding region impairs T cell receptor ( TCR ) signaling by reducing expression of a Src-kinase regulatory phosphatase ( PTPRE ) in vitro . Since TCR signaling is a critical first step in T cell activation , differentiation , and effector function , its inhibition may contribute towards HCV persistence in vivo . The effect of HCV infection on PTPRE expression in vivo has not been examined . Here , we found that PTPRE levels were significantly reduced in liver tissue and peripheral blood mononuclear cells ( PBMCs ) obtained from HCV-infected humans compared to uninfected controls . Loss of PTPRE expression impaired antigen-specific TCR signaling , and curative HCV therapy restored PTPRE expression in PBMCs; restoring antigen-specific TCR signaling defects . The extent of PTPRE expression correlated with the amount of sequence complementarity between the HCV E2 vsRNA and the PTPRE 3’ UTR target sites . Transfection of a hepatocyte cell line with full-length HCV RNA or with synthetic HCV vsRNA duplexes inhibited PTPRE expression , recapitulating the in vivo observation . Together , these data demonstrate that HCV infection reduces PTPRE expression in the liver and PBMCs of infected humans , and suggest that the HCV E2 vsRNA is a novel viral factor that may contribute towards viral persistence .
Hepatitis C virus ( HCV ) persistently infects more than 120 million people globally , and chronic viremia frequently leads to cirrhosis and hepatocellular carcinoma [1–8] . Although numerous factors appear to contribute to viral persistence , the mechanisms by which HCV evades immune responses are incompletely understood . Prior studies found that HCV-infection is associated with reduced T cell function in vitro , impaired HCV-specific intrahepatic and peripheral T cell response ex vivo , delayed onset of HCV-specific humoral and cellular immunity in vivo , and impaired immune responses to HBV and adenoviral vaccination [2 , 8–20] . We recently reported that incubation of peripheral blood mononuclear cells ( PBMCs ) with plasma derived HCV , infectious cell culture derived HCV , and serum exosomes containing HCV RNA reduced IL-2 release and CD69 upregulation by T lymphocytes following activation through the T cell receptor ( TCR ) [11] . Expression of the HCV envelope ( E2 ) coding RNA in Jurkat cells was sufficient to reduce TCR signaling , and to reduce phosphorylation of the lymphocyte-specific , protein tyrosine Src kinase ( Lck ) . Deletion mutagenesis of HCV E2 RNA reducing Lck activation demonstrates that a short RNA region is sufficient to reduce TCR signaling [11] . This E2 RNA sequence contains a conserved 8 base region complementary to two sites in the 3’UTR of the Src regulatory phosphatase PTPRE ( protein tyrosine phosphatase receptor epsilon ) [11] . PTPRE activates signaling by Src family tyrosine kinases [21–23] , and previous studies demonstrate that inhibition of Src-kinase signaling promotes HCV replication [24 , 25] . Expression of HCV vsRNA is sufficient to reduce PTPRE protein levels in Jurkat cells , and mutation of conserved residues in the HCV E2 short RNA region restore PTPRE levels and TCR-mediated Lck activation [11] . PTPRE specificity was confirmed by placing the PTPRE 3’UTR sequences after GFP , and showing that HCV E2 expression regulated GFP expression in this system . Furthermore , replacement of the PTPRE targeting sequence in HCV E2 with a sequence targeting CXCR4 restored PTPRE levels and reduced CXCR4 expression . Thus , a virus ( HCV E2 ) RNA-derived , short RNA ( vsRNA ) regulates PTPRE and reduces TCR signaling in vitro [11] . Although DNA viruses and retroviruses generate functional vsRNAs [26 , 27] , the ability of strictly cytoplasmic RNA viral genomes to be processed into functional vsRNAs is controversial [28–30] . Short RNA species are found in HCV and other cytoplasmic RNA virus infected cells [11 , 31–35]; however , there are no data demonstrating that these vsRNAs are functional during human infection . Here , we expand the previous in vitro characterization of the HCV vsRNA effect on TCR signaling by showing that synthetic HCV genomic and vsRNA regulate PTPRE via one of the two potential target sites with complementarity within the PTPRE 3’UTR , and that HCV regulates TCR and PTPRE expression in human liver tissue and PBMCs during HCV infection . Importantly , curative HCV therapy restored both PTPRE levels and T cell activation following TCR stimulation . The data provide the first in vivo evidence of a functional vsRNA generated from the HCV genome , and identify PTPRE as a novel cellular factor regulating T cell activation .
To determine if HCV RNA-containing sera inhibits antigen-specific TCR signaling , PBMCs from three healthy blood donors were incubated in sera obtained from 5 HCV infected donors before and following curative HCV therapy . Sera pooled from 5 HCV uninfected individuals served as the negative control . Following overnight incubation , cells were stimulated with either viral T-cell antigenic peptides from CMV , EBV , and influenza ( CEF peptides; Anaspec ) or anti-CD3 antibody . Incubation of PBMCs in HCV infected patient serum reduced , but did not abolish IL-2 release following antigen-specific T cell receptor stimulation ( representative donor PBMCs in Fig 1A ) . Following curative HCV treatment , IL-2 release by cells incubated in the five treated HCV patients was not different than IL-2 released by cells incubated in pooled sera from five HCV-negative subjects or in cells that were not incubated in human serum ( Fig 1A ) . HCV RNA positive serum also reduced TCR signaling induced by anti-CD3 stimulation , and as expected anti-CD3 was more potent in inducing IL-2 than the antigen-specific stimulation ( Fig 1B ) . Although markedly different concentrations of IL-2 were released by PBMCs obtained from different blood donors following TCR stimulation , the fold-change in IL-2 following TCR stimulation followed the same pattern of inhibition by HCV RNA-containing sera . Following curative therapy , the same patient’s sera did not inhibit IL-2 release ( Fig 1C and 1D ) . Previous studies found that serum from HCV-infected individuals also regulates TCR-mediated IL-2 release in a CD4+ T cell line ( Jurkat cells ) [11] . Jurkat cells were incubated in HCV RNA-positive sera before or following direct anti-HCV therapy , and PTPRE expression was measured by immune blot ( Fig 2A ) . PTPRE was reduced in Jurkat cells incubated in serum from HCV infected people prior to treatment , but this reduction was lost following treatment ( Fig 2B ) . Furthermore , Jurkat cells incubated in Huh7 . 5 cell culture-derived infectious HCV particles ( HCVccs ) also reduced PTPRE expression relative to that expressed in Jurkat cells incubated in post-treatment serum , or in control Jurkat cells incubated in a pool of HCV negative donors ( Fig 2B ) . Together , these data show that serum from HCV-infected individuals reduces both TCR-signaling as measured by IL-2 release and PTPRE expression . Further , HCVcc particles lacking other serum factors similarly reduce PTPRE expression , and as previously shown , reduces TCR signaling [11] . Stable expression of a short region of HCV E2 RNA in Jurkat cells negatively regulates TCR signaling , PTPRE expression , and Lck phosphorylation following TCR activation [11] . Here , we examined transfection of in vitro transcribed , full-length , infectious HCV RNA or a synthetic RNA duplex comprised of the HCV vsRNA sequence to determine if transfection of HCV RNA was sufficient to regulate PTPRE expression . Bioinformatic analyses identified several genes predicted to be targeted by the HCV vsRNA based on the putative seed sequence , including PTPRE , Vesicle-associated membrane protein-A ( VAPA ) , and growth factor receptor-bound protein 2 ( Grb2 ) [36–38] . Like PTPRE , the VAPA 3’UTR contains two sequences with at least 7 bases complementary to a conserved 8 nt HCV RNA sequence within the HCV vsRNA , while Grb2 contains one such target site . VAPA is a proviral factor required for HCV replication , thus reducing its expression would be deleterious for HCV , and Grb2 is a positive regulator of Src kinase signaling , thus inhibition could contribute to impaired TCR signaling [39–42] . Because HCV is hepatotropic , and due to poor transfection efficiency of Jurkat cells , we transfected the HCV permissive hepatocyte cell line Huh 7 . 5 with full length HCV genomic RNA ( HCVgRNA ) transcribed from an infectious clone ( kindly provided by Drs . Rice and Wakita ) [43] . PTPRE levels were reduced in the HCVgRNA transfected cells compared to sham transfected cells . In contrast , VAPA and Grb2 expression levels were not altered ( Fig 3A ) . Alignment of the 29 base HCV vsRNA sequence identified in HCV infected Huh7 . 5 cells by Andrew Fire’s laboratory [31] , found 38% complementarity between the HCV vsRNA sequence and site 1 of the PTPRE 3’ UTR ( Fig 3B ) , both sites of VAPA 3’UTR ( Fig 3C ) , and the Grb2 3’ UTR sequence ( Fig 3D ) . However , there was 56% complementarity between the vsRNA sequence and site 2 on the PTPRE 3’ UTR ( Fig 3B ) . Furthermore , all of the target sequences contained 7 bases complementary to the conserved 8 base HCV sequence except site 2 of PTPRE . PTPRE 3’UTR site 2 contained 8 bases complementary to the vsRNA sequence of all 627 isolates listed in the Los Alamos database ( http://hcv . lanl . gov/content/sequence/HCV/ToolsOutline . html ) . In addition , PTPRE Site 2 was complementary to 9 HCV E2 conserved bases in the majority of these isolates ( Fig 3 ) . These results suggest that the number of bases within the target sequence complementary to the HCV sequence and the flanking HCV sequences may contribute to target gene specificity . To determine which of the two PTPRE 3’UTR target sites interacted with the HCV vsRNA , the PTPRE site 1 sequence and the PTPRE site 2 sequence were independently inserted into the 3’UTR region of GFP as illustrated ( Fig 4A ) . These plasmids and the parent GFP expression plasmid were used to generate human embryonic kidney ( HEK ) 293 cell lines stably expressing GFP as previously described [11] . Each cell line was transfected with a synthetic , genome-length , HCV RNA transcript or the transfection reagent . Alternatively , each cell line was incubated in HCVccs ( 1 . 4 x 106 infectious units ) . GFP ( Fig 4B ) and PTPRE expression ( Fig 4C ) were monitored 48 hours post transfection or HCVcc incubation using immunoblot and flow cytometry analyses , respectively . HCV gRNA only reduced GFP expression in cells expressing GFP with PTPRE site 2 3’UTR target sequences ( Fig 4B ) . Since site 1 has 38% complementarity and site 2 has 56% complementarity with the HCV E2 vsRNA , these data provide additional support for the hypothesis that the percent complementarity between the vsRNA and PTPRE 3’UTR is critical for gene regulation , and may explain why VAPA and Grb2 were not regulated by HCV RNA . As expected , HCV gRNA and HCVccs reduced PTPRE expression in all three cell lines compared to control cells ( Fig 4C ) . HCVcc’s did not reduce GFP in any of these cell lines , presumably due to the lower concentration of HCV RNA present in this preparation and the high levels of GFP expression . To further examine RNA-mediated regulation of PTPRE , two RNA duplexes containing the conserved HCV vsRNA targeting sequence were synthesized . The 8 nt HCV sequence was placed at either the 5’ end ( vsRNA-1 ) or the 3’ end ( vsRNA-2 ) of the HCV sequence ( Fig 5A ) . These vsRNAs were transfected into Huh7 . 5 cells , and both reduced PTPRE but not VAPA levels compared to control cells transfected with non-specific siRNA ( Fig 5B ) , suggesting that the location of the 8 nt seed sequence within the vsRNA may not be critical for PTPRE inhibition . PTPRE is expressed by hepatocytes in liver tissue and in lymphocytes ( web-based protein atlas ) ( http://www . proteinatlas . org/ENSG00000132334-PTPRE/tissue ) . HCVgRNA and synthetic HCV vsRNA duplexes are sufficient to reduce PTPRE levels in cells of hepatocyte origin ( Figs 2 , 3 and 4 ) . Furthermore , serum-derived HCV RNA present in virions or serum extracellular vesicles are transferred into hepatocytes and lymphocytes resulting in reduced PTPRE protein levels and productive infection in vitro [11 , 44–46] . Since HCV replicates primarily in hepatocytes during human infection [47] , we examined PTPRE levels in liver explant tissues obtained from HCV-infected and HCV-uninfected individuals . All liver tissue was evaluated by a pathologist with extensive experience in hepatic pathology , and fibrosis and inflammation scores used the metavir system . Inflammation was graded A0 = no activity , A1 = mild activity , A2 = moderate activity , and A3 = severe activity . Fibrosis was scored as F0 = no fibrosis , F1 = portal fibrosis without septa , F = portal fibrosis with few septa , F3 = numerous septa without cirrhosis , and F4 = cirrhosis . S1 Table summarizes the age , gender , diagnosis , fibrosis and inflammation scores for the subjects . The number of subjects with grade 3 or 4 fibrosis were equal in the HCV and the non-HCV liver tissues ( n = 3 ) . PTPRE levels were significantly lower in tissue obtained from HCV-infected humans compared to liver tissues from people with liver disease other than HCV infection when normalized to GAPDH ( Fig 6A and 6B ) , and PTPRE levels did not correlate with inflammation or fibrosis score ( Fibrosis data shown in S1 Fig ) . Since PTPRE activates Src-kinases , and previous studies found an inverse relationship between HCV replication and Src-kinase signaling [24 , 25] , PTPRE may be a previously unrecognized viral restriction factor in HCV infection . Interestingly , PTPRE levels were lower in Huh7 . 5 and Huh7D human hepatoma cell lines compared to the Huh7 cell line that they were clonally derived from ( Fig 4C ) [48] , and HCV replicates significantly higher in Huh7 . 5 and Huh7D cells compared to the parental Huh7 cell line [48] . Early studies suggested that a mutation in RIG-I in Huh7 . 5 cells may contribute to enhanced HCV replication; however , subsequent studies found that Huh7D cells do not have the RIG-I mutation , yet support HCV replication as well as Huh7 . 5 cells [48] . PTPRE variant 1 is a transmembrane protein while PTPRE variant 2 lacks the transmembrane sequence , and is strictly cytoplasmic [49] . Both Huh7 . 5 and Huh7D cell lines had lower levels of both PTPRE variant-1 ( transmembrane , open arrow ) and variant-2 ( cytosolic , closed arrow ) compared to Huh7 cells ( Fig 4C ) . Thus , there is an association between reduced PTPRE levels and HCV replication in hepatoma cell lines in vitro and PTPRE levels are reduced in liver tissue from HCV infected people compared to HCV uninfected , suggesting that PTPRE may interfere with HCV replication by promoting Src-kinase signaling . HCV RNA is present in , or bound to PBMCs and platelets [50–52] , and HCV infection is associated with impaired IL-2 and IFN-γ responses following stimulation [53] . Incubation of healthy donor PBMCs in HCV RNA-containing particles leads to reduced IL-2 release and surface expression of T cell activation markers [11] . Thus , we examined PTPRE expression in lymphocytes obtained from HCV-infected individuals before and following curative HCV therapy , and compared the results with PTPRE levels in HCV uninfected subjects . S2 Table summarizes the age , gender , diagnosis , fibrosis , and inflammation scores for the subjects . PTPRE expression was significantly lower in HCV-infected PBMCs compared to controls , and rose to levels comparable or higher than HCV uninfected controls following curative HCV therapy ( Fig 7A–7C ) . Lck activation ( phosphorylation of Y394 ) following TCR stimulation is required for TCR-mediated activation and proliferation [54] . We examined the ability of anti-CD3 stimulation to phosphorylate Lck in PBMCs from subjects before and after HCV therapy . Lck phosphorylation following 5 minutes TCR stimulation was significantly higher in subjects cured of HCV by therapy compared to pre-treatment levels in the same subjects ( Fig 7D ) . These findings are consistent with recent studies demonstrating that curative anti-HCV therapy restores immune cell function in HCV-infected humans by other measures [55 , 56] . Andrew Fire’s group identified HCV vsRNAs in HCV-infected Huh7 . 5 cells [31] , one of which was the vsRNA we identified that reduces PTPRE protein expression in vitro ( vsRNA sequences kindly provided by Drs . Fire and Parameswaran ) [11] . Since PBMC PTPRE levels varied somewhat among HCV-infected individuals ( as in Fig 7B ) , we sequenced the HCV E2 RNA present in serum obtained from ten subjects to determine if there is a relationship between PTPRE expression and HCV sequence diversity ( Fig 8A , underlined ) [31] . Examining the patient’s 29 base E2 sequence that is detected as a vsRNA in HCV-infected cells [11] , the 8 nt region complementary to PTPRE 3’UTR was highly conserved . However , there were numerous sequence polymorphisms in the flanking sequences ( Fig 8A ) . To quantify this , the percent of bases in the HCV vsRNA sequences from each of the ten subjects complementary to the two PTPRE 3’UTR target sequences were correlated with the expression of PTPRE in their PBMCs ( Fig 8B ) . The greater the percent complementarity between each subjects’ E2 RNA sequence with the PTPRE 3’ UTR , the lower the level of PTPRE expression detected ( R2 0 . 56 , p<0 . 01 Spearman Correlation ) .
Virus derived small RNAs ( vsRNAs ) encoded by DNA viruses and retroviruses play an important role in viral replication and may contribute to immune evasion [26 , 27] . Although vsRNAs are found in cells infected with cytoplasmic RNA viruses in vitro [11 , 31–35 , 57 , 58] , their role in human infection is not characterized , and their significance is debated [28 , 30] . HCV is unusual among cytoplasmic RNA viruses in that it establishes persistent infection in the majority of infected people [2 , 59] . Previous studies found that HCV infection impairs T cell function , and presumably this contributes to viral persistence [7–11 , 59] . Although several mechanisms may contribute to HCV immune evasion , we recently found that expression of full-length HCV E2 coding RNA with a frame-shift to abolish translation in Jurkat cells reduced TCR signaling and Lck activation following TCR stimulation with anti-CD3/CD28 [11] . Placing the PTPRE 3’UTR after GFP , the expression of HCV E2 RNA regulated GFP in transient transfection experiments , and expression of the RNA with 4 bases substituted restored TCR activity and PTPRE levels . Finally , when the sequence that is complementary to the PTPRE 3’UTR was replaced with a sequence targeting CXCR4 , the RNA reduced CXCR4 and not PTPRE [11] . We also demonstrated that HCV RNA-containing plasma , HCVccs , and plasma-derived HCV RNA-containing micro-vesicles impaired IL-2 release by Jurkat cells and primary human PBMCs and purified T cells following stimulation with anti-CD3 antibody [11] . Here , we expand the earlier findings to show that HCV containing serum inhibits both antigen-specific ( CEF-mediated ) TCR signaling , and PTPRE expression , and that T cell function and PTPRE levels are restored following curative HCV therapy , providing novel insights into antigen-specific and non-specific modulation of T cells by HCV RNA . We also demonstrated that in vitro transcribed full-length HCV genome and short synthetic HCV E2 vsRNAs were sufficient to regulate PTPRE expression following transfection into Huh7 . 5 cells , and that one of the two complementary sequences in the PTPRE 3’UTR is sufficient to regulate upstream protein expression . We also observed that PTPRE levels were reduced in liver biopsy tissues and cell lines of hepatocyte origin that support HCV replication , and that HCV infection was associated with both reduced PTPRE levels in PBMCs , and with reduced Lck activation following TCR stimulation . PTPRE levels and Lck phosphorylation were restored by curative HCV therapy . Further supporting a functional role for the HCV vsRNA , the extent of sequence complementarity between the HCV E2 RNA sequences correlated directly with the level of reduction of PTPRE expression in lymphocytes . Although complete blockade of TCR signaling would render an infected person severely immune compromised , HCV effects on TCR signaling are incomplete . This regulator of TCR signaling likely contributes to both establishment of infection and persistent viremia . Although there are many vsRNAs detected in HCV infected cells for which no function has been determined [31] , these data suggest that vsRNAs may serve as an unexplored mechanism of HCV regulation of host gene expression . Nevertheless , several important questions remain . For example , how does HCV RNA interact with and regulate cellular gene targets ? We speculate that the HCV RNA-containing virions , lipo-particles , and exosomes released from the liver during viral replication deliver viral RNA to uninfected hepatocytes and lymphocytes . Early studies showed that hepatic lymphocytes are more impaired in proliferative activity than circulating lymphocytes [8] , and we propose that this may be explained by the greater exposure of hepatic lymphocytes to HCV RNA . Given that plasma HCV RNA concentrations are typically > 1 million copies/ml , circulating lymphocytes are also exposed to HCV RNA-containing particles [11] . Another question relates to the concentration of viral RNA and whether it is sufficient to regulate cellular genes [28] . It is important to note that standard methods to detect HCV RNA will not detect vsRNAs , and thus are likely to underestimate the concentration of small RNA present in plasma and lymphocytes . The findings reported here clearly demonstrate that HCV RNA regulates PTPRE and Lck activation in vitro , and that HCV infection regulates TCR activation and PTPRE expression in vivo . Among potential targets identified by bioinformatics , VAPA , a proviral factor required for HCV replication [39–42] , and Grb 2 were not reduced by HCV vsRNA despite having putative binding sites in their 3’UTRs . The extent of complementarity between the HCV vsRNA and the VAPA and Grb2 3’ UTRs was the same as the PTPRE site 1 ( 38% ) . Neither VAPA nor Grb2 were regulated by HCV RNA , and placing PTPRE site 1 downstream of GFP did not lead to downregulation of GFP , suggesting that sequence diversity outside the conserved seed sequence that reduced the amount of complementarity with the potential target sequences is critical for the specificity of target gene regulation . Further supporting this hypothesis , HCV RNA-containing sera reduced GFP expression when PTPRE site 2 was placed downstream of the GFP coding region , and the extent of PTPRE reduction correlated with the complementarity between the HCV E2 RNA sequence detected in clinical isolates and the PTPRE 3’UTR ( Fig 8B ) . Our data also identified PTPRE as a novel factor regulated by HCV vsRNA in hepatocytes . This phosphatase activates Src-kinase signaling , and previous studies demonstrate an inhibitory role of Src-kinases in HCV replication [24 , 25] . Thus , vsRNA mediated inhibition of PTPRE expression in hepatocytes may promote viral replication in addition to contributing to T cell dysfunction . In summary , these data indicate that PTPRE plays an important role in T cell function and potentially HCV replication , and may serve as an attractive target for anti-HCV or immunomodulatory therapeutics .
HCV-infected subjects recruited from the University of Iowa Hepatology Clinic or healthy blood donors were invited to participate in this study . Characteristics of subjects are described in S2 Table . Liver biopsy protein was extracted by sonication in protein extraction buffer ( Tris-HCl , NP-40 , NaCl , EDTA , protease inhibitors , pH 7 . 5 ) , and cellular lysate protein concentrations were determined by Pierce BCA Protein Assay Kit . Human cell lines Huh7 cells ( obtained from the American Type Culture Collection ) Huh-7 . 5 ( kindly provided by Dr . Charles Rice ) , and Huh7D ( kindly provided by Dr . Dino Feigelstock ) , were cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum , 1% penicillin-streptomycin and 1% L-glutamine at 37°C in a 5% CO2 . HCV genomic RNA ( gRNA ) was transcribed from J6/JFH infectious clone as described [43] . PBMC isolation was performed as previously described [11 , 60] . Coding sequence for eGFP were ligated into a modified pTRE2-HGY plasmid ( Clontech , Inc . ) expressing GFP with an EMC IRES element directing translation as previously described [61] . The two putative target sites within the PTPRE 3’UTR were inserted after the GFP open reading frame ( Site 1 = TGCAGTTGGGTTCAAATGCCAAATAGTGATTAGAAGACGA ( 38% complementary to HCV vsRNA ) ; Site 2 = ATAGTGTTCGACTTCAAATGCCACGACGCGGCCG ( 56% complementary to HCV vsRNA ) and PTPRE sequences were confirmed by sequencing plasmid DNA ( University of Iowa DNA Core Facility ) . Jurkat ( tet-off ) cell lines ( Clontech , Inc ) were transfected ( Nucleofector II , Lonza Inc . ) and cell lines were selected for hygromycin and G418 resistance . GFP positive cells were bulk sorted ( BD FACS Aria , ( University of Iowa Flow Cytometry Facility ) and GFP expression was assessed by flow cytometry ( BD LSR II ) . All cell lines were maintained in RPMI 1640 supplemented with 10% heat-inactivated fetal calf serum , 2mM L-glutamine , 100 IU/ml penicillin , and 100 μg/ml streptomycin with hygromycin and G418 ( 200 μg/ml ) . PBMCs ( 2×106 cells/ml ) obtained from HCV-negative donors were resuspended in 200 μl serum obtained from HCV-infected donors before or after curative HCV therapy and incubated for 24 hrs prior to stimulation with plate-bound anti-CD3 ( 100 ng/ml , OKT3 clone , eBioscience ) . Alternatively , antigen-specific TCR-mediated activation was stimulated using pooled synthetic peptides ( 20 μg/ml ) with sequences derived from human cytomegalovirus , Epstein-Barr virus , and influenza viruses ( CEF control peptides , AnaSpec , EGT Group ) [62 , 63] . TCR-mediated signaling was determined 16 hours post-TCR stimulation by measuring IL-2 using ELISA as described [64 , 65] , or by measuring activated Lck protein as described below . Each experiment was performed in three replicate cultures , and in a minimum of three healthy donor PBMCs with consistent results . Plasma RNA was isolated from HCV-infected humans ( QIAmp Viral RNA Kit , Qiagen ) and cDNA generated using random hexamer primers as described [66] . HCV E2 was amplified using either genotype specific or degenerate primers: Sense 5’-WCDGGHCAYCGMATGGCD TGGGA and antisense 5’-GCAGAAGAACACGAGGAAGGASA . PCR products were cloned into the TA cloning vector pCR2 . 1 ( Invitrogen ) and automated DNA sequences obtained by the University of Iowa DNA Core Facility [67] . Huh7 . 5 cells were transfected by electroporation ( Bio-Rad Gene Pulser Xcell ) using 10μg HCV genomic RNA ( gRNA ) . HCV and control vsRNA duplexes were purchased from Integrated DNA Technologies and used at 1μM concentrations . Cells and HCV RNA were mixed in cold PBS ( 500μL ) transferred to a 4 mm gap-width electroporation cuvette and pulsed once at 270V and 950 μF capacitance . Transfected cells were maintained in complete medium for 96 hours at 37°C . Cell lysates were separated by SDS-PAGE gel electrophoresis , transferred to nitrocellulose membranes , and proteins detected by chemiluminescence as described [65 , 68] . Primary antibodies included phospho-Lck Y394 ( R&D Systems ) , PTPRE ( 4B2 ) and GAPDH ( Origene ) , PTPRE ( Rabbit; Abcam ) , or Actin ( Sigma ) . Immunoblots were quantified using ImageJ . Statistics were performed using GraphPad software V4 . 0 ( GraphPad Software Inc . ) . Student’s t test was used to compare results between groups . P values less than 0 . 05 were considered statistically significant . This study was approved by the University of Iowa Institutional Review Board ( IRB-01 ) and all subjects provided written informed consent . | The mechanism by which hepatitis C virus ( HCV ) establishes persistent human infection is complex and incompletely understood . Recent studies identified virus-derived small RNAs ( vsRNAs ) in HCV-infected cells; however , their biological significance is unclear . One HCV vsRNA arising from the E2 coding region reduces expression of a Src-kinase regulatory phosphtase ( PTPRE ) both in hepatocytes and lymphocytes in vitro , and leads to impaired T cell function . Here , we show that PTPRE expression is reduced in liver tissues and peripheral blood mononuclear cells ( PBMCs ) obtained from HCV-infected humans . Furthermore , serum from HCV infected individuals reduced antigen-specific TCR signaling , and curative anti-HCV therapy restored PTPRE expression in HCV-infected humans coincident to rescuing antigen-specific TCR-signaling defects . Transfection of a hepatocyte cell line with HCV genomic RNA or synthetic vsRNA duplexes inhibited PTPRE expression , recapitulating the in vivo observations . Together , these data suggest that HCV genomic RNA is processed into short , regulatory HCV RNA sequences that regulate PTPRE levels in HCV-infected humans , contributing to HCV immune evasion in vivo . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"blood",
"cells",
"transfection",
"complement",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"hepacivirus",
"pathogens",
"immunology",
"microbiology",
"fibrosis",
"viruses",
"developmental",
"biology",
"rna",
"viruses",
"tcr",
"signaling",
"cascade",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"t",
"cells",
"hepatitis",
"c",
"virus",
"hepatitis",
"viruses",
"molecular",
"biology",
"immune",
"system",
"biochemistry",
"signal",
"transduction",
"t",
"cell",
"receptors",
"cell",
"biology",
"flaviviruses",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"immune",
"receptors",
"cellular",
"types",
"cell",
"signaling",
"organisms",
"signaling",
"cascades"
]
| 2017 | Hepatitis C virus infection inhibits a Src-kinase regulatory phosphatase and reduces T cell activation in vivo |
Nodal and Activin are morphogens of the TGFbeta superfamily of signaling molecules that direct differential cell fate decisions in a dose- and distance-dependent manner . During early embryonic development the Nodal/Activin pathway is responsible for the specification of mesoderm , endoderm , node , and mesendoderm . In contradiction to this drive towards cellular differentiation , the pathway also plays important roles in the maintenance of self-renewal and pluripotency in embryonic and epiblast stem cells . The molecular basis behind stem cell interpretation of Nodal/Activin signaling gradients and the undertaking of disparate cell fate decisions remains poorly understood . Here , we show that any perturbation of endogenous signaling levels in mouse embryonic stem cells leads to their exit from self-renewal towards divergent differentiation programs . Increasing Nodal signals above basal levels by direct stimulation with Activin promotes differentiation towards the mesendodermal lineages while repression of signaling with the specific Nodal/Activin receptor inhibitor SB431542 induces trophectodermal differentiation . To address how quantitative Nodal/Activin signals are translated qualitatively into distinct cell fates decisions , we performed chromatin immunoprecipitation of phospho-Smad2 , the primary downstream transcriptional factor of the Nodal/Activin pathway , followed by massively parallel sequencing , and show that phospho-Smad2 binds to and regulates distinct subsets of target genes in a dose-dependent manner . Crucially , Nodal/Activin signaling directly controls the Oct4 master regulator of pluripotency by graded phospho-Smad2 binding in the promoter region . Hence stem cells interpret and carry out differential Nodal/Activin signaling instructions via a corresponding gradient of Smad2 phosphorylation that selectively titrates self-renewal against alternative differentiation programs by direct regulation of distinct target gene subsets and Oct4 expression .
Morphogens are secreted signaling molecules that orchestrate the spatial distribution and sequence of cellular differentiation events throughout embryonic development . The specific cell types , their localization and order of induction from recipient stem cell populations are determined by the concentration gradient of morphogens diffusing from the source of secretion . Previous studies have proposed some of the models by which morphogen gradients are initiated , established and stabilized including the level of receptor occupancy , positive/negative feedback and feed forward mechanisms [1]–[3] . However , little is understood about the transcriptional mechanisms responding to variable receptor activation and how they permit pluripotent stem cells to interpret signaling levels and direct the appropriate differentiation programs during mammalian development . Nodal and Activin are morphogens of the TGFβ superfamily of signaling molecules . In Xenopus embryos , Activin is a potent concentration-dependent inducer of mesoderm , mesendoderm and endoderm in animal cap cells [2] , [4] , [5] . Nodal has also been shown to be a classical morphogen in zebrafish where it functions in a concentration gradient independently of any relaying mechanisms [6] . In the early mouse embryo , mutations that perturb the level of Nodal/Activin signaling show that the pathway plays crucial roles in the induction of the primitive streak/mesoderm , mammalian organizer ( node ) , mesendoderm and endoderm during the establishment of the anterior-posterior axis [7]–[11] . In contrast to in vivo evidence that Nodal/Activin signaling predominantly promotes differentiation events , the pathway also paradoxically has important roles in the maintenance of self-renewal and pluripotency . Indeed Activin A is frequently used directly in culture for the continued propagation and expansion of human embryonic and mouse epiblast stem cells [12]–[15] . The signaling level of the Nodal/Activin pathway is determined by the overall activity of its components many of which have been identified . Both the Nodal and Activin ligands bind to the same type I/II serine-threonine receptor kinase complexes consisting of ActRIIA/B and Alk4/5/7 respectively in the mouse [16] . Nodal requires the cofactors Cripto/Criptic for receptor activation as opposed to Activin that can bind directly to the receptors and is inhibited by Cripto [17]– . Upon ligand docking , the Type I receptors phosphorylate the downstream signal transducers Smad2 and Smad3 ( Smad2/3 ) which form hetero- or homodimers and trimers [20] . Both Smad2/3 are also phosphorylated by crosstalk with EGF/ERK/MAPK signaling [21]–[23] but only the serine residues of the SSXS motif on the extreme carboxy terminus are specifically phosphorylated by Nodal/Activin/TGFbeta signaling . This phosphorylation is important for the translocation of Smad2/3 to the nucleus in association with Smad4 [24] , [25] where the complex recruits a number of transcription factors including FoxHI , p53 , β-catenin and Jun/Fos for the direct regulation of target genes [20] . Specificity of the Smads for their direct target genes is partly conferred by a DNA domain in the MH1 region to the Smad-binding DNA element ( SBE ) consisting of a basic CAGA sequence or its complement [26] . The other partner transcription factors within the complex are required for additional target gene affinity and specificity . While Smad2/3 share more than 90% protein homology , they are not functionally equivalent . Full-length Smad2 differs from Smad3 as the presence of an inhibitory domain in the MH1 region prevents direct DNA binding while Smad3 can bind directly to SBE boxes [27] . However , an alternatively spliced variant of Smad2 that lacks the inhibitory domain can bind DNA directly and has been shown to be the isoform that accounts for all developmental Smad2 functions in vivo [28] . The developmental roles of Smad2/3 are also disparate . Smad2 knockout mouse embryos fail to form mesoderm and endoderm due to defects in primitive streak specification after implantation at 6 . 5 dpc [29] closely phenocopying Nodal mutants [10] . In contrast , Smad3 mutant mice are born alive and are fertile but develop chronic intestinal inflammation leading to colorectal cancer [30] . This suggests that Smad2 is the primary transcriptional mediator of early developmental events while Smad3 is involved in immune function and possibly acts as a tumor suppressor postnatally . Our focus here is to clarify how mechanistically different levels of Nodal/Activin signaling lead to different embryonic stem ( ES ) cell fate decisions . ES cells were differentiated using three different quanta levels of Nodal/Activin signaling . We showed that ES cells are able to arbitrate between three distinct cell fate decisions . Maintenance of endogenous Nodal/Activin signaling is required for self-renewal of ES cells where any perturbation leads to an exit from self-renewal and pluripotency programs towards mesendoderm induction at high signaling and trophectoderm differentiation at low signaling . One obvious question to resolve is whether different levels of Nodal/Activin signaling recruit different sets of genes . While genome wide transcriptome studies have suggested possible Nodal/Activin targets , the identity of many transcriptional targets directly regulated by Smad2/3 remains unknown . One ChIP-chip study to date has been performed to address endogenous Smad2/3 binding in transformed human keratinocytes [31] while none have been carried out in the context of stem cell fate decisions , graded Nodal/Activin signaling or examining beyond promoter regions . Here we performed quantitative chromatin immunoprecipitation ( ChIP ) of phospho-Smad2 ( pSmad2 ) during graded Nodal/Activin signaling followed by massively parallel sequencing ( ChIP-Seq ) covering the full extent of pSmad2 binding to the ES cell genome including 5′/3′ UTRs , exons/introns and gene deserts . PSmad2 binding and regulation of direct target gene expression does not vary uniformly across the genome but changes in both a qualitative and quantitative manner with different signaling levels . Some targets are up- or downregulated proportionate to the activity of the Nodal/Activin pathway . However , separate subsets of target genes are regulated only during high or low signaling conditions . The downstream consequences of this is differential expression of the target genes that combine dose-dependent genes with different subsets of genes activated or repressed specifically for each signaling level . Thus ES cells carry out alternative cell fate decisions via the recruitment of target gene subsets in a pSmad2 dose-dependent manner . To reconcile some of the conflicting functions of Nodal/Activin signaling in self-renewal and pluripotency versus differentiation cell fate decisions , we examined the regulation of the Oct4 pluripotency and self-renewal master gene . Oct4 was directly regulated by pSmad2 binding in the promoter region independent of all other cis regulatory elements . Consistent with the modulation of pSmad2 binding , both endogenous mRNA and protein levels of Oct4 were also repressed by inhibition of Nodal/Activin signaling . Hence pSmad2 is a direct upstream regulator of Oct4 transcription where it permits an exit from maintenance of the stem cell state towards mesendoderm or trophectoderm differentiation programs as specified by the signaling level . In conclusion , the molecular switching of binding locations and target genes by pSmad2 across the ES cell genome in a dose-dependent manner provides a mechanism for the shift in the balance between maintenance of the stem cell state and the opposing induction of differentiation . Key signaling pathways have been predominantly studied in a binary context where they are either present or absent in a biological system . This view has only been able to account for some of their many and often conflicting roles . Our findings challenge this view and support multi-level signaling in stem cells where different signaling strengths can engender different cell fate decisions reflective of the in vivo development of embryos directed not just by Nodal/Activin signaling but possibly Hedgehog , FGF , Wnt and other morphogen pathways .
The direct cellular function of the Nodal/Activin pathway notably of the downstream components Smad2/3/4 is for the regulation of transcription . To address the relation between graded signaling and how they affect transcription , we quantified the changes in expression of known target genes under different signaling levels in chemically defined KSR media conditions . Pluripotent mouse embryonic stem ( ES ) cells were used to assess the mechanism of morphogen activity as they can differentiate into all tissue types of the adult and express all components of the pathway permitting response to manipulated Nodal/Activin signaling . Some of the known target genes include Pitx2 and Lefty2 which are responsible for the establishment of left-right asymmetry during early embryogenesis , a key developmental role of Nodal/Activin signaling [32] . In addition , both Lefty2 and Smad7 function as inhibitors of the pathway in a negative feedback mechanism for the attenuation of Nodal/Activin signaling strength [33] , [34] . Although direct Smad2/3 binding and regulation of the Pitx2 and Lefty2 genes have not yet been demonstrated , in vivo reporter assays suggest that specific enhancers are responsive to Nodal/Activin signaling and are active only on the left side of the embryo [35] , [36] . Moreover , these enhancers have been shown to contain FoxH1 binding sites , a known key transcriptional copartner of Smad2/3 . Smad7 has been shown to be a direct target of Smad2/3/4 binding in the promoter region by gel shift assays [37] , [38] and it antagonizes the interaction of Smad2/3 with the Type I kinase receptors [39] during negative feedback . Using real-time PCR quantitation , the expression of the 3 target genes was examined in the ES cells following the induction of high signaling by direct treatment with Activin in a time-course . In the reciprocal experiment , the small chemical inihibitor SB-431542 that specifically prevents the kinase domains of the Type I kinase receptors from phosphorylating Smad2/3 [40] was used to generate low Nodal/Activin signaling conditions . Pitx2 , Lefty2 and Smad7 were up- and downregulated in direct correlation with the level of signaling under chemically defined conditions compared to the DMSO carrier control representing endogenous or medium signaling ( Figure 1 ) . Over the course of 24 hours , the maximum expression of Pitx2 and Lefty2 occurred at 18 hours ( Figure 1A and 1B ) while that of Smad7 ( Figure 1C ) occurred earlier at 12 hours . We therefore conclude based on these known target genes that Nodal/Activin signal transduction and its effects on transcription require up to 18 hours to fully develop and any earlier time points result in weaker inductions . We further confirmed that Pitx2 , Lefty2 and Smad7 are direct targets of the Nodal/Activin pathway by conducting chromatin immunoprecipitation of phosphorylated Smad2 in the ES cells under the same chemically defined conditions at 18 hours followed by quantification of the enriched genomic DNA fragments by real-time PCR using tiling primers ( Figure S1 and Table S3 ) . The antibody used for the pulldown was raised against the phosphorylated serines 465 and 467 on the carboxy-terminus of Smad2 that are specifically targeted by TGFbeta signaling and not by EGF/ERK/MAPK signaling . At 18 hours where there is maximum expression of the 3 target genes , there was also a robust divergence in the level of pSmad2 binding according to the signaling level for the enhancers of Pitx2 and Lefty2 ( Figure S1A and S1B ) . Interestingly pSmad2 binding was invariant on the known TGFbeta response element of the Smad7 promoter ( Figure S1C ) . This suggested that Nodal/Activin target genes had different binding efficiencies for pSmad2 at each Nodal/Activin signaling level and this was not uniformly changed for all target genes . In conclusion , we confirm that Pitx2 , Lefty2 and Smad7 were direct targets of Nodal/Activin signaling and graded pSmad2 binding . Differential signaling sustained for 18 hours also leads to the maximum level of differential gene expression with clear changes in pSmad2 binding on the Pitx2 and Lefty2 genes . Given the downstream changes in pSmad2 binding and transcription of the known direct target genes , we next addressed how extracellular signal levels are translated into intracellular levels of signal transduction . We hypothesized that this could be directly related to changes in pSmad2 levels in ES cells as a consequence of Type I receptor kinase activity . Hence ES cells subjected to differential morphogen signaling conditions may be able to produce different amounts of pSmad2 in cells generating a corresponding differential level of intracellular signaling that leads to differential transcription . It has recently been shown that overexpression of the constitutively active Alk4 type I kinase receptor is sufficient to drive phosphorylation of Smad2 independent of all other Nodal/Activin receptor complex components [41] . Here we show that direct treatments of the ES cells with Activin and the specific Type I receptor kinase inhibitor SB-431542 in chemically defined conditions also tightly regulates receptor complex activity and produces the phosphorylation of Smad2 in a signaling dependent manner ( Figure 2A ) . During Activin stimulation ( high signaling ) for 18 hours , there is a defined 2-fold increase in pSmad2 levels while repression with 10 µM SB ( low signaling ) leads to a 2-fold decrease that is within the limits of physiological change compared to the DMSO vehicle control ( equivalent of medium signaling ) . Differential signaling had no effect on the equilibrium of total Smad2 suggesting that only phosphorylation and not regulation of the total Smad2 population is mediated by Nodal/Activin signaling . Hence extracellular signaling levels are translated into an equivalent gradient of intracellular Smad2 phosphorylation in ES cells . Subsequently we addressed the long-term consequences of increased and decreased signaling on ES cell fate decisions by examining how manipulation of the pathway recapitulates in vivo cell fate decisions by direct treatment with Activin or SB for 6 days . Analysis of a broad range of early cell fate markers ( Figure 2B ) shows that enhanced Nodal/Activin signaling promotes mesendoderm differentiation in ES cells with strong upregulation of mesendodermal lineage genes including Gsc , Mixl , Eomes and Fgf8 . The marker for mesoderm , Brachyury ( T ) , was also strongly induced although this was not reflected by the other mesodermal markers such as Flk1 and Tbx6 . This was consistent with the finding that T is also co-expressed in mesendoderm in vivo at the anterior primitive streak [42] . Taken together , this suggests that high signaling induced by Activin predominantly drives mesendoderm differentiation . Conversely , inhibition of the pathway with SB ( Figure 2B ) led to the upregulation of trophectoderm specific markers including Dlx3 , Esx and Hand1 and a less significant induction of extraembryonic primitive endoderm markers such as Gata4/6 and Pdgfra . Similar results were obtained when the ES cells were treated with recombinant Lefty1 protein for the same period of time ( data not shown ) suggesting that the trophectoderm induction was specific to low Nodal/Activin signaling . Interestingly there was no induction of mesendodermal markers as in the Activin treatment and instead some of these such as Gsc , Mixl and Fgf8 were strongly downregulated . Together , these results suggest that perturbation of the level of Nodal/Activin signaling and consequently endogenous Smad2 phosphorylation led to an exit from self-renewal in ES cells towards highly divergent cell fate decisions of either mesendoderm or trophectoderm differentiation . To confirm these results , fluorescent immunostaining was carried out to assess the protein markers of trophectodermal and mesendodermal lineages ( Figure S2 ) after differentiation in serum containing media . The cell fates obtained under these conditions are similar to the results from the marker analysis performed in chemically defined conditions . Differentiated cells staining positive for Mixl and Lim1 in the nucleus could be detected in Activin cultures . Similarly , Hand1 and placental Cadherin ( P-cad ) positive giant cells could also be derived from SB treated ES cells . Control treatments with a low dose of DMSO carrier ( 1/5000 dilution ) contained large populations of ES cells that stained strongly for Oct4 and SSEA-1 . These results confirmed that the level of Nodal/Activin signaling is responsible for at least 3 cell fate decisions . The endogenous level of signaling is permissive for self-renewal and maintenance of pluripotency , an increase in signaling leads to the induction of mesendoderm like cells while reduction of signaling results in trophectoderm differentiation . We hypothesized that for divergent differentiation programs to be initiated in ES cells , differential gene expression mediated by pSmad2 transcription would be a pre-requisite , which is in turn dependent on the level of Nodal/Activin signaling . Each discrete signaling threshold should induce an independent and unique transcriptional signature distinct from other thresholds . To determine the genetic targets regulated downstream of Nodal/Activin signaling and their pattern of expression , microarray analysis was carried out to examine genome-wide gene expression following Activin , DMSO or SB treatments in chemically defined KSR media for 18 hours . No significant changes in gene expression out of 26 , 000 probes could be detected between the DMSO and KSR media control suggesting that the effect of the low concentration of DMSO was negligible on ES cells ( Figure 3A ) . In contrast , Activin and SB treatments induced specific changes in gene expression compared to the DMSO and KSR media controls . Most significantly , we were able to identify subsets of target genes that were regulated by one signaling level and not the other consistent with our hypothesis of threshold specific target gene regulation . For example , 19 genes including Gdf15 , Msmb and Orai3 were consistently upregulated in Activin treated cells while showing no significant changes in SB . In contrast , a larger subset of 131 target genes were specifically up- and down-regulated only in the SB treatment and not in Activin . A core subset of 12 targets was co-regulated by both high and low signaling changing their expression in correlation with the treatment including the known Nodal/Activin target genes Lefty1/2 and Pitx2 that were upregulated by Activin and downregulated in SB . Interestingly , the number of SB regulated targets significantly exceeds that of Activin targets , suggesting that endogenous Nodal/Activin signaling in ES cells is high or near saturation levels such that a 2-fold increase in pSmad2 could only induce a smaller subset of genes compared to a 2-fold downregulation . Higher doses of Activin treatments and greater than 2-fold increases in pSmad2 may be required to mirror the strength of SB inhibition providing an explanation for asymmetric up- or downregulation of gene expression during different levels of signaling . Some of the target genes driven by Nodal/Activin signaling were indeed implicated in the mesoderm , endoderm and trophectoderm lineages . Fgf15 plays an important role in the development of cardiac mesoderm [43] and Chst15 is specifically expressed in definitive endoderm in vivo [44] with both targets being upregulated by Activin . For SB treatments , Gata3 , Tcfap2c and Igf2 were specifically upregulated . Gata3 is a driver of trophectoderm development [45] , [46] while Tcfap2c is expressed specifically in the placenta where it regulates essential ADA expression [47] , [48] and Igf2 is an imprinted gene that modulates nutrient supply between the placenta and fetus [49] , [50] . Together these target genes support some of the mesendodermal and trophectodermal differentiation programs that may be initiated at 18 hours after the induction of differential Nodal/Activin signaling . With longer-term graded Nodal/Activin signaling over 6 days differentiation , it is likely that additional target genes reinforcing the specification of both lineages may be brought into play over time . Lefty1 , Pitx2 , Fgf15 and Spsb1 were validated by RT-PCR ( Figure 3B ) to be co-regulated target genes of high , medium and low signaling displaying a gradient of expression following the signaling level . Cripto , Bcar3 , Nphs1 and Cdh3 were targets that were predominantly downregulated by SB inhibition of signaling showing no significant change during Activin stimulation . Conversely , the ID1/2/3 family of transcriptional repressors and Serping1 are specifically upregulated only by the SB treatment showing no difference in response to either Activin or the DMSO control . Hence we conclude that different thresholds of Nodal/Activin signaling are indeed able to regulate the expression of specific subsets of target genes providing an important explanation for the establishment of divergent differentiation programs . While whole genome microarrays are able to identify the putative subsets of target genes differentially expressed during specific Nodal/Activin signaling levels , this does not provide a molecular mechanism for how different target genes can be directly regulated by the same pathway at different signaling strengths . To address this question , we examined the recruitment of the pSmad2 transcription factor to target genes after subjecting ES cells to Activin , SB or DMSO control treatments in chemically defined KSR media that produce 2-fold up- and downregulation of Smad2 phosphorylation by 18 hours . ChIP-Seq of pSmad2 was employed to identify where pSmad2 was binding on a whole genome scale in parallel cultures of ES cells under the 3 signaling conditions . ChIP samples from each condition were sequenced to a similar depth of 10-13 million tags . Interestingly the number and magnitude of pSmad2 binding events did not correspond to the 2-fold up- or downregulation of pSmad2 in ES cells under Activin and SB treatments . In fact the greatest number of binding peaks ( 7423 ) occurred in the control DMSO condition that maintains self-renewal and pluripotency of ES cells ( Figure 4A ) . When homeostatic Nodal/Activin signaling was perturbed by Activin and SB treatments , the number of binding events decreased to 5094 and 4859 respectively suggesting that any change in the levels of endogenous pSmad2 from the ES cell undifferentiated condition also caused a dynamic change in pSmad2 binding across the ES cell genome . The lower numbers may also be reflective of the transition where pSmad2 is dissociating from former target genes and establishing the recruitment of new genes . This was further supported by the percentage of overlapping peaks that were common to the 3 treatments being relatively small at 10 . 3% with a significantly larger number of unique peaks appearing in specific treatments ( 37 . 25% in DMSO , 20 . 44% in SB and 19 . 5% in Activin out of 12979 total peaks in the union ) . A previous study has profiled Smad2/3 binding sites using promoter arrays in human keratinocytes [31] . However , reporter assays on Nodal/Activin responsive target genes such as Lefty1/2 , Nodal [36] , [51] and Pitx2 [35] suggest that Smad2/3 may also regulate DNA elements in the introns rather than at the promoter region . Consistent with the reporter assay studies , our ChIP-Seq data showed that the majority of pSmad2 binding ( Figure 4B ) occurs in introns ( ∼30% ) with only a minority of sites in the proximal promoters ( ∼10% ) . Furthermore , there was a significant shift in pSmad2 binding from the distal 5′ and 3′ regions towards the promoters of genes in the SB treatment compared to DMSO and Activin ( Figure 4B ) . Examination of binding specifically in the promoter region showed a clear preference for pSmad2 to associate in the +/−600 bp proximal region of transcriptional start sites ( TSS ) with a steady decrease in binding further away from the TSS ( Figure 4C ) . In addition , the increase in number of pSmad2 binding peaks during low signaling with the SB treatment can be confirmed in the promoter region both up- and downstream of the TSS . In conclusion , pSmad2 binding , similar to the changes in gene expression identified by microarrays , also demonstrates binding to distinct subsets of genomic locations at different signaling levels . We next examined the relationship governing the degree of pSmad2 binding and the level of transcription across the genome ( Figure 5A ) . In all 3 conditions , a clear trend emerges suggesting that more pSmad2 binding drives higher levels of gene expression . However , the possibility that pSmad2 is not driving expression but preferentially associates with more transcriptionally active genes and open chromatin cannot be excluded . To distinguish between the 2 possibilities , we examined the trend between pSmad2 binding events and differential gene expression from the microarray analysis in the 3 signaling conditions . Indeed , a significant majority ( 64 . 2% ) of microarray target genes had pSmad2 binding within +/−50 kb and all displayed >1 . 5 fold change in binding in each signaling condition or had different number of binding events or changed the location of pSmad2 binding ( Table S1 ) suggesting that the pattern of gene expression was indeed dynamically driven by pSmad2-DNA interactions . To account for how pSmad2 is able to switch binding locations during differential Nodal/Activin signaling , we examined its preference for specific DNA motifs under each condition . It is known that Smad2/3 are able to bind directly the basic CAGA motif and at the same time they possess a number of partner transcription factors that modulate the specificity and strength of binding . Here we see that there is strong pSmad2 association with the basic CAGA SBE specifically in the Activin treatment ( Figure 5B ) . This was also confirmed when we examined the strong CAGAC canonical SBE as defined by the TRANSFAC PWM database which also appears with high frequency at the center of pSmad2 ChIP-seq peaks in the Activin treatment and not in DMSO , SB or the random mouse genome sequence control ( Figure 5C ) . This suggests that both CAGA and CAGAC displayed graded pSmad2 binding that varied with the signaling level and were preferentially bound in the Activin condition . To compare the contribution of CAGA against non-CAGA sequences towards pSmad2 binding , the top 10 de novo motifs in each condition were identified using the Weeder program ( Figure 5D ) . Motifs that occurred with significant frequency but were not enriched in the center of pSmad2 ChIP-Seq peaks were excluded to remove the influence of comotifs around the peaks . A number of non-CAGA motifs that occurred with similar or greater frequency than CAGA were isolated . Interestingly , these de novo motifs also showed a graded effect on pSmad2 association similar to the CAGA SBE . Other non-CAGA motifs were preferentially bound by pSmad2 only in the DMSO and SB condition and depleted during the high signaling Activin condition . This suggested that while CAGA binding was significant , binding to non-CAGA sequences accounted for the majority of pSmad2 association within the ES cell genome suggesting that this was primarily mediated by transcriptional co-partners . Indeed , when the top consensus motifs in the center of all ChIP-seq peaks in each signaling condition and in the combined dataset were studied ( Figure 5E ) , there was a strong enrichment for motifs belonging to transcription partners such as E2f and Ap1 instead of Smad binding CAGA boxes . To confirm the association of the putative transcriptional cofactors and establish their identity , we expanded the analysis to TRANSFAC co-motifs occurring within +/− 1 kb range of pSmad2 binding sites ( Table S2 ) . A large number of known pSmad2 transcription partners such as Ap1 , Sp1 and E2f are indeed associated within the vicinity of pSmad2 peaks regardless of the level of Nodal/Activin signaling . However , there were additional co-motifs bound by transcription factors such as Oct4 , Stat3 and p53 that only appear prominently in Activin treatments and Hes1 , Lrf and Plzf appearing in SB . This is supportive of an exchange of transcription partners in association with pSmad2 that was governed by the level of Nodal/Activin signaling which was likely to be responsible for the change in specificity of pSmad2 transcriptional complexes for target gene subsets and their level of expression . Furthermore , while pSmad2 does bind to its own CAGA sequence , transcriptional copartners played a greater role both in binding affinity and specificity of pSmad2 protein complexes for the ES cell genome . To investigate the different models of pSmad2 binding during differential Nodal/Activin signaling , we examined the ChIP-Seq profiles including those of the transcriptionally regulated microarray targets and identified at least 4 types of pSmad2 binding . The first model is that of “graded” target genes that follow closely the changes in Nodal/Activin signaling with increased binding and transcription during high signaling , have moderate response in endogenous baseline signaling and showed a loss of binding with decreased mRNA levels during signaling repression . This category of pSmad2 binding comprises 23 . 87% of high confidence ChIP-Seq peaks corresponding to 16 . 28% of target genes associated within +/−50 kb of these peaks ( Figure S4 ) . Radil a Rap GTPase effector that plays a role in the migration of neural crest progenitors [52] exemplified such pSmad2 binding and transcriptional regulation ( Figure 3A , Figure 6A , and Table S1 ) in the first intron with normalized relative enrichments of 107 tags in Activin compared to 51 in the DMSO control and complete loss of binding indistinguishable from background sequencing levels in SB . The known target gene Pitx2 showed reproducible results with the ChIP data obtained by real-time PCR ( Figure S1A ) both in terms of the binding location in the intronic enhancer as well as the level of pSmad2 enrichments under graded Nodal/Activin signaling . There were normalized enrichments of up to 201 tags in Activin , 156 in DMSO control and again complete loss of binding in SB ( Figure S3A ) . Interestingly , Pitx2 had 2 graded binding sites , one of which is in the known intronic region and a novel site in the 3′ region . The graded binding in the Pitx2 locus also correlates with transcriptional consequences showing strong induction/inhibition of Pitx2 mRNA levels from 0 to 24 hours ( Figure 1A ) . The two inducers of the mesendoderm cell fate Mixl [53] and Nodal [54] also show evidence of graded pSmad2 binding within 50 kb of the genes ( Figure S5A and S5B ) suggesting that they may be directly regulated by Nodal/Activin signaling for this purpose . The binding location in the first intron of Nodal also corresponds to the intronic enhancer previously described to be important for left side expression in the early embryo via the Nodal/Activin signaling autoregulatory loop [55] , [56] confirming that Nodal is itself a direct target . It was also unclear if pSmad2 binding and regulation of target genes only exists in a 1-to-1 relationship or if the same binding sites were capable of regulating multiple targets in the genomic vicinity . While Lefty2 was a known direct target with pSmad2 binding in its promoter region ( Figure S1B ) , for the first time , to our knowledge , we characterized an important pSmad2 transcriptional hotspot in the entire 100 kb Lefty1/2 locus where all the genes within this region were co-regulated by pSmad2 binding suggesting a coordinated mode of transcriptional regulation ( Figure S3B ) . This was further confirmed in the microarray analysis ( Figure 3A ) demonstrating that Lefty1/2 , Pycr2 and Tmem63a display the same pattern of gene expression following a graded response to Nodal/Activin signaling . This was consistent with the real-time PCR quantification of the pSmad2 pulldown of the Lefty2 promoter ( Figure S1B ) that corresponds to the most upstream pSmad2 binding site in the Lefty1/2 hotspot as did a time course profiling of Lefty2 expression from 0 to 24 hours ( Figure 1B ) . In the second model of pSmad2 binding , we describe “low signaling dominant” conditions that permit pSmad2 binding but less so under other signaling levels . The Id1/2/3 ( Figure 6B , Figure S6 , and Table S1 ) family of transcriptional repressors shows pSmad2 binding to these genes only in the SB treatments and not in Activin or the DMSO control . Statistically , 32 . 73% of pSmad2 binding sites display this mode of behavior associated with 23 . 44% of target genes ( Figure S4 ) . In contrast , the third model showed the opposite “high signaling dominant” mode of binding such as in the case of 220011C2Rik ( Figure 6C and Table S1 ) where pSmad2 only binds strongly in the Activin condition but to a lesser degree in DMSO or SB also resulting in transcriptional consequences ( Figure 3A and Figure S4 ) . Another known component of the mesendodermal cell fate Fgf8 [57] also shows strong pSmad2 binding in the promoter region specifically during high signaling ( Figure S5C ) . Intriguingly , findings in the chick embryo show that Fgf8 also plays important roles in left-right asymmetry where it can be induced by Activin [58] in agreement with our results . In the fourth model which accounts for the regulation of the largest proportion ( 33 . 69% ) of target genes associated with pSmad2 ChIP-Seq peaks ( ) , the same target gene may be regulated by “multimodal pSmad2 binding” events . Copz2 has two pSmad2 association sites in the intron and promoter region ( Figure 6D ) . The promoter site only binds pSmad2 in the SB condition while the intronic enhancer shows a graded response to the signaling level . In the case of the known target gene Smad7 , we have shown that the pSmad2 binding peak in the promoter region is invariant in all 3 signaling conditions ( Figure S1C ) which could not explain how Smad7 was differentially expressed during graded Nodal/Activin signaling ( Figure 1C ) . In confirmation with these results , the ChIP-Seq data showed the same pSmad2 association on the Smad7 proximal region with no change in binding under all 3 signaling conditions . Surprisingly , we discovered a previously undescribed pSmad2 regulatory element in the distal Smad7 promoter region that binds pSmad2 in a graded manner ( Figure S3C ) and could account for why Smad7 was responsive to different Nodal/Activin signaling levels . Hence pSmad2 binding in the Smad7 proximal region may not be the dominant regulatory region for Nodal/Activin signaling but may depend instead on the dynamically changing pSmad2 distal promoter element for Smad7 regulation . Indeed the proximal promoter element may be more of a Smad3 regulated region instead of Smad2 as previously described [38] . In conclusion we demonstrate that pSmad2 dependent binding and transcription during graded Nodal/Activin signaling occurs in the ES cell genome in a graded , low or high signaling dominant , many-to-one or one-to-many multimodal manner in relation to the target genes that they regulate . The mesendodermal and trophectodermal cell fate decisions brought about by graded Nodal/Activin signaling strikingly resemble the ES cell response to a less than 2-fold up- or downregulation of the Oct4 master regulator of stemness in driving differentiation towards similar cell fates [59] . Furthermore , an important mechanism for trophectoderm differentiation depends on the Oct4 repression of Cdx2 expression and the induction of this lineage is thought to be indicative of loss of stemness [60] . We therefore hypothesized that Oct4 may be a key downstream target under Nodal/Activin control during the specification of divergent cell fate decisions and investigated how the pathway may be governing Oct4 . We discovered that the Oct4 locus was rich in multiple pSmad2 binding events from ChIP-Seq profiling ( Figure 7A ) . During graded Nodal/Activin signaling in chemically defined conditions however , only a pSmad2 peak in the promoter region of Oct4 showed a similarly graded response suggesting that this was the functional Nodal/Activin signaling response element . We examined the transcript levels of endogenous Oct4 expression ( Figure 7D ) upon inhibition of Nodal/Activin signaling with SB in serum containing media and found that it was also significantly downregulated within 24 hours . In agreement with the transcript data , Oct4 protein levels were similarly downregulated in SB treated ES cells ( 7E ) . Analysis of the 503 bp promoter region encompassing the beginning and end of the pSmad2 binding peak showed that it contained eight CAGA sites or their inversion ( Figure 7B ) . To determine if this regulatory sequence was indeed a Nodal/Activin response element of the Oct4 promoter , we cloned this into luciferase reporter constructs and transfected ES cells subjected to the 3 signaling conditions with Activin , DMSO and SB ( Figure 7C ) in serum containing media . The reporter activity of the wild type Oct4 promoter construct was >100X higher than that of the empty reporter construct in the DMSO control signaling condition suggesting that the 503 bp sequence had strongly driven Oct4 promoter activity in ES cells . Crucially , the Oct4 promoter reporter displayed a specific graded response to Nodal/Activin signaling while the control empty reporter did not . To confirm that the Oct4 response to graded Nodal/Activin signaling was functionally driven by pSmad2 binding , we determined the exact SBE responsible for Oct4 inducibility ( Figure 7C ) . Mutagenesis experiments on the Oct4 promoter region in luciferase assays revealed that the strong CAGAC consensus SBE site in the middle of the 503 bp fragment was indispensable for graded Oct4 promoter activity . Loss of this site completely abolished the promoter response to both high and low Nodal/Activin signaling . Further point mutations of two minimal CAGA SBEs flanking the CAGAC site led to no further significant effects on the Oct4 promoter . We therefore conclude that Oct4 is a direct target of pSmad2 binding and Nodal/Activin signaling regulates both its promoter activity and endogenous expression . The 503 bp Oct4 promoter response element with the essential CAGAC SBE was sufficient and independent of all other pSmad2 binding events in the Oct4 locus or other DNA regulatory elements in cis that may be mediated by Nodal/Activin signaling . The regulation of Oct4 is well known for its importance in cell fate decisions and its downregulation during loss of Nodal/Activin signaling is significant not only as an impetus for trophectoderm differentiation but also reconciles the alternative role of Nodal/Activin signaling in maintaining self-renewal and pluripotency .
The molecular basis of extracellular signaling instructions governing differential cell fate decisions in the Nodal/Activin pathway has been postulated but not shown conclusively . Primarily , the transcriptional events occurring at the interface between pSmad2 signal transduction from the activated cell surface receptors to manipulation of the global stem cell transcriptome driving specific lineage programs have not been well characterised . This study provides an important insight into how quantitative signaling is translated into qualitative cell fate decisions by showing for the first time , to our knowledge , that the same transcription factor pSmad2 is able to bind and transcriptionally regulate different subsets of target genes in a dose-dependent manner . The specification of cell fate decisions is governed by 2 distinct events . The first requires an exit from self-renewal and maintenance of stemness programs by direct control of pSmad2 over key pluripotency factors . Previous studies have revealed that Nanog is a direct target of Smad2/3 transcription in human ES cells [61] . Here we show an additional level of control over the stem cell program by direct transcriptional regulation of the Oct4 master pluripotency gene by pSmad2 . The second event requires an entry into a specific differentiation program that is in turn brought about by direct and indirect pSmad2 regulation of differentiation genes such as Gata3 , Tcfap2c and Igf2 that are known to be important factors for trophectoderm cell fates . This cell fate decision is further reinforced by loss of Oct4 with inhibition of Smad2 phosphorylation as the former is known to be a potent repressor of the trophectoderm gene Cdx2 in the blastocyst [60] . The pSmad2 binding target genes driving mesendodermal differentiation include Mixl , Fgf8 and Nodal itself , while other genes such as Chst15 expressed in definitive endoderm and Fgf15 for cardiac mesoderm have also been identified as strong Nodal/Activin transcriptional targets . It is likely that over the course of long-term differentiation for 6 days , additional target genes may be recruited for the specification of both lineage decisions that may not be apparent at the 18 hours time point in this study which may be too early for endpoint differentiation . Indeed , strong regulation of Mixl and Fgf8 and to a lesser extent Nodal could be detected at 3 and 6 days ( Figure 2B and data not shown ) of treatment in correlation with the level of Nodal/Activin signaling . Consistent with the role of Nodal/Activin as morphogens , we found that many components of the pathway were themselves feedback targets that were directly regulated by pSmad2 binding in ChIP-Seq and/or differentially expressed in our microarray analysis . These include the negative feedback inhibitors such as Lefty1/2 and Smad7 which are already known targets of Nodal/Activin signaling . In this study , graded pSmad2 binding could be detected in the intronic region of Tmepai ( Figure S6 ) which sequesters Smad2/3/4 from receptor kinase activity [62] . Similarly , SnoN [63] and Ski [64] also present graded intronic binding of pSmad2 ( Figure S6 ) and both function as transcriptional repressors of Smad2/3/4 . There were also positive feedback components such as Nodal , its cofactor Cripto and FoxH1 the transcriptional copartner of Smad2 ( Figure 8 and Figure S6 ) that show graded binding in the intron and promoter regions . The preponderance of the negative components in the autoregulatory loop of Nodal/Activin signaling is significant , as it suggests that the pathway mainly dampens and attenuates its own signaling via negative feedback and less so by positive feedback loops mediated by Nodal , Cripto and FoxH1 . One of the intriguing findings is that extracellular signaling gradients were translated into a gradient of Smad2 phosphorylation that we have now shown to be able to recruit different target genes in a dose-dependent manner . This was possibly achieved by an exchange of transcriptional copartners that permits the shifting of the pSmad2 transcriptional complex to different target gene subsets as suggested by the differential recruitment of non-CAGA motifs and comotifs under each signaling condition . The fact that pSmad2 contains only CAGA sequence binding domains and not transcription activation domains suggest that it is further dependent upon copartners for transcription , binding affinity and specificity . In some cases graded pSmad2 transcription complex binding drives graded target gene response that follows signaling strength with high fidelity . In other cases , the target genes are only regulated and responsive at defined signaling thresholds ( Figure 8 ) . The consequence is that a relatively modest stimulation with Activin leading to a physiological 2-fold increase in Smad2 phosphorylation eventually drives mesendodermal differentiation while the reciprocal SB inhibition resulting in a 2-fold decrease of pSmad2 is able to promote trophectoderm cell fates . During this process , the master regulator of pluripotency Oct4 is itself titrated by the same Nodal/Activin signaling gradients in the ES cells undergoing differentiation . Hence the same pathway is able to tilt the balance in favor of maintenance of pluripotency or mediate an exit from self-renewal and entry into a specific lineage program . In conclusion , this study for the first time , to our knowledge , reconciles the multiple divergent roles of Nodal/Activin signaling in both pluripotency and differentiation with pSmad2 playing a central role in the cell fate decision making process .
E14 TG2A mouse embryonic stem cells ( ATCC ) were propagated in FBS media consisting of 20% ES cell-qualified FBS in DMEM supplemented with 100 µM non-essential amino acids , 100 U/ml penicillin , 100 µg/ml streptomycin , 2 mM GlutaMAX-I ( Invitrogen ) , 55 µM β-mercaptoethanol ( Sigma ) and 1X homemade Leukemia inhibitory factor ( LIF ) . For the establishment of Nodal/Activin signaling gradients in chemically defined conditions , KSR media containing 20% Knockout Serum Replacement ( KSR , Invitrogen ) in place of FBS with all other components of ES media excluding LIF were used . For acute ( 0 to 48 hours ) signaling conditions , 25000 ES cells/cm2 were plated for 18 hours in FBS media followed by adaptation of the cells to chemically defined conditions with 10 µM SB-431542 ( Tocris ) in KSR media for 6 hours as previously described [41] . High signaling was induced by treatment with KSR media containing 25 ng/ml Activin ( R&D Systems ) or low signaling with 10 µM SB and maintenance of endogenous signaling with control KSR media or 1/5000 dilution of DMSO vehicle as indicated . For long-term differentiation , 2000 ES cells/cm2 were plated and 18 hours later directly treated with Activin , DMSO and SB in FBS media without LIF or KSR media for 6 days with media change everyday . The DMSO vehicle used to dissolve SB can induce differentiation and loss of pluripotency in ES cells [65] , [66] . In the microarray analysis of the 3 signaling conditions , the effect of DMSO on differential gene expression was determined by comparing against the unsupplemented KSR media control ( Figure 3A ) . The SB inhibitor was used at a high stock concentration of 50 mM permitting 5000X dilution of DMSO in ES cell cultures which was well below the limit required for differentiation . The cultures and treatments were carried out for the microarray study in 4 biological replicates consisting of ES cells at 4 different passages from P20 to P24 to identify and eliminate any cell culture variation effects from analysis . ES cells were lysed in RIPA buffer ( 150 mM NaCl , 1% NP-40 , 0 . 5% Sodium Deoxycholate , 0 . 1% SDS , 50 mM Tris pH 8 . 0 ) for protein extracts . SDS-PAGE was performed on 10% polyacrylamide gels and transferred on Immun-Blot PVDF membranes ( Bio-rad Laboratories ) followed by probing with 1∶1000 dilutions of rabbit anti-Smad2 ( pSer465/467 , Calbiochem ) , rabbit anti-Smad2 ( Invitrogen ) , mouse anti-Pcna ( Santa Cruz Biotechnology ) and goat anti-Oct4 ( Santa Cruz Biotechnology ) . Secondary antibodies used were 1∶1000 donkey anti-rabbit IgG-HRP ( GE Healthcare ) , 1∶1000 goat anti-mouse IgG-HRP and 1∶2500 donkey anti-goat IgG-HRP ( Santa Cruz Biotechnology ) . Densitometry measurements of protein bands on western blots were acquired using Photoshop CS3 ( Adobe Systems Incorporated ) . For gene expression , total RNA was extracted from cells using the RNeasy Mini kit ( Qiagen ) as per manufacturer instructions . This was reverse transcribed into cDNA using the High Capacity RNA-to-cDNA Master Mix ( Applied Biosystems ) . Quantitative real-time PCR was performed on the 7900HT Fast Real-Time PCR System ( Applied Biosystems ) or the Biomark System ( Fluidigm Corporation ) on cDNA or ChIP DNA according to manufacturer instructions . For RT-PCR , products amplified for 25 to 33 cycles were resolved on a 2 . 5% agarose gel . Primer sequences for both ChIP-qPCR and gene/marker expression can be found in Table S3 . Total RNA was reverse transcribed into cDNA and in vitro transcribed into biotin-labeled cRNA using the Illumina TotalPrep RNA Amplification kit ( Ambion ) . This was hybridized on MouseRef-8 v2 . 0 Expression BeadChips ( Illumina ) . Raw intensity values were subjected to background subtraction on the BeadStudio Data Analysis Software ( Illumina ) and normalized using the cross-correlation method [67] . Differential gene expression was identified based on a fold change cutoff of >1 . 5 compared to the DMSO control . The microarray data was deposited in NCBI GEO with accession number GSE23239 . Chromatin Immunoprecipitation ( ChIP ) using a certified ChIP-grade rabbit polyclonal anti-Smad2 ( phospho S465+S467 ) antibody ( ab16509 , Abcam ) was carried out in ES cells under chemically defined high , medium or low Nodal/Activin signaling conditions according to the Agilent Mammalian ChIP-on-chip protocol v9 . 1 up to the ChIP DNA purification step . Adapter ligation , library amplification and size selection in the 200–300 bp range were performed according to the Illumina ChIP Sample Prep protocol ( #11257047 , Rev . A ) . Massively parallel sequencing was carried out for ChIP samples in all 3 signaling conditions with their respective input DNA controls on the Genome Analyzer ( Illumina ) up to a sequencing depth of at least 10×106 tags pass filter . The ChIP-Seq data was deposited in NCBI GEO with accession number GSE23581 . Details of the ChIP-Seq , motif and statistical analysis can be found in the Text S1 . The 503 bp fragment of the mouse Oct4 promoter region corresponding to chr17:35640683–35641185 was cloned into the pGL4 . 23[luc2/minP] Firefly luciferase reporter construct ( Promega ) to generate pGL4 . 23 Oct4 . This construct was point mutated by oligo cloning into unique StuI and NsiI sites to produce pGL4 . 23 m4 Oct4 ( CAGAC mutated to CATGC ) and pGL4 . 23 m345 Oct4 ( TCTGGGCAGACGGCAGA mutated to TATGGGCATGCGGCATA ) . The constructs were transfected into mouse ES cells in an 80∶1 ratio with the pGL4 . 75[hRluc/CMV] Renilla luciferase co-transfection control ( Promega ) using Lipofectamine 2000 ( Invitrogen ) . A control transfection was included with an 80∶1 ratio of the empty pGL4 . 23 vector to pGL4 . 75 Renilla control . Immediately after lipofections , the ES cells were pretreated with FBS media without LIF and with 10 µM SB for 6 hours . Subsequently the cells were split into replicates and plated in FBS media without LIF containing 25 ng/ml Activin , 1/5000 DMSO vehicle or 10 µM SB , which induces high , medium or low Nodal/Activin signaling respectively for 8 hours . The cells were washed once in PBS and lysed in 1XPassive Lysis Buffer and luciferase assays were performed using the Dual-Luciferase Reporter Assay System on the GloMax 96 Microplate Luminometer with Dual Injectors ( Promega ) . | Nodal and Activin are extracellular signaling molecules that diffuse from the source of secretion and induce recipient stem cells to become new cell types according to a concentration gradient . In the early embryo , they are important for the specification of tissues at the correct place and time , but paradoxically they drive the opposite function in embryonic and epiblast stem cells where they maintain the stem cell state instead of promoting differentiation . The molecular basis of how the level of signaling determines stem cell fate decisions remains poorly understood . We found that Smad2 , the main transcription factor of the Nodal/Activin pathway was phosphorylated according to the level of signaling . By mapping where phospho-Smad2 binds in the embryonic stem cell genome and how this affects transcription of the associated target genes , we show that phospho-Smad2 can recruit and regulate different sets of target gene depending on the signaling level . Moreover , phospho-Smad2 also directly regulates Oct4 , a master gene controlling the stem cell state thereby reconciling the opposing functions of the Nodal/Activin pathway in differentiation versus self-renewal programs . The pathway can mediate the exit from self-renewal via Oct4 and simultaneously drives differentiation towards particular lineages by recruiting the relevant gene subsets for this purpose . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genome",
"expression",
"analysis",
"signaling",
"networks",
"signaling",
"in",
"selected",
"disciplines",
"sequence",
"assembly",
"tools",
"cell",
"differentiation",
"dna",
"transcription",
"genome",
"sequencing",
"developmental",
"biology",
"genome",
"analysis",
"tools",
"developmental",
"signaling",
"stem",
"cells",
"sequence",
"analysis",
"signaling",
"in",
"cellular",
"processes",
"embryonic",
"stem",
"cells",
"gene",
"expression",
"regulatory",
"networks",
"smad",
"signaling",
"biology",
"molecular",
"biology",
"transcriptomes",
"microarrays",
"signal",
"transduction",
"tgf-beta",
"signaling",
"cascade",
"genomics",
"molecular",
"cell",
"biology",
"computational",
"biology",
"genetics",
"and",
"genomics",
"cell",
"fate",
"determination",
"signaling",
"cascades"
]
| 2011 | Graded Nodal/Activin Signaling Titrates Conversion of Quantitative Phospho-Smad2 Levels into Qualitative Embryonic Stem Cell Fate Decisions |
Soil-transmitted helminths ( STHs ) are a major health concern in tropical and sub-tropical countries . Oesophagostomum infection is considered endemic to West Africa but has also been identified in Uganda , East Africa , among primates ( including humans ) . However , the taxonomy and ecology of Oesophagostomum in Uganda have not been studied , except for in chimpanzees ( Pan troglodytes ) , which are infected by both O . bifurcum and O . stephanostomum . We studied Oesophagostomum in Uganda in a community of non-human primates that live in close proximity to humans . Prevalence estimates based on microscopy were lower than those based on polymerase chain reaction ( PCR ) , indicating greater sensitivity of PCR . Prevalence varied among host species , with humans and red colobus ( Procolobus rufomitratus ) infected at lowest prevalence ( 25% and 41% by PCR , respectively ) , and chimpanzees , olive baboons ( Papio anubis ) , and l'hoest monkeys ( Cercopithecus lhoesti ) infected at highest prevalence ( 100% by PCR in all three species ) . Phylogenetic regression showed that primates travelling further and in smaller groups are at greatest risk of infection . Molecular phylogenetic analyses revealed three cryptic clades of Oesophagostomum that were not distinguishable based on morphological characteristics of their eggs . Of these , the clade with the greatest host range had not previously been described genetically . This novel clade infects humans , as well as five other species of primates . Multiple cryptic forms of Oesophagostomum circulate in the people and primates of western Uganda , and parasite clades differ in host range and cross-species transmission potential . Our results expand knowledge about human Oesophagostomum infection beyond the West African countries of Togo and Ghana , where the parasite is a known public health concern . Oesophagostomum infection in humans may be common throughout Sub-Saharan Africa , and the transmission of this neglected STH among primates , including zoonotic transmission , may vary among host communities depending on their location and ecology .
Soil-transmitted helminths ( STHs ) are parasitic nematodes that cause infection via eggs and larvae , which are shed in feces and persist in the soils of tropical and sub-tropical countries [1] . STHs infect over one billion people worldwide [2] and may cause a combined disease burden as substantial as that caused by malaria or tuberculosis [3] . Nevertheless , these parasites are largely neglected in research , perhaps in part because the diseases they cause are suffered by the world's most impoverished populations [1] . Although roundworms ( Ascaris lumbricoides ) , hookworms ( Necator americanus and Ancylostoma duodenale ) and whipworms ( Trichuris trichiura ) are of global importance , other “lesser” parasites are localized to specific regions [1] . This includes Oesophagostomum spp . , a genus of nodule-causing worms with L3 larvae that are infective via ingestion after 4–7 days [4]–[7] . The human burden of Oeosphagostostomum infection is considered localized to West Africa , specifically the countries of Togo and Ghana [5] , [8] , [9] . A variety of mammals , including pigs , ruminants [10] , [11] , and non-human primates are frequently parasitized by Oesophagostomum . Infections in wild primates appear to be asymptomatic; clinical signs and mortality due to Oesophagostomum have only been recorded in captive settings [10] , [12] . Eight species of Oesophagostomum have been recorded in wild primates , of which the three most common , O . bifurcum , O . stephanostomum , and O . aculeatum , are able to infect humans [5] , [7] , [11] , [13] . Of these , O . bifurcum appears to be the only species to regularly parasitize humans , with human infections by other species considered incidental [5] , [9] . In Togo and Ghana , the majority of human Oesophagostomum cases occur within endemic foci [5] , [8] and affect 20% and 90% of the population , respectively , with prevalence highest in rural areas [12] , [14] , [15] . The only known species to cause infection within these countries is O . bifurcum , which also infects the region's non-human primates , including patas monkeys ( Erythrocebus patas ) , mona monkeys ( Cercopithecus mona ) , and olive baboons ( Papio anubis ) [4] , [10] , [16] . However , previous research has indicated that O . bifurcum is not commonly transmitted among primate species ( including humans ) because different parasite variants within the species are adapted to specific hosts [4] , [16] , [17] . In Uganda , a number of primate species harbor Oesophagostomum , as evidenced by microscopic detection of eggs in feces . These include members of the primate subfamilies Cercopithecinae and Colobinae , as well as chimpanzees ( Pan trogolodytes ) [18]–[20] . There have also been reports of oesophagostomiasis in human patients in Uganda , although no such reports , to our knowledge , have been published since the 1980s [5] , [21] , perhaps due to under-reporting or improvements in treatment . With the exception of chimpanzees , which are infected with both O . bifurcum and O . stephanostomum [22] , the species of Oesophagostomum infecting Ugandan primates and humans remains unknown . In this study , we examined Oesophagostomum within the primate community of Kibale National Park , Uganda , using a combination of microscopic and molecular methods . Species-specific identification of eggs by microscopy alone is difficult , because eggs are similar morphologically to other STHs , including hookworms , Trichostrongylus spp . , and the “false hookworm” Ternidens deminutus [5] , [9] , [23]–[25] . In other studies , coproculture of L3 larvae or necropsy to isolate adult worms have been used to identify these parasites to species [25] , [26] . Here , we used molecular methods to detect and sequence Oesophagostomum DNA directly from feces; such methods have proven informative for other similar studies [27] , [28] . In addition , we used phylogenetic comparative methods to ascertain whether primate host traits explain variation in prevalence of Oesophagostomum infection among host species . Our sampling and analyses included nearby human populations to assess whether Oesophagostomum is a public health concern in the region , as well the parasite's local propensity for zoonotic transmission .
Prior to data collection , all protocols were reviewed and approved by the Uganda National Council for Science and Technology and the Uganda Wildlife Authority , as well as by the Institutional Review Board and the Animal Care and Use Committees of McGill University and the University of Wisconsin-Madison . Due to low literacy rates , oral informed consent was obtained from all adult subjects and a parent or guardian of all minor participants by trained local field assistants and documented by witnessed notation on IRB-approved enrollment forms . Kibale National Park ( 0°13′–0°41′N , 30°19′–30°32′E ) is a 795 km2 semi-deciduous protected area in Western Uganda . Primate research has occurred in Kibale for over four decades , focusing on chimpanzees and red colobus monkeys ( Procolobus rufomitratus ) [29] , [30] . As a result , a number of primate groups are habituated to human presence , and many individuals are recognizable based on a combination of physical attributes and collars affixed as part of a larger project on primate health and conservation [31] . Samples from monkeys in the Kanyawara area of Kibale National Park were collected from red-tailed guenons ( Cercopithecus ascanius ) , blue monkeys ( Cercopithecus mitis ) , l'hoest monkeys ( Cercopithecus lhoesti ) , grey-cheeked mangabeys ( Lophocebus albigena ) , olive baboons ( Papio anubis ) , red colobus , and black and white colobus ( Colobus guereza ) ( Figure 1 ) . Chimpanzee samples were collected from Kanyanchu , an area that has a habituated chimpanzee community as a result of tourism ( Figure 1 ) . All samples were collected non-invasively immediately after defecation and placed into sterile tubes . Date , location , species , age and sex category , and social group membership were recorded . Human samples were collected after the receipt of Institutional Review Board-approved informed consent following World Health Organisation protocols . Samples collection occurred in three villages: Ibura , Kanyansohera , and Kasojo , which are less than 5 km from the border of the park ( Figure 1 ) . Individuals between the ages of 2 and 70 were suitable participants of this study . Consenting participants were given instructions on how to collect the sample , which was then retrieved for processing within one day . Samples were subjected to a modified ethyl acetate concentration method , recommended in the approved guidelines of the Clinical and Laboratory Standards Institute for the identification of intestinal-tract parasites [32] , [33] . Concentration by sedimentation was performed in the field using one gram of undiluted feces without fixture in formalin , as formalin is a known inhibitor of the polymerase chain reaction [34] . All materials were sterilized prior to use , and care was exercised throughout the procedure to prevent contamination . Sediments were left uncapped for two hours after completion of the procedure to allow ethyl acetate that may inhibit polymerase chain reaction ( PCR ) to volatilize . Sediments were then suspended in 2 mL RNALater nucleotide stabilization solution ( Sigma-Aldrich , St . Louis , MO , USA ) and frozen at −20°C until shipment to North America . Thin smears from sedimented feces were used for microscopy [35] . All eggs of the genus Oesophagostomum were identified at 10× objective magnification on a Leica DM2500 light microscope . Data were recorded on size , shape , color and internal contents of eggs . Images were captured at 40× objective magnification of all specimens using an Infinity1 CMOS digital microscope camera and Infinity Camera v . 6 . 2 . 0 software ( Lumenera Corporation , Ottawa , ON , Canada ) . Samples were considered negative after the entire sediment sample was scanned and no eggs were found . We note that while identification of Oesophagostomum eggs was based on a rigorous set of characteristics , this genus cannot easily be distinguished from hookworm infection by eggs alone . However , hookworms have not been found in previous surveys of the gastrointestinal parasites of this primate community [19] , [20] , suggesting that eggs identified with morphological characteristics of both Oesophagostomum and hookworm were almost certainly Oesophagostomum . DNA was extracted from 200 µL of sedimented feces using a ZR Fecal DNA MiniPrep Kit ( Zymo Research Corporation , Irvine , CA , USA ) , following manufacturer protocols . External PCR was performed targeting the ribosomal internal transcribed spacer 2 gene using primers NC1 ( 5′-ACGTCTGGTTCAGGGTTGTT-3′ ) and NC2 ( 5′-TTAGTTTCTTTTCCTCCGCT-3′ ) , which generated products that ranged in size from 280 to 400 bp , suggesting that , as expected , the primer set detected a number of parasitic helminths present in the samples [27] , [36] . Subsequently , an internal , semi-nested PCR generating amplicons of predicted size 260 bp was performed using primer NC2 and newly designed Oesophagostomum-specific primer , OesophITS2-21 ( 5′-TGTRACACTGTTTGTCGAAC-3′ ) . Primer OesophITS2-21 was generated by aligning publicly available sequences of the Oesophagostomum internal transcribed spacer 2 gene [26] , [36]–[39] , and GenBank accession numbers HQ283349 , HQ844232] . In total , eight species of Oesophagostomum were represented in the alignment . Other species of varying relatedness , including other members of the taxa Chabertiidae ( Chabertia ovina , Accession No . JF680981; Ternidens deminutus , Accession No . HM067975 ) , Strongylidae ( Strongylus vulgaris [40] ) , and Strongylida ( Necator americanus [36] , and Ancylostoma duodenale [41] ) were also included . Priming regions were selected to be identical among all species of Oesophagostomum but divergent from the other genera . Primer ITS2-21 was highly specific as confirmed by sequencing , since all PCR products matched Oesophagostomum despite the fact that a number of other parasites ( including Strongyloides , Necator and Trichuris ) , were identified in the same samples during microscopic examination . External PCR was performed in 25 µL volumes using the FailSafe System ( Epicentre Biotenchnologies , Madison , WI , USA ) with reactions containing 1× FailSafe PCR PreMix with Buffer C , 1 Unit of FailSafe Enyme Mix , 2 . 5 picomoles of each primer ( NC1 and NC2 ) , and 1 µL of template . Reactions were cycled in a Bio-Rad CFX96 platform ( Bio-Rad Laboratories , Hercules , CA , USA ) with the following temperature profile: 94°C for 1 min; 45 cycles of 94°C for 15 sec , 50°C for 30 sec , 72°C for 90 sec; and a final extension at 72°C for 10 min . Internal PCR was performed in 25 µL volumes using the DyNAzyme DNA Polymerase Kit ( Thermo Scientific , Asheville , NC , USA ) with reactions containing 0 . 5 Units of DyNAzyme I DNA Polymerase , 1× Buffer containing 1 . 5 mM MgCl2 , 2 . 5 picomoles of each primer ( OesophITS2-21 and NC2 ) , and 1 µL of template . Reactions were cycled with the following temperature profile: 95°C for 1 min; 45 cycles of 95°C for 15 sec , 55°C for 30 sec , 70°C for 90 sec; and a final extension at 70°C for 5 min . Amplicons were electrophoresed on 1% agarose gels stained with ethidium bromide , and purified from gels using the Zymoclean Gel DNA Recovery Kit ( Zymo Research Corporation , Irvine , CA , USA ) according to the manufacturer's instructions . Products were Sanger sequenced in both directions using primers OesophITS2-21 and NC2 on ABI 3730xl DNA Analyzers ( Applied Biosystems , Grand Island , NY , USA ) at the University of Wisconsin-Madison Biotechnology Center DNA Sequencing Facility . Sequences were hand-edited and assembled using Sequencher v4 . 9 ( Gene Codes Corporation , Ann Arbor , MI , USA ) and all ambiguous bases were resolved by repeat PCR and re-sequencing , as described above . All new sequences were deposited in GenBank , under Accession Numbers KF250585 - KF250660 . Sequences were aligned using the computer program ClustalX [42] with minor manual adjustment . Published reference sequences were included to identify putative species ( AF136575 , Y11733 , AF136576 ) and as outgroups ( HQ844232 , Y11738 , Y11735 , Y10790 , AJ006149 ) , and were trimmed to the length of the newly generated sequences using Mesquite v . 2 . 75 [43] . Trimmed sequences yielded the same tree topology as did untrimmed sequences ( by neighbor-joining method; results not shown ) , suggesting that the amplified region was sufficient for taxonomic discrimination . Phylogenetic trees were reconstructed using maximum likelihood in MEGA v . 5 . 05 [44] and the Hasegawa-Kishino-Yano substitution model [45] . Phylogenetic support was assessed using 1 , 000 bootstrap replicates . To estimate Oesophagostomum genetic diversity , percent nucleotide-level sequence identity among sequences was calculated as the uncorrected pairwise proportion of nucleotide differences ( p-distance ) in MEGA v5 . 05 [44] . Diagnostic performance of microscopy versus PCR was estimated by calculating sensitivity ( i . e . , true positive rate ) and specificity ( i . e . , true negative rate ) using MedCalc v . 12 . 5 . 0 ( MedCalc Software , Ostend , Belgium ) . Prevalence of infection was calculated as the number of samples found to be positive for Oesophagostomum divided by the total number of samples collected , with 95% confidence intervals calculated using the modified Wald method [46] . To determine whether prevalence differed among primate host species , a chi-square test was conducted in Quantitative Parasitology v3 . 0 [47] . To explore variation in prevalence among hosts while controlling for their phylogenetic non-independence , a phylogenetic least squared regression ( PGLS ) was conducted in R [48] using the ape [49] and caper [50] libraries . Prevalence of Oesophagostomum was included as the dependent variable , and various primate life history traits were independent variables: terrestriality ( predominantly terrestrial versus predominantly arboreal ) , maximum home range [51]–[56] , maximum group size [51] , [53] , [55] , [57]–[60] , percentage time spent in polyspecific associations [61] , [62] , average female body mass , and average daily travel distance ( the latter was log transformed since the relationship was close to exponential ) [55] , [56] , [62]–[64] . Humans were omitted from the PGLS analysis because many of these traits vary widely among human populations , making accurate estimations problematic . To determine the degree to which each Oesophagostomum lineage ( i . e . , taxonomic unit ) identified by DNA sequencing was host restricted , we calculated the phylogenetic dispersion of infected hosts using the net relatedness index ( NRI ) in R [48] using the ape [49] and picante libraries . Mean pairwise distance ( MPD ) was weighted by the ratio of occurrence of each Oesophagostomum within each lineage , and compared to null expectation in 1000 randomly assembled communities . Results are reported as standard effects sizes , with values close to 1 indicating phylogenetic evenness ( i . e . , Oesophagostomum lineages infect a greater diversity of hosts than would be expected by chance ) , while values <0 . 05 indicate phlylogenetic clustering ( i . e . , Oesophagostomum lineages are host-specific ) .
A total of 318 fecal samples from primates , including humans , were collected ( Table 1 ) . Of these , 112 were positive for Oesophagostomum by microscopy , for a community-wide prevalence of infection of 35 . 2% ( Table 1 ) . All eggs identified by microscopy were similar in internal and external morphology in samples from all primate species ( Figure 2 ) . Eggs were 65–80 by 35–50 µm in size , which is consistent with previous results from this community [19] , [20] ( Figure 2 ) . PCR generated single , clear amplicons of expected size ( 260 bp ) in 222 samples , indicating positive detection of Oesophagostomum DNA , for an overall prevalence of 69 . 8% . No amplicons were present in remaining samples . Resulting DNA sequences overlapped 100% with published sequences and contained no insertions or deletions , making alignment unequivocal . When PCR results were compared to microscopy , the overall sensitivity of PCR was 100% ( 95% CI 96 . 8%–100 . 0% ) , but specificity was only 47 . 5% ( 95% CI 40 . 5%–54 . 7% ) . Thus , PCR did not classify any microscopy-positive samples as negative but identified 109 microscopy-negative samples as positive . Prevalence of Oesophagostomum infection ( as determined by both microscopy and PCR ) varied significantly among host species ( microscopy: chi-square = 54 . 31 , df = 8 , P<0 . 0001; PCR: chi-square = 112 . 2 , df = 8 , P<0 . 0001 ) . Both microscopy and PCR identified humans as having the lowest prevalence of infection ( 8 . 3% and 25 . 0% , respectively ) , followed by red colobus ( 17 . 2% and 40 . 6% , respectively ) . Chimpanzees , l'hoest monkeys , and olive baboons had the highest prevalence by both methods , with 100% prevalence by PCR in all three species ( although sample sizes were low in some cases; Table 1 ) . PGLS indicated that terrestriality , maximum home range , maximum group size , percent time spent in polyspecific associations , and average female body mass were not significant univariate predictors of Oesophagostomum prevalence ( all P>0 . 05 from PGLS with lambda = ML; Table 2 ) . However , log daily travel explained nearly 55% of the variation in prevalence among host species ( P<0 . 05 , R2 = 0 . 546 , from PGLS with the ML estimate of lambda = 0 ) . In a multivariate model , both group size and log daily travel were significant predictors of prevalence , with group size showing a negative relationship and log daily travel a positive relationship ( Table 2 ) . This two-predictor model including group size and daily travel explained over 75% of the variation in Oesophagostomum prevalence among species ( model P<0 . 01 , R2 = 0 . 7701; Table 3 ) . From 222 positive samples , 76 were randomly selected for sequencing to represent as even a number of positive samples per host species as possible . All 76 sequences most closely matched published Oesophagostomum ITS-2 DNA sequences using the BLASTn tool on the National Centre for Biotechnology Information website . Phylogenetic analysis resolved these sequences into three clades ( Figure 3 ) . Clade 1 contained all 12 sequences from olive baboons , one sequence from l'hoest monkeys , one sequence from grey-cheeked mangabeys , three sequences from red colobus and one sequence from red-tailed guenons . These sequences were identical to published reference sequences for O . bifurcum [26] , [36] . Five additional sequences from l'hoest monkeys sorted into clade 1 and were 97 . 1% similar to this same O . bifurcum reference sequence . Clade 2 contained all eight sequences from chimpanzees , five sequences from blue monkeys , two sequences from black and white colobus , two sequences from grey-cheeked mangabeys , three sequences from red colobus , and twelve sequences from red-tailed guenons . All sequences in clade 2 were identical to an O . stephanostomum reference sequence [26] . Clade 3 was composed of two nearly identical branches ( 99 . 4% identity ) that contained all six sequences from humans , as well as sequences from three blue monkeys , three black and white colobus , five grey-cheeked mangabeys , two red colobus , and two red-tailed guenons . These sequences were 92 . 4–93 . 0% and 93 . 0–93 . 6% similar to O . bifurcum , and O . stephanostomum , respectively , but were not identical to any published reference sequence . Host species were phylogenetically clustered within O . bifurcum clade 1 ( NRI = −1 . 76 , P<0 . 05 ) . Clade 2 ( O . stephanostomum ) did not vary significantly from the null expectation of no host clustering , NRI = 0 . 86 , P = 0 . 75 ) . Clade 3 was marginally phylogenetically over-dispersed with respect to distribution of host species ( NRI = 1 . 24 , P = 0 . 04 ) .
Here we evaluate the prevalence of Oesophagostomum infection in wild primates and humans in Western Uganda using both microscopy and PCR . Our results clearly show that prevalence varied significantly among host species . Humans had the lowest prevalence of infection likely because of avoidance behaviors such as sanitation practices [65] , [66] and because of the common use of antihelminthics in the region . Red colobus and black and white colobus also had comparatively low prevalence of infection , as found in previous studies [16] , [20] , [67] . This observation may reflect colobine gastrointestinal physiology , which is characterized by folivory and foregut fermentation [68] , and the associated regular ingestion of plant secondary compounds that may suppress infection by pathogenic organisms [69] . Conversely , the high prevalence of infection in chimpanzees , olive baboons , and l'hoest monkeys may reflect reduced physiological barriers to infection or increased susceptibility . To explain this interspecific variation in prevalence , we examined correlations between life history variables and prevalence among host species . We found that two variables , daily travel distance and group size , explain over 75% of the variance in Oesophagostomum prevalence among host species . Surprisingly , body mass , the strongest predictor of helminth species richness elsewhere , was not significant here [67] . Previous studies have concluded that group-living animals with small home ranges are likely to suffer high intensities of infection due to frequent environmental re-exposure [70]–[72] . Our results indicate the opposite in the case of Oesophagostomum: smaller primate groups with large daily travel distances had higher prevalence . Animals with larger day ranges may encounter greater habitat variation [73] , which may increase exposure to Oesophagostomum from environmental sources . In addition , previous research has implicated terrestriality as an important factor affecting the prevalence of trematode parasites in primates [74] . In our study , the three host species with highest Oesophagostomum prevalence ( chimpanzees , olive baboons and l'hoest monkeys ) were also the only three predominantly terrestrial species . Although this trend was not statistically significant , it is possible that terrestrial primates contact soil more frequently , and thus the infective stages of STHs . Although group size was not a significant predictor of prevalence in univariate analyses , our multivariate analysis found smaller groups with large daily travel distances to be at greatest risk of infection . This finding contrasts with previous studies showing that increased intragroup contact increases exposure [71] , [75] . In Kibale , positive associations between group size and parasite richness have been documented for protozoan parasites in mangabeys [76] . However parasite richness is not necessarily associated with prevalence . Small primate groups might maintain high intra-group infection rates for certain parasites if transmission within the group is frequent , thus maintaining high prevalence ( as seen here ) without correspondingly high parasite richness . Our study detected substantial cryptic phylogenetic diversity in Oesophagostomum infecting Ugandan primates . Currently , the principal human Oesophagostomum species is considered to be O . bifurcum [5] , while other great apes harbor O . stephanostomum [6] , [12] , [18] , [77] . Recently , however , chimpanzees inhabiting a northern sector of Kibale were identified as positive for O . bifurcum , making this the first discovery of O . bifurcum in non-human apes . The same study identified chimpanzees also infected or co-infected with O . stephanostomum [22] . In our phylogenetic analysis , we identified both O . bifurcum and O . stephanostomum in the Kibale primate community . However , we found only O . stephanostomum in chimpanzees , although the possibility of undetected O . bifurcum infections cannot be ruled out . In addition , we identified a third Oesophagostomum lineage that did not cluster with any published sequence and thus may represent a previously uncharacterized taxon . It is possible that this new taxon has remained undetected in previous molecular investigations . We examined the OB primer that has been used previously to identify O . bifurcum [36] and conclude that it would probably not amplify our newly identified taxon due to mismatched bases at both the 5′ and 3′ ends of the primer . It is therefore possible that the new taxon we identified exists elsewhere ( e . g . in Togo and Ghana ) but has been not been detected or differentiated from other members of the genus . However , we caution that these inferences are based on a short region of a single gene , and that sequencing additional genes as well as morphological characterization of L3 larvae and adults will be necessary to confirm these findings . Nonetheless , our results suggest a heretofore unappreciated degree of hidden genetic diversity within this well-described genus of parasites that are known to infect humans . Interestingly , all Oesophagostomum sequences recovered from humans clustered with the previously undescribed third taxon , and not with published O . bifurcum sequence from humans elsewhere in Africa ( clade 1 ) [36] . In Ghana , geographic separation between humans and non-human primates infected with Oesophagostomum , despite apparently conducive environments for zoonotic transmission , motivated efforts to determine the host range of the parasite using molecular methods [28] . Genome-wide analyses ( amplified fragment length polymorphism , random amplification of polymorphic DNA ) suggested that O . bifurcum clusters into distinct groups by host species , thus suggesting that zoonotic transmission is uncommon [4] , [17] . By contrast , in our study area , no such geographic separation exists between humans and non-human primates . In this setting , we found that both humans and non-human primates were infected with the novel Oesophagostomum clade 3 , which is phylogenetically over-dispersed compared to the other Oesophagostomum clades . While our conclusions await verification from more detailed examination of the Oesophagostomum genome , our results nevertheless suggest that this novel clade may be broadly transmissible among species of distantly related primate hosts , including humans . The Kibale ecosystem is known for its high degree of spatio-temporal overlap between humans and non-human primates and its ensuing high rates of transmission of diverse pathogens across primate species [78]–[81] . Our results provide further evidence for cross-species pathogen transmission between wild primates and humans in this region . Our paired analyses applying both microscopy and PCR to the same samples indicate that traditional methods based on microscopy may significantly underestimate prevalence . Concentration methods followed by microscopic visualisation of eggs in thin smears are considered definitive diagnostic methods for soil-transmitted helminth ( STH ) infections [82] , [83] . Previous studies that have used fecal sedimentation and microscopy have reported Oesophagostomum infection prevalence estimates between approximately 3% and 10% in wild Ugandan primates [19] , [20] . These values are considerably lower than what we report here using molecular methods; however , our results parallel other studies that have estimated prevalence using molecular methods [16] , [22] . Not surprisingly , we find that PCR is more sensitive than microscopy , perhaps because it can detect Oesophagostomum infection even when eggs are not present . For example , tissues or secretions shed by adult worms into the intestinal lumen would be detected by PCR , as would eggs that have hatched into L1 larva prior to fixation during the sedimentation procedure . To our knowledge , ours is the first study in several decades to report human Oesophagostomum infection in Uganda , a country that is over 3 , 000 km from known foci of infection in West Africa [5] . Given that the prevalence of Oesophagostomum was 25% in our sample of people , we suspect that this parasitic infection occurs more commonly across Sub-Saharan Africa than previously thought and may be causing infections that are untreated or misdiagnosed . Our finding of a previously genetically uncharacterized lineage of Oesophagostomum that may be transmitted among primate species underscores that the diversity ( genetic and otherwise ) of this parasite genus may be under-sampled in Africa . Further ecological studies of Oesophagostomum in Uganda and elsewhere are needed to quantify the degree of enzootic versus zoonotic transmission . Regardless of the outcome of such research , our results suggest that Oesophagsotomum should be considered a pathogen of concern beyond its accepted foci of infection in Togo and Ghana , and perhaps across all of equatorial Africa . | Nodule worms infect the gastrointestinal tracts of a number of mammalian species , including humans and other primates . This study sought to identify the species of nodule worms causing infections within and around an East African national park in Uganda where monkeys and apes co-occur and overlap with people . Some primates , particularly those traversing large distances in small groups , were most susceptible to nodule worm infection . Additionally , molecular analyses identified three separate groups of nodule worm that could not be distinguished based on microscopic examination of their eggs . One of these groups was found in humans as well as other primates and had not previously been genetically characterized . These results suggest that certain types of nodule worm may be restricted to particular hosts , while others may be transmitted among primates , including humans . Nodule worms are currently thought to be a human health concern only in some West African countries . This research suggests that nodule worms have a broader geographic impact in humans than previously appreciated . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"social",
"and",
"behavioral",
"sciences",
"medicine",
"veterinary",
"science",
"biology"
]
| 2014 | Nodule Worm Infection in Humans and Wild Primates in Uganda: Cryptic Species in a Newly Identified Region of Human Transmission |
Organismal development and many cell biological processes are organized in a modular fashion , where regulatory molecules form groups with many interactions within a group and few interactions between groups . Thus , the activity of elements within a module depends little on elements outside of it . Modularity facilitates the production of heritable variation and of evolutionary innovations . There is no consensus on how modularity might evolve , especially for modules in development . We show that modularity can increase in gene regulatory networks as a byproduct of specialization in gene activity . Such specialization occurs after gene regulatory networks are selected to produce new gene activity patterns that appear in a specific body structure or under a specific environmental condition . Modules that arise after specialization in gene activity comprise genes that show concerted changes in gene activities . This and other observations suggest that modularity evolves because it decreases interference between different groups of genes . Our work can explain the appearance and maintenance of modularity through a mechanism that is not contingent on environmental change . We also show how modularity can facilitate co-option , the utilization of existing gene activity to build new gene activity patterns , a frequent feature of evolutionary innovations .
For our study we consider a network of genes . Each gene's activity state is regulated by other genes in the network . The genotype of an individual is defined as the set of the interactions among its genes . We represent this set of interactions as a matrix . Non-zero elements in indicate activation ( ) or repression ( ) of gene exerted by gene . The state of the network at time is described by a vector . A certain gene at time can be either active ( ) or inactive ( ) . We model the change in the activity of the genes in the network according to the difference equation ( 1 ) where equals if , and it equals in all other cases . Despite its simplicity , variants of this model have been successfully used to study how robustness can evolve in gene regulatory networks [22]–[24] , how robustness can aid in evolutionary innovation [25] , [26] , and how recombination can produce negative epistasis [27] . Moreover , similar models have been successfully used to predict the dynamics of developmental processes in plants and animals [28] , [29] . For our purpose , we consider that a phenotypic trait is defined by an attractor , a stable gene activity pattern resulting from the dynamics of a gene regulatory network . Attractors are often associated with developmental end-states and ‘outputs’ of developmental mechanisms [22] , [30]–[32] . In order to study the evolution of modularity in gene regulatory networks , we implemented evolutionary simulations that consisted of iterative rounds of mutation and selection in populations of networks . In these simulations , we compared a set of reference gene activity patterns to actual network attractors , so that networks with attractors that were similar to the selected activity patterns had higher fitness than others ( see Methods ) . To quantify the modularity of networks in our model , we used an algorithm [33] that identifies modules as non-overlapping densely connected groups of nodes with sparser connections between groups ( see Methods ) . Thus , if genes in individual modules interact with many genes outside their module , the autonomy of the modules decreases , which would be reflected in a lowered modularity score .
To find out whether specialization can increase modularity , we studied 200 independent evolving populations of gene regulatory networks ( eq . 1 ) . Each of these populations was started with identical networks , and was subject to 500 generation cycles of mutations and selection towards attainment of a fixed-point attractor I ( see Methods for details ) . The number of generations was chosen to ensure that networks that stably attain I can arise in the population . After gene activity pattern I had evolved , we allowed the population to evolve for 1500 more generations , but selecting for attainment of gene activity pattern I and a new pattern II during this time . Under this selection regime , the fittest networks were those capable of stably attaining I and II from different initial conditions that may occur in different parts of a multicellular organism . In other words , selection maintained the ability to attain I while at the same time favoring acquisition of II . Pattern II was chosen such that half of the network genes had identical ( shared ) expression states in I and II , and the other half differed in their activity state ( Figure 1A ) . We chose such activity patterns because we hypothesized that interactions between genes with shared activity states and the rest of the genes would obstruct either i ) the constant activity state of the former , or ii ) the capacity of the latter to acquire different activity states independently of genes with constant activity states . If so , interactions between the different sets of genes may be selected against , thus resulting in two sets of genes with only sparse connections between them . In most of the 200 evolving populations , modularity increased after evolving towards the attainment of both I and II . We observe this increase both in the networks with the highest fitness in the population ( Figure 1B; Wilcoxon signed-rank test; ; ) , and when averaged over all networks in a population ( Figure 1C; Wilcoxon signed-rank test; ; ) . Figures 1D , E show an example of how modularity increases after selection for attainment of activity patterns I and II . Modularity does not increase when selection for II is absent , nor when networks evolve in the absence of selection ( Figure S2 ) . The increase in modularity is not transient because it is maintained around the same level , at least for 10 , 000 additional generations , when selecting for both I and II ( Figure 2 ) . We next verified that our results were insensitive to changes in model assumptions and parameters . We first decreased the mutation rate , and even though the time required to evolve activity patterns I and II then increases , modularity still increases significantly ( ; Figure S3A ) . Modularity increases as well when is increased ( ; Figure S3B ) . We next asked whether our observations were sensitive to the assumption that individual gene activity patterns contribute to fitness additively . Changing this assumption to multiplicative fitness contributions still leads to a significant increase in modularity ( Figure S4 ) . In addition , the increase in modularity also occurs for networks containing more genes ( ; Figure S5 ) , suggesting that such behavior does not depend on the number of genes in a network . In a next analysis , we asked whether the increase in modularity depends on the identity of gene activity states I and II . We found that it does not , as long as some genes have the same activity state in the two patterns . For example , modularity also increases when the activity patterns differ in the activity of either three or seven genes ( Figure S6A , B ) . Moreover , modularity increases when both the first and the second gene activity patterns are randomly chosen , except that pairs with fewer than two different activity states are discarded ( Figure S6C , based on 100 populations with different pairs of activity patterns ) . In contrast , modularity does not increase when all genes in the activity states I and II differ in their expression ( Figure S6D ) . This result is not due to a lack of adaptation , since networks able to attain both activity patterns arise in all evolving populations . Taken together , these observations show that modularity does not only increase for specific gene activity patterns , but that it is a generic evolutionary response . Moreover , the distinction between two sets of genes , those with identical and those with different activity in both expression patterns , is essential for the evolution of modularity . That modularity increases only in this case suggests that modules arise as a means of diminishing the effects of genes with unchanging activity on genes with changing expression in I and II , and vice versa . If so , modules should correspond to sets of genes that are required to switch their activity in a concerted manner . The following section shows that this is the case . Having established that the evolution of modularity requires genes with both shared and different activity states , we next asked whether the partitioning of modules is congruent with these two sets of genes . In other words , does one module tend to involve the genes with shared activity states , whereas another involves genes with different activity states in I and II ? We evolved 300 network populations , first towards activity pattern I and later towards both I and II , depicted in Figure 1A . Throughout evolution , we determined for one of the best adapted networks in each population: i ) the frequency at which two genes with activity states shared in I and II occur within the same module , ii ) the frequency at which two genes with different activity states in I and II occur within the same module , iii ) the frequency with which a specific gene with a shared activity state and a gene with a non-shared activity state are in the same module ( Figure 3A ) . As selection for I and II occurs , and increase , while decreases ( Figure 3B ) . This observation tells us that genes with activity states that change concertedly throughout all the selected activity patterns – be they shared or not – will tend to be included in the same module , and kept apart from other genes . This is exemplified in Figure 1D , E , which compares one of the optimal networks after selection for I with one of the optimal networks after selection for both I and II . The latter is partitioned into modules in which genes with shared and distinct activity states in I and II lie apart . Thus , the structure of modules reflects the manner in which selection has molded the traits , as has been previously suggested [2] . We also tested whether modularity arises only where selection favors the attainment of two gene activity patterns , or whether it increases further with even more gene activity patterns . To this end , we analyzed 100 evolving populations in which selection first favored a gene activity pattern I ( 500 generations ) , then an additional pattern II ( I+II , next 1 , 500 generations ) , and then a third pattern III ( I+II+III , last 3 , 000 generations ) . The patterns share the activity of some genes and differ in others . As selection for the third pattern begins , more and smaller groups of genes arise whose activity changes in a concerted manner ( Figure 4A ) . Interactions between different such groups would obstruct evolutionary adaptation . Such interactions should thus be selected against , resulting in a further increase in modularity . Our observations confirm this hypothesis . After selection for patterns I and II , we observed a significant first increase in modularity ( Wilcoxon signed-rank test; ; ) . Modularity increased further after selection for pattern III ( Figure 4B; Wilcoxon signed-rank test; ; ) . In addition , we observed an increased number of modules in networks with high fitness after selection for patterns I and II . Moreover , this number increases further after selection for patterns I , II and III ( Figure S8B ) . This result suggests that the increase in modularity after selection for the three patterns occurs because of the appearance of new modules , and is not a mere consequence of the consolidation and refinement of previously evolved modules . We also analyzed how the probability of two genes being part of the same module changes across evolution . We found that the frequency of two genes occurring in the same module in the fittest networks of each evolving population changes according to whether those genes change their activity concertedly across the selected patterns ( Figure S8C ) . For example , as we depict in Figures 4 and S8A , the activity of genes 5 and 6 changes concertedly across all activity patterns: if in one pattern gene 5 is active , then gene 6 is inactive in that same pattern , and vice versa . The frequency with which those genes lie in the same module increases across evolution . In contrast , the activity of genes 0 and 6 changes concertedly when selecting for patterns I and II , but not when also selecting for activity pattern III . Thus , the probability of those genes occurring in the same module increases prior to selection for pattern III . After selection for pattern III starts , the probability that genes 0 and 6 lie in the same module decreases abruptly ( Figure S8C ) . These results show that the modules that arise after selection for the third pattern also tend to coincide with sets of genes whose activity states change concertedly throughout the selected patterns . Computational cost did not allow exploration of further increases in modularity via selection of additional gene activity patterns . However , our observations already suggest that modularity will increase as long as there is an increase in the number of gene groups for which concerted activity changes are favored . A question recurring in the literature is how modularity may increase evolvability by facilitating co-option , the combination of previously evolved modules to perform new functions [19]–[21] , [34]–[36] . We addressed how the previous evolution of modules in gene regulatory networks biases future evolutionary potential by asking whether gene networks acquire new gene activity patterns faster if these patterns use gene activity states associated with previously evolved modules . Specifically , we selected networks for their ability to stably attain three gene activity patterns I , II and III ( Figure 5A ) . We chose the specific combination of patterns in Figure 5A because: i ) it promotes the evolution of a module including genes 0–4 and another module including genes 5–9 , as shown above , and , ii ) it allows the inclusion of an additional activity pattern ( IV ) that is composed entirely of activity states associated with previously evolved modules ( Figure 5A , B ) . After 3 , 000 generations , we subjected networks in 100 evolving populations to selection favoring such an additional gene activity pattern IV ( Figure 5B ) . Importantly , this pattern shares the activity states of genes 0–4 with III , and the activity state of genes 5–9 with II . Thus , gene activity pattern IV may evolve by combining previously evolved modules in a new manner . In addition , we repeated this approach in 100 “control” populations where the fourth favored gene activity pattern was randomly chosen with equal probability for genes being active and inactive . Notice that we do not expect that selection for activity pattern IV increases modularity , because the inclusion of this pattern does not cause an increase in the number of gene groups with concerted activity changes . Rather , we hypothesize that modularity facilitates the evolutionary acquisition of such an activity pattern , as compared to other activity patterns . We found that networks with high fitness arise much more rapidly when IV is the new gene activity pattern . This indicates that pattern IV is much easier to attain than random gene activity patterns in populations of networks that have previously been selected for their ability to attain I , II and III ( Figure 5C ) . The same trend occurs when not just the networks with highest fitness are considered , but also when we analyze mean population fitness ( Figure S7 ) . We note that in our analysis selection favors the attainment of IV to the same extent as the attainment of any one random gene activity pattern in the control populations . This means that our observations are not simply caused by a greater increase in fitness conveyed by IV . The fitness increase rather depends on how easily the new gene activity patterns can be constructed: it is easier to evolve gene activity patterns that combine activity states of previously evolved modules .
In sum , we showed here that modularity arises in gene networks when they acquire the ability to attain new activity patterns that share the activity state of some genes with old patterns . Our observations indicate that selection to attain the new activity patterns can cause modularity to arise in gene regulatory networks when pleiotropic effects obstruct adaptation [2] , [8] , [11] . Such pleiotropic effects are caused by interactions between ( i ) genes whose activity is shared between different patterns , and ( ii ) genes whose activity is specific to one pattern: If changes in the latter affect the former , evolutionary acquisition of the new pattern is hindered . Thus , the scenario we propose favors networks with few interactions between genes with an unchanging activity state and genes that adopt new regulatory functions . In this way , genes that have correlated activity states come to lie in the same network module ( Figures 3 and S8C ) . Our results suggest that modularity increases as long as selection favors new activity patterns involving more and smaller groups of genes whose activity changes in a concerted manner ( Figures 4 and S8 ) . Empirical falsification ( or validation ) of the mechanism that we propose ideally requires comparative analyses of the structure of gene regulatory networks in several related species . Such information might not be available soon . However , existing information from various sources suggests that the mechanism we propose could be important . Specifically , the evolutionary acquisition of new gene activity states by regulatory networks is ubiquitous in evolution , and nowhere more than in the evolution of development . It occurs wherever new cell types , organs , or body structures , arise from previously undifferentiated ones . Many examples in the literature suggest that some genes exhibit specialized activity in different parts of an organism , whereas others present shared activity patterns . Indeed , gene functions may be inferred via correlated gene expression patterns in conventional or high-throughput expression analyses [37]–[39] . For example , the activity of the same genes patterns both vegetative and floral meristems in the plant Arabidopsis thaliana . Floral identity genes are active exclusively in floral meristems , so that the floral structure is determined by both the floral identity genes and the shared patterning genes [40]–[42] . In the sea urchin Strongylocentrus purpuratus , some differentiation genes are active in the micromer lineage that produces the euechinoid exclusive embryonic skeleton and also in the independently derived juvenile skeletogenic centers that produce the adult skeleton [43] . Some other genes of the gene network that specifies the skeletogenic micromere lineage are active in those cells but not in the juvenile skeletogenic centers . Examples include genes involved in induction of neighbouring cells or in triggering the initial stages of micromere specification [43] , [44] . Another example involves the cellular level . Mammalian brown fat cells share some traits and gene activity patterns with white fat cells , and others with muscle cells [45] . More generally , evolutionarily derived cell types usually perform just a fraction of the functions that ancestral cell types performed [46] , a trend that will lead to similar activity states for some genes and different states for others in sister cell types . In a similar vein , evolutionary specialization of initially homogeneous metameric units is likely to occur mainly by modifications ( such as changes in the transcriptional circuitry ) that result in metamers with different activity states of some genes but not of others; otherwise , differentiated metameric units would be hardly recognizable as such . For example , in D . melanogaster , limbs are positioned and patterned by mechanisms that are reiterated along the body , however limb identity relies on segment-specific mechanisms [47] . Moreover , in heteronomous arthropods , in which the morphology of segments along an individual is very distinct , processes underlying segmentation and limb differentiation interact less than in homonomous arthropods , in which the segments along a body are very similar [47] . Segment formation is performed throughout the organism ( shared ) , and , in heteronomous taxa , limb identity determination is specialized according to the place where a limb develops . Thus , when there is specialization in limb identity , the two processes are more independent , in contrast to taxa that lack this specialization . Co-option , the recruitment of previously evolved modules to perform new functions , is a common feature of evolutionary innovations [20] , [21] , [34] , [36] , [43] . A case in point regards the gene network regulating pharyngeal dentition in fish , which is co-opted to also generate oral dentition [36] . Another example is the gene network that patterns the insect wing blade . It is co-opted to determine the localization of eyespots in butterfly wings [34] . Our work shows that a modular network may readily generate new gene activity patterns that make use of gene activity states of previously evolved modules . The existence of such structured , or “facilitated” variation has been known for a long time [48]–[51] . Our work provides a candidate mechanism to create such variation , namely via network modularity that results from specialization in gene activity . Our observations could thus help explain the repetitive co-option of several modules , such as that responsible for proximal-distal polarity in lateral appendages and body outgrowths [20] , [21] , or the achaete and scute module that operates in a wide range of developmental processes in animals [19] . An alternative hypothesis for the evolution of modularity is the ‘modularly-varying goals’ scenario [10] . This scenario requires that populations are exposed to evolutionary goals that fluctuate over time , so that modularity can arise and be maintained . In contrast , our scenario requires specialization of gene activity , that is , new gene activity patterns must be attained while old activity patterns are preserved . Relatedly , the modularly-varying goals scenario requires genetic changes for evolutionary adaptation after an evolutionary goal changes . In contrast , our mechanism requires one genotype to produce different activity patterns under different conditions , conditions that may occur in different parts of a multicellular organism . In other words , in our scenario , modularity arises to avoid obstruction to attain different selected patterns within the same genotype . Our scenario may thus be more appropriate for traits where environmental demands are not constantly fluctuating , such as in the development of many morphological traits in plants and animals . Thus far , we motivated our approach with the development of multicellular organisms . However , the approach could also explain modularity in unicellular organisms . For example , the metabolic networks of bacteria living in changing environments tend to be more modular than those of bacteria living in stable environments [12] . Similar patterns may exist for gene regulatory networks . If so , the modularly varying goals scenario is not their only possible explanation . Unicellular organisms respond to changing environments by tuning their gene activity pattern . In other words , they usually have adaptively plastic phenotypes . For example , different sets of genes are activated or repressed when yeast cells are exposed to different environments [52]–[54] . Evolving the ability to switch gene expression according to the environment requires producing several alternative activity patterns , as we propose here . Importantly , some yeast genes change their expression concertedly in several environments , whereas others have responses that are specific to any one environment [52]–[54] . This observation suggests that the activity of some genes is shared across alternative activity patterns while the activity of other genes is particular to certain environments , as our model demands . In sum , because organisms in changing environments are required to produce different gene activity patterns according to the environment , our scenario can explain the evolution of modularity both in fluctuating and non-fluctuating environments . A question that remains unanswered is whether our model applies to genotype-phenotype maps different from those of gene regulatory networks . A prominent example is metabolic networks , whose phenotypes are patterns of metabolic fluxes through network reactions . Our framework may apply to some instances of modularity in metabolic systems , as the following example illustrates . The main requirement of our model is an increase in the number of functions that a network must perform ( i . e . in the number of selected gene activity patterns ) . The appearance of new functions in a metabolic network usually involves the production of new metabolites . Hintze and Adami [55] performed evolutionary simulations of an artificial metabolism in which the fittest metabolic networks were able to produce an increasingly diverse spectrum of metabolites . This selection regime resulted in increased modularity of metabolic networks , an observation consistent with the mechanism that we propose for gene regulatory networks . Our work aimed at conceptual clarity by using only few essential assumptions in explaining the evolution of modularity . We therefore neglected many processes that doubtlessly play a major role in the evolution of regulatory gene networks . For example , we did not consider mutations changing the number of genes in a network , even though processes such as gene loss or duplication may be frequently involved in the appearance of new gene activity patterns . Similarly , the appearance of new body structures or cell types requires interactions among cells , tissues and organs . Such interactions ensure the proper placement of cells with the combination of general and specialized gene activity that is characteristic of specialization . The incorporation of these and other processes in future work will deepen our understanding of the evolution of modularity , and thus of evolvability .
We here identify modularity using one [33] , [56] of several algorithms aimed at identifying structural modules , densely connected groups of nodes with sparser connections between groups . The measure of modularity in this algorithm is a score that compares the abundance of intra-module connections between a given network to that of random networks with the same degree distribution [57] . is defined as: ( 2 ) where denotes one of the prospective modules in a network , stands for the total number of edges in the network , represents the number of edges within module , and is the sum of the number of connections that each node in module has [33] , [56]–[58] . The algorithm we use [33] identifies a partitioning of networks into modules that maximizes . We use this algorithm because of its computational efficiency and accuracy [33] , [56] . We also explored different algorithms [57] , [58] and found that our results hold regardless of these choices . Typical values of partitions that maximize intra-module connections in random networks vary depending on the number of nodes , edges and connectivity distribution [59] . For example , the maximum value of a network varies as a function of the total number of edges in it [60] . Hence , a fair comparison of modularity in different networks requires first addressing how atypical is in the best partition of each network when compared with random networks with the same attributes . Following [10] we use for normalization the equation: ( 3 ) where is the modularity returned by the Newman algorithm [33] , [56] for a certain network , stands for the average value of 1 , 000 random networks with the same number of genes and edges and the same degree distribution as the original network . values for these random networks are also calculated using the Newman algorithm . is the maximal value in these 1 , 000 random networks . The normalized modularity tells us how modular a network is in comparison to random networks with the same attributes . Non-normalized and normalized values render equivalent results in our analysis ( Figure S1 ) . Therefore , we restrict ourselves to report results for normalized modularity . The fitness function we use compares a set of reference gene activity patterns to actual network attractors . Our fitness measure also incorporates the likelihood that an attractor is attained in the face of perturbations . In doing so , it takes into account not only the identity of an attractor but also its robustness , an important feature for the stability and reproducibility of developmental processes [13] , [18] , [61]–[63] . For each gene activity pattern that contributes to fitness and for each network in our analysis , evaluation of fitness involved the following steps: i ) The initial state of the gene network at time 0 was chosen to be a perturbation of the target pattern , drawn from a probability distribution where the initial state of each gene differs from that of with probability . ii ) We carried out network dynamics ( eq . 1 ) until some new attractor was reached; iii ) We recorded the Hamming distance ( ) separating from , and calculated the contribution to fitness of this developmental trajectory as ; modifications of by varying the exponent produce equivalent results . iv ) We repeated steps i ) –iii ) 500 times to determine 500 values ( ) . Notice that several of such 500 values would correspond to the same initial condition , and that the distribution of possible initial conditions is biased towards gene activity patterns similar to the reference pattern . This reflects our assumption , for the sake of simplicity , that selection favors similar initial conditions leading to the same selected activity pattern . We also assumed that gene activity patterns that are similar to the reference pattern are more likely to be required as initial conditions . Relaxation of such assumptions by variation in did not modify our results . We then calculated the network's fitness as ( 4 ) where is the arithmetic mean of all . Wherever fitness needed to be evaluated for multiple gene activity patterns , we calculated the arithmetic mean of over these multiple patterns . Notice that selection is pushing the acquisition of different gene activity patterns that would appear under different conditions ( such as different parts of the organism ) . Hence , the optimal networks will be those with dynamics that lead to different attractors matching the reference activity patterns , and not those with a single attractor that is a combination of the reference patterns . Had we used multiplicative contributions to fitness then the benefits that result from attaining a gene activity pattern would have depended on the acquisition of all other activity patterns . Because our simulations start with selection for a single activity pattern , it was preferable to assume otherwise . Using additive contributions to fitness guarantees that networks that are not able to attain the new gene activity pattern still have a chance to contribute to the next generation . However , usage of multiplicative fitness contributions does not affect our results qualitatively . For each simulation of gene network evolution , we first built a 10 node network and added 20 interactions at random to its interaction matrix . These interactions were activating or repressing , with equal probability . To construct the initial population we exposed 100 copies of this initial network to random mutation . Mutations occurred independently among different genes . A mutation of a gene either added a positive or negative interaction affecting the gene's activity , or eliminated one of the interactions that regulated the gene . Such mutations can be interpreted as changes in the regulatory regions of a gene , adding or eliminating cis-regulatory elements . Most of our results are based on a probability of a mutation occurring in a gene ( ) of 0 . 05 . This value of allowed adaptation within a tractable number of generations . Variation in did not affect our results , but only affected the time required for adaptation . For a gene undergoing mutation , we defined the probability of losing an interaction as ( 5 ) and the probability of acquiring a new interaction as . Here represents the number of regulators of the gene , and equals the number of genes in the network , and hence , the maximum number of regulators of any gene . This procedure results in networks that evolve towards low connectivities of 2–3 regulators per gene . Such low connectivity is often observed in transcriptional regulation networks of plants , animals , fungi and bacteria [64] . Loss of interactions may also help explain the observation that loss of gene expression is more common than acquiring new expression patterns in the evolution of gene families [65] . In our evolutionary simulations , we kept populations of constant size ( 100 individuals ) and imposed iterative rounds of mutations and selection to those populations . Every new generation , we sampled networks from the ones in the previous generation with probability proportional to the networks' fitness . Specifically , we defined the probability of copying network from generation into a network in the new generation ( ) as , where stands for the fitness of network . Each of the new networks underwent mutation with a probability of per gene . Finally , we evaluated the fitness of each new network . We iterated these steps through the end of the simulation . All simulation code ( written in C++ ) took advantage of the LEDA library of C++ data types [66] . | Throughout life's history , organisms have produced evolutionary innovations , features that are useful when facing new ecological and environmental challenges . A property that aids in the production of such innovations is modularity . Modular systems consist of groups of molecules with many interactions within a group but fewer interactions between groups . Such modularity increases the chances of innovation , because it allows changes inside one module without perturbing others , and because it permits redeployment of modules to create new biological functions . We simulate the evolution of gene networks known to be important in development to show that modularity increases when selection favors specialization in gene activity . Specialization occurs wherever new cell types , organs , or other body structures arise . In the course of this process gene networks acquire the ability to produce new gene activity patterns specific to these structures . We also demonstrate how modularity favors the evolution of new gene activity patterns that make use of already existing modules . Because specialization in gene activity is very common in evolution , the mechanism that we put forward may be important for the origins of modularity in gene regulatory networks . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"computational",
"biology/evolutionary",
"modeling",
"developmental",
"biology/developmental",
"evolution",
"evolutionary",
"biology/developmental",
"evolution"
]
| 2010 | Specialization Can Drive the Evolution of Modularity |
Human networks of sexual contacts are dynamic by nature , with partnerships forming and breaking continuously over time . Sexual behaviours are also highly heterogeneous , so that the number of partners reported by individuals over a given period of time is typically distributed as a power-law . Both the dynamism and heterogeneity of sexual partnerships are likely to have an effect in the patterns of spread of sexually transmitted diseases . To represent these two fundamental properties of sexual networks , we developed a stochastic process of dynamic partnership formation and dissolution , which results in power-law numbers of partners over time . Model parameters can be set to produce realistic conditions in terms of the exponent of the power-law distribution , of the number of individuals without relationships and of the average duration of relationships . Using an outbreak of antibiotic resistant gonorrhoea amongst men have sex with men as a case study , we show that our realistic dynamic network exhibits different properties compared to the frequently used static networks or homogeneous mixing models . We also consider an approximation to our dynamic network model in terms of a much simpler branching process . We estimate the parameters of the generation time distribution and offspring distribution which can be used for example in the context of outbreak reconstruction based on genomic data . Finally , we investigate the impact of a range of interventions against gonorrhoea , including increased condom use , more frequent screening and immunisation , concluding that the latter shows great promise to reduce the burden of gonorrhoea , even if the vaccine was only partially effective or applied to only a random subset of the population .
In 2017 the WHO added Neisseria gonorrhoeae to its priority list of bacterial pathogens in response to the global spread of antibiotic resistance [1] . The bacteria have developed resistance to every therapy used against them , from penicillin through to third-generation cephalosporins [2 , 3] . At a time when resistance to first line therapy ( ceftriaxone 250-500mg in combination with azithromycin 1-2g ) is increasingly observed [4] , it is more important than ever to understand the transmission dynamics of the infection , and how interventions might be used to reduce the burden on antibiotic treatment [5] . It has been well documented that heterogeneity in sexual activity levels has an impact on disease transmission , with individuals who have many partners bearing much of the burden of disease [6–9] . However , the risk of acquiring and passing on a sexually transmitted infection ( STI ) depends not only on an individual’s sexual risk profile , but also on their position in the wider sexual network [10 , 11] . Furthermore the structure of an underlying network affects the probability that an infection that is introduced to the network leads to an outbreak , as well as the size and longevity of any outbreaks that occur [12 , 13] . As such , it is important to take into account the structure of the underlying sexual network when modelling STI outbreaks . The distribution of the number of sexual contacts within a network is known as its degree distribution . Several studies have shown that real world sexual networks often have degree distributions that obey a power-law [14–16] , where the probability of having k partners over a given period of time is proportional to k−γ . The constant γ is usually between 1 and 4 , and different values have been observed in heterosexual and same-sex networks , as well as between genders [15] . Power-law networks exhibit high levels of heterogeneity , with the majority of individuals having a relatively small number of contacts , while a few have many . The standard method to simulate power-law networks is to use a system of preferential attachment , in which individuals are added one by one , connecting with a higher probability with existing individuals who already have a large number of partners [17] . Once all individuals have been added , the network thus created is guaranteed to have a static power-law distribution . It is important to note that even though the preferential attachment algorithm is dynamic in nature , the dynamic method used is purely a technique for generating a static network and does not in any way reflect the dynamics known to occur in real world sexual networks . Furthermore , transmission of infection occurs only once the network has been generated , with all partnerships being in place constantly from the beginning of the simulation of infection transmission until the end . An alternative method of producing static networks with a power-law degree distribution has been proposed based on each network node having an intrinsic fitness parameter , and a function that determines the probability that a network connection exists between any two nodes depending on their fitness [18 , 19] . In a sexual network , this can be thought of as each individual having an inherent propensity to seek new partnerships , relative to others in the network , with the probability of occurrence of each possible partnership depending on the mutual attraction of two individuals . In a sexual network model , the rate of infection of an individual depends on whether their sexual partners are infectious , rather than on the prevalence of infection in the pool of potential partners , as in compartmental models . Compartmental models that do not explicitly represent partnerships have been shown to underestimate the importance of core groups of highly sexually active individuals in sustaining STI transmission , while overestimating the contribution of long-term partnerships and low-activity individuals [20] . Furthermore , several studies have shown that , in order to explain observed patterns of infection , it is important to take into account not only the network structure but also the duration of partnerships , and the gaps between them [21 , 22] . It may therefore be necessary to use a dynamically evolving network to correctly simulate the spread of STI outbreaks . The power-law network methods described above [17–19] produce networks that are static and do not capture the dynamics of real networks . Conversely , several dynamic algorithms have been proposed where relationships are formed and dissolved over time [21 , 23 , 24] , but they do not explicitly aim for the degree distribution over a year to be power-law distributed , as observed in real networks [14–16] . Since no algorithm has yet been designed to simulate a dynamic sexual network with the correct real-world properties of a power-law distribution of number of sexual partners over a year , the difference between such a realistic dynamic network and a more approximate static network has not been assessed . Here we present a novel approach to dynamic network simulation using stochastic partnership formation and breakdown based on individuals’ intrinsic properties . We demonstrate that our method produces power-law networks , and that it can simulate a population reflecting the observed network characteristics in UK men who have sex with men ( MSM ) . We then simulate an outbreak of gonorrhoea in three types of network: a fully-connected static network , a heterogeneous static network , and our novel heterogeneous dynamic network , showing important differences between all three models in terms of the resulting patterns of transmission . We estimate the resulting offspring distribution ( the number of secondary cases caused by each primary case ) and generation time distribution ( time between infection of a primary case and infection of a secondary case ) to assess the likelihood of super-spreading events predicted by each network structure , as well as the predicted duration of infection . We also compare the impact of the number of sexual partnerships on the probability of infection and transmission under each network structure , providing a basis for risk assessment . Finally , using the dynamic network model we investigate the impact of a range of interventions against gonorrhoea , including increased condom use , more frequent screening and a hypothetical vaccine .
We first analysed the number of partners reported by MSM in the third National Survey of Sexual Attitudes and Lifestyles ( Natsal-3 ) , a population-based survey conducted in 2010-2012 [25–27] . 15 . 4% individuals reported zero partners , and amongst the remainder the distribution of number of partners approximately followed a power-law distribution ( Fig 1A ) . We used Bayesian inference to estimate the exponent γ of this power-law distribution , and found a posterior mean of γ = 1 . 81 ( 95% credible interval: [1 . 69 , 1 . 96] ) . This is comparable to estimates calculated based on the previous Natsal data , collected in 1990-1991 and 1999-2001 , and the London Gay Men’s Sexual Health Survey [28] , which were 1 . 57 ( 95% CI: [1 . 43 , 1 . 72] ) , 1 . 75 ( 95% CI: [1 . 57 , 1 . 95] ) and 1 . 87 ( 95% CI: [1 . 80 , 1 . 94] ) respectively [15] . We performed the same analysis based on data collected between June and November 2004 as part of the national Gonococcal Resistance to Antimicrobials Surveillance Programme ( GRASP ) run by Public Health England ( PHE ) from individuals diagnosed with gonorrhoea in London [29 , 30] . Fewer individuals had only one partner in the last three months than would be expected under a power-law distribution; however a power-law tail was observed for MSM having more than one partner . The inferred scale-parameter γ for gonorrhoea infected individuals was significantly lower than that in the Natsal-3 data at 1 . 60 ( 95% credible interval: [1 . 56 , 1 . 65] ) with non-overlapping credible intervals ( Fig 1B ) . We developed a new algorithm to simulate dynamic sexual networks in which relationships are being formed and broken down over time . To incorporate sexual behaviour heterogeneity , each individual in the network is characterised by a parameter λ that represents the propensity to make and break relationships . This λ parameter is analogous to the fitness property used in a previously published method to generate static power-law networks [18 , 19] . We built upon this work to create a dynamically evolving network model , with mathematical properties such that individuals are involved over one year in a power-law distributed number of partnerships ( see Materials and methods ) . In our model long-term partnerships form less frequently than short-term partnerships . High degree individuals have a higher turnover of partners than low degree individuals , rather than accumulating more long-term concurrent partners . To demonstrate the ability of our algorithm to simulate realistic networks , we generated dynamic sexual networks of size N = 10 , 000 over one year using a power-law exponent γ equal to 1 . 7 , 1 . 8 and 1 . 9 ( Fig 2 , S1 Fig ) . The network size was chosen to represent MSM aged between 15 and 65 in a UK city such as Brighton or central Manchester [31 , 32] . Our algorithm also requires to set the parameter k0 which determines the proportion of individuals that do not have a sexual partnership during the year . Using values of k0 equal to 0 . 4 , 0 . 5 and 0 . 6 respectively , we were able to produce networks exhibiting a power-law distribution of partnerships and proportion of individuals with zero partners over one year that were comparable to the 15 . 4% proportion in the Natsal-3 data ( Fig 2 ) . Finally , a third parameter ϕ in our method determines the rate of partnership breakdown , which in turn decreases the level of partnership concurrency in the network without affecting the distributions of partner numbers . The full mathematical description of parameters k0 and ϕ can be found in the Materials and methods section . To assess the importance of the underlying sexual network structure and dynamism in the way gonorrhoea outbreaks spread , we performed a comparison of simulated gonorrhoea outbreaks on three types of networks ( Fig 3A ) : a fully connected network , a static power-law network and our new dynamic power-law network . On each type of network we used the same model of gonorrhoea outbreak , adapted from a recent study [33] as illustrated in Fig 3B and described in the Materials and Methods section . The flow parameters were calibrated for each of the three types of network in order to produce outbreaks of the same realistic size over a year ( cf Materials and methods section ) . The resulting parameter values are summarised in Table 1 with no significant difference between the three models for any parameter except the rate of transmission per partnership β , which takes widely different values as expected . From the resulting simulations we analysed the offspring distribution , defined as the number of onward transmissions attributable to every case infected in the first year ( Fig 4A ) and we extracted the generation times , defined as the length of time from acquisition of infection to onward transmission to an uninfected partner ( Fig 4B ) . The offspring distributions we derive are the average of 100 model realisations , based on all individuals that become infected in each simulation , and as such are conditional on the probability of infection . Both the offspring distribution and the generation time distribution exhibited important differences depending on the underlying type of sexual network considered ( Fig 4 ) . For all three network structures ( fully connected , static , and dynamic ) mean numbers of offspring per infected individual were around 1 . 2 with overlapping 95% ranges ( Fig 4A , X axis ) . This equality is due to the calibration of the models , which required outbreaks to be of similar sizes . However , we found that the variance in the simulated offspring distributions differed depending on network structure ( Fig 4A , Y axis ) . Simulated outbreaks in both static and dynamic power-law networks had greater variance in the offspring distribution than the fully connected networks ( 4 . 2; 95% range: [1 . 9 , 7 . 7] ) , due to the effect of heterogeneity in contact patterns . However , dynamic partnership formation and dissolution partially mitigated the impact of the network structure on the offspring distribution . The sample variance of the offspring distributions in the static power-law network ( 10 . 9; 95% range: [5 . 7 , 20 . 3] ) was on average greater than in the dynamic network ( 6 . 8; 95% range: [3 . 4 , 12 . 0] ) , suggesting that adopting a static power-law network in disease models would overstate the importance of super-spreading events . The distribution of mean generation times in the simulations is also affected by the underlying network structure ( Fig 4B , X axis ) . Outbreaks in the dynamic network structure have a mean generation time of 63 days ( 95% range: [33 , 101] ) . Generation times are overestimated when partnership dynamics are ignored , as in the case of the static power-law network structure ( 77 days; 95% range: [49 , 124] ) , an effect which is exacerbated when heterogeneity in sexual activity levels is omitted , as in the fully connected network ( 109 days; 95% range: [54 , 168] ) . In order to maintain persistence of the outbreak at realistically low prevalence , as is observed in gonorrhoea , outbreaks in the fully connected network overstate the proportion of asymptomatic infections compared to the dynamic network , 44 . 6% ( 95% range: [17 . 2% , 59 . 7%] ) compared to 29 . 5% ( 95% range: [6 . 4% , 58 . 8%] ) [6] . This larger untreated asymptomatic reservoir also serves to increase the variance of the simulated generation times from 6 , 960 ( 95% range: [2 , 100 , 17 , 530] ) in the dynamic networks to 13 , 500 ( 95% range: [3 , 850 , 26 , 200] ) in the fully connected networks ( Fig 4B , Y axis ) . For both the static and dynamic power-law networks we investigated the relationship between the number of sexual partners that an individual has over one year , the probability of becoming infected , and the number of transmission events arising from those individuals who become infected . Fig 5 shows the proportion of infected individuals , the probability of an individual becoming infected in first year and the mean onward transmissions for infected individuals , split by the number of partners over one year . The proportion of infected individuals having fewer than three partners per year was lower than would be expected under a power-law distribution for both static and dynamic networks , however the distribution exhibited power-law behaviour for more highly active infectees ( Fig 5A ) . This is similar to the pattern exhibited in the GRASP London data ( Fig 1B ) . Compared to the dynamic network the static network structure appears to overestimate the burden of infection in individuals with more than 11 partners per year , while underestimating the burden in individuals with fewer partners . Under the static network structure an individual’s probability of infection increases linearly with their annual number of partners ( Fig 5B ) . When we allow for partnership dynamics , the probability of infection increases linearly at first , albeit at a slower rate than the static network , then levels off in individuals having more than five partners per year , eventually approximating the probability of infection observed in a fully connected network . This suggests that a static network structure may underestimate the risk of infection for individuals with few sexual partners . There is a similar relationship between the expected number of onward transmission events from individuals and their number of partners over the year in the static and dynamic network structures ( Fig 5C ) . An infected individual’s expected number of transmissions in the static network increases linearly with their total number of partnerships . For individuals with up to five partners per year , the dynamic network also shows a strong linear relationship between the expected offspring and number of partners . However , for individuals with more than five partners per year the relationship is less strong with a much greater variance in mean number of offspring . In the dynamic network the expected number of offspring is greater than one in individuals with at least three partners , whereas the mean offspring per person in the static network only becomes greater than one in individuals having more than 90 partners . The static network therefore likely overestimates the importance of very highly active individuals in maintaining transmission . Using the dynamic network model we investigated the impact of a range of interventions and preventative measures against gonorrhoea , including: 20% increased condom use ( resulting in a reduction in the transmission rate β ) , 20% increased sexual health screening ( an increase in parameter η ) , and the impact of a hypothetical gonorrhoea vaccine , deployed either at random to 20% of the network or targeted to a “core group” of 20% of individuals with the highest propensity to form partnerships ( λ ) . In practice these people would be identified by clinics through repeat infections and/or co-infections , self-reported larger numbers of sexual partners , or though being named repeatedly as a contact by other patients with infection . It should be noted however that the original conception of the core group was in relation to a simple compartmental model with core and non-core parts of the population [52] , whereas in a network model such as ours , the boundary of the core group is not clearly defined . Recent estimates suggest that a meningococcal B vaccine may be 31% [21% , 39%] effective against gonorrhoea [53] . The impact of vaccinating 20% of individuals at random with a vaccine that is 100% effective is comparable to vaccinating 65% [51% , 95%] of individuals with a vaccine of similar effectiveness . We assessed the one-year impact of these four measures on the probability of a outbreak stemming from a single introduction of gonorrhoea into a dynamic sexual network ( Fig 6A ) , the total number of gonorrhoea diagnoses ( Fig 6B ) , and the number of sexual health clinic visits from both screening and symptomatic treatment-seeking ( Fig 6C ) . The baseline proportion of simulated outbreaks persisting for at least one year from a single introduction of gonorrhoea was 31% ( 95% range: [15% , 49%] ) . All of the interventions we considered reduced the probability of an outbreak , with vaccination having the greatest impact; a fully effective vaccine administered to 20% of individuals in a randomised strategy reduced the probability of an outbreak by around a third to 21% ( 95% range: [7% , 40%] ) , a targeted strategy had a greater effect , reducing the probability to 19% ( 95% range: [5% , 36%] ) . The non-vaccine interventions had less of an impact . Increasing condom usage by 20% reduced the probability of an outbreak by around 10% to 28% ( 95% range: [13% , 47%] ) ; A 20% increase in the rate of screening for asymptomatic cases had a similar effect , reducing the probability of an outbreak to 29% ( 95% range: [11% , 49%] ) . The expected size of the visible outbreak in a population of 10 , 000 , as measured by the total number of infected individuals diagnosed and receiving treatment , was reduced by a fifth from 98 ( 95% range: [25 , 269] ) to 79 ( 95% range: [19 , 217] ) with a 20% increase in condom-use , and could be halved using vaccination: down to 47 cases ( 95% range: [14 , 114] ) using the randomised strategy and 45 cases ( 95% range: [14 , 105] ) by targeting the most sexually-active individuals . However , a 20% increase in the screening rate resulted in a 5% increase in the visible outbreak size to 103 cases ( 95% range: [25 , 295] ) , due to more asymptomatic cases receiving treatment . There was a similar pattern in the burden of sexual health services , while increased condom use decreased the total number of clinic visits by 11% from 16 , 939 ( 95% range: [7 , 950 , 28 , 810] ) to 15 , 053 ( 95% range: [7 , 064 , 25 , 602] ) . A 20% increase in the rate of screening , both of uninfected and asymptomatically infected individuals increased the total clinic visits by 20% , because the majority of testing is prompted by screening rather than symptomatic treatment seeking . The number of sexual health clinic visits remained stable in the vaccine scenarios , however the financial and administrative cost of initiating either a targeted or randomised vaccination programme must be considered once a vaccine candidate has been developed . It is important to note that while the targeted strategy is more effective , requires the ability to identify and vaccinate the 20% most sexually active individuals in a given population . It is unclear how effectively this could be done in practice , but could perhaps be offered at the GUM clinic at the same time as testing .
Real-world sexual networks are dynamic by nature , and we have developed a method that can reproduce observed power-law distributions of numbers of sexual partners in the last year in a dynamic network , which accounts for heterogeneity of individual behaviour in the way relationships form and break down . Our model allows the user to specify the power-law distribution via the exponent γ , to vary the proportion of individuals having no partners via the parameter k0 , and to set the average length of partnerships , via the parameter ϕ . Varying the length of partnerships for a given degree distribution impacts the pattern of partnership concurrency in the network . The longer the average partnership , the higher the degree of concurrency . While other models may be able to produce a power law distribution over time with the right choice of formation and breakdown partnership functions [24] , we are the first to focus on this property and to demonstrate how it can be achieved . We implemented this dynamic simulation algorithm into a R package called simdynet which is freely available at https://github . com/lwhittles/simdynet . Taking an outbreak of gonorrhoea as a case study , we found that failing to allow for sexual network structure ( i . e . using the fully-connected network ) resulted in an overestimation of the duration of carriage and asymptomatic reservoir . When network structure , but not dynamics of sexual partnership formation and breakage , was accounted for ( i . e . using the static network ) the model overstated the likelihood of super-spreading events and the burden of disease among individuals with high numbers of partners , compared to a dynamic model . While it is important to take heterogeneity into account , the traditional formulation of a core group [6 , 21] might approximate the true transmission dynamics of gonorrhoea better than using a static power-law network . Our findings add support to previous modelling work that suggested that having more sexual partners does not greatly impact the rate at which antibiotic resistant gonorrhoea can spread [9] . We used our realistic dynamic power-law network model to investigate the impact of a range of interventions against gonorrhoea , including increased condom use , more frequent screening and immunisation . Our results confirm that vaccination shows great potential to reduce the burden of gonorrhoea [54]: if a random 20% of individuals were immune , then the probability of outbreaks persisting at least a year would be reduced by 16% with the outbreak size reduced on average by 31% . ( Fig 6 ) . Such a level of protection could be achieved either through vaccination of a small portion of the population with a highly effective vaccine , or by widespread use a less effective vaccine . For example , a recent retrospective case-control study has shown that the MeNZB vaccine against meningitis is cross-protective against gonorrhoea , with an estimated effectiveness between 20% and 40% [53 , 55] . The dynamic sexual network we implemented is highly realistic , but also too computationally expensive to be used in many applications . For example if the population under study is very large , keeping track of every partnership formation and break down is clearly inefficient , especially for the study of the early stages of an outbreak of a new resistant strain in which only a small subset of individuals are being affected . However , we have computed from the full dynamic model the offspring distribution and distribution of generation time ( Fig 4 ) . These estimates allow for our model to be approximated as a stochastic branching process [56 , 57] , where infected individuals transmit to a number of secondary cases drawn from the offspring distribution , and the intervals of time between each primary and secondary cases are drawn from the generation time distribution . The advantage of such a model formulation is that it is much simpler than the full dynamic sexual network we described in this paper , but retains the same basic properties in terms of the transmission process . Furthermore , a branching model is at the basis of several recently developed methods to reconstruct transmission trees from genomic data , such as outbreaker [58] , TransPhylo [59 , 60] and phybreak [61] . Our estimates of the generation time distribution and offspring distribution therefore pave the way for these genomic epidemiology methods to be applied to the reconstruction of transmission in gonorrhoea outbreaks [42 , 62–64] . We estimated that on average the mean and variance of the generation time distribution were equal to 63 days and 6980 days2 , respectively ( Fig 4B ) , which can be emulated using a Gamma distribution with shape and scale parameters equal to 0 . 57 and 110 . 48 , respectively . The resulting 95% quantile range of the generation time stretches up to 298 days , which is in good agreement with an analysis of genomes from pairs of known sexual contact , in which the greatest observed time to most recent common ancestor was 8 months [42] . Since gonorrhoea can often remain asymptomatic , any outbreak reconstruction would need to account for the possibility of unsampled cases acting as intermediates in the transmission chains [42 , 58 , 60] . Accurate inference of unsampled cases requires in turn a good prior knowledge of the generation time distribution like the one we estimated here based on a dynamic power-law network . Both the mean and variance of the generation time distribution are overestimated when considering a fully-connected or static network ( Fig 4B ) which would likely result in an underestimation of the role played by unsampled cases . The offspring distribution was estimated to have a mean and variance equal to 1 . 2 and 6 . 8 , respectively ( Fig 4A ) . Since the variance is greater than the mean , the offspring distribution is over-dispersed compared to a Poisson process , which indicates the presence of super-spreaders [65–67] , although this transmission heterogeneity is not as pronounced as would be implied by an unstructured or static network model ( Fig 4A ) . In a branching model , over-dispersion can be implemented using a Negative-Binomial distribution for the number of offspring , which in this context is often parametrised in terms of its mean and dispersion parameter k , with lower values of k indicating more over-dispersion [60 , 65 , 68 , 69] . Here we estimated that k = 0 . 257 , which with a mean offspring number of 1 . 2 gives a 95% quantile ranging up to nine secondary cases , compared to only four cases for a Poisson distribution with the same mean . We ensured that our estimates were appropriate for use in simple branching process models by confirming that there is no correlation between successive offspring distributions ( S2 Fig ) . These estimates of the generation time distribution and offspring dispersion parameter pave the way for future studies of genomic epidemiology in gonorrhoea outbreaks .
The third National Survey of Sexual Attitudes and lifestyles in the UK ( Natsal-3 ) , was conducted between September 2010 and August 2012 in 15 , 000 adults aged between 16 and 74 [25–27] . We extracted the number of same-sex partners over one year for the 188 men who reported sexual contact with another man within the past five years . 15 . 4% of MSM reported having no same sex partners over the past year . In addition to the Natsal data we examined the number of partners reported by 691 MSM diagnosed with gonorrhoea in a collection of 2 , 045 isolates sampled between June and November 2004 from 13 major sexual health clinics throughout London as part of the national Gonococcal Resistance to Antimicrobials Surveillance Programme ( GRASP ) , run by Public Health England ( PHE ) [29] . PHE has produced a GRASP report annually since 2000 to monitor trends in resistance and susceptibility to the drugs used to treat gonorrhoea in England and Wales , which is used to inform national treatment guidelines and strategy . The GRASP data from London represent 54% of the 3 , 754 cases reported across the city at that time [30 , 42 , 70] . We fitted the power-law distribution using Bayesian inference , implemented via a Monte Carlo Markov Chain , to obtain obtained posterior estimates of γ based on the Natsal-3 and GRASP London datasets , using an uninformative γ ∼ U [ 1 , 10 ] prior . Five chains with over-dispersed starting points were run for 100 , 000 iterations after a 10 , 000 iterations burn-in period and thinned by a factor of 100 . The convergence of the MCMC was assessed by visual inspection of the trace plots , and confirmed to have a Gelman-Rubin criterion of < 1 . 1 [71 , 72] . In order to investigate the patterns of transmission under different network structures , we consider a stochastic individual-based model of gonorrhoea , adapted from [33] . Individuals are initially uninfected ( U ) and become infected probabilistically at rate β due to sexual contact with an contagious individual as dictated by the underlying sexual network structure . Infected individuals initially pass through an incubation period ( I ) which they leave at rate σ . A proportion ψ of those infected then develop symptoms ( S ) , whereas the remainder enters an asymptomatic stage ( A ) . In men , gonococcal infection can occur in the rectum , pharynx and/or urethra , resulting in different rates of onward transmission and probabilities of developing symptoms [73] . We do not explicitly model separate anatomical sites of infection , therefore the rate of transmission , β , and the likelihood of developing symptoms , ψ , should be seen as an average for any infection site . Asymptomatic individuals ( A ) undergo screening and receive treatment at rate η , otherwise recovery from asymptomatic infection happens ( either naturally or following unrelated antibiotic treatment ) at rate ν . The symptomatic individuals ( S ) seek treatment at rate μ . Individuals who have been treated recover from the infection and become uninfected again at rate ρ . The contagious population is denoted C = I + S + A , since individuals in treatment are assumed either to no longer be contagious or to abstain from sexual activity in accordance with treatment guidelines [51] . Using the Gillespie algorithm described in S2 Appendix we generated dynamic sexual networks exhibiting a power-law distribution of partnerships and proportion of individuals with zero partners over one year that were comparable to the Natsal-3 data . A thousand sets of parameters were sampled by Latin Hypercube Sampling from input ranges selected based on published sources ( cf Table 1 and S3 Appendix ) . For each underlying network structure ( fully connected , static and dynamic ) we seeded infection 100 times and simulated over three years for each parameter set . Parameter sets that produced outbreaks persisting at least one year in fewer than 20% of simulations were discarded , as were parameter sets that resulted in total diagnoses exceeding 400 cases per year on average . Thus only parameter sets that resulted in realistic gonorrhoea outbreaks were retained . We considered a suite of interventions and preventative measures against gonorrhoea , including: increased condom use ( resulting in a reduction in the transmission rate β ) , increased sexual health screening ( an increase in η ) , and the impact of a hypothetical gonorrhoea vaccine , deployed either at random or targeted to the individuals with the highest propensity to form new partnerships ( λ ) . We assessed the one-year impact on the probability of a outbreak stemming from the introduction of gonorrhoea into a dynamic sexual network , the number of sexual health clinic visits from both screening and symptomatic treatment-seeking , and the total number of gonorrhoea diagnoses . | The formation and dissolution of sexual relationships in human populations constitute an ever-changing network of links between individuals through which sexually transmitted diseases spread . To study this phenomenon , we developed a dynamic simulation algorithm that can reproduce the same distribution of sexual contacts as observed in real populations . We applied our algorithm to the study of gonorrhoea outbreaks and showed that it results in significantly different patterns of transmission compared to models where the sexual network does not change or is ignored . We show how our model can be incorporated into existing algorithms of outbreak investigation based on genomic sequencing data . We also apply our model to the evaluation of a range of interventions frequently proposed to limit the spread of gonorrhoea transmission , and in particular we quantify the potential of vaccination strategies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"immunology",
"microbiology",
"vaccines",
"preventive",
"medicine",
"antibiotic",
"resistance",
"probability",
"distribution",
"mathematics",
"sexually",
"transmitted",
"diseases",
"network",
"analysis",
"pharmacology",
"infectious",
"disease",
"control",
"vaccination",
"and",
"immunization",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"antimicrobial",
"resistance",
"epidemiology",
"men",
"who",
"have",
"sex",
"with",
"men",
"probability",
"theory",
"people",
"and",
"places",
"microbial",
"control",
"population",
"groupings",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"sexuality",
"groupings"
]
| 2019 | A dynamic power-law sexual network model of gonorrhoea outbreaks |
Dengue , an arboviral disease , is a public health problem in tropical and subtropical regions worldwide . In Brazil , epidemics have become increasingly important , with increases in the number of hospitalizations and the costs associated with the disease . This study aimed to describe the direct costs of hospitalized dengue cases , the financial impact of admissions and the use of blood products where current protocols for disease management were not followed . To analyze the direct costs of dengue illness and platelet transfusion in Brazil based on the World Health Organization ( WHO ) guidelines , we conducted a retrospective cross-sectional census study on hospitalized dengue patients in the public and private Brazilian health systems in Dourados City , Mato Grosso do Sul State , Brazil . The analysis involved cases that occurred from January through December during the 2010 outbreak . In total , we examined 8 , 226 mandatorily reported suspected dengue cases involving 507 hospitalized patients . The final sample comprised 288 laboratory-confirmed dengue patients , who accounted for 56 . 8% of all hospitalized cases . The overall cost of the hospitalized dengue cases was US $210 , 084 . 30 , in 2010 , which corresponded to 2 . 5% of the gross domestic product per capita in Dourados that year . In 35 . 2% of cases , blood products were used in patients who did not meet the blood transfusion criteria . The overall median hospitalization cost was higher ( p = 0 . 002 ) in the group that received blood products ( US $1 , 622 . 40 ) compared with the group that did not receive blood products ( US $550 . 20 ) . The comparative costs between the public and the private health systems show that both the hospitalization of and platelet transfusion in patients who do not meet the WHO and Brazilian dengue guidelines increase the direct costs , but not the quality , of health care .
Dengue fever ( DF ) is an important public health concern in tropical and subtropical regions worldwide , with approximately 100 million dengue infections and 24 , 000 deaths occurring annually worldwide [1] , [2] . Recently , the rates of severe illness and hospitalizations related to dengue have increased , threatening public health and negatively influencing the growth of developing countries , and particularly those in Latin America [3] , [4] , [5] . Dengue is an arboviral disease transmitted to humans by Aedes aegypti [6] , and the dengue virus ( DENV ) is a positive-strand RNA virus that belongs to the Flaviviridae family . Four genetically distinct DENVs ( DENV-1 , DENV-2 , DENV-3 and DENV-4 ) cause dengue with a wide clinical spectrum of symptoms , including fever and severe dengue ( SD ) [7] , [2] . Sequential heterotypic infections are also common in dengue-endemic areas [8] . The economic literature on the costs of DF is recent and minimal . The results are often conflicting because studies have used inconsistent assumptions . Moreover , these studies have failed to report the differences between the public and the private health care systems and the costs of hospitalization and platelet transfusion in cases in which the World Health Organization ( WHO ) guidelines were not followed [3] , [9] , [10] . In 2010 , Brazil had 94 , 887 hospitalizations and 673 deaths due to dengue and 60 . 4% of worldwide reported cases of dengue illness . The State of Mato Grosso do Sul had the second highest incidence rate of dengue in Brazil , with 2 , 593 . 6 cases per 100 , 000 inhabitants . The prevalent dengue serotype was DENV-1 , with co-circulation of DENV-2 and DENV-3 [11] , [1] . We designed the present study to describe and compare the direct public and private medical costs of hospitalized dengue cases and the costs of platelet use and hospitalization with or without adherence to the criteria recommended in the WHO guidelines .
We conducted the study in Dourados , the second largest city in the State of Mato Grosso do Sul , in the Midwest region of Brazil . This city is located 235 km from the capital , Campo Grande , at a latitude of 22°13′18 . 54″ South and a longitude of 54°48′23 . 09″ West . The estimated population of Dourados City is 196 , 035 inhabitants , 181 , 005 of whom reside in the urban area [12] , [13] . Dourados , a center for public health consortia , is of economic and political importance and links 30 counties ( Figure 1 ) [12] , [13] . Preventing and controlling dengue and other epidemic outbreaks , along with all associated health care in Brazil , is the federal government's responsibility . County governments are responsible for administering the public health care system , with technical and financial assistance from the federal government and states . Moreover , hospitalization is provided by private and public health care systems . Four hospitals were involved in the study . One was a public ( university ) hospital , and the other three were private . This is the unique public hospital in Dourados supported by the Unified Health System ( Sistema Único de Saúde – SUS ) , only this hospital has pediatric and neonatal intensive care units ( ICUs ) and is reference for hospitalization dengue cases . The other three hospitals treat patients with private health ( PH ) care , which consists of PH plans ( PHPs ) and payment with own resources ( OR ) . These hospitals only have an adult ICU , and only one of hospitals has more than 50 beds . This study included a retrospective survey with a cross-sectional design , and we used a bottom-up approach to determine the direct medical costs of the hospitalization of dengue cases using the Health System Agency Funding perspective [14] , [15] , [16] . The population consisted of all mandatorily reported dengue cases and hospital admissions from January to December 2010 in Dourados . We obtained the cases from the official database , or the National System for Reportable Diseases ( Sistema Nacional de Agravos de Notificação - SINAN ) [17] , and from the medical records of suspected cases of dengue at the hospitals . We included all of the hospitalized dengue cases entered into the Information System for Disease Notification ( SINAN ) Dourados from January to December 2010 , and we excluded dengue patients who were discharged because of a change in diagnosis . The SINAN includes all cases reported as suspected dengue cases . Subsequently , the epidemiological surveillance team performed an investigation to confirm or exclude each case using the WHO criteria for inclusion [2] . We obtained access to the SINAN database from the Management Epidemiology and Information Municipal Health Secretariat of Dourados City and extracted the following variables: name , date of birth , sex , date and place of hospitalization . Based on these variables , we requested the medical records from the department of medical records at each institution , and the review of medical records by researchers occurred between 2012 and 2013 . In this study , we defined clinical cases of dengue as patients who were identified by physicians and who exhibited the following symptoms and clinical signs: febrile illness presenting with at least 1 clinical manifestation suggestive of dengue illness , including headache , retro-orbital pain , myalgia , joint pain , rash or any bleeding symptom [2] . For each dengue case , an NS1 ELISA , RT-PCR laboratory confirmation and serotyping tests were performed [18] , [19] , [20] . In 2009 , to identify patients at an increased risk of complications from dengue , the WHO revised its 1997 classification of DF , DHF and dengue shock syndrome ( DSS ) , which was very rigid and limited to evaluating patients with severe clinical presentations that frequently did not meet the criteria . The new classification was based on a list of clinical warning signs suggestive of a severe disease outcome and classified dengue into dengue without warning signs ( DWWS ) ; dengue with warning signs ( DWS ) , such as abdominal pain or tenderness , persistent vomiting , clinical fluid accumulation , mucosal bleeding , lethargy/restlessness , liver enlargement greater than 2 cm and an increase in the hematocrit concurrent with a rapid decrease in the platelet count; and severe dengue ( SD ) [2] . The criteria used for SD were as follows: shock , fluid accumulation with respiratory distress , severe bleeding , an AST or ALT level greater than or equal to 1 , 000 , impaired consciousness and severe involvement of the heart and other organs [21] , [22] , [23] . In 2010 , the Brazilian Ministry of Health continued to classify dengue cases using the 1997 WHO classification . Thus , we reclassified past dengue cases according to the new classification proposed by the WHO in 2009 using information extracted from medical and laboratory records . The criteria for hospitalization based on the WHO recommendations are as follows: i ) patients with co-existing conditions that may make dengue or its management more complicated ( such as pregnancy , infancy , old age , obesity , diabetes mellitus , renal failure or chronic hemolytic disease ) ; ii ) patients with certain social circumstances ( such as living alone or living far from a health facility , without reliable means of transport ) ; and iii ) patients who require emergency treatment and urgent referral due to SD [2] . Platelet transfusion is indicated when severe thrombocytopenia ( platelet count <20 , 000 mm3 ) is present , with suspected bleeding in the central nervous system and/or major bleeding from the gastrointestinal tract ( and/or the vagina in adult females ) [2] . We classified the cases that did not satisfy the recommended WHO criteria as i ) admission without criteria ( AWC ) and/or ii ) platelet transfusion without criteria . For sociodemographic characterization , we considered the following variables: age , sex , race and education . We also examined clinical characteristics and outcomes by analyzing the type of hospital or intensive therapy clinic , the final disease classification , the case outcome ( cure or death ) and the use or lack of use of hospitalization and platelet transfusion criteria . To analyze the direct medical costs , we examined the hospitalization duration , complementary examinations , medications , medical fees , inputs and the type of health care system ( public or private ) [15] , [24] . We obtained the costs of each hospitalization directly from the hospitals' own records . To measure the costs of assistance , we considered 2 services: public ( SUS ) and private ( PH ) [25] . The direct medical costs included payments for hospital health care , medical services and prescriptions that were made by OR , PH and the SUS . The payments made by the SUS for hospitalized cases are based on illness type using an established price to account for hospital health care , medical services and prescriptions , except complex medical procedures and laborious assays , which are paid for separately . Therefore , the SUS transfers a fixed amount to accredited hospitals using Hospital Admission Authorization ( Autorização de Internação Hospitalar - AIH ) , or US $135 . 4 per hospitalization for dengue cases [26] , [27] . This amount includes the costs of daily hospitalization , medications , supplies , laboratory tests and X-rays and remains constant , regardless of the number of hospitalization days or tests required . However , additional fee is paid when computed tomography ( CT ) ; magnetic resonance imaging ( MRI ) ; or other special procedures , such as a transfusion or ICU use , are required . These additional costs may contribute to the variations in the total amount paid by the SUS to the hospitals for each case [26] , [27] . We compiled the PHP and OR payment amounts from private health care companies and/or the hospitals' financial departments . To calculate the health care costs of the 4 hospitals in this study , we used the final financial data present in the medical records , which were based on specific tables . We then determined the economic value of dengue and compared it with Brazil's gross domestic product ( GDP ) per capita for 2010 ( US $330 . 4 ) [13] . The Kolmogorov-Smirnov and Shapiro-Wilk tests showed that the data were not normally distributed ( p<0 . 0001 ) , so we expressed the categorical variables as proportions and the continuous variables as the median and interquartile range ( IQR; 75th and 25th percentiles ) . We used the Mann-Whitney U-test or the Kruskal-Wallis test to compare medians and the chi-squared test to compare proportions , with a significance level of 95% . We double-typed the data using EpiData version 3 . 1 ( Lauritsen JM ( Ed . ) , Odense , Denmark ) and performed statistical analyses using SAS 9 . 1 ( SAS Institute , Cary , NC ) . The values were converted from reals ( R$ ) to US dollars ( US$ ) by the Brazilian Central Bank on December 31 , 2010 ( US $1 . 00 = R $1 , 695 ) . We also stratified the dengue costs by cost type , financing source , sociodemographic characteristics , age group and outcomes ( hospitalization and/or platelet transfusion with or without use of the WHO criteria ) . The project protocols were approved by the Committee of Ethics and Research at the Federal University of Grande Dourados ( UFGD; protocol number 003/2011 ) . During all data collection , we guarantee anonymity creating alphanumeric codes to identify each patient in the study .
We stratified all sociodemographic characteristics and outcomes by health care system , and these characteristics are presented in Table 1 . We observed the highest incidence of hospitalized dengue ( 53 . 5% ) in the 15- to 60-year-old age group , with a median age of 39 . 5 years ( IQR , 19 to 57 years ) . However , in the public health care system , children younger than 15 years old were most severely affected , which was a significantly higher rate than that in the private health care system ( p<0 . 0001 ) , and the education level was lower in the public health care system ( p<0 . 0001 ) . We observed the highest incidence of ‘without WS’ hospitalizations in the private health care system . However , we observed no differences in SD hospitalizations or in clinical outcomes between the public and the private health care systems . The median cost of all reported dengue hospital admissions ( n = 288 ) was US $259 . 9 ( US $179 . 2 to US $621 . 2 ) ( Table 2 ) . The median values in the different age groups were US $201 . 1 ( US $184 . 7 to US $378 . 7 ) for children younger than 15 years , US $260 . 7 ( 169 . 8 to 605 . 8 ) for the 15- to 60-year-old age group and US $382 ( US $189 . 5 to US $762 ) for individuals older than 60 years ( p = 0 . 003 ) . The individuals older than 60 years stayed in the hospital for a median time of 4 days ( IQR , 2 to 6 days , p = 0 . 003 ) ( Table 2 ) . The major components of the cost analysis included the length of hospital stay , medical fees and prescriptions ( medications ) . Moreover , we determined that the cost of dengue treatment varied greatly among patients . Table 3 compares the dengue case costs in the public and private health care systems and provides p values . When we compared the length of hospitalization and the costs of the hospitalized cases according to illness categories ( WHO criteria ) and the health care systems , private had significantly higher cost than public , except for SD cases ( p = 0 . 3307 ) ( Table 4 ) . In this study , 51 individuals were AWC , which represented 18 . 1% of all dengue cases ( n = 288 ) . These cases represented 11 . 1% ( US $23 , 343 . 7 ) of the total hospital health care costs . 74 . 5% of the AWC patients ( n = 38 ) were hospitalized in a private institution . Table 5 presents the median costs and the length of hospital stay based on the health care system type ( public or private ) and the use of the WHO criteria . Platelet transfusion occurred in 17 . 7% ( 51/288 ) of the dengue cases , 35 . 2% of which ( 18/51 ) did not meet the WHO criteria . Additionally , 16 of these patients were hospitalized in the private health care system , and 2 were hospitalized in the public health care system ( Table 5 ) . The total costs were higher ( p = 0 . 040 ) in the group using platelets that did not meet the WHO criteria ( 18/51 ) , or US $1 , 090 . 6 ( IQR , US $1 , 454 . 3 to US $506 . 6 ) , compared with the total costs in the group that was administered platelets and met the WHO criteria ( n = 33 ) , or US $595 . 6 ( IQR , US $1 , 185 . 7 to US $344 . 8 ) . However , when we separately analyzed these parameters based on health insurance ( public or private ) , there was no statistically significant difference ( Table 5 ) . A comparison of the length of the hospital stay and costs stratified by classification ( DWWS , DWS and SD ) and the type of health care system showed significant differences ( Table 6 ) , including between SD and DWS ( p< . 0001 ) and between SD and DWWS ( p = 0 . 0002 ) . However , when we compared DWS and DWWS , there was no statistically significant difference ( p = 0 . 6447 ) .
In this study , we determined that the direct medical costs related to dengue equaled 2 . 5% of the public domestic product per capita of Dourados in 2010 , totaling US $210 , 084 . 3 . Additionally , the median cost of hospitalizations ( US $259 . 9 ) was equivalent to 78 . 7% of the median monthly GDP ( US $330 . 4 ) of the city population studied [13] . The cost analyses indicated that the cost in the private sector was 280% higher compared with the cost in the public sector . This difference is likely due to the fact that the SUS transfers a fixed amount to accredited hospitals using AIH ( US $135 . 4 per dengue case hospitalization ) [27] . However , this amount does not represent the real cost to the SUS of a patient with dengue and demonstrates underfunding and inadequate transfer by the SUS . Created in 1988 by the Brazilian Federal Constitution , the SUS is based on the principles of universality and equality , without any conditions , and guarantees free access to health care for approximately 190 million Brazilians [28] , [26] . However , in epidemic years , the demand is often higher than the availability of health care services . Therefore , the SUS is unable to meet the demand , which forces patients to use private institutions [29] . Previous studies have indicated that the population that uses the SUS has a low level of education , fewer financial resources and a low sociodemographic profile [30] , [31] , [32] . In our study , we assumed that the poorest individuals are the least studied and most need the SUS . The prevention programs conducted by the PH sector were introduced very recently , and measures to prevent DF are nonexistent [28] , [33] . We determined that the direct medical costs of the hospitalized cases in this study ( US $259 . 9 ) were lower compared with costs in studies examining all of Brazil ( US $381 ) [34] , [4] . This difference is likely due to the fact that the authors of those studies excluded the SUS financial data and used the amounts paid by the Brazilian PHPs , whose table values are far greater than those of the SUS . However , if we consider only the private sector amounts , the costs calculated in our study were higher ( US $515 . 8 ) . To our knowledge , no studies have compared the quality of the clinical management of DF regarding hospitalization criteria . However , studies of other diseases have indicated that approximately 30 to 40% of patients do not receive care according to scientific evidence and suggested guidelines . Moreover , approximately 20 to 25% of the care is not needed or is potentially harmful . Furthermore , patients who are hospitalized unnecessarily are exposed to inherent risks , such as iatrogenic or infectious diseases , which increase health care costs and the risk of death [35] , [36] , [37] , [38] . Blood components are expensive and potentially dangerous and have a short expiration date , and their availability is often limited [39] , [10] . In our study , the frequency of platelet use without meeting the recommended clinical criteria ( n = 18 ) was similar to that reported in other recent studies [40] , [41] . However , we determined that the median cost of hospitalization among patients using platelets without meeting the recommended criteria ( US $1 , 090 . 6 ) was 83 . 1% higher compared with that among patients who received blood components based on the WHO protocol ( US $595 . 6 ) . This difference may be due to the fact that most cases occurred in the private sector , which has higher costs . However , we observed a significant difference ( p = 0 . 002 ) between the groups that used ( US $1 , 622 . 4 ) and did not use ( US $550 . 2 ) blood products . The clinical management of patients with DWS and DWWS is different according to the WHO classification , as DWS requires more imaging methods , laboratory tests and medicines depending on how serious the case is [2] . However , our study showed no significant difference between the cases of DWWS and DWS ( p = 0 . 6447 ) , perhaps due to non-adherence to the revised WHO guideline recommendations or other reasons not available in this study . Our study was limited by using a secondary database . We could not evaluate indirect medical and nonmedical costs , expenditures on prevention and vector control or family income and financial impacts , and we did not determine why physicians did not follow the WHO recommendations . Furthermore , we assume the possibility of selection bias due to the refusal of 1 hospital to participate in the study ( n = 185 ) . The new WHO guidelines better classify the severity of dengue cases [42] , [43] , and our study demonstrated that the use of WHO recommendations may result in savings by reducing both unnecessary hospitalizations and the use of blood products . Therefore , training needs to be offered to health care professionals to improve adherence to the revised WHO guidelines . | The costs of dengue outbreaks and hospitalizations have recently increased . Endemic in many tropical and subtropical parts of the world , dengue outbreaks occur each year and require appropriate economic studies to determine the potential financial and public health impacts of dengue management policies . Economic literature on this topic is rare , and results are conflicting , because inconsistent assumptions are used . Health economics research specific to dengue is critical for controlling and preventing this disease . In Brazil , health care is the federal government's responsibility and is provided by the private and public health care systems; however , during an outbreak , both systems become overcrowded . Municipal governments are responsible for administering health care , with technical and financial cooperation from the government and states . The data presented here reveal the direct hospitalization costs , the private and public health care systems' costs and the impact of using the WHO guidelines on both systems . Together , these data will aid health care workers , researchers and health policy makers in financing the prevention , control and treatment of dengue fever . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"socioeconomic",
"aspects",
"of",
"health",
"medicine",
"and",
"health",
"sciences",
"health",
"economics",
"dengue",
"fever",
"neglected",
"tropical",
"diseases",
"tropical",
"diseases",
"viral",
"diseases",
"health",
"care"
]
| 2014 | Direct Costs of Dengue Hospitalization in Brazil: Public and Private Health Care Systems and Use of WHO Guidelines |
Dilated cardiomyopathy ( DCM ) is a structural heart disease with strong genetic background . Monogenic forms of DCM are observed in families with mutations located mostly in genes encoding structural and sarcomeric proteins . However , strong evidence suggests that genetic factors also affect the susceptibility to idiopathic DCM . To identify risk alleles for non-familial forms of DCM , we carried out a case-control association study , genotyping 664 DCM cases and 1 , 874 population-based healthy controls from Germany using a 50K human cardiovascular disease bead chip covering more than 2 , 000 genes pre-selected for cardiovascular relevance . After quality control , 30 , 920 single nucleotide polymorphisms ( SNP ) were tested for association with the disease by logistic regression adjusted for gender , and results were genomic-control corrected . The analysis revealed a significant association between a SNP in HSPB7 gene ( rs1739843 , minor allele frequency 39% ) and idiopathic DCM ( p = 1 . 06×10−6 , OR = 0 . 67 [95% CI 0 . 57–0 . 79] for the minor allele T ) . Three more SNPs showed p < 2 . 21×10−5 . De novo genotyping of these four SNPs was done in three independent case-control studies of idiopathic DCM . Association between SNP rs1739843 and DCM was significant in all replication samples: Germany ( n = 564 , n = 981 controls , p = 2 . 07×10−3 , OR = 0 . 79 [95% CI 0 . 67–0 . 92] ) , France 1 ( n = 433 cases , n = 395 controls , p = 3 . 73×10−3 , OR = 0 . 74 [95% CI 0 . 60–0 . 91] ) , and France 2 ( n = 249 cases , n = 380 controls , p = 2 . 26×10−4 , OR = 0 . 63 [95% CI 0 . 50–0 . 81] ) . The combined analysis of all four studies including a total of n = 1 , 910 cases and n = 3 , 630 controls showed highly significant evidence for association between rs1739843 and idiopathic DCM ( p = 5 . 28×10−13 , OR = 0 . 72 [95% CI 0 . 65–0 . 78] ) . None of the other three SNPs showed significant results in the replication stage . This finding of the HSPB7 gene from a genetic search for idiopathic DCM using a large SNP panel underscores the influence of common polymorphisms on DCM susceptibility .
Dilated cardiomyopathy ( DCM ) is a common form of heart muscle disease with a prevalence of 1∶2 , 500 in the general population . It represents a major cause of cardiovascular morbidity and mortality and is characterized by systolic dysfunction as well as dilation and impaired contraction of the ventricles , often leading to chronic heart failure and eventually requiring cardiac transplantation [1] . In about 35% of cases DCM is a familial disease [2] . However , in the sporadic form of DCM , i . e . after exclusion of affected family members and all detectable causes ( also called idiopathic DCM ) , a genetic component is discussed , but can thus far not be assigned to single gene defects . Knowledge of genetic risk factors for both , familial and non-familial forms of DCM is important to initiate treatment prior to symptomatic onset of the disease , to delay its occurrence or possibly halt its progression . To date , only a few common susceptibility alleles for sporadic DCM were identified from candidate-gene approaches , but could not be confirmed in replication samples [2] , [3] , this being a common problem of single gene based analyses [4] . In contrast , unbiased genome-wide association studies ( GWAS ) allow the identification of genetic risk factors even outside of known genes , but higher power is needed to compensate for multiple testing [5] . No comprehensive GWAS was performed to date on sporadic form of DCM . The cardiovascular gene-centric 50K single nucleotide polymorphism ( SNP ) ITMAT-Broad-CARe ( IBC ) array represents an established compromise between GWAS and hypothesis-driven candidate gene approach by analyzing polymorphisms in more than 2 , 000 genes known or predicted to be involved in cardiovascular phenotypes [6] . In this study , we conducted a screening based on the cardiovascular 50K SNP array with three independent replication studies to reveal insight in genetic contribution to idiopathic DCM . The four samples from Germany and France included 1 , 910 sporadic DCM cases and 3 , 630 healthy controls individuals . We identified a common intronic variant in HSPB7 , encoding a cardiovascular small heat shock protein , to be associated with sporadic form of DCM .
In our screening case-control sample , DCM cases were more likely men , were slightly younger and less frequently smokers , had a lower BMI and a higher prevalence of hypertension , hypercholesterolemia as well as type 2 diabetes ( Table 1 ) . After quality control , 30 , 920 SNPs were available for analysis with 23 , 307 independent markers ( defined as SNPs with pairwise r2<0 . 8 based on linkage disequilibrium ( LD ) in the control group ) . Therefore , we set a significance threshold to 0 . 05/23 , 307 = 2 . 15×10−6 to account for the multiple testing . In association analyses of this stage 1 study applying logistic regression adjusted for gender , four SNPs , namely rs1739843 ( HSPB7 , intron 2 ) , rs11701453 ( RUNX1 , intron 1 ) , rs7597774 ( ADD2 , intron 1 ) and rs2229714 ( RPS6KA1 , 3′ untranslated region ) showed a p-value below this threshold ( 3 . 16*10−8 , 1 . 65*10−7 , 2 . 05*10−7 , and 1 . 51*10−6 , respectively ) . Results were similar when additionally adjusting for age ( e . g . for rs1739843 p = 2 . 40*10−8 ) . None of the four polymorphisms showed deviation from Hardy-Weinberg equilibrium . The lowest p-value for association with DCM was observed for a SNP located in HSPB7 intron 2 ( rs1739843 ) leading to a protective effect of the minor allele ( OR = 0 . 67 [95% CI 0 . 58–0 . 77] ) . Analysis of the region around the SNP rs1739843 using HapMap data ( release #22 ) revealed the presence of six genes and 27 polymorphisms in LD with the lead SNP ( r2-value>0 . 5 ) ( Figure 1A ) . Nine of these SNPs were present on the cardiovascular 50K array after quality control and were located in HSPB7 gene as well as two genes downstream , CLCNKA and CLCNKB ( Figure 1B; Table S1 ) . In this sample , the genomic inflation factor λ was 1 . 285 for the highest 90% of the 30 , 920 observed p-values . When correcting rs1739843 for this λ factor , the p-value was 1 . 06*10−6 and OR = 0 . 67 [95% CI 0 . 57–0 . 79] ( Table 2 ) . The four SNPs with uncorrected p<2 . 15×10−6 in the initial scan ( rs1739843 , rs11701453 , rs7597774 and rs2229714 ) were analyzed using logistic regression adjusted for gender in three independent replication samples . First , n = 564 additional German DCM patients and n = 981 controls were genotyped for the four SNPs . Marker rs1739843 showed strong association with DCM ( p = 2 . 07*10−3 , OR = 0 . 79 [95% CI 0 . 67–0 . 92] ) . Conversely , for rs2229714 ( p = 0 . 075 , OR = 1 . 20 [95% CI 0 . 98–1 . 47] ) , rs11701453 ( p = 0 . 373 , OR = 1 . 09 [95% CI 0 . 90–1 . 31] ) and rs7597774 ( p = 0 . 621 , OR = 1 . 04 [95% CI 0 . 89–1 . 22] ) the initial association results were not replicated . Second , a French replication sample ( France 1 ) consisted of n = 433 cases and n = 395 controls . Only rs1739843 showed association with DCM after adjustment for gender ( p = 3 . 73*10−3 , OR = 0 . 74 [95% CI 0 . 60–0 . 91] ) . For the other SNPs , no significant association was seen in this sample . Third , in an independent second French replication sample ( France 2 ) , again only rs1739843 showed association with DCM after adjustment for gender ( p = 2 . 26*10−4 , OR = 0 . 63 [95% CI 0 . 50–0 . 81] ) . Replication results are summarized in Table 2 . None of the four polymorphisms showed deviation from Hardy-Weinberg equilibrium in any replication samples . In a combined analysis of the screening step , corrected for the λ factor of 1 . 285 , and the three follow-up studies ( n = 5 , 540 ) , the SNP rs1739843 reached a p-value of 5 . 28*10−13 ( OR = 0 . 72 [95% CI 0 . 65–0 . 78] ) for association with idiopathic DCM ( Table 2 , Figure 2 ) . There was no between-study heterogeneity for this effect ( I2 = 6 . 9% , p = 0 . 36 ) . To reveal potential causal variants , the coding region of HSPB7 was resequenced in a total of 48 DCM patients . We detected three known synonymous variants ( rs945416 , rs732286 and rs1739840 ) . The synonymous variants rs945416 ( position 19 , serine ) and rs732286 ( position 33 , alanine ) are in high LD with rs1739843 ( r2 = 0 . 96 , HapMap CEU data release #24 ) . SNP rs1739840 ( position 117 , threonine ) is not available in HapMap . In the initial sample of 664 DCM patients , all three synonymous polymorphisms are in perfect LD to each other and to rs1739843 as shown by genotyping . Neither missense nor splice site de novo mutations were identified by sequencing . Synonymous SNP rs11807575 , as well as non-synonymous variants rs77021870 and rs74626772 were listed in databases , but not found to be polymorphic in our sample . Since the design of the 50K human gene-centric bead chip ( IBC array ) aims at a large-scale gene-based approach , we screened candidate genes which are known for or potentially involved in susceptibility to DCM in our initial screening sample utilizing the information on 30 , 920 SNPs . We established a list of previously reported genes for DCM by searching PubMed and OMIM databases ( http://www . ncbi . nlm . nih . gov/ ) for “CARDIOMYOPATHY , DILATED” and “GENETIC” . A total of 315 SNPs including 234 independent SNPs ( defined as SNPs with pairwise r2<0 . 8 based on LD in the control group ) were located in or near ( +/−10kb ) the chosen candidate genes , representing 1 . 01% of array content . DCM association results for these SNPs were obtained from our screening study on 664 cases and 1 , 874 controls ( Table 3 , more details in Table S2 ) . On a single candidate gene level , polymorphisms in or near ABCC9 , DES , MYH6 and TPM1 showed nominal significance after Bonferroni correction for the number of SNPs tested in gene regions ( p = 0 . 010 , p = 0 . 022 , p = 0 . 005 and p = 0 . 018 , respectively ) . However , none of these SNPs remained significant after correction for the 234 independent markers tested in this DCM candidate gene approach . Our study was powered to detect moderate to large effects ( e . g . for OR>1 . 3 and MAF = 30% or OR>1 . 5 and MAF = 20% or OR>1 . 7 and MAF = 10% , the power was 56% , 96% and 97% for two-sided p<0 . 05/234 = 2 . 14*10−4 , respectively ) .
In the present case-control study , we evaluated the relationship of common SNPs with sporadic DCM using a large-scale screening approach . Our comprehensive strategy set out to analyze the human gene-centric 50K bead chip ( IBC array ) , which focuses on loci with a potential functional link to cardiovascular disease ( CVD ) and covers more than 45 , 000 SNPs from about 2 , 000 genes [6] . Our study identified a polymorphism ( rs1739843 ) in intron 2 of the HSPB7 gene being associated with susceptibility to DCM in a German case-control sample with three replication steps . Recently , Cappola et al . reported an association between rs1739843 and both , ischemic and non-ischemic heart failure , applying the same gene-centric 50K bead chip [7] . They found a protective effect of the minor allele , which is in conformity with our results on DCM . As DCM is a potential preliminary stage for non-ischemic heart failure , these independent findings point to a possible common pathophysiologic cascade . However , a second association signal for heart failure located in the FRMD4B region ( rs6787362 , minor allele frequency ( MAF ) 10 . 4% ) identified by Cappola et al . [7] could not be detected in our DCM case-control sample ( p = 0 . 64 ) . Our study had a power of 99% to find a nominal association between DCM and rs6787362 with p<0 . 05 and an OR = 0 . 67 . The finding on HSPB7 is also in-line with a previously reported large-scale re-sequencing approach in four biologically relevant cardiac signaling genes , which detected HSPB7 sequence diversity in sporadic cardiomyopathy [8] . Our data together with the results from Cappola et al . [7] and Matkovich et al . [8] , substantiate the importance of rs1739843 or related polymorphisms in the HSPB7 locus for DCM and heart failure and possibly underscore a common genetic basis for these related phenotypes . Matkovich et al . further report that none of the detected HSPB7 gene variants altered amino acid sequence [8] , which is also consistent with the fact that we found neither missense nor splice site mutations in the HSPB7 sequence . Therefore , the biological mechanism explaining the association between the polymorphism rs1739843 and DCM risk remains still unclear . The three detected synonymous variants ( rs945416 , rs732286 and rs1739840 ) are in high LD with each other as well as with our lead SNP rs1739843 and lie on one LD block . Therefore , it could be hypothesized that these SNPs represent causal risk factors for DCM , as described for the P-glycoprotein encoding gene MCP1 and affected drug and inhibitor interactions [9] . Synonymous SNPs lead to changes in codon usage and may cause functional implications by conformational changes in protein structure due to translation efficiency . Alternatively , a de novo splice site could be created by a SNP or other ( unmapped ) polymorphisms outside the HSPB7 coding region may alter its gene expression . Clearly , functional studies would be required to prove these hypotheses . Besides the HSPB7 gene , where the lead SNP is located , also five genes ( CLCNKA , CLCNKB , C1orf64 , ZBTB17 and SPEN ) lying on the same LD block may potentially be responsible for the association with DCM . CLCNKA and CLCNKB encode for two members of the family of voltage-gated chloride channels . These proteins are predominantly expressed in the kidney and participate in renal salt reabsorption [10] . The function of C1orf64 is currently unknown . ZBTB17 , also known as MIZ-1 , encodes a zinc finger protein involved in the regulation of c-myc [11] . SPEN ( RBM15C or MINT ) encodes a conserved transcriptional repressor that controls the expression of regulators in diverse signaling pathways [12] , [13] . HSPB7 , encoding the small heat shock protein cvHsp ( also known as HspB7 ) , is the functionally most plausible candidate gene in this genomic region . It is known to be expressed in cardiovascular and insulin-sensitive tissues [14] . In general , the expression and activation of heat shock proteins is influenced by elevated temperatures as well as ischemia , hypoxia and acute cellular stress [15] , [16] . In the aging skeletal muscle increase of cvHsp protein content was observed [17] . cvHsp was shown to be constitutively localized under non-stressful conditions to nuclear splicing speckles and may influence mRNA processing [18] . Recent data suggest co-localization between cvHsp and α-B-crystallin in the z-band of cardiac tissue and interaction with other small heat shock proteins [19] . However , further investigations like genomic fine-mapping and subgroup analyses in the context of cardiomyopathies are needed . Genetic analyses in familial forms of DCM led to the identification of risk loci showing X-linked , autosomal dominant or autosomal recessive patterns of inheritance [2] , [20] , [21] . Some of the DCM causing genes or plausible candidate genes were also covered by the 50K bead chip , wherefore we specifically tested those SNPs lying in risk gene regions ( 10 kb upstream and downstream , respectively ) . In these analyses , no significant association with any of the gene variants was found , indicating that in sporadic cases of DCM probably other pathways are involved than in familial DCM . However , less frequent variants may have been missed due to insufficient power of our screening sample . Furthermore , the distinction between familial and sporadic forms of DCM is , to a certain degree , somewhat arbitrary . Screening of family members is rarely done in clinical routine , but when carried out on a systematic basis , up to 7% of previously healthy first-degree relatives have reduced left ventricular function or dilation without presence of cardiac symptoms [22] . Therefore , it might be anticipated that genetic testing could help to identify individuals at risk in familial DCM but also in families of patients affected by so-called idiopathic forms of the disease . Already known genetic factors account for only a fraction of DCM heritability [20] . Given a 1 . 5-fold increased risk of DCM among heterozygous subjects in our screening sample ( 48% in the general population-based KORA study ) and a 2 . 25 times increased risk among homozygous subjects ( 34% in KORA ) , 49% of DCM cases would be attributable to the SNP rs1739843 ( or correlated polymorphisms ) with 19% attributable to heterozygous and 30% to homozygous carriers , respectively . Therefore , the genetic component seems to comprise a large proportion for this disease . However , with the prevalence of the idiopathic form of the disease being about 1∶2 , 700 [23] , a genetic screening of the general population would include four cases out of 10 , 000 screened persons and two of these would have the disease due to this SNP . Therefore , the great potential of this variant might rather be screening of high risk populations , or this pathway indicates potential drug targets . Further investigations should aim ( 1 ) to identify additional variants underlying DCM susceptibility with otherwise unknown etiology and ( 2 ) to analyze potential influence of these common alleles as modifiers for familial forms of DCM . Taken together for both , modifiers of familial forms and susceptibility alleles in idiopathic DCM , knowledge of genetic background will support preventive medical measures in the future . Some limitations of our study should be mentioned . First , we conducted a large-scale SNP analysis focused on genes potentially involved in cardiovascular traits . Therefore , on the one hand we were able to detect associations between DCM and polymorphisms only in these pre-selected genes . On the other hand , the 50K human CVD bead chip allows comprehensive gene-based analysis with more than 2 , 000 well covered loci . Second , our sample size only allowed to detect moderate to large effects ( e . g . for OR>1 . 3 and MAF = 30% or OR>1 . 5 and MAF = 20% or OR>1 . 7 and MAF = 10% , the power was 19% , 75% and 80% for p<2 . 15*10−6 , respectively ) . Therefore , we may have overlooked real association signals in our screening step . Third , there could be some population stratification in our initial screen sample . However , the observed λ could also be caused - in part - by underlying association due to the analysis of pre-selected loci known or suggested to be involved in cardiovascular phenotypes . The fact that the association between rs1739843 in HSPB7 and idiopathic DCM was replicated in three independent samples strongly enhances the confidence in our results .
The ethics committees of the participating study centers approved the study protocol and all participants gave their written informed consent . The study was in accordance with the principles of the current version of the Declaration of Helsinki . Cases for the initial German screening study were recruited from the German Heart Institute ( Berlin ) , and controls were from a population-based German KORA study ( follow-up survey F3 , Augsburg ) [24] . Phenotypic details are summarized in Table 1 . Controls ( n = 1 , 874 ) had no medical history for coronary artery disease ( CAD ) , myocardial infarction or DCM; mean age was 62±11 years and slightly more women ( n = 986 ) than men ( n = 888 ) were present in the control group . Inclusion criteria for DCM cases were the following: reduced systolic function ( left ventricular ejection fraction ( LVEF ) <45% ) without angiographically assessed evidence of major CAD , significant valvular heart disease ( >grade 2 , i . e . such as mitral or aortic regurgitation ) , hypertensive heart disease , congenital heart disease , myocarditis ( by endomyocardial biopsy , when available ) or other secondary forms of heart failure . Patients with a positive family history were also excluded from this study . In DCM cases ( n = 664 ) , mean LVEF was 24±3% and mean age of disease diagnosis was 46±11 years . For the first replication step , additional German DCM cases ( mean age 53±13 years; n = 564 , n = 440 men , n = 124 women ) were recruited from different German study centers: Berlin , n = 64; Lübeck , n = 96 ( Angio-Lueb ) ; Marburg , n = 61 ( EUROGENE ) ; Münster , n = 101 ( EUROGENE ) ; Regensburg , n = 150 ( EUROGENE ) ; Regensburg , n = 92 ( GoKard ) . Independent German KORA controls from surveys S1 and S2 ( n = 981 , n = 539 men , n = 442 women ) had a mean age of 52±10 years [24] . Inclusion and exclusion criteria were identical to the initial case-control sample . A second replication study ( France 1 ) was recruited in France ( CARDIGENE ) [25] , [26] . The French cases were of white European origin ( all born in France , from parents born in France or neighboring countries ) with a diagnosis of DCM , i . e . enlarged left ventricle end-diastolic volume/diameter >140 ml/m2 on ventriculography or >34 mm/m2 on echocardiography and LVEF ≤40% confirmed over a six-month period , in the absence of causal factors such as CAD or sustained hypertension , intrinsic valvular disease , documented myocarditis , congenital malformation , insulin-dependent diabetes . Only apparently sporadic DCM cases without additional ( first degree ) relative with DCM were included ( but 8% were in fact with familial form after careful cardiac examination in relatives ) . Recruitment was performed in ten hospitals in six regions in France ( Lille , Lyon , Nancy , Nantes , Paris-Ile de France , Strasbourg ) from September 1994 to February 1996 . A total of 433 patients ( 229 had undergone a cardiac transplantation ) were included ( n = 345 men , n = 88 women ) . Mean age of patients was 45±11 years , mean LVEF was 23±7% and mean end-diastolic volume was 195±67 ml/m2 . Controls ( n = 395 ) were age- and gender-matched ( n = 310 men , n = 85 women ) . The third replication sample was also of French origin ( France 2 ) . Inclusion criteria were identical to the France 1 sample . A total of 249 patients from EUROGENE and PHRC were included ( n = 198 men , n = 51 women ) . Mean age of patients at diagnosis was 51±10 years . Controls ( n = 380 ) were free of medical history for CAD , myocardial infarction or DCM and mean age was 46±11 years ( n = 301 men , n = 79 women ) . Initial genotyping was carried out using the 50K gene-centric human CVD bead chip version 1 ( IBC v1 array ) ( Illumina , San Diego , CA , USA ) [6] following the manufacturer's protocol . Data were analyzed ( calling and sample clustering ) and exported employing BeadStudio analysis software ( Illumina ) . From the initial 45 , 707 SNPs , those markers with low call rates ( <95% ) or low frequency ( MAF<1% ) were excluded . Minimal call rate per individual was 90% . We used identity-by-descent methods to exclude unknown first-degree relation of participants . Replication samples were taken from human CVD bead chip data or genotyped with 5′ exonuclease TaqMan technology ( Applied Biosystems , Foster City , CA , USA ) as previously described [27] . A by-design assay for rs1739843 was used with primer sequences 5′-CTCTGCCATCACCATCTCACA-3′ and 5′-GGCAGAGGGAGCCTGAG-3′ and probe sequences 5′-VIC-AGGGTGGGAGGTGACAG-NFQ-3′ and 5′-FAM-AGGGTGGGAGATGACAG-NFQ-3′ ( site of rs1739843 is underlined; fluorescence dyes VIC and FAM on 5′ end and non-fluorescence quencher ( NFQ ) on 3′ end are indicated ) . All other assays were obtained pre-designed directly from Applied Biosystems . Detailed information on assays used in France 2 sample are available at http://genecanvas . ecgene . net/infusions/genecanvas/Polymorphisms/PolymorphismsList . php . SNP rs1739843 was re-genotyped using the by-design TaqMan assay in initial case sample ( n = 664 ) to check for discrepancies between human CVD bead chip and TaqMan genotypes . A >99 . 8% concordance of genotypes was found . For all genotyped samples a call rate >97% for each SNP assay was reached . Polymerase chain reaction ( PCR ) primer were generated using Primer3Plus ( http://www . bioinformatics . nl/cgi-bin/primer3plus/primer3plus . cgi ) [28] to cover the coding parts of the three HSPB7 exons ( GenBank accession No . NM_14424 . 4 ) . The primer sequences and PCR amplification products are listed in Table S3 . Included intronic regions were 267 bp for 5′ end of intron 1 , 156 bp for 3′ end of intron 1 , 136 bp for 5′ end of intron 2 , and 89 bp for 3′ end of intron 2 , respectively . PCR cycling conditions consisted of an initial denaturation at 95°C for 9 min , followed by 40 cycles with denaturation at 95°C for 30 s , annealing at 60°C for 30 s , and elongation at 72°C for 30 s , with a final elongation step at 72°C for 7 min . After PCR amplification , primers and dNTPs were removed using ExoSAP-IT ( USB Europe , Staufen , Germany ) following the manufacturer's instructions . The purified PCR products were directly sequenced using the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit Version 3 . 1 on the ABI 3730 ( Applied Biosystems , Foster City , CA , USA ) . For initial screening and replication analyses , logistic regression adjusted for gender was used . P-values , odds ratios ( OR ) and their 95% confidence intervals ( CI ) were reported . The inflation factor λ was computed in the 50K initial screening analysis for logistic regression analysis assuming a χ2 distribution with two degrees of freedom of the minus two-times logep measures ( 90% highest p-values ) . The p-values and CI from initial screening analysis were genomic-control corrected using this λ factor via standard errors ( standard error[corrected] = sqrt ( λ ) *standard error ) and beta estimates ( 95%CI beta[corrected] = beta±1 . 96*standard error[corrected] ) . Deviation from Hardy-Weinberg equilibrium was calculated with an exact test [29] . Statistical and association analyses were performed using JMP 7 . 0 . 2 ( SAS Institute Inc , Cary , NC , USA ) and PLINK v1 . 07 ( http://pngu . mgh . harvard . edu/~purcell/plink/ ) [30] , respectively . Power analysis was carried out using Quanto 1 . 2 . 4 ( http://hydra . usc . edu/gxe/ ) . We combined the initial scan results corrected for λ with the replication studies' results using a fixed effect model . Annotation of association results on a genome level was performed with WGAViewer software ( http://people . genome . duke . edu/~dg48/WGAViewer/ ) [31] . LD patterns were calculated using HapMap releases #22 and #24 ( http://www . hapmap . org/ ) [32] . | Dilated cardiomyopathy is a severe disease of the heart muscle and often leads to chronic heart failure , eventually with the consequence of cardiac transplantation . Identification of genetic disease markers in at-risk persons could play an important role in preventive health care . Several mutations in familial forms of the disease are described . Here , we examine the role of common genetic variants on the sporadic form of dilated cardiomyopathy . By screening about 2 , 000 candidate genes previously related to cardiovascular disease in more than 1 , 900 cases and 3 , 600 controls , we show that a polymorphism in the HSPB7 gene ( rs1739843 ) is strongly associated with susceptibility to dilated cardiomyopathy . We also show that the effect on disease risk is present in both German and French cohorts . Therefore , this study is an important step towards revealing insight in the genetic background of the sporadic form of dilated cardiomyopathy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cardiovascular",
"disorders/myopathies",
"genetics",
"and",
"genomics/gene",
"discovery",
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"cardiovascular",
"disorders/heart",
"failure"
]
| 2010 | Genetic Association Study Identifies HSPB7 as a Risk Gene for Idiopathic Dilated Cardiomyopathy |
The rate of meiotic recombination varies markedly between species and among individuals . Classical genetic experiments demonstrated a heritable component to population variation in recombination rate , and specific sequence variants that contribute to recombination rate differences between individuals have recently been identified . Despite these advances , the genetic basis of species divergence in recombination rate remains unexplored . Using a cytological assay that allows direct in situ imaging of recombination events in spermatocytes , we report a large ( ∼30% ) difference in global recombination rate between males of two closely related house mouse subspecies ( Mus musculus musculus and M . m . castaneus ) . To characterize the genetic basis of this recombination rate divergence , we generated an F2 panel of inter-subspecific hybrid males ( n = 276 ) from an intercross between wild-derived inbred strains CAST/EiJ ( M . m . castaneus ) and PWD/PhJ ( M . m . musculus ) . We uncover considerable heritable variation for recombination rate among males from this mapping population . Much of the F2 variance for recombination rate and a substantial portion of the difference in recombination rate between the parental strains is explained by eight moderate- to large-effect quantitative trait loci , including two transgressive loci on the X chromosome . In contrast to the rapid evolution observed in males , female CAST/EiJ and PWD/PhJ animals show minimal divergence in recombination rate ( ∼5% ) . The existence of loci on the X chromosome suggests a genetic mechanism to explain this male-biased evolution . Our results provide an initial map of the genetic changes underlying subspecies differences in genome-scale recombination rate and underscore the power of the house mouse system for understanding the evolution of this trait .
Meiotic recombination fulfills dual roles in genetics and evolution . In many species , including mammals , the proper segregation of homologous chromosomes at the first meiotic division is contingent on the presence of at least one well-positioned crossover per homologue pair [1]–[3] . The improper patterning of recombination events across chromosomes can lead to aneuploidy , a significant risk factor for fetal loss and developmental disability in humans [4] . In addition , recombination influences the evolutionary dynamics of populations by rearranging existing patterns of allelic variation to generate novel multi-locus genotypes . This genetic shuffling can increase the efficacy of natural selection by decoupling high fitness alleles from linked deleterious variation [5]–[7] . At the same time , recombination can facilitate the removal of deleterious variation from the gene pool [8] , [9] . The amount of recombination per unit DNA ( i . e . the rate of recombination ) exhibits tremendous variation among species and between individuals . In mammals , the mean rate of recombination across species genomes varies by an order of magnitude [10]–[12] . Likewise , there is considerable heterogeneity in the global crossover rate among individual humans [13]–[17] , house mice [18] , [19] , dogs [20] , cows [21] , and shrews [22] . Fine-scale recombination rates display similar trends: the genomic locations of recombination hotspots are not conserved between humans and chimpanzees [23] , [24] , and hotspots that segregate as presence/absence polymorphisms are common in human populations [25] , [26] and among closely related laboratory strains of house mice [27]–[29] . Classical genetic experiments established that the rate of recombination is a complex genetic trait [30]–[33] . More recently , specific genes that influence genome-scale recombination rate variation in humans have been identified [17] , [34] , [35] . Prdm9 , a meiosis-specific histone H3 methyltransferase , was recently found to control the genome-wide distribution of recombination hotspots in mice [36] and humans [37]–[39] . Despite these exciting advances in our understanding of population variation in recombination rate , genetic explanations for the large differences in recombination rate between species remain elusive . Fundamental questions have never been addressed experimentally: How many loci contribute to species divergence in recombination rate ? What are their effect sizes and modes of inheritance ? Where in the genome do species recombination rate modifiers reside ? Do loci that control the positioning and activity of recombination within species also contribute to recombination rate differences between species ? Answers to these questions are essential for understanding how the rate of recombination evolves . To date , most genetic studies of recombination rate variation have measured recombination rates using patterns of marker inheritance in large human pedigrees [16] , [17] , [35] or in experimental crosses [29] , [40] , [41] . However , recombination rates are estimated from genetic data with considerable statistical uncertainty owing to the independent assortment of recombinant chromatids at meiosis [42] . In pedigree-based studies , this error is further compounded by the limited number of meiotic transmissions surveyed per individual . In addition , the inability to eliminate environmental contributors to phenotypic variation in humans adds even more noise to recombination rate estimates . These sources of error result in a marked loss of statistical power to find genomic regions contributing to recombination rate variation through linkage or association analysis . A powerful alternative approach to the genetic dissection of recombination rate variation is to combine experimental crosses with cytological measures of recombination rate [43] . In particular , the immunolocalization pattern of the mismatch repair protein MLH1 along the mature synaptonemal complex has been shown to accurately and faithfully reproduce the distribution of meiotic crossovers in late pachytene spermatocytes [14] , [44]–[46] and oocytes [47] . The MLH1 method for measuring recombination rate offers several notable advantages over traditional genetic approaches . First , because crossovers are directly observed , recombination rate estimates are not affected by binomial sampling of recombinant chromosomes at meiosis . Second , large numbers of spermatocytes or oocytes can be analyzed to yield precise estimates of the global recombination rate for single individuals . Third , though laborious , this method is inexpensive compared to the costs of genotyping many offspring from a single individual ( as required by pedigree-based methods ) . We use the MLH1 immunocytological method to demonstrate that males from wild-derived inbred strains of the house mouse subspecies Mus musculus musculus have markedly increased genome-scale recombination rates relative to the closely related subspecies M . m . castaneus and M . m . domesticus . We identify multiple genetic determinants of this substantial divergence in global recombination rate , providing an initial portrait of the genetic basis of species differences in this key genomic parameter .
A representative image of a late pachytene spermatocyte stained with fluorescently labeled antibodies against MLH1 and a protein component of the synaptonemal complex ( SYCP3 ) is shown in Figure 1 . We used the MLH1 immunostaining assay to measure genomic recombination rates in two wild-derived inbred strains from each of three distinct subspecies of house mice ( Mus musculus musculus , M . m . castaneus , and M . m . domesticus ) [48] . We observed a striking difference in mean MLH1 foci count between M . m . musculus and both M . m . domesticus and M . m . castaneus males ( Figure 2; Wilcoxon signed rank test , P<10−16 ) . On average , M . m . castaneus and M . m . domesticus have 21–23 MLH foci per meiosis , whereas M . m . musculus males undergo >26 crossovers . Several patterns suggest that the distribution of MLH1 foci along the synaptonemal complex ( SC ) faithfully mirrors the distribution of meiotic crossovers in the wild-derived inbred strains we examined . First , SCs lacking a MLH1 focus were rare in our survey ( 0 . 4% ) , consistent with the obligate chiasma requirement for homologue disjunction in mammals [1] , [2] . Second , on SCs with two or more MLH1 foci , foci were distantly spaced . This patterning matches expectations under a model of positive crossover interference , a process that is known to be important in house mice [42] . Third , we very seldomly observed cells with two or more SCs lacking a MLH1 focus . Pachytene spermatocytes nearly always had a full complement of foci , indicating that MLH1 protein loads on and off sites of recombination repair along the SC in a highly concerted fashion . Fourth , our cytology maps approximate the total male mouse genetic map length estimated from genetic data . Assuming that each MLH1 focus corresponds to a map distance of 50 cM , wild-derived inbred strains included in our survey have map lengths that range from 1085 cM–1500 cM . This range includes the estimate of total male genetic map length from the standard mouse map ( 1375 cM ) [49] . Although the small fraction of crossover events that are not dependent on MLH1 will be missed by this method [50] , our observations suggest that the total number of MLH1 foci in a meiotically dividing cell provides a reliable estimate of the genomic recombination rate . Mus musculus subspecies radiated nearly simultaneously from a common ancestor ∼500 , 000 years ago [51] , [52] . The striking increase in mean MLH1 foci count in inbred strains of M . m . musculus relative to the M . m . domesticus and M . m . castaneus strains suggests that considerable divergence in male recombination rate has accrued along the M . m . musculus lineage . Under a neutral model of phenotypic evolution , the expected recombination rate divergence between subspecies is ≈2Vmt , where Vm is the per-generation rate at which phenotypic variance increases via neutral mutation and t is the divergence time in generations [53] . Given that t is roughly equal for pairwise comparisons between house mouse subspecies and assuming constancy of Vm over this short evolutionary period , absolute divergence in recombination rate should be approximately equal among subspecies pairs . Clearly , our data are not consistent with this theoretical prediction . At mutation-drift equilibrium , the amount of within subspecies polymorphism for recombination rate is ≈2NeVm , where Ne is the effective population size [53] . Curiously , M . m . musculus has a smaller estimated Ne than either M . m . domesticus or M . m . castaneus ( ∼60 , 000 , 100 , 000 , and 200 , 000 , respectively; [52] ) yet displays the highest level of polymorphism for recombination rate ( ∼3 MLH1 foci between CZECHI and PWD ) . The higher polymorphism for recombination rate in M . m . musculus and greater divergence for recombination rate in comparisons with this subspecies are consistent with several evolutionary hypotheses . Recombination rates may have experienced a relaxation of selective constraint along the M . m . musculus lineage . Alternatively , recombination rates in M . m . domesticus and M . m . castaneus may have been subject to stronger purifying selection . These findings could also be explained by higher mutational variance for recombination rate in M . m . musculus . We caution that these observations derive from comparisons of just two wild-derived inbred strains per subspecies . A detailed survey of polymorphism and divergence in recombination rate in natural populations of these three subspecies will be required to determine the underlying evolutionary processes at work . To investigate the genetic basis of the rapid divergence in genomic recombination rate in M . m . musuclus , we conducted an F2 intercross between wild-derived inbred strains PWD/PhJ ( M . m . musculus ) and CAST/EiJ ( M . m . castaneus ) . We measured the global rate of recombination in 276 F2 males by averaging total autosomal MLH1 foci counts from at least 15 spermatocytes per animal ( mean = 20 . 4 cells ) . F2 mice vary substantially in the global number of crossovers ( Figure 3 ) . Importantly , the variation in MLH1 foci number among cells from a single male is far less than the variation in mean MLH1 foci count among animals , indicating the presence of segregating genetic differences in this inter-subspecific F2 population . Most males have recombination rates that fall within the range defined by the two parental means , although 9% of individuals lie outside these values . The continuous nature of this variation provides evidence for multiple recombination rate modifiers segregating between the parental PWD and CAST strains . The high broad-sense heritability of recombination rate in this cross ( H2 = 0 . 93 ) motivates the application of genetic mapping approaches to link variation in recombination rate with genetic variation at specific genomic loci . We genotyped our F2 population at 222 informative SNPs distributed across the genome . Using standard interval mapping [54] with a permutation-derived threshold for declaring statistical significance ( genome-wide α = 0 . 05 ) [55] , we identified two genomic regions linked to variation in mean MLH1 foci count . One of these QTL localizes to the proximal half of chromosome 7 and the second QTL lies on the X chromosome ( Figure 4 ) . Although there is a clear peak in the X chromosome LOD profile centered on ∼30 cM , the entire chromosome displays strong statistical evidence for linkage to variation in global recombination rate ( Figure 4 ) . QTL genotype at this single , large-effect locus explains 46% of the variance in mean MLH1 foci count among F2 males ( adjusted R2 = 0 . 46 from a linear regression ) . Interestingly , the allele from the low recombination rate CAST parent confers the increase in recombination rate at this X-linked locus , opposite the pattern seen at the chromosome 7 QTL ( Table 1 ) . Consistent with this result , we uncover a striking difference in genomic recombination rate between reciprocal F1 males ( Figure 6 ) . Male F1 animals that receive their X chromosome from a CAST mother ( CASTxPWD F1s ) have ∼5 more MLH1 foci per meiosis than F1 males carrying the PWD X chromosome ( PWDxCAST F1s ) . Single QTL mapping approaches , including standard interval mapping , formally assume that only one QTL in the genome affects the phenotype . When QTL of moderate to large effect exist , accounting for the phenotypic variance they explain can enhance statistical power to find additional loci . The discovery of the major effect QTL on the X chromosome prompted us to use an approach that could adjust for the presence of this locus to enable the simultaneous detection of multiple additional QTL . We applied a model-based multiple QTL mapping strategy [56] to identify the set of genetic loci that best explain segregating variation in mean MLH1 foci count among F2 males . Using a forward/backward model selection approach , with model discrimination performed via penalized LOD scores to control the rate of false inclusion [57] , we identify six autosomal QTL and two X-linked QTL for genomic recombination rate ( Figure 5 ) . As expected , the two QTL identified in the single QTL scan are among those recovered in the multiple QTL mapping analysis . The six autosomal loci show mostly additive effects , with CAST alleles at the chromosome 4 , 11 , and 17 QTL exhibiting slight dominance over PWD alleles ( Table 1 ) . At each autosomal locus , the high recombination rate PWD parent confers the high recombination rate allele . Consistent with single QTL scan results and reciprocal F1 phenotypes , the high recombination rate allele at both X-linked QTL derives from the low recombination rate CAST parent ( Table 1 ) . Combined , these eight QTL explain 74% of the phenotypic variance among F2s , individually accounting for 0 . 9–3 . 4 MLH1 foci ( 2–35% of the total phenotypic variance; Table 1 ) . These effect sizes are probably overestimated , as the conflation of QTL detection and estimation on a common dataset leads to a systematic upward bias [58] . Although these eight QTL explain a large fraction of F2 variation in global recombination rate , they account for a lesser percentage of the observed difference between the parental PWD and CAST strains . Combined , the six autosomal QTL explain a difference of 8 . 5 MLH1 foci between the inbred parents – more than the observed difference of 8 foci . However , the two X-linked QTL account for ∼4 MLH1 foci in the opposite direction . Summing these effects suggests that our multiple QTL model explains approximately half of the difference in mean MLH1 foci count between PWD and CAST males . Clearly , additional QTL for mean MLH1 foci number segregate between these strains . These undetected QTL likely have small to moderate effect sizes , as power calculations indicate that our study is only sufficiently powered ( 80% power ) to find QTL with additive effects >0 . 9 [59] . The early stages of female meiosis , including recombination , occur in the fetal ovary . These temporal aspects of oogenesis complicate cytological analysis of recombination in females . For this reason , we limited our genetic mapping efforts to males . However , the genome-wide rate of recombination is a sexually dimorphic trait in many mammals , including house mice . The female standard mouse genetic linkage map is 9% longer than the corresponding male map [49] , and marked sex-specific recombination trends are observable on finer physical scales of measurement [29] , [49] , [60] , [61] . The non-random concentration of QTL with transgressive effects to the X chromosome , coupled with the noted sex differences in this trait , led us to investigate variation in global recombination rate in females from the two parental inbred strains and hybrid F1s . We applied the MLH1 immunostaining procedure to oocytes harvested from day 17–20 post-conception PWD and CASTxPWD F1 female fetuses ( n = 3 and n = 2 animals , respectively ) . Mean MLH1 foci counts from CAST females have been reported previously [62] . Although CAST and PWD males differ in their global crossover count by 8 MLH1 foci , PWD females have only 2 foci more than CAST females ( Figure 6 ) . PWD and CASTxPWD F1 females have indistinguishable crossover counts . Overall , there is surprisingly little variation for mean MLH1 foci count among females , indicating that evolutionary divergence in recombination rate has occurred primarily in males . These findings suggest that several of the QTL detected in our inter-subspecific F2 male population may be sex-limited in their expression or have polarizing effects in males versus females [17] . Interestingly , PWD females have a lower mean MLH1 foci count than PWD males ( Figure 6 ) . This finding presents an intriguing directional reversal of the global recombination rate sex dimorphism widely observed in house mice [49] , [61] , nominating the PWD strain as an excellent model for understanding the causes of sex differences in this phenotype .
Our genetic study of variation in mean MLH1 foci number in an inter-subspecific panel of F2 males identifies QTL for global recombination rate divergence . Our findings provide initial clues toward the genetic mechanisms of species divergence in this trait . First , the discovery of eight QTL jointly explaining 74% of the variance among inter-subspecific F2 males indicates that observed patterns of recombination rate evolution are dominated by loci with large phenotypic effects . The X-linked QTL at 33 cM is the strongest modifier of global recombination rate identified to date , explaining 35% of the variation in our inter-subspecific F2 panel ( Table 1 ) . The presence of such a large effect locus provides a clear genetic mechanism for rapid phenotypic evolution between species . Second , the autosomal loci we identify display mainly additive inheritance . This finding extends studies of within species recombination rate variation in humans [34] , indicating that additive alleles contribute to both within and between species differences in recombination rate . Third , at least two of the QTL we identify exert trans effects on recombination rate . Males do not recombine along their X chromosome , indicating that the two X-linked QTL act strictly in trans . This result corroborates findings from genetic studies of fine-scale recombination rate control: Prdm9 regulates recombination hotspot activity across the mouse genome [36] , indicating that trans regulatory mechanisms are important for both the fine- and broad-scale control of recombination . Fourth , our multiple QTL map points to the presence of high and low recombination rate alleles in the two parental strains ( Table 1 ) . A similar pattern has been previously reported for recombination rate variation in Drosophila melanogaster [32] and is often observed in the evolution of complex traits [63] . Finally , our study uncovers a prominent role for the X chromosome in the evolution of recombination rate . Combined , the two X-linked loci in our multiple QTL model account for a difference of 4 MLH1 foci ( 200 cM ) between males hemizygous for CAST versus PWD alleles . Recessive X-linked loci subject to positive selection will reach fixation more rapidly than autosomal loci because their expression is unmasked in hemizygous males [64] . We speculate that selection on the X-linked modifiers identified in our F2 male population may have played a leading role in the rapid evolution of recombination rate in this sex . Recently , Murdoch et al . [43] used the approach applied in our study – genetic mapping of MLH1 foci count in F2 males – to identify seven QTL conferring recombination rate differences between the C57BL6 and CAST inbred mouse strains . Five of these loci map to chromosomes that harbor QTL in our study , including the X chromosome ( chromosomes 3 , 4 , 15 , 17 , and X ) . Interestingly , the CAST genotype at the X-linked QTL was associated with only a moderate increase in F2 recombination rate in this study , as opposed to the large effect observed in our cross ( males with the CAST X have ∼1 focus more than males with the C57BL6 X chromosome; in comparison , males with the CAST X have ∼2 . 5 foci more than males with the PWD X chromosome ) . If the large-effect X-linked QTL at 33 cM identified here and the X-linked QTL identified by Murdoch et al . [43] are the same locus , it would appear that genetic background strongly affects its expression . Our application of multiple QTL mapping did not identify any genetic interactions ( even when the penalty to QTL inclusion was relaxed; data not shown ) , but we acknowledge limited power to find interacting QTL with our small sample size . Allelic incompatibilities that decrease reproductive fitness in hybrids commonly evolve between incipient species [65]–[67] . These genetic incompatibilities often affect hybrid fitness by hindering progression through meiosis , including impairment of chromosome synapsis and recombination [e . g . 68] . Importantly , we detect no epistasic interactions between subspecies-specific alleles in our cross . Several additional observations indicate that observed F2 variation in mean MLH1 foci count is not due to hybridization-related defects in meiosis . First , our immunostaining assay allowed us to identify diplotene stage cells in all F1 and F2 animals , ruling out wide-spread activation of the pachytene meiotic checkpoint as an underlying mechanism of possible hybrid sterility [69] . Second , we observed no overt defects in chromosome pairing or synapsis in any hybrid animals . If CAST and PWD hybrids suffer fitness reductions , the molecular mechanism ( s ) of infertility must act after the completion of recombination at prophase I . It is difficult to imagine how any problems that surface late in meiosis ( or possibly in spermatogenesis ) could affect recombination . Third , the distribution of mean MLH1 foci counts in our F2 population is centered on the mid-parent mean ( Figure 3 ) , an unlikely result in the event of hybrid dysgenesis in the phenotype . Finally , we note that M . m . musculus and M . m . castaneus hybridize in nature [70] and F1s from both directions of our CAST and PWD cross were fertile . Together , these observations provide a compelling case that patterns of inter-subspecific hybrid variation in mean MLH1 foci count reflect underlying subspecies differences in the rate of recombination per se . Our analysis uncovered two striking differences in recombination rate evolution between the sexes . First , the magnitude of evolutionary change is much greater in males than females . Second , there is a reversal in the direction of the sex-dimorphism for recombination rate between PWD and CAST . It is tempting to consider these results , combined with the localization of both transgressive QTL to the X chromosome , as inter-related findings . In particular , they seem to raise the possibility that divergence in recombination rate among house mouse subspecies has been shaped by conflicting evolutionary pressures on the sexes . If natural selection favors distinct recombination rates in males and females [71] , [72] , modifiers might preferentially aggregate on the X chromosome [73] . Recessive X-linked loci will differ in expression between males and females , thereby imposing a reduced fitness burden on the opposite sex . This scenario is speculative , especially given that the dominance effects of X-linked recombination rate modifiers identified here cannot be determined from hemizygous F2 males . An extension of our QTL analysis to include mean MLH1 foci counts in F2 females will offer further clues into the evolution and genetic basis of these intriguing observations . Although the rate of recombination differs between males and females of many mammalian species [13] , [16] , [19] , [74] , [75] , the causes of this pattern remain poorly understood . Sex differences in crossover interference [76] , features of the meiotic cell cycle [77] , [78] , the strength of epistatic selection in haploid gametes [72] , and the genetic architecture of recombination [17] , [35] may play contributing roles . Few X-linked recombination rate modifiers have been previously identified [43] , [79] , but our recent findings suggest that sex-linked loci are pervasive components of the genetic architecture of recombination rate evolution in house mice [this study]; [ 48 , 80] . Further examination of the genetic basis of recombination rate should allow the relative importance of sex chromosome evolution and other causes of sexual dimorphism to be determined . Prdm9 is the only gene known to contribute to species differences in recombination rate . Human and chimpanzee alleles of Prdm9 are predicted to recognize and bind distinct DNA sequence motifs that may be important for recombination hotspot initiation [37] , [81] , [82] . These observations have led to the hypothesis that rapid evolution at Prdm9 underlies abrupt shifts in the distribution of recombination hotspots between species [37] , [81] . Although the CAST and PWD strains used in our study have different functional variants of Prdm9 [36] , we do not find QTL that co-localize with this gene . Prdm9 modifies the activity of multiple hotspots in mice [36] and in humans [38] , but it does not appear to have detectable effects on the global level of recombination in either species [this study; 37] . Taken together , these findings suggest that recombination rate evolution on fine and broad scales could be controlled by separate genes [83] . While examining the genetic control of hotspot activity can deliver mechanistic insights into recombination rate evolution , the total number of recombination events in a meiotically dividing cell – the phenotype examined here – is more likely to be a functionally relevant measure [84] . For example , female reproductive output is associated with global recombination rate in humans [15] , [16] , whereas the presence or absence of recombination activity in individual hotspots has yet to be linked to variation in fitness . In fact , the rapid evolutionary turnover of recombination hotspots within [25] , [26] , [39] , [85] , [86] and between species [23] , [24] seems to argue against a selective advantage of particular hotspot locations over others . In contrast , the genomic rate of recombination is subject to evolutionary constraints imposed by its essential functions in mammalian meiosis . A minimum of one crossover per chromosome is required for the proper disjunction of homologs in mice whereas high recombination rates may elevate the frequency of deleterious rearrangements [87] . Our study nominates eight genomic regions contributing to evolutionary divergence in genomic recombination rate . Future work will be required to determine whether the causal alleles are fixed or shared between M . m . musculus and M . m . castaneus . The observation that independent wild-derived inbred strains of M . m . musculus and M . m . castaneus conform to the recombination pattern established by PWD and CAST ( Figure 2 ) suggests that at least some of these QTL represent subspecies differences . The QTL identified in our analysis have broad peaks , each spanning a genomic interval that includes hundreds of genes . As a first step toward the identification of the causal variant ( s ) underneath the large X-linked QTL at 33 cM , we assayed transcript abundance between PWD and CAST alleles at 12 candidate genes . We found a suggestive difference in allele-specific expression at one gene , Brcc3 , a component of the BRCA1-BRCA2 complex involved in double-strand break repair ( Figure S1; Text S1; Table S1 ) [88] . These considerations nominate Brcc3 as a putative candidate gene for divergence in recombination rate . However , fine-mapping strategies will be required to test this hypothesis and to further localize the genetic changes that contribute to the increased global recombination rate in PWD . Genetic and ecological resources will facilitate the fine-mapping of QTL identified in our experimental intercross . The Collaborative Cross , an eight-way recombinant inbred line panel currently under development , includes inbred strains CAST and PWK [89] , a close relative of PWD . The increased mapping resolution and ability to measure mean MLH1 foci count on multiple animals with identical genotypes are key advantages of this resource that will aid efforts to fine-map those QTL that are common between PWK and PWD . In addition , populations of M . m . musculus and M . m . castaneus hybridize in nature , forming a fourth widely recognized subspecies of house mouse , M . m . molossinus [70] . The genetic properties of these natural hybrid populations are not well characterized , but lower levels of linkage disequilibrium could allow genomic windows containing causative loci to be narrowed through association studies or admixture mapping [90] . Identifying the determinants of the marked divergence in male recombination rate between M . m . musculus and M . m . castaneus promises to reveal the mechanisms of sex-limited evolution in this important phenotype .
Wild-derived inbred strains of Mus musculus castaneus ( CAST/EiJ ) and Mus musculus musculus ( PWD/PhJ and CZECHI/EiJ ) were purchased from the Jackson Laboratory ( Bar Harbor , Maine , USA ) and housed in the University of Wisconsin School of Medicine and Public Health mouse facility according to animal care protocols approved by the University of Wisconsin Animal Care and Use Committee . Pups were weaned into same-sex groupings at 21 days , with males subsequently separated into individual cages prior to 56 days . Animals were provided with food and water ad libitum . A total of 315 F2 males were sacrificed at 70 ( ±3 ) days of age ( 305 CAST/EiJ×PWD/PhJ and 10 PWD/PhJ×CAST/EiJ ) . Males from inbred strain CIM were purchased from Dr . Francois Bonhomme's stock repository at the Universite Montpellier II . Animals were sacrificed shortly after arrival to the University of Wisconsin-Madison ( aged 24 . 5–35 . 5 weeks ) . Spermatocyte spreads were prepared as described [91] . Briefly , the left testis of sexually mature males was removed , weighed , and rinsed in sterile 1× PBS . The outer tissue coating of the testis was punctured to allow a small volume of seminiferous tubules to be extracted . Tubules were incubated in a hypotonic solution ( 30 mM Tris , 50 mM sucrose , 17 mM citric acid , 5 mM EDTA , 2 . 5 mM dithiothreitol , 0 . 5 mM phenylmethanesulfonylfluoride ) for approximately 45 minutes at room temperature . Tubules were then transferred to a small volume ( 20 µl ) of 100 mM sucrose solution deposited on a clean glass slide and shredded using fine-gauge forceps . Tubular remnants were removed and an additional 20 µl of 100 mM sucrose added to the cell slurry . The solution was agitated by pipetting and 20 µl deposited onto each of 2 3×1″ glass slides coated with 100 µl 1% paraformaldehyde supplemented with TritonX-100 ( 0 . 15%; pH = 9 . 2 ) . The slides were gently rocked to distribute cells across their surface and allowed to dry overnight in a room temperature humid chamber . Dried slides were then washed briefly in 0 . 4% PhotoFlo ( Kodak ) , air dried , and subjected to immunostaining . The immunostaining protocol was adapted from Anderson et al . [45] and Koehler et al . [46] . A 10× concentration of antibody dilution buffer ( ADB ) was prepared ( 2 . 5 mL normal donkey serum ( Jackson ImmunoResearch ) , 22 . 5 mL 1× PBS , 0 . 75 g bovine serum albumin ( Fraction V; Fisher Scientific ) , and 12 . 5 µl TritonX-100 ) and sterilized by vacuum filtration ( 0 . 45 µm; Millipore ) . Slides were blocked in 1× ADB ( diluted in 1× PBS ) for approximately 30 minutes then lightly drained by touching the edge of the slide to a clean paper towel . All antibody dilutions were made into 1× ADB and all incubations were performed in a 37 C humid chamber . A 60 µl aliquot of primary antibody cocktail ( 1∶50 rabbit polyclonal antibody against MLH1 ( Calbiochem ) and 1∶50 goat polyclonal antibody against SYCP3 ( SantaCruz Biotechnology ) ) was dispensed on each slide . Slides were cover-slipped , sealed with rubber cement , and incubated overnight . The rubber cement was then carefully removed and coverslips were soaked off in 1× ADB . Slides were washed twice for 30 minutes each in 1× ADB . A 60 µl volume of 1∶100 Alexa 488 donkey anti-rabbit secondary antibody ( Molecular Probes ) was deposited on each slide . Slides were cover-slipped , sealed with rubber cement , and incubated overnight . After soaking off coverslips in 1× ADB , 60 µl of 1∶100 Alexa 568 donkey anti-goat secondary antibody ( Molecular Probes ) was applied to each slide . Slides were sealed with a parafilm “coverslip” and incubated for 2 hours . Slides were then washed three times for one hour each in 1× PBS , air-dried , and mounted in a drop of ProLong Gold antifade media ( Molecular Probes ) . Cells were visualized using a Zeiss Axioskop microscope equipped with an AxioCam HRc camera and a 100× objective lens . Late pachytene cells that were damaged during preparation , displayed bulbous chromosome termini ( indicative of transition into diplotene ) , lacked clear cell boundaries , or displayed flagrant defects in synapsis were not imaged . Images were captured in AxioVision ( Rel . 4 . 8 ) software and stored as moderate resolution tiff files . Images were subsequently cropped and the fluorescent intensity adjusted using ImageJ software . Numbers of autosomal MLH1 foci in late pachytene cells were manually scored . Only cells characterized by ( i ) the complete merger of SYCP3 signals from the two homologues , ( ii ) a full complement of chromosomes , ( iii ) clear , brightly stained MLH1 foci , and ( iv ) minimal background fluorescence were scored . We retained only cells with at least one MLH1 focus on each synaptonemal complex , excepting the possibility of one achiasmate bivalent; cells with more than two synaptonemal complexes lacking a MLH1 focus were extremely rare and likely represent staining artifacts . Approximately 20 cells were scored per animal . We were unable to obtain a sufficient number of high quality images for 39 of the 315 F2 animals . DNA from each F2 animal was extracted from liver tissue using a Wizard Genomic DNA Purification Kit ( Promega ) following manufacturer's protocols . 295 SNPs distinguishing PWD/PhJ and CAST/EiJ alleles were identified from Phase 4 of the Perlegen mouse resequencing project ( Frazer et al . 2007 ) and genotyped using the Sequenom iPLEX ( San Diego , CA ) MassARRAY system as previously described [92] . Raw genotype data were cleansed of putative genotype errors and non-Mendelian inheritances as described [80] . A total of 222 high quality SNPs , with an average call rate of 94 . 2% per SNP , were retained . A F2 genetic linkage map was constructed using the est . map function in the qtl add-on package for R [93] . Recombination fractions were converted to map distances using the Carter-Falconer mapping function [42] , [94] . We assumed no genotype error for map construction . Although a few base miscalls might have survived our data cleaning procedure , including a very small number of errors will have a negligible effect on map length estimation . Multiple QTL mapping was performed using the forward/backward model selection algorithm implemented in the R/qtl command stepwiseqtl . Model fitting was performed via extension of Haley-Knott regression [95] , with genotype probabilities calculated along a 1 cM grid . Model comparisons were conducted using a penalized LOD score approach with penalties calculated from 1000 permutations of the data [93] . Because biases may be introduced in the stepwise addition of new QTL to the model [93] , we repeated the model search multiple times . Each search converged on an identical model of eight QTL and zero epistatic interactions . | Homologous recombination is an indispensable feature of the mammalian meiotic program and an important mechanism for creating genetic diversity . Despite its central significance , recombination rates vary markedly between species and among individuals . Although recent studies have begun to unravel the genetic basis of recombination rate variation within populations , the genetic mechanisms of species divergence in recombination rate remain poorly characterized . In this study , we show that two closely related house mouse subspecies differ in their genomic recombination rates by ∼30% , providing an excellent model system for studying evolutionary divergence in this trait . Using quantitative genetic methods , we identify eight genomic regions that contribute to divergence in global recombination rate between these subspecies , including large effect loci and multiple loci on the X-chromosome . Our study uncovers novel genomic loci contributing to species divergence in global recombination rate and offers simple genetic explanations for rapid phenotypic divergence in this trait . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"linkage",
"(genetics)",
"trait",
"locus",
"phenotypes",
"heredity",
"genetics",
"biology",
"evolutionary",
"biology",
"quantitative",
"traits",
"cytogenetics",
"evolutionary",
"genetics",
"genetics",
"and",
"genomics",
"complex",
"traits",
"genotypes"
]
| 2011 | Genetic Analysis of Genome-Scale Recombination Rate Evolution in House Mice |
Buruli ulcer is a stigmatising disease treated with antibiotics and wound care , and sometimes surgical intervention is necessary . Permanent limitations in daily activities are a common long term consequence . It is unknown to what extent patients perceive problems in participation in social activities . The psychometric properties of the Participation Scale used in other disabling diseases , such as leprosy , was assessed for use in former Buruli ulcer patients . Former Buruli ulcer patients in Ghana and Benin , their relatives , and healthy community controls were interviewed using the Participation Scale , Buruli Ulcer Functional Limitation Score , and the Explanatory Model Interview Catalogue to measure stigma . The Participation Scale was tested for the following psychometric properties: discrimination , floor and ceiling effects , internal consistency , inter-item correlation , item-total correlation and construct validity . In total 386 participants ( 143 former Buruli ulcer patients with their relatives ( 137 ) and 106 community controls ) were included in the study . The Participation Scale displayed good discrimination between former Buruli ulcer patients and healthy community controls . No floor and ceiling effects were found . Internal consistency ( Cronbach's alpha ) was 0 . 88 . In Ghana , mean inter-item correlation of 0 . 29 and item-total correlations ranging from 0 . 10 to 0 . 69 were found while in Benin , a mean inter-item correlation of 0 . 28 was reported with item-total correlations ranging from −0 . 08 to 0 . 79 . With respect to construct validity , 4 out of 6 hypotheses were not rejected , though correlations between various constructs differed between countries . The results indicate the Participation Scale has acceptable psychometric properties and can be used for Buruli ulcer patients in Ghana and Benin . Future studies can use this Participation Scale to evaluate the long term restrictions in participation in daily social activities of former BU patients .
Buruli ulcer ( BU ) is a neglected tropical disease caused by Mycobacterium ulcerans ( MU ) and has been reported in more than 30 countries predominantly in tropical or subtropical areas [1] . Most burdened region is West Africa of which Côte d'Ivoire , Ghana and Benin have reported large number of cases in 2012 [2] . BU presents as a small nodule or a plaque sometimes accompanied with edema . Later , the lesion breaks open , and a characteristic lesion presenting with undermined edges appears [3] . BU affects skin , subcutaneous tissue , muscles and sometimes bone resulting in scarring , deformities , and contractures; sometimes , amputation is necessary [4] , [5] . This in turn may lead to long-term physical disability such as restriction in movement of joints [6] and functional limitation [7] . Apart from these problems , stigmatisation due to superstitious beliefs or inappropriate or delayed treatment may lead to considerable impact on functioning in social life [8] , [9] , such as change in occupation , unemployment , school dropout , and economic burden [10] , [11] . The World Health Organization's ( WHO ) model of International Classification of Functioning , Disability and Health ( ICF ) [12] describes disabilities in terms of impairments , activity limitations and participation restrictions . Participation restrictions are being viewed in this model through a social perspective and are defined as; ‘any problem an individual may experience in involvement in life situations’ [12] . Visible signs , activity problems and stigma are associated with an increased risk for participation restrictions [13] , [14] . Various instruments have been developed to measure participation restrictions for populations in affluent and industrialized countries . The Participation Scale ( P-scale ) designed to measure the severity of participation restrictions as perceived among persons with disabilities , is the only questionnaire developed for the context of low and middle-income countries [15] . The P-scale has been applied to study the psychometric properties among persons with leprosy , poliomyelitis and spinal cord injuries aged at least 15 years [15] . The instrument is considered culturally free , generic , easy to administer by local staff , and has been recommended for patients with BU [15] , however properties in such patients have not yet been tested . Therefore , the aim of this study was to test the psychometric properties of the P-scale among former BU patients in Ghana and Benin .
Former patients with BU were identified using medical records kept by the Agogo Presbyterian Hospital in Ghana and the Centre de Dépistage et de Traitement de l'Ulcère de Buruli in Lalo , Benin . These patients were clinically diagnosed as BU patients in accordance with the WHO case definition [16] . Participants had to be at least 15 years; with treatment for BU between 2005–2011 to be completed at least 3 months before the start of the current study . We also included at least 50 healthy community controls in each of the two countries to test the discriminative potential of the P-scale . To strive for an equal distribution in patients and controls regarding location , age and sex , healthy community controls were recruited from villages located in the study area; we attempted to have these control participants match with our former patients for age ( +5/−5 years ) and sex . Finally , former patients with BU were asked to help identify a relative to be approached who could be included in the study to act as representative to be able to test one of the hypothesis for the construct validity of the P-scale . The English P-scale ( version 6 . 0 ) was translated into Twi ( one of the local languages in Ghana ) and French ( Benin ) and back translated . To ensure correct translation into the local language in Benin , Fon , an interview conducted by a native speaker was recorded and back translated by another native speaker without knowledge of the initial French version of the P-scale . Thereafter the translated , back translated and initial versions were compared . Each item was checked for the correct words used that were appropriate for the target group and whether the meaning of the question was maintained . Both in Ghana and Benin , pilot studies were conducted to study understandability of the wording , the question and the response options . Examples not suitable for the West African context , like bazaars , or tea/coffee shops were removed . In both countries , face validity of the P-scale was established among medical doctors , nurses , a physical therapist , a social scientist and a BU coordinator . The conceptualization of the construct participation restriction , the relevance and importance of each item , the understandability of the peer concept , and the response option for the target population was discussed . In both countries , two native language speaking interviewers participated in a training on the objective of the interview , and conducting the interviews in accordance with the available manuals , the P-scale Users Manual ( version 6 . 0 ) and BUFLS manual ( 2012 ) to prevent biases during interviewing . The P-scale was adapted for relatives . Each question was adapted in such a way that the relative could answer for the former BU patient . For example the first question of the P-scale was changed into; Does he/she have equal opportunity as his/her peers to find work ? How big is that problem to him/her ? To establish discrimination , it was determined a significant difference ( α value of 0 . 05 ) in P-scale sum scores between former patients with BU and healthy controls indicates the instrument has discriminative value . Floor and ceiling effects were determined to be present when ≥15% of the respondents scored the lowest ( 0 ) or the highest ( 90 ) possible score on the P-scale . A Cronbach's alpha was calculated to determine internal consistency . A value of ≥0 . 70 was considered sufficient [24] . To determine construct validity , hypotheses were formulated a priori accounting for both countries together . To positively rate construct validity ≥75% ( 5 out of the 6 ) of the a priori hypothesis need to be confirmed [24] . The hypotheses formulated are: The Medical Ethical Review Committees of the Kwame Nkrumah University of Science and Technology; School of Medical Sciences , Komfo Anokye Teaching Hospital in Ghana ( ref: CHRPE/RC/127/12 ) and Ministry of Health in Benin ( ref: N01961/MS/DC/SGM/DRF/SRAO/SA ) approved the study . Before the interview started , the aim of the interview was explained and written informed consent was obtained from the participants . The participants were informed on the voluntary participation and confidentiality of the study . Data were analyzed using Statistical Package for the Social Sciences ( SPSS ) 20 . 0 . To correct for missing values of P-scale questions of former BU patients , imputation using individual mean subscale scores was used . In case a missing value represented a 1-item subscale , the total mean score was calculated to fill out the missing value . Non-parametric analysis was computed as all questionnaire data was positively skewed . The Mann-Whitney U test or Chi-square was used to analyze possible differences in clinical signs , socio-economic , and demographic factors between countries and P-scale scores . To control for differences across countries , analysis for Ghana and Benin were performed separately . When striking differences between countries were not found analysis was performed taking both countries together . Cronbach's alpha , inter-item correlation and item-total correlation were used to analyze internal consistency . Spearman's rho correlations and confidence intervals were used to test the associations between the P-scale and the scores of the BUFLS , EMIC and P-scale administered among relatives . Calculation of confidence intervals into z-scores allowed interpretation as effect sizes [25] to study differences across countries . An effect size of 0 . 10 is considered small with a negligible practical importance , an effect size of 0 . 30 is considered medium with a moderate practical importance , and an effect size of 0 . 50 is considered large and of crucial importance [25] . Mann-Whitney U test was computed to analyze differences in P-scale scores between former patients with BU and healthy controls .
Age and sex did not differ significantly between former patients with BU and healthy controls ( Table 1 ) . P-scale scores among former patients with BU ( median 13 , IQR; 4;29 ) were significantly higher ( P< . 001 ) than that of healthy controls ( median 2 , IQR; 0;6 ) . Similar results were found when looking at discriminative value of the P-scale in both countries . Overall , no floor and ceiling effects were found as 6% scored the lowest possible sum score ( 0 ) and none of the participants scored the highest possible ( 90 ) sum score . The median scores of the separate items according to the different subscales of the P-scale across countries are shown in Table 2 . For Ghana the mean inter-item correlation was 0 . 29 ( range −0 . 13–0 . 71 ) while for Benin this was 0 . 28 ( range −0 . 22–0 . 95 ) . In Ghana , lowest item-total correlation ( 0 . 10 ) was found for the item ‘confident to try to learn new things’ of and highest item-total correlation ( 0 . 69 ) was found with ‘work hard as your peers do’ . In Benin , lowest item-total correlation ( −0 . 08 ) was found with ‘comfortable meeting new people’ and highest ( 0 . 79 ) was found with ‘mobility house/village as other people’ . The 18 item P-scale had a Cronbach's alpha of 0 . 88 ( 124 participants ) . A Cronbach's alpha of 0 . 88 was found among participants in Ghana ( n = 57 ) and was similar ( 0 . 87 ) in Benin ( n = 67 ) . Hypothesis 1: P-scale scores of former patients with BU with visible deformities ( n = 18 , median 19 , IQR; 11; 45 ) were significantly ( P = . 023 ) higher than those without visible deformities ( n = 84 , median 8 . 5 , IQR; 3; 25 ) . ( Hypothesis accepted ) . Hypothesis 2: P-scale scores of former patients with BU with joint involvement ( n = 51 , median 18 , IQR; 8; 32 ) were nearly significantly ( P = . 056 ) higher than those without joint involvement ( n = 80 , median 9 . 5 , IQR; 3; 27 ) . ( Hypothesis rejected ) . Hypothesis 3: P-scale scores of former patients with BU that changed occupation were higher ( n = 8 , median 16 . 5 , IQR; 5; 38 ) compared to those that continued the same occupation ( n = 117 , median 13 , IQR; 5; 29 ) but this difference did not reach statistical significance ( P = . 809 ) . ( Hypothesis rejected ) . Hypothesis 4: A Spearman's rank correlation of 0 . 67 was found between the BUFLS and P-scale scores . ( Hypothesis accepted ) . Hypothesis 5: A Spearman's rank correlation of 0 . 53 was found between the EMIC and the P-scale . ( Hypothesis accepted ) . Hypothesis 6: A Spearman's rank correlation of 0 . 80 was found between the P-scale sum scores of former BU patients and relatives . ( Hypothesis accepted ) . The associations between functional limitations , perceived stigma and participation restrictions separately for Ghana and Benin are depicted in Table 3 . A small effect size [25] was found with respect to functional limitation and participation restrictions implying that associations between constructs are almost similar across Ghana and Benin . When looking at the difference in association between perceived stigma and participation restrictions in both countries , a medium effect size appeared meaning that the difference in association between both countries is of moderate importance . Finally , the association between the participation score as answered by a relative of the patient and the patient differed largely between Ghana and Benin . In Ghana a Spearman's rank correlation of 0 . 70 was found between the P-scale sum scores of former BU patients and relatives , while in Benin a Spearman's rank correlation of 0 . 92 was found between the P-scale sum scores of former BU patients and relatives .
This study aimed to test the psychometric properties of the P-scale among former patients with BU in Ghana and Benin . The findings indicate acceptable psychometric properties for assessing participation restrictions in former patients with BU aged at least 15 years . Findings revealed good discriminative potential between former BU patients and controls , which are similar to those in the initial development study of the P-scale of van Brakel et al . ( 2006 ) . Internal consistency was good and consistent with those found in previous studies [13] , [17] , [18] . Floor and ceiling effects were absent , as in earlier studies [13] , [17] . Our study confirmed 4 out of the 6 hypotheses . Moderate to strong relations were found with functional limitation , stigma and participation problems indicated by relatives . Associations with functional limitation and stigma are corresponding with associations previously found [13] , [14] . When looking at country level , similar associations between functional limitations and participation restrictions emerged . However correlations between perceived stigma and participation restrictions were moderately different , and the association between relatives and patients was strikingly different . It is possible that translation of the instrument and observer differences could have resulted in distinct associations in our study . However , in both countries , after translation and back-translation of the P-scale , the correct formulation of the content of the questions was discussed extensively . In addition , in Benin and Ghana a pilot study was conducted to ensure reliability between the trained native-speaking interviewers and interviews were reviewed regularly during data collection . However the equivalence of pervious scores and scores in this study should be compared and a factor analysis should be performed to assess measurement invariance which we initially planned , but the sample sizes of both groups ( Ghana and Benin ) were too small . It is also plausible that the experience of participation restrictions and stigma may be different in Ghana and Benin due to cultural differences , and that evaluation of relatives regarding the experience of participation restrictions for former patients with BU could have been interpreted differently due to differences in type of relatives in Ghana and Benin or local cultural differences . The findings of this study indicate the P-scale has acceptable psychometric properties and can be used for BU patients in Ghana and Benin , but the cross-cultural applicability of the instrument should deserve further study . Recently , it was also recommended that the P-scale should be studied more extensively on cultural validity in a new cultural context [26] . In addition future studies should focus on psychometric testing using pre-post scores to provide information regarding the responsiveness of the P-scale , test-retest reliability and intra- rater and inter-rater reliability . Since many of the BU patients are children , further research is needed in order to also develop a suitable instrument for children up to 15 years old measuring participation restrictions and explore the extent of participation problems among these children . Gathering knowledge on participation problems among BU patients is essential in order to be able to develop a suitable intervention . | Buruli ulcer is a stigmatising condition caused by infection with Mycobacterium ulcerans . Besides the long term medical consequences , Buruli ulcer may lead to participation restrictions in social life . The Participation Scale intends to assess perceived participation restrictions; however , this instrument has been developed in patients affected by leprosy and other disabling conditions , and has never been used before among Buruli ulcer patients . We aimed to analyze the reliability and validity of the Participation Scale among former Buruli ulcer patients in Ghana and Benin . This study included former Buruli ulcer patients from 2 different treatment sites , along with their relatives and healthy community controls residing in similar geographical areas . Former Buruli ulcer patients were interviewed using the Participation Scale , Buruli Ulcer Functional Limitation Score , and the Explanatory Model Interview Catalogue to measure stigma . Relatives and healthy community controls were interviewed using the Participation Scale . We tested the Participation Scale for discrimination , floor and ceiling effects , internal consistency , inter-item correlation , item-total correlation and construct validity . The results of the analysis suggest that the Participation Scale has acceptable psychometric properties . As such , the instrument can be used to assess participation restrictions among former Buruli ulcer patients in Ghana and Benin . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"sociology",
"social",
"sciences",
"physiotherapy",
"health",
"care"
]
| 2014 | Psychometric Properties of the Participation Scale among Former Buruli Ulcer Patients in Ghana and Benin |
A vaccine to prevent infection and disease caused by Plasmodium vivax is needed both to reduce the morbidity caused by this parasite and as a key component in efforts to eradicate malaria worldwide . Vivax malaria protein 1 ( VMP001 ) , a novel chimeric protein that incorporates the amino- and carboxy- terminal regions of the circumsporozoite protein ( CSP ) and a truncated repeat region that contains repeat sequences from both the VK210 ( type 1 ) and the VK247 ( type 2 ) parasites , was developed as a vaccine candidate for global use . We conducted a first-in-human Phase 1 dose escalation vaccine study with controlled human malaria infection ( CHMI ) of VMP001 formulated in the GSK Adjuvant System AS01B . A total of 30 volunteers divided into 3 groups ( 10 per group ) were given 3 intramuscular injections of 15μg , 30μg , or 60μg respectively of VMP001 , all formulated in 500μL of AS01B at each immunization . All vaccinated volunteers participated in a P . vivax CHMI 14 days following the third immunization . Six non-vaccinated subjects served as infectivity controls . The vaccine was shown to be well tolerated and immunogenic . All volunteers generated robust humoral and cellular immune responses to the vaccine antigen . Vaccination did not induce sterile protection; however , a small but significant delay in time to parasitemia was seen in 59% of vaccinated subjects compared to the control group . An association was identified between levels of anti-type 1 repeat antibodies and prepatent period . This trial was the first to assess the efficacy of a P . vivax CSP vaccine candidate by CHMI . The association of type 1 repeat-specific antibody responses with delay in the prepatency period suggests that augmenting the immune responses to this domain may improve strain-specific vaccine efficacy . The availability of a P . vivax CHMI model will accelerate the process of P . vivax vaccine development , allowing better selection of candidate vaccines for advancement to field trials .
Malaria is a devastating parasitic disease transmitted through the bite of infected Anopheles mosquitoes . Outside sub-Saharan Africa , Plasmodium vivax is the most prevalent of all human malarias with approximately 2 . 48 billion people at risk [1] and an estimated 16 million cases in 2013 ( WHO World Malaria Report , 2014 ) . Unlike Plasmodium falciparum , P . vivax produces liver stages ( hypnozoites ) that , initially dormant , can reactivate several weeks to months after the primary infection causing symptomatic disease [2 , 3] . This propensity to relapse stands as a significant barrier to efforts to eradicate this species of malaria [3] . Additionally , P . vivax is increasingly reported as the causative agent of symptoms associated with severe malaria as well as chloroquine resistance [4–7] . A vaccine to prevent infection and disease caused by P . vivax is urgently needed to reduce morbidity of the disease and accelerate elimination of this parasite . The circumsporozoite protein ( CSP ) is the most abundant sporozoite protein present on the sporozoites of all Plasmodium species and has been shown to have great potential as a vaccine target [8 , 9] . Antibodies to the repeat region of P . falciparum CSP have been shown to be associated with protection [10–12] . Unlike P . falciparum , the repeat region of P . vivax CSP exhibits sequence heterogeneity resulting in immunologically distinct populations indicating that a vaccine based on one strain may not be sufficient to protect against all circulating strains [13] . To take into account the diversity of P . vivax strains , we developed vivax malaria protein 001 ( VMP001 ) as a candidate vaccine for P . vivax malaria . The vaccine antigen VMP001 is an Escherichia coli produced synthetic chimeric recombinant protein that incorporates the three major domains of CSP but is distinct from the native molecule [14 , 15] . This synthetic construct includes the amino ( N- ) and carboxy ( C- ) terminal parts of CSP and a truncated repeat region that contains repeat sequences from the immunologically divergent VK210 ( type 1 ) and the VK247 ( type 2 ) strains of parasites . The VMP001 antigen was adjuvanted with AS01B , a proprietary liposome-based adjuvant system from GSK Biologicals that contains the immunostimulants monophosphoryl lipid A ( MPL ) and QS-21 , a triterpene glycoside purified from the bark of Quillaja saponaria [9] . This adjuvant system has been used in other malaria vaccine candidates , including RTS , S [9] . We report the results of a first in humans phase 1 clinical trial using VMP001/AS01B in terms of reactogenicity , immunogenicity , and efficacy against a P . vivax sporozoite challenge in healthy , malaria-naive adults .
The study ( ClinicalTrials . gov identifier NCT01157897 ) , sponsored by the Office of the Surgeon General , U . S . Army , was conducted following scientific and ethical review by the WRAIR scientific review committee , WRAIR institutional review board ( IRB ) , the USAMRMC’s Human Subjects Research and Review Board as well as the Western IRB and assigned protocol numbers WRAIR 1692 , HRPO A-16037 . The protocol was conducted under the U . S . Food and Drug Administration ( FDA ) Investigational New Drug ( IND ) application #14380 . Written informed consent was obtained from all volunteers prior to screening and enrollment . Subjects were healthy malaria naïve men and women aged 18–55 years . All subjects had normal blood levels of glucose-6-phosphate dehydrogenase ( G6PD ) , and were either homozygous or heterozygous positive for Duffy antigen . Subjects agreed to be available for the duration of study with no plans to travel to a malaria endemic area or outside the Washington , DC area until a treatment course was completed following CHMI . This was a phase 1 , non-randomized , open label , dose-escalation study in 36 adults . Thirty volunteers , divided into 3 cohorts ( 10 in each group ) , were vaccinated with 3 doses of VMP001/AS01B . Cohorts 1 , 2 , and 3 received 15 μg , 30 μg , or 60 μg , respectively , of VMP001in 500 μL of AS01B at each immunization . The first and second immunizations in each cohort were separated by 28 days , and the third dose for all cohorts was normalized such that the interval between the last immunization and day of challenge was 2 weeks ( Fig 1 ) . Controlled human malaria infection ( CHMI ) using P . vivax infected Anopheles dirus mosquitoes was performed in the volunteers from all cohorts that completed all 3 immunizations ( n = 27 ) and a control group ( n = 6 ) who were not administered investigational product ( Figs 1 and 2 ) . The VMP001 antigen [14 , 15] and the AS01B adjuvant system [9] have been described previously . The VMP001 recombinant subunit protein was produced in and purified from E . coli [15] and reconstituted in 500 μl of AS01B . AS01B is an Adjuvant System containing 50 μg 3-O-desacyl-4’- monophosphoryl lipid A ( MPL , produced by GSK ) and 50 μg Quillaja saponaria Molina , fraction 21 ( QS-21 , licensed by GSK from Antigenics Inc , a wholly owned subsidiary of Agenus Inc . , a Delaware , USA corporation ) , in a liposomal formulation [9] . Lyophilized VMP001 was reconstituted with liquid AS01B and administered in doses of 15 μg , 30 μg , or 60 μg ( 500 μl/dose ) by slow intramuscular injection within 1 hour of reconstitution . Vaccine tolerability was assessed by evaluating local reactogenicity and systemic symptoms , as well as changes in biochemical and hematologic laboratory parameters . Clinical assessments of subjects were performed 30 min after each vaccination and again 1 , 2 , and 6 days later . Local and systemic solicited adverse events ( AEs ) were collected during this 7 day period . Severity of adverse reactions was classified as grade 1 ( mild ) , grade 2 ( moderate ) and grade 3 ( severe ) . Unsolicited AEs were recorded during the 28 days following vaccinations 1 and 2 and during the 14 days after vaccination 3 and reported to a safety monitoring committee ( SMC ) . Biochemical and hematologic laboratory parameters were measured prior to administration and 6 days after each vaccination . Complete blood cell counts with white blood cell differential counts were performed in addition to serum levels of creatinine , alanine aminotransferase , and aspartate aminotransferase . During the challenge phase , solicited AEs were recorded beginning on the day of CHMI . Clinical assessments continued on days 1 and 3 post-CHMI , and then daily beginning on day 5 until the subject had 3 consecutive negative daily blood smears following the initiation of treatment for malaria infection . All unsolicited events were documented for 28 days beginning on the day of CMHI . Safety laboratories were measured on the day of CHMI , at the time of initiating anti-malaria therapy , and on days 28 and 42 post-CMHI . Serious AEs ( SAEs ) were reported from the time of subject enrollment until study closure . Episodes of P . vivax relapse were recorded up to the conclusion of the study period , approximately 5 years post-CHMI .
This study was conducted at the WRAIR Clinical Trials Center . A total of 30 subjects received at least one vaccination . Of those , 27 subjects that completed the three dose vaccination regimen , and 6 infectivity control subjects were challenged by the bites of five P . vivax infected mosquitoes . ( Figs 1 and 2 ) . Immunizations of subjects in all three dose cohorts were well tolerated and no safety halting criteria were met . There were no clinically concerning imbalances observed between groups . Similar frequencies of solicited local and solicited general AEs were reported in the three cohorts ( Fig 3 ) . There were no AEs in any vaccination cohort that led to withdrawal of any subjects . One SAE ( ductal carcinoma in situ of the breast ) occurred during the study and it was determined that the event was not related to the study vaccine or CHMI . Mild ( grade 1 ) to moderate ( grade 2 ) pain in the days following immunization was the most frequently reported local solicited AE , occurring in 100% of subjects who received at least one vaccination . No subjects experienced severe ( grade 3 ) pain . Grade 3 solicited local AEs included erythema at the injection site in 4 subjects , 1 in cohort 2 and 3 in cohort 3 . The grade 3 erythema lasted for ≤2266 3 days in 3 of the subjects and in 7 days for one subject . Erythema was not associated with more severe pain or any functional impairment . Fatigue and headache were the most common solicited systemic AEs following each vaccination , increasing in frequency from vaccination 1 to vaccination 3 . The number of subjects experiencing fatigue and headache increased from 17% to 52% and 17% to 45% , respectively , following vaccination 1 to vaccination 3 . Myalgia and arthralgia occurred in their highest frequency following vaccination 3 , occurring in 41% and 21% of subjects , respectively . Gastrointestinal AEs , to include nausea , diarrhea and abdominal pain , ( 19% ) and fever ( 10% ) were also recorded most frequently following vaccination 3 . All other systemic AEs occurred in 10% or less of the subjects following any vaccination . All solicited systemic AEs resolved within the 7-day follow-up period . There was one episode of fever that met the criteria for a grade 3 solicited systemic AE which resolved in less than 24 h . There appeared to be a trend towards increased numbers of solicited AEs associated with each subsequent vaccination . Throughout the vaccination period , eleven mild ( grade 1 ) biochemical or hematologic laboratory adverse events were documented , none of which were determined to be related to vaccination . There were no grade 2 or 3 laboratory abnormalities during this timeframe . This study represents only the second site to implement a P . vivax CHMI model which , unlike P . falciparum , requires mosquitoes that have been fed on blood directly obtained from an infected human donor rather than from in vitro cultured parasites . To ensure subject safety , donor blood was subjected to transfusion-grade screening for blood-borne infections as well as vector-borne infections . All study volunteers were screened to ensure they had normal concentrations of G6PD to prevent hemolytic anemia during radical cure therapy with primaquine . The challenge was well tolerated with no untoward reactogenicity following mosquito bites and 100% of the subjects that were exposed became parasitemic . No untoward SAEs were observed following the challenge and treatment phases . On days 65 and 79 days post-CHMI , 2 subjects experienced P . vivax relapse [17] which we hypothesize was associated with their inability to metabolize PQ into sufficiently high enough concentrations of its active form leading to drug failure . One subject experienced a total of two relapses while the second subject experienced three relapses . No further relapses were observed up to the end of the study period , approximately 5 years post CHMI . Seroconversion to VMP001 following the second vaccination was set as the progression criteria for proceeding to CHMI . Subjects were considered to have seroconverted if , at a serum dilution of 1:100 , the OD414 of the test sample obtained two weeks post-2nd immunization was significantly different by paired t-test compared to serum obtained prior to 1st immunization ( data not shown ) . All subjects seroconverted , with anti-VMP001 antibodies detectable in 100% of vaccinees at 2 weeks post vaccine dose 2 . Having met the progression criteria , subsequent ELISA data was reported as antibody titer , defined as the reciprocal of the serum dilution giving an OD414 of 1 . 0 , and responses were measured at 2 weeks following each immunization as well as at 1 and 6 months post challenge . The highest geometric mean titer ( GMT ) of anti-VMP001 antibody were noted 2 weeks after dose 2 in cohorts 2 and 3 ( 74 , 608 and 61 , 711 , respectively ) and on the day of challenge ( DOC; 2 weeks after dose 3 ) in cohort 1 ( 61 , 203 ) ( Fig 4 ) . Peak GMTs of anti-VMP001 antibody were not significantly different between the groups . While there was a decrease in antibody titers in all three cohorts in the weeks following the 2nd immunization , titers were boosted minimally ( for cohort 3 ) to significantly ( for cohort 1 ) , following the 3rd immunization . As a result , GMTs were not statistically significantly different between cohorts 1 , 2 , and 3 on the DOC ( Fig 4 ) . Antibody titers showed a 5 to 8-fold decrease 6 months post-challenge compared to the pre-challenge titers . However , these titers remained significantly higher than those measured following the 1st immunization . Antibody fine-specificities were evaluated to determine if the antibody responses were skewed to any specific region of the protein . Antibodies were detected to all regions of the protein , i . e . to the N-terminal , central repeat region as well as the C-terminal region . There were no significant differences in the GMT of anti-C term or anti-N term antibody between any groups on the DOC . The DOC GMT of anti-Type 1 repeat antibody was significantly higher in group 2 ( GMT 16 , 554 ) compared to groups 1 and 3 ( GMT 4 , 412 and 5 , 922 respectively ) ( Fig 5 ) . There were no significant differences in GMT of anti-Type 1 repeat antibody between groups 1 and 3 on the DOC . CD4+ T cell responses to VMP001 were detected in all individuals , with a majority ( 60% ) showing peak response 14 days after the 2nd immunization . All vaccinated individuals produced IL-2 , 93% produced TNF-α and 55% produced IFN-γ following stimulation with VMP001 . Cytokine positive cells were predominantly IL2+ single positive or IL2+TNF-α+ double positive ( Fig 6 ) . Smaller frequencies of triple positive cells that also expressed IFN-γ were also detected ( Fig 6 ) . Cytokine profiles did not show marked differences post challenge . The responses were predominantly directed towards the N-term region ( 90% volunteers ) . Only 17% of vaccinated subjects responded to the repeat and C-term regions . There were no detectable CD8+ T cell responses in any volunteer at any of the time-points tested . Following CHMI , all infectivity control subjects ( 100% ) became infected . All 27 ( 100% ) immunized subjects who underwent CHMI also became infected ( Fig 7 ) . Vaccine efficacy was 0% . The median prepatent period of all immunized subjects was 11 . 9 days and that of infectivity controls was 10 . 7 days . A significant delay in the median prepatency period was noted in all cohorts as compared to the infectivity controls ( cohort 1 , 10 . 9 days , p = 0 . 0261; cohort 2 , 12 . 6 days , p<0 . 0001; cohort 3 , 12 . 4 days , p = 0 . 0003 ) . The median prepatency period for cohort 3 was significantly longer than that of cohort 1 ( p = 0 . 035; Log-rank test ) ; otherwise , there was no statistically significant delay in median prepatency periods between the other vaccinated groups . Vaccinated subjects with a prepatency period greater than 2 standard deviations above the mean of the prepatency period of the infectivity controls were categorized as having a significant delay in the onset of parasitemia . The median time to parasitemia for the delayed group ( n = 16 ) was 12 . 8 days and 10 . 9 days in the subjects without delay ( n = 11 ) ( Fig 8A ) . On the day of the CHMI there was a significant correlation ( r = 0 . 51 , p = 0 . 006 ) in all vaccinated subjects between anti-type 1 peptide antibody titer and net time to parasitemia ( time to parasitemia of vaccinated subjects—time to parasitemia of control subjects ) . The GMT of anti-type 1 repeat antibody was significantly greater in the delayed group ( 10 , 489 , 95% CI = 6 , 154–17 , 878 ) as compared to the group who did not experience delay ( 4 , 279 , 95% CI = 2 , 684–6 , 822 ) ( p = 0 . 025 , 2-tailed Mann Whitney t-test ) ( Fig 8B ) . There were no other significant differences in the GMT of anti-VMP001 , anti-N term , or anti-C term antibody between the delayed and no delay groups . No significant differences were identified between the cohorts in terms of CMI response at any of the time points and there was no association between CMI and delay in prepatency period .
The development of a controlled human malaria infection challenge model to evaluate P . vivax vaccines is challenging and fraught with numerous technical obstacles . While P . falciparum has been adapted for use in CHMI primarily because of the ability to grow mature gametocytes that infect Anopheles mosquitoes to produce infectious stage sporozoites , the same methodology cannot be used for P . vivax . To date , P . vivax has proven refractory to continuous in vitro culture . Therefore , infection of permissive Anopheline mosquito species ( i . e . An . dirus and An . albimanus ) —that themselves are difficult to maintain and/or infect in insectaries—relies on the identification and consent of naturally-infected patients to present to medical treatment facilities to donate blood for initiating infection in mosquitoes . Nevertheless , the successful implementation of P . vivax CHMI previously reported in Colombia has opened the door to test vaccine efficacy by incorporating an infected mosquito challenge [19] . In preparation for conducting a vaccine efficacy study we established a CHMI model for P . vivax at WRAIR , in collaboration with AFRIMS , using mosquitoes that were infected in Thailand ( Chuang et al . in preparation . Here we report the initial safety , immunogenicity , and efficacy data for the candidate vaccine VMP001/AS01B . This represents the first human efficacy study incorporating a CHMI for any P . vivax vaccine . This vaccine formulation , administered in 3 increasing antigen doses ( 15 μg , 30 μg , or 60 μg ) was well tolerated and the adverse event profile was consistent with other vaccines containing the AS01B adjuvant system [10 , 20 , 21] . The vaccine induced strong antibody and CD4+ T cell immune responses in all antigen dose groups . While all groups demonstrated greater than 50-fold boosting in antibody titers between the first and second immunization , only the low dose cohort showed a modest 2 . 8-fold increase in titers between the second and third immunization , thus matching the titers of the other two cohorts , which did not show any boost in antibody titers post third immunization . A possible explanation for the absence of antibody boosting following the second vaccination in the medium- and high-dose cohorts could be that the antibody titers were already saturated for these cohorts and either the vaccine dose or the range of intervals between the 2nd and 3rd vaccination ( 6 and 4 weeks , respectively ) did not allow for sufficient enhancement in titers . No subjects were sterilely protected following CHMI; however , all dose groups experienced a statistically significant delay in mean prepatency period as compared to the infectivity control group suggesting an anti-parasite effect elicited by the vaccine . A two day delay in prepatent period reflects a significant decrease in liver-stage parasites [22] [23] , indicating that the immune responses generated by the vaccine is able to induce partial protection in vaccinated subjects . This would be consistent with the previously described correlation of high P . falciparum CSP repeat-specific antibody titer with sterile protection in malaria naïve adults [10] and children in endemic regions [11 , 12] . We have previously reported an efficacy study in Aotus monkeys that were immunized with VMP001 formulated in Montanide ISA 720 and CpG . Following an intravenous challenge a vaccine efficacy of 66 . 7% was observed and this protection was associated with anti-type 1 antibodies [24] . This observation supports the results observed in the current study . The lack of protection in humans may be due to a lower magnitude of anti-type 1 antibody titers in comparison to those observed in the Aotus study which was conducted with a different adjuvant formulation ( Yadava , A . manuscript in preparation ) . As we consider strategies to improve vaccine efficacy , alternate approaches , such as particulate delivery to improve immunogenicity and epitope-display , and well as alterations in schedule and dosing to improve qualitative and quantitative responses are points to ponder . We have developed CSV-S , S , a particulate formulation containing VMP001 which , similar to the P . falciparum CSP-based vaccine RTS , S , is co-expressed as a hepatitis B fusion particle . Analysis of the fine specificity of antibody responses in serum from rhesus monkeys immunized with CSV-S , S demonstrated significantly higher antibody response to the type 1 repeat peptide as well as greater responses to a smaller AGDR sequence within the type 1 peptide , suggesting that a particle formulation may improve the humoral response to the repeat sequences over soluble protein alone in the presence of adjuvants that are appropriate for human use [16] . The significant association of type 1 repeat-specific antibody responses with delay in prepatency period suggests that new vaccine strategies that enhance immune responses to this region might further improve vaccine efficacy against these strains of P . vivax . In addition to the modulation of immune responses to the repeat region by particulate formulations , an alternate strategy is to increase the number of repeat motifs in the vaccine construct . The resulting increase in epitope density may result in enhanced anti- repeat-specific responses . Finally , alterations in schedule and dosing to optimize antibody affinity to protective epitopes may improve vaccine efficacy as has been reported for the RTS , S vaccine ( Regules et al . in preparation ) . The logistical difficulties of performing P . vivax CHMI have slowed the developmental efforts for a vaccine against this parasite . We demonstrate that this challenge model , although complex , is feasible and can provide rapid assessment of vaccine efficacy . An unexpected outcome from the CHMI in this study identified two subjects who experienced multiple relapses from latent hypnozoites parasites despite adherence the optimal radical cure therapy . The investigation into the cause and follow-up of these two subjects has been reported previously and identified an association between the cytochrome P450 isoenzyme 2D6 ( CYP2D6 ) phenotype and the metabolism of PQ . It appears that CYP2D6 poor metabolizers are unable to convert the parent drug PQ into its active metabolite responsible for anti-hypnozoite activity and are , therefore , more likely to experience PQ failure and P . vivax relapse [13] . We propose that in addition to the Duffy blood group and G6PD testing , a laboratory screening test be used to characterize volunteers’ CYP2D6 genotype/phenotype in order to exclude subjects who , based on their genetic background , would be more likely to fail PQ therapy and experience relapse . Decreases in P . falciparum malaria especially in Southeast Asia have not been associated with commensurate decreases in P . vivax malaria . Recent strategies to eliminate P . vivax by targeting the reservoir of latently infected patients with antimalarial 8-aminoquinolines alone are unlikely to achieve elimination because of both the safety and lack of active drug metabolites in a significant proportion of the population [13] . A highly protective and durable pre-erythrocytic CSP-based P . vivax vaccine would have a dual beneficial effect of preventing not only the initial infection but also secondary relapses from hyponozoites thus inhibiting the establishment of latent infection . | Plasmodium vivax malaria has several unique features . Two of the main features are the inability to culture this parasite in vitro and its propensity to form dormant stages within the liver , which can only be treated with a single class of drugs that are contraindicated for a proportion of the population . Therefore , vaccines will play an important role in preventing this geographically widespread malaria species . In this clinical trial , we tested increasing amounts of the vaccine candidate VMP001/AS01B for safety and immunogenicity . In order to test if the vaccine can afford protection , we challenged the volunteers via the bite of infected mosquitoes , the first time such a human infection model has been used to evaluate vaccine efficacy for P . vivax malaria . While the vaccine did not protect any of the vaccinated subjects , this study resulted in some important findings , including the observation that a significant proportion of the subjects displayed a trend towards a delay in infection in individuals that correlated with antibodies to the repeat region of the vaccine antigen . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"immune",
"physiology",
"enzyme-linked",
"immunoassays",
"plasmodium",
"immunology",
"tropical",
"diseases",
"parasitic",
"diseases",
"animals",
"parasitology",
"vaccines",
"preventive",
"medicine",
"apicomplexa",
"antibodies",
"vaccination",
"and",
"immunization",
"insect",
"vectors",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"immune",
"system",
"proteins",
"proteins",
"epidemiology",
"immunoassays",
"disease",
"vectors",
"insects",
"immune",
"response",
"arthropoda",
"biochemistry",
"mosquitoes",
"physiology",
"biology",
"and",
"life",
"sciences",
"malaria",
"organisms"
]
| 2016 | Phase 1/2a Trial of Plasmodium vivax Malaria Vaccine Candidate VMP001/AS01B in Malaria-Naive Adults: Safety, Immunogenicity, and Efficacy |
Zika virus ( ZIKV ) and dengue virus ( DENV ) are genetically and antigenically related flaviviruses that now co-circulate in much of the tropical and subtropical world . The rapid emergence of ZIKV in the Americas in 2015 and 2016 , and its recent associations with Guillain-Barré syndrome , birth defects , and fetal loss have led to the hypothesis that DENV infection induces cross-reactive antibodies that influence the severity of secondary ZIKV infections . It has also been proposed that pre-existing ZIKV immunity could affect DENV pathogenesis . We examined outcomes of secondary ZIKV infections in three rhesus and fifteen cynomolgus macaques , as well as secondary DENV-2 infections in three additional rhesus macaques up to a year post-primary ZIKV infection . Although cross-binding antibodies were detected prior to secondary infection for all animals and cross-neutralizing antibodies were detected for some animals , previous DENV or ZIKV infection had no apparent effect on the clinical course of heterotypic secondary infections in these animals . All animals had asymptomatic infections and , when compared to controls , did not have significantly perturbed hematological parameters . Rhesus macaques infected with DENV-2 approximately one year after primary ZIKV infection had higher vRNA loads in plasma when compared with serum vRNA loads from ZIKV-naive animals infected with DENV-2 , but a differential effect of sample type could not be ruled out . In cynomolgus macaques , the serotype of primary DENV infection did not affect the outcome of secondary ZIKV infection .
The spread of Zika virus ( ZIKV ) from Africa to Asia and the Americas has led to the co-circulation and co-infection of ZIKV with other endemic arboviruses including dengue virus ( DENV ) [1–4] . ZIKV exists as a single serotype while DENV consists of four antigenically similar serotypes ( DENV-1-DENV-4 ) [5 , 6] . Primary infection with one of the four DENV serotypes typically confers antibody-mediated lifelong protection against reinfection with the same serotype . However , primary DENV infection may protect against , have no effect on , or enhance subsequent infection with a heterotypic serotype [7] . In humans , reinfection with a heterotypic DENV serotype is associated with higher viral load and an elevated risk of dengue hemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) [8] . Secondary DENV infections in macaques share some similarities with human infections including a trend toward higher peak DENV viral load [9] . In addition , a single rhesus macaque was previously reported as developing clinical responses consistent with dengue fever/DSS in response to secondary DENV-2 infection , including leukocytosis , thrombocytopenia , and elevated hematocrit approximately 5 days after viremia was first detected [9] . The similarity of DENV and ZIKV antigenic epitopes makes it difficult to distinguish between these viruses serologically in people living in DENV/ZIKV endemic areas [10–12] , suggesting that pre-existing immunity to either virus might affect the course of infection with the other . Cross-reactive DENV antibodies have been hypothesized as a factor driving the association of ZIKV with Guillain-Barré and adverse pregnancy outcomes [11] . Enhancement of ZIKV in the presence of anti-DENV antibodies has been demonstrated both in vitro and in vivo , in particular in murine models [13–16] . Experimental infections in mice have shown enhancement of , and protection from , ZIKV infection in the context of DENV immunity [15 , 17–19] . In humans to date , enhancement of ZIKV infection by pre-existing DENV immunity has not been observed [20] . In fact , a recent study in Salvador , Brazil showed an association between high anti-DENV total IgG titers and a decreased risk of ZIKV infection and symptoms [21] . Likewise , clinical data collected from pregnant women in Rio de Janeiro also suggest that there may be no association between the presence of DENV antibodies and ZIKV-associated pregnancy outcomes; however , DENV seroprevalence in the study population was >88% [22] . Macaque monkeys are commonly used in biomedical research as models for human diseases [23] . One key advantage of macaques over other animal models for studying vector-borne flaviviruses is their susceptibility to infection without requiring immunological manipulation . Primary ZIKV infections in macaques range from subclinical infections to mild fever , conjunctivitis , and rash [24–30] . In primary DENV infection in macaques , severe disease , including hemorrhage , has only been induced using a high dose intravenous inoculation [31] . On the other hand , infections with doses designed to mimic mosquito-borne transmission of DENV are clinically inapparent , necessitating the use of secondary clinical and laboratory parameters such as viral RNA ( vRNA ) loads , complete blood count ( CBC ) tests , and serum chemistry panels to study disease enhancement [32] . Multiple nonhuman primate studies utilizing a variety of DENV serotypes and ZIKV strains have been done over variable periods of time following primary infection to investigate the impact of primary DENV-1 , DENV-2 , and DENV-4 infection on secondary ZIKV infection; however , none to date have documented ZIKV enhancement [30 , 32] . Notably , no studies have looked at prior infection with DENV-3 . Our study more than doubles the total number of macaques used to study DENV/ZIKV interactions and incorporates data from all four DENV serotypes . Here we examined whether laboratory markers of clinical illness and vRNA loads are enhanced during secondary heterologous flavivirus infections in macaques . We also differentiate changes in clinical and laboratory parameters that are due to sedation , stress , and frequent venipuncture on animals from the impact of ZIKV infection , which were not controlled in previous studies . This work is timely given the efforts to develop an effective vaccine for ZIKV , as well as the introduction of a tetravalent DENV vaccine candidate , and other DENV vaccines in areas where ZIKV and DENV are now co-endemic [33 , 34] . This study adds new , more comprehensive and controlled information about the impact of prior DENV infection on secondary ZIKV disease in the macaque model .
The antigenic similarities between DENV and ZIKV have led to concerns that prior infection with , or vaccination against , one virus impacts the severity of disease upon secondary infection with the other virus [49] . Several lines of laboratory evidence support the possibility that DENV-specific antibodies can enhance ZIKV replication , serving as the impetus for the current study . Here , we directly compare the influence of all four DENV serotypes on subsequent ZIKV disease . The presence of heterotypic binding but non-neutralizing antibodies in human secondary DENV infections can be associated with increased viral load and disease severity [42 , 50] . However , within the context of secondary ZIKV infection , we did not observe a difference in either the magnitude or duration of ZIKV vRNA detection in the plasma of cohort A animals compared with ZIKV control animals ( Fig 3 ) . Consistent with these data , there were no significant differences in clinical laboratory parameters , body weight , or temperature between cohort A and negative control animals , although in some instances , we did observe differences compared with ZIKV control animals . Overall , the laboratory parameters suggest that the frequency of blood draws and animal manipulation , every day for the first 10 days of the study , may explain transient chemistry panel and CBC test perturbations . Alternatively , we may have observed natural fluctuations in parameters on an individual level that were only detectable due to the high frequency of sampling . This observation highlights the importance of including negative control animals for data interpretation . All cohort A animals had clinically in-apparent ZIKV infections , consistent with historical data for ZIKV infections in our animals [35 , 51] . Taken together , these results suggest that within a year of DENV-3 infection there is no evidence of protection from nor enhancement of ZIKV infection in rhesus macaques . These findings are consistent with those reported by Pantoja et al . and by McCracken et al . for rhesus macaques infected with ZIKV after DENV-1 , DENV-2 , DENV-4 , or yellow fever virus ( YFV ) infection [30 , 32] . Cohort B animals allowed us to examine the influence of primary ZIKV infection on secondary DENV infection . We found that the peak DENV-2 plasma vRNA loads of cohort B animals were higher than the DENV-2 serum vRNA loads for four DENV control animals , though this is likely due to the different sample types tested for vRNA in the two groups ( Fig 5 , S4 Fig ) . There were no clinical signs of severe DENV infection nor were any of the clinical and laboratory parameters perturbed in a way that suggested enhanced disease in any of the animals . The presence of heterotypic Ab is associated with increased DENV disease in secondary DENV infections [52] . Interestingly , prior to DENV-2 infection , we detected low levels of bAb in all three cohort B animals , as well as very low levels of nAb in two animals . Despite this , we observed no outward signs of more severe DENV disease in cohort B animals based on the clinical and laboratory parameters we examined . Overall , our results show that prior ZIKV exposure does not confer protection from infection nor enhancement of DENV-2 disease in rhesus macaques . This may be different in animals exposed to DENV-2 many years after ZIKV infection . Our findings seemingly contradict those reported by George et al . who observed elevated body temperature , neutropenia , lymphocytosis , and hyperglycemia , as well as significantly enhanced peak DENV-2 plasma viremia , in rhesus macaques infected with DENV 56 days after primary ZIKV exposure [43] . To examine whether pre-existing immunity to any of the four DENV serotypes and additionally , to model a natural exposure route within the context of secondary ZIKV infection , we compared ZIKV infections across four groups of MCM . Each group was previously exposed to a different serotype of DENV by SC inoculation at a consistent time point . Even though ZIKV-bAb were present in the plasma of all cohort C animals and nAb were detected in the serum of cohort C-1 ( previously infected with DENV-1 ) , we observed no evidence of protection from , nor enhancement of , ZIKV disease in any animals . One notable limitation of our findings for cohort C is the lack of ZIKV-alone or mock-inoculated control animals of the same species . The MCM with prior DENV exposure were available opportunistically for ZIKV infection , but DENV-naive MCM were not available to use as simultaneous control animals . While this may limit our ability to directly compare ZIKV infections in DENV-exposed with DENV-naive animals , we are still able to examine whether different DENV serotype exposures differentially influence ZIKV outcome . Because DENV-3 exposure in rhesus macaques did not result in protection or enhancement of ZIKV disease relative to DENV-naive control animals , and there was no difference in ZIKV outcome between DENV-3 and any other DENV serotype in MCMs , we postulate that other DENV serotypes likely also do not alter ZIKV disease relative to DENV-naive animals . In addition , the vRNA loads presented here were consistent in both magnitude and duration with previously published studies of ZIKV infection in cynomolgus macaques [26 , 27] . Likewise , the delay in the time to peak ZIKV RNA loads observed for the mosquito-infected animals in cohort C compared with SC-inoculated animals is consistent with those described previously for rhesus macaques infected with ZIKV via mosquito bite [48] . Interestingly , when we compared the AUC for each cohort C animal’s vRNA load , mosquito-infected animals had higher AUC values than SC-inoculated animals , suggesting that the mosquito infections may have resulted in higher vRNA loads and/or a longer duration of vRNA detection in these animals . We previously observed a similar trend with 2 of 4 rhesus macaques infected with ZIKV via mosquito bite having peak ZIKV RNA loads approximately 0 . 5–1 log10 higher than SC-inoculated macaques [48] . Potentially , the difference in magnitude could be the result of mosquito-infected animals receiving multiple infectious bites . Cohort C1 had detectable ZIKV nAb prior to infection; however , we did not identify statistically significant differences in plasma vRNA loads between groups based on DENV serotype exposure history ( Fig 6 ) . Likewise , there was no consistent viral load pattern based on the estimated number of mosquito bites/probes each animal received ( S1 Table ) . Overall , our studies of secondary ZIKV infection within a year of primary DENV infection in macaques suggest that there is no effect of pre-existing DENV immunity on ZIKV infection in macaque monkeys , consistent with other macaque studies [30 , 32] . This is in contrast with in vitro and murine studies that have shown significant enhancement of secondary ZIKV infection in the presence of anti-flavivirus antibodies [12 , 15 , 53 , 54] . Although we did not examine the potential enhancement activity of the antisera from our animals in mice , it is likely that we would observe outcomes similar to the aforementioned studies; this could be assessed in more detail in future studies . Whether this is consistent with secondary ZIKV infections in humans is unknown , but recent human cohort studies suggest more similarities with macaques [5 , 21 , 22 , 54] . In addition , there was , at most , a minimal difference in peak DENV vRNA load between animals with and without prior ZIKV exposure , suggesting that prior ZIKV exposure may only minimally affect DENV infection , at least in macaques . Primary and secondary DENV infections in macaques are largely subclinical while in humans , approximately 25% of DENV infections are estimated to be symptomatic [55] . Likewise , based on a meta-analysis that included 23 epidemiological studies by Haby et al . in June of 2018 , approximately 60% of human ZIKV infections are estimated to be symptomatic while macaques may show few signs of disease [24–30 , 56] . However , macaques remain an important immunocompetent model for human disease , in particular , in pregnancy studies where no other animal model as closely mimics human pregnancy and congenital Zika syndrome [57–62] . Although we found no evidence of enhanced ZIKV infection in DENV-immune macaque monkeys , it is important to note that this was in nonpregnant animals . These data therefore do not address the potential impact of pre-existing DENV immunity on ZIKV infection during pregnancy . Pregnancy is associated with major immunological changes that are likely related to the maintenance of an allogeneic fetus , including changes in the systemic cytokine milieu , impaired B cell lymphopoiesis , and high numbers of tolerogenic and regulatory T and B cells [63–66] . Thus , it will be imperative to determine if pregnancy-specific cofactors impact potential interactions between ZIKV pathogenesis and DENV immunity . It is also important to note that the DENV strains utilized for these experiments are not necessarily representative of the viruses currently circulating in Asia and the Americas . Future studies examining sequential infections with contemporary DENV strains may yield additional insights .
The macaques used in this study were cared for by the staff at the Wisconsin National Primate Research Center ( WNPRC ) in accordance with recommendations of the Weatherall report and the principles described in the National Research Council's Guide for the Care and Use of Laboratory Animals [67] . The University of Wisconsin—Madison , College of Letters and Science and Vice Chancellor for Research and Graduate Education Centers Institutional Animal Care and Use Committee approved the nonhuman primate research covered under protocol number G005401-R01 . The University of Wisconsin—Madison Institutional Biosafety Committee approved this work under protocol number B00000117 . The use of mice to infect mosquitoes with ZIKV in this study was approved by the University of Wisconsin-Madison , School of Veterinary Medicine Institutional Animal Care and Use Committee under protocol number V005519 . Mice were housed at the University of Wisconsin-Madison Mouse Breeding Core within the School of Medicine and Public Health . Once infected with ZIKV , they were housed in the Department of Pathobiological Sciences BSL-3 Insectary facility . All animals were housed in enclosures with required floor space and fed using a nutritional plan based on recommendations published by the National Research Council . Animals were fed a fixed formula , extruded dry diet with adequate carbohydrate , energy , fat , fiber , mineral , protein , and vitamin content . Macaque dry diets were supplemented with fruits , vegetables , and other edible objects ( e . g . , nuts , cereals , seed mixtures , yogurt , peanut butter , popcorn , marshmallows , etc . ) to provide variety to the diet and to inspire species-specific behaviors such as foraging . To further promote psychological well-being , animals were provided with food enrichment , structural enrichment , and/or manipulanda . Environmental enrichment objects were selected to minimize chances of pathogen transmission from one animal to another and from animals to care staff . While on study , all animals were evaluated by trained animal care staff at least twice each day for signs of pain , distress , and illness by observing appetite , stool quality , activity level , physical condition . Animals exhibiting abnormal presentation for any of these clinical parameters were provided appropriate care by attending veterinarians . Prior to all minor/brief experimental procedures , macaques were sedated using ketamine anesthesia and monitored regularly until fully recovered from anesthesia . Mice were anesthetized using isoflurane prior to inoculation and CO2 was used as the euthanasia method . Nine male and four female Indian-origin rhesus macaques ( Macaca mulatta ) and fifteen male Mauritian cynomolgus macaques ( Macaca fascicularis ) comprising the experimental cohorts utilized in these studies were housed and cared for at the WNPRC . Animals were observed daily and samples including blood , body weight , and body temperature measurements were collected as described previously with a timeline as shown in Fig 1 [47] . Historical data for ZIKV control and DENV control rhesus macaques were collected for previous , unrelated studies . ZIKV strains used in these studies included: Zika virus/H . sapiens-tc/FRA/2013/FrenchPolynesia-01_v1c1 ( ZIKV-FP ) and Zika virus/H . sapiens-tc/PUR/2015/PRVABC59-v3c2 ( ZIKV-PR ) . ZIKV-FP was originally obtained from Xavier de Lamballerie ( European Virus Archive , Marseille , France ) . ZIKV-PR was obtained from Brandy Russell ( CDC , Ft . Collins , CO ) . Both ZIKV strains were prepared as described previously [24 , 47] . The DENV-3 strain used to infect cohort A animals was dengue virus/H . sapiens-tc/IDN/1978/Sleman/78 ( DENV-3 throughout the text ) . The DENV-2 strain used to infect cohort B was dengue virus/H . sapiens-tc/NGU/1944/NGC-00982_p17c2 ( NGC ) and was obtained from Brandy Russell ( CDC , Ft . Collins , CO ) ( DENV-2 throughout the text ) . The four DENV strains used to infect the MCM ( cohort C ) include: DENV-1 , dengue virus/H . sapiens-tc/NRU/1974/WP74 , DENV-2 ( NGC as above ) , DENV-3 ( Sleman/78 as above ) , and DENV-4 , dengue virus/H . sapiens-tc/IDN/1978/1228 . DENV-1 and DENV-3 were originally obtained from the NIH while DENV-2 and DENV-4 were obtained from the CDC ( Ft . Collins , CO ) . All four viruses for cohort C were prepared by Takeda Vaccines , Inc . ( Cambridge , MA ) . Cohort A animals ( n = 3 rhesus macaques ) were infected subcutaneously ( SC ) with 0 . 5 mL of 6 x 105 PFU/0 . 5mL of DENV-3 . Approximately 6–12 months later , they were SC-inoculated with 1 mL of 1 x 104 PFU/mL of ZIKV-FP [24 , 34] ( Table 1 and S1 Table ) . cohort B animals ( n = 3 ) were SC- inoculated twice , 70 days apart , with 1 mL of 1 x 104 PFU/mL ZIKV-FP and approximately one year later they were infected SC with 1 mL of 1 x 105 PFU/mL of DENV-2 ( Table 1 and S1 Table ) . Fifteen Mauritian cynomolgus macaques ( MCM , cohort C ) were SC-inoculated with 0 . 5 mL of 1 x 105 PFU/0 . 5 mL DENV and approximately one year later with ZIKV-PR by mosquito-bite ( n = 13 ) , or a 1 mL SC inoculation ( n = 2 , 1 x 104 PFU/mL ) ( Table 1 and S1 Table ) . Rhesus macaque negative control animals ( n = 3 ) were SC-inoculated with 1 mL sterile PBS . Rhesus macaque ZIKV control animals were previously SC-inoculated with ZIKV ( 104 PFU/mL ) for other studies [24 , 34 , 50] . DENV control animal serum vRNA data were collected as part of a separate and unrelated study and were obtained from Takeda Vaccines , Inc . and included five rhesus macaques infected via SC inoculation with 1 x 105 PFU/0 . 5mL DENV-2 . Detailed descriptions of each cohort can be found in S1 Table and the sampling timeline after each infection is shown in Fig 1 . Aedes aegypti ( black-eyed Liverpool ( LVP ) strain ) used in this study were obtained from Lyric Bartholomay ( University of Wisconsin-Madison , Madison , WI ) and maintained at the University of Wisconsin-Madison as previously described [68] . Ae . aegypti LVP are ZIKV transmission competent [47 , 69] . Mosquitoes were exposed to ZIKV by feeding on isoflurane anesthetized , ZIKV-infected Ifnar1-/- mice as described previously [47] . These mice yielded an average infectious blood meal concentration of 1 . 45 x 106 PFU/mL ( ± 0 . 218 , n = 4 ) . Blood-fed mosquitoes were maintained as described previously [47] and randomized prior to exposure to two groups of ZIKV-naive , anesthetized MCM ( cohort C ) . ZIKV saliva titers collected from blood fed mosquitoes ranged from 101 . 48 to 103 . 26 PFU ( S8 Fig ) . Complete blood count ( CBC ) tests were assessed from EDTA-treated whole blood using a Sysmex XS-1000i hematology analyzer and manual slide evaluations as described previously [58] . Hemoglobin ( HB ) , hematocrit ( HCT ) , platelet ( PLT ) , and white blood cell ( WBC ) counts were compared between cohort A and ZIKV and negative controls , cohort B and negative controls , and cohort C DENV serotype exposure groups . Serum chemistry panels were evaluated using a Cobas 6000 analyzer ( Roche Diagnostics , North America ) . Serum aspartate aminotransferase ( AST ) , alanine aminotransferase ( ALT ) , creatinine ( CR ) , alkaline phosphatase ( ALP ) , creatine phosphokinase ( CPK ) , and lactate dehydrogenase ( LDH ) levels were compared between groups as described for CBC test parameters . Titers of ZIKV or DENV neutralizing antibodies were determined using plaque reduction neutralization tests ( PRNT ) on Vero cells ( ATCC #CCL-81 ) with a cutoff value of 50% ( PRNT50 ) [70] . Neutralization curves were generated in GraphPad Prism ( San Diego , CA ) and the resulting data were analyzed by nonlinear regression to estimate the dilution of serum required to inhibit 50% Vero cell culture infection . DENV-specific neutralizing antibodies were quantified in the serum of cohort C animals using a luciferase-expressing dengue reporter virus particle ( RVP ) neutralization assay for all four serotypes of DENV . Serum samples and positive controls were heat inactivated at 56°C for 30 minutes and diluted four-fold in Opti-MEM media ( Thermo Fisher Scientific , Inc . , Waltham , MA ) . Assays were conducted in triplicate 384-well plates using human lymphoblastoid cell line ( Assay-Ready Frozen Instant Cells of Raji-DC-SIGNR; Raji cells , acCELLerate GmbH , Hillsborough , NJ ) expressing flavivirus attachment factor lectin DC-SIGNR ( CD209L ) . Plates containing diluted serum and dengue RVP were incubated at 37°C , 5% CO2 for 1 hour to allow formation of immune-complexes to reach equilibrium . Thereafter , 15μL of Raji-R cells diluted in Opti-MEM at 4x105cells/mL seeding density were added to all wells of the plates . Plates were then incubated at 37°C , 5% CO2 for 72 ± 2 hours . Following incubation , plates were equilibrated to room temperature for 15 minutes followed by addition of 30 μL of Renilla-Glo ( Promega Co . , Madison , WI ) detection reagent ( diluted 1:100 in Renilla-Glo buffer ) . After 15 minutes , the plates were read using the Perkin Elmer EnSpire Luminescence program ( Perkin-Elmer , Inc . , Waltham , MA ) . Raw data were transposed in Microsoft Excel ( Microsoft Co . , Redmond , WA ) to fit the format requirements of GraphPad Preference intervalSM ( GraphPad Software , San Diego , CA ) . The titer of each sample was determined by calculating EC50 values using sigmoidal dose response nonlinear regression analysis . Plasma and PBMC were isolated from EDTA-treated whole blood on Ficoll paque at 1860 x rcf for 30 minutes as described in Dudley et . al . [24] . Serum was isolated from clot-activator tubes without additive . Viral RNA ( vRNA ) was extracted as previously described with a Maxwell 16 MDx instrument ( Promega , Madison , WI ) and evaluated using qRT-PCR [24 , 34 , 50] . RNA concentration was determined by interpolation onto an internal standard curve of seven ten-fold serial dilutions of a synthetic ZIKV RNA segment based on ZIKV-FP . The limit of detection of this assay is 100 copies vRNA/ml plasma or serum . For DENV challenged animals ( cohort B and DENV controls ) , vRNA in serum or plasma samples was measured using qRT-PCR at Takeda Vaccines . Viral RNA was extracted from 140μl of each sample using the QIAamp viral RNA kit ( Qiagen , Valencia , CA ) . The vRNA was eluted in 60 μl elution buffer and stored at -80°C until use . Viral RNA was quantified in a singleplex qRT-PCR with a primer/probe set targeting the 3’ non-coding region of DENV using a standard curve derived from in vitro transcribed cDNA clones and quantified as previously described [71] . All qRT-PCR reactions were performed in a final volume of 25 μl using the ABI 4X TaqMan Fast Virus 1-Step Master Mix . The reactions contained 5 μl extracted vRNA , 0 . 4 μM of each primer , and 0 . 2 μM probe . The reaction was conducted in the ABI 7500DX using a cycling protocol as follows: cycle 1–50°C for 5 minutes , cycle 2–95°C for 20 seconds , repeat cycle 3 , 45 times—95°C for 3 seconds and 55°C for 30 seconds . The qRT-PCR limit of detection of 3 . 6 log10 copies vRNA/mL was determined by testing nine replicates per dilution of the standard curve and selecting the concentration with a 100% detection rate as well as a low ( ≤ 0 . 5 ) cycle threshold standard deviation of the replicates . High-binding 96-well ELISA plates ( Greiner ) were coated with 30ng 4G2 antibody ( clone D1-4G2-4-15 ) in carbonate buffer ( pH 9 . 6 ) overnight at 4°C . Plates were blocked in Tris-buffered saline containing 0 . 05% Tween-20 and 5% normal goat serum ( cat . # G6767 , Sigma-Aldrich , St . Louis , MO ) for 1 hour at 37°C , followed by incubation with either ZIKV ( PRVABC59 , BEI ) or DENV-2 ( New Guinea C , BEI ) for 1 hour at 37°C . Heat inactivated plasma was tested at an 1:12 . 5 starting dilution in 8 serial 4-fold dilutions in duplicate , incubating for 1 hour at 37°C . Horseradish peroxidase ( HRP ) -conjugated goat anti-monkey IgG antibody ( Abcam , Cambridge , MA ) was used at a 1:2 , 500 dilution , followed by the addition of SureBlue reserve TMB substrate ( KPL , Gaithersburg , MD ) . Reactions were stopped by stop solution ( KPL , Gaithersburg , MD ) . Optical densities ( OD ) were detected at 450 nm . The limit of detection was defined as an OD value of the 1:12 . 5 dilution greater than three times the background OD of ZIKV/DENV naive macaque plasma . The log10 50% effective dilutions ( ED50 ) were defined as the plasma dilutions at which there was a 50% decline in the maximum IgG virion binding based on the OD . Whole virion IgG binding responses were compared between 0 and 28 dpi time points . Plasma samples from cohort A rhesus macaques were analyzed using Milliplex map Nonhuman Primate Cytokine Magnetic Bead Panel Premixed 23-Plex Assay ( EMD Millipore Corporation , Billerica , MA ) . Assays were run on a Bio-Plex 200 system and analyzed using Bio-Plex Manager Software version 6 . 1 . 1 ( Bio-Rad Laboratories , Hercules , CA ) . Standard curves were calculated using a logistic-5PL regression method using Bio-Plex Manager software version 6 . 1 . 1 ( Bio-Rad Laboratories , Hercules , CA ) . The following 23 cytokines and chemokines are included in the panel: G-CSF , GM-CSF , IFN-γ , IL-1ra , IL-1β , IL-2 , IL-4 , IL-5 , IL-6 , IL-8 , IL-10 , IL-12/23 ( p40 ) , IL-13 , IL-15 , IL-17 , IL-18 , MCP-1 ( CCL2 ) , MIP-1α ( CCL3 ) , MIP-1β ( CCL4 ) , sCD40L , TGF-α , TNF-α , and VEG . An additional dilution of the standard , beyond what is suggested in the manufacturer’s protocol , was included and used in analysis when detectable above background . Samples from animals 489988 , 756591 , 875914 , 850585 , 411359 , and 321142 were assessed using an 8-point standard curve . 448436 , 774011 , and 829256 were assessed using a 7-point standard curve . With the exception of the additional dilution of the standard , the assay was performed according to the manufacturer's protocol and used the provided serum matrix as a background control . To minimize plate effect as a confounder in analyses , all plasma samples from individual animals were assayed on a single plate . With the exceptions of IL-1ra , IL-8 , IL-2 , IL-15 , MCP-1 ( CCL-2 ) , and sCD40L , the cytokine and chemokine levels were not discernible above background levels of fluorescence and were not interpretable . Body weight and body temperature measurements were collected as shown in Fig 2 to assess weight loss and fever as proxies for disease . Body weights were monitored by WNPRC animal care and veterinary staff throughout the studies . Body weight data were not collected for cohort B animals . Temperatures were compared with the WNPRC reference ranges for the appropriate macaque species when determining the presence and/or absence of fever at each time point . WNPRC veterinary staff were consulted in determining whether an individual animal’s body temperature outside the reference ranges was clinically significant . Longitudinal vRNA loads ( vRNA/mL plasma or serum ) were compared between ZIKV control animals and cohort A , between DENV control animals and cohort B , and between cohorts C1-C4 over time , using Student’s t-tests ( cohorts A and B ) or analysis of variance ( ANOVA ) ( cohort C ) after calculating the area under the curve ( AUC ) for each animal’s vRNA load trajectory in R Studio ( v . 1 . 1 . 383 ) . Peak vRNA loads were compared between ZIKV control animals and cohort A , between DENV control animals and cohort B , and between C1-C4 using Student’s t-tests or non-parametric equivalents ( cohorts A and B ) or ANOVA ( cohort C ) in R Studio ( v . 1 . 1 . 383 ) . CBC test and serum chemistry panel parameters were normalized to baseline ( pre-infection ) levels by calculating fold changes from the baseline . The magnitude of laboratory values over the 28 day follow-up period was quantified by calculating the area under the curve ( AUC ) using the trapezoidal rule for each animal and laboratory parameter based on the fold changes from day 0 to day 28 . Because the AUC values were non-normally distributed , all AUC values were log-transformed before conducting the analysis . Analysis of variance ( ANOVA ) was used to compare the log-transformed AUC values between groups . Longitudinal changes of the fold changes were compared between groups using a linear mixed effects model with animal specific random effects . Multiple comparisons between groups were conducted using Tukey’s Honestly Significant Difference ( HSD ) to control the type I error . All reported p-values are two-sided and p < 0 . 05 was used to define statistical significance . Statistical analyses of CBC test and serum chemistry panel data were conducted using SAS software v . 9 . 4 ( SAS Institute Inc . , Cary NC ) . Differences in IL-1ra , IL-8 , IL-2 , IL-15 , MCP-1 , and sCD40L values were normalized to baseline values , transformed to positive values , and compared between negative control , ZIKV control , and cohort A animals using repeated measures ANOVA followed by pairwise comparisons where appropriate using Tukey’s HSD in R Studio ( v . 1 . 1 . 383 ) . Body weights were compared between cohort A , ZIKV control , and negative control animals using a mixed effects model with weight as the dependent variable , dpi as the fixed effect , and animal ID as a random effect . Longitudinal body weights were compared between DENV serotype exposures for cohort C using two-way ANOVA in the lme4 package [72] . Longitudinal body temperatures were compared between cohort A and ZIKV control and negative control animals , between cohort B and negative control animals using mixed effects models with temperature as the dependent variable , cohort/group/serotype as the fixed effect , and animal ID as the random effect using the lme4 package [72] . Final mixed effects models were chosen based on the minimization of Akaike information criteria ( AIC ) . Body temperatures were compared between cohort C DENV serotype exposure groups using repeated measures ANOVA . All body weight and temperature data were analyzed in R Studio ( v . 1 . 1 . 383 ) . Complete datasets for these studies have been made publicly available in a manuscript-specific folder on the Zika Open Research Portal ( https://go . wisc . edu/6wrw87 ) . Authors declare that all other data for these study findings are available via this portal or through supplementary information files from this article . | Pre-existing immunity to one of the four DENV serotypes is known to increase the risk of severe disease upon secondary infection with a different serotype . Due to the antigenic similarities between ZIKV and DENV , it has been proposed that these viruses could interact in a similar fashion . Data from in vitro experiments and murine models suggests that pre-existing immunity to one virus could either enhance or protect against infection with the other . These somewhat contradictory findings highlight the need for immune competent animal models for understanding the role of cross-reactive antibodies in flavivirus pathogenesis . We examined secondary ZIKV or DENV infections in rhesus and cynomolgus macaques that had previously been infected with the other virus . We assessed the outcomes of secondary ZIKV or DENV infections by quantifying vRNA loads , clinical and laboratory parameters , body temperature , and weight for each cohort of animals and compared them with control animals . These comparisons demonstrated that within a year of primary infection , secondary infections with either ZIKV or DENV were similar to primary infections and were not associated with enhancement or reduction in severity of disease based on the outcomes that we assessed . | [
"Abstract",
"Introduction",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"enzyme-linked",
"immunoassays",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"immunology",
"microbiology",
"vertebrates",
"animals",
"mammals",
"primates",
"viruses",
"animal",
"models",
"rna",
"viruses",
"experimental",
"organism",
"systems",
"antibodies",
"immunologic",
"techniques",
"viral",
"load",
"old",
"world",
"monkeys",
"research",
"and",
"analysis",
"methods",
"rhesus",
"monkeys",
"immune",
"system",
"proteins",
"monkeys",
"animal",
"studies",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"immunoassays",
"blood",
"plasma",
"macaque",
"biochemistry",
"eukaryota",
"blood",
"anatomy",
"flaviviruses",
"virology",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"amniotes",
"organisms",
"zika",
"virus"
]
| 2019 | Primary infection with dengue or Zika virus does not affect the severity of heterologous secondary infection in macaques |
Leishmania ( L . ) species are the causative agent of leishmaniasis . Due to the lack of efficient vaccine candidates , drug therapies are the only option to deal with cutaneous leishmaniasis . Unfortunately , chemotherapeutic interventions show high toxicity in addition to an increased risk of dissemination of drug-resistant parasites . An appropriate laboratory animal based model is still missing which allows testing of new drug strategies in the context of human immune cells in vivo . Humanized mice were infected subcutaneously with stationary phase promastigote L . major into the footpad . The human immune response against the pathogen and the parasite host interactions were analyzed . In addition we proved the versatility of this new model to conduct drug research studies by the inclusion of orally given Miltefosine . We show that inflammatory human macrophages get infected with Leishmania parasites at the site of infection . Furthermore , a Leishmania-specific human-derived T cell response is initiated . However , the human immune system is not able to prevent systemic infection . Thus , we treated the mice with Miltefosine to reduce the parasitic load . Notably , this chemotherapy resulted in a reduction of the parasite load in distinct organs . Comparable to some Miltefosine treated patients , humanized mice developed severe side effects , which are not detectable in the classical murine model of experimental leishmaniasis . This study describes for the first time L . major infection in humanized mice , characterizes the disease development , the induction of human adaptive and innate immune response including cytokine production and the efficiency of Miltefosine treatment in these animals . In summary , humanized mice might be beneficial for future preclinical chemotherapeutic studies in systemic ( visceral ) leishmaniasis allowing the investigation of human immune response , side effects of the drug due to cytokine production of activated humane immune cells and the efficiency of the treatment to eliminate also not replicating ( “hiding” ) parasites .
Over the last decades inbred mice were used to investigate the mechanisms of adaptive and innate immune responses against the obligatory intracellular parasite Leishmania ( L . ) major . From these studies fundamental paradigms regarding the elimination of the parasites were generated [1] . It could be demonstrated that in resistant C57BL/6 mice the elimination of intracellular parasites depends on the induction of an interferon-γ ( IFN-γ ) -driven T helper ( Th ) 1 immune responses . This Th1-associated cytokine production is crucial for the activation of infected macrophages to produce leishmanicidal pathways such as inducible nitric oxide synthase ( iNOS ) and subsequently nitric oxide radicals ( NO ) [2] . In contrast non-healing BALB/c mice mount a Th2 response associated with high levels of IL-4 and IL-13 [3]–[5] . Of note , similar to the experimental model , healing of cutaneous leishmaniasis in humans is also associated with a Th1-type immune response [5] . However , leucocytes from humans and rodents vary in their phenotype and function . For instance , the number of granulocytes in peripheral blood , which are the first line of defense against Leishmania infection , differs between human ( 50–70% ) and mouse ( 10–25% ) [6] . Furthermore , the relevance of the leishmanicidal molecules such as NO that is known to be substantially involved in the elimination of L . major parasites in the murine model of experimental leishmaniasis , is still not completely understood in the human system [7]–[11] . This might be due to the fact that murine macrophages respond to classical iNOS-inducing stimuli such as LPS and IFN-γ with iNOS expression and accumulation of NO whereas in humans alternative stimuli ( such as IFN-α/β , IL-4 , and anti-CD23 ) appear to be more efficient for monocyte and macrophage stimulation [2] , [12]–[13] . In addition antimicrobial peptides e . g . defensins which have the potential to reduce intracellular pathogens [14] differ in number , sequence , genomic location , and activation between human and mice [2] , [12]–[13] . For instance human defensins are present in human neutrophils but not in mice [15] whereas the mouse Paneth cells in the crypts of the small intestine express more than 20 different defensins but only two defensins are described in human cells [16] . Furthermore , differences in the maturation and regulation of T cells [17]–[18] , in immunoglobulin subtypes , receptors , and their function [18] impede the transfer of scientific results gained from animal experiments to the human system . NOD/scid-IL2Rγ−/− ( NOG or NSG ) mice were first described and generated by crossing IL2Rγ−/− mice with NOD-scid mice [19]–[20] . These mice lack not only B and T cells , but also feature considerably higher engraftment rates with human hematopoietic cells , extended lifespan , low levels of murine innate immunity and no NK cell activity [21] compared to previous mouse models . Upon transplantation of human CD34+ hematopoietic stem cells , NSG mice develop all human subsets of the immune system like T , B , and NK cells as well as myeloid cells and are able to mount an innate and adaptive immune response to antigens [19] , [21] , [22] . Therefore , mice bearing a human immune system might offer the opportunity to bridge the gap between animal models and clinical studies and are increasingly integrated in infectious disease studies and the investigation of human medication and vaccine projects [23]–[33] . The present study characterizes the human immune response in humanized mice after subcutaneous infection with L . major parasites . Our data reveal that human macrophages harbor Leishmania parasites . Unexpectedly , the parasitic replication was not limited to the site of infection but also observed in visceral organs such as spleen and liver . Although Leishmania-specific human T cells were generated , the humanized mice succumbed to the cutaneous infection . To demonstrate the versatility of Leishmania-infected humanized mice for future drug studies , we complemented the infection studies with systemic Miltefosine treatment . Miltefosine efficiency was first implemented in tumor treatments and later described as a new line of drugs against Leishmania infections tested in humans [34] and mice [35] . It amends the second line drugs of aromatic diamidines , amphotericin B , and pentavalent antimonials which induce rising resistance and are toxic with severe sometimes life-threatening side effects [36] . In conclusion we could demonstrate that human immune cells interact with protozoan parasites in a murine environment . This novel experimental model might be beneficial for the investigation of drug efficiency to eliminate the parasite in the context of human immune cells possibly involved in severe side effects such as organ damage .
Cord blood samples were taken with approval from the Ethics Committee of the University Regensburg ( permission no . 08/021 ) . All patients included in the experiments provided written informed consent . All experiments were performed in accordance with relevant institutional and national guidelines , regulations and approvals . Mice ( NOD-scid IL2Rγnull; ( NSG ) ) used for the experiments were obtained from Jackson Laboratories , and bred and kept in a specialized pathogen-free facility . Newborn mice were transplanted with 3×105 CD34+ hematopoietic stem cells isolated from Cord blood as described before [37] . Hematopoietic cells were injected into the liver of neonatal mice . For infection studies , a total of 32 humanized mice , between three and five months of age were analyzed for their reconstitution levels shortly before L . major infection . All animal work was approved by the local veterinary authorities from the district government of Upper Palatinate/Bavaria based on the international European guidelines and national regulations of the German animal protection act ( permission no . 54-2531 . 2-18/08 ) . Humanized mice were infected via s . c . injection of 3×101–3×106 stationary phase promastigote L . major ( MHOM/IL/81/FE/BNI ) parasite into the right hind footpad adjusted to a final volume of 30 µl . BALB/c and C57BL/6 mice were infected with 3×106 stationary phase promastigote L . major and served as controls . Disease progression was monitored on a daily basis . Footpad swelling and weight loss was monitored once a week . Treatment survey was implemented by administering Miltefosine ( 2 . 5 mg/kg; Cayman Chemical ) orally on a daily basis using a gavage tube starting two weeks post infection . This initial condition of infection is important to mimic the scenario in L . major patients before starting Miltefosine treatment . Parasite load was determined using qPCR-analysis as described before [38] . SYBR Green ( BioRAD , Germany ) was used for product detection and quantified on total human-beta-actin 5′-GGG TGT AAC GCA ACT AAG TCA T-3′ ( forward ) and 5′-TGG ACA TCC GCA AAG ACC TG-3′ ( reverse ) or mouse ß-actin 5′-GGA TGC CAC AGG ATT CCA TAC CCA-3′ ( reverse ) and 5′-TCA CCC ACA CTG TGC CCA TCT ACG A-3′ ( forward ) . Amplification and detection were performed using the Multicolor Real-Time PCR Detection System ( BioRAD , Germany ) . Standards , samples and non template controls were analyzed in triplicate for each run . As described in [38] the copies of Leishmania DNA ( Units ) were determined within the desired samples . DNA isolated from a distinct number of promastigote Leishmania parasites were used to determine the number of parasites/Unit Leishmania DNA . According to that internal standard we calculated the number of parasites within the sample . The data are presented as number of parasites/ng human β-actin ( humanized mice ) or number of parasites/ng mouse β-actin ( C57BL/6 and BALB/c mice ) . For FACS analyses mononuclear cells were isolated from the indicated tissues as previously described [37] . Spleen cells from infected humanized mice ( six weeks post infection ) and uninfected controls were isolated as described above and labeled with Carboxyfluorescein succinimidyl ester ( Invitrogen , Oregon , USA ) as described before [38] . For stimulation , 5×105 cells/well from each individual mouse were incubated with lysed L . major promastigote total antigen ( ratio: ten parasites per cell ) , PMA ( final concentration 50 ng/ml ) , or with ConA ( final concentration 1 µg/ml ) in microtiter plates in RPMI 1640 supplemented with 10% FCS ( Sigma ) , glutamine , Hepes , and Penicillin-Streptomycin ( Seromed-Biochrom , Berlin , Germany ) . After 72 h , cells were harvested and the percentage of proliferating B cells , CD4+ and CD8+ T cells was determined by flow cytometry . For the analysis of human cytokine production in L . major infected humanized mice , spleen cells ( three weeks post infection ) were harvested and re-stimulated as described above . Supernatants were collected 72 hours later and the release of cytokines was measured in a 5-colour-Multiplex Human Cytokine Panel-assay ( Milllipore , Billerica , MA , USA ) according to manufacturer's protocol . Reconstitution with human immune cells was determined by flow cytometry using a LSR-II flow cytometer running the Diva software package ( BD Biosciences , San Jose , USA ) . To reduce non-specific binding , cells were incubated with mouse and human IgGs ( 10 ug/ml , Sigma ) on ice for 10 minutes before staining with human specific monoclonal antibodies ( mAb ) or appropriate isotype mAb . Samples were stained using the following human specific mAb: anti-CD3-FITC ( IgG2aκ , clone HIT3a ) , anti-CD19-PE ( IgG1κ , Clone: HIB19 ) , anti-CD33-PerCP-Cy5 . 5 ( IgG1κ , clone P67 . 6 ) , anti-CD45-APC ( IgG1κ , clone HI30 ) , anti-CD4-PE ( IgG1κ , clone SK3 ) , and anti-CD8-PE ( IgG1κ , clone HIT8a ) from BD Biosciences . For further characterization we used anti-CD3-PerCP ( IgG2a , clone OKT3 ) , anti-HLA-DR-PE-Cy7 ( IgG2bκ clone LN3 ) , anti-CD86-PE ( IgG2bκ clone IT2 . 2 ) , anti-CD45RA-PE-Cy7 ( IgG2aκ , clone HI100 ) , and anti-CD27-Biotin ( IgG1κ , clone O323 ) from eBioscience ( San Diego , USA ) . Biopsies from organs of interest were embedded in Killik cryostat embedding medium ( BioOptica , Milano , Italy ) and stored at −80°C . Cryo sections ( 7 µm ) were thawed onto poly-L-Lysin slides ( Thermo Scientific , Germany ) and processed as described before [38] . The sections were first stained with rabbit-anti-L . major serum in PBS/BSA and 0 . 1% saponin ( Roth , Karlsruhe , Germany ) . After washing in PBS containing 0 . 01% Tween-20 , slides were incubated with goat anti-rabbit AlexaFlour 546 ( Invitrogen , Darmstadt , Germany ) . The antibodies anti-hCD45 ( APC ) antibody ( IgG1κ , clone HI30 ) and anti-hCD68-PE ( IgG2bκ , clone Y1/82A ) were used to stain for human leukocytes and macrophages . All antibodies were tested for human-specificity on not humanized NSG mice . Nuclei were stained using DAPI ( Sigma-Aldrich , Deisenhofen , Germany ) . After mounting with Permafluor ( Thermo Scientific ) , sections were analyzed using an immuno-fluorescence microscope ( Zeiss , Jena , Germany ) equipped with high-sensitivity gray scale digital camera ( Openlab System; Improvision , Heidelberg , Germany ) . Separate images were collected for each section , analyzed and merged false-color afterwards . Final image processing was performed using Adobe Photoshop Elements ( Adobe Systems GmbH , München , Germany ) . Alanin-aminotransferase ( ALT/GPT ) , and aspartat-aminotransferase ( AST/GOT ) were measured in serum using the analytic instrument COBAS INTEGRA 800 ( Roche Diagnostics , Germany ) . Normal values published for mice: ALT = 17–77 ( U/I ) and AST = 54–298 ( U/I ) and for humans: ALT = <50 and AST = <50 . Spleen and foot tissue or spleen cells were homogenized in a Tissue Lyser ( Qiagen , Hilden , Germany ) , total RNA was extracted using the TriFast reagent ( Peqlab , Erlangen , Germany ) and contaminating DNA was removed with DNase I ( Ambion DNAfree , Invitrogen , Karlsruhe , Germany ) . Subsequently , 2–5 µg RNA were reverse transcribed using the High Capacity cDNA Reverse Transcription Kit ( Invitrogen ) . To assess the amount of iNOS cDNA the Applied Biosystems HT7900 Taqman quantitative PCR system ( Invitrogen ) was used . Human and mouse NOS2 cDNA was measured in triplicates with the following gene-specific Applied Biosystems Taqman assays ( Invitrogen ) : huNOS2 ( Hs01075529_m1 ) , mNOS2 ( Mm00440485_m1 ) . The gene for human glyceraldehyde 3-phosphate dehydrogenase ( GAPDH , Hs02758991_g1 ) or for mouse hypoxanthine guanine phosphoribosyl transferase-1 ( HPRT-1 , Mm00446968_m1 ) was used as endogenous control for calibration of mRNA levels , respectively . Quantitative analyses were performed using the SDS 2 . 3 software ( Applied Biosystems/Invitrogen ) . The mRNA levels were calculated by the following formula: relative expression = 2- ( CT ( Target ) -CT ( Endogenous control ) ) ×f , with f = 104 as an arbitrary factor . Statistical analysis was performed using Graph Pad's Prism . All data are represented as as mean ± SEM , and were tested for statistical significance using Student's t test , ANOVA , Bonferroni posttest , Log-rank or Tukey's Multiple Comparison Test , as indicated in the figure legends .
Shortly before infection humanized mice were tested for efficient reconstitution with human immune cells ( Figure S1 ) . To prove the concept that humanized mice develop a local inflammation at the site of infection we infected them subcutaneously with different numbers of L . major parasites . Depending on the amount of parasites injected , differences in footpad swelling ( Figure 1A ) and weight loss ( Figure 1C ) were found . Humanized mice infected with a high dose ( 3×106 ) of L . major showed an earlier , massive weight loss starting at day 35 followed by intermediate dose ( 3×103 ) which resulted in a delayed , less pronounced weight loss at day 49 post infection ( Figure 1B ) . Due to the severity of disease , some of the animals infected with 3×106 or 3×103 Leishmania parasites died between day 14 and day 49 and the remaining mice were sacrificed at day 50 and 56 , respectively ( Figure 1D ) . In contrast to the high dose infection , low dose infection ( 3×101 ) caused only a slight weight loss ( Figure 1C ) and all mice survived until day 56 ( Figure 1D ) . Non-infected humanized mice served as controls and did not show any sign of weight loss or disease and survived until day 56 on which they were sacrificed ( data not shown ) . In comparison high dose ( 3×106 ) infected BALB/c showed the highest local inflammation reaction and had to be sacrificed after day 29 ( Figure 1B ) . Six weeks post infection , quantification of the number of Leishmania parasites within the samples of humanized mice revealed that the site of infection represents the organ with the highest parasites load ( Figure 1E ) . Visceral organs such as the spleen and liver were also infected but with lower concentration of parasites when normalized to human β-actin ( Figure 1E ) . High dose infected BALB/c showed also systemic spreading of the parasites with the highest concentration in the footpad ( Figure 1F ) . Considering the high load of L . major parasites in the liver of humanized mice , we assessed the liver damage by measuring the levels of the liver enzymes Aspartat-Aminotransferase ( AST/GOT ) and Alanin-Aminotransferase ( ALT/GPT ) in the serum . These data revealed that humanized mice show an increase of GOT ( ø 541 U/l+/−220 SEM ) and GPT ( ø70 U/l+/−19 SEM ) after infection with L . major compared to uninfected controls GOT ( ø 77 , 2 U/l+/−6 SEM ) and GPT ( ø 21 U/l+/−2 , 6 SEM ) . In contrast , in resistant C57BL/6 mice ( GOT: 88 U/l+/−3 SEM; GPT: 15 U/l+/−0 , 6 SEM ) as well as susceptible BALB/c ( GOT: 198 U/l+/−47 SEM; GPT: 71 U/l+/−15 SEM ) infected with L . major parasites no signs of liver damage were detectable . Thus , only humanized mice infected with L . major showed an increase in GOT levels compared with normal values of GOT and GPT in humans and mice ( see Material & Method section ) . Notably , these infection experiments clearly demonstrate that humanized mice show signs of inflammation at the site of infection indicating that they respond to the protozoan parasites L . major . Furthermore , we show that comparable to some human patients [39] , [40] humanized mice develop a mixture of cutaneous and visceral manifestation of leishmaniasis . The cellular components involved in that process are presented below . To address the question whether human host cells are infected with the parasites , immunohistological analyses were performed . Consistent with the PCR data we were able to show that L . major antigens were present in liver and spleen of humanized mice ( Figure 2A and B ) . Furthermore , L . major antigens were detectable within human CD45+ hematopoietic cells ( Figure 2A and 2B ) . However , there are also murine macrophages and granulocytes ( hCD45− cells ) still present in humanized mice [21] . Therefore , one cannot exclude the possibility that besides human phagocytes , mouse phagocytes were also infected with L . major parasites . To determine the origin of the infected phagocytes in detail , cryosections from infected footpads and from infected visceral organs ( e . g . spleen ) were characterized . These data revealed that Leishmania-antigens were detectable within nucleated cells ( Figure 2C ) . Moreover , we confirmed that Leishmania-antigens were detectable in CD68+ human macrophages ( Figure 2C; insert CI ) and CD68− cells ( Figure 2C; insert CII ) . However , randomized spot tests indicated that over 80% of infected cells were human CD68+ macrophages ( data not shown ) . Thus we conclude that human macrophages as well as mouse phagocytes can engulf Leishmania parasites within the infected tissue of humanized mice . To address the questions whether other myeloid-derived human cells respond to the Leishmania-infection , antigen-presenting cells such as dendritic cells ( Lin−CD11c+HLA-DR+ ) were characterized in infected humanized mice . These data demonstrate that activation markers such as CD86 and HLA-DR were increased in humanized mice infected with L . major compared to uninfected controls ( Figure S2 A–C ) . Therefore , we conclude that myeloid cells derived from human stem cells interact and respond to the parasites in vivo . For functional studies of the leishmanicidal capacity induced at the side of infection and in visceral organs such as the spleen , we analyzed the mRNA expression levels of human and murine derived iNOS three weeks after infection . Naive humanized mice showed a baseline expression of murine iNOS within the footpad and spleen tissue which was significantly increased in the footpad after the infection with L . major ( Figure S2D ) whereas no alterations of mouse iNOS mRNA levels were observed in the spleen . Human iNOS mRNA expression was below the detection limit in all samples ( Figure S2D ) . The fact that mouse iNOS is induced upon infection further supports the idea that not only human macrophages but also murine phagocytes get infected and activated in humanized mice . L . major parasites disseminate within the humanized mice and interact with human leukocytes . We further aimed to answer the question whether the parasites or their corresponding antigens are immunogenic in terms of being able to initiate an adaptive human-derived immune response . From Leishmania-infected patients it is known that macrophages represent the dominant cell population within cutaneous lesions , followed by CD3+ T cells [41] . Considering our histological findings , demonstrating a dense dermal infiltration of CD68+ macrophages building clusters in the infected area ( data not shown ) , we focused our interest on the activation status of human lymphocytes . Comparable to the biopsy data from patients suffering from cutaneous leishmaniasis [41] , we were able to show that CD3+ T cells dominate the lymphocyte population at the site of infection whereas B cells represent the minority of the skin-infiltrating lymphocytes ( Figure 3B ) . Further distinction of the invading CD3+ T cells into CD4+ T helper cells versus CD8+ cytotoxic T cells revealed the dominance of the CD4+ T cell population ( Figure 3C ) . Additional analysis of the infiltrating T cells illustrated a memory phenotype ( CD27+ CD45RA− ) of both T cell subpopulations ( Figure 3D ) . This increased population of memory phenotype indicates that T cell activation took place after infection with L . major parasites . Due to a relatively weak formation of lymph nodes and Peyer's patches in humanized mice [22] , we analysed human lymphocytes isolated from spleens of infected animals . The majority of immune cells in the spleen were CD3+ T cells , mainly CD4+ ( 70% ) in infected humanized mice as well as in controls . Further analysis of CD4+ and CD8+ T cells revealed that there was a significant switch towards the memory phenotype ( CD45RA−CD27+; Figure 4A ) . As shown in Figure 4B , immune cells isolated from uninfected controls did not proliferate after stimulation with soluble Leishmania antigen ( SLA ) whereas spleen cells derived from infected animals showed proliferation of CD4+ as well as CD8+ T cells . The highest proliferation rate was induced by polyclonal stimulation with phorbol myristate acetate ( PMA ) and Concanavalin A ( ConA ) ( Figure 4B+C ) . Thus , human T cells are still responsive to different stimuli even though they were embedded in a murine environment . Furthermore , the cytokine response of splenocytes from infected humanized mice was characterized . These data revealed that a restimulation with SLA results in a significantly increased IFN-γ ( Th1 type cytokine ) production whereas IL-4 and IL-10 ( Th2 type cytokine ) were hardly detectable ( Figure 4D ) . The proinflamatory cytokines TNF and IL-6 are not increased after restimulation with SLA . Based on this cytokine profile we conclude that the splenocytes release cytokines representing a Th1-type immune response . In addition we tried to induce a DTH reaction in humanized mice infected with L . major to measure the migration capacity of Th1 type effector cell to the site of antigen inoculation [42] , [43] . Remarkably , no clinical signs of swelling and redness could be measured after subcutaneous injection of Leishmania antigen ( data not shown ) . Cutaneous leishmaniasis caused by L . major is characterized by skin ulcers , papules or nodules . However , exceptions from these classical cutaneous manifestations have been reported . Those patients show unusual forms of cutaneous leishmaniasis with abnormal liver function [39] , [44] . As already shown above humanized mice develop a sever course of disease not sufficient to eliminate the parasites . Thus , humanized mice might represent an experimental model for Leishmania-caused disease , which do not show any self-healing capacity . Accordingly , we tested a drug-based treatment against the parasites in the humanized mouse model . We used Miltefosine because it is already established for treatment of visceral and cutaneous leishmaniasis [34] . To mimic the clinical situation , Miltefosine was given orally starting at day 14 post infection when clinical signs such as swelling and redness at the site of infection were detectable . Miltefosine treatment of humanized and BALB/c mice did not reduce the local inflammatory process at the site of infection ( Figure 5A ) . Nevertheless , humanized mice showed a reduced local inflammation at the site of infection compared to BALB/c . Furthermore , BALB/c mice did not loose weight ( with or without treatment ) whereas humanized mice lost weight when treated with Miltefosine ( Figure 5B ) . Additionally , only humanized mice treated with Miltefosine revealed increased liver enzyme levels ( Figure 5 C ) . For the characterization of the therapy efficiency we quantified the number of parasites in the organs of interest . These data revealed that BALB/c mice show significant reduction of the parasite load at the site of infection and visceral organs and bone marrow ( Figure 5E ) . In contrast the Miltefosine therapy in humanized mice showed only slight reduction of the parasites in the liver and footpad ( Figure 5D ) . However , during Miltefosine therapy we observed adverse effects , which are also described in humans [34] such as weight loss ( Figure 5B ) and significantly increased release of liver enzymes ( GOT and GPT ) , which did not occur in infected and treated BALB/c mice ( Figure 5C ) . Notably , Miltefosine treatment neither affected the overall reconstitution , the T and B cell distribution nor the priming of human T cells in the spleen ( Figure S3A+B ) .
Humanized mice in triple deficient NOD/scid-IL2γ−/− ( NSG ) mice develop all cellular components of the human immune system such as T , B , and NK cells as well as myeloid cells and are able to mount an adaptive and innate immune response to antigens [19] , [21] . Based on this innovative concept to create a human immune system in a murine environment , scientific issues regarding virus-caused infectious diseases [24] , [45]–[47] , therapy and vaccine studies [28]–[33] , graft versus host disease ( GvHD ) [48] , and tumor diseases [37] were already studied . To characterize the interactions between human immune cells and obligatory intracellular protozoan parasites under in vivo conditions we infected humanized mice with L . major parasites . In general these parasites induce cutaneous leishmaniasis . However , exceptions of this classic form with mild visceral tendency have been reported in patients before [40] , [49]–[51] . This capacity for visceralization was also described in L . tropica infected patients [52] . The L . major strain ( MHOM/IL/81/FE/BNI] used in our experiments was isolated from a LCL patient [53] . Of note , under experimental conditions this L . major strain can result in visceral manifestations in resistant C57BL/6 and susceptible BALB/c mice as well [50] , [54]–[60] . Thus , it cannot be generally excluded hat L . major parasites might show the tendency of mild visceralization in parallel to the cutaneous manifestation . Here we questioned for the first time whether L . major parasites can replicate within humanized mice . Furthermore , we tested this novel experimental model - sharing murine and human components – for its capacity to characterize human adaptive and innate immune response to L . major and the efficiency of treatments like such as hexadecylphosphocholine Miltefosine . Initial experiments revealed that cutaneous inoculation of L . major parasites resulted in an inflammation at the site of infection . Additionally the severity of disease correlates with the dose of parasites injected . Thus , humanized mice develop a local inflammation caused by the protozoan parasites . Based on the reconstitution of NSG mice with naïve human leucocytes it is most likely that human cells are involved in the parasite-induced inflammatory response . Thus , we further investigated the phenotype and origin of potential host cells at the site of infection of humanized mice . It is important to mention that humanized mice still possess mouse myeloid cells such as macrophages and granulocytes . And indeed , our data revealed that Leishmania-antigens can be detected in human as well as murine myeloid cells indicating that mouse myeloid cells represent target cells for Leishmania in humanized mice . As suggested by several findings infected mouse myeloid cells might function as a niche for Leishmania allowing immune evasion . First , the human adaptive immune system is not able to control or stimulate mouse immune cells due to mismatched MHC expression and the species-specificity of human IFN-γ [61]–[64] . Second , most of those myeloid mouse cells show functional weaknesses as for instance insufficient differentiation of fully functional DC from NSG bone marrow cells [21] . And third , it has been shown that in NOD-scid mice the remaining mouse macrophages show an impaired functionality [61] . However , we detected a 10-fold increase in the expression of murine iNOS mRNA in the footpad tissue of humanized mice three weeks after infection with L . major whereas no increase of iNOS mRNA was found in the spleen . This clearly indicates that mouse macrophages somehow get activated during infection , although the upregulation of iNOS mRNA in humanized mice is ∼factor 100 lower than that seen in infected BALB/c mice . In accordance , no iNOS mRNA increase was visible in the spleen of BALB/c mice upon infection [unpublished data , U . Schleicher and C . Bogdan , Erlangen ) . With regard to parasite control this lack of iNOS increase in the spleen is of minor importance as rather NADPH-oxidase-dependent than iNOS-dependent mechanism are required for the resolution of the parasites in this organ [65] . In contrast to mouse iNOS mRNA no induction of human iNOS mRNA was detectable in infected humanized mice . Since we demonstrated that human macrophages get infected in humanized mice this result rather supports the idea that iNOS expression by macrophages or other cells in response to infections does not occur in the human immune system [15] . That this statement might not hold true in human cutaneous leishmaniasis is supported by results demonstrating iNOS mRNA and protein expression in skin lesions of CL patients [11] , [66] . Together , it can be therefore speculated that the atypical visceralisation of L . major parasites within humanized mice might be at least partially due to the insufficient production of murine derived leishmanicidal molecules and the lack of human iNOS induction . Considering that infected macrophages need to be activated by an efficient Th1-cell response [67] the human-derived T cell response was further analyzed . Inflammatory human T cells could be detected at the site of infection . Most of them represent the phenotype of a memory T cells and the majority of these cells belong to the CD4+ T helper cell population , known to play a major role in the activation of infected macrophages and subsequent killing of the intracellular Leishmania parasites [68] . The fact that CD4+ as well as CD8+ T cells were efficiently primed in vivo indicates that human antigen presenting cells are functional and that the induced human T cell response is Leishmania-specific . Based on the fact that splenocytes of infected humanized mice respond with a pronounced human IFN-γ production after restimulation with soluble Leishmania antigen ( SLA ) we conclude that a human Th1-like immune response was induced . From the experimental model of leishmaniasis it is known that resistant C57BL/6 mice develop a delayed hypersensitivity ( DTH] response after s . c . injection of SLA whereas susceptible BALB/c mice do not show any signs of a DTH response [42] , [43] . This lack of DTH reaction was also detectable in infected humanized mice and might represent the dysfunction of the induced Th1-type effector cells to get recruited to the site of antigen inoculation . Based on the fact that humanized mice do not show signs of a DTH reaction we can conclude that Th1 cells are efficiently primed within the spleen but failed to migrate to the site of the infection . This might also explain the relative low parasite density within the spleen compared to the parasite load detectable at the site of infection . To investigate the versatility of our model for future drug studies , we treated infected humanized mice and BALB/c mice with Miltefosine . In BALB/c mice the application of Miltefosine induced a reduction of the parasite load predominantly in visceral organs such as liver and spleen in the absence of liver damage and weight loss . In contrast humanized mice respond different to the Miltefosine therapy . They show signs of liver damage and significant weight loss after treamtent . Thus the Miltefosine-derived side effect must be the consequence of the presence of human immune cells . This is in agreement with published data [69]–[70] demonstrating signs of liver toxicity in patients after Miltefosine therapy . The observed Miltefosine-based side in humanized nice might be due to the fact that Miltefosine can activate human immune cells to release pro inflammatory cytokines [71]–[72] that in turn cause the observed liver damage . Thus , humanized might be a useful tool to test possible side effects which are not detectable in classical animal models such as BALB/c , before starting clinical trials . The leishmanicidal effect of Miltefosine in humanized mice is visible but not comparable to control BALB/c mice . This might be due to the lack of mouse IFN-γ which in turn can not activate infected mouse macrophages . In conclusion , we have generated a novel Leishmania major infection model using humanized mice which develop a systemic non-self-healing disease . Humanized mice might therefore be especially interesting for studies on visceral leishmaniases offering an additional challenge for new treatment strategies: the elimination of resting ( “hiding” ) Leishmania parasites within the mouse immune cells . In addition , side effects induced by activated human immune cells ( cytokine release ) which do not occur in wild type mice can now be illuminated in a small animal model which might help to predict possible side effects before the drugs get tested in clinical trials . | As many as 12 million people suffer from Leishmania ( L . ) infection worldwide with about one to two million newly infected people every year . Due to the lack of vaccine strategies , the only option is chemotherapeutic intervention which can cause serious side effects . Therefore , new prevention or treatment strategies are urgently needed in addition to an appropriate animal model for testing . We infected humanized mice in the footpad with stationary phase promastigote L . major and analyzed the human innate and adaptive immune response by flow cytometry , histology , and quantitative PCR . Infected macrophages were detectable at the site of infection and in lymphoid organs . Additionally , we were able to measure Leishmania-specific T cell priming in humanized mice . However , these human immune defense mechanisms were not sufficient to prevent systemic spreading and lethality . By the inclusion of Miltefosine , we tested this novel model for its versatility in conducting drug studies . The oral treatment was able to reduce parasitic load but also revealed side effects which are described in humans but not in mice . Therefore , we propose humanized mice as a novel model , which offers the opportunity to study new therapy strategies in chronic leishmaniasis in the context of a human immune system . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"parastic",
"protozoans",
"leishmania",
"immunology",
"biology",
"microbiology",
"protozoology",
"host-pathogen",
"interaction",
"immune",
"response"
]
| 2012 | Leishmania major Infection in Humanized Mice Induces Systemic Infection and Provokes a Nonprotective Human Immune Response |
Human immunodeficiency virus ( HIV ) infection is often accompanied by infection with other pathogens , in particular herpes simplex virus type 2 ( HSV-2 ) . The resulting coinfection is involved in a vicious circle of mutual facilitations . Therefore , an important task is to develop a compound that is highly potent against both viruses to suppress their transmission and replication . Here , we report on the discovery of such a compound , designated PMEO-DAPym . We compared its properties with those of the structurally related and clinically used acyclic nucleoside phosphonates ( ANPs ) tenofovir and adefovir . We demonstrated the potent anti-HIV and -HSV activity of this drug in a diverse set of clinically relevant in vitro , ex vivo , and in vivo systems including ( i ) CD4+ T-lymphocyte ( CEM ) cell cultures , ( ii ) embryonic lung ( HEL ) cell cultures , ( iii ) organotypic epithelial raft cultures of primary human keratinocytes ( PHKs ) , ( iv ) primary human monocyte/macrophage ( M/M ) cell cultures , ( v ) human ex vivo lymphoid tissue , and ( vi ) athymic nude mice . Upon conversion to its diphosphate metabolite , PMEO-DAPym markedly inhibits both HIV-1 reverse transcriptase ( RT ) and HSV DNA polymerase . However , in striking contrast to tenofovir and adefovir , it also acts as an efficient immunomodulator , inducing β-chemokines in PBMC cultures , in particular the CCR5 agonists MIP-1β , MIP-1α and RANTES but not the CXCR4 agonist SDF-1 , without the need to be intracellularly metabolized . Such specific β-chemokine upregulation required new mRNA synthesis . The upregulation of β-chemokines was shown to be associated with a pronounced downmodulation of the HIV-1 coreceptor CCR5 which may result in prevention of HIV entry . PMEO-DAPym belongs conceptually to a new class of efficient multitargeted antivirals for concomitant dual-viral ( HSV/HIV ) infection therapy through inhibition of virus-specific pathways ( i . e . the viral polymerases ) and HIV transmission prevention through interference with host pathways ( i . e . CCR5 receptor down regulation ) .
Human immunodeficiency virus ( HIV ) infection is commonly associated with other sexually transmitted infections such as herpes simplex virus type 2 ( HSV-2 ) . Such infections with HSV-2 may facilitate the risk of HIV acquisition and often worsen the clinical course of the HIV disease [1]–[5] . In fact , HIV-1 has been recovered frequently from genital herpes lesions in co-infected individuals [6] . Although HSV target cells in tissues are still poorly understood and it is not known whether macrophages are important targets for HSV in vivo , both HIV-1 and HSV-2 can infect macrophages . Cells of the monocyte/macrophage ( M/M ) lineage reside in genital mucosal tissues and are thought to be reservoirs of HIV-1 in the genital tract [7] , [8] . Also , there is evidence that HSV infection can also stimulate macrophages in vitro and induce HIV-1 replication in these cells [9] . Thus , it would be beneficial if a microbicide has efficient suppressive activity against both HIV-1 and HSV-2 . Highly specific drugs , such as the acyclic nucleoside phosphonate ( ANP ) analogue 9- ( 2-phosphonylmethoxypropyl ) adenine [ ( R ) PMPA; tenofovir] against HIV [10] and the nucleoside analogue 9- ( 2-hydroxyethyloxymethyl ) guanine ( acyclovir; ACV ) against HSV [11] , have been developed . Unexpectedly , treatment of HIV-1/HSV-2–coinfected individuals with acyclovir diminishes both HSV-2 and HIV-1 load [12]–[15] , while topically applied tenofovir diminishes transmission not only of HIV-1 [16] but also of HSV-2 [17] . Both drugs have been found to be directly active against these two viruses [18] , [19] , although suppression of HIV-1 by acyclovir and of HSV-2 by tenofovir is suboptimal . It would be advisable if antiviral agents can be designed and developed that concomitantly display pronounced inhibitory activity against both pathogens . Therefore , the different subclasses of acyclic nucleoside phosphonates were revisited because it has been previously shown that several members of the ANPs display potent activity against a variety of DNA viruses , including herpesviruses and hepatitis B virus , and retroviruses [20]–[22] . From a wide screen of hundreds of ANPs , a compound designated 6-phosphonylmethoxyethoxy-2 , 4-diaminopyrimidine ( PMEO-DAPym ) ( Fig . 1 ) emerged as a novel prototype drug that concomitantly act as an efficient inhibitor of both HIV-1 and HSV-2 replication , but that was also surprisingly endowed with a capacity to interact with HIV entry . This combination of unique properties in one single molecule makes it a promising new-generation multitargeted antiviral for dual-viral ( HSV/HIV ) infection therapy and HIV transmission prevention .
We compared the activity of PMEO-DAPym with those of tenofovir ( Fig . 1 ) , a commonly used anti-HIV nucleotide reverse transcriptase ( RT ) inhibitor ( NtRTI ) that also shows anti-HSV-2 activity [10] , [19] and 9- ( 2-phosphonylmethoxyethyl ) adenine ( PMEA; adefovir ) ( Fig . 1 ) , the prototype compound of another subclass of the acyclic nucleoside phosphonates , which is used for the treatment of hepatitis B virus infections [22] but has also shown to efficiently suppress HIV and HSV in cell culture [20] , [21] . Human CD4+ T-lymphocyte CEM cell cultures were inoculated with HIV-1 and HIV-2 and exposed to adefovir , tenofovir , or PMEO-DAPym . All the compounds potently and comparably suppressed HIV replication with 50% effective concentration ( EC50 ) values that ranged between 0 . 36 and 1 . 9 µg/ml ( Table 1 ) . The anti-HSV activities of PMEO-DAPym , adefovir , and tenofovir were evaluated in human embryonic lung HEL cell cultures infected with wild-type laboratory HSV-1 and HSV-2 strains . PMEO-DAPym and adefovir were 20- to 40-fold superior to tenofovir against HSV: the EC50 for adefovir and PMEO-DAPym ranged between 2 . 6 and 6 . 3 µg/ml compared with 105–122 µg/ml for tenofovir ( Table 1 ) . Comparable data were obtained when tested against thirteen clinical wild-type HSV-1 and HSV-2 isolates: on average the EC50 for PMEO-DAPym was ∼2 µg/ml , and for adefovir ∼6 µg/ml , while for tenofovir it was ∼130 µg/ml ( Table 1 , Fig . 2A , B ) . Also , both adefovir and PMEO-DAPym were 20- to 65-fold more effective than tenofovir against a broad variety of thirteen clinical mutant thymidine kinase-deficient ( TK− ) HSV-1 and HSV-2 strains . These clinical strains had been isolated from patients treated with anti-herpetic nucleoside analogues . The EC50's of PMEO-DAPym ( average EC50: 2 . 4 µg/ml ) against these TK− HSV isolates were lower ( higher potency ) than those of acyclovir ( EC50: 17±1 . 9 µg/ml for HSV-1 ( TK− ) and 21±2 . 5 µg/ml for HSV-2 ( TK− ) . ( Fig . 2C , D ) . None of the compounds showed microscopical signs of cytotoxicity in the HEL cell cultures at the highest concentrations tested ( i . e . 100 µg/ml ) . Primary human keratinocytes ( PHK ) are one of the main targets for HSV infection in vivo . Whereas 20 µg/ml tenofovir hardly affected HSV-1- and HSV-2-infected organotypic epithelial raft cultures of PHKs ( <0 . 5 log10 reduction ) , a similar concentration of adefovir and PMEO-DAPym exhibited a striking 4 to 5 orders of magnitude reduction of herpesvirus replication ( Fig . 3 ) . In general , PMEO-DAPym inhibited the replication of the HSV-2 clinical isolates at EC99's ( viral load reduced by 2 orders of magnitude ) between 1 and 7 µg/ml ( Fig . 4 ) , while EC99 inhibitory concentrations for tenofovir were in the range of 250 to 300 µg/ml [19] . Also , PMEO-DAPym or adefovir solely applied prior to virus infection resulted in a marked suppression of HSV replication . Acyclovir did not create such an anti-herpetic blockade upon preincubation ( Table 2 ) . The highest tested concentrations of PMEO-DAPym ( i . e . 50 µg/ml ) that fully suppressed HSV infection in the organotypic epithelial raft cultures of the PHKs had no effect on cell differentiation as ascertained by histological examination . Monocyte/macrophage ( M/M ) lineage cells can be infected by HSV-2 and HIV-1 . M/M reside in genital herpes lesions and are thought to be reservoirs of HIV [6] , [23] . In HSV-2-infected M/M , 2- to 10-µg/ml PMEO-DAPym , adefovir and tenofovir , virtually completely inhibited HSV-2 replication . Lower concentrations were dose-dependently inhibitory , with EC50 values ranging between 0 . 02 and 0 . 2 µg/ml ( Table 1 , Table 3 ) . No effect on the viability of M/M cultures was observed for PMEO-DAPym at 100 µg/ml . Human lymphoid tissue explants that retain tissue cytoarchitecture and many of its physiological functions represent one of the most adequate laboratory models for studying viral pathogenesis [24] . Adefovir and PMEO-DAPym inhibited HSV-2 replication in this tonsilar system at an EC99 ( i . e . , viral load reduced by 2 orders of magnitude ) of 5 µg/ml and 2 µg/ml , respectively , as determined with an RT-PCR assay ( Fig . 5 ) . HSV-2 inhibition by 4 to 5 orders of magnitude was observed for both drugs at 30 µg/ml . In contrast , tenofovir comparably inhibited HSV-2 replication at much higher concentrations ( 240 µg/ml ) [19] . Also in cervico-vaginal explants , both adefovir and PMEO-DAPym at 1 µg/ml markedly inhibited HSV-2 infection ( Fig . 6 ) . To mimic more closely the in vivo situation , we coinfected human ex vivo tissue with HIV-1 and HSV-2 . In both lymphoid and cervico-vaginal tissues , the replication of both viruses was efficiently suppressed by 1 µg/ml adefovir or PMEO-DAPym as observed from the production of HSV-2 DNA and HIV-1 p24 Ag in the culture supernatants ( Fig . 6 . and 7A , B ) . In coinfected lymphoid tissues , the HSV-2 load was reduced by 1 to 1 . 5 orders of magnitude by adefovir and PMEO-DAPym , while HIV-1 was partially inhibited by either of these drugs by 2 orders of magnitude ( Fig . 7B ) . Although extremely potent ( 99 . 34% inhibition at 3 µg/mL ) , PMEO-DAPym was still less efficient to inhibit HSV-2 in human lymphoid tissue compared to ACV ( Table 4 ) . Athymic nude mice lumbosacrally scarificated with HSV-1 or HSV-2 were used for testing adefovir and PMEO-DAPym ( Fig . 8 ) . Placebo-treated mice developed lesions at the lumbosacral area leading to paralysis of the hind legs , and animal death occurred within 7 to 8 days . Treatment with either one of the drugs significantly delayed virus-related morbidity and prolonged the survival of the mice ( p<0 . 05 ) ( Fig . 8A , B ) . Although adefovir better suppressed HSV infection than PMEO-DAPym in the virus-infected mice , both drugs were superior to tenofovir in their antiviral efficacy [19] . Thus , PMEO-DAPym and to a certain extent adefovir demonstrated a unique potency in efficiently suppressing HIV and HSV in in vitro , ex vivo , and in vivo experimental systems . The pharmacologically active diphosphorylated metabolites of adefovir and PMEO-DAPym efficiently inhibited HIV RT– and HSV DNA polymerase–catalysed [3H]dATP incorporation at 50% inhibitory concentrations ( IC50 ) of 0 . 19 and 0 . 43 µg/ml for HIV-1 RT and 0 . 43 and 0 . 52 µg/ml for HSV DNA polymerase , respectively ( Tables 5 & 6 ) . IC50's for DNA polymerization by incorporation of other [3H]dNTPs by HSV DNA polymerase were , respectively , 1 . 5–2 . 1 and 3 . 1–5 . 2 µg/ml for adefovir-diphosphate and PMEO-DAPym-diphosphate , whereas in this system the IC50 for tenofovir-diphosphate was markedly higher ( 8 . 9–32 µg/ml ) ( Table 6 ) . Thus , efficient suppression of both HIV RT and HSV DNA polymerase by PMEO-DAPym explains its concomitant anti-HIV-1 and anti-herpetic activity . Although the suppression of HIV-1 RT and HSV DNA polymerase seems to be sufficient to explain the dual-targeted anti-viral activity of the tested compounds , we discovered an additional immune-modulatory mechanism for PMEO-DAPym that may further contribute to the anti-HIV-1 activity of this drug: PMEO-DAPym strikingly induces the release of anti-HIV-1 CC chemokines in peripheral blood mononuclear cells ( PBMCs ) in a dose-dependent manner ( Fig . 9A ) . At a concentration as low as 20 µg/ml , PMEO-DAPym induced 7 . 5 ng/ml MIP-1β in the PBMC cultures at 24 h after drug exposure , while at 100 and 500 µg/ml , it induced the release of 20- and 27-ng/ml MIP-1β , respectively . These PMEO-DAPym concentrations did not affect PBMC viability as measured by the trypan blue dye exclusion method and flow cytometric analysis . PMEO-DAPym also dose-dependently induced secretion of MIP-1α and RANTES , although at concentrations inferior to those of MIP-1β . PMEO-DAPym at 100 µg/ml caused a >9-fold increase in MIP-1β and MIP-1α mRNA expression ( Fig . 9B ) , pointing to the necessity of triggering new mRNA synthesis to enable the upregulation of these CC-chemokines . Neither adefovir nor tenofovir induced notable production of CC-chemokines even at 500 µg/ml . Note that the secretion of the CC-chemokines was not the result of the stimulation of activation markers , since none of the cell activation markers included in our studies ( i . e . HLA-DR , CD25 and CD69 ) ( Fig . 10 ) , neither CD3 , CD28 , CD38 , CD54 , CD71 , CD80 and CD86 ( data not shown ) were upregulated by this drug after 72 h exposure to PBMC . Importantly , PMEO-DAPym-induced production of CC chemokines was associated with a decrease of CCR5 expression on the cell surface by ≥75% at 500 and 100 µg/ml , and by 60% at 20 µg/ml ) ( Fig . 9A , 11 ) . Neither adefovir nor tenofovir induced a marked CCR5 decrease . The observed downregulation of CCR5 by PMEO-DAPym proved to be induced by the released CC-chemokines themselves , as the conditioned medium from PMEO-DAPym-exposed PBMC markedly and dose-dependently decreased CCR5 expression in freshly exposed PBMCs within 1 h of exposure time . This effect was not due to residual amounts of PMEO-DAPym in the conditioned medium , since exposure of PBMC to this drug for 1 h was not sufficient to downregulate CCR5 ( data not shown ) . Exogenous recombinant MIP-1 α ( isoform ) LD78β showed an effect on the PBMC cultures similar to that of PMEO-DAPym . In contrast , conditioned medium from either adefovir- or tenofovir-exposed PBMC was unable to decrease CCR5 expression in freshly exposed PBMCs ( Fig . 12 ) . No downregulation of CXCR4 or CXCR3 was observed ( data not shown ) , pointing to an important degree of ( co ) receptor selectivity by PMEO-DAPym . It should also be mentioned that exposure of PBMC to longer time periods than 24 h ( i . e . 72 h ) still showed a significant drop of CCR5 expression in the PBMC cultures . Indeed , CCR5 expression after 72 h was 10% , 38% and 60% of control after addition of 500 , 100 and 20 µg/ml PMEO-DAPym . Thus , the target cells do not compensate for the drug-induced CCR5 drop , and PMEO-DAPym differs in this respect with some co-receptor antagonists for which CCR5 down-regulation recovered after longer time periods . Beside β-chemokine induction , PMEO-DAPym also triggered the expression of several other chemo/cytokines and growth factors in a dose-dependent manner ( Fig . S1 ) . These stimulatory effects were less pronounced in PHA-exposed PBMC ( Fig . S2 ) .
We demonstrated the potent anti-HIV and -HSV activity of PMEO-DAPym in a diverse set of clinically relevant in vitro , ex vivo , and in vivo systems including ( i ) CD4+ T-lymphocyte ( CEM ) cell cultures , ( ii ) embryonic lung ( HEL ) cell cultures , ( iii ) organotypic epithelial raft cultures of primary human keratinocytes ( PHKs ) , ( iv ) primary human monocyte/macrophage ( M/M ) cell cultures , ( v ) human ex vivo lymphoid tissue , and ( vi ) athymic nude mice . In all assay systems the drugs were administered prior to , or at the time of , virus infection which is relevant from a microbicidal application viewpoint , for which the drugs are preferentially already present at the time of infection . However , it might well be possible that these drugs are also effective when exposed shortly after virus infection given the DNA polymerase/HIV reverse transcriptase as being one of the intracellular targets of PMEO-DAPym . In addition to our findings that PMEO-DAPym can efficiently suppress a wide variety of HSV clinical isolates including acyclovir-resistant virus strains , it has previously been demonstrated that PMEO-DAPym suppresses a wide variety of HIV-1 clinical strains belonging to different HIV-1 clades ( i . e . A , B , C , A/E , G ) [25] . This study also revealed that PMEO-DAPym showed a somewhat more favorable cross-resistance profile to various mutant HIV-1 isolates than adefovir and tenofovir . Likewise , we could demonstrate that the antivirally active PMEO-DAPym diphosphate metabolite potently inhibited mutant HIV-1 RT enzymes that harbor the tenofovir-characteristic K65R and K70R mutations ( IC50 in the submicromolar range ) ( data not shown ) explaining the efficient suppression of a variety of ( mutant ) clinical virus isolates . Thus , PMEO-DAPym may have a rather high genetic barrier suppressing a wide variety of clinical HSV and HIV-1 clade isolates , including clinically relevant mutant viruses . We investigated the underlying basis of the antiviral activity of PMEO-DAPym and found several different mechanisms that together explain the unique dual antiviral activity of this new compound: PMEO-DAPym efficiently suppressed HIV reverse transcriptase ( RT ) – and HSV-1 DNA polymerase–catalysed viral replication . Moreover , it also induced secretion of beta-chemokines , in particular MIP-1β , MIP-1α , and RANTES , whose upregulation is associated with a pronounced downmodulation of the HIV-1 coreceptor CCR5 , resulting in inhibition of HIV entry . PMEO-DAPym did not show benefit over adefovir in HSV-infected mice ( Fig . 8 ) . These experiments have been performed in immunodeficient athymic nude mice . However , PMEO-DAPym was approximately equally effective as adefovir in human cervico-vaginal tissues ex vivo ( Fig . 6 ) and slightly superior to adefovir in human tonsillar tissue ex vivo ( Fig . 7 ) . The main conclusion of the antiviral data in the ex vivo and in vivo models is that PMEO-DAPym is superior to tenofovir in its antiherpetic activity . Since it is approximately equal to tenofovir for its anti-HIV activity , PMEO-DAPym can therefore be considered as an efficient dual-targeted antiviral . Thus , this study revealed an additional and surprising property of PMEO-DAPym that is of clinical importance especially in view of its potential microbicide application . Indeed , the immunomodulatory/immunostimulatory properties of ANPs such as PMEO-DAPym may become highly relevant in case of microbicide drug development . A growing body of evidence in fact suggests that dual antiviral mechanisms should be required for effective mucosal protection [26] , including stimulation of innate antiviral factors such as β-chemokines that down-regulate the HIV CCR5 co-receptor [27] , [28] . It was shown that RANTES , MIP-1α and MIP-1β are significantly associated with protection against rectal mucosal SIV infection [29] . RANTES derivatives are currently considered for microbicide formulation . Dudley et al . [30] could show that chemically modified RANTES , PSC-RANTES , binds to CCR5 , inhibits HIV-1 entry and dose-dependently protects rhesus macaques from SHIV infection as a vaginal microbicide . Ahmed et al . [31] demonstrated that spontaneous production of RANTES and IFN-γ correlated with protection against SIVsm challenge of Cynomolgus macaques . Beside the pronounced stimulatory effect of PMEO-DAPym on MIP-1α and MIP-1β , and to a lesser extent RANTES , it also stimulates the production of a variety of other interleukins and several growth factors . Its stimulatory effect on the mitogenic IL-2 and IL-4 chemokines , as well as IL-7 that inhibits apoptosis of HIV-infected cells and stimulates T-cell proliferation [32] , is very minor . Also , PMEO-DAPym did not affect IP-10 levels . This chemokine is a potent chemotactic factor for lymphocytes and monocytes [33] and thus , the drug is not expected to attract these cells to the genital sub-mucosa through IP-10 induction . Instead , the induction of proinflammatory interleukins such as IL-6 and IL-8 is more pronounced in cells treated with PMEO-DAPym . Also production of G-CSF and IL-1β was markedly enhanced in the drug-exposed PBMC cultures . In contrast with the activation of P2Y2 nucleotide receptors with UTP resulting in an intracellular Ca++ response to activation of CXCR2 in HEK cells [34] , the broad cellular chemokine response triggered by PMEO-DAPym does not induce Ca++ release in drug-exposed PBMC ( data not shown ) . However , the relevance of the stimulatory effect on these interleukins in PBMC by PMEO-DAPym suggesting potential side effects of the drug is currently still unclear . In fact , PRO-2000 , a polyanionic drug that has been exposed to healthy individuals for a prolonged time period as a potential microbicide drug also markedly stimulated IL-1ra , IL-6 and IL-8 , G-CSF and GM-CSF [35] in PBMC , but was never found to display any signs of toxicity or HIV-promoting transmission in the drug-treated individuals [36] . The fact that ( i ) the CC chemokines RANTES , MIP-1α and MIP-1β act as potent natural inhibitors of HIV-1 [37] , [38] , ( ii ) such chemokines are indeed protective to SIV challenge in monkeys by acting as virus entry inhibitors [29] , [39] , ( iii ) ANPs , in particular PMEO-DAPym , may behave as natural chemokine production enhancers in PBMC and M/M in addition to their viral reverse transcriptase/DNA polymerase inhibition , ( iv ) the increased β-chemokine production found in PMEO-DAPym-exposed PBMC cultures coincided with CCR5 co-receptor down-regulation and ( v ) the supernatants of these drug-exposed cell cultures were able to induce an antiviral refractory state when administered to new non-treated cell cultures , point to an additional and important antiviral ( i . e . HIV ) potential of PMEO-DAPym through its immunomodulatory properties . In our experiments , it was shown that 500 and 100 µg/ml PMEO-DAPym down-modulated CCR5 by ≥75% . This extent of CCR5 down-regulation might be sufficient to affect virus infection , since studies with anti-CCR5 antibodies have demonstrated that their exposure to PBMC can drop the mean surface expression of CCR5 from CD4+ T-lymphocytes or U87-CCR5 cell cultures to 19–37% of control . In such cases , the antibodies could efficiently neutralize R5 virus infection [40] . Although the exact molecular mechanism of action of PMEO-DAPym is yet to be revealed , it is clear that its action on CCR5 down-regulation by stimulation of β-chemokine mRNA and subsequent enhanced β-chemokine induction is different from the previously reported down-modulation of CCR5 ( and concomitant anti-HIV activity ) by rapamycin . The latter drug , that disrupts IL-2 receptor signaling , interferes with CCR5 expression at the transcriptional level , and the reduced expression of CCR5 on PBMCs is then associated with increased extracellular levels of MIP-1α and MIP-1β [41] . It should be kept in mind that intravaginal 1% gel applications ( as used in the CAPRISA 004 trial ) afford high local drug concentrations at the cervicovaginal tissue that easily exceed those used in our study [42] , [43] . Therefore , it is expected that such locally high drug concentrations may efficiently induce β-chemokine production that , in turn , will be able to down-regulate CCR5 expression , creating a local transient antiviral state in addition to efficient inhibition of the viral DNA polymerases of HIV and HSV . It should be emphasized that our study has not addressed other key aspects that are important for eventual clinical efficacy of a drug , such as the optimal drug concentration that can be reached , the drug formulation , its pharmacokinetics and pharmacodynamics , and potential drug-resistance development . These aspects are somewhat complicated given the fact that PMEO-DAPym targets two different viruses and regarding the anti-HIV activity , PMEO-DAPym is directed against two different events in the viral infection/replication cycle . The antiherpetic activity of PMEO-DAPym is inferior to that of acyclovir ( Table 1 , 3 and 4 ) . However , in human tissues ex vivo at the concentration of 3 µg/mL both drugs suppressed HSV-2 by more than 99% . Future modifications of the drug may bring it to an anti-HSV activity even closer to that of acyclovir , the highly efficient anti-HSV compound and the current drug of choice to treat HSV-2 infection . The key accomplishments in designing PMEO-DAPym is its anti-HIV activity comparable to that of tenofovir combined with a marked ( ∼30- to 70-fold ) superior activity against HSV compared with tenofovir . Since tenofovir proved already effective to prevent HSV infection by 51% in the CAPRISA trial , it is expected that PMEO-DAPym will perform markedly better than tenofovir against HSV infection/transmission based on our in vitro , ex vivo and in vivo data . Although it would be preferable to develop a drug that concomitantly has potent anti-HIV and anti-HSV activity as the best in their class , it might be sufficient to have moderate activity against one of these viruses in case of microbicides given the topical administration modality that allows to reach high local ( and therefore very effective ) drug concentrations , as already shown for tenofovir in the CAPRISA trial [16] , [17] . Also an important advantage of PMEO-DAPym is its suppression of CCR5-tropic HIV , the type of HIV-1 that is predominantly transmitted . In conclusion , we defined a distinct new subclass of ANPs , structurally and functionally different from tenofovir and adefovir , that has significant advantage over the commonly used drugs . The prototype drug , PMEO-DAPym , efficiently suppresses HSV DNA polymerase and also retains the ability of tenofovir and adefovir to markedly suppress HIV-1 RT . In addition , and unlike adefovir and tenofovir , the drug also induces anti-HIV CC-chemokines , which downmodulate the CCR5 coreceptor . Thus , PMEO-DAPym combines various anti-HIV and anti-HSV activities in one molecule and concomitantly targets HIV entry and viral polymerase-catalysed HIV/HSV replication . The efficiency of this new class of antivirals ( even at concentrations much lower than the ones achievable in vaginal application ) [43] makes it a promising new-generation multitargeted and multifunctional antiviral for dual-viral ( HSV/HIV ) infection therapy and HIV transmission prevention .
All animal work was approved by the Katholieke Universiteit Leuven Ethics Committee for Animal Care and Use ( Permit number: P097-2010 ) . All animal guidelines and policies were in accordance with the Belgian Royal Decree of 14 November 1993 concerning the protection of laboratory animals and the European Directive 86-609-EEC for the protection of vertebrate animals used for experimental and other scientific purposes . Virus isolates were obtained as part of a translational research program ( www . regavir . org ) granted by the Belgian Ministry of Health as part of the National Cancer Plan for the diagnosis of drug resistance in herpesviruses . All viruses were obtained and used as approved by the Belgian IRB equivalent ( Departement Leefmilieu , Natuur en Energie , protocol SBB 219 2011/0011 , and the Biosafety Committee KU Leuven ) . Human embryonic lung HEL-299 fibroblasts were obtained from ATCC . Primary human keratinocytes ( PHKs ) were isolated from neonatal foreskins and cultured as previously described [44] . The TZM-Bl cells [45] were kindly provided by Dr . G . Van Ham ( ITG , Antwerp , Belgium ) . Buffy coat preparations from healthy donors were obtained from the Blood Transfusion Center in Leuven , Belgium . PBMC were isolated by density gradient centrifugation over Lymphoprep ( d = 1 . 077 g/ml ) ( Nycomed , Oslo , Norway ) and cultured in cell culture medium ( RPMI 1640 ) containing 10% FCS and 2 mM L-glutamine . The healthy donors were anonymous . The laboratory HSV-1 strain KOS and the HSV-2 strain G were used as reference herpesviruses . Several clinical isolates of wild-type HSV-1 [RV-6 , RV-132 , RV-134 , C559143 , RV-174 , RV-175] , thymidine kinase-deficient ( TK− ) HSV-1 [RV-28 , RV-36 , RV-117 , 328058 , RV-179 , RV-294] , wild-type HSV-2 [RV-24 , RV-124 , RV-194 , NA , PB , NS , HSV-47] , and HSV-2 TK− [RV-101 , RV-129 , BA 19026589 , LU C557672 , HSV-44 , RV-184 , RV-185] from virus-infected individuals in Belgium were used . Viral TK sequences were deposited in Genbank ( GenBank accession JN415116-JN415119 for HSV-1 mutants and JN415120–JN415126 for HSV-2 mutants ) . HIV-1 strains IIIB and Ba-L were provided by Drs . R . C . Gallo and M . Popovic ( at that time at the National Institutes of Health , Bethesda , MD ) and HIV-2 ( ROD ) was obtained from Dr . L . Montagnier ( at that time at the Pasteur Institute , Paris , France ) . The sources of the compounds were as follows: acyclovir [ACV , 9- ( 2-hydroxyethoxymethyl ) -guanine] , GlaxoSmithKline , Stevenage , UK; PMEA [adefovir , 9-[2- ( phosphonylmethoxyethyl ) -adenine]; and ( R ) -PMPA [tenofovir , ( R ) -9-[2- ( phosphonylmethoxypropyl ) adenine]] , Gilead Sciences , Foster City , CA . PMEO-DAPym was synthesized by A . Holý ( Prague , Czech Republic ) [46] , [47] . PMEO-DAPym was also synthesized and provided by Shanghai Medicilon Inc . , Shanghai , China . Tenofovir diphosphate ( tenofovir-DP ) and adefovir diphosphate ( adefovir-DP ) were obtained from Moravek Biochemicals , Brea , CA , and the diphosphate of PMEO-DAPym was synthesized by A . Holý , Prague , Czech Republic . The stock solutions of PMEA , PMPA , and PMEO-DAPym ( 10 mg/ml ) were tested for endotoxin content with the Limulus Amebocyte Lysate assay ( Cambrex Bioscience , Verviers , Belgium ) and were found to contain <1 ng/ml endotoxin . The HSV-induced cytopathic effect ( CPE ) was evaluated in HSV-infected HEL and PHK cultures as described [44] , [48] . Briefly , cells were infected with each viral strain at 100 tissue culture infective dose-50% ( TCID50 ) ( 1 TCID50 being the 50% tissue culture infective dose , or virus dose required to infect 50% of the virus-exposed cell cultures ) and cultured in 96-well microtiter plates for 3 days in the presence of several concentrations of the test compounds . After the incubation period at 37°C , CPE was visually assessed , and the 50% effective concentration ( EC50 , the compound concentration required to reduce viral CPE by 50% ) was determined . These assays were carried out in HEL cell monolayers at different times post infection . Cells were grown in 24-well microtiter plates and infected with one HSV-1 ( RV-174 ) or two HSV-2 clinical isolates ( NS and RV-124 ) at the indicated multiplicity of infection ( m . o . i . ) . After 2 h at 37°C , the cells were washed and medium containing different concentrations of PMEO-DAPym ( in duplicate ) was added . Following 24 , 48 , and 72 h of incubation , we released the viruses by freeze-thawing and then titrated them using a plaque assay in HEL cells . The EC90 and EC99 are defined as the drug concentrations causing a 90% ( one order of magnitude ) or 99% ( two orders of magnitude ) reduction , respectively , in virus production as measured following viral titration by plaque assay . We obtained human PBMC from the blood of healthy seronegative donors using Ficoll–Hypaque density gradient centrifugation . The PBMC were resuspended in RPMI 1640 medium supplemented with 20% heat-inactivated serum and then seeded into 48-well plates ( 1 . 8×106 cells/well ) . M/M were separated by adherence onto plastic . After 5 days , non-adherent cells were carefully removed by repeated gentle washings with warm medium , and adherent M/M were cultured for an additional 3 days to mature and to form a monolayer . M/M were estimated to be 105 cells/well at the time of infection . To evaluate the anti-HSV-2 activity of PMEO-DAPym on human macrophages , we added the compound to macrophages 1 h before infection at a variety of concentrations ( 0 . 0032 , 0 . 016 , 0 . 08 , 0 . 4 , 2 , 10 , or 50 µg/ml ) . Similarly , macrophage cultures were treated with the same concentrations of adefovir and tenofovir . Macrophage cultures were then infected with HSV-2 ( G ) ( 100 TCID50 ) in the presence of the compounds . After 2 h of adsorption , the cultures were extensively washed to remove any residual virus particles . Fresh complete medium and compounds , at the established concentrations , were then added to the cultures . Appropriate positive ( infected but not treated ) and mock-infected negative ( uninfected and untreated ) M/M controls were run for each experiment as well . All assays were performed in triplicate . The HSV-2-induced cytopathic effect was found to be complete 120 h after virus challenge of infected but untreated M/M [49] . We determined the amount of infectious virus in the supernatants 5–6 days after infection using the cytopathic effect assay . Tenfold serially diluted supernatants were added to confluent monolayers of Vero cells in 96-well plates ( 100 µl/well , 6 parallel wells ) . The plates were incubated at 37°C for 4 or 5 days , at which time the cytopathic effect was demonstrated . The titers of virus produced were calculated according to the Reed and Muench method and expressed as 50% tissue culture infective dose per mL ( TCID50/mL ) . Primary human keratinocytes ( PHKs ) were seeded on top of collagen gels in 24-well microtiter plates and maintained submerged for 24–48 h . The collagen rafts were then raised and placed into stainless-steel grids at the interface between air and the liquid culture medium . The epithelial cells were allowed to stratify . Rafts were infected with 5 , 000 PFU of HSV-1KOS or HSV-2G at 10 days post lifting and treated with adefovir or PMEO-DAPym . Five days later , one series of rafts was fixed in 10% buffered formalin , embedded in paraffin , and stained with hematoxylin and eosin for histological evaluation . Another series of rafts was used to quantify virus production . For that purpose , each raft was frozen in 3 ml of phosphate buffered saline ( PBS ) and thawed to release the virus from the infected epithelium . We clarified supernatants by centrifugation at 1 , 800 rpm and titrated them using a plaque assay in HEL cells . Titers were calculated as plaque forming units ( PFU ) per ml of virus suspension . Virus production per raft was then calculated . Two rafts per drug concentration were used to determine the effects of the compounds on virus yield . Human tonsillar tissues were obtained from patients undergoing routine tonsillectomy at the Children's National Medical Center ( Washington , DC ) under IRB-approved protocol . Cervical tissues were obtained through the National Disease Research Interchange ( NDRI , Philadelphia , PA ) . Tissues were dissected into blocks of about 2×2×2 mm and placed onto collagen sponge gels in culture medium at the air-liquid interface as described earlier [24] . For tonsillar tissue for each experimental condition , 27 tissue blocks ( 9 blocks/well filled with 3 ml of medium ) were inoculated with HSV-1 ( strain F ) or HSV-2 ( strains G and MS ) . In coinfection experiments , tissue blocks were sequentially infected with HSV-2 ( G ) and HIV-1 ( IIIB ) ( obtained from the Rush University Virology Quality Assurance Laboratory , Chicago , IL ) . Drugs ( adefovir and PMEO-DAPym ) were added to the culture medium 12 h prior to viral infection and replenished at each culture medium change . For cervico-vaginal tissue , 16 blocks were infected by immersion in HSV-2G-containing medium and maintained on the gelfoam rafts . Adefovir was added at the time of infection and replenished at each culture medium change . HSV replication was evaluated from the release of viral DNA into the culture medium as measured with quantitative real-time PCR [18] . We evaluated HIV-1 replication from the release of p24 capsid antigen using a bead-based assay [50] . Female adult NMRI athymic nude mice or hairless mice ( weighing ∼20 g and approximately 4 weeks old ) were scarified on the lumbosacral area over a surface of about 1 cm2 with 5×103 PFU of HSV-1 ( KOS ) or 5×102 PFU of HSV-2 ( G ) . Formulations of adefovir and PMEO-DAPym ( 0 . 3% and 1% ) in a gel ( identical to that used in the CAPRISA 004 trial ) were applied topically twice a day for a period of 5 days starting 1–2 h post infection . In each experiment , a group of animals was treated with a placebo formulation that contained exactly the same vehicle but without drug . All animal procedures were approved by the KU Leuven Animal Care Committee . Development of lesions and mortality were recorded over a 1-month period . We estimated survival rates according to the Kaplan-Meier method and compared them using the log-rank test ( Mantel-Cox ) with GraphPad Prism . Animals' care was in accordance with institutional guidelines and the ethical committee of the KU Leuven . The reaction mixture ( 40 µl ) for the HSV-1 DNA polymerase and HIV-1 RT assays contained 4 µl of Premix ( 200 mM Tris . HCl , pH 7 . 5; 2 mM DTT; 30 mM MgCl2 ) , 4 µl of BSA ( 5 mg/ml ) , 1 . 6 µl of activated calf thymus DNA ( 1 . 25 mg/ml ) , 0 . 8 µl of dCTP ( 5 mM ) , 0 . 8 µl of dTTP ( 5 mM ) , 0 . 8 µl of dGTP ( 5 mM ) , 2 µl of radiolabeled [3H]dATP ( 1 mCi/ml ) ( 3 . 2 µM ) , 18 µl of H2O , and 4 µl of adefovir-pp or PMEO-DAPym-pp at different concentrations ( i . e . , 200 , 20 , 2 , and 0 . 2 µM ) . In the HSV DNA polymerase assays , the inhibitory effect of adefovir-pp or PMEA-DAPym-pp on herpesvirus DNA polymerase-catalyzed polymerization was also determined in the presence of radiolabeled [3H]dGTP ( 1 mCi/ml; 2 . 8 µM ) , [3H]dTTP ( 1 mCi/ml; 1 µM ) , or [3H]dCTP ( 1 mCi/ml; 2 . 5 µM ) as competing dNTP in the presence of 0 . 8 µl ( 5 mM ) of the other dNTPs in the reaction mixture as described above . The reaction was started by the addition of 4 µl of recombinant HSV-1 DNA polymerase ( kindly provided by M . W . Wathen , at that time at Pfizer , Kalamazoo , MI ) or 4 µl of recombinant HIV-1 RT ( in 20 mM Tris . HCl , pH 8 . 0; 1 mM DTT; 0 . 1 mM EDTA; 0 . 2 M NaCl; 40% glycerol ) , and the reaction mixture was incubated for 60 min ( HSV-1 DNA polymerase ) or 30 min ( HIV-1 RT ) at 37°C . Then , 1 ml of ice-cold 5% TCA in 0 . 02 M Na4P2O7 . 10 H2O was added to terminate the polymerisation reaction , after which the acid-insoluble precipitate ( radiolabeled DNA ) was captured onto Whatman glass fiber filters ( type GF/C; GE Healthcare UK Limited , Buckinghamshire , UK ) and further washed with 5% TCA and ethanol to remove free radiolabeled dNTP . Radioactivity was determined in a Perkin Elmer Tri-Carb 2810 TR liquid scintillation counter . PBMC were cultured in the presence of adefovir , tenofovir , and PMEO-DAPym and incubated at 37°C in a humidified , 5% CO2-controlled atmosphere . The expression of cellular activation markers was measured after 3 days of culture . Briefly , after washing the cells with PBS containing 2% FCS , we incubated them with PerCP-conjugated anti-CD4 mAb ( clone SK3 ) in combination with PE-conjugated anti-CD25 , anti-CD69 , or anti-HLA-DR mAbs for 30 min at room temperature . For aspecific background staining , cells were stained in parallel with Simultest Control IgG γ1/γ2a and PerCP-conjugated mouse IgG1 . Finally , the cells were washed with PBS , fixed with 1% formaldehyde solution , and analyzed with a FACSCalibur ( BD Biosciences , San Jose , CA ) ; data were acquired with CellQuest software and further analyzed with the FLOWJO software ( Tree Star , San Carlos , CA ) . The expression of the chemokine receptors was measured after 24 h , and PBMC were incubated with PerCP-conjugated anti-CD4 mAb ( clone SK3 ) in combination with APC-conjugated anti-CXCR4 mAb ( clone 12G5 ) and PE-conjugated anti-CCR5 ( clone 2D7 ) or anti-CXCR3 ( clone 1C6 ) . To evaluate the biological effects of the β-chemokines produced by drug-treated PBMC , freshly isolated PBMC were incubated for 1 h at 37°C with supernatants collected from PBMC cultures treated with compounds for 24 h . Then , cells were incubated with PerCP-conjugated anti-CD4 mAb ( clone SK3 ) in combination with PE-conjugated anti-CCR5 mAb ( clone 2D7 ) for 30 min at room temperature . As a control , cells were also incubated for 1 h at 37°C with the LD78β isoform of MIP-1 alpha ( PeproTech , London , United Kingdom ) , which is described as potently downregulating the CCR5 receptor [51] . All mAbs were purchased from BD Biosciences ( Erembodegem , Belgium ) . Freshly isolated PBMC were cultured in the presence of adefovir , tenofovir , and PMEO-DAPym , and culture supernatants were collected after 24 h . We determined the cytokine production profile using the Bio-Plex 200 system ( Bio-Rad , Hercules , CA ) and a Bio-Plex Human chemokine assay according to the manufacturer's instructions and as described in detail earlier [52] . The assay kit contains beads conjugated with mAbs specific for interferon-inducible protein-10 ( IP-10 ) , macrophage inflammatory protein-1α ( MIP-α ) , and MIP-1β , and regulated on activation normal T-cell expressed and secreted ( RANTES ) . For each chemokine , eight standards ranging from approximately 1 . 5 pg/ml to 32 , 000 pg/ml were constructed , and the minimum detectable dose was between 1 . 5 pg/ml and 8 pg/ml . Standard curves and the concentrations of chemokines within samples were generated with the Bio-Plex Manager 4 . 1 software . We isolated total RNA from approximately 106 PBMC using the RNeasy Mini Kit ( QIAGEN , Hilden , Germany ) . To eliminate potential genomic DNA contamination , the samples were treated with DNase I ( Roche ) . Total RNA ( 300 ng ) was reverse-transcribed to cDNA by means of Moloney murine leukemia virus reverse transcriptase ( Invitrogen ) and Hexamer Primer ( Invitrogen ) according to the manufacturer's instructions . For RT-PCR , the primers used were 5′-GCAACCAGTTCTCTGCATCA-3′ ( sense ) and 5′-TTCTGGACCCACTCCTCACT-3′ ( antisense ) for human CCL3/MIP-1α; 5′-AAGCTCTGCGTGACTGTCCT-3′ ( sense ) and 5′-CCAGGATTCACTGGGATCAG-3′ ( antisense ) for human CCL4/MIP-1β; 5′-CGCTGTCATCCTCATTGCTA-3′ ( sense ) and 5′-ACTCCCGAACCCATTTCTTC-3′ ( antisense ) for human CCL5/RANTES; 5′-ATCCTCACCCTGAAGTACCCCA-3′ ( sense ) and 5′-GAAGGTCTCAAACATGATCTGGGT-3′ ( antisense ) for human β-actin; and 5′-CGAGATCCCTCCAAAATCAA-3′ ( sense ) and 5′-ACAGTCTTCTGGGTGGCAGT-3′ ( antisense ) for human glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) . The reaction mixtures contained dNTPs ( Invitrogen ) at 100 nM , each of the forward and reverse primers at 0 . 5 µM , and 0 . 5 U of SuperTaq DNA polymerase ( HT Biotechnology , Cambridge , UK ) in a total volume of 50 µl . After electrophoresis through a 2% agarose gel , the amplified cDNA fragments were visualized with ethidium bromide . | To contain the HIV-1 epidemic , it is necessary to develop antivirals that prevent HIV-1 transmission . It is well known that HIV infection might be accompanied by other pathogens , which often are engaged with HIV-1 in a vicious circle of mutual facilitation . One of the most common of these pathogens is herpes simplex virus ( HSV ) type 2 . Since there is an urgent need for a next generation antivirals that are multi-targeted , we can now report on the development of the first antiviral of this new generation that efficiently suppresses both HIV-1 and HSV-2 . We found that the dual-targeted antiviral drug affects several targets for viral replication . It uniquely combines in one molecule three important abilities: ( i ) to efficiently suppress HSV-encoded DNA polymerase , ( ii ) to efficiently suppress HIV-1-encoded reverse transcriptase , and ( iii ) to stimulate secretion of CC-chemokines that downregulate the HIV-1 coreceptor CCR5 . The compound suppresses both viruses in a wide-range of in vitro , ex vivo , and in vivo experimental models . The ability of one molecule to suppress HIV-1 and HSV-2 by combining direct activity against their two key enzymes and indirect immunomodulatory effects is unique in the antiviral field . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"medicine",
"infectious",
"diseases",
"drug",
"research",
"and",
"development",
"herpes",
"simplex",
"drugs",
"and",
"devices",
"hiv",
"viral",
"diseases",
"drug",
"discovery"
]
| 2013 | A Multi-targeted Drug Candidate with Dual Anti-HIV and Anti-HSV Activity |
Due to the recent advances in high-throughput sequencing technologies , it becomes possible to directly analyze microbial communities in human body and environment . To understand how microbial communities adapt , develop , and interact with the human body and the surrounding environment , one of the fundamental challenges is to infer the interactions among different microbes . However , due to the compositional and high-dimensional nature of microbial data , statistical inference cannot offer reliable results . Consequently , new approaches that can accurately and robustly estimate the associations ( putative interactions ) among microbes are needed to analyze such compositional and high-dimensional data . We propose a novel framework called Microbial Prior Lasso ( MPLasso ) which integrates graph learning algorithm with microbial co-occurrences and associations obtained from scientific literature by using automated text mining . We show that MPLasso outperforms existing models in terms of accuracy , microbial network recovery rate , and reproducibility . Furthermore , the association networks we obtain from the Human Microbiome Project datasets show credible results when compared against laboratory data .
Microbes play an important role both in environment and human life . However , the way microbes affect the human health remains largely unknown . Knowledge of the microbial interactions can provide a solid foundation to model the interplay between the ( host ) human body and the microbial populations; this can serve as a key step towards precision medicine [1] . Unfortunately , understanding microbes interactions is difficult , as most microbes cannot be easily cultivated in standard laboratory settings . However , the recent increase of quality and reduced costs of sequencing technologies ( e . g . , shotgun or PCR directed sequencing [2] ) enable researchers to collect information from the entire genome of all microbes under different environment conditions . As a result , various datasets ranging from earth ecosystem to human microbiome have been made publicly available under the Human Microbiome Project [3] or the Earth Microbiome Project [4] . In this paper , we aim at analyzing the networks of associations ( putative interactions ) among the microbes of human microbiome in order to understand how microbes can affect the human health . To this end , there exist several challenges: First , the amount of sequenced data that corresponds to human microbiome available from public websites is scarce . To date , one of the largest metagenomic datasets of human niches is the NIH Human Microbiome Project ( HMP ) [3] which only provides a few hundreds of healthy individual samples ( n ) of various body sites , while the number of measured microbes ( p ) usually ranges from hundreds to thousands . As a consequence , the number of associations ( p ( p − 1 ) /2 ) is much greater than the number of samples ( i . e . , high-dimensional data ) . Another big challenge stems from the nature of the data itself . Sequencing data only provides the relative abundance of various species; this is because the sequencing results are a function of sequencing depth and the biological sample size [5] . Therefore , from a statistical standpoint , the relative taxon abundance falls into the class of compositional data [6]; this causes statistical methods such as Pearson or Spearman correlations ( which work with absolute values ) to generate spurious results . To infer microbe associations for both compositional and high-dimensional data , several algorithms have been developed . A pioneering method called SparCC [7] applies log-ratio transform on compositional data and directly approximates the correlation among microbes based on sparsity assumption of microbial associations . However , SparCC does not consider the influence of errors in compositional data; this may reduce the correlation estimation accuracy . More precisely , SparCC approximates the basis variance ( i . e . , the variance of compositional data ) under the assumption that average correlations are small . Second , the iterative procedure used to estimate the magnitude of correlations can exceed value 1; this may cause poor approximations if one tries to remedy the problem by setting up the threshold value to 1 or -1 for the estimated correlations; these series of approximations may reduce the correlation estimation accuracy quite significantly . SPIEC-EASI [8] calculates the covariance of the log-ratio transformed data to approximate the covariance of the absolute abundance of microbes; then , it uses either neighborhood selection ( mb ) [9] or graphical Lasso ( gl ) [10] to estimate the conditional dependencies among microbes . CCLasso [11] is similar to SPIEC-EASI which applies log-ratio transform on compositional data and imposes a l1 penalty on the inverse covariance matrix of the microbes and then solves it to obtain a sparse covariance matrix . However , it is not clear whether or not CCLasso can obtain a consistent estimator on the inferred microbial covariance without showing consistency analysis ( see consistency analysis for graphical Lasso in S1 File section 8 ) . We note that although the above methods can estimate the covariance among microbes under the sparsity assumption , they still have major difficulties to infer the associations among microbes given such high-dimensional data . To solve the problem caused by high-dimensional data , we propose to integrate multiple levels of biological information to enhance the model accuracy on inferring microbial associations . Indeed , an increasing amount of scientific literature provides a large amount of data which can be mined not only for the co-occurrence of microbes , but also to predict microbes associations directly . For instance , pioneering work [12] considers automated analysis of the co-occurrence of bacterial species through statistical testing approaches ( e . g . , Fisher’s exact test ) . Recently , Lim et al . [13] incorporated machine learning techniques to automatically identify and extract microbial associations directly from the abstracts of scientific papers . Finally , Wang et al . [14] and Li et al . [15] use prior biological knowledge to reconstruct genes interaction networks . To the best of knowledge , we are the first to consider experimentally verified biological knowledge as a priori information to derive microbial association networks . To this end , we transform the original problem of microbial associations estimation into a graph structure learning problem where nodes represent microbes and edges represent ( pairwise ) associations among microbes . With this new problem formulation , the graphical Lasso algorithm becomes suitable to infer the microbial association network . We also integrate the text mining results from the scientific literature as prior knowledge for inferring the microbes graph structure; the proposed algorithm Microbial Prior Lasso ( MPLasso ) turns out to be more accurate than other existing methods on inferring the microbial associations . The proposed MPLasso pipeline is shown in Fig 1 . We assess the performance of MPLasso in the presence of prior knowledge by first comparing it against other previously proposed methods ( e . g . , CCLasso , REBACCA [16] , SparCC , SPIEC-EASI , and CCREPE [17] ) through synthetic data generated from different graph structures ( run time comparisons of existing methods are summarized in S1 File section 1 and S1 Table ) . We show that our proposed MPLasso outperforms all these methods in terms of area under the precision-recall curve ( AUPR ) and accuracy ( ACC ) of network associations prediction . Next , we evaluate the HMP datasets of two different sequencing techniques ( shotgun and 16S ribosomal RNA ( rRNA ) ) at five different body sites and compare the reproducibility of the estimated results . Taken together , our contributions are three fold:
In this paper , we consider high-throughput comparative metagenomic data obtained from the next-generation sequencing ( NGS ) platforms . More specifically , two types of gene sequencing data are considered: 16S rRNA and shotgun data . Shotgun data analyses are accomplished by unrestricted sequencing of the genome of all microorganisms present in a sample; on the contrary , the domain of 16S rRNA is restricted to bacteria and archaea . Data obtained from the human microbiome project ( HMP ) have a curated collection of sequence of microorganisms associated with the human body from both shotgun and 16S sequencing technologies . For the 16S rRNA data , we consider the high-quality sequencing reads in 16S variable regions 3-5 ( V35 ) of HMP healthy individuals from Phase one production study ( May 1 , 2010 ) . The taxonomy classification of the 16S rRNA are performed using either mothur ( HMMCP ) [18] or QIIME ( HMQCP ) [19] pipelines . The resulting table for operational taxonomic units ( OTUs ) at each body site of the human samples can be obtained from http://hmpdacc . org/HMMCP/ and http://hmpdacc . org/HMQCP/ . For the shotgun data ( HMASM ) , we obtain data from http://hmpdacc . org/HMASM/ and use the trimmed sequences as inputs to the metaphlan2 [20] pipeline which can generate the OTU abundance for each sample . The OTU table can be represented by a matrix D ∈ ℕn×p where ℕ represents the set of natural numbers . d i = [ d 1 i , d 2 i , … , d p i ] denotes the p-dimensional row vector of OTU counts from the ith sample ( i = 1 , … , n ) . To account for different sequencing depths for each sample , the raw count data ( di ) are typically transformed into relative abundances ( x ) by using log-ratio transform [6] . Statistical inference on the log-ratio transform of the compositional data ( x ) can be shown to be equivalent to the log-ratio transform on the unobserved absolute abundance ( d ) as: log ( x i x j ) = log ( d i / m d j / m ) = log ( d i d j ) . Here , we apply the centered log-ratio ( clr ) transform as follows: c = clr ( x ) = [ log ( x 1 m ( x ) ) , log ( x 2 m ( x ) ) , … , log ( x p m ( x ) ) ] ( 1 ) where m ( x ) = ( ∏ i = 1 p x i ) 1 p is the geometric mean of the composition vector x . The resulting vector c is constrained to be a zero sum vector . The covariance matrix of the clr transform C = Cov[clr ( c ) ] can be related to the covariance matrix of the log-transformed absolute abundances Γ = Cov[log D] via the relationship [6 , 8] C = UΓU , where U = I p - 1 p J , where Ip is the p-dimensional identity matrix , and J is the p-dimensional all-ones vector . For the case where p > > 0 , the finite sample estimator ( C ^ ) serves as a good approximation of Γ ^; therefore , the finite sample estimator ( C ^ ) serves as the basis on inferring the correlations among microbes . To account for the zero counts in samples , we add pseudo count to the original count data to avoid numerical issues when using the clr transform . To infer the pairwise associations among microbes , we can transform the original inferring problem into a graph learning problem where each node represents an OTU ( e . g . , taxon ) and each edge represents a pairwise association between microbes; the resulting graph is an undirected graph G = ( V , E ) , where V and E represent the node and edge sets , respectively . Suppose the observed data ( d ) are drawn from a multivariate normal distribution N ( d|μ , Σ ) with mean μ and covariance Σ . The inverse covariance matrix ( precision matrix ) Ω = Σ−1 encodes the conditional independence among nodes . More specifically , if the entry ( i , j ) of the precision matrix Ωi , j = 0 , then node i and node j are conditionally independent ( given the other nodes ) and there is no edge among them ( i . e . , Ei , j = 0 ) . However , microbial data usually come with a finite amount of samples ( n ) but with high dimensionality ( p ) ; this makes the graph inferring problem intractable since the number of variables ( p ( p - 1 ) 2 ) is greater than n . To solve this problem , an important assumption that needs to be made is to assume that the underlying ( true ) graph is reasonably sparse . One suitable algorithm to select the precision matrix under sparsity assumption is to utilize the graphical Lasso proposed previously [8 , 10] . As shown in Fig 2 , we propose to utilize the information obtained from the scientific literature in order to construct the prior matrix P ∈ ℝp×p , where each entry Pi , j ∈ [0 , 1] represents the prior probability of associations between taxon i and taxon j . We can impose different amounts of penalties on the precision matrix; this is different from the standard formulation where the penalty ( ρ ) imposed on the precision matrix is the same . Therefore , by incorporating the prior information into the penalty matrix ( P ) , the proposed MPLasso can be formulated as follows: Ω ^ = arg max Ω { log det ( Ω ) - tr ( Ω C ^ ) - ρ | P ⊗ Ω | 1 } ( 2 ) where C ^ is the empirical covariance of the microbial data , and Ω is the precision matrix of the estimated associations among microbes . Here det and tr denote the determinant and the trace of a matrix , respectively . |Ω|1 is the L1 norm , i . e . , the sum of the absolute values of the elements of Ω and ⊗ represents the component-wise multiplication . When the value of Pi , j is large , this directly puts a heavy penalty and represents a weaker association between taxa and vice versa . This way , by imposing the prior information , we can accurately infer the associations among microbes . We extract two types of data to be used as priors for our model . One type of data is from the microbial co-occurrence in literature that examines the number of abstracts where two taxa appear together and compares this to random chance . The second type of data is from the machine learning-based method that extracts the full details of the interaction , including the sign and direction of the interaction . To acquire the prior knowledge ( P ) of microbial associations from reported experiments and published papers , we utilize the PubMed database ( https://www . ncbi . nlm . nih . gov/pubmed/ ) that contains a massive amount of papers with abstracts . For the 16S rRNA data where the taxonomy level can only be achieved at the genus level , we adopt the statistical testing method ( i . e . , Fisher’s exact ) [12] to identify the pairwise associations derived from the microbial co-occurrence in literature . On the other hand , for the shotgun data where the taxonomy level can be up to species level , we adopt both the microbial co-occurrence in literature and the machine-learning-based methods [13] to obtain such associations . We modify the code available on https://github . com/CSB5/atminter that utilizes the Entrez search system to query all the possible combinations of taxon-taxon pairs from the data . More specifically , the query “taxon i AND taxon j” for genus ( species ) level are performed on PubMed database in order to obtain the number of papers that corresponds to this query term . Acquisitions of abstract’s content follow a similar way where the query term follows the format “species i AND species j” for each pair of species . Note that , all text-mining procedures are completely automated; that is , users only need to specify the species pairs and the tool will extract the information automatically ( and comprehensively ) from the PubMed database . To select the optimal penalty parameter ( ρ ) , we use the Bayesian information criterion ( BIC ) [23] which is a standard method for model selection . The BIC for Gaussian graphical models takes the form: B I C = - 2 l n ( Ω ) + | E | log ( n ) ( 3 ) where |E| is the number of edges in the association network , n is the sample size , and l n ( Ω ) = n 2 [ log ( det ( Ω ) ) - tr ( Ω C ^ ) ] . Based on ( 3 ) , we choose ρ that minimizes BIC .
To show the effectiveness of our proposed model , we first compare our model against several state-of-the-art models: CCREPE , SparCC , REBACCA , CCLasso , SPIEC ( mb ) and SPIEC ( gl ) . All these codes have been implemented using the R language . We set up p-value at 0 . 05 for CCREPE and the threshold of correlation for SparCC at 0 . 1 ( see S1 File section 1 for precise simulation settings for each algorithm ) . For MPLasso in real datasets , the true underlying network is only partially known and contains spurious information . To assess our algorithm performance with imperfect prior information , we consider prior information with different precision levels , where the precision level is defined as the number of true edges over the total number of edges in the prior information . The total number of edges in the prior network is set to be equal to the number of edges in the true underlying network . Therefore , a precision level of 0 . 1 indicates that 10% of the edges in the prior network are true edges , whereas the other 90% are spurious ones ( see S1 File section 7 for details of introducing priors ) . We report the results we obtained for 0 . 5 precision level in the synthetic experiments while more results for different precision levels can be found in S1 File section 3 and S2 Fig . Emboldened by the success of our proposed algorithm on synthetic data , we have applied MPLasso to infer the associations among microbes for HMP data . Acquisitions and preprocessing for both 16S rRNA and shotgun sequencing data are described in Material and methods section . We report the same three body sites ( i . e . , buccal mucosa , supragingival plague , and tongue dorsum ) of each pipeline and filter out OTUs that appear in less than 10% of total samples—two more body sites ( i . e . , stool and anterior names ) are reported in S1 File section 5 and S8 Fig . The total number of samples and OTUs are summarized in Table 2 and S8 Table . We use the clr transformation in ( 1 ) and add pseudo count 0 . 1 to all the samples , then normalize the counts to get compositional data . However , there is no true correlation network of taxon-taxon associations in real data as opposed to synthetic data . To assess and compare the performance among different methods in real data experiments , we measure the reproducibility of the resulting networks . More specifically , we define the “gold standard” network as the one that uses the full dataset . The reproducibility is defined as the number of edges that had been correctly estimated when using only half of the samples in the full dataset compared to the “gold standard” network . We randomly select half of the samples in the full dataset of each body site and then average over 20 independent simulations . We compare the reproducibility of the MPLasso against SPIEC ( gl ) which has a better performance than other existing algorithms on synthetic datasets as well as CCLasso which has a better performance than other correlation based methods in [11] . The reproducibility results are summarized in Table 2 . MPLasso has a better reproducibility over SPIEC ( gl ) and CCLasso; this implies that MPLasso is not only more robust , but also more accurate at inferring edges . We also consider reproducibility on different percentages of highly connected nodes in S9 Table . Only when we consider as little as only 25% of high degree nodes , CCLasso has a better performance ( but even so for 2% only , on average ) . We also summarize the statistics of the non-associated pairs found by the Fisher’s exact test , potential associated pairs , associated pairs found by MPLasso , and recovered associated pairs in S10 Table ( see also S10 Fig and S1 File section 9 ) . As shown , the known associations obtained from Fisher’s exact test is around 50% over all possible pairs of associations ( i . e . , around 50% prior information ) . The recovery rate of associated pairs of MPLasso is around 80% . For comparison , we also include the recovery rate of associated pairs for CCLasso and SPIEC ( gl ) algorithms in S11 Table . As we can see by comparing the S10 and S11 Tables , CCLasso tends to discover more edges than MPLasso and SPIEC ( gl ) . Although CCLasso can obtain similar results on the recovery rate of associated pairs , it does not perform as well as MPLasso when considering the recovery rate ( i . e . , reproducibility ) of both associated and non-associated pairs ( see Table 2 ) . In other words , CCLasso finds a greater amount of false positive taxa pairs when compared to MPLasso; this is evaluated through AUPR in synthetic experiments shown in Fig 3 . To compare the estimated association networks at each body site for different pipelines ( i . e , HMASM , HMMCP and HMQCP ) , we select the “top players” ( i . e . , high degree nodes ) and arrange them using a counterclockwise layout as shown in Fig 5 . For the genus level data , since we only utilize the Fisher’s exact test ( that only requires the information of the number of abstracts ) , we can use contents of published scientific literature to validate the inferred associations . In contrast , for the species level data , the machine learning-based approach has already used the contents of abstract to obtain the prior information; therefore , it is inappropriate to use any papers that appear in the PubMed search results to validate the inferred associations . To circumvent the potential circular validation , we only use the scientific literature that has not yet been used to create the prior information . For the buccal mucosa ( BucMuc ) , the association pair 〈Streptococcus mitis , Actinomyces naeslundii〉 , which was found in HMASM ( Fig 5 ( a ) ) , has been shown to have associations [26] . Additionally , the associations are also detected at genus level data as shown in Fig 5 ( b ) . Note that the top degree nodes in HMMCP and HMQCP has 70% in common ( i . e . , belongs to same genus ) which implies that the microbial composition of BucMuc is relatively robust . For the supragingival plague ( SupPla ) , the “top players” in species level data ( Fig 5 ( d ) ) mainly come from two genera: Actinomyces and Prevotella which can be widely found in SupPla and also correspond well with the HMMCP dataset ( Fig 5 ( e ) ) . Similarly , the species level associations in tongue dorsum ( TonDor ) is dominated by Actinomyces as shown in Fig 5 ( g ) ; this is because Actinomyces possess 10 different strains out of the total 103 taxa , yet this does not imply that all members of a particular genus group should be associated . Although not seen in Fig 5 ( h ) and 5 ( i ) , genus Actinomyces is also a high degree node in the association network of the genus data . One noticeable observation in the species level dataset ( HMASM ) is that the same genus belongs to the same community which means that edges are mostly found within OTUs of the same taxonomic group . This phenomenon is called assortativity and it has been widely observed in other microbial network studies [17] . However , this does not imply that all members of the same taxon should be ecologically associated . To quantify the similarity of high degree nodes that are found both in HMMCP and HMQCP datasets , we compute the correlation between node degrees at different body sites by utilizing the Spearman correlation method ( see S1 File section 6 ) . We found that TonDor has lower correlations ( ∼0 . 5 ) than other body sites ( ∼0 . 7 ) ; this can be directly observed from Fig 5 ( h ) and 5 ( i ) that have a few high degree genera in common .
The MPLasso R package can be downloaded from here https://github . com/ChiehLo/MPLasso_RPackage | Microbial communities exhibit rich dynamics including the way they adapt , develop , and interact with the human body and the surrounding environment . The associations among microbes can provide a solid foundation to model the interplay between the ( host ) human body and the microbial populations . However , due to the unique properties of compositional and high-dimensional nature of microbial data , standard statistical methods are likely to produce spurious results . Although several existing methods can estimate the associations among microbes under the sparsity assumption , they still have major difficulties to infer the associations among microbes given such high-dimensional data . To enhance the model accuracy on inferring microbial associations , we propose to integrate multiple levels of biological information by mining the co-occurrence patterns and interactions directly from large amount of scientific literature . We first show that our proposed method can outperform existing methods in synthetic experiments . Next , we obtain credible inference results from Human Microbiome Project datasets when compared against laboratory data . By creating a more accurate microbial association network , scientists in this field will be able to better focus their efforts when experimentally verifying microbial associations by eliminating the need to perform exhaustive searches on all possible pairs of associations . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"sequencing",
"techniques",
"microbiome",
"applied",
"mathematics",
"microbiology",
"cloning",
"random",
"variables",
"covariance",
"simulation",
"and",
"modeling",
"algorithms",
"non-coding",
"rna",
"mathematics",
"statistics",
"(mathematics)",
"network",
"analysis",
"shotgun",
"sequencing",
"molecular",
"biology",
"techniques",
"dna",
"cloning",
"cellular",
"structures",
"and",
"organelles",
"microbial",
"genomics",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"medical",
"microbiology",
"molecular",
"biology",
"research",
"assessment",
"probability",
"theory",
"ribosomes",
"biochemistry",
"rna",
"ribosomal",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"reproducibility",
"statistical",
"data"
]
| 2017 | MPLasso: Inferring microbial association networks using prior microbial knowledge |
Dengue is a mosquito-borne disease caused by one of four serotypes of Dengue virus ( DENV-1–4 ) . Severe dengue infection in humans is characterized by thrombocytopenia , increased vascular permeability , hemorrhage and shock . However , there is little information about host response to DENV infection . Here , mechanisms accounting for IFN-γ production and effector function during dengue disease were investigated in a murine model of DENV-2 infection . IFN-γ expression was greatly increased after infection of mice and its production was preceded by increase in IL-12 and IL-18 levels . In IFN-γ−/− mice , DENV-2-associated lethality , viral loads , thrombocytopenia , hemoconcentration , and liver injury were enhanced , when compared with wild type-infected mice . IL-12p40−/− and IL-18−/− infected-mice showed decreased IFN-γ production , which was accompanied by increased disease severity , higher viral loads and enhanced lethality . Blockade of IL-18 in infected IL-12p40−/− mice resulted in complete inhibition of IFN-γ production , greater DENV-2 replication , and enhanced disease manifestation , resembling the response seen in DENV-2-infected IFN-γ−/− mice . Reduced IFN-γ production was associated with diminished Nitric Oxide-synthase 2 ( NOS2 ) expression and NOS2−/− mice had elevated lethality , more severe disease evolution and increased viral load after DENV-2 infection . Therefore , IL-12/IL-18-induced IFN-γ production and consequent NOS2 induction are of major importance to host resistance against DENV infection .
Dengue fever ( DF ) and its severe forms , dengue hemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) , are mosquito-borne diseases caused by one of four serotypes of Dengue virus ( DENV-1–4 ) . Fifty to 100 million cases of DF are estimated annually mostly in tropical and subtropical regions of the world [1]–[3] . According to the World Health Organization ( WHO ) , around 500 , 000 patients develop the severe forms of dengue and 20 , 000 deaths are estimated to occur each year . DHF is defined by the WHO as fever with hemorrhagic manifestations , thrombocytopenia , and hemoconcentration or other signs of plasma leakage [2] . Treatment of DF and of the severe forms of dengue infection is largely supportive . The large number of infected individuals , the lack of clinical or laboratory markers that indicate which patients will develop severe disease and the lack of specific treatment place an enormous burden on health systems of low income countries [2] . The pathogenesis of DENV infection remains poorly understood and involves a complex interplay of viral and host factors . Risk factors for severe disease include age [1] , [4] , viral serotype [1] , [5] and genotype [1] , [6] , [7] , and the genetic background of the host [1] , [8] , among others . Retrospective and prospective human studies have demonstrated that secondary infection by a heterologous serotype is the single greatest risk factor for DHF/DSS [9]–[11] . However , severe disease may also occur after primary infection [5] , [12] , [13] . In both cases , there appears to be a correlation between disease severity and viral load [9]–[13] . In addition , the immunopathogenesis of DENV probably involves the effects of cytokines on both infected and bystander immune cells , hepatocytes , and endothelial cells [2] , [3] , [13] . There are several studies which show enhanced levels of IFN-γ in dengue-infected humans but the precise role of IFN-γ in clinical dengue is somewhat controversial [14]–[16] . There are studies which suggest that levels of this cytokine may correlate positively with disease in humans [16] , but other studies have shown that increased IFN-γ production correlated with higher survival rates in DHF patients [15] . In experimental systems , non-adapted viruses usually are unable to reach high viral loads , except in mice deficient for IFN receptors , suggesting that IFN-γ and its receptors are necessary for host resistance to Dengue infection [17]–[19] . However , the major cell types producing IFN-γ , mediators controlling production of this cytokine and major effector mechanisms triggered by IFN-γ are not known . Optimal IFN-γ production in various infections models in mice is controlled by cytokines , especially IL-12 and IL-18 [20] , [21] . The IFN-γ produced may then upregulate inducible nitric oxide synthase ( NOS2 ) , resulting in high levels of NO production by dendritic cells and macrophages [22] . NO is known to possess potent antiviral activities [22] . Therefore , in order to examine the role played by these molecules during dengue disease we conducted infection experiments in mice infected with an adapted strain of DENV-2 . This unique DENV-2 strain was chosen because it was previously shown to induce in immunocompetent mice a disease that resembles severe dengue cases in humans [23]–[25] , what does not happen with most non-adapted strains usually utilized in experimental settings [2] , [3] . We show that disease is more severe and there are higher viral loads after DENV-2 infection of IFN-γ-deficient mice . Furthermore , we demonstrate that the combined action of IL-12 and IL-18 is necessary for optimal IFN-γ production and control of DENV-2 infection . Finally , we show that IFN-γ controls expression of NOS2 and NO production after DENV-2 infection and that NO production is crucial for resistance of the murine host to infection with DENV .
This study was carried out in strict accordance with the Brazilian Government's ethical and animal experiments regulations . The experimental protocol was approved by the Committee on the Ethics of Animal Experiments of the Universidade Federal de Minas Gerais ( CETEA/UFMG , Permit Protocol Number 113/09 ) . All surgery was performed under ketamine/xylazine anesthesia , and all efforts were made to minimize suffering . The guidelines followed by this Committee are based on the guidelines of Animal Welfare Act ( AWA ) and associated Animal Welfare Regulations ( AWRs ) and Public Health Service ( PHS ) Policy . Mice deficient for IFN-γ and NOS-2 were obtained from The Jackson Laboratory and were bred and maintained at the Gnotobiology and Immunology Laboratory of Instituto de Ciências Biológicas . Mice deficient for IL-12p40 were kindly provided by Dr . J . Magran through Dr . L . V . Rizzo ( Instituto de Ciências Biomédicas ( ICB ) , University of São Paulo , São Paulo , Brazil ) and were bred and maintained at the Gnotobiology and Immunology Laboratory of Instituto de Ciências Biológicas . Mice deficient for IL-18 [26] were kindly provided by Dr . F . Q . Cunha and were bred and maintained at the Gnotobiology and Immunology Laboratory of Instituto de Ciências Biológicas . Mice deficient for IL-23p19 [27] were bred and maintained at the animal facility of the Transgenose Institute ( CNRS , Orleans ) . All mice were on C57BL/6J genetic background ( back-crossed at least 10 times ) and wild-type control C57BL/6J ( WT ) mice were used , except for IL-18-deficient mice , that were on the BALB/c background and were compared to their proper WT littermates . For experiments , 7–10 weeks old mice were kept under specific pathogen–free conditions , in filtered-cages with autoclaved food and water available ad libitum . An adapted Dengue virus 2 ( DENV-2 ) strain was obtained from the State Collection of Viruses , Moscow , Russia [23] . Briefly , the virus had undergone two passages in the brain of BALB/c suckling mice . Five days after infection , brains of moribund mice were harvested for preparing 10% ( w/v ) brain suspension in modified Eagle's medium ( MEM ) . After that , eight sequential passages through BALB/c mice of different ages ( 1–4 weeks old ) by intraperitoneal ( i . p . ) injection were performed . Two sequential passages were carried out for each age group of . After each passage , the brains of the moribund mice were harvested for preparing 10% brain suspension and then used for the next passage . The last passage of DENV-2 strain P23085 was performed in neonatal mice to produce stocks which were stored as 10% brain suspension at −70°C . Sequences of portions of E and NS1 genes of the adapted virus were deposited previously at GenBank under the accession number AY927231 [22] . Virus adaptation was performed in a biosafety level-3 ( BSL-3 ) facility of the SRC VB «Vector» , Russia , Koltsovo . After adaptation , monolayers of Aedes albopictus C6/36 cell line were infected with DENV-2 strain P23085 at a multiplicity of infection ( MOI ) of 0 . 05 PFU/cell and incubated at 28°C for 5–7 days . The cultured medium was harvested after cytopathic effect was noticed and cell debris removed by centrifugation . The virus supernatant was collected and stored at −70°C until use . The cultured medium of mock-infected monolayers of Aedes albopictus C6/36 cell line was used as control of the infection . To calculate virus titer , expressed as LD50 , in the harvested cultured medium , groups of ten mice were inoculated i . p . with serial dilutions of the virus and lethality recorded . The titer of our DENV-2 stock was 105 LD50/ml of suspension , as calculated in 8–10-week-old BALB/c mice . 1LD50 corresponded to 20 PFU of the adapted DENV-2 strain . For infection experiments , the virus-containing cell-supernatant was diluted in endotoxin-free PBS and injected i . p . into mice . For the evaluation of lethality , mice were inoculated i . p . with DENV-2 virus and lethality rates evaluated every 12 h . The various other parameters were evaluated at 3 , 5 or 7 days after i . p . inoculation of the virus . In all experiments using genetically deficient mice , experiments with the relevant WT controls were performed in parallel . Non-infected ( NI ) animals were inoculated with suspension from non-infected cell supernatant diluted in a similar manner . In the experiments involving genetically deficient mice , the NI group represents the pooled results obtained from the analysis of deficient mice and WT non-infected mice . Results were pooled for ease of presentation . In some experiments IL-18 was neutralized by daily i . p . injection of 250 µg of recombinant human IL-18BP per animal ( hIL-18 bp ) , starting 1 hour after DENV-2 inoculation until day 4 after virus inoculation . The dose was chosen based in a previous study [28] . Control animals received PBS . The hIL-18 bp isoform was a kind gift of Dr . Amanda Proudfoot from Merck-Serono Pharmaceuticals ( Geneve , Switzerland ) . Murine bone marrow cells were isolated from femurs and were differentiated into myeloid DCs after culturing at 2×106 cells/ml for 10 days in RPMI supplemented with 10% FBS and 4% J558L cell-conditioned medium as a source of GM-CSF as described [29] . DCs were plated in 96-well microculture plates ( at 2×105 cells/well in DMEM supplemented with 2 mM l-glutamine and 2×10−5 M 2-ME ) and for infection , cells were incubated with 50 µL of the cell supernatant suspension containing DENV-2 at 0 , 01 MOI in the presence or not of recombinant murine IFN-γ ( 100 U/ml ) . Negative controls were stimulated with sterile cell supernatant obtained from mock infected cells . Mice were assayed for viral titers in spleen . For virus recovery in spleen , the organ was collected aseptically and stored at −70°C until assayed for DENV-2 virus . Tissue samples were weighed , grounded by using a pestle and mortar and prepared as 10% ( w/v ) homogenates in minimal essential medium ( MEM ) without fetal bovine serum ( FBS ) . Viral load in the supernatants of tissue homogenates assessed by direct plaque assays using LLC-MK2 cells cultured in agarose overlay . Briefly , organ homogenates were diluted serially and a 0 . 4 ml volume placed in duplicate into each of 6-wells of LLC-MK2 cell monolayers and incubated for 1 h . An overlay solution containing 2× MEM and 1% agarose in equal volumes was added to each well and the cultures incubated for 7 days . Cultures were stained with crystal violet for enumeration of viral plaques . Cell monolayers incubated with tissue homogenates of not infected mice were used as control for the assay . The results were measured as plaque forming units ( PFU ) per gram of tissue weight . The limit of detection of the assay was 100 PFU/g of tissue . The concentration of cytokines ( TNF-α , IFN-γ , IL-6 , IL-12p40 , IL-12p70 and IL-18 ) in serum or tissue samples was measured using commercially available antibodies and according to the procedures supplied by the manufacturer ( R&D Systems , Minneapolis , except for IL-18 , manufactured by BD Pharmingen ) . Serum was obtained from coagulated blood ( 15 min at 37° , then 30 min a 4°C ) and stored at −20°C until further analysis . One hundred milligrams of tissues ( liver and spleen ) was homogenized in 1 ml of PBS containing anti-proteases ( 0 . 1 mM phenylmethilsulfonyl fluoride , 0 . 1 mM benzethonium chloride , 10 mM EDTA and 20 KI aprotinin A ) and 0 . 05% Tween 20 . The samples were then centrifuged for 10 min at 3000 g and the supernatant immediately used for ELISA assays . The detection limit of the ELISA assays was in the range of 4–8 pg/ml . Cell-free culture medium was obtained by centrifugation and assayed for nitrite content , determined by the Griess method [30] . For this assay , 0 . 1 ml of culture medium was mixed with 0 . 1 ml of Griess reagent in a multiwell plate , and the absorbance at 550 nm read 10 min later . The NO2− concentration ( µM ) was determined by reference to a NaNO2 standard curve . Blood was obtained from the brachial plexus in heparin-containing syringes at the indicated times . The final concentration of heparin was 50 u/ml . Platelets were counted in a Coulter Counter ( S-Plus Jr ) . Results are presented as number of platelets per µl of blood . For the determination of the hematocrit , a sample of blood was collected into heparinized capillary tubes and centrifuged for 10 min in a Hematocrit centrifuge ( HT , São Paulo , Brazil ) . Aspartate transaminase activity was measured in individual serum samples , using a commercially available kit ( Bioclin , Belo Horizonte , Brazil ) . Results are expressed as the U/dL of serum . Total RNA was isolated from Spleen of mice for evaluation of NOS2 mRNA expression . RNA isolation was performed using Illustra RNAspin Mini RNA Isolation Kit ( GE Healthcare ) . The RNA obtained was resuspended in diethyl pyrocarbonate treated water and stocked at −70°C until use . Real-time RT-PCR was performed on an ABI PRISM 7900 sequence-detection system ( Applied Biosystems ) by using SYBR Green PCR Master Mix ( Applied Biosystems ) after a reverse transcription reaction of 2 µg of total RNA by using M-MLV reverse transcriptase ( Promega ) . The relative level of gene expression was determined by the comparative threshold cycle method as described by the manufacturer , whereby data for each sample were normalized to hypoxanthine phosphoribosyltransferase and expressed as a fold change compared with non-infected controls . The following primer pairs were used: hypoxanthine phosphoribosyltransferase , 5′-GTTGGTTACAGGCCAGACTTTGTTG-3′ ( forward ) and 5′-GAGGGTAGGCTGGCCTATAGGCT-3′ ( reverse ) ; and nos2 , 5′- CCAAGCCCTCACCTACTTCC -3′ ( forward ) and 5′- CTCTGAGGGCTGACACAAGG -3′ ( reverse ) . Spleen cells were evaluated ex vivo for extracellular molecular expression patterns and for intracellular cytokine expression patterns . Briefly , spleens were removed from infected mice at the indicated timepoints . Then cells were isolated , and immediately stained for surface markers , fixed with 2% formaldehyde and then permeabilized with a solution of saponin and stained for 30 min at room temperature , using conjugated anti-IFN-γ monoclonal antibodies . Preparations were then analyzed using a FACScan ( Becton Dickinson ) , and 50 000 gated events on total lymphocyte/monocyte population were acquired for later analysis . Figure S1A shows the gating strategy utilized for IFN-γ+ population analysis in CD4+ cells . Briefly , lymphocyte/monocyte population was isolated in gate R1 . At this region , the cell population positive for the surface marker of interest was isolated ( R2 ) and among cells in this region , IFN-γ+ cells were obtained ( R3 ) . Analogous strategies were utilized for the other several populations studied . The antibodies used for the staining were rat immunoglobulin controls , anti-CD4-PE , anti-CD8-PE , anti-NK1 . 1-PE , anti-CD3- PE-Cy5 and anti-IFN-γ-FITC ( all from Biolegend Inc ) . Analysis was conducted using the software Flow Jo 7 . 2 ( Tree Star Inc ) . A portion of liver was obtained from killed mice at the indicated time points , immediately fixed in 10% buffered formalin for 24 hours and tissues fragments were embedded in paraffin . Tissue sections ( 4 µm thick ) were stained with hematoxylin and eosin ( H&E ) and examined under light microscopy or collected in serial sections on glass slides coated with 2% 3-aminopropyltriethylsilane ( Sigma Aldrich , St . Louis , MO ) . The latter sections were deparaffinized by immersion in xylene , and this was followed by immersion in alcohol and then incubation with 3% hydrogen peroxide diluted in Tris-buffered saline ( TBS ) ( pH 7 . 4 ) for 30 minutes . The sections were then immersed in citrate buffer ( pH 6 . 0 ) for 20 minutes at 95°C for antigen retrieval . The slides were then incubated with the rabbit polyclonal anti-NOS2 ( N-20 , sc-651 , Santa Cruz Biotechnology , Santa Cruz , CA ) diluted 1∶100; at 4°C overnight in a humidified chamber . After washing in TBS , the sections were treated with a labeled streptavidin-biotin kit ( LSAB , K0492 , Dako , Carpinteria , CA ) . The sections were then incubated in 3 , 3′-Diaminobenzidine ( K3468 , Dako ) for 2 to 5 minutes , stained with Mayer's hematoxylin and covered . Negative controls were obtained by the omission of primary antibodies , which were substituted by 1% PBS-BSA . Results are shown as means ± SEM . Differences were compared by using analysis of variance ( ANOVA ) followed by Student-Newman-Keuls post-hoc analysis . Differences between lethality curves were calculated using Log rank test ( Graph Prism Software 4 . 0 ) . Results with a P<0 . 05 were considered significant .
An initial set of experiments were carried out to assess the kinetics of IFN-γ production and major IFN-γ producing cell types after DENV-2 infection . As shown in Figure 1 , there was an increase in serum and splenic levels of IFN-γ from the 5th day of infection ( Figure 1A ) . Levels of IFN-γ enhanced further at day 7 in both serum and spleen ( Figure 1A ) . In spleen , IFN-γ staining was detected in about 10% of total cells in the 5th day after inoculation and reached about 15% at the 7th day post infection ( Figure 1B and Figure S1B ) . CD3−NK1 . 1+ NK cells and CD3+NK1 . 1+ NKT populations presented increased proportions of IFN-γ staining at the 5th day post infection ( Figure 1B and Figure S1E and S1F ) . In addition , there was increase in expression of IFN-γ on all cell populations analyzed at day 7 after infection ( Figure 1B ) . Significantly , over 30% of CD4+ T cells , 25% of CD8+ T cells , 40% of CD3−NK1 . 1+ NK cells and CD3+NK1 . 1+ NKT cells were IFN-γ+ at day 7 after infection ( Figure 1B and Figures S1C–F ) . When the gate was set at IFN-γ+ cells , the majority of IFN-γ+ cells were CD8+ T cells ( 30±3% ) and CD4+ T cells ( 25±1% ) . To investigate the role played by IFN-γ during DENV infection , WT and IFN-γ-deficient ( IFN-γ−/− ) mice were inoculated DENV-2 and lethality rates and disease course evaluated . As seen in Figure 1C , 100% of IFN-γ−/− mice were dead before the seventh day of infection , and only 15% of WT mice had succumbed to infection . This early lethality of IFN-γ−/− mice was characterized by more severe manifestation of disease after DENV infection . Three days after infection , IFN-γ−/− mice already presented reduced platelets counts ( Figure 1D ) , and at the 5th day of infection , there was marked thrombocytopenia ( Figure 1D ) and significant increase in hematocrit values ( Figure 1E ) in IFN-γ−/− mice when compared to WT mice . In addition to the alterations seen in hematological parameters , there was enhanced production of pro-inflammatory cytokines after infection . As shown in Figures 1F and 1G , there were no detectable levels of TNF-α and IL-6 in serum of WT mice at day 5 after DENV-2 infection . However , both cytokines were significantly elevated in serum of infected IFN-γ−/− mice ( Figures 1F and 1G ) . Infected-IFN-γ−/− mice showed hepatic injury , as assessed by increased AST activity in plasma of IFN-γ−/− mice in the 5th day of infection ( Figure 1H ) . There was also marked changes in liver architecture . WT mice inoculated with DENV-2 had little changes in liver , as assessed by histology . In contrast , there were signs of congestion and hepatocyte degeneration and necrosis in infected IFN-γ−/− mice ( Figure 1I ) . In addition to the greater disease severity observed , IFN-γ−/− mice presented greater viral replication after infection than in WT mice . At the 3rd day of infection , IFN-γ−/− mice presented a 10 fold increase in DENV-2 viral loads in spleen and DENV-2 titers in spleen of infected-IFN-γ−/− mice were above 1 . 5 log greater than in infected-WT mice in the 5th day of infection ( Figure 1J ) . Therefore , the data depicted here show IFN-γ is expressed and plays an important role in host defense against DENV infection . Our next objective was to evaluate the roles of IL-12 and IL-18 in controlling IFN-γ production by the murine host during DENV infection . After DENV-2 infection , there were detectable levels of both IL-12p70 and IL-12p40 in the spleen of WT mice already in the 3rd day of infection ( Figure 2A ) . The concentration of both cytokines was increased in the 5th and remained above background levels at the 7th day of infection ( Figure 2A ) . This early production is consistent with a putative role of IL-12 in inducing IFN-γ production . Consistently with the latter possibility , there was a drastic reduction in IFN-γ production after DENV-2 infection of IL-12p40−/− mice , which are deficient for both IL-12 and IL-23 production ( Figures 2B and 2C ) . In keeping with the relevance of IFN-γ during dengue infection and reduced IFN-γ production , there was enhanced lethality rates ( Figure 2D ) , increased thrombocytopenia ( Figure 2E ) and enhanced hemoconcentration ( Figure 2F ) after DENV-2 infection of IL-12p40−/− mice . There were higher concentrations of TNF-α ( Figure 2G ) and IL-6 ( Figure 2H ) in spleen and more severe hepatic injury in IL-12p40−/− than WT mice after infection ( Figure 2I and 2J ) . Finally , IL-12p40 deficiency resulted in greater loads of DENV-2 in spleen at the 7th day after infection , when compared with WT-infected mice ( Figure 2K ) . The reduction of IFN-γ production and the more severe disease seen in IL-12p40−/− mice seem to be specifically due to IL-12 deficiency as IL-23p19−/−-deficient mice produced similar amounts of IFN-γ after DENV-2 infection ( Supplementary Figure S2A ) and presented a disease of similar intensity ( Figure S2B and S2C ) and unaltered viral loads ( Figure S2 D ) when compared to infected-WT mice . Another cytokine shown to induce IFN-γ production during infections is IL-18 [21] . In the present study , IL-18 concentrations rose rapidly in liver at the 3rd day of DENV-2 infection , but returned to basal levels in the subsequent timepoints evaluated ( Figure 3A ) . There was marked reduction of IFN-γ production in spleen and serum of DENV-2-infected IL-18−/− mice when compared with WT infected mice ( Figure 3B and 3C , respectively ) . Available IL-18−/− mice were in the BALB/c background which we have previously shown to be more susceptible to DENV2-induced disease and lethality [24] . Indeed , all WT mice in the BALB/c background were dead by day 10 of DENV-2 infection using an inoculum that caused little lethality in C57Bl/6 mice ( compare Figures 3D and 1C ) . All IL-18−/− mice also succumbed to infection but mice died earlier than WT controls after DENV-2 infection ( p = 0 . 0237 ) ( Figure 3D ) . Although the degree of thrombocytopenia was similar in both strains of mice ( Figure 3E ) , hemoconcentration was greater in IL-18−/− than WT infected mice ( Figure 3F ) . Levels of TNF-α ( Figure 3G ) and IL-6 ( Figure 3H ) and severity of liver injury ( Figure 3I and 3J ) occurred to a greater extent in spleens of IL-18−/− than WT infected mice ( Figure 3G and 3H ) . Significantly , enhanced clinical disease and earlier deaths were accompanied by elevation in viral loads in spleen of IL-18−/− mice ( Figures 3K ) . The phenotype of either IL-12−/− or IL-18−/− mice were not as severe as the phenotype of IFN-γ−/− mice . For example , whereas viral loads were already approximately 2 log greater at day 5 in IFN-γ−/− mice , this was not the case in IL-12−/− or IL-18−/− mice ( Figures 2J and 3J ) . Indeed , IFN-γ production was not abolished in IL-12−/− or IL-18−/− mice and viral loads were only significantly different from WT at day 7 after infection ( see Figures 2J and 3J ) . In order to block simultaneously the action of both IL-12 and IL-18 , IL-12p40−/− mice were treated with IL-18 bp at doses shown to block IL-18 action [28] . Treatment of IL-12p40−/− mice with IL-18 bp also resulted in total abrogation of IFN-γ levels in serum ( Figure 4A ) or spleen ( Figure 4B ) of infected mice . Treatment of IL-12p40−/− with IL-18 bp also resulted in marked enhancement of viremia already at day 5 after infection ( Figure 4C ) , results which are similar to those obtained in IFN-γ−/− mice ( Figure 1I ) and substantially different from results observed at day 5 in IL-12p40−/− mice or mice treated with IL-18 bp alone ( Figure 4C ) . Moreover , treatment of IL-12p40−/− with IL-18 bp resulted in thrombocytopenia , which was similar to that observed in IL-12p40−/− or IL-18 bp-treated mice ( Figure 4D ) , and hemoconcentration , which was greater than in the other groups ( Figure 4E ) . Levels of IL-6 in plasma were also further enhanced by the treatment of IL-12p40−/− mice with IL-18 bp than in either condition alone ( Figure 4F ) . The enhanced viral load and greater disease severity already at day 5 resulted in greater lethality rates in IL-12p40−/− mice treated with IL-18 bp than in either condition alone or WT mice ( Lethality rate at day 7: WT mice , 0%; IL-18bp-treated mice , 0%; IL-12p40−/− mice , 33%; IL-12p40−/− mice+IL-18 bp , 83% , n = 6 ) . In concert , the data presented above suggest that IL-12 and IL-18 act together to induce optimal IFN-γ production during dengue infection in mice . Nitric Oxide production by phagocytes is a well known effector mechanism induced by IFN-γ during host response to infections [22] . To assess whether this pathway is relevant in host response to DENV infection , we evaluated NOS2 expression after DENV-2 infection . As shown in Figure 5A , there was increase in NOS2 mRNA expression in spleen already at day 5 day but expression rose rapidly at day 7 after DENV2 infection of WT mice ( Figure 5A ) . Evaluation of NOS2 staining in the liver by immunohistochemistry showed significant NOS2 expression , virtually only in infiltrating leukocytes , at day 7 after infection ( Figure 5B , C ) . Consistently with the ability of IFN-γ to induce NOS2 , there was no production of NO by dendritic cells infected with DENV-2 , in vitro ( Figures 5D ) . However , treatment of dendritic cells with IFN-γ prior to infection resulted in production of significant amounts of NO ( Figure 5D ) . In addition , expression of NOS2 was greatly decreased in spleen of IFN-γ−/− mice after DENV-2 infection ( Figure 5E ) . As IL-12 and IL-18 cooperate for optimal induction of IFN-γ ( results above ) , we evaluated whether treatment of IL-12p40−/− mice with IL-18 bp would also results in reduced NOS2 expression in spleen . As seen in Figure 5E , concomitant absence of both IL-12 and IL-18 led to impaired NOS2 expression in spleen that was quantitatively similar to results obtained in IFN-γ−/− mice ( Figure 5E ) . To assess the role played by NOS2-induced NO during DENV infection , NOS2−/− mice were inoculated with DENV-2 and lethality rates and hematological alterations monitored . As shown in Figure 6A , NOS2−/− mice were markedly susceptible to DENV infection , as all knockout animals but none of WT mice were dead by the 10th day of infection . Thrombocytopenia ( Figures 6B ) was more intense earlier but hemoconcentration was similar in both groups ( Figure 6C ) . There was enhanced splenic production of TNF-α ( Figure 6D ) and IL-6 ( Figure 6E ) and greater hepatic injury ( Figure 6F and 6G ) after DENV-2 infection of NOS2−/− than WT mice . Importantly , viral loads in spleen after DENV-2 infection were significantly greater in NOS2−/− than WT mice ( Figures 6H ) . Of note , all alterations seen in NOS2−/−-infected mice were not due to reduction in IFN-γ production after infection . Indeed , IFN-γ levels in spleen and serum were similar in WT and NOS2−/− infected mice ( Figures 6I and 6J ) . Therefore , NOS2-derived NO production is driven by IFN-γ and is essential for host protection during DENV primary infection .
The major findings of the present study can be summarized as follows: 1 ) IFN-γ production is essential for host resistance to DENV infection . NK and NKT cells are the sources of IFN-γ during the early periods of infection and are followed by CD4+ and CD8+ T cells , which are the main producers at the peak of host response to infection; 2 ) production of IL-12 and IL-18 precedes IFN-γ and optimal IFN-γ production relies on the combined action of IL-12 and IL-18; and 3 ) IFN-γ is essential for NOS2 induction and NOS2 plays an important role in controlling virus replication . These studies , therefore , indicate that IL-12/IL-18-induced IFN-γ production and consequent induction of NOS2 are essential for murine host response to DENV infection . Previous studies support a protective role played by IFN-γ during host response to DENV infection . For example , Shresta and coworkers have shown that IFN-γ receptor-deficient mice were more susceptible to DENV-induced lethality than WT-infected mice , despite no differences in viral loads in several target organs between both groups [17] . The increased susceptibility was enhanced further when type I IFN receptor was also absent , and deficiency in both cytokine receptors resulted in disseminated viral replication [17] . In this respect , IFN receptors-deficient mice ( AG129 strain ) are known to be permissive for replication of DENV clinical isolates in peripheral tissues and CNS , and represent a well established experimental model of DENV infection [17]–[19] . In the present work , we have demonstrated that IFN-γ is produced as early as the fifth day of infection in WT mice and lack of IFN-γ action culminated in early lethality to a sublethal inoculum . These data establish IFN-γ as essential for host control of DENV replication and resistance to infection . The correlation between increased IFN-γ production and higher survival rates in DHF patients [15] also supports this idea . Of note , enhanced viral replication in IFN-γ-deficient mice was associated with more severe disease manifestation , as showed by enhanced hematological alterations and hepatic injury . More severe disease was also noticed in DENV-infected AG129 mice , characterized by paralysis and elevated hematocrit [17] . Importantly , Gunther and colleagues have demonstrated in a human challenge model of DENV infection that only sustained IFN-γ production was associated with protection against fever and viremia during the acute phase of illness [31] . These data suggest that IFN-γ is important to prevent worsening of disease . In humans , epidemiological studies have shown that a substantial number of patients with severe disease have evidence of a previous infection with a distinct serotype [1]–[3] , [9]–[11] , [32] . Several hypotheses have been raised to explain this immune-mediated enhancement of disease severity . For example , it has been hypothesized that subneutralizing levels of antibodies facilitate the entry of viral particles in permissive cells ( a phenomenon termed antibody-dependent enhancement - ADE ) , enhancing viral load , and exacerbating disease manifestation [33] . Experimental DENV models support this hypothesis and suggest that disease severity is directly associated with enhanced viral replication during infection [34] , [35] . Of note , infected IFN-γ-deficient mice , as well as IL-12p40−/− and IL-18−/− infected mice , presented elevated viral loads , in parallel with elevated hematocrits , thrombocytopenia , and liver injury . Therefore , we may suggest that the worse outcome seen in mice with reduced IFN-γ production after infection is due to inability in control of DENV replication , leading to viral burden and enhancement of disease . Mice in which IFN-γ production was decreased or deficient had a significant increase in levels of pro-inflammatory mediators after DENV infection . Indeed , both TNF-α and IL-6 production were enhanced in DENV-2 infected IFN-γ−/− , IL-12p40−/− , and IL-18−/− mice , when compared with WT controls . Increased levels of these cytokines have been associated with severity of dengue manifestation in humans [36]–[38] . Hence , enhanced TNF-α release by T cells during secondary stimulation with DENV antigens was found in hospitalized patients with more severe disease evolution [39] . In addition , the ratio of TNF-α-producing to IFN-γ-producing T cells among peripheral blood mononuclear cells from dengue-vaccine recipients was shown to be greater after in vitro stimulation with antigen from heterologous dengue serotypes [39] , suggesting that increased amounts of TNF-α alters response to infection and may result in more-severe disease manifestation . Findings in murine experimental models support this idea [40] . Altogether , these findings in humans suggest that IFN-γ production is associated with protective responses to DENV infection and that severe disease may occur due to absence of proper IFN-γ release and to enhanced TNF-α production during response , although it remains to be shown if enhanced TNF-α production seen in DENV infected IFN-γ−/− mice was due to T cells or to any other cellular population . Interestingly , enhanced viral load have also been associated with increased pro-inflammatory response during mouse experimental infection by West Nile virus [41] , another important flavivirus that is pathogenic to humans . The latter findings support the hypothesis that increased virus replication in the absence of IFN-γ production leads to increased pro-inflammatory mediators response . TNF-α blockade in experimental models of DENV infection resulted in prevention of disease [19] , [23] and TNF-α action has been implicated in increased vascular permeability after infection in experimental settings [13] . Of note , inhibition of other pro-inflammatory mediators produced in the evaluated experimental model of DENV infection , including PAF and MIF , is associated with reduced production of TNF-α and IL-6 and milder disease manifestation , reduced hypotension and vascular permeability after DENV infection [13] , [24] , [25] . Hepatic injury was also enhanced in IFN-γ−/− mice infected with DENV . Data from our laboratory suggest that enhanced liver injury during experimental DENV infection involves both productive viral infection of hepatocytes and immunopathological mechanisms , such as enhanced leukocyte arrest and activation in hepatic tissue ( our unpublished data , manuscript in preparation ) . Therefore , the elevation of pro-inflammatory cytokine production and consequent liver injury seen in the absence of IFN-γ appears to account for the worse outcome after DENV infection in mice . Several studies have demonstrated the IFN-γ-inductive role played by IL-12 and IL-18 during experimental models of viral infections [20] , [21] , [42] . Here , we have shown that IL-12 and IL-18 were produced early after DENV infection . The kinetics of production of these cytokines was compatible with their inductive role of IFN-γ production . In support of the latter possibility , IL-12p40−/− and IL-18−/− mice presented marked reduction in IFN-γ production after DENV infection . In addition , absence of one of these cytokines led to worsening of dengue disease , despite a small delay in peak of DENV-induced alterations . Of note , only during simultaneous blockade of both IL-12 and IL-18 , there was complete abrogation of IFN-γ production . Interestingly , IL-12−/− mice treated with IL-18 bp presented marked enhancement of splenic viral loads already at the 5th day post DENV-2 infection and disease seen in these mice was very similar to that found in infected IFN-γ−/− mice . Thus , IL-12 and IL-18 act synergistically to induce IFN-γ production during DENV infection . Of note , IL-18 production has been shown to be dependent on inflammasome complex activation [43] , suggesting that this molecular scaffold may play a role in the control of IFN-γ production and in host resistance to DENV infection . IL-18 is known to augment IL-12-induced IFN-γ production by T and NK cells [20] , [21] , [42] , [44] , and absence of IFN-γ in infected mice is known to abolish both NK cell and CTL responses during viral infections [42] , [44] . Our data suggest that , upon infection , NK and NKT cells are the cell populations involved in early IFN-γ production and that CD8+ and CD4+ T cells are the main IFN-γ producers at later moments of response to infection ( 7th day ) . IFN-γ production by CD4+ T cells during experimental DENV infection has been previously demonstrated [45] . In addition , CD8 T cell activation has been associated to protection to DENV primary infection in mice [46] , [47] . Our data showing a significant increase in IFN-γ+ NK and NKT cells and the finding that IFN-γ−/− mice succumb very early to infection suggest a important role for these cell populations in mediating resistance to DENV infection during its initial phases . Of note , NK cell activation early after experimental DENV infection has been previously demonstrated [44] . Interestingly , increased percentages of NK cells and of activated NK cells were also associated with milder DF , whereas reduced cell counts , low percentages and lack of activation markers ( comparable to healthy controls ) were associated with evolution to DHF in patients [48] , [49] . Altogether , these observations suggest that sequential and coordinated IFN-γ production by these lymphocytes populations during DENV infection is an event of extreme importance for host resistance to disease . However , it remains to be shown the antigenic specificity of these IFN-γ-producing lymphocytes in the studied experimental settings . In addition , whether these cells are poly-functional and secrete other cytokines or present other effector functions remain to be studied . In this regard , it has been demonstrated that development of subclinical secondary infection in school children is associated with increased proportions of DENV-specific TNF-α , IFN-γ and IL-2-producing CD4+ and CD8+ T cells [50] , suggesting that poly-functional responses correlate with protection to severe disease manifestation . On the contrary , cytokine-producing T cells ( especially TNF-α and/or IFN-γ ) were associated with DHF development in patients and these DHF associated , cytokine-producing T cells were shown to be negative for CD107a staining , suggesting that these lymphocyte populations represent mono-functional or oligo-functional T cells [51] . Therefore , assessment of the pattern of T cell cytokine production and of the mechanisms controlling such polyfunctionality ( whether IL-12 and or IL-18 are involved in such control ) may provide important information regarding protective versus pathogenic responses to DENV infection and may bear relevance during development of vaccinal strategies . At the moment , these subjects have been matter of ongoing analysis in our experimental infection model . Apart from promotion of NK and CTL responses , IFN-γ seems to be important for viral clearance by induction of NO production . It has been shown that NOS2 expression is increased upon DENV infection in humans and that this expression in peripheral blood monocytes of DF patients was found to correlate with the late acute phase of disease and preceded the clearance of DENV from monocytes [52] . Hence , NO production was associated with less severe form of dengue disease in humans [53] . Here , we demonstrate that NOS2 expression is increased during DENV infection and that this expression is controlled by IFN-γ production , once IFN-γ−/− and IL-12p40−/− mice treated with IL-18 bp presented reduced NOS2 expression . In addition , IFN-γ stimulation was necessary for NO production by DENV-infected DCs , in vitro . Importantly , blockade of NOS2 action was associated with enhanced viral loads after infection , and more severe disease manifestation , even in the presence of high levels of IFN-γ . Of note , NO is able to inhibit DENV replication in human cells in vitro [54] , [55] , an effect associated with inhibition of DENV associated polymerase activity [54]–[56] . Thus , NOS2-mediated NO production is pivotal for resistance to DENV infection and this seems to be a major pathway involved in IFN-γ-mediated resistance to disease . However , in the absence of NOS2 , animals die with a slower kinetics than IFN-γ−/− mice , suggesting that mechanisms in addition to NOS2-mediated NO production may be relevant for IFN-γ-mediated host protection to infection . This could involve the presence of CTL responses and NK cells , but not NKT cells , which seem to play detrimental role in experimental DENV infection [57] . These IFN-γ-dependent and NOS2-independent mechanisms are currently being investigated in our laboratory . However , other studies have demonstrated a pathogenic role for NO during DENV infection . Utilizing human cell lines and experimental mouse infection , it has been shown that overproduction of NO could lead to endothelial cell damage , and cross-reactive antibodies against endothelial cells , present during DENV infection , were found to induce cell damage in an NO-dependent manner [58] . For example , Yen and coworkers have found that tissue hemorrhage after experimental DENV infection was dependent upon reactive nitrogen species production by endothelial cells . This event was associated with increased endothelial cell apoptosis during infection [59] . Although NOS2 inhibition resulted in reduced hemorrhage , viral replication was not evaluated . In addition , the increased hemorrhage displayed after NO production seemed to be an endothelial cell-associated phenomenon and was potentiated by TNF-α and reactive oxygen species ( ROS ) . On the contrary , IFN-γ-mediated NO inhibition of viral replication was demonstrated especially in leukocytes population both in human and mouse settings [52]–[56] . Our results showed that NOS2 staining during DENV-2 infection in the present model was mainly associated to leukocytes . These findings suggest that NO may have a dual role during DENV infection and that this is associated with the cell populations involved in NO production and on the presence of additional inflammatory mediators . NO production by infected leukocytes may be associated to control of viral replication and prevention of disease evolution , while NO production by endothelial cells , especially in the presence of TNF-α and ROS , would favor cell death and more severe disease manifestation . Additional experiments evaluating cell-specific NOS2-deficient mice will help in answering the latter hypothesis and aid in defining other roles of NO in the context of experimental dengue . In conclusion , we have demonstrated that IFN-γ production is essential for host resistance to DENV infection . IFN-γ production upon infection is controlled by concomitant production of IL-12 and IL-18 and the IFN-γ-dependent mechanisms associated to resistance to dengue disease involve NOS2 up-regulation and consequent NO production . In the absence of these molecules , there is enhancement of viral burden and more severe manifestation of dengue disease . Thus , IFN-γ induction helps to orchestrate immune response maturation , control of viral replication and regulation of inflammatory response during host response to DENV infection , defining the outcome of dengue disease . Despite extrapolation of this experimental scenario to human infection requires further investigation , we may suggest that strategies that improve the production of IFN-γ-mediated immunity by the host could be useful during the control of primary infection by Dengue virus . | Dengue fever and its severe forms , dengue hemorrhagic fever and dengue shock syndrome , are the most prevalent mosquito-borne diseases on Earth . It is caused by one of four serotypes of Dengue virus ( DENV-1–4 ) . At present , there are no vaccines or specific therapies for dengue and treatment is supportive . Host response to infection is also poorly understood . Here , using a DENV-2 strain that causes a disease that resembles the severe manifestations of Dengue in humans , we demonstrate that IFN-γ production is essential for the host to deal with infection . We have also shown that IFN-γ production during DENV infection is controlled by the cytokines IL-12 and IL-18 . Finally , we show that one of the mechanisms triggered by IFN-γ during host response to DENV infection is the production of Nitric Oxide , an important virustatic metabolite . Mice deficient for each of these molecules present marked increase in DENV replication after infection and more severe disease . Altogether , this study demonstrates that the IL-12/IL-18-IFN-γ-NO axis plays a major role in host ability to deal with primary DENV infection . These data bear relevance to the understanding of antiviral immune responses during Dengue disease and may aid in the rational design of vaccines against DENV infection . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"animal",
"models",
"model",
"organisms",
"emerging",
"infectious",
"diseases",
"immunity",
"virology",
"microbial",
"pathogens",
"immunology",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"immune",
"response"
]
| 2011 | IFN-γ Production Depends on IL-12 and IL-18 Combined Action and Mediates Host Resistance to Dengue Virus Infection in a Nitric Oxide-Dependent Manner |
Most common methods for inferring transposable element ( TE ) evolutionary relationships are based on dividing TEs into subfamilies using shared diagnostic nucleotides . Although originally justified based on the “master gene” model of TE evolution , computational and experimental work indicates that many of the subfamilies generated by these methods contain multiple source elements . This implies that subfamily-based methods give an incomplete picture of TE relationships . Studies on selection , functional exaptation , and predictions of horizontal transfer may all be affected . Here , we develop a Bayesian method for inferring TE ancestry that gives the probability that each sequence was replicative , its frequency of replication , and the probability that each extant TE sequence came from each possible ancestral sequence . Applying our method to 986 members of the newly-discovered LAVA family of TEs , we show that there were far more source elements in the history of LAVA expansion than subfamilies identified using the CoSeg subfamily-classification program . We also identify multiple replicative elements in the AluSc subfamily in humans . Our results strongly indicate that a reassessment of subfamily structures is necessary to obtain accurate estimates of mutation processes , phylogenetic relationships and historical times of activity .
Repetitive elements may comprise two-thirds or more of most vertebrate genomes [1] , and most repeat sequence is derived from transposable elements ( TEs ) . To obtain an accurate picture of the structure and evolutionary history of vertebrate genomes , it is therefore necessary to have a good understanding of the origins and expansion histories of TEs . Early studies attempted to reconstruct the relationships among TEs by dividing extant TE sequences into subfamilies on the basis of shared high-frequency diagnostic nucleotide variants [2]–[7] . Many of these early studies , particularly in primates , were interpreted as supporting a “master gene model” , in which one or a few source elements produce large numbers of inert copy elements that are incapable of replication [8] , [9] . According to this model , each subfamily represents the descendants of a particular master gene , the sequence of which is assumed to be the subfamily consensus . Later studies found evidence for multiple source elements within subfamilies [10]–[12] , however , and recent empirical studies have shown that up to hundreds of elements are capable of replication when placed in a laboratory system [13] . This research suggests that subfamily classification based on diagnostic nucleotides provides only a coarse picture of what may be an intricate web of familial relationships among the TEs in the genome . However , no previously established method can accurately reconstruct relationships among thousands of TE sequences . Our group is particularly interested in utilizing TEs to understand the genomic mutation process . In theory , TEs are extremely useful for this purpose , as mutations that accumulate after a duplication occurs should typically be almost entirely neutral , and therefore serve as an accurate reflection of the mutation process unfiltered by selection [14] . However , in the course of using TEs to investigate evolutionary processes , we discovered inconsistencies that suggested that subfamily consensus sequences produced by CoSeg , a popular program for TE subfamily classification , are not reliable for use as ancestral sequences . The main problem is that at many positions in TE alignments , far more sequences than expected differ from the subfamily consensus sequence . This leads to high apparent mutation rates at these positions if the subfamily consensus is assumed to be the ancestor of all elements in the subfamily . Instead , we inferred that many of the elements in the subfamily were produced by source elements that already differed from the subfamily consensus at one or more sites but were not identified by CoSeg . An additional limitation of CoSeg and all other current subfamily-classification methods is that they assign elements to subfamilies deterministically , without accounting for inference uncertainty . This is especially problematic for TE evolutionary studies because similarities between ancestral TEs may make it impossible to precisely determine the ancestry of any given element . These problems limit the utility of TEs for investigating evolutionary processes , and thus strongly motivate the development of a new approach . Here , we propose a novel Bayesian Markov chain Monte Carlo ( MCMC ) method that predicts which sequences replicated during a TE family's evolutionary history , and reconstructs the ancestral relationships among replicating and non-replicative sequences . The method returns the posterior probability that each TE sequence was replicated from each of a set of plausible ancestral sequences , as well as the probability that each candidate ancestral sequence replicated at all . To our knowledge , the only other method specifically designed to reconstruct ancestral TE relationships that is not based on heuristic subfamily classification is that of Cordaux and colleagues [10] . These authors build a median joining network [15] of the extant elements , a maximum-parsimony based method . Although this method was an important contribution , it is deterministic , only applicable to a small number of sequences , and shares the general problems [16] of maximum-parsimony phylogenetic methods . Some authors apply phylogenetic techniques designed for inferring species relationships , such as neighbor joining methods , to reconstruct TE relationships [17] , [18] . These methods implausibly assume bifurcating trees , though a single source TE may replicate itself many times . We applied our approach to two TE families: the gibbon-specific LAVA TEs [19] and the Sc subfamily of Alu . The gibbon LAVAs are a novel class of element found exclusively in gibbon ( Hylobatidae ) species , and are composed of portions of other TEs usually found in primate genomes: L1ME5 , AluSz6 , and SVA_A [19] . The LAVA elements are an attractive system for understanding the evolution of TEs because their recent origin ( sometime after the Gibbon divergence from other hominids 15–18 million years ago ) and limited diversification [19] make the analysis of their relationships more tractable . In contrast , the AluSc family is an older inactive Alu subfamily ( estimated to be at least 35 million years old [20] ) . Using our new method , we evaluated whether the likely number of replicating ancestral sequences in each TE family or subfamily differed from the number of subfamilies returned by CoSeg , whether the subfamilies previously identified are compatible with predicted ancestral relationships , and whether our method solved the problem of unrealistically high implied mutation rates at some sites . Finally , we suggest new subfamily designations in the gibbon LAVA TE family based on their probable relationships .
Most methods to characterize TE relationships first divide a TE family into subfamilies . Subfamily-classification methods group sequences on the basis of their nucleotide identity at “diagnostic” sites [21] , [22] , for example by recursively splitting subfamilies that fail a test of homogeneity [22] . By far the most popular automated subfamily classification method is CoSeg [23] , a wrapper for the AluCode program [24] that is integrated with the widely-used RepeatMasker TE identification program [25] . The CoSeg/AluCode method tends to identify more subfamily structure than previous approaches , so we decided to compare results from our new program exclusively to CoSeg results . The AluCode algorithm used by CoSeg iteratively identifies sequences in a family or proposed subfamily that contains pairs of sites with nucleotide variants that co-occur more frequently than would be expected by random mutation from the subfamily consensus sequence . This pair of sites is then used to divide sequences into two new subfamilies , which may be further split by the same criteria , and so on to completion . The observation of overrepresented nucleotides at a pair of sites suggests that some sequences currently assigned to a subfamily were produced by a progenitor sequence that diverged at these sites prior to replicating . This justifies introducing a new subfamily to contain the descendants of that progenitor . After generating subfamilies , CoSeg links them using a minimum spanning tree of the subfamily consensus sequences , which is intended to represent the subfamily phylogeny . The CoSeg algorithm was applied to 986 aligned LAVA elements ( 401 bp ) to obtain 14 subfamilies . We noticed that some sites showed higher levels of divergence from the CoSeg-defined subfamily consensus sequences than might be expected due to mutation alone . In earlier work on human Alu and opossum SINE1 TEs , we had observed similarly aberrant sites [26] . These sites suggest the existence of undiscovered replicative sequences that carry the divergent variant , so we hypothesized that CoSeg subfamily classification might be too conservative about adding new subfamilies to give a realistic picture of ancestral replicative sequence structure in LAVA . CoSeg implements a number of conservative measures that guide the splitting . For example , it only allows each site to be used once to split subfamilies . Additionally , split decisions are only made on the basis of a strict significance test , which means that subfamilies with high support for existing may still be rejected . To determine the plausibility that the CoSeg subfamily consensus sequences represent all of the ancestral sequences of the TEs in the data , we developed a re-sampling test . Null expectations were obtained by resampling substitutions from the consensus sequence of each subfamily , accounting for variation in mutation rates among sites and mutation types . The substitution resampling process was replicated 1000 times to get a predicted distribution of each nucleotide at each site for each subfamily under the assumption that all differences between ancestors and descendants are due to mutation . The expected sums of deviations from these expectations were compared to the observed deviations from expectation among the real by-site nucleotide distributions in each CoSeg-inferred subfamily . Applying this test to the LAVA CoSeg subfamilies , we found that , in 12 of the 14 CoSeg subfamilies , deviation from expectations exceeded the deviation among any of the 1000 resampling replicates ( Figure 1 ) . Thus , we can reject the hypothesis that the sequence data can be explained solely by substitutions from the subfamily consensuses , and infer that there are likely to be many more ancestrally-replicative sequences than identified by CoSeg . To better understand the evolution of TEs , what is needed is a method that directly addresses which sequences were historically replicative and which sequences descended from each replicative element . To achieve this , we developed a novel Bayesian Markov chain Monte Carlo ( MCMC ) approach that jointly considers the evidence for replication of all plausible ancestral TE sequences in a family . We will refer to this method as AnTE . The first step in this method is to identify plausible discriminatory sites that separate ancestral replicative elements . We call them “discriminatory” sites to distinguish them from “diagnostic” sites that are used to deterministically classify sequences in subfamily-based methods . Discriminatory sites will tend to vary more than other sites , because replicative sequences that differ from the consensus at such sites will increase the frequency of the variant as they proliferate . Initially , the plausible discriminatory sites were identified as those sites with variant frequencies more than three standard deviations greater than the mean frequency of that variant among all sites with the same consensus base ( see Methods for a full description of discriminatory site identification ) . The next step of the AnTE algorithm is to construct a pool of candidate replicating ancestors; the probability that each candidate is a true ancestor can then be evaluated using the MCMC . By definition , ancestors differ from the consensus only at discriminatory sites , so only the discriminatory site sequence needs to be considered . Initially , the set of candidate ancestor discriminatory site sequences was constructed to be the set of all discriminatory site sequences observed in the sequence data . During the burn-in period of the chain , discriminatory site combinations outside the initial set of candidate ancestors were added if their inclusion improved the likelihood of the model . The MCMC estimates posterior distributions for three sets of parameters: the relative rates of replication for each candidate ( a rate of 0 indicates that the candidate is not ancestral ) , the times at which each candidate with non-zero replicative rate was actively replicating , and rate parameters for a nucleotide substitution rate matrix that determines the probability of transitioning between any pair of nucleotides over a time period . For any step in the MCMC procedure , the likelihood of generating the sequence data was calculated based on the inferred ancestors ( i . e . , sequences with non-zero replicative frequency ) , their replicative frequencies and times of activity at that step , and the substitution rate matrix . A prior was set on the total number of replicative sequences by giving a likelihood penalty for each candidate with non-zero replicative rate . The likelihood of each sequence observed in the data or inferred by the model was calculated based on summing the probability ( see Methods , Equation 2 ) that it was produced by mutation from each inferred ancestral sequence , weighted by the replicative frequency of that candidate ancestor . The posterior probability distributions of the replicative frequency for each candidate sequence , whether it replicated at all , and which ancestors it was derived from , were then calculated by averaging these probabilities over all steps in the post convergence portion of the MCMC . Separate chains were run on the LAVA and AluSc datasets for five different prior distributions of the total number of replicative sequences , set by applying a penalty on each additional ancestor inferred by the model . These penalties consisted of 0 , 2 , 4 , 6 , or 8 log points per ancestor . In LAVA , 38–43 ( 99% credible region ) replicative sequences were inferred even under the harsh 8 log penalty , many more than the 14 subfamilies identified by the CoSeg program ( Table 1 and Figure 2a ) . More replicative sequences were also identified for AluSc than the three subfamilies given by CoSeg , though the total number was much less than for LAVA , with 6–11 replicative sequences inferred among all priors considered ( Figure 2b ) . The same substitution resampling method applied to the CoSeg subfamilies earlier was applied to the results from each AnTE run , testing whether mutation alone can explain the differences between inferred ancestral sequences and their descendants ( Table 1 ) . Based on this analysis , we reject the mutation-only hypothesis for the LAVA runs with 8 ( p<0 . 001 ) , 6 ( p = 0 . 004 ) , or 4 ( p<0 . 001 ) log penalty , inferring that these runs fail to identify some true ancestral sequences . We fail to reject the mutation-only hypothesis for the 2 log penalty run ( p = . 064 ) and the 0 log penalty run ( p = . 090 ) . Thus , we select the results from the 2 log penalty chain as a conservative estimate of the number of replicative sequences in the history of LAVA , and use it in all further analyses of LAVA . Results for this chain are given in Supplementary Tables 1 and 2 . The 99% credible region for the number of replicative elements in the 2 log penalty run is 50–60 , suggesting 50 as a reasonable lower bound for the total number of replicative sequences . For AluSc , mutation appears to be a sufficient explanation for the differences from the inferred ancestors for all priors considered ( Table 1 ) . Since the number of sequences identified in AluSc was relatively insensitive to the prior , we present results for the same 2 log penalty as used for LAVA to facilitate comparison ( Supplementary Table 3 ) . Network representations of the relationships among the elements of the LAVA and AluSc families are shown in Figures 3–5 . These networks show the predicted ancestral relationships among all sequences with more than 50% probability of being replicative ( shown most clearly in Figure 3a and 5a ) . The arrows on the graph indicate the predicted original source of each replicative sequence , with cycles representing uncertainty about the direction of original descendancy . Note that later copies of that sequence may have arisen from other ancestors , including possible back mutation from one of its descendants . Each node in the graph represents a particular sequence , with the diameter of the node proportional to its estimated frequency of replication . There are four sequences inferred to have at least a 5% probability of being the LAVA root according to the AnTE algorithm . We compared these sequences to the segment of the human genome homologous to the 3′ end of LAVA [19] . One of these four plausible root sequences ( Figure 3 and 4 , marked with an arrow ) has only 2 differences from the human sequence among 73 discriminatory sites; among all other candidates with >50% probability of being replicative , there are 4–28 differences ( mean 12 . 1 ) . Thus , the marked sequence is the probable ancestral LAVA , and the inferred root from AnTE is consistent with the homology data . The assignment of CoSeg subfamilies to nodes in the ancestry networks of LAVA ( Figure 3 ) and AluSc ( Figure 5 ) indicates that most CoSeg subfamilies are represented by multiple ancestral replicative sequences . Although CoSeg subfamilies tend to cluster together in the network , replicative sequences from three LAVA subfamilies ( colored in , purple , magenta and light blue in the graph ) are disjointed , with intervening replicative sequences from other subfamilies ( or that are not assigned to a subfamily at all ) . Additional discrepancies can be found when considering the CoSeg subfamily assignments of all sequences , not just replicative sequences ( Figure 3b ) . Among descendants of all ancestors with CoSeg subfamily assignment , 57 LAVA sequences ( 6 . 5% ) and 19 Alu sequences ( 2% ) are assigned to different subfamilies than their most probable ancestor . Based on this result , and considering the ancestral relationships inferred by the AnTE MCMC , we propose a subfamily organization for LAVA with 9 new subfamilies ( Figure 4; see Figure S1 for legend ) . This subfamily scheme was designed based on the desiderata of a ) relatively few subfamilies; 2 ) matching the CoSeg subfamilies where possible , to facilitate comparison; and 3 ) minimizing the number of sequences with uncertain subfamily assignment . The low mixing of colors in Figure 4b indicates that we have largely achieved our goal , although there is unavoidable uncertainty at most boundaries between subfamily groups . We want to emphasize here that the utility of the subfamilies is entirely organizational and aesthetic . We recommend that any analytical inference be carried out on the full ancestral probability network , and that it should sum over all ancestral uncertainty rather than arbitrarily assigning uncertain sequences to one ancestor or another and subsequently treating the assignment as though it were data . We estimated the number and rate of substitutions between ancestral and descendent sequences at each site . This analysis indicates that , contrary to the assumption of the CoSeg algorithm , substitutions at individual sites repeatedly discriminate among replicative sequences . In LAVA , there are multiple substitutions among replicative sequences at between 38%–45% ( 95% credible region ) of the discriminatory sites . The CoSeg algorithm does not allow sites to discriminate between subfamilies more than once; this is intended to prevent the creation of new subfamilies from elements formed by recombination between sequences from separate subfamilies . However , it is reasonable to expect that substitutions that create new replicative sequences may occur multiple times . From a mechanistic perspective , discriminatory sites may be less likely to affect replicative function , whereas non-discriminatory sites may be more likely to affect replicative function . If there are only a limited number of sites that don't affect function , all of the mutations among replicative sequences will be focused on those sites . To test whether all sites are equally likely to be discriminatory , we considered a null model in which the probability of substitution between ancestral replicative sequences is proportional to the probability of substitution to extant sequences at that site . We randomly re-sampled all substitutions on the tree of replicative LAVA sequences to obtain a null distribution for the number of substitutions per site . Although 33–42 sites ( MCMC 95% credible region ) had exactly one substitution among the actual replicative sequences , 51–93 sites had a single substitution in 1000 draws of the null model . Thus , there is an excess of sites with multiple substitutions among ancestors in the observed data compared to the null hypothesis of no constraint . We conclude that , as expected , some variants are not neutral with regard to replication . To further explore this question , we created a simple model of constraint on replicative elements that allows for two types of sites: constrained sites , which eliminate replicative capacity entirely if mutated , and unconstrained sites , which have no effect on replicative capacity . We tested this model for different m , the number of constrained sites among the 330 sites analyzed ( microsatellites , CpG sites , and large insertions were removed prior to MCMC analysis and therefore were not considered ) . As before , substitutions were drawn to match the number among replicative sequences , but no substitution was allowed at m random sites separately selected for each draw . Taking the upper bound inference of 42 sites with single substitutions , the lowest m for which at least 5% of 1000 draws had 42 or fewer sites with one substitutions was 163 , leaving 167 sites unconstrained . This analysis suggests that only around half the tested sites are effectively neutral to replicative function . The LAVA sequence is divided by a VNTR ( variable number of tandem repeats ) region of up to 2000 bp . Our main analysis focused on the region 3′ from the VNTR , as many LAVA loci lack all or part of the VNTR and 5′ region . The full-length 5′ region is around 350 bp , and we found 337 loci with intact 5′ regions . Analysis of these sequences revealed three separate clusters defined by presence or absence of two large interior segments of around 100 bp each . We used AnTE to reconstruct the ancestral relationships separately within each of these three clusters . These ancestral networks largely agree with the analysis of the 3′ region: the first cluster consists mostly of sequences from the adjacent green , purple , and brown subfamilies from Figure 4 ( Figure S2A ) ; the second cluster consists mostly of green and grey subfamilies ( Figure S2B ) , and the third cluster is composed mostly of the older red , yellow , pink , and blue subfamilies ( Figure S2C ) . However , 26 sequences ( 7 . 7% ) are assigned ancestors on the 5′ network that are distantly related to ancestors in the 3′ network . A probable explanation for this discrepancy in placement between the 3′ and 5′ ancestral networks is recombination across the VNTR . Aside from these putative recombinants , the network structure within the three 5′ clusters is largely in agreement with the structure of the 3′ network ( compare Figure 4 and Figure S2 ) . Ancestral inference of TEs that inserted prior to a speciation event can be validated by comparing homologous elements between two species . To see this , consider that if the ancestor is correct , then the number of shared differences from the ancestor at each site will be approximately proportional to the time between insertion and speciation ( ) . The number of unique differences in each branch will be approximately proportional to the time between speciation and the present ( ) . Sequences that differ from the predicted ancestor upon insertion will have an inflated number of shared differences from the predicted ancestor . This will lead to a higher estimate of than at non-discriminatory sites . Taking the AluSc consensus sequence as the presumed ancestor , we found that five of the six discriminatory sites inferred by our method exceeded the mean ratio by 3-fold or greater ( Figure S3 ) , whereas all of the non-discriminatory sites have lower ratios . To validate each branch on the tree in Figure 5b , we separately considered the descendants of each predicted ancestral sequence ( the “test” ancestor ) along with all of the descendants of its ancestor ( the “parent” ) . Considering the ratios assuming the parent sequence was the true ancestral sequence , a positive validation result would consist of a high ratio ( exceeding the 3x threshold ) for the site that discriminated the test ancestor from the parent . All predicted ancestors were validated by this test . No non-discriminatory sites exceeded the 3-fold threshold except a single CpG site ( position 1 ) , which is possibly a true discriminatory site that was undetected because we eliminated CpG sites in the AnTE analysis . It is also notable that in this branch-validation analysis , the discriminatory site with the lowest ratio in the overall consensus analysis ( Figure S3 ) was validated , but the two non-discriminatory sites that had higher ratios were not .
We have confirmed here that the CoSeg subfamily classification method fails to identify many highly-probable ancestral sequences in both LAVA and AluSc , and therefore that CoSeg subfamily consensus sequences are problematic for use as presumed ancestors in divergence and substitution analysis . In contrast , the AnTE method we developed and describe here provides a detailed picture of TE evolutionary history , providing ancestral sequences , the times of replicative activity of these sequences , and their replication frequency . The AnTE method is fast and enables the probabilistic evaluation of relationships between thousands of elements within subfamilies and between subfamilies . The AnTE program , relevant datasets , and user instructions are available at www . EvolutionaryGenomics . com . Though the AnTE method identifies more sequences than previous approaches in both subfamilies studied , many more ancestrally-replicative sequences were identified for LAVA ( 50–60 ) than for AluSc ( 6–7 ) from similar-sized sequence datasets . Our analysis suggests that most AluSc sequences derive from a single ancestor , while the most successful LAVA source sequence is responsible for only 13% of extant LAVA elements . The two datasets are not directly comparable , as most of the LAVA sequences identified in the Gibbon genome were used for our analysis of LAVA , while only a small subset of AluSc was used , and AluSc itself is a subfamily of the much larger Alu TE family . Nevertheless , this large difference between families suggests differing evolutionary dynamics . The method presented in this paper has some limitations that should be addressed in future work . Firstly , it assumes that all differences between sequences and their ancestors are the result of mutation , rather than recombination or gene conversion . We found strong evidence of recombination across the large VNTR region in LAVA in 7 . 7% of full sequences , but no obvious evidence of recombination between distant ancestral sequences within the regions either 5′ or 3′ from the VNTR . However , we cannot rule out the possibility that some sequences are a result of recombination events between closely-related subfamilies . Second , our method , like most phylogenetic methods , assumes site-independence . We excluded CpG sites from our analysis because their elevated mutation rate violates site independence . CpG sites are common in both LAVA and Alu , and it is possible that some are discriminatory sites that can help distinguish true ancestral sequences . Methods that allow the relaxation of site-independence assumptions would also allow large deletions and microsatellites to be used as TE subfamily markers . Here , we had to analyze the clusters separated by large deletions in independent analyses . Third , our method accounts for the activity periods of transposable elements in a simplistic way , assuming a single time point of activity rather than representing a distribution of replication rates across time . One obvious but non-trivial improvement that could be made would be to better estimate the distribution of replication times for each ancestral subfamily , such as has been done for Alu subfamilies [26] . Despite the assumptions made in creating subfamilies using previous approaches , they have often been used in studies of TE evolution . For example , most methods for estimating the age of subfamilies are based on some measure of divergence between subfamily consensus sequences and the members of the subfamily [27]–[30] . Our findings suggest that this prior widespread use of subfamily consensus sequences as the single ancestral subfamily source sequence to analyze TE mutation patterns [14] has led to over-estimation of substitution rates and TE divergence times , and to incorrect inference of substitution patterns . AnTE can be used to improve such analyses , and may be useful to revise existing subfamily nomenclature based on more realistic estimates of ancestral replication patterns , as we have done with the gibbon LAVA elements . Overall , we expect that such approaches will be central for evaluating genome structural evolution and using TEs to understand genome-wide mutation processes .
The human genome was downloaded from the RepeatMasker [25] website . The 2006 build of the human genome was masked based on Repbase [31] version 20090604 using version RepeatMaskerOpen-3 . 2 . 8 of RepeatMasker . The annotated Alu sequences were extracted from the genome and sorted by subfamily classification . A total of 34 , 515 AluSc sequences were identified in this way . Of these , 1200 were selected at random for ancestry determination and manually aligned . For all human Alu elements , the corresponding Alu elements from rhesus macaque were obtained using Galaxy [32] . The “extract pair wise MAF blocks” tool from Galaxy was used to get the sequence matches of each of macaque to human Alu elements . The “Stitch MAF blocks” tool was used to obtain the correspondence between matches among genomes to the human Alu coordinates . To ensure accurate alignment , macaque AluSc sequences with less than 80% identity to their human homologue were removed from analysis . We identified LAVA sequences in the Gibbon genome using the probability-based oligonucleotide clustering method P-clouds [33] . The published LAVA consensus sequence , which contains only the region 3′ of the VNTR [19] , was segmented into regions which were used to form clouds . We then searched the genome for locations that matched the cloud data . Identified locations were merged if the distance between them was less than the length of the region in the consensus sequence . This resulted in 1136 sequences will full 3′ regions . Sequence for the region 5′ of the VNTR was obtained by building clouds from the region upstream of the VNTR in these sequences . Locations matching these clouds were then merged to the 5′ sequences to obtain full length sequences . This process identified 338 sequences with complete 5′ regions . Alignments for both the 3′ and 5′ regions were constructed manually . An assumption of our model is that the substitution process at each site is independent of all other sites . This assumption is clearly violated by large insertions/deletions , CpG sites , and microsatellites . Therefore , CpG and gap sites within the consensus , as well as microsatellite regions , were excluded from all analyses . All sequences with gaps larger than four nucleotides in their alignment to the consensus were also excluded from analysis . This left 986 LAVA and 972 AluSc sequences for the main analysis . Alignments before and after processing are provided in Supplementary Data Files S1–S10 . We define “discriminatory sites” as those sites which differ among historically replicative sequences . Since only discriminatory sites are informative in ancestry determination , our first goal is to predict these sites . Two features distinguish discriminatory from non-discriminatory sites . First , discriminatory sites will tend to have a higher frequency of a particular variant than expected by mutation alone . At non-discriminatory sites , all variation is due to substitution; at discriminatory sites , replication of a sequence that already differs at that position will also increase the frequency of the variant . Second , discriminatory sites will show association with each other , because discriminatory variants arise in particular backgrounds of variation at other discriminatory sites . We predicted which sites are discriminatory as follows: First , a nucleotide substitution probability matrix was derived by counting the number of differences from the consensus sequence to each of the elements in the sequence database . Each nucleotide difference count between each pair of bases or gaps , was used to obtain relative substitution probabilities from to , ( 1 ) Sites with mutations exceeding the mean rate of any type of mutation by more than three standard deviations were then identified as an initial set of predicted discriminatory sites . For each predicted discriminatory site , we then tested for association with all other sites using a Monte Carlo chi-square test . All sites with p-values < . 01 for association with any of the initially-predicted site were added to the pool of discriminatory sites . Note that , as described below , each candidate ancestral sequence is evaluated by MCMC for the probability it is a true ancestor . Therefore , we are not concerned with including some false discriminatory sites , as the strength of evidence for each site will be reflected in the final results . A set of candidate ancestral sequences was constructed based on the predicted discriminatory sites . By definition , ancestral sequences do not differ at non-discriminatory sites , so all ancestors were assumed to agree with the consensus except at discriminatory sites . For AluSc , the small number of discriminatory sites allowed inclusion of all possible discriminatory site combinations as ancestral sequences . For LAVA , all discriminatory site combinations observed in the data were included as an initial set of candidate ancestral sequences . Since some ancestors may have had combinations of discriminatory site which no longer exist , we added new plausible candidates during the burn-in phase of the MCMC , as described below . The TE ancestry model consists of three sets of parameters: , the replicative frequency of each candidate ancestor; , the estimated time of replicative activity of each ancestor , and rate parameters for a nucleotide substitution rate matrix . The ancestral frequencies were modeled as discrete variables with constant sum equal to the total number of sequences in the data . The parameters approximate the time of replicative activity for each candidate as a single time point , in which that candidate produced all descendants . For computational efficiency , these time parameters were restricted to 1001 equally-spaced points between 0 and 1 , with 0 defined as the present and 1 as the time of activity of the root sequence . Flat priors were assumed for all parameters except , for which a penalty is applied for each candidate with nonzero replicative frequency . The size of this penalty was varied across runs to reflect different beliefs about the prior probability any given sequence is replicative . The likelihood of generating any TE sequence in the dataset , given all parameters , is defined as: ( 2 ) where is the number of ancestral candidates , is the jth candidate ancestral sequence , is the replicative frequency of candidate j , and is the probability of transitioning from sequence to sequence in time period . This sequence transition probability is the product of the transition probabilities at each site between the base in and the base in at that site . The transition probabilities between each pair of nucleotides over time are obtained from the matrix exponential . The overall likelihood of the data , , is the product of the likelihood of all sequences which exist according to these parameters , both current and ancestral . Note that for any i such that , there is no implication that candidate ever existed , so we need only consider the likelihood all candidate ancestors i for which . For any such sequence , other than the root of the family: ( 3 ) For , the sequence transition probability is the probability of transitioning from sequence to sequence over time period , calculated , as described above , by taking the product of nucleotide transition probabilities over all sites . For , this probability is zero , since ancestral sequences cannot produce descendants which were active earlier than they were . The root sequence is defined to have likelihood 1 . The substitution rate matrix is defined by 10 rate parameters according to a general strand-symmetric model , giving the substitution rates between all pairs of nucleotides and single-nucleotide insertion/deletions . The Markov chain was run using the Metropolis-Hastings method [34] to sample all parameters . The chain was initialized by randomly selecting half of the candidate sequences as replicative , and their initial frequencies were assigned according to a multinomial distribution with equal prior probabilities for each selected candidate . Two types of proposals were used to efficiently sample , the replicative frequencies of the candidate ancestors . In the first proposal type , two candidate ancestors are selected at random; the frequency of the first is increased by one and the frequency of the second is reduced by one . Proposals are always rejected if acceptance would lead to negative values . In the second proposal type , the frequency of two randomly-chosen candidate ancestors is swapped . The parameters were also sampled by two proposal types . In the first , a candidate ancestor j was selected at random . A random integer n was drawn from 0 to 1000 , and is set to n/1000 . In the second , candidate ancestors and were selected at random , and their associated parameters and were swapped . The substitution rate parameters were sampled by a single proposal , in which the current rate was added to a draw from a normal distribution with mean 0 and standard deviation . 01 . As all proposals are symmetric , the chain satisfies detailed balance if the acceptance probability for the moves from to follows the Metropolis-Hastings [34] acceptance proposal , where is the likelihood of the set of all parameters . ( 4 ) The first 10 million generations of the Markov chain were considered a burn-in stage , used to obtain an equilibrium sample of parameters prior to sampling the posterior . For the LAVA sequences , this stage was also used to add plausible candidates to the pool of candidate ancestral sequences for inclusion in the model . The initial set of candidate ancestral sequences was the set of discriminatory site sequences observed in the data . However , it was necessary to account for the possibility that some ancestral sequences were not represented; for every candidate sequence in the pool at any sampling point , every possible nucleotide change in the sequence was evaluated for whether the overall likelihood of the data would increase if that change were made , keeping everything else constant . If the likelihood increased for a given nucleotide change , a new candidate sequence , differing only by that nucleotide change , was added to the ancestral pool . New candidates were tested for addition every 100 , 000 steps from step 2 . 5 million to step 7 . 5 million in the burn-in . After burn-in , the Markov chains were run for 10 million generations and sampled every 10 , 000 generations . Good mixing was verified by running three replicates with each replicate starting from a random parameter set , and confirming that the within-replicate variance was at least 99% of the overall variance . Given a proposed ancestral reconstruction for a set of TE sequences , we developed a test of the hypothesis that mutation alone can explain the variation between descendants and their proposed ancestors . If the mutation hypothesis is true , we expect the substitution process at a given site to be independent of the ancestral sequence once the ancestral nucleotide and the site position are accounted for . Therefore , we can reject the mutation-only hypothesis if the descendants of a proposed ancestor have a higher frequency of a variant than can be explained by mutation alone . Such a result suggests the existence of one or more intermediate sequences between the ancestor and some of its proposed descendants that vary from the proposed ancestor at the high-frequency variant sites . The basis of the test is to “redraw” the substitutions of each sequence in the data . First , the number of substitutions of each type at each site between all proposed ancestors and descendants were counted . For each sequence in the data , a new sequence was constructed from its proposed ancestral sequence by adding a number of substitutions equal to the number of differences between and . These substitutions were drawn randomly according to the following process . First , a site is selected for substitution . The probability of selecting any site which was in nucleotide state in is weighted by the fraction of sequences which has a substitution at site from an ancestor in state out of all sequences whose ancestor was in state at site . Then , the particular substitution is selected , with the probability of each substitution type weighted by the frequency of that substitution from the ancestral nucleotide at that site according to the proposed ancestral reconstruction . This process is repeated until has a number of differences from equal to the number of differences between and . Note that this redraw process accounts for differences in substitution probability at a site based on ancestral nucleotide at that site and position . The redraw process is conducted 1000 times . For each redraw , a 3-dimensional matrix is constructed giving the number of each variant at each site among descendants of each ancestor . The entries in these matrices are averaged among redraws to give a matrix of expected values . For each redraw , the sum of absolute differences between observed and expected values is computed over the entire matrix . Finally , the sum is computed by the same calculation based on the actual substitutions according to the proposed ancestral reconstruction . If the mutation hypothesis is true , should fall within the distribution of the values . This redraw test was run on both the CoSeg-inferred ancestors and the ancestors inferred from the AnTE algorithm . To draw a deterministic ancestral reconstruction from the probabilistic output of AnTE , a step of the MCMC after convergence was selected at random , and all sequences were assigned ancestors based on their probability of descendance according to the parameters at that step . A relevant question in understanding TE ancestry is whether only a limited number of sequences can be successful in the replication process . If so , it is expected that mutations at constrained sites will lead to inactive copies that will not replicate further . Such sites will be non-discriminatory , while sites that do allow substitutions among ancestral replicators may become discriminatory sites . To assess whether there was support for constraint at some sites , we tested whether the substitution patterns matched either of two models of sequence constraint in replicative TEs . In the null model , no constraint was assumed , so the expected relative frequency of substitutions at a site among replicative elements equaled the relative frequency overall . In the test model , it was assumed that m sites were completely constrained , so that any differences from consensus at that site prevented replication . A random tree of ancestral relationships was drawn from the MCMC data by selecting a step of the MCMC at random , and assigning ancestors to all data sequences and inferred ancestral sequences randomly , with the probability of assignment to each ancestor weighted by the probability of descendence from that ancestor according to the parameters at that step . As this tree gives the ancestral sequence for all sequences in the data , we can use it to derive the substitutions between ancestors and descendants , distinguishing between substitutions to replicative and non-replicative sequences . The test statistic was the number of sites with no substitutions among replicative sequences; i . e . , the number of discriminatory sites . We generated distributions of according to the assumptions of each model , and then compared these to the posterior distribution of implied by the MCMC results . First , 1000 trees of TE relationships were drawn randomly . For each tree , the number of substitutions at each site was calculated , both for all elements and restricting to replicative elements . Additionally , the number of sites with no substitutions was calculated to get the distribution of according to the MCMC results . Then , to generate a distribution of according to each model , for each tree we drew from a multinomial distribution with number of trials equal to the total number of substitutions among replicative elements according to that tree . For the first model , the vector of probabilities in the multinomial distribution is the relative frequency of substitution at each site . For the second model , sites were selected from the sites for which no substitutions occurred among replicative elements according to the tree . These sites were assigned a substitution probability of zero , and the other probabilities were normalized to sum to 1 before drawing from the multinomial distribution . Thus , from 1000 draws of a tree , we obtain distributions of according to the MCMC results , the no-constraint model , and models for each possible value of , from 1 to the total number of sites . We reject a model if fewer than 5% of values fell within the 95% confidence region for from the MCMC . The best fit for the second model was defined as the that minimized the absolute difference of the ordered values from the MCMC and the model . The AluSc subfamily predates the split between human and rhesus macaque . We used the homologous AluSc sequence to validate the ancestors inferred by AnTE . We define as the time between insertion of an AluSc sequence and the split between macaque and human , and as the time between the split and the present . Given that the ancestral nucleotide at a position is , we can estimate the probability that neither , one , or both of the macaque and human sequence have substituted away from . Assuming low rates of substitution , and no back-mutation , the probability of substitution is approximately proportional to time . The probability that both descendant sequences are still is then: ( 5 ) where and are the present-day bases in human and macaque , respectively , is the base the TE has upon insertion , and is the mutation rate . Similarly , the probability that one of the two descendant sequences has substituted away is: ( 6 ) By inserting the proportion of sequences with 0 or 1 substitutions into the above equations and solving for and , we can obtain an estimate for and at every site , under the hypothesis that all sequences are descended from the consensus . Though we expect to differ between sites , estimates of the ratio should be similar if the hypothesis holds . If the hypothesis is false , then at sites where some of the sequences already differed from the consensus when they were inserted , we expect estimates of this ratio to be higher than at other positions , to account for the greater number of sequences for which macaque and human share a difference from the consensus . Thus , a relatively high estimate indicates a discriminatory site . Given a tree of relationships among AluSc sequences , we estimate for every position among all descendants ( immediate or distant ) of each ancestor . We consider a branch in the tree validated if the sites which distinguish the descendant node from the ancestral node all have ratios at least 3-fold greater than the mean ratio . | The most common entities in vertebrate genomes are transposable elements ( TEs ) , DNA sequences that have been repeatedly copied and inserted into new locations throughout the genome . Some TEs have been replicated hundreds of thousands of times , and their ecology and evolutionary history within a genome is thus critical to understanding how genome structure evolves . It was once thought that only a few “master gene” copies could replicate , while the rest were inactive ( dead on arrival ) , but recent computational and laboratory studies have indicated that this is not the case . However , previous methods for reconstructing TE evolutionary history were not designed to solve the problem of determining the ancestral source sequence for large numbers of elements . Here , we present a new method that is . Our method surveys all likely TE ancestors and determines the probability that each modern element arose from each of its plausible ancestors . We applied our method to the gibbon-derived LAVA TE family and to the human AluSc subfamily and inferred many more source elements than indicated by previous methods . This new method will help us better understand TE evolution , including both the impact of sequence on replication and the substitution process after replication . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"genomics",
"genome",
"evolution",
"genome",
"analysis",
"evolutionary",
"processes",
"evolutionary",
"modeling",
"genetics",
"biology",
"and",
"life",
"sciences",
"comparative",
"genomics",
"molecular",
"genetics",
"computational",
"biology",
"evolutionary",
"biology",
"evolutionary",
"genetics",
"evolutionary",
"theory"
]
| 2014 | Inference of Transposable Element Ancestry |
In understanding the etiology of breast cancer , the contributions of both genetic and environmental risk factors are further complicated by the impact of breast developmental stage . Specifically , the time period ranging from childhood to young adulthood represents a critical developmental window in a woman’s life when she is more susceptible to environmental hazards that may affect future breast cancer risk . Although the effects of environmental exposures during particular developmental Windows of Susceptibility ( WOS ) are well documented , the genetic mechanisms governing these interactions are largely unknown . Functional characterization of the Mammary Carcinoma Susceptibility 5c , Mcs5c , congenic rat model of breast cancer at various stages of mammary gland development was conducted to gain insight into the interplay between genetic risk factors and WOS . Using quantitative real-time PCR , chromosome conformation capture , and bisulfite pyrosequencing we have found that Mcs5c acts within the mammary gland to regulate expression of the neighboring gene Pappa during a critical mammary developmental time period in the rat , corresponding to the human young adult WOS . Pappa has been shown to positively regulate the IGF signaling pathway , which is required for proper mammary gland/breast development and is of increasing interest in breast cancer pathogenesis . Mcs5c-mediated regulation of Pappa appears to occur through age-dependent and mammary gland-specific chromatin looping , as well as genotype-dependent CpG island shore methylation . This represents , to our knowledge , the first insight into cellular mechanisms underlying the WOS phenomenon and demonstrates the influence developmental stage can have on risk locus functionality . Additionally , this work represents a novel model for further investigation into how environmental factors , together with genetic factors , modulate breast cancer risk in the context of breast developmental stage .
In the United States , breast cancer is the most frequently diagnosed cancer and second leading cause of cancer death among women [1] . Its etiology is complex , consisting of the interaction of both genetic and environmental risk factors whose contribution to overall risk can vary depending on the developmental context of the individual . In general , time periods in which women are more susceptible to initiating events affecting their long term breast cancer risk are broadly referred to as Windows of Susceptibility ( WOS ) [2] . In humans , the best documentation of a WOS can be found in studies of radiation exposure in women . Women exposed to radiation between 0 and 30 years of age during either the atomic bombings of Japan or for the treatment of Hodgkin’s lymphoma had an increased risk of developing breast cancer later in life compared to women >30 years of age at time of exposure [3 , 4] . This time period , therefore , represents one of the WOS , and encompasses ages spanning childhood , adolescence , and young adulthood in women . Animals studies performed in rats to model the human WOS phenomenon [5] further suggest the existence of at least two mechanistically distinct susceptibility windows within the larger human WOS , namely , the sexually immature WOS ( iWOS ) and the adolescent WOS ( aWOS ) . This division of the WOS is most evident in work by Ariazi et al . [6] on a carcinogen-inducible model of breast cancer , where administration to developmentally immature ( 3 week ) and adolescent-aged ( 7 week ) rats resulted in differential carcinoma development depending on age of administration and the carcinogen used . Additionally , although over 80 genetic loci affecting breast cancer susceptibility have been identified in human genome-wide association studies ( GWAS ) [summarized in 7] , their function in relation to developmental stages has not been characterized . In general , while the effects of window specific exposures are well documented , the cellular mechanisms responsible for their function and governing their interactions with environmental and genetic risk factors are poorly understood . To begin to understand the complex interactions between WOS , genetics , and the environment , we turned to a comparative genomics approach , utilizing a rat model of breast cancer . The rat is an excellent model for this type of study , as not only does its mammary gland and mammary tumor development mimic that of the human condition [8] , but , as previously mentioned , it too displays the WOS phenomenon [5 , 6] . Additionally , inbred rat strains vary in their susceptibility to carcinogen-induced mammary cancer , allowing for the identification of genetic susceptibility loci through quantitative trait loci ( QTL ) analysis . This approach was applied in our lab , utilizing the mammary cancer resistant Wistar-Kyoto ( WKy ) and susceptible Wistar-Furth ( WF ) inbred rat strains resulting in the identification and subsequent fine-mapping of the Mammary Carcinoma Susceptibility 5c , Mcs5c , locus [9–11] . Mcs5c maps to a 170kb region located in a large gene desert on rat chromosome 5 that shares homology with mice and humans ( Fig 1 ) . In both chemical carcinogen and oncogene-induced models of mammary cancer , congenic lines homozygous for the resistant WKy Mcs5c allele showed an approximately 50% reduction in carcinoma number compared to susceptible WF-homozygous controls [11] . Using the Mcs5c locus as a model , we sought to examine the interaction between a genetic risk factor and WOS . We have characterized an 8 . 5kb temporal control element ( TCE ) within Mcs5c affecting the expression of neighboring gene Pregnancy-associated plasma protein A , Pappa , in a genotype-dependent manner in mammary epithelial cells ( MECs ) . The function of the Pappa/PAPP-A protein makes it an attractive candidate for involvement in both the WOS phenomenon and breast cancer development . PAPP-A is a protease that acts to positively regulate bioavailability and signaling of the Insulin-like growth factors , IGFs , through the cleavage of IGF binding proteins 2 , 4 and 5 , IGFBP2/4/5 [12–15] . The specific role of PAPP-A in normal breast development has not been studied , but the IGF-I pathway , in general , is an essential component of breast/mammary gland development , as evident by the severe mammary gland defects of Igf-I and Igf-I receptor ( Igf1r ) knockout mice [16–18] . The role of IGF-I in breast cancer development is supported by numerous studies which associate the IGF-I signaling pathway with breast cancer initiation and progression [19] . Indeed , in transgenic mice , overexpression of IGF-I in the mammary gland resulted in increased susceptibility and decreased latency to spontaneous and carcinogen-induced mammary adenocarcinomas [20] . Limited studies of PAPP-A function in cancer have demonstrated that increased PAPP-A activity enhanced tumor growth in ovarian and lung cancer cell lines [21 , 22] , and inhibition of its proteolytic function reduced tumor growth in a murine mammary cancer cell line [23] . Furthermore , TCGA data [24] found PAPP-A to be altered in 6% of invasive breast carcinomas , with amplification/mRNA upregulation identified as the most common genetic alterations , and found co-amplification of neighboring loci , encompassing the homologous MCS5C locus , occurring in approximately 1–2% of cases ( accessed via www . cbioportal . org; [25 , 26] ) . In this study , we have identified age-specific differences in Mcs5c activity which support the existence of mechanistically distinct susceptibility windows . We have functionally characterized the non-coding Mcs5c locus , finding that it acts during the aWOS to regulate Pappa expression through age-dependent chromatin looping and genotype-dependent DNA methylation . To our knowledge , this study represents the first identification of a molecular mechanism underlying the aWOS phenomenon and highlights the ability of developmental age to influence the activity of a susceptibility locus .
To determine if Mcs5c exerts its effect on carcinoma multiplicity via the mammary gland , transplant experiments were performed . Donor mammary gland tissue from either the Mcs5c resistant 5C-27 line or a Mcs5c susceptible control line was transplanted onto the interscapular fat pad of recipient rats from both genotypes , creating four donor-recipient groups . This direct transplant design allowed for the detection of mammary gland-host interactions and did not result in differential tissue rejection rates , as the lines are isogenenic except at the Mcs5c locus . Transplant tissue rejection rates were not statistically significant between transplant groups consisting of donors and recipients with the same genotype versus groups with different genotypes ( Chi-squared test , X = . 10 , df = 1 , p-value = 0 . 75 ) . Results from the mammary gland transplant experiment are shown in Fig 2 . Resistant and susceptible rats receiving resistant donor tissue had a transplant site carcinoma incidence of 21% and 27% , respectively ( n = 76 , 49 ) , while resistant and susceptible rats receiving susceptible tissue had incidences of 42% and 38% , respectively ( n = 69 , 39 ) . Recipient rats of either genotype that received susceptible donor tissue had higher transplant site carcinoma incidences than those that received tissue from resistant rats . In this way , the carcinoma phenotype was dependent on the donor tissue genotype and was not influenced by the recipient’s genotype , suggesting that Mcs5c acts within the mammary gland . Indeed , logistic regression analysis found a statistically significant donor effect ( p-value = 0 . 0043; recipient effect p-value = 0 . 825 ) . Thus , it was concluded that Mcs5c acts in a mammary gland autonomous manner to influence carcinoma multiplicity . Quantitative real-time PCR ( qPCR ) was used to investigate expression levels of nearby genes in mammary epithelial cells ( MECs ) of Mcs5c resistant and susceptible rats at 4–12 weeks of age . This age range was chosen as it captures multiple mammary gland developmental windows , including the iWOS ( 4 weeks ) , aWOS ( 6–9 weeks ) , and adult ( 12 weeks ) time periods . Pappa , located over 517kb away from Mcs5c , was found to be differentially expressed in MECs in an age-dependent manner ( Fig 3 ) . In general , Pappa expression levels were dynamic in Mcs5c susceptible MECs during development , while Mcs5c resistant expression remained relatively steady over time . Compared to Mcs5c resistant rats , Pappa expression was increased in susceptible rats by 43% at 6 weeks ( Mann-Whitney U test , p-value = 0 . 015 , n = 13 and 15 , respectively ) , 14% at 7 weeks ( p-value = 0 . 05 , n = 23 and 19 ) , and 31% at 9 weeks ( p-value = 0 . 0003 , n = 23 and 18 ) . Differential expression disappeared by 12 weeks of age ( n = 9 and 18 ) , at which point the mammary gland is fully developed and rats are past the aWOS stage [5] . Expression trends were reversed in 4 week old rats , with susceptible animals showing a sharp decrease in expression relative to resistant rats ( p-value = 8e-5 , n = 9 and 8 , respectively ) . Mcs5c , therefore , appears to functioning during both the iWOS and aWOS . Unfortunately , we were unable to obtain robust antibodies for analysis of Pappa protein levels in mammary gland tissue . Differential expression in MECs was not observed for neighboring genes Tenascin C , Tnc , and Tumor Necrosis Factor ( Ligand ) Superfamily , Member 15 , Tnfsf15 , during the aWOS . However , differential expression of Tnfsf15 was observed in 4 week-old , immature MECs , highlighting the complexity and age-specific nature of Mcs5c locus activity ( S1 Fig ) . Genotype dependent differential expression seen in MECs led to the hypothesis that Mcs5c contained a long-distance acting regulatory element influencing Pappa expression . Such a relationship could be mediated by a physical association between the two regions , resulting in the looping out of intervening DNA sequence . Chromosome conformation capture ( 3C ) was used to identify such an interaction . To create 3C templates , MECs were isolated from the mammary glands of Mcs5c resistant and susceptible animals at 4 , 6 , 7 , and 12 weeks of age . Two fixed bait regions located at the Pappa locus were chosen for extensive analysis of potential interactions with Mcs5c . These regions , P3-1 and P4-1 , span approximately 2 . 4kb and 2kb in size , respectively , with P3-1 encompassing Pappa exon one and a conserved CpG island , and P4-1 falling within the first intron ( Fig 4A and 4C ) . These two regions were chosen for analysis as their degree of sequence conservation suggested that they may be functionally relevant in transcriptional regulation of the Pappa gene ( Fig 4A ) . Bait region P3-1 was negative for any interaction with Mcs5c at 4 , 7 , and 12 weeks of age ( S2 Fig ) . Conversely , 3C analysis using bait region P4-1 revealed an 8 . 5kb region within Mcs5c that displayed a high relative interaction frequency ( IF ) in 6 and 7 week templates , indicative of a physical interaction between the two regions occurring over a distance of 590kb ( Fig 4D ) . 4 and 12 week templates had a much lower IF at this -590 region , leading to the formation of two distinct , age-dependent interaction groups displaying either a strong ( 6 and 7 week ) or weak ( 4 and 12 week ) IF . The difference in IF for these two groups was statistically significant ( Mann-Whitney U test , p-value = 1 . 02e-10 , n = 27 and 38 biological replicates , respectively ) . For all ages , there was no difference in IF between genotypes , indicating that the interaction is age-dependent but not genotype-dependent . We will therefore refer to the -590 looping region of Mcs5c as the temporal control element ( TCE; chr5:84 , 428 , 694–84 , 437 , 192; RGSC 5 . 0/rn5 ) . Three additional Pappa bait regions were tested for interactions with the Mcs5c TCE at 4 and 6 weeks of age ( S2 Fig ) . Two of these regions , P4-1A and P4-2 , were negative , while the more proximal P3-3 region displayed an aWOS-specific looping interaction that mimicked the TCE/P4-1 interaction . This indicates that the Mcs5c TCE may utilize a more complex looping scheme to facilitate Pappa regulation , and defines the TCE as a functionally important region within Mcs5c . To determine if these interactions are also tissue-specific , 3C profiles were analyzed from 4 and 7 week colon epithelial cells and 7 week liver hepatocytes from Mcs5c resistant rats . The Mcs5c TCE did not interact with P4-1 ( Fig 4E ) or P3-3 ( S2 Fig ) in these tissues , implying that the interactions between Pappa and the Mcs5c TCE are tissue-specific in addition to age-dependent . Sequencing of the resistant WKy and susceptible WF TCE alleles revealed 10 variants between the two ( S7 Table ) , and although our 3C results showed that age-specific looping occurs independent of genotype , we speculate that one or more variants may be involved in genotype-dependent expression differences observed during this time period . CpG island ( CGI ) shores are regions located approximately 2kb away from CGIs , and have increasingly been identified as the sites of tissue specific differential methylation associated with gene expression changes [27] . The Pappa looping fragment , P4-1 , resides in a CGI shore region ( Fig 5A ) . As this region is a target site of Mcs5c TCE looping , we hypothesized that Mcs5c may affect Pappa expression through an epigenetic mechanism targeted to the P4-1 fragment . Methylation levels for 12 CG dinucleotides within and proximal to P4-1 were examined in MECs of Mcs5c resistant and susceptible rats at 4 , 6 , 7 , 9 and 12 weeks of age using custom designed pyrosequencing assays ( Fig 5A ) . Selection of these timepoints allowed for the examination of methylation patterns before , during , and after the aWOS . In general , methylation levels were dynamic across this region , with sites 2–4 consistently displaying the lowest methylation levels ( average = 13% methylated ) and sites 9–12 displaying the highest levels ( average = 68% methylated ) ( S4 Table ) . Additionally , there appeared to be few age-specific differences in methylation levels for animals within the aWOS , therefore data for 6 , 7 , and 9 week old rats were combined within genotypes . Of the 12 sites examined , 6 showed statistically significant genotype-dependent differences in methylation levels after adjusting for multiple comparisons ( Mann-Whitney U-test with Bonferroni correction ) . The percent change in methylation levels along with p-values are shown in Table 1 . All statistically significant , genotype-dependent methylation differences occurred during the aWOS and were directionally identical , with methylation levels decreased in Mcs5c susceptible MECs . The percent decrease in methylation levels ranged from 5 . 0%– 22 . 7% . Additionally , a number of other sites displayed a similar trend , although these differences were not significant after Bonferroni correction . At ages outside of the aWOS , there were no statistically significant genotype-dependent differences in methylation , although sites 1 and 2 displayed a non-significant trend of increased methylation in Mcs5c susceptible MECs at the 4 week time point . We also investigated the methylation state of the Pappa CGI using 2 pre-made pyrosequencing assays ( Fig 5A ) . Methylation levels for both assays were assessed in 4 week old animals , while one assay was examined at the remaining timepoints . For all CGI assays and timepoints , there were no genotype-dependent differences in methylation levels and , in general , the Pappa CGI is hypomethylated at all ages , with site specific methylation levels ranging from 0 . 16% - 8 . 41% ( S5 Table ) . The observation of decreased shore methylation and increased Pappa expression in Mcs5c susceptible MECs strongly supports the canonical role of DNA methylation in gene regulation , that is , that the two are negatively correlated . Indeed , for 6 week MECs , for which we had both DNA and RNA samples , Pappa expression was negatively correlated with the average methylation percentage of the 6 significant shore sites ( Fig 5B; Pearson correlation coefficient , R , = -0 . 67 , n = 18 , p-value = 0 . 0023 ) . By contrast , no correlation was observed between Pappa expression and the average methylation percentage of the CGI-2 assay sites ( Fig 5C; Pearson correlation coefficient , R , = 0 . 16 , n = 18 , p-value = 0 . 52 ) . The identification of genotype-dependent methylation differences during the aWOS suggests that Mcs5c facilitates genotype-dependent Pappa expression differences observed during this time period through epigenetic modification of the Pappa CGI shore . In an effort to causally tie the Mcs5c TCE to Pappa expression and CGI shore methylation , the entire 8 . 5kb region was targeted for deletion in the rat mammary carcinoma cell line , LA7 . Two CRISPR guides were used to target the region , and clones were screened via PCR across the cut site , with validation by sequencing ( S1 Table ) . We were unable to identify a clone with all copies of the TCE removed , despite much effort . This was likely due to the aneuploid nature of LA7 cells , and mutations incurred at CRISPR guide target regions ( S2 & S3 Tables ) . Copy number analysis of 9 positive CRISPR edited clones showed that we were able to delete a majority of TCE copies , reducing the copy number by 3 . 5-fold across all clones ( Fig 6A ) . 3C analysis of positive clones indicated that removal of multiple TCE copies resulted in decreased TCE/P4-1 looping , but did not alter TCE/P3-3 looping , which remained consistent with WT levels ( Fig 6B ) . Interestingly , this suggests that the looping mechanisms responsible for these interactions are functionally distinct . Expression analysis revealed a significant reduction in Pappa expression in CRISPR clones compared to wild-type LA7 cells , with Pappa decreased 4-fold across all clones ( Fig 6C ) . A Pearson correlation coefficient was computed to determine the relationship between Pappa expression and TCE copy number , and a positive correlation between the two was observed ( Fig 6D; R = 0 . 6245 , n = 13 , p-value = 0 . 0225 ) . Conversely , no change in Tnc or Tnfsf15 expression was observed with TCE knockdown , and expression levels were not correlated to TCE copy number ( S3 Fig ) . These data support our hypothesis that Mcs5c contains a long-range regulatory element , and emphasizes the functionality of the TCE/P4-1 chromatin loop to Pappa gene expression . Our in vivo analysis highlighted the importance of Pappa CGI shore methylation to Pappa expression , and we sought to verify this relationship in our in vitro model as well . Treatment of wild-type LA7 cells with the DNA methylation inhibitor 5-aza-2’-deoxycytidine ( 5-aza-dC ) resulted in a 15-fold increase in Pappa expression ( Fig 6E ) , indicating that DNA methylation plays a role in Pappa regulation . To more specifically address the relationship between TCE/P4-1 looping and Pappa methylation , CGI and CGI shore methylation levels were analyzed in wild-type LA7 cells and CRISPR edited clones ( S10 Table ) . Two CGI shore sites showed statistically significant differences in methylation , with a 3 . 9% decrease and a 25% increase in methylation levels observed in CRISPR clones at sites 5 site 12 , respectively ( S10 Table and Fig 6F ) . Methylation changes at site 12 were more pronounced , and were investigated further . We found site 12 methylation levels to negatively correlate with TCE copy number ( Fig 6G; R = -0 . 8034 , n = 13 , p-value = 0 . 0009 ) . Site 12 methylation levels were also negatively correlated with Pappa expression ( Fig 6H; R = -0 . 6022 , n = 17 , p-value = 0 . 011 ) , mimicking the observed in vivo relationship . Altogether , these data suggest a functional chain of events whereby the TCE , via the TCE/P4-1 loop , affects Pappa CGI shore methylation levels which then , in turn , affect Pappa expression .
Previous work on Mcs5c had fine-mapped the locus to a 170kb non-coding region on rat chromosome 5 . This locus resulted in an approximately 50% decrease in both chemical carcinogen and oncogene-induced mammary carcinoma development when homozygous for the resistant WKy allele . The gene Tnc was initially identified as a possible target of Mcs5c activity , with genotype-dependent differential expression observed in the thymus and ovaries exclusively following carcinogen exposure [11] . However , in this study , we have shown that the Mcs5c locus affects carcinoma multiplicity in a mammary gland autonomous manner ( Fig 2 ) . This suggests that the previous non-mammary gland expression differences observed following carcinogen exposure do not play a role in carcinoma initiation , and are either irrelevant or secondary to initial carcinoma development that is dependent on mammary gland intrinsic factors . While these hypotheses warrant further investigation , a reevaluation of gene expression within the mammary gland was conducted , revealing genotype-dependent and age-specific differential expression of Pappa in mammary epithelial cells ( MECs; Fig 3 ) . Specifically , Mcs5c susceptible MECs from 6 to 9 week old rats showed increased expression of Pappa compared to Mcs5c resistant rats . Importantly , the 6 to 9 week age range encompasses a time period of rapid mammary gland development and maturity , falling within the aWOS [5] . We hypothesized that Pappa expression changes were mediated by a regulatory element within Mcs5c , and our experimental results support this hypothesis , identifying a complex set of mechanisms underlying Mcs5c-mediated regulation of Pappa . Through 3C experiments , we have identified a region within Mcs5c , the temporal control element ( TCE ) , that physically interacts with the Pappa locus at two regions , P4-1 and P3-3 , in an aWOS- and MEC-specific manner over distances of 590kb and 580kb , respectively ( Fig 4 and S2 Fig ) . The importance of the TCE/P4-1 long-range looping interaction to Pappa expression was demonstrated in vitro , where removal of TCE copies resulted in a reduction of TCE/P4-1 , but not TCE/P3-3 , looping , and correlated with decreased Pappa expression ( Fig 6 ) . These data indicate that the two observed TCE chromatin interactions are functionally distinct , and demonstrates a strong positive regulatory relationship between the Mcs5c TCE and Pappa expression , which appears to be dependent on TCE/P4-1 chromatin looping . The Mcs5c TCE/Pappa P4-1 interaction , therefore , represents another example of a long-distance acting regulatory region , akin to those identified for the Shh [28] and Sox9 [29] genes , as well as the previously characterized Mcs1a locus [30] . Importantly , as looping occurs in a genotype-independent manner , additional mechanisms must be responsible for the differential expression observed between Mcs5c resistant and susceptible rats . With the intronic P4-1 looping region falling in a CGI shore , DNA methylation of this region became a mechanistic candidate to explain observed expression differences . The importance of differentially methylated CGI shores to gene expression was first highlighted by Irizarry and colleagues in 2009 [27] . Since then , many studies have shown an association between differentially methylated shore regions and gene expression changes [31–38] . Our study identified 6 CG dinucleotides within and proximal to the P4-1 looping region that were differentially methylated between Mcs5c resistant and susceptible MECs ( Table 1 ) . Significant methylation differences were observed exclusively during the aWOS , and a negative correlation between shore methylation levels and Pappa expression strongly suggest that DNA methylation plays a role in differential Pappa expression ( Fig 5B ) . This correlation was recapitulated in our in vitro model , where differential shore methylation was also negatively correlated with TCE copy number ( Fig 6G and 6H ) . As copy number acts as an indicator of TCE/P4-1 looping frequency in this model , this suggests a functional relationship between looping and shore methylation , where the TCE/P4-1 loop acts to facilitate differential methylation that , in turn , regulates Pappa expression . Given the inherent difficulties of modeling an age-dependent phenomenon in vitro , these results must be interpreted cautiously; however , we feel that the similarities between our in vitro and in vivo results indicate that these mechanisms are robust and functionally relevant to Mcs5c-mediated Pappa regulation . Overall , we have identified two mechanisms associated with Mcs5c regulation of Pappa expression during the aWOS , chromatin looping and DNA methylation . Our in vitro experiments have indicated the importance of the TCE/P4-1 loop for Pappa expression and shore methylation; however , in vivo analyses have shown that these actions are mechanistically distinct , as Pappa expression and differential methylation , but not looping , are genotype-dependent . An unresolved issue is precisely how Mcs5c is mediating these activities . We hypothesize that the genotype-independent TCE/P4-1 loop serves to facilitate the recruitment of transcription factors , cofactors , and/or methyltransferases that act separately or together to directly regulate Pappa methylation and expression during the aWOS . The binding of these regulatory factors would be affected by one or more variants within the TCE without affecting chromatin looping . Wright et al . [39] identified a similar interaction at the c-MYC locus , where an enhancer-associated SNP affected transcription factor binding without altering chromatin structure . Sequencing of the resistant WKy and susceptible WF Mcs5c TCE alleles ( chr5:84 , 428 , 694–84 , 437 , 192; RGSC 5 . 0/rn5 ) has revealed 10 candidate polymorphisms ( S7 Table ) for future investigation of their effect on protein binding and subsequent Pappa expression and methylation changes . Mcs5c activity during the aWOS stands in stark contrast to that observed during the iWOS ( 4 weeks ) . Specifically , differential Pappa expression during the iWOS is reversed compared to the aWOS ( Fig 2 ) , TCE/Pappa looping is lacking ( Fig 4D and S2 Fig ) , and there are no statistically significant CGI shore methylation differences ( Table 1 ) . These data indicate that the regulatory actions of Mcs5c are dependent on developmental context , a phenomenon observed at other regulatory regions , most notably the β-globin locus control region [40] . Age-specific differences in Pappa expression , looping , and methylation could be explained by interactions with proteins specific to these developmental time points . Identifying proteomic differences between the immature and adolescent mammary gland will be crucial in understanding the players driving window-specific mechanistic differences in Mcs5c activity . We hypothesize that age-specific protein expression results in an alternative looping interaction between Mcs5c and Pappa during the iWOS . Differential expression of Tnfsf15 exclusively at the 4 week time point ( S1 Fig ) indicates that Mcs5c may exhibit a more complex chromatin interaction during the iWOS , regulating multiple genes simultaneously . Additionally , a trend towards increased methylation in Mcs5c susceptible MECs is functionally consistent with the reduction of Pappa expression observed during this time period . It is possible that these sites are indicative of significant methylation differences occurring at sites not examined in this study , both at the Pappa locus as well as Tnfsf15 , and shore methylation may still , therefore , be relevant to Mcs5c activity during the iWOS . Altogether , we have functionally characterized the Mcs5c locus , finding that it acts via two distinct mechanisms to influence Pappa expression in an age-dependent manner during a well-characterized breast cancer WOS ( Fig 7 ) . This work highlights the importance of characterizing genetic risk factors in the context of developmental windows of susceptibility ( G x WOS ) , and emphasizes the complex interaction between genetic , environmental , and age-specific risk factors . Mcs5c susceptible rats showed increased expression of Pappa in MECs and an increased susceptibility to carcinogen-induced mammary carcinogenesis , supporting a protective benefit of reduced Pappa levels during adolescent development . Decreased levels of Pappa in the developing mammary gland would result in reduced Igf-I bioavailability through a reduction in Igfbp cleavage [12] . Given that the Igf-I signaling pathway acts to promote proliferation and inhibit apoptosis during mammary gland development [41] , it is therefore likely that a reduction in free Igf-I would reduce the proliferative index of MECs . As the effects of many environmental mutagens , such as radiation and chemical carcinogens , are dependent on interactions with the DNA of proliferating cells [42] , this would result in fewer targets for mutagenesis , and represents one possible method by which reduced Pappa expression during the aWOS may result in a mammary carcinoma resistant phenotype . Understanding the mechanisms behind G x WOS interactions and how environmental risk factors influence these interactions will play a crucial role in breast cancer risk assessment , and in the identification of targets and strategies for cancer prevention in young women . There is growing concern over the impact adolescent exposure to a broad range of environmental factors may have on long-term breast cancer risk [43] . Our work has demonstrated a functional relationship between genetic risk factor activity and developmental stage , and it is likely that environmental risk factors may further confound such interactions . Indeed , CGI shore methylation has been found to be affected by environmental factors such as ELS and diet [32 , 33 , 44 , 45] . We believe that the Mcs5c locus will serve as a robust model to study how environmental factors affect breast cancer risk by influencing G x WOS interactions , and may encourage the characterization of other such cancer susceptibility loci in this context .
Congenic rat lines were maintained in an AAALAC-accredited facility as previously described [11] . All protocols were approved by the University of Wisconsin–Madison School of Medicine and Public Health Animal Care and Use Committee . Congenic rat lines are defined as having the resistant Wistar-Kyoto ( WKy ) Mcs5c allele introgressed on a susceptible Wistar-Furth ( WF ) background . The resistant congenic line used in this study , 5C-27 , is WKy-homozygous for a genomic region that includes the entirety of the Mcs5c locus ( Fig 1 ) [11] . Susceptible control animals are WF-homozygous at the Mcs5c locus . Mcs5c WKy-homozygous congenic rats from line 5C-27 were used as resistant donors and recipients ( Mcs5c resistant ) , and Mcs5c WF-homozygous rats were used as susceptible controls ( Mcs5c susceptible ) . Abdominal and inguinal mammary glands were collected from female donor rats aged 30–35 days old , scissor minced , and split into four equal volumes . One volume was then grafted onto the interscapular white fat pad of four different 30–35 day old female recipient rats . Three weeks after transplantation ( 51–56 days ) , recipients were administered the chemical carcinogen 7 , 12-dimethylbenz ( a ) anthracene ( DMBA ) , as a single oral dose dissolved in sesame oil at 65 mg/kg of body weight to induce mammary carcinoma formation . At 15 weeks post-DMBA , animals were removed from the study and carcinomas present at the transplant site greater than 3x3mm were counted . Generally , recipient rats developed ≤1 carcinoma at the transplant site , so incidence values were used as opposed to multiplicity . Fat pads were whole mounted and stained with aluminum carmine to verify transplant mammary gland growth . Four transplant groups were studied , with resistant 5C-27 or susceptible donor glands transplanted into both resistant and susceptible recipients ( R->R , R->S , S->R , S->S ) . The tissue rejection rate for transplant groups consisting of donors and recipients with the same genotype ( R->R , S->S ) was compared to the rejection rate for groups with differing donor and recipient genotypes ( R->S , S->R ) via a Chi-squared test . The effect of donor and recipient genotype on carcinoma incidence , converted to a binary response value , was analyzed using logistic regression with two independent variables ( donor genotype and recipient genotype ) and no interaction term . For all experiments , mammary epithelial cell ( MEC ) isolation began with fresh mammary glands ( abdominal and inguinal , with lymph nodes removed ) that were finely minced and digested for 2 hours at 37°C in 10 mL of GIBCO Dulbecco’s modified Eagle’s medium/F12 ( DMEM/F12; Life Technologies ) containing 0 . 01 g/mL of type III collagenase ( Worthington ) . Cell pellets were collected by centrifugation and resuspended in 5 mL DMEM/F12 . The suspension was loaded onto a 40μm nylon filter to eliminate stromal cells and collect mammary ductal fragments , consisting of an enriched MEC population . DNA was isolated from cells via the DNeasy Blood and Tissue kit ( Qiagen ) . To isolate RNA , cells were homogenized in TRI Reagent ( Ambion ) , followed by RNA extraction using the MagMAX-96 for Microarrays Total RNA kit ( Ambion ) . LA7 cells used for downstream analysis were collected via treatment with 0 . 25% trypsin/EDTA ( Life Technologies ) . RNA was extracted using the RNeasy Mini Kit ( Qiagen ) and DNA was extracted using the DNeasy Blood and Tissue kit ( Qiagen ) . MECs were collected at 4–12 weeks of age from female Mcs5c resistant and susceptible control rats . RNA was isolated as described above . For in vivo and in vitro expression analysis , cDNA was prepared from 1–2μg total RNA using Superscript II reverse transcriptase ( Invitrogen ) . Gene expression was quantified using pre-designed or custom made TaqMan qPCR assays ( Pappa , Rn01458295_m1 , FAM; Tbp , Rn01455646_m1 , VIC; Tnc , probe-FAM 5’ CGAGAGCTGTGATTAGA 3’ , primers 5’ GGCTGTCAGAAGGCCAGATG 3’ and 5’ TGCCATGAAGGGATTTGAAGA 3’; Tnfsf15 , Rn00595596_m1 , FAM ) and run on an ABI Prism 7900HT ( Applied Biosystems ) . Tbp was chosen as the reference gene as its expression has been found to be relatively stable across a variety of rat tissues and during different stages of the estrous cycle in the mammary gland [46] . cDNA was diluted 1:4 or 1:8 and run using reaction conditions described previously [11] . Transcript quantities were calculated as described in Smits et al . [30] , using a standard curve method to calculate Ct values and extrapolate quantity values . Sample measurements are an average of 3–4 technical replicates and data were analyzed using SDS software version 2 . 2 . 2 ( Applied Biosystems ) . Sample templates were prepared from MECs , colon epithelial cells , liver hepatocytes , and LA7 cells . MECs were isolated from 4 , 6 , 7 , and 12 week old Mcs5c resistant and susceptible rats and the resulting cell suspension was diluted in PBS and fixed via the addition of 1 . 5% formaldehyde . Colon epithelial cells were isolated from 4 and 7 week old resistant rats , processed as described in Whitehead et al . [47] , and fixed in formaldehyde . To isolate hepatocytes , the livers of 7 week old resistant rats were digested via cannulation of the portal vein and blanching of the liver using a pre-warmed solution of HBSS ( Gibco ) + 0 . 5mM EGTA followed by digestion via pre-warmed DMEM-low glucose ( Gibco ) + 1000CDU/mL Collagenase type IV ( Worthington ) . Digested livers were collected in DMEM/F12 + 10% FBS on ice and cells dispersed manually . The suspension was filtered through a 100μm nylon filter and the filtrate spun for 2 minutes at 50xg . Supernatant was removed and cell pellets were washed until media became clear , followed by fixation in formaldehyde . Bacterial artificial chromosomes ( BACs ) encompassing the rat Mcs5c and Pappa promoter regions ( CH230-433D12 , CH230-498D4 , CH230-256M9 , and CH230-244C7 ) were ordered from Children’s Hospital Oakland Research Institute ( CHORI ) and used as positive control templates . Subsequent template preparation for all cell types and for BAC controls continued as described in detail in Smits et al . [48] . The restriction enzyme used was BglII . Chromatin interactions were detected via PCR , with bait primers located at the Pappa gene tested against Mcs5c primers spanning the entire locus ( Fig 4A and 4B ) . Primer sequences are listed in S6 Table . Reaction components were 1X Herculase reaction buffer , 0 . 2mM dNTPs , 0 . 4μM primers , 0 . 3μμL Herculase Enhanced polymerase ( 5U/μL , Agilent ) in a total volume of 25μμL . The amount of DNA template to add and optimal annealing temperatures were determined empirically . PCR reactions were performed using the following cycling conditions: 95°C for 1 min , 36 cycles of 95°C for 30 s , Ta for 30 s , 72°C for 20 s , followed by a final extension of 72°C for 8 min . Reactions were analyzed by agarose gel electrophoresis and visualized by ethidium bromide staining . Band intensities were quantified using ImageQuant software ( GE Healthcare ) . A relative interaction frequency ( IF ) was calculated by dividing the band intensity of the sample templates by that of the BAC control . Sequencing of the 8 . 5kb Mcs5c looping region ( TCE; chr5:84 , 428 , 694–84 , 437 , 192; RGSC 5 . 0/rn5 ) identified in 3C experiments was performed on MEC DNA from Mcs5c resistant and susceptible rats to assess polymorphisms between the WKy and WF alleles . Sequencing primers are listed in S7 Table . Traditional Sanger sequencing was performed at the University of Wisconsin–Madison Biotechnology Center DNA sequencing facility as described in Smits et al . [30] . The rat mammary carcinoma cell line , LA7 , was obtained from the American Type Culture Collection and maintained in DMEM/F12 supplemented with 100 IU/mL penicillin , 100 μg/mL streptomycin ( Life Technologies ) , 5% FBS ( HyClone ) , and 0 . 005mg/mL insulin . Gene expression analysis proceeded as described above , and copy number analysis was performed via SYBR Green qPCR ( Life Technologies ) . For 5-aza-2’-deoxycytidine ( 5-aza-dC; Sigma ) experiments , cells were treated with 1μM 5-aza-dC for 48hrs followed by cell collection and processing . For quantification of Mcs5c TCE copies , a primer set located within the CRISPR targeted region was used ( 5’ CAATCACGTTCACTGTGGGT 3’ and 5’ TCACCTCACACTACCCCATG 3’ ) and as a control region , a primer set located within the non-targeted Pappa gene was used ( 5’ TCCTCCTGACCACTCTGAGA 3’ and 5’ CCCTACAAACAGCAGAGGGA 3’ ) . The CRISPR-Cas9 plasmid pSpCas9 ( BB ) -2A-Puro ( PX459 ) was provided by Dr . Feng Zhang ( Addgene plasmid #48139 ) [49] . Guide sequences were designed using the CRISPR Design Tool ( http://crispr . mit . edu ) and phosphorylated and annealed guide oligos were inserted into the PX459 plasmid via a combination digestion/ligation reaction . 100ng PX459 plasmid was mixed with 2μL of oligos ( diluted 1:250 ) , 1μL Fast Digest BbsI ( Thermo Scientific ) , 1X Fast Digest Buffer , 1mM DTT , 1mM ATP , and 1500 units T7 ligase ( New England BioLabs ) and incubated in a thermocycler for 5 minutes at 37°C followed by 5 minutes at 23°C for a total of 6 cycles . The resulting reaction was then treated with Exonuclease V ( NEB ) according to the manufacturer’s protocol . The product was transformed into competent cells , and colonies expanded and verified via sequencing of the guide insertion site . LA7 cells were transfected with two CRISPR guides flanking the 8 . 5kb target region . Transfection was performed via electroporation using a Nucleofector II Device and Amaxa Cell Line Nucleofector Kit V ( Lonza ) , according to the manufacturer’s instructions . Stable clones were isolated following puromycin selection , and clonal colonies were expanded . Removal of the targeted region was determined via PCR screening and sequencing . Primers used to create guides and screen clones are listed in S8 Table . DNA was isolated from wild-type LA7 and CRISPR edited cells as well as Mcs5c resistant and susceptible MECs at 4 , 6 , 7 , 9 , and 12 weeks of age . Bisulfite conversion was carried out on 500ng of DNA using the EZ DNA Methylation-Lightening kit ( Zymo Research ) , according to the manufacturer’s instructions . Four primer sets were designed to amplify the 12 CpG sites of interest within the Pappa CpG island ( CGI ) shore . Their sequences , along with the sequencing primers used for pyrosequencing , are listed in S9 Table . Optimal amounts of template DNA , MgCl2 , primers , and annealing temperature were experimentally determined ( S9 Table ) . In general , PCR reactions were performed using the following cycling conditions: 95°C for 5 min , 50 cycles of 95°C for 15 s , Ta for 30 s , 72°C for 30 s , followed by a final extension of 72°C for 5 min . 15μL of PCR product was used for pyrosequencing on a PyroMark MD instrument ( Qiagen ) , with 2–3 technical replicates per sample . Data were analyzed using PyroMark CpG software ( v 1 . 0; Qiagen ) . For analysis of the Pappa CGI , 2 pre-made assays were obtained from Qiagen ( CGI-1 , Rn_D3ZNQ7_01_PM; CGI-2 , Rn_D3ZNQ7_02_PM ) , with PCR conditions following manufacturer’s recommendations . Both pre-made assay CGI-1 and CGI-2 amplified 5 CpG sites within the Pappa CGI , for a total of 10 sites in the island examined . For statistical analysis of methylation differences , the non-parametric Mann-Whitney U test was used , with a Bonferroni correction applied for multiple comparisons . | A woman’s lifetime risk of developing breast cancer is affected by both genetic and environmental risk factors that can be further exacerbated by breast developmental stage . Time periods conferring increased risk are referred to as Windows of Susceptibility ( WOS ) and , generally speaking , the molecular mechanisms responsible for their effect on breast cancer risk are unknown . Our work presented here on the characterization of the rat Mammary Carcinoma Susceptibility 5c , Mcs5c , locus has identified a region within Mcs5c that interacts with the neighboring gene , Pappa , in an age-dependent manner to influence gene expression via genotype-dependent DNA methylation . Importantly , Mcs5c-mediated gene regulation occurs specifically within a WOS , and these finding represent the first identified molecular mechanisms by which a WOS influences the ability of a locus to affect mammary/breast cancer risk . This work highlights the importance developmental stage can have on genetic risk factor function , and we anticipate that the Mcs5c locus will serve as a model for future studies on WOS in combination with genetic and environmental risk factors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biotechnology",
"genome",
"engineering",
"medicine",
"and",
"health",
"sciences",
"breast",
"tumors",
"reproductive",
"system",
"engineering",
"and",
"technology",
"synthetic",
"biology",
"carcinomas",
"cancers",
"and",
"neoplasms",
"cloning",
"synthetic",
"bioengineering",
"crispr",
"oncology",
"epigenetics",
"dna",
"molecular",
"biology",
"techniques",
"dna",
"methylation",
"chromatin",
"synthetic",
"genomics",
"research",
"and",
"analysis",
"methods",
"bioengineering",
"synthetic",
"genome",
"editing",
"chromosome",
"biology",
"gene",
"expression",
"breast",
"cancer",
"exocrine",
"glands",
"chromatin",
"modification",
"dna",
"modification",
"molecular",
"biology",
"genetic",
"loci",
"breast",
"tissue",
"biochemistry",
"cell",
"biology",
"nucleic",
"acids",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"mammary",
"glands"
]
| 2016 | The Non-coding Mammary Carcinoma Susceptibility Locus, Mcs5c, Regulates Pappa Expression via Age-Specific Chromatin Folding and Allele-Dependent DNA Methylation |
Pathogenic and commensal Neisseria species produce an Adhesin Complex Protein , which was first characterised in Neisseria meningitidis ( Nm ) as a novel surface-exposed adhesin with vaccine potential . In the current study , the crystal structure of a recombinant ( r ) Nm-ACP Type I protein was determined to 1 . 4 Å resolution: the fold resembles an eight-stranded β-barrel , stabilized by a disulphide bond between the first ( Cys38 ) and last ( Cys121 ) β-strands . There are few main-chain hydrogen bonds linking β4-β5 and β8-β1 , so the structure divides into two four-stranded anti-parallel β-sheets ( β1-β4 and β5-β8 ) . The computed surface electrostatic charge distribution showed that the β1-β4 sheet face is predominantly basic , whereas the β5-β8 sheet is apolar , apart from the loop between β4 and β5 . Concentrations of rNm-ACP and rNeisseria gonorrhoeae-ACP proteins ≥0 . 25 μg/ml significantly inhibited by ~80–100% ( P<0 . 05 ) the in vitro activity of human lysozyme ( HL ) over 24 h . Specificity was demonstrated by the ability of murine anti-Neisseria ACP sera to block ACP inhibition and restore HL activity . ACP expression conferred tolerance to HL activity , as demonstrated by significant 3–9 fold reductions ( P<0 . 05 ) in the growth of meningococcal and gonococcal acp gene knock-out mutants in the presence of lysozyme . In addition , wild-type Neisseria lactamica treated with purified ACP-specific rabbit IgG antibodies showed similar fold reductions in bacterial growth , compared with untreated bacteria ( P<0 . 05 ) . Nm-ACPI is structurally similar to the MliC/PliC protein family of lysozyme inhibitors . However , Neisseria ACP proteins show <20% primary sequence similarity with these inhibitors and do not share any conserved MliC/PliC sequence motifs associated with lysozyme recognition . These observations suggest that Neisseria ACP adopts a different mode of lysozyme inhibition and that the ability of ACP to inhibit lysozyme activity could be important for host colonization by both pathogenic and commensal Neisseria organisms . Thus , ACP represents a dual target for developing Neisseria vaccines and drugs to inhibit host-pathogen interactions .
Lysozymes are ubiquitous enzymes with N-acetylmuramoyl hydrolase activities , which hydrolyse the bacterial cell wall polymer peptidoglycan ( PG ) . PG is the major structural component of the bacterial cell wall: its major function is to provide resistance against turgor pressure and its cleavage results in bacteriolysis [1] . Therefore , host lysozymes are an important component of innate immunity , contributing to a first line of defence against bacterial colonization or infection . In humans , C-type lysozyme can be found on all mucosal surfaces and secretions [2 , 3] , including the respiratory airway [4] , the digestive tract [5] , milk [6] as well as in serum [7] . Bacteria engaging in commensal or pathogenic interactions with a human or animal host have evolved various strategies to evade lysozymal activity . One mechanism of lysozymal resistance that both Gram-positive and Gram-negative bacteria use is PG modification , which has been demonstrated by several pathogens , including Streptococcus pneumoniae , Staphylococcus aureus , Listeria monocytogenes , Neisseria meningitidis and Neisseria gonorrhoeae [8] . A second mechanism to protect bacteria against host lysozyme involves the production of lysozyme inhibitors . Expression of lysozyme inhibitors likely contributes to bacterial colonization and infection . These enzymes can also function as important mediators in bacteria–bacteria interactions , by modulating the generation of PG fragments and by providing protection against other bacterial lysozymes in bacterial competition [1] . Furthermore , bacterial lysozyme inhibitors have been shown to control autolysis by inhibiting the lytic activity of transglycoslyases , which are enzymes involved in the biosynthesis and maintenance of PG [9] . To date , these lysozyme inhibitors have been identified only in Gram-negative bacteria such as Escherichia coli [10 , 11] , Salmonella enteriditis [11] , Pseudomonas aeruginosa [11 , 12] and , more recently , Mycobacterium tuberculosis [13] , among others [1] . These inhibitors are located either in the periplasm or anchored to the luminal face of the outer membrane ( OM ) . However , lysozyme inhibitor ( s ) for several important Gram-negative pathogens , such as Legionella spp . , Campylobacter spp . , or Neisseria spp . have not been reported to date . For these pathogens , PG modification has been described [14 , 15] , although the existence of other mechanisms to counter host lysozymal activity cannot be excluded . Members of the genus Neisseria colonize mucosal surfaces: Neisseria meningitidis ( Nm , meningococcus ) , the causative organism of meningococcal meningitis and sepsis [16] , and the commensal organism Neisseria lactamica , both colonize the human nasopharyngeal mucosal epithelium [17] . Neisseria gonorrhoeae ( Ng , gonococcus ) is closely related to the meningococcus , but colonizes the mucosal epithelium of the reproductive tract of both men and women and causes the sexually transmitted disease gonorrhoea [18 , 19] . Recently , we described a novel adhesin termed the Adhesin Complex Protein ( ACP—NMB2095/NEIS2075 ) , a conserved protein found in the OM of meningococci , gonococci , N . lactamica and all other Neisseria spp . [20] . We identified that the majority of sequenced meningococcal isolates contained one of two distinct Nm-ACP type proteins: Nm-ACPI and Nm-ACPII , encoded by Alleles 1 and 2 respectively . Nm-ACP acts as a minor adhesin , mediating the adherence of meningococci principally with epithelial cells . A recombinant Nm-ACPI protein was also capable of inducing cross-protective bactericidal antibodies , suggesting that it might be a potential vaccine candidate for inclusion in second-generation meningococcal vaccines . In the current study , we have examined further the biological properties of Neisseria ACP proteins by determination of the crystal structure of Nm-ACPI . We show that it has some structural similarity to the MliC/PliC protein family , which are membrane-bound or periplasmic inhibitors of C-type lysozyme . We also present data that show that Neisseria ACP acts as a lysozyme inhibitor , fulfilling this ubiquitous function for which no protein has been reported so far for any of the Neisseria species , although the mechanism of inhibition appears to be different from the MliC/PliC family .
We previously described expression of the Nm-ACPI protein from N . meningitidis strain MC58 in E . coli as a recombinant ( r ) full length protein ( 124 residues ) with an N-terminal hexa-histidine tag linked to the leader peptide . The molecular mass ( Mr ) of this expressed protein was ~17 . 8 kDa; the protein was insoluble and was therefore purified under denaturing conditions [20] . Here , we optimized the expression of rNm-ACPI protein by inserting the full length sequence for expression and export to the periplasm; the signal peptide sequence is cleaved by the E . coli signal peptidase and the periplasm provides an oxidizing environment for the formation of a disulphide bond ( see below ) . rNm-ACPI was purified in soluble form by serial column chromatography steps , to give a homogeneous product that eluted as a single peak from a size exclusion chromatographic column , at an elution volume consistent with a monomer ( S1A and S1B Fig ) . Mass spectrometry was used to verify that the predicted signal peptide was cleaved after export to the periplasm and gave a mass of 12 , 306 Da , consistent with a mature polypeptide of 103 residues plus the C-terminal hexa-histidine tag . Crystallization trials were initiated on the purified protein , and native data were collected from a single crystal to 1 . 4 Å resolution ( Table 1 and Table 2 ) . The structure was solved by single wavelength anomalous diffraction from iodide ions , following a soak of a crystal in 0 . 4 M potassium iodide . The overall fold of Nm-ACPI is close to an 8-stranded β-barrel , stabilized by a disulphide bond between cysteine residues on the first and last β-sheet ( Fig 1A and 1B ) . However , there are few main-chain hydrogen bonds linking β4-β5 and β8-β1 , meaning that the structure divides into two four-stranded anti-parallel β-sheets ( β1-β4 and β5-β8; Fig 1C ) . The interaction between β8 and β1 is stabilised by the disulphide bridge between Cys38 and Cys121 . The computed surface electrostatic charge distribution showed that the face of the β1-β4 sheet is predominantly basic , whereas the β5-β8 sheet is more apolar , apart from the loop between β4 and β5 ( Fig 1C ) . Using the DALI fold comparison server [21] , we searched for proteins with similar structures: top ‘hits’ from the search were PliC from Salmonella typhimurium ( St-PliC ) , MliC from Pseudomonas aeruginosa ( Pa-MliC ) and PliC from Brucella abortus ( Ba-PliC ) . These form part of the MliC/PliC family of lysozyme inhibitor proteins from Gram-negative bacteria [22–24] . All three structures superposed onto Nm-ACPI with r . m . s . deviations of less than 2 . 0Å , and all preserved the same eight stranded β-fold ( Fig 1D ) , although sequence identity with Nm-ACPI was less than 20% in each case . A structure-based sequence alignment showed that the disulphide bridge between β1 and β8 was preserved in each case , although the sizes of the loops varied ( Fig 2 ) . In addition , we noticed that loop 4 from the MliC/PliC family proteins , which occupies the lysozyme active site , corresponded to loop 4 of Nm-ACPI , but was longer in the latter by 3 residues . The sequence motifs SxSGAxY and YxxxTKG , which were conserved in the MliC/PliC family proteins , were not retained in Nm-ACPI ( Fig 2 ) . These observations suggested potentially a different mode of Nm-ACPI binding to lysozyme . Since Nm-ACPI is similar in structure to the MliC/PliC family proteins , we used in vitro lysozyme inhibition assays to test the hypothesis that it could act as an inhibitor of C-type lysozyme isolated from hen egg-white ( Hewl ) and from human neutrophils ( HL ) . The Hewl assay uses a Micrococcus lysodiekticus suspension in which the bacterial cell walls are labelled with fluorescein to such a degree that the fluorescence is quenched . The suspension acts as the lysozyme substrate , such that release of the quenched fluorophore upon hydrolysis of peptidoglycan by Hewl action gives higher fluorescence signal intensities . Addition of doses of rNm-ACPI as low as 12 . 5 μg/100μL assay volume , containing 10U of Hewl , led to almost complete inhibition of Hewl activity , as judged by this lysozyme assay ( Fig 3A ) . Even at the lowest dose of rNm-ACPI tested , ~1 . 6μg /100μL assay volume , an ~70% reduction of Hewl activity was observed ( P<0 . 05 ) ( Fig 3A ) . As expected , treatment with a positive control antibody H04 VH-Ab , which binds and occupies the active site of lysozyme , inhibited Hewl activity completely , similar to rNm-ACPI treatment ( P<0 . 05 ) ( Fig 3B ) . By contrast , treatment with a negative control HEL4 VH-Ab , which binds to lysozyme but not at the active site , showed no inhibitory effect on Hewl activity ( P>0 . 05 ) ( Fig 3B ) . Next , the same assay was used under identical conditions to compare the inhibition of HL and Hewl activity by rNm-ACPI , using equimolar quantities of protein and lysozyme ( Fig 3C ) : in these experiments , Hewl activity was reduced by ~82% , whereas HL activity was virtually completely eliminated ( >99% reduction ) , reflecting a statistically significant higher efficiency of rNm-ACPI inhibition of the human enzyme ( P<0 . 05 ) . In addition to enzyme inhibition , the binding of rNm-ACPI to HL was analysed by MicroScale Thermophoresis ( MST ) , a method which uses a microscopic temperature gradient combined with detection of HL , which had been pre-labelled with NT-647-NHS fluorescent dye . The data were consistent with a single binding isotherm and a Kd value of 11 μM ( = 13 . 5μg/100μL assay volume ) was recorded ( Fig 3D ) . Ostensibly this relatively weak binding affinity is at odds with the more potent inhibition seen at lower concentrations in Fig 3A . However , inhibition in the enzyme assay in Fig 3A is carried out in the presence of substrate , which is not the case for the experiment in Fig 3D . It is possible that the Kd value of ACP for lysozyme is influenced by the presence of lysozyme substrate ( e . g . if ACP also binds to cell wall components ) . To exclude the possibility that lysozyme antimicrobial activity was responsible for the observed killing of M . lysodiekticus , we treated bacterial suspensions with HL that had been boiled for 1 h , a treatment that has been reported to destroy peptidoglycan hydrolase activity , whilst maintaining antimicrobial activity [25–27] . We confirmed that boiled HL had no activity against M . lysodiekticus ( Fig 3E ) , indicating that lysozyme does not induce M . lysodiekticus cell death by antimicrobial activity of the molecule . We examined also the kinetics of rNm-ACPI inhibition of HL activity by measuring lytic absorbance over time . In preliminary experiments to quantify HL concentrations that produced significant lysis , suspensions of M . lysodeikticus cells ( initial OD λ595nm of 1–1 . 2 , equivalent to 1 mg/ml ) were treated with increasing concentrations of HL ( 0–5 μg/ml ) and absorbance was measured every 5 min for a period of 2 h . A final HL concentration of 2 μg/ml was chosen as optimal for inducing a significant ( P<0 . 05 ) linear decrease in absorbance values of >50% after 2 h incubation with 1 mg/ml M . lysodeikticus cells ( S2 Fig ) . These assay parameters were used to examine the dose-dependent kinetics of rNm-ACPI inhibition of HL activity . A dose of 0 . 1 μg/ml rNm-ACPI conferred ~30–60% protection of M . lysodeikticus cells against HL lysis by 2 h ( Fig 4A ) . Higher doses were even more effective , with ≥0 . 25 μg/ml of rNm-ACPI providing ~80–100% protection by 2 h . Notably , similar levels of inhibition of HL activity by rNm-ACPI were observed after 24 h incubation , with the M . lysodeikticus cells still intact after prolonged exposure to enzyme ( Fig 4B ) . Doses of rNm-ACPI <0 . 1 μg/ml were not protective against HL lysis ( Fig 4A and 4B ) ( P>0 . 05 ) . Next , we examined the specificity of the rNm-ACPI inhibitory activity towards HL by using murine antisera raised against native protein to restore the activity of HL . Decomplemented murine antisera raised against the protein using a variety of adjuvants as described elsewhere [20] , were added to a mixture of rNm-ACPI ( 0 . 5 μg/ml ) , M . lysodeikticus cell suspension ( 1 mg/ml ) and HL ( 2 μg/ml ) . Antisera generated by rNm-ACPI immunisation delivered in Al ( OH ) 3 or liposome formulations significantly prevented rNm-ACPI protein from inhibiting HL activity by ≥ 95% ( P<0 . 05 ) after 2 h ( Fig 5A ) . In comparison , antisera raised to rNm-ACPI in saline prevented rNm-ACPI protein from inhibiting HL activity by ≥ 50% ( P<0 . 05 ) after 2 h ( Fig 5A ) , which is likely due to lower levels of antibody production through immunization ( S3 Fig ) . Furthermore , sham sera or normal mouse serum ( NMS ) did not prevent rNm-ACPI from inhibiting HL activity ( Fig 5B ) . The specificity of rNm-ACPI as an inhibitor of HL activity was also supported by the observation that a heterologous recombinant outer membrane protein , Nm-Macrophage Infectivity Potentiator ( rNm-MIP , NMB1567/NEIS1487 ) did not inhibit lysis of M . lysodeikticus induced by either Hewl ( Fig 3B ) or HL ( Fig 5A ) . Amino acid sequence alignment of the Nm-ACPI ( encoded by Allele 1 ) , Nm-ACPII ( encoded by Allele 2 ) and Neisseria gonorrhoeae Ng-ACP ( encoded by Allele 10 ) proteins showed that the Loop 4 predicted binding interface ( Fig 2 ) for the meningococcal proteins were identical ( S4 Fig ) , but two amino acid changes were observed in Ng-ACP ( Nm Val80 to Ng Met80 and Nm Glu81 to Ng Asp81 ) . Thus , to test the hypothesis that the inhibitory function of rNm-ACPI for lysozyme was a general property of Neisseria ACP proteins , similar experiments were done with rNm-ACPII ( represented by N . meningitidis strain MC161 ) and rNg-ACP ( represented by N . gonorrhoeae strain FA1090 ) . The same assay parameters used with rNm-ACPI protein ( Fig 4 and Fig 5 ) were used to examine the dose-dependent kinetics of rNm-ACPII and rNg-ACP inhibition of HL-induced lysis of M . lysodeikticus . Inhibition of HL activity by rNm-ACPII protein was indistinguishable from rNm-ACPI ( S5 Fig and S6 Fig ) and the same pattern of lysozyme inhibition in vitro is likely to be the case for N . lactamica ACP protein , which is identical to Nm-ACPI . Kinetic analyses demonstrated that rNg-ACP inhibited HL in an identical manner to rNm-ACP proteins , i . e . a dose of 0 . 1 μg/ml rNg-ACP conferred ~30–60% protection of M . lysodeikticus cells against HL lysis by 2 h ( Fig 6A ) . Higher doses were even more effective , with ≥0 . 25 μg/ml of rNg-ACP providing ~80–100% protection by 2 h ( Fig 6A ) and similar levels of inhibition of HL activity by rNg-ACP were observed after 24 h incubation ( Fig 6B ) . Doses of rNg-ACP and rNm-ACPII <0 . 1 μg/ml were not protective against HL lysis ( Fig 6 ) ( P>0 . 05 ) . Attempts to co-crystallize rNm-ACPI with Hewl or HL were unsuccessful , yielding only crystals of rNm-ACPI alone . Superposition of rNm-ACPI onto coordinates of members of the MliC/PliC family , where structures were available in complex with lysozyme , gave steric clashes which suggested that the mode of binding of rNm-ACPI to lysozyme was different . For example , loop 4 in rNm-ACPI is longer than its equivalent in Ba-PliC , leading to potentially unfavourable interactions with HL ( denoted by the arrow in Fig 7A ) . To obtain a more plausible model of the Nm-ACPI:HL complex , 10 , 000 independent docking simulations between rNm-ACPI and HL were carried using RosettaDock [28] . The resulting model ( Fig 7B ) gave a complex in which the HL molecule was significantly further away from ACP than from Ba-PliC , although this could be affected by loop flexibility which is not accounted for by RosettaDock . In this modelled complex , the β4-β8 β-sheet on Nm-ACPI forms an extensive interaction surface with HL , which could inhibit substrate access to the enzyme active site . To obtain independent experimental evidence to validate the model of the complex , NMR chemical shift mapping was used to identify regions of rNm-ACPI that were perturbed on binding to lysozyme . Using 1H , 15N , 13C triple resonance experiments , over 80% of the non-proline backbone amide resonances of rNm-ACPI were assigned for the sample conditions used in the experiments . At the high protein concentrations used for NMR data collection , we found that complexes of rNm-ACPI with HL precipitated . The complex of rNm-ACPI with Hewl was , however , soluble and therefore tractable to chemical shift analysis . To identify the lysozyme-binding site ( s ) on the surface of Nm-ACPI , 1H-15N HSQC spectra of 15N labelled Nm-ACPI alone and in complex with Hewl were compared ( S7 Fig ) . Small but selective chemical shift changes are observed for a number of Nm-ACPI resonances . Most significant are the shift perturbations observed for a contiguous stretch of peptide , spanning residues D74 to T82; when mapped onto the structure of Nm-ACPI , these peptides span loop 4 ( Fig 7B and 7C ) . Shift changes were also observed for isolated amino acids; these were likely due to small differences in sample pH values . Furthermore , sequence alignment of HL and Hewl demonstrated that the residue ranges that interacted most closely with Nm-ACPI in the model with HL were almost identical ( Fig 7D ) . The identification of loop 4 within the β4-β8 region as the principal interaction site between rNm-ACPI and Hewl/HL therefore agrees well with the model proposed from the docking simulations . Our data demonstrated that purified Neisseria ACP proteins inhibited lysozyme activity using in vitro enzyme function assays . We next examined whether expression of ACP by live Neisseria spp . was necessary for bacterial survival in the presence of HL . Thus , we compared the growth of Neisseria acp knock-out strains ( meningococcus MC58ΔacpI and MC161ΔacpII and gonococcus FA1090Δacp ) with the corresponding wild type and complemented strains in the presence of HL . Δacp meningococci and Δacp gonococci showed significant ( P<0 . 05 ) ~3–5 fold and ~5–9 fold reductions in CFU after 8 h , respectively , whereas wild-type bacteria were relatively insensitive to the effects of HL ( Fig 8A; Table 3 ) . Complementation of ACP expression restored resistance to HL at levels similar to wild-type ( Fig 8A; Table 3 ) . However , Δacp strains were not completely killed . We next tested the possibility that the observed increased sensitivity to lysozyme of the Δacp meningococci and Δacp gonococci could have been due to decreased membrane integrity that allowed ingress of extracellular lysozyme . Initially , we quantified the Minimum Inhibitory Concentrations ( MIC ) of vancomycin and streptomycin antibiotics for wild-type , Δacp and Δacp-complemented meningococci and gonococci ( Table 4 ) . Differences in antibiotic susceptibility would be expected if deletion of ACP was causing increased OM permeability . However , there were no significant differences in the MIC values for vancomycin against wild-type , Δacp and Δacp-complemented MC58 and MC161 meningococci ( 64–96 μg/mL ) , or for wild-type , Δacp and Δacp-complemented FA1090 gonococci ( which were more sensitive than meningococci at 6–8 μg/mL ) . The MIC values for streptomycin were also similar for the meningococcal wild-type , Δacp and Δacp-complemented variants ( 16–24 μg/mL ) , whereas wild-type FA1090 gonococci and the corresponding Δacp and Δacp-complemented variants were relatively insensitive to this antibiotic ( MIC>1024 μg/mL ) . Examination of membrane permeability with the Baclight viability dyes demonstrated that there were no significant differences in the percentages of propidium iodide-labelling of wild-type , Δacp and Δacp-complemented variants of MC58 ( 11–13% , P>0 . 05 ) and MC161 ( 8–14% , P>0 . 05 ) meningococci or FA1090 gonococci ( 11–19% , P>0 . 05 ) ( Fig 9 ) . To our knowledge there are no published general protocols available for generating N . lactamica gene knock-out mutants . A method has been presented at a conference that reported successful transformation of a kanamycin resistance gene into N . lactamica , with the use of hypermethylated ( i . e . restriction resistant ) PCR products as donor material leading to a 1000-fold increase in transformation efficiency [29] . Nevertheless , as robust and reproducible mutagenesis methods are not generally available , we used an alternative strategy to investigate the role of ACP for survival of N . lactamica commensal organism in the presence of HL . An assay was developed in which purified anti-Nm-ACPI rabbit IgG was reacted with live wild-type N . lactamica strain Y92-1009 in order to block ACP function . In the presence of this specific antibody and HL , we observed a significant ~3–8 fold reduction ( P<0 . 05 ) in N . lactamica CFU over an 8h incubation . In general , the N . lactamica Y92-1009 wild type strain was relatively insensitive to HL , similar to wild-type meningococci and gonococci ( Fig 8B; Table 5 ) . Addition of pre-immunization rabbit IgG with and without HL or anti-Nm-ACPI rabbit IgG alone did not affect wild-type bacterial growth .
In the current study , our data suggest that Neisseria ACP , a surface-exposed OM molecule with a role in mediating meningococcal adhesion and capable of inducing a functional bactericidal antibody response [20] , is involved also in inhibiting C-type human lysozyme activity . To the best of our knowledge , this represents the first report of ACP functioning as a novel lysozyme inhibitor for both pathogenic and commensal Neisseria spp . Moreover , examination of ACP allelic variation amongst 12 , 483 Neisseria isolates ( http://pubmlst . org/neisseria/ , analysed on December 2016 ) shows that there is a high degree of conservation of this lysozyme inhibitor . There are 153 different acp alleles encoding 43 non-redundant ACP protein sequences ( S1 Table ) : Allele 2 is expressed by the largest number of isolates that are almost exclusively N . meningitidis strains ( 7966/8568 ( 93% ) of meningococcal isolates and 7966/12483 ( 64% ) of all Neisseria spp . isolates ) . The second most commonly expressed ACP protein by N . meningitidis is encoded by Allele 1 ( 570/8568 , representing 7% of meningococcal isolates ) , which is also the major ACP protein expressed by commensal N . lactamica ( 100/139 , representing 72% of all N . lactamica isolates ) . Allele 10 and Allele 6 are expressed predominantly by N . gonorrhoeae strains ( 3021/3737 ( 81% ) of gonococcal isolates and 3021/12483 ( 24% ) of total Neisseria spp . isolates for Allele 10; and 553/3737 ( 15% ) of gonococcal isolates and 553/12483 ( 4% ) of total Neisseria spp . isolates for Allele 6 ) ( S1 Table ) . Alignment of all 43 non-redundant Neisseria ACP protein sequences shows a high degree of amino acid sequence conservation within the mature proteins ( amino acids 22–124 ) ( S8 Fig ) . Notably , Neisseria ACP amino acid sequences encoded by Alleles 1 and 2 show ≥99% identity and there is only one single amino acid substitution between Type I ( Asp25 ) and Type II ( Asn25 ) proteins ( S8 Fig ) . Alleles 6 and 10 encoded gonococcal ACP proteins are 98% identical , with the same single amino acid substitution ( Asp25 to Asn25 ) and a deletion of Ala22 in Ng-ACP protein encoded by Allele 10 ( S8 Fig ) . Thus , the four major proteins encoded by acp Alleles 1 , 2 , 6 and 10 show ~94% amino acid sequence similarity and cover 98% of Neisseria isolates in the http://pubmlst . org/neisseria/ database . To our knowledge , the current study is the first to determine the structure of the Neisseria ACP protein by crystallography and suggests potential differences in the mechanism of lysozyme inhibition compared with known lysozyme inhibitor families found in other Gram-negative bacteria . These families have been classified according to substrate specificity , the presence of specific sequence motifs and structural topology , i . e . —Ivy from E . coli [10] , MliC/PliC from S . enteritidis , E . coli and P . aeruginosa [11 , 24] , PliI from Aeromonas hydrophila [30] , PliG from E . coli [31] and the recently discovered PliG-type Tsi3 from P . aeruginosa [12 , 32] . Bacterial lysozyme inhibitors have conserved motifs within the C-terminal region , which are located within protruding loops involved in enzyme inhibition . Previous structural studies for MliC/PliC family lysozyme inhibitors , e . g . P . aeruginosa MliC in complex with Hewl ( PDB 3F6Z ) [24] , E . coli MliC [33] and S . typhimurium PliC ( PDB 3OE3 ) [22] have shown that they all fold into eight-stranded anti-parallel β-barrels and share two conserved sequence motifs . The first conserved region ( SxSGAxY ) , present in the MliC/PliC , PliI and PliG family proteins , is mainly located on the fourth loop between β4-β5 strands: the second conserved region ( YxxxTKG ) is located on the sixth β-strand of proteins and is only found in the MliC/PliC family [1] . Structural data supported by mutational analysis has provided insight into the molecular mechanism by which these proteins interact with and inhibit lysozyme , and a ‘double key-lock’ mechanism has been suggested [24] . In our current study , the Nm-ACPI crystal structure , has demonstrated that Neisseria ACP proteins are structurally similar to the MliC/PliC family , but do not share primary sequence similarity nor any described sequence motifs with this family [11] , or with any other characterized bacterial lysozyme inhibitor reported to date [1] . Although a central role for the Neisseria ACP loop 4 is suggested for binding to and inhibition of lysozyme , this loop is three amino acid residues longer than the MliC/PliC loop 4 . A structural model of Nm-ACPI in complex with HL , obtained by docking simulation , showed loop 4 is predicted to insert into the HL active cleft , and the β4-β8 β-sheet on Nm-ACPI forms an extensive interaction surface with lysozyme , which could inhibit substrate access to the enzyme active site . NMR-specific chemical shift changes of Nm-ACPI in the presence of HL supported our proposed mode of interaction . These observations , together with the fact that Neisseria ACP proteins cannot be classified into any of the known and well-characterised lysozyme inhibitor families due to the lack of significant primary sequence homology and absence of conserved specific sequence motifs , support our hypothesis that this protein is a novel lysozyme inhibitor that binds to and inhibits lysozyme via different sequence motif ( s ) . More generally , our study supports the notion that lysozyme inhibitors in Gram negative bacteria are more diverse than previously thought . For example , recent studies have shown that there are mechanistic differences in lysozyme inhibition , even between members of the same family , e . g . the identification of a new binding interface in Brucella abortus PliC complexed with HL , which was disordered in P . aeruginosa MliC complexed with Hewl [23] . In addition , for lysozyme inhibitors belonging to the PliI or PliG family , only the first motif loop is conserved and surface-exposed and predicted to inhibit lysozyme by inserting into the active-site cleft [22] . To try to confirm our model of the complex , we attempted mutagenesis of a number of sites within ACP and characterized a panel of single mutants . However , there was no single site mutation that had a significant effect on ACP activity , although a triple mutant ( Asn79Ala/Tyr84Ala/Gly95Ala ) showed ~39% reduced activity compared with the wild-type ( S2 Table ) . Since mutagenesis of individual sites had little effect and the observation with a triple mutant is insufficient on its own to validate the model , we hypothesise that the energetics of binding are distributed across several different residues . Instead , the NMR data provide much stronger validation for the model . Our study has shown that ACP inhibits death by lysozyme of Micrococcus and Neisseria spp . In addition to its muramidase activity , it is known that lysozyme also displays antimicrobial peptide ( AMP ) activity . However , several lines of evidence support our conclusion that Neisseria ACP inhibits peptidoglycan degradation by lysozyme . First of all , we demonstrated conclusively by MST and NMR that Neisseria ACP bound to lysozyme . In particular , the NMR data are a good test of specificity , with observed chemical shift changes for a limited range of residues , which is characteristic of specific binding . By contrast , non-specific binding would have been manifested by chemical shift changes by many different surface residues , as the lysozyme would have multiple binding sites on ACP . Secondly , we confirmed that boiled lysozyme did not affect Micrococcus integrity [34] , indicating that lysozyme antimicrobial activity was not responsible for the observed Micrococcus cell death in our in vitro assays . Moreover , lysozymal AMP activity , in common with other AMPs , does not degrade peptidoglycan directly [35 , 36] . Thirdly , direct peptidoglycan degradation by lysozyme is reported with the in vitro assay using fluorescein-labelled cell walls of Micrococcus . In this assay , cell wall peptidoglycan-fluorescein is the lysozyme substrate and increase in fluorescence intensity is due to the enzymatic activity of lysozyme and not AMP activity . Thus , the observed reduction in fluorescence in the presence of Neisseria ACP was attributable to the capacity of ACP to inhibit peptidoglycan degradation by lysozyme . In addition , we demonstrated that deletion of acp gene was associated with increased sensitivity of Neisseria spp . to lysozyme in vivo , without affecting bacterial membrane integrity or permeability . Taken together , these lines of evidence support our conclusion that ACP inhibits lysozymal degradation of peptidoglycan . We have shown also that meningococcal ACP proteins are OM-located and surface-exposed [20] ( S3 Fig ) . By contrast , other Gram-negative bacteria lysozyme inhibitor proteins are present either in the periplasm or anchored to the luminal face of the OM . Neisseria ACP proteins do not possess a recognisable lipobox and are unlikely to be anchored to the OM in a lipidated form , unlike membrane-bound MliC family proteins such as P . aeruginosa-MliC and Mycobacterium tuberculosis lipoprotein LprI [13] . Interestingly , the face associated with the N-terminal region of Neisseria ACP and opposite to the predicted HL-binding site , is highly positively charged ( Fig 1 and Fig 7 ) , which could provide ACP with the ability to associate with negatively charged surfaces and possibly mediate attachment to the membrane or capsule surface . Electrostatic binding interactions have been reported for other bacterial proteins , e . g . E coli protein antibiotic colicin N [37] and P . aeruginosa OprH [38] both interact with lipo-oligosaccharide chains , and P . aeruginosa G-type lysozyme Tse3 interacts with the inner leaflet of the OM [32] . The ability of ACP proteins to inhibit lysozyme function clearly provides Neisseria spp . with the potential to avoid a key vertebrate innate immune defence mechanism found on all mucosal surfaces and secretions [2 , 3] and in the respiratory airway [4] . Our study shows that ACP proteins conferred increased lysozyme tolerance to both pathogenic and commensal Neisseria spp . and showed higher activity against human lysozyme compared to avian lysozyme . A hypothetical model can be proposed to describe how expression of ACP enables Neisseria spp . interaction with the host . Neisseria ACP protein ( s ) functions as an adhesin primarily for epithelial cells [20] and is likely to contribute to mechanisms of adherence of meningococci and commensal Neisseria spp . to nasopharyngeal mucosal epithelium and gonococci to urogenital mucosal epithelia . Evasion of bacteriolysis during host colonization by Neisseria spp . is likely to be effected by multiple and possibly non-redundant mechanisms . Thus , a combination of PG acetylation [14 , 15] , ACP expression and other recently described mechanisms such as expression of LtgA and LtgD lytic transglycosylases [34] , enables both pathogenic and commensal Neisseria species to limit human innate immune clearance by lysozyme and contribute to establishing colonization . C-type lysozyme is present also in blood [7] and for potentially invasive species such as meningococci , it is possible that expression of ACP may be important for survival . In summary , we have identified a new function for Neisseria ACP as an OM-located , surface-exposed lysozyme inhibitor , in addition to its role as an adhesin and as a potential vaccine candidate . ACP is found in all pathogenic and commensal Neisseria spp . , and despite structural similarity to the MliC/PliC family proteins , it shares no conserved motifs involved in lysozyme inhibition . Thus , ACP could represent a distinctive protein different to the known families of bacterial lysozyme inhibitors [1] . Our data suggest that the ability of ACP to inhibit lysozyme activity could be important for host colonization , not only by pathogenic meningococci and gonococci , but also by commensal organisms . Thus , ACP represents a dual target for developing Neisseria vaccines and drugs to inhibit host-pathogen interactions .
Neisseria meningitidis strains MC58 ( B: 15:P1 . 7 , 16b ) and MC161 ( C: 2–37 , P1 . 5–1 , 10–4 ) have been described previously [39 , 40] . Neisseria gonorrhoeae strain FA1090 was purchased from the American Type Culture Collection ( ATCC 700825 ) and Neisseria lactamica strain Y92-1009 ( sequence type 3493 , clonal complex [CC] 613 ) was produced by the Current Good Manufacturing Practices pharmaceutical manufacturing facilities at Public Health England ( Porton Down , United Kingdom ) . Wild type , mutant and complemented strains were grown on supplemented GC agar plates ( with the addition of selective antibiotics and/or isopropyl-β-1-D-thiogalactopyranoside ( IPTG; Sigma-Aldrich ) when necessary ) and incubated at 37°C with an atmosphere of 5% ( v/v ) CO2 [41] . Escherichia coli DH5α or XL-10 Gold ultra-competent E . coli cells ( Stratagene ) ( cloning ) and BL21 ( DE3 ) pLysS ( New England Biolabs ) ( protein expression ) strains ( Invitrogen , UK ) were grown on Luria-Bertani ( LB ) agar , LB , SOB or Terrific Broth with or without addition of selective antibiotics and IPTG when necessary . Whole suspensions of MC58 and MC161 were prepared in water and OM were prepared by extraction of whole bacteria with 0 . 2 M lithium acetate buffer , pH 5 . 8 and differential ultracentrifugation , as described previously [42 , 43] . Construction of MC58ΔacpI mutant strain has been described previously [20] and construction of MC161ΔacpII mutant strain was done using the same protocol . Briefly , primers KO2095F ( 5’-CGGGCTGAACCAGATAGACT-3’ ) and KO2095R ( 5’-GCTCCAGTTTGGTACGGAGA-3’ ) were used to amplify the 2 . 9 kb DNA segment from the genomic DNA of the acp- mutant 35/11 [44] . In this strain , the nm-acp ORF was interrupted by the insertion of a mini-transposon ( 1 . 6 kb ) [44] , while the amplified PCR product from the wild-type MC58 or MC161 strains was 1 . 3 kb . The 2 . 9 kb PCR product was gel purified , cleaned , and used for transformation of naturally competent MC161 strain [45] . Transformed bacteria were screened by PCR , and the selected MC161ΔacpII colonies were confirmed as Nm-ACP- by Western blot analysis using murine anti-Nm-ACP sera . To generate a Δacp isogenic deletion mutant in gonococcal strain FA1090 , the target gene was replaced by a kanamycin ( Kan ) resistance cassette ( following the method described by Echenique-Rivera et al . [46] , with modifications ) . Approximately 400-bp fragments of the flanking regions of the target gene were amplified by PCR from FA1090 genomic DNA . A 400-bp fragment upstream of ng-acp ( F1 ) was generated using a forward primer ( F1-Fw; 5’-TAGACTTCTGGGGCAAGGTC-3’ ) and a reverse primer ( F1-Rv; 5’-GGCTATTCTAGATTTTATTCCTTTGGATAGATG-3’ ) , carrying the restriction site for XbaI ( underlined ) . A 400-bp fragment downstream of the target gene ( F2 ) was amplified using a forward primer ( F2-Fw; 5’GGCTATTCTAGATCAGGCAACAAAAAACAGCG-3’ ) , also incorporating a restriction site for XbaI , and a reverse primer ( F2-Rv; 5’-GGTACGGAGATTGTCGCCC-3’ ) . The PCR products were purified using a Wizard SV Gel and PCR Clean-up System ( Promega ) , digested with XbaI restriction enzyme ( Promega ) at 37°C for 3 h , and purified from an agarose gel band . The purified digested PCR fragments were then ligated overnight at 4°C and used as a template for amplification by PCR using F1-Fw and F2-Rv primers . The ~800-bp PCR product was purified and cloned into the pGEM-T Easy vector ( Promega ) , which was then used to transform competent E . coli DH5α cells . Purification of the construct ( pGEM::F1+F2 ) was performed by mini-prep from a positive transformed colony , and the plasmid was then digested with XbaI in order to insert a Kan cassette in between fragments F1 and F2 . The kanamycin resistance cassette was previously amplified by PCR with primers KanFw ( 5’-GGTTCTAGATTCAGACGGCGTGATCTGATCCTTCAACTC-3’ ) and KanRv ( 5’- GGTTCTAGATTAGAAAAACTCATCGAGCATC-3’ ) , which incorporated an XbaI restriction site and a DNA uptake sequence ( in italics ) . The PCR product was purified , digested with XbaI and purified from an agarose gel band before ligation to linearized pGEM::F1+F2 ( o/n at 4°C ) . This ligation product ( pGEM::F1+Kan+F2 ) was used to transform E . coli DH5α cells and subsequently the naturally competent FA1090 strain . Mutagenesis was achieved by heterologous allelic exchange . Transformed colonies were screened by PCR , and the selected FA1090Δng-acp strain was confirmed by Western blotting using cross-reacting murine anti-Nm-ACP sera . Complementation of MC161Δnm-acpII and FA1090Δng-acp knock-out strains was performed as previously described for the complementation of MC58Δnm-acpI mutant strain [20] . Primers Com2095F ( 5’-GGCTATTTAATTAAATGAAACTTCTGACCACCGC-3’ ) and Com2095R ( 5’- TTAACGTGGGGAACAGTCTT-3’ ) were used to amplify nm-acpII and ng-acp genes from MC161 or FA1090 genomic DNA respectively , cloned into pGCC4 vector in between the PacI and PmeI restriction sites ( under transcriptional control of a lac promoter ) and confirmed by sequencing using an upstream primer , LacPFw ( 5’-CGGTTCTGGCAAATATTCTG-3’ ) . The resulting constructs pGCC4-nm-acpII and pGCC4-ng-acp were transformed into MC161Δnm-acpII or FA1090Δng-acp knock-out strains respectively using the method of Stohl and Seifert [45] , and the complementary strains were identified by PCR screening and Western blotting with murine anti-Nm-ACP sera . The nm-acpI and II gene sequences , optimized for E . coli expression and encoding the entire coding sequence for NMB2095 protein ( NEIS2075 , http://pubmlst . org/neisseria/ , 375 bp ) , were synthesized in vitro ( GeneArt , Invitrogen ) . Primers gACP-Fw ( 5’- GGCTATCATATGAAACTTCTGACCAC -3’ ) and gACP-Rv ( 5’- GGCTATCTCGAGACGTGGGGAACAG -3’ ) were used to amplify the complete NEIS2075 ORF from N . gonorrhoeae strain FA1090 genomic DNA . All acp genes were cloned into the pET22b ( + ) system ( Novagen ) in between the NdeI and XhoI restriction sites providing a C-terminal hexa-histidine tag . Recombinant plasmid ( pET-nm-acpI , pET-nm-acpII and pET-ng-acp ) was transformed into either E . coli XL-10 Gold ultra-competent or E . coli DH5α cells for plasmid amplification and subsequently into competent E . coli BL21 ( DE3 ) pLysS cells for protein expression . In order to produce ultra-high purity recombinant ( r ) Nm-ACPI for X-ray crystallography , E . coli BL21 ( DE3 ) pLysS carrying plasmid pET-nm-acpI was grown in Terrific Broth and induced at an optical density ( OD ) λ600nm of 0 . 8 by adding IPTG to a final concentration of 0 . 5 mM and left to grow for 12–14 h at 16 0C . To produce 15N- and 13C/15N-labelled rNm-ACPI for NMR studies , E . coli cells were grown in M9 minimal media ( 6 g/l Na2HPO4 , 3 g/l KH2PO4 pH 7 . 4 , 0 . 5 g/l NaCl , 1 g/l 15NH4Cl , 0 . 2 mM/l CaCl2 , 1 mM/l MgSO4 and 20 ml of 20% 13C Glucose ) supplemented with 200 mg/l 15N- or 15N 13C Isogro media ( Sigma-Aldrich ) . Stable isotope-labelled rNm-ACPI was subsequently purified following the same procedure as for the unlabelled protein ( see below ) . Purified protein was dialysed into 50 mM sodium phosphate buffer pH 6 . 0 for NMR spectroscopy . Bacteria were harvested by centrifugation at 10 , 000 x gav for 20 min at 4 0C and subsequently suspended in lysis buffer ( 50 mM HEPES/NaOH buffer , pH 7 . 5 , 200 mM NaCl , 5% v/v glycerol ) supplemented with 0 . 2 mg DNase ( Sigma ) and 1x EDTA-free protease inhibitor cocktail ( complete , Roche ) . Bacteria were broken using a probe sonicator set at 30% amplitude for 5 min with 5 s ON and 10 s OFF pulses to release soluble rACP proteins , which were recovered from the supernatant fraction after centrifugation at 27 , 000 x gav for 30 min at 4 0C . The supernatant was passed through a 0 . 45 μm filter then stored at 4 0C with 2 ml Ni-IDA ( nickel iminodiacetic acid ) resin ( Generon ) for 1 hr with gentle mixing . After storage , the mixture was passed through a gravity flow column to pack the rACP-bound resin . Unbound proteins were washed with 20 column volumes of wash buffer ( 50 mM HEPES/NaOH buffer , pH 7 . 5 , 200 mM NaCl , 40 mM imidazole , 5% v/v glycerol ) , while bound rNm-ACPI was eluted with increasing concentrations of imidazole up to 500 mM in wash buffer . Protein samples obtained were analysed using 12% ( w/v ) SDS-PAGE ( Invitrogen ) . rNm-ACPI was subsequently dialysed against 25 mM HEPES/NaOH buffer , pH 7 . 5 at 4 0C for 12–14 h and was purified further by ion exchange chromatography with a cation exchanger , Resource S column ( GE Healthcare ) using a linear gradient from dialysis buffer A to buffer B ( 25 mM HEPES/NaOH , pH 7 . 5 , 1 M NaCl ) . Eluted rNm-ACPI was concentrated with 5 kDa cut-off Vivaspin centrifugal concentrators ( Sartorius ) at 6 , 000 x gav then further purified with a size exclusion chromatography column , Superdex S-75 Hi-Load 16/600 column ( GE Healthcare ) on an AKTA-FPLC system ( GE Healthcare ) using 25 mM HEPES/NaOH buffer , pH 7 . 5 and 150 mM NaCl as running buffer . Eluted protein was analysed for purity by SDS-PAGE ( S1 Fig ) , and concentrated . Protein concentration was determined at 1600–1690 cm-1 wave number using a Direct Detect FTIR spectrometer ( EMD Millipore ) . Protein identification of purified rNm-ACPI for crystallography was confirmed by mass spectrometry of excised SDS-PAGE bands . Excised gel bands were dehydrated using acetonitrile and subsequently trypsin digested overnight at 37°C . LC-MS/MS was done using an UltiMate 3000 Rapid Separation Liquid Chromatography system ( Dionex ) coupled to a LTQ Velos Pro ( Thermo Fisher Scientific ) mass spectrometer . Peptide mixtures were separated using a gradient from 92% A ( 0 . 1% formic acid in water ) and 8% B ( 0 . 1% formic acid in acetonitrile ) to 33% B , in 44 min at 300 nL min-1 , using a 250 mm x 75 μm i . d . 1 . 7 mM BEH C18 , analytical column ( Waters ) . Peptides were selected for fragmentation automatically by data dependant analysis . Using Mascot ( Matrix Science ) , data produced were searched against the UNIPROT database with taxonomy of Neisseria spp . and were subsequently validated using Scaffold ( Proteome Software ) . rNm-ACPII and rNg-ACP proteins were purified also using Ni-IDA affinity chromatography under native conditions , but bound proteins were eluted using 50 mM NaH2PO4 , 300 mM NaCl , 250 mM imidazole buffer , pH 8 . 0 , and subsequently dialyzed against PBS , pH 7 . 4 for 48 h . Protein concentration was determined using the BCA Protein Assay Kit ( Pierce ) . The molecular mass ( Mr ) of mature rNm-ACPI , rNm-ACPII and rNg-ACP proteins , without their leader peptide sequence and each with a C-terminal hexa-histidine tag , was ~12 . 3 kDa , as shown by SDS-PAGE ( S1 Fig ) . All recombinant proteins were stored at -80 0C until needed . Crystallization was performed by the sitting drop method using MRC 96 Well 2 Drop Crystallization Plates ( Molecular Dimensions ) using a Mosquito crystallization robot ( TTP Laptech ) . Wild type rNm-ACPI crystals were grown at 20 0C by mixing 200 nl of protein solution ( 20 mg/ml in 25 mM HEPES/Na pH 7 . 5 , 37 mM NaCl ) with 200 nl precipitant solution containing 0 . 2 M lithium sulfate , 0 . 1 M sodium acetate pH 4 . 5 and 50% ( w/v ) PEG 400 ( JCSG plus , Molecular Dimensions ) , with a reservoir volume of 50 μl . Crystals were cryoprotected by washing for 5 min in well solution supplemented with 20% ( v/v ) glycerol , before flash cooling in liquid nitrogen . For phasing , crystals were incubated for 10 min in the well solution supplemented with freshly prepared 0 . 4 M KI , which was also included in the cryoprotectant . Diffraction data were collected at Diamond Light Source beamlines I04-1 and I03 . Datasets were processed by automated pipeline implemented in xia2 [47] , using XDS [48]; data collection statistics for native and KI-soaked crystals are summarised in Table 1 . Reflection data from native and KI-soaked crystals , were merged and scaled using Aimless [49] , as implemented in the CCP4 suite [50] . For phasing , automated heavy atom substructure identification was combined with experimental phasing and model building using Autosol [51] , as implemented in PHENIX [52] . Autosol identified and refined the locations of 8 iodine sites ( figure of merit 0 . 289 ) ; following density modification by RESOLVE , a readily interpretable electron density map was produced . An initial model was produced by Autobuild [53] , which built the majority of the residues . The structure was completed by manual model building in Coot [54] , and refined using REFMAC [55] and PDB_REDO [56] . The relevant phasing statistics and refinement parameters are detailed in Table 2 . Biomolecular interaction between rNm-ACPI and HL ( Sigma-Aldrich ) was studied using the Monolith NT . 115 microscale thermophoresis instrument , ( NanoTemper Technologies GmbH ) and a light-emitting diode ( LED ) filter of 605–645 nm excitation and 560–685 nm emission . The rNm-ACPI protein ( 2 . 5 mM ) was dialysed overnight at 4 0C into MST buffer ( 50 mM Tris pH 7 . 4 , 150 mM NaCl , 10 mM MgCl2 and 0 . 05% ( v/v ) Tween-20 ) . Recombinant ( r ) HL ( 20 μM ) was labelled using the Monolith NT Protein Labeling Kit ( NanoTemper Technologies ) according to the manufacturer’s protocol . In brief , 100 μl of 20 μM rHL was incubated at room temperature with 100 μl of 60 μM NT-647-NHS fluorescent dye ( NanoTemper Technologies ) reconstituted in DMSO . Labelled rHL protein ( 3 . 33 μM ) was recovered with 600 μl of MST buffer using a buffer exchange column . The binding affinity measurements between rNm-ACPI and rHL were carried out as described elsewhere [57] . Two fold serial dilutions up to 16 dilutions of 2 . 5 mM rNm-ACPI were prepared in MST buffer and mixed in a 1:1 ratio with 8 . 3 nM NT647-labelled rHL to yield a final volume of 20 μl per dilution . The reaction mixtures were loaded into premium-coated capillary tubes ( NanoTemper Technologies ) . The interaction was analysed by MST at 20% and 40% MST power and a light-emitting diode ( LED ) intensity of 20% . Interaction by thermophoresis was analysed after 20 s laser-on time and dissociation constants ( Kd ) were calculated from concentration dependent changes in normalised fluorescence ( Fnorm ) of NT647-HL after 5 s of thermophoresis using the NTAffinity Analysis software . Experiments were carried out in triplicate and arithmetic means and standard deviations of readings used . The structure of Nm-ACPI was superposed using CCP4MG [58] onto PliC from Brucella abortus in the complex with HL ( PDB accession code 4ML7 ) . This modelled complex of Nm-ACPI was used as the starting point for 10 , 000 independent docking simulations carried out using RosettaDock [28] . The complex with the lowest energy was selected . NMR spectra were recorded at 25°C on a Bruker DRX 800 spectrometer equipped with CryoProbes . Sequence-specific backbone resonance assignment of rNm-ACPI was obtained using multi-dimensional heteronuclear NMR experiments HNCACB and CBCA ( CO ) NH . 15N-rNm-ACPI in complex with Hewl was prepared by mixing 15N-rNm-ACPI with unlabelled Hewl followed by SEC purification . For comparison of the 15N-1H HSQC spectra obtained from the free rNm-ACPI and in complex with Hewl , the data were collected from the same batch of 15N-labelled rNm-ACPI , in order to minimise any differences in conditions between the free and complexed protein . Data were processed using the Bruker Software TopSpin and analysed using CCPN software [59] . BALB/c mice ( H-2d haplotype ) were bred within the animal facilities of the university under standard conditions of temperature and humidity with a 12 h lighting cycle and with food and water available ad libitum . Groups of five BALB/c mice of approximate equal sizes and weights ( 6–7 weeks of age ) were immunized intraperitoneally with purified mature rNm-ACP proteins ( Types I and II ) in saline solution , liposomes or Al ( OH ) 3 formulations , prepared using methods described previously [20] . Within each group , individual mice were immunized with 20 μg of recombinant protein on days 0 , 14 , and 28 . Groups of five mice were also sham immunized ( no protein ) and one group was kept for normal mouse serum ( NMS ) . Mice were terminally bled by cardiac puncture under anesthesia on day 42 . All sera were stored at −20°C until required and decomplemented by heating at 56°C in a water bath for 30 min before use . The quality and specificity of murine anti-rNm-ACPI and anti-rNm-ACPII sera were assessed by i ) Enzyme-Linked ImmunoSorbent assay ( ELISA ) reactivity against recombinant protein and MC58 and MC161 OM preparations and ii ) Western immunoblotting on wild type MC58 and MC161 OM preparations as described previously [60] ( S3 Fig ) . Binding of antibodies to bacterial surfaces was demonstrated by Fluorescence-Activated Cell Sorter ( FACS ) analysis , as described previously [20] ( S3 Fig ) . This study complied with the animal experimentation guidelines of the Home Office ( HO ) , with approval granted under the Animals ( Scientific Procedures Act , 1986 ) with a HO project licence number PPL 30/3126 . The study was approved by the Animal Welfare and Ethics Review Board ( AWERB ) at the authors’ institution ( University of Southampton , no number assigned ) . Animal health and welfare was assessed daily by qualified animal technicians and no animals suffered significant adverse effects . Polyclonal IgG antibodies were purified from rabbit antisera that were generated to rNm-ACPI in our previous study [20] , using the Pierce Fab Preparation Kit , following the manufacturer’s protocol with some variations . Briefly , rabbit pre- and post-Nm-ACPI immunisation sera were desalted with a Zeba Spin Desalting Column and incubated for 10 min in a Nab Protein A Plus Spin Column at room temperature . Purified IgG antibodies were recovered with Elution buffer pH 2 . 8 ( which contains primary amine ) and immediately neutralized with 1M Tris buffer , pH 8 . 5 . Elution samples were then concentrated and dialyzed against PBS , pH 7 . 4 with Amicon Ultra-4 centrifugal filters ( Merck Millipore ) . Protein concentration was finally determined by absorbance at λ280nm using an estimated extinction coefficient of 1 . 4 and finally analysed by non-reducing and non-boiled SDS-PAGE . Wild-type N . meningitidis MC58 , MC161 , N . gonorrhoeae FA1090 and N . lactamica , and acp knock-out and complemented strains , were grown overnight at 37°C with an atmosphere of 5% ( v/v ) CO2 [41] on supplemented GC agar plates with the addition of the corresponding antibiotic when necessary . Bacteria were suspended in 1 ml of supplemented GC broth , diluted 1/100 in fresh supplemented GC broth and incubated for a further 2–3 h . To the cultures was then added 250 μM IPTG , to allow expression of rNm-ACPI , rNm-ACPII or rNg-ACP proteins from a lac promoter on the chromosomally complemented strains . IPTG was also added to the corresponding wild type and knock-out strains , as well as to wild type N . lactamica , to ensure identical culture conditions . At an optical density ( ODλ600nm ) of ~ 0 . 6 , cells were diluted serially in supplemented GC broth to 1 x 104 CFU/ml , with and without human lactoferrin ( L1294 –Sigma-Aldrich ) at a final concentration of 3 mg/ml and/or 10 μg/ml of HL . It has been reported that lactoferrin is needed to permeabilize the OM of Gram-negative bacteria [64 , 65] , but in our study the observed effects of lysozyme on the Neisseria OM were similar in the presence or absence of lactoferrin [25] . For wild type N . lactamica , an assay was developed in which bacteria were incubated with and without HL ( 10 μg/ml ) in the presence or absence of purified anti-rNm-ACPI rabbit IgG antibodies ( 90 μg/ml ) . For both assays , culture medium samples were diluted serially in supplemented GC broth at time ( t ) = 0 h and at t = 8 h , plated onto GC agar plates and incubated at 37°C with an atmosphere of 5% ( v/v ) CO2 overnight . Fold changes in bacterial growth at 8 h were calculated using the following equation: N0/N , where N0 is the number of CFUs after 8 h incubation without HL , and N is the number of CFUs in the presence of HL at the same time point . PCR-based site directed mutagenesis was used to generate single , double and triple Nm-ACPI mutants in pET22b expression vector using the following primers: N79Afwd 5'-CGGTGAATCTGGATAAAAGCGATGCTGTGGAAACCTTCTATGG-3'; N79Arev 5'-CCATAGAAGGTTTCCACAGCATCGCTTTTATCCAGATTCACCG-3'; Y84Afwd 5’-GCGATAATGTGGAAACCTTCGCTGGTAAAGAAGGTGGTTACG-3’; Y84Arev 5’-CGTAACCACCTTCTTTACCAGCGAAGGTTTCCACATTATCGC-3’; G95Afwd 5'-CTTTTACCATCCATAACAGCGGTGCCTAAAACGTAAC-3' and G95Arev 5'-GTTACGTTTTAGGCACCGCTGTTATGGATGGTAAAAG-3' . For each mutant , two synthetic mutagenic oligonucleotide primers ( synthesized by Eurofins ) were used to substitute the respective amino acid residue with alanine . The forward and reverse mutagenic primers were 37 to 43 bp long and designed to have the substitution at the centre of the sequence . PCR amplification was carried out in a PCR-thermal cycler ( Prime ) , and subjected to the following thermal cycle reaction: 95 0C for 2 min ( 1 cycle ) ; 95 0C for 50 s , 62 0C for 50 s , 72 0C for 6 min ( 20 cycles ) ; 72 0C for 7 min ( 1 cycle ) . Plasmids generated were incubated with DpnI enzyme ( New England Biolabs ) at 37 0C for 2 hr and transformed into XL-10 gold ultra-competent E . coli cells for colony growth on LB-agar plates supplemented with 100 mg/ml ampicillin . The sequence of each mutant generated was confirmed by nucleotide sequencing ( GATC Biotech ) and subsequently expressed and purified as described for wild-type Nm-ACPI . Wild-type Neisseria meningitidis strains MC58 and MC161 and wild-type Neisseria gonorrhoeae strain FA1090 and the corresponding Δacp and complemented strains were grown overnight on supplemented GC agar medium and colonies suspended in pre-warmed supplemented GC broth at an ODλ600nm of 0 . 1 ( equivalent to ~0 . 5×108 CFU/mL ) . A 100μL aliquot of each bacterial suspension was diluted with an equal volume of medium and spread evenly onto the surface of fresh GC agar plates and allowed to dry . Etest strips with vancomycin and streptomycin ( Biomerieux ) were placed aseptically onto each agar plates and the plates incubated at 37°C with 5% ( v/v ) CO2 . The minimum inhibitory concentration ( MIC ) of each antibiotic for each organism was determined following the Etest reading guide from the manufacturer . Wild-type Neisseria meningitidis strains MC58 and MC161 and wild-type Neisseria gonorrhoeae strain FA1090 and the corresponding Δacp and complemented strains were grown overnight on supplemented GC agar medium and colonies suspended in 0 . 5mL volumes of pre-warmed supplemented GC broth . From these initial suspensions , 100μL aliquots were used to inoculate 5mL of GC broth and the cultures incubated at 37°C with shaking ( 200 rpm ) until mid-logarithmic growth phase was reached ( ODλ600nm >0 . 4 ) . Bacteria were centrifuged ( 5500 g , 3 min ) , the pellet washed once in sterile saline ( 0 . 9% w/v NaCl ) and each pellet suspended in a final volume of 1mL of saline , to which was added 3μL of a 1:1 mixture of the propidium iodide ( PI , red ) and Syto9 ( green ) dyes from the LIVE/DEAD Baclight Bacterial Viability Kit ( Molecular Probes ) . Mixtures were kept at room temperature for 15 min in the dark and then 5μL volumes applied in duplicate to microscope slides under cover slips and viewed with a fluorescence microscope ( Leica DMRB , Leitz ) . Approximately 5–6 independent fields of view were examined per sample , counting ~100 colonies per field . A minimum of n = 3 biological replicates were done per bacterium . The percentage of bacteria positive for red PI staining , indicating bacteria with permeable membranes , was calculated by dividing PI-positive bacteria by total ( red and Syto9 green ) bacteria . Data were compared using either a two-sample t-Test or a paired t-Test , with P values <0 . 05 considered significant . | The genus Neisseria contains two major human pathogens: N . meningitidis ( Nm ) causes meningitis and sepsis , and N . gonorrhoeae ( Ng ) causes the sexually transmitted disease gonorrhoea . In addition , the genus contains a larger number of commensal organisms , including N . lactamica ( Nl ) . Common to all of these organisms is the ability to colonize exposed mucosal epithelia . Recently , we identified a novel surface-exposed adhesin in Neisseria spp . , the Adhesin Complex Protein ( ACP ) , which was capable also of generating a functional bactericidal antibody response in mice . In the current study , we have determined the crystal structure of a recombinant ( r ) Nm-ACP and shown that it shares structural homology to bacterial lysozyme inhibitors . We demonstrate that Neisseria ACP functions as an inhibitor of mammalian lysozyme but the mechanism appears to be different from other bacterial family lysozyme inhibitors . Expression of ACP enables Neisseria spp . to tolerate human lysozyme . We propose that ACP-mediated inhibition of lysozyme activity could be important for host colonization by both pathogenic and commensal Neisseria organisms and that ACP represents not only a target for developing Neisseria vaccines but also drugs to inhibit host-pathogen interactions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"crystal",
"structure",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"condensed",
"matter",
"physics",
"microbiology",
"neisseria",
"gonorrhoeae",
"physiological",
"processes",
"sequence",
"motif",
"analysis",
"crystallography",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"tissue",
"repair",
"neisseria",
"solid",
"state",
"physics",
"sequence",
"alignment",
"neisseria",
"meningitidis",
"proteins",
"medical",
"microbiology",
"bioinformatics",
"microbial",
"pathogens",
"recombinant",
"proteins",
"physics",
"lysis",
"(medicine)",
"biochemistry",
"database",
"and",
"informatics",
"methods",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
]
| 2017 | Structure of the Neisseria Adhesin Complex Protein (ACP) and its role as a novel lysozyme inhibitor |
Land plants rely mainly on gravitropism and phototropism to control their posture and spatial orientation . In natural conditions , these two major tropisms act concurrently to create a photogravitropic equilibrium in the responsive organ . Recently , a parsimonious model was developed that accurately predicted the complete gravitropic and proprioceptive control over the movement of different organs in different species in response to gravitational stimuli . Here we show that the framework of this unifying graviproprioceptive model can be readily extended to include phototropism . The interaction between gravitropism and phototropism results in an alignment of the apical part of the organ toward a photogravitropic set-point angle . This angle is determined by a combination of the two directional stimuli , gravity and light , weighted by the ratio between the gravi- and photo-sensitivities of the plant organ . In the model , two dimensionless numbers , the graviproprioceptive number B and the photograviceptive number M , control the dynamics and the shapes of the movement . The extended model agrees well with two sets of detailed quantitative data on photogravitropic equilibrium in oat coleoptiles . It is demonstrated that the influence of light intensity I can be included in the model in a power-law-dependent relationship M ( I ) . The numbers B and M and the related photograviceptive number D are all quantitative genetic traits that can be measured in a straightforward manner , opening the way to the phenotyping of molecular and mechanical aspects of shoot tropism .
Plants are constantly moving to reach the light and to maintain the architecture and posture of their aerial organs: stems , branches , leaves… These movements are active , generally powered by differential growth and are controlled by environmental signals [1] . Tropisms are the class of movements that are oriented by a vectorial environmental factor . Light is the main source of energy for plants and is a major cue for tropic movement . In phototropism , shoots grow in the direction of the light source . Under natural conditions on Earth , gravity is unavoidable and drives gravitropism , in which shoots usually grow against the direction of gravity . Proprioception , the ability of plants to perceive their own deformations , has recently been identified as a major factor in tropic movement; it stimulates the straightening of a curved organ , a response that can be thought of as autotropism [2–4] . Such classification of control mechanisms into three different tropic drivers is however merely a conceptual convention: it is likely that , in natural conditions , the three processes interact constantly . Plant shoots generally exhibit negative gravitropism , positive phototropism , and negative autotropism in the same organs at the same time . The way phototropism and gravitropism interact to control the movement of plant shoots is acknowledged to be important in determining plant habit in natural conditions but is not well understood [5] . Advanced genetic studies in the model species Arabidopsis thaliana , which used mutants with impaired gravitropism or phototropism [6 , 7] , have shown that the two processes interact in a complex manner to control movement in wild-type plants [8] . However , part of this complexity lies in the variety of phototropic responses observed in response to different qualities and intensities of the light stimulus . For example , the effects of low-fluence pulses differ from those of continuous light [7]; but only the latter are directly relevant to plant growth under natural conditions , and will therefore be considered here . The focus of the current study is the regulation and the control of organ movements during the interaction of tropisms . To do so , it is proposed to extend the recent dynamical model for gravitropism called the AC model [2] to phototropism . In this model , active tropic bending is controlled by the additive ( but opposing ) effects of graviception and proprioception , expressed by ∂C ( s , t ) ∂ t = - β A ( s , t ) - γ C ( s , t ) for s > L - L g z and 0 otherwise ( 1 ) where s is the position along the organ , Lgz is the length of the growth zone , L is the length of the entire organ , t is time , A ( s , t ) is the local angle of the organ to the vertical , C ( s , t ) is the local curvature ( i . e . , the spatial rate of change of A along s ) and the parameters β and γ are , respectively , the gravi- and proprio-ceptive sensitivities . The AC model was shown to explain the complex kinematics of gravitropism in eleven species covering a broad taxonomical range of angiosperms , major growth habits and organ types . [3 , 4] Although graviception and proprioception act additively , their control over the dynamics of tropic movement and the steady-state final shape actually depends only on the ratio between gravisentitivity and propriosentitivity , scaled to the size of the growth zone . In the model , this is formalized through the definition of the dimensionless graviproprioceptive bending number B = β Lgz/γ . B fully defines both the time to reach the steady state and the final shape at steady state , and can be measured in simple morphometric experiments . As B is dimensionless , it can be used to make quantitative comparisons between experiments involving very different sizes and growth velocities . This enables universal behaviors and control mechanisms to be identified . To understand and describe properly the interaction between gravitropism and phototropism , successive steps are undertaken: First , the hypotheses that lead the construction of the model are discussed , then general specifications of the geometry of the organ and of the gravity and light fields are established . In the following part , the construction of model is given . Starting with the simple case of phototropism in isolation which allows to dissect the interaction between photosensitivity and propriosensitivity . Finally , the dynamics of the interaction among photoception , graviception and proprioception are explored , including the influence of light intensity . All these models are investigated for different localization of photoception , at the tip or distributed along the organ . Quantitative experimental data on photogravitropic equilibrium are used to test the validity of the model . These results and the implications for plant biology are discussed .
In order to formulate hypotheses regarding the extension of the AC model to include phototropism , it is now useful to review our knowledge of the distribution of sensing mechanisms and differential growth responses along the plant organ , the mechanisms of photo- gravi- and proprio-ception , as well as their possible interactions . The localization of gravisensitivity has been established in more detail than the localization of photosensitivity . The plant perception of gravity is related to the presence of statoliths within specialized cells called statocysts . In aerial organs , statocysts and statoliths are found throughout the growth zone , and both perception and the bending response are local [9] . Our knowledge of the mechanisms of proprioception is limited , but it is known to be exclusively local and likely to involve cytoskeleton remodeling [1 , 2] . In phototropism , blue-light-sensitive phototropins sense the direction of incoming light , but they interact with other photoreceptors in a network that remains to be fully elucidated [10–13] . Photoreceptor localization studies were pioneered by Darwin [14] . By masking different parts of coleoptiles , Darwin found that the perception of light occurred at the coleoptile apex . As tropic bending occurred all along the growth zone of the organ , a secondary basipetal signal was postulated to be involved , now elucidated as a lateral redistribution of the polar transport of the plant hormone auxin [10] . Very few studies have further systematically analyzed the localization of photoreceptors or even phototropic responses along plant organs . It has been found , however , that apical phototropic sensing is not universal . For example , the hypocotyl of Arabidopsis exhibits photoreceptors all along the growth zone , allowing for distributed local photoception and local phototropic growth responses [11 , 15] . The functional significance of such differences in the localization of photosensitivity among organs has not been investigated . A few quantitative studies of the effects of continuous light [16–18] have shown that phototropism and gravitropism seem to act in an additive manner . When a plant organ is lit from a direction that differs from that of gravity , the plant undergoes an active movement until a steady-state shape called the photogravitropic equilibrium is reached [17 , 19] . The photogravitropic equilibrium angle ( PGEA ) of the organ tip was found to follow the phenomenological equation P G E A = k log I I 0 - g sin A 0 ( 2 ) where I is the fluence rate of illumination , I0 is the light-sensing threshold , g is the gravitational force , and A0 is the initial angle of inclination towards gravity . The term gsinA0 is the sine law of gravitropism [4] . This sine law was also used when defining the gravisensing term in the AC model for gravitropism1 but in this case it has been reduced to a linear term using the approximation sin ( A ( s , t ) ) ≈ A ( s , t ) + O ( A3 ) . The term k log ( I/I0 ) reflects the phototropic stimuli . Both phototropic and gravitropic sensing have been shown to control the relocalization of polar auxin transporters , controlling the formation of lateral gradients in auxin concentration and hence differential growth . These molecular dynamics might explain why the effects of the two stimuli on PGEA are additive [7 , 20 , 21] . Thus far , the dynamics of the processes leading to this photogravitropic equilibrium have not been analyzed or modeled in detail . Building on this current knowledge , we propose herein to extend the graviproprioceptive AC modeling approach to study phototropism and its interactions with gravitropism . We focus on investigating dynamic control of photosensitivity , gravisensitivity and propriosensitivity throughout the whole movement . More precisely , our working hypotheses are as follows: ( H1 ) The action of the tropic motor is fully driven by the perception-regulation process and results in a change in the local curvature . ( H2 ) The angles formed between the axis of the growing organ and the respective gravity and light fields influence graviception and photoception , and hence influence gravity- and light components driving the tropic response . ( H3 ) Proprioception takes place regardless of whether external tropic signals are present . The occurrence of proprioception has been shown for gravitropism [2 , 22] . Concerning phototropism , plants growing in a clinostat ( an apparatus that suppresses graviception ) were also shown to straighten after a transient light pulse [23] . Thus , we assume that each constituent element of the organ perceives its own local deformation during active bending , namely the curvature , and responds in order to restore local straightness [2] . ( H4 ) The different types of perception act additively within the model , meaning that their respective contributions have equivalent roles in driving the movement . It is therefore expected that much of the apparent complexity of the motion is due to spatio-temporal integration of several responses over the changing geometry of the organ with respect to the light and gravity fields . In order to formulate testable and refutable predictions regarding the dynamic control of tropic movement , we combine these working hypotheses into successive variants of a dynamic model , extending the graviproprioceptive AC modeling approach . These models are expected to provide relevant dimensionless numbers that control the dynamics , as well as a means of estimating their values on the basis of experimental measurements . This will enable the model and hypotheses to be evaluated experimentally , and eventually open the way to obtaining accurate phenotypes of tropisms . We have chosen to neglect the effects related to the elongation of the organ . Recent work has shown that elongation tends to destabilize tropic movements and should increase the oscillations that take place during these movements [24] . We have found , however , that the proprioceptive sensitivities of modern angiosperms have evolved such that these effects are fully controlled by the plants . Therefore , these growth-related effects can be neglected in the description of tropic movement [24] . Similarly , even though plants are expected to bend under their own weight , we propose that the effects of a plant’s weight on its shape can also be neglected [25] . This is based on the fact that the AC model , which takes into account the plant’s perception as a sole driving process , has proved to be sufficient to explain the movement of a wide variety of organs and species of many orders of magnitude of size . To finish , it is important to note that the AC model accounts for orthogravitropism , wherein the organ aligns with the direction of gravity . Some organs align in a direction different from that of gravity , called the gravitropic set-point angle ( GSA ) [26 , 27] . It is possible to adapt the AC model for such cases by modifying the model’s graviceptive term , −β A ( s , t ) →−β ( A ( s , t ) −GSA ) such that the organ aligns with the GSA . However , the AC model has not been tested for non-orthogravitropic organs . And this study concentrates on the case of ortho-gravi and photo-tropisms ( a very common case in primary shoots ) . As stated before , the present study involve successive variants of the model . To unease their handling we named the successive models presented below according to i ) the sensory processes that are involved and their locations ( biological specification ) —e . g . , photoproprioceptive; and ii ) the driving variables under investigation ( mathematical specification ) ( in accordance with the naming convention in [2] , e . g . model AC is driven par the inclination angle A versus the vertical and the curvature C all along the axis , whereas model AaC is driven by the apical angle of inclination and by the curvature all along the axis ) . The reader may find a list of name of the variables and parameters in Table 1 and of the model in Table 2 . One drawback of most studies on phototropism is that they have overlooked the true geometry of the organ , measuring only the orientation of the apical tip of the organ [28–31] . One of the greatest insight of the AC model was to show that the geometry of the organ and of the source field is central to describe the movement properly . It is then important to have a clear description of the studied geometry . Like the AC model , the models developed herein describe the shape of the organ in terms of its median line , i . e . , its central axis ( Fig . 1 ) . We parameterize the position along this median line according to the curvilinear abscissa s going from the base s = 0 to the apex s = L . The angle A ( s , t ) then describes the local orientation of the median with respect to the vertical pointing up ( and hence to the direction of the gravity vector g ) at time t . The angle of the apex at the tip of the organ , Aa , is then equal to A ( L , t ) . The local curvature C ( s , t ) is the spatial rate of change of A ( s , t ) along s , and from differential geometry we know that C ( s , t ) = ∂ A ( s , t ) ∂ s or A ( s , t ) = A 0 + ∫ 0 s C ( l , t ) d l ( 3 ) At the scale of the plant , gravity acceleration g is homogeneous , uniform , and invariant to translation and to rotation around the main direction of the gravity field ( Fig . 1 ) . The case of light perception seems to be less straightforward . The geometry of light fields in nature can be quite diverse . At the scale of the plant , the light field produced by the sun on the earth is supposed to be homogeneous , uniform , and invariant to translation and to rotation around the main direction of the light field , but its orientation relative to the gravity field varies over the course of the day . In the lab , a point light source can emit a spherical field whose intensity ( irradiance ) decreases with the square of the distance . Furthermore , multiple point sources might be used , thereby increasing the complexity of the analysis . Thus far , most controlled experiments have used a distant punctual source with a collimated beam [16–18 , 32] . For simplicity , we assume that the light field l has the same symmetry as the gravity field . This assumption holds true if the source of light is far from the organ . To further reduce the complexity of the analysis , we do not consider curvature outside the plane; that is , we explore the case in which the main direction of the plant’s organ , the gravity field g and the light field l are all in the same plane Pgl ( Fig . 1 ) . This again was the case in most controlled experimental studies ( e . g . [17] ) . Therefore , the direction of the lighting in the Pgl plane is defined by the angle Ap and by a radiance ( or light intensity ) I0 ( Fig . 1 ) . The simplifications we adopt enable us to assess the model by comparing it with experimental data , while still providing insights regarding the regulation of the movement . Moreover , the assumption that the initial direction of the organ , the direction of gravity and that of the light are in the same plane is fulfilled for organs that are already aligned with the direction of gravity . Coleoptiles and other plant shoots appear to sense light direction via the light gradient across the organ [33] . Treatments that change the steepness of this gradient ( e . g . , infiltration with dyes or changes in the amounts of natural pigments ) alter the phototropic response: the steeper the gradient , the more the coleoptile bends towards the light . This gradient depends on internal light transfer through plant tissues but also depends linearly on the irradiance of the light impinging on one side of the surface of an organ . This irradiance is given by Lambert’s cosine law of irradiance of geometrical optics ( e . g . [17] ) , which states that the irradiance falling on any surface varies in proportion to the cosine of the incident angle . I ( s , A P ) = I 0 cos π 2 - ( A ( s , t ) - A P ) = - I 0 sin A ( s , t ) - A P ( 4 ) As the AC model is a first-order model , and in the limit of small angles A ( s , t ) −AP , we can use the approximation sin ( A ( s , t ) −AP ) ≈ A ( s , t ) −AP + O ( A3 ) , so that the photosensitivity is proportional to the angle between the organ and the light direction i . e . , A ( s , t ) −AP ( the influence of the intensity of the light irradiance will be considered in a later section ) . Although this approximation is only valid for small angles , such an approximation gives a good and efficient understanding of the dynamics of the system . The influence of each term on the tropic movement can be easily understood and discussed . We may now start with the simplest case of phototropism without gravitropism . Using equation 4 , a phototropic model can be constructed . Two sub-models are considered , when the perception is apical and when the perception is local . The previous model can now be easily extended to take into account the perceptions of both light and gravity . As described in Fig . 1 . A , the direction of the light stimulus is now considered to form an angle AP with the vertical , the direction of the gravity . The orientation angle A ( s , t ) along the organ is still measured with respect to the vertical , and hence to the direction of gravity . If AP ≠ 0 the symmetry of the system is broken . When only one of the two tropisms influences the system , rotation around the axis defined by the direction of the corresponding field should have no effect on the system . When the two tropisms act simultaneously , however , the angle between the two fields prevents such global symmetry . It should also be noted that if the gravitropic term or the phototropic term tends to 0 , the system converges to the photo-proprioceptive models outlined above . As for the phototropic models , effects due to local and apical perception are considered and discussed . A series of experiments were conducted or reprocessed from the bibliography in order to assess the Photo-Gravitropic models against experimental data in continuous light and 1g conditions ( i . e . reminiscent of natural outdoor conditions on Earth ) : an experiment in the typical set-up used for phototropic studies , but tracking the entire kinematics of the movement ( LLE experiment ) and a set of experiments conducted by [17] and focusing only on the PhotoGravitropic Equilibrium tip Angle ( PGEA ) , but at various tilting angles and light intensities ( referred to as “Galland’s experiments” in the following ) .
A major prediction of the of AR-based models is that even when photoception and graviception act in different directions ( i . e . , AP ≠ 0 ) , they seem to act together to bring the organ towards a single direction defined by the angle AR . Additionally the ARC model predicts that the whole coleoptile curves in the direction AR which is first achieved at the tip and the curvature then concentrates near the base . These two predictions are independent of the position at which light is perceived , i . e . , at the tip of the organ or along the entire organ , and is thus common to all the variants of the ARC model , ( i . e . the local photoception model ARC and apical photoception model A R a C and their versions including explicit light intensity effects AR ( I ) C and A R a ( I ) C ) . For the sake of simplicity , whenever this common core is addressed , the notation ARC will be used to refer to all of them . When illuminated by a source of light from a direction that differed from that of gravity ( LLE experiment ) , the coleoptile curved and finally reached a steady state . The apical part aligned with an angle that was an intermediary between the direction of gravity ( A = 0 ) and the direction of the light ( A = π/2 ) ( Fig . 3 . A and B ) . This shows qualitative agreement with the ARC model . Additionnally the transient kinematic pattern in which i ) the whole coleoptile curves until the direction AR is first achieved at the tip and ii ) the curvature then concentrates near the base . is qualitatively observed in ( Fig . 3 . B and C ) . However , this kinematic experiment also revealed some limits of grass coleoptile as a “model system” for tropic studies . Indeed , when a coleoptile is illuminated , photomorphogenetic effects take place , slowing down the expansion of the coleoptile and increasing that of the inner leaf . After approximately 10h , the leaf inside the coleoptile pierces the coleoptile . The dynamics of the coleoptile then ceases , and there is no further elongation of coleoptile . The dynamics of the movement is then fully determined by the leaf that was inside and this dynamics differs from that of the coleoptile . Moreover the influence of the inner leaf may explain why the rather steady tip angle ( between 3 and 6 h in the example Fig . 3 . A and B ) is subsequently varied again ( between 7 and 10h ) , and it is thus difficult to determine whether the coleoptile has had time to reach a steady state before been disrupted . However , the apical tip is predicted in the model to converge to the direction defined by AR before the shape reaches a steady state , and we saw that this is consistent with what is observed experimentally . We may thus assume that in Galland [17] estimate of AR could be measured even when the full steady state of the rest of the coleoptile could not be well defined . We therefore tested further the validity of the functional form of AR-based models through the analysis of the apical tip angle at PhotoGravitropic Equilibrium ( PGEA ) in the work by Galland [17] . If the ARC core-model is correct , then the experimental PGEA should be equal to AR , and according to equation 14 , it should depend on the angle between the direction of gravity and direction of the light AP , and on the ratio between the gravisensitivity and the photosensitivity levels . Indeed AR can be expressed in the model as a function the photograviceptive number M: A R = A P 1 1 + M ( 25 ) It follows directly that when photoception dominates , the organ bends in the direction of the light , M < < 1 , AR = AP . However when graviception dominates , M > > 1 , AR = 0 ( Fig . 4 ) . It is then possible to express M directly as a function of AR/AP . M = A P A R - 1 ( 26 ) A direct quantitative assessment of the functionnal form of the AR ( I ) C and A R a ( I ) C models—and of the competing M ∼ Φ ( I ) sub-models—vs . experimental data can then be conducted through the analysis of the experimental relation between the steady-state tip angle PGEA and the fluence rate of the incident light I . Indeed AP is known , and we have seen that we may assume AR = PGEA . Then using equation 27 an estimate of M can be obtained . And combining equation 27 with equation 23 , a prediction of the model is then that a plot of M = AP/AR − 1 as a function of I should display either a power-law dependancy , equation 24 , or a log-dependancy , equation 25 . Moreover , these relations should fit to a single curve M ∼ Φ ( I ) , independent of the initial tilting angle A0 . Note however that due to the linearization of the sine terms of the angle dependency of gravi- and photoception ( equations 1 and 4 ) this should be only valid for inclination angles A0 < 90° . Fig . 2 shows the compiled data from the first experimental protocol ( PROT1 ) [17] , reprocessed for comparison with the predictions of equation 27 . The results of fitting equations 24 and 25 to the experimental data are also shown in Fig . 2 A and B . And Table 4 provides the values of the coefficients a and b of the curves fitting the power law ( equation 24 ) for each initial angle A0 . In the log-log plot in Fig . 2 . A , the results of each individual experiment at a given A0 fit well to a straight line M ∼ I−b with values of b between 0 . 36 and 0 . 44 ( Table 4 ) . Except in cases in which A0>90° , the coleoptile response is independent of the initial angle A0 , as predicted by the ARC model ( in Fig . 2 . A ) , and these data collapse on a single master curve . On the semi-log plot , however , no convincing fit can be identified . The power law relationship therefore seems to better describe the results than the logarithmic relationship does . Under the power law dependency of M on light intensity I , the coefficient of determination was R2 ∼ 0 . 91 , such that our model captured 91% of the total changes in the steady-state tip angle attributable to changes in the initial angle A0 and fluence rate I . The response for higher angles does not follow the first order relation expressed in the AC model anymore , but the sine laws . It is then expected that , for inclination angles A0 > 90° , the experimental data are adrift from this fit and vary non-monotonously , as A varies in a way that can no longer be accounted for by the ARC model . This is indeed observed in Fig . 2 . The second protocol ( PROT2 ) provides a different way to measure the same parameter than PROT1 , even if the protocol is different . The plot of AR/AP = A0/AP as a function of the experimental values of I should then collapse on the single master curve defined with PROT1 in Fig . 2 . C . Again , the model predictions were fair at small angles , AR/AP ≤ 0 . 4 , but for larger angles the experimental data diverged non-monotonously from the model . It is interesting to note that the results fit well to both the logarithmic law and the power law [17] . Finally a last testable prediction of the ARC model is that the plot of AP/AR as a function of M , and hence of the light fluence rate I , should be S-shaped with two asymptotes: i ) when the light intensity is low , gravity dominates and AR = 0 and ii ) when the light intensity is high . Moreover the equilibrium AR is shifted toward the orientation of incident light such that AR = AP ( Fig . 4 ) . These predictions are also consistent with the experimental data shown in Fig . 2 . D . Now that experimental agreement has been found with features of the photogravitropic models , it is interesting to discuss more in details the properties of the models and the insights it provides on the photo- and gravi-tropic control . This will be conducted through considering variant models of increasing complexity . In order to reach insights on the dynamic of the interactions between photo - , gravi- ( and proprio- ) -ceptions on the control of the tropic movement , we will consider the case of a tropic motion at constant incident light intensity I , therefore studying the ARC model ( the influence of I through changes in M , the ratio between photoception and graviception , having been clarifyied previously ) .
This study shows that a minimal modeling approach used to study shoot gravitropism , which gave rise to the AC model [2] , can be readily extended to study the interactions among photoception , graviception and proprioception . The extended model , the ARC model , is minimal but complete . Simple dimensionless control parameters can be estimated from the ARC model , and the model is tractable and easy to understand . Furthermore , it is possible and straightforward to include the dependency of photosensitivity on light intensity I , thereby producing an augmented ARC model . Predictions from the latter model were assessed against detailed quantitative experimental results [17] . The model agreed very well with the data for small initial angles . All the experimental curves at various inclination angles and light intensities collapse to a single curve when using the ARC model with a power-law dependency between the photograviceptive control number M and the intensity of light I . Although the assessment is not comprehensive , the evidence presented supports the validity of the model . The hypotheses that the action of the tropic motor is fully driven by perception-regulation processes ( H1 ) , that the angles between the gravity and the light fields are first-order variables influencing the tropic dynamical movement ( H2 ) , that the two types of perceptions act additively ( H4 ) , and that the balance between gravisensing and photosensing depends on the perception of light intensity ( H5 ) are upheld in the model . The simplified versions and limiting cases of the ARC model served as useful tools for analyzing the dynamical interactions associated with perception during tropic movement , yielding five major insights . i ) Despite having different set-points , gravitropism and phototropism act together to align the organ with the direction defined by AR ( equation 26 ) . AR is determined by the dimensionless number M and hence by the ratio between gravisensitivity and photosensitivity . As in the AC model , the tropic movement is controlled globally through the different types of tropic perception , which together drive the local curving velocity . The final steady-state shape reflects the ratio of the respective sensitivities to light and gravity [2] . This may seem to be at odds with the phenomenological model for PGEA proposed by Galland , summarized in equation 2 [17] , which states that the global steady-state shape at photo-gravitropic equilibrium depends additively on the phototropic and gravitropic stimuli , not on their ratio . However this phenomenological model for PGEA becomes problematic when considering the limiting cases . According to equation 2 , in darkness ( i . e . , when I = 0 ) the PGEA = −g sin ( A0 ) . In other words , the gravitropic equilibrium orientation is a function of the initial angle . However , when there is no light , the set-point becomes simply the gravitropic set-point angle ( GSA ) [28] , and the equilibrium shape is driven by the GSA and by the graviproprioceptive number B [2] . For ortho-gravitropic organs ( i . e . , in which the GSA is vertical ) in plants in which graviception dominates proprioception ( where the value of B is large ) , experimental observations confirm that the equilibrium tip angle is indeed vertical , that is , PGEA = 0 for any initial angle A0 [2 , 22] . The final gravitropic steady-state shape does not depend on sin ( A0 ) at all . Rather , the sine law dependency applies to the transient rate of changes in curvature , not to the final steady-state shape [1 , 2 , 4 , 34] . In the other limiting case , when photoception dominates gravisensing ( e . g . , when the light is very bright , in microgravity or in agravitropic mutants ) , the set-point angle is determined by the direction of the light alone . The steady-state tip angle should therefore tend to AP . The expression of the AR in equation 26 accommodates this condition , whereas the phenomenological model for PGEA in equation 2 predicts a tip angle that would depend on the light intensity with no saturation . The phenomenological model for PGEA [17] is thus not consistent for these two limiting cases . The expression of the AR as the ratio between the sensitivities overcomes these problems directly and is also consistent with a complete dynamical model , the ARC model . AR is thus a better indicator than the previously defined PGEA [17] . In order to to unify the notation , we propose to follow [26] , and to refer to this angle AR as the PhotoGravi Set Point Angle ( PGSA ) . ii ) The relation between the intensity of light ( I ) and the tropic response is better described by a power law , ν ∼ I−b , than by a logarithmic law [17] . The power-law relationship was found to be more robust , fitting all the different experiments over more orders of magnitude of light intensity . For small angles , it was possible to collapse the different experimental results at various inclination angles and various light intensities into a single master curve . As the power law , the log law only relies on two parameters , however the log law also requires the assumption of a threshold in the perception of light . The power law is then more parsimonious . Further studies should be performed to investigate the photochemical basis of such an invariant scaling of the photo-gravisensitivity balance with light intensity , and to confirm the values of the exponent of the power law , estimated here as a ∼ 0 . 4 . iii ) The respective effects of local distributed perception along the organ and of apical perception are clearly defined for the first time . The AaC model , which deals with purely apical perception , predicts that the shape formed by the responding organ is an arc of a circle . It is reasoned that all the cells located along the organ receive the same secondary signal from the apical sensory apparatus , so every part of the organ curves to the same extent in order to bring the apical part into the direction of light . This prediction could be assessed experimentally on aerial organs but would require agravisensitive experimental conditions in which the A R a ( I ) C model can be reduced to the AaC model . Experimental conditions that approximate the absence of gravisensitivity may be achieved using microgravity , clinostat experiments [32 , 41] , or mutant genotypes with impaired gravisensing [30 , 41] . The relevant control parameter is the dimensionless photoceptive number D , which measures the ratio between photoception and proprioception . D can be easily estimated by measuring the apical angle of the steady state and the basal angle A0 , two variables routinely measured in most experiments on tropism . iv ) Local perception ( gravitropism ) and apical perception ( in species with apical phototropic sensing ) interact so that aerial organs can forage for light without losing mechanical stability . Indeed , according to the A R a ( I ) C model , the plant strives to align with the direction set by the ratio between photoception and graviception , but the curvature concentrates near the base while the apical part straightens . The length of the curved zone is under graviproprioceptive control only . This could be a functional adaption: Although the shoot must grow toward the light , the posture of the stem also needs to be controlled to ensure long-term stability . This may also explain how the apical part of some plants such as sunflowers can track the sun daily , while the plants continue to maintain control over their posture . v ) Information from the AR ( I ) C model can be used to design methods for high-throughput phenotyping of the complete tropic control of organs by measuring the dimensionless parameters B , D and M of individual plants . B can be determined by measuring a plant’s entire shape in simple gravitropic experiments in the darkness [2] . D can technically be measured in non-gravisensing conditions . However , it is simpler and faster to directly estimate M from the PGSA . Indeed , M is the ratio between gravisensitivity and photosensitivity , so the experiment can be conducted in gravisensing conditions . The apical angle at the steady state is the only measurement needed in order to estimate M . Once B and M are known , the value of D can be readily calculated , and the entire set of parameters controlling the dynamics of the system is then quantified . The dimensionless numbers B , D and M are real quantitative genetic traits that can be used to phenotype tropic mutants in genetically amenable plant models [30 , 41] and subsequently identify the genetic and molecular mechanisms controlling tropic movements in plants . The main limitation of the AR ( I ) C model in its present form is that it only deals with a geometrically simple light field , i . e . , with uniform light and with the gravity field g , where the plant and the light field l lie within the same plane . It is possible , for example , to create spherical light fields by setting up point sources of light in proximity to the plant . The model presented here can be extended to include this case by making the light sensitivity parameter ν dependent on the spatial position ν ( x , t ) . It is likely that the geometry of light fields in nature is even more diverse . For example , the direction of the sun’s light changes over the course of the day . The direction of the light may also be outside the plane defined by the plants and the gravitational field . The modeling framework developed herein could be extended to deal with more complex light-field geometries , but this would require further mathematical and programming development . Such extensions would enable diverse movements to be simulated and would provide a more complete understanding of the ecological function of tropisms [5] beyond the core example of tropic control presented in this study . It should also be noted that the use of etiolated organs ( e . g . , coleoptiles and hypocotyls ) can complicate the analysis of phototropism . These organs are subject to photomorphogenetic effects that modify their behavior , independently of phototropic movements . Hypocotyl geometry is modified by the opening of the hook [42 , 43]; in addition , it has been shown that pretreatment with light affects hypocotyl behavior [44] . Fully developed organs , which are less susceptible to photomorphogenetic effects , are therefore a powerful tool in the study of phototropism . Thus , kinematic analysis of the inflorescence of Arabidopsis thaliana might yield powerful insights regarding the phototropism process [19] . Finally and more generally , our modeling approach may also be used to study the tropic responses of other organisms , such as fungal stripes , sporangiophores or hyphae [18 , 45] , or of other plant organs such as roots . Plant roots might provide an interesting system in which to test the effects of apical perception and to assess the validity of the AaC model . As graviception in roots is purely apical [41] , the AaC model can be readily extended to root gravitropism with the caveat that the extent of proprioception in roots remains unknown . This model could then be assessed by measuring the steady-state curvature ( predicted to be constant along the organ ) in hydroponics experiments to avoid confounding effects of interactions with substrates . If the model is validated , it may be possible to confirm a unified theory of the tropic movements of fixed organisms in natural conditions . | Although plants are mostly seen as static , they are constantly moving to adapt to changes in their stature and to their environment . Gravity and light , among others , are major factors that sculpt the shapes of plants . Plants tend to grow in the direction of the light to get access to their energy resource . At the same time , however , they need to maintain their balance and control their posture . In a recent study , we showed that postural control is regulated by two types of perception: graviception and proprioception . We extend that study to include light perception in order to obtain a unified description of plant tropism . As the system is highly dynamic , a model is required in order to evaluate hypotheses against experimental data . Our results show that the direction of plant growth is determined by the combined influence of light and gravity , whereas postural control is still regulated by gravity and proprioception . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | A Unified Model of Shoot Tropism in Plants: Photo-, Gravi- and Propio-ception |
Recently , a number of Global Health Initiatives ( GHI ) have been created to address single disease issues in low-income countries , such as poliomyelitis , trachoma , neonatal tetanus , etc . . Empirical evidence on the effects of such GHIs on local health systems remains scarce . This paper explores positive and negative effects of the Integrated Neglected Tropical Disease ( NTD ) Control Initiative , consisting in mass preventive chemotherapy for five targeted NTDs , on Mali's health system where it was first implemented in 2007 . Campaign processes and interactions with the health system were assessed through participant observation in two rural districts ( 8 health centres each ) . Information was complemented by interviews with key informants , website search and literature review . Preliminary results were validated during feedback sessions with Malian authorities from national , regional and district levels . We present positive and negative effects of the NTD campaign on the health system using the WHO framework of analysis based on six interrelated elements: health service delivery , health workforce , health information system , drug procurement system , financing and governance . At point of delivery , campaign-related workload severely interfered with routine care delivery which was cut down or totally interrupted during the campaign , as nurses were absent from their health centre for campaign-related activities . Only 2 of the 16 health centres , characterized by a qualified , stable and motivated workforce , were able to keep routine services running and to use the campaign as an opportunity for quality improvement . Increased workload was compensated by allowances , which significantly improved staff income , but also contributed to divert attention away from core routine activities . While the campaign increased the availability of NTD drugs at country level , parallel systems for drug supply and evaluation requested extra efforts burdening local health systems . The campaign budget barely financed institutional strengthening . Finally , though the initiative rested at least partially on national structures , pressures to absorb donated drugs and reach short-term coverage results contributed to distract energies away from other priorities , including overall health systems strengthening . Our study indicates that positive synergies between disease specific interventions and nontargeted health services are more likely to occur in robust health services and systems . Disease-specific interventions implemented as parallel activities in fragile health services may further weaken their responsiveness to community needs , especially when several GHIs operate simultaneously . Health system strengthening will not result from the sum of selective global interventions but requires a comprehensive approach .
Since 2000 , global health initiatives ( GHIs ) have become a dominant international aid strategy , drawing on effective methods to control specific diseases , and accounting for a substantial increase of resources for global health [1] . Very soon , however , concerns emerged that , beyond anticipated benefits for targeted diseases , GHIs might erode health systems' capacity to respond to general health needs [2]–[7] . Early criticisms of GHI included the distortion of national policies as well as the creation of parallel bodies and processes burdening the health system [8] . Conversely , GHIs realized rapidly that their intervention capacity was limited by countries' weak health systems [6] . While it is now acknowledged that GHIs and health systems are influencing each other [8] , [9] , health systems and GHIs advocates still tend to have divergent views , partly framed in the long-standing horizontal-vertical discussion [10] . A WHO collaborative group was assigned in 2008 to assess the interactions between health systems and GHIs [9] , and its findings were discussed during a policy dialogue meeting in Venice in June 2009 [11] . The report draws attention to the paucity of evidence to help understanding the interactions between GHIs and health systems . So far , most studies have dealt with global interventions in the field of HIV/AIDS control [8] . These research results may however not be applicable to other GHIs , as distinct objectives , policies , structures and operational processes of GHIs are likely to produce distinct effects on health systems [9] . Another limitation is that most studies focus on national level , while empirical evidence at point-of-delivery level remains particularly narrow [8] , [9] . In recent years , growing awareness of NTDs , coupled with the availability of relatively low-cost control strategies , have led to important new global initiatives , including the WHO Neglected Tropical Disease Program , the Schistosomiasis Control Initiative ( SCI ) , the Global Network for Neglected Tropical Diseases ( GNNTD ) , the Neglected Tropical Disease Initiative ( NTDI ) , and others [12]–[14] . The emphasis of the current global NTD control strategy is on mass drug administration . Based on geographic overlap and co-endemicity of NTDs , it addresses simultaneously up to five NTDs ( lymphatic filariasis , onchocerciasis , schistosomiasis , soil-transmitted helminthiasis , and trachoma ) with a package of 4 drugs ( ivermectin or diethylcarbamazine , praziquantel , albendazole or mebendazole , and azithromycin ) . Mass chemotherapy conducted over successive years is expected to eliminate or reduce NTDs to a prevalence rate at which they no longer pose a threat to public health . The case for efficiency of the intervention is based on the “integration” of 5 diseases , but also on the fact that drugs are donated by pharmaceutical companies or available as generics , and distributed by community volunteers [15] . Mali , which already had experience with distinct campaigns for trachoma and schistosomiasis , was the first country to implement this integrated NTD control program in 2007 with USAID financial support . Since the knowledge on possible side effects of drug combinations was deemed insufficient , drugs were distributed serially for precautionary reasons , in a total period of 7 weeks between April and June 2007 . Each single drug distribution was followed by a 2 weeks break ( week 1: azitromicine , week 4: albendazole and ivermectine , week 7: praziquantel ) . For temporary financial constraints , the campaign was launched only in 3 regions , the plan for the rest of the country being postponed to end 2007 . The campaign occurred simultaneously in the 3 regions and concerned 24 districts; in some of these districts , this coincided with a Vitamin A campaign . The present study analyzes interactions between the NTD Control Program and Mali's health system , with a special focus on the district and service delivery level . Health districts in Mali are based on a network of health centers , each covering a defined health area; a typical health centre is staffed by a qualified nurse and 2 to 4 auxiliaries . A district executive team runs the referral hospital and provides technical support to the health centers . The aim of this exploratory study was to assess the program implementation processes on the field , and to identify plausible positive and negative effects for the health care system .
To gain more insight into the interactions between the NTD campaign and the local health system in Mali , we carried out an exploratory qualitative study , which is a common approach to study situations with little prior knowledge and few definite hypotheses [16] . We chose to observe NTD campaign interactions with health services in reasonably well functioning settings , in order to avoid that our findings could be attributed to major failures of the local health care system . Using ‘purposeful’ sampling [16] , [17] , we identified two rural districts from two different Regions that were typical of “good” rural districts in terms of output indicators ( utilization rate for curative services and coverage rates for preventive activities ) . Our approach combined three standard qualitative data collection methods , i . e . participant observation , in depth interviews with key informants and document analysis [16] . Interviews are valuable to gain information on feelings , thoughts and opinions , but less useful to describe events , behaviour or settings , as responses tend to be distorted by personal biases , lack of awareness , recall errors or selective accounts [16] . Participant observation is more appropriate to understand context issues , events and processes , but also has limitations including atypical behaviour of those being observed and selective perceptions of the observer [16] . Official documents provide a range of helpful information , including on planned processes and their rationale , but are selective and do not necessarily reflect actual processes . By using different methods we intended to compensate for limitations of each of them and to cross- check and triangulate our findings [16] , [17] . Participant observation was conducted during two weeks in May – June 2007 by a researcher ( AC ) with a public health background , who was familiar with health systems in various sub-Saharan countries . In each of the two districts , she accompanied the district medical officer and/or health centre staff in their follow up of campaign activities at health centre and community level . This allowed for observations of mass drug administration in 16 health areas ( 8 areas per district ) ; the selection of these areas was opportunistically determined by the district's agenda . Observation focused primarily on contextual issues , on procedures ( e . g . task allocation , place of distribution , contents of information , dosage ) and on behaviour of staff , drug distributors and community members . During observations , the researcher also used situational conversations , i . e . asking on-the-spot questions and discussing with district authorities and health centre staff in a naturalistic and informal way [18] , [19] . Besides routine drug administration , participant observation including situational conversations also entailed a “training of trainers” session for district health teams at Regional level , two district executive team meetings , and a community meeting . Structured interviews were not conducted at local level due to heavy time constraints for staff during the campaign . But in depth interviews were conducted with key informants , including ten Ministry of Health officers and ten representatives of support agencies acquainted with the Malian health system . Key informants were identified through snowball sampling [16]–[17] . These interviews sought to elicit information on campaign processes which could not be directly observed - such as decision making at national level , planning and financing - and to explore informants' views on interactions between the NTD control program and local health services . Further information was collected by consulting Malian official documents , website search and literature review . We sought to triangulate the data as much as possible and to check the same information at different sources . Observation and interview data were recorded in field note transcripts that were reviewed independently by two researchers . The study being exploratory , categories for analysis were largely inductive , though some were based on interactions of other GHIs with health systems reported in the literature [2]–[7] . They included implementation processes ( training , drug procurement and distribution , monitoring ) , task distribution , positive and negative effects on service delivery , and decision making processes . Throughout these issues we looked for emerging recurrent patterns and variations among situations and/or informants . Results were validated and complementary information was generated during three feedback sessions held in November 2007 with health authorities at peripheral , regional and national level . Most data are based on field observation which had no potential harmful effects on patients or other vulnerable persons . Data collection was complemented by in depth interviews with twenty key informants . We asked about their opinions which were mostly public . Nevertheless we asked for oral consent after explaining the purpose and methods of our study , and ensured confidentiality , as some of the key informants who were high officials of the MOH or international organizations preferred not to be identified . This consent was witnessed by at least one person other than the principal investigator . No written consent of key informants was asked , as this would have been unusual in the context and might have biased the interview process , otherwise quite informal . Another ethical concern for this health systems research was governance: the national authorities of the Ministry of Health in Mali gave permission , regional and district authorities were supportive of this research , and provisional results were shared and discussed with authorities before broader dissemination . Potential ethical issues were examined with the head of the IRB of ITM ( Antwerp ) and it was decided that there was no need for a full review . Neither was it required under the Malian legislation ( see law n°2009/63/4L ) , which rules biomedical research but not health systems research , and was not applicable at the time of the study .
Access to mass chemotherapy for targeted NTDs clearly improved according to all interviewees . Some informants however regretted that the control program only included drug distribution and some health education on NTDs , and did not address other NTD disease control strategies , such as curative care ( e . g . eye surgery for trachoma ) or sanitation . Several informants also criticized the high priority given to targeted diseases , while more common health problems received little attention; they worried about the campaign mobilising energy and diverting staff's attention from routine care delivery . These interview results were in line with our direct field observations: routine care was cut down or even totally interrupted in most health centres observed , as a consequence of health centre nurses being absent from their station and not being replaced by other staff . During the first round of the NTD control program , head nurses were required to devote 10 full working days for program-related training and supervision , in addition to monitoring and drug supply activities ( Table 2 ) . The mass distribution schedule also required health centre staff to postpone or reorganise planned routine outreach immunisation sessions . Some informants thought that , in a context of low service utilisation , the campaign was at least a way to bring services closer to underserved populations . Observation however did not suggest that the campaign had positive effects on nontargeted services . We observed missed opportunities for curative care: children queuing for NTD prophylactic drugs and presenting obvious other illnesses and need for care ( e . g . abscesses or trauma ) were not identified as such by the attending staff . Not all health centers responded similarly to these interferences: 2 of the 16 health centers managed to keep their curative consultations and immunization services running normally and to use the campaign to support the overall development of their health centre . For instance , one nurse passed his NTD training on to other team members , another took advantage of NTD training of community volunteers to discuss other health issues than the targeted NTDs , and supervision of the campaign at village level became an opportunity for health education on other topics . These 2 health centres differed from the others in terms of human resources: both were well staffed and had no vacant positions . They were managed by a qualified nurse , who was on the job for over 5 years , and reputed for dynamism , professionalism and leadership , both at regional and central level . Moreover , utilisation rates in these centres were above the national average ( >0 . 20 new cases/inhabitant/year ) , preventive coverage rates were considered good ( >75% ) , and they benefited from a supportive community organisation . Disparities between health centres were also seen with respect to the health centre capacity to implement the NTD control program: operational problems were observed in all health centres , except for the two more robust ones . These problems included errors in the population census , poor community mobilisation , drug dosing errors and omission of side effect monitoring . While these problems were not captured by the monitoring system of the NTD campaign , which indicators were limited to treatment and geographical coverage , they nevertheless suggest quality problems in campaign implementation processes . Finally , several interviewees highlighted possible negative effects of free distribution of drugs on care seeking: as sick patients have to pay for drugs in routine circumstances: patients might , so they feared , wait for a next edition of the campaign rather than seek care . While mass drug distribution itself relied on community volunteers , the NTD campaign nevertheless implied increased workload for both district and health centre staff to ensure drug supply , campaign monitoring , training and supervision . Informants thought that the training component was a positive effect of the program . A training cascade was organized , starting with a “training of trainers” where national program coordinators trained district authorities . The cascade continued with district authorities training health centre nurses , who trained community volunteers . The training consisted in transmission of information on epidemiology , diagnosis and treatment of each of the targeted diseases; several participants said it was a repetition of previous training sessions . Training and supervision activities entitled staff to receive allowances , which represented approximately an 80% increase of a district medical officer monthly salary ( increase of 122 000 F CFA for an average salary of 150 000 F CFA ) , and a 45% increase of a health centre nurse salary ( increase of 46 000 F CFA for an average salary of 100 000 F CFA ) . Some informants considered these incentives as contributing to the motivation and retention of health staff , but others reported that these allowances distracted staff's attention from their core activities , for which no extra allowances were provided . Allowances were also given to community volunteers . Most informants considered this as necessary to attract volunteers . We witnessed a local dispute around the selection of volunteers , suggesting that becoming a volunteer was much sought after . Still , informants worried about the sustainability of allowances for volunteers ( which were supposed to be paid by the communities themselves after the first round of the NTD control program ) , as well as about the inconsistency of allowance amounts between various donors , generating growing demands from the side of volunteers . Preparatory to mass drug distribution , census data were collected which can be used for purposes other than NTD control . Improved information on NTD treatment coverage was also made available . The NTD campaign however introduced a parallel monitoring and evaluation system . At district level , 12 new forms were introduced for drug supply management , 15 new forms for monitoring and supervision of campaign activities , and 1 new form for clinical complications of filariasis ( Table 3 ) . For each drug distributed , one report per village , one per health centre and one per district was required . Following donor instructions , a specific timetable for reporting was installed: village data were processed daily at health centre level , and transmitted to district level . Districts reported campaign results weekly to the regional level . NTD campaign drugs were mostly donated by pharmaceutical companies , which increased their availability for preventive chemotherapy nationwide . But a parallel drug procurement system was established to ensure rapid distribution of drugs from national to regional and district level . As storage space at national and regional level was insufficient , trucks were especially rented for the occasion . Drug management forms and processes , distinct from the national ones , were created specifically for campaign implementation ( see Table 3 ) . Besides , some informants stressed imbalances in drug availability: while Azitromicine was distributed at no cost during the campaign , patients suffering from trachoma had no access to this drug in routine conditions , as it was substituted by payable tetracycline ointment . One local informant reported complaints from the community on this issue . According to official documents [20] , the provisional budget for mass drug administration – cost of drugs not included - was approximately US $ 12 million spread over 5 years , most of which was financed by USAID , with complements from other agencies . This budget covered drug distribution ( 29% of the budget ) , training of staff and volunteers ( 26% ) , supervision and evaluation ( 17% ) , drug procurement ( 9% ) , health education ( 7% ) , institutional strengthening ( 10% ) and intersectoral collaboration ( 2% ) . Not included in the budget were State supported staff salaries and infrastructure used for the campaign . Institutional strengthening consisted mostly in intervention related equipment and staff , and several informants considered that the budget left no room for institutional strengthening beyond campaign related needs . Investments in general equipment were limited to motorbikes for districts; the increase of storage capacity at national level was not approved , nor was the acquisition ( rather than rental ) of trucks for drug procurement , which several informants regretted . Though acknowledging the relevance of the intervention , they expressed concerns about longer-term financing for sustainable campaign results . Though the decision to implement the program was taken by Malian authorities , several informants considered that national authorities had little space for negotiation , as most decisions were taken at supranational level by donors and their grantees . Indeed USAID financing was not directly allocated to the country , but to sub-grantees which , in the case of Mali were the International Trachoma Initiative ( ITI ) , replaced by Helen Keller International at the end of 2007 . A steering group with coordination and decision-making power was set up in parallel to the existing coordination structures in the Malian Ministry of Health . It included MoH officials and ITI staff . The strategic NTD control plan for 2007–2011 was designed at National Health Directorate level . Several informants considered the design of a single NTD control plan as a positive effect of the program , as it stimulated coordination between previously standalone program coordinators . However , the strategic plan had to adapt to donor and grantees requirements . Predefined strategies , earmarked funding and budgets kept tight for the sake of demonstrating efficiency left only limited margins of maneuver . Several informants were also critical about distorting effects of the program on national priorities . They recognized the importance of NTDs and Mali's longstanding experience with mass treatment targeting distinct NTDs: onchocerchiasis since 1988 , trachoma since 2004 , schistosomiasis since 2005 , and attempts to integrate mass treatment for onchocerchiasis , filariasis and helminthiasis started in 2005 . Donor websites emphasize that the program met ongoing country efforts and contributed to scale them up at national level . But informants also reported that these ongoing efforts partly resulted from previous external financing opportunities as well . They considered that the accumulation of donor conditions was hampering resource allocation following national strategic orientations . Finally , the top-down implementation process of the NTD campaign was felt by most informants to contradict local district leadership , central to Malian health policy ( PRODESS II ) , and to interfere with planned activities . Indeed , as information concerning the campaign reached regional and district authorities with short notice , district authorities had to modify or adjourn their calendar of health centre supervisions with no space for negotiation .
This study is the first to address interactions between the integrated NTD control program and a country health system . It provides insights on positive and negative effects of the integrated NTD control program at point-of-delivery and on district systems . A previous study documented community resistance to free drug distribution for schistosomiasis and soil transmitted helminths in Uganda , but did not address health system effects [21] . A few limitations of this study should not remain unnoticed . Data collection was inevitably influenced by the presence of the researcher and relations between her and people in the field [17] . It is however plausible that investigator effects led authorities and staff to show the best of their performance and refrain from critical comments: biases are more likely to minimise rather than maximise problems . Furthermore this qualitative study was based on a limited number of interviews and contextualised observation units , and not intended to be generalisable as in quantitative research , though readers may assess their applicability to their own settings [17] Purposeful sampling [16]–[17] does not provide guarantees that the districts and health centres observed are typical of the country . As we selected “better performing” districts in terms of output indicators , we assume that campaign related problems are not more severe in these districts than elsewhere in the country , but this needs to be probed . Another reason for caution in transferring our findings to other settings or countries is that the Malian campaign was the very first NTD control program integrating five diseases and may have suffered from startup problems , avoided in further editions . As most USAID-supported NTD campaigns are based on similar principles and processes as in Mali , our study however provides plausible hypotheses to be tested in other contexts . The purpose of our study was exploratory , which implies that more research is needed , both qualitative and quantitative . We suggest that our findings are helpful in framing further research questions . A large part of current understanding of interactions between GHI and health systems is based on HIV/AIDS related programs . Though the NTD control program appears as a “small” GHI compared to others such as Global Fund or PEPFAR [1] , its analysis brings new insights to the ongoing debate . The NTD control program , like other preventive programs , focuses on protection rather responsiveness to patient demand [22] . This distinguishes it from HIV/AIDS related GHIs scaling up access to ARV treatment for individual patients . We found that the integrated NTD program did not include curative care for NTDs , but also that extra workload and staff absences for program purposes disrupted access to general curative care at health centre level . A Bill and Melinda Gates Foundation study [23] reported extra workload for district staff in Angola and Tanzania resulting from donor requirements , but did not assess effects on service delivery . Evidence on effects of workload generated by GHI on access to health centre care is so far extremely limited [24] . A key finding of our study is the emerging differentiation between health centres within the same district in their capacity to cope with program's interferences . Only the most resilient services , characterized by a qualified , stable and motivated workforce , managed to maintain routine activities , and even to use the program as an opportunity for overall quality improvement . This finding is consistent with the notion that positive effects of GHIs are more likely to occur when the health system is robust [9] , [25] . But it expands it from country health system to district and health service level , emphasizing the importance of human resources for robustness of services . Future research should further explore the relation between human resources characteristics and absorption capacity of programs at health service level . Indeed , this might bear consequences for the adaptation of programs to specific local health systems and services . Other findings of our research are coherent with known effects of other GHIs on health systems and show a mixture of positive and negative effects . Like other programs , uptake in targeted services was scaled up , but duplications occurred , especially for drug procurement and monitoring and evaluation [8]–[9]; these parallel systems , meant to improve campaign efficiency , increased workload and total costs for the health system . Also like other GHIs , the NTD control program influenced priority setting [7]: pressure to absorb donated drugs and reach short term chemotherapy coverage results contributed to the distraction of energies away from other NTD control strategies such as treatment and sanitation , and more generally from overall health systems strengthening . From a health systems perspective , the question is however not so much whether the balance of a specific GHI is positive or negative . The problem for Mali , as for other countries , is the cumulative effect of a large number of GHIs and other campaigns , each with implications at all levels . Besides NTD campaigns , Malian health services also run National Immunisation Days , Vitamin A and bed-net distribution as well as eradication or elimination campaigns of polio , tetanus or yellow fever . An estimation of time spent outside the health centre by head nurses of a Malian rural district in 2006 showed absences reaching 54% of working days; half of this time was dedicated to campaigns and trainings linked to vertical programs [24] . As the head nurse is usually the only staff qualified to provide first line curative care in Mali , disruptions in consultation schedules erode service responsiveness and community's confidence in their local health centre [26] . Another cumulative effect is the growing mobilisation of communities to meet top down defined targets , to the expense of an empowerment approach to community participation [21] , [27] . The need for health system strengthening is increasingly acknowledged , also by promoters of integrated NTD control [28] . Most GHIs claim to contribute to health system strengthening with additional resources and capacity strengthening , but these interventions are mostly selective , targeting those system functions essential for implementation of their own program [22] . This was also the case for the Malian NTD control program . The prospect of adding vitamin A , bed nets and vaccines to the present campaign model [28] might contribute to the improvement of the protective function of health systems , but not to their responsiveness to population's demand for curative care , which may be even further undermined [22] . The control of NTDs in vulnerable communities is a necessity . But so is health systems strengthening , in order to respond adequately to other health problems and to ensure sustainable achievements , including of NTD control . A major challenge is how to engage in disease control – NTD and other diseases - without negatively impacting on existing health systems . Increased knowledge on interactions with the health system is needed to allow GHIs to plan for positive effects and alleviate potential negative effects . Presently , short term and quick win interventions are given priority , but more long term strategies are also needed . Health system strengthening should rely on country-specific development plans aligned with national policy , and requires a comprehensive approach across diseases and health problems and coordination among GHIs . For example , program specific in-service training should be organised in ways mitigating potential interruptions of service provision , but investments are also needed in pre-service education for qualified staff . The accumulation of program specific extra allowances , making targeted interventions more popular than routine activities [29] , could gradually be replaced by comprehensive human resource management at national and district level . Parallel drug supply should be limited to exceptional emergencies , and investments redirected to reinforce national drug supply systems . There are signs of GHIs learning from experience and gradually modifying some of their processes [8] . They also show increasing willingness to reduce fragmentation and to review processes [11] . Still the chaotic architecture for development assistance for health remains a major obstacle for health system strengthening . Progress towards effective and inclusive health systems will not result from the sum of selective GHI interventions . | Prevention of neglected tropical diseases was recently significantly scaled up in sub-Saharan Africa , protecting entire populations with mass distribution of drugs: five different diseases are now addressed simultaneously with a package of four drugs . Some argue however that , similarly to other major control programs dealing with specific diseases , this NTD campaign fails to strengthen health systems and might even negatively affect regular care provision . In 2007 , we conducted an exploratory field study in Mali , observing how the program was implemented in two rural areas and how it affected the health system . At the local level , we found that the campaign effects of care delivery differed across health services . In robust and well staffed health centres , the personnel successfully facilitated mass drug distribution while running routine consultations , and overall service functioning benefitted from programme resources . In more fragile health centres however , additional program workload severely disturbed access to regular care , and we observed operational problems affecting the quality of mass drug distribution . Strong health services appeared to be profitable to the NTD control program as well as to general care . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"public",
"health",
"and",
"epidemiology/preventive",
"medicine",
"public",
"health",
"and",
"epidemiology/global",
"health"
]
| 2010 | Interactions between Global Health Initiatives and Country Health Systems: The Case of a Neglected Tropical Diseases Control Program in Mali |
Trypanosoma brucei , the agents of African trypanosomiasis , undergo density-dependent differentiation in the mammalian bloodstream to prepare for transmission by tsetse flies . This involves the generation of cell-cycle arrested , quiescent , stumpy forms from proliferative slender forms . The signalling pathway responsible for the quorum sensing response has been catalogued using a genome-wide selective screen , providing a compendium of signalling protein kinases phosphatases , RNA binding proteins and hypothetical proteins . However , the ordering of these components is unknown . To piece together these components to provide a description of how stumpy formation arises we have used an extragenic suppression approach . This exploited a combinatorial gene knockout and overexpression strategy to assess whether the loss of developmental competence in null mutants of pathway components could be compensated by ectopic expression of other components . We have created null mutants for three genes in the stumpy induction factor signalling pathway ( RBP7 , YAK , MEKK1 ) and evaluated complementation by expression of RBP7 , NEK17 , PP1-6 , or inducible gene silencing of the proposed differentiation inhibitor TbTOR4 . This indicated that the signalling pathway is non-linear . Phosphoproteomic analysis focused on one pathway component , a putative MEKK , identified molecules with altered expression and phosphorylation profiles in MEKK1 null mutants , including another component in the pathway , NEK17 . Our data provide a first molecular dissection of multiple components in a signal transduction cascade in trypanosomes .
Cells respond to their external environment in order to regulate their proliferation , developmental fate , specialisation or death . This can be in response to environmental cues such as temperature , pH or light , or can be driven by chemical signals generated by other cells of the same species or from competing or co-operating cells occupying the same niche [1] . To respond to such signals , single-celled and multicellular organisms have evolved elaborate signalling pathways in which surface receptors often transduce a signal to protein kinases and protein phosphatases , these transduction cascades eventually driving changes in gene expression either in their nucleus , or by generating phenotypic responses through changes in the abundance or activity of mRNAs or proteins [2–4] . The organisation of these signalling cascades is relatively well conserved in overall structure in eukaryotic organisms , although the individual receptors and transducer kinases and phosphatases are different . In particular , the ability of cells to become quiescent through exiting the cell cycle is a central feature of eukaryotic life , enabling cells to withstand periods of nutrient restriction , or to prepare for cell differentiation [5] . Parasitic protozoa also respond to extracellular signals to regulate their proliferation , tropism and development to ensure their successful transmission to new hosts [6–8] . Among major pathogens such as malaria , many signalling proteins have been identified and individual components have been assigned functions in different cellular processes [9 , 10] . However , in most cases the interactions between different components of the signalling pathways are unknown and the connections between regulators of particular processes are not understood , preventing assembly of a coherent regulatory pathway despite high throughput mutant selection and analysis [11] . The ability to assemble these pathways can also be limited by the evolutionary divergence of protozoan pathogens such that the conventional framework established in the crown group eukaryotes may not apply [12 , 13] . In consequence , the environmentally regulated control of virulence , transmission competence , tissue tropism or metabolic adaptation is poorly understood in eukaryotic microbial pathogens . Environmental sensing has particular importance in the life cycle of Trypanosoma brucei in the bloodstream of their mammalian hosts . These parasites are responsible for African trypanosomiasis in humans and animals [14 , 15] and live extracellularly in the bloodstream , adipose tissue [16] and skin [17] of hosts where they evade immune destruction by a sophisticated antigenic variation process [18] . This exchange of antigen types during chronic infections contributes to the waves of parasitaemia that characterise infection with these parasites [19 , 20] . However , a further contributor is environmentally regulated , that being the development of the parasite in the mammalian bloodstream in preparation for its transmission by tsetse flies [21] . Specifically , as trypanosomes proliferate they generate a soluble factor , stumpy induction factor ( SIF ) , that accumulates as parasite numbers increase [22 , 23] . At a given density , the SIF signal stimulates the bloodstream parasites to exit their proliferative cell cycle and differentiate to morphologically stumpy forms that are adapted for transmission by tsetse flies [24] . Therefore , this signal-response pathway serves two purposes- it restricts proliferation of the parasite in the host , prolonging host survival , and it also optimises the parasite for its uptake by the disease vector [21] . The soluble signal driving parasite development is not identified despite many years of effort . However , the pathway that transduces the signal is quite well characterised at least in terms of its molecular composition . This is because the action of SIF can be mimicked in vitro by cell permeable cAMP analogues which cause cells to arrest but not undergo full development to stumpy forms [23 , 25] . This enabled us to carry out a genome-wide RNAi screen for molecules whose silencing renders parasites unresponsive to the SIF mimic signal [26] . Analysis of the genes contributing to resistance identified approximately 30 components and , for several of them , their involvement in physiological stumpy formation was confirmed in vivo . Thus , where cells silenced the expression of identified genes they became highly virulent in mice because they did not arrest with the accumulation of SIF . This validated the selectional screen and provided a large compendium of molecules linked to developmental signalling in the parasite . Furthermore , analysis of the molecules identified revealed that they fell into a potential hierarchy with signal processing molecules ( specific to cell permeable cAMP ) identified , as well as protein kinases and phosphatases and predicted post transcriptional gene expression regulators [24] . Complementing this screen for activators of stumpy formation , a number of inhibitors of stumpy formation have been described [27 , 28] , including a novel component of the TOR family whose gene silencing drives parasites to become arrested as stumpy like forms , although this was only tested in monomorphic parasites that are naturally incompetent to undergo this developmental transition [29] . Signalling may involve AMPK ( also identified in the genome-wide screen for positive regulators of stumpy formation ) , which becomes phosphorylated during the slender to stumpy differentiation and whose activity regulates quiescence [30] . Despite the identification of molecules linked to stumpy formation , their positioning with respect to one another is unknown . Here we have applied extragenic suppression as a strategy to dissect the quorum sensing signalling pathway in African trypanosomes . This has exploited our ability to create null mutants for pathway components that have lost the ability to respond to the SIF signal and to simultaneously express in the same cells other pathway components under doxycycline regulatable expression both in vitro and in vivo . This approach has determined the dependency relationships between distinct components of the pathway and the existence of complex non-linear relationships in the pathway . The further analysis of one of the null mutants in the signalling pathway has provided a molecular characterisation of the difference between differentiation competent cells and those lacking a central developmental signalling component .
To investigate the dependency relationships between identified components of the SIF signalling pathway in Trypanosoma brucei we used pairwise gene deletion and inducible overexpression to assess the ability of one gene to compensate for the loss of another ( Fig 1A ) . This required the creation of null mutants for the target genes because the kinetics of simultaneous inducible RNA interference and inducible ectopic expression complicate the interpretation of any resulting phenotypes . It also required the use of pleomorphic trypanosome lines since these are competent for developmental progression to stumpy forms in vivo , unlike monomorphic lines which are far more easily manipulated but developmentally incompetent . The creation of null mutants in T . brucei EATRO 1125 AnTat1 . 1 90:13 involved a sequential allelic replacement using nested integrative cassettes to replace each target gene copy in the trypanosome’s diploid genome . Thereafter , ectopic overexpression in the null mutant lines was achieved through integration of a construct allowing doxycycline inducible expression in vivo , this generating a protein with a TY epitope tag permitting its expression to be monitored ( Fig 1B ) . Each generated cell line was assessed in vivo for their ability to produce stumpy forms , this involving scoring the virulence of the parasites in mice , their accumulation as quiescent G1 arrested cells as parasite numbers increased ( indicated by an accumulation of cells with 1 kinetoplast and 1 nucleus ) , their morphological differentiation to stumpy forms and their expression of the stumpy form specific cell surface marker PAD1[31] . The ability of cells to generate procyclic parasites once harvested from infections and exposed to cis aconitate ( CA ) was also assessed , this confirming the developmental competence of the parasites: stumpy forms uniformly express the procyclic form surface marker EP procyclin within 2–4 hours of exposure to CA whereas slender forms either do not differentiate , or differentiate asynchronously and inefficiently over 24 hours [32] . In all cases , triplicate infections were carried out for both ‘induced’ and ‘non induced’ infection profiles . Our earlier RNAi analysis had indicated that the predicted RNA binding protein RBP7 ( encoded by two tandemly arranged gene copies , RBP7A and RBP7B ) was required for the sensitivity of trypanosomes to cell permeable cAMP and for the SIF-induced differentiation of pleomorphic trypanosomes to stumpy forms in vivo[26] . To confirm this , we initially created pleomorphic null mutants for RBP7AB using integrative constructs targeting the 5’UTR of RBP7A and the 3’UTR of RBP7B , the resulting cells being validated by genomic PCR amplification of a fragment of the coding regions using gene specific primers ( S1A Fig; other null mutants described in this study are validated in S1B and S1C Fig ) . Thereafter , the RBP7AB null mutants were assessed in vivo for their developmental capacity in parallel with wild type T . brucei EATRO 1125 AnTat1 . 1 90:13 which are differentiation competent . Fig 2A demonstrates that the null mutant parasites proliferated in vivo until infections were terminated on day 4 on humane grounds due to their rapidly ascending parasitaemia . In contrast , the parental control infection showed restricted growth from day 3 and began to develop morphological stumpy forms on day 4 . Consistent with this , the control cells showed a significantly higher proportion of parasites with a 1 kinetoplast 1 nucleus configuration of DNA containing organelles[33] reflecting G1 arrest ( 89%±0 . 9 vs . 69±2 . 3% 1K1N in the null mutant on day 4; p = 0 . 0014 , Fig 2B ) and , when harvested , expressed the stumpy form marker PAD1 on 72±8 . 5% of cells compared with 21±5 . 5% of the RBP7AB null mutant cells ( Fig 2E ) . Furthermore , upon incubation with 6mM CA , 6±3 . 8% of the RBP7AB null mutant parasites expressed EP procyclin after 4 hours , contrasting with 70±3 . 4% of the control cells ( Fig 2F ) . All of these parameters confirmed that RBP7AB gene deletion reduces the ability of the parasites to arrest as stumpy forms in vivo , matching our previous RNAi analysis[26] . However , whilst the inability to completely eliminate stumpy formation seen previously could have been attributable to incomplete gene silencing using RNAi lines , the use of null mutants demonstrated that RBP7AB loss reduced differentiation capacity , but did not completely ablate either stumpy formation or differentiation competence . Having established the effect of RBP7AB deletion on stumpy formation , we analysed the interaction between RBP7 and YAK ( Tb927 . 10 . 15020 ) , encoding a predicted protein kinase of the DYRK family . As with RBP7AB , our earlier analysis had demonstrated that YAK RNAi reduces stumpy formation [26]and consistent with this , a null mutant for YAK resulted in parasites that were virulent in vivo ( Fig 2C ) and showed reduced accumulation in the 1K1N configuration ( 74±1 . 2% vs . 89±0 . 96% in the control , p = 0 . 0008; Fig 2D ) . The cells also did not express PAD1 as assessed by flow cytometry ( 0 . 7±0 . 1% vs . 72±8 . 5%; Fig 2E ) and less than 1% of cells expressed EP procyclin 4 hours after exposure to CA ( 0 . 4±0 . 1%; Fig 2F ) . Hence , contrasting with RBP7AB null mutants , where inefficient differentiation was observed , YAK null mutants had very significantly abrogated their ability to differentiate to stumpy forms . The ability of RBP7B ectopic expression to restore differentiation competence to the RBP7AB and YAK null mutant was then examined . Initially , the consequences for induction of RBP7B ectopic expression were evaluated in vitro by growth analysis . S2A and S2B Fig , demonstrate that when RBP7B ectopic expression was induced in either the RBP7AB null mutant line , or the YAK null mutant line , growth was inhibited there being equivalent RBP7B expression in each mutant ( S2C Fig ) . This matched our prior demonstration that RBP7B ectopic expression in wild type T . brucei EATRO 1125 AnTat1 . 1 90:13 cells caused arrested growth in vitro and in vivo , the latter being accompanied by an accumulation of cells with 1K1N configuration and increased expression of EP procyclin when exposed to 6mM CA [26] . It also indicated that the ectopic expression of Ty-tagged RBP7B alone could restore differentiation in the RBP7AB null mutant . The consequences of RBP7B ectopic expression in the YAK knockout line were then assessed in vivo ( Fig 3A ) . Here , the uninduced parasites were highly virulent such that infections were humanely terminated as the parasitaemia ascended beyond 3x108 parasites/ml . In contrast , with RBP7B ectopic expression , the parasites showed slow growth reaching 1x108 parasites/ml on day 4 of infection and a small proportion of cells ( ~10% ) cells with an intermediate/stumpy morphology were observed ( Fig 3B ) . Furthermore , they accumulated with a 1K1N configuration , indicative of cell cycle arrest ( 86±1% induced vs . 74±5% uninduced on day 4 , p = 0 . 019; Fig 3C left panel ) . When PAD1 was examined , the capacity of some cells to express this marker was observed ( 9±1 . 4 induced vs . 0 . 3±0 . 1% uninduced on day 4 , p = 0 . 0029; Fig 3C right panel ) , this reflecting the proportion of morphologically stumpy forms seen in the population at this relatively early time point . Hence , RBP7B ectopic expression can restore growth arrest and the ability to undergo at least some stumpy transformation in YAK KO cells , indicating that RBP7B induced arrest and stumpy formation is not dependent on YAK . In addition to RBP7AB and YAK , genome-wide RNAi screens for resistance to the cell permeable cAMP analogue pCPTcAMP identified protein phosphatase 1 as potentially necessary for stumpy formation , this being confirmed in vivo by the simultaneous RNAi-mediated silencing of three PP1 genes ( PP1-4 , 5 , 6; respectively , Tb927 . 4 . 3640 , Tb927 . 4 . 3630 , Tb927 . 4 . 3620; >94 . 90% identity ) , whereupon stumpy formation was lost[26] . The creation of null mutants for the PP1-4 , 5 , 6 genes , individually or together , was not successful potentially due to the tandem arrangement of these genes . However , it was possible to explore the ability of PP1-6 ectopic expression to restore stumpy formation in the RBP7 and YAK null mutant lines evaluated above . Initially we investigated the consequences of PP1-6 ( Tb927 . 4 . 3620 ) ectopic expression in a wild type T . brucei EATRO 1125 AnTat1 . 1 90:13 background in vitro ( S3A Fig ) and in vivo ( Fig 4A ) . Here , induction of PP1-6 expression resulted in the rapid cessation of parasite growth ( Fig 4A , left ) and the appearance of morphologically stumpy forms in rodent infections , despite the low parasitaemia of the induced population . The induced parasites also showed an accumulation of cells with 1K1N configuration ( p<0 . 0001: S4A Fig ) and elevated expression of PAD1 ( Fig 4D ) . Their capacity for differentiation to procyclic forms in vitro as assessed by EP procyclin expression was also significantly enhanced ( p = 0 . 0002; Fig 4A , right ) . Having demonstrated that PP1-6 ectopic expression could precipitate premature differentiation to stumpy forms in a parental background , inducible expression of PP1-6 was tested in the RBP7AB and YAK null mutant lines . In an RBP7AB null mutant background , PP1-6 ectopic expression was accompanied by reduced parasitaemia ( Fig 4B ) and the appearance of morphologically intermediate-stumpy forms on day 4 of infection ( Fig 4C ) . This was matched by the expression of PAD1 ( Fig 4D ) , the accumulation of 1K1N cells ( 92±4% vs . 73±1%; p = 0 . 0095; S4B Fig ) , and increased capacity for differentiation to procyclic forms when exposed to CA ( 82±4% vs . 33±10% , p = 0 . 0114; Fig 4B , right ) . Hence , ectopic expression of PP1-6 restored differentiation competence to RBP7AB null mutants and demonstrated that PP1-6 induced differentiation was RBP7 independent . With a YAK null mutant background the picture was more complex . In vitro , reduced growth accompanied PP1-6 overexpression ( S3B Fig ) and in vivo , the induction of PP1-6 expression caused the parasitaemia to arrest ( Fig 5A ) although this was not accompanied by a large accumulation of 1K1N cells ( 65±2 . 3% -DOX vs . 73±1% +DOX; p = 0 . 029; ( S4C Fig ) ) nor effective expression of PAD1 ( 4±1% -DOX vs . 16±3% +DOX , p = 0 . 015 as determined by immunofluorescence; Fig 5B shows a western blot of PAD1 expression ) . Furthermore , the capacity of the cells to differentiate to procyclic forms was much less than when PP1-6 was expressed in the parental background ( 19±6 . 7% compared to 91±1 . 4% , p = 0 . 05; Figs 4A and 5A right panels ) . Also , when the morphology of parasites induced to ectopically express PP1-6 was examined in the YAK null mutant line , stumpy cells were not observed ( Fig 5C ) . Rather , the parasites remained slender but demonstrated aberrant nuclear morphologies , this appearing elongate and often curved or irregular in shape ( Fig 5C , lower DAPI stained samples ) . This morphological phenotype was quantitated using two descriptors- aspect ratio and solidity . The aspect ratio defines the length as a proportion of the width of the organelle , whereas the solidity measures the area of the nucleus divided by its convex area , providing a measure of how concave the nucleus was . An analysis of more than 100 cells ( +dox , 102; -dox , 145 ) demonstrated that the doxycyline treated cells induced to express PP1-6 in the YAK background showed a significantly increased aspect ratio and decreased solidity ( p<0 . 00005 ) ( Fig 5D ) . This nuclear phenotype was not present when PP1-6 was ectopically expressed in the parental cell line , indicating that it was generated only in the absence of YAK or with the high level of PP1-6 expression in that line . Overall this analysis demonstrated that PP1-6 can drive premature differentiation in parental cells and RBP7AB null mutants but not in the YAK null mutant , where aberrant cells were generated . This indicates that PP1-6 induced stumpy formation is RBP7AB independent and YAK dependent . In addition to YAK kinase , a member of the NEK kinase family was also identified in the screen for stumpy inducers , NEK17 [26 , 34] ) . NEK17 is encoded by a cluster of three almost identical genes ( Tb927 . 10 . 5950; Tb927 . 10 . 5940; Tb927 . 10 . 5930; e = 0 . 0 ) indistinguishable by RNAi . Therefore , the ability of one representative , Tb927 . 10 . 5950 to restore stumpy formation to the RBP7AB and YAK null mutants analysed above was explored by ectopic expression . In vitro , the inducible expression of NEK17 in the parental T . brucei EATRO 1125 AnTat1 . 1 90:13 background prevented population growth ( S5A Fig ) . When the same assay was carried out in the RBP7AB null mutant , however , the parasites grew as well as uninduced parasites ( S5B Fig ) despite the effective expression of NEK17 , which was at higher level than in the parental background ( S5D Fig , left panel ) . In contrast , when examined in a YAK null mutant background , NEK17 ectopic expression resulted in slowed parasite growth ( S5C Fig ) , and expression of the ectopically expressed protein was at the elevated levels seen in the RBP7AB null mutant ( S5D Fig , right panel ) . In addition to cell growth in vitro , the ability of NEK17 expression to drive physiological stumpy formation was analysed in mouse infections . As in culture , NEK17 overexpression in RBP7AB null mutants did not arrest growth in vivo; the parasites continued to proliferate , albeit more slowly ( Fig 6A ) , and remained slender in morphology ( Fig 6B ) despite effective expression of the protein ( Fig 6C ) . The cells also did not accumulate in a 1K1N cell cycle configuration ( Fig 6D ) nor express PAD1 at an elevated level ( Fig 6E ) . When NEK17 was expressed in a YAK KO background in vivo cells showed greatly reduced growth ( Fig 7A ) , as seen in vitro ( S5C Fig ) . However , this was not linked to G1 arrest; rather there was appearance of the aberrant cytological configuration 1K2N ( Fig 7B ) and PAD1 expression on some cells albeit at low frequency ( ~5% ) ( Fig 7C ) . Hence , in an RBP7AB null mutant background , neither arrest nor morphological change or PAD1 expression was seen , demonstrating arrest mediated by NEK ectopic expression is RBP7AB dependent . In contrast , NEK17 expression arrested growth independently of YAK but this caused only very limited development to stumpy forms and cell cycle abnormalities , demonstrating that effective NEK17 induced differentiation is also YAK dependent . So far , our studies had focused on interactions between different positive regulators of stumpy formation . However , inhibitors of stumpy formation have also been identified . Amongst these , TbTOR4 is a ‘slender retainer’[24] , whose depletion causes stumpy forms to appear in monomorphic populations which normally cannot undergo this development [29] . To explore whether the same effect was present in physiologically-relevant pleomorphic cells , we generated an inducible RNAi line able to reduce the expression of TbTOR4 . This line was then analysed in mice , where induction of TbTOR4 RNAi generated reduced growth over 6 days ( S6A Fig ) when the levels of TbTOR4 mRNA were reduced ( S6B Fig ) . This indicated that , as in monomorphs , TbTOR4 depletion arrested cells prematurely , albeit incompletely , perhaps related to the remaining TbTOR4 after RNAi induction . The effect of TbTOR4 is expected to be close to the top of a signal transduction pathway if organised similarly to other eukaryotes . Also expected to be close to the top of a conventional signalling pathway are MAP kinase kinase kinases ( MEKK ) and MAP kinase Kinases ( MEK ) . One such molecule was identified in the genome-wide screen for stumpy inducers , MEKK1 ( Tb927 . 2 . 2720 ) whose Leishmania major orthologue LmjF02 . 0570 is annotated as a Ste11/MEKK ( the T . brucei protein is also named FlaK [34] for its flagellar location; www . Tryptag . org ) . Therefore , we investigated the dependency , if any , between MEKK1 and TbTOR4 by creating a MEKK1 null mutant and then activating TbTOR4 RNAi in the same cell line . This was anticipated to determine whether the absence of MEKK1 overrides TbTOR4 silencing-induced arrest , or not . As expected , the MEKK1 null mutant was virulent in mice such that infections had to be terminated by day 5 ( Fig 8A , MEKK1 , middle panel ) , this being consistent with their incapacity to generate stumpy forms . However , when RNAi against TbTOR4 was induced in the MEKK1 null mutant background , the growth of the null mutant was reduced , indicating that TbTOR4 silencing overrides the virulence generated by MEKK1 deletion ( Fig 8A; MEKK1 KO TOR4 RNAi ) . This was supported when the accumulation of cells in a 1K1N configuration was examined ( Fig 8B ) . Thus , MEKK1 null mutants showed reduced levels of 1K1N cells compared to parental T . brucei EATRO 1125 AnTat1 . 1 90:13 cells ( Fig 8B centre ‘MEKK1KO’ ) , whereas the induction of TbTOR4 RNAi in the MEKK1 null mutant resulted in significantly more 1K1N cells on day 3 and day 4 of infection with respect to when TbTOR4 RNAi was not induced ( Fig 8B right ) . This demonstrated that TbTOR4 depletion results in the arrest of parasites independently of the presence of MEKK1 . Interestingly , however , when the expression of PAD1 was investigated , it was found that TbTOR4 RNAi in the MEKK1 null mutant resulted in much less PAD1 expression than when TbTOR4 was depleted in the presence of MEKK1 ( Fig 8C ) . Thus , on day 4 of infection , TbTOR4 RNAi in the parental cell line generated 38 . 4±7 . 6% PAD1+ve cells , whereas TbTOR4 RNAi in the MEKK null mutant at the same time point , few ( 11 . 2±2 . 2% ) cells were PAD1 +ve ( Fig 8C ) . A similar phenomenon was observed on day 5 of infection ( TbTOR4 RNAi induced in parental cells , 57 . 4±10 . 2%; TbTOR4 RNAi induced in the MEKK1 null mutant , 19 . 7±1 . 6% ) . Apparently therefore , TbTOR4 depletion causes the arrest of cells independently of the presence of MEKK1 , but MEKK1 is necessary for the efficient expression of the stumpy marker PAD1 . Since arrest precedes the expression of PAD1 protein in the development pathway that results in the formation of stumpy cells , this indicates that both TbTOR4 and MEKK1 are involved in cell cycle arrest , but that MEKK1 is required for efficient development to stumpy forms . Our genetic dissection above attempted to identify dependency relationships between different components of the differentiation control pathway . However , these studies do not inform on molecular interactions between the components , nor distinguish direct or indirect molecular consequences of the perturbations generated . To explore molecular changes associated with a loss of developmental competence in slender forms we analysed the phosphoproteomic changes that accompany deletion of MEKK1 as this was expected to operate early in the signalling pathway . Hence , parental T . brucei EATRO 1125 AnTat 1 . 1 90:13 and the MEKK1 null mutant cells were cultured in duplicate at equivalent cell density in vitro and protein extracted and analysed after isobaric tandem mass tagging ( Fig 9A ) . In total 4767 unique proteins and 2499 unique phosphoproteins were identified; correlations between the replicates were >99% at the peptide level , and at the phosphopeptide level were 0 . 91 ( T . brucei EATRO 1125 AnTat1 . 1 90:13 ) and 0 . 88 ( MEKK1 null mutant ) respectively , demonstrating excellent reproducibility ( Fig 9B ) . The datasets ( S1 Table ) were then analysed for peptides with >1 . 5-fold changes in phosphorylation , with an adjusted P value of <0 . 05 , regardless of the direction of change ( i . e . more phosphorylated or less phosphorylated in the MEKK1 null mutant compared to T . brucei EATRO 1125 AnTat1 . 1 90:13 cells ) ( Fig 9C ) . This identified 19 proteins , of which 17 were 1 . 5-fold less phosphorylated in the null mutant and 2 were 1 . 5-fold more phosphorylated . Most interestingly , in the MEKK1 null mutant we observed significantly reduced phosphorylation of peptides derived from another component of the quorum sensing signalling pathway , NEK 17 ( Log2–1 . 68; adj P = 0 . 0128 ) . Further , analysis of the phosphosite change revealed that NEK 17 was less phosphorylated in the absence of MEKK1 at the activation loop threonine 195 ( S7 Fig ) , consistent with reduced kinase activity when cells are less able to generate stumpy forms . Although indirect phosphorylation through other molecular components of the pathway is possible , this invokes the potential for MEKK1 to directly phosphorylate and activate NEK17 in the trypanosome quorum sensing signalling pathway . Other proteins with reduced phosphorylation in the absence of MEKK1 included the kinetoplastid kinetochore protein KKT4 and RNA regulators ALPH1 and RBP31 .
Genome-wide RNAi screens identify the genes whose silencing renders parasites resistant to an imposed selection[35] . This approach was applied to identify genes that confer resistance to cell permeable cAMP , acting as an in vitro proxy for SIF-induced differentiation in vivo[26] . The resulting screen identified approximately 30 genes , many of which were subsequently validated in individual RNAi lines , and tested for their inability to undergo natural stumpy formation in vivo . This confirmed the involvement of many of the hits from the original selection as molecules linked to developmental competence in bloodstream form trypanosomes . However , whilst the list of genes identified suggested the existence of a potential signalling pathway , the ordering and interactions between components of the pathway could not be assumed . Furthermore , it was unclear whether resistance to cell permeable cAMP was enacted through a simple processional linear pathway or whether there was more complexity . One established approach to address this question in conventional genetic systems has been to explore the ability of one molecular mutant to suppress a second mutant and thereby gain understanding of their relative positioning with respect to one another; this also often highlights direct molecular interactions between the respective molecules[36–38] . An alternative approach , epistasis or extragenic suppression , can use the ectopic expression of a wild type or mutant protein to restore a phenotype lost by mutation of another component in a pathway . This also orders the molecules with respect to one another , but does not necessarily imply direct molecular interaction[37] . Here we have used extragenic suppression and phosphoproteomic analysis of a null mutant for a signalling component to dissect the pathway responsible for the generation of stumpy forms in response to the quorum sensing signal , stumpy induction factor , and assigned pairwise dependency relationships between several molecules involved in the process . The analysis of gene function in trypanosomes has been enormously assisted by the use of inducible gene silencing via RNA interference[39] . However , this has limitations where gene silencing is incomplete , and because the kinetics of gene silencing are such that loss of protein function might not be observed for over 24 hours . These limitations thwarted our early attempts to understand molecular ordering in the stumpy formation pathway , since the ectopic expression of rescue molecules occurred more rapidly than the effective depletion of the partner molecule . Although this issue is now less problematic due to the development of alternative inducible systems based on vanillic acid [40] or cumate [41] control , such that independent gene silencing and ectopic expression can be activated , null mutants provide the best background for complementation studies despite the technical challenge associated with this in pleomorphic cell lines . This was highlighted in our analysis of the differentiation capacity of RBP7AB null mutants that retained the capacity for expression of PAD1 in a small proportion of cells , and these were able to differentiate to procyclic forms albeit inefficiently . With the creation of null mutants , low level protein remaining after incomplete gene silencing by RNAi can be eliminated as an explanation for this inefficient differentiation and it is clear that loss of these molecules does not completely eliminate differentiation capacity . The importance of the use of null mutants for the interpretation of pathway dependencies has been well recognised in the genetic dissection of cell lineage determination in C . elegans[37] . Analysis of the combination of gene knock out and ectopic expression for several components identified from our RNAi screen highlighted potential features of the control pathway ( Fig 10A ) . Thus , both RBP7AB and YAK knockout lines prevented efficient development to stumpy forms , and this was restored when RBP7B was ectopically expressed in either line . This confirms the involvement of both genes in the stumpy formation in vivo and potentially places RBP7B downstream of YAK , an order that might be anticipated given the predicted role of RBP7B as a gene expression regulator[24 , 42] . However , when PP1-6 was ectopically expressed in the same null mutant lines , stumpy formation was observed for the RBP7AB null mutant indicating that PP1-6 can promote stumpy formation independently of the presence of RBP7AB . For YAK , the picture was more complex . Although growth inhibition was observed in vitro , analysis of PPI-6 expression in the YAK KO line in vivo highlighted that growth retardation was not accompanied by the hallmarks of stumpy formation: the cells did not activate the expression of PAD1 , nor were they capable of efficient differentiation to procyclic forms . Instead , the parasites exhibited a slender morphology and cytological abnormalities , with the nucleus becoming elongate and misshapen . Interestingly , this phenotype was not seen when PP1-6 was expressed in a parental background or the RBP7AB null mutant but was restricted to the YAK null mutant . The trypanosome genome contains 7 PP1 genes , of which a cluster of 3 genes , indistinguishable by RNAi , were identified in the screen for regulators of stumpy formation ( PP1-4 , PP1-5 , PP1-6 ) [26] . The predicted encoded proteins share 95–99% identity . RNAi silencing of all PP1 genes in procyclic forms generates slow growth ( 40% of control growth ) and an accumulation of mitotic and post mitotic cells[43] , whereas specifically depleting PP1-1 to PP1-3 in procyclic forms generates altered positioning of the nucleus and kinetoplast[44] . Okadaic acid treatment ( inhibiting PP1 and PP2 activities in vitro ) also caused an accumulation of multinucleate cells[45] . Hence , the phenotype observed with PP1-6 expression in the YAK null mutant background is similar to organellar defects seen with other perturbations of PP1 genes , and may represent activity of PP1-6 against inappropriate targets in the absence of YAK . Contrasting with PP1-6 , the slow growth and differentiation phenotype induced by NEK17 ectopic expression was RBP7AB dependent , being absent in the RBP7AB null mutant . NEK17 is a member of the expanded NEK family in T . brucei which includes NEK12 . 2 ( RDK2 ) , that has been proposed to act as a negative regulator of differentiation to procyclic forms such that its depletion precipitates procyclic formation[34] . When ectopically expressed in a YAK null mutant , NEK17 inhibited the growth of parasites but only a few stumpy cells were generated . This indicates NEK17 induced stumpy formation is RBP7AB-dependent and that , as with PP1-6 , its normal function may also be YAK-dependent . Interestingly , the expression level of NEK17 that was detectable in the parental and YAK or RBP7AB null mutants differed . Thus , parental cells , with an intact stumpy formation pathway consistently expressed less NEK17 than either of the null mutant lines . Although individual differences between distinct cell lines cannot be eliminated , NEK17 may be subject to regulatory feedback such that its expression is restricted in developmentally-competent cells . As well as promoters of the differentiation step , inhibitors of the differentiation from slender to stumpy forms have been identified . In particular , a component of a novel TORC complex , TbTOR4 , has been identified that prevents the differentiation of monomorphic cells to stumpy forms[29] . To evaluate this finding in a physiologically relevant context , we initially explored the consequences of TbTOR4 inducible depletion in pleomorphic bloodstream forms and confirmed that they too exhibited slow growth and premature G1 arrest upon TbTOR4 loss , although this was incomplete probably due to incomplete transcript depletion . Thereafter , we explored the interaction between TbTOR4 and the predicted MAP Kinase Kinase Kinase ( MEKK1 ) since this too would be predicted to operate early in a signalling cascade if structured in a similar way to conventional eukaryotic pathways . As expected , the MEKK1 null mutant was virulent in vivo and did not generate stumpy forms . However , when TbTOR4 was depleted in the MEKK null mutant , slowed growth and G1 arrest was observed similar to the depletion of TbTOR4 alone . This demonstrates that the growth effects of TbTOR4 depletion act independently of MEKK1 . However developmental competence after TbTOR4 depletion was not complete in the absence of MEKK1; instead , the cells expressed low levels of PAD1 . This may reflect parasites entering cellular quiescence reversibly before committing to irreversible stumpy formation and full PAD1 expression . This would be consistent with our previous mathematical modelling of the transition from proliferative slender forms where cells undergo commitment to differentiation only after cell cycle arrest[46] . It is also consistent with PAD1 being a marker for later stages in the differentiation programme , its mRNA being expressed prior to the expression of PAD1 protein that accompanies morphological transformation[21 , 46] . This highlights that TbTOR4 may be needed to prevent reversible cell cycle arrest , similar to the signal-induced reversible quiescence in several eukaryotic systems[47–51] . This is also compatible with the effects of AMPK on trypanosome cellular quiescence observed previously [30] . In contrast , MEKK1 is needed for both effective cell cycle arrest and stumpy formation . The molecular consequences of depletion of MEKK1 were also evaluated by phosphoproteomic analysis of isobaric mass tag labelled samples . In this analysis , it was important to use pleomorphic slender cells grown in culture so that the mutant and parental ( MEKK1+ ) population could be assessed at an equivalent cell density and growth phase , maximising the ability to detect phosphorylation differences between cells with or without MEKK1 . As a protein kinase , direct substrates of MEKK1 might be detected in the null mutant line as molecules demonstrating reduced phosphorylation , although phosphorylation changes in either direction could also result indirectly from the absence of MEKK1 . Interestingly , we found that another SIF pathway component , NEK17 , showed less phosphorylation in the MEKK1 null mutant line . Furthermore , analysis of the phosphorylation site indicated that it was positioned at the activation loop threonine 195 of the NEK kinase and therefore potentially able to regulate its activity . This provides support for NEK17 being positioned downstream of MEKK1 in the differentiation control pathway although direct interaction remains to be demonstrated . Also identified were several hypothetical proteins , and also proteins linked to gene expression ( ALPH1 , an mRNA decapping enzyme[52] , and the post transcriptional repressor RBP31[53] ) and cytoskeletal or protein trafficking control . Regulation of development through the action of RNA binding proteins would be expected given the emphasis of post transcriptional control of the trypanosome genome , such that differential phosphorylation of these proteins may affect their ability to bind RNA or interact with other regulatory or RNA binding proteins to retain cells in a slender form , or permit differentiation to stumpy forms . All of these molecules change phosphorylation status in the absence of MEKK1 but their involvement as slender retainers or stumpy inducers awaits individual analysis . In Fig 10A we summarise the data generated in our study and in Fig 10B and 10C present alternative ( non-exhaustive ) models for the combinatorial interactions explored . In one interpretation ( Fig 10B ) , there are two branches leading to G1 arrest and then stumpy formation , one dependent upon YAK , the second dependent on RBP7 . Thus , PP1-6 can drive stumpy formation in the absence of RBP7AB , but not in the absence of YAK , where cell cycle defects are generated . In contrast , NEK17—potentially a direct substrate of MEKK1—is dependent upon both RBP7 and YAK , such that RBP7 is needed to generate arrest and stumpy formation , whereas the absence of YAK generates cell cycle defects and the cells remain slender . Although bifurcated , each branch is not redundant , however , because the loss of individual pathway components on either branch ( e . g . YAK or RBP7AB ) prevents efficient stumpy formation . Perhaps , therefore , the amount of signal through each branch determines the differentiation response . Notably , the ability of RBP7AB null mutants to differentiate at very low level may be contributed by SIF-independent differentiation , a mechanism proposed to operate with disruption of VSG expression site activity[54] . An alternative pathway structure ( Fig 10C ) is unbranched , with NEK17 and PP1-6 acting downstream of YAK but with each dependent on the presence of YAK for their normal activity; in the absence of YAK , cell cycle defects and growth inhibition arise when either NEK17 or PP1-6 are ectopically expressed . Inevitably a complication in the interpretation of these interactions is that the molecules may have other functions in the cell that are disrupted upon their expression or deletion , exemplified by the inability to knock down the expression of PP1-4 , 5 , 6 without inducing a strong growth phenotype in bloodstream form cells[26] . In conclusion , our work has highlighted ( i ) the potential for the differentiation signalling pathway to be non-linear , and ( ii ) that cell cycle arrest does not inevitably lead to stumpy formation , as exemplified by the TbTOR4-induced arrest , that does not generate PAD1+ cells unless there is MEKK1 present . It has also demonstrated that to accurately interpret the phenotypes resulting from perturbation in molecules proposed to be involved in differentiation requires pleomorphic cell lines and detailed monitoring of physiological stumpy formation through several cytological parameters ( G1 arrest , PAD1 expression , morphology and differentiation capacity or procyclic formation ) . When combined with extragenic suppression and phosphoproteomic analysis in a null mutant background , the systematic piecing together of the components of the signalling pathway becomes possible , as do the wider molecular events linked to remaining as slender forms , or committing to development .
All animal experiments were carried out after local ethical approval at the University of Edinburgh Animal Welfare and Ethical Review Body ( approval number PL02-12 ) and were approved under the United Kingdom Government Home office licence P262AE604 to satisfy requirements of the United Kingdom Animals ( Scientific Procedures ) Act 1986 . Infections were carried out in adult MF1 mice . Experiments were carried out using the Trypanosoma brucei brucei EATRO 1125 AnTat1 . 1 . 90:13 line[55] . In vitro , parasites were cultured in HMI-9[56] . Transfection of pleomorphic AnTat 90:13 strains was performed using either cells grown in vitro or harvested from mouse blood[57] . In the latter case , whole blood was harvested from a mouse showing slender parasitaemia and added to 25 ml HMI-9 + 10% FCS and allowed to settle for 4–6 hours . 20 ml of this culture was then carefully pipetted from the top and transferred to a new flask containing another 20 ml media . Any remaining blood was allowed to settle overnight . The following day media was pipetted from the top , transferred to a new flask and used for transfection . 2-3x107 parasites at a density of 6x105-1 . 2x106/ml were centrifuged for 5 minutes at 1000g and washed in TDB ( Trypanosome Dilution Buffer; Trypanosome Dilution Buffer: 5 mM KCl , 80 mM NaCl , 1 mM MgSO4 , 20 mM Na2HPO4 , 2 mM NaH2PO4 , 20 mM glucose , pH7 . 4 ) . They were resuspended in 50–100 μl Amaxa buffer ( Lonza ) mixed with the DNA . The sample was then transferred to a cuvette and electroporated in an Amaxa Nucleofector II using program Z-001 –Free Program Choice . The cells were immediately transferred to a flask containing 30ml pre-warmed media and serially diluted 1:10 and 1:100 in 2 additional flasks . After incubation at 37°C for minimum 6 hours , antibiotic was added to each flask for selection of transfectants . Antibiotic concentrations used for parasite transfection were: Hygromycin ( 0 . 5μg/ml ) , puromycin ( 0 . 5μg/ml ) , phleomycin ( 1 . 5μg/ml ) , G418 ( 2 . 5μg/ml ) , Blasticidin ( 10μg/ml ) . The cultures were plated in 1ml volumes in 24-well plates . Positive transfectants were identified microscopically and diluted into antibiotic-containing media 4–8 days post transfection . For inducible expression and/or RNAi induction , Tetracycline was used at 1μg/ml . Constructs were generated using standard molecular biology protocols or by Gibson assembly . For the latter , the assembly reaction comprised 50–100 ng of vector with 2–3 fold excess of each insert in a 10 μl volume . 10 μl of 2x Gibson Assembly Master Mix was then added and the reaction incubated at 50°C for 15 minutes . Null mutants were created by replacing the YFP and TY tags of the pEnT6B-Y and pEnT6P-Y vectors[58] with fragments of the 3’ and 5’UTRs of each target gene . Restriction digestion of these plasmids between the 3’UTR and 5’UTR sequences produced 2 linear constructs , containing resistance markers to blasticidin and puromycin respectively , which were capable of homologous recombination with , and replacement of , target gene alleles . The pEnT6B-Y construct contained external UTR fragments located distal to the ORF relative to the internal UTR fragments in the pEnT6P-Y construct . This meant that when the pEnT6B-Y construct was transfected into the AnTat 1 . 1 90:13 strain in the first of 2 sequential transfections , one allele of the target gene including the internal UTR fragments was deleted . When the second pEnT6P-Y construct was transfected , recombination occurs only with the remaining target gene allele , producing a double knockout ( dKO ) line resistant to both antibiotics . Ectopic expression was achieved using as a construct backbone pDEX577Y[58] , that integrates in to the 177bp repeat minichromosome region . Drug selection was mediated by a bleomycin resistance cassette transcribed by a constitutive rRNA promoter; inducible expression was achieved using a T7 promoter with a tetracycline operator sequence . Parasites purified from whole blood were resuspended at 3x106/ml in SDM-79 +10% FCS media supplemented with 6mM cis-aconitate ( Sigma ) to induce procyclic differentiation and incubated at 27°C/5% CO2 . After 0 , 4 and 24 hours , 1 ml culture was removed , washed and fixed in 2% formaldehyde/0 . 05% gluteraldehyde for EP procyclin expression analysis by flow cytometry . Samples were prepared in 5ml FACS tubes ( BD Biosciences ) . Cells were fixed in 2% formaldehyde/0 . 05% gluteraldehyde at 4°C overnight . Samples were centrifuged at 2000g for 7 minutes , washed in PBS and blocked in 200 μl 2% BSA/PBS for 40 minutes or overnight . They were centrifuged again and resuspended in primary antibody in 2% BSA/PBS for 1 hour or overnight at 4°C . Cells were then washed again and resuspended in fluorescently conjugated secondary antibody for 40 minutes at 4°C . The cells were washed again and finally resuspended in 500 μl 0 . 2 μg/ml 4’ , 6-diamidino- 2-phenylindole ( DAPI ) /PBS . Antibody concentrations are listed below . At least 2x105 cells were centrifuged at 1000g for 5 minutes , washed in ice-cold PBS and resuspended in 125 μl ice-cold PBS . 75 μl 8% paraformaldehyde in PBS was added and the tube mixed by inverting . The sample was incubated on ice for 10 minutes . 500 μl PBS was added and the tube centrifuged at 1000g for 5 minutes . The cells were resuspended in 0 . 1 M glycine/PBS and stored at 4°C . Samples were resuspended in 15 μl PBS and pipetted onto slide wells pre-treated with 15 μl 0 . 1 mg/ml poly-L-lysine ( Biochrom ) . Slides were incubated in a humid chamber for 1 hour to allow the cells to adhere . For intracellular proteins , 0 . 05% Triton X-100 was added for 5 minutes . The wells were washed 3 times in PBS and then blocked in 20% FCS/PBS for 45 minutes . Excess liquid was removed and primary antibody in 20% FCS/PBS was applied for 1 hour . The wells were washed 5 times and then fluorescent dye conjugated secondary antibody in 20% FCS/PBS added for 1 hour . Secondary antibody was aspirated and 10 μg/ml DAPI applied for 1 minute . The wells were washed 5 times in PBS . The slides were removed from the humid chamber and dried for 10 minutes at 37°C , then mounted and dried . Antibody concentrations used are shown in Table 1 . Lysates were prepared by centrifuging cells at 1000g for 5 minutes , washing in PBS and resuspending in Laemmli buffer at 3x105 cells/μl . For PAD1 expression analysis , samples were sonicated using a Bandelin Sonorex Ultrasonic bath to reduce viscosity . All other samples were incubated at 95°C for 5 minutes . Lysates were stored at -20°C until use . Protein was transferred from acrylamide gels to nitrocellulose membrane ( GE Healthcare ) using a Bio-Rad Mini Trans-Blot cell according to the manufacturer’s instructions , using cold transfer buffer ( 25 mM Tris , 200 mM glycine , 20% methanol ) and run at 200 mA for 90 minutes with stirring . For reaction , the membrane was blocked at room temperature with agitation for at least 30 minutes in 5% milk in PBS-0 . 1% Tween 20 ( PBS-T ) . It was then incubated in primary antibody in 5% milk in PBS-T for minimum 1 hour with agitation and washed 3 times in PBS-T . Secondary detection of the antibody used one of two systems . Either the membrane was incubated in secondary antibody conjugated to horseradish peroxidase ( HRP ) diluted in 5% milk in PBS-T for 1 hour with agitation and washed three times with PBS-T; the membrane was then reacted with Pierce ECL2 Western Blotting substrate for 2 minutes and exposed to X-ray film ( Fujifilm ) . Alternatively , the signal was visualised using the LI-COR Odyssey system . In this case , the secondary antibody was conjugated to a fluorescent dye and diluted in 50% Odyssey Blocking Buffer/50% PBS-T . After 1 hour incubation in secondary antibody , the membrane was washed twice in PBS-T and once in PBS , and scanned using a LI-COR Odyssey imager . For Northern blotting , RNA was extracted from Trypanosoma brucei using the QIAGEN RNeasy kit according to the manufacturer’s instructions and run on a formaldehyde gel in MOPs buffer . RNA was then transferred from the gel to a positively charged nylon membrane and cross-linked using a UV Stratalinker . The blot was then pre-hybridised in 10 ml formaldehyde hybridisation buffer ( 5XSSC , 50% formamide , 0 . 02% SDS , 2% DIG block ) at 68°C for 1 hour in a hybridisation oven ( Techne Hybridiser HB-1D ) . This was then replaced with 7 ml hybridization buffer containing 1 μl DIG labelled riboprobe ( Roche; prepared according to the manufacturer’s instructions ) , which had been denatured for 5 mins at 95°C , and hybridised overnight . The blot was washed at 68°C in the hybridisation tube , first in 2x SSC/0 . 1% SDS three times for 30 minutes and then once in 0 . 5x SSC/0 . 1% SDS for 30 minutes . It was then removed from the tube and rinsed in maleic acid buffer with 0 . 3% Tween 20 for 1 minute at room temperature . Signal was detected according to the Roche DIG-labelling protocol . For preparation of phosphoproteomic samples , low binding tips and low binding eppendorfs were used . Samples were extracted from 2 replicates each of T . brucei EATRO1125 AnTat 1 . 1 90:13 or MEKK1 KO lines . For each , 2 . 7x108 cells at 9x105/ml were spun at 1000 g for 10 minutes at 4°C , washed three times in PBS and resuspended in 100 μl lysis buffer ( 4% SDS; 25 mM Tris ( 2-carboxyethyl ) phosphine ( TCEP ) ( Thermo ) ; 50 mM N-ethylmaleimide ( Thermo ) ; 150 mM NaCl;1x PhosSTOP phosphatase inhibitor ( Roche ) ; 10 mM Na2HPO4 pH6 ) . The samples were sonicated using a Bioruptor ( Diagenode; 10 cycles of 30 seconds alternating sonication/rest at 18°C ) . The samples were then chloroform:methanol precipitated . After 10 minutes incubation at 65°C , 400 μl methanol was added and the samples vortexed . 100 μl of chloroform was added and the samples vortexed again . 300 μl of ddH2O was added and the samples vortexed for 1 minute . They were then centrifuged for 5 minutes at 9000g and the upper phase aspirated . A further 300 μl of methanol was added and the samples vortexed and centrifuged as before . The supernatants were aspirated and the pellets air dried . For tryptic digestion , the pellets were resuspended in 150 μl 8M urea/0 . 1 M Tris pH8/1 mM CaCl2 and quantified by Bradford assay . The protein concentration was determined by comparison to a bovine serum albumin ( BSA ) standard curve ( 0 . 1–2 . 5 mg/ml ) . 125 μl of Bradford Reagent ( Sigma ) was added to 2 . 5 μl protein sample in triplicate in a 96 well plate and absorbance read at 595 nm using a Fluostar Omega plate reader ( BMG Labtech ) . 800 μg of protein for each sample was then diluted with 0 . 1 M Tris pH8/1 mM CaCl2 to 1 M urea and 8 μl LysC ( Wako ) added . The samples were digested overnight at 37°C . 8 μg trypsin ( Pierce ) in 50 mM acetic acid was then added and the samples incubated for 4 hours at 37°C . Finally , 1% v/v trifluoroacetic acid was added . Labelling and mass spectrometry were carried out at the FingerPrints Proteomics Facility at the University of Dundee . The samples purified by solid phase extraction and then quantified by bicinchoninic acid assay prior to labeling with isobaric tandem mass tags ( 6-plex TMT ) . The samples were pooled and fractionated into fractions using hydrophilic interaction liquid chromatography . 5% of each fraction was analysed by nLC-MS/MS ( nano liquid chromatography-tandem mass spectrometry ) using a Thermo Q-Exactive HF mass spectrometer to generate proteomic quantitation . Phosphopeptides were enriched in the remaining 95% of each fraction and these samples were then also analysed by nLC-MS/MS to generate quantitation of phosphopeptides . Statistical analysis of the fold change in phosphorylation between MEKK1 KO and parental replicates was performed using the R package limma [59 , 60] . This uses the empirical Bayes method which calculates a moderated t-statistic . In this method , posteriors residual SDs replace ordinary SDs , effectively compressing the variance of the peptides with similar means towards a common value , allowing a more stable statistical interference when only few measurement are available [61] . Prior the statistical analysis , data were normalised using the voom methodology included in the limma library . All phosphoproteins that showed an adjusted p-value < 0 . 05 and a greater than 1 . 5 fold change compared to parental T . brucei replicates were considered to be differentially regulated . Statistical analysis for all parasitaemia , cell cycle , PAD1 IFA , FACS data collected in triplicate was carried out using either Minitab version 15 or GraphPad Prism version 6 . Where only a single timepoint was analysed , the data was analysed by students t test . In the case of multiple timepoints , the data was analysed by general linear model with Tukey’s test for multiple comparisons . P values of less than 0 . 05 were considered statistically significant . | African trypanosome parasites respond to density sensing information in the bloodstream of their mammalian hosts to generate their transmission stage , the stumpy form . Components of this ‘quorum sensing’ signalling cascade are known but their interactions and ordering are not . Here we have dissected the dependency relationships between molecules in the pathway by combinatorial gene knockout and ectopic expression , as well as by detailed phosphoproteomic analysis of one component . Our results provide a first analysis of the signal pathway architecture , revealing that it is non-linear . Moreover , phosphoproteome analysis reveals pathway hierarchy through identifying that the phosphorylation of a NEK kinase component of the pathway is reduced when a predicted upstream kinase is absent . This provides a framework for the coherent dissection of a signal transduction cascade in these parasites that use quorum sensing to control disease spread . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"parasitic",
"cell",
"cycles",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"cell",
"differentiation",
"parasitology",
"trypanosoma",
"brucei",
"developmental",
"biology",
"protozoans",
"epigenetics",
"genetic",
"interference",
"proteins",
"gene",
"expression",
"life",
"cycles",
"biochemistry",
"rna",
"trypanosoma",
"eukaryota",
"cell",
"biology",
"nucleic",
"acids",
"post-translational",
"modification",
"genetics",
"biology",
"and",
"life",
"sciences",
"trypanosoma",
"brucei",
"gambiense",
"organisms",
"parasitic",
"life",
"cycles"
]
| 2018 | Non-linear hierarchy of the quorum sensing signalling pathway in bloodstream form African trypanosomes |
Accurate and reliable forecasts of seasonal epidemics of infectious disease can assist in the design of countermeasures and increase public awareness and preparedness . This article describes two main contributions we made recently toward this goal: a novel approach to probabilistic modeling of surveillance time series based on “delta densities” , and an optimization scheme for combining output from multiple forecasting methods into an adaptively weighted ensemble . Delta densities describe the probability distribution of the change between one observation and the next , conditioned on available data; chaining together nonparametric estimates of these distributions yields a model for an entire trajectory . Corresponding distributional forecasts cover more observed events than alternatives that treat the whole season as a unit , and improve upon multiple evaluation metrics when extracting key targets of interest to public health officials . Adaptively weighted ensembles integrate the results of multiple forecasting methods , such as delta density , using weights that can change from situation to situation . We treat selection of optimal weightings across forecasting methods as a separate estimation task , and describe an estimation procedure based on optimizing cross-validation performance . We consider some details of the data generation process , including data revisions and holiday effects , both in the construction of these forecasting methods and when performing retrospective evaluation . The delta density method and an adaptively weighted ensemble of other forecasting methods each improve significantly on the next best ensemble component when applied separately , and achieve even better cross-validated performance when used in conjunction . We submitted real-time forecasts based on these contributions as part of CDC’s 2015/2016 FluSight Collaborative Comparison . Among the fourteen submissions that season , this system was ranked by CDC as the most accurate .
Seasonal influenza epidemics cause widespread illness which is associated each year with an estimated 250 000 to 500 000 deaths worldwide [1] and 3000 to 56 000 deaths in the United States alone [2–4] . In contrast to influenza “pandemics” , which are rare global outbreaks of especially novel influenza A viruses [5 , 6] , seasonal epidemics ( i . e . , non-pandemics ) , while still having worldwide reach , occur annually in the United States and other countries with ( generally ) temperate climates . Time series of influenza prevalence in these areas are typically low and flat for the majority of the season , but trace a single , sharp peak sometime during winter , with significant variability in timing and intensity . Accurate and reliable forecasts of seasonal epidemics can help policymakers plan countermeasures such as vaccination campaigns , and increase awareness and preparedness of hospitals and the general public . The Centers for Disease Control and Prevention ( CDC ) monitors influenza prevalence with several well-established surveillance systems [7]; the recurring nature of seasonal epidemics and availability of historical data provide promising opportunities for the formation , evaluation , and application of statistical models . Starting with the 2013/2014 “Predict the Influenza Season Challenge” [8] and continuing each season thereafter as the Epidemic Prediction Initiative’s FluSight project [9] , CDC has solicited and compiled forecasts of influenza-like illness ( ILI ) prevalence from external research groups and worked with them to develop standardized forecast formats and quantitative evaluation metrics . Various approaches to influenza epidemic forecasting are summarized in literature reviews [10–12] and descriptions of the CDC comparisons [8 , 9] . Some common approaches are described below , with references to work applicable to the current FluSight project and related seasonal dengue forecasting tasks , emphasizing more recent work that may not be listed in the above three literature reviews: We present a novel phenomenological approach to epidemiological forecasting using “delta densities” , which assumes an autoregressive dependency structure similar to those of some time series models , but uses a kernel density estimation approach to model these dependencies rather than the common choice of linear relationships plus Gaussian noise . This technique is similar to the method of analogues [19] in that it uses an instance-based , nonparametric estimation procedure , but provides distributional forecasts of entire trajectories rather than point predictions of individual observations . The kernel conditional density estimation ( KCDE ) forecasting method [22] attacks many of the same issues encountered when applying kernel density estimation methods to seasonal epidemic data , but models the dependency structure of future weeks with a copula , while delta density chains together 1-week-ahead simulations . Compared to approaches that treat the entire season as a unit , such as deterministic , single-strain , fully-mixed compartmental models [13] or our previous empirical Bayes approach based on modifying past seasons’ data [21 , 32] , this method forms a larger library of possible trajectories by piecing together local models , which appears to help forecast performance , even though the trajectories considered may seem less reasonable on average . Our second contribution is an adaptively weighted ensemble approach to combining the output of different forecasting methods given their historical and/or cross-validation forecasts . We first implemented this method in preparation for the 2014/2015 FluSight comparison , mixing together our empirical Bayes forecasting method with two baselines ( a uniform distribution and an empirical distribution for each target ) , and later applied it while participating in the Dengue Forecasting project [33] and following FluSight comparisons ( adding up to 9 additional components including delta density based methods ) , and found it improved our forecasts in all cases . Other epidemic forecasting teams have also reported success with concurrently or subsequently developed stacking generalization [34 , 35] ensemble approaches to the FluSight forecasting tasks using Bayesian model averaging [36] , the fixed weighting scheme that we examine below [37] , and alternative adaptive weighting schemes based on gradient tree boosting [37] , as well as with earlier ensemble approaches to short-term point predictions [20] . Methodologically , our adaptively weighted ensemble framework differs from these alternatives in that it selects a weighting over components for a particular forecast using “plug-in” statistical estimators for the optimal weights given the context of the forecast being prepared . Like the adaptive approaches presented in [37] , component weights for each forecast are selected using regression , but the type of regression used and the manner of incorporating additional information , such as the forecast week , are distinct .
Recording every case of influenza is not practicable; infections are often asymptomatic [38] or symptomatic but not clinically attended [39] , laboratory testing may not be performed for clinically attended cases or give false negative results , and reporting of lab-confirmed cases is not mandatory in most instances . Forecast comparisons are instead based on syndromic clinical surveillance data from the U . S . Outpatient Influenza-like Illness Surveillance Network ( ILINet ) [7 , 40] , a group of health care providers that voluntarily report statistics regarding ILI , where ILI is defined as a 100°F ( 37 . 8°C ) fever with a cough and/or sore throat without a known cause other than influenza . CDC aggregates these reports and estimates the weekly percentage of patients seen that have ILI , %ILI , across all health care providers using a measure called weighted %ILI ( wILI ) . CDC hosts the latest ILINet report and other types of surveillance data through Fluview Interactive , a collection of web modules [41]; we provide current and historical ILINet reports and some other data sources through our delphi-epidata API [42] and epivis visualizer [43] . The FluSight project focuses on in-season distributional forecasts and point predictions of key targets of interest to public health officials: When making distributional forecasts , wILI values are discretized into CDC-specified bins and a probability assigned to each bin , forming a histogram over possible observations . The width of the bins was set at 0 . 5 %wILI for the 2015/2016 comparison and 0 . 1 %wILI for the 2016/2017 comparison; we use a width of 0 . 5 %wILI for analysis of the 2015/2016 comparison prospective forecasts , and a width of 0 . 1 %wILI for retrospective evaluation . CDC typically presents wILI values rounded to a resolution of 0 . 1 %wILI; some targets and evaluations are based on these rounded values . We focus on three metrics for evaluating performance of a forecast for a given target: Unibin log score: log p ^ i , where p ^ i is the probability assigned to i , the bin containing the observed value . We use this score for ensemble weight selection and most internal evaluation as it has ties to maximum likelihood estimation , and is “proper score” [44] . A score for a ( reported ) distributional prediction p ^ is called “proper” if its expected value according to any ( internal ) distributional prediction q ^ is maximized when reporting p ^ = q ^ , i . e . , forecasters can maximize their expected scores by reporting their true beliefs . We refer to the “unibin log score” simply as the “log score” except for when comparing it with the multibin log score , which is defined next . The exponentiated mean log score is the ( geometric ) average probability assigned to events that were actually observed . The exponentiated difference in the mean log scores of method A and method B is an estimate of the ( geometric ) expected winnings of unit-sized bets of the form “this bin will hold the true value” when bets are placed optimally according to the forecasts of A , and ( relative ) prices are set optimally according to the forecasts of B . Multibin log score: log ∑ i near observed value p ^ i , where the i’s considered are typically bins within 0 . 5 %wILI of observed values for a wILI target , or within 1 week for a timing target . The multibin log score was designed by FluSight hosts in consultation with participants , and the judgment “near observed value” was selected as a level of error that would not significantly impact policymakers’ decisions . The exponentiated mean multibin log score is the ( geometric ) average amount of mass a forecaster placed within this margin for error of observed target values . Absolute error: | y ^ - y | , where y ^ is the point prediction and y is an observed value . ( In the case of onset , we consider point predictions for the value of onset conditioned on the fact that an onset actually occurs . We do not consider absolute error for onset in instances where no onset occurred . Some methods considered would sometimes fail to produce such conditional onset point predictions when they were confident that there was no onset , but these methods are not included in any of the figures containing absolute errors . ) The FluSight 2015/2016 forecast comparison evaluations were based solely on the multibin log score [45] . The “flu season” is typically defined as epi week 40 of one year through 20 of the next; we also include data from the rest of the year as part of the season for the purpose of fitting models . In all mathematical notation , we will number the first week of the season as 1 rather than using the corresponding epi week . Let W 1 . . t t denote the t-th CDC report of the current season , containing wILI values for weeks 1 through t , inclusive , which is normally published on Friday of week t + 1; T be the number of weeks in the current season ( either 52 or 53 ) ; we omit all details regarding differing season lengths , presenting forecasting methods and labeling epi week plot axes as if all seasons were of length T; Y1 . . T be the ground truth wILI for the current season: the wILI values used for forecast evaluation , specifically the epi week 28 report for the FluSight comparison , or later revisions as they are available for cross-validation analysis; Y 1 . . T s be the ground truth wILI for past season s; and Zt be a vector containing the forecasting targets of interest at the t-th wILI report of the current season: Yt+1 . . t+4 and the seasonal onset , peak week , and peak percentage; for the FluSight comparison , forecasts for these targets were typically due on Monday of week t + 2 , and allowed to use ILINet and any other data released before the deadline . Our goal is to forecast Zt given W 1 . . t t and previous reports . This can be broken down into multiple steps , such as: “Backcast” updates to the data through time t , producing a distribution over Y1 . . t based on the value of W 1 . . t t and previous reports . Connect the backcast for Y1 . . t with corresponding forecasts for Yt+1 . . T , yielding a distribution for the entire trajectory Y1 . . t . Calculate the distribution for Zt corresponding to this distribution over Y1 . . t . We first introduce the delta density method , which forecasts Yt+1 . . T given Y1 . . t ( step 2 ) . We then discuss a separate procedure for combining multiple forecasts into an adaptively weighted ensemble , forecasting Zt given either Y1 . . t or W 1 . . t t ( steps 2–3 or 1–3 ) . We also outline a method for estimating the distribution of Y1 . . t given W 1 . . t t ( step 1 ) , and analyze its performance when used in conjunction with the delta density method . Consider the task of estimating the density function f Y t + 1 . . T ∣ Y 1 . . t using an instance-based approach . Kernel density estimation and kernel regression use smoothing kernels to produce flexible estimates of the density of a random variable ( e . g . , f Y t + 1 . . T ) and the conditional expectation of one random variable given the value of another ( e . g . , E [ Y t + 1 . . T ∣ Y 1 . . t ] ) , respectively; we can combine these two methods to obtain estimates of the conditional density of one random variable given another . One possible approach would be to use the straightforward estimate f ^ Y t + 1 . . T ∣ Y 1 . . t ( y t + 1 . . T ∣ y 1 . . t ) = ∑ s = 1 S I 1 . . t ( y 1 . . t , Y 1 . . t s ) O t + 1 . . T ( y t + 1 . . T , Y t + 1 . . T s ) ∑ s = 1 S I 1 . . t ( y 1 . . t , Y 1 . . t s ) O t + 1 . . t ( y t + 1 . . t , Y t + 1 . . t s ) , where {1 . . S} is the set of fully observed historical training seasons , and I1 . . t and Ot+1 . . T are smoothing kernels describing similarity between “input” trajectories and between “output” trajectories , respectively . However , while basic kernel smoothing methods can excel in low-dimensional settings , their performance scales very poorly with growing dimensionality . During most of the season , neither Y1 . . t nor Yt+1 . . T is low-dimensional , and the current season’s observations are extremely unlikely to closely match any past Y 1 . . t s or Y t + 1 . . T s . This , in turn , can lead to kernel density estimates for Yt+1 . . T based almost entirely on the single season s with the closest Y 1 . . t s when conditioning on Y1 . . t , and excessively narrow density estimates for Yt+1 . . T even without conditioning on Y1 . . t . So , instead of applying kernel density estimation directly , we first break the task down into a sequence of low-dimensional sub-tasks . We avoid the high-dimensional output problem by chaining together estimates of f Δ Y u ∣ Y 1 . . u - 1 for each u from t + 1 to T , where ΔYu = Yu − Yu−1; estimating these single-dimensional densities requires relatively little data . However , this reformulation exacerbates the high-dimensional input problem since we are conditioning on Y1 . . u−1 , which can be considerably longer than Y1 . . t . We address the high-dimensional input problem by approximating f Δ Y u ∣ Y 1 . . u - 1 with f Δ Y u ∣ R u , where Ru is some low-dimensional vector of features derived from Y1 . . u−1 . Smoothing kernel methods are used to approximate the conditional density functions using data from past seasons . We use two sets of choices for the approximate conditional density function and summary features to form two versions of the method . This same approach can be applied to estimate the distribution of residuals of a wILI point predictor . Suppose that we have observed our goal is to estimate the conditional distribution of given Y 1 . . t 1 and X 1 . . t 2 , using data from past seasons . This can be achieved by chaining together draws from conditional density estimates of ( Y − X ) u ∣ Ru for u from t1 + 1 to t2 , where Ru is a function of Y1 . . u−1 and X 1 . . t 2 . The delta density method can be seen as a special case where t1 = t; t2 = T; X1 . . t = Y1 . . t , past values of Y which are treated as known and are simply duplicated in the simulated trajectories; and Xt+1 . . T = Yt . . T−1 , values of Y which begin as unknown but are filled in as needed by previous simulation steps , giving ( Y − X ) t+1 . . T = ΔYt+1 . . T . We use the residual density method to backcast Y1 . . t from W 1 . . t t and as the basis for another forecaster in the ensemble . Fig 2 shows sample forecasts over wILI trajectories generated by each of these approaches and compares them to some alternatives described in S1 Appendix . Forecasting systems that select effective combinations of predictions from multiple models can improve on the performance of the individual components , as demonstrated by their successful application in many domains . For each probability distribution and point prediction in a forecast , we treat the choice of an effective combination as a statistical estimation problem , and base each decision on the models’ behavior in leave-one-out cross-validation forecasts . Additional cross-validation analysis indicates that this approach achieves performance comparable to or better than the best individual component . Two important features of ILINet data to consider in models and forecast evaluation are 1 . timeliness and accuracy of initial wILI values for each week and subsequent updates to these values , and 2 . changes in behavior on and around major holidays . We examine these details of the data generation process , describe how they are addressed in the delta density model , and demonstrate the importance of considering the update procedure when performing retrospective evaluation and prospective forecasting .
During the 2015/2016 FluSight comparison , we submitted weekly , prospective forecasts from three forecasting systems: Delphi-Stat: an adaptively weighted ensemble of instance-based statistical forecasting methods , and the topic of this paper; Delphi-Archefilter: forms an empirical ( rather than mechanistic ) process model describing wILI trajectories , and incorporates both wILI and multiple forms of digital surveillance data using statistical filtering techniques [50]; and Delphi-Epicast: wisdom-of-crowds approach based on combining predictions submitted by several human participants [72] . Our past and ongoing forecasts , as well as Python [73] and R [46] code for components of the systems used to generate them , are publicly available online [74–76] . Changes made to Delphi-Stat throughout the 2015/2016 season are described in S5 Appendix . These three forecasting systems were ranked as the top three in the 2015/2016 comparison in terms of overall multibin score , with Delphi-Stat at the top . Fig 4 shows the performance of the three Delphi forecasting systems , broken down by evaluation metric and forecasting target . S1 and S2 Figs . show the multibin scores broken down by location and by forecasting week . Delphi-Stat had consistently strong aggregate multibin scores across different targets , locations , and forecasting weeks , and the best overall multibin log score of all FluSight 2015/2016 submissions . Delphi-Stat’s unibin log score evaluations relative to the other two Delphi systems seem similar to or better than the corresponding multibin log score evaluations , as Delphi-Stat has the best unibin score of the three for each target rather than just overall; this observation seems natural since Delphi-Stat was developed to optimize unibin log score , and may suggest that optimizing for multibin log score rather than unibin log score when selecting ensemble weights or as a post-processing step could produce multibin log score improvements . However , the system’s point predictions , while optimized for the mean absolute error metric , were less accurate than ( but still competitive with ) the other two when averaged across all predictions . Prospective forecast evaluation ensures that performance estimates are truly out-of-sample , not inflated by design decisions or model fits that are influenced by the evaluation data; however , such evaluation data is not readily generated , as it is expensive in terms of physical time: new wILI observations arrive once per week , and performance can vary significantly from season to season and from week to week . The evaluations from the 2015/2016 comparison may be noisy due to these season-to-season fluctuations . To address this issue , we use pseudo-out-of-sample retrospective analysis to provide more stable estimates of performance . Specifically , we use leave-one-season-out cross-validation: for each evaluation season s , we form and evaluate retrospective forecasts for s at every evaluation week using all training seasons except for s as inputs to the forecasting methods as if they were past seasons . ( We exclude seasons prior to 2010/2011 from the evaluation set because records of HHS region ILINet data revisions are only available beginning in late 2009 . We exclude seasons prior to 2003/2004 from the training set because year-round ILINet observations , which are required by some of the ensemble components , started in 2003 . The 2009/2010 season—containing the peak of the 2009 pandemic according to our adjusted definition of “season”—is also removed from the training set . Finally , we do not include the season currently underway ( S + 1 ) in evaluation or training as it has not been completely observed . ) Using cross-validation prevents most direct model fitting to evaluation data , and basing design decisions on motivations other than the effects on cross-validation evaluation helps limit fitting through iterative design . Fig 5 shows the distribution of log scores for several forecasting methods , described earlier in the text and in S1 Appendix , and the three ensemble approaches specified earlier in the text . Except for the uniform distribution and ensembles , all forecasting methods miss some possibilities completely , reporting unreasonable probabilities less than exp ( −10 ) ≈ 0 . 0000454 for events that actually occurred . In these situations , the log score has been increased to the cap of −10 ( as CDC does for multibin log scores ) . Delta and residual density forecasting methods ( Delta density , Markovian; Delta density , extended; and BR , residual density ) are less likely to commit these errors than other non-ensemble , non-uniform approaches , and have higher average log scores . Ensemble approaches combine forecasts of multiple components , missing fewer possibilities , and ensuring that a reasonable log score is obtained by incorporating the uniform distribution as a component . For the full Delphi-Stat ensemble , the main advantage of the ensemble over its best component appears to be successfully filling in possibilities missed by the best component with other models to avoid -10 and other low log scores appears , while for ensembles of subsets of the forecasting methods , there are other benefits; S3 Appendix shows the impact of these missed possibilities and the log score cap . Fig 5 also includes estimates of the mean log score for each method and rough error bars for these estimates . We expect there to be strong statistical dependence across evaluations for the same season and location , and weaker dependencies between different seasons and locations; thus , the most common approaches to calculating standard errors , confidence intervals , and hypothesis test results will be inappropriate . Properly accounting for such dependencies and calibrating intervals and tests is an important but difficult task and is left for future investigation . We use “rough standard error bars” on estimates of mean evaluations: first , the relevant data ( e . g . , all cross-validation evaluations for a particular method and evaluation metric ) is summarized into one value for each season-location pair by taking the mean of all evaluations for that season-location pair; we then calculate the mean and standard error of the mean of these season-location values using standard calculations as if these values were independent . Under some additional assumptions which posit the existence of a single underlying true mean log score for each method , these individual error bars—or rough error bars for the mean difference in log scores between pairs of ethods—suggest that the observed data is unlikely to have been recorded if the true mean log score of the extended delta density method were greater than that of the adaptively weighted ensemble , or if the true mean log score of the “Empirical Bayes A” method were greater than the extended delta density method . The mean and rough standard error estimates in Fig 5 also appear in tabular form in S4 Appendix . Methods that model wILI trajectories and “pin” past wILI to its observed values have a large number of log scores near 0 because they are often able to confidently “forecast” many onsets and peaks that have already occurred; ensemble methods also have a large number of log scores near 0 . Note that these scores are closer to 0 for ensembles that optimize weighting of different methods than for the ensemble with uniform weights . For this particular set of forecasting methods , targets , and evaluation seasons: the uniformly weighted ensemble has lower average log score than the best individual component ( extended delta density ) , using the stacking approach to assign weights to ensemble components improves ensemble performance significantly and gives higher average log score than the best individual component , the adaptive weighting scheme does not provide a major benefit over a fixed-weight scheme using a single set of weights for each evaluation metric . When given subsets of these forecasting methods as input , with regard to average performance: the uniformly weighted ensemble often outperforms the best individual , but is sometimes slightly ( ≈ 0 . 1 log score ) worse; the stacking approach improves upon the performance of the uniformly weighted ensemble; and the adaptive weighting scheme’s performance is equal to or better than that of the fixed-weight scheme , sometimes improving on the log score by ≈ 0 . 1 . The adaptive weighting scheme’s relative performance appears to improve with more input seasons , fewer ensemble components , and increased variety in underlying methodologies and component performance . These trends suggest that using wider RelevanceWeight kernels , regularizing the component weights , or considering additional data from 2003/2004 to 2009/2010 , for which ground truth wILI but not weekly ILINet reports are available , may improve the performance of the adaptive weighting scheme . In addition to these avenues for possible improvement in ensemble weights for the components presented in Fig 5 , the adaptive weighting scheme provides a natural way of incorporating forecasting methods that generate predictions for only a subset of all targets , forecast weeks , or forecast types ( distributional forecast or point prediction ) . For example , in the 2015/2016 season , we incorporated a generalized additive model that provided point predictions ( and later , distributional forecasts ) for peak week and peak height given at least three weeks of observations from the current season . Fig 6 shows a subset of the cross-validation data used to form the ensemble and evaluate the effectiveness of the ensemble method , for two sets of components: one using all the components of Delphi-Stat , and the other incorporating three of the lower-performance components and a uniform distribution for distributional forecasts . The Delphi-Stat ensemble near-uniformly dominates the best component , extended delta density , in terms of log score , and has comparable mean absolute error overall . The ensemble approach produces greater gains for the smaller subset of methods , surpassing not only its best components , but all forecasting methods in the wider Delphi-Stat ensemble except for the delta density approaches . Fig 7 shows cross-validation performance estimates for the extended delta density method based on three evaluation schemes: Ground truth , no nowcast: the ground truth wILI for the left-out season up to the forecast week is provided as input , resulting in an optimistic performance estimate; Real-time data , no nowcast: the appropriate wILI report is used for data from the left-out season , but no adjustment is made for possible updates; this performance estimate is valid , but we can improve upon the underlying method; Backcast , no nowcast: the appropriate wILI report is used for data from the left-out season , but we use a residual density method to “backcast” updates to this report; this performance estimate is valid , and the backcasting procedure significantly improves the log score; Backcast , Gaussian nowcast: same as “Backcast , no nowcast” but with another week of simulated data added to the forecast , based on a Gaussian-distributed nowcast; and Backcast , Student t nowcast: same as “Backcast , Gaussian nowcast” but using a Student t-distributed nowcast in place of the Gaussian nowcast . Backcast , ensemble nowcast: same as the previous two but using the ensemble nowcast ( which combines “no nowcast” with “Student t nowcast” ) . For every combination of target and forecast week , using ground truth as input rather than the appropriate version of these wILI observations produces either comparable or inflated performance estimates . Using the “backcasting” method to model the difference between the ground truth and the available report helps close the gap between the update-ignorant method . The magnitude of the performance differences depends on the target and forecast week . Differences in mean scores for the short-term targets are small and may be reasonably explained by random chance alone; the largest potential difference appears to be an improvement in the “1 wk ahead” target by using backcasting . More significant differences appear in each of the seasonal targets following typical times for the corresponding onset or peak events; most of the improvement can be attributed to preventing the method from assigning inappropriately high probabilities ( often 1 ) to events that look like they must or almost certainly will occur based on available wILI observations for past weeks , but which are ultimately not observed due to revisions of these observations . The magnitude of the mean log score improvement depends in part on the resolution of the log score bins; for example , wider bins for “Season peak percentage” may reduce the improvement in mean log score ( but would also shrink the scale of all mean log scores ) . Similarly , the differences in scores may be reduced but not eliminated by use of multibin scores for evaluation or ensembles incorporating uniform components for forecasting . Using the heavy-tailed Student t nowcasts or nowcast ensemble appears to improve on short-term forecasts without damaging performance on seasonal targets . The performance of the nowcast ensemble is further explored in S5 , S6 , S7 , S8 , S9 , S10 , S11 and S12 Figs . The Gaussian nowcast has a similar effect as the other nowcasters except on the “1 wk ahead” target that it directly predicts: its distribution is too thin-tailed , resulting in lower mean log scores than using the forecaster by itself on this target .
The delta density forecasting method , stacking-based adaptively weighted ensemble , distributional “backcasts” of wILI updates , and nowcasts from ILI-Nearby provide significant improvements upon other individual forecasting approaches that we considered . Promising avenues for further improvements include refining the methodology to rely less on arbitrary and heuristic feature , kernel , bandwidth , and parameter selections , regularization of ensemble weights , incorporating conditional density estimators from statistical literature , and using additional data sources and finer-resolution data models . | Seasonal influenza is associated with 250 000 to 500 000 deaths worldwide each year ( WHO estimates ) . In the United States and other temperate regions , seasonal influenza epidemics occur annually , but their timing and intensity varies significantly; accurate and reliable forecasts that quantify their uncertainty can assist policymakers when planning countermeasures such as vaccination campaigns , and increase awareness and preparedness of hospitals and the general public . Starting with the 2013/2014 flu season , CDC has solicited , collected , evaluated , and compared weekly forecasts from external research groups . We developed a new method for forecasting flu surveillance data , which stitches together models of changes that happen each week , and a way of combining its output with other forecasts . The resulting forecasting system produced the most accurate forecasts in CDC’s 2015/2016 FluSight comparison of fourteen forecasting systems . We describe our new forecasting methods , analyze their performance in the 2015/2016 comparison and on data from previous seasons , and describe idiosyncrasies of epidemiological data that should be considered when constructing and evaluating forecasting systems . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"density",
"influenza",
"applied",
"mathematics",
"simulation",
"and",
"modeling",
"algorithms",
"seasons",
"probability",
"distribution",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"materials",
"science",
"materials",
"physics",
"infectious",
"disease",
"control",
"information",
"technology",
"data",
"processing",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"epidemiology",
"mathematical",
"and",
"statistical",
"techniques",
"probability",
"theory",
"physics",
"infectious",
"disease",
"surveillance",
"earth",
"sciences",
"kernel",
"methods",
"disease",
"surveillance",
"physical",
"sciences",
"viral",
"diseases",
"material",
"properties",
"statistical",
"methods"
]
| 2018 | Nonmechanistic forecasts of seasonal influenza with iterative one-week-ahead distributions |
Zinc finger nucleases ( ZFNs ) have been used successfully to create genome-specific double-strand breaks and thereby stimulate gene targeting by several thousand fold . ZFNs are chimeric proteins composed of a specific DNA-binding domain linked to a non-specific DNA-cleavage domain . By changing key residues in the recognition helix of the specific DNA-binding domain , one can alter the ZFN binding specificity and thereby change the sequence to which a ZFN pair is being targeted . For these and other reasons , ZFNs are being pursued as reagents for genome modification , including use in gene therapy . In order for ZFNs to reach their full potential , it is important to attenuate the cytotoxic effects currently associated with many ZFNs . Here , we evaluate two potential strategies for reducing toxicity by regulating protein levels . Both strategies involve creating ZFNs with shortened half-lives and then regulating protein level with small molecules . First , we destabilize ZFNs by linking a ubiquitin moiety to the N-terminus and regulate ZFN levels using a proteasome inhibitor . Second , we destabilize ZFNs by linking a modified destabilizing FKBP12 domain to the N-terminus and regulate ZFN levels by using a small molecule that blocks the destabilization effect of the N-terminal domain . We show that by regulating protein levels , we can maintain high rates of ZFN-mediated gene targeting while reducing ZFN toxicity .
Homologous recombination is a natural mechanism that cells use for a variety of processes including double strand break ( DSB ) repair [1] . To repair a DSB by homologous recombination , the cell usually uses the sister chromatid as a donor-template but can use other pieces of DNA such as extrachromosomal DNA . Gene targeting uses homologous recombination to make a precise genomic change and is commonly used experimentally in a variety of cells including yeast and murine embryonic stem cells . However , the spontaneous rate of homologous recombination is too low in mammalian somatic cells ( 10−6 ) to be commonly used experimentally or therapeutically [2]–[5] . The rate of gene targeting , however , can be increased ( to over 10−2 ) by creating a gene specific DSB [2] , [6]–[10] . Zinc Finger Nucleases ( ZFNs ) can create site-specific DSBs and have been shown to increase the rate of gene targeting by over 5 orders of magnitude [11]–[13] . ZFNs are chimeric proteins that consist of a specific DNA binding domain made up of tandem zinc finger binding motifs fused to a non-specifc cleavage domain from the FokI restriction endonuclease ( the development of which is reviewed in [14] ) . By changing key residues in the DNA binding domain , ZFN binding specificity can be altered providing a generalized strategy for delivering a site-specific DSB . However , many ZFNs have been shown to have cytotoxic effects [2] , [15]–[17] . Several studies suggest that this toxicity is caused by “off-target” DSBs . For example , a zinc finger protein containing no nuclease domain was not toxic when transfected into HEK293 cells ( unpublished data ) . Similarly , Beumer et al . ( 2006 ) have shown that ZFNs containing point mutations to inactivate the nuclease domain do not exhibit cytotoxicity in flies [18] . There have been two published strategies for reducing the number of “off-target” breaks: ( 1 ) increase the specificity of the ZFN by protein engineering or ( 2 ) force heterodimerization of the ZFN pairs [16] , [19]–[24] . Here , we explore a third strategy to reduce cytotoxicity by small molecule regulation of ZFN protein levels . By creating ZFNs from zinc finger DNA binding domains that are more specific , toxicity is reduced . While on-target cutting is generated by heterodimerization of a ZFN pair at its target site ( at least 18 base pairs ) , off-target cutting can be mediated by either homodimer pairs or heterodimer pairs . Modifications in the nuclease to prevent homodimerization results in ZFNs with reduced toxicity [16] , [20] , [21] . We found , however , that this reduction can come at a cost of reduced activity in stimulating gene targeting [16] ( Wilson et al . , manuscript submitted ) . It has been shown that the rate of gene targeting can be increased , up to a point , by increasing the amount of transfected ZFN expression plasmids [16] , [18] . However , very high levels of ZFN expression cause toxicity without increasing targeting rates [16] , [18] . These observation lead to the hypothesis that reduced toxicity could also be obtained by being able to regulate ZFN expression . This “Goldilocks” phenomenon means that being able to titrate the amount of ZFN protein is critical to optimizing ZFN mediated gene targeting . Porteus and Baltimore demonstrated that maximal DSB-mediated gene targeting occurs within 60 hours of transfection of DNA [2] . Expression of ZFNs outside this window will increase toxicity without increasing targeting . We hypothesized that if we could narrow the time of ZFN protein expression , we could reduce toxicity while maintaining high rates of targeting . In this study , we use two previously described strategies to regulate protein levels and apply them to ZFNs . We show that by regulating protein levels , we reduce the number of “off-target” DSBs and reduce toxicity , while maintaining high ZFN-stimulated gene targeting activity .
The ability to regulate ZFN protein levels could theoretically give optimal rates of gene targeting with minimal toxicity . Degradation signals or “degrons” are specific domains that confer instability on a protein [25] . The N-end rule correlates the in vivo half-life of a protein to the N-terminal amino acid; some residues are destabilizing while other residues are stabilizing [26] , [27] . Normal N-terminal processing precludes simply adding a desired residue to the N-terminus of a protein . By adding a ubiquitin moiety ( Ub ) to the N-terminus of a protein , the N-terminal amino acid of a protein can be controlled . In eukaryotes , the Ub-X-POI ( where POI is protein of interest ) fusion is cleaved by Ub-specific processing proteases immediately before X ( where X is an amino acid residue ) [28] . This cleavage leaves the X residue as the N-terminal amino acid and thus affects protein stability . It has been established by several groups that an N-terminal arginine is a degradation signal [26] , [28] . It is also possible to create poorly cleavable or uncleavable Ub-X-POI fusions . If the ubiquitin protein is not cleaved from the POI , the protein can undergo ubiquitin fusion degradation [29] . That is , the Ub-X-POI fusion can be further ubiquitinated and thereby “marked” for degradation by the proteasome . This allows for another strategy to create short-lived POIs . By substituting the last residue of the ubiquitin moiety from glycine to valine and using a valine linker ( Ub-VV-POI ) , the ubiquitin moiety can no longer be cleaved from the POI [28] . We created a pair of Ub-VV-ZFNs and Ub-R-ZFNs fusion proteins ( Figure 1A ) to destabilize a pair of previously validated ZFNs targeting the GFP gene that contained the wildtype FokI domain [16] . The Ub-VV-ZFNs were made to take advantage of the potential destabilizing effect of a covalently linked N-terminal ubiquitin , and the Ub-R-ZFNs were made to take advantage of the potential destabilizing effect of an N-terminal arginine . Expression of the ZFN chimeras in transiently transfected HEK293 cells was examined by Western blot analysis ( Figure 1B ) . The size of the Ub-VV-ZFNs corresponded with the expected size of an uncleaved fusion protein . The size of the Ub-R-ZFNs corresponded with the size of the unmodified ZFNs , confirming that the ubiquitin moiety was cleaved . The addition of the proteasome inhibitor MG132 can increase the levels of Ub-X-POI fusion proteins [28] . We therefore examined the expression of ubiquitin linked ZFNs and unmodified ZFNs in the presence and absence of MG132 ( Figure 1B ) . The addition of the proteasome inhibitor had little effect on the unmodified ZFNs . In contrast , addition of MG132 to cells transfected with ubiquitin modified ZFNs produced a striking increase in the expression relative to the level of expression of the modified proteins in the absence of MG132 . It is interesting to note that even the untreated Ub-R-ZFNs had an increase in protein levels relative to the unmodified ZFNs ( Figure 1B , compare Ub-R-ZFNs , +MG132 at 24 hours to unmodified ZFNs at 24 hours ) . Although expression of Ub-R-ZFNs is higher than the unmodified ZFNs , the UB-R-ZFNs levels decrease more rapidly suggesting that the N-terminus arginine is , in fact , destabilizing . We hypothesize that the addition of the ubiquitin moiety to the N-terminus aided in protein folding of the ZFNs and thus produced higher expression . We compared the activity of the ubiquitin linked ZFNs to the unmodified ZFNs using a GFP gene targeting assay . In this assay , gene targeting is measured by the correction of a chromosomally integrated mutated GFP target gene [2] . We normalized the gene targeting rate for each condition to the rate obtained for the optimal amount of the unmodified ZFNs as previously determined [16] . We first compared activities of the Ub-VV-ZFNs and Ub-R-ZFNs with increasing amounts of DNA in the absence of proteasome inhibitor to that of the unmodified proteins ( Figure 2A and 2B ) . In the absence of drug , both the Ub-VV-ZFNs and the Ub-R-ZFNs , at all DNA concentrations tested , produced lower amounts of gene targeting compared to rates produced using the unmodified pair . The Ub-VV-ZFNs produced lower rates of gene targeting in the absence of drug compared to the Ub-R-ZFNs . We next evaluated the rate of gene targeting produced by the Ub-modified proteins in the presence of MG132 compared to when the drug was absent ( Figure 2C and 2D ) . Both the Ub-VV-ZFNs and Ub-R-ZFNs produced increased rates of gene targeting in the presence of MG132 compared to when the drug was absent . The rate of gene targeting produced by the Ub-VV-ZFNs in the presence of drug was not as high , however , as rates produced by the unmodified protein . In contrast , in the presence of drug , the Ub-R-ZFNs produced equivalent rates of gene targeting as compared to the unmodified proteins . In order to determine if the ubiquitin modification of these ZFNs reduced the cytotoxicity associated with unmodified ZFNs , we used a flow cytometry based cell survival assay ( the “toxicity assay” ) [16] . In this assay , we use a non-toxic endonuclease , I-SceI ( hereafter called Sce ) , as the standard for a non-toxic nuclease to which we normalize relative amounts of toxicity . The percent of surviving cells transfected with a potentially toxic nuclease is compared to the percent of surviving cells transfected with Sce . A lower percent of surviving cells is a sign of greater toxicity . As shown in Figure 2E , the percent survival relative to Sce of the unmodified ZFNs is about 50% . Both the Ub-VV-ZFNs and the Ub-R-ZFNs examined in this experiment produced lower toxicity and therefore a higher percentage of survival compared to the unmodified proteins . At 20 nanograms ( ng ) of Ub-R-ZFNs in the presence of drug , there was no observable toxicity in this assay . This is also the amount at which equivalent rates of gene targeting were obtained relative to the unmodified proteins ( Figure 2D ) . In summary , we found that the VV-linked versions minimized toxicity at the cost of reduced targeting efficiency . In contrast , using the R-linked versions , we could decrease toxicity without losing targeting efficiency . An alternative strategy to using ubiquitin involves linking a destabilization domain to the POI . This destabilization domain was engineered by making mutants of the FKBP12 protein , which is constitutively and rapidly degraded in mammalian cells [30] . Fusion of this destabilization domain to another protein confers instability to the fusion protein . In order to stabilize the protein , Banaszynski et al . ( 2006 ) , developed a synthetic ligand ( called Shield1 ) that binds the destabilization domain and protects the fusion protein from degradation [30] . We made a pair of chimeric proteins that linked the destabilization domain ( dd ) to the N-terminus of the ZFNs , containing the wildtype FokI domain , that target the GFP gene ( “dd-ZFNs” , Figure 3A ) . We examined the expression of the dd-ZFNs and unmodified ZFNs in transfected HEK293 cells by Western blot analysis ( Figure 3B ) . In the absence of Shield1 , the dd-ZFNs were destabilized as shown by reduced expression relative to the unmodified ZFNs . Upon addition of Shield1 for the first 24 hours post transfection , however , the dd-ZFNs were stabilized to relatively equivalent levels of expression as the unmodified ZFNs at 24 hours . The amount of protein expressed at 32 hours post transfection after drug treatment , however , was substantially reduced when compared to the unmodified ZFNs at the same time point ( Figure 3B ) . To examine the activity of the dd-ZFNs , we used the GFP gene targeting assay . At high concentrations of DNA , the rate of gene targeting stimulated by the dd-ZFNs in the absence of drug almost reached the rate stimulated by the unmodified ZFNs ( Figure 4A ) . Because of this high rate of targeting in the absence of drug , we chose to continue the experiments with 5 or 20 nanograms of transfected DNA . We conducted a series of experiments to characterize the timing and dosing of the drug in order to determine the drug conditions needed to obtain optimal rates of gene targeting . After 24 hours of exposure to Shield1 , the rate of gene targeting induced by the dd-ZFNs is equivalent to the rate stimulated by the unmodified ZFNs ( Figure 4B ) . We found that additional exposure to the drug , beyond 24 hours , did not further increase these rates ( data not shown ) . We next evaluated the dosing of the Shield1 drug with respect to gene targeting . At 1000 nM of Shield1 , we observed equivalent rates of gene targeting , but there was a dose-dependent decrease in targeting as the dose was lowered ( Figure 4C ) . With Shield1 present at 1000 nM for the first 24 hours , we observed that using either 5 or 20 nanograms of the dd-ZFNs could produce rates of gene targeting equivalent to the optimal rates obtained with the unmodified ZFNs in HEK293 cells ( Figure 4D ) . To determine if this method could be used in other cell types , we measured the gene targeting rates in HeLa and 3T3 cells stably transfected with the GFP gene targeting system . As shown in Figure 4E and 4F , the addition of Shield1 to cells transfected with the dd-ZFNs resulted in an increase in the rate of gene targeting relative to when the drug was absent . In the presence of Shield1 , the rates of gene targeting in both the 3T3 cells and HeLa cells at 20 ng were equivalent to the rates produced using the unmodified ZFNs at 20 ng with no drug treatment ( Figure 4E and F ) . To determine if linking the destabilization domain to the ZFNs reduced the cytotoxicity associated with the unmodified ZFNs , we used the toxicity assay . Strikingly , in the presence of drug at 5 or 20 ng of dd-ZFNs , toxicity relative to Sce appears to be negligible , and there is a significant reduction in toxicity compared to the unmodified ZFNs at 20 ng ( Figure 4G ) . We found that ZFNs with a N-terminal FKBP12 domain that is not destabilizing have greater toxicity than the dd-ZFNs , suggesting that the decreased toxicity is not simply the result of improved protein folding ( data not shown ) . Previous studies have suggested that the cytotoxicity associated with unmodified ZFNs is due to the creation of off-target DSBs [16]–[18] . When a DSB occurs , a signaling cascade is activated including the phosphorylation of H2AX and the recruitment of an array of proteins , including 53BP1 , to the site of the DSB that can be detected as foci by immunofluorescence [16] , [31] . We have previously shown that ZFNs that produce larger numbers of foci are more toxic than ZFNs that produce fewer foci [16] . Although the unmodified ZFN pair used in this study shows cytoxicity in the toxicity assay , this pair did not show an increased number of foci per cell relative to Sce when this assay was performed in human foreskin fibroblasts . To sensitize the assay , we used cells mutated in Ku80 , a gene important in the non-homologous end-joining pathway of DSB repair , which are known to have delayed repair of DSBs [32] . In this cell line , GFP transfected cells and cells transfected with Sce alone had an average of about 4 foci per cell ( Figure 5 ) . As a further control , we transfected cells with a plasmid encoding Caspase Activated DNAse ( CAD ) , an endonuclease that cleaves DNA non-specifically . CAD-transfected cells had an average of about 12 foci per cell ( Figure 5 ) . To aid in our comparison of the unmodified ZFNs and dd-ZFNs , we used higher amounts of DNA than determined in Figure 3 in order to amplify the number of DSBs visualized . We did however maintain the 1∶4 ratio ( 5 ng∶20 ng vs . 75 ng∶300 ng ) of dd-ZFN DNA concentration with respect to unmodified ZFN DNA concentration for this comparison . Cells transfected with the unmodified ZFNs had an average of 10 foci per cell ( comparable to the CAD transfected cells , Figure 5B ) . In contrast , the dd-ZFN transfected cells had only about 4 foci per cell ( comparable to the GFP-alone and Sce transfected cells ) . In summary , linking the destabilization domain of a modified FKBP12 protein to the N-terminus of ZFNs resulted in a way to regulate the expression level of the ZFNs that maintained high rates of gene targeting while minimizing toxicity .
Homologous recombination is the most precise way to manipulate the genome and is a powerful experimental tool in several different systems . ZFNs have been shown to increase the rate of gene targeting in a wide variety of experimental systems previously not amenable to genome manipulation by homologous recombination [2] , [10] , [13] , [17] , [23] , [24] , [33] . In addition to the problem of designing ZFNs to recognize target sites [34] , another limitation has been concern about off-target effects [2] , [15]–[17] . Improvements in toxicity have been attained by increasing the specificity of ZFNs and by modifications of the nuclease domain [16] , [19]–[24] , [35] . Further strategies to minimize ZFN toxicity , however , could further broaden the window between the desired and undesired genomic effects of ZFNs . In whole organisms such as flies and zebrafish , high levels of ZFN expression led to abnormal developmental mutations [18] , [24] , [33] . Reducing ZFN toxicity by regulating ZFN expression could hypothetically help attenuate these abnormalities . In this work , we show that small molecule regulation of ZFN expression can result in an improved toxicity profile without sacrificing gene targeting activity . The use of the destabilization domain may not be necessary when making gene modified cell lines ( where one can characterize a single clone ) but instead will be useful when treating a large population of cells that may be infused into a patient ( as would be done in gene therapy ) where isolation of a single clone is either not feasible or desirable . The standard strategy to control protein levels is to use transcriptional based methods ( examples include the TetOn or TetOff: Clonetech , Ponasterone System; Stratagene , and Dimerizer System; Ariad ) . Gene targeting induced by ZFNs is already a three-component system ( ZFN-1 , ZFN-2 , and a repair/donor molecule ) . Adding an inducible transcriptional regulator as a fourth component to make the system more complex was not desirable , particularly as the technology moves into cell types that are more difficult to transfect or infect . The ERT2 domain , a modified ligand binding domain from the estrogen receptor , has been successfully used to control protein activity by modulating the location of the protein . Unfortunately , we found that attaching the ERT2 domain to ZFNs did not stimulate gene targeting with presence of tamoxifen ( data not shown ) . An alternative strategy is to use a post-translational method of regulating ZFN level . In this strategy , a destabilized ZFN is created by adding a destabilizing domain and then levels of ZFNs are controlled by adding a small molecule to block the destabilization effects . By fusing a ubiquitin domain to the N-terminus through a non-cleavable linker ( Ub-VV-ZFN ) , we made ZFNs that could be regulated by proteasome inhibition , which resulted in decreased toxicity . When we fused the ubiquitin domain to the N-terminus of the ZFN through a cleavable linker leaving a destabilizing arginine at the N-terminus , we created ZFNs that were regulated by proteasome inhibition resulting in decreased toxicity and maintained high rates of gene targeting . Because proteasome inhibitors such as bortezomib are FDA approved for use in humans , this strategy has long-term promise . We did find the window of exposure to MG132 , the proteasome inhibitor used in this study , in which we got good induction without cytotoxicity was narrow . Finally , when we fused a modified FKBP12 domain to the N-terminus of the ZFN , we created ZFNs that were regulated by the small molecule Shield1 , which resulted in reduced ZFN toxicity and maintained high rates of targeting . Despite using amounts of Shield1 for prolonged periods ( up to 48 hours ) , we did not observe any discernable toxicity . Moreover , by expression microarray analysis , Shield-1 has almost no effect on gene expression [36] . Thus , the Shield1/FKBP12 system may ultimately be the better system despite Shield1 not being currently FDA approved for use in humans . Regulating ZFN expression also gave insight into the kinetics of gene targeting . Previously , Porteus and Baltimore found that maximal gene targeting was measured at 60 hours after transfection [2] . In this work , we demonstrate that ZFNs need only be expressed for less than 32 hours after transfection to attain maximal gene targeting ( measured at 72 hours post-transfection ) . These experiments define a window for ZFN expression , here defined as 0–32 hours but perhaps even shorter , in which expression of ZFNs beyond the window does not increase targeting but does increase toxicity . We have no explanation for the 32-hour window for gene targeting based on experimental data . One would expect , for example , that gene targeting events should increase as long as ZFNs are present , but we do not observe this [2] . A hypothesis is that the repair substrate/donor may not be available ( for example dilution , sequestration , or modification ) for the repair of a double strand break by homologous recombination after this window . This hypothesis will have to be experimentally tested in the future . Previously , we used human diploid fibroblasts to measure 53BP1 foci created by off-target DSBs . The unmodified ZFNs used in this study did not show significantly increased numbers of foci in that cell line , presumably because the cells were efficient at repairing DSBs . To sensitize the assay , we used murine Ku80−/− cells that are deficient in DSB repair . Using these sensitized cells resulted in a higher number of background 53BP1 foci , but also allowed us to detect subtle differences in ZFN toxicity between dd-ZFNs and untagged ZFNs that we could not detect using cells that were not deficient in DSB repair . As ZFNs continue to improve , the use of sensitized assays to quantitate improvements will be an important strategy . We have utilized strategies in which a drug stabilizes the protein rather than a drug to destabilize the protein . This “drug-on” system has several advantages . First , it means that the drug only needs to be administered for a brief period ( the window to maximize gene targeting activity ) . This brief administration is advantageous because it is cheaper and because it minimizes the potential side-effects of the drug itself . Second , when the drug is absent , the ground state of ZFN expression will be low , thus reducing the potential side-effects of the ZFNs . In summary , we have found that small molecule regulation of ZFN expression is an effective way to reduce cytotoxicity without compromising targeting efficiency . This strategy may be particularly beneficial to using ZFN mediated genome modification in a wide variety of cell types , including human stem cells .
All plasmids were made using standard cloning techniques and molecular biology as previously described [37] . The unmodified ZFNs were selected by the B2H design strategy and fused to the wildtype FokI nuclease domain as described earlier and called “GFP1 . 4-B2H” and “GFP2-B2H” [16] . For the Ub-X-ZFN versions , the ubiquitin open reading frame was amplified by PCR from pUb-R-GFP [28] with sense primer 5′-ACTGGGATCCTCTAGATCCACCATGCAGATCTTCGTGAAG-3′ and the antisense primers 5′-ACTGGGATCCAAGCTTCCCCACCACACCTCTGAGACGGAGTAC-3′ for the Ub-VV-ZFNs , or 5′-ACTGGGATCCAAGCTTCCCTCTGCCACCTCTGAGACGGAGTAC-3′ for the Ub-R-ZFNs ( restriction site underlined , variable codons in bold ) and cloned into the ZFN expression plasmid using the BamHI site . Directionality was determined by XbaI digest . To create the dd-ZFNs , the L106P destabilization domain was PCR amplified using primers 5′-ACGTGCGGCCGCACCATGGGAGTGCAGGTGGAAACCATCTCC – 3′ and 5′-ACTGGGATCCTTCCGGTTTTAGAAGCTCCAC-3′ . The resulting fragment was digested with NotI and BamHI and cloned in-frame to the N-terminus of the GFP-ZFNs in a CMV expression vector . For all constructs the N-terminal domains and junctions were confirmed by sequencing . All cell culture experiments were performed in HEK293 cells except where identified . Cells were cultured in a humidified incubator at 37°C with 5% CO2 in DMEM supplemented with 10% bovine growth serum ( Hyclone , Logan , UT , USA ) , 2 mM L-glutamine , 100 IU/ml penicillin , and 100 mg/ml streptomycin . Stable cell lines were constructed as previously described [15] . Transient transfections were performed using the calcium phosphate technique as previously described and produced transfection efficiencies between 10–35% [38] . For experiments using MG132 ( carboxybenzyl-leucyl-leucyl-leucinal; Sigma-Aldrich , St . Louis , MO ) , 10 uM drug was added to cells from 18–22 hours post-transfection unless otherwise noted . We determined this window and concentration of the proteasome inhibitor empirically to maximize stimulation of gene targeting while minimizing toxic effects of the drug ( data not shown ) . For experiments using Shield1 ( Clontech , Mountain View , CA ) , 1000 nM drug was added to cells from time of transfection and left on for 24 hours unless otherwise noted . As discussed in the results , we determined the dose and timing of the drug empirically to maximize gene targeting activity and minimizing toxicity . Gene targeting experiments were performed in triplicate as previously described using calcium phosphate transfection [15] . Transfection efficiencies were determined at day 2 post-transfection , and the rates of gene targeting were determined by flow-cytometry and analyzed on a FACS Calibur ( Becton-Dickinson , San Jose , CA , USA ) at day 3 ( day of transfection is considered day 0 ) . Gene targeting rates are calculated as GFP positive cells per million cells transfected because the background rate of spontaneous gene targeting using this system is approximately one event per million cells . Gene targeting rates are then normalized to the percent gene targeting obtained using 20 ng of ZFN-1 and ZFN-2 as these conditions have given the highest rates of gene targeting for the unmodified proteins . The absolute rate of gene targeting using ZFN-1/ZFN-2 at 20 ng in HEK293 cells was about 20 , 000 GFP positive cells per million cells transfected . Both a HeLa and 3T3 cell line were created using electroporation that stably incorporated the GFP gene targeting system . Gene targeting experiments were performed in triplicate as previously described [15] . Lipofectamine 2000 Reagent ( Invitrogen ) was used to transfect cells using Invitrogen's suggested protocol . pcDNA6/V5-HisA plasmid DNA was added as stuffer DNA when necessary to raise the total DNA to 800 ng per well . 1000 nM Shield1 was added to drug-treated wells at the time of transfection . 24 hours later , Shield1 was removed and the medium was replaced with fresh , supplemented Dulbecco's Modified Eagle's Medium . The absolute rate of gene targeting using ZFN-1/ZFN-2 at 20 ng in HeLa cells stably transfected with GFP gene targeting reporter was about 2 , 000 . The absolute rate of gene targeting using ZFN-1/ZFN-2 at 20 ng in 3T3 cells stably transfected with the GFP gene targeting reporter was about 20 , 000 . For time course blots , cells were harvested at indicated times post-transfection . Each sample was counted and lysate volumes were adjusted to give equal amounts of cells per volume . Equal amounts of total lysates were subjected to SDS-PAGE , wet transferred to PVDF membranes and incubated with specific antibodies . ZFNs were detected using an anti-Flag M2 monoclonal antibody ( 1∶10 , 000 , Sigma-Aldrich ) , and β-actin was detected using a rabbit anti-actin antibody ( 1∶5 , 000 , Sigma-Aldrich ) . The blots were further incubated with HRP-conjugated secondary antibodies and visualized using Western blotting luminal reagent ( Santa Cruz Biotechnology , Santa Cruz , CA ) . Toxicity assays were performed as previously described [16] . Briefly , HEK293 cells were transfected in triplicate by calcium phosphate technique with 200 ng of a GFP expression plasmid and with varying amounts of each nuclease expression plasmid ( two plasmids total ) . At day two post-transfection , a fraction of transfected cells was analyzed by flow-cytometry and the percentage of GFP positive cells was determined . At day six post-transfection , the percentage of GFP positive cells was determined by flow-cytometry . To calculate the percent survival relative to Sce , a ratio of ratios was calculated as previously described [16] . The ratio after nuclease transfection was normalized to the ratio after Sce transfection and this determined the percent survival compared to Sce . In control experiments , we showed that Sce expression had no effect on cell survival compared to cells transfected with an empty expression vector . Cell Culture: Ku80−/− mouse 3T3 cells were maintained in Dulbecco's Modified Eagle's Medium ( Hyclone , Logan , UT ) supplemented with 20% fetal calf serum and 2 mmol/l L-glutamine . The cells were maintained in a humidified incubator with 5% CO2 at 37°C . Transfection of 3T3 Ku80−/− cell line: Mouse 3T3 cells that are Ku80−/− were used in these studies because they repair DNA breaks more slowly , providing a more sensitive assay for monitoring DNA damage . Ku80−/− cells were seeded in 4-well Lab-Tek II Chamber Slides ( Nalge Nunc , Rochester , NY ) at 40 , 000 cells per well . 24 hours later , cells in each well were lipofected with 200 ng GFP DNA and 75 ng or 300 ng of each nuclease . pcDNA6/V5-HisA plasmid DNA was added as stuffer DNA when necessary to raise the total DNA to 800 ng per well . Lipofectamine 2000 Reagent ( Invitrogen ) was used to transfect cells using Invitrogen's suggested protocol . 1000 nM Shield1 was added to drug-treated wells at the time of transfection . 24 hours later , Shield1 was removed and the medium was replaced with fresh , supplemented Dulbecco's Modified Eagle's Medium . 48 hours after lipofection the cells were fixed , stained and visualized . 53BP1 foci were counted only in cells that were brightly GFP positive because these were the ones transfected with the GFP and the nuclease ( s ) . Immunofluorescence staining was carried out as performed in [16] . Briefly , cells were washed in phosphate buffered saline , fixed in cold 4% paraformaldehyde , washed again , and then permeabilized with . 5% Triton X-100 . Cells were re-washed , blocked in 5% bovine serum albumin , and then incubated with rabbit anti-53BP1 antibody ( Cell Signaling , Danvers , MA ) . After another set of washes , cells were incubated with goat anti-rabbit Rhodamine Red-X antibody ( Invitrogen , Carlsbad , CA ) . Cells were washed again and then mounted in Vectashield mounting medium containing 4 , 6-diamidino-2-phenylinodole ( Vector Laboratories , Burlingame , CA ) . Images were captures using an epifluorescence microscope equipped with a Q-Fire charge-coupled device camera ( Olympus America , Melville , NY ) and QCapture Software ( QImaging , British Columbia , Canada ) . Images were merged using ImageJ Software ( NIH , ver . 1 . 40 g ) . | Zinc finger nucleases ( ZFNs ) are a powerful tool to create site-specific genomic modifications in a wide variety of cell types and organisms and are about to enter human gene therapy clinical trials . An important aspect of using ZFNs for use in gene therapy is to minimize off-target effects . We made ZFNs that contain destabilizing domains on their amino-terminus . The expression level of the modified ZFNs could be increased transiently by the addition of a small molecule , either a proteasome inhibitor or Shield1 . We demonstrate that off-target effects can be reduced without compromising gene targeting efficiency by using small molecules to limit the maximal expression of the ZFNs to a narrow window . The ability to regulate ZFN expression using small molecules provides a new strategy to minimizing off-target effects of ZFNs and may be an important way of ultimately using ZFNs for clinical use in gene therapy protocols . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biotechnology",
"genetics",
"and",
"genomics/gene",
"therapy",
"genetics",
"and",
"genomics",
"genetics",
"and",
"genomics/gene",
"expression"
]
| 2009 | Attenuation of Zinc Finger Nuclease Toxicity by Small-Molecule Regulation of Protein Levels |
Prevention of viral-induced respiratory disease begins with an understanding of the factors that increase or decrease susceptibility to viral infection . The primary receptor for most adenoviruses is the coxsackievirus and adenovirus receptor ( CAR ) , a cell-cell adhesion protein normally localized at the basolateral surface of polarized epithelia and involved in neutrophil transepithelial migration . Recently , an alternate isoform of CAR , CAREx8 , has been identified at the apical surface of polarized airway epithelia and is implicated in viral infection from the apical surface . We hypothesized that the endogenous role of CAREx8 may be to facilitate host innate immunity . We show that IL-8 , a proinflammatory cytokine and a neutrophil chemoattractant , stimulates the protein expression and apical localization of CAREx8 via activation of AKT/S6K and inhibition of GSK3β . Apical CAREx8 tethers infiltrating neutrophils at the apical surface of a polarized epithelium . Moreover , neutrophils present on the apical-epithelial surface enhance adenovirus entry into the epithelium . These findings suggest that adenovirus evolved to co-opt an innate immune response pathway that stimulates the expression of its primary receptor , apical CAREx8 , to allow the initial infection the intact epithelium . In addition , CAREx8 is a new target for the development of novel therapeutics for both respiratory inflammatory disease and adenoviral infection .
Adenoviruses ( AdV ) are a common cause of upper and lower respiratory tract infections . Although most AdV infections are self-resolving , some may lead to acute respiratory distress syndrome , a serious and frequently fatal respiratory condition [1 , 2] . Epidemic AdV infections occur in closed communities , among children , and military recruits , and are most severe , often lethal , in immunosuppressed individuals [1–3] . In addition , AdV is frequently associated with exacerbation of inflammatory airway diseases such as asthma , cystic fibrosis ( CF ) , and chronic obstructive pulmonary disease ( COPD ) [4–7] . No specific therapeutics exist to treat or prevent AdV infection; thus , the discovery of novel strategies to limit viral infection in susceptible populations would be an important advancement . Human AdV is a non-enveloped double-stranded DNA virus that can be grouped into seven species ( A through G ) , with >60 types identified [2 , 8] . All species , except group B , use the coxsackievirus and adenovirus receptor ( CAR ) as a primary receptor for cell attachment via the AdV fiber knob ( FK ) [9–12] . In polarized epithelial cells , CAR is found below the tight junction seal that separates the air-exposed apical surface from the basolateral surface [13] . Until recently , it was believed that AdV must breach the epithelial tight junction barrier to access CAR and initiate viral infection in the lungs [13] . It is now known that CAR has another transmembrane isoform that is able to localize at the apical surface of polarized airway epithelia and mediate AdV infection [14–16] . Whereas the basolateral isoform is composed of the first seven exons of the human CXADR gene ( CAREx7 or hCAR1 ) , the apical isoform occurs via splicing from a cryptic site within the seventh exon to the eighth and final exon ( CAREx8 ) . The two nearly identical proteins vary only in the last 26 ( CAREx7 ) or 13 aa ( CAREx8 ) of the proteins . The abundance of apical CAREx8 and the amount of AdV infection are tightly regulated by the cellular scaffold protein MAGI-1 and are increased by side-stream tobacco smoke [15 , 16] . Determining other cellular and environmental factors that regulate CAREx8 will provide insight into what controls the susceptibility of the host epithelium within an individual to viral infection . The factors that predispose both healthy and immunocompromised individuals to AdV infection are complex , and likely related to the co-evolution of the host and pathogen . Similar to many other proinflammatory pathogens , AdV is a proinflammatory virus that can stimulate the secretion of proinflammatory cytokines , including interleukin-8 ( IL-8 ) , by airway macrophages and the epithelial cells within the lung epithelium [17 , 18] . IL-8 exposure in turn favors AdV infection of the airway epithelium [17] . How the proinflammatory cytokines enhance AdV infection remains unclear . IL-8 is a potent neutrophil chemoattractant that initiates transepithelial migration . Previous studies have shown that basolateral CAREx7 interacts with a neutrophil surface protein , junctional adhesion molecule-like protein ( JAML ) , and that blocking the interaction interferes with the efficiency of neutrophil transmigration [19] . The extracellular domain of CAR binds to JAML , and since the extracellular domain of CAREx7 and CAREx8 are identical , this suggests that CAREx8 might also bind to neutrophils via JAML . We hypothesized that the apical expression of CAREx8 is stimulated by IL-8 in order to function as a receptor that tethers neutrophils at the apical surface of epithelia . We further hypothesized that AdV may have co-opted this potential innate immune function of CAREx8 in order to facilitate AdV entry from the apical surface of a polarized epithelium . Finally , considering that neutrophils mainly target bacterial pathogens and antibody or complement bound molecules [18] , we hypothesize that IL-8 and neutrophils contribute to AdV infection . Consistent with this , AdV is frequently isolated from patients with inflammatory respiratory diseases . In this study , we show for the first time that IL-8 increases the protein synthesis and apical localization of CAREx8 in polarized cells via activation of the AKT/S6K pathway and inactivation of GSK3β . Apical CAREx8 tethers infiltrating neutrophils on the apical surface of polarized epithelia , a novel biological function of CAREx8 , and adherent neutrophils at the apical surface enhance AdV infection . Taken together , AdV uses the host innate immune response , triggered by either invading microbes or other IL-8 stimulants entering the airway , to facilitate entry into host cells . Understanding the intricate interplay between the host innate immune system and different types of pathogens is critical in order to develop targeted therapies that prevent infection and disease progression .
To investigate the effect of IL-8 on AdV infection in polarized epithelia , we first used polarized Calu-3 airway epithelial model cells . Polarized Calu-3 epithelia were treated with increasing concentrations of IL-8 ( 0–100 ng/ml ( 0–12 . 5 nM ) ) for 4 h , followed by apical infection with recombinant , replication-defective , AdV type 5 ( AdV5 ) . Quantitative PCR ( qPCR ) analysis for AdV5 genomes ( Vg ) was performed by determining the copy number of the AdV5 hexon gene relative to a cellular housekeeping gene after DNA extraction . QPCR showed that AdV entry was increased in response to IL-8 treatment in a dose-dependent manner ( Fig . 1A ) . Viral entry reached its maximum and plateaued at 3 , 10 and 30 ng/ml of IL-8 , with ∼5-fold increase in Vg when compared to control ( 0 ng/ml IL-8; p<0 . 05 ) , followed by a decrease at 100 ng/ml . However , there was no significant change in the transepithelial resistance ( TER ) indicating that the effect of IL-8 on viral entry was not due to decreased integrity of the epithelial junctions ( Fig . 1B ) . These data suggest that the increase in epithelial susceptibility to AdV entry upon IL-8 exposure may be due to specific cellular effects , such as increased primary receptor expression at the apical surface of the polarized epithelium . The primary receptor for AdV5 is CAR; therefore , we investigated the expression of CAR in the presence of IL-8 . In particular , we examined the expression of the apical isoform of CAR , CAREx8 , since it is known to be present at the air-exposed surface of airway epithelia [14] . Polarized Calu-3 cells were treated with IL-8 at varying concentrations ( Fig . 1C , D ) and for varying time points ( Fig . 1E ) . IL-8 increased the expression of CAREx8 in both a concentration and time-dependent manner . IL-8 had its maximal effect on CAREx8 expression at 30 ng/ml ( Fig . 1C , D ) . By contrast , IL-8 did not affect the amount of total CAR , which is predominantly composed of the basolaterally-sorted CAREx7 isoform [14 , 15] . IL-8 also did not affect the junction-adhesion protein E-cadherin or actin ( loading control ) ( Fig . 1C ) . Consistent with increased CAREx8 protein levels in lysates , apical-surface specific biotinylation assays showed that CAREx8 localization at the apical surface increased in response to IL-8 in a dose-dependent manner , with a maximum increase at 30 ng/ml ( Fig . 1D ) . By contrast , no biotinylated cytosolic actin was detected . IL-8 had its maximal effect on CAREx8 protein expression between 4–12 h at 30 ng/ml and returned to baseline levels within 24 h ( Fig . 1E ) . To investigate acute effects on the epithelium , further experiments were carried out with 30 ng/ml of IL-8 for 4 h . Well-differentiated primary airway epithelia obtained from healthy human donors were used to validate the results found in Calu-3 epithelia . Similar to Calu-3 epithelia , apical treatment with 30 ng/ml of IL-8 for 4 h resulted in a robust increase in CAREx8 protein expression ( Fig . 1F ) and apical localization ( Fig . 1G ) . As expected , the protein levels of the basolateral junctional-adhesion protein E-cadherin and cytosolic actin did not change ( Fig . 1F ) and were not detected upon apical surface-specific biotinylation ( Fig . 1G ) . Neutrophils at the apical surface of an epithelium play a critical role in pathogen clearance [20] . Since CAREx8 protein expression in polarized epithelia is activated by the neutrophil chemoattractant IL-8 and since the extracellular region of epithelial CAREx7 that binds JAML on the surface of neutrophils is identical to CAREx8 , we hypothesized that CAREx8 has a role in tethering neutrophils to the apical surface of the epithelium . To test this hypothesis , polarized primary human airway epithelia ( Fig . 1H and I ) or polarized Calu-3 cells ( S1A–S1B Fig ) were pre-stimulated with IL-8 . Neutrophils isolated from the healthy human donors were fluorescently labelled and added to the apical surface of the epithelia for a neutrophil adhesion assay . Neutophil binding was determined by total fluorescence integrated density over 5–10 images per condition . IL-8 treatment resulted in a significant 2–2 . 5 fold increase in the adhesion of primary neutrophils . Neutrophil adhesion was completely blocked when the IL-8 pre-stimulated polarized epithelia were pre-treated with AdV5 fiber knob ( FK ) , the capsid protein that binds to the extracellular domain of CAR with high affinity ( Fig . 1H and I ) . These data indicate that CAREx8 is an important component of IL-8 stimulated apical neutrophil adhesion . To confirm that CAREx8 is responsible for increased AdV5 transduction and that CAREx8 tethers neutrophils at the apical surface of polarized epithelia , model epithelial cells stably expressing CAREx8 under a Doxycycline ( DOX ) inducible promoter were generated . Control cells stably expressing CAREx7 or mCherry were also generated from the same parental Tet-on MDCK cell line . MDCK cells were chosen because these cells are well characterized , grow quickly , and polarize rapidly into an epithelium with an expected distribution of cellular proteins [21–23] . Polarized epithelia from cell lines derived from single-cell clones with stable integration of FLAG-tagged CAREx8 , FLAG-tagged CAREx7 , or mCherry , under the DOX sensitive PTight promoter were characterized and compared in the absence of DOX . Clones were selected that had similar growth and polarization characteristics , including the ability to form tight junctions , distribution of apical , basolateral , and tight junction proteins , and polarity of baseline AdV5 transduction . MDCK cells stably expressing mCherry , FLAG-tagged CAREx8 , or FLAG-tagged CAREx7 , demonstrated a DOX-dose dependent increase of mCherry fluorescence ( Fig . 2A ) , or CAREx8 or CAREx7 protein levels relative to actin ( Fig . 2B ) . To confirm the polarity of protein expression with the polarized MDCK epithelium , apical surface-specific biotinylation was performed . In contrast to CAREx7 in MDCK-CAREx7 epithelia , CAREx8 protein was detected at the apical surface of MDCK-CAREx8 epithelia at low doses of DOX and expression was saturated above 100 ng/ml of DOX ( Fig . 2C ) . To characterize the susceptibility of the MDCK stable cell lines to AdV infection , cells were polarized and infected with AdV5-β-Gal from the apical surface . Data from quantitative PCR ( viral genomes , Vg; Fig . 2D ) and transduction ( β-Gal expression; Fig . 2E ) showed a dose-dependent increase in adenoviral entry into MDCK-CAREx8 epithelia exposed to low levels of DOX , which was not observed in MDCK-CAREx7 and MDCK-mCherry epithelia . These data show that apical AdV entry and transduction is highly sensitive to the induction of apical CAREx8 expression in polarized epithelia . Consistent with the findings by Western blot ( Fig . 2B , C ) , the MDCK-CAREx8 epithelia demonstrated a plateau in viral genome entry and transduction above 100 ng/ml DOX treatment , suggesting that there may be cellular limits to the amount of CAREx8 expressed within a cell , the amount of CAREx8 available at the apical surface , or limitations to viral entry at the apical surface . Since IL-8 induces CAREx8 protein expression and increases neutrophil retention at the apical surface of polarized epithelia ( Fig . 1 ) , we hypothesized that induction of CAREx8 protein expression in the absence of IL-8 would be sufficient to increase the binding of neutrophils at the apical surface of polarized epithelia . To test this , polarized MDCK-CAREx8 , -CAREx7 and-mCherry epithelial cells were induced with increasing concentrations of DOX for 24 h and a neutrophil adhesion assay was performed . Increasing apical CAREx8 protein levels in MDCK-CAREx8 epithelia correlated directly with increased neutrophil adhesion on the epithelial cell surface ( Fig . 3A ) . By contrast , MDCK-CAREx7 and-mCherry DOX-induced epithelia only showed baseline neutrophil adhesion ( Fig . 3B , C ) . These data suggest that CAREx8 is able to tether neutrophils at the apical epithelial cell surface . To confirm that this was a CAREx8-mediated effect , purified AdV5 fiber knob ( FK ) , which has a 500–1000 fold higher affinity for the overlapping CAR-JAML or CAR-CAR binding site [24–26] , was used to compete with the putative interaction between epithelial apical CAREx8 and neutrophil JAML . AdV5 FK decreased neutrophil adhesion in a dose-dependent manner in both mock and DOX-induced MDCK-CAREx8 cells , including a complete block of neutrophil adhesion at the highest concentration of AdV5 FK ( Fig . 3D ) . In contrast , FK from AdV3 , a group B AdV that does not use CAR as a primary receptor [27] , did not block neutrophil adhesion ( Fig . 3D ) . Taken together , these data show that CAREx8 tethers neutrophils at the apical epithelial cell surface and that AdV may potentially be able to out-compete neutrophils to bind apical CAREx8 . Next , we sought to determine the fate of neutrophils that transmigrate from the physiologically relevant basal surface to the apical surface in the presence or absence of DOX-induced CAREx8 . To do this , fluorescently-labeled neutrophils were added to the basal surface of epithelia and stimulated to transmigrate to the apical surface by adding the neutrophil chemoattractive bacterial peptide fMLP to the apical surface . Two populations of cells were quantified: 1 ) neutrophils that transmigrated through but remained adhered to the apical epithelial surface and 2 ) neutrophils that completely transmigrated through and detached from the epithelium ( Fig . 4A , B respectively ) . DOX-induced MDCK-CAREx8 epithelia retained ∼3 times as many transmigrated neutrophils on the apical surface as compared to uninduced MDCK-CAREx8 epithelia , and MDCK-CAREx7 , or-mCherry epithelia regardless of DOX-induction or not ( Fig . 4A ) . In contrast , ∼3 times as many neutrophils transmigrated through induced MDCK-CAREx7 epithelia relative to all other conditions ( Fig . 4B ) . This is consistent with the known role for basolateral CAREx7 in facilitating neutrophil transepithelial migration [19] . These data confirm that CAREx8 is able to enhance the adhesion of transmigrating neutrophils at the apical surface and also indicate that each CAR isoform plays a distinct role in neutrophil recruitment . Neutrophils are part of the innate immune system and the first cells recruited to sites of injury or pathogenic invasion . In order to understand the contribution of neutrophils bound to the apical surface to AdV infection , increasing amounts of primary human neutrophils ( 0–1 x 107 cells ) were allowed to bind to the apical surface of mock-induced or DOX-induced polarized MDCK-CAREx8 epithelia . Unbound neutrophils were removed by washing and AdV5-β-Gal was added to the apical surface for 1 h at 37°C . Viral entry was quantified 24 h later by qPCR . Neutrophils enhanced AdV entry by approximately 2–3 fold ( Fig . 5A , white bars ) and , consistent with a significant increase in neutrophil binding , AdV entry was increased by an additional 2-fold when CAREx8 expression was induced with DOX ( Fig . 5A , grey bars ) . To determine whether neutrophil-enhanced AdV entry was dependent on viral dose , 2 X 106 neutrophils were allowed to bind CAREx8 , or control CAREx7 and mCherry , mock-induced or DOX-induced epithelia followed by apical transduction with increasing MOI of AdV5-β-Gal ( Fig . 5B-D ) . To control for baseline infection , mock-induced epithelia having no neutrophils were also similarly infected with AdV5-β-Gal . Neutrophils increased apical AdV entry by 3–10-fold in MDCK-CAREx8 epithelia at all MOI ( Fig . 5B ) . Except at MOI 1 , this increase was further amplified by at least 3 fold in the presence of DOX , indicating that both neutrophils and the level of apical CAREx8 play a major role in AdV entry . In the case of uninduced mCherry epithelia ( Fig . 5C ) , neutrophils significantly increased AdV entry in a similar manner as uninduced MDCK-CAREx8 cells , while MDCK-CAREx7 epithelia followed this trend ( Fig . 5D ) . However , no significant change in AdV entry occurred in the presence of DOX indicating the importance of CAREx8 expression . Taken together , these data show that neutrophils facilitate viral entry into the polarized MDCK epithelium , particularly upon induction of apical CAREx8 expression . To confirm that the effect of neutrophils on AdV entry depends on CAR , polarized MDCK-CAREx8 cells were treated with either AdV5 FK or AdV3 FK , followed by neutrophil adhesion and infection with AdV5-β-Gal . We observed that AdV5 FK blocked AdV5-β-Gal entry by ∼7-fold in the absence of adhered neutrophils ( p<0 . 0001 , Fig . 5E ) . In the presence of adhered neutrophils , AdV5 FK , but not AdV3 FK , blocked AdV5-β-Gal entry by ∼25-fold ( p<0 . 0001 ) . The difference in fold change reflects the increased AdV-β-Gal entry in the presence of neutrophils . These results indicate that neutrophils promote adenoviral entry via CAREx8 . To further confirm that neutrophils were not simply disrupting the epithelial tight junction , TER was measured in the presence or absence of apically adhered neutrophils . Interestingly , a trend towards increased transepithelial resistance was observed when compared to MDCK-CAREx8 cells without neutrophils ( Fig . 5F ) . A lack of tight junction disruption is consistent with the evidence that increasing the basolateral CAREx7 isoform does not further augment viral infection in the presence of neutrophils ( Figs . 2B and 5D ) . To determine the mechanism by which IL-8 stimulates endogenous CAREx8 protein expression , transcription of CAREx8-specific mRNA was first investigated in polarized Calu-3 cells ( S2A Fig ) and in polarized primary human airway epithelia ( Fig . 6A ) . CAREx8 , CAREx7 , and E-cadherin mRNA levels did not significantly change within 4 h of IL-8 treatment ( Fig . 6A ) or when treated with different IL-8 concentrations ( S2A Fig ) indicating that the increase in CAREx8 was by post-transcriptional mechanisms . Accordingly , co-treatment of polarized Calu-3 ( S2B Fig ) or primary human airway epithelia ( Fig . 6B , quantitated in S3A Fig ) with IL-8 and the protein synthesis inhibitor cycloheximide ( CHX ) abolished the IL-8 mediated increase in CAREx8 expression indicating that IL-8 acutely stimulates de novo CAREx8 protein synthesis . We then asked which signaling proteins downstream of IL-8 stimulation are involved in the IL-8-mediated post-transcriptional increase of CAREx8 . It is known that IL-8 activates AKT , leading to the downstream activation of ribosomal S6 protein kinase ( S6K ) and protein translation [28] . Consistent with this , a robust activation of both AKT ( phospho-AKT-T308; Figs . 6C and S3B ) and S6K ( phospho-S6K T389; Figs . 6D and S3C ) was observed in response to IL-8 treatment . To determine whether the IL-8-mediated increased CAREx8 protein expression is downstream of AKT and S6K activation , polarized epithelia were incubated with IL-8 , chemical inhibitors for AKT ( Ly294002; Figs . 6E and S3D ) or S6K ( RO318220; Figs . 6F and S3E ) , or a combination of IL-8 and each inhibitor . Whereas IL-8 increased CAREx8 protein expression and each inhibitor alone did not affect CAREx8 protein expression , the inhibitors were able to block the IL-8-mediated increase in CAREx8 protein levels ( Figs . 6E , F , S3D , S3E ) . To further test the role of S6K in the regulation of CAREx8 protein expression , Myc-tagged S6K was expressed in Calu-3 cells by plasmid transfection ( Figs . 6G and S3F ) . Overexpression of Myc-S6K increased CAREx8 protein expression to a level similar to IL-8 treatment . Interestingly , we did not observe an additive effect between Myc-S6K and IL-8 treatment indicating that S6K is a major regulator of CAREx8 protein translation , and potentially that the amount of CAREx8 mRNA is limited . Taken together , these data show that IL-8 regulates the expression of CAREx8 via the AKT/S6K pathway . GSK3β is a multifunctional , constitutively active kinase that plays a role in multiple cellular pathways , including post-transcriptional regulation of protein expression [29] . We have previously shown that GSK3β negatively regulates CAREx8 expression and inhibition of GSK3β increases CAREx8 protein levels [16] . Although to our knowledge , GSK3β is not a known target of IL-8 signaling pathways , we investigated the activity of GSK3β upon IL-8 treatment . IL-8 treatment of polarized epithelia increased the inactivated form of GSK3β ( phospho-GSK3β-S9; Figs . 6H and S3G ) . To further validate the involvement of GSK3β inhibition in the increase of CAREx8 protein expression , epithelia were treated with GSK3β inhibitors ( SB415286 or LiCl ) for 4 h in the presence or absence of IL-8 . We observed that both GSK3β inhibitors increased CAREx8 protein expression to a level similar to that observed with IL-8 treatment ( Figs . 6I and S3H ) . No further increase in CAREx8 protein expression was observed with the addition of GSK3β inhibitor to IL-8 . Taken together , these data indicate that IL-8 regulates CAREx8 protein expression by inhibiting GSK3β . Since inhibition of S6K reverses the stimulatory effect of IL-8 on CAREx8 protein expression ( Fig . 6F ) and inhibition of GSK3β augments CAREx8 protein expression to the same extent as IL-8 treatment ( Fig . 6I ) , we asked whether these two signaling proteins lay in the same or different pathways . Polarized cells were treated with IL-8 while inhibiting GSK3β ( SB415286 ) and S6K ( RO318220 ) , individually and combined , and compared to mock treated cells . The data showed an increase in CAREx8 protein levels upon treatment with the combination of IL-8 , GSK3β inhibitor and S6K inhibitor ( Figs . 6J , lane 4 , and S3I ) indicating that blocking GSK3β relieves the inhibition of CAREx8 protein translation that is either downstream of or independent from the S6K pathway . To determine whether the effects of the above pathways on CAREx8 protein expression alter AdV infection , polarized epithelia were treated with IL-8 alone or in the presence of inhibitors for AKT ( Ly294002 ) , S6K ( RO318220 ) , or GSK3β ( SB415286 ) for 4 h . Inhibitors and IL-8 were removed and AdV5β-Gal was added to the apical surface for 1 h at 37°C . Viral entry was quantified by qPCR 24 h later ( Fig . 6K ) . Consistent with decreased CAREx8 protein expression upon inhibition of AKT and S6K , viral entry decreased and S6K inhibition completely reversed the effect of IL-8 stimulation ( first four bars , Fig . 6K ) . Consistent with the finding that GSK3β inhibition does not further increase CAREx8 protein expression upon IL-8 treatment , viral entry was identical with or without GSK3β inhibitor ( Fig . 6K , last bar compared to second bar ) . Given the above data , and taken together with current literature , we propose the model that activation of AKT by IL-8 exposure activates S6K which either directly , or via inactivation of GSK3β , is able to augment translation of the pool of mRNA present for CAREx8 ( Fig . 6L ) .
Viruses are sophisticated biological entities that can often initially infect epithelial cells without damaging the tight junction barrier integrity [13 , 30] . In the absence of preexisting immunity , viruses have evolved mechanisms to avoid inciting a robust inflammatory response and epithelial damage until replication has occurred so that progeny virions can co-opt the inflammatory response to enhance viral dissemination . Many inflammatory factors have been shown to modulate viral infections [31] and AdV infections are common in patients with inflammatory respiratory diseases such as COPD , CF , and asthma [4–7] . In this study , we show that the level of the apical AdV receptor , CAREx8 , is a major predictor of the susceptibility of an epithelium to AdV infection and that IL-8 and neutrophils , components of the innate immune system , enhance AdV entry . The proinflammatory cytokine IL-8 has previously been shown to increase the susceptibility of an airway epithelium to AdV infection potentially by translocation of an AdV5 co-receptor , αvβ3 integrin , to the apical surface of airway epithelia [17] . Consistent with this , we found that IL-8 exposure increased the levels of AdV5 co-receptor β1 integrin [32] at the apical surface of Calu-3 cells ( S4A Fig ) . Several co-receptors have been described for AdV5 and integrins have specifically been shown to facilitate adenoviral endocytosis and endosomal escape [32–34] . However , CAR is the primary receptor that mediates efficient virus attachment , a crucial step that occurs prior to integrin binding and viral entry [9 , 32–35] . We show for the first time that physiologically relevant levels of IL-8 stimulate the protein expression and the apical localization of the primary apical AdV receptor , CAREx8 , in polarized human airway epithelia . Consequently , this enhances Ad5 FK-sensitive AdV infection from the apical surface of the epithelium ( Figs . 1A and 5E ) . Interestingly , the IL-8-mediated effect was reduced at 100 ng/ml concentration . This finding is consistent with physiological studies that have demonstrated a bell-shaped dose response to IL-8 for neutrophil migration due to receptor saturation and desensitization [36] . It is also possible that IL-8-mediated signaling may undergo negative feedback to inhibit IL-8 signaling by downregulating the IL-8 receptor [37] . IL-8 appears to be CAREx8-specific since it did not affect the expression of total CAR , which is predominantly CAREx7 [14] . This is consistent with these two CAR isoforms having different biological functions within a polarized epithelium . IL-8 also has an acute effect that stimulates maximal CAREx8 expression between 4–12 h ( Fig . 1E ) , suggesting that CAREx8 might be crucial in facilitating early innate immunological responses . It is possible that prolonged IL-8-mediated signaling or apical CAREx8 expression would lead to excessive levels of neutrophils at the apical surface , and adverse immunological complications due to prolonged inflammation . Future work will focus on elucidating the effect of CAREx8 on inflammation and bacterial clearance , particularly in the presence of inflammatory diseases , such as CF . We hypothesized that the endogenous biological function of CAREx8 at the apical epithelial cell surface is to tether infiltrating neutrophils transmigrating from the basolateral interstitial space . We report for the first time that stimulation of cells with IL-8 or overexpression of CAREx8 increases neutrophil adhesion at the apical surface of epithelia ( Figs . 1H , 3 and S1 ) . Consistent with a major role for apical CAREx8 in neutrophil adhesion to the apical surface , neutrophil binding could be blocked completely by AdV5 FK , but not by FK from a non-CAR binding AdV ( Fig . 3D ) . Moreover , apical surface adhesion of CHO cells , normally lacking CAR and JAML expression , to polarized MDCK cells was significantly enhanced by over expression of CAREx8 or JAML . This indicates that cells expressing either adhesion molecule could adhere to apical CAR ( S4B Fig ) . Future studies will compare the importance of CAREx8 to epithelial ICAM-1 , which is the major rhinovirus receptor and has been identified as a neutrophil binding partner when epithelia are stimulated by IFNγ and TNFα [38 , 39] . Finally , we show that neutrophils transmigrating through the epithelium bind to the apical surface upon induction of CAREx8 ( Fig . 4A ) . Taken together , an endogenous biological function of CAREx8 is to tether infiltrating neutrophils at the epithelial apical surface . Enhanced adhesion of neutrophils to the epithelial apical surface may serve several important biological functions . For example , retention would prevent transmigrating neutrophils from being washed away into the airway lumen , and local retention in the region of IL-8 secretion would maintain focused inflammation that would prevent damage to neighboring regions . Retention would also allow neutrophils to achieve the critical concentration required to efficiently kill invading pathogens [20] and form a defensive barricade that prevents further infection of the epithelium . It is also possible that apical CAREx8 contributes to tight junction integrity as neutrophils break through to the apical surface . Each of these possibilities will be examined in future work . Although neutrophils are normally expected to facilitate the clearance of microbial pathogens , we have discovered a new repercussion for the accumulation of neutrophils at the epithelial apical surface: adhered neutrophils enhance AdV infection ( Fig . 5 ) . We observed that upon addition of neutrophils there is nearly a log increase in the AdV5 FK-sensitive viral entry ( Fig . 5B-E ) . This effect appears to be CAREx8 specific since an additional increase in AdV entry was not observed when the expression of CAREx7 or mCherry was turned on in MDCK-CAREx7 and MDCK-mCherry cells , respectively . We propose that AdV may have evolved to co-opt the innate immune response of the host in order to enhance entry into polarized epithelia . This is consistent with the fact that AdV early protein E1A stimulates the host cell to secrete IL-8 [6 , 40] . In addition , it is possible that under pathological conditions , such as CF , where excess neutrophils accumulate on the cell surface , AdV entry might be augmented even further . There could be several mechanisms by which the neutrophils might be promoting viral infection . For example , it is possible that the apically adhered neutrophils cause epithelial cell signaling which culminates in the loosening of the junctions to enable increased neutrophil recruitment [38] . This is not likely given that AdV infection is highly efficient from the basolateral surface and overexpression of basolateral CAREx7 does not enhance apical AdV infection ( Fig . 5D ) . Moreover , there was no change in TER in the presence of adhered neutrophils ( Fig . 5F ) . Neutrophils are known to secrete IL-8 [41] and therefore may stimulate apical CAREx8 synthesis and localization . While this would accommodate additional infiltrating neutrophils , AdV FK has greater affinity for CAR than neutrophil JAML and would be able to take advantage of the additional receptors to enter the epithelium . Apically adhered neutrophils may also release inflammatory mediators that alter fluid phase endocytosis from the apical surface to facilitate viral entry . Future experiments will focus on elucidating the exact mechanism ( s ) behind neutrophil-enhanced viral entry . Understanding this may lead to novel therapies to inhibit AdV infection or reduce the toxic effects of chronic inflammation . Several major steps of the mechanisms underlying the IL-8-mediated increase in CAREx8 expression have been identified . IL-8 regulates CAREx8 expression by post-transcription mechanisms ( Figs . 6 and S3 ) . Other mechanisms , such as protein stabilization , an increase in the half-life of CAREx8 , or enhanced trafficking to the apical surface , may also contribute to the effect of IL-8 and will be examined in the future . However , the reversal of the effect of IL-8 by protein synthesis inhibition and the effect of plasmid-expressed S6K , an enzyme known to directly upregulate the expression of several proteins by post-transcriptional mechanisms [28 , 42] , demonstrate an important role for IL-8 signaling in de novo CAREx8 protein synthesis . Our data show that IL-8 triggers the activation of AKT and activation of its proximal target , S6K . This is consistent with previous studies demonstrating that IL-8 signaling via AKT and S6K post-transcriptionally upregulates the protein synthesis of cyclin D1 [28 , 42] . In addition , consistent with our previous studies that show that GSK3β negatively regulates CAREx8 expression [16] , we demonstrate that IL-8 signaling results in the inhibition of GSK3β ( Fig . 6H ) and upregulation of CAREx8 expression . To our knowledge , this is the first time that GSK3β inhibition has been shown to occur upon IL-8 exposure . Based on our data and the literature , GSK3β inhibition is most likely downstream of S6K activation [43] . Importantly , we demonstrated that the treatment of Calu-3 cells with IL-8 in the presence of AKT/S6K inhibitors decreases AdV entry , while the inhibition of GSK3β augments AdV entry . Taken together ( Fig . 7 ) , these data indicate that IL-8 , potentially derived from stimulated resident macrophages or the epithelium itself , activates specific signaling pathways within polarized epithelial cells ( Fig . 6L ) that lead to increased apical CAREx8 and retention of transmigrating neutrophils at the apical surface . AdV has likely evolved to hijack this innate pathway and induced apical CAREx8 expression for entry into a polarized epithelium from the apical surface . Previously , it was assumed that AdV must breach the tight junction barrier to access its primary receptor . This study provides a novel mechanism and an explanation as to how the virus can infect the intact epithelium without breaching the barrier . Further elucidating these mechanisms in both healthy and diseased individuals will yield a greater understanding of the susceptibility of the airway epithelium to invading viral pathogens and interventions that reverse this effect . Moreover , if CAREx8 is upregulated in diseased conditions , novel therapies that target the CAR-neutrophil interaction may present a new anti-inflammatory treatment for inflammatory airway disease .
Primary human airway tracheal epithelial cells were isolated from the lungs of healthy human donors under IRB approval by the Institutional Review Board of the University of Iowa ( IRB ID No . 9507432 ) and according to the principles expressed in the Declaration of Helsinki . Primary human airway epithelia were isolated from discarded and de-identified trachea and bronchi of donor lungs and were analyzed anonymously . This study used discarded lung tissue , thus the IRB deemed consent was not needed . Calu-3 cells , cultured as described [16] , or MDCK , cultured as described [44] , were plated on 6 and 24 well dishes ( Thermo Fisher Scientific ) , respectively , and allowed to polarize for at least 4 days , or 12 mm millicells ( Millipore ) with 0 . 4 μm pores for standard polarization experiments or with 3 μm pores for transmigration studies at a density of 2 . 5 X 105 cells/millicell and grown at the air-liquid interface until TER was >600 Ω•cm2 as measured by a chopstick ohmmeter ( World Precision Instruments , Sarasota , FL ) . Primary human airway tracheal epithelial cells were a kind gift from Dr . Joseph Zabner , University of Iowa Cells and Tissues Core , Iowa City , IA . Primary airway epithelial cells , cultured and expanded as described [45] , were seeded on millicells and allowed to differentiate , as described [14 , 46] , for >2 weeks and TER >600 Ω•cm2 . Myc-tagged S6K was cloned into pRK5 plasmid under the CMV promoter . IL-8 was purchased from Gold Biotechnology ( St . Louis , MO ) , cycloheximide was from Sigma , and all other inhibitors were from Tocris Bioscience ( Bristol , United Kingdom ) . Plasmid for HA-tagged AdV5 FK was a kind gift from Dr . Glen Nemerow and Tina-Marie Mullen ( The Scripps Research Institute , CA ) . Purified AdV3 FK was a kind gift from Dr . André Lieber ( University of Washington , Seattle , WA ) . Total CAR ( 1605p ) and CAREx8-specific ( 5678p ) Abs have previously been described [44] . Ab for actin was from Sigma , E-cadherin from Life Technologies , FLAG from AbCam , AKT , S6K , GSK3β , and phosphospecific antibodies from Cell Signaling Technology ( Danvers , MA ) . Viruses were purchased from the University of Iowa Gene Transfer Vector Core . Epithelia were infected with recombinant AdV5-β-Gal at a multiplicity of infection ( MOI ) of 100 plaque forming units ( pfu ) per cell , or as indicated in the text , for 1 h at 37°C , washed with PBS , and lysed 24 h later for β-Gal protein expression and DNA isolation for qPCR for the AdV5 hexon gene , GAPDH , or MDCK actin , as previously described [16] and detailed Supplemental Experimental Procedures ( S1 Text ) . AdV5-β-Gal is replication defective and the copy number of the AdV5 hexon gene 24 h post-infection is indicative of the total number of AdV5 genomes present in a cell . Consistent with previous studies [47] , no significant amount of cell surface bound AdV5 is observed after 24 h , as measured by the trypsinization of virus off of epithelia prior to DNA extraction , indicating that the viral DNA isolated by this assay is within the epithelial cells ( S4C Fig ) . Western blot analysis and cell surface biotinylation with Sulfo-NHS-SS-Biotin ( Thermo Scientific ) were performed as previously described [14–16] and as detailed in the Supplemental Experimental Procedures . The Lenti-X Tet-On advanced inducible expression system was used according to the manufacturer’s protocol ( Clontech Laboratories ) and as detailed in Supplemental Experimental Procedures . MDCK-CAREx8 , -CAREx7 and-mCherry cells , either mock- or DOX-induced , were polarized on a 24-well dish or on millicells , as above . Neutrophils were isolated , as described previously [48] , from the peripheral blood of healthy donors who signed an Institutional Review Board-approved consent form . A neutrophil adhesion assay was performed as described [49] . Briefly , freshly isolated neutrophils from the peripheral blood of healthy human donors were stained with 1 . 5 μM calcein green for 30 min at 37°C . Stained neutrophils in 300 μl HBSS were added to the apical epithelial cell surface , spun down at 140 x g for 4 min without any centrifuge break , and allowed to adhere for 15 min at 37°C in a CO2 incubator prior to washing and imaging or addition of AdV5-β-Gal . Neutrophil binding to the apical surface of the epithelium is stable for 1–2 h after which they detach . After 1 h AdV5-β-Gal infection , epithelia are washed multiple times to ensure removal of neutrophils prior DNA extraction 24 h post infection . To block neutrophil adhesion or AdV5-β-Gal infection , the epithelial cells were incubated with either purified AdV5 FK or AdV3 FK for 10 min at room temperature prior to the addition of the neutrophils , as above , and washed 3 times with HBSS+/+ to remove the unbound neutrophils . Neutrophils were imaged using fluorescence microscopy ( Nikon Eclipse TE 2000–5 ) and the fluorescence intensity was quantified using the Metamorph software program ( Metamorph Meta Imaging Series 6 . 1 ) . Polarized Calu-3 cells on the 24 well dish were treated with IL-8 ( 30 ng/ml ) for 4 h and washed to remove IL-8 prior to neutrophil adhesion assay as described above . Neutrophil transmigration assay was performed , as previously described [50] , in the physiologically relevant basal-to-apical surface direction . Briefly , 106 fluorescently-labelled neutrophils were added to the upper chamber ( basolateral surface ) of MDCK-stable cells polarized on millicells ( 3 μM pore ) in an inverted fashion [51] and stimulated to migrate in response to 100 nM n-formyl-methionyl-leucyl-phenylalanine ( fMLP; AbCam ) added to the apical surface for 1 h at 37°C . Post-neutrophil transmigration , the neutrophils that successfully transmigrated to the bottom chamber ( transmigrated neutrophils ) were imaged with a fluorescence microscope . Apically-adhered neutrophils were detached , as previously described [49] . Briefly , millicells were transferred to a fresh 24-well dish , spun at 50 x g for 5 min , imaged and quantified as above . All experiments were performed in triplicate . Microsoft Excel , Graph Pad Prism V5 , or SPSS were used to perform statistical analyses . Statistical significance was evaluated using ANOVA or t-test , as indicated . | Respiratory viral infection is one of the leading causes of morbidity and mortality worldwide . Interventions that are able to limit viral infection will enhance human health and productivity . However , the mechanisms that control our susceptibility to viral infection and the factors that allow viral pathogens to breach the exterior epithelial barrier to initiate infection are not well understood . Here we find that adenovirus , a common cold virus and a potential gene therapy vector , uses a cellular receptor that is induced by the host innate immune response . Moreover , neutrophils , cells that are meant to protect the host in the early phase of an innate immune response , instead facilitate adenovirus infection . It has been known for over 15 years that adenovirus itself can induce an innate immune response and specifically induce host cell secretion of IL-8 , a critical chemokine that attracts neutrophils to sites of infection . However , until now , it has been unclear how IL-8 induction might benefit the virus . Our data indicate that adenovirus evolved to use our innate defense system to enhance entry into the epithelium and identifies the apical adenovirus receptor as a new target that may modulate inflammatory disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Adenovirus Entry From the Apical Surface of Polarized Epithelia Is Facilitated by the Host Innate Immune Response |
Hantaviruses infect humans via inhalation of virus-contaminated rodent excreta . Infection can cause severe disease with up to 40% mortality depending on the viral strain . The virus primarily targets the vascular endothelium without direct cytopathic effects . Instead , exaggerated immune responses may inadvertently contribute to disease development . Mononuclear phagocytes ( MNPs ) , including monocytes and dendritic cells ( DCs ) , orchestrate the adaptive immune responses . Since hantaviruses are transmitted via inhalation , studying immunological events in the airways is of importance to understand the processes leading to immunopathogenesis . Here , we studied 17 patients infected with Puumala virus that causes a mild form of hemorrhagic fever with renal syndrome ( HFRS ) . Bronchial biopsies as well as longitudinal blood draws were obtained from the patients . During the acute stage of disease , a significant influx of MNPs expressing HLA-DR , CD11c or CD123 was detected in the patients’ bronchial tissue . In parallel , absolute numbers of MNPs were dramatically reduced in peripheral blood , coinciding with viremia . Expression of CCR7 on the remaining MNPs in blood suggested migration to peripheral and/or lymphoid tissues . Numbers of MNPs in blood subsequently normalized during the convalescent phase of the disease when viral RNA was no longer detectable in plasma . Finally , we exposed blood MNPs in vitro to Puumala virus , and demonstrated an induction of CCR7 expression on MNPs . In conclusion , the present study shows a marked redistribution of blood MNPs to the airways during acute hantavirus disease , a process that may underlie the local immune activation and contribute to immunopathogenesis in hantavirus-infected patients .
Hantaviruses pathogenic to humans are rodent borne , but do not cause disease in their natural hosts . However , transmission to humans via inhalation of aerosolized virus-contaminated rodent excreta may lead to severe disease and death , thus representing a severe threat to public health [1 , 2] . Hantaviruses in Europe and Asia primarily cause hemorrhagic fever with renal syndrome ( HFRS ) whereas hantaviruses in the Americas cause hantavirus pulmonary syndrome ( HPS ) , with case fatality rates of 0 . 1–10% and 40% respectively [3] . Puumala virus ( PUUV ) , the endemic strain in Sweden , has an incubation time of 2–3 weeks and can cause a mild form of HFRS , also referred to as nephropathia epidemica [2 , 4 , 5] . In humans , hantaviruses infect the vascular endothelium without causing cytopathic effects [6] . Yet , increased vascular permeability is a hallmark of hantavirus diseases . It has been suggested that an immune-mediated dysregulation of endothelial permeability might contribute to disease pathogenesis [1 , 3 , 7–9] . Hantavirus immunopathogenesis is most likely a complex multifactorial process involving both innate [10–12] and adaptive immune cells [13–15] . Cytotoxic T lymphocytes ( CTLs ) and natural killer ( NK ) cells as well as pro-inflammatory cytokines such as tumor necrosis factor ( TNF ) produced by these lymphocytes have been implicated in causing capillary leakage [16] . Supporting this notion , stronger CTL responses have been associated with a more severe disease outcome and even death [14 , 17–20] . Monocytes and dendritic cells ( DCs ) , together termed mononuclear phagocytes ( MNPs ) , are able to present viral antigens to T cells , thus initiating and regulating virus-specific immune responses [21 , 22] . In human blood , monocytes can be further subdivided into classical , intermediate and non-classical monocytes based on varying expressions of CD14 and CD16 [23] . During both bacterial and viral infections , intermediate and non-classical monocytes in blood of patients have been reported to increase in numbers [24–27] . Kwissa et al . further illustrated in acute dengue virus infection that the expansion of intermediate monocytes correlated with formation of plasmablasts , important for development of humoral responses [27] . Indeed , a robust production of hantavirus-specific plasmablasts in circulation , as well as IgG and IgM antibodies in serum may be necessary for patient recovery and even survival [28–30] . DCs , which are superior to monocytes in priming naïve T cell responses , consist of plasmacytoid DCs ( PDCs ) and myeloid DCs ( MDCs ) that can be further separated into CD1c+ MDCs and CD141+ MDCs [31 , 32] . During viral infections , DCs are rapidly mobilized to peripheral tissues where they replenish the tissue-resident DCs that first encounter the pathogens and either die due to infection or migrate to draining lymph nodes [33–35] . In humans , both monocyte-derived cells as well as DCs have been observed in respiratory compartments at steady state with the capacity to detect and respond to invading pathogens [31 , 36–39] . Since hantaviruses are transmitted predominantly via inhalation , studying immunological events in the airways where viral replication is first initiated is of importance to understand the processes leading to immunopathogenesis . Furthermore , pulmonary dysfunction has been reported in HFRS patients [40–42] . Expansion of cytotoxic CD8+ T cells in the airways of hantavirus-infected patients has been described as contributing to disease severity [13 , 14] . This suggests that DCs may be involved in promoting the recruitment and local activation of T cells . However , little is known on the contributions of MNPs in human hantavirus-infected patients in vivo , especially at the site of entry . In order to investigate the involvement of monocytes and DCs in local immune events in the airways , we obtained endobronchial biopsies from 17 PUUV-infected HFRS patients during the acute phase of disease and compared them to samples from uninfected controls ( UC ) . We illustrated a significant infiltration of CD8+ T cells and MNPs into the bronchial tissue during acute HFRS compared to UC . As hantaviruses establish a systemic infection , we characterized MNPs in longitudinal peripheral blood samples from the same patients . Concurrent with the increase of MNPs in the airways , we observed a dramatic depletion of circulating monocytes and DCs during the acute phase of disease . By investigating the distribution of monocytes and DCs in different anatomical compartments during an acute viral infection in humans , we gained insights into the potential roles of specific cell subsets based on their migratory patterns and tissue-specific locations during the course of disease .
Although HFRS mainly manifests in the kidneys , respiratory involvement has been increasingly documented in these patients , including pulmonary edema that may lead to respiratory failure [14 , 28 , 40] . Of the 17 PUUV-infected HFRS patients included in this study , 10 experienced respiratory symptoms such as dry cough and dyspnea , and 5 needed oxygen treatment ( S1 Table ) . Given that the airways are the initial site of infection after inhalation of hantavirus-contaminated rodent excreta , little is known regarding the early immune response taking place locally . In the current study , bronchoscopy was performed on PUUV-infected HFRS patients in order to sample their airways during the acute phase of disease . As soon as platelet counts had stabilized and the patients were able to withstand the procedure , endobronchial biopsies and bronchoalveolar lavage ( BAL ) were collected from each patient ( median 9 days after onset of symptoms ) . In addition , longitudinal peripheral blood samples were collected from these patients during both the acute phase ( 2–14 days after onset of symptoms ) and convalescent phase ( >15 days after onset of symptoms ) of HFRS . Results from blood and lung samples collected from HFRS patients were compared throughout the study with samples from UC , who similarly underwent bronchoscopies and blood draws ( Fig 1A ) . In 15 patients , viral load was detected in BAL cells , suggesting local viral replication . However , viral load alone could neither explain the respiratory symptoms experienced by 10 patients , nor the need for oxygen treatment by 5 patients ( Fig 1B ) . As CTLs have been implicated in contributing to hantavirus pathogenesis [13 , 14 , 16 , 43] , we investigated the absolute numbers of CD8+ T cells in the airways of patients . As previously described within the same study cohort [14] , more CD8+ T cells were detected in BAL of patients that required oxygen treatment when compared to those who did not need oxygen treatment ( p<0 . 05 ) ( Fig 1C ) . Granzyme B , a cytotoxic protease released by CTLs and NK cells that can induce apoptosis , was also detected in BAL fluid of patients , with a trend towards higher amounts of granzyme B in patients requiring oxygen treatment ( n = 5 ) , although the difference was not significant ( p = 0 . 08 ) ( Fig 1D ) , possibly due to the relatively large variation between individuals . In order to assess whether CD8+ T cells were also present in bronchial tissue of patients , we performed immunohistochemistry on sections of endobronchial biopsies ( Fig 1E ) . Indeed , significantly more CD8+ T cells were detected in biopsies taken from patients with acute HFRS than in biopsies from UC ( Fig 1F ) . This prompted us to examine aspects of local T cell activation by investigating monocytes and DCs in the airways during hantavirus disease . By immunohistochemistry , we investigated the distribution and frequency of cells expressing surface markers for monocytes and DCs on sections of endobronchial biopsies . We found significantly more HLA-DR+ cells ( p<0 . 01 ) in biopsies taken from patients with acute HFRS than in biopsies from UC ( Fig 2A and 2B ) . The increased HLA-DR staining , especially in the lamina propria , suggested an infiltration of MNPs . Additionally , the pulmonary epithelium also displayed increased HLA-DR staining in tissues from acute HFRS patients , possibly from local inflammation leading to upregulation of HLA-DR on epithelial cells [44] in addition to infiltration of immune cells to the site of infection . We also observed a significant increase in the number of cells expressing the myeloid cell marker CD11c during acute HFRS compared to UC ( Fig 2C ) . Detailed analysis of bronchial tissue showed significantly increased numbers of CD11c+ cells in the lamina propria ( p<0 . 05 ) and epithelium ( p<0 . 05 ) of hantavirus-infected patients compared to UC ( Fig 2D ) . In addition , the number of cells expressing the PDC marker CD123 was also significantly higher in the lamina propria ( p<0 . 05 ) of these patients ( Fig 2E and 2F ) . To address whether patients with high numbers of CD8+ T cells in the bronchial tissue also had high numbers of MNPs , we performed a Spearman correlation test and observed a positive association between CD8+ cells and CD11c+ cells ( p = 0 . 08 ) ( Fig 2G ) , and a significant correlation between CD8+ cells and CD123+ cells ( p<0 . 05 ) ( Fig 2H ) . Taken together , we observed an infiltration of MNPs into the airways during acute HFRS coinciding with the presence of CD8+ T cells at the site of infection . However , hantavirus infection is typically systemic and not limited to the airways [3] . Indeed , viral RNA copies can be detected in the plasma early during disease onset , but also on the day of bronchoscopy when the bronchial biopsies were obtained ( Fig 2I and S1 Fig ) . Thus , we next explored the possibility that blood DCs and monocytes exposed to virus or virus-induced cytokines may have received signals to migrate into the airways , contributing to the significant influx of MNPs in the bronchi . Since monocytes participate in inflammation [22] , especially following viral infection , we hypothesized that the number of monocytes would expand in peripheral blood during the acute phase of HFRS , as has been observed in other acute viral infections [27 , 45 , 46] . To investigate how hantavirus infection may affect monocytes in circulation , the frequencies of classical ( CD14+ CD16- ) , intermediate ( CD14+ CD16+ ) and non-classical ( CD14- CD16+ ) monocytes were analyzed using flow cytometry ( Fig 3A and S2 Fig ) . The absolute number of classical monocytes per microliter of blood decreased significantly ( p<0 . 001 ) during acute HFRS compared to samples from UC ( Fig 3B and S2 Table ) , and subsequently normalized during the convalescent phase of the disease ( S3 Table ) . Similarly , the numbers of intermediate monocytes and non-classical monocytes were also significantly reduced ( p<0 . 001 ) during acute HFRS compared to UC ( Fig 3B and S2 Table ) . Interestingly , for all the monocyte subsets observed , the low numbers of cells in circulation coincided with high viral load ( viral RNA copies per mL of plasma ) as assessed by quantitative reverse transcriptase polymerase chain reaction ( qRT-PCR ) ( Fig 3C and S1 Fig ) . During convalescence , when virus was no longer detectable in plasma , the number of monocytes returned to comparable values as those in UC , except for intermediate monocytes . In summary , we observed a loss of monocytes in the peripheral blood of patients during the acute phase of HFRS . The loss of monocytes in circulation during acute HFRS led us to also assess whether DCs , key determinants of viral disease outcome due to their capacity to initiate and activate T cell responses [47] , would also be affected by hantavirus infection . We first analyzed the two MDC subsets found in human blood: CD1c+ MDCs and CD141+ MDCs ( Fig 4A ) . We observed a dramatic reduction of both MDC subsets during acute HFRS as compared to UC ( Fig 4A ) . The reduction in absolute numbers was statistically significant for CD1c+ MDCs ( p<0 . 01 ) and CD141+ MDCs ( p<0 . 001 ) ( Fig 4B ) . On average , the number of CD1c+ MDCs was as low as 1 . 5 cells per microliter of blood during the early acute phase as compared to 16 cells per microliter found in UC ( S2 Table ) . The number of CD141+ MDCs , already rare under steady state conditions , decreased by 97% during acute HFRS ( S3 Table ) . For both CD1c+ and CD141+ MDCs , the cell numbers normalized during the convalescent phase of the disease ( Fig 4B ) . Of importance , we excluded the possibility that the reduced number of DCs in acute HFRS blood samples was a consequence of the cells being more fragile to freeze-thawing , by confirming similarly low frequencies of DCs and monocytes in fresh samples from patients with acute HFRS ( S3 Fig ) . As with the monocytes , we compared the kinetics of plasma viral load with the absolute numbers of MDCs in blood and found that high viral load coincided with low numbers of CD1c+ and CD141+ cells ( Fig 4C ) . We also assessed the numbers of PDCs in blood during acute and convalescent HFRS . PDCs are the major producers of antiviral type I interferon ( IFN ) in the body and are important in the defense against viral pathogens , despite their low frequency . Yet , levels of IFN-α are not elevated in blood of HFRS patients [48] . Here , we found that the number of blood PDCs , as defined by their CD123 and CD303 expression , was significantly reduced during acute HFRS as compared to UC ( Fig 5A ) . The drop in absolute PDC number ( p<0 . 001 ) was maintained also during early convalescence at days 15–21 after the onset of HFRS , but eventually returned to normal values ( Fig 5B and S2 Table ) . Again , the loss of blood PDCs during acute HFRS coincided with high viral load ( Fig 5C ) . Together , a massive depletion of both MDCs and PDCs was observed in peripheral blood during acute HFRS . As hantaviruses are not known to cause cytopathic effects , the loss of monocytes and DCs in circulation could reflect a redistribution of MNPs from circulation into the airways , as we had observed an infiltration of MNPs in the bronchial tissue ( Fig 2 ) . To assess whether trafficking of blood DCs and monocytes to lymph nodes [49 , 50] or other tissues [51 , 52] could account for the reduced numbers of monocytes and DCs in blood during acute HFRS , we measured the surface expression of the chemokine receptor CCR7 on the few MNPs still in circulation . At steady state , few or no cells expressed CCR7 , as exemplified by the UC ( Fig 6A ) . However , during acute HFRS , a subset of the cells remaining in peripheral blood expressed CCR7 on their surfaces that progressively disappeared over time , as exemplified by CD1c+ MDCs ( Fig 6A ) . Intermediate monocytes and non-classical monocytes presented with a higher frequency of CCR7+ cells in HFRS patients compared to UC ( p<0 . 05 ) during acute HFRS ( Fig 6B ) . CD1c+ MDCs ( p<0 . 01 ) also upregulated CCR7 expression pattern throughout the acute phase , while the CD141+ MDCs showed no or very modest upregulation of CCR7 ( Fig 6B ) . Although a subset of PDCs upregulated CCR7 during early acute HFRS ( p<0 . 001 ) , the frequency of CCR7+ PDCs returned back to low levels in the late stage of acute disease ( days 11–14 ) , at frequencies similar to those in UC ( Fig 6B ) . Although the overall maturation profile of monocytes and DCs in circulation as determined by upregulation of co-stimulatory molecules CD70 and CD86 was not pronounced , CCR7+ cells had a higher expression of CD70 ( classical monocytes and CD1c+ MDCs ) and CD86 ( CD1c+ MDCs ) than the CCR7- cells , consistent with a more mature phenotype ( Fig 6C and 6D ) . Taken together , the data suggest that although the majority of monocytes and DCs are absent in circulation during acute HFRS , the cells that remain in blood appear to have received signals to upregulate migratory receptors such as CCR7 , facilitating migration to lymph nodes or peripheral tissues . Finally , we established an experimental system to address the loss of MNPs in blood of patients with acute HFRS . Classical monocytes and CD1c+ MDCs were isolated from blood of healthy volunteers ( Fig 7A ) and exposed to PUUV or UV-inactivated PUUV in vitro . We measured the frequency of cells expressing PUUV antigens over time by immunofluorescence staining using human anti-PUUV serum ( Fig 7B ) . Forty hours post infection , 1 . 4% of classical monocytes and 0 . 4% of CD1c+ MDCs were infected by PUUV ( Fig 7C ) , whereas PUUV antigen was undetectable in cells that were either uninfected or exposed to UV-inactivated PUUV . In addition to detecting viral proteins , PUUV RNA was detected by qRT-PCR in classical monocytes and CD1c+ MDCs exposed to replicating PUUV , at approximately 200-fold higher than in cells exposed to UV-inactivated PUUV ( Fig 7D ) . However , no replicating viruses were detected in supernatants of infected cells , suggesting that PUUV replication is restricted in these cells ( S4 Fig ) . Neither replicating nor UV-inactivated PUUV decreased the viability of these cells compared to uninfected cells ( Fig 7E ) , typical of the non-cytopathic effects of hantavirus [1] . Interestingly , exposure to PUUV improved the viability of classical monocytes significantly at 40 hours ( p<0 . 01 ) and 60 hours ( p<0 . 001 ) ( Fig 7E ) . When exposed to Hantaan virus ( HTNV ) that causes severe HFRS , we similarly observed that classical monocytes and CD1c+ MDCs were susceptible to infection in vitro , without the induction of cell death ( S5 Fig ) . Both classical monocytes and CD1c+ MDCs also responded to PUUV exposure by modulating their expression of chemokine receptors over time ( S6 Fig ) . CCR2 , important for mobilization of monocytes from bone marrow to peripheral tissues [22 , 53] , was downregulated on both classical monocytes ( p<0 . 05 ) and CD1c+ MDCs after 40 hours of PUUV exposure ( p<0 . 05 ) compared to uninfected controls ( Fig 7F and 7G ) . CCR4 and CCR6 , chemokine receptors that have been associated with tissue homing , were upregulated on classical monocytes ( p<0 . 05 ) and CD1c+ MDCs respectively after 12 hours of PUUV exposure compared to cells that were not exposed to virus ( Fig 7F and 7G ) . Both classical monocytes and CD1c+ MDCs ( p<0 . 05 ) also responded to PUUV exposure by upregulating the migratory chemokine receptor CCR7 typically expressed by mature cells ( Fig 7H and 7I ) . In line with their increased CCR7 expression , both classical monocytes and CD1c+ MDCs also upregulated the co-stimulatory molecule CD86 ( p<0 . 05 ) upon exposure to replicating PUUV , but not to UV-inactivated PUUV ( Fig 7F and 7G ) . In summary , both classical monocytes and CD1c+ MDCs were susceptible to PUUV infection in vitro , but infection did not result in cell death . The observed regulation of chemokine receptor expression on classical monocytes and CD1c+ MDCs upon exposure to PUUV in vitro provides a platform for further investigations into the molecular mechanisms governing the redistribution of MNPs observed in patients with acute HFRS .
In this study , we explored the involvement of monocytes and DCs during hantavirus infection by characterizing MNPs at the initial site of infection: the airways , where virions enter their human host upon inhalation of aerosolized hantavirus-containing rodent excreta . Viral RNA , as previously shown [14] , can be detected in BAL cells of patients during acute HFRS . An expansion of cytotoxic CD8+ T cells in BAL has been shown to correlate with disease severity [14] . In addition to CD8+ T cells lining the airways as reflected by BAL sampling , we now demonstrate the presence of CD8+ T cells in the bronchial tissue of hantavirus-infected patients . In the same HFRS patients , high levels of HLA-DR+ or CD11c+ cells were observed in the bronchial biopsies , suggesting an influx of monocytes and/or MDCs . An increase in the number of CD123+ cells also suggests an infiltration of PDCs into the airways . The presence of monocytes and DCs in the airways might explain the observed increase in CD8+ T cells present in the airways during acute HFRS . During influenza virus infection in mice , recruitment of DCs to the lungs is necessary for mounting adaptive immune responses needed for efficient viral clearance [33–35] . Specifically , local interaction in the lungs between antigen-bearing DCs is required for protective CD8+ T cell responses [54] . A careful investigation of how the MNPs interact with CD8+ T cells in the airways of hantavirus-infected patients would facilitate understanding of whether MNPs contribute to pathogenesis or immunity , by activating or controlling CTL activity . As hantavirus infection is systemic [3] , we also characterized the absolute numbers of six distinct MNP populations in the blood of patients over the course of disease , from acute infection to convalescence . We found a depletion of all populations , especially MDCs , in the peripheral blood of PUUV-infected patients during the acute phase of HFRS , coinciding with the presence of viral RNA in blood . During acute HFRS , CCR7 was upregulated on several monocyte and DC populations , indicating a mobilization of cells from the blood toward lymph nodes or peripheral tissues . In vitro , our data further demonstrated that classical monocytes and CD1c+ MDCs were susceptible to PUUV infection and remained alive . Although Markotic et al . suggested a differentiation of monocytes into DC-like cells after hantavirus infection in vitro [55] , we were not able to identify monocytes expressing DC markers by flow cytometry upon PUUV exposure . Our data corroborated earlier findings by Raftery et al . and Temonen et al . suggesting that human DCs and monocytes may contribute to pathogenesis: both monocytes and DCs remain alive after infection , potentially leading to viral dissemination due to their migratory properties [56 , 57] . Hantavirus infection is not marked by a general loss of immune cells or leukopenia in circulation of patients , since these viruses do not cause obvious cytopathic effects [56 , 58] . Instead , the numbers of NK cells in blood are expanded in PUUV-infected patients [11 , 13 , 59] . We anticipated that a similar expansion of MNPs would be detected in blood from our patients as has been reported in those with other acute viral infections , as both monocytes and DCs are mobilized from the bone marrow to partake in the innate response to a viral infection [27 , 45 , 46 , 60 , 61] . In contrast , we found a depletion of all monocyte and DC subsets in blood during acute HFRS . Although Tang et al . reported an expansion of intermediate monocytes in the blood of HFRS patients [62] , no such increase was observed in the present study . The cause of this disparity could be technical due to our gating strategy that excluded all lineage+ and HLA-DR- cells . Alternatively , biological differences between the virus strains could yield differing results , as their patient cohort was infected with HTNV whereas our patients were infected with PUUV , the endemic hantavirus strain in Sweden . In line with our findings , the loss of DCs has been documented in patients with acute influenza A virus ( IAV ) and human immunodeficiency virus ( HIV ) infections , related to depletion and impaired function of DCs during acute infection [63–66] . While all DC populations in peripheral blood returned to normal values during the convalescent phase of HFRS , we noted that the absolute numbers of intermediate monocytes and non-classical monocytes remained low , even more than 100 days after onset of disease ( S3 Table ) . Monocytes arise from bone marrow precursors , differentiating from classical monocytes via intermediate monocytes to non-classical monocytes in their lifetime [67 , 68] . We speculate that in PUUV-infected patients , there may be a delay in the developmental progression of circulating classical monocytes , even during convalescence . The prolonged expansion of NK cells in the circulation of HFRS patients [11] could provide a source of IFN gamma , a cytokine known to activate classical monocytes to a more inflammatory phenotype [69] . Additionally , the emerging concept of trained immunity [70] suggests that monocyte precursors in the bone marrow may be epigenetically modified upon exposure to hantavirus such that they remain poised for future infections . Our data from experiments performed in vitro suggest that even if blood MNPs were susceptible to the virus , hantavirus infection did not lead to cell death . A potential explanation for the stark depletion of DCs observed in blood could be that these cells have migrated out of circulation . The chemokine receptor CCR7 controls the homing of DCs to lymph nodes , where priming of T cells and initiation of adaptive immune responses can occur [49 , 50 , 71 , 72] . Increased CCR7 expression on blood CD1c+ MDCs during acute HFRS could indicate that these cells have migrated to the lymph nodes , in response to either direct viral infection , as we could show in vitro , or to pro-inflammatory cytokines in serum of patients [73] . Specifically for hantavirus infections , activation of CD4+ T cells have been shown to be instrumental in viral control and improved clinical outcome [74] . By infecting monocytes and DCs in vitro , we have developed a platform for further dissection of the underlying mechanisms by which exposure to hantavirus determines cellular trafficking . For instance , the chemokine receptor expression may indicate where blood MNPs traffic to during acute HFRS . Accumulation of monocytes in the brain during West Nile virus infection has been related to expression of CCR2 , which is important for the egress of monocytes from the bone marrow into tissue [60] . Detection of hantavirus RNA in the bone marrow of a patient [75] suggests that hantavirus could impede the release of monocytes into the bloodstream by downregulating expression of CCR2 , as our data suggest . In other respiratory diseases such as chronic obstructive pulmonary disease ( COPD ) , the increased presence of DCs in the lungs correlated with the upregulation of CCR6 on DCs and an increase of the CCR6 ligand ( CCL20 ) in the airways [76] . In mice , expression of CCR4 on T cells imprints them to home to the lungs upon influenza infection [77] . However , these scenarios have not been carefully investigated in the homing of monocytes and DCs into the lungs during viral infection in humans . In conclusion , blood monocytes and DCs were dramatically depleted during the acute phase of HFRS caused by PUUV . The high numbers of CD8+ T cells in the airways [14 , 16] , correlating with respiratory symptoms experienced by patients , may have been promoted by an infiltration of MNPs into the airways , as demonstrated in bronchial biopsies of hantavirus-infected patients in this study . As the viral load subsides in the blood , the numbers of blood monocytes and DCs also return to normal values . By establishing in vitro hantavirus infections of MNPs , the descriptive nature of patient data can be complemented in future in vitro studies to elucidate the mechanisms of how hantavirus infection can orchestrate the mobilization of monocytes and DCs from the blood into peripheral tissues such as the respiratory tract , and lymphoid organs . A better understanding on the role of monocytes and DCs during hantavirus infection is valuable in the development of immunomodulatory strategies to treat hantavirus-infected patients .
The study protocol was approved by the regional Ethical Review Board at Umeå University , Umeå , Sweden . Written informed consent was obtained from study subjects , all of whom were adults . Peripheral blood , bronchoalveolar lavage ( BAL ) and endobronchial biopsies were prospectively obtained from 17 hospitalized PUUV-infected patients between 2008 and 2011 . The criteria for study enrollment were described earlier [14] . Peripheral blood samples for flow cytometry analysis were collected during the acute phase ( 2–14 days after disease onset; median 6 days ) . Follow-up samples were taken throughout the first weeks of infection as well as the convalescent phase ( >15 days after disease onset ) . Patients were monitored using qRT-PCR to assess plasma viral load until two consecutive measurements were negative ( median 11 days ) [78] . Briefly , viral RNA was extracted from plasma and cDNA was generated . qRT-PCR was performed in triplicate using primers designed based on PUUV RNA sequences . No fatal cases were observed in this study cohort . Twelve uninfected age- and sex-matched blood donors were included in this study . They underwent bronchoscopy for the collection of endobronchial biopsies and bronchoalveolar lavage ( BAL ) as well as peripheral blood draw . Standard clinical procedures , including differential cell counts , were used to obtain clinical data for all subjects used in this study ( S1 Table ) [79] . Bronchoscopy was performed on all patients 6 to 14 days ( median 9 days ) after onset of symptoms . Patients underwent bronchoscopy as soon as their clinical condition allowed them to withstand the procedure . This included absence of hypotension or hypoxemia as well as improvements of coagulation parameters to avoid bleeding . Bronchoscopy was performed as soon as possible and when blood platelet count was higher than 100 x 109 . At that time , all patients were still in need of hospital care due to the acute infection . In brief , patients and UCs were treated with oral midazolam ( 4–8 mg ) and intravenous glycopyrronium ( 0 . 2–0 . 4 mg ) 30 minutes ( min ) before the bronchoscopy . For topical anesthesia , lidocaine was applied , and additional lidocaine was administered in the larynx and bronchi during the procedure . A flexible video bronchoscope ( Olympus BF IT200 ) was inserted through the mouth via a plastic mouthpiece . From each patient , four to six endobronchial biopsies were taken from the main carina and the main bronchial divisions on the left side using fenestrated forceps ( Olympus FB-21C ) . BAL was obtained with saline solution ( 3 x 60 mL ) from the contralateral side . BAL samples were filtered through a 100 μm nylon filter ( Syntab ) and centrifuged at 400 x g for 15 min at 4°C . Endobronchial biopsies were processed and embedded into glycol methacrylate resin ( Polyscience ) , as previously described [80] . Sections from biopsies ( 2 μm ) were stained in duplicates with anti-CD8 , HLA-DR , CD11c , and CD123 ( all BD Biosciences ) followed by the rabbit anti-mouse ( Dako ) biotinylated secondary antibody . The immunostaining was performed as previously described [13] . All sections were visualized with 3-amino-9-ethylcarbazole ( AEC ) , and cell nuclei were counterstained with Mayer hematoxylin ( Histo Lab ) . Finally , all sections were analyzed using a high-resolution digital scanner , NanoZoomer-XR ( HAMAMATSU ) to convert them into digital images . A blinded analysis was performed using the scanned sections and NanoZoomer Digitial Pathway View2 software ( NDP View; HAMAMATSU ) . The number of positive cells was expressed as cells/mm and cells/mm2 of epithelium and lamina propria , respectively . Quantification of HLA-DR molecules was carried out with a Leica DMR-X microscope ( Leica Microsystems GmbH ) coupled to computerized image analysis ( Leica Qwin 5501W; Leica Imaging Systems ) as described previously [81] . For isolation of peripheral blood mononuclear cells ( PBMCs ) , whole blood from PUUV-infected patients and UC was collected in CPT tubes ( BD ) and centrifuged according to manufacturer’s instructions . The separated suspension of PBMCs was harvested and then washed in PBS . PBMCs were frozen in 90% human albumin ( Octapharma ) , 10% DMSO ( WAKO-Chemie Medical ) , and 50 IE heparin ( LEO Pharma ) , and stored in liquid nitrogen for later analysis . Absolute numbers of all monocyte and DC subsets were calculated by using the absolute lymphocyte and monocyte counts obtained on the automated hematology analyzer and the percentages of events in each respective gate obtained from flow cytometry data . Monocytes and primary CD1c+ MDCs were isolated from buffy coats obtained from Karolinska University Hospital ( Stockholm , Sweden ) as previously described [82] . Monocytes were isolated using the human monocyte enrichment kit ( RosetteSep; StemCell Technologies ) according to the manufacturer’s instructions . The blood was diluted , carefully layered on Ficoll-Paque PLUS ( GE Healthcare Biosciences ) and centrifuged for 20 min at 1800 x g at room temperature . For isolation of human CD1c+ MDCs , magnetic labeling using CD1c+ MDC isolation kit ( Miltenyi Biotec ) was used on enriched populations of monocytes . Monocytes and MDCs were cultured in RPMI1640 ( Sigma-Aldrich ) with 10% fetal bovine serum ( FBS ) , 1% penicillin/streptomycin and 1% L-Glutamine ( all Invitrogen ) . Cells were cultured at 1x106 cells per mL of complete medium . MDCs were additionally supplemented with 2 ng/mL GM-CSF ( R&D Systems ) . PUUV strain Kazan and HTNV strain 76–118 were propagated on Vero E6 cells ( ATCC Vero C1008 ) as previously described [48] . The virus stocks were titrated on Vero E6 cells for calculations of multiplicity of infection ( MOI ) . UV inactivation of hantaviruses was performed for 25 seconds using a VL215G Vilber Lourmat UV lamp ( Torcy ) , as a negative control for productive infection . Cells were exposed to medium alone ( uninfected ) , infected with hantaviruses or exposed to UV-inactivated hantaviruses at an MOI of 7 . 5 for 2 hours ( h ) . Cells were then washed and incubated for 12–60 h at 37°C . Supernatants were collected after centrifugation and were stored at −80°C until further analysis . Cells were stained for flow cytometric analysis . Cell suspensions were stained with Live/Dead Aqua fixable dead cell stain kit ( Invitrogen ) to exclude dead cells . Non-specific binding was prevented by adding FcR blocking reagent ( Miltenyi Biotec ) followed by surface staining with conjugated Abs ( S5 Table ) . Briefly , cells were stained for 15 min at 4°C in FACS buffer ( PBS with 2% fetal bovine serum ) and fixed in 1% paraformaldehyde ( PFA ) . For chemokine receptor staining , cells were stained for 15 min at 37°C prior to addition of cell surface antibodies for another 15 min at RT . For HTNV-infected cells , intracellular staining with anti-nucleocapsid protein ( N ) antibody ( B5D9 , Progen ) was assessed using a standard protocol . In brief , cells were stained for surface markers , fixed and permeabilized using Transcription Factor Staining Buffer Set ( eBioscience ) . Cells were analyzed by flow cytometry using a LSRII instrument or LSRFortessa ( both BD Biosciences ) and data were analyzed using FlowJo X software ( Tree Star ) . At 60 h post infection , RNA from DCs and monocytes infected with PUUV in vitro was isolated using 450 μl TriPure Isolation Reagent ( Roche Diagnostics ) . The relative levels of PUUV RNA in PUUV-infected DCs and monocytes was assessed using a qRT-PCR assay , as previously described [83] . β-actin mRNA levels were measured in parallel , using a commercially available TaqMan gene expression assay ( 4333762; Applied Biosystems ) . The expression of PUUV S segment RNA was calculated against the housekeeping gene β-actin: 2-[Ct ( PUUV gene ) -Ct ( B-Actin ) ] . At 40 h post infection , classical monocytes and CD1c+ myeloid DCs were adhered for 20 min on Alcian blue-coated coverslips at 100 000 cells per condition . Cells on the coverslips were gently washed in PBS and fixed with pre-warmed 4% paraformaldehyde for 20 min at room temperature . Cells were then blocked with PBS containing 1% normal goat serum and permeabilized with 0 . 1% Triton-X 100 ( Sigma ) and stained with human anti-PUUV serum for 1 hour . Secondary antibodies against human IgG conjugated to Alexa Fluor 488 were used . Additionally , CD1c+ MDCs were co-stained with anti-HLA-DR conjugated to Alexa Fluor 647 . Coverslips were mounted on glass slides with Prolong Diamond Antifade mountant with DAPI ( Molecular Probes ) . Confocal images were acquired on a Zeiss LSM700 using a 10x objective . PUUV+ cells were enumerated out of 1000–2000 cells per condition using FIJI ImageJ software ( NIH ) . Levels of granzyme B in BAL fluid were measured using the commercially available Human Granzyme B ELISA kit ( Abcam ) . For all patient data generated ex vivo , mean cell counts of monocyte and DC populations were modeled using Poisson regression with patient-specific random intercept and robust standard errors . The proportions of CCR7+ CMs , IM , NCM , CD1c+ MDCs , CD141+ MDCs , and PDCs were modeled using logistic regression with patient-specific random intercept . Random intercepts were used to account for the potential dependence among repeated blood measurements over time . Number of days since symptoms' onset in HFRS patients was the predictor of interest and was categorized as acute phase ( 2–14 days ) or convalescent phase ( >15 days ) . UC served as the reference group . Correlations were analyzed using Spearman’s rank correlation coefficient . For in vitro experiments , statistical significance was assessed using paired t-test . Comparisons for IHC data are by Mann–Whitney U-test . Data were analyzed using GraphPad Prism version 6 . 0 ( GraphPad Software ) and Stata version 14 . 1 ( StataCorp , College Station , TX ) . All the reported p-values are two-sided and p-values<0 . 05 was considered statistically significant . | Inhalation of hantavirus-infected rodent droppings can cause a wide range of disease ranging from mild symptoms to deaths in humans . Central to hantavirus disease is vascular leakage that can manifest in different organs , including the lungs . Although the virus can infect endothelial cells lining the blood vessels , it does not cause cell death . Instead , activation of the immune system in response to viral infection has been implicated in causing vascular leakage . In this study , we investigated how monocytes and dendritic cells ( DCs ) are involved in hantavirus disease , given their capacity to activate other immune cells . We obtained unique clinical material from 17 Puumala virus-infected patients including mucosal biopsies from the airways as well as multiple blood draws over the course of disease . In the airways of these patients , we observed an infiltration of monocytes and DCs . In parallel , there was a dramatic depletion in peripheral blood—more than ten-fold—of monocytes and DCs that was sustained throughout the first two weeks of disease . Taken together , this study provides novel insights into immune mediated processes underlying human hantavirus pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"blood",
"cells",
"flow",
"cytometry",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"biopsy",
"pathogens",
"immunology",
"microbiology",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"viruses",
"hemorrhagic",
"fever",
"with",
"renal",
"syndrome",
"hantavirus",
"rna",
"viruses",
"cytotoxic",
"t",
"cells",
"bunyaviruses",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"spectrum",
"analysis",
"techniques",
"white",
"blood",
"cells",
"animal",
"cells",
"medical",
"microbiology",
"microbial",
"pathogens",
"t",
"cells",
"spectrophotometry",
"cytophotometry",
"blood",
"cell",
"biology",
"monocytes",
"anatomy",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases",
"organisms"
]
| 2017 | Human hantavirus infection elicits pronounced redistribution of mononuclear phagocytes in peripheral blood and airways |
Anopheles darlingi , the main malaria vector in the Neotropics , has been considered to be highly anthropophilic . However , many behavioral aspects of this species remain unknown , such as the range of blood-meal sources . Barrier screens were used to collect resting Anopheles darlingi mosquitoes from 2013 to 2015 in three riverine localities ( Lupuna , Cahuide and Santa Emilia ) in Amazonian Peru . Overall , the Human Blood Index ( HBI ) ranged from 0 . 58–0 . 87 , with no significant variation among years or sites . Blood-meal analysis revealed that humans are the most common blood source , followed by avian hosts ( Galliformes-chickens and turkeys ) , and human/Galliforme mixed-meals . The Forage Ratio and Selection Index both show a strong preference for Galliformes over humans in blood-fed mosquitoes . Our data show that 30% of An . darlingi fed on more than one host , including combinations of dogs , pigs , goats and rats . There appears to be a pattern of host choice in An . darlingi , with varying proportions of mosquitoes feeding only on humans , only on Galliformes and some taking mixed-meals of blood ( human plus Galliforme ) , which was detected in the three sites in different years , indicating that there could be a structure to these populations based on blood-feeding preferences . Mosquito age , estimated in two localities , Lupuna and Cahuide , ranged widely between sites and years . This variation may reflect the range of local environmental factors that influence longevity or possibly potential changes in the ability of the mosquito to transmit the parasite . Of 6 , 204 resting An . darlingi tested for Plasmodium infection , 0 . 42% were infected with P . vivax . This study provides evidence for the first time of the usefulness of barrier screens for the collection of blood-fed resting mosquitoes to calculate the Human Blood Index ( HBI ) and other blood-meal sources in a neotropical malaria endemic setting .
The Human Blood Index ( HBI ) , formerly known as the anthropophilic index or human blood ratio , is the proportion of recently-fed mosquitoes , usually vector species that have taken a human blood-meal [1] . This index is a very important component of the formulae used to determine vectorial capacity and varies depending on mosquito species , collection area and season or time of collection [2] . From an epidemiological standpoint , it is crucial to be able to accurately identify mosquito blood-meals for studies of transmission dynamics of viral and parasitic pathogens [3] . For example , in Equatorial Guinea , the calculation of this index before and after indoor interventions to reduce malaria did not detect any mosquito behavioral differences , and researchers concluded that control strategies in this region were ineffective [4] . In Central Kenya , anthropophily decreased in An . gambiae after the introduction of long lasting insecticide nets ( LLINs ) and zooprophylaxis [5] . However , in southern Zambia , after two years of LLIN intervention , the main vector , Anopheles arabiensis , remained highly anthropophilic [6] . In Tanzania the HBI showed a change in the main blood-source in An . arabiensis but not in An . funestus after the use of spatial repellent coils [7] . Another index to quantify host selection patterns is the incidence of multiple blood-meals from the same host species ( cryptic ) or from two or more different host species ( patent ) [8] . Evidence that malarial mosquitoes take partial blood-meals from multiple hosts may be interpreted as interrupted blood-feedings that could increase the probability of both acquiring and transmitting Plasmodium [9] . On the other hand , Burkot and colleagues [10] contend that fewer gametocytes would be ingested per meal , resulting in lower mosquito infection rates . Anopheles darlingi , the primary regional malaria vector in the Amazon Basin , is anthropophilic in the Iquitos region [11] , although both human biting rate ( HBR ) and entomological inoculation rate ( EIR ) vary widely [12] depending on the setting [13–15] . The An . darlingi feeding site in this region is exophagic and/or endophagic , depending on local circumstances ( e . g . , vegetation cover , type of house ) and host availability [11 , 12 , 14 , 15] . In 2015 , Loreto Department reported 95% of the total malaria cases in Peru ( 59 , 349 of 62 , 220 total ) with Plasmodium vivax as the most prevalent human parasite followed by P . falciparum , with 46 , 924 and 12 , 425 cases , respectively [16] . Parker and collaborators [13] demonstrated that high HBR , EIR , and infectivity of An . darlingi are a signature of remote riverine malaria hot spots and hyperendemicity in certain areas of the Peruvian Amazon , upending previous notions that transmission is hypoendemic throughout the peri-Iquitos region [11 , 12] . Recent studies also detected very high seasonal HBR and moderate EIR in the peri-Iquitos region [14 , 15] . Most malaria cases occur during the rainy season , from December to June [17] and a correlation was detected between An . darlingi abundance and peak river levels , but there was no significant correlation between river level and malaria case numbers [12 , 14 , 15] . In this last study , mosquitoes positive for Plasmodium were collected in peridomestic areas within approximately 10 m of the main house entrance , ( a caveat being that very few An . darlingi were found indoors despite extensive searching ) , suggesting that most malaria is transmitted exophagically , where humans have little protection against mosquito bites . Despite being the dominant malaria vector in Amazonia , few studies have documented the blood-meal sources for An . darlingi . In Amapá state , Amazonian Brazil , an ELISA analysis found that 13 . 1% of blood-meals were human; most resting An . darlingi had fed on cattle , pigs and dogs [18] . Notwithstanding the relatively low level of HBI , these communities are endemic for malaria , and An . darlingi is considered to be the most effective local vector [19] . In Peru , no studies have been published on the identity of An . darlingi blood-meals , but potential non-human hosts in rural residences near Iquitos include common peridomestic animals , dogs and chickens , and several potential wild mammalian hosts [12] . Although resting mosquitoes are optimal for calculating HBI , adequate sample sizes can be difficult to obtain in some habitats [18–20] . Little information exists on host preference and resting behavior of An . darlingi . The location of resting sites of An . darlingi could be useful for focal vector control if such mosquitoes are clustered non-randomly in the landscape . The development of barrier screens as a method for collecting anophelines outdoors has been tested successfully in the South East Pacific [20] and recently in southern Zambia [21] . This study was designed to address the following questions regarding An . darlingi feeding behavior in the Peruvian Amazon: i ) are barrier screens a useful tool to collect resting blood-fed An . darlingi in the area; ii ) what is the degree of anthropophily ( HBI ) in An . darlingi in contrast to more opportunistic behavior; iii ) what is the influence of available host biomass and iv ) is there evidence of seasonal age-structure in An . darlingi .
This study was approved by the Human Subjects Protection Program of the University of California San Diego , La Jolla , California and by the Ethical Boards of Universidad Peruana Cayetano Heredia and Asociación Benéfica PRISMA , Lima , Peru . The strategy of the barrier screen method of collecting mosquitoes outdoors is to intercept and capture mosquitoes transiting between blood feeding and resting sites [20] . Two possible scenarios can be identified: 1 ) intercepting mosquitoes entering a village seeking a blood-meal after emergence or oviposition; and 2 ) intercepting blood-fed mosquitoes leaving the village and seeking resting sites for egg development ( swamp , creek , stream , forest ) . In this Peruvian study , barrier screens were placed to intercept mosquitoes flying between house-forest and house-river depending on the specific characteristics of the locality . Mosquito collections were performed in three villages in Loreto Department: Lupuna ( LUP ) and Cahuide ( CAH ) in the peri-Iquitos area , and Santa Emilia ( SEM ) , in a remote area ~150 km from Iquitos ( Fig 1 ) . Detailed descriptions of these villages are in [15 , 22] . In 2013 , from March to May , a pilot study was conducted using a single screen in LUP and CAH placed at different points within each village ( between the creek/river and village houses ) . Specimens were collected for 4 nights ( 6PM- 6AM ) each month . Each barrier screen was constructed from a lightweight window screen mesh approximately 15 m long and 2 m high ( S1 Fig ) . Screens were then attached to poles with thin wire . Permission from the inhabitants/owners was obtained prior to any activity , including setting up the barrier screens and performing mosquito collections . Resting mosquitoes from the barrier screens were sampled by manually searching the surface of the screens with a mouth aspirator every hour for 15 minutes on each side , and the location ( next to house , forest or river ) and height ( ˃ or ˂ 1m above ground ) of mosquitoes was recorded . Mosquitoes were captured and stored by hour of collection and screen side separately . In 2014 ( monthly ) and 2015 ( January-June ) , the design was slightly modified to include four barrier screens in LUP and CAH to better represent the An . darlingi population in each locality . When multiple screens were used per village , data from each screen was maintained separately . In SEM , a remote village along the Nahuapa River , collections were performed with two barrier screens for two nights in May-June 2014 and May-September 2015 . Additionally , in 2015 , daytime mosquito collections ( 6AM-6PM ) with barrier screens were performed two days monthly from January-June in LUP and CAH , and from May-July in SEM . Screen orientation , wind speed and direction were recorded for every collection with a Windmate 300 Wind/Weather Meter . A census questionnaire of domestic hosts present in the study villages was performed in October 2014 in LUP and CAH and May 2015 in SEM ( S1 Table , Fig 2 ) . Because the first study was performed a year prior and the animal composition could have changed , the questionnaire included a retrospective question to assess the presence of potential past hosts . All specimens collected were morphologically identified using entomological keys [23–25] and abdominal status recorded ( unfed , blood-fed or gravid ) . Mosquitoes were stored and labeled individually with silica gel and placed at 4°C until subsequent analysis . To estimate the female age composition of the population , in March-April 2014 and February-June 2015 in LUP and CAH a proportion of females were dissected to determine the parity rates per hour , trap and side of trap [26] . Parity is also used as an indicator of mosquito survival under natural conditions . Mosquito longevity ( life expectancy ) was estimated using Davidson’s methodology ( 1954 ) Age=1loglP , where l is the natural logarithm of the constant P ( daily survival rate ) . ( P ) was calculated P = PRgc , where PR is the ratio of parous mosquitoes and the total number of females dissected , and gc is the duration of the gonotrophic cycle in days [27] . A limitation of this calculation is the assumption of accurate estimates of the length of the gonotrophic cycle . We have assumed that two or more blood-meals are required for the first oviposition and that the temporal feeding pattern is not regular , and therefore , we followed the method of calculations proposed by Garret-Jones and Grab [28] . Various studies have estimated the gonotrophic cycle of An . darlingi to be 2–3 days [29 , 30 , respectively] . Recently , it was calculated to be 2 . 19 days in the rainy season and 2 . 43 in the dry season [31] . Calculations in our study were performed using the 2 . 19 day estimate based on the timing of our An . darlingi collections ( the rainy season ) . Individual An . darlingi were bisected between the head/thorax and abdomen and DNA was extracted manually using the DNeasy Blood & Tissue kit ( Qiagen ) . A PCR-RFLP protocol was performed to detect the most common host in the area [32] for all mosquito abdomens in 2013–2015 , except for a subsample ( 60% ) of mosquitoes collected in LUP 2014 ( due to a extended sample size ) . In addition , samples were tested for Galliformes ( Gallus gallus and turkeys; see census and proportion of chickens; Fig 2 , S1 Table ) following [33] , rat and didelphis [34] , and monkey [35] . A subsample of the unidentified blood samples was sequenced for the mitochondrial COI gene [36] and then compared with sequences in GenBank using BLASTn ( http://www . ncbi . nmln . nih . gov ) or BOLD SYSTEMS v2 . 5 ( http://www . barcodinglife . org ) . The best match with identity of 95% or above was recorded . Detection of Plasmodium infection was conducted using real-time PCR of the small subunit of the 18S rRNA , with a triplex TaqMan assay ( Life Technologies ) , as described in [37] . First , DNA was extracted from each specimen of An . darlingi , then the RT-PCR was conducted on pools of DNA of head/thoraces of five mosquitoes , and finally the pools were analyzed for detection of P . vivax and P . falciparum . Specimens from positive pools were tested individually to calculate infection rate ( IR ) . HBI was calculated as the proportion of mosquitoes fed on a specific host divided by the number of mosquitoes analyzed ( mixed blood-meals were added to totals of each host ) . To adjust the HBI , mosquitoes with unidentified blood-meals were excluded . This index was calculated monthly in each locality and Chi-square ( χ2 ) analyses were performed to compare statistical differences temporally and among sites . Host data recorded in the census was used for the calculation of the forage ratio ( wi ) [38 , 39] and selection index ( Bi ) [40] , to quantify the preference of mosquitoes for available blood resources . The forage ratio for species i was calculated as wi=oipi , where oi is the proportion of host species i in the blood-meals , and pi is the proportion of available host in the environment . Forage ratios >1 . 0 indicate preference and < 1 . 0 avoidance and selection of another host; ~1 . 0 means neither preference nor avoidance . The selection index Bi was calculated with the formula Bi=wi∑i=1nwi , where wi is the forage ratio for species i and n is the number of blood sources available . Wind speed was measured at 6:00pm , 12:00am , and 6:00am each collection night in LUP , CAH , and SEM in 2015 . For each collection night , mosquito density was aggregated into four 3-hour collection periods ( 6-9pm , 9pm-12am , 12-3am , and 3-6am ) . The wind speed at 6:00pm was assigned to the 6-9pm collection time , the wind speed at 12:00am was assigned to the 9pm-12am and 12-3am collection times , and the wind speed at 6:00am was assigned to the 3-6am collection time . The mosquito density was plotted against wind speed for each collection period at each location ( n = 48 collection periods each for LUP and CAH , and 40 collection periods for SEM ) using the ggplot2 package in RStudio v0 . 98 . 1091 [41] . A null-model analysis was used to test whether An . darlingi feeding habits were random or structured among the three villages , as in [36] and [42] . All specimens with identified blood-meals from 2013–2015 for LUP , 2013–2015 for CAH , and 2014–2015 for SEM were included , and specimens with mixed blood-meals were counted once for each host identified in the blood-meal . We calculated a C-score comparing the blood- meal sources of mosquitoes from the three villages using Ecosim 7 . 0 and we used the R bipartite package [43] to generate a host-vector quantitative interaction network for the three localities , as in [36] .
In 2013 , all specimens caught on the screens were collected and identified to determine the potential use of screens for collecting not only Anophelinae but also other Culicidae , potential vectors of parasites and arboviruses . A total of 322 mosquitoes in LUP and 514 in CAH were collected in 6 nights ( 72 h collection ) ( Table 1 ) ; 94 . 4% ( 304/18 ) of mosquitoes collected in LUP and 89 . 7% ( 461/53 ) of all mosquito species in CAH were females . Anopheles darlingi comprised 78 . 9% and 61 . 5% of these collections in LUP and CAH , respectively , and Culex quinquefasciatus was the second most common species identified in both localities ( Table 1 ) . Only one additional species of anopheline , Anopheles forattini , was identified ( in LUP ) . With respect to screen position , in LUP 63 . 4% of the An . darlingi were collected on the side facing the houses ( In ) and 36 . 6% on the side facing the creek ( Out ) , although this difference was not significant ( Kolmogorov-Smirnov test; p = 0 . 4 ) . On both sides of the screen , most of the specimens were collected <1m from the ground ( Below; Table 2 ) ( range 76 . 5–90 . 2% ) . In CAH , 61 . 8% of the mosquitoes were collected on the house side and 38 . 2% on the creek side , and 93 . 1% and 84 . 5% ( In and Out , respectively ) were caught <1 m from the ground . No differences were found between LUP and CAH for side of the barrier screen . Only 1 . 62% in LUP and 6 . 57% in CAH of the An . darlingi females were determined by visual inspection to be blood-fed , with no differences between screen sides ( Table 3 ) . In 2014 , using multiple barrier screens per locality , a total of 4 , 593 An . darlingi females were collected in LUP , 175 in CAH and 216 in SEM ( Table 2 ) . One specimen of Anopheles dunhami in LUP and eighteen Anopheles benarrochi B in SEM were also identified as in [14] . In LUP , no significant differences were detected between the sides of four screens tested independently . However , when data were grouped over months there was a significant difference between mosquitoes collected on the side of the houses ( In ) and creek/vegetation side ( Out ) ( Wilcoxon test; p = 0 . 0313 ) . In CAH , the four barrier screens were not homogeneous , with significant differences in number of mosquitoes collected from each side ( K-S; In: p = 0 . 0082 and Out: p = 0 . 0356 ) , and when In/Out were compared by month ( K-S; p = 0 . 0022 ) . There were also significant differences between collections in LUP and CAH ( K-S , p = 0 . 0336 ) . In SEM , captures in May ( two screens ) and in June ( four screens ) , were not significantly different between screens . In 2015 , in LUP , 1 , 019 female mosquitoes were collected , 233 in CAH and 277 in SEM . Most specimens were captured resting < 1m from the ground with little variation among years and sites ( Table 2 ) . Differences in mosquito density by time of collection and side of barrier screen were tested ( Fig 3 ) with time of collection split into four three-hour periods ( 6-9pm , 9pm-12am , 12-3am , and 3-6am ) . In both LUP and CAH in 2015 , there was a significant difference in the distribution of mosquito collection location ( side of screen ) by time period ( Kruskal-Wallis p < 0 . 0001 for both sites ) , with higher proportions of mosquitoes found on the In ( facing house ) side of the screen from 9pm-12am and 12-3am than from 6-9pm and 3-6am . In LUP and CAH in 2013 and 2014 , and in SEM in 2015 , there was no significant difference in mosquito density by time of collection ( Kruskal-Wallis p>0 . 05 ) . Plots of mosquito density against wind speed for each locality in 2015 are shown in Fig 4 . Overall , there was a negative but non-significant correlation between mosquito density and wind speed ( Pearson’s r = -0 . 09 , p = 0 . 3 ) . The correlation between mosquito density and wind speed was also negative in LUP ( Pearson’s r = -0 . 25 , p = 0 . 1 ) and SEM ( Pearson’s r = -0 . 27 , p = 0 . 09 ) , but was positive in CAH ( Pearson’s r = 0 . 14 , p = 0 . 34 ) ( Fig 4 ) . To investigate the diurnal behavior of An . darlingi , barrier screen collections were performed in LUP and CAH from January to June , and in SEM from May to June from 6AM to 6PM twice January-June 2015 . In LUP a total of 59 An . darlingi were collected during this period and female activity was reported from 6AM to 9AM and from 2PM to 5PM . In CAH , the number of collected specimens was 23 , with an activity similar to LUP . In SEM , 33 mosquitoes were collected , with an extension of the flying activity until 8AM , and beginning again in the evening at 4PM . In LUP , 20 . 3% , in CAH , 34 . 8% and in SEM 54 . 5% of diurnal An . darlingi specimens were collected on the house side ( In ) . A total of 583 An . darlingi females from LUP were dissected in 2014 ( 12% of the total ) and 19 in CAH ( 11% ) ; in 2015 , n = 633 in LUP ( 62% ) and n = 153 ( 65% ) in CAH were dissected . The monthly mean parity rate in LUP in 2015 was ~ 55% ( range 45 . 6–66 . 7 ) and in CAH it was ~ 51% ( range 27 . 8–64 . 5 ) ( Table 4 ) . No significant differences were found between months or between localities , although in February , the rate was slightly higher compared to June . Mosquito age in LUP in March—April 2014 was 7 . 47 and 14 . 21 days , respectively , whereas in 2015 it ranged from 14 . 21–23 . 90 days . In CAH , mosquitoes collected in March 2014 were estimated to survive 14 . 98 days , and between 3 . 73–20 . 24 days in 2015 ( Table 4 ) . Blood-meal source was determined for 4 , 417 An . darlingi females ( S2 Table ) . A total of 3 , 214 mosquitoes from LUP , 729 from CAH and 474 from SEM were analyzed . Single-host blood-meals were the highest percentage among the blood-meals detected ( 69 . 98% ) and human was the most common blood source ( 42 . 5% ) , followed by Galliformes ( 25 . 1% ) and dog ( 1 . 42%; Fig 5 ) . Only 4% of the samples could not be identified to blood-meal source . Multiple blood-meals were found in 1 , 272 mosquitoes and accounted for 30% of the blood- meals , with 1 , 262 double feeds in the three localities , and triple feeds ( n = 10 ) only identified in LUP . In total , seventy-three samples with non-identified blood-meal source by PCR-RFLP , were sequenced for 16S ribosomal DNA [36] and mammalian cytochrome-b [32] . Only ten were identified as of human origin with the 16S protocol , whereas 23 were consistent with human for cytochrome-b . The distribution of blood-meal source in An . darlingi presented little temporal or spatial variation . Evaluation of the proportion of feeds on single different hosts showed that in LUP , no significant differences between years were detected by one-way ANOVA analysis; paired Wilcoxon-tests were not significant when comparing years 2013–2014 with 2015 or 2013 and 2014 . In CAH , no significant differences between the years 2013–2014 , 2014–2015 or among the 3 years were found . In SEM , a non-parametric Mann-Whitney test was not significant comparing 2014 and 2015 . For locality comparison , data from the same years and different localities were compared . In 2013 , there were no significant differences between LUP and CAH , and in 2014 and 2015 a one-way ANOVA test did not show differences between sites . HBI was calculated monthly ( S3 Table ) and annually ( Table 5 ) per locality . In 2013 , no significant differences were detected in LUP or CAH . Mean HBI per year was non-significant among localities ( LUP , CAH , SEM ) and years 2014–2015 . The Forage Ratio and Host Selection Index were calculated , accounting for single and multiple blood-meals ( Table 6 ) . Humans were the preferred source , closely followed by Galliformes , in all three settings for both years . When the Forage Ratio was analyzed , the weight per host was used instead of the numerical presence at the site [36] ( S4 Table ) , Galliformes were by far the preferred host , with humans as the second most favoured . For example in LUP , the Galliforme forage ratio ranged from 10 . 35 to 17 . 96 and the human forage ratio from 0 . 58–0 . 72 . The null model test indicated that the mosquito feeding patterns were aggregated among the localities , indicating that diet overlapped more than expected between the localities , although this finding was only marginally significant ( C-score: 0 . 33 , p = 0 . 08 ) . The quantitative interaction network of blood-meal source by locality ( Fig 6 ) supported patterns of organization based on the above-mentioned trophic preferences ( humans and Galliformes ) from the three mosquito populations ( LUP , CAH , SEM ) . A total of 5 , 387 , 362 and 455 mosquitoes in LUP , CAH and SEM , respectively , collected on barrier screens , were tested for Plasmodium . The Infection rate ( IR ) of mosquitoes varied among sites and seasons , ranging from 0 . 20–3 . 85 in LUP , 0 . 51–14 . 3 in CAH and 0–2 . 04 in SEM ( Table 7 ) . A logistic regression model analysis determined that IR was significantly higher in CAH ( p = 0 . 02 ) and SEM ( p = 0 . 003 ) vs . LUP . No specimens from the diurnal collections in the three localities ( n = 116 ) were positive for P . vivax , independent of the collection season .
Ours is the first study to conclusively demonstrate that An . darlingi readily feeds on Galliformes . Overall , the feeding preference of An . darlingi in the Peruvian Amazon is more variable than previous studies have assumed . In addition , a consistent pattern of blood-meal source was observed at each site every year of collection: mosquitoes feeding only on humans , only on chickens , or on both hosts . This consistency could suggest the co-occurrence of different subpopulations within a metapopulation , with local adaptation as the main driving force . A single metapopulation was initially detected in An . darlingi in the Iquitos area with AFLPs [44] and microsatellite markers [45] . However , using 2x the number of microsatellites , a population replacement event was detected between 2006 and 2012 and two subpopulations were detected , one significantly more prevalent in highway compared with riverine habitat [20] . This recent genetic structure could explain some of the heterogeneity in feeding preferences of An . darlingi among localities [45 , 46] . Additional studies , focused on intrinsic host preference , vector density and social practices of the human population might elucidate the basis for the described behavior and whether some An . darlingi populations are under selective pressure for host preference or whether this pattern is strongly correlated with host availability . Similar HBI across the dry and rainy seasons and between populations infers that mosquitoes maintain their host preference behavior independent of local ecological conditions . In an earlier investigation of HBI of An . darlingi in riverine villages in Amapá State , Brazil [18] , researchers reported high among-village variance ( HBI 0 . 131–0 . 435 ) and ~10% of mixed blood-meals overall , mainly from cattle and pigs . In contrast , in our study , there was virtually no variance in HBI among localities , HBIs were higher ( 0 . 58–0 . 79 ) and ~30% of blood-meals were mixed , with Galliformes as the primary alternate host . Because HBI is an integral parameter of the vectorial capacity formula ( the daily rate of malaria transmission from a single infected human , assuming every bite from an infected mosquito leads to transmission ) [2] , our data suggest that An . darlingi is a more effective vector in the peri-Iquitos area compared with Amapá state , Brazil . Curiously , in Tanzania , An . arabiensis avoids , and may be repelled by , the volatiles of chickens [47] . Subgenera Nyssorhynchus ( An . darlingi ) and Cellia ( An . arabiensis ) were estimated to have diverged ~94 million years ago [48]; therefore their olfactory responses are expected to have evolved differentially . The present study provides evidence of the successful use of barrier screens to collect blood-fed An . darlingi mosquitoes in Amazonian Peru . Initially , in 2013 , we conducted preliminary barrier screen collections with Procopack aspirators in LUP and CAH from 5 to 8 AM for 6 days/collection in March-May in at least 10 houses each time , but only one An . darlingi specimen was caught . Interestingly , in Iquitos the Procopack effectively collected indoor resting Culicidae including Aedes aegypti and Culex pipiens complex [49] . One explanation for our failure to find An . darlingi using the Procopack despite extensive searching could be due to its singular resting and biting behavior in this region . Anopheles darlingi resting behavior varies across its range [50]: in Venezuela , Guyana [51] and in Brazil , in Matto Grosso and in southern Amazonas [52 , 53] it rests during the day inside houses ( endophily ) . In contrast , in Suriname , using exit traps , a peak departure from the dwelling was observed at sunrise [54] and in Brazil An . darlingi was resting indoors only at night [55] . In Amapá state , Brazil , resting mosquitoes were collected after sunrise ( 6AM-7AM ) under houses and in peridomestic vegetation [18] . In French Guiana , no resting An . darlingi were collected indoors after pyrethroid spray , from pit-shelters or in the shade in the peridomestic area [56] . In our study , overall differences detected between screen sides may reflect the relative nearness of screens to houses , resulting in the interception of a higher proportion of blood fed An . darlingi leaving the peridomestic area , compared with questing females , entering the village from numerous resting and/or breeding sites . In CAH , we hypothesize that additional differences among screens and between months could result from a much smaller population of An . darlingi intercepted in this village . Our results constitute a major accomplishment: the use of barrier screens in this setting to overcome the difficulty of performing host-independent sampling for determining blood-meal sources . The success of individual mosquito blood-meal identification in this study ( range of 92 . 2–99 . 3% ) , was remarkably high when compared to visually blood-fed mosquitoes ( 0 . 92%-14 . 44% ) . When analysis is restricted only to the latter , information from partial blood-meals or partially digested blood is missed , leading to underestimation of the proportion of host sources ( up to 18 . 7% ) ; hence , a miscalculation of HBI [57] . One limitation of our study was the lack of identification of potential wild animal hosts; use of novel targeted high-throughput sequencing [58] would rectify this . In LUP , the age of the mosquito population at each time point is enough to sustain the sporogonic cycle of P . vivax ( range 7 . 24–9 . 13 days; calculated by the Moshkovsky method in [31] ) , whereas in CAH the population is , in general , younger , but with non-dangerously aged mosquitoes only in May and June . The proportion of young females might be explained by differential dispersal and aggregation of different age classes of An . darlingi populations , as previously reported for An . farauti in Papua New Guinea [59] . Use of 2 . 19 days of the gonotrophic cycle [31] could have produced a miscalculation in the age parameter . For instance , gravid females may experience delays while searching for suitable oviposition sites or there could be variation in extrinsic environmental conditions within this population of An . darlingi [60] . Because of the natural development of the parasite within the mosquito , a longer life-span is related to a higher potential to transmit malaria [61] . Parity is also associated with seasonality , i . e . , mosquitoes generally survive longer during the rainy season [62 , 63] , but see [64] . Overall , our study provides unreported information of the blood-meal preferences of An . darlingi in the peri-Iquitos area , which will be the base-line to compare potential changes in the behavior of these mosquito populations . HBI , together with other malaria metrics such as HBR or EIR , should be taken into consideration for surveillance and epidemiological studies of malaria transmission . | Anopheles darlingi is the major malaria vector in the Amazon . This species has been commonly described as highly anthropophilic throughout its geographic range , although little is known about its feeding preferences . Scant information is available regarding the origin of An . darlingi blood-meals . In the context of malaria elimination programs , the Human Blood Index ( HBI ) may provide crucial information regarding mosquito-human contact related to transmission dynamics . Additionally , collection of resting An . darlingi is challenging , mainly because the resting behavior of this species has not been well characterized . Our study , conducted from 2013–2015 in three localities in Loreto Department in the Peruvian Amazon , showed for the first time the efficacy of the barrier screen methodology for collecting recently blood-fed An . darlingi in a neotropical setting for the purpose of identifying the source of their blood-meals . Our data show that An . darlingi feeds on humans , Galliformes , dogs , pigs and goats , and that 30% of the mosquitoes fed on more than one type of host . Despite this opportunistic feeding behavior , however , An . darlingi is primarily anthropophilic . We hypothesize that mosquito population structure is associated with feeding preferences , which may affect the pattern of malaria transmission in the area . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"plasmodium",
"atmospheric",
"science",
"tropical",
"diseases",
"vertebrates",
"parasitic",
"diseases",
"animals",
"parasitology",
"apicomplexa",
"insect",
"vectors",
"birds",
"behavior",
"epidemiology",
"gamefowl",
"pathogenesis",
"fowl",
"disease",
"vectors",
"insects",
"hematology",
"arthropoda",
"mosquitoes",
"blood",
"anatomy",
"meteorology",
"host-pathogen",
"interactions",
"physiology",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"malaria",
"wind",
"amniotes",
"organisms"
]
| 2017 | Intensive trapping of blood-fed Anopheles darlingi in Amazonian Peru reveals unexpectedly high proportions of avian blood-meals |
HIV-1 infection is associated with a progressive loss of T cell functional capacity and reduced responsiveness to antigenic stimuli . The mechanisms underlying T cell dysfunction in HIV-1/AIDS are not completely understood . Multiple studies have shown that binding of program death ligand 1 ( PD-L1 ) on the surface of monocytes and dendritic cells to PD-1 on T cells negatively regulates T cell function . Here we show that neutrophils in the blood of HIV-1-infected individuals express high levels of PD-L1 . PD-L1 is induced by HIV-1 virions , TLR-7/8 ligand , bacterial lipopolysaccharide ( LPS ) , and IFNα . Neutrophil PD-L1 levels correlate with the expression of PD-1 and CD57 on CD4+ and CD8+ T cells , elevated levels of neutrophil degranulation markers in plasma , and increased frequency of low density neutrophils ( LDNs ) expressing the phenotype of granulocytic myeloid-derived suppressor cells ( G-MDSCs ) . Neutrophils purified from the blood of HIV-1-infected patients suppress T cell function via several mechanisms including PD-L1/PD-1 interaction and production of reactive oxygen species ( ROS ) . Collectively , the accumulated data suggest that chronic HIV-1 infection results in an induction of immunosuppressive activity of neutrophils characterized by high expression of PD-L1 and an inhibitory effect on T cell function .
Neutrophils , the most abundant leukocyte population , are traditionally recognized as essential effector cells of the innate immune system in the host defense against invading pathogens [1] . In recent years , a new appreciation of the role of neutrophils in interacting with and regulating the adaptive arm of the immune system has emerged [1] , [2] . Neutrophils co-localize and actively communicate with T cells at sites of infection and migrate to the draining lymph nodes where they are involved in the induction and regulation of cellular and humoral immune responses by exerting pro-inflammatory or anti-inflammatory function [2]–[4] . Accumulating evidence supports the role played by neutrophils in the negative regulation of T cell function via production of reactive oxygen species ( ROS ) and arginase-1 [2] , [5]–[7] . A recent study has identified an immunosuppressive population of CD16+CD62Llow neutrophils that is induced in human volunteers following injection of a low dose of bacterial lipopolysaccharide and inhibits T cell function by local release of hydrogen peroxide into the immunological synapse between the neutrophil and T cell [7] . A population of cells referred to as myeloid-derived suppressor cells ( MDSCs ) has been identified in peripheral blood mononuclear cells ( PBMCs ) in multiple pathological conditions involving inflammation including cancer , chronic bacterial and viral infection , trauma , and sepsis [6] , [8] . MDSCs have been shown to serve as a negative feedback mechanism preventing potential damage caused by acute and chronic inflammation . Data recently obtained in sepsis , chronic inflammatory conditions and several types of cancers demonstrate the presence of a population of MDSCs of granulocytic origin ( G-MDSCs ) . G-MDSCs likely originate from circulating neutrophils that acquire low density neutrophil ( LDN ) phenotype and co-segregate in the PBMC fraction on a density gradient [6] , [8]–[10] . It is unclear at present whether LDN/G-MDSCs originate by granulopoiesis from dedicated suppressive progenitors in the bone marrow or whether they represent a functional subset of neutrophils that acquired the immunosuppressive phenotype in response to specific signals in the periphery [6] . G-MDSCs display a remarkable ability to suppress T cell-mediated immune responses by multiple mechanisms including release of arginase-1 resulting in a depletion of arginine and downregulation of TCR ζ chain , production of reactive oxygen species ( ROS ) , production of regulatory cytokines , and induction of regulatory T cells [6] , [8] . CD8+ and CD4+ T cells play a key role in controlling HIV-1 replication and progression to AIDS . However , HIV-1 infection is associated with a progressive loss of T cell functional capacity including decreased responsiveness to antigenic stimuli , lowered capacity to produce cytokines , and reduced proliferative and cytotoxic activity [11]–[15] . Loss of CD4+ T cells and functional impairment of HIV-1-specific CD8+ and CD4+ T cells eventually results in a failure of host immune system to maintain control of HIV-1 leading to an accelerated disease progression . HIV-1-specific T cells from rapidly progressing patients exert decreased cytotoxic and proliferative activity and produce reduced levels of TNFα , IL-2 , IFNγ , and CD107a compared to T cells from non-progressors [12] , [13] . T cell exhaustion in HIV-1 infection is associated with increased expression of programmed death-1 ( PD-1 ) and CD57 on the surface of CD4+ and CD8+ T cells [14]–[17] . Binding of PD-1 on T cells to the inhibitory ligand PD-L1 expressed on cells of myeloid lineage including myeloid dendritic cells ( DCs ) , monocytes , and macrophages negatively regulates T cell proliferation and production of effector cytokines [18]–[21] . In simian immunodeficiency virus ( SIV ) -infected rhesus macaques , high PD-1 expression is associated with an impaired response of SIV-specific T cells during both acute and chronic infection . Importantly , blocking of PD-L1/PD-1 axis in vivo in chronic SIV infection restores the function of SIV-specific cellular and humoral immune responses , improves viral control , and reduces immune activation [22]–[24] . These observations support a critical role of PD-L1/PD-1 interaction in the regulation of immune environment in HIV-1 infection and serve as a basis for ongoing clinical trials assessing the therapeutic potential of blocking PD-L1/PD-1 signaling in chronically infected individuals . Here we demonstrate that neutrophils in the blood HIV-1-infected individuals express high levels of surface PD-L1 and suppress the function of T cells via ROS and PD-L1/PD-1 pathways . The upregulated expression of PD-L1 on neutrophils correlates with the presence of LDNs that co-segregate in the mononuclear cell fraction on a density gradient and express G-MDSC phenotype . Neutrophil PD-L1 is induced directly by HIV-1 virions , IFNα , TLR-7/8 ligand R848 , and LPS . The presented data suggest that the induction of PD-L1 on neutrophils , the most abundant leukocyte population , in concert with high PD-1 expression on T cells significantly contributes to the ongoing T cell exhaustion and immune suppression in HIV-1 infection .
To characterize the phenotype of neutrophils in freshly obtained blood of HIV-1-infected individuals , multiparameter flow cytometry analysis of SSChigh CD15+ CD33+/dim CD11b+ neutrophil population was performed to determine the levels of surface markers including CD11b , CD15 , CD16 , CD33 , CD80 , CD86 , CD115 ( M-CSFR ) , HLA-DR , and PD-L2 ( Figure S1 ) . No significant differences in the levels of expression of these markers were observed between HIV-1 patients and uninfected controls . In contrast , circulating neutrophils from HIV-1-infected patients expressed significantly elevated levels of PD-L1 compared to neutrophils from healthy uninfected donors , irrespective of antiretroviral therapy ( ART ) status ( p = 0 . 02 and 0 . 002 in individuals on and off ART , respectively; Fig . 1A , C ) . Neutrophil PD-L1 expression was significantly higher in patients with HIV-1 viral load >2 , 000 copies of viral RNA ( vRNA ) per ml of plasma compared to patients that successfully controlled viral replication ( p = 0 . 04; Fig . 1E ) . No increase in PD-L1 expression was observed in elite controllers ( EC ) restricting viral proliferation below 50 vRNA copies/ml in an absence of ART . The extent of the increase of the neutrophil PD-L1 expression was comparable to the increase observed on CD14+ monocytes from HIV-1-infected individuals ( Fig . 1B , D , F ) , as published previously [25] , [26] . The analysis revealed a trend to a direct correlation between HIV-1 viral load and PD-L1 expression on blood neutrophil population; however , the trend has not reached statistical significance ( p = 0 . 1 ) . To address whether neutrophil PD-L1 expression is modulated by ART , fresh blood from 5 HIV-1-infected subjects was analyzed before and after the initiation of ART resulting in a successful control of HIV-1 replication . PD-L1 expression was significantly reduced following ART implementation on both neutrophil and monocyte populations ( Figure 1G , H ) . Previous studies have identified a population of suppressive neutrophils that acquire the phenotype of low-density neutrophils ( LDN ) , co-segregate with PBMCs on a density gradient , and display the phenotype of G-MDSCs [6] , [8]–[10] , [27] . Here we show that HIV-1-infected individuals display higher frequency of LDNs in PBMCs compared to healthy donors ( p<0 . 001; Fig . 2A ) . Importantly , elevated PD-L1 expression on neutrophils in whole blood correlates with the frequency of CD15+ LDNs in PBMCs ( R = 0 . 6; p = 0 . 01; data not shown ) . Surface expression of PD-L1 on LDNs was not significantly different from that on blood neutrophils . LDNs expressed elevated levels of CD15 , CD33 , and CD66b and lower levels of CD62L , CD80 , CD114 , and CXCR4 compared to whole blood neutrophils ( N . B . , Z . H . , unpublished data ) [28] . Depletion of CD15+ LDN cells from PBMCs of HIV-1-infected donors resulted in an increase in the percentage CD8+IFNγ+ T cells ( Fig . 2B , C; p = 0 . 02 ) , a trend towards an increase in the frequency of CD4+IFNγ+ T cells ( 3-fold mean increase; p = 0 . 6 ) , and an increase in IFNγ production ( Fig . 2D; p = 0 . 007 ) in response to stimulation with HIV-1 Gag overlapping peptide pool . Similar results were obtained following non-specific stimulation with PHA or microbeads coated with antibodies against CD3 and CD28 antigens suggesting that the LDN-mediated inhibition is independent of specific antigen presentation [6] , [7] , [10] . Analysis of freshly prepared PBMCs from HIV-1-infected individuals demonstrated elevated levels of staining with DCF-DA in LDNs , an indicator of ROS production , compared to uninfected volunteers ( p = 0 . 008 ) and a significant correlation between DCF-DA staining and elevated levels of PD-L1 on neutrophils ( R = 0 . 9; p = 0 . 01; data not shown ) . It has been previously demonstrated that the production of ROS represents a major mechanism of neutrophil and LDN/G-MDSC –mediated suppression of T cell function [6] , [7] . Since the PD-L1 on the surface of monocytes , DCs , and other cells of myeloid lineage negatively regulates T cell proliferation and production of effector cytokines [18]–[21] , we hypothesized that direct PD-L1/PD-1 interaction may contribute to neutrophil-mediated suppression of T cell function . To address the specific contribution of PD-L1 to neutrophil-mediated T cell suppression , polymorphonuclear cells ( PMNs ) purified from the blood of HIV-1-infected patients were depleted of any residual CD14+ monocytic cells and incubated for 24 hours with CD3/CD28-activated autologous T cells at 5∶1 ratio resulting in about 70% decrease in IFNγ production ( Fig . 2E ) . Importantly , the inhibition of IFNγ production was partially reversed in the presence of ROS scavengers in combination with antibodies blocking PD-L1 ( p = 0 . 05 ) but not control isotype antibodies . Reversal of inhibition by blocking PD-L1/PD-1 interaction was contingent on elevated PD-1 expression on T cells and PD-L1 expression on neutrophils ( MFI >4 , 500; three independent experiments using separate donors ) . PD-L1 blocking had little effect on neutrophils ( PMNs ) from HIV-1-infected patients or uninfected volunteers with low PD-L1 expression ( MFI <3 , 000 ) . The signaling events leading to increased PD-L1 expression on myeloid cells in HIV-1 infection are not completely understood . PD-L1 expression on monocytes and plasmocytoid dendritic cells ( pDCs ) was shown to be directly induced by HIV-1 virions and ligands of the TLR-7 and -8 receptors [25] , [26] . Furthermore , recognition of the single-stranded RNA of the HIV-1 genome by TLR7 and 8 in pDCs results in a production of IFNα that directly induces PD-L1 expression on monocytes and other cell types [29] . We determined the effect of HIV-1 virions , TLR-7/8 ligand R848 , and IFNα on PD-L1 expression on neutrophils . Stimulation with IFNα or TLR-7/8 ligand R848 results in a significantly increased expression of PD-L1 on LDNs that are present at low frequency in PBMCs of healthy donors as well as on purified CD15+ PMNs ( >95% purity; Fig . 3A , B ) . To directly test the effect of HIV-1 virions on PD-L1 expression , PBMCs or purified PMNs from healthy donors were incubated with AT-2-inactivated HIV-1 MN virions or control microvesicle preparation absent of viral proteins or RNA . Neutrophil ( LDN and PMN ) expression of PD-L1 was increased in a dose-dependent manner in response to the treatment with AT-2 HIV-1 ( Fig . 3C; p = 0 . 03 and 0 . 01 for CD15+ LDNs or PMNs , respectively ) . In addition , treatment with HIV-1 virions resulted in a decreased expression of CD62L on LDNs; other neutrophil surface proteins were not significantly modulated ( Figure S2A ) . Relative induction ( fold of increase ) of PD-L1 expression following stimulation with R848 or AT2 HIV-1 was higher in LDNs than in purified PMNs; this may reflect a contributing effect of a factor or factors produced by other cell population . A direct correlation between neutrophil PD-L1 expression and IFNα concentration in plasma was detected in patients with <500 CD4+ T cells/ml of blood ( R = 0 . 7; p = 0 . 04 ) . To assess the potential contribution of IFNα to the induction of PD-L1 , PBMCs from HIV-1-seronegative donors were cultured with R848 or HIV-1 in the presence of antibodies blocking the cellular receptor for IFNα ( IFNAR ) . The presence of IFNAR-blocking antibody partially inhibited HIV-1- and R848- induced PD-L1 expression on whole blood neutrophils ( indicating a partial contribution of IFNα to PD-L1 induction by TLR-7/8 ligands ( p = 0 . 04 and 0 . 003 , respectively; Figure S2B ) . Translocation of LPS and other microbial products from gut lumen across the damaged intestinal epithelial barrier contributes to the systemic immune activation observed in HIV-1 infection [30] . Interaction between LPS , CD14 , myeloid differentiation-2 ( MD-2 ) , and TLR-4 results in an activation of nuclear factor kappa-light-chain-enhancer of activated B cells ( NFκB ) signaling pathway and shedding of soluble CD14 ( sCD14 ) from myeloid cells . Plasma levels of sCD14 indicate the degree of bacterial translocation and independently predict disease progression in HIV-1 patients [31]–[33] . In agreement with previously studies , we observed elevated levels of plasma sCD14 in the cohort of HIV-1-infected volunteers compared to healthy uninfected donors ( p<0 . 001 ) . Pillay et al . have recently observed that an injection of low dose of bacterial lipopolysaccharide in human volunteers induces a suppressive subpopulation of neutrophils [7] . We have therefore addressed the hypothesis that LPS directly modulates PD-L1 expression on neutrophils . PD-L1 expression on PMNs of healthy donors was significantly increased following the stimulation with LPS ( Fig . 3D , E; p = 0 . 005 ) and reversed in the presence of polymyxin B ( Figure S2C ) . Although higher concentration of LPS was used in an in vitro experiment than is typically found in vivo , the effect of LPS on neutrophils is enhanced in vivo by an interaction with LPS-binding protein [34] . This data suggests that bacterial translocation may contribute to the elevated PD-L1 expression on neutrophils of HIV-1-infected patients at sites of high local concentration of LPS . In addition , treatment with LPS resulted in an upregulation of DC-SIGN and down-regulation of CD16 but no modulation of CD62L ( Figure S2A ) . T cell exhaustion in HIV-1 infection is associated with elevated expression of PD-1 on CD4+ and CD8+ T cells [16] , [17] . CD57 is expressed primarily on T cells at late or terminal stages of differentiation and marks a state of replicative senescence characterized by a loss of the proliferative and target cell killing capacity [35] , [36] . Here we show that PD-L1 expression on circulating neutrophils closely correlates with the expression of PD-1 on CD8+ and CD4+ T cells ( Fig . 4A; p = 0 . 005 and <0 . 001 , respectively ) and with the expression of CD57 on CD4+ T cells ( Fig . 4B; p = 0 . 03; gating strategy in Figure S3 ) . Interestingly , correlations of PD-1 and CD57 expression on CD4+ and CD8+ T cells with PD-L1 expression on neutrophils was stronger than the respective correlations with PD-L1 levels on monocytes ( Figure S4 ) . Neutrophil-derived arginase-1 is known to down-regulate the CD3ζ chain via the depletion of L-arginine resulting in T cell hyporesponsiveness in HIV-1-infected individuals [37] , [38] . Interestingly , high PD-L1 expression on neutrophils correlated with reduced CD3ζ expression on T cells ( Fig . 4C ) as well as with increased levels of arginase-1 in plasma ( Fig . 5A ) . Collectively , this data suggests that high PD-L1 expression on neutrophils correlates with the dysregulation of T cell function in HIV-1-infected subjects . Although the etiology of LDNs is unclear , some studies indicate that the LDN phenotype is acquired following neutrophil degranulation resulting in co-segregation with the PBMC fraction on density gradient [6] , [9] , [10] . The content of neutrophil primary , secondary , and tertiary granules is released in response to activation of neutrophils at the site of infection and inflammation and contributes to the creation of an antimicrobial milieu at the inflammatory site [1] . Plasma levels of neutrophil granule proteins myeloperoxidase ( MPO ) , neutrophil gelatinase-associated lipocalin ( NGAL ) , and arginase-1 are significantly higher in HIV-1-infected patients than in uninfected volunteers ( E . S . H . , Z . H . , unpublished data ) . Importantly , increased PD-L1 expression on neutrophils in whole blood correlated with plasma levels of neutrophil degranulation markers arginase-1 , MPO , and NGAL ( p = 0 . 05 , 0 . 002 , and 0 . 005 , respectively; Fig . 5A–C ) . In addition , circulating neutrophil PD-L1 expression in viremic ART-naive HIV-1-infected patients directly correlated with plasma levels of G-CSF ( R = 0 . 5; p = 0 . 05 ) , GM-CSF ( R = 0 . 7; p = 0 . 02 ) , and monocyte chemoattractant protein-1 ( MCP-1/CCL2 ) ( R = 0 . 7; p = 0 . 003; Figure 5D–F ) . These proteins have been demonstrated to play important roles in the recruitment , activation , and chemoattraction of neutrophils to the inflamed tissue . Collectively , these results indicate that elevated PD-L1 expression on neutrophils is closely associated with the production of factors involved in neutrophil recruitment and increased rate of neutrophil degranulation in vivo [1] , [2] .
The study presented here reveals four major findings: ( i ) neutrophils in blood of HIV-1-infected individuals express elevated level of PD-L1; ( ii ) the level of neutrophil PD-L1 expression correlates with the expression of PD-1 on CD8+ and CD4+ T cells and CD57 on CD4+ T cells and decreases following ART; ( iii ) PD-L1 on neutrophils is induced by multiple stimuli including HIV-1 , TLR-7/8 ligand R848 , IFNα , and LPS; and ( iv ) PD-L1/PD-1 pathway contributes to the suppression of T cell function by neutrophils . Taken together , these findings are consistent with a hypothesis that HIV-1 infection and ongoing microbial translocation induce neutrophils with an immunosuppressive activity that significantly contributes to the suppression of T cell function in HIV-1-infection . This novel mechanism of immune suppression mediated by neutrophils may alter our understanding of HIV-1 pathogenesis and result in a design of novel therapies targeting the loss of immune function in HIV-1-infected individuals . The results presented here are consistent with previous studies demonstrating the suppressive activity of activated neutrophils [2] , [6]–[9][10] . Since neutrophils readily interact with T cells in inflamed tissue and secondary lymphoid organs [2]–[4] , [7] , [39] , [40] , neutrophil PD-L1 is likely to significantly contribute to the PD-1-mediated suppression of T cell function . The data presented here are strongly supported by a report published at the time of submission of this manuscript demonstrating that IFNγ-stimulated neutrophils suppress T cell proliferation via the expression of PD-L1 [41] . Interestingly , PD-L1 expression is significantly elevated in the suppressive CD16+CD62Llow subpopulation of neutrophils that is induced in human volunteers following injection of a low dose of LPS [7] , [41] . Neutrophil-mediated immune suppression can be highly beneficial in acute sepsis [42] and acute viral infection [43] , [44] where it limits the damage caused by the host's inflammatory response and prevents excessive tissue damage . However , it exerts a detrimental effect in the context of prolonged immune activation such as chronic viral infections and cancer by inducing long-term attenuation of T cell functionality [1] , [5] , [7] . It is likely that immune suppression mediated by PD-L1 on neutrophils plays a role in the pathogenesis of other viral and bacterial infections . Increased expression of PD-L1 on neutrophils was recently described in patients with active tuberculosis [45] . Future in vivo studies utilizing murine and/or simian models will be critical to delineate the significance of neutrophil mediated suppression of T cell function via the PD-L1/PD-1 pathway . The results presented here are consistent with the study of Volbrecht et al . demonstrating an expansion of CD15+ CD33+/dim CD11b+ G-MDSC population in HIV-1 infection [27] and with the studies by Cloke et al . describing a population of activated low-density granulocytes/neutrophils ( LDNs ) in PBMCs of HIV-1-infected patients [28] , [46] . Similarly , studies in cancer patients showed an expansion of CD14−CD15+CD11b+ G-MDSC population with the ability to inhibit T cell function via ROS- and arginase-1-dependent mechanisms [6] , [9] , [10] , [47] . Distinct subpopulations of circulating neutrophils were identified by several studies; however , the phenotype and physiological function of these populations remain elusive [48] . In contrast to several previous studies in cancer but consistent with a recent study by de Kleijn et al . [41] , we demonstrate that certain phenotypic changes , such as elevated PD-L1 expression , occur in entire circulating neutrophil population and are not restricted to the subpopulation of LDN/G-MDSC neutrophils co-segregating with PBMCs . Although the relative contribution of LDNs versus blood neutrophils to immune regulation in HIV infection is unclear , we propose that the PD-L1-mediated suppression of T cell function is not restricted to LDNs and can be mediated by a significant part of the entire circulating neutrophil population . We propose that neutrophils represent a highly sensitive sensor of chronic inflammation and function as a negative feedback mechanism curbing the detrimental impact of systemic immune activation . Association between PD-L1high phenotype and systemic neutrophil activation and degranulation in vivo is supported by correlations with plasma levels of markers of neutrophil degranulation ( MPO , NGAL , and arginase-1 ) , production of factors involved in neutrophil recruitment and activation , and production of ROS ( Fig . 5 ) . These results are in agreement with a recent study suggesting that circulating neutrophils in HIV-1-infected patients become progressively more activated and degranulated with increased disease severity [28] . Thus , elevated PD-L1 expression may represent an early stage of neutrophil activation preceding the acquisition of LDN/G-MDSC phenotype [49] . Neutrophils are highly sensitive to experimental manipulation and become easily activated during preparation [6] , [7] . In this study we utilized techniques minimizing the stress associated with neutrophil enrichment such as the use of isotonic red blood cell lysis buffer . Collectively , these techniques resulted in a neutrophil population that was highly viable at 24 hours of in vitro culture ( Figure S5 ) . Experiments elucidating the role of PD-L1 in T cell suppression were performed in the presence of ROS scavengers . Production of ROS appears to be a major mechanism of neutrophil and G-MDSC-mediated suppression of T cell function [6]–[8] . Importantly , PD-L1-mediated suppression appears to function independently of ROS and is contingent on elevated PD-1 expression on T cells . Release of arginase-1 from neutrophils , depletion of arginine , and subsequent downregulation of TCR CD3ζ chain represent another potential mechanism of neutrophil-mediated immune suppression [9] , [10] , [37] , [47] , [50] . Supporting this mechanism is the data demonstrating a correlation between PD-L1 expression on neutrophils , elevated concentration of arginase-1 in plasma ( Fig . 5A ) , and reduced CD3ζ expression on T cells ( Fig . 4C ) . This data is in agreement with a report by Cloke et al . demonstrating that the level of arginase-1 activity in PBMCs of HIV-seropositive patients increases with disease severity , inversely correlates with CD3ζ expression on T cells , and that the main source of arginase-1 are neutrophils co-purifying in the PBMC fraction [28] , [37] , [46] . Depletion of suppressive CD15+ neutrophil population co-segregating in the PBMC fraction results in a reversal of suppression and increased cytokine production by T cells ( Fig . 2 ) . Multiple studies have described limited production of IFNγ , TNFα , IL-2 and other factors by antigen-specific T cells from viremic HIV-1 patients [12] . Such studies may be confounded by the presence of suppressive LDNs limiting the functionality of T cells in in vitro assay . The upregulation of neutrophil PD-L1 expression in HIV-1-infected patients is likely caused by a combined effect of several factors . We show that inactivated HIV-1 virions and a TLR-7/8 ligand directly upregulate neutrophil PD-L1 ( Fig . 3 ) . PD-L1 expression is higher on neutrophils obtained from viremic patients and becomes significantly reduced following the initiation of ART ( Fig . 1 ) . This data suggests that PD-L1 is directly induced by the virus , consistent with previous reports in monocytes and dendritic cells [25] , [26] . However , other factors may contribute to this process . HIV-1 infection is associated with an extensive damage to the intestinal mucosal barrier and ensuing translocation of microbial products including LPS [30] . LPS cognate receptor TLR-4 is expressed at high levels on the surface of neutrophils and mediates the recognition of gram-negative bacteria [51] . Since LPS upregulates neutrophil PD-L1 ( Fig . 3 ) , microbial translocation may directly contribute to the induction of PD-L1high neutrophil phenotype . This view is strongly supported by a recent report demonstrating that an injection of low dose of LPS into the circulation of human volunteers causes an induction of a suppressive neutrophil population inhibiting T cell function via PD-L1/PD-1 [7] , [41] . Boasso et al . have shown that HIV-1-induced PD-L1 upregulation on monocytes is in part dependent on IFNα [29] . Consistent with this report , we show that blocking IFNα receptor partially blocked PD-L1 upregulation ( Figure S2B ) . Importantly , high levels of IFNα are detected in both plasma and lymphoid tissues during different stages of HIV-1 infection [52] , [53] and therefore may contribute to the increased levels of PD-L1 on neutrophils . The novel model of immune suppression mediated by neutrophils via the PD-L1/PD-1 pathway presented in this study enhances our understanding of T cell exhaustion in HIV-1 infection and highlights the need to target immunosuppressive pathways such as PD-L1/PD-1 in future therapeutic approaches in HIV-1-infection . Blocking the activation and suppressor function of neutrophils could improve immune competence in patients with AIDS .
All procedures involving human subjects were approved by the Institutional Review Board of the University of Alabama at Birmingham . All participants in this study were adults . Informed consent was obtained from all participants . Blood was collected from healthy donors ( HD ) and HIV-1 infected donors using acid citrate dextrose ( ACD ) collection tubes . 16 HIV-1-infected subjects on antiretroviral therapy ( ART ) ( median HIV-1 viral load = 190 ( 20–10 , 700 ) vRNA copies per ml; median CD4+ T cell count = 444 [102–1 , 385] per µl of blood ) ; and 21 HIV-1-infected subjects off ART therapy ( median viral load = 19 , 900 [71–1 , 040 , 000]; median CD4+ T cell count = 657 [189–1 , 763] ) were recruited for the purpose of this study . Peripheral blood mononuclear cells ( PBMCs ) were isolated by density centrifugation using Lymphocyte Separation Media ( MP Biomedicals; Solon , OH ) . Polymorphonuclear cells ( PMNs ) were isolated by density centrifugation using Ficoll-Paque Premium ( GE Healthcare ) . Briefly , after centrifugation , the mononuclear cell layer was removed and the granulocyte layer was collected and resuspended . The erythrocytes were lysed with isotonic NH4Cl erythrocyte lysis buffer ( 170 mM NH4Cl , 10 mM KHCO3 , 20 µM EDTA , pH 7 . 3 ) [7] . This procedure results in ≥95% purity of neutrophils as determined by the expression of CD15 marker . Cells were cultured in RPMI 1640 supplemented with 5% human A/B serum ( Atlanta Biologicals; Atlanta , GA ) , 100 I . U . /mL penicillin , 100 µg/mL streptomycin , 2 mM L-glutamine , and 1× minimal essential amino acids ( Life Technologies , Grand Island , NY ) . Positive selection of CD3+ T cells and depletion of CD14+ monocytes were performed using anti-CD3 FlowComp or anti-CD14 Dynabeads magnetic microbeads , respectively ( Life Technologies; Grand Island , NY ) . All cell culture reagents were obtained from Mediatech Inc . ( Manassas , VA ) , unless indicated otherwise . Antibodies , beads and columns for cell purification were obtained from Life Technologies/Invitrogen ( Grand Island , NY ) . Antibodies for flow cytometry were purchased from eBioscience ( San Diego , CA ) , unless listed otherwise . HIV-1 MN ( X4-tropic ) virus inactivated with aldrithiol-2 ( AT-2 ) and control microvesicles from uninfected cell cultures were kindly provided by AIDS and Cancer Virus Program , SAIC Frederick , Inc . , National Cancer Institute ( Frederick , MD ) . HIV-1 Consensus B Gag specific peptide pool ( 15-mers ) was provided by the NIH AIDS Research and Reference Reagent Program ( Germantown , MD ) . Plasma levels of myeloperoxidase ( MPO ) , arginase-1 , and neutrophil gelatinase- associated lipocalin ( NGAL ) were determined using ELISA according to the manufacturer's protocol ( Hycult Biotech , Uden , The Netherlands ) . To analyze the phenotype of neutrophils and other cells in blood , 50 µL of freshly obtained whole blood ( within 2 hours past blood draw ) , was incubated for 20 minutes with indicated antibodies , lysed with erythrocyte lysis buffer , and analyzed on LSRII ( BD Biosciences , San Diego , CA ) . PBMCs were blocked in PBS complemented with 10% human A/B serum ( Atlanta Biologicals; Atlanta , GA ) for 20 min prior to staining in staining buffer ( PBS containing 2% FBS ) . To analyze the expression of CD3ζ-PE , T cells were stained with CD3-efluor450 and CD8-PerCP-Cy5 . 5 , permeabilized , and stained with CD3ζ-PE antibody ( clone 6B10 . 2 ) . Intracellular cytokine staining was performed using the Cytofix/Cytoperm Fixation/Permeabilization Solution Kit ( BD Biosciences; San Jose , CA ) . Briefly , cytokine production was stimulated with 2 µg/mL HIV-1 consensus B Gag specific peptide pool ( 15-mers ) ( NIH AIDS Research and Reference Reagent Program; Germantown , MD ) for 24 hours in PBMCs or PBMCs depleted of CD15+ neutrophils . T cells were stained with CD3-efluor450 and CD8-PerCP-Cy5 . 5 , permeabilized , and stained with IFNγ-APC . Whole blood , PBMCs , or PMNs were cultured with human interferon-α ( IFNα ) ( 1 , 000 U/ml , Alpha A/D hybrid , #11200 , PBL Interferon Source; Piscataway , NJ ) , R848 ( 5 µg/ml , InvivoGen , San Diego , CA ) , AT-2 HIV-1MN ( 3–1 , 500 ng/ml p24 capsid equivalent ) , control microvesicles , or lipopolysaccharide ( LPS ) ( 100 ng/ml , Escherichia coli 0111:B4; Sigma ) . Cells were then blocked in PBS with 10% human serum for 20 min at 4°C , resuspended in 100 µl of master mix containing staining buffer ( PBS with 2% FBS ) and antibodies: CD15-FITC ( Biolegend ) , CD14- PerCP-Cy5 . 5 , and PD-L1-APC ( Biolegend ) . Media for experiments using AT-2MN was supplemented with FBS instead of human serum to avoid potential blocking effects of human serum [54] . Blocking of human IFNα was performed by pre-incubating PBMCs or PMNs with 5 µg/ml anti-IFNAR ( clone MMHAR-2; Invitrogen ) for 30 min before addition of AT-2 HIV or R848 . Purified T cells were stimulated with plate bound anti-CD3 ( 2 µg/mL ) and soluble anti-CD28 ( 2 µg/mL ) antibodies for 24 hours . CD15+ Neutrophils were co-cultured with CD3+ T cells at a 5∶1 ratio in media supplemented with catalase ( 1000 U/mL ) and superoxide dismutase ( 200 U/mL; Sigma ) to neutralize reactive oxygen species . PD-L1 dependent T cell suppression was neutralized by addition of anti-PD-L1 antibody ( clone 29E . 2A3 , Biolegend ) ; isotype control antibody served as a control treatment ( IgG2b; MCP-11; Biolegend ) . IFNγ ELISA was performed in cell culture media following manufacturer's protocol ( eBioscience , San Diego , CA ) . Data was analyzed using Student's t-test , Mann-Whitney rank sum test , and Wilcoxon signed-rank test as appropriate . Paired t-test was used on populations passing the Kolmogorov-Smirnov normality distribution test . Correlations were performed using Spearman rank order test or by Pearson product-moment correlation test for populations that passed D'Agostino & Pearson omnibus normality test . A standard level of statistical significance α = 0 . 05 was used; all reported p-values are two-sided . GraphPad Prism 5 ( GraphPad Software Inc . , LaJolla , CA ) statistical and graphing software packages were used . | Despite 30 years of intensive research , our understanding of how HIV-1 virus undermines the ability of the immune system to fight common infections is limited . Although we know that T cells , a key cell population that normally fights invading pathogens , lose their ability to function in HIV-1-infected individuals , we do not fully understand why . Here , we found that HIV-1 virus activates another type of cells , neutrophils , the most common type of white cell in the blood . Activated neutrophils negatively affect the function of T cells and prevent them from producing cytokines , protective proteins that serve as messengers orchestrating the immune response to bacteria and viruses . This newly identified mechanism of immune suppression mediated by neutrophils may alter our understanding of HIV-1 pathogenesis and result in a design of novel therapies targeting the loss of immune function in HIV-1/AIDS . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"infectious",
"diseases",
"immunology",
"biology",
"microbiology"
]
| 2014 | Immune Suppression by Neutrophils in HIV-1 Infection: Role of PD-L1/PD-1 Pathway |
Upon apoptotic stimuli , epithelial cells compensate the gaps left by dead cells by activating proliferation . This has led to the proposal that dying cells signal to surrounding living cells to maintain homeostasis . Although the nature of these signals is not clear , reactive oxygen species ( ROS ) could act as a signaling mechanism as they can trigger pro-inflammatory responses to protect epithelia from environmental insults . Whether ROS emerge from dead cells and what is the genetic response triggered by ROS is pivotal to understand regeneration of Drosophila imaginal discs . We genetically induced cell death in wing imaginal discs , monitored the production of ROS and analyzed the signals required for repair . We found that cell death generates a burst of ROS that propagate to the nearby surviving cells . Propagated ROS activate p38 and induce tolerable levels of JNK . The activation of JNK and p38 results in the expression of the cytokines Unpaired ( Upd ) , which triggers the JAK/STAT signaling pathway required for regeneration . Our findings demonstrate that this ROS/JNK/p38/Upd stress responsive module restores tissue homeostasis . This module is not only activated after cell death induction but also after physical damage and reveals one of the earliest responses for imaginal disc regeneration .
Tissues and organs need to function reliably regardless of adverse environmental conditions . Injuries , disease , infection and environmental insults are stressors causing cell damage that can be repaired via homeostatic machinery . Thus , optimal health is largely dependent upon tissue homeostasis , which involves cell replacement and tissue repair . Although many signaling pathways have been proposed to respond to environmental insults , the early activation of those signals is poorly understood . Response to damage can involve oxidative stress and , subsequently , the stimulation of stress-activated protein kinases . The production of reactive oxygen species ( ROS ) by various redox metabolic reactions , which has generally been considered to be deleterious , is now emerging as an active participant in cell signaling events [1 , 2] . ROS are byproducts of aerobic metabolism that include superoxide O2- , peroxide H2O2 and hydroxyl radicals OH· . ROS , and in particular H2O2 are required for inflammatory cell recruitment [3 , 4] . Amphibian and zebrafish injuries produce the ROS necessary to promote proliferation and regeneration [5–8] . In mammalian cells , ROS are known to act as second messengers to activate diverse redox-sensitive signaling transduction cascades , including the stress-activated MAP kinases p38 and the Jun-N Terminal kinase ( JNK ) [9–11] . ROS-mediated p38 activation occurs during the inflammatory response in rats [12] and during the loss of self-renewal and differentiation in glioma-initiating cells [13] . It has also been found that p38 and JNK are differentially required during repair . In endothelial cells , TNF-α stimulates repair through the positive action of JNK and negative regulation of p38 [14] , whereas in corneal repair , p38 , and not JNK , is required for epithelial migration [15] . In Drosophila both MAPK have been associated with stress responses [16] . Drosophila p38 pathway responds to different environmental stimuli and stressors [17 , 18] . Moreover , increasing ROS beyond basal level triggers precocious differentiation of Drosophila hematopoietic progenitors through JNK signaling [19] . The JNK signaling pathway has emerged as an early response to cell death and physical damage and appears to play a critical role in compensatory proliferation , regeneration and wound healing [20–28] . Moreover , upon apoptotic stimulus p53 and JNK are activated by the caspase Dronc and function upstream of pro-apoptotic genes , creating an amplifying loop that ensures cell death [29–33] . One of the early known responses to cell death is the transcriptional activation of the phosphatase puckered ( puc ) , a downstream effector of the JNK pathway and a powerful negative regulator of the same pathway . Interestingly puc has been found in surviving cells of nearby tissue after cell death [23 , 27] and after physical injury [22 , 34] . JNK activation of the cytokines unpaired ( upd ) , a family of cytokines linked to the human interleukin-6 , is necessary for hyperproliferation in Drosophila tumors and for wound healing [34–36] . Thus , we hypothesize here that the activation of JNK , which is amplified in dying cells , is in some way propagated to nearby surviving tissue where beneficial low levels of JNK promote upd expression . Apoptotic cells have been observed in the early regeneration of different animals and are thought to provide signals that regulate wound healing and regeneration [37–39] . As apoptosis has been associated with oxidative stress and cytokines act as a functional link between oxidative stress and compensatory proliferation in mammals [40] , we decided to investigate whether ROS occur upstream from the stress-activated protein kinases p38 and JNK and cytokines during tissue repair . We took advantage of the regeneration capacity of Drosophila imaginal disc epithelium ( reviewed in [41 , 42] ) to address these questions . Imaginal discs are larval epithelial sacs that possess a robust ability for homeostatic cell renewal to overcome the effect of stressors . We report here that , either by inflicting a physical lesion or after inducing cell death , imaginal disc cells produce ROS that are linked to the activation of p38 and JNK stress MAP kinases . In addition , JNK and p38 activity in the living tissue triggers transcription of the cytokine unpaired ( upd ) , which acts as a ligand of the JAK/STAT signaling pathway and promotes regeneration of the missing part .
To monitor ROS after tissue damage we used CellROX Green , a cell-permeant fluorogenic probe that is non-fluorescent in the reduced state and exhibits bright fluorescence upon oxidation . We found high levels of CellROX Green near the wound edges of physically cut wing imaginal discs . Only a few of the CellROX Green positive cells were TO-PRO-3 positive cells ( dying cells ) , indicating that most ROS-producing cells were alive ( Fig 1A ) . We examined the production and propagation of ROS over time . Few minutes after cut ( 0–5’ ) some cells at the wound edges were CellROX Green positive , indicating that the oxidative burst is rapidly occurring after damage ( Figs 1B and 1C and S1 ) . Ex vivo imaging showed that fluorescence propagates to the neighboring cells during the first 30’ after damage ( Figs 1B and S1 ) . We next monitored ROS production after controlled induction of cell death ( also known as genetic ablation ) , which can be used as a type of insult to study cellular responses . Apoptosis was induced using patched ( ptc ) -Gal4 to drive expression of the pro-apoptotic gene reaper ( rpr ) under the control of a UAS ( henceforth ptc>rpr ) ; the Gal4/UAS system was controlled by the temperature-sensitive form of Gal80 ( Gal80TS ) , which inhibits Gal4 and enables examination of regeneration after cell death [23 , 24] . As previously described , ptc>rpr discs show a stripe of apoptotic cells that eventually extrudes basally and is replaced apically by living cells ( Fig 1D ) [23] . CellROX Green was strongly incorporated into the ptc>rpr apoptotic cells ( TO-PRO-3 positive ) ( Fig 1E ) . Strikingly , living cells adjacent to the apoptotic zone also showed ROS , albeit at much lower levels than in dead cells ( Fig 1E and 1F ) . Similar observations were obtained using 2' , 7'-dichlorodihydrofluorescein diacetate ( H2DCFDA ) which upon oxidation is converted to the highly fluorescent DCF . Indeed , cut or rpr-ablated discs , showed high levels of fluorescence on the wound edges , in the apoptotic cells and also in the living cells near the apoptotic ( S1B , S1C and S1D Fig ) . Thus , these results showed that both physical injury and genetically induced apoptosis are insults that result in the production of ROS . Oxidative burst following death or damage could propagate from dying to living cells in sub-toxic doses and initiate repair . To explore this issue , we decided to deplete ROS production and examine adult wings after cell death . We first checked whether antioxidants ( vitamin C , Trolox or N-acetyl cysteine [NAC] ) are capable of blocking ROS production . We incubated cut discs in Schneider’s medium containing antioxidants , and found strong reduction of CellROX Green fluorescence ( S2 Fig ) . Next , we studied the effects of ROS scavenging on regeneration . We used a Gal4 construct under the control of a wing-specific enhancer ( salE/Pv-Gal4 ) , which allows analysis in adult wings while not affecting the rest of the organism , to activate UAS-rpr ( henceforth salE/Pv>rpr ) . To deplete intracellular ROS , salE/Pv>rpr larvae were fed with food supplemented with antioxidants ( Fig 2A ) . ROS scavengers in salE/Pv>rpr controls kept at 17°C to prevent cell death did not show any alteration of wing morphology ( S2B Fig ) . Conversely , a salE/Pv>rpr control group without scavengers moved to 29°C for 11 h showed complete wing regeneration ( Fig 2B ) . However , the salE/Pv>rpr experimental group with ROS scavengers and induced cell death showed incomplete regeneration in about 50% of the cases ( Fig 2B and 2C ) . We considered incomplete regeneration when some veins or intervein sectors were missing . To discard that these effects could be caused by differences in survival or developmental delay , we checked whether proliferation is impaired after ROS depletion . We counted the number of mitoses after cell death induction in discs from NAC-fed larvae and found a significant decrease compared to discs from larvae fed in the absence of antioxidants ( Fig 2D ) . The number of mitoses in controls fed with or without antioxidants and kept at 17°C to block cell death did not vary ( Fig 2D ) . In addition , we used enzymatic manipulation of ROS . Superoxide dismutase ( Sod ) catalyzes the dismutation of superoxide anion into oxygen and hydrogen peroxide . In the presence of hydrogen peroxide , Catalase ( Cat ) catalyzes its breakdown into water and oxygen . Thus , overexpression of Sod or Cat will remove their respective ROS substrates , whereas simultaneous activation of Sod and Cat will enhance the depletion of both O2- and H2O2 . UAS-Sod , UAS-Cat or UAS-Sod:UAS-Cat were ectopically expressed under the nub-Gal4 driver , which operates throughout the wing pouch ( Fig 2E ) . To induce cell death , we used an independent transactivator based on the LexA/lexO binary system . We generated a salE/Pv-LHG transgene , which includes a Gal80 suppressible form of LexA [43] , to conditionally express lexO-rpr in the salE/Pv domain . This combination permits control of the temporary expression of two binary systems ( salE/Pv-LHG lexO-rpr and nub-Gal4 UAS-transgene ) by tubGal80TS ( Fig 2E ) . This design has the advantage of simultaneously activating two wing-specific transgenes ( nub-Gal4 and salE/Pv-LHG ) in overlapping domains , therefore hindering early ROS . After salE/Pv-LHG lexO-rpr genetic ablation and nub-Gal4 UAS-transgene expression we allowed the larvae to develop to adulthood and found a drop in the number of regenerated wings ( Figs 2F and 2G and S2C and S2D ) . Together , these results indicate that chemical and enzymatic ROS scavengers interfere with regeneration . To determine whether ROS act on JNK during wing disc repair , we first monitored the activity of this pathway in wing discs after cell death . We used two different reporters to monitor JNK activity: puc-lacZ , which marks puc-expressing cells [44] , and the TRE-DsRed . T4 reporter , which monitors the JNK substrate AP1 transcription factor ( hereafter TRE-red reporter ) ( Fig 3A and S3A Fig ) [45] . In ptc>rpr discs , we found high levels of TRE-red reporter in the basal apoptotic zone and , to a lesser extent , in the apical living cells ( Fig 3A and 3B ) . In contrast , puc-lacZ positive cells were found in the apical zone , as described previously [23] , and rarely in the apoptotic zone . Some puc positive cells incorporated EdU , supporting that JNK is also induced in living cells ( Fig 3C ) . As NAC is an excellent source of sulfhydryl SH- groups and efficiently promotes scavenging of free radicals [46] , it was the most suitable antioxidant to determine the relationship between ROS and JNK in stressed imaginal discs . To test NAC effects on JNK , we used the TRE-red reporter because is more rapidly and extensively expressed than puc-lacZ ( Figs 3A and S3B ) and because its activity is blocked in JNK mutants [45] or after chemical JNK inhibitors ( S3C Fig ) . We found that the mean pixel intensity of the TRE-red reporter in ptc>rpr wing discs from animals grown in NAC-supplemented food was lower than in the same zone of individuals fed with standard food ( Fig 3D and 3E ) . Moreover , discs cultured ex vivo in which NAC was added into the medium resulted in a drop of TRE-red activity after physical injury ( Fig 3F and 3G ) . These observations indicate that activation of JNK is ROS dependent . Another potential response to ROS increase is the activation of the p38 signaling cascade [10 , 17 , 18] . Active p-38 signaling can be monitored using anti-phosphorylated p38 ( P-p38 ) . We found that discs fixed and incubated with anti-P-p38 a few minutes after physical injury showed P-p38 staining around the wound . P-p38 localization was variable and depended on the severity of the injury . In contrast , intact discs immediately stained after fixing did not show P-p38 ( Fig 4A ) . However , discs cultured for 3 to 8 hours with or without injury showed P-p38 staining throughout the disc . This general staining is likely due to the stress generated by culturing , and contrasts with the fast local P-p38 response around the damaged zone . We next wondered whether the boost in ROS that propagates to the surviving tissue triggers p38 activation . We observed that the early P-p38 staining was blocked in discs cut and cultured ex vivo in medium containing NAC ( Fig 4B and 4C ) . We also analyzed p38 activation after inducing cell death and found P-p38 only in living cells but never in the basal apoptotic zone ( Fig 4D ) . In the absence of cell death , no P-p38 was detected . Blocking of ROS production with NAC resulted in a significant drop in P-p38-labeled cells ( Fig 4D and 4E ) . In addition , we used the double transcriptional trans-activator system consisting of the salE/Pv-LHG lexO-rpr to induce apoptosis and simultaneously interfere with ROS production by inducing UAS-Sod:UAS-Cat in the anterior ( ci-Gal4 ) compartment ( Fig 4G ) . The results showed a strong reduction of P-p38 in the anterior ( ci-Gal4 UAS-Sod:UAS-Cat ) compartment in comparison to the posterior . To test whether an independent source of ROS could activate P-p38 in discs , we fed larvae for 2h with food supplemented with 1% H2O2 and checked for P-p38 . Intact discs ( no cut , no cell death ) from these larvae resulted in high levels of P-p38 as well as high CellROX Green fluorescence ( S4 Fig ) . Together , these observations show that chemical ( NAC ) or genetic ( UAS-Sod:UAS-Cat ) ROS scavengers inhibit P-p38 and therefore indicate that oxidative stress is required for p38 activation . We next scored wing regeneration after salE/Pv>rpr induction of cell death in different mutant backgrounds of the p38 pathway . As most of the alleles are lethal or semilethal in homozygosis [47] , we tested them in heterozygosis . Alleles of two Drosophila p38 genes , p38a and p38b , were used in this work . We found that heterozygous p38bd27 animals regenerated entire wings ( Fig 5A ) . However , a severe effect was observed with p38a1 as the resulting wings lacked some sectors and presented notches in the margin . Drosophila p38 signaling is activated by MKK3/licorne ( lic ) -mediated phosphorylation [48] . Heterozygous licd13 showed all wing sectors albeit wings were smaller than controls . However , double heterozygotes for licd13 and p38bd27 were unable to regenerate some wing sectors . We also tested Atf2PB , a hypomorphic allele of the ATF2 transcription factor downstream of p38 [49] , either in homozygosis or in double heterozygous combinations ( licd13 Atf2PB or p38bd27 Atf2PB ) . We found defects in size and pattern after salE/Pv>rpr induction . Regeneration was severely impaired in double heterozygotes for p38a1 ( MAPK ) and Atf2PB ( Fig 5A ) . We also blocked the pathway with UAS-RNAi constructs for lic , p38b , p38a and Atf2 and analyzed the adult wings . These transgenes were activated in the anterior compartment ( ci>RNAi ) and cell death was induced in the salE/Pv domain ( salE/Pv-LHG lexO-rpr ) . We found a reduction of individuals capable to fully regenerate wings for those RNAi’s ( S5A Fig ) . To gain further insight into the requirement for p38 , we chemically blocked the pathway using the imidazole drug SB202190 , a specific cell permeable p38 MAP kinase inhibitor that has been reported to do not interfere JNK or ERK kinases and is known to prevent phosphorylation of Atf2 in Drosophila S2 cells [50] . We first tested the specificity of the SB202190 on P-p38 in rpr-ablated discs and found significant differences between individuals fed with the drug and controls . In contrast , the differences on TRE-red reporter were not significant ( S5C Fig ) . This indicates that SB202190 strongly blocked P-p38 and weakly the TRE-red . SalE/Pv>rpr larvae grown at 17°C to prevent cell death and fed with food containing SB202190 ( 0 . 12 , 1 . 0 or 5 . 0 μM ) emerged into normal adults ( S5B Fig ) . However , salE/Pv>rpr-induced larvae fed with SB202190 developed wings lacking some sectors . The highest percentage of aberrant wings was found using 5 μM SB202190 ( Fig 5B ) . This observation confirms that activation of p38 is required for wing repair . To assess the relationship between JNK and p38 , we tested p38 activation in wounded null hemizygous JNKK hemipterous ( hepr75 ) discs . P-p38 was localized near the wound after physical injury ( Fig 6A ) . This contrasts with the decrease in P-p38 when the MAPK kinase lic , which is the p38 activating kinase , was interfered with RNAi in injured discs ( S6A Fig ) . Moreover , P-p38 staining was localized in hepr75 discs after ptc>rpr induction , as in the wild type ( Fig 6B , compare with Fig 4 ) , indicating that JNK and p38 act independently . In addition , we fed animals with the JNK Inhibitor IX , which abolishes TRE-red reporter expression and inhibits regeneration ( S3C and S3D Fig ) , and found that P-p38 after rpr-ablation was not affected ( S6B Fig ) . To confirm that JNK and p38 act independently , we blocked the p38 pathway and checked for TRE-red reporter activity . As the p38a1 allele in heterozygosis strongly affects regeneration ( Fig 5 ) , we used this null allele in homozygosis and tested TRE-red activity after physical injury . Our results showed that TRE-red is induced at the wound edges of p38a1-/- mutant discs ( Fig 6C and 6D ) . Together , these results demonstrate that p38 and JNK stress responses act independently in damaged imaginal discs . The evidence that JNK is active in the living tissue located near damaged zones arises from the expression of puc and TRE-red reporters ( Fig 3A and 3B ) , and also because inhibition of JNK results in defects in repair [21 , 23 , 25 , 26 , 51] . Moreover , JNK activation promotes upd expression in different contexts [28 , 52] . We wondered whether those low non-deleterious JNK levels are capable of triggering tissue repair through upd expression . Upd cytokines are ligands that associate with the receptor domeless ( dome ) to stimulate the kinase activity of the receptor associated protein kinase hopscotch ( hop ) , which in turn phosphorylates dimers of the transcription factor STAT92E [53] . We found upd and upd3 expression near the wound after both physical injury and cell death ( Fig 7A , 7B , 7F and 7G ) . This injury-induced upd expression was blocked in JNKK hepr75 mutants ( Fig 7C and 7D ) and by JNK Inhibitor IX ( S3C Fig ) , which is consistent with previous observations [28 , 35] . To study the requirement for JAK/STAT for regeneration , we used the salE/Pv-LHG lexO-rpr to induce apoptosis and simultaneously interfered with the receptor dome using the dominant negative form UAS-domeDN driven by ci-Gal4 . These wings lacked most of the tissue where cell death was induced and dome was blocked ( Fig 7J and 7K ) , indicating that JAK/STAT signaling is needed for tissue recovery . Moreover , heterozygous alleles for the JAK/STAT pathway resulted in partial disruption of adult wing recovery after cell death ( S7 Fig ) . We next analyzed if JAK/STAT signaling requires ROS in this context . Two different reporters ( 10XSTAT92E-GFP and upd-Gal4 UAS-myrtomato ) were used in physically injured discs and showed reduced expression after NAC feeding ( Fig 7E ) . In addition , ptc>rpr induced discs from NAC fed larvae showed a reduction of upd expression ( Fig 7H and 7I ) . Thus , this expression is ROS dependent after both physical injury and cell death . We speculated that if ROS operate upstream upd , the impairment of regeneration resulting from NAC feeding should be rescued by activating Upd . To this aim , we used the double transactivation system to induce cell death in NAC-fed individuals and concomitantly activate upd expression ( Fig 7L ) . Analysis of the resulting wings showed that upd ectopic expression rescued the NAC inhibition phenotype ( Fig 7M and 7N ) . These observations demonstrate that ROS function upstream of JAK/STAT during repair . We wondered whether p38 is also required for upd expression in damaged discs . Expression of upd or upd3 was severely reduced in p38a1-/- wound edges ( Figs 8A–8C and S8 ) . This suggests that in addition to JNK , p38 is essential for upd expression upon stress . Finally , we argued that if p38 is required for repair through upd , its ectopic expression should rescue the impaired regeneration after inhibition of p38 . We , again , used the double transactivation system to induce cell death in SB202190-fed individuals , to block p38 phosphorylation and alongside activate upd expression ( Fig 8D ) . Indeed , we found that the number of wings that regenerated after p38 inhibition increased ( Fig 8E and 8F ) . Altogether these results position Upd cytokines downstream from the ROS/p38/JNK module .
In this work , we demonstrate a stress-responsive module activated upon cell death or physical damage . This module consists of ROS dependent stimulation of non-deleterious levels of JNK and p38 MAP kinases necessary for the expression of Upd and JAK/STAT signaling which drives regeneration . Non-lethal levels of JNK may have multiple functions , among them cytoskeleton organization [44 , 54] , healing and initiation of regenerative growth [21 , 23–26 , 51 , 55 , 56] . Thus , this early responsive module is crucial to maintain tissue in a healthy condition , trigger tissue repair and restore homeostasis . In an apoptotic context , Rpr dimerizes and , through direct binding , brings the Drosophila inhibitor of apoptosis protein-1 ( DIAP1 ) to mitochondria , concomitantly promoting DIAP1 auto-ubiquitination and destruction [57 , 58] . Rpr action on the mitochondria results in alteration of cytochrome C driven by caspases [59] and in mitochondrial disruption [60] . The ROS dyes used here detect a wide range of ROS , and therefore we cannot discriminate between membrane oxidases or mitochondrial origin . However , since Rpr acts on mitochondria , mitochondrial alterations could cause the burst of ROS in apoptotic cells . Of note , we observed that high ROS levels are associated with high levels of JNK in apoptotic cells . It has been proposed that ROS can mediate the activation of JNK [61] by quenching the MAP kinase phosphatases [62] . Conversely , low levels of ROS detected in nearby surviving tissue correlate with low non-deleterious levels of JNK and activation of MAP kinase phosphatases . Thus , puc MAP kinase phosphatase could protect the living cells close to the damage from the noxious effects of high JNK . Indeed , living cells near the wound retain low levels of JNK , not sufficient to kill but necessary for tissue recovery . Additionally , the caspase Dronc , which acts downstream from Rpr , has functions beyond apoptosis [63] . Dronc is involved in the activation of JNK and p53 , which activate the pro-apoptotic genes , creating an amplification loop that ensures apoptosis [27 , 29–31 , 64] . The JNK/p53 driven apoptosis stimulates proliferation of the nearby tissues [29–31 , 65] . Although still unclear , it has been proposed that apoptotic cells can release the products of mitogenic genes such as wingless ( wg ) and decapentaplegic ( dpp ) [33 , 66 , 67][31 , 68] . Alternatively , we show here that ROS operate as signals responding to insults ( apoptosis , mechanical stress ) that turn on the homeostatic machinery to compensate the epithelial damage . This fits with a scenario in which ROS are able to either diffuse from cell to cell or perhaps to propagate their production to several rows of cells . Indeed , ROS have been proven to cross cell membranes , to spread through gap junctions [69–71] and to enter into the cell through specific membrane aquaporin channels [71 , 72] . Therefore , ROS behave as an efficient paracrine signal that ultimately will result in Upd activation . In addition to JNK , ROS are stressors involved in p38 activation [73] . ROS may activate the p38 pathway through the oxidative modification of intracellular kinases such as redox-sensitive activating protein-1 ASK1 [74] . We showed here that not only JNK but also p38 is required for regeneration . Moreover , the p38a1 allele seems to particularly affect upd expression and regeneration . This concurs with the finding that Drosophila p38a is more susceptible to environmental stressors , such as oxidative stress [18] . However , other p38 kinases could contribute to tissue regeneration . Indeed , heterozygous alleles of the p38 activating kinase lic , which normally do not show patterning defects after rpr-mediated ablation , can result in incomplete regeneration when a dose of p38b is missing ( Fig 5A ) . Moreover , RNAi of p38b also can show defective regeneration individuals ( S2 Fig ) . In addition , we cannot discard that p38c , which has been recently found involved in intestinal immune homeostasis [75] , may also function in imaginal disc regeneration . We have found that both hepr75 and p38a1 inhibit upd expression . But hepr75 mutants , which block JNK signaling , do not affect p38 phosphorylation and viceversa , p38a1 mutants , which block at least the p38a branch of the p38 kinase , do not interfere with the TRE-red reporter expression . This suggests that ROS activate p38 and JNK independently and that both MAP kinases act on upd expression to drive tissue repair . Thus , ROS signaling operates through these two MAP kinase pathways that in turn will converge to stimulate the transcriptional expression of the cytokines ( Fig 9 ) . JNK and p38 are not only activated after cell death but also after physical injury . Beneficial ROS production is an ubiquitous reaction associated with inflammatory responses to wounding [4 , 6 , 76] . Recent findings show that ROS produced in dynamic epithelia operate as a tuning mechanism for reorganization of epithelia [77] . Therefore , it could be that changes in mechanical stress generated during wounding and epithelial disruption ( mechanical stretching ) results in ROS production . Some dead cells were also found after physical injury . Thus , a partial contribution of dead cells in addition to the stress due to epithelial disruption can account for the oxidative burst generated after physical injury . In summary , an early boost of oxidative stress is required to activate p38 and JNK in apoptotic cells or near the wound . Moreover , upd is turned on downstream JNK and p38 . Thus , downstream of the stress response module , cytokines operate to control tissue growth during regeneration .
The Drosophila melanogaster strains used were ptc-Gal4 [78] , tubGal80TS [79] , UAS-rpr [80] , ci-Gal4 [81] , nub-Gal4 [82] sal-Gal4 and salE/Pv-Gal4 [83] , p38bd27 , licd13 [47] , dATF2PB [49] , p38a1 , [17] , LexO-rCD2::GFP [43] , TRE-DsRed . T4 [45] as AP1 reporter , puc-lacZ [44] , pucE69-A-Gal4 [84] , UAS-upd [85] , upd-Gal4 ( from D . Harrison ) , 10XSTAT92E-GFP [86] , en-Gal4 , UAS-GFP , UAS-myrtomato , UAS-Sod . A ( sod1 ) , UAS-Cat . A , UAS-domeDN , hop2 , hop27 , stat92e06346 ( Bloomington Stock center ) , stat92e397 [87] , and hepr75 [88] . Transgenic Drosophila shRNAi lines were obtained from the Vienna Drosophila RNAi Center ( VDRC ) . Canton S was used as the wild type control . Wing discs were dissected from third instar larvae in Schneider’s insect medium ( Sigma-Aldrich ) and a small fragment was removed with tungsten needles . Discs were cultured in Schneider’s insect medium supplemented with 2% heat activated fetal calf serum , 2 . 5% fly extract and 5 μg/ml insulin , for different periods of time ( from 1 to 10 hours ) at 25°C . Ex vivo images were taken using a Leica SPE confocal microscope and processed with Fiji software . The salE/Pv-LHG construct was created cutting the wing specific enhancer of spalt , salE/Pv [83] from pC4LacZ-Spalt PE EcoRI/BamHI and cloning this fragment into the plasmid attB-LHG containing a Gal80-suppressible form of LexA transcriptional activator ( LHG ) [43] . LHG contains both the binding domain of LexA and the activator domain of Gal4 , which is recognized by the inhibitor Gal80TS . The LexO-rpr strain was obtained subcloning the pro-apoptotic gene reaper ( rpr ) from pOT2-rpr ( IP02529 ) EcoRI/XhoI in the pLOTattB plasmid [89] carrying the lexA operator LexO . Transgenic flies were performed with standard protocols . Cell death was genetically induced as previously described [23 , 90] . We used two different drivers to induce cell death . The first , ptc-Gal4 which is expressed in a narrow stripe in the center of the disc . This strain was used to induce cell death in imaginal discs ( UAS-rpr ) , because the dead domain can be easily discerned from the neighboring living domain . The second , salE/Pv-Gal4 strain , which consists of sal wing enhancer with expression confined to the wing [83] has been used in this work to score adult wing parameters . The UAS line used to promote cell death was UAS-rpr , and the system was controlled by the thermo sensitive repressor tubGal80TS . We also used the salE/Pv-LHG and LexO-rpr strains for genetic ablation using the same design as for Gal4/UAS . Embryos were kept at 17°C until the 8th day/192 h after egg laying ( equivalent to 96 hours at 25°C ) to prevent rpr expression . They were subsequently moved to 29°C for 11 hours and then back to 17°C until adulthood . Controls without rpr expression were always treated in parallel . In dual transactivation experiments , we used the salE/Pv-LHG LexO-rpr to ablate the salE/Pv domain , whereas Gal4 was used to express different transgenes under the control of nub-Gal4 or ci-Gal4 . In the experiments on antioxidants ( Fig 2 ) and upd ( Figs 7 and 8 ) overexpression , larvae were transferred to NAC- ( 100 μg/ml ) or SB202190- ( 5 μM ) supplemented food 24 h before cell death induction . All experiments for ROS detection were done in living conditions . To detect the presence of ROS we used CellROX Green Reagent ( Life Technologies ) , which is an indicator of oxidative stress in living cells . For both genetic ablation and physical injury experiments , third instar discs were dissected in Schneider´s medium immediately after cell death or injury and incubated for 15 minutes in medium containing 5 μM CellROX Green Reagent , followed by three washes . Samples were protected from light throughout . Then they were mounted using culture medium supplemented with 1 μM TO-PRO-3 ( Life Technologies ) nucleic acid stain . As TO-PRO-3 only enters dead cells , we used it to distinguish dead cells from living cells in the ex vivo experiments . Images were taken using a Leica SPE and SPII confocal microscope . Grey values of regions of interest ( ROI ) were measured using Fiji software . ROIs were established at the wound edges of injured discs ( examples in S2 Fig ) , or in rectangles as indicated in Fig 1E . Pixel intensities were collected and analyzed from raw images taken under the same laser confocal conditions . Thermal LUT images were rendered from slices taken from the confocal using the Interactive 3D Surface Plot tool of the Fiji software ( ImageJ ) . We also used the cell-permeant 2' , 7'-dichlorodihydrofluorescein diacetate ( H2DCFDA 5μM , Life Technologies ) which upon oxidation is converted to the highly fluorescent 2' , 7'-dichlorofluorescein ( DCF ) . To visualize the ROS images in Fig 1 after genetic ablation , the whole stacks were subject to the Enhance Contrast tool at 0 . 4 pixel saturation in whole stack normalization . For physical injury images , thermal LUT images were obtained from raw stacks . To prevent ROS production , we used two protocols . The first was mainly used for rpr-ablation discs . It consisted in that antioxidants were supplemented into standard fly food . As antioxidants we used vitamin C ( 250 μg/ml ) , Trolox ( an analog of vitamin E; 20 μg/ml ) and N-acetyl cysteine ( NAC ) ( 100 μg/ml ) , all from Sigma-Aldrich . To score adult wings , larvae were transferred from vials containing standard food to vials containing food with the desired antioxidant concentration . Antioxidant treatment was administered at 168 h of development at 17°C ( equivalent to 84 h AEL at 25°C ) . After 24 hours , experimental larvae were moved to 29°C for 11 hours to promote rpr apoptosis . Meanwhile one control consisted of larvae maintained at 17°C and another control consisted of larvae transferred to a vial with standard food and moved to 29°C for the same period as in the experimental group . After rpr induction temperature was returned to 17°C to allow tissue recovery . This protocol was applied for Figs 2A–2D , 3D , 3E , 4E , 4F and 7E . The second was used for ex vivo cultured discs . Wing imaginal discs were incubated for 30 minutes in Schneider’s insect medium supplemented with NAC 100 μg/ml . Then , they were transferred to Schneider’s containing CellROX Green ( S2 Fig ) . NAC incubated discs were used for monitoring TRE-red ( Fig 3F and 3G ) or for P-p38 antibody staining ( Fig 4B and 4C ) . In S2 Fig medium was supplemented with NAC , Trolox or Vit C . The imidazole drug SB202190 ( Sigma-Aldrich ) was added to standard fly food to prevent p38 activation . We used three different concentrations ( 0 . 12 μM , 1 μM and 5 μM ) , and DMSO as the control . To inhibit chemically JNK we used the JNK Inhibitor IX ( 5 μM , Selleckchem ) which is a thienylnaphthamide compound that is a selective and potent inhibitor of the ATP binding site of JNK . The timing and protocol followed to inhibit both pathways was the same as that to scavenge ROS . Third instar larvae were transferred to vials containing 1% H2O2 , 1 , 3% low melting agarose and 5% sucrose . To avoid loss of oxidative capacity , H2O2 was added at a temperature under 45°C . Larvae were fed for 2h in this medium prior dissection and fixation of the discs . Controls without H2O2 were done in parallel . For testing the capacity to regenerate we used adult wings emerged from salE/Pv>rpr individuals , in which patterning defects can be easily scored . Flies were fixed in glycerol:ethanol ( 1:2 ) for 24 h . Wings were dissected on water and then washed with ethanol . Then they were mounted on 6:5 lactic acid:ethanol and analyzed and imaged under a microscope . Definition of regenerated/non-regenerated wings: when veins or interveins were missing , we considered them as defective in their capacity to restore the normal pattern . Therefore , the % of regenerated wings ( Figs 2 , 5 , 7 , 8 , S2 , S3 , S5 and S7 ) was calculated after the number of wings with the complete set of veins and interveins . For each sample of “regenerated wings” we scored the percentage of individuals that belong to the “regenerated wings” class and calculated the standard error of sample proportion based on binomial distribution ( regenerate complete wing or not ) SE = √p ( 1-p ) /n , where p is the proportion of successes in the population . Ratios between wing areas ( Fig 5 ) were used as an indication of the size achieved after cell death for each genetic background , and consisted of a comparison between wing size with and without rpr induction . Immunostaining and FISH were performed using standard protocols . Primary antibodies used in this work were P-p38 ( rabbit 1:50 , Cell Signaling Technology ) , phospho-Histone-H3 ( rabbit 1:1000 , Millipore ) , ß-galactosidase ( rabbit 1:1000 , ICN Biomedicals ) , Upd ( rabbit 1:800 , gift from D . Harrison ) and cleaved caspase-3 ( rabbit 1:100 , Cappel ) . Fluorescently labeled secondary antibodies were from Life Technologies and Jackson Immunochemicals . Discs were mounted in SlowFade ( Life Technologies ) supplemented with 1 μM TO-PRO-3 ( Life Technologies ) to label nuclei . Note that in fixed tissues all nuclei are TO-PRO-3 labeled , whereas in ex-vivo culture only nuclei of dead or dying cells are TO-PRO-3 labeled . The number of mitosis after analyzing the stacks of confocal images was calculated using Fiji software ( Cell counter plug-in ) . Mitosis were counted for the entire anterior compartment of the wing pouch for each disc . For apoptotic cell detection , we used both anti cleaved caspase 3 or TUNEL assay . For TUNEL we used the fluorescently labeled dUTP ChromaTide BODIPY FL-14-dUTP ( Life Technologies ) and incorporated using terminal deoxynucleotidyl transferase ( Roche ) . EdU was incorporated using the Click-iT EdU Imaging Kit ( Life Technologies ) . Wing discs were dissected after cell death induction and incubated in Schneider’s insect medium supplemented with 1 mg/ml EdU for 5 minutes . Following EdU incorporation , discs were fixed and immunostained . Riboprobes for upd and upd3 were synthesized using cDNA clones from DGRC AT1366 and FI03911 . Fig 1 A , B . Wild type D , E , F . ptc>rpr: UAS-rpr/+; ptc-Gal4/+; tubGal80TS/+ Fig 2 A , B , C . salE/Pv>rpr: UAS-rpr/+; salE/Pv-Gal4/+; tubGal80TS/+ D . ptc>rpr: UAS-rpr/+; ptc-Gal4:tubGal80TS/+ E , F , G . salE/Pv>rpr nub>GFP: w; nub-Gal4/UAS-GFP; salE/Pv-LHG:tubGal80TS/lexO-rpr salE/Pv>rpr nub>Cat: w; nub-Gal4/UAS-Cat; salE/Pv-LHG:tubGal80TS/lexO-rpr salE/Pv>rpr nub>Sod: w; nub-Gal4/UAS-Sod; salE/Pv-LHG:tubGal80TS/lexO-rpr salE/Pv>rpr nub>Sod:Cat: w; nub-Gal4/UAS-Sod:Cat; salE/Pv-LHG:tubGal80TS/lexO-rpr Fig 3 A , B . ptc>rpr: UAS-rpr/+; ptc-Gal4:tubGal80TS/TRE-DsRed . T4; puc-LacZ/+ C . ptc>rpr: UAS-rpr/+; ptc-Gal4:tubGal80TS/+; puc-LacZ/+ D , E . ptc>rpr: UAS-rpr/+; ptc-Gal4:tubGal80TS/TRE-DsRed . T4 F , G . w; TRE-DsRed . T4 Fig 4 A , B , C . Wild type . D , E , F . ptc>rpr: UAS-rpr/+; ptc-Gal4/+; tubGal80TS/+ G . w; ci-Gal4/UAS-Sod:Cat; salE/Pv-LHG:tubGal80TS/lexO-rpr Fig 5 A . Control: w; +; salE/Pv-LHG:tubGal80TS/lexO-rpr ( control for lexO-rpr on the third chromosome ) and w; lexO-rpr/+; salE/Pv-LHG:tubGal80TS/+ ( control for lexO-rpr on the second chromosome ) licd13/+: licd13/+; +; salE/Pv-LHG:tubGal80TS/lexO-rpr p38bd27/+: w; p38bd27/+; salE/Pv-LHG:tubGal80TS/lexO-rpr p38a1/+: w; lexO-rpr/+; p38a1/salEPv-LHG:tubGal80TS licd13/+p38bd27/+: licd13/+; p38bd27/+; salEPv-LHG:tubGal80TS/lexO-rpr dATF2PB/+: w; Atf2PB/+; salE/Pv-LHG:tubGal80TS/lexO-rpr dATF2PB-/-: w; Atf2PB/Atf2PB; salE/Pv-LHG:tubGal80TS/lexO-rpr licd13/+ dATF2PB/+: licd13/+; Atf2PB/+; salE/Pv-LHG:tubGal80TS/lexO-rpr p38bd27/+ dATF2PB/+: w; p38bd27/Atf2PB; salE/Pv-LHG:tubGal80TS/lexO-rpr dATF2PB/+p38a1/+: w; Atf2PB/lexO-rpr; p38a1/salEPv-LHG:tubGal80TS B . salE/Pv>rpr: UAS-rpr/+; salE/Pv-Gal4/+; tubGal80TS/+ Fig 6 A . hepr75: hepr75/Y B . ptc>rpr hepr75: hepr75/Y; ptc-Gal4:tubGal80TS/+; UAS-rpr/+ C . wt: w; TRE-DsRed . T4 p38a-/-: w; TRE-DsRed . T4; p38a1/p38a1 Fig 7 A , B . Wild type . C , D . hepr75: hepr75/Y E . upd-Gal4/+; UAS-myrtomato/10XSTAT92E-GFP F , G , H , I . ptc>rpr: UAS-rpr/+; ptc-Gal4/+; tubGal80TS/+ salE/Pv>rpr: UAS-rpr/+; sal-Gal4/+; tubGal80TS/+ J , K . salE/Pv>rpr ci>GFP: w; ci-Gal4/lexO-rpr; salE/Pv-LHG:tubGal80TS/UAS-GFP ci>domeDN salE/Pv>GFP: w; ci-Gal4/lexO-rCD2::GFP; salE/Pv-LHG:tubGal80TS/UAS-domeDN salE/Pv>rpr ci> domeDN: w; ci-Gal4/lexO-rpr; salE/Pv-LHG:tubGal80TS/UAS-domeDN L , M , N . ci-Gal4 UAS-upd: w; ci-Gal4/UAS-upd; salE/Pv-LHG:tubGal80TS/lexO-rCD2::GFP salE/Pv-LHG lexO-rpr: w; ci-Gal4/UAS-GFP; salE/Pv-LHG:tubGal80TS/lexO-rpr ci-Gal4 UAS-upd salE/Pv-LHG lexO-rpr: w; ci-Gal4/UAS-upd; salE/Pv-LHG:tubGal80TS/lexO-rpr Fig 8 A , B , C . Wild type . p38a-/-: w; +; p38a1/p38a1 D , E , F . ci-Gal4 UAS-upd: w; ci-Gal4/UAS-upd; salE/Pv-LHG:tubGal80TS/lexO-rCD2::GFP salE/Pv-LHG lexO-rpr: w; ci-Gal4/UAS-GFP; salE/Pv-LHG:tubGal80TS/lexO-rpr ci-Gal4 UAS-upd salE/Pv-LHG lexO-rpr: w; ci-Gal4/UAS-upd; salE/Pv-LHG:tubGal80TS/lexO-rpr S1 Fig A , C , D . Wild type . B . ptc>rpr: UAS-rpr/+; ptc-Gal4:tubGal80TS/+ S2 Fig A . wt B . salE/Pv>rpr: UAS-rpr/+; salE/Pv-Gal4/+; tubGal80TS/+ C . salE/Pv>rpr nub>GFP: w; nub-Gal4/UAS-GFP; salE/Pv-LHG:tubGal80TS/lexO-rpr salE/Pv>rpr nub>Cat: w; nub-Gal4/UAS-Cat; salE/Pv-LHG:tubGal80TS/lexO-rpr salE/Pv>rpr nub>Sod: w; nub-Gal4/UAS-Sod; salE/Pv-LHG:tubGal80TS/lexO-rpr salE/Pv>rpr nub>Sod:Cat: w; nub-Gal4/UAS-Sod:Cat; salE/Pv-LHG:tubGal80TS/lexO-rpr D . salE/Pv>rpr nub>GFP: w; nub-Gal4/UAS-GFP; salE/Pv-LHG:tubGal80TS/lexO-rpr salE/Pv>rpr nub>Cat: w; nub-Gal4/UAS-Cat; salE/Pv-LHG:tubGal80TS/lexO-rpr nub>Cat: w; nub-Gal4/UAS-Cat; salE/Pv-LHG:tubGal80TS/lexO-GFP salE/Pv>rpr nub>Sod: w; nub-Gal4/UAS-Sod; salE/Pv-LHG:tubGal80TS/lexO-rpr nub>Sod: w; nub-Gal4/UAS-Sod; salE/Pv-LHG:tubGal80TS/lexO-GFP salE/Pv>rpr nub>Sod:Cat: w; nub-Gal4/UAS-Sod:Cat; salE/Pv-LHG:tubGal80TS/lexO-rpr nub>Sod:Cat: w; nub-Gal4/UAS-Sod:Cat; salE/Pv-LHG:tubGal80TS/lexO-GFP S3 Fig A . Wild type . B . w; TRE-DsRed . T4/+; puc-Gal4:UAS-GFP/+ C . ptc>rpr: UAS-rpr/+; ptc-Gal4:tubGal80TS/TRE-DsRed . T4 D . salE/Pv>rpr: UAS-rpr/+; salE/Pv-Gal4/+; tubGal80TS/+ S4 Fig Wild type S5 Fig A . salE/Pv>rpr: w; ci-Gal4/UAS-GFP; salE/Pv-LHG:tubGal80TS/lexO-rpr ci>RNAi p38a: w; ci-Gal4/UAS-RNAi p38a; salE/Pv-LHG:tubGal80TS/lexO-GFP salE/Pv>rpr ci>RNAi p38a: w; ci-Gal4/UAS-RNAi p38a; salE/Pv-LHG:tubGal80TS/lexO-rpr ci>RNAi p38b: w; ci-Gal4/UAS-RNAi p38b; salE/Pv-LHG:tubGal80TS/lexO-GFP salE/Pv>rpr ci>RNAi p38b: w; ci-Gal4/UAS-RNAi p38b; salE/Pv-LHG:tubGal80TS/lexO-rpr ci>RNAi Atf2: w; ci-Gal4/UAS-RNAi Atf2; salE/Pv-LHG:tubGal80TS/lexO-GFP salE/Pv>rpr ci>RNAi Atf2: w; ci-Gal4/UAS-RNAi Atf2; salE/Pv-LHG:tubGal80TS/lexO-rpr ci>RNAi lic: w; ci-Gal4/UAS-RNAi lic; salE/Pv-LHG:tubGal80TS/lexO-GFP salE/Pv>rpr ci>RNAi lic: w; ci-Gal4/UAS-RNAi lic; salE/Pv-LHG:tubGal80TS/lexO-rpr B . salE/Pv>rpr: UAS-rpr/+; salE/Pv-Gal4/+; tubGal80TS/+ C . salE/Pv>rpr: UAS-rpr/+; salE/Pv-Gal4/+; tubGal80TS/+ ptc>rpr: UAS-rpr/+; ptc-Gal4:tubGal80TS/TRE-DsRed . T4 S6 Fig en>RNAi lic: w; en-Gal4:UAS-GFP/UAS-RNAi lic S7 Fig Control: UAS-rpr/+; salE/Pv-Gal4/+; tubGal80TS/+ hop2/+: UAS-rpr/hop2; salE/Pv-Gal4/+; tubGal80TS/+ hop27/+: UAS-rpr/hop27; salE/Pv-Gal4/+; tubGal80TS/+ stat92e397/+: UAS-rpr/+; salE/Pv-Gal4/+; tubGal80TS/stat92e397 stat92e06346/+: UAS-rpr/+; salE/Pv-Gal4/+; tubGal80TS/stat92e06346 S8 Fig Wild type . p38a-/-: w; +; p38a1/p38a1 | Regenerative biology pursues to unveil the genetic networks triggered by tissue damage . Regeneration can occur after damage by cell death or by injury . We used the imaginal disc of Drosophila in which we genetically activated apoptosis or physically removed some parts and monitored the capacity to repair the damage . We found that dying cells generate a burst of reactive oxygen species ( ROS ) necessary to activate JNK and p38 signaling pathways in the surrounding living cells . The action of these pathways is necessary for the activation of the cytokines Unpaired ( Upd ) . Eventually , Upd will turn on the JAK/STAT signaling pathway to induce regenerative growth . Thus , we present here a module of signals that depends on oxidative stress and that , through the p38-JNK interplay , will activate cytokine-dependent regeneration . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | ROS-Induced JNK and p38 Signaling Is Required for Unpaired Cytokine Activation during Drosophila Regeneration |
Sensing of viral RNA by RIG-I-like receptors initiates innate antiviral response , which is mediated by the central adaptor VISA . How the RIG-I-VISA-mediated antiviral response is terminated at the late phase of infection is enigmatic . Here we identified the protein kinase A catalytic ( PKAC ) subunits α and β as negative regulators of RNA virus-triggered signaling in a redundant manner . Viral infection up-regulated cellular cAMP levels and activated PKACs , which then phosphorylated VISA at T54 . This phosphorylation abrogated virus-induced aggregation of VISA and primed it for K48-linked polyubiquitination and degradation by the E3 ligase MARCH5 , leading to attenuation of virus-triggered induction of downstream antiviral genes . PKACs-deficiency or inactivation by the inhibitor H89 potentiated innate immunity to RNA viruses in cells and mice . Our findings reveal a critical mechanism of attenuating innate immune response to avoid host damage at the late phase of viral infection by the house-keeping PKA kinase .
Innate immune response is the first line of host defense against invading microbial pathogens . The structurally conserved components of microbes called pathogen-associated molecular patterns ( PAMPs ) are recognized by host pattern-recognition receptors ( PRRs ) , which initiates signaling pathways that lead to induction of type I interferons ( IFNs ) , proinflammatory cytokines and other downstream effector genes [1] . During RNA virus infection , viral RNAs , including the invading viral RNAs and RNA intermediates produced during viral replication , act as PAMPs that are mostly recognized by the cytoplasmic RIG-I-like receptor ( RLR ) family members including RIG-I and MDA5 [2] . Although RIG-I and MDA5 sense distinct types of viral RNAs , they utilize a common adaptor protein called VISA ( also known as MAVS , IPS-1 or Cardif ) to transmit signals [3–6] . Upon binding to viral RNAs , RLRs are recruited to VISA located on the mitochondrial outer membrane , and this induces aggregation and activation of VISA [7] . VISA then acts as a central platform for recruitment of downstream signaling components , including TRAF2/3/5/6 , cIAP1/2 and WDR5 [6 , 8–10] . In these complexes , TRAF6 functions redundantly with TRAF2 and TRAF5 to activate IRF3 and IKK [11] . These processes lead to eventual transcription of downstream antiviral genes , including type I IFNs , proinflammatory cytokines and other effectors [12 , 13] . Protein phosphorylation and dephosphorylation play important roles in innate immune responses to RNA viruses by regulating the activation and deactivation of multiple RLR-mediated signaling components , such as RIG-I , VISA , TRAF3 , TBK1 and IRF3 [14–18] . In some cases , the enzymes that are responsible for their modifications are unknown . The RLR-mediated signaling pathways are also heavily regulated by other post-translational modifications , such as ubiquitination , sumoylation , methylation and [19–23] . How post-translational modifications cross-talk to regulate innate antiviral response remains enigmatic . Protein kinase A ( PKA ) is one of the first identified protein kinases , which is critically important for many divergent cellular processes , such as metabolism , cell cycle , cell migration , differentiation and apoptosis . PKA exists as a tetrameric holoenzyme with two regulatory subunits and two catalytic subunits in its inactive form . Cyclic adenylyl monophosphate ( cAMP ) causes dissociation of the inactive holoenzyme into a dimer of regulatory subunits bound to four cAMP and two free monomeric catalytic subunits [24] . Four regulatory subunits ( PKARIα , PKARIβ , PKARIIα and PKARIIβ ) and three catalytic subunits ( PKACα , PKACβ and PKACγ ) have been identified in humans . PKACα and PKACβ are ubiquitously expressed in most examined tissues , but PKACγ is specifically expressed in testis . Human PKACα and PKACβ are highly homologous , which share ~93% sequence identity at the amino acid level [25] . In this report , we identified PKACα and PKACβ as two redundant negative regulators of RNA virus-triggered induction of downstream antiviral genes . Viral infection activated PKACs , which in turn phosphorylated VISA at T54 , leading to impairment of VISA aggregation and its K48-linked polyubiquitination and degradation by the E3 ligase MARCH5 . We also showed that PKACs-deficiency or inactivation potentiated innate immunity to RNA viruses in cells and mice . Our findings reveal a critical mechanism of attenuating innate immune response at the late phase of viral infection and establish an un-described function for PKA in innate antiviral response .
VISA is a central adaptor protein in innate immune response to RNA virus . To identify potential kinases that regulate VISA-mediated signaling , we screened a cDNA library contains 352 kinase clones . We found that PKACα and PKACβ markedly inhibited VISA-mediated activation of the IFN-β promotor in HEK293 cells ( Fig 1 , panel A ) . PKACα and PKACβ also dose-dependently inhibited Sendai virus ( SeV ) -induced activation of the IFN-β promoter and ISRE , an enhancer motif for activated IRF3 . Overexpression of PKACα and PKACβ activated NF-κB ( Fig 1 , panel B ) , which is consistent with previous reports that PKACα and PKACβ can phosphorylate p65 on S276 [26 , 27] . Overexpression of PKACα and PKACβ also inhibited SeV-induced transcription of downstream genes such as IFNB1 , ISG15 and IKBA ( Fig 1 , panel C ) . In contrast , the testis specific PKA catalytic subunit PKACγ , the PKA regulatory subunits PKARIα and PKARIIβ , or the catalytic inactive mutants of PKACα ( K73A ) and PKACβ ( K73A ) [28] , did not inhibit SeV-induced activation of the IFN-β promoter ( Fig 1 , panel D ) . These results suggest that PKACα and PKACβ can inhibit RNA virus-triggered and VISA-mediated induction of downstream antiviral genes . We next determined whether endogenous PKACα and PKACβ are involved in regulation of virus-induced signaling . We found that knockdown of either PKACα or PKACβ by RNAi had no marked effects on SeV-induced activation of the IFN-β promoter , ISRE and NF-κB in reporter assays . However , simultaneous knockdown of both PKACα and PKACβ markedly potentiated SeV-induced activation of the IFN-β promoter , ISRE and NF-κB ( Fig 2 , panel A ) . Consistently , simultaneous but not individual knockdown of PKACα and PKACβ potentiated SeV-induced transcription of IFNB1 and CXCL10 genes ( Fig 2 , panel B ) . Because PKACα and PKACβ are highly conserved at both amino acid and mRNA sequence levels , we constructed two more RNAi plasmids ( PKACs-RNAi #1 and #2 ) , each of them simultaneously target both of human PKACα and PKACβ mRNAs . Simultaneous knockdown of the two catalytic subunits PKACα and PKACβ ( referred below as PKACs ) by these two RNAi plasmids potentiated SeV-induced activation of the IFN-β promoter , ISRE and NF-κB ( Fig 2 , panel C ) , as well as SeV or vesicular stomatitis virus ( VSV ) -induced transcription of IFNB1 , ISG56 and TNFA genes ( Fig 2 , panel D&E ) . The degrees of potentiation were correlated to the knockdown efficiencies of the RNAi plasmids ( Fig 2 , panel C , D&E ) . Knockdown of PKACs also potentiated IFN-β promoter activation triggered by poly ( I:C ) transfected into HEK293 cells ( Fig 2 , panel F ) . In addition , knockdown of PKACs also markedly enhanced SeV or VSV-induced phosphorylation of TBK1 , IRF3 and IκBα ( Fig 2 , panel G&H ) . However , knockdown of PKACs had no marked effects on IFN-α-induced activation of STAT1/2 and IFN-γ-induced activation of the IRF1 promoter ( Fig 2 , panel I ) . These data suggest that PKACα and PKACβ negatively regulate RNA virus-induced expression of downstream genes in a redundant manner . To investigate how PKACs function following viral infection , we determined whether viral infection triggers the accumulation of cellular cAMP . We found that SeV infection caused a transient decrease of cellular cAMP level at the early phase of infection ( 3 h ) but marked increase at the late phase of infection ( Fig 3 , panel A ) . The cAMP levels in RIG-I-deficient cells at the late phase of infection were not increased ( Fig 3 , panel B ) , suggesting that the increase of cAMP level at the late phase of viral infection was dependent on RIG-I-mediated signaling . In addition , SeV infection induced the phosphorylation of PKACs on T197 ( Fig 3 , panel C ) , which is a hallmark of PKACs activation [29 , 30] . These results suggest that viral infection leads to increase of cellular cAMP levels and activation of PKACs at the late phase of infection in a RIG-I-dependent manner . Additionally , we determined the effects of exogenous cAMP on virus-induced expression of downstream antiviral genes . We found that introduction of exogenous cAMP into the cells abolished SeV-induced transcription of IFNB1 gene , and knockdown of PKACs reversed the effects of exogenous cAMP ( Fig 3 , panel D ) . These results suggest that viral infection induced increase of cAMP levels and activation of PKACs , which in turn inhibit virus-triggered induction of downstream antiviral genes in a negative feed-back manner . We next determined the molecular mechanisms responsible for the inhibitory effects of PKACs on virus-triggered induction of downstream genes . In transient transfection and co-immunoprecipitation experiments , PKACα interacted with VISA , while PKACβ interacted with VISA , TRAF3 and TRAF6 . Neither PKACα nor PKACβ interacted with RIG-I ( Fig 4 , panel A ) . Cellular fractionation experiments indicated that PKACs were localized in the cytosol and at the mitochondria ( Fig 4 , panel B ) . Endogenous co-immunoprecipitation experiments indicated that PKACs were constitutively associated with VISA before and after SeV infection ( Fig 4 , panel C ) . In reporter assays , knockdown of PKACs enhanced upstream components RIG-I- , MDA- and VISA- but not downstream components TBK1- , IRF3- or IRF7-mediated ISRE activation ( Fig 4 , panel D ) . H89 is a specific and potent PKA inhibitor [31] , which completely reversed the inhibitory effects of PKACs on VISA-mediated activation of the IFN-β promoter ( Fig 4 , panel E ) . In the same experiments , H89 had no effects on the inhibition of VISA-mediated activation of the IFN-β promoter by DYRK2 , a kinase that negatively regulates virus-triggered signaling by targeting TBK1 for phosphorylation [16] . Collectively , these results suggest that PKACs inhibit virus-triggered induction of downstream antiviral genes by targeting VISA . Consistently , overexpression of PKACα or PKACβ but not their kinase inactive mutants caused a shift of VISA to higher molecular weight species ( Fig 5 , panel A ) . These higher molecular weight species of VISA were recognized by an antibody to phosphorylated serine and/or threonine ( p-S/T ) and removed by treatment with lambda protein phosphatase ( λ-PPase ) ( Fig 5 , panel B ) . These results suggest that PKACs phosphorylate VISA . We next determined the residues of VISA that are phosphorylated by PKACs . Prediction by GPS3 . 0 program indicates that VISA contains four consensus PKA phosphorylation residues , including T54 , S100 , T234 and S238 . Among the phosphorylation sites , T54 is highly conserved in mammals and the only residue located in the N-terminal CARD-like domain of VISA . Mutagenesis indicated that PKACα phosphorylated wild-type VISA and the VISA mutants VISA ( S100A ) , VISA ( T234A ) and VISA ( S238A ) but not VISA ( T54A ) ( Fig 5 , panel C ) . Reporter assays indicated that mutation of S100 , T234 and S238 of VISA to either alanine ( A ) or aspartic acid ( D ) had no marked effects on its ability to activate the IFN-β promoter . However , mutation of T54 of VISA to D , which mimics its phosphorylation , dramatically impaired its ability to activate downstream signaling . Unexpectedly , mutation of T54 of VISA to A , which mimics its un-phosphorylated form , also impaired its activity , though to a lesser degree ( Fig 5 , panel D ) . We further investigated the functions of the T54 mutants of VISA by reconstituting them into VISA-deficient HEK293 cells . We found that VISA ( T54D ) completely lost the ability to mediate SeV-triggered induction of downstream IFNB1 and ISG56 genes , while VISA ( T54A ) partially maintained the ability in comparison to wild-type VISA ( Fig 5 , panel E ) . Consistently , SeV-induced phosphorylation of TBK1 and IRF3 was partially and completely impaired in VISA ( T54A ) - and VISA ( T54D ) -reconstituted cells respectively in comparison to wild-type VISA-reconstituted cells ( Fig 5 , panel F ) . These results suggest that T54 is probably the target residue of PKACs . To determine whether PKACs indeed target T54 of VISA for phosphorylation , we generated a rabbit polyclonal antibody to a peptide containing phosphorylated T54 ( p-VISA-T54 ) . Immunoblot analysis indicated that PKACs but not their kinase inactive mutants caused phosphorylation of VISA at T54 ( Fig 5 , panel G ) . The phosphorylation of VISA at T54 was dramatically enhanced following SeV infection for 10 hours , while knockdown of PKACs impaired the phosphorylation of VISA at T54 and increased induction of RIG-I ( Fig 5 , panel H ) . In addition , knockdown of PKACs potentiated wild-type VISA and its S100A , T234A and S238A mutants but not T54A mutant mediated activation of the IFN-β promoter ( Fig 5 , panel I ) . Taken together , these results suggest that PKACs inhibit SeV-triggered induction of downstream genes by direct phosphorylation of VISA at T54 . It has been demonstrated that VISA forms prion-like aggregates on the mitochondrial membrane to activate innate immune response after viral infection [7] . We next investigated whether phosphorylation of VISA at T54 by PKACs impairs the formation of VISA aggregates . Co-immunoprecipitation experiments indicated that the self-association of VISA ( T54D ) was markedly decreased compared to wild-type VISA , VISA ( T54A ) , VISA ( S100A ) , or VISA ( S100D ) ( Fig 6 , panel A ) . Reconstitution experiments indicated that SeV-induced aggregation of VISA ( T54D ) was decreased in comparison to wild-type VISA and VISA ( T54A ) ( Fig 6 , panel B ) . In addition , knockdown of PKACs increased the formation of VISA aggregates following SeV infection ( Fig 6 , panel C ) . Confocal microscopy indicated that VISA ( T54A ) formed aggregates more dramatically than wild-type VISA at the early phase of infection ( 2–8 h ) and some aggregates remained even at the late phase of infection ( 24 h ) . Interestingly , VISA ( T54D ) colocalized with mitochondria but did not form aggregates before and after viral infection . In addition , VISA ( T54D ) was also not degraded after viral infection ( Fig 6 , panel D&E ) . These results suggest that PKACs-mediated phosphorylation of VISA at T54 impairs its aggregation and activation . In our experiments , we routinely found that PKACα and PKACβ but not their kinase inactive mutants caused down-regulation of VISA ( Fig 5 , panel A ) , while knockdown of PKACs up-regulated the levels of endogenous VISA ( Fig 5 , panel H ) . Kinetic experiments indicated that the levels of VISA were gradually down-regulated from 2–24 h after viral infection , and knockdown of PKACs slowed SeV-triggered down-regulation of VISA ( Fig 7 , panel A ) . We therefore tested the hypothesis that phosphorylation of VISA at T54 by PKACs impairs its aggregation and primes it for polyubiquitination and proteasomal degradation . We found that overexpression of PKACs enhanced K48- but not K63-linked polyubiquitination of VISA ( Fig 7 , panel B ) . Conversely , knockdown of PKACs inhibited SeV-induced K48- but not K63-linked polyubiquitination of VISA ( Fig 7 , panel C ) . Reconstitution experiments indicated that SeV-induced K48-linked polyubiquitination and degradation of VISA ( T54A ) was impaired in comparison to wild-type VISA ( Fig 7 , panel D ) . These results suggest that phosphorylation of VISA at T54 by PKACs primes it for K48-linked polyubiquitination and degradation . Previous studies have identified AIP4 , MARCH5 and RNF5 as E3 ubiquitin ligases that catalyze K48-linked polyubiquitination of VISA [32–34] . We found that individually knockdown of the examined three E3 ligases did not affect the phosphorylation of VISA by PKACα . However , knockdown of MARCH5 but not RNF5 or AIP4 dramatically inhibited the degradation of VISA mediated by PKACα ( Fig 7 , panel E ) . In addition , knockdown of MARCH5 but not RNF5 and AIP4 inhibited the synergistic activation of the IFN-β promoter induced by SeV and PKACs knockdown ( Fig 7 , panel F ) . Previously , it has been demonstrated that MARCH5 targets K7 and K500 of VISA for K48-linked polyubiquitination and degradation [33] . We found that PKACα caused degradation of wild-type VISA but not VISA ( T54A ) or VISA ( K7/500R ) , in which either the PKACs-mediated phosphorylation or MARCH5-mediated K48-linked polyubiquitination residues are mutated ( Fig 7 , panel G ) . Collectively , these results suggest that PKACs-mediated phosphorylation of VISA at T54 primes it for K48-linked polyubiquitination and degradation by MARCH5 . Finally , we investigated whether PKACs regulate innate antiviral response in immune cells and in vivo . We found that induction of Ifnb1 , Ifna4 and Il6 mRNAs by either SeV or encephalomyocarditis virus ( EMCV ) in bone marrow-derived dendritic cells ( BMDCs ) was markedly increased by knockdown of PKACs ( Fig 8 , panel A ) . In addition , mice treated with the PKA inhibitor H89 produced higher levels of serum cytokines including IFN-α4 , IFN-β and IL-6 upon EMCV infection ( Fig 8 , panel B ) and were more resistant to EMCV-induced death ( Fig 8 , panel C ) . These results suggest that PKA negatively regulates innate immune responses to RNA viruses in mice .
Proper and efficient innate immune response at the early phase of infection is critical for clearance of viruses , while timely termination of innate antiviral response at the late phase of infection is important for avoiding harmful immune damage and death of the host . How the innate antiviral response is delicately regulated has been heavily investigated in the past decade . In this report , we found that the house-keeping kinase PKA played an essential role in attenuating innate immune response to RNA virus by inactivating the central adaptor protein VISA in the virus-triggered signaling pathways . Overexpression of PKACα and PKACβ , but not PKACγ , the PKA regulatory subunits PKARIα and PKARIIβ , or the catalytic inactive mutants of PKACα ( K73A ) and PKACβ ( K73A ) markedly inhibited SeV-triggered and VISA-mediated induction of downstream antiviral genes . Interestingly , knockdown of either PKACα or PKACβ had no marked effects on SeV-triggered signaling , but simultaneous knockdown of both PKACα and PKACβ dramatically potentiated SeV-triggered induction of downstream genes . These results suggest that PKACα and PKACβ play redundant roles in inhibiting innate antiviral response , which is consistent with their redundant roles in regulation of many other cellular processes . Our experiments suggest that PKACs inhibit innate antiviral response by targeting the central adaptor protein VISA , which is mostly localized at the mitochondria . Consistent with previous reports [35] , we found that a fraction of PKACs was located in the mitochondria , and associated with VISA constitutively before and after viral infection . Overexpression of PKACs phosphorylated VISA at T54 and caused its degradation , whereas knockdown of PKACs inhibited SeV-induced phosphorylation of VISA at T54 and up-regulated its protein level . These results suggest that PKACs negatively regulate SeV-triggered induction of downstream genes by phosphorylating VISA at T54 . Consistently , reconstitution experiments indicated that mutation of T54 of VISA to D , which mimics its phosphorylated status , abolished VISA activity , and this mutant failed to mediate SeV-triggered signaling and induction of downstream antiviral genes . Un-expected , mutation of T54 of VISA to A , which mimics its un-phosphorylated status , partially inhibited its activity . The exact reasons responsible for this observation is currently unknown . The simplest explanation is that mutation of T54 of VISA to A causes its conformational changes that partially affect its activity . In fact , there are numerous cases that mutation of T to A or D does not act in an opposite way . For examples , both S366A and S366D mutants of STING/MITA , or both S527A and S527D mutants of TBK1 have greatly reduced ability to mediate IFN-β induction [16 , 36] . Our experiments suggest that phosphorylation of VISA by PKACs causes its inactivation by at least two processes . Firstly , the self-association , as well as SeV-induced aggregation of VISA ( T54D ) , was decreased in comparison to wild-type VISA or VISA ( T54A ) , whereas knockdown of PKACs increased VISA aggregation . Since virus-induced aggregation of VISA is an essential event for its activation , PKACs may inhibit virus-triggered signaling by phosphorylating VISA and impairing its aggregation . Secondly , phosphorylation of VISA by PKACs increased its K48- but not K63-linked polyubiquitination , whereas knockdown of PKACs decreased SeV-induced K48-linked polyubiquitination of VISA . In addition , SeV-induced K48-linked polyubiquitination and degradation of VISA ( T54A ) were abolished in comparison to wild-type VISA . These results suggest that phosphorylation of VISA at T54 by PKACs primes it for K48-linked polyubiquitination and degradation . We further showed that the E3 ubiquitin ligase MARCH5 but not RNF5 or AIP4 was responsible for mediating PKACs-primed K48-linked polyubiquitination of VISA . This is consistent with previous reports that MARCH5 is a mitochondrial-associated E3 ligase that negatively regulates virus-triggered induction of downstream genes at the late phase of infection [33] . It is possible that multiple E3 ligases are involved in regulation of VISA-mediated signaling in distinct cellular compartments and/or different phases of viral infection . Previously , the involvement of PKA in the regulation of innate immune response has not been reported . We found that SeV infection caused decrease of cAMP levels at the early phase of infection , but induced increase of cAMP levels at the late phase of infection , which was correlated to the increased PKA activity at the late phase of infection . Since virus-induced increase of cAMP levels at the late phase of infection was abrogated in Rig-I-/- cells , we conclude that virus-triggered induction of cAMP and PKA activity is dependent on RIG-I-mediated pathways . It has been well established that binding of a ligand to a G-protein coupled receptor ( GPCR ) activates adenylyl cyclase ( AC ) , which catalyzes the synthesis of cAMP from ATP . It is possible that the RIG-I pathways directly or indirectly through GPCR activate an AC , which leads to induction of cAMP . In light of the observation that PKACs are constitutively associated with VISA in the mitochondria , our results suggest that virus-triggered induction of cAMP modulates VISA activity in a temporal manner . In un-infected cells , basal cAMP maintains VISA activity at a steady level . At the early phase of infection , cAMP levels are down-regulated , which decreases PKACs activity and promotes VISA activity for efficient induction of downstream antiviral genes . At the late phase of infection , cAMP levels and PKACs activity are increased , which inactivates VISA and attenuates innate antiviral response . Therefore , PKA attenuates innate antiviral response in a feed-back negative regulatory manner . Our experiments suggest that PKA is not only important for negative regulation of innate antiviral response in cells , it is also essential for attenuating innate antiviral response in mice . We found that inhibition of PKA by the specific inhibitor H89 markedly potentiated SeV- and EMCV-induced expression of type I IFNs and IL-6 in the sera , and potentiated EMCV-induced death of infected mice . Because Prkaca and Prkacb double knockout is lethal in mice [37] , we are currently unable to directly determine the effects of PKACα and PKACβ deficiency on innate antiviral response in animals . Nevertheless , our studies provide solid evidences for the feed-back negative regulation of VISA-mediated innate antiviral response by the house-keeping kinase PKA , and certainly help to understand how innate immune response is terminated at the late phase of viral infection to avoid host damage .
All animal experiments were performed in accordance with the Wuhan University animal care and use committee guidelines . Lipofectamine 2000 ( Invitrogen ) ; RNase inhibitor ( Thermo ) ; SYBR ( Bio-Rad ) ; mouse monoclonal antibodies against Flag , HA , and β-actin ( Sigma ) , TBK1 , phospho-TBK1 ( Ser172 ) and phospho-Ser/Thr ( Abcam ) , phospho-PKA substrate and phospho-IRF3 ( Ser396 ) ( Cell Signaling Technology ) , rabbit polyclonal antibodies against IRF3 ( Santa Cruz Biotechnology ) and rabbit polyclonal antibodies against VISA ( Bethyl ) were purchased from the indicated manufacturers; mouse anti- PKACα/β antisera were raised against recombinant human full-length PKACβ; SeV , EMCV and HSV-1 were previously described [38 , 39]; HEK293 and THP1 cells ( ATCC ) were purchased from the indicated manufactures; HEK293T cells were originally provided by Dr . Gary Johnson ( National Jewish Health ) . NF-κB , ISRE , STAT1/2 , IFN-β promotor , and IRF1 promoter luciferase reporter plasmids , mammalian expression plasmids for Flag- or HA-tagged RIG-I , MDA5 , VISA , MITA , TBK1 and IRF3 were previously described [40–42] . Flag- or HA-tagged PKACα , PKACβ , PKACγ and their mutants were constructed by standard molecular biology techniques . Double-strand oligonucleotides corresponding to the target sequences were cloned into the pSuper . retro RNAi plasmid ( Oligoengine ) . The targeting sequences are as following . Human PKACα: 5′-GGGTGATGCTGGTGAAACA-3′; Human PKACβ: 5′-GAAGAGTCATGTTGGTAAA-3′ . The targeting sequences for both human PKACα and PKACβ ( PKACs ) : #1: 5′-GAAGGTTCAGTGAGCCCCA-3′; #2: 5′-TAGCCAAAGCCAAAGAAGA-3′ . siRNA oligonucleotides sequence for both mouse PKACα and PKACβ ( PKACs ) : 5′-TAGCCAAAGCCAAAGAAGATT-3′ . Cells ( 1×105 ) were seeded in 24-well plates and transfected the following day by standard calcium phosphate precipitation . In the same experiment , empty control plasmid was added to ensure that each transfection receives the same amount of total DNA . To normalize for transfection efficiency , 0 . 01 μg of pRL-TK or pRL-SV40 ( Renilla luciferase ) reporter plasmid was added to each transfection [43 , 44] . Luciferase assays were performed using a dual-specific luciferase assay kit ( Promega ) . For co-immunoprecipitation , cells ( 1×107 ) were lysed in 1 mL NP-40 lysis buffer ( 20 mM Tris-HCl [pH 7 . 4] , 150 mM NaCl , 1 mM EDTA , 1% Nonidet P-40 , 10 μg/mL aprotinin , 10 μg/mL leupeptin , and 1 mM phenylmethylsulfonyl fluoride ) . For direct analysis of protein expression , cells were lysed with SDS-PAGE loading buffer followed by ultra-sonication . Co-immunoprecipitation and immunoblot analysis were performed as previously described [45 , 46] . SDD-AGE was performed as previously described [47] . Ubiquitination assays were performed as previously described [48 , 49] . GraphPad Prism software was used for all statistical analyses . Quantitative data displayed as histograms are expressed as means ± SD ( represented as error bars ) . Data were analyzed using a Student’s unpaired t test or multiple t test . The number of asterisks represents the degree of significance with respect to p values . Statistical significance was set at a p < 0 . 05 . Mice in each sample group were selected randomly in mouse experiments . The sample size ( n ) of each experimental group is described in the figure legend . | VISA is a central adaptor protein required for innate immune response to RNA virus . Phosphorylation of VISA by protein kinase A leads to its polyubiquitination and degradation by the E3 ligase MARCH5 at the late phase of viral infection , which provides a critical control mechanism for the host to avoid excessive and harmful immune response . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"viral",
"transmission",
"and",
"infection",
"molecular",
"probe",
"techniques",
"immunology",
"microbiology",
"plasmid",
"construction",
"viruses",
"rna",
"viruses",
"dna",
"construction",
"epigenetics",
"molecular",
"biology",
"techniques",
"bioenergetics",
"mitochondria",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"immunoblot",
"analysis",
"genetic",
"interference",
"proteins",
"gene",
"expression",
"molecular",
"biology",
"immune",
"response",
"biochemistry",
"rna",
"cell",
"biology",
"post-translational",
"modification",
"nucleic",
"acids",
"virology",
"genetics",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles",
"organisms"
]
| 2017 | PKACs attenuate innate antiviral response by phosphorylating VISA and priming it for MARCH5-mediated degradation |
Seroprevalence and incidence of toxoplasmosis in women of child bearing age has remained a contentious issue in the Indian subcontinent . Different laboratories have used different patient recruitment criteria , methods and variable results , making these data difficult to compare . To map the point-prevalence and incidence of toxoplasmosis in India . In this cross-sectional study , a total of 1464 women of fertile age were recruited from 4 regions using similar recruitment plans . This included women from northern ( 203 ) , southern ( 512 ) , eastern ( 250 ) and western ( 501 ) regions of India . All samples were transported to a central laboratory in Delhi and tested using VIDAS technology . Their age , parity , eating habits and other demographic and clinical details were noted . Most women were in the 18–25 years age group ( 48 . 3% ) , followed by 26–30 years ( 28 . 2% ) and 31–35 years ( 13 . 66 ) . Few ( 45 ) women older than 35 yr . were included . Overall prevalence of anti-Toxoplasma IgG antibodies was seen in 22 . 40% , with significantly more in married women ( 25 . 8% ) as compared to single women ( 4 . 3% ) . Prevalence increased steadily with age: 18 . 1% in the 18–25 yr . age group to 40 . 5% in women older than 40 yr . The prevalence was high ( 66% ) in those who resided in mud houses . Region-wise , the highest prevalence was observed in South India ( 37 . 3% ) and the lowest ( 8 . 8% ) in West Indian women . This difference was highly significant ( P<0 . 001 ) . Prevalence was 21 . 2% in East India and 19 . 7% in North India . The IgM positivity rate ranged from 0 . 4% to 2 . 9% in four study centers . This pan-India study shows a prevalence rate of 22 . 4% with a wide variation in four geographical regions ranging from as low as 8 . 8% to as high as 37 . 3% . The overall IgM positivity rate was 1 . 43% , indicating that an estimated 56 , 737–176 , 882 children per year are born in India with a possible risk of congenital toxoplasmosis .
Toxoplasma gondii ( T . gondii ) infection is a significant member of the TORCH group of diseases which cause congenital abnormalities , and even fetal loss . TORCH group infectious agents also consist of Rubella , Cytomegalovirus , Herpes viruses and Treponema pallidum . In India , awareness about these infections that cause congenital conditions is poor [1]–[4] . Most women who seek medical attention , or are referred by obstetricians , are those who have had an undesirable pregnancy outcome [5] , [6] . Toxoplasmosis is caused by the protozoan parasite T . gondii . It has a wide host range , infecting most warm-blooded species but the life cycle is completed only in felids [7] , [8]; Indeed only cats can shed the environmentally-resistant stage of the parasite ( oocyst ) in their feces [9] , [10] . Humans usually become infected by ingesting food or water contaminated with cat faeces containing oocysts [11] , [12] or by eating under-cooked meat containing the encysted stage of the parasite ( tissue cysts ) [13] , [14] . Infection acquired during pregnancy can be transmitted to the fetus , sometimes with serious consequences . There are numerous serological surveys of T . gondii infection in pregnant women in India , but most of them were based on convenience sampling , and often selectively in women with bad outcome of pregnancy [15]–[18] . Here , we present the first designed survey for determining the prevalence rate of anti-T . gondii antibodies in Indian women of reproductive age from four geographic regions: East , West , North and South India .
Blood sampling was performed from October 2011 to October 2012 . The Institutional Review Board of AIIMS approved this study ( IEC/NP-92/2011 ) . Informed written consent was obtained from all participating women who agreed to participate in the study . The study followed the STROBE guidelines ( Supplementary file S2 ) . Serum samples were assayed for anti-Toxoplasma IgG and IgM antibodies by a commercially available Vitek Immuno-Diagnostic Assay System ( VIDAS , BioMerieux SA , France ) , strictly following the manufacturer's instructions . All IgG and IgM positive samples were tested for IgG avidity using the same technology ( VIDAS , BioMerieux SA , France ) , as published previously [3] . An avidity index of <0 . 200 indicates low avidity; an index of 0 . 200–0 . 299 indicates borderline avidity , and an index of >0 . 300 denotes high avidity for IgG . High avidity enables exclusion of a recent infection of <4 months duration . More details are provided in a related publication [3] . Data was entered in Excel sheet and imported into SPSS statistical program for analysis . For statistical evaluation of binomial data , the χ2 test with 95% confidence intervals according to Clopper and Pearson were used; P values <0 . 05 were considered statistically significant . Incidences ( IgM positivity ) and prevalence ( IgG positivity ) rates are expressed as percentages . To estimate the approximate number of babies born with risk of congenital T . gondii infection in India per annum , following formula was used . Total population and live birth rates were taken from the Government of India official website [19] . For example , we had 434 women in their first trimester , of whom 7 ( 1 . 61% ) were IgM positive . In the first trimester of pregnancy the rate of congenital transmission is reported to be 13% [6] . Hence , presuming that all women were in their first trimester , the approximate number of children born with a risk of congenital T . gondii infection would be Using the same formula , out of 177 women in their second trimester , 4 ( 2 . 25% ) women were IgM positive . Hence presuming a transmission rate of 29% in the second trimester [6] , the approximate number of children born with a risk of congenital T . gondii infection would be
Data on 1464 women of reproductive age which ranged from 18 to 45 years ( mean ± SD , 26 . 9±5 . 9 ) from four geographically distinct regions of India are presented here . Of these , 250 ( 17 . 1% ) were from East India ( Assam ) , 203 ( 13 . 8% ) from North India ( Delhi and national capital region ) , 499 ( 34 . 1% ) from Western India ( Gujarat ) , and 512 ( 34 . 9% ) from South India ( Karnataka ) . Region-wise , the mean age of the women in the East was 24 . 2±4 . 2 yr . , 29 . 3±5 . 8 yr . in the North , 25 . 2±5 . 9 yr . in the West and 29 . 2±5 . 7 yr . in the South . The difference was insignificant . The distribution of various age groups is shown in figure 1 . Out of 1464 women , 233 ( 15 . 9% ) were single with a mean age of 22 . 4±3 . 8 ( 18–41 yr . age range ) while 1231 ( 84 . 08% ) were married with a mean age of 27 . 8±5 . 9 ( 18–45 yr . age range ) . Of the 1231 married women , 297 ( 24 . 1% ) were nulliparous with a mean age of 30 . 2±6 . 3 and 934 ( 75 . 9% ) were parous with a mean age of 27 . 1±5 . 6 ( Figure 2 , Flow chart 1 ) . All single women were nulliparous and non-pregnant . Of the 934 parous women , 471 ( 50 . 4% ) were single gravida and 463 ( 49 . 6% ) were multigravida . Their mean age was 25 . 3±5 . 0 and 28 . 8±5 . 6 , respectively . The number of gravida was as high as 7 ( 6 women; 1 . 3% ) . Of the 934 gravida women 356 ( 38 . 2% ) had at least one live birth , while 153 ( 16 . 4% ) had all adverse pregnancy outcomes . The remaining 424 ( 45 . 4% ) were primigravida . Overall , 751 women were pregnant at the time of sample collection . Of these , 434 ( 57 . 8% ) were in their first trimester , 176 ( 23 . 4% ) in their second and 141 ( 18 . 8% ) in their third trimester . The overall seroprevalence was 22 . 4% ( 328 of 1464 ) . The prevalence rates varied significantly across the 4 regions , with the highest ( 37 . 3% ) in South India and the lowest in West India ( 8 . 8% ) . The difference was highly significant ( Figure 3 ) . We also observed a significant difference in the prevalence rates of anti-Toxoplasma antibodies between single women ( 4 . 29% , 95% CI; a range of 1 . 9% to 6 . 9% ) and married women ( 25 . 0% , 95% CI; a range of 22 . 6% to 27 . 4% ) ( p<0 . 005 ) . However , seroprevalence increased with age ( Figure 4 , trend line ) . Prevalence was lower than 11 . 7% among those under 20 yr . of age , but steadily increased to 40 . 4% in those who were older than 41 yr . Region-wise proportion of anti-Toxoplasma IgG antibody positive women in various age groups is shown in table 2 . Most ( 74 . 2% ) of the 328 seropositive pregnant women were multigravida and only a quarter ( 25 . 8% ) were primigravida ( Figure 5 , Flow chart 2 ) . Of the 208 seropositive pregnant women at the time of sampling , two thirds ( 138/208 or 66 . 35% ) were in their first trimester while 36/208 ( 17 . 30% ) were in their second trimester and 34/208 ( 16 . 35% ) in their third . Only 21 women out of 1464 ( 1 . 43% ) had anti-Toxoplasma IgM antibodies . Trimester-wise , 434 women were in their first trimester and 7 ( 1 . 61% ) of them were IgM positive , while 177 women were in their second trimester and 4 ( 2 . 25% ) of these had IgM antibodies . None of the 141 pregnant women in their third trimester was IgM positive . Ten of 479 non-pregnant women had IgM antibodies . IgM positivity was highest ( 2 . 9% , 15 out of 512 ) in South India , followed by 0 . 8% in Eastern ( 2 out of 250 ) , 0 . 6% in Western ( 3 of 499 ) and 0 . 4% , in North ( 1 out of 203 ) India . The region-wise pattern was similar to the prevalence rate of IgG antibodies . All IgM positive women were also IgG positive and none , except for one woman , was IgG negative and IgM positive . From South India , 191 women were seropositive , of which 15 ( 7 . 8% ) were IgM positive , while in North India only 1 out of 40 ( 2 . 5% ) and in East India 2 out of 53 ( 3 . 7% ) IgG positive women were also IgM positive . However , in West India the IgM positivity rate amongst the IgG positives was 6 . 8% ( 3 out of 44 ) . This change was significant ( <0 . 001 ) when estimated out of the total study subjects as opposed to only out of IgG positives . Out of 21 IgM positive subjects , 6 had low avidity , 8 were borderline and 7 showed high avidity . All low avidity subjects were followed up for 6 months . Of these , 3 were pregnant and 2 delivered normal babies while one newborn had congenital hydrocephalus and microphthalmia . Follow-up samples from babies could not be tested . All low avidity seropositive women were from South India . Most women ( 97 . 1% ) belonged to low or lower middle income groups . The majority of women from South and East India resided in mud-plastered houses and consumed tube well/hand pump water without using any disinfectant or filter . The difference in the living conditions was highly significant ( p<0 . 0001 ) between South India and West India and also between South India and North India ( p<0 . 001 ) , while the difference between North and East India was not significant ( Figure 3 ) . General socio-behavioral characteristics of these women are shown in table 3 . Most women ( 1253 of 1464; 85 . 6% ) were involved in housekeeping . In spite of having a rural background ( 75 . 1% ) , our study showed that 90 . 6% ( 1328 ) of the women had elementary education . A history of contact with animals was found in 426 ( 29 . 1% ) women but pets were significantly more common ( 53 . 5% ) in South Indian households ( Table 3 ) . Consumption of raw salad was common ( 1360/1464; 92 . 89% ) across the country . Of the 308 women from North India who are excluded from our final data analysis , 178 were parous and 130 were nulliparous . Their mean age was 29 . 1 yr . ( range of 15–52 yr . ) . Of the parous women , 78 ( 43 . 8% ) were primigravida and 45 ( 25 . 3% ) were two gravida . Fifty two ( 29 . 2% ) women were multigravida and the number of gravida was as many as eight .
This community based study provides a significant and much needed resource material specific to India . Our study showed that age was one major variable of a higher prevalence rate of toxoplasmosis . The prevalence was also higher in those who were married and multigravida than those who were unmarried/single . This higher prevalence was not associated with the number of gravida per se but rather to higher mean age [27 . 8±5 . 9 ( range 18–45 yr . ) ] of those who were married than those who were unmarried/single women [22 . 4±3 . 8 ( range 18–41 yr . ) ] . This conclusion is further validated with observations that age of women from South India was higher than other regions and prevalence was also significantly higher in women of this region . There are also anecdotal reports associating multiple sexual exposures with high prevalence and possibility of sexual transmission of toxoplasmosis [20] . Over all prevalence of toxoplasmosis in the present study was significantly lower than reported earlier from India [3] . One plausible explanation for this difference could be that in the present study we included all women of fertile age from various cohorts , as compared to an earlier study in which we included only pregnant women . However , even if we combine those with suspected TORCH infections , who were otherwise excluded from final data analysis , the overall prevalence rate in North Indian women was 17 . 4% , which was significantly lower than the 45% reported by us 10 years ago , in the same population using the same diagnostic techniques . Whether it was due to improved awareness and social hygiene in the last 10 years , or due to selection biases in the two studies , cannot be ascertained with certainty . The difference in the prevalence rate between women from South and West India was highly significant . It could be mainly due to socio-cultural and climatic factors . The climatic conditions in South India favour sustenance and proliferation of Toxoplasma oocysts . Also a highly significant number of households owned cats in this region . Moreover , as a social culture , South Indians do not wear shoes and most often are barefoot or wear sleepers only . This might increase chances of transferring T . gondii oocysts from soil and water to their food [9]–[13] . Western India , on the other hand , is a dry arid climatic zone where temperature in May-June averages 46°C . Socio-culturally also , the population in West India must wear shoes due to the high temperature and sandy soil . These climatic conditions are detrimental to T . gondii to maintain its life cycle . We [21] and others [22] , [23] have previously demonstrated the role of environmental conditions on the prevalence of toxoplasmosis . Although consumption of raw/undercooked meat and exposure to soil through farming or gardening have been associated with a higher risk of infection in various studies [1] , [4] , [21] , no such correlation was observed in the multivariate analysis done in the present study . In India , consumption of raw or undercooked meat is extremely rare , hence this route of infection is theoretically negligible . The type of food , social considerations and quality of water consumed were the most likely factors associated with high prevalence of toxoplasmosis in South India , besides higher age . Water and food-borne outbreaks of toxoplasmosis have been well documented worldwide [11] , [12] and also from India [24] . Anti-toxoplasma IgM antibodies were considered indicative of a recent infection for several decades , but then it was realized that these antibodies could persist for several months , even years after the primary infection [2] , [3] , [5] . Hence in the 1990s a new test , based on affinity levels of IgG antibodies binding with antigen , was developed , and known as the avidity test . Determination of avidity helps in determining if the infection is of recent origin or more than 4 months old [25]–[27] . The IgM positivity and low avidity rates we observed in this study may seem very low , but cumulative figures are alarming . India has a population of more than 1 . 22 billion and the live birth rate is 22 . 22/1000 per annum [19] . Taking these IgM rates into consideration , a conservative estimate of child births with a possible risk of congenital toxoplasmosis [6] would be between 56737 and 176882 . This would translate into health and rehabilitation expenses to treat and rehabilitate the congenitally-infected children , many of whom may remain asymptomatic for several years . Unfortunately , many of such congenitally-infected adolescents and adults are not diagnosed accurately , whether the infection was acquired in-utero or after birth . Our calculations , which are based only on IgM positivity rates , are not a very reliable marker of recent infection , as discussed in the previous paragraph . Therefore , these estimates need more validation studies from India for transplacental transmission rates of T . gondii using IgG avidity or molecular methods , such as PCR . Due to a high false IgM positivity rate , we also conclude that carrying out only IgM testing without IgG testing is not an advisable approach of investigating toxoplasmosis , and all patients must be tested first for IgG and , if found positive , samples would be subjected to IgM and/or avidity tests . Our study had some limitations . We attempted to extract a maximum of information regarding pan-India seroprevalence of toxoplasmosis and correlate it with environmental and socio-cultural considerations . It would have been ideal to include more centers and more samples but this was not feasible due to time and financial constraints . Also , we could not do multiple follow-up sampling to find out true incidence rates in various seasons and in IgM positive women . | Toxoplasmosis is a protozoan parasitic disease commonly transmitted and propagated by cats as family pets . Infection acquired during pregnancy can lead to congenital abnormalities in the fetus , still birth or intrauterine death . Seroprevalence and incidence of toxoplasmosis in Indian women of child bearing age has remained a contentious issue . Different laboratories have used different patient recruitment criteria , methods and variable results , making these data unreliable . There is no published pan-India seroprevalence study . Hence , a seroprevalence study was undertaken comprising 1464 women of reproductive age representing four distinct geographical regions of India . This resulted in an estimated prevalence of 22 . 4% ( 328 ) ; the highest prevalence being in South India ( 37 . 3% ) followed by East India ( 21 . 2% ) and North India ( 19 . 7% ) . West Indian women had the lowest seroprevalence ( 8 . 8% ) . This difference was highly significant . In our analysis we determined the possible risk-factors of infection in these women . These included lower socioeconomic status , residing in mud plastered houses , consumption of raw salad , drinking untreated water , owning pets and advanced age . Overall , the incidence rate of toxoplasmosis was 1 . 43% . Extrapolating the data , we estimate that between 56 , 737 and 176 , 882 children a year may be born in India with a possible risk of congenital toxoplasmosis , which can manifest itself in-utero or several years after birth . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"veterinary",
"diseases",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"medicine",
"women's",
"health",
"veterinary",
"microbiology",
"obstetrics",
"and",
"gynecology",
"epidemiology",
"biology",
"and",
"life",
"sciences",
"veterinary",
"science",
"pediatrics"
]
| 2014 | Serologic Prevalence of Toxoplasma gondii in Indian Women of Child Bearing Age and Effects of Social and Environmental Factors |
Advanced cholangiocarcinoma continues to harbor a difficult prognosis and therapeutic options have been limited . During the course of a clinical trial of whole genomic sequencing seeking druggable targets , we examined six patients with advanced cholangiocarcinoma . Integrated genome-wide and whole transcriptome sequence analyses were performed on tumors from six patients with advanced , sporadic intrahepatic cholangiocarcinoma ( SIC ) to identify potential therapeutically actionable events . Among the somatic events captured in our analysis , we uncovered two novel therapeutically relevant genomic contexts that when acted upon , resulted in preliminary evidence of anti-tumor activity . Genome-wide structural analysis of sequence data revealed recurrent translocation events involving the FGFR2 locus in three of six assessed patients . These observations and supporting evidence triggered the use of FGFR inhibitors in these patients . In one example , preliminary anti-tumor activity of pazopanib ( in vitro FGFR2 IC50≈350 nM ) was noted in a patient with an FGFR2-TACC3 fusion . After progression on pazopanib , the same patient also had stable disease on ponatinib , a pan-FGFR inhibitor ( in vitro , FGFR2 IC50≈8 nM ) . In an independent non-FGFR2 translocation patient , exome and transcriptome analysis revealed an allele specific somatic nonsense mutation ( E384X ) in ERRFI1 , a direct negative regulator of EGFR activation . Rapid and robust disease regression was noted in this ERRFI1 inactivated tumor when treated with erlotinib , an EGFR kinase inhibitor . FGFR2 fusions and ERRFI mutations may represent novel targets in sporadic intrahepatic cholangiocarcinoma and trials should be characterized in larger cohorts of patients with these aberrations .
Biliary tract cancers ( BTC ) comprise malignant tumors of the intrahepatic and extrahepatic bile ducts . Known risk factors for BTC are the liver flukes O . viverrini and C . sinensis in high prevalence endemic regions in southeast Asia [1]–[3] , as well as primary sclerosing cholangitis [4]–[7] , Caroli's disease [8] , hepatitis B and hepatitis C [9]–[14] , obesity [13] , hepatolithiasis [15] , [16] and thorotrast contrast exposure [17] , [18] . Surgical approaches such as resection and liver transplantation represent the only curative treatment approaches for BTC [19] . Unfortunately , most patients present with surgically unresectable and/or metastatic disease at diagnosis . Systemic therapy with gemcitabine and cisplatin has been established as the standard of care for patients with advanced disease , but is only palliative [20] , emphasizing the imminent need for novel therapies . Multiple studies have reported the presence of mutations/allelic loss of known cancer genes in BTC [21]–[39] and recently , a prevalence set of 46 patients was used to validate 15 of these genes including: TP53 , KRAS , CDKN2A and SMAD4 as well as MLL3 , ROBO2 , RNF43 , GNAS , PEG3 , XIRP2 , PTEN , RADIL , NCD80 , LAMA2 and PCDHA13 . Recent studies have also identified recurrent mutations in IDH1 ( codon 132 ) and IDH2 ( codons 140 and 172 ) with a prevalence of 22–23% associated with clear cell/poorly differentiated histology and intrahepatic primary [40] , [41] . Fusions with oncogenic potential involving the kinase gene ROS1 have been identified in patients with BTC with a prevalence of 8 . 7% in a recent study [42] . Less frequently , mutations in sporadic BTC have been reported in EGFR [43] , [44] , BRAF [45] , NRAS [40] , [46] , PIK3CA [40] , [46] , [47] , APC [40] , CTNNB1 [40] , AKT1 [40] , PTEN [40] , ABCB4 [48] , ABCB11 [49] , [50] , and CDH1 [51] as well as amplifications in ERRB2 [52] . Recently , two independent studies reported the presence of FGFR fusions in cholangiocarcinoma; a single case with FGFR2-AHCYL1 [53] as well as several cases identifying FGFR2-BICC1 fusions [53] , [54] . Arai et al . evaluated the presence of FGFR2 fusions in a cohort of 102 cholangiocarcinoma patients observing that the fusions occurred exclusively in the intrahepatic cases with a prevalence of 13 . 6% [53] . Due to the presence of known dimerization motifs in the fusion partners , Wu et al . conducted mechanistic studies that demonstrated the in vitro interaction of FGFR2-BICC1 and other fusions that was not observed in the presence of wildtype FGFR2 [54] . Furthermore , overexpression of the FGFR2-BICC1 and other selected fusions resulted in altered cell morphology and increased cell proliferation [54] . These data led to the conclusion that the fusion partners are facilitating oligomerization , resulting in FGFR kinase activation in tumors possessing FGFR fusions . In addition , in vitro and in vivo assessment of the sensitivity of cell lines containing an FGFR2 fusion to an FGFR inhibitor demonstrated sensitivity to treatment only in the fusion containing cells [53] , [54] , suggesting the presence of FGFR fusions may be a useful predictor of tumor response to FGFR inhibitors . To comprehensively explore the genetic basis of sporadic intrahepatic cholangiocarcinoma ( SIC ) , with emphasis on elucidation of therapeutically relevant targets , we performed integrated whole genome and whole transcriptome analyses on tumors from 6 patients with advanced , sporadic intrahepatic cholangiocarcinoma ( SIC ) . Notably , recurrent fusions involving the oncogene FGFR2 ( n = 3 ) were identified . A patient whose tumor presented with an FGFR2-MGEA5 fusion has demonstrated preliminary evidence of anti-tumor activity manifest as stable disease accompanied by CA19-9 reduction and tumor necrosis to ponatinib , a pan-FGFR inhibitor ( in vitro FGFR1 IC50≈24 nM , FGFR2 IC50≈8 nM , FGFR3 IC50≈8 nM and FGFR4 IC50≈34 nM ) . In another patient whose tumor possessed an FGFR2-TACC3 fusion , preliminary anti-tumor activity of pazopanib ( in vitro FGFR2 IC50≈350 nM ) was also noted . After progression on pazopanib , the same patient also responded to ponatinib and again demonstrated tumor shrinkage . Additionally , a non-FGFR fusion patient was found to have allele-specific preferential expression of a loss of function mutation in ERRFI1 , a direct negative regulator of EGFR activation . Similarly , rapid and robust disease regression was noted in the patient with an ERRFI1 mutant tumor when treated with erlotinib , an EGFR kinase inhibitor . Results suggest that these novel targets in the EGFR and FGFR pathways may be therapeutically relevant in patients with sporadic cholangiocarcinoma .
We identified 327 somatic coding mutations , with an average of 55 mutations/tumor ( range 34–112 ) , within our cohort ( Table 1 , Figure 1 ) . Nonsynonymous single nucleotide variations were the predominant class in all of the patients . Patients 1 and 2 accumulated a high number of synonymous mutations in comparison to the other patients . Patient 5 carried the most stops gained likely contributing to a higher number of pseudogenes in comparison to the others and was also the only patient to carry several predicted high impact mutations affecting splice site acceptor regions ( Figure 1 , light green , percentage <5% ) . In addition , patient 6 also carried a codon change plus insertion variation . Sequencing statistics are provided in Table 2 . Genes with mutations in more than one case included CSPG4 ( n = 2 ) , GRIN3A ( n = 2 ) and PLXBN3 ( n = 2 ) ( Table S1 ) ; with half of these predicted to be potentially damaging by SIFT [55] , Polyphen [56] , Mutation Assessor [57] and Mutation Taster [58] . While there was overlap in the somatic landscape of SIC with liver-fluke associated cholangiocarcinoma , hepatocellular cancer and pancreatic cancer , most of the aberrations detected in our study were distinct ( Table 3 ) . More importantly , using previously published methods [59] , we identified molecular fusions involving FGFR2 that were felt to be therapeutically relevant in 3 patients . Additionally , these fusions were validated with a break apart Fluorescent In situ Hybridization ( FISH ) assay ( Figure 2 ) . Notably , the patients who did not harbor the FGFR2 fusions were negative using the same assay . Two of the three patients with FGFR2 fusions ( Patients 4 and 6 ) were treated with FGFR inhibitors while the third patient ( Patient 5 ) , experienced clinical decline prior to the availability of results and as such did not receive any further therapy . Furthermore , overexpression of an SNV in ERRFI1 ( E384X ) , a negative regulator of EGFR , was detected in a non-FGFR2 translocation patient's tumor . Taken together , our results constitute important therapeutically actionable alterations in patients with advanced SIC ( Text S1 ) . Comparative pathway analysis of genes carrying small scale nucleotide variations ( SsNVs ) has implicated several major pathways , possibly interacting as a network , that are predicted to underlie disease in all of our studied biliary carcinoma patients . These shared pathways include EGFR , EPHB , PDGFR-beta , Netrin-mediated and Beta1 integrin mediated signaling pathways ( Figure 3 and Tables S2 and S3 ) . Interestingly , most of these pathways have known roles in mediating epithelial-to-mesenchymal cell transitions , which occur frequently during development as well as tumorigenesis . Cell growth and motility is inherent to the successful progression of both biological processes . Studies of the nervous system and lung development have shown that Netrins act to inhibit FGF7 and FGF10 mediated growth or cell guidance [60] . In addition , Netrin-1 has a known role in mediating cell migration during pancreatic organogenesis [60] . Furthermore , Netrin-1 acts as a ligand for α3β1 and α6β4 integrins , both of which are involved in supporting adhesion of developing pancreatic epithelial cells with Netrin-1 although it is thought that α6β4 plays the principle role during this process [60] . Interestingly , α3β1 has been hypothesized to play a role during the process of angiogenesis , when chemoattractants and chemorepellents act to guide filopodia during migration [60] . The α3β1 integrin receptor may act together with additional pathways proposed to play a role during angiogenesis such as VEGF , PDGFR-beta [61] , and EphrinB [62] as well as tumorigenesis [60] . Patients 3 and 4 also shared several genes acting in cadherin signaling pathways ( Tables S3 , S4 ) , which are important for maintaining cell-cell adhesion and are known to be intimately integrated with EGFR and FGFR signaling pathways [63] . In addition to the variations identified in genes acting in EGFR and/or FGFR signaling pathways , we also report multiple sSNVs and copy number variations ( CNVs ) ( Figure 4 ) in genes such as HDAC1 , TP53 , MDM2 and AKT1 , acting in interaction networks or regulatory pathways involving the fusion partner genes in patients 5 ( BICC1 ) , and 6 ( TACC3 ) ( Table 4 ) . Known mutations in BICC1 have been shown to disrupt canonical Wnt signaling [64] and genes , such as BCL9 , involved in this pathway are known to regulate a range of biological processes such as transcription and cell proliferation and carry variations in patient 5 ( Table 4 ) . CSPG4 , a target that is being investigated for antibody-based immunotherapy in preclinical studies of triple negative breast cancer [65] , is involved in the Wnt signaling pathway , and carries variations in both patients 1 and 2 , however , it is not mutated in patient 5 . TACC3 is known to mediate central spindle assembly and multiple genes including CDCA8 , BUB1 , and TACC1 , belonging to the TACC3 interaction network exhibit aberrant copy number in patient 6 ( Table 4 ) . A recent study has also implicated TACC3 in EGF-mediated EMT when overexpressed [64] , and we find that the PLCG1 , MAP2K1 , and MAPK8 genes , which act in both FGFR and EGFR regulatory pathways , exhibit CNV in patient 6 . We also note that the DNAH5 gene encoding a dynein protein which is part of the microtubule-associated motor protein complex carries two G→C missense mutations in patient 6 ( Table S1 ) . Several genes carrying more than one variation in either the same patient or different patients also included genes with known roles similar to genes in FGFR/EGFR pathways including axon guidance , invasive growth , or cell differentiation ( NAV3 , LAMC3 , PLXNB3 , and PTPRK ) ( Table S1 ) . In the case of patient 4 , our studies suggest that the primary effect of the FGFR2-MGEA5 fusion is on FGFR2 related signaling , since changes in expression were observed in FGF8 ( p<0 . 05 ) and the genome of this patient also carries a 4-bp insertion ( ∧GTGT ) in the FGFR4 gene ( Table S1 ) . Patient 4 is a 62 year-old white female found to have a left-sided intrahepatic mass with satellite lesions , with metastasis to regional lymph nodes ( Table 5 ) . A biopsy of the liver mass revealed the presence of a poorly differentiated adenocarcinoma that was consistent with intrahepatic cholangiocarcinoma ( CK7+ , CEA+ , CK20+ , Hep-par 1− , TTF-1− ) ( Table 6 ) . She received gemcitabine and cisplatin and obtained clinical benefit in the form of stable disease for 6 months , followed by disease progression . She was re-treated with gemcitabine and capecitabine systemic therapy and attained stable disease for 6 months , followed by disease progression . A clinical trial of pegylated hyaluronidase ( PEGPH20 ) produced only stable disease for 4 months , followed again by disease progression . At this juncture , she underwent a liver biopsy to obtain tissue for whole genome characterization of her tumor . She was found to have an FGFR2-MGEA5 fusion ( Table 7 , Figure 2 ) and ponatinib monotherapy was pursued as salvage treatment . Evaluation of pre-treatment immunohistochemistry demonstrated increased expression of FGFR2 and FGFR3 ( Figure 5 ) and Clinical Laboratory Improvement Amendments ( CLIA ) validation by quantitative PCR revealed increased expression of FGFR3 ( Table S5 ) . In order to further validate the activation of the receptor , we conducted immunohistochemistry ( IHC ) of pFRS2 Y436 and pERK ( MAPK ) that revealed strong expression of pFRS2 Y436 and pERK ( Figure 6 ) , thus confirming activation of the receptor . Ponatinib was initiated at 45 mg given orally on a daily schedule . Approximately 6 weeks after initiation of therapy she was noted to have anti-tumor activity that was characterized by central necrosis of a caudate liver lobe mass , shrinkage of metastatic lymph nodes involving the right cardiophrenic angle , central necrosis and shrinkage of a metastatic supraceliac axis lymph node ( Figure 7 ) and reduction in CA 19-9 from 1408 U/ml to 142 U/ml . Per RECIST criteria , she exhibited stable disease with a 14% decrease in the sum of largest diameters but with tumor necrosis and reduction in the CA19-9 tumor marker ( 89 . 8% ) . While the evidence is preliminary in nature , it was felt that the combination of tumor shrinkage not meeting the RECIST criteria definition of partial response , tumor necrosis and reduction in CA19-9 constituted preliminary evidence of anti-tumor activity . She has experienced no treatment related toxicities thus far and remains on therapy of approximately 3 . 5 months duration thus far . The FGFR2 fusion partner observed in this patient , MGEA5 , is an enzyme responsible for the removal of O-GlcNAc from proteins [66] . Interestingly , soft tissue tumors myxoinflammatory fibroblastic sarcoma ( MIFS ) and hemosiderotic fibrolipomatous tumor ( HFLT ) both share a translocation event resulting in rearrangements in TGFBR3 and MGEA5 [67] , [68] . Associated with this translocation event is the upregulation of NPM3 and FGF8 [68] , of which both genes are upregulated in this patient ( fold change: NPM3 = 6 . 17865 , FGF8 = 1 . 79769e+308 ) . In breast cancer , grade III tumors had significantly lower MGEA5 expression than grade I tumors with a trend of decreasing expression observed with increasing tumor grade [66] . In summary , MGEA5 may play an important role in carcinogenesis as an FGFR fusion partner . Patient 6 is a 43 year-old white female who underwent a right salpingo-oophorectomy and endometrial ablation in the context of a ruptured ovarian cyst ( Table 5 ) . Postoperatively she developed dyspnea and was found to have pulmonary nodules as well as a 5 cm left sided liver mass . Pathological evaluation of the liver mass was consistent with a moderately differentiated intrahepatic cholangiocarcinoma ( CK7+ , CK20− , TTF-1− ) in the absence of any known risk factors ( Table 6 ) . She was treated systemically with gemcitabine and cisplatin and had stable disease for approximately 6 months , but was subsequently found to have disease progression . She was treated with FOLFOX for 7 months and again attained stable disease as best response to therapy but eventually experienced disease progression . Upon disease progression , she was enrolled on a clinical study with the multi-kinase inhibitor pazopanib that is FDA-approved for the treatment of advanced renal cancer or sarcoma – and fortuitously has nanomolar activity against FGFR2 ( in vitro IC50 to FGFR2≈350 nM ) [69] . Transcriptome analysis revealed the presence of an FGFR2-TACC3 fusion ( Table 7 , Figure 2 ) . Evaluation of post-pazopanib tissue by immunohistochemistry revealed increased expression of FGFR2 and FGFR3 ( Figure 5 ) Further evaluation of phosphorylation of downstream targets FRS2 Y436 , and ERK ( MAPK ) revealed strong expression of pERK and moderate expression of pFRS2 Y436 ( Figure 6 ) , confirming activation of the receptor . She had been treated with pazopanib 800 mg orally daily for 4 months and demonstrated tumor shrinkage , which in retrospect , was postulated to be secondary to the FGFR2 inhibitory activity of pazopanib ( Figure 8A ) . By RECIST criteria v1 . 1 , the patient had a partial response to therapy as evidenced by a 71% decrease in the sum of diameters . Subsequently , the same patient was treated with a dedicated pan-FGFR inhibitor , ponatinib , ( 45 mg daily orally; in vitro IC50 : FGFR1≈24 nM , FGFR2≈8 nM , FGFR3≈8 nM and FGFR4≈34 nM ) . She again attained minor tumor shrinkage ( stable disease by RECIST criteria v1 . 1 , decrease of 4% in sum of largest diameters ) in multiple lesions after 2 months of therapy , despite undergoing a 50% dose reduction for abdominal pain felt to be related to drug ( Figure 8B ) . She remains on therapy approximately 4 months since the initiation of ponatinib . As such , anti-tumor activity was obtained with two distinct FGFR inhibitors in the same patient . The FGFR2 fusion partner observed in this patient's tumor , TACC3 , is overexpressed in many tumor types with enhanced cell proliferation , migration , and transformation observed in cells overexpressing TACC3 [70] . Furthermore regulation of ERK and PI3K/AKT by TACC3 may contribute in part to epithelial-mesenchymal transition ( EMT ) in cancer [70] , a significant contributor to carcinogenesis . Interestingly , TACC3 has been identified as a fusion partner to FGFR3 in bladder cancer , squamous cell lung cancer , oral cancer , head and neck cancer and glioblastoma multiforme [54] . Patient 3 was a 50 year-old white male who presented with fevers and night sweats ( Table 5 ) . He was found to have a 4 cm tumor in his liver determined to be a poorly differentiated intrahepatic cholangiocarcinoma ( CK7+ , CK20− , TTF1− , CD56− , synatophysin− , Hep-par 1− ) with sclerotic features ( Table 6 ) . No overt risk factors for cholangiocarcinoma were identified . A left hepatectomy was undertaken three months later . In addition to the primary tumor in segment 4 , limited resections of segments 6 and 8 were undertaken to remove two tumor nodules . He was soon noted to have increased hypermetabolism in the left lower cervical , upper mediastinal and abdomino-retroperitoneal lymph nodes related to metastatic disease from his cholangiocarcinoma . He was treated with gemcitabine and cisplatin for 9 months and obtained stable disease as his best response , followed by eventual progression . He received treatment with pegylated hyaluronidase ( PEGPH20 ) in the setting of an investigational study for one month and had no response to therapy . A biopsy of a left supraclavicular lymph node was obtained two months prior to the initiation of PEGPH20 in the context of a clinical study employing whole genome analysis for putative therapeutic target selection . Since our study goal was to identify potential therapeutically relevant events , the novel loss of function mutation in ERRFI1 ( E384X ) detected in Patient 3's metastatic , recurrent/refractory SIC ( Table S1 ) warranted additional examination . Specifically , the allelic fraction of the DNA mutation constituted only 11% of the sequencing reads , is consistent with tissue heterogeneity , and constituted 78% of the sequencing reads within the RNASeq data . Such allele specific expression of the mutated allele from the same tissue specimen suggests nearly complete loss of function of ERRFI1 in this patient's tumor . Notably , the patient's tumor did not harbor any mutations or amplifications in other EGFR signaling members such as EGFR and BRAF . Upon availability of CLIA validated sequencing data ( Table S5 ) , the patient was treated with erlotinib 150 mg orally/daily . After 3 months , RECIST v1 . 1 partial response evidenced by a decrease of 58% in the sum of largest diameters was observed ( Figure 9 ) . Evaluation of pretreatment tumor tissue by immunohistochemistry revealed increased expression of EGFR pathway members ( Figure 10 ) .
Integrated analysis of sporadic intrahepatic cholangiocarcinoma ( SIC ) genomic and transcriptomic data led to the discovery of FGFR2 fusion products in three of six assessed patients ( Table 7 , Figures 4 and 11 ) . Members of the FGFR family ( FGFR1-4 ) have been associated with mutations , amplifications and translocation events with oncogenic potential [71] . FGFR fusions with oncogenic activity have been previously identified in bladder cancer ( FGFR3 ) [72] , lymphoma ( FGFR1 and FGFR3 ) [73] , [74] , acute myeloid leukemia ( FGFR1 ) [75] , multiple myeloma [76] , myeloproliferative neoplasms [77] , and most recently glioblastoma multiforme ( FGFR1 and FGFR3 ) [78] . FGFR2 , FGFR3 and FGFR4 have been found to be overexpressed in IDH1/IDH2 mutant biliary cancers [79] , a context seen within Patient 1 in our study ( Tables S1 and S6 , Figure 5 ) ; although , no fusion events were depicted in these studies or in Patient 1 . Although the gene partner fused to FGFR2 was different for each patient ( MGEA5 , BICC1 and TACC3 ) , the breakpoints in FGFR2 all occurred within the last intron distal to the last coding exon and terminal protein tyrosine kinase domain ( Figure 11 ) . All three fusions were validated at the DNA and/or RNA level ( Table 8 ) . Amongst these fusions , the FGFR2-BICC1 fusion has recently been independently identified in SIC [53] , [54] . For this particular fusion product we observed , and validated , the presence of two fusion isoforms ( FGFR2-BICC1 and BICC1-FGFR2 ) . Interestingly , BICC1 is a negative regulator of Wnt signaling [80] and when comparing expression of tumor and normal tissue we observed differentially expressed Wnt signaling genes , APC ( fold change -4 . 75027 ) , GSK3B ( fold change -3 . 35309 ) , and CTNNB1 ( fold change -1 . 73148 ) , yet when the expression was compared to other cholangiocarincomas , no difference was observed . The FGFR genes encode multiple structural variants through alternative splicing . Notably , RNASeq data revealed that the FGFR2-IIIb isoform was present in all fusions detected in our study and has been shown to have selectivity for epithelial cells as opposed to the FGFR2-IIIc isoform , which is found selectively in mesenchymal cells [81] . Paradoxically , wildtype FGFR2-IIIb has been described as a tumor suppressor in pre-clinical systems of bladder cancer and prostate cancer [82] , [83] . As such , FGFR signaling appears context-dependent and exhibits variability in disparate tumor types . Importantly , one critical study has shown that FGFR2 carboxy-terminal deletion mutants induce ligand-independent transformation and clonogenic growth [84] . This is important because all of the fusion events within our study would lead to loss of the carboxy-terminus of FGFR2 . Furthermore , a very recent study that described FGFR fusions in solid tumors illustrated that FGFR fusion partners in SIC resulted in dimerization domains , and suggested that activation occurred through ligand independent dimerization and oligomerization [54] . It is likely that both loss of the carboxy terminus and the addition of dimerization domains leads to oncogenic FGFR2 activity in these tumors . Comparative pathway analysis of genes carrying mutations/aberrant in copy number identified additional potential therapeutic targets belonging to , or intimately integrated with , the EGFR and FGFR signaling pathways ( Figure 3 , Tables S2 , S3 , S4 ) . Interestingly , most of these pathways also have known roles in mediating epithelial-to-mesenchymal cell transitions , which occur frequently during development as well as during tumorigenesis [60] . Patients 3 and 4 harbored aberrations in several genes acting in cadherin signaling pathways ( Tables S3 , S4 ) , which are important for maintaining cell-cell adhesion [63] . The preliminary anti-tumor activity noted in a patient with FGFR2-MGEA5 ( Patient 4 ) and FGFR2-TACC3 fusion ( Patient 6 ) represent the first reports of application of FGFR inhibitors to the treatment of patients with cholangiocarcinoma harboring these alterations . These results suggest that oncogenic activation of FGFR2 represent a potential therapeutically actionable event . The FGFR tyrosine kinase inhibitors ( TKI ) dovitinib [85] and NVP-BGJ398 [86] are currently in clinical development and the FGFR TKI ponatinib [75] , [87] was recently approved by the FDA for use in treating T315I mutant chronic myelogenous leukemia . FGF7 ( keratinocyte growth factor ) has been previously linked to poor prognosis in patients with biliary tract cancer and a small molecule FGFR kinase inhibitor , Ki23057 , has demonstrated efficacy in preclinical models [88] . It should be recognized that small molecule tyrosine inhibitors are almost universally promiscuous with regards to specificity and typically significant off-target effects are resultant . Off target efficacy resulting from inhibition of angiogenic kinases in addition to FGFR2 inhibition could explain the anti-tumor activity exhibited in patient 6 , as pazopanib has been shown to have nanomolar range potency towards VEGFR1-3 , PDGFRA/B and CKIT as well [89] . Larger trials , preferably of a randomized nature with a control arm , need to be conducted to truly define the role of FGFR inhibitors in the treatment of patients with cholangiocarcinoma , particularly those harboring FGFR2 fusions . While our results provide impetus and enthusiasm towards this end , at this stage they should be considered preliminary in nature . The preliminary anti-tumor activity observed in patient 6 with both pazopanib , and subsequently ponatinib , is particularly intriguing , but also raises important questions . There was an initial response to pazopanib , followed by disease progression . This is a phenomenon observed with the clinical application of most targeted therapeutic approaches . Potential explanations include tumor heterogeneity resulting from clonal selection , transcriptional up-regulation of escape pathways , epigenetic mechanisms and other yet undefined mechanisms of resistance to therapy . The patient did not have additional known alterations in key oncogenic pathways in genes such as BRAF , KRAS , EGFR and PIK3CA , which if present , could provide a putative basis for eventual escape from FGFR pathway inhibition . It is unclear why patient 6 initially responded to pazopanib followed by resistance and subsequently responded to ponatinib , another FGFR inhibitor . Putative explanations include the higher potency of ponatinib observed in vitro to FGFR2 ( IC50≈8 nM for ponatinib vs . 350 nM for pazopanib ) and resistance being defined as >20% increase in sum of largest diameters per RECIST v1 . 1 standard criteria that triggered a discontinuation from pazopanib and recapturing of anti-tumor activity by subsequent inhibition of the FGFR pathway which still maintained therapeutic relevance in that patient at a later time point . ERRFI1 has a role as a negative regulator of EGFR dependent skin morphogenesis [90] , [91] , uterine steroid hormone responsiveness [92] and as a tumor suppressor gene [90] , [93] , [94] . ERRFI1 is an endogenous inhibitor of EGFR , ERRB2 , ERRB3 and ERRB4 through direct interaction with the kinase domains of these proteins [95] , [96] and endocytosis/lysosomal degradation of ERBB receptors [97] . ERRFI1 deletions have been found in glioblastoma multiforme and breast cancer [98]–[100] . Other mechanisms of ERRFI1 loss include methylation , acetylation and loss of function mutations [98] , [101] , [102] . Consistent with a driver role of this mutation , previously germline homozygous disruption of ERRFI1 in mice induces hyperplasia and adenoma formation in the epithelium and development of spontaneous adenocarcinomas of the lung , gallbladder and biliary tract [103] . The tyrosine kinase inhibitor gefitinib has demonstrated anti-tumor activity in mice in spontaneous tumors driven by ERRFI1 germline loss [91] . Our results suggest immediate and actionable implications for SIC patients with tumors harboring ERFFI1 loss of function mutations or FGFR fusions , given the clinical availability of FDA-approved EGFR and FGFR tyrosine kinase inhibitors . Antibodies specific to FGFR2-IIIb have also shown preclinical efficacy and may serve as an additional platform for therapeutic development in this context [104] . Additional studies to characterize the prevalence of these aberrations in both sporadic and liver fluke associated BTC will need to be conducted . Nevertheless , our results suggest that prospective clinical studies designed to treat patient's tumors harboring these novel genomic aberrations utilizing targeted agents on an individualized basis should be pursued more fully through larger clinical studies in order to explore the precise extent of clinical benefit that this tailored approach may have in patients with primary or advanced BTC . Additionally , post-treatment biopsies to assess pathway down-regulation in patients 4 and 6 ( treated with FGFR inhibitors ) and patient 3 ( treated with EGFR inhibitor ) are not available , as the treatment was not conducted in the setting of a protocol that would allow for the collection of additional research biopsies . Incorporation of post-treatment biopsies in carefully designed prospective studies will be critical towards defining the association between the use of FGFR and EGFR inhibitors in appropriately selected patients with relevant genomic aberrations .
Clinical information was assimilated from patient records from the Mayo Clinic . Informed consent was obtained for each patient on two ongoing research protocols approved by the Mayo Clinic Institutional Review Board ( 10-006180 and 10-002879 ) . Clinicopathological features collected included age , gender , stage , histological grade , sites of metastasis , tumor sample assessment for overall cellularity/necrosis and percent tumor cellularity , prior therapies and risk factors ( hepatitis B and C , Caroli's disease , obesity , hepatolithiasis and cholelithiasis , primary sclerosing cholangitis , thorotrast exposure and H . pylori , H . bilis , S . typhi and S . paratyphi infections ) . All patients were known to not have had prior exposure to liver flukes that have been implicated in biliary carcinogenesis ( O . viverrini and C . sinensis ) . Tissue specimens were collected fresh frozen and maintained below −80°C until nucleic acid extraction . A board certified pathologist who is experienced in biospecimen studies , evaluated a portion of each specimen to confirm the presence of tumor , the degree of necrosis and the percent cellularity . Libraries with a 1% phiX spike-in were used to generate clusters on HiSeq Paired End v3 flowcells on the Illumina cBot using Illumina's TruSeq PE Cluster Kit v3 ( catalog# PE-401-3001 ) . Clustered flowcells were sequenced by synthesis on the Illumina HiSeq 2000 using paired-end technology and Illumina's TruSeq SBS Kit . Mutations of potential clinical relevance were confirmed in a Clinical Laboratory Improvement Amendments ( CLIA ) laboratory with Sanger sequencing or quantitative PCR . The immunohistochemistry was performed following the procedures described previously [110] . Briefly , slides were dewaxed , rehydrated and antigen retrieved on-line on the BondMax autostainer ( Leica Microsystems , INC Bannockburn , IL ) . Slides were then subjected to heat-induced epitope retrieval using a proprietary EDTA-based retrieval solution . Endogenous peroxidase was then blocked and slides were incubated with the following antibodies: FGFR2 ( BEK , Santa Cruz , catalog# sc-20735 ) , FGFR3 ( C-15 , Santa Cruz , catalog# sc-123 ) , panAKT ( Cell Signaling Technology , catalog# 4685 , pAKT ( Cell Signaling Technology , catalog# 4060 ) , EGFR ( Cell Signaling Technology , catalog# 4267 , pEGFR ( Cell Signaling Technology , catalog#2234 ) , MAPK/ERK1/2 ( Cell Signaling Technology , catalog# 4695 ) , pMAPK/pERK ( Cell Signaling Technology , catalog# 4376 ) and pFRS2 Y436 ( Abcam , catalog# ab78195 ) . Sections were visualized using the Polymer Refine Detection kit ( Leica ) using diaminobenzidine chromogen as substrate . FISH was performed on formalin-fixed paraffin-embedded ( FFPE ) specimens using standard protocols and dual-color break-apart rearrangement probes specific to the FGFR2 gene ( Abbott Molecular , Inc . Des Plaines , IL ) located at 10q26 . The 5′ FGFR2 signal was labeled with Spectrum Orange ( orange ) and the 3′ FGFR2 signal was labeled with Spectrum Green ( green ) . | Cholangiocarcinoma is a cancer that affects the bile ducts . Unfortunately , many patients diagnosed with cholangiocarcinoma have disease that cannot be treated with surgery or has spread to other parts of the body , thus severely limiting treatment options . New advances in drug treatment have enabled treatment of these cancers with “targeted therapy” that exploits an error in the normal functioning of a tumor cell , compared to other cells in the body , thus allowing only tumor cells to be killed by the drug . We sought to identify changes in the genetic material of cholangiocarcinoma patient tumors in order to identify potential errors in cellular functioning by utilizing cutting edge genetic sequencing technology . We identified three patient tumors possessing an FGFR2 gene that was aberrantly fused to another gene . Two of these patients were able to receive targeted therapy for FGFR2 with resulting tumor shrinkage . A fourth tumor contained an error in a gene that controls a very important cellular mechanism in cancer , termed epidermal growth factor pathway ( EGFR ) . This patient received therapy targeting this mechanism and also demonstrated response to treatment . Thus , we have been able to utilize cutting edge technology with targeted drug treatment to personalize medical treatment for cancer in cholangiocarcinoma patients . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"oncology",
"genome",
"sequencing",
"medicine",
"cancer",
"genetics",
"genetics",
"cancer",
"treatment",
"chemotherapy",
"and",
"drug",
"treatment",
"biology",
"genomics",
"cancers",
"and",
"neoplasms",
"gastrointestinal",
"tumors",
"genomic",
"medicine",
"pharmacogenomics",
"cholangiocarcinoma"
]
| 2014 | Integrated Genomic Characterization Reveals Novel, Therapeutically Relevant Drug Targets in FGFR and EGFR Pathways in Sporadic Intrahepatic Cholangiocarcinoma |
Codon usage bias in prokaryotic genomes is largely a consequence of background substitution patterns in DNA , but highly expressed genes may show a preference towards codons that enable more efficient and/or accurate translation . We introduce a novel approach based on supervised machine learning that detects effects of translational selection on genes , while controlling for local variation in nucleotide substitution patterns represented as sequence composition of intergenic DNA . A cornerstone of our method is a Random Forest classifier that outperformed previous distance measure-based approaches , such as the codon adaptation index , in the task of discerning the ( highly expressed ) ribosomal protein genes by their codon frequencies . Unlike previous reports , we show evidence that translational selection in prokaryotes is practically universal: in 460 of 461 examined microbial genomes , we find that a subset of genes shows a higher codon usage similarity to the ribosomal proteins than would be expected from the local sequence composition . These genes constitute a substantial part of the genome—between 5% and 33% , depending on genome size—while also exhibiting higher experimentally measured mRNA abundances and tending toward codons that match tRNA anticodons by canonical base pairing . Certain gene functional categories are generally enriched with , or depleted of codon-optimized genes , the trends of enrichment/depletion being conserved between Archaea and Bacteria . Prominent exceptions from these trends might indicate genes with alternative physiological roles; we speculate on specific examples related to detoxication of oxygen radicals and ammonia and to possible misannotations of asparaginyl–tRNA synthetases . Since the presence of codon optimizations on genes is a valid proxy for expression levels in fully sequenced genomes , we provide an example of an “adaptome” by highlighting gene functions with expression levels elevated specifically in thermophilic Bacteria and Archaea .
Due to non-random use of synonymous codons , protein coding sequences contain a layer of information on the DNA level that is not reflected at the protein sequence level . The principal determinant of codon usage in prokaryotes are nucleotide substitution patterns [1] , [2] that vary greatly across genomes , as evidenced in the range of genomic G+C content spanned by the sequenced organisms . There is also significant variation in direction and strength of these nucleotide substitution biases along the prokaryotic chromosome [3] with a general tendency toward A+T-enrichment near the replication terminus . Another common intra-genomic trend in nucleotide composition concerns the distinction between the two DNA strands where the leading strand is ‘GC-skewed’ , i . e . enriched in G over C and T over A [4] mostly due to deamination of cytosine in single-stranded DNA exposed during replication . Such biases in mutational processes may result from the nature of chemical changes to the nucleotides , but also from biases in errors of DNA replication and repair , and appear to be an important contribution to the background substitution patterns . In addition to the mutational biases , an adaptive component has also been proposed for specific nucleotide compositions , e . g . [5] and also for dinucleotides [6] . We refer the reader to a review of the organizational features of prokaryotic genomes with respect to local sequence composition and gene distribution [7] . In addition to the nucleotide substitution patterns , a competing influence on silent sites is selection acting to make protein translation more ‘efficient’ ( in this context implying ‘faster’ ) and more accurate; although the term ‘efficiency’ is technically a misnomer [8] , we use it for sake of consistency with previous literature . Traditionally , this effect was linked to abundances of tRNA isoacceptors for a particular codon [9] , in agreement with a model where the speed of translational elongation is limited by availability of charged tRNA molecules [10] . Translational selection is also reflected in biased codon use that guards against missense and nonsense errors in proteins [11] . More recently , other more subtle translation-related determinants of codon usage have been observed , for instance the ‘load minimization’ where codons whose mutated forms cause less structural disruption to proteins are preferred [12] and the selective charging of tRNAs which promotes use of starvation-insensitive codons in amino acid biosynthetic pathways [13] . Some correlations have been observed between codon usage and protein structural features [14] , and a synonymous mutation in a human gene was shown to produce a phenotype via an altered protein structure [15] . Selection for translational efficiency and accuracy would be expected to affect strongly a small set of highly abundant proteins , a typical representative being the ribosomal protein ( RP ) genes . The portion of a genome undergoing some degree of translational optimization may , however , be larger , and choice of genes within this subset was speculated to be related to the environment of a particular organism [16] . A number of prokaryotic genomes have been reported to show no influence of translational selection at all , most notably the slow-growing pathogens Borrelia burgdorferi [17] and Helicobacter pylori [18] or the insect endosymbionts Buchnera [19] , Wigglesworthia [20] and Blochmannia floridanus [21] . However , in Buchnera a correlation was found between measured tRNA abundances and codon composition of highly expressed genes [22] . Three previous multiple-genome analyses detected evidence of translational selection in approx . 25% [23] , 50% [24] or 70% [25] of the prokaryotes analyzed , the authors' conclusions depending heavily on the mathematical apparatus employed . A multitude of statistics have been invented specifically for codon usage analyses and implemented in software [26] , and many of these statistics can be generalized to measures of pairwise distances between codon frequency vectors . A prominent example is the popular ‘codon adaptation index’ [27] that measures the distance to a predefined set of highly expressed genes . Interestingly , several authors that relied on a codon distance measure have found that gene functions close in codon usage to RP genes in E . coli also have RP-like codon usage in some of the organisms which are supposed to lack translational selection , see e . g . the glycolysis genes in H . pylori [28] or respiration and ATP synthase genes in B . floridanus [29] . Correspondence analysis , an unsupervised dimensionality reduction technique followed by visualization or clustering , has often been used in single-genome studies to detect dominant trends in codon usage patterns . This approach has lead to qualitatively different results regarding presence or absence of translational selection depending on how the data was normalized , as demonstrated for B . burgdorferi [30] , and for a larger number of genomes using a related technique of principal component analysis [31] . Our motivation for the present work was to reconcile the inconsistencies in the literature concerning the prevalence of translational selection among and within genomes , and its relationship to microbial ecology and physiology . To this end , we introduce a supervised machine learning-based computational framework that couples a classifier to standard statistical tests , an approach that exhibits an increased accuracy over commonly used unsupervised techniques , and the ability to control for a strong confounding factor – the nucleotide substitution patterns – that shape codon usage , but in a manner not related to protein translation .
In contrast to previous approaches , we introduce a supervised machine learning-based framework for detecting the presence and the extent of translational selection in 461 prokaryotic genomes . Our method is based on the Random Forest ( RF ) classifier [32] which we evaluate in the task of discriminating a group of genes affected by selection for translational efficiency and/or accuracy , using only codon frequencies . The group of ribosomal protein ( RP ) genes is assumed to be highly expressed and therefore a representative subset of genes under such selective pressures; we back this assumption by a survey of RP mRNA abundances in a phylogenetically diverse set of organisms ( Table S4 ) . We demonstrate RF to be a more accurate tool ( Figure 1 ) in comparison to three previous pairwise distance-based approaches [27] , [33] , [34] . Additionally , the RF predictions correlate with experimental measurements of protein concentrations in Escherichia coli slightly better than previous methods do ( Figure 1 , Table S1 ) . We also show the widely used CAI method [27] to be suboptimal for genomes with imbalanced G+C content ( Figure 1 ) , as previously speculated by its author [35] and as evidenced by its inability to predict gene expression in the A+T rich eukaryote Plasmodium falciparum [34] . For each genome , we train two series of RF classifiers to discriminate ribosomal protein genes: first , a series of ‘baseline’ classifiers that have at their disposal the description of regional nucleotide substitution patterns as mono- and di-nucleotide frequencies in non-coding DNA in the genes' vicinity , followed by a second series of classifiers that introduces additional information about codon frequencies of genes ( see Materials and Methods and Figure S1 ) . In 460 of 461 examined genomes , the codon frequencies consistently facilitated classification over the baseline ( Table S3 ) , providing strong evidence that translational selection is , in fact , ubiquitous among prokaryotes . This trend also holds true in genomes which previous large scale studies [23]–[25] have found as lacking translational selection ( Figure 2 ) . The only genome where our method did not detect a translation-related codon bias was Saccharophagus degradans 2–40 , although even this result may change depending on the size of the ‘window’ of ncDNA examined ( Text S1 ) . This genome was previously found to exhibit extensive mosaicism in G+C content [36] , probably due to large amounts of recently horizontally transferred ( HT ) DNA which might not have had sufficient time to ‘ameliorate’ [37] to match the new host's translational apparatus . S . degradans genome serves as an example how strong local variation in background nucleotide frequencies , here caused by HT events , might obscure translation-related codon usage biases even if they did exist by boosting the accuracy of the baseline classifier; an empirical evaluation of influence of HT on our findings is given in Materials and Methods and in Text S2 . Between repetitions of the RF training on the same genome , accuracy of classifiers obtained using codon frequencies was also generally uncorrelated with the accuracy of baseline classifiers obtained without codon frequencies ( Figure 2 ) , further indicating that the codon usage brings into the datasets information independent of that encoded in intergenic DNA . The accuracies of codon-trained RF models also tend to deviate less between runs . On a side note , the unexpectedly strong RF accuracy obtained solely from the description of local nucleotide composition ( Figure 2 , Table S3 ) underscores the need to control for this confounding factor in codon usage analyses . It may prove fruitful to reinvestigate whether the so-called ‘genomic signatures’ – dinucleotide frequencies in DNA – are indeed invariant within bacterial and archaeal genomes , as claimed previously [38] . On the other hand , the strong RF accuracy that we observed with intergenic DNA might in significant part be attributed to the use of ncDNA windows that overlap for neighboring genes ( see Materials and Methods ) . Expectedly , the accuracy of codon-trained RF classifiers generally reflects the intensity of codon biases within a genome ( Figure S3 , Dataset S1 ) , but the accuracy is also bound to be related to the proportion of genes within a genome that are affected by translational selection; see section below . The increase in accuracy for a genome additionally depends on the baseline model derived from local and between-strand variation in background nucleotide composition . Therefore , the magnitude of the increase does not have a straightforward interpretation in itself; rather , if the increase over the baseline has sufficient statistical support , it may be concluded that a translation-related codon usage bias is present , be it strong or weak . The classifiers' predictions on a per-gene level provide estimates of similarity to the ribosomal protein genes . We declare a gene to have optimized codon usage ( OCU ) if this similarity exhibits a statistically significant increase after codon frequencies are introduced to the classifier ( Figure 1 ) . Using a conservative estimate ( see Materials and Methods ) we find that genomes contain on average 13 . 2% of OCU genes ( Figure 3 , Dataset S1 ) , with a minimum of 5 . 4% in the metabolically versatile , free living Pseudomonas flourescens Pf-5 , and a maximum of 33 . 0% in the highly reduced genome of the obligate parasite Aster yellows witches-broom phytoplasma . These estimates of the extent of translational selection within bacterial and archaeal genomes are broadly comparable to the results of a study in eukaryotes [39] that reported purifying selection at synonymous sites in ∼28% of the analyzed mouse-rat orthologs . Given these findings , it does not seem generally safe to assume that silent sites of all protein coding genes evolve neutrally , regardless of the domain of life under scrutiny . The division of genes into the OCU and non-OCU groups does not imply that there is a clear-cut boundary between codon frequencies of the two groups . Rather , we would expect a gradient of codon usages to exist , where the genes labeled as OCU are those above the detection threshold of our method . This concept builds on an approach formulated by Karlin , Mrazek and colleagues [29] , [33] , [40] where a subset of genes in the genome is assigned the “PHX” ( predicted highly expressed ) label by codon usage similarity to a set of RP and other translation-related protein genes . There are , however , three important distinguishing features of OCU assignments: ( a ) they are based on a RF classifier that outperforms the ‘codon bias’ distance measure used for PHX assignments ( Figure 1 ) , ( b ) OCU is separated from non-OCU by a significance call of a statistical test instead of relying on an arbitrary threshold , and ( c ) OCU assignments are made using a control for local nucleotide substitution patterns which are a strong confounder in codon usage analyses . In mammalian genomes , the presence of translationally selected codon usage is still an unresolved issue , see [41] for a recent analysis . The method we have here employed to prokaryotes could potentially be useful for future investigations on mammalian genomes where non-coding DNA is plentiful and local variation in GC content ( isochores ) greatly complicates analysis . We have demonstrated that in almost all examined genomes the RP genes can be discerned by their codon usage , even after local or strand-specific nucleotide composition are controlled for . To verify that this codon bias of the RP-like genes ( = OCU genes ) is indeed due to translational selection , we examine the correlation of OCU/non-OCU assignments to gene expression data ( Text S1 , Appendix A ) from 19 phylogenetically diverse species . We find that OCU genes record microarray signal intensities on average 2 . 4-fold higher than non-OCU genes ( 2 . 2x if RP genes are excluded ) , ranging from 1 . 2x to 3 . 7x; compare this to the 6 . 0x difference between RP – representing the most highly expressed genes – and the average measurement ( Table S5 ) . The ratio of means is significantly greater than unity in all genomes at p<0 . 01 ( permutation test ) . Note that the relationship of the microarray signal intensities to mRNA abundances may be highly non-linear; in a study of gene expression in human tissues , the signal of Affymetrix microarrays was found to be roughly proportional with log-transformed counts of mRNA molecules obtained with Illumina sequencing [42] . In all 19 organisms we examined , the distribution of microarray signal intensities for OCU genes was significantly shifted towards higher values ( p<0 . 01 , Baumgartner-Weiss-Schindler permutation test [43] ) . The trend remains in the four of the 19 genomes where translational selection was previously considered to be ineffective ( Figure 4 , full data in Table S5 ) . These estimates of correlation are likely conservative as gene expression will match codon usage better under certain growth conditions , presumably those that were dominant during the organism's evolutionary past [44] . It is not trivial to surmise these conditions in advance , and a dataset that matches them may not be available . Expression measurements taken under conditions of stress or starvation , for instance in the stationary phase , are expected to correlate less strongly with codon usage , as evidenced previously for Bacillus subtilis and Escherichia coli [34] . We looked for further evidence that OCU assignments were indeed due to translational selection by examining whether the OCU genes show a preference for putatively optimal codons in two-fold degenerate amino acids , where we assumed an optimal codon to match the tRNA anticodon by canonical base pairing ( without wobble ) , as proposed in an early investigation of codon usage in yeast genes [45] . The presence or absence of tRNA genes with specific anticodons given in GtRNAdb database [46] indicates that the optimal codons in prokaryotes are almost always either C/A-ending , or undefined if the genome contains tRNA genes with both anticodons for the two-fold degenerate amino acid . We found that OCU genes prefer the putatively optimal codon 6 . 2x more frequently than the suboptimal one in Bacteria ( p<10−30 , sign test ) and 3 . 1x more frequently in Archaea ( p = 10−12 , sign test; Figure 5; Table S6 ) . All amino acids contribute approximately equally to this effect ( Figure 5 ) , with the exception of Cys which is rare and therefore hard to show a preference for or against , and Lys , for which an optimal codon cannot be defined in a majority of genomes . In spite of the dominant trend of OCU preference for optimal codons defined through tRNA content , some examples of amino acid-genome combinations where this regularity is reversed do exist; full data in Dataset S1 . We speculate that these exceptions might stem from chemical modifications of nucleosides in the tRNA . A rich structural diversity of the modifications exists in Bacteria and Archaea [47] , many affecting the anticodon or its vicinity , thereby modulating the codon-anticodon interaction; reviewed in [48] , [49] . The modifications prevent wobble matching to the wrong amino acid or enhance wobble matching to the correct amino acid , which may , in turn , render inapplicable the definition of cognate codons as the optimal codons . The potential for such modifications to elevate translation speed of ‘suboptimal’ codons over ‘optimal’ ones has been demonstrated experimentally on Drosophila tRNAHis [50] and on E . coli tRNAGlu [51] . Currently , the tRNA nucleoside modifications are fully known only for few organisms [52]; it will be interesting to see how the future experimental data on the modifications will align with the OCU-preferred codons . Equally relevant to this issue is a recent paper discussing the choice of optimal codons in genomes [53] , stating that optimal codons are dictated mainly by the direction of nucleotide substitution patterns evident in the overall genomic G+C content . Moreover , the authors have also found that in a large majority of bacterial genomes the gene set with the most highly biased codon usage is enriched with RPs and translation elongations factors , and that this codon usage cannot be reproduced from composition of intergenic DNA [53] . This finding is consistent with the notion of translationally selected codon usage as a prevalent phenomenon among prokaryotic genomes . The proportion of OCU genes correlates inversely to genome size ( Figure 3A , Spearman's ρ = −0 . 71 , Dataset S1 ) , and we note a relationship of % OCU to lifestyle of bacteria , free-living vs . host-associated . These effects are largely a consequence of the changes in proportions of gene functional categories in genomes with regard to size ( Figure 3B ) [54] , [55] and lifestyle , as the % OCU is readily predictable from frequencies of selected Gene Ontology ( GO ) categories ( Dataset S1 ) , even after controlling for genome size ( Figure 3C ) . Note that our procedure to estimate of the extent of translational selection within genomes – expressed as % OCU genes – was not designed to measure the strength of this selection within a genome . Ideally , the genes' OCU assignments should be independent of the strength of the selection , previously recognized to differ greatly between genomes depending on the growth rate of the organism , or the composition of the cellular tRNA pool [56] . Therefore , the high % OCU in small genomes is not an indication that translational selection is stronger or weaker in these organisms , rather it is largely consequential to the proportion of underlying gene functional categories . If the extent of codon optimization within a genome is dictated mainly by its content of gene functions , one would expect the individual GO categories to have general preferences towards enrichment or depletion from OCU genes that are conserved across organisms , as we have verified ( Figure 6 , Table S7 ) . The most prominent trend in our results is consistent with previous research [28] , [33] , with translational selection readily acting on genes involved in protein production and in energy metabolism . Genes that are rarely OCU are involved in regulation , DNA replication and repair , sensing of stimuli , and most kinds of transport , except the electron transport chain and ATP synthesis-coupled H+ transport . Very similar trends are observed when focusing only on a subset of genomes previously cited as lacking translation-related codon usage biases ( Table S8 ) , corroborating the idea that translational selection is operative even in these genomes . We compared Gene Ontology categories enriched with OCU genes between Bacteria and in Archaea ( Figure 6 , box ) , revealing a general agreement between the two domains of life . Two prior studies highlight an unusual reduction in frequency of codon-optimized ribosomal proteins as specific to Archaea [40] and also Archaea-specific optimizations in a DNA replication and repair protein PCNA [57] while our results do not support these observations . We suspect them to be artifacts of a methodology that does not control for the local variation in background nucleotide composition , coupled with the fact that in Bacteria – but not in Archaea – ribosomal protein genes are often collocated on the chromosome as they tend to share operon membership . Several eukaryotes have histone proteins which are highly biased towards the use of optimal codons [58] . Here , we report that Bacteria have nucleoid-associated proteins ( Figure 6 , callout ) that are frequently OCU , identifying another instance of equivalent gene functions being translationally optimized in different domains of life . Furthermore , the archaeal chromatin protein AlbA is commonly OCU ( 40 out of 51 occurences , 4 . 6x enrichment , p = 10−21 by Fisher's exact test ) . These findings are supported by a quantitative proteomics experiment [59] that places the nucleoid-associated proteins Fis , H-NS and HU among the top 10 most abundant non-ribosomal proteins in the E . coli cytosol . Aminoacyl tRNA synthetases ( aa-tRS ) are infrequently OCU ( enrichment = 0 . 24x ) , consistent with their mRNA levels close to the genomic average ( Table S4 ) . A prominent exception is the Asn-tRNA charging enzyme ( Table S9 , enrichment = 1 . 05x ) , which might signal erroneous homology-based transfers of functional annotation involving some instances of this protein and the evolutionarily and structurally related [60] amino acid biosynthetic enzyme , asparagine synthetase A ( AsnA , enrichment = 1 . 44x ) . A different NH4+-assimilating enzyme , glutamine synthetase , is also enriched with OCU genes ( GlnA , enrichment = 2 . 12x ) , in contrast to a general avoidance of codon optimization in amino acid biosynthesis genes ( enrichment = 0 . 50x ) . This leads us to speculate about an additional physiological role for GlnA and/or AsnA in Bacteria that would involve detoxification of ammonia by the energetically costly incorporation into amino acids , as has recently been demonstrated to occur in the yeast S . cerevisiae [61] . Aerobic respiration normally produces oxygen radical species which are then detoxified , or their damaging affects averted by scavenging of free iron , by an array of proteins that we found to be frequently OCU across Bacteria ( Table S10 ) . The abundance of proteins defending from oxidative stress therefore seems not to be specific to the radioresistant organism Deinococcus radiodurans , as claimed previously [33] . OCU genes are , however , conspicuously rare in the catalase genes in Bacteria ( Table S10 ) , possibly due to their activity being necessary only at supraphysiological levels of H2O2 , as previous experimental work indicates is actually the case in E . coli [62] . We have shown that a number of gene functional categories exhibit strong preferences toward or against translational optimization across all Bacteria and Archaea . If organisms defined by a specific lifestyle show a tendency contrary to the general trend , and this tendency is constrained to a gene functional category , this correlation could be biologically meaningful . An example illustrative of this principle was previously brought forward in work by Karlin [40] and Carbone [28] , where the glycolysis genes were claimed to be more frequently codon-optimized in anaerobic bacteria . Indeed , we detected this association ( GO∶6096 , enrichment = 2 . 1x , p = 10−33 ) alongside with a – not unexpected – increase in OCU frequency of carbohydrate transporters ( GO∶8643 , enrichment = 1 . 8x , p = 10−14 ) . We also found an increase in OCU frequency of ferritin in aerobes ( COG∶1145 , enrichment = 3 . 0x , p = 10−17 ) , consistent with ferritin's role in protection against oxidative stress mediated by soluble ferrous ions . In specific Archaea and Bacteria , the ability to thrive at high temperatures was expected to leave a distinct ‘signature’ in genome composition , implying that a gene complement responsible for this phenotype could be delineated . This is , however , generally not the case [54] . Perhaps the genome-encoded determinants of thermophily are discernible on a more fine-grained level , encompassing adaptation through changes in gene regulation and/or recruitment of existing genes for alternative physiological roles; let us name this set of alterations evolved in response to specific environmental challenges an ‘adaptome’ of an organism with respect to an environment . The hallmark feature of translational selection is that it affects highly expressed genes most strongly . The detection of codon optimizations on individual genes can thus be used as a proxy for the genes' expression level , offering an insight into adaptive changes in an organism's physiology; we would not expect the codon optimizations , as we measure them , to reflect other kinds of functional differences between proteins . Based on the correlations observed independently in Archaea and in Bacteria , we conjecture about two metabolic adjustments that would aid in protection of proteins and DNA against thermal denaturation . Phosphorylation is normally used to regulate protein activity , and regulatory proteins are typically not highly abundant in cells . Consequently , they were rarely labeled as OCU ( Figure 6 ) . However , genomes of thermophilic microbes tend to contain comparatively more OCU genes within the ‘protein phosphorylation’ functional category ( Table 1 ) than the genomes of mesophiles . This difference would be explained if the addition of phosphate groups was a very commonly occurring process that affected a considerable fraction of total cellular protein , for instance if the phosphates served a structural role in many different proteins . Previous comparative analyses of structures of thermophile proteins versus their mesophile counterparts indicated that a typical characteristic of thermophile proteins is an increase of charged residues on the protein surface [63] , [64] . Attaching charged phosphate groups ( Table 1 ) to existing amino acids would lead to a similar effect , perhaps even to a stronger degree as the phosphate carries a higher charge than the side chains of charged amino acids . On the other hand , the activity level of such phosphorylation might be controlled through acetylation enzymes ( Table 1 ) that would compete for the same substrates ( amino acid side chains ) . A putative DNA thermoprotective mechanism could be inferred from an unexpectedly high frequency of codon optimization within thermophile genes annotated as response regulators for two-component systems and other transcriptional regulators . Again , we would generally not expect high expression levels from regulatory proteins , unless they were to perform a different role in the cell , either alone or in addition to their original function . We speculate that a subset of these proteins with DNA binding domains ( candidates in Table 1 ) might play a role in formation of a chromatin-like structure that would act to preserve DNA geometric properties , protect it from chemical damage or aid in repair under high temperatures . To our knowledge , there is currently no experimental data that would directly support this hypothesis; however , on the other end of the temperature spectrum , stabilization of DNA and RNA secondary structures occurs and is known to be counteracted by overproduction of nucleic acid binding proteins and RNA helicases when mesophiles are brought into cold conditions [65] . Our results indeed show that psychrophilic Bacteria have increased translational optimization of genes with ATP-dependent helicase activity ( GO∶8026 , enrichment = 3 . 1x , p = 10−4; Dataset S2 ) . We have performed numerous other statistical comparisons involving 35 distinct microbial lifestyles or phenotypes and information on genes' codon optimizations within functional categories . An exhaustive listing of significant results is available in Dataset S2 and on the authors' website at http://www . adaptome . org; here , care should be taken in interpretation as the lifestyles/phenotypes are intercorrelated and also not independent from phylogenetic subdivisions . We hope this data will stimulate further research and help direct experimental work to elucidate environmental adaptations of microbes . Moreover , since the ‘adaptomes’ are purely sequence-derived , an equivalent – or improved – computational methodology can be applied to genomes as soon as they are sequenced , cutting time and effort required to understand the physiology of novel organisms .
We have downloaded 621 fully sequenced prokaryotic genomes from the NCBI Entrez Genome FTP site [66] and removed multiple strains of a single species to retain the strain best covered by Gene Ontology annotations , leaving 461 genomes . Information about lifestyles of organisms was assembled from the JCVI Genome Properties [67] , and the NCBI Entrez Microbial Genome Properties [68] , and curated . The original data used in all computations is freely available via the authors' web site at http://www . adaptome . org/ . To construct datasets – one per genome – we declare all genes coding for ribosomal proteins ( RP ) to be the ‘positive class’ , including the rare cases ( approx . 0 . 6 occurences per genome ) of multicopy RP genes . All other protein-coding genes are the ‘negative class’ . Genes shorter than 80 codons were excluded from computation . A gene is represented by a series of codon frequencies for all degenerate codon families ( excluding the stop codons ) where the frequencies of codons for a single amino acid are normalized to add up to one . Codon frequencies of amino acids absent from a protein are coded by a ‘missing value’ symbol . Ten-fold crossvalidation was run to determine performance of the Random Forest ( RF ) classifier in discriminating the RP genes by their codon frequencies; the area-under-ROC-curve ( AUC ) score [69] was recorded . The AUC is a measure independent of class sizes that ranges from 0 . 5 for a random classification model to 1 . 0 for a perfect model . The RF algorithm [32] produces an ensemble of decision tree classifiers , where each decision tree is constructed by recursively partitioning the data by attribute value tests ( forming ‘nodes’ ) so as to reduce the entropy of the class label in the resulting partitions ( ‘branches’ ) . In RF , trees are constructed on bootstrap samples of the entire dataset , and choice of attributes at each node is restricted to introduce variability . The final predictions of a RF model are obtained by averaging over individual trees ( ‘voting’ ) . Regarding the specific implementation of the RF algorithm , we used FastRandomForest [70] . RF was compared to three pairwise distance measures for vectors of codon frequencies: ( i ) the “codon bias between gene groups” ( CB ) is essentially a weighted Manhattan distance employed by Karlin and colleagues [33] for finding ‘predicted highly expressed’ genes in microbial genomes; ( ii ) the “codon adaptation index” ( CAI ) is an established surrogate for gene expression under optimal growth conditions of Escherichia coli and Saccharomyces cerevisiae [27]; and ( iii ) the “measure independent of length and composition” ( MILC ) [34] is a corrected χ2-type statistic devised to address methodological deficiencies in other approaches such as CB . For a thorough description and formulae for calculation of these measures , see Text S1 . We have incorporated the three distance measures into a ‘nearest class centroid’-type classifier , analogous to uses of CB and CAI in the literature , and compared AUC scores of the RF classifier to the nearest centroid classifiers ( Figure 1 ) . As another verification of the RF classifier , we have compiled protein abundance data in the E . coli cytoplasm from two quantitative proteomics experiments , Ishihama et al . [59] and Lu et al . [71] . After retaining data for 369 proteins that occur in both studies , we computed Spearman's rank correlation of the methods' output ( probability of belonging to positive class , in crossvalidation ) , and protein abundances ( Figure 1 ) ; also the correlations with the full experimental data are given in Table S1 . We encode the information about nucleotide substitution patterns underlying the sequence of each gene by computing mononucleotide and dinucleotide frequencies in the non-coding regions of DNA neighboring the translated part of the gene . Genes for functional RNA molecules such as tRNA and rRNA are also treated as coding DNA and thus do not contribute toward composition of non-coding ( intergenic ) DNA . The size of the neighborhood window was set to either 5 , 10 or 20 kilobases upstream from the gene's start codon , and 5 , 10 or 20 kilobases downstream from the stop codon . The window size of 10 kb upstream +10 kb downstream guarantees that in 99% of the genomes ( 457 out of 461 ) , 99% of the genes have at least 142 non-coding nucleotides available for estimation of nucleotide substitution patterns ( Table S2 ) . To detect if translational selection acts on a genome , the RF classifier is first trained to distinguish RP genes ( ‘positive class’ ) based solely on the mono- and di-nucleotide frequencies of genes' neighboring non-coding DNA within a given window size . Fifty runs of four-fold crossvalidation are used to estimate the accuracy of the classifier , and the AUC score for each of the 50 runs is recorded . The crossvalidation is stratified , meaning that genes are sampled so that the proportions of the RP genes are conserved in the training and the testing parts of each crossvalidation split . The 50 runs of crossvalidation are then repeated for a second time , however now the codon frequencies are also included in the dataset for the RF classifier training , in addition to description of the intergenic regions . The sign test [72] is used to compare AUC scores obtained without codon frequencies to AUC scores obtained with codon frequencies , for each genome ( Figure S1 ) . A summary of results for window size 10k is presented in Table S3 . Note that the described approach to determine mono/di-nucleotide frequencies implies that for neighboring genes the corresponding ncDNA windows will overlap , meaning the description of background nucleotide composition will not be independent between the neighbors . Consequently , the estimate of crossvalidation accuracy ( AUC ) will be somewhat optimistic for the baseline , ncDNA-only models . This should , however , not be problematic as our conclusions regarding presence of translational selection can be biased only to the conservative side due to this issue – having ncDNA models of higher AUC means it can be only more difficult to surpass the models' AUC using the ncDNA+codon models . During the procedure described above which involves two rounds of RF classifier training – without and with codon frequencies – the per-gene probabilities of belonging to positive class are recorded for each of the 50 runs of crossvalidation , and compared between the two rounds of crossvalidation ( Figure S1 ) . A sign test [72] is used to determine if an increase in probability occurs more frequently than expected by chance; if it does , the gene is labeled as having optimized codon usage ( OCU ) . At this point , we combine the OCU assignments obtained with the three values of the window size parameter ( 5 , 10 and 20 kilobases ) into a consensus set by determining the median p-value of sign tests of the three window sizes for each gene . The agreement between window sizes is analyzed in Text S2 . The p-values of OCU assignments for each gene are available as Dataset S1 and from the authors' website at http://www . adaptome . org . Additionally , the full Java source code that performs all calculations described in the Materials and Methods section will be made freely available from the same website in the near future , or on request from the authors . We have set the default p-value threshold to 10−15 , corresponding to exactly 50 out of 50 sign test ‘wins’ for the dataset with codon frequencies . This sign test p-value should be regarded as somewhat optimistic because the repeated runs of crossvalidation are not independent , being based on repeated sampling from the same set of genes . To obtain a conservative estimate , we employed a corrected paired t-test [73] intended for comparison of classification algorithms using repeated runs of crossvalidation . Note that we here compare RF models derived from a specific dataset , and not the different variants of the underlying RF algorithm itself , and therefore we would expect the p-value obtained with this test to be pessimistic for our experimental setup [73] . After this corrected t-test , the median p-value for OCU genes in 10 representative genomes ( Text S2 ) was 6·10−6 , while for 95% of the OCU genes , p<2·10−3 . When testing for difference of AUC scores with/without codons , the median p-value for all 461 genomes obtained by the corrected paired t-test was p = 10−13; for 95% of the genomes , p<4·10−5 ( compare to median p = 10-15 by sign test ) . We test the robustness of the genes' OCU assignments by performing several computational experiments . First experiment is a simulation to see if a methodological bias exists where changes in the positive-to-negative class size ratio would affect frequency of OCU labels; we demonstrate the extent of such changes to be minor ( Text S2 ) and insufficient to explain the anticorrelation between genome size and % OCU described in Results ( Figure 3A ) . The second experiment verifies if an outlier in the positive class – a RP gene with atypical codon usage – would affect OCU assignments . Our RF-classifier based methodology for OCU detection was remarkably robust ( Text S2 ) to such errors in annotation that might stem from e . g . pseudogenes or from the uncommon occurrence of RP gene horizontal transfer [74] . The third issue concerns non-coding DNA , which is generally not abundant in prokaryotes and therefore non-negligible parts of it may be occupied by regulatory elements subject to selection . This should have little impact on our results , as our representation of the ncDNA is not complex enough to permit distinction between ncDNA containing specific regulatory motifs , unless the motifs were to differ strongly in their overall composition and occupy the major part of the ncDNA . Even if this were the case , our procedure would err on the conservative side , leading to an overestimation of the influence of nucleotide substitution patterns and consequentially a lower % OCU , and only if there was a correlation between the composition of regulatory regions and expression levels . We have empirically verified to what extent selection on ncDNA affects our results by re-running the computational framework ( Figure S1 ) on a subset of genomes while excluding the ncDNA regions 20 bp upstream of the translation start codon; this region was shown to be under much stronger selection than the rest of the prokaryotic ncDNA [75] as it contains translation-related sequence elements . The OCU assignments were sufficiently robust to this perturbation ( Text S2 ) as the resultant changes are commensurate to the variability of the method itself . A fourth experiment concerns the influence of putatively horizontally transferred ( HT ) regions on our results . Our method will treat a HT region in the same as it treats the regions that differ due to ( endogenous ) local or strand-specific variation in background nucleotide composition , i . e . regardless of the biological mechanism that causes a deviation in the background nucleotide frequencies , we aim to detect shifts away from this background and toward RP-like codon usage . We empirically evaluated the influence of HT on our results by taking 10 representative genomes and masking the segments marked as HT in the IslandViewer database [76] by any of the three underlying HT-detection algorithms to have highest possible coverage . Then we repeated the OCU finding procedure and compared assignments to the original ones ( Text S2 ) . Again , the changes induced by this filtering were generally commensurate to the between-run variability of the method , save for a slight increase in %OCU in several genomes , meaning the original estimate of %OCU may be conservative with regard to horizontal gene transfer . We use a series of Fisher's exact tests for association of two categorical variables [77] to describe distribution of OCU genes along Gene Ontology [78] categories or COG orthologous groups [79] in all analyzed genomes , and in genomes grouped by environmental and phenotypic contexts ( lifestyles ) . We perform two kinds of tests: ‘Test A’ operates across all genes in all genomes , and compares distribution of optimized genes within a GO category/COG group to the distribution of optimized genes outside the GO/COG , and is iterated over all GOs/COGs; ‘Test B’ operates within a single GO category/COG group , and compares organisms with a specific lifestyle to all organisms which are known not to possess the lifestyle . ‘Test B’ is iterated over all possible combinations of GO/COG and lifestyle; we discuss only the thermophilic lifestyle in the manuscript , while data on other lifestyles is given in Dataset S2 . We set the threshold p-value to 10−3 , yielding a 1 . 6% false discovery rate among the GO categories/COGs found to be enriched or depleted in OCU genes in Archaea , and a false discovery rate of 1 . 7% in Bacteria . All tests passing these criteria were additionally screened to retain only tests with a sufficient magnitude of enrichment or depletion of optimized genes ( >1 . 50x or <0 . 67x ) between GO categories/COGs ( test A ) or organism groups ( test B ) . All test results passing the thresholds for gene group size , statistical significance and magnitude of enrichment/depletion are available in Dataset S2 and from the authors' Web site http://www . adaptome . org/ . Filtered subsets of the GO categories with reduced redundancy are also available as a part of Dataset S2; they were prepared using the REViGO tool available at http://revigo . irb . hr/ . A more detailed description of the employed computational methods and procedures is provided in Text S1 and summarized in a flowchart diagram in Figure S1 . | Synonymous codons are not equally common in genomes . The main causes of unequal codon usage are varying nucleotide substitution patterns , as manifested in the wide range of genomic nucleotide compositions . However , since the first E . coli and yeast genes were sequenced , it became evident that there was also a bias towards codons that can be translated to protein faster and more accurately . This bias was stronger in highly expressed genes , and its driving force was termed translational selection . Researchers sought for effects of translational selection in microbial genomes as they became available , employing a flurry of mathematical approaches which sometimes led to contradictory conclusions . We introduce a sensitive and accurate machine learning-based methodology and find that highly expressed genes have a recognizable codon usage pattern in almost every bacterial and archaeal genome analyzed , even after accounting for large differences in background nucleotide composition . We also show that the gene functional category has a great bearing on whether that gene is subject to translational selection . Since presence of codon optimizations can be used as a purely sequence-derived proxy for expression levels , we can delineate “adaptomes” by relating predicted gene activity to organisms' phenotypes , which we demonstrate on genomes of temperature-resistant Bacteria and Archaea . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"biology/microbial",
"physiology",
"and",
"metabolism",
"biochemistry/molecular",
"evolution",
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics",
"computer",
"science/applications",
"genetics",
"and",
"genomics/comparative",
"genomics",
"genetics",
"and",
"genomics/functional",
"genomics",
"genetics",
"and",
"genomics/gene",
"expression",
"ecology/environmental",
"microbiology",
"computational",
"biology/genomics",
"biochemistry/transcription",
"and",
"translation",
"computational",
"biology/systems",
"biology"
]
| 2010 | Translational Selection Is Ubiquitous in Prokaryotes |
In metazoan integrin signaling is an important process of mediating extracellular and intracellular communication processes . This can be achieved by cooperation of integrins with growth factor receptors ( GFRs ) . Schistosoma mansoni is a helminth parasite inducing schistosomiasis , an infectious disease of worldwide significance for humans and animals . First studies on schistosome integrins revealed their role in reproductive processes , being involved in spermatogenesis and oogenesis . With respect to the roles of eggs for maintaining the parasite´s life cycle and for inducing the pathology of schistosomiasis , elucidating reproductive processes is of high importance . Here we studied the interaction of the integrin receptor Smβ-Int1 with the venus kinase receptor SmVKR1 in S . mansoni . To this end we cloned and characterized SmILK , SmPINCH , and SmNck2 , three putative bridging molecules for their role in mediating Smβ-Int1/SmVKR1 cooperation . Phylogenetic analyses showed that these molecules form clusters that are specific for parasitic platyhelminths as it was shown for integrins before . Transcripts of all genes colocalized in the ovary . In Xenopus oocytes germinal vesicle breakdown ( GVBD ) was only induced if all members were simultaneously expressed . Coimmunoprecipitation results suggest that a Smβ-Int1-SmILK-SmPINCH-SmNck2-SmVKR1 complex can be formed leading to the phosphorylation and activation of SmVKR1 . These results indicate that SmVKR1 can be activated in a ligand-independent manner by receptor-complex interaction . RNAi and inhibitor studies to knock-down SmILK as a representative complex member concurrently revealed effects on the extracellular matrix surrounding the ovary and oocyte localization within the ovary , oocyte survival , and egg production . By TUNEL assays , confocal laser scanning microscopy ( CLSM ) , Caspase-3 assay , and transcript profiling of the pro-apoptotic BCL-2 family members BAK/BAX we obtained first evidence for roles of this signaling complex in mediating cell death in immature and primary oocytes . These results suggest that the Smβ-Int1/SmVKR1 signaling complex is important for differentiation and survival in oocytes of paired schistosomes .
Communication of cells with their environment is an essential requirement to regulate fundamental biological processes such as cell growth and differentiation . Different types of membrane-linked receptors mediate these communication processes , sometimes in a solitary , single receptor-mediated way , sometimes in a cooperative , multiple receptors-mediated way . The latter leads to the integration of different signaling cascades to execute one or more complex operations [1–4] . Schistosomes are parasitic platyhelminths causing schistosomiasis , one of the most threating infectious diseases worldwide after malaria [5–7] . As the only members of the trematodes , schistosomes have evolved separate sexes . The pathology of the disease is caused by eggs which are produced by paired schistosome females in the final host . Egg production is a complex process that involves not only the participation of different cell types , oocytes and vitellocytes . It also comprises the participation of different organs , ovary and vitellarium , whose development in the female depends on a close and permanent pairing contact with the male [8–11] . Although this nearly unique way of regulating sexual development in the animal kingdom is long known [12] and fundamental for the reproductive biology of schistosomes as well as for the pathogenic consequences of schistosomiasis , understanding the underlying molecular principles is still in its infancy . A number of signaling cascades have been uncovered that are involved in the control of gonad differentiation in paired schistosome females [13 , 14] . In S . mansoni , a kinase complex of three different cellular tyrosine kinases ( CTKs ) was postulated , whose members were able to interact with different receptors such as β integrin ( Smβ-Int1 ) and venus kinase receptors ( SmVKRs ) [15–18 ) . The latter represent an unusual type of receptor tyrosine kinases ( RTKs ) consisting of an intracellular tyrosine kinase ( TK ) domain with homology to that of insulin receptors ( IR ) and an extracellular venus-flytrap ( VFT ) module , whose structure is similar to the ligand binding domain of G protein-coupled receptors ( GPCRs ) of the C class [19 , 20] . RNAi-mediated knockdown of Smβ-Int1 and SmVKRs exhibited their roles in oogenesis and egg formation of S . mansoni females [17 , 21] . As potential ligands , L-Arginine ( L-Arg ) and calcium ions were discovered , which activated SmVKR1 and SmVKR2 , respectively , when they were expressed in Xenopus oocytes [22] . Studies in Aedes aegypti have substantiated roles of VKRs for reproduction . AaeVKR expression was found in the ovaries of blood-fed adult females and its activation by the neuroparsin , ovary ecdysteroidogenic hormone , was demonstrated [23] . Since neuroparsins are neuropeptides specific for arthropods [24] it still remains elusive whether and which other molecules except ions and amino acids may be able to activate schistosome VKRs [25] . Physical associations were documented between integrins and GFRs [26] . The latter include RTKs , whose activities can be likewise influenced by integrins [27 , 28] . As shown in skin fibroblasts , interactions with integrins support the activation of the GFRs even in the absence of a ligand [29] . Among other functions the αvβ3 integrin was found to directly associate with the insulin-like IGF1 receptor in vascular cells [30] . Such integrin-GFR interactions are mediated by bridging molecules such as ILK ( integrin-linked kinase ) , PINCH ( particularly interesting new cysteine-histidine-rich protein ) and Nck2 ( non-catalytic region of tyrosine kinase adaptor protein ) . They are central parts of an integrin-actin hub mediating many protein interactions that regulate processes such as pericellular matrix deposition , cell morphology , motility and apoptosis [31–33] . Aims of our study were to investigate whether Smβ-Int1 and SmVKR1 , which colocalize in the ovary of S . mansoni females and whose RNAi-mediated knock-downs led to similar phenotypes [17 , 21] , may interact to govern differentiation processes in this organ . Our findings provide first evidence for this cooperation and for a Smβ-Int1-induced activation of SmVKR1 , which is independent from an extracellular VKR ligand . Furthermore , our data suggest that Smβ-Int1/SmVKR1 cooperatively control the differentiation status of oocytes by regulating cell death-associated processes .
In eukaryotic systems integrin-GFR cooperation can be accomplished by ILK , PINCH , and Nck2 . As cytoplasmic molecules they bind to the intracellular parts of integrin ( ILK ) or GFR ( Nck2 ) , or simultaneously to both receptors with PINCH as bridging molecule connecting ILK and Nck2 [32 , 33] . To investigate the possibility of such an interaction in S . mansoni , we first searched for orthologs in the schistosome database [34 , 35] . Based on comparisons to orthologs from human , potential candidate genes were identified and analyzed in silico . Deletion clones , including those potentially originating from alternative splicing events , were excluded from further analyses . Finally , full-length cDNAs of the longest variants of SmILK ( Smp_079760 ) , SmPINCH ( Smp_020540 . 2 ) , and SmNck2 ( Smp_014850 ) were amplified by RT-PCR , cloned , and sequenced . Detailed sequence analyses showed the cloned cDNAs of SmILK , SmPINCH , and SmNck2 isolated from the Liberian strain of S . mansoni [36] were 100% identical to those of the Puerto Rican strain used for genome sequencing [34 , 35 , 37] . BLAST analyses showed 97% and 89% identity at the cDNA level to ILK orthologs of S . haematobium ( XM_012945977 . 1 ) and S . japonicum ( AY810458 . 1 ) , respectively . Furthermore , SmILK exhibited all typical domains for this class of enzymes such as three N-terminal ankyrin repeat domains as well as one C-terminal kinase-like domain ( S1A Fig ) . The latter is considered as a catalytically inactive domain , which makes ILK a potential pseudokinase without catalytic but with structural importance [38 , 39] . As zinc-finger adaptor protein , PINCH contains five Lim ( similar to Lin11 , Isl-1 and Mec-3 proteins ) domains including eight zinc-binding residues [40] . SmPINCH follows this characteristic structural organization ( S2A Fig ) . At the cDNA level SmPINCH showed 91% and 81% identity to PINCH orthologs of S . haematobium ( XP_012797616 . 1 ) and S . japonicum ( AAX26687 . 2 ) , respectively . Nck2 , finally , represents another adaptor protein consisting of three SH3-domains and one C-terminal SH2-domain . The latter is important for binding to GFRs whereas one or more of the SH3-domains can support GFR binding or mediate interactions to downstream partners such as PINCH [41] . The occurrence of all these domains at comparable positions ( S3A Fig ) indicated that SmNck2 is an ortholog of Nck2 proteins . BLAST analyses showed 94% and 85% identity of SmNck2 at the cDNA level to Nck2 orthologs of S . haematobium ( XM_012936558 . 1 ) / S . japonicum ( AY809191 . 1 ) , respectively . Phylogenetic analyses of the three molecules with orthologs of vertebrates and invertebrates demonstrated that the schistosome ILK , and Nck2 formed separate clusters together with other parasitic platyhelminths , and schistosome Nck2 was part of a trematode cluster separate from the cestodes and other invertebrates ( S1B–S3B Figs ) . This observation coincides with previous findings made for the schistosome α and β integrins , which according to phylogenetic analyses constitute parasite-specific clades separate from free-living flatworms and further metazoan integrins [17] . In situ-hybridization localized the transcripts of SmILK , SmPINCH and SmNck2 in the ovary and the vitellarium of the female as well as in the testis of the male ( Fig 1 ) . Ovary and testis transcription were independently confirmed by gonad RNA-specific RT-PCRs [42] showing amplification products of the expected sizes ( S4 Fig ) . In each case the in situ-hybridization signals appeared to be stronger in the large part of the bulb-like ovary which contains mature primary oocytes . Furthermore , SmILK and SmPINCH transcripts were found in the parenchyma of both genders and in the subtegumental area , within the gastrodermis of males , and around the ootype , although not as dominant as in the gonads . Sense transcripts of all three genes as controls showed varying degrees of week signals ( very low in case of SmNck2 ) . This indicates antisense regulation , a finding made for different schistosome genes before including integrins and further molecules involved in reproduction [17 , 43 , 44] . To elucidate the roles of SmILK , SmPINCH , and SmNck2 in complex formation with Smβ-Int1 and SmVKR1 , we started a series of biochemical experiments in Xenopus oocytes . Previous studies had demonstrated the efficiency of expression of schistosome genes in this system and , furthermore , the possibility to study kinase activities by their capacities to induce resumption of meiosis and germinal vesicle breakdown ( GVBD ) [15 , 45] . Activation by L-Arg of the SmVKR1 kinase led to GVBD , which failed when a dead-kinase mutant of SmVKR1 was used [22] . No GVBD was observed in Xenopus oocytes when a wildtype form of SmILK was expressed ( Table 1 ) . This is in agreement with the present view that SmILK may represent a pseudokinase ortholog of eukaryote ILKs lacking catalytic activity [38 , 39] . Also PINCH and Smβ-Int1 failed to induce oocyte maturation . According to previously published data [22] , the wildtype form of SmVKR1 induced 90% GVBD only in the presence of its activating ligand L-Arg . The constitutively active SmVKR1 mutant induced GVBD , whereas a dead kinase mutant did not . When Smβ-Int1 , SmILK , and SmPINCH were coexpressed , no GVBD was observed . However , when these three proteins were co-expressed with SmVKR1 , GVBD was obtained independently of the addition of L-Arg ( Table 1 , Inj 11 ) . This suggested that SmVKR1 kinase activation could be induced by its participation to the complex with Smβ-Int1 , SmILK and SmPINCH . However , in this injection ( no . 11 ) GVBD was activated in the absence of SmNck2 . This finding led to the questions whether S . mansoni Nck2 is dispensable for complex formation , or whether Xenopus Nck2 may have rescued complex formation in this case ? When deletion mutants of SmILK ( SmILKΔAnk1 , missing the first ankyrin repeat necessary for interaction with PINCH; [32 , 33] ) or SmPINCH ( SmPINCHΔLIM4 , missing the fourth Lim domain necessary for interaction with Nck2/GFR; [32 , 33] ) or SmNck2 ( SmNck2ΔSH3 , missing the SH3 domain necessary for interaction with PINCH; [33] ) were used , activation of SmVKR1 was no more observed . The result with the deletion mutant of SmNck2 ( Table 1 , Inj 15 ) indirectly indicated the presence of Xenopus Nck2 in the complex and a competitive situation between Xenopus Nck2 and SmNck2ΔSH3 when the latter protein was present . Furthermore , adding the ILK-inhibitor QLT-0267 ( 1 μM ) also prevented GVBD in oocytes expressing the wildtype forms of Smβ-Int1 , SmILK , SmPINCH , SmNck2 , and SmVKR1 . These data suggest a direct interaction of these proteins , and also that Smβ-Int1-SmILK-SmPINCH-SmNck2-SmVKR1 complex formation is able to induce GVBD in Xenopus oocytes in the absence of a ligand for SmVKR1 . This interaction appeared to be specific for SmVKR1 , since other RTKs such as SmVKR2 [22] , SER ( S . mansoni EGF Receptor; [45 , 46] or the insulin receptor orthologs SmIR 1 and SmIR2 [47] were not activated by this complex . Furthermore , since GVBD was supposed to be dependent on the kinase activation of SmVKR1 , we checked the autophosphorylation status of SmVKR1 by Western blot analysis . GVBD occurred only when SmVKR1 was phosphorylated ( see below ) . To confirm the existence and the function of this complex , the HA-tagged intracellular part of Smβ-Int1 [17] was co-expressed in Xenopus oocytes together with V5-tagged variants of SmILK ( wildtype and SmILKΔAnk1 ) , SmVKR1 ( dead kinase and constitutively active mutants ) , or Flag-tagged SmPINCH ( wildtype and SmPINCHΔLIM4 ) and SmNck2 ( SmNck2ΔSH3 ) for co-immunoprecipitation . In this series of experiments , L-Arg was not used for stimulating SmVKR1 activity ( Table 2 ) . Besides investigating the GVBD-inducing activity of appropriate combinations of complex members in their wildtype or mutated forms , oocyte lysates were immunoprecipitated with an α-HA antibody , and Western blot analyses were performed with α-HA or α-V5 to investigate the presence of members complexed with Smβ-Int1 . The results showed V5-tagged SmILK , SmPINCH , and SmVKR1 in HA-tagged precipitates only when the wildtype forms were used ( Fig 2 , lane 11 ) . A replication of the experiment demonstrated complex formation also when the constitutively active SmVKR1 variant was used ( Fig 3 , lane 14 ) . As expected , no V5-tagged precipitates were detected , when the deletion variants SmILKΔAnk1 or SmPINCHΔLIM4 or the dead kinase variant of SmVKR1 were used . These results corresponded to the GVBD results obtained , confirming the formation of a complex of these four proteins . However , complex formation was possible due to the presence of Xenopus Nck2 ( Fig 3 ) , which was detected by Western blot analysis . This confirmed the previous interpretation of the GVBD experiment ( Table 1 ) . To investigate SmVKR1 activation upon complex formation , the phosphorylation status of this receptor was investigated . After confirming that SmNck2 is also part of the immunoprecipitated protein complex ( Fig 4A ) , Western blot analyses showed that SmVKR1 phosphorylation ( without adding L-Arg ) occurred only when it was coexpressed together with the wildtype forms of Smβ-Int1 , SmILK , SmPINCH , and SmNck2 ( Fig 4B ) . When deletion mutants of individual complex partners or the ILK inhibitor QLT-0267 were used , no SmVKR1 phosphorylation was detected . This is in perfect agreement with the GVBD results obtained with the same combinations of molecules ( Table 1 ) . In a previous study it was shown that upon SmVKR1 stimulation with L-Arg signaling pathways known to be involved in RTK signaling were activated in Xenopus oocytes . Among these were ERK , JNK , and Akt pathways [21] . To find out whether Smβ-Int1/SmVKR1 complex formation without L-Arg induction activates the same signaling cascades in Xenopus oocytes , we performed cotransfection experiments and subsequent phosphorylation assays . Indeed , the obtained results showed that SmVKR1 in cooperation with all complex partners induced the phosphorylation of ERK , JNK , and AKT in a ligand-independent manner ( Fig 5 ) . In analogy to the previous results , no phosphorylation of these signaling molecules was observed when one of the complex members was used in its mutated form or when the ILK-inhibitor QLT-0267 was applied . The effect of Smβ-Int1/SmVKR1 complex formation on the phosphorylation of ERK , JNK , and AKT resembled the activation of Xenopus oocyte receptors by insulin or the natural ligand progesterone . Because SmILK is one of the decisive complex partners mediating Smβ-Int1 cooperation with SmVKR1 we functionally analyzed this molecule in more detail . RNAi-mediated SmILK knock-down experiments were performed with S . mansoni couples in vitro , and the knock-down value determined by qPCR to be nearly 90% ( S5 Fig ) . Following treatment with SmILK-dsRNA , pairing stability was not affected , and the amount of couples was similar to the untreated control group . However , egg production per ( remaining ) couple of the treated group significantly decreased during the observation period from 48 h post treatment on compared to the control ( Fig 6 ) . Inhibiting ILK was also achieved by QLT-0267 , and following treatment with different concentrations ( 50–200 μM ) a negative effect on pairing stability was observed . Furthermore , also egg production per remaining couple decreased in a concentration-dependent manner from 48 h post treatment on ( Fig 7 ) . Morphologically , CLSM analysis showed effects of QLT-0267 on oogenesis in paired females . This inhibitor caused not only a reduction of oocyte number and the mislocalization of oocytes of various stages of differentiation in the different parts of the ovary but also oocyte degeneration ( Fig 8 ) . The intensity of the phenotype increased with QLT-0276 concentration . A similar oocyte-related observation was made by RNAi in ILK-dsRNA treated paired females ( Fig 8 ) , although the strength of the observed phenotype ( less oocytes , mislocalization , degeneration ) was weaker compared to inhibitor treatment . Previous studies in cancer cells provided evidence that among other functions ILK is involved in cytoskeletal reorganization and cell survival , and its deregulation can contribute to errors in cell division and genomic instability [48] . Microtubule disruption was shown to induce cytoskeleton as well as cell adhesion changes . This led to focal adhesion kinase hydrolysis and the onset of apoptosis , a phenotype that was rescued by ILK overexpression [49] . Because there is evidence that apoptosis has a biological function for the maintenance of the maturation state of the reproductive organs of paired females [50] , we investigated whether SmILK may be involved in this processes in S . mansoni . To this end we compared paired females treated with QLT-0267 and DMSO as control and performed immunolocalization with a β-tubulin antibody ( S6 Fig ) . Under inhibitor influence the number of immature and primary oocytes was reduced in inhibitor-treated females . Compared to the control , primary oocytes clustered closer together , they appeared more compact , and some appeared as rounded up ( Fig 9 ) . In a previous study first hints were obtained that laminins as extracellular matrix proteins may interact with Smβ-Int1 [17] . To investigate whether there is also an influence on components of the extracellular matrix we immunolocalized laminin in paired females treated with QLT-0267 or DMSO ( S6 Fig ) . Indeed , a concentration-dependent decrease of laminin staining was observed within the epithelium surrounding the ovary of treated females ( Fig 10 ) . TUNEL assays finally confirmed apoptotic processes in ovaries of females treated with QLT-0267 . TUNEL-positive cells occurred mainly within the smaller part of the ovary containing immature oocytes ( Fig 11 ) . To get further support for apoptotic processes induced by QLT-0267 in females , caspase-3 activity was determined in inhibitor-treated females . Following treatment the level of caspase-3 activity increased significantly ( Fig 12 ) . Next we investigated whether the expression of genes involved in early steps of apoptosis is affected . To induce the mitochondrial apoptosis pathway , a number of pro-apoptotic BCL-2 ( B cell lymphoma 2 ) proteins collaborate with the outer mitochondrial membrane to permeabilize it . BAK ( BCL-2 Antagonist Killer 1 ) and BAX ( BCL-2 Associated X protein ) are pro-apoptotic BCL-2 family members which are essential for the permebilization of the mitochondrial outer membrane [51] . We selected these genes because presumptive orthologs exist in the genome of S . mansoni ( BAK , Smp_095190; BAX , Smp_072180 ) . Therefore , we investigated the transcript profiles of Smp_095190 and Smp_072180 in schistosome females after treatment of couples for 72 h with 50 μM QLT . Compared to a DMSO control , we detected an upregulation of BAX after treatment , whereas BAK transcription remained constant . Upon RNAi both schistosome orthologs BAK and BAX were transcribed at higher levels ( S7 Fig ) .
Although much research has been performed on integrins and integrin signaling in different organisms , there is not much known about their roles in platyhelminths . Here we report on an integrin-signaling complex in S . mansoni consisting of Smβ-Int1 , SmILK , SmPINCH , SmNck2 , and SmVKR1 . According to phylogenetic analyses , SmILK , SmPINCH , SmNck2 form clusters that are specific for parasitic platyhelminths as it was shown for integrins before [17] . Together with the exclusive role of VKRs [25] , it appears likely that parasites have modified the function of insulin-like signaling as well as integrins and their interacting partners for specific signaling purposes . Among these , at least one deals with the reproductive biology of platyhelminths . In this context schistosomes exhibit remarkable features because of the pairing-dependent development and maintenance of the differentiation status of female gonads . The involvement of schistosome VKRs and integrins for this physiological process has already been demonstrated [17 , 21] , and studies on the VKR ortholog AaeVKR of A . aegypti support the assumption of a specific role of VKRs for oogenesis and/or egg formation [23] . Our study provides first evidence for cooperation between integrin and VKR signaling in S . mansoni . This interaction is mediated by SmILK , SmPINCH , and SmNck2 , cytoplasmic molecules with bridging function . Their colocalization with Smβ-Int1 and SmVKR1 , especially in the ovary , indicated potential functions for the reproductive biology of schistosomes . In all three cases the intensities of localization signals in the ovary were higher in its posterior part which contains mature primary oocytes . Smβ-Int1 and Smα-Int1 were localized in the ovary—also dominating in its posterior part - , the vitellarium , the testes , the ootype-surrounding area , the subtegument , and within the parenchyma [17] . SmVKR1 expression was localized mainly in the female ovary , especially in mature , primary oocytes in the posterior part . In addition , SmVKR1 was also localized around the ootype and in the parenchyma of males [21] . Thus SmILK , SmPINCH , SmNck2 colocalized widely with Smβ-Int1 and SmVKR1 including their preferential occurrence in mature , primary oocytes , a prerequisite for potential interactions . Different experiments with Xenopus oocytes expressing the complex members alone or in defined combinations of wildtype or mutated forms finally confirmed also by co-immunoprecipitation that a Smβ-Int1-SmILK-SmPINCH-SmNck2-SmVKR1 complex can be formed . GVBD assays demonstrated the biochemical function of this complex and the potentiality of SmVKR1 to be activated inside of the complex and to induce—in the absence of its ligand—processes leading to GVBD , which was confirmed by the results obtained . This suggests a new mode of SmVKR1 activation , which is achieved in a ligand-independent fashion by indirect cooperation with a β-integrin receptor . As mediators , GFR-specific and β integrin-specific adapter molecules operate , in this case SmILK , SmPINCH , and SmNck2 . The participation of SmNck2 was shown indirectly by the use of a deletion mutant that negatively influenced GVBD , by Western blot analysis confirming the presence of Xenopus Nck2 in complexes without SmNck2 , and finally by co-immunoprecipitation of SmNck2 after its addition . In Xenopus oocytes , ligand-activated RTKs as IRs trigger the activation of Erk MAPK and PI3K/Akt/mTOR and JNK pathways resulting in meiotic maturation [52] . As shown for L-Arg-activated SmVKR1 [21] , in our actual study the phosphorylation of Erk1/2 , Akt , and JNK in Xenopus oocytes was achieved also by Smβ-Int1/SmVKR1 complex formation without ligand activation . Thus similar to IR activation , complex-activated SmVKR1 induced signaling processes involved in protein synthesis and cellular growth associated with Xenopus oocyte maturation , which substantiates the IR-like function of SmVKR1 but also its conjunction with oogenesis . Functional analyses of SmILK in Xenopus oocytes or as a member of the complex by RNAi and inhibitor studies finally indicated that this molecule represents a pseudokinase being involved in different processes in schistosomes . Among these is inside-out signaling in the ovary because the extracellular matrix as part of the epithelium surrounding the ovary was changed upon inhibiting SmILK as shown by laminin immunolocalization . Furthermore , SmILK appeared to control oocyte localization within the ovary , and oocyte survival . Inhibiting SmILK activity led to the reduction of the amount of immature oocytes and the degeneration of mature , primary oocytes . This is in part explained by apoptotic processes , for which evidence was obtained by TUNEL assays in case of immature oocytes , by determining caspase-3 activity which increased following inhibitor treatment , and by transcriptional analysis of the schistosome orthologs of BAK and BAX , two pro-apoptotic genes [51] . A recently conducted RNA-seq study revealed that both genes were expressed in schistosome females and within the ovary . Interestingly , the profiling of transcript abundance revealed that both genes were more abundantly transcribed in the ovaries of unpaired , immature females . After pairing , transcript abundance of both genes decreased in the ovary ( [53]; S7 Fig ) . This supports the conclusion that apoptosis plays a role in oocyte differentiation , and that males exert a regulatory influence on this—suppressing apoptosis in the gonads of their female partners during a constant pairing contact ( Fig 13 ) . Degenerated primary oocytes were also detected that did not respond to TUNEL staining . Thus it seems feasible that further , apoptosis-independent processes leading to cell death contribute to oocyte degeneration . In summary , the results presented here strongly suggest that the Smβ-Int1/SmVKR1 complex in the ovary of paired schistosomes is important for the maintenance of the differentiation status of oocytes and their survival . Against the background of the unusual reproductive biology of schistosomes this conclusion supports findings of a previous , independent study showing that apoptosis is used to control vitelline cell survival in a pairing-dependent manner in S . mansoni [50] . In this context our results match to a scenario of cell death processes controlling gonad maintenance in schistosome females and thus contribute to the understanding of biological processes controlling reproductive biology in this exceptional parasite . It has been hypothesized before that SmVKR1 , possibly activated by L-Arg delivered with the male seminal fluid [21] , is responsible for meiosis resumption and/or oocyte migration in schistosome females . In view of the new results it appears feasible that integrin-signaling contributes to this process providing a SmVKR ligand-independent alternative for activation ( Fig 13 ) . This could be achieved by mechanosensory forces . Indeed , integrins have been shown to sense , sort , and transduce mechanical forces into cellular responses . This form of integrin-based mechanotransduction contributes among others to cell growth , cell migration , gene expression including the activation of kinases , but also to apoptosis [54–57] .
Animal experiments using Syrian hamsters ( Mesocricetus auratus ) as model hosts were performed in accordance with the European Convention for the Protection of Vertebrate Animals used for experimental and other scientific purposes ( ETS No 123; revised Appendix A ) and were approved by the Regional Council ( Regierungspraesidium ) Giessen ( V54-19 c 20/15 c GI 18/10 ) . S . mansoni was maintained in Biomphalaria glabrata as the intermediate host , with Syrian hamsters ( Mesocricetus auratus ) as the definitive hosts [36] . Adult worms were obtained by hepatoportal perfusion at day 46 or day 67 ( in case of single sex infection; [58] ) post-infection , respectively , and kept in M199 medium ( Gibco ) supplemented with 10% newborn calf serum and 1% ABAM-solution ( 10 , 000 units penicillin , 10 mg streptomycin and 25 mg amphotericin B per ml ) at 37°C and 5% CO2 for 24h until the experiments started . Couples were cultivated in 6-well plates in groups of eight per well ( n = 3 ) and 3 ml supplemented M199 medium for RNA interference ( RNAi ) experiments ( see below ) or inhibitor studies . The latter were performed with the integrin-linked kinase ( ILK ) inhibitor QLT-0267 ( Dermira , Inc . , USA; [59] ) which was added at final concentrations and period of times as indicated . It targets the ATP-binding site of ILK and was shown to be as effective as siRNA-mediated depletion of ILK [60] . Since ILK exerts no catalytic function , the inhibitory effect of QLT-0267 was explained by an impairment of the stability of ILK [61] . Equivalent volumes of dissolvent DMSO was used as control . Worms were monitored by bright-field microscopy ( CX21 , Olympus; Labovert FS , Leitz ) over periods of 24 h– 96 h to analyze pairing stability , egg production , gut peristalsis and movement . For cloning of the full-length cDNAs of SmILK ( Smp_079760 ) , SmPINCH ( Smp_020540 . 2 ) , and SmNck2 ( Smp_014850 ) , total RNA was isolated from adult schistosomes using Trizol reagent ( Invitrogen ) . Residual DNA was removed by DNase digestion ( RNAeasy kit , Qiagen ) following the manufacturer’s instruction . RNA quality was checked by Bioanalyzer microfluidic electrophoresis ( Agilent Technologies ) . Starting RT-PCR the synthesis of cDNA was performed with 1 μg RNA using QuantiTect Reverse Transcription Kit ( Qiagen ) . PCR reactions were performed in a final volume of 25 μl using primer end concentrations of 800 nM , denaturation at 95°C for 30 sec , annealing at 54°–64°C depending on the primer combinations ( S1 Table ) , and elongation at 72°C for up to 2 min , and using FirePol-Taq ( Solis biodyne ) . As vectors for cloning , pACT2 ( Clontech ) , pcDNA3 . 1 ( Invitrogen ) , or pBridge ( Clontech ) were used for directional cloning via restriction enzyme sites . Full-length SmILK cDNA was cloned via NotI and XbaI into pcDNA3 . 1 , full-length SmPINCH cDNA via EcoRI/PstI into pBridge , and full-length SmNck2 via BamHI and XbaI into pcDNA3 . 1 ( S1 Table ) . Primers designed for RT-PCRs to generate these cDNAs contained appropriate restriction sites for cloning . The sequence integrities of all cloned cDNAs were verified by sequencing ( LGC Genomics , Berlin ) . Ovaries of female worms were isolated using the combined detergents/enzyme-based organ isolation protocol [42] . In short , isolated adult females ( about 50 each ) were transferred into 2 ml-reaction vessels and washed twice with 2 ml of non-supplemented M199-medium at room temperature . The medium was removed , and 500 μl of tegument solubilisation ( TS ) -solution was added ( 0 . 1% of each following compounds in DEPC ( diethylpyrocarbonate ) /PBS ( phosphate-buffered saline ) : Brij 35 ( Roth ) , Nonidet P40-Substrate ( Fluka ) , Tween80 ( Sigma ) and TritonX-405 ( Sigma ) , pH 7 . 2–7 . 4 ) followed by incubation in a thermal shaker ( TS-100 , Biosan ) for 5 min at 1 , 200 rpm at 37°C . Shaking was repeated twice , and the solution was replaced after each cycle . Then the worms were rinsed three times with M199 and subsequently treated with elastase ( 300 μl elastase solution: 5 U/ml in M199; Sigma ) at 37°C and 650 rpm in the thermal shaker to release the ovaries . Digestion was monitored by bright-field microcopy ( Leica ) and stopped when the gonads were released from the disrupted and digested worm carcasses . Finally , the gonads were manually collected by pipetting and transferred into supplemented M199 medium . Sample preparation of S . mansoni adults was conducted as described previously [62] . In short , schistosome pairs were fixed in Bouin's solution ( picric acid/acetic acid/formaldehyde; 15/1/5 ) followed by embedding in paraplast ( Paraplast plus , Sigma ) . Sections of 5 μm thickness were incubated in xylol and after rehydration , the sections were treated with proteinase K ( 1 μg/ml ) and dehydrated . As probe , in vitro-generated transcripts were synthesized and labeled with digoxigenin as suggested by the manufacturer ( Roche ) . The correct sizes of labeled sense and antisense transcripts were checked by gel electrophoresis , and the RNA quality was tested by blotting and detection of digoxigenin using alkaline phosphatase-conjugated anti-digoxigenin antibodies , naphtol-AS-phosphatase , and Fast Red TR ( Sigma ) . In situ-hybridization was performed at 57°C for 16 h . Afterwards , the sections were washed up to 0 . 5 × SSC ( 75 mM NaCl , 7 . 5 mM sodium citrate , pH 7 . 0 ) , and detection of alkaline phosphatase was performed as mentioned above . The intracellular part of Smβ-Int1 containing the C-terminus with an HA-tag at its N-terminus was subcloned into pcDNA 3 . 1 ( Invitrogen ) as described earlier [17] . Capped messenger RNA ( cRNA ) encoding Smβ-Int1 C-term was synthesized in vitro ( T7 mMessage machine Kit , Ambion , USA ) following a previously established protocol [45] . Furthermore , V5-tagged SmILK and SmPINCH and Flag-tagged SmNck2 were cloned the same way into pcDNA 3 . 1 , and their sequence identities confirmed by commercial sequencing . Also cRNAs were prepared from these clones as well as from V5-tagged SmVKR1 variants ( wildtype SmVKRwt , dead kinase mutant SmVKRdk [= KO] , and a constitutively active mutant SmVKR1YYRE [= XE] ) cloned in pcDNA 3 . 1 as reported in a previous study [22] . Interaction studies between these proteins ( see results ) were done by co-injecting different cRNA combinations into Xenopus oocytes as reported before [17 , 45] . Expressed proteins were detected by immunoprecipitation and Western blot analyses . Following the standard procedure [45] , 30 oocytes were lysed in 300 μl of buffer ( 50 mM HEPES , pH 7 . 4 , 500 mM NaCl , 5 mM MgCl2 , 1 mg/ml bovine serum albumin , 10 μg/ml leupeptin , 10 μg/ml aprotinin , 10 μg/ml soybean trypsin inhibitor , 10 μg/ml benzamidine , 1 mM PMSF , 1 mM sodium vanadate ) after 5 h or 15 h of expression . Following centrifugation at 4°C for 15 min and 10 , 000 g , the resulting supernatants were incubated with anti-HA ( 1:100; Invitrogen ) or anti-V5 ( 1:100; Invitrogen ) then added to protein A-Sepharose beads ( 5 mg , Amersham Biosciences ) for 1 h at 4°C . After washing three times , immune complexes were eluted from the beads in Laemmli buffer and analyzed by SDS-PAGE ( 7 . 5%–15% polyacrylamide gels ) . Western blot analyses were performed using anti-V5 ( 1: 50 , 000 ) , anti HA ( 1: 50 , 000 ) , anti-Flag ( 1: 1 , 000 ) , anti-human nck2 ( 1: 1 , 000 , nck2 ( 8 . 8 ) : sc-20020 , Santa Cruz Biotechnology ) , or PY20 ( 1: 10 , 000; anti-phosphotyrosine , BD Biosciences ) antibodies . The following primary antibodies were applied to confirm the presence of total or phosphorylated ERK2 , JNK and Akt kinases: anti-ERK2 ( 1: 10 , 000; Santa Cruz Biotechnology ) , anti-phospho p44/p42 MAPK ( ERK1/2; Thr 202/Tyr 204; 1: 10 , 000; Cell Signalling Technology ) , anti-c-jun N-terminal kinase JNK ( 1: 10 , 000; Sigma ) , anti-active JNK polyclonal antibody ( 1: 8 , 000; Promega ) , anti-Akt1 ( C-20; 1: 5 , 000; Santa Cruz Biotechnology ) , anti-phospho Akt ( Thr308; 1: 5 , 000; Upstate Biotechnology ) and anti-phospho Akt ( Ser 473; 1: 5 , 000; Upstate Biotechnology ) . Mouse , rabbit or goat Trueblot secondary antibodies ( eBioscience ) were used as secondary antibodies and chemoluminescence was detected using the advanced ECL detection system ( Amersham Biosciences ) . Following standard protocols for RNAi in adult schistosomes [15 , 63] , double-stranded RNA of approximately 500 bp was synthesized ( nucleotide position 559–1016 ) using the MEGAscript RNAi kit ( Life Technologies ) . Gel electrophoresis in 1 . 2% agarose-MOPS was conducted to prove for single RNA bands of the correct size . Schistosome couples in groups of eight pairs ( n = 3 ) were electroporated in the presence of 25 μg dsRNA and subsequently soaked in vitro for 96 h . Every 24 h the treated worms were inspected and different parameters evaluated such as pairing stability , egg production , gut peristalsis and movement . Schistosome samples were collected and transferred into PeqGOLD TriFast ( Peqlab ) . After storage at -80°C or immediately after transfer , RNA isolation was done following the manufacturer’s instructions ( Peqlab ) . About 500 ng total RNA was used for cDNA synthesis using the Quantitect Reverse Transcription Kit ( Qiagen ) . For PCR , 1 μl of a 1:20 dilution of cDNA was tested using exon-spanning PDI ( protein dilsufide isomerase ) 5’/3’ primers; forward: 5´-AAATGATGCCCCGACTTACC-3´ and reverse: 5´- TCATCCCAAACTGGAGCAAG-3`[62 , 64] ) to confirm that the genomic DNA was properly removed . For quantitative RT PCR ( qRT-PCR ) , a RotorGene-Q PCR cycler ( Qiagen ) was used and all reactions were set up in triplicates . Each reaction had a final volume of 25 μl; 5 μl of a 1:20 cDNA served as template and 125 nM ( final concentration ) of each primer were added to 12 . 5 μl of 2x PerfeCTa SYBR Green super mix ( Quanta ) . No template controls ( NTC ) were included in each run . RNAi-mediated knockdown of gene expression was analysed by absolute quantification . Therefore , a standard curve on diluted gel eluate was included in each run [65] . For qPCR analyses to study transcript profiles of SmBAK ( Smp_095190 ) , SmBAX ( Smp_072180 ) , SmmTor ( Smp_122910 ) , and SmSod ( Smp_056440 ) several genes were tested for their suitability as reference . Based on transcriptome data [53] the gene Smp_008900 ( annotated as eukaryotic translation initiation factor 4 gamma ) fulfilled this criterion and was further tested under various condition using an absolute quantification approach . To this end , the Smp_008900 amplicon was cloned into pDrive ( Qiagen ) and served as template in dilution series . Different cDNAs of electroporated and inhibitor-treated S . mansoni couples confirmed constant numbers of Smp_008900 transcripts ( primers , see S1 Table ) . Sliced specimens of 4 μm thickness on slides were deparaffinized , dehydrated and then equilibrated in proteinase-K buffer ( 100 mM Tris/Cl , 50 mM EDTA pH 8 . 0 ) for 5 min . Subsequently , treatment with proteinase K ( 1 μg/ml ) was performed for 20 min at 37°C . Afterwards , slides were rinsed once with 1x PBS and immersed twice with 500 μl washing buffer provided by the fluorometric DNA-fragmentation detection kit III ( F-dUTP; Promokine ) for 5 min at ambient temperature . The staining solution containing FITC-labeled dUTP was prepared according to the manufacturer’s instruction ( Promokine ) . A control solution was prepared without TdT enzyme . Subsequently , specimens were immersed with 100–200 μl of the staining solution and kept in the dark . Following incubation for 60 min at 37°C , the slides were rinsed twice with 500 μl rinse buffer ( Promokine ) for 5 min at ambient temperature and counterstained with 200 μl propidium iodide/RNase A solution ( Promokine ) for 15 min . Slides were finally mounted with FluoroMount ( Roth ) and analyzed within 3 hours after staining by fluorescence microscopy ( Ex/Em = 488/520 nm for FITC , and 488/623 nm for PI ) . Following perfusion , S . mansoni couples were taken in culture and treated with DMSO or with QLT-0267 ( 100 μM ) for 72 h . Following treatment the couples were carefully separated with featherweight forceps , and females and males transferred separately into 1 . 5 ml tubes with 500 μl 1x PBS for washing . After sedimentation , the supernatant was replaced by 50 μl cold ( 4°C ) cell lysis buffer of the caspase-3 colorimetric assay kit ( Promokine ) and the samples kept on ice . After 10 min incubation , worms were homogenized with sterile pestils and kept on ice for further 10 min . Debris were subsequently sedimented by 10 . 000 x g centrifugation for 2 min at 4°C and afterwards placed back on ice . Subsequently , 20 μl of the supernatant were mixed with 30 μl of pre-transferred cell-lysis buffer in 96-well plates , and 50 μl of two-fold reaction buffer complemented with 10 μM DTT was added . Two hours incubation at 37°C allowed cleaving the p-nitroanilide-labeled substrate DEVD ( Promokine ) that was added at a final concentration of 200 μM . Samples were read at 405 nm with a Varioscan plate reader ( Thermo Fisher Scientific ) . Samples were normalized to the protein concentration that was determined using the BCA assay kit ( Pierce ) promptly after the caspase-3 assay was set up . The results relied on the analysis of three biological replicates obtained from independent perfusions . In contrast to previously published protocols [50] caspase-3 activity was not detected when the homogenization step was omitted . Adult worms were stained with carmine red for a general morphological analysis by CLSM according to previously published protocols [14 , 66] . For microscopy ( CLSM; Leica TSC SP2 microscope ) and documentation the probes were excited with a 488 nm He/Ne laser and emission was captured with a 470 nm long-pass filter in reflection mode as described before [14] . Fluorescence microscopy of antibody-stained worm sections ( 5 μM ) was done using an Olympus IX 81 inverted microscope . Anti-laminin ( antibodies-online . com; LN , ABIN268409 ) and anti-β-tubulin ( antibodies-online . com; anti-TUBB , ABIN269949 ) antibodies were used in concentrations of 1: 5 , 000 each , as recommended by the manufacturer ( Novus Biologicals ) . As secondary antibody , a fluorescence-labeled goat anti-rabbit antibody was used ( LI-COR Bioscience , IRDye 680LT; 1: 5 , 000 ) . These antibodies were tested on lysates of adult schistosomes by Western blot analyses as described before [42] using 15 μg protein each , which had been size-separated by SDS-PAGE using 7 . 5%–15% polyacrylamide gels depending on the size of the protein to be detected . The following public domain tools were used: BLASTx ( http://www . ncbi . nlm . nih . gov/BLAST ) , SchistoDB ( http://schistodb . net/schisto/; [67] ) , and WormBase ParaSite ( release 6 , April 2016; http://parasite . wormbase . org/; [37] ) . The online-tool SMART ( http://smart . embl-heidelberg . de/ ) [68] was used to predict protein domains . Primer3Plus was used for primer design ( http://www . bioinformatics . nl/cgi-bin/primer3plus/primer3plus . cgi ) and Oligo Calc for analyzing primer properties ( http://www . basic . northwestern . edu/biotools/oligocalc . html ) . | Parasites of the genus Schistosoma cause schistosomiasis , a life-threatening infectious disease for humans and animals worldwide . Among the remarkable biological features of schistosomes is the differentiation of the female gonads which is controlled by pairing with the male and a prerequisite for egg production . Eggs , however , are not only important for the maintenance of the life-cycle; they also cause the pathological consequences of schistosomiasis . Part of the eggs gets trapped in host tissues such as liver and spleen and trigger inflammatory processes , finally leading to liver cirrhosis . Research activities of the last decade have indicated that different families of cellular and receptor-type kinases but also integrins contribute to the control of mitogenic activity and differentiation the female goands . In this context an unusual class of receptor tyrosine kinases ( RTKs ) has been identified , the venus kinase receptors ( SmVKRs ) . By biochemical and molecular approaches we demonstrate that SmVKR1 activation can be achieved by cooperation with a signaling complex consisting of the beta integrin receptor Smβ-Int1 and the bridging molecules SmILK , SmPINCH , SmNck2 . Besides unravelling a novel way of SmVKR1 activation , we provide evidence that this complex controls the differentiation status of oocytes by regulating cell death-associated processes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"helminths",
"cell",
"processes",
"vertebrates",
"animals",
"xenopus",
"animal",
"models",
"germ",
"cells",
"oocytes",
"model",
"organisms",
"amphibians",
"integrins",
"immunoprecipitation",
"experimental",
"organism",
"systems",
"co-immunoprecipitation",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"cell",
"adhesion",
"extracellular",
"matrix",
"precipitation",
"techniques",
"xenopus",
"oocytes",
"ovaries",
"anatomy",
"cell",
"biology",
"ova",
"apoptosis",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"frogs",
"organisms"
]
| 2017 | Evidence for Integrin – Venus Kinase Receptor 1 Alliance in the Ovary of Schistosoma mansoni Females Controlling Cell Survival |
Yersinia pestis , the agent of plague , is transmitted to mammals by infected fleas . Y . pestis exhibits a distinct life stage in the flea , where it grows in the form of a cohesive biofilm that promotes transmission . After transmission , the temperature shift to 37°C induces many known virulence factors of Y . pestis that confer resistance to innate immunity . These factors are not produced in the low-temperature environment of the flea , however , suggesting that Y . pestis is vulnerable to the initial encounter with innate immune cells at the flea bite site . In this study , we used whole-genome microarrays to compare the Y . pestis in vivo transcriptome in infective fleas to in vitro transcriptomes in temperature-matched biofilm and planktonic cultures , and to the previously characterized in vivo gene expression profile in the rat bubo . In addition to genes involved in metabolic adaptation to the flea gut and biofilm formation , several genes with known or predicted roles in resistance to innate immunity and pathogenicity in the mammal were upregulated in the flea . Y . pestis from infected fleas were more resistant to phagocytosis by macrophages than in vitro-grown bacteria , in part attributable to a cluster of insecticidal-like toxin genes that were highly expressed only in the flea . Our results suggest that transit through the flea vector induces a phenotype that enhances survival and dissemination of Y . pestis after transmission to the mammalian host .
Arthropod-borne transmission of bacterial pathogens is somewhat rare but has evolved in a phylogenetically diverse group that includes the rickettsiae , Borrelia spirochetes , and the gram-negative bacteria Francisella tularensis and Yersinia pestis , the plague bacillus . Y . pestis circulates among many species of wild rodents , its primary reservoir hosts , via flea bite . As it alternates between fleas and mammals , it is postulated that Y . pestis regulates gene expression appropriately to adapt to the two disparate host environments , and that different sets of genes are required to produce a transmissible infection in the flea and disease in the mammal . Many important Y . pestis virulence factors that are required for plague in mammals have been identified , and most of them are induced by a temperature shift from <26°C to 37°C , which mimics the transition from a flea to the warm-blooded host [1] . To date , only three transmission factors ( genes specifically required to produce a transmissible infection in the flea ) have been characterized . One , the yersinia murine toxin ( ymt ) gene , encodes a phospholipase D that is required for survival in the flea midgut [2] . The other two , ( hmsHFRS and gmhA ) , are responsible for an extracellular polysaccharide and a lipopolysaccharide ( LPS ) core modification that are required for normal biofilm formation and blockage in the flea [3] , [4] . Biofilm development in the flea digestive tract is important for biological transmission [5] , [6] , [7] . After being taken up in a blood meal , Y . pestis proliferates in the lumen of the flea midgut to form cohesive multicellular biofilm aggregates . In some infected fleas , the proventricular valve between the midgut and esophagus is colonized . The subsequent growth and consolidation of the adherent Y . pestis biofilm amongst the rows of cuticle-covered spines that line the proventriculus interferes with normal blood feeding , resulting in regurgitation of bacteria and transmission . Fleas with a completely blocked proventriculus make prolonged , repeated attempts to feed , increasing the opportunities for transmission . Formation of a Y . pestis biofilm in vitro and in the flea proventriculus depends on synthesis of an extracellular polysaccharide matrix ( ECM ) that is synthesized only at temperatures below 26°C [3] , [7] . In common with many other bacteria , ECM synthesis in Y . pestis is controlled by intracellular levels of cyclic di-GMP , which are determined by competing activities of the hmsT diguanylate cyclase and hmsP phosphodiesterase gene products [8] , [9] . Bacterial adhesins are typically required for initial adherence and autoaggregation in biofilm development [10] , but such factors have yet to be identified in Y . pestis . In a previous study , we reported the in vivo gene expression profile of Y . pestis during bubonic plague in rats [11] . In this study , we characterized the Y . pestis transcriptome in blocked Xenopsylla cheopis rat fleas , an important vector of plague to humans . Comparing the Y . pestis gene expression profile in the flea to those of in vitro biofilm and planktonic cells cultured at the low temperature typical of the flea implicated several genes in a flea-specific adaptive response and in proventricular blockage . In addition , comparing the gene expression patterns in the flea and in the rat bubo confirmed that distinct subsets of genes are differentially expressed during the Y . pestis life cycle . Notably , several genes with known or predicted roles in protection against the mammalian innate immune system and in pathogenesis were upregulated in the flea , suggesting that transit through the insect vector preinduces a phenotype that enhances Y . pestis survival and dissemination in the mammal after flea-borne transmission .
Little is known about the environmental conditions in the flea digestive tract , how Y . pestis adapts to them , or the physiological state of the bacteria at transmission when they exit the flea and enter the mammal . Adult fleas are obligate blood feeders and take frequent blood meals , consisting primarily of protein and lipid with relatively little carbohydrate . Flea proteases , lipases , and other digestive enzymes begin to process the blood meal in the midgut immediately after feeding , yielding amino acids and peptides , glycerol , fatty acids , and simple carbohydrates [12] . This provides the “medium” for Y . pestis growth , but these and other factors such as pH , oxygen tension , osmolarity , and flea antibacterial immune components are poorly defined . During the first week after being ingested in an infectious blood meal , Y . pestis grows rapidly in the flea midgut to form large bacterial aggregates . Bacterial load peaks at about 106 cells per flea as the Y . pestis biofilm accumulates in the proventriculus to cause blockage , and then plateaus [2] , [3] . In this study , we determined the Y . pestis gene expression profile in infective , blocked fleas , in which the proventriculus was occluded with a mature bacterial biofilm . Y . pestis KIM6+ , which lacks the 70-kb virulence plasmid that is not required for flea infection or blockage [3] was used for this analysis . Blockage occurred between 1 . 5 and 3 . 5 weeks after the initial infectious blood meal , during which time the fleas fed on uninfected mice twice weekly . The Y . pestis in vivo biofilm transcriptome was compared to the transcriptomes of in vitro biofilm and planktonic cultures grown at 21°C , the same temperature at which the fleas were maintained . Expression of 55% of Y . pestis ORFs was detected in the flea samples; and 74 to 79% in the in vitro biofilm , exponential phase planktonic and stationary phase planktonic cultures . Principal component analysis to visualize overall clustering of the microarray data showed that the transcriptional profiles were reproducible and discrete for the in vitro and in vivo conditions ( Fig . 1A ) . Profiles of the exponential and stationary phase planktonic cultures clustered most closely , whereas the profiles from in vitro and in vivo biofilm growth were more distinct from each other and from the planktonic culture profiles . There were 214 Y . pestis genes whose expression was significantly upregulated and 56 genes downregulated in the flea compared to all in vitro growth conditions ( Fig . 1B; Tables S1 and S2 ) . Quantitative RT-PCR analysis of a subset of Y . pestis genes differentially expressed in the flea was confirmatory of the microarray results ( Fig . S2 ) . Of the 214 genes upregulated in the flea gut compared to all in vitro conditions , 78 are metabolic genes , 60 of which are involved in uptake and catabolism of amino acids and carbohydrates ( Table S1 ) . In particular , genes involved in transport and catabolism of the L-glutamate group of amino acids ( Gln , His , Arg , and Pro ) were specifically upregulated in the flea ( Fig . 2 ) . The degradation of these amino acids gives rise to L-glutamate and the TCA cycle intermediates succinate , formate , and α-ketoglutarate . The gabD and gabT genes involved in the production of succinate from γ-aminobutyrate ( GABA ) , another member of the L-glutamate group , were also highly induced in the flea . The gabD gene functions to produce succinate from both GABA and hydroxyphenylacetate ( HPA ) , an aromatic degradation product of Tyr and Phe; and the HPA transport ( hpaX ) and catabolism ( hpaCBIFHDE ) genes of Y . pestis were also highly upregulated in the flea gut ( Table S1 , Fig . 2 ) . As Y . pestis does not have homologs of genes required to produce GABA or HPA , these metabolites may be taken up from the flea digestive tract . Alternatively , the gabD and gabT gene products might act in the reverse direction to synthesize GABA , which has osmoprotective properties [13] . The central role of the L-glutamate family of amino acids may also confer this advantage in the flea gut , because Glu and Pro are osmoprotectants . Interestingly , both glutamate and GABA are important neurotransmitters at the neuromuscular junction of insects , and the concentration of glutamate is very low in insect hemolymph , suggesting that it is converted to glutamine before it is absorbed [14] . Insect midgut epithelium is typified by multiple amino acid transporters with specific substrates and rapid absorption kinetics , but different amino acids enter the hemocoel at different rates and amounts [14] , [15] . Thus , Y . pestis metabolism in the flea may reflect the available pool of amino acids in the midgut . In contrast to the amino acids , hexoses do not appear to be an important energy source during infection of the flea . Only the genes encoding for chitobiose phosphotransferase ( PTS ) uptake and utilization systems ( chbBC; chbF ) , and for a PTS system of unknown specificity ( frwBCD ) were significantly upregulated in the flea [16] , [17] . Chitobiose could be present in the flea gut due to turnover of the chitin layer on the proventricular spines . Expression of the glucose PTS system was only slightly increased relative to LB cultures , and other PTS systems were downregulated ( Table S2 ) . Glycolytic pathways were not upregulated in the flea; instead , available hexoses and the gluconeogenesis pathway may be used to synthesize polysaccharide components required for cell growth . Upregulation of the actP and acs genes in the flea , which direct the uptake of acetate and its conversion to acetyl-CoA , also suggests that insufficient acetyl-CoA is produced by glycolysis to potentiate the TCA cycle . The switch from acetate secretion to acetate uptake is typical of growth in a glucose-limited , amino acid rich environment [18] . In contrast to hexose uptake systems , Y . pestis genes that encode permeases for the pentoses ribose , xylose , and arabinose were induced in the flea gut . Acquisition of pentoses from the environment may be important because Y . pestis does not possess glucose 6-phosphate dehydrogenase activity , the first step of the pentose phosphate pathway [19] . Although the flea gut contains lipid derived from the blood meal , Y . pestis does not appear to use it as a major energy source . None of the fatty acid uptake or catabolism genes were upregulated in the flea compared to growth in LB . However , genes for glycerol and glycerol-3-phosphate uptake and utilization were upregulated , suggesting that flea digestion products derived from blood glycerolipids may be used by Y . pestis . In summary , Y . pestis appears to use amino acids , particularly the L-glutamate family , as primary carbon , nitrogen , and energy sources in the flea . Amino acid carbon is presumably funneled into the TCA cycle , the genes for which are highly expressed in the flea ( Table S3 ) . Because blockage of the flea vector is essentially a biofilm phenomenon , Y . pestis genes whose expression patterns are significantly upregulated in the flea and flowcell biofilms relative to planktonic cultures ( Table S4 ) might indicate that they are transmission factors . Several studies comparing the transcriptional profiles of Escherichia coli and other gram negative bacteria during biofilm and planktonic growth in vitro have been published [20] , [21] , [22] , [23] . Certain genes whose mutational loss resulted in an altered biofilm phenotype have been identified in these studies; but in general a consistent , distinct biofilm gene expression profile has not emerged . This is probably because different media and experimental systems have been employed and the fact that a biofilm consists of a physiologically heterogeneous community [24] , [25] . Nevertheless , common biofilm-related adaptations include the repression of motility and the induction of specific adhesins , an extracellular polysaccharide matrix ( ECM ) , and an envelope stress response ( ESR ) [10] , [23] . However , Y . pestis is constitutively nonmotile , and synthesis of the Hms-dependent biofilm ECM is regulated post-translationally [26] . The ymt gene was among the most highly expressed genes in the flea ( Table S3 ) , but neither it nor the known transmission factors ( hmsHFRS , hmsT , hmsP , and gmhA ) showed significantly higher expression in the flea than in vitro at 21°C , indicating that they are induced primarily by low temperature , and not by environmental factors specific to the flea gut . Y . pestis homologs of two genes with previously identified roles in biofilm , yidE , which encodes a hyperadherence factor in E . coli [27] , and cpxP , a member of the cpxPAR ESR system , were upregulated in the flowcell; but predicted adhesin genes were not upregulated . The transcriptional profile of Y . pestis in blocked fleas showed greater similarity to the transcriptional profile reported for E . coli in mature , four-day-old in vitro biofilms [23] . In addition to yidE and cpxP , other Y . pestis predicted adhesins and components of an ESR were upregulated in the flea . The Y . pestis homologs of Pseudomonas aeruginosa cupA1 and cupA3 in a predicted fimbrial biosynthesis operon and yapL , a predicted autotransporter adhesin similar to E . coli tibA , were specifically upregulated in the flea ( Table S1 ) . The cupA fimbrial locus and tibA are important for surface adherence and for biofilm formation in P . aeruginosa and E . coli , respectively [28] , [29] . Evidence for induction of an ESR in the flea included the high expression levels of rpoE , the gene for the alternate transcription factor σE ( as well as the anti-σE negative regulator genes rseA and rseB ) , cpxP; and pspA and pspG , components of the phage-shock protein ( Psp ) response ( Tables S1 and S3 ) . These genes were also found to be upregulated in mature E . coli biofilms [23] , suggesting that the three prominent ESR systems are important for integrating signals required for survival in a biofilm . Because homologs of the yidE , cpxP , tibA ( yapL ) , cupA fimbriae , and pspABC genes were upregulated in the flea and have been shown to be involved in biofilm formation in other bacteria [23] , [27] , [28] , [29] , we made a series of Y . pestis strains containing deletions of these loci . However , the single loss of any of these genes did not result in a noticeable defect in biofilm formation in vitro , or in flea infection or blockage ( data not shown ) . These genes may contribute to biofilm formation , but are not individually essential for this phenotype . Although genes in the polyamine transport gabTpotDBC locus are among the most highly induced genes in the flea ( Table S1 ) and polyamines are essential for Y . pestis biofilm formation [30] , we have previously reported that a Y . pestis Δpot mutant has no defect in flea infection or blockage [31] . This is likely due to the fact that Y . pestis is able to synthesize polyamines de novo . With this study , the in vivo transcriptome of Y . pestis in blocked fleas and in the rat bubo [11] have now both been characterized . A comparison of normalized gene expression levels from the two data sets provides insight into the biology of the flea-mammal life cycle . About 15% of Y . pestis genes showed significantly higher relative expression levels or expression only in the flea than in the bubo; 24% were more highly expressed in the bubo than in the flea; and 61% were not differentially expressed in the two hosts ( Fig . 3 ) . Several virulence factors were differentially regulated in the two hosts , but others were not ( Table 1 ) . In addition to the known temperature-induced virulence factors , iron acquisition systems , including the ybt and yfe operons that are required for virulence; and oxidative and nitrosative stress response genes , including the hmp virulence factor , are highly upregulated in the rat bubo , but not the flea . The analysis also reinforces the model that Y . pestis produces a hexaacylated lipid A in the flea , and that the change to the less immunostimulatory tetraacylated form occurs only after transmission [32] . Other virulence and transmission factors were not differentially regulated , including the hms genes; and the Y . pestis plasminogen activator ( pla ) , critical for dissemination from extravascular tissue at the fleabite site [33] , and ymt were highly expressed in both hosts ( Table S3 and [11] ) . The Y . pestis outer surface protein gene yadB , recently shown to be required for dissemination and bubonic plague pathogenesis from a subcutaneous inoculation site [34] , was significantly upregulated in both the flea and the bubo compared to in vitro conditions ( Tables 1 , S1 ) . Expression of genes in the pH 6 antigen locus ( psaEFABC ) , responsible for the synthesis and transport of the PsaA fimbriae that enhance resistance to phagocytosis by macrophages [35] , [36] , were higher in the bubo than the flea , although the usher protein gene psaC was upregulated in the flea compared to in vitro growth ( Tables 1 , S1 ) . The psa locus is regulated by RovA [36] . Consistent with these findings , rovA expression was downregulated in the flea; whereas expression of rovM , a negative regulator of rovA [37] , was upregulated . The transcriptional regulator gene phoP of the PhoPQ two-component regulatory system and the PhoP-regulated mgtC gene were expressed at levels >2-fold higher in fleas than in any other condition ( Tables 1 , S1 , S3 ) . PhoP and MgtC are established virulence factors known to be important for survival of Y . pestis and other gram-negative bacteria in macrophages and for resistance to cationic antimicrobial peptides ( CAMPs ) of the mammalian innate immune response [38] , [39] , [40] . The PhoPQ system is induced in low Mg2+ or low pH environments , or by exposure to CAMPs [41] , [42] , [43] . The Mg2+ concentration and pH of the flea digestive tract have not been defined , so the inducing stimulus is unknown , but CAMPs are induced and secreted into the gut by blood feeding insects when they take a blood meal containing bacteria [44] , [45] . X . cheopis fleas encode homologs of the insect CAMPs cecropin and defensin , and mount an inducible antibacterial response to infection ( unpublished data ) . Thus , the PhoPQ regulatory system may be induced by the flea's immune system in response to Y . pestis in the midgut . Despite the upregulation of phoP in the flea , with the notable exception of mgtC there was little correlation between predicted PhoP-regulated genes in vitro and genes upregulated in the flea [39] , [46] , [47] . Differential regulation of members of the PhoP regulon may occur depending on the inducing stimulus , however [48] . Soon after transmission , Y . pestis would be expected to encounter rapidly-responding phagocytic cells in the dermis . To assess the overall effect of the flea-specific phenotype on this encounter , we compared the interaction of Y . pestis recovered from infected fleas and from in vitro cultures with murine bone marrow macrophages . Bacteria from fleas showed significantly lower levels of phagocytosis ( Fig . 4A ) . We have previously reported analogous findings using human polymorphonuclear leukocytes ( PMNs ) [7] . The yit and yip genes in a Y . pestis locus ( y0181–0191 ) that encode predicted insecticidal-like toxins of the toxin complex ( Tc ) family and three linked phage-related genes were upregulated 4- to 50-fold in the flea midgut ( Tables 1 and S1 ) . We previously reported that the genes for these Tc-like proteins are highly expressed in fleas , but that their products are nontoxic to fleas [49] . yitR , a LysR-type regulator that activates the Tc-like yit genes [50] , was upregulated >10-fold in the flea , but its expression was not detected in the rat bubo ( Table 1 ) . The specific induction in the flea of yitR and genes in the adjacent Tc-like yit and yip loci suggests that they are involved in adaptation to and colonization of the flea . However , deletion of yitR or yitA-yipB ( y0183–y0191 ) does not affect the ability of Y . pestis KIM6+ to infect or block fleas ( data not shown ) . These observations , and the fact that the Yersinia Tc proteins have toxicity to certain eukaryotic cell lines in vitro [50] , [51] , prompted us to investigate a possible post-transmission antiphagocytic role for these proteins in the mammalian host . To determine if the insecticidal-like toxins were involved in resistance to phagocytosis , we repeated the macrophage experiments with a Y . pestis ΔyitR mutant , which as expected showed greatly reduced expression of the yit and yip genes in vitro and in the flea ( Fig . 4B ) . Loss of yitR significantly reduced the increased resistance to phagocytosis of Y . pestis isolated from infected fleas ( Fig . 4C ) . Since the yit and yip genes are not required for Y . pestis to produce a transmissible infection in fleas , it was possible to compare the virulence of wild-type and ΔyitR Y . pestis following transmission by fleabite . The incidence rate and time to disease onset were identical for both Y . pestis strains , demonstrating that expression of yit and yip is not essential for flea-borne transmission or disease ( data not shown ) . On average , the mice challenged with Y . pestis ΔyitR-infected fleas , both those that developed disease and those that did not , received a higher cumulative number of bites from blocked fleas than the mice challenged with Y . pestis-infected fleas , but this difference was not statistically significant ( Fig . 5 ) . However , it was not possible to detect any relatively minor difference in LD50 because the number of bacteria transmitted by a blocked flea varies widely [1] , [52] . Even a small decrease in LD50 provided by the Yit-Yip proteins would be significant at the ecological level in the maintenance of plague transmission cycles , because the transmission efficiency of blocked fleas is very low– often only a few or no bacterial cells are transmitted in an individual fleabite [52] . Because phoP is required by Y . pestis to produce a transmissible infection in fleas ( unpublished data ) , it was not possible to similarly assess the effect on disease transmission of phoP induction in the flea . When Y . pestis is transmitted into the dermis by an infected flea , it is immediately exposed to the mammalian innate immune system . The most important antiphagocytic virulence factors , the cytotoxic Yersinia outer proteins ( Yops ) , part of the T3SS encoded by the Y . pestis virulence plasmid and the F1 capsule encoded by the pMT1 plasmid , are not present at this initial stage of infection . Their expression is strictly temperature-regulated and are not produced in vivo until 3–5 hours after the temperature shift to 37°C that accompanies transmission [1] , [3] , [53] , [54] . Consequently , Y . pestis grown at <28°C in vitro are initially susceptible to in vivo uptake and killing by phagocytes until the Yop and F1 virulence factors are produced , effectively preventing further phagocytosis [53] , [54] . Our results indicate that Y . pestis entering the mammal from an infective flea is relatively resistant to macrophages , as well as PMNs [7]; a vector-specific phenotype that is not related to the T3SS or capsule . Coming from the flea , Y . pestis is also associated with the biofilm ECM , identical or closely related to the poly-β-1 , 6-N-acetyl glucosamine ECM of staphylococcal biofilms , which has been shown to provide protection from innate immune components [55] , [56] . In addition , although the antiphagocytic F1 capsule and Psa fimbriae do not appear to be produced in the flea , upregulation in the flea of most F1 genes in the cafRcaf1M1A1 locus and the Psa usher protein gene psaC ( Tables 1 , S1 ) suggests that components of the F1 and Psa translocation system are made , which may prime Y . pestis for rapid secretion of these extracellular virulence factors after transmission . The upregulation of the innate immunity resistance genes phoP and mgtC suggest that those Y . pestis that are phagocytized may be prepared for resistance to CAMPs and intracellular survival while still in the flea vector . Finally , the major essential virulence factors yadBC and pla , essential for Y . pestis dissemination from the dermis , were maximally or very highly expressed in the flea ( Tables 1 , S3 ) . Besides degrading plasminogen , the Pla protease may also inactivate CAMPs , particularly when the F1 capsule is not present [57] , which matches the phenotype of Y . pestis in the flea . In summary , Y . pestis appears to be prepared for pathogenesis in the mammal while still in the flea vector . The biofilm phenotype of Y . pestis and the virulence factors upregulated or highly expressed in the flea may enhance the earliest stages of plague pathogenesis while the full complement of temperature-shift-regulated virulence factors is still being induced . Increased resistance to innate immunity that is preinduced in the flea vector may be critical to productive transmission because blocked fleas transmit relatively few bacteria , often below the LD50 of Y . pestis grown in vitro at <28°C [1] , [52] .
All animals were handled in strict accordance with good animal practice as defined by NIH animal care and use policies and the Animal Welfare Act , USPHS; and all animal work was approved by the Rocky Mountain Laboratories Animal Care and Use Committee . Y . pestis KIM6+ , which lacks the 70-kb virulence plasmid that is not required for flea infection or blockage , was used for gene expression analyses . A KIM6+ strain with an in-frame deletion that eliminated amino acids 28–281 of the predicted 291 amino acid residue yitR ( y0181 ) gene product was produced by allelic exchange , using the pCVD442 suicide vector system [11] . This mutant was complemented by electroporation with a recombinant pWKS130 plasmid containing the wild-type yitR promoter and orf . The ΔyitR mutant was also transformed with pWKS130 alone to generate an empty vector control strain . For in vitro planktonic samples , bacteria were grown from frozen stocks in brain heart infusion ( BHI ) medium at 28°C , followed by two successive transfers in Luria Bertani broth supplemented with 100 mM MOPS , pH 7 . 4 ( LB/MOPS ) at 21°C . An inoculum of 104 cells/ml was added to 50 ml of LB/MOPS and incubated at 21°C with shaking at 250 rpm until exponential ( OD600 = 2 . 5 ) or stationary phase ( OD600 = 4 . 5 ) . Approximately 0 . 5 ml of the exponential phase culture and 0 . 25 ml of the stationary phase culture was resuspended in 1 ml and 0 . 5 ml , respectively , of RNAprotect bacterial reagent ( Qiagen; Valencia , CA ) , incubated for 5 min at room temperature , and centrifuged at 21°C for 5 min prior to RNA extraction . For in vitro biofilms , 400 µl of a 107/ml bacterial suspension was injected into a flowcell ( Stovall; Greensboro , NC ) that was connected to a reservoir of LB/MOPS at 21°C . Following a 30 min incubation period to allow the bacteria to adhere to the glass surface of the flow cell , LB/MOPS was pumped through the flow cell at a rate of 0 . 3 ml/min . After 48 hours , the flowcell was disconnected and the thick Y . pestis biofilm was harvested and treated with 0 . 5ml of RNAprotect similarly to the planktonic cultures . X . cheopis fleas were infected with Y . pestis KIM6+ by using a previously described artificial feeding system [3] . The infectious blood meal was prepared by growing Y . pestis KIM6+ overnight at 37°C in BHI medium , without aeration . A cell pellet containing 109 bacterial cells was resuspended in 1 ml PBS and added to 5 ml heparinized mouse blood . Fleas that took a blood meal were maintained at 21°C and 75% relative humidity , fed twice weekly on uninfected mice , and monitored for proventricular blockage as previously described [3] . On the day blockage was diagnosed , the digestive tract was dissected out and macerated in RNAprotect , a process that required about 1 min . Thirty midguts from blocked fleas were pooled for each of the two biological replicates . Midguts from 60 uninfected fleas were also collected as controls to assess background hybridization of flea RNA to the microarray . A flea-borne transmission model [58] was used to determine Y . pestis infectivity after challenge by flea bite . Fleas were infected with Y . pestis 195/P , a fully virulent wild-type strain , or with a Y . pestis 195/P ΔyitR mutant constructed as described above . Between 2–3 weeks after infection , the time required for Y . pestis to block fleas with a proventricular biofilm , groups of 20–40 fleas were applied to a restrained mouse and allowed to feed for 60 min . The fleas were then recovered and examined under a dissecting microscope to determine how many had taken a normal blood meal ( unblocked or non-infective fleas ) and how many were blocked ( infective fleas ) . After challenge , mice were monitored and euthanized upon the appearance of signs of terminal illness . Mice that did not develop any symptoms after one week following a challenge were re-challenged . A total of 9–10 BALB/cAnN and 10 RML Swiss-Webster mice were challenged with each strain . RNA was isolated from six independent samples from in vitro and flow cell cultures and two independent samples from pooled blocked fleas ( Fig . S1 ) using the RNeasy Mini Kit ( Qiagen ) . Flea-derived RNA samples were secondarily split into three technical replicates each . RNA integrity was verified on a Bioanalyzer 2100 ( Agilent Technologies; Santa Clara , CA ) . Total RNA ( 100 ng ) was amplified and labeled with modified biotin-11-CTP ( Perkin Elmer; Waltham , MA ) and biotin-16-UTP ( Roche Molecular Biochemicals , Pleasanton , CA ) by using the Message-Amp II-Bacteria amplified antisense RNA ( aRNA ) kit ( Ambion; Austin , TX ) . Amplified RNA was then fragmented using Ambion's Fragmentation reagent ( Applied Biosystems ) , hybridized to the RML custom Affymetrix GeneChip that contains sequences for all Y . pestis predicted ORFs , and scanned . The amplification step did not affect the relative transcript signals obtained by microarray ( data not shown ) . Affymetrix GeneChip Operating Software ( GCOS v1 . 4 , GEO platform GPL2129 , http://www . affymetrix . com ) was used for initial analysis of the microarray data at the probe-set level . All * . cel files , representing individual biological replicates , were scaled to a trimmed mean of 500 using a scale mask consisting of only the Yersinia pestis KIM6+ probe-sets to produce the * . chp files . A pivot table with all samples was created including calls , call p-value and signal intensities for each gene . The pivot table was then imported into GeneSpring GX 7 . 3 ( http://www . chem . agilent . com ) , where hierarchical clustering ( condition tree ) using a Pearson correlation similarity measure with average linkage was used to produce the dendrogram indicating that biological replicates grouped together . The pivot table was also imported into Partek Genomics Suite software ( Partek Inc . ; St . Louis , MO ) to produce a principal components analysis ( PCA ) plot as a second statistical test for the grouping of biological replicates . ANOVA was run from this data set to produce a false discovery rate report producing false positive reduced p-values for each comparison of interest . The correlated replicates of all test conditions and controls were combined , and quality filters based upon combined calls and signal intensities were used to further evaluate individual gene comparisons . Present and marginal calls were treated as the same whereas absent calls were negatively weighted and eliminated from calculations . Ratios of test/control values and associated t-test and ANOVA p-values values of all individual genes passing the above filters were determined using GeneSpring , SAM , and Partek software . The microarray data have been deposited in the NCBI GEO public database ( accession number GSE16493 ) . To compare differential in vivo gene expression patterns in the flea and the rat , the average hybridization signal for each individual Y . pestis gene was divided by the average signal of all 4 , 683 genes on the microarray for both the flea microarray ( this study ) and the rat bubo microarray [11] data sets . Gene by gene comparisons of these normalized expression data sets were used for Fig . 3 and Tables 1 , S5 , and S6 ) . Murine bone marrow-derived macrophages were prepared as described [59] , [60] and cultured in Dulbecco's Modified Eagles medium ( DMEM ) supplemented with 5 mM L-glutamine , 25 mM HEPES , 10% heat-inactivated fetal bovine serum , 5 mM non-essential amino acids , and 10 ng/ml CSF-1 ( PeproTech; Rocky Hills , NJ ) . 1-ml suspensions of Y . pestis KIM6+ containing pAcGFP1 ( Clontech; Mountain View , CA ) from 21°C stationary phase LB/MOPS cultures , or from triturated midguts dissected from fleas 2 to 3 weeks after infection were treated for 15 sec in a FastPrep FP120 using lysing matrix H ( Qbiogene; Carlsbad , CA ) to disrupt bacterial aggregates , quantified by Petroff-Hausser direct count , and diluted in DMEM to ∼1×106 bacteria/ml . 0 . 1 ml of bacterial suspension was added to tissue culture plate wells containing ∼1×105 differentiated primary macrophages cultured on 12 mm glass coverslips in 1 ml DMEM . The plates were not centrifuged after addition of the bacteria , and midgut triturate from an equivalent number of uninfected fleas was added to the in vitro-derived bacterial suspensions used for these experiments . After 1 h incubation at 37°C and 5% CO2 , the medium was removed and the cells washed , fixed in 2 . 5% paraformaldehyde for 10 min at 37°C , and then rewashed . Extracellular bacteria were labelled by indirect immunofluorescence as described [60] using a 1∶50 , 000 dilution of hyperimmune rabbit anti-Y . pestis polyclonal antibody [7] and a 1∶400 dilution of AlexaFluor 568-conjugated goat anti-rabbit antibody ( Invitrogen; Carlsbad , CA ) . The percentage of extracellular bacteria was determined by dividing the number of red-fluorescent bacteria by the total number ( red- and green only-fluorescent ) bacteria associated with individual macrophages . To calculate differential resistance to phagocytosis for a given strain , the average percent extracellular LB-grown bacteria was subtracted from the average percent extracellular flea-derived bacteria . Results from 2–3 independent experiments performed in triplicate were analyzed by unpaired two-tailed t-test . Independent RNA samples were prepared from blocked fleas and in vitro biofilm and planktonic cultures as described for the microarray experiments , except that the RNA was not amplified . Samples were treated with rDnase I ( Ambion ) and confirmed by PCR to be free of genomic DNA contamination . cDNA was synthesized from the RNA and used for quantitative PCR on an ABI Prism 7900 sequence detection system ( Taqman , Applied Biosystems ) . The reactions contained oligonucleotide primers and probes designed using Primer Express version 2 . 0 software ( Applied Biosystems ) and the Taqman Universal PCR Master Mix ( Applied Biosystems ) . For each primer-probe set assay , a standard curve was prepared using known concentrations of Y . pestis KIM6+ genomic DNA and used to transform CT values into relative DNA quantity . The quantity of cDNA for each experimental gene was normalized relative to the quantity of the reference gene crr ( y1485 ) , and the ratio of the normalized quantity of each gene in the flea samples to the normalized quantity in the in vitro samples was calculated ( Fig . S2 ) . Primer and probe sets used are listed in Table S7 . | Bubonic plague cycles depend on the ability of Yersinia pestis to alternately infect two very different hosts—a mammal and a flea . Like any arthropod-borne pathogen , Y . pestis must sense host-specific environmental cues and regulate gene expression accordingly to produce a transmissible infection in the flea after being taken up in a blood meal , and again when it exits the flea and enters the mammal . We examined the Y . pestis phenotype at the point of transmission by in vivo gene expression analyses , the first description of the transcriptome of an arthropod-borne bacterium in its vector . In addition to genes associated with physiological adaptation to the flea gut , several Y . pestis virulence factors required for resistance to innate immunity and dissemination in the mammal were induced in the flea , suggesting that the arthropod life stage primes Y . pestis for successful infection of the mammal . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/functional",
"genomics",
"microbiology/microbial",
"evolution",
"and",
"genomics",
"microbiology/innate",
"immunity",
"genetics",
"and",
"genomics",
"evolutionary",
"biology",
"microbiology/microbial",
"physiology",
"and",
"metabolism",
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis"
]
| 2010 | Transit through the Flea Vector Induces a Pretransmission Innate Immunity Resistance Phenotype in Yersinia pestis |
Mutation rate varies greatly between nucleotide sites of the human genome and depends both on the global genomic location and the local sequence context of a site . In particular , CpG context elevates the mutation rate by an order of magnitude . Mutations also vary widely in their effect on the molecular function , phenotype , and fitness . Independence of the probability of occurrence of a new mutation's effect has been a fundamental premise in genetics . However , highly mutable contexts may be preserved by negative selection at important sites but destroyed by mutation at sites under no selection . Thus , there may be a positive correlation between the rate of mutations at a nucleotide site and the magnitude of their effect on fitness . We studied the impact of CpG context on the rate of human–chimpanzee divergence and on intrahuman nucleotide diversity at non-synonymous coding sites . We compared nucleotides that occupy identical positions within codons of identical amino acids and only differ by being within versus outside CpG context . Nucleotides within CpG context are under a stronger negative selection , as revealed by their lower , proportionally to the mutation rate , rate of evolution and nucleotide diversity . In particular , the probability of fixation of a non-synonymous transition at a CpG site is two times lower than at a CpG site . Thus , sites with different mutation rates are not necessarily selectively equivalent . This suggests that the mutation rate may complement sequence conservation as a characteristic predictive of functional importance of nucleotide sites .
The functional and phenotypic effects of mutations and , consequently , the strength of negative selection vary widely among nucleotide sites in any genome . At the opposite ends of the continuum , mutations at some sites are effectively neutral , while mutations at some other sites are lethal . Nucleotide sites can be subdivided , according to their molecular function , into classes with different typical strengths of negative selection . Generally , rapidly evolving segments of intergenic regions and introns , as well as most of synonymous coding sites , are controlled by only weak selection or even by no selection at all . Slowly evolving segments of intergenic regions and introns , as well as UTRs and non-synonymous coding sites , are under much stronger selection ( e . g . , [1]–[8] ) . However , even within such functional classes , the strength of negative selection varies widely among individual sites ( e . g . , [9]–[12] ) . The rate of spontaneous mutation is also not uniform across individual sites [13]–[15] . The standard deviation of the mutation rate at a site may be comparable to its mean . Moreover , some rare hot-spot sites may mutate much more frequently than an average site . Thus , the mutation rate at a site depends both on its local sequence context ( e . g . , [16]–[19] ) and on its global location within the genome [13]–[15] , although these dependencies are rather different in different groups of organisms [19] , [20] . In particular , in mammals the 5′CpG3′ context substantially increases the rate of transversions , and especially transitions [16]–[19] , [21] . Mutation and selection are generally thought to be independent evolutionary forces [22] . In other words , the rate with which a mutation occurs is routinely assumed to be independent of the effect of this mutation on fitness . Inferences of the strength of selection on specific genes and sites within genes usually rely on this assumption . Although selection for reduced mutability is stronger at sites where mutations are more deleterious [23] , it is hard to imagine adaptive fine-tuning of mutation rates at the level of individual nucleotide sites . Thus , one might expect selective constraint and mutability to vary more or less independently across individual sites . However , another phenomenon may lead to a seemingly counterintuitive association between stronger negative selection and higher mutation rates . Sites that are under weak or no selection are free to evolve and to get rid of hypermutable contexts . In contrast , negative selection will preserve such contexts at functionally important sites , provided that they confer a higher fitness . In particular , non-synonymous [24] and even synonymous [21] , [25] coding sites of mammalian genomes are enriched , relative to what is expected at a neutral mutational equilibrium , by CpG contexts , leading to a substantially higher mutation rate within coding exons than within introns . Here we consider human non-synonymous coding sites and subdivide them into just two classes – those within and those outside CpG contexts , because in mammals this context exerts by far the strongest influence on the mutation rate [19] . Then , we compare the rates of human-chimpanzee divergence [26] and the levels of intrahuman polymorphism at coding sites that are within vs . outside CpG context . We have found that the strength of negative selection acting at non-synonymous coding sites is substantially higher within hypermutable CpG contexts .
It is well known that in mammals CpG context substantially increases the mutation rate; however , the exact magnitude of this effect has not been established with certainty . We used three sources of information on the impact of CpG context on the rates of transitions and transversions: 1 ) direct data on Mendelian diseases in humans [18] , 2 ) Bayesian Markov Chain Monte Carlo analysis of evolution of several species of mammals [19] , and 3 ) parsimony-based analysis of human-chimpanzee-orangutan genome alignments ( Table 1 ) . The third analysis must underestimate the impact of CpG context on transversion and especially transition rates , because two nucleotide substitutions , one on the edge leading to a sister species ( human or chimpanzee ) and the other on the edge leading to the outgroup ( orangutan ) , can happen within a CpG context . Such occurrences will lead to underestimation of the fraction of sites that were within CpG context in the common ancestor of human and chimpanzee and , thus , of the fraction of allele substitutions that destroy a CpG context . Indeed , this underestimation is evident from Table 1 . Thus , below we will use the mean values of the first two estimates and will assume that in humans CpG context increases the rate of transitions by the factor of 14 . 5 , and the rate of transversions by the factor of 3 . 5 . We used human-chimpanzee-orangutan alignments of coding sequences to compare the rates of a particular nucleotide substitution that causes a particular amino acid replacement within vs . outside CpG context ( CpG vs . ⌝CpG ) . For example , a P→L replacement , caused by a C→T transition , can occur within ( CCG→CTG; the site of substitution is boldfaced ) or outside ( e . g . , CCC→CTC ) CpG context . The common ancestor of humans and chimpanzees , as revealed by the orangutan outgroup , carried , at all the loci we studied , TargetP→L CpG = 18 , 088 of CCG codons , and TargetP→L⌝CpG = 185 , 826 of CCA , CCT , or CCC codons ( Table 2 ) . There were 215 and 284 P→L replacements ( ReplacementsP→L CpG and ReplacementsP→L⌝CpG ) , caused by C→T transitions , within CpG and outside CpG contexts , respectively . Thus the impact of CpG context on the rate of P→L replacements in the course of human-chimpanzee divergence is ( 1 ) This analysis relies on the identification of the human-chimpanzee ancestral state using orangutan as outgroup . To test whether possible erroneous identifications affect our results , we repeated the same analysis using the macaque outgroup , which must lead to more errors , because macaque is about three times more distant from the human-chimpanzee last common ancestor than orangutan . Also , all the analyses were performed only for human and chimpanzee coding sequences , under the assumption that the proportion of CpG context within these sequences is at equilibrium . Estimates of the impact of CpG context on the rates of evolution obtained in this way were only slightly higher than estimates obtained using the orangutan outgroup ( data not reported ) . For intraspecies nucleotide diversity , the number of SNPs that involve a particular amino acid change within and outside CpG context were used in equation ( 1 ) , instead of the corresponding numbers of substitutions ( Table 2 ) . The direction of an amino acid change associated with a particular SNP was determined by the orthologous chimpanzee sequence . We assumed that the ratio of CpG vs . ⌝CpG target sizes for a particular amino acid replacement was the same as for human-chimpanzee divergence . Indeed , the SNPs we used were obtained by resequencing of ∼11 , 000 human loci [27] so that we can expect the nucleotide composition of this sample to be close to that of all protein-coding loci . The data on the impacts of CpG context on human-chimpanzee divergence and on intrahuman diversity are shown in Table 2 and in Figure 1 . Thus , the impact of CpG context on the rate of divergence , i . e . the average ratio of the rates of divergence within vs . outsides CpG contexts , was 7 . 1 for transitions and 2 . 5 for transversions . The average ratio of values of intrahuman diversities for non-synonymous SNPs within vs . outsides CpG contexts was 11 . 2 for transitions and 2 . 4 for transversions ( Table 3 ) . If macaque instead of orangutan is used as an outgroup , the observed impacts of CpG context on the rates of divergence decline only slightly ( 6 . 8 instead of 7 . 1 for transitions , and 2 . 1 instead of 2 . 5 for transversions ) . We applied several tests to evaluate the significance of the difference of the impact of CpG context on non-synonymous divergence and diversity . This difference is insignificant for transversions and highly significant for transitions , according to the χ2 test ( p = 2 . 8·10−16 ) . However , the χ2 test does not stratify data according to amino acid replacements , which is essential in our case . We used two approaches to perform stratified analysis of contingency tables . First , we combined p-values of separate tests for each amino acid replacement , using Stouffer ( p<2 . 2·10−16 ) and Fisher ( p = 2 . 7·10−16 ) methods . We also applied Cochran-Mantel-Haenszel test , the standard test for stratified analysis of contingency tables ( p = 4 . 6·10−16 ) . We measured the impacts of CpG context on rates of evolution and nucleotide diversity at synonymous coding and at non-coding sites ( Table 3 ) . As it was the case for non-synonymous sites , we assumed parsimony . Thus , the data on rates of evolution at non-coding sites shown in Table 3 are taken from ( ( human-chimpanzee ) -orangutan ) comparison shown in Table 1 . We can see that the impacts of CpG context on non-coding human-chimpanzee divergence and intrahuman nucleotide diversity are rather close to the corresponding impacts on the mutation rate , which is consistent with effective neutrality of most of the non-coding DNA in humans . The figures in Table 3 are likely to be slightly underestimated , due to substitutions in the outgroup lineage . In contrast to non-coding sites , at synonymous sites the impacts of CpG context on human-chimpanzee divergence and intrahuman nucleotide diversity due to transitions , but not to transversions , are substantially lower than the corresponding impacts on the mutation rates , although still higher than the corresponding impacts at non-synonymous sites . This implies that some selection acts on synonymous transitions within CpG context , and that this selection is weaker than the corresponding selection at non-synonymous sites . Several analyses revealed weak selection favoring Cs and Gs at synonymous sites [25] , [28] .
Our results show that negative selection is stronger within CpG contexts than in less mutable sites at identical codon positions . We can see that the per nucleotide site rate of transitions , accepted in the course of human-chimpanzee divergence , is on average 7 . 1 times higher within CpG contexts than outside CpG contexts ( Table 3 ) . A comparison of this figure with the impact of CpG on the corresponding mutation rate ( Table 1 ) suggest that a transition that occurred within CpG context gets fixed in the course of human-chimpanzee divergence with a probability of 7 . 1/14 . 5 = 0 . 49 of the probability of fixation of a transition that occurred outside CpG context . Thus , nucleotides within CpG context are protected by a stronger selection . In the case of SNPs , we observed a similar but weaker effect . On average , non-synonymous SNPs caused by transitions are 11 . 2 times more common within CpG context than outside of it . Thus , a non-synonymous transition mutation that occurred within CpG context is observed as a SNP with a chance that constitutes only 11 . 2/14 . 5 = 0 . 77 of the chance of observing a transition that caused the same amino acid replacement but occurred outside CpG context . In other words , in the case of transitions , CpG context increases the level of intrahuman diversity and in particular the rate of non-synonymous divergence less than proportionally to its impact on the mutation rate . This demonstrates that negative selection at non-synonymous sites within CpG context is stronger than at sites outside it . This seemingly counterintuitive pattern probably has a simple evolutionary explanation: nucleotide sites that are not under strong negative selection will eventually lose most of their hypermutable CpG contexts . Thus , hypermutable contexts must be disproportionally common at sites under strong negative selection . It is not surprising that a stronger negative selection within CpG contexts affects the rates of evolution more than it affects intraspecies diversity . Indeed , a substantial fraction of SNPs that segregate within a population are nevertheless subject to negative selection that is strong enough to prevent their fixation [22] . The large difference between the impacts of CpG context on polymorphism and divergence suggests that the observed effect is mostly due to nucleotide sites under weak selection , which affects divergence more than polymorphism . Such sites are abundant in human protein coding genes [9]–[11] , [29] . Predictably , the impacts of CpG context at mostly selectively neutral noncoding sites do not differ substantially from its impacts on the mutation rate . In contrast , coding synonymous sites within CpG contexts evolve slower and are less diverse within humans than what would be expected on the basis of the mutation rates alone . This is not surprising because the impact of CpG context must be sensitive to even weak selection [25] , [28] . Indeed , CpG contexts are greatly underrepresented at purely neutral sites , but even a rather weak selection is expected to increase their prevalence substantially , as long as the coefficient of selection is of the order of the reciprocal of the effective population size or higher [22] . CpG contexts are much more common within synonymous sites than within non-coding sites [25] . CpG context exerts a much weaker influence on the rate of transversions than on the rate of transitions ( see Table 1 ) . Thus , it is not surprising that the effects , which we can easily observe in the case of transitions , are not visible in the case of transversions . More data are needed to determine if these effects , however weak , are still present in the case of transversions . Our estimates of the impact of CpG context on divergence ( Tables 2 and 3 ) are probably too low due to substitutions in the outgroup lineage . However , these estimates depend only slightly on whether orangutan or macaque is used as an outgroup , although in the second case the prevalence of multiple substitutions at a site should be much higher . Also , the estimates computed from only human and chimpanzee genomes assuming equilibrium of the CpG content are only slightly higher than the estimate obtained using an outgroup . Further , the estimate of the impact of CpG context on human-chimpanzee divergence due to transitions at non-synonymous sites is much lower than the corresponding estimate for non-coding sites computed using the same outgroup ( Table 3 ) . This indicates that the low impact of CpG contexts not just an artifact of the assumption of parsimony . Even under the impossible assumption that every site that is located within CpG context in either human or chimpanzee sequence was also located within CpG context in their last common ancestor , the resulting estimate of the impact of this context on the rate of divergence equals 12 and is still lower than CpG impact on raw mutation rate . The analysis of intrahuman diversity relies on the chimpanzee sequence for determining the identity of ancestral alleles . Misidentification of ancestral alleles would result in an underestimation of the impact of CpG context because ancestral CpGs would preferentially evolve in the chimpanzee lineage . To evaluate a possible extent of this bias we repeated the analysis using major and minor alleles instead of inferred ancestral and derived alleles . The resulting estimate of the impact of CpG context on non-synonymous transitions is 11 . 5 , which is only slightly higher than 11 . 2 ( Table 2 ) . Negative selection can also be detected in polymorphism data independently of intraspecies nucleotide diversity through changes in the distribution of allele frequencies , because such selection causes an excess of low-frequency alleles . In particular , minor allele frequencies of non-synonymous SNPs that affect slowly evolving ( conserved ) protein sites are reduced [30] , [31] . The excess of rare alleles was not statistically significant in the two datasets of human SNPs used in this study . The effect of weak negative selection on allele frequency distribution is expected to be much smaller than on divergence and data on rare SNPs in protein coding regions are sparse . Thus , the analysis of allele frequency distribution may lack statistical power . Our analysis suggests that mutation rates can be used in computational methods to predict which amino acid replacements are deleterious [32]: a replacement that occurred at a highly mutable site is more likely to be deleterious . Currently , prediction methods rely on the properties of an encoded amino acid sequence , its conservation between species , and the properties of the corresponding protein . Our analysis suggests that taking the DNA-level features of an amino acid replacement into account will increase the accuracy of prediction of its effect on protein function .
To determine the impact of CpG context on mutation rates we constructed a human-chimpanzee-orangutan alignment for a ∼1 Mb piece of orangutan genomic sequence ( gi:119380173 ) , and analyzed it assuming parsimony . To study the impact of CpG context on the rate of evolution , we constructed human-chimpanzee-orangutan and human-chimpanzee-macaque alignments of coding regions of individual genes by finding the orthologous macaque gene for each UCSC human-chimpanzee pair with the by-directional best BLAST hits approach [33] . We also repeated the analysis on just two sequences assuming equilibrium CpG content ( data not shown ) . This analysis resulted in similar estimates . For the analysis of intrahuman diversity we used a comprehensive and systematic Applera dataset [27] . Chimpanzee nucleotides corresponding to human SNP positions were identified using the SNP UCSC genome track [34] . Applera set is gene centric . Therefore , for the analysis of non-coding diversity , we used randomly ascertained SNPs from the Perlegen set [35] . We also verified that coding SNPs from the Perlegen dataset produced estimates highly similar to those based on the Applera dataset . We analyzed each population separately and excluded SNPs , which were fixed in the population and could not be mapped to chimpanzee nucleotides ( ≈4 . 6% ) . Statistical analysis was carried out using R statistical package v2 . 7 . 0 [36] . p-Values for individual amino acid residue contingency tables were computed by Monte Carlo simulations with the number of replicates B = 106 . To obtain combined p-values we used Stouffer's z-scores [37] and Fisher's sum of logs of p [38] methods . Cochran-Mantel-Haenszel test of conditional independence [39] was utilized to ensure there was no three-way interaction with the amino acid residue type . | Mutations occur in some sites in the genome more frequently than in others . Similarly , mutations in some sites have greater consequences than in others . The effect of mutations might not be independent of the frequency with which mutations occur . Indeed , sites where mutations happen frequently will be preserved if the effects of these mutations are severe or will otherwise be allowed to mutate if there are no consequences for the organism . We compared both human–chimpanzee differences and sequence variation among humans in protein coding genes . We found that highly mutable nucleotide sites , such as the dinucleotide CpG , are on average more important and more frequently preserved by natural selection . Using this information , together with other features such as sequence conservation , opens a new perspective to predict the effect of human mutations , including their potential involvement in diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"computational",
"biology/population",
"genetics",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics",
"computational",
"biology/comparative",
"sequence",
"analysis",
"computational",
"biology/molecular",
"genetics",
"evolutionary",
"biology/genomics",
"computational",
"biology/genomics",
"genetics",
"and",
"genomics/bioinformatics",
"genetics",
"and",
"genomics/population",
"genetics"
]
| 2008 | Hypermutable Non-Synonymous Sites Are under Stronger Negative Selection |
Infections with high-risk human papillomaviruses ( HPVs ) are causally involved in the development of anogenital cancer . HPVs apparently evade the innate immune response of their host cells by dysregulating immunomodulatory factors such as cytokines and chemokines , thereby creating a microenvironment that favors malignancy . One central key player in the immune surveillance interactome is interleukin-1 beta ( IL-1β ) which not only mediates inflammation , but also links innate and adaptive immunity . Because of its pleiotropic physiological effects , IL-1β production is tightly controlled on transcriptional , post-translational and secretory levels . Here , we describe a novel mechanism how the high-risk HPV16 E6 oncoprotein abrogates IL-1β processing and secretion in a NALP3 inflammasome-independent manner . We analyzed IL-1β regulation in immortalized keratinocytes that harbor the HPV16 E6 and/or E7 oncogenes as well as HPV-positive cervical tumor cells . While in primary and in E7-immortalized human keratinocytes the secretion of IL-1β was highly inducible upon inflammasome activation , E6-positive cells did not respond . Western blot analyses revealed a strong reduction of basal intracellular levels of pro-IL-1β that was independent of dysregulation of the NALP3 inflammasome , autophagy or lysosomal activity . Instead , we demonstrate that pro-IL-1β is degraded in a proteasome-dependent manner in E6-positive cells which is mediated via the ubiquitin ligase E6-AP and p53 . Conversely , in E6- and E6/E7-immortalized cells pro-IL-1β levels were restored by siRNA knock-down of E6-AP and simultaneous recovery of functional p53 . In the context of HPV-induced carcinogenesis , these data suggest a novel post-translational mechanism of pro-IL-1β regulation which ultimately inhibits the secretion of IL-1β in virus-infected keratinocytes . The clinical relevance of our results was further confirmed in HPV-positive tissue samples , where a gradual decrease of IL-1β towards cervical cancer could be discerned . Hence , attenuation of IL-1β by the HPV16 E6 oncoprotein in immortalized cells is apparently a crucial step in viral immune evasion and initiation of malignancy .
High-risk human papillomaviruses ( HPVs ) are causally responsible for anogenital cancer , both in women and men [1] , [2] . While in the latter , penile and anal carcinomas are relatively rare , HPV infection is also linked in both genders to more than 50% of all oropharyngeal squamous cell carcinomas [1] , [3] , [4] . The transforming potential of these viruses is mediated by the E6 and E7 oncoproteins that are responsible for sustaining a proliferative phenotype mainly by promoting degradation of the cellular tumor suppressor proteins p53 and pRb , respectively [1] , [2] . During the last years , however , it became evident that viral oncoproteins not only affect cell cycle regulatory mechanisms and apoptosis , but also have a negative impact on the innate immune response of their host and in turn on the respective premalignant microenvironment where unscheduled growth of persistently infected cells is finally taking place [5] , [6] . Monitoring chemotactic and pro-inflammatory genes in a top-down approach , genome-wide transcriptome analyses and the subsequent in silico topological reconstruction of the cellular immune network shows that high-risk HPVs always target highly interconnected nodes of the antiviral defense interactome , leading either to cell lysis and virus spread , to viral persistence or ultimately to malignant transformation [5]–[9] . In other words , considering virus-host interactions as a result of a long-lasting evolutionary selection process , HPVs have developed sophisticated strategies to circumvent innate immunity long before the adaptive immune response is activated [10]–[12] . Concerning our understanding about the role of the individual oncoproteins and their cross-talk with the host cell interactome , E6 and E7 either directly or indirectly interfere with innate immunosurveillance [5] , [6] , [11] . For instance , the high-risk HPV E6 oncoprotein both inactivates type I interferon ( IFN ) signaling ( e . g . keratinocyte-specific IFN-κ ) [13] and downstream pathways such as chemokine expression which is required to attract and activate specific subsets of effector leukocytes , cells from the monocyte/macrophage lineage as well as natural killer cells [14] . E7 expression , on the other hand , can inhibit the function and nuclear translocation of p48 ( ISGF3γ ) , one component of the IFN-stimulated gene factor 3 ( ISGF3 ) trimeric complex formed between p48 , STAT1 and STAT2 . If p48 is missing , the transcription of IFN-regulated genes via its binding to cognate IFN-stimulated response elements is also abrogated [15] . Another key player that is located in the center of the antiviral and pro-inflammatory network is interleukin-1β ( IL-1β ) [16] . This cytokine is a potent activator of immune responses directed against viral and bacterial infections [17] . It mediates the migration of leukocytes , induces fever and promotes the activation and polarization of T cells [18] . While the precursor protein of IL-1α ( pro-IL-1α ) is cleaved by calpain , pro-IL-1β is processed by caspase-1 to its biologically active secreted form [18] , [19] . The activating caspase-1 in turn is a component of a multi-protein complex , called NALP3 [NACHT- , leucine-rich repeat ( LRR ) -and PYD-containing protein 3] inflammasome that consists of the NOD-like receptor protein NALP3 , the adaptor protein ASC ( apoptosis-associated speck-like protein containing a caspase recruitment domain ) and pro-caspase-1 [20] . NALP3 inflammasome assembly and the subsequent autoproteolytic cleavage of pro-caspase-1 is not only activated upon viral or bacterial infection [20]–[22] , but also by numerous exogenous and endogenous danger signals such as extracellular ATP ( a known inducer of potassium efflux via the P2X7 receptor [23] ) , hyaluronan [24] , amyloid-β fibrils [25] and intracellular reactive oxygen species ( ROS ) [26] . While IL-1α is mainly found to be membrane-associated , IL-1β is active in an auto-/paracrine manner when secreted [16] , [18] . Mature IL-1β and IL-1α share the same IL-1 receptor which is constitutively expressed on many cell types . Upon binding , NF-κB and the mitogen-activated protein ( MAP ) kinases p38 and JNK are activated to induce a wide range of genes involved in the pro-inflammatory response [16] , [18] , [27] . Abnormal IL-1β release or lack of IL-1β expression changes the conditions towards a pathological microenvironment , resulting either in chronic inflammation or in the absence of a proper immune surveillance against infections , respectively [16] , [18] , [19] , [28] . Apart from immunological effector cells such as neutrophils , macrophages and dendritic cells , also keratinocytes are a potent source of IL-1β [29] , [30] . These cells are important sensors of pathogens and danger signals that mediate immune responses , supporting the notion that keratinocytes , which are the main target of HPV infection , also have to be considered as central nonprofessional immune competent cells of the mucosa [10] , [31] . Although IL-1β was recently identified to be targeted by high-risk HPV as a central hub within the network of innate immunity [5] and down-regulated in cervical tumors [32]–[34] , our knowledge about its function and regulation in the context of HPV-induced carcinogenesis is still rudimentary . To get insights on how this central cytokine is regulated , we monitored the processes that modulate the NALP3 inflammasome complex and in turn IL-1β secretion using primary and HPV16-immortalized human keratinocytes as well as HPV-positive cervical carcinoma cells as experimental model system . Although the IL1B gene is still transcribed in non-malignant cells , the protein levels of pro-IL-1β as well as the secretion of mature IL-1β is strongly impaired in E6- and E6/E7- immortalized keratinocytes . Excluding different pathways regulating protein degradation ( e . g . via lysosomes or autophagy ) , finally we show that the E3 ubiquitin ligase E6-AP ( = E6-associated protein ) and p53 are controlling proteasomal degradation of pro-IL-1β in high-risk HPV-positive cells . We conclude that an inflammasome-independent , but proteasome-mediated degradation of pro-IL-1β represents an early viral immune evasion mechanism occurring during high-risk HPV immortalization , ultimately culminating in the complete silencing of the gene itself in malignant cells . Additionally , transcriptional silencing of the IL1B gene also seems to be a favored mode of escape in hyperproliferative cells infected by HPV6 and HPV11 , where post-translational degradation of IL-1β was much less efficient . Since a gradual absence of IL-1β was also found in cervical tissue sections , inactivation of IL-1β signaling apparently inhibits its central role in the balance between inflammation and antiviral immunity against an HPV infection , thereby significantly contributing to the development of cervical cancer .
In order to analyze the influence of high-risk HPV on the capacity to express and secrete mature IL-1β in the context of different transformation stages of an HPV infection , we used primary human keratinocytes ( PK ) , immortalized PK encoding individual oncoproteins of HPV16 ( immE6 , immE7 , and immE6/E7 ) as well as HPV16/18-positive cervical tumor cells ( CaSki , SiHa and HeLa ) . The cells were incubated with recombinant GFP-expressing adenoviruses , which are known to induce strong IL-1β secretion via activation of the NALP3 inflammasome [21] , [35]–[37] . As shown in Fig . 1A , both PK and immE7 responded effectively by releasing mature IL-1β after a one hour exposure to adenoviral infection . ImmE6 and immE6/E7 cells only released small amounts of IL-1β , while cervical carcinoma cell lines completely failed to secrete IL-1β . The deficiency in mature IL-1β release in these cells was not due to reduced or missing adenovirus infectivity , since similar fluorescence levels of GFP could be detected in all cell lines 24 h post infection ( Fig . S1A ) . Additionally , to test the biological activity of the secreted IL-1β in functional terms , human umbilical vein endothelial cells ( HUVEC ) were incubated for 12 hours with conditioned medium obtained from adenovirus-infected and non-infected cells . Here , only supernatants from infected immE7 keratinocytes contained biologically active IL-1β as demonstrated by substantial induction of IL-1β-responsive genes such as IL-6 and CCL-20 in HUVEC cells used as indicator cell line ( Fig . S1B ) [38] , [39] . To evaluate whether the lack of IL-1β secretion was the result of a diminished synthesis of the protein itself , the intracellular levels of IL-1β were monitored by ELISA . Here , a strong reduction of IL-1β in immE6 and immE6/E7 cells and its complete absence in cervical carcinoma cell lines could be discerned ( Fig . 1B ) . Since the ELISA did not discriminate between pro-IL-1β and its mature form , Western blot analysis was performed . As depicted in Fig . 1C , only PK and immE7 cells constitutively expressed substantial amounts of the 31 kDa pro-IL-1β form , while it was reduced in E6-positive immortalized keratinocytes or even absent in the tumor cell lines . When compared to primary human keratinocytes , parallel performed transcriptional analysis in immortalized cells by quantitative RT-PCR ( q-PCR ) even revealed an approximately two to four fold up-regulation of the basal pro-IL-1β mRNA levels ( see Discussion ) , while in the respective malignant cells CaSki , SiHa and HeLa ( Fig . 1D ) as well as in the HPV-negative cervical cancer cell line C-33 A ( Fig . S1C ) , transcription was either strongly reduced or absent . Hence , when tumor progression occurs , cervical cancers apparently use an additional strategy to inhibit the inflammatory response by diminishing IL-1β expression at the transcriptional level via gene silencing . Moreover , while still transcribed in immortalized keratinocytes , the same transcriptional down-regulation was also detected in cervical carcinoma cells when other cytokines such as IL-1α , IL-18 and IL-33 were monitored by q-PCR ( Fig . S1D ) . NALP3 inflammasome assembly , caspase-1 activation and subsequent IL-1β release can be triggered by a plethora of environmental insults , including viral infections [20]–[22] . Hence , a possibility that may explain the divergence in IL-1β levels and its release ( Fig . 1A and B ) could be a dysfunction in caspase-1 activity . Indeed , there are several reports that show that certain viruses , such as orthopoxviruses and influenza virus can interfere with signaling downstream of the inflammasome by modulating caspase-1 function [27] . Consistent with this notion is that caspase-1 knock-out mice have a higher susceptibility to various infections [17] . We therefore first examined the transcriptional levels of caspase-1 in PK and immortalized keratinocytes ( Fig . 2A ) , where no substantial variances could be detected . Monitoring the basal activity of caspase-1 in PK and immortalized cells using the specific caspase-1 substrate R110-YVAD [40] , again no significant alterations could be observed , although the activity was slightly higher in immortalized cells when compared to PK ( Fig . 2B ) . On the other hand , since there exist naturally occurring genetic variants of human caspase-1 that can differ in their ability to activate IL-1β [41] , we monitored whether inhibition of caspase-1 led to an increase of the intracellular levels of pro-IL-1β in immE6 and immE6/E7 cells . However , as depicted in Fig . 2C , this was not the case . Hence , these data suggest that the reduced levels of pro-IL-1β observed in immE6 and immE6/E7 cells are not the result of caspase-1 dysregulation . Autophagy is considered a master regulator of cellular homeostasis which is critical for survival under nutrient deprivation [42] , [43] . Moreover , autophagy induction is able to diminish inflammation by targeting inflammasome components and pro-IL-1β for autophagosome/lysosomal degradation [20] , [44]–[47] . Conversely , inhibition of autophagy potentiates inflammasome activity and prevents autophagosome/lysosome-mediated pro-IL-1β degradation [44]–[47] . A specific marker for autophagosome formation is the cytoplasmic microtubule-associated protein 1 light chain 3 ( LC3-I ) which is coupled to phosphatidylethanolamine to generate LC3-II [42] , [43] . The latter accumulates at the inner membrane of the autophagosome to form so-called LC3-punctae which can be quantified by high throughput fluorescence microscopy analysis [48] , [49] . We assessed basal autophagy by using a GFP-LC3 construct that was transduced into immortalized keratinocytes by lentiviral gene transfer . The number of LC3-punctae was then quantified in the presence of bafilomycin which stabilizes LC3-II within the autophagosomes by preventing their acidification [48] , [49] . Here , significantly increased levels of LC3-punctae in immE6 and immE6/E7 cells were noticed ( Fig . S2A ) . However , treatment with bafilomycin or 3-Methyladenin ( 3-MA ) which is known to prevent autophagy via the inhibition of type III phosphatidylinositol 3-kinases ( PI-3K ) [50] , did not recover intracellular levels of IL-1β in E6-positive cells as depicted by confocal microscopy ( Fig . 3A ) or ELISA ( Fig . 3B ) . This observation revealed that the reduced protein levels of pro-IL-1β in immE6 and immE6/E7 are not a consequence of higher basal autophagocytic activity . Consistent with previous reports [44] , [45] , also experimental induction of autophagy by starvation demonstrated that pro-IL-1β is degraded in all cell lines ( Fig . S2B ) . Additionally , we addressed whether enhanced basal lysosomal activity may contribute to increased pro-IL-1β degradation in immE6 and immE6/E7 cells , since the maturation of autophagosomes is a step-wise process that finally culminates in the fusion with lysosomes to generate autolysosomes [43] , [51] , [52] . For this reason , we assessed the lysosome number per cell using LysoTracker Red ( Fig . S2C ) . To measure lysosomal activity , cells were loaded with the bovine serum albumin conjugate DQ™-Red BSA , where intensely fluorescent fragments are detectable upon cleavage of the conjugate . Here , bafilomycin was used as a negative control to prevent DQ™-Red BSA cleavage via lysosomes [53] . As depicted in Fig . S2C , immE6 and immE6/E7 cells exhibited not only more lysosomes and enhanced cleavage of DQ™-Red BSA ( Fig . 3C ) , but also a higher activity of the lysosomal enzyme cathepsin B ( Fig . 3D ) . However , even a wide spectrum of different inhibitors of lysosomal enzymes was not able to increase the levels of pro-IL-1β in these cells ( Fig . 3E ) . One can therefore conclude that the observed differences in lysosomal and cathepsin B activity cannot account for the low basal amounts of pro-IL-1β in immE6 and immE6/E7 cells compared to immE7 cells ( Fig . 1C ) . Having excluded autophagy and lysosomal degradation as potential mechanisms for the reduced pro-IL-1β levels in E6-positive cells ( Fig . 3 ) , we reasoned that a post-translational degradation process may account for the reduced intracellular levels of pro-IL-1β in E6-positive cells compared to immE7 cells . In cells infected with high-risk HPVs , the E6 oncogene is known to interact with the E3 ubiquitin ligase E6-AP forming a trimeric complex with p53 [54] , [55] . This targets p53 for degradation through the ubiquitin-proteasome pathway , resulting in the loss of apoptotic functions and relieve of cell cycle control [54] , [55] . Since E6-AP was shown to ubiquitinate a variety of other cellular proteins [56] , we incubated immortalized cells for six hours with the proteasome inhibitor MG132 and subsequently measured the intracellular levels of IL-1β by ELISA . As shown in Fig . 4A , MG132 treatment increased the amount of IL-1β protein in immE6 and immE6/E7 cells to similar quantities as found in immE7 cells . Note that elevation of IL-1β on the protein level was not due to an increase of the corresponding mRNA after MG132 treatment ( Fig . S3A ) , further supports the notion that a post-translational control mechanism is responsible for IL-1β degradation in immortalized keratinocytes . Next , we analyzed whether IL-1β is a direct substrate for ubiquitination which ultimately targets proteins for proteasomal degradation . For this purpose , immortalized keratinocytes were treated for six hours with the deubiquitinase inhibitor PR-619 [57] ( Fig . 4B ) . Under these conditions , both IL-1β and p53 were degraded in immE7 cells , thereby mimicking the effect of enhanced p53 and pro-IL-1β degradation observed in the presence of the E6 oncoprotein . Additionally , to directly confirm ubiquitination of pro-IL-1β , the TUBE ( Tandem Ubiquitin-Binding Entities ) technique was applied [58] , which allows the specific pull-down of ubiquitinated proteins from cell lysates and their detection by Western blotting ( Fig . 4C ) . Using p53 as a positive control ( Fig . S4A ) , poly-ubiquitinated pro-IL-1β could be visualized in all cell lines , where the band intensities directly reflected the overall steady-state amount of pro-IL-1β protein in the respective samples ( Fig . 4C , see also Fig . 1C for comparison ) . To prove whether E6-AP contributed to the degradation of pro-IL1β , we used siRNA delivery to knock-down E6-AP in immE6 , immE7 and immE6/E7 cells ( Fig . S3B ) . As visualized by confocal microscopy , knock-down of E6-AP leads to an accumulation of pro-IL-1β in the cytosol of transfected cells ( Fig . 4D ) that could also be detected by Western blot ( Fig . 4E ) . Measurement of intracellular IL-1β amounts by ELISA ( Fig . 4F ) revealed that the knock-down of E6-AP in E6-positive cells re-constituted the levels of IL-1β to a similar extend as obtained after MG132 treatment ( Fig . 4A ) . As shown for MG132 treatment ( Fig . S3A ) , knock-down of E6-AP had also no effect on IL1B transcription ( Fig . S3B ) , again demonstrating that E6-AP depletion resulted in post-translational stabilization of pro-IL-1β . To test whether the knock-down of E6-AP and subsequent activation of the NALP3 inflammasome leads to the secretion of mature IL-1β , cells were treated with the ionophore Nigericin [23] . As presented in Fig . 4G , Nigericin treatment alone had no significant effect on immE6 and immE6/E7 cells , but induced a strong IL-1β release in immE7 cells that was similar to that observed after adenoviral infection ( Fig . 1A ) . Notably , in immE6 and immE6/E7 cells , delivery of siRNA against E6-AP and the subsequent treatment with Nigericin lead to a secretion of IL-1β that was comparable to that of immE7 cells ( Fig . 4G ) . This finding demonstrated that the elevation of intracellular pro-IL-1β levels alone was sufficient to restore the secretion capacity of E6-positive cells . Since both MG132 treatment and E6-AP knock-down also restored p53 protein , we next analyzed whether the increase in pro-IL-1β levels was a p53-dependent process . For this purpose , double knock-downs of E6-AP and p53 were performed ( Fig . 4H ) . While E6-AP siRNA was able to strongly elevate the levels of pro-IL-1β in E6- and E6/E7 positive cells , co-delivery of siRNA directed against p53 completely abrogated this effect . In contrast , knock-down of p53 in E7 cells only lead to a minor reduction , probably resulting in the autophagosomal degradation of pro-IL-1β upon p53 inactivation [59] . One can therefore conclude that pro-IL-1β is post-translationally controlled in immE6 and immE6/E7 cells by the interplay between E6-AP , p53 and HPV16 E6 . However , co-immunoprecipitation studies in HPV16-positive SiHa cells did not show a direct interaction between p53 and pro-IL-1β under our experimental conditions where p53 and E6-AP association could be detected ( Fig . S4D ) . Since siRNA directed against the E6 oncoprotein strongly affects the viability of the respective cell lines [60] , we alternatively transfected immE7 cells with HA-tagged HPV16 E6 and analyzed the levels of pro-IL-1β by Western blot . The delivery of E6 not only reduced p53 levels in a dosage-dependent manner , but also lead to a rapid decrease of the steady state amounts of pro-IL-1β 24 hours after transfection ( Fig . 5A ) . This suggests that the expression of E6 appears to immediately interfere with the processing and half-life of pro-IL-1β in a dominant fashion when delivered into immE7 cells . Transfection of HPV18 E6 protein into immE7 cells showed the same effect , whereas the reduction of pro-IL-1β after ectopic expression of E6 of low-risk HPV6 or HPV11 was only marginal ( Fig . S4B ) . Conversely , we raised the question whether the stability of ectopically expressed pro-IL-1β is also affected when introduced into SiHa cervical carcinoma cells , which lack endogenous transcription of the corresponding gene ( Fig . 1D and S1C ) . Using increased p53 levels as an internal control for effective inhibition of the proteasome , ectopically expressed pro-IL-1β could only be stabilized in the presence of MG132 , but was barely detectable when the proteasome inhibitor was omitted ( Fig . 5B ) . Carrying out an equivalent experimental set-up but in the presence of E6-AP siRNA , the same effect was obtained ( Fig . 5C ) . As depicted by Western blotting , the knock-down of E6-AP resulted not only in a strong restoration of endogenous p53 , but also in the stabilization of ectopically expressed pro-IL-1β . These data indicate that the stability of pro-IL-1β is functionally linked to an E6-AP/p53-dependent pathway . In order to determine which region of pro-IL-1β confers its instability in the presence of E6 , we transfected SiHa cells with pro-IL-1β deletion mutants that either expressed only the N-terminus ( IL-1βΔC-HA ) that is lacking the region corresponding to the mature form of pro-IL-1β or an N-terminally truncated version of the protein ( IL-1βΔN ) that contains the region of the mature form . As depicted in Fig . 5D , cells that were transfected with IL-1βΔC-HA did not display any protein expression unless the proteasome was inhibited . Conversely , transfection of IL-1βΔN yielded a strong signal even in untreated cells whose intensity did not increase further upon MG132 treatment . This finding demonstrates that the N-terminus of pro-IL-1β is required to target the protein for proteasomal degradation , whereas the C-terminus of IL-1β appears to be very stable in the presence of HPV16 E6 . To confirm whether our in vitro data are also of clinical relevance , sections of normal cervical tissue , different progression states ( cervical intraepithelial neoplasia , CIN I-III ) and cervical cancer biopsies were studied by immunohistochemistry . Considering representative examples depicted in Fig . 6A , immunohistochemical examination of four out of five normal tissue samples showed ubiquitous staining for IL-1β throughout the whole section from parabasal to suprabasal cells with the most intensive signals coinciding with the basal layer . While three out of five CIN I lesions still show a positive , but diffuse staining , sections with medium and higher degree of neoplasia ( CINII/III lesions ) as well as samples from cervical cancer patients gradually lack detectable IL-1β expression . In parallel experiments where pro-IL-1β RNA was extracted from cervical smears ( characterized by cytology and HPV DNA status ) and analyzed by qPCR ( expressed as box-and-whisker diagram , Fig . 6B ) , the same tendency towards transcriptional silencing of IL1B could be discerned . Notably , low-risk types such as HPV6 obviously prefer the mode of gene suppression instead of post-translational labilization of the IL-1β precursor protein to escape immune surveillance ( see Fig . S4B ) , since five out of five genital warts showed a complete shut-off in IL-1β transcription when compared to primary keratinocytes which were used as a positive control ( Fig . S4C ) .
The innate and adaptive immune system plays a critical role in the prevention , limitation and clearance of high-risk HPV infections , which are known to be the primary cause of anogenital cancer [61] . Due to their low permissive cycle , viral burst size and the absence of a distinctive viremia , the humoral response against viral capsids is weak and can only be detected around six months after a natural infection [62] , [63] . Conversely , this period of time is apparently sufficient to establish a persistent infection , thereby escaping from innate immune surveillance as a first line of antiviral defense . During this time interval , high-risk HPVs , like other potential tumor viruses , attack several hubs within the interactome of their target cell to circumvent innate immunity and prevent cell elimination [5] , [8] , [64] . There are several strategies how this can be mediated including targeting of cytokines [13] and chemokines [14] as well as by inhibition of antigen presentation [65] . In addition , the viral oncogenes promote cell cycle progression [66] and therefore favor the accumulation of forthcoming premalignant cells that are prone to convert towards malignancy [1] , [66] . Beside type I interferons [67] , pro-inflammatory cytokines such as IL-1β are also crucial effector molecules that are required to mount a proper innate immune response against viral infections [27] . In view of its strong immune stimulatory effects and its pleiotropic function , IL-1β release is tightly controlled at multiple levels , such as transcription , post-translational maturation in an inflammasome/caspase-1-dependent manner and secretion of the biologically mature protein [20] , [21] , [27] , [28] , [36] , [68] . In the present report we show that intracellular stability of pro-IL-1β is impaired by a post-translational control mechanism in HPV16-immortalized human keratinocytes ( Fig . 4 ) , therefore limiting their capacity to release mature IL-1β upon infection with adenovirus ( Fig . 1A ) or after treatment with Nigericine ( Fig . 4G ) which activates the NALP3 inflammasome . Similar to our previous studies on the regulation of the keratinocyte-specific interferon-κ [13] , the E6 oncoprotein is playing a central role in this process , since the intracellular levels of pro-IL-1β ( Fig . 1B ) , its secretion as mature IL-1β ( Fig . 1A ) and its biological activity were not impaired in immE7 cells ( Fig . S1B ) . Moreover , abrogation of IL-1β availability during HPV-induced carcinogenesis seems to be a two-step mechanism , since cervical carcinoma cells ( Fig . 1D ) as well as human cervical tumor tissues ( Fig . 6B ) showed either a strongly reduced or even an absence of the corresponding mRNA . The observed higher levels of IL-1β mRNA in immortalized cells in contrast to primary keratinocytes ( Fig . 1D ) could be due to a general viral oncogene-mediated increase of NF-κB signaling [69] that may activate the gene itself , which is known to be NF-κB responsive [70] . Nonetheless , the impairment of IL-1β release in immortalized cells and its transcriptional inactivation in malignant cells and genital warts ( Fig . S4B ) is obviously advantageous , since it also interrupts the well-known regulatory circuit in stimulating the expression of MCP-1 [71] , [72] , a chemokine , attracting cells of the monocyte/macrophage lineage and activating them to release growth-inhibitory cytokines [73] , [74] . Consistent with this notion is also a growth retarding effect on cervical carcinoma cells upon heterotransplantation into nude mice when IL-1β was ectopically expressed [75] . To get insight into the molecular pathway by which IL-1β is regulated in HPV-positive cells , we investigated whether modulation of the inflammasome may account for the decreased IL-1β levels in E6-positive cells . Examples in which way this can be mediated have been reported for several DNA and RNA viruses such as herpesviruses that interfere with the assembly of the inflammasome , influenza viruses encoding a caspase-1 specific inhibitor and orthopoxviruses that secrete an IL-1β binding protein that acts as a decoy receptor [27] . Pro-IL-1β is processed to its mature form by caspase-1 after NALP3 activation and the subsequent autoproteolytic cleavage of pro-caspase-1 [21] . Studies in caspase-1 knock-out mice have shown that these animals have an increased susceptibility to a variety of infections , because IL-1β maturation is impaired [17] . Conversely , caspase-1 seems not to be involved in the immune surveillance against Chlamydia trachomatis [76] , since its activity is even required for efficient and optimal growth of chlamydial inclusions in cervical epithelial cells [77] This aspect is interesting especially in the context of an epidemiologically significant association between Chlamydia infection , persisting HPV and the development of cervical cancer [78] . Hence , our finding that a similar caspase-1 activity can be observed within all immortalized cells independent of which viral oncoprotein is expressed ( Fig . 2B and C ) may provide an explanation for a selective benefit of such a co-habitation in maintaining caspase-1 activity , but shutting down IL-1β secretion also in infected individuals . In attempts to identify the mechanism involved in pro-IL-1β regulation , we also investigated autophagy-related processes in our cell system ( Fig . 3A and B ) . Autophagy is considered to be an evolutionary highly conserved catabolic pathway , targeting proteins or cell organelles for lysosomal degradation . Moreover , autophagy is also interconnected to many intracellular processes such as energy sensing , apoptosis and inflammasome signaling [20] , [53] . Nonetheless , despite increased basal levels of LC3 punctae in autophagosomes ( Fig . S2A ) and enhanced lysosomal activity in immE6 and immE6/E7 keratinocytes ( Fig . 3C ) , inhibition of autophagy ( Fig . 3A and B ) or treatment with various inhibitors of lysosomal proteases failed to increase intracellular pro-IL-1β levels ( Fig . 3E ) . On the other hand , incubation of immE6 and immE6/E7 cells in the presence of MG132 could elevate intracellular pro-IL-1β protein to levels comparable to immE7 keratinocytes ( Fig . 4A ) . Indeed , precedential cases have been reported in a human monocytic cell line ( THP-1 ) or in murine bone marrow-derived macrophages where the proteasome inhibitor MG132 was also capable to significantly raise the intracellular half-life of pro-IL-1β [79] , [80] . This indicates that the intracellular amounts of pro-IL-1β can also be regulated by proteasomal degradation in a NALP3-independent manner [17] . The ubiquitin proteasome pathway targets cellular proteins for proteasomal degradation by site-specific poly-ubiquitination of lysine residues [81] . This process is reversible and balanced by a tightly controlled interplay between ubiquitin ligases and deubiquitinating enzymes [82] , [83] . The treatment with PR-619 , a non-selective reversible inhibitor of deubiquitinases , lead to a strong reduction of both pro-IL-1β and p53 levels in immortalized keratinocytes , showing common pathways controlling their intracellular half-life ( Fig . 4B ) . Similar to p53 , poly-ubiquitinated pro-IL-1β could be visualized after MG132 treatment ( Fig . 4C ) , confirming a proteasome-dependent degradation process that is increased in the presence of the E6 oncogene . In cells infected with high-risk HPVs , the viral oncoprotein E6 targets the E3 ubiquitin ligase E6-AP , forming a trimeric complex with p53 and facilitating its ubiquitination and subsequent degradation via the ubiquitin proteasome pathway [55] . Notably , E6-AP also has an additional function , being a transcriptional co-activator of steroid hormone receptor [84] . This is important in the context of HPV-induced carcinogenesis , since the long-term use of oral contraceptives significantly increases the risk for the development of cervical cancer in HPV-positive women [85] , [86] . To distinguish whether the increased amounts of pro-IL-1β after MG132 treatment were due to a direct involvement of E6-AP or simply due to an extended half-life of a labile protein that controls pro-IL-1β stability , siRNA experiments were performed . As shown by confocal microscopy ( Fig . 4D ) , Western blot ( Fig . 4E ) and ELISA ( Fig . 4F ) , the specific knock-down of E6-AP restored intracellular pro-IL-1β levels and even rescued the secretion of mature IL-1β after NALP3 inflammasome stimulation ( Fig . 4G ) . Remarkably , this process is directly depending on the availability of p53 , because its simultaneous knock-down counteracts intracellular pro-IL-1β accumulation , as achieved by preceding inactivation of E6-AP alone ( Fig . 4H ) . Since neither the treatment with MG132 nor the knock-down of E6-AP lead to an increase in IL-1β mRNA levels ( Fig . S3A and B ) , a role of p53 as a transcriptional activator of the IL1B gene can be excluded . Although co-immunoprecipitation experiments demonstrate not direct binding between p53 and pro-IL-1β , under our experimental conditions where E6-AP and p53 interact ( Fig . S4A and D ) , the latter seems to be involved in the half-life control of pro-IL-1β . Hence , mass spectrometric analyses [87] should identify accessory proteins of pro-IL-1β that modulate its intracellular stability in high-risk HPV-positive cells . Nevertheless , since p53 can be also considered as a central hub within the cellular network , its inactivation by E6 and the successive downstream effects also on the keratinocyte-specific interferon-κ and chemokines such as MCP-1 [14] allows the virus to efficiently circumvent essential components of the innate immunity in a one-step mode by reducing p53 availability . These findings identify a novel role of p53 and E6-AP within the cellular interactome , which apart from inactivating p53 [55] or proteins involved in cell cycle control or differentiation [56] , also controls the levels of IL-1β , a central cytokine that is necessary to orchestrate a potent innate immune response [20] , [28] . Moreover , the importance of E6-AP in the development of HPV-induced carcinomas has been elegantly shown in an HPV16-transgenic mouse model after cross-breeding with E6-AP null animals where the loss of E6-AP completely abrogated the E6-induced development of cervical cancer upon chronic estrogen treatment [88] . In order to study the role of E6 in greater detail , transfection of HA-tagged HPV16 E6 expression vectors into E7-positive keratinocytes strongly down-regulated the levels of both pro-IL-1β and p53 in a dose-dependent manner ( Fig . 5A ) . The same effect could be noted in immE7 cells that expressed the E6 proteins of HPV18 , while E6 of the low-risk types ( HPV6 or HPV11 ) had only a minor influence on pro-IL-1β stability ( Fig . S4B ) . In contrast , monitoring the levels of p53 after HPV6 and HPV11 E6 expression in comparison to empty vector transfected E7-positive keratinocytes as reference , a discernible effect could be observed ( Fig . S4B ) . While some studies claimed that E6 proteins from low-risk HPV are neither able to bind nor degrade p53 [54] , [89] , [90] , other groups showed the opposite , although with weaker efficiency than high-risk HPV E6 [91]–[93] . Discrepancies like this may be explained either by the usage of different cell lines , the respective gene dosage or the techniques to monitor the final read-out . For our transfection experiments we used human keratinocytes as recipients which are the natural target cells for an HPV infection . One possible explanation for the decrease of p53 in our experiments could be that the presence of HPV16 E7 is somehow predisposing p53 for reduction of its half-life upon low-risk HPV E6 delivery , since E7 can up-regulate SIRT , an aging-related NAD-dependent deacetylase . This can lead to p53 deacetylation [94] thereby inherently affecting its stability [95] . As shown in Fig . 5B and C , ectopic expression of pro-IL-1β in combination with MG132 treatment or E6-AP siRNA knock-down was able to stabilize pro-IL-1β in cervical carcinoma cells . This indicates that even though the endogenous gene is no longer transcribed , the post-translational process that controls the half-life of pro-IL-1β is still active in the presence of E6 . One can therefore propose that there exists a two-step silencing mechanism of IL-1β towards cervical cancer: i ) on the post-translational level during immortalization and ii ) transcriptionally by gene silencing in malignant cells . Since the absence of transcription was also noted in the HPV-negative C-33 A cell line ( Fig . S1C ) , U2OS , HEK 293 and H1299 cells ( data not shown ) , this second step of IL-1β inactivation is likely not HPV-dependent , but might merely reflect a selective growth advantage in different types of tumors . The fact that E6-positive SiHa cells still maintained the E6-mediated post-translational degradation of ectopically expressed pro-IL-1β allowed us to assess the region of pro-IL-1β that confers its instability in these cells . Different in vitro studies provided some hints that mature IL-1β is less susceptible to various proteases than its pro-form [96] . To characterize the region that is responsible for degradation , N- or C-terminally truncated mutants of pro-IL-1β were expressed in SiHa cells ( Fig . 5D ) . As already anticipated the deletion mutant containing amino acids 1-116 of pro-IL-1β was highly susceptible to proteasomal degradation and could only be stabilized by MG132 treatment , whereas the N-terminally truncated mutant protein ( lacking amino acids 1-76 and containing the full-length region of mature IL-1β ) was readily expressed even in untreated cells . We are currently analyzing whether one or more of the five lysine residues within the N-terminal pro-protein region of IL-1β are responsible for its increased degradation in the presence of E6 . Finally , to estimate the clinical relevance of our in vitro data , initial validation of human biopsy material shows a progressive loss of IL-1β , as monitored both by immunohistochemical staining ( Fig . 6A ) and quantitative RT-PCR of tissue samples ( Fig . 6B ) , representing different grades of intraepithelial neoplasia ( CIN I–III ) and cervical cancer . We currently do not know at which time there is a switch from post-translational labilization of pro-IL-1β towards silencing of the gene itself during multi-step progression to cervical cancer . However , our recent studies indicate that the IL-1B gene gets converted into heterochromatin , a situation already described for the keratinocyte-specific interferon-κ [13] . Notably , monitoring IL-1β transcription in genital warts infected by HPV6 , the same situation could be discerned ( Fig . S4C ) . Persistent genital warts are characterized by a lack of infiltrated immune cells where low numbers of intraepithelial CD8+ T cells and mononuclear cells are present mainly in the stroma [10] . Since IL-1β mediates the migration of leukocytes and promotes the activation of T cells [18] , its consecutive transcriptional inactivation apparently provides a selective advantage also for low-risk HPV-infected cells to escape immune surveillance , since their ability to labilize IL-1β in a post-translational manner is only marginal when compared to HPV16 or HPV18 E6 ( Fig . S4B ) . A broader study of tissue specimen will answer the question whether the status of IL-1β expression can also be used as an additional marker for progressing lesions . Taken together , our data reveal an effective and novel mechanism how high-risk HPV circumvent the function of IL-1β . As a consequence , attenuated innate immune response against HPV-infected cells may favour viral persistence , which is an important step in the initiation of cellular transformation and tumorigenesis . A detailed understanding of how high- risk HPV E6 can interfere with the expression and maturation of different pro-inflammatory cytokines , such as IL-1β via E6/E6-AP/p53 interaction or via de novo methylation as recently shown for the keratinocyte-specific type I interferon IFN-κ [13] , should allow the development of new strategies to treat existing HPV lesions before their progression to invasive tumors .
Samples were donated from HPV infected patients or health donors using protocols approved by the Charité Campus Benjamin Franklin , Berlin , Germany and the Deutsche Klinik Bad Münder , Hannover , Germany . Subsequently the samples were analysed anonymously where informed consent was not required . HPV16-immortalized keratinocytes expressing the E6 and/or E7 oncogene [97] were cultivated in Keratinocyte-SFM medium containing rhEGF and bovine pituitary gland extract ( Life Technologies ) . Tumor cell lines CaSki , SiHa , HeLa and C-33 A were grown in Dulbecco's Modified Eagle Medium ( Sigma ) containing 10% fetal calf serum ( Linaris ) . Human Umbilical Vein Endothelial Cells ( HUVEC ) ( EMD Millipore ) were cultivated in EndoGRO-Low Serum complete medium ( EMD Millipore ) . Human neonatal Epidermal Keratinocytes ( Life Technologies ) were cultivated in EpiLife medium ( Life Technologies ) containing 60 µM calcium supplemented with Human Keratinocyte Growth Supplement ( Life Technologies ) . Infection using GFP-adenovirus 5 ( Vector Biolabs ) was carried out as described previously [98] . Briefly , 100 MOI of virus was added to the primary keratinocytes and HPV16-positive cells in the respective medium and incubated for 1 h . The cells were washed twice with phosphate buffered saline and cultivated again for 24 h . Subsequently the supernatants were collected and stored at −80°C until used to determine the amount of secreted IL-1β by ELISA . ELISAs were performed using the Human IL-1 beta ELISA Ready-SET-Go ! Kit ( eBioscience ) according to the manufacturer's instructions . For intracellular protein analyses , 15–30 µg of total protein was applied in triplicates to the coated plate . Secreted IL-1β levels were measured by applying 100 µl of supernatants of the respective medium to the coated ELISA plate . After treatments or transfections , ∼1×106 cells were collected , washed in 1×PBS and resuspended in RIPA buffer ( 20 mM Tris pH 7 . 5; 150 mM NaCl; 1 mM Na2EDTA; 1 mM EGTA; 1% NP-40; 1% sodium deoxycholate ) including 1× complete protease inhibitor cocktail ( Roche ) . Samples were incubated for 30 min on ice and subsequently centrifuged for 30 min at 4°C and 13 , 000 rpm . The supernatants were quantified using Bio-Rad Protein Assay Dye Reagent Concentrate ( Bio-Rad ) . 50–80 µg of denatured proteins was used for Western blotting . After transfer , the filters were incubated with the following antibodies: anti-human interleukin-1β ( IL-1b-I ) , 3415-3-250 , ( MABTECH ) , anti-p53 antibody ( DO-1 ) , sc-126 ( Santa Cruz Biotechnology ) , anti-E6AP antibody ( H-182 ) , sc-25509 ( Santa Cruz Biotechnology ) , anti-actin clone C4 ( MP Biomedical ) , anti-LC3A D50G8 ( Cell Signalling ) , rat anti-HA clone 3F10 ( Roche ) . After transfections or treatments , ∼1×106 cells were collected , washed in 1×PBS and resuspended in non-denaturing lysis buffer ( 20 mM HEPES , 0 . 15 M NaCl , 5 mM EDTA , 10% Glycerol , 0 . 5% Triton X-100 ) including 1× complete protease inhibitor cocktail ( Roche ) . Samples were incubated for 30 min on ice and subsequently centrifuged for 30 min at 4°C and 13 , 000 rpm . The supernatants were quantified using Bio-Rad Protein Assay Dye Reagent Concentrate ( Bio-Rad ) and 500 µg of protein was used for immunoprecipitation . Non-specific binding in the cell extracts was removed by adding 50 µl of protein G–Sepharose beads ( Santa Cruz Biotechnology ) equilibrated with non-denaturing lysis buffer and rocking for 1 h at 4°C . After removing the beads , the cell extract was incubated with 5 µg of anti-p53 monoclonal antibody ( DO-1 ) or 5 µg of normal mouse IgG sc-2025 ( Santa Cruz Biotechnology ) used as control for 6 h at 4°C . Then , 10 µl of protein G–Sepharose beads equilibrated with non-denaturing lysis buffer was added to each reaction mixture and rocked for 12 h at 4°C . The beads were washed five times with non-denaturing lysis buffer and analyzed by Western blotting using interleukin-1β ( IL-1b-I ) , anti-p53 antibody ( DO-1 ) and anti-E6AP antibody ( H-182 ) . Additional co-immunoprecipitation experiments were performed using 4 µg of interleukin-1β antibody or 0 . 8 µg of GFP antibody ( Roche ) for immunoprecipitation . Western blot analyses employed the previously described antibodies according to [99] . HA-tagged pro-IL-1β was detected with an antibody directed against the HA-tag ( Roche ) . RNA was extracted from cells using the RNeasy Mini Kit ( Qiagen ) according to the manufacturer's instructions . RNA concentrations were determined photometrically and 1 µg of RNA was reverse transcribed using RevertAid Reverse Transcriptase ( Fermentas ) and dT22 primers according to the manufacturer's protocol . A 1∶5 dilution of the resulting cDNA was used for semi-quantitative and quantitative PCR analyses . Semi-quantitative RT-PCRs were performed by using DreamTaq Green DNA polymerase ( Fermentas ) according to the manufacturer's instructions . The following primers were used: GAPDH forward: 5′-GCCTTCCGTGTCCCCACTGC-3′ , GAPDH reverse: 5′-GCTCTTGCTGGGGCTGGTGG-3′ , Caspase-1 forward: 5′-TCTTCCTTTCCAGCTCCTCA-3′ , Caspase-1 reverse: 5′-CGCTGTACCCCAGATTTTGT-3′ , IL-1β forward: 5′- GGGCCTCAAGGAAAAGAATC-3′ , IL-1β reverse: 5′-AGCTGACTGTCCTGGCTGAT-3′ , IL-6 forward: 5′-TCGAGCCCACCGGGAACGAA-3′ , IL-6 reverse: 5′-GCAGGGAAGGCAGCAGGCAA-3′ , CCL-20 forward: 5′-GGCGAATCAGAAGCAAGC-3′ , CCL-20 reverse: 5′- TTCCATTCCAGAAAAGCCAC-3′ , E6-AP forward: 5′-GCGGGGGCGACGACAGGTTA-3′ , E6-AP reverse: 5′-TGCAGCTTCTCCATCCTGCAAGC-3′ . Quantitative real-time PCR ( qPCR ) was performed with an ABI 7300 qPCR cycler ( Applied Biosystems ) using Maxima SYBR Green/ROX qPCR Master Mix ( Fermentas ) and the following primers: IL-1β forward: 5′-AGGCACAAGGCACAACAGGCT-3′ , IL-1β reverse: 5′-GGTCCTGGAAGGAGCACTTCAT-CTG-3′ , IL-1α forward: 5′-TGGTAGTAGCAACCAACGGGA-3′ , IL-1α reverse: 5′-ACTT-GATTGAGGGCGTCATTC-3′ , IL-18 forward: 5′-ATCGCTTCCTCTCGCAACAA-3′ , IL-18 reverse: 5′-TCCAGGTTTTCATCATCTTCAGC-3′ , IL-33 forward: 5′-TAGGAGAGAAACC-ACCAAAAGG-3′ , IL-33 reverse: 5′-ACTTTCATCCTCCAAAGCAAAAGT-3′ . As a normalization control , TBP ( = TATA-Box binding protein ) -specific primers were used . TBP forward: 5′-GAGTCGCCCTCCGACAAAG-3′ and TBP reverse: 5′-GTTTCCTCTGGGATTCCATCG-3′ . Caspase-1 activity assay was performed by staining the cells for 4 h at 37°C with 20 µM of the cell-permeant R110-YVAD ( Rhodamine 110-Based Caspase-1 substrates; Life Technologies ) . The fluorescence emission of caspase-1 ( Excitation: 485 nm/Emission: 528 nm ) were read in a Synergy 2 multireader ( BioTek ) without fixation . For the pharmacological inhibition of caspase-1 , 250 nM of the caspase-1 inhibitor I cell permeable ( Calbiochem ) for 5 h was used . The GFP-LC3 expressing lentivirus was a kind gift from Dr . Chiramel and Dr . Bartenschlager from Heidelberg University . Shortly , the lentiviral particles were mixed with Polybrene 5 µg/ml ( Sigma ) and incubated with the immortalized keratinocytes for 6 h and then replaced by normal culture medium . 72 h after infection , the target cells were seeded in 6 well plates and selection with puromycin ( 2 to 3 µg/ml ) was performed for 12 days . The pooled puromycin resistant clones were tested by fluorescence microscopy and Western blot analysis . Quantification of lysosomes was performed using LysoTracker Red and lysosome activity with DQ Red BSA ( Life Technologies ) . Cathepsin B activity was monitored by incubating the cells with Magic Red ( Immunochemistry technologies ) according to the manufacturer's instructions . Inhibitors of autophagy used: 1 mM 3-methyladenine ( Enzo Life Sciences ) incubated for 8 h and 100 nM bafilomycin ( Enzo Life Sciences ) incubated for 8 h . Inhibitors of lysosome/protease activity: 20 µM of Calpain inhibitor PD 150 , 606 ( Adipogen ) , 25 µM of cathepsin B inhibitor CA-074 ( Enzo Life Sciences ) , 10 µM of Leupeptin ( Biomol ) , 20 µM Vincristine ( Enzo Life Sciences ) . All treatments were carried out for 6 h . Starvation was done by cultivating the cells in HEPES buffered saline solution ( HBSS ) for 8 h . For proteasomal inhibition cells were treated with 5 µM of the proteasome inhibitor MG132 ( Calbiochem ) for 6 h . Quantification of autophagy , lysosome amount and activity as well as cathepsin B activity were performed using high throughput high resolution fluorescence microscopy analysis ( BD pathway , Beckton Dickenson ) using the filter set Ex: 516 nm/Em: 590 nm . The images were analyzed using the cell imaging analysis program ( CellProfiler ) . For inflammasome stimulation , the cells were incubated for 6 h with 50 µM of Nigericin ( Enzo Life Sciences ) . Inhibition of deubiquitinases was carried out by incubating cells with 10 µM of PR-619 ( Life sensors ) for 6 h . The complete pro-IL-1β protein coding region including the stop codon was amplified from an IL-1β open reading frame “gateway” clone ( Core facility , DKFZ ) with Phusion polymerase ( Finnzymes ) using the following primers: pIL-1β forward: 5′-CTCGAGGCCGC-CATGGCAGAAG-3′ and pIL-1β reverse: 5′-CTCGAGTTAGGAAGACACAAATTGCATG-3′ . HA-tagged IL-1β was generated using the pIL-1β forward primer with the pIL-1β-HA reverse:5′-GAATTCGAAGACACAAATTGC-3′ . Truncated IL-1β expression plasmids either lacking the C-terminus from amino acid 117 to 269 ( IL-1βΔC-HA ) or lacking the N-terminal part from amino acid 1 to 76 ( IL-1βΔN ) were constructed using the following primers: IL-1βΔC forward 5′-CTCGAGGCCGC-CATGGCAGAAG-3′ , IL-1βΔC reverse: 5′-TCTCGAATTCGCATCGTGCACATAAGCC-3′ or IL-1βΔN forward: 5′-GAGACTCGAGGCCATGCTGGTTCCCTGC-3′ , IL-1βΔN reverse , 5′-CTCGAGTTAGGAAGACACAAATTGCATG-3′ , respectively . Amplified DNA was cloned into the pPK-CMV-E3 expression vector , where the fragments are inserted to a HA ( hemagglutinin ) fusion tag ( PromoKine ) . pPK-CMV-E3 was also used to express the full length E6 open reading frame of the different HPV types where the following primers were used: HPV6E6 forward: 5′-GAGACTCGAGGCCGCCATGGAAAGTGC-3′ , HPV6E6 reverse: 5′-TCTCGGATCCGGGTAACATGTCTTC-3′ , HPV11E6 forward: 5′- GAGACTCGAGGCCGCCATGGAAAGTAAAG-3′ , HPV11E6 reverse: 5′-TCTCGGATC-CGGGTAACAAGTCTTC-3′ , HPV18E6 forward: 5′-GAGACTCGAGGCCGCCATGGCGC-GCTTTGAG-3′ , HPV18E6 reverse: 5′-TCTCGAATTCAGTACTTGTGTTTC-3′ . HPV16E6 in pPK-CMV-E3 and p53-YFP were provided by Dr . Dorothea Muschik . Cervical carcinoma cells were transfected with expression plasmids using TurboFect Transfection Reagent ( Fermentas ) , while immortalized keratinocytes were transfected using Lipofectamine 2000 Transfection Reagent ( Life Technologies ) according to the manufacturers' instructions . E6-AP or p53 knock-down was performed by transfection of Silencer Select siRNA directed against human UBE3A ( S14604 ) or human TP53 ( S605 ) ( both Ambion ) using Lipofectamine 2000 Transfection Reagent ( Life Technologies ) according to the manufacturer's instructions . Cells were incubated for 24 h and then transfected a second time under the same conditions . Cells were harvested after 48 h for RNA , or after 72 h for protein extraction , respectively . After different incubation conditions , immortalized keratinocytes were fixed in 4% paraformaldehyde in PBS for 20 min at 25°C and washed for 5 min in 0 . 1 M glycine prior to permeabilization with 0 . 25% Triton X-100 in PBS for 15 min . For immunostaining , cells were incubated with the anti-human interleukin-1β ( IL-1b-I ) antibody ( MABTECH 3415-3-250 ) diluted 1∶1000 . The primary antibody was visualized after washing with PBS and subsequent incubation using the following secondary antibodies: rabbit anti-mouse IgG-Alexa-488 ( cat: A11029 ) or IgG-Alexa-633 ( cat: A21052; Invitrogen ) for 1 h . Nuclei were stained by Hoechst solution in PBS ( 1∶10000; Sigma Chemicals ) . Fluorescence signals were obtained with a Zeiss Confocal Laser Scan Microscope ( Zeiss ) and the images were analyzed using the Zen imagine program ( Zeiss ) . Ubiquitinated proteins were isolated from cells using the agarose-conjugated tandem ubiquitin-binding entities ( TUBEs ) technique ( Life Sensors ) according to the manufacturer's instructions . Briefly , MG132-treated keratinocytes ( 5×106–1×107 cells ) were lysed in TUBE lysis buffer ( 50 mM Tris-HCl , pH 7 . 5; 0 . 15 M NaCl; 1 mM EDTA; 1% NP-40; 10% glycerol ) containing 1× complete protease inhibitor cocktail ( Roche ) , 20 µM MG132 and 50 µM PR-619 and were pre-cleared using protein A/G Plus Agarose beads ( Santa Cruz Biotechnology ) . 2 mg of pre-cleared cell lysates were incubated with 30 µl of equilibrated agarose-conjugated TUBEs for 4 h at 4°C on a rotating platform , washed three times in 1×TBST and eluted by incubation at 95°C for 5 minutes in 1×SDS loading dye . Eluates were subjected to SDS-PAGE on 10% acrylamide gels and analyzed by Western blotting Cervical smear ( n = 8 ) fixed in PreservCyt medium ( Becton Dickinson ) at 4°C were collected in a routine colposcopy clinic in Bad Münder , Germany . All samples were HPV16 DNA-positive identified both by the BSGP5+/6+-PCR/Multiplex HPV Genotyping ( MPG ) assay that homogenously amplifies all known genital HPV types generating biotinylated amplimers of ∼150 bp from the L1 region [100] and a MPG assay with bead-based xMAP Luminex suspension array technology , which is able to simultaneously detect 51 HPV types and the β-globin gene [101] , [102] . Colposcopy directed biopsies were taken from all 8 patients and histologically defined as cervical intraepithelial neoplasia ( CIN1 ) ( n = 1 ) , CIN2 ( n = 2 ) and CIN3 ( n = 5 ) . Furthermore , cervical cancer ( CxCa ) samples ( n = 3 ) and samples negative for intraepithelial lesion and malignancy ( Nil/M ) ( n = 5 ) were added from a population-based HPV prevalence study conducted in Ulaanbaatar , Mongolia in 2005 [103] . The 3 CxCa samples were well characterized by histology , the presence of HPV DNA and E6 transcription [104] . Genital benign warts ( n = 5 ) were collected and transported in RNAlater RNA Stabilization Reagent at room temperature in a routine gynecological examination at the Bad Münder clinic , Germany and tested for HPV6 and HPV11 mRNA positivity by qPCR analysis using the following primers: HPV6-E7 forward: 5′-TTCGACTGGTTGTGCAGTGT-3′ , HPV6-E7 reverse 5′-GCGCAGATGGGA-CACACTAT-3′ , HPV11-E7 forward 5′-GACCCTGTAGGGTTACATTGC-3′ , HPV11-E7 reverse 5′-AGTGTGCCCAGCAAAAGGTC-3′ . RNA was extracted from 16 cervical smears obtained from Germany and Mongolia using the MagNA Pure 96 device ( Roche Applied Science , Germany ) according to the manufacturer's instructions . Briefly , 4 ml PreservCyt volume was centrifuged for 10 min at 4 , 700 rpm . The pellet was resuspended in 200 µl of buffer and processed into the MagNA Pure 96 system; the RNA was eluted in 50 µl of elution buffer and stored at −70°C until use . Concentration of the RNA was measured using a NanoDrop 2000 . RNA from 5 genital warts was extracted cells using the RNeasy Mini Kit ( Qiagen ) as describe previously . A total of 25 cases , including normal cervical epithelium , CIN I , CIN II , CIN III and cervical cancer ( 5 cases of each grade ) were analyzed . Formalin-fixed , paraffin-embedded sections ( 4 µm thick ) were mounted on super-frost slides , dewaxed with xylene and gradually rehydrated . The sections were boiled for 20 min in Target Retrieval Solution ( DAKO ) . Activity of endogenous peroxidase was blocked by 5 min incubation in Peroxidase-Blocking Solution ( DAKO ) . The sections were then exposed to normal goat serum ( Invitrogen ) for 20 min . Immunohistochemical reactions were performed using a two-step staining technique ( DAKO Envision system-HRP ( DAB ) ) . The samples were first incubated with the anti-human interleukin-1β ( IL-1b-I ) , 3415-3-250 , ( MABTECH ) diluted 1∶100 for 2 h and subsequently with the corresponding secondary antibody for 30 min at room temperature . All sections were stained with DAB ( DAKO ) for 10 min and counterstained with Meyer's hematoxylin . In every case , control reactions were included in which the specific antibody was substituted by the primary mouse negative control ( DAKO ) . To evaluate the statistical differences between analyzed groups in figure 6 , a two-sided unpaired t-test was applied . The statistical differences between analyzed groups in figure 3 were assesses by an ANOVA test . The following human proteins are available in the “Swiss-Prot” database ( http://www . uniprot . org ) : IL-1β ( P01584 ) , Caspase-1 ( P29466 ) , TBP-1 ( P20226 ) , E6-AP ( Q05086 ) , p53 ( P04637 ) , actin ( P60709 ) and GAPDH ( P04406 ) . | Persistently high-risk HPV-infected individuals have an increased risk to develop anogenital cancer . HPV encodes the viral proteins E6 and E7 that interact with and induce the degradation of the cell cycle regulators p53 and pRb , respectively , priming immortalized keratinocytes towards malignant transformation . In early antiviral immune response , IL-1β is an important factor for the initiation of inflammation and activation of immune cells such as macrophages and T cells . Our study describes a post-translationally controlled pathway where E6 mediates proteasomal degradation of IL-1β in HPV16-immortalized human keratinocytes . This process depends on the cellular ubiquitin ligase E6-AP and p53 highlighting a novel molecular mechanism of a virus-host interaction that is critical for evading innate immune defense . IL-1β dysregulation is also found in tissue sections which represent different stages of virus-induced carcinogenesis , underlining the clinical relevance of our findings . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"immunity",
"virology",
"immunology",
"biology",
"microbiology"
]
| 2013 | Post-Translational Control of IL-1β via the Human Papillomavirus Type 16 E6 Oncoprotein: A Novel Mechanism of Innate Immune Escape Mediated by the E3-Ubiquitin Ligase E6-AP and p53 |
Mouse apolipoprotein B mRNA-editing enzyme catalytic polypeptide-like editing complex 3 ( mA3 ) , an intracellular antiviral factor , has 2 allelic variations that are linked with different susceptibilities to beta- and gammaretrovirus infections among various mouse strains . In virus-resistant C57BL/6 ( B6 ) mice , mA3 transcripts are more abundant than those in susceptible BALB/c mice both in the spleen and bone marrow . These strains of mice also express mA3 transcripts with different splicing patterns: B6 mice preferentially express exon 5-deficient ( Δ5 ) mA3 mRNA , while BALB/c mice produce exon 5-containing full-length mA3 mRNA as the major transcript . Although the protein product of the Δ5 mRNA exerts stronger antiretroviral activities than the full-length protein , how exon 5 affects mA3 antiviral activity , as well as the genetic mechanisms regulating exon 5 inclusion into the mA3 transcripts , remains largely uncharacterized . Here we show that mA3 exon 5 is indeed a functional element that influences protein synthesis at a post-transcriptional level . We further employed in vitro splicing assays using genomic DNA clones to identify two critical polymorphisms affecting the inclusion of exon 5 into mA3 transcripts: the number of TCCT repeats upstream of exon 5 and the single nucleotide polymorphism within exon 5 located 12 bases upstream of the exon 5/intron 5 boundary . Distribution of the above polymorphisms among different Mus species indicates that the inclusion of exon 5 into mA3 mRNA is a relatively recent event in the evolution of mice . The widespread geographic distribution of this exon 5-including genetic variant suggests that in some Mus populations the cost of maintaining an effective but mutagenic enzyme may outweigh its antiviral function .
The family of apolipoprotein B mRNA-editing enzyme catalytic polypeptide-like editing complex 3 ( APOBEC3 ) proteins consists of cytidine deaminases that function as cellular restriction factors against various exogenous and endogenous viruses [1]–[17] . Seven APOBEC3 paralogues have been identified on human chromosome 22 , while only a single copy of the Apobec3 gene is found in the mouse genome [10] , [18] , [19] . Among the human APOBEC3 enzymes , APOBEC3G ( hA3G ) is the best characterized member and is known to inhibit HIV-1 replication when the virus lacks the functional accessory protein , viral infectivity factor ( Vif ) [reviewed in 20] . In the absence of Vif , hA3G is incorporated into newly generated virions budding from virus-producing cells and exhibits its antiviral effect in subsequently infected cells . Thus , during reverse transcription in the target cells , the virion-incorporated hA3G catalyzes C-to-U deamination on the minus strand of nascent viral DNA , resulting in G-to-A mutations on the plus strand of the double-stranded viral DNA , which can be detrimental to viral replication [7] , [9] , [10] , [21] , [22] . In addition , a deaminase-independent antiviral mechanism exerted by hA3G has also been reported [23] , [24] . In contrast to its human counterparts , mouse APOBEC3 ( mA3 ) restricts HIV-1 regardless of the presence of Vif , as well as mouse mammary tumor virus ( MMTV ) , ecotropic murine leukemia viruses ( MuLVs ) , Friend MuLV ( F-MuLV ) and Moloney MuLV ( M-MuLV ) , along with endogenous mouse retroviruses including the AKR ecotropic virus ( AKV ) [5] , [25]–[30] . This suggests that APOBEC3 enzymes protect host genomes from the retroviruses they commonly encounter , although some retroviruses , like HIV-1 , have evolved to counter the intracellular restriction mechanisms of their natural hosts . Friend virus ( FV ) is an acutely leukemogenic retroviral complex composed of replication-competent F-MuLV and replication-defective spleen focus-forming virus ( SFFV ) . Susceptibilities to FV-induced disease development differ among various inbred strains of mice , and these are controlled by several host factors that either directly affect FV replication or influence host immune responses to the viral antigens [31] , [32] . We and others have reported that the mouse Apobec3 locus is polymorphic , and its genotypes are associated with the levels of viremia after F-MuLV or FV inoculation [26] , [27] . Mice of the prototypic FV-resistant strains C57BL/6 ( B6 ) and C57BL/10 exhibit restricted replication of F-MuLV and earlier production of FV-neutralizing antibodies , while FV-susceptible BALB/c and A strains are less restrictive of F-MuLV replication and show delayed production of neutralizing antibodies , all of which are linked with Apobec3 genotypes [26] , [27] , [33] , [34] , although the production of neutralizing antibodies is also influenced by genotypes at the major histocompatibility complex and the Tnfrsf13c loci , the latter of which encode the receptor for B-cell activating factor belonging to the tumor necrosis factor family [31]–[33] , [35] . Mouse APOBEC3 and hA3G contain two cytidine deaminase domains ( CDDs ) , each harboring the conserved zinc-coordinating motif; however , deaminase activity is exerted only by the N-terminal CDD of mA3 and the C-terminal CDD of hA3G [36] . We showed that the increased efficiency of B6 mA3 in inhibiting F-MuLV replication is associated with differences in the primary amino acid sequence within the active N-terminal CDD [27]; the functional importance of these residues was further implicated by the demonstration that they have been under positive selection in Mus [37] . In addition to the above sequence differences in the protein-coding regions , efficient virus restriction is also associated with higher levels of mA3 transcripts in FV-resistant B6 than in -susceptible BALB/c mice [27] , [29] , [38] , [39] . This enhanced transcription was linked with the presence of the long terminal repeat ( LTR ) of an endogenized xenotropic MuLV in the B6 , but not in the BALB/c , Apobec3 locus [37] . A third factor associated with virus resistance is the presence or absence of exon 5 , which encodes a 33-amino acid segment separating the C-terminal and N-terminal CDDs [27] . The mA3 isoform of the FV-resistant strains of mice lacks exon 5 , while the predominant transcript in FV-susceptible mice is the exon 5-containing isoform [27] , [29] , [37] . Genetic sequences controlling mA3 splicing have not been identified , although it has been pointed out that polymorphisms between the B6 and BALB/c Apobec3 alleles at the end of intron 4 include a putative splice acceptor site and possible mRNA branch selection site structures [38] . It has also not been shown whether the well-confirmed differences in transcript levels result in altered expression levels of mA3 protein in FV-resistant and -susceptible strains of mice . In the present report , we show that mA3 protein is indeed more abundant in B6 than in BALB/c mice . This difference is due in part to more efficient translation of the exon 5-deficient message . We further show extensive functional evidence that two distinct polymorphisms within the Apobec3 locus regulate exon 5 inclusion during its splicing: the previously predicted [38] TCCT repeat numbers in intron 4 and a newly identified single nucleotide polymorphism ( SNP ) within exon 5 . We also describe the linkage between these splicing regulatory sequences in wild mouse species , their acquisition in Mus evolution , and their distribution in wild mouse populations .
It has been reported that mA3 mRNA expression is higher in B6 than in BALB/c mice , and the mA3 transcripts detected in B6 mice are predominantly the exon 5-lacking Δ5 isoform , while the majority of mA3 transcripts in BALB/c mice contain exon 5 ( 5+ ) [27] , [29] , [38] . To more accurately describe the above quantitative differences in mA3 transcripts , we performed two types of PCR analyses with two different sets of specific primers ( Figure 1A ) . The primer set c–d , which is the same as the one previously used [27] , detected both the 5+ and Δ5 transcripts in reverse-transcription PCR ( RT-PCR ) assays; however , mA3 transcripts were barely detectable after 30 cycles of amplification in BALB/c mice while the Δ5 transcript was readily detectable in B6 mice ( Figure 1B ) . After 35 cycles of amplification , both the 5+ and Δ5 transcripts became detectable in BALB/c mice , although the 5+ mRNA was more abundant . Quantitative real-time PCR ( qPCR ) assays revealed mA3 expression levels that were approximately 18-times higher in B6 than in BALB/c spleens , although the difference in transcript lengths might have influenced the efficiencies of amplification in the real-time PCR reactions . To more precisely quantify the mA3 transcripts , we utilized the second primer set , a–b , which generates amplicons of the same size from both alleles ( Figure 1A ) . The qPCR assay performed by using this latter primer set clearly demonstrated that B6 mice expressed >7-times higher amounts of mA3 transcript than BALB/c mice did ( Figure 1C ) , consistent with the previous report [27] . The normalization of mA3 mRNA levels with TATA box-binding protein or GAPDH transcripts instead of β-actin gave similar results ( data not shown ) . Protein levels of mA3 in the above prototypic FV-resistant and -susceptible strains of mice were also compared . The spleen lysate from the mA3-knockout mice [40] was used as a negative control . In B6 spleens , an immunoreactive protein corresponding to Δ5 mA3 was detected as a prominent band , but the higher molecular weight 5+ mA3 was also faintly detectable ( Figure 1D ) , even though the 5+ message was barely detectable by RT-PCR in the present study ( Figure 1B ) and in the previous reports [27] , [29] . On the other hand , only the 5+ mA3 protein was detected , with a much lower intensity , in BALB/c spleens . The immunoblotting assays thus demonstrated that the level of total mA3 protein expression in the spleens of B6 mice is much higher than that in BALB/c mice . The mA3 expression data shown in Figure 1 indicate that levels of mA3 protein expression roughly correlate with the corresponding transcript levels in mice with different Apobec3 alleles . However , we previously observed that even when FLAG-tagged mA3 was expressed under the control of the cytomegalovirus ( CMV ) promoter , the levels of expression of the 5+ mA3 protein tended to be much lower than those of the Δ5 protein , despite similar levels of mRNA expression in transfected cells [27] . This implies that exon 5 might affect either the efficiency of mA3 translation or protein stability . To evaluate these possibilities , 5+ and Δ5 mA3 expression plasmids were constructed with either B6- or BALB/c-derived mA3 cDNA ( Figure 2A ) . The mA3 cDNA were fused with a FLAG tag at their N-terminus and the transcription was driven by the CMV promoter . After transfection of 293T cells with either one of the FLAG-tagged expression constructs , we quantified mA3 transcripts by qPCR and mA3 protein by immunoblotting ( Figure 2 , B–D ) . Based on the co-expressed luciferase activities , similar transfection efficiencies for the 5+ and Δ5 plasmids were confirmed , and the data from the qPCR showed that the 5+ mA3 transcript was more highly expressed than the Δ5 mRNA for both alleles . Nevertheless , the expression levels of the 5+ mA3 protein were lower compared to those of the Δ5 counterpart regardless of the allelic differences , although B6 Δ5 protein was expressed more abundantly than BALB/c Δ5 ( Figure 2 , B and C ) . These results indicate that the inclusion of exon 5 does not compromise mA3 mRNA expression , but decreases the steady-state levels of mA3 proteins for both allelic variants . As higher mRNA levels of the 5+ mA3 resulted in lower protein levels in comparison with those for the Δ5 isoform in transfected cells , we next examined the steps at which the synthesis or degradation of mA3 protein was affected by the presence of exon 5 . To this end , we transfected 293T cells with an expression plasmid harboring the 5+ or Δ5 mA3 cDNA , and cycloheximide was added to stop protein synthesis . The inhibitor of nuclear factor-κB , IκBα , was utilized as a control , as this protein is known to be in the process of constant degradation and regeneration under serum-containing culture conditions [41] . As expected , immunodetectable IκBα decreased upon cycloheximide treatment , while the solvent alone did not affect the protein content ( Figure 3A ) . On the other hand , mA3 expressed in the same cells did not exhibit any reduction upon cycloheximide treatment regardless of the presence or absence of exon 5 . Further , the Δ5 protein was again detected at higher levels than the 5+ protein . These results support the conclusion that the effect of exon 5 inclusion in reducing the amount of expressed mA3 protein is not due to accelerated protein degradation . We next utilized an in vitro transcription and translation procedure to determine if exon 5 affects the translation of mA3 ( Figure 3B ) . When the 5+ or Δ5 mA3 templates were added to the in vitro transcription and translation reaction , similar amounts of each transcript were detected by RT-PCR assays both at 30 min and 60 min after the beginning of incubation . However , the tempos of appearance and amounts of the 5+ mA3 protein were different from those of the Δ5 counterpart: a large amount of the Δ5 protein was detected at as early as 30 min after the beginning of incubation , while the 5+ mA3 was undetectable at the same time-point . The 5+ mA3 protein became detectable after 60 min of incubation , but its protein level was still markedly lower than that of the Δ5 counterpart . These results collectively indicate that the inclusion of exon 5 modulates the translation efficiency of mA3 rather than its protein degradation . As the inclusion of exon 5 is associated with a reduced level of mA3 protein , we next attempted to identify genetic polymorphisms that affect the splicing patterns of mA3 transcripts in terms of exon 5 inclusion . One possible allelic difference in the mouse Apobec3 locus putatively associated with its splicing patterns is the possible pre-mRNA branch site polymorphisms found in the intron upstream of exon 5: a T/C SNP that lies within a preferred branch site sequence , TA ( T/C ) CAAC , and TCCT repeat numbers between this and the acceptor site [38] . As described previously [38] , the intron 4 of the BALB/c allele contains a tandem repeat of the TCCT sequence near the intron 4/exon 5 boundary , while the B6 allele contains only a single TCCT copy , changing the length of the putative pyrimidine-rich lariat intermediate . The adjacent T/C SNP at position 741 from the first nucleotide of exon 4 is in linkage with the TCCT repeat number . To directly investigate the possible effect of these polymorphisms , we constructed splicing assay plasmids that harbor the Apobec3 genomic fragment encompassing exons 4 through 7 from either the B6 or BALB/c allele . The resultant plasmids were designated B6 exon 4–7 and BALB exon 4–7 ( Figure 4 ) . Three mutants of each of these constructs were made with different combinations of the position 741 T/C SNP and the TCCT copy number variation as depicted in Figure 4B . Because of the possible presence of species-specific regulatory factors , the resultant genomic constructs were transfected into BALB/3T3 instead of 293T cells along with the luciferase expression plasmid as a control for transfection efficiency . To avoid the amplification of the endogenous mA3 message expressed in BALB/3T3 cells [27] , we utilized the primers g and h ( Figure 4B ) for RT-PCR assays , which were designed to hybridize to the T7 promoter and V5 tag regions of the expression vector . Transfection with the B6 exon 4–7 plasmid resulted in the expression of only the Δ5 transcript , while the BALB exon 4–7 generated both the 5+ and Δ5 transcripts with much higher intensity of the 5+ one ( Figure 4C ) , reproducing the splicing patterns observed in B6 and BALB/c spleens , respectively ( Figure 1B ) . The ratio between the 5+ and Δ5 transcripts was reduced in the samples transfected with BALB ΔTCCT or BALB C741T ΔTCCT plasmid harboring only a single copy of TCCT , which was further confirmed by utilizing the primer i hybridizing to the sequence within exon 5 along with primer h ( Figure 4C ) . Quantitative real-time PCR analyses were done by utilizing another primer ( primer j ) , designed to hybridize with the sequence within exon 5 , and further confirmed reduced levels of the exon 5-containing message expressed from the modified BALB exon 4–7 constructs lacking the TCCT repeat . These results imply that the TCCT repeat observed in the BALB/c allele , but not the T/C substitution at position 741 , at least partly facilitates the exon 5 inclusion . The lack of an effect of the T/C SNP within the putative branch site sequence was further confirmed by the abundant expression of the 5+ message from the BALB C741T construct . However , both the 5+ and Δ5 transcripts were still produced from the BALB ΔTCCT plasmid , suggesting that the TCCT repeat number is not the only determinant controlling the exon 5 inclusion . In fact , despite the presence of the repeated TCCT , B6 +TCCT and B6 T741C +TCCT did not generate the 5+ message , while transfection with the corresponding BALB C741T and the wild-type BALB exon 4–7 clearly resulted in the generation of the 5+ mRNA . These results indicate that polymorphisms other than those in intron 4 might play more important roles in determining exon 5 inclusion into mA3 messages . As abundant expression of the exon 5-containing message was found with the BALB C741T but not with the B6 +TCCT construct , we were prompted to examine other polymorphisms downstream of intron 4 that might be involved in the splicing of exon 5 . To explore other possible sequence variations required for exon 5 inclusion , we focused on intron 5 , since this region contains numerous polymorphisms between reported genomic sequences of different mouse strains ( Figure 5A ) . Splicing assay vectors harboring the genomic DNA fragment containing exons 5 and 6 along with the entire intron 5 , from either the B6 or the BALB/c allele , were constructed and designated B6 exon 5–6 or BALB exon 5–6 ( Figure 5B ) . BALB/3T3 cells transfected with the BALB exon 5–6 plasmid generated the properly spliced product of the expected size ( Figure 5C ) , indicating that the splice donor site downstream of exon 5 and the acceptor site upstream of exon 6 in the BALB/c allele are both functional . In contrast , no spliced product was detectable in cells transfected with the B6 exon 5–6 plasmid despite a higher transfection efficiency . These results indicate that the B6 allele may not carry a functional splice donor site at the exon 5/intron 5 boundary , as the splice acceptor site upstream of exon 6 seems intact and thus can generate a message corresponding to the mA3 Δ5 when exon 4 and intron 4 are included ( Figures 4 and 5C ) . To narrow down the region affecting the intron 5 splicing , serial deletion constructs were produced and subjected to the splicing assay ( Figure 5B ) . The intron 5-Δ3′ is a deletion construct which lacks the 3′ half of intron 5 but retains the 261-bp sequence adjacent to exon 6 to include the putative acceptor site . The cells transfected with the BALB intron 5-Δ3′ plasmid produced the spliced message as efficiently as those transfected with the parental BALB exon 5–6 construct; however , the B6 intron 5-Δ3′ plasmid generated a barely detectable band representing spliced message ( Figure 5D ) . Quantitative real-time PCR assays revealed markedly lower levels of properly spliced message generated from the B6 intron 5-Δ3′ plasmid compared to that generated from the BALB/c counterpart . These data show that the 5′ half of intron 5 is responsible for the observed differences in splice site functions between the B6 and BALB/c alleles . Thus , further deletions progressively closer to exon 5 were introduced into the intron 5 fragment . As expected , the shorter the included intron 5 fragment , the more abundant the spliced product generated was , regardless of the expressed Apobec3 alleles , indicating that the intron length of the primary transcript influences the splicing efficiency ( Figure 5D ) . Nevertheless , the spliced product was generated much more efficiently from the BALB/c allele than from the B6 allele with all examined construct pairs , despite comparable levels of transfection efficiency , indicating that each shorter intron fragment still retained the polymorphism responsible for the inefficient splicing of the B6 mRNA . To exclude the possibility that the 3′ region of intron 5 and/or exon 6 from the B6 allele might harbor inhibitory sequences that interfere with splicing , we constructed chimeras between the BALB and B6 exon 5–6 plasmids by exchanging the corresponding genomic DNA fragments at approximately position 1100 within intron 5 ( Figure 5E ) . BALB/B6 chimera harboring the 5′ donor site sequence from the BALB/c allele and 3′ acceptor site from the B6 allele generated properly spliced mRNA as efficiently as the BALB exon 5–6 plasmid did , while the reciprocal construct harboring the B6 donor and BALB/c acceptor sequences did not ( Figure 5E ) , indicating that there are no inhibitory elements in the fragment harboring the 3′ intron 5 and exon 6 from the B6 allele . Thus , we continued to narrow down the sequence affecting the intron 5 splicing toward the 5′ end of the genomic constructs . Even the shortest construct pair , B6 100bp intron 5 and BALB 100bp intron 5 , which harbor the 5′ intron 5 fragment a mere 100-bp from the exon 5 boundary , still exhibited a readily discernible difference in the amounts of the spliced messages , and this was also true for another pair of deletion constructs that harbor a 100-bp acceptor region fragment of 3′ intron 5 ( Figure 5F ) . These results clearly indicate that the region including exon 5 and the 5′ 100-bp of intron 5 carries the primary determinants responsible for the different efficiencies in splicing of intron 5 shown by the B6 and BALB/c alleles . Within the above narrowed-down region of exon 5 and the 5′ 100 bp of intron 5 , there are only 4 SNPs between the B6 and BALB/c alleles , as previously described [38]: the T/C SNP at 14-bp downstream from the first nucleotide of exon 5 , G/C at 88-bp downstream from the same first nucleotide , G/C at 153-bp downstream within intron 5 , and A/C at 163-bp downstream within intron 5 ( Figure 6A ) . We therefore produced a series of point mutants based on the B6 intron 5–100 ( 3′-100bp ) and BALB/c intron 5–100 ( 3′-100bp ) constructs to precisely identify the critical SNP responsible for the functionalities of the splice donor site . In the first set of experiments , shown in Figure 6B , each indicated nucleotide within the BALB/c genomic sequence was substituted with the corresponding nucleotide found in the B6 allele . The substitution of the C at position 14 to the B6-type T slightly reduced the amount of the spliced message; more importantly , however , the similar substitution of C at position 88 to the B6-type G totally abrogated the intron 5 splicing . Simultaneous substitutions of C to G at position 153 and G to A at 163 did not affect the generation of the spliced message . Transfection efficiencies were essentially equivalent for all samples and even higher than those for the C88G construct as confirmed by cotransfecting the luciferase expression plasmid ( data not shown ) . These results indicate that the G/C SNP at position 88 in exon 5 is most critical for the splice donor function . In a reciprocal set of experiments , we attempted to alter splicing of the transcript of the B6 allele by replacing the polymorphic nucleotides with the BALB/c-types ( Figure 6C ) . When all 4 SNPs in the B6 intron 5–100 fragment were replaced with those corresponding to the BALB/c allele , abundant generation of the spliced message was observed ( Figure 6C ) . A combined substitution of T to C and G to C at positions 14 and 88 , respectively , also resulted in efficient splicing of intron 5 , and a single substitution of T to C at position 14 alone slightly increased the generation of the spliced message from the plasmid harboring the truncated intron . Importantly , a single substitution of G to C at position 88 resulted in as abundant expression of the spliced message as that observed with the BALB/c allele . Thus , these results collectively indicate that the G at position 88 is critical for the exclusion of exon 5 , although the C at position 14 may also play a minor role in the inclusion of this exon . To determine the association of the above 4 SNPs with splicing efficiency under more physiological conditions , we further introduced point mutations into the B6 exon 5–6 plasmid harboring the entire intron 5 of 6kb in length , and the resultant plasmids were subjected to the splicing assays ( Figure 7A ) . In contrast to the result with the B6 100bp intron 5 T14C mutant ( Figure 6C ) , we could not detect the generation of spliced mRNA after transfection of the B6 exon 5–6 T14C mutant ( Figure 7B ) . However , the single substitution at position 88 from the B6-type G to BALB/c-type C led to readily detectable production of the spliced mRNA , despite the general inefficiency in splicing of the full-length intron 5 depicted in Figure 5 . Quantitative real-time PCR assays revealed levels of expression of the exon 5-containing spliced message from the B6 exon 5–6 G88C mutant that were about 70% of those expressed from the BALB exon 5–6 plasmid . Single nucleotide substitutions at two other positions did not result in a detectable generation of the spliced mRNA from the B6 allele ( Figure 7B ) . Combined mutations at positions 14 and 88 within exon 5 on the B6 exon 5–6 construct ( B6 exon 5–6 T14C G88C ) and further addition of the nucleotide replacements at positions 153 and 163 also resulted in detectable generation of the spliced message , but did not noticeably enhance the splicing over what resulted from the G88C substitution alone ( Figure 7C ) . To further compare the effects of the TCCT repeat number and exon 5 G/C SNP at position 88 , we constructed B6 and BALB exon 4–7 plasmids that harbored reciprocal substitutions at these two polymorphic sites ( Figure 8 ) . Introduction of an additional copy of TCCT into the B6 exon 4–7 construct did not result in the generation of 5+ message , while the combination of the additional TCCT and G88C substitution resulted in the generation of exon 5-containing message at levels comparable to those expressed from the BALB exon 4–7 plasmid . On the other hand , a deletion of a TCCT copy from the BALB exon 4–7 construct resulted in much reduced expression of the 5+ message , and the combination of ΔTCCT and C88G substitution totally abrogated the generation of exon 5-containing message . Thus , these results clearly demonstrate that the most critical polymorphism for exon 5 inclusion into the mA3 mRNA is the G/C SNP at position 88 within exon 5 , but for the full-level expression of the 5+ mRNA as observed in BALB/c mice the intron 4 TCCT repeat is also required . Because the inclusion of exon 5 into mA3 message depends on the intron 4 TCCT repeat and the exon 5 G/C polymorphism at position 88 ( G/C88 ) , we screened additional inbred strains and wild mouse species for these sequence variations and for the presence or absence of exon 5 in the mA3 message . We sequenced exon 5 and segments of the flanking introns from 39 mice that represent different taxa or members of the same species trapped in different geographic locations ( Table S1 ) , as well as those from the inbred laboratory strains B10 . A and A/WySn , prototypic strains with FV-restrictive and -permissive phenotypes [26] , [27] , [31] , [32] . The complete dataset is shown in Figure S1 . Genomic sequencing showed that A/WySnJ mice share the TCCT duplication and exon 5 C88 SNP with BALB/c , while B10 . A mice are identical to B6 mice at both sites ( Figure 9A ) ; this is consistent with previous observations on mA3 expression levels and inclusion of exon 5 in these strains of mice [27] . Among the 39 wild mice , the TCCT duplication was found in 10 mice that possessed the exon 5 C88 SNP , and all of the 29 other mice with G88 had a single TCCT copy as representatively shown in Figure 9A . The observed linkage between the repeated TCCT and C88 is reasonable , as both are required for efficient expression of 5+ message from the BALB/c allele ( Figures 4 , 6 , and 8 ) . RNA samples from 23 of the above mice were typed for the splicing phenotype . The presence or absence of exon 5 in mA3 mRNA correlated with the above identified sequence polymorphisms: exon 5 was present only in 7 mice , all of which had the linked TCCT duplication and C88 , consistent with our functional assays . On the other hand , there were at least 7 discrepancies with respect to exon 5 inclusion and the C14 SNP confirming that this SNP has no major role in exon 5 splicing . The coding sequence for exon 5 was present in all 39 Mus DNA specimens , as well as in rat DNA specimens ( GenBank EDM15775 ) as described previously [29] . There were frameshift mutations within exon 5 in M . setulosus and both M . cervicolor subspecies , all of which lack exon 5 in mA3 mRNA , and the exon 5-containing splice variant was found only in mA3 mRNA of M . spretus and 3 of the house mouse species ( M . domesticus , M . musculus , M . castaneus ) ( Figure 9B ) . The wild mice were also tested for the presence or absence of another genetic variation associated with variable expression levels of mA3 , the exon 2-associated xenotropic MuLV LTR insertion [37] . Results indicated that the presence of this LTR ( LTR+ ) , like the inclusion of exon 5 due to the above allelic variation ( exon 5+ ) , was introduced into the Apobec3 locus at about the time of the house mouse radiation ( Figure 9B ) . These two features are not found together in any mouse , and the LTR+ and exon 5+ variants are both found in mice classed as M . musculus or M . domesticus , while M . castaneus mice have both variants as well as the ancestral Δ5 LTR− mA3 type . To better understand this species distribution , we typed additional samples of European M . domesticus and M . musculus for the presence of the LTR and exon 5 ( Figure 9C ) . The geographic distribution of the trapping sites for these 19 house mice and for the exon 5+ M . spretus indicated that the exon 5+ mA3 was characteristic of M . domesticus of north Africa and western Europe , whereas LTR+ mA3 was found in eastern Europe ( P = 0 . 049 by Fisher's exact test ) . Four of the 9 eastern European M . musculus mice carried the M . domesticus variant , which is consistent with previous reports that gene flow across the hybrid zone is biased in the direction from M . domesticus to M . musculus ( [42] , for example ) . Additional samples identified as M . domesticus were trapped in the Americas ( HAF , CL , WSA , PGN2 , SC1 , BQC , SAF , JJD , PERA; see Table S1 ) , and these carried either the exon 5+ mA3 ( like M . domesticus ) or the LTR+ mA3 ( like M . musculus ) ; this is likely due to the fact that Mus are non-native species that were introduced into the Americas by passive transport , and although generally classed as M . domesticus , these mice show evidence of hybridization with other introduced house mouse species [43] , [44] .
Polymorphisms in mA3 are responsible for the fact that the B6 and C57BL/10 mouse strains are more restrictive to the replication of both beta- and gammaretroviruses than the BALB/c and A strains [25]–[29] . These differences in mA3 antiviral activities have been associated with sequence differences in the N-terminal region of mA3 [27] , different levels of its mRNA expression [27] , [29] , [38] , [39] , and the presence or absence of exon 5 in mA3 mRNA and protein [27] , although the relative importance of these three factors for antiviral activity has not been established . In the present study , we focused on the role of exon 5 and have shown that the Δ5 mRNA directs more efficient mA3 protein synthesis than the exon 5-containing message . We have also identified 2 genetic determinants responsible for the inclusion of this exon into mA3 , a TCCT repeat in intron 4 and a C88 substitution in exon 5 relative to the B6 allele . We further showed that these two determinants were coordinately acquired within the last 0 . 5 million years by house mouse species of Mus . Previous studies have recognized differences in mRNA expression levels of mA3 between B6 and BALB/c mice [27] , [29] , [38] , [39] . Higher constitutive expression of mA3 mRNA is thought to contribute to better antiviral activity as suggested by the facts that higher levels of protein production result in more efficient incorporation of mA3 into viral particles [45] and that lipopolysaccharide ( LPS ) -induced enhancement of mA3 expression results in better restriction of MMTV in vivo [46] . It has also been shown that increased APOBEC3G and APOBEC3F expression is associated with lower viral load in rhesus macaques infected with simian immunodeficiency virus [47] . Higher protein levels of hA3G in CD14+ monocytes are also associated with HIV-1-exposed but uninfected status in humans [48] . The increased level of Apobec3 gene transcription has been associated with the presence in the B6 allele of an intact MuLV LTR , a sequence capable of driving enhanced transcription [37] . While increased levels of the mA3 transcript could be solely responsible for the presently observed increase in mA3 protein levels in B6 mice ( Figure 1D ) , it was also reported that , when driven by the same strong promoter , the B6 mA3 cDNA lacking exon 5 produced more protein product than the exon 5-containing BALB/c cDNA and resulted in more efficient incorporation into viral particles [27] , [45] . Thus , it was unclear whether the experimentally observed increased translation was due to the lack of exon 5 in the B6 cDNA or to the other sequence differences that distinguish the two Apobec3 alleles [27] , [37] , [38] . Here we have clearly shown that the Δ5 mRNA directs much more efficient protein synthesis than the 5+ mA3 mRNA regardless of other allelic differences ( Figures 2 and 3 ) , and the combination of the higher mRNA expression levels with preferential generation of the Δ5 isoform in mRNA splicing results in much higher mA3 protein levels in B6 in comparison with BALB/c mice ( Figure 1 ) . The mechanisms by which the presence of exon 5 in the mRNA interferes with protein synthesis are currently unclear: however , predictions of the secondary structure for the mA3 mRNA suggest that the portion encoding exon 5 may form stable stem-loop structures ( Figure S2 ) , which might interfere with efficient translation . The predicted stem-loop structure is thermodynamically more stable for B6 than for BALB/c exon 5 , and this portion is actually spliced out from the major transcript in B6 mice . It is possible , therefore , that the elimination of this exon from mA3 mRNA confers a functional advantage to B6 mice through more efficient translation of the Δ5 mRNA and resultant higher levels of mA3 protein expression . It should be noted , however , that introduction of this exon between the two CDDs does not physically disrupt the functional domains , as the 5+ product of the BALB/c allele still exerts deaminase activity and restricts AKV replication in vitro [29] . The possible effect of this 33-amino acid exon on interactions between the N-terminal and C-terminal CDDs and on mA3 oligomerization remains to be elucidated . Our results also show that the ancestral Mus Apobec3 gene locus lacked both the MuLV LTR and the functional polymorphisms that determine exon 5 inclusion ( Figure 9 ) . These two genetic features were acquired independently at about the time of the house mouse radiation 0 . 5–1 . 0 million years ago , and are found in distinct lineages of wild mice and different inbred strains of laboratory mice . It should be noted that all 4 house mouse species originated and diverged from an ancestral population on the Indian subcontinent that carries many of the alleles found in peripheral Eurasian populations [49] , [50] . The observed distribution of the LTR+ and exon 5+ mA3 variants among the house mouse lineages is consistent with the appearance of these variants in this ancestral population . It is thus likely that M . domesticus progenitors carrying the exon 5+ Apobec3 allele moved west to north Africa and western Europe , and M . musculus ( LTR+ ) moved north to Russia , eastern Europe and north China . M . castaneus , which carries these two variants as well as the ancestral Δ5 LTR− mA3 , migrated east to Thailand and China . Only the ancestral mA3 type is found in M . molossinus , which is a natural hybrid of M . musculus and M . castaneus [44] , and these Japanese mice presumably inherited their Apobec3 gene from their M . castaneus progenitors . Numerous studies indicate that host antiviral factors co-evolve with the pathogens they restrict , and this “arms race” is responsible for mutational changes in APOBEC3 in primates and mice as evidenced by detectable positive selection [37] , [51] . The acquisition of the MuLV LTR that is associated with enhanced mA3 transcription in virus-infected mice also makes sense as an evolutionary adaptation to pathogen infection [37] . Thus , the B6 Apobec3 allele has acquired two advantageous features that contribute to enhanced retrovirus restriction: high mRNA expression directed by the intron 2 LTR that is in linkage with the positively selected , more functional amino acid sequence in the N-terminal deaminase domain [27] , [37] . The BALB/c allele , in contrast , acquired the polymorphisms that direct exon 5 inclusion as shown in the present study , along with coding sequence polymorphisms associated with lower antiviral activity [27] . The far-flung distribution of exon 5+ mA3 throughout western Europe ( Figure 9C ) is surprising , because this allele , having a reduced level of mA3 protein expression with inefficient restriction of MuLV replication at least in vitro [27] , would seem to be evolutionarily deleterious . The observed distribution of exon 5+ mA3 suggests either that the exon 5+ variant provides sufficiently protective levels of antiviral activity in virus-infected mice and is therefore not subject to purifying selection , that these particular mice have not been exposed to significant challenge by mA3-sensitive pathogens , that this mA3 variant provides some other , unrecognized selective advantage , or that this allele is tightly linked to another advantageous polymorphism . All 4 house mouse species possessing the exon 5+ genetic variation carry beta- and gammaretroviruses along with other retroelements that are subject to APOBEC3 restriction [25]–[30] , [52] . In this regard , although the BALB/c mA3 is less antiviral than the product of the B6 allele , it does have measurable antiviral activity [29] . Further , analysis of 54 germline MuLV proviruses of three envelope types in the sequenced B6 genome demonstrated evidence of APOBEC3-mediated editing of the polytropic and modified polytropic , but not the xenotropic , proviruses [30] . The mA3-edited polytropic and modified polytropic viruses originated in M . domesticus [52] , [53] , mice that carry the exon 5+ mA3 , suggesting that this mA3 variant may be effectively antiviral in vivo . Finally , in mA3-deficient B6 mice , the absence of this protein differentially affects FV replication depending on target cell types [54] . As cell type-dependent and pathogen-induced changes in mA3 expression may differ between mice of the ancestral Δ5 LTR− , exon 5+ , and LTR+ variants , mice with the exon 5+ Apobec3 allele may still restrict mA3-sensitive pathogens in critical target cells . It is also possible that some mA3 polymorphisms have been selected for reasons other than their antiviral functions , and these functions may take precedence in mice not threatened with retrovirus assault . It is unlikely that differences in mA3 expression levels and/or its amino acid sequence provide a significant selective advantage in terms of normal development , survival , or fertility , as Apobec3-knockout mice show no increased propensity for tumor development or disease , and both male and female mA3-deficient mice were fertile [40] . On the other hand , the presence of active cytidine deaminases can have costs that may become more significant when endogenous and exogenous retroviral activity is low . Transgenic mice overexpressing mouse activation-induced deaminase ( AID ) or rabbit apolipoprotein B mRNA-editing enzyme catalytic polypeptide 1 ( APOBEC1 ) , two other members of the AID/APOBEC family of cytidine deaminases , developed neoplastic diseases and showed evidence of significant editing of various expressed genes [55] , [56] . This suggests that when the retroviral threat is low , the consequences of possessing highly active mutagenic enzymes , like the Δ5 mA3 expressed in B6 mice , can outweigh their advantages as antiviral factors . In this regard , it should be noted that infectious MuLV has not been isolated from European M . domesticus , which carry the less highly expressed 5+ mA3 , whereas the more active Apobec3 alleles are found in all 3 of the other species of Mus , all of which have been found to carry infectious MuLVs [57] . Finally , it has also been observed that there is a similarly wide-spread distribution in humans of a deletion of a genomic segment harboring the Apobec3b gene locus [58] . At least one copy of this deletion is present in >40% of humans . The selective advantage of this genetic variant has not been determined , but its retention may suggest a possible advantage in reducing the genotoxic activity of this cytidine deaminase , or alternatively , it may be linked to positively selected variants in one or more of the 6 linked human APOBEC3 paralogues [59] , [60] . Further comparative analyses of the geographic and species distributions of natural mouse pathogens and the various mA3 variants may help define previously unrecognized targets of APOBEC3-mediated restriction and provide further insight into the coevolution of pathogens and this host restriction factor .
The studies utilizing laboratory animals were carried out in strict accordance with the Act on Welfare and Management of Animals of the government of Japan and the Regulations for the Care and Use of Laboratory Animals of Kinki University . The protocol was approved by the Institutional Animal Experimentation Committee of Kinki University School of Medicine ( Permit Number: KAME-19-029 ) . All surgery was performed under sodium pentobarbital anesthesia , and all efforts were made to minimize suffering . All studies involving wild mice were performed in compliance with the US Government Principles for the Utilization and Care of Vertebrate Animals used in Testing , Research , and Training; the Public Health Service Policy on Humane Care and Use of Laboratory Animals; The Animal Welfare Act and amendment laws; the Animal Care Policies of the USDA; The Guide for the Care and Use of Laboratory Animals ( 7th Edition; National Research Council ) ; and the guidelines of the Committee on the Care and Use of Laboratory Animals under an NIAID-approved animal study protocol , and all studies and procedures were reviewed and approved by the Institutional Animal Care and Use Committee of the NIH ( Permit Number: ASP LMM 1 ) . C57BL/6NCrslc , BALB/cCrslc , and B10A/SgSnslc mice were purchased from Japan SLC , Inc . , Hamamatsu , Japan . Breeding pairs of A/WySnJ mice were purchased from The Jackson Laboratory , Bar Harbor , ME . The mA3-deficient strain on the B6 background has been described [27] , [40] . All laboratory mice were housed and bred in the Experimental Animal Facilities at Kinki University School of Medicine under specific pathogen-free conditions . The isolation of genomic DNA from spleens was carried out with DNeasy blood and tissue kit ( Qiagen , Inc . , Hilden , Germany ) according to the manufacturer's instructions . DNA and RNA were separately isolated from animals and cell lines developed from wild mice and wild mouse-derived breeding colonies or inbred strains ( Table S1 ) . Many wild-derived mice were obtained from M . Potter ( National Cancer Institute , Bethesda , MD ) . CAST/Rp mice were obtained from R . Elliott ( Roswell Park Cancer Institute , Buffalo , NY ) . Cells from some wild mouse species were obtained from J . Rodgers ( Baylor College of Medicine , Houston , TX ) and from J . Hartley , M . Lander or S . Chattopadhyay ( National Institute of Allergy and Infectious Diseases , Bethesda , MD ) [61] , [62] . M . cervicolor popaeus mice and tissue samples were obtained from R . Callahan ( National Cancer Institute ) . Mice or DNA samples of inbred lines of M . castaneus ( CAST/EiJ ) and M . molossinus were obtained from The Jackson Laboratory . M . musculus DNA samples were obtained from S . Chattopadhyay and H . Morse ( National Institute of Allergy and Infectious Diseases ) . DNA samples from wild-trapped European M . domesticus were provided by M . Nachman ( University of Arizona , Tucson ) . DNA samples from 5 wild-derived strains ( BLG2 , NJL , MSM , HMI , PGN2 ) were obtained from the National Institute of Genetics , Mishima , Japan . A set of Nannomys DNAs was obtained from Y . Cole and P . D'Eustachio ( Departments of Biochemistry and Medicine , New York University , NY ) ; these mice were classed into 4 species on the basis of skeletal features by J . T . Marshall ( Smithsonian Natural History Museum , Washington , DC ) . DNA was isolated from cultured tail biopsies , spleen , or liver by standard protocols , and RNA was isolated from the spleen or liver using TRI-Reagent ( Molecular Research Center , Cincinnati , OH ) or by a guanidine chloride extraction method [63] . DNA containing the mouse Apobec3 exon 5 and associated intron sequences was amplified using either one of the following forward primers: 5′-GGACAATGGTGGCAGGCGATTC-3′ , 5′-GCATCTTTGTGGATGGGG-3′ , and the reverse primer 5-TCATTCCTCAATGCTCCTCC-3′ . PCR products were cloned into pCR2 . 1-TOPO ( Invitrogen , Carlsbad , CA ) before sequencing . The Apobec3 region was amplified from genomic DNA of B6 , BALB/c , B10A , and A/WySnJ mice using the forward primer 5′-TTACAAATTTTAGATACCAGGATTCTAAGCTTCAGGAG-3′ and the reverse primer 5′-GTCCTTTATGTGGGTTCCAAGGACC-3′ . PCR products were treated with ExoSAP-IT ( USB , Cleveland , OH ) and directly sequenced with the above reverse primer . To determine the BALB/c genomic sequence of the Apobec3 intron 5 , the BALB exon 5–6 and BALB intron 5-Δ3′ plasmids ( Figure 5 ) were sequenced by using BigDye Terminator V3 . 1 Cycle Sequencing Kit ( Applied Biosystems , Foster City , CA ) and an ABI PRISM 3100 Genetic Analyzer ( Applied Biosystems ) using the following primers: T7 promoter forward , 5′- TAATACGACTCACTATAGGG-3′; and V5 reverse , 5′-CGTAGAATCGAGACCGAGGAGAGGGTTAGGGATAGGC-3′ . The secondary structure of a portion of the mA3 mRNA exon 5 was predicted with mfold [64] , [65] . BALB/3T3 and human 293T cells were cultured in Dulbecco's modified Eagle medium supplemented with 10% heat-inactivated fetal bovine serum ( Invitrogen ) . These cells were seeded at 1 . 0×105/well in a well of 6-well plates one day prior to transfection . DNA transfection was performed by using Lipofectamine 2000 ( Invitrogen ) according to the manufacturer's protocols . Mouse spleen cells were harvested as described previously for protein and RNA extractions [27] . At least three independent transfection experiments were performed in this study and representative results are shown in the figures . The expression plasmids , pFLAG-CMV2-mA3bΔ5 harboring the mA3 cDNA derived from the Δ5 transcript of the B6 allele , pFLAG-CMV2-mA3d harboring the cDNA derived from the 5+ transcript of the BALB/c allele , pFLAG-CMV2-mA3dΔ5 harboring the cDNA derived from the Δ5 transcript of the BALB/c allele , and control pFLAG-CMV2-GFP have been described [27] . The genomic DNA encoding the mA3 exon 5 was amplified from B6 genome with the following primers: 5′-ACCTTGCTACATCTCGGTCCCTTCCAGC-3′ and 5′-CTGCCCTCCACCCAGAACCTCGTCTCTGG-3′ . The above pFLAG-CMV2-mA3bΔ5 was used as a PCR template with primers 5′-GCGAATGGACCCGCTAAGTGAAGAGG-3′ and 5′-CTCAGAATCTCCTGAAGCTTAGAATCCTGG-3′ to amplify the linearized plasmid with a gap between exons 4 and 6 . The two PCR products above were treated with T4 polynucleotide kinase ( TAKARA Bio , Otsu , Japan ) and fused by using the DNA Ligation Kit ver . 2 . 1 ( TAKARA Bio ) to construct the plasmid pFALG-CMV2-mA3b , which expresses the 5+ mA3 cDNA derived from the B6 allele . B6 or BALB/c genomic fragments harboring exons 4–7 and the intervening introns were amplified by PCR using either the B6 or BALB/c genomic DNA as a template and the common primers 5′-CACCAATTTAAAAAGTGTTGGAAGAAG-3′ and 5′-GTGGGAGGTCCATGACGTCCACCAGGATCCC-3′ . Each amplified DNA product was cloned into a pcDNA3 . 2/V5/GW/D-TOPO cloning vector ( Invitrogen ) and designated as B6 or BALB exon 4–7 , respectively . The above B6 or BALB exon 4–7 was used as a template with the primers 5′-CACCACCTTGCTACATCTCGGTCC-3′ and 5′-GCAGAGATGCTTGACTCGTTGGTTG-3′ or 5′-CACCACCTTGCTACATCCCGGTCC-3′ and 5′-GCAGAGATGCTTGACTCGTTGGTTG-3′ , respectively , for the amplification of B6 or BALB/c exons 5 and 6 and the intervening intron 5 . Each PCR product was cloned into the pcDNA3 . 2/V5/GW/D-TOPO vector and designated as B6 or BALB exon 5–6 . The DNA fragments harboring sequentially deleted Apobec3 intron 5 were prepared by PCR using either one of the above B6 or BALB exon 5–6 plasmids as a common template with the primer pairs A–F listed in Table S2 . Each amplified DNA product was cloned into the pcDNA3 . 2/V5/GW/D TOPO vector to generate the expression plasmid shown in Figure 5B . Reciprocal chimeras between the above B5 and BALB exon 5–6 plasmids were generated by amplifying a linearized plasmid DNA lacking the 3′ intron 5 and exon 6 using primer pair G , and by amplifying the insert fragment using primer pair H . Site-directed mutagenesis was performed by employing the QuikChange Site-Directed Mutagenesis Kit ( Stratagene , La Jolla , CA ) using the following templates and primers , listed separately in Table S2: BALB 100bp intron 5 ( 3′-100bp ) plasmid was used as a common template for the preparation of BALB C14T , BALB C88G , and BALB C153GG163A with primer pairs I , J , K , respectively: B6 100bp intron 5 ( 3′-100bp ) plasmid was used as a template to make B6 T14C or B6 G88C with primer pair L or M , respectively . The resultant B6 T14C plasmid was used as a template with the same primer pair M employed for the generation of B6 G88C to make B6 T14C G88C . B6 T14C G88C was then used as a template for the generation of B6 T14C G88C G153C A163G by using primer pair O . Similarly , the above-used primer pairs were also utilized for the generation of B6 exon 5–6 T14C , B6 exon 5–6 T14C G88C , and B6 exon 5–6 T14C G88C G153C A163G mutants by using B6 exon 5–6 as a template . The B6 exon 5–6 plasmid was also used as a template with primer pairs M , P , and Q for making B6 exon 5–6 G88C , B6 exon 5–6 G153C , and B6 exon 5–6 A163G , respectively . To make BALB ΔTCCT , 4 consecutive nucleotides within the repeat sequence , TCCT , were deleted from the BALB exon 4–7 plasmid by mutagenesis using primer pair R , listed in Table S2 . BALB ΔTCCT or BALB exon 4–7 plasmid was used as a template with primer pair S to make BALB C741T ΔTCCT or BALB C741T , respectively . To introduce the TCCT sequence by mutagenesis and make B6+TCCT , B6 exon 4–7 plasmid was used as a template with primer pair T . B6+TCCT or B6 exon 4–7 was used as a template with primer pair U to generate B6 T741C+TCCT or B6 T741C , respectively . Similarly , the above produced B6 exon 4–7+TCCT or BALB exon 4–7 ΔTCCT plasmid was used as a template and G88C or C88G substitution was introduced with primer set M or J , respectively . All resultant plasmids were entirely sequenced by using BigDye Terminator V3 . 1 Cycle Sequencing Kit with an ABI PRISM 3100 Genetic Analyzer . In order to normalize the transfection efficiency , a plasmid expressing the luciferase gene , pluc , based on the expression vector pGL3 ( Promega , Madison , WI ) , was utilized . All the primers used in this study were purchased from Operon Biotechnologies , Tokyo , Japan . For the quantification of mA3 transcripts in mouse spleens , total RNA was extracted from each spleen with RNeasy Mini Kit ( Qiagen ) . The RNA was then subjected to reverse transcription with PrimeScript RT reagent Kit ( TAKARA Bio ) . Real-time PCR reactions were carried out with SYBR Premix Ex Taq II ( TAKARA Bio ) on an Applied Biosystems 7900HT Fast Real-Time PCR System ( Applied Biosystems ) with two different sets of Apobec3-specific primers: set 1 , 5′-GTGTTGGAAGAAGTTTGTGG-3′ ( primer a ) and 5′-CCTGAAGCTTAGAATCCTGG-3′ ( primer b ) ; and set 2 , 5′-TTACAAATTTTAGATACCAGGATTCTAAGCTTCAGGAG-3′ ( primer c ) and 5′-TTGGTTGTAAAACTGCGAGTAAAATTCCTCTTCAC-3′ ( primer d ) . The data were normalized with expression levels of β-actin mRNA to obtain ΔCt values , and ΔΔCt values were calculated . For the detection of endogenous mA3 transcripts in mouse spleen cells , the RT products and primer sets used for the real-time PCR were also utilized for PCR using KOD Dash DNA polymerase ( Toyobo , Osaka , Japan ) . The PCR products were separated by 1% agarose gel electrophoresis and detected by staining with ethidium bromide . For splicing assays , BALB/3T3 cells were transfected with 1 µg of each plasmid harboring a genomic DNA fragment and 0 . 5 µg of pluc to normalize the transfection efficiency . Total RNA was extracted from the transfected cells using the RNeasy Mini Kit at 24 hours post-transfection . The total RNA was treated with DNase I , reverse transcribed with SuperScript III First Strand Synthesis System ( Invitrogen ) , and the resultant cDNA was subjected to PCR detection using KOD Dash DNA polymerase with the following primers: 5′-TAATACGACTCACTATAGGG-3′ ( primer g ) and 5′-CGTAGAATCGAGACCGAGGAGAGGGTTAGGGATAGGC-3′ ( primer h ) , designed to hybridize the T7 promoter and V5 tag regions of the vector , respectively . Primer i , 5′-GGTCTCCCAGAGACGAGGTTCTG-3′ , was used to detect only the exon 5-containing transcript . For real-time PCR quantification of the mA3 transcripts containing exon 5 , primer j ( 5′-GTGGATGAAGAGCTGGAAGGGACCG-3′ ) was used along with the above primer g . Transcripts containing exon 6 were similarly quantified by using primer k ( 5′-CAACCAACGAGTCAAGCATCTCTGC-3′ ) and the above primer h . For RT-PCR detection of mA3 mRNA in cells transfected with an mA3 expression plasmid , the same RT product as was used for real-time PCR analyses was utilized with the above-described primer set 1 ( a–b ) . The RT products made by PrimeScript RT reagent Kit were supposed to be relatively short in length because a mixture of oligo-dT and random 6-mers were used as primers in the RT reaction . Thus , in order to detect the full-length and Δ5 transcripts , the following primers , 5′-GGGAATTCGATGGGACCATTCTGTCTGGGATGCAGCCATCGC-3′ ( primer e ) and 5′-GGGTCGACTCAAGACATCGGGGGTCCAAGCTGTAGGTTTCC-3′ ( primer f ) , were used along with newly synthesized RT products generated by the SuperScript III First Strand Synthesis System ( Invitrogen ) . To quantify the mA3 transcripts in 293T cells transfected with an mA3 expression plasmid , one-third of the transfected cells were used for total RNA isolation using the RNeasy Mini Kit at 24 hours after transfection . The purified RNA was treated with 5 units of DNase I ( TAKARA Bio ) for 1 hour at 37°C to digest the transfected DNA and then reverse transcribed with the PrimeScript RT reagent Kit . The real-time PCR was performed as described above . For RT-PCR analyses of mA3 transcripts generated by the in vitro transcription/translation system , the reaction was stopped by the addition of lysis buffer included in RNeasy Mini Kit after 30 or 60 min of incubation , and the transcribed products were purified with the above kit . After the treatment of the purified RNA with DNase I , the RT-PCR reaction was carried out with the above primer set 1 . After electrophoresis , gel images were recorded with a FluorChem™ IS-8900 transilluminator and band intensities were analyzed with AlphaEase FC Stand Alone software ( Alpha Innotech , San Leandro , CA ) . Apobec3 splicing patterns for wild-derived mice were identified by RT-PCR to amplify a segment of mA3 RNA spanning exon 5 from total RNA using forward primer 5′-GGACCATTCTGTCTGGGATGCAGCCATCG-3′ and reverse primer 5′-GGTTGTAAAACTGCGAGTAAAATTCC-3′ . Luciferase assays for normalization of transfection efficiencies were performed by utilizing the Luciferase Assay System ( Promega ) . The enzymatic activities were measured by Wallac 1420 ARVO™ MX-2 Multilabel Counter ( Perkin Elmer ) . A spleen from an Apobec3 knock-out mouse [40] was homogenized in 1ml ice-cold phosphate-buffered saline ( PBS ) . Four ml of ice-cold acetone was added to the homogenate , mixed , and incubated on ice for 30 min . The lysate was centrifuged at 10 , 000×g at 4°C for 10 min . The pellet was washed with ice-cold acetone once and dried completely at room temperature to make spleen extract powder . Two µl of anti-APOBEC3 NT antibody specific for the N-terminal portion of mA3 ( Millipore , Billerica , MA ) was added to 0 . 5ml of KTBT buffer ( 50 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 10 mM KCl , 1% Triton X-100 ) containing 10% ( w/v ) unimmunized sheep serum as a carrier , into which 3mg of the spleen extract powder was dissolved . The mixture was incubated overnight at 4°C with gentle rotation . After centrifugation at 10 , 000×g at 4°C for 10 min , the supernatant was used as a pre-absorbed anti-mA3 antibody . Western blotting analyses were conducted as described previously [27] with some modifications . Briefly , proteins were extracted with a lysis buffer ( 1% Nonidet P-40 , 25 mM Tris-HCl , pH 7 . 5 , 140 mM NaCl , 1 mM EDTA , 10 mM Na4P2O7 ) containing protease inhibitors from Complete , Mini , EDTA-free Protease Inhibitor Cocktail Tablets ( Roche Applied Science , Mannheim , Germany ) and a phosphatase inhibitor , PhosSTOP ( Roche Applied Science ) . Total protein concentrations were determined by Bradford assay ( Nacalai Tesque , Kyoto , Japan ) and the extracts were mixed with sodium dodecyl sulfate ( SDS ) -polyacrylamide gel electrophoresis ( PAGE ) sample buffer and heated at 95°C for 5 min . The proteins were separated by SDS-PAGE , transferred to Immobilon-P membrane ( Millipore ) , and the blotted membranes were blocked with 5% ( w/v ) skim milk ( Wako Pure Chemicals , Osaka , Japan ) in Tris-buffered saline with 0 . 05% Tween 20 ( TBST ) . The blocked membranes were incubated with the primary antibody at 4°C overnight . Membranes were then washed with TBST and incubated with horse radish peroxidase ( HRP ) -conjugated secondary antibody for 2 hours at room temperature , washed again with TBST , and the bound antibodies were detected using ECL plus reagent ( GE Healthcare , Tokyo , Japan ) . The images were captured with a LAS-1000 Plus ( Fujifilm , Tokyo , Japan ) and the band intensities evaluated with Image Gauge ver . 3 . 12 ( Fujifilm ) . A 1/5 dilution of the above-mentioned pre-absorbed mA3 antibody was made with IMMUNO SHOT ( COSMO BIO , Tokyo , Japan ) and was used as a primary antibody . Anti-FLAG M2 monoclonal antibody ( mAb ) ( Sigma-Aldrich ) , anti-IκBα mAb ( L35A5 ) ( Cell Signaling Technology , Beverly , MA ) , anti-actin antibody ( C-11 ) ( Santa Cruz , CA ) , and the His-probe ( H-15 ) ( Santa Cruz ) were diluted at 1∶1000 with TBST . HRP-conjugated rabbit anti-mouse IgG ( Zymed , South San Francisco , CA ) and HRP-conjugated Goat anti-rabbit IgG antibodies ( Invitrogen ) were also diluted at 1∶1000 and used as secondary antibodies to detect each appropriate primary antibody . 1 . 0×105 of 293T cells were transfected with 0 . 1 µg of an mA3 expression plasmid , 0 . 01 µg pFLAG-CMV2-GFP , and 0 . 1 µg of pluc . After 24h , the cells were treated with 10 µg/ml of cycloheximide or its solvent dimethyl sulfoxide ( DMSO ) as a control for 0 , 2 , or 4 hours . The cells were washed and resuspended in PBS . One-tenth of the cell suspension was subjected to luciferase assays to normalize transfection efficiencies , and the remaining cells were dissolved in the SDS-PAGE sample buffer . The normalized amount of cell lysates were separated by SDS-PAGE followed by immunoblotting as described above . The FLAG-mA3 cDNA was amplified with highly proofreading Pfu Turbo DNA Polymerase ( Stratagene ) from the plasmids pFLAG-CMV-mA3d and pFLAG-CMV-mA3dΔ5 [27] , with a forward primer harboring the T7 promoter sequence , 5′-GGATCCTAATACGACTCACTATAGGGAACAGCTGGGATGGGACCATTCTGTCTGGGATGC-3′ and a reverse primer harboring the His-Tag sequence , 5′-TCAATGGTGATGGTGATGATGAGCAGCAGCAGACATCGGGGGTCCAAGCTGTAGG-3′ . The PCR products were subjected to reactions for in vitro transcription and translation using TNT T7 Quick for PCR DNA ( Promega ) according to the manufacturer's protocol . Half of the generated products were mixed with the SDS-PAGE sample buffer and analyzed by immunoblotting . The remaining products were used for RNA purification with RNeasy Mini Kit ( Qiagen ) followed by RT-PCR to detect mA3 transcripts . | Susceptibility to acutely leukemogenic Friend virus ( FV ) retrovirus infection varies among different mouse strains and is governed by several genetic factors , one of which is allelic variations at the mouse Apobec3 locus . FV-resistant C57BL/6 ( B6 ) mice express higher amounts of Apobec3 transcripts than susceptible BALB/c mice . We previously showed that the differences in N-terminal amino acid sequences between B6 and BALB/c APOBEC3 proteins partly account for the distinct antiretroviral activities . In addition , B6 and BALB/c mice express major Apobec3 transcripts of different sizes: the exon 5-lacking and the full-length transcripts , respectively . Here we asked if exon 5 has any role in the antiviral activity of mouse APOBEC3 and found that the presence of this exon resulted in a profound decrease in the efficiency of protein synthesis without affecting the mRNA expression levels . We also identified two genomic polymorphisms that control the inclusion of exon 5 into the Apobec3 message: the number of TCCT repeats in intron 4 and a single nucleotide polymorphism within exon 5 . The distribution of these functional polymorphisms among Mus species and wild mouse populations indicates that the exon 5 inclusion occurred recently in Mus evolution , and the full-length variant may have selective advantages in some mouse populations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"organismal",
"evolution",
"viral",
"transmission",
"and",
"infection",
"virology",
"genetics",
"biology",
"evolutionary",
"biology",
"microbiology",
"genetics",
"and",
"genomics"
]
| 2012 | Two Genetic Determinants Acquired Late in Mus Evolution Regulate the Inclusion of Exon 5, which Alters Mouse APOBEC3 Translation Efficiency |
Schistosomiasis is a disease of world-wide importance and is caused by parasitic flatworms of the genus Schistosoma . These parasites exhibit a unique reproduction biology as the female's sexual maturation depends on a constant pairing-contact to the male . Pairing leads to gonad differentiation in the female , and even gene expression of some gonad-associated genes is controlled by pairing . In contrast , no morphological changes have been observed in males , although first data indicated an effect of pairing also on gene transcription in males . To investigate the influence of pairing on males , we performed a combinatory approach applying SuperSAGE and microarray hybridization , generating the most comprehensive data-set on differential transcription available to date . Of 6 , 326 sense transcripts detected by both analyses , 29 were significantly differentially transcribed . Besides mutual confirmation , the two methods complemented each other as shown by data comparison and real-time PCR , which revealed a number of genes with consistent regulation across all methods . One of the candidate genes , follistatin of S . mansoni ( SmFst ) was characterized in more detail by in situ hybridization and yeast two-hybrid ( Y2H ) interaction analyses with potential binding partners . Beyond confirming previously hypothesized differences in metabolic processes between pairing-experienced ( EM ) and pairing-unexperienced males ( UM ) , our data indicate that neuronal processes are involved in male-female interaction but also TGFβ-signaling . One candidate revealing significant down-regulation in EM was the TGFβ-pathway controlling molecule follistatin ( SmFst ) . First functional analyses demonstrated SmFst interaction with the S . mansoni TGFβ-receptor agonists inhibin/activin ( SmInAct ) and bone morphogenic protein ( SmBMP ) , and all molecules colocalized in the testes . This indicates a yet unknown role of the TGFβ-pathway for schistosome biology leading to male competence and a possible influence of pairing on the male gonad .
Schistosoma mansoni is a species of parasitic flatworms causing schistosomiasis , an infectious disease of worldwide importance for man and animals . Besides vertebrates as final hosts , the parasites' life cycle includes a snail intermediate host , and both are infected by aquatic larval stages . Schistosomiasis occurs in 78 , mainly tropical and sub-tropical countries with about 600 million people at risk , of which 243 million required regular treatment in 2011 . Thus it is one of the most prevalent parasitemias in the world , second only to malaria [1]–[4] . Pathology is induced by eggs deposited in the bloodstream by paired females , each producing up to 300 eggs per day [5] . A necessity for egg production is the completion and maintenance of the full development of female gonads . For this , the female depends on a constant pairing-contact to a male partner , an exceptional phenomenon in nature . Thus male worms have a key-role in the reproduction biology of schistosomes . Besides causing morphologic alterations comprising a significant increase of the body size of a paired female , which originates from mitogenic processes and differentiation of the reproductive organs ovary and vitellarium , the male even controls gene expression in its partner [6] , [7] . While these effects on females have been a strong focus for research throughout the last decades [8]–[11] , only few studies concentrated on pairing-dependent processes in the male . Authors of early studies presumed that different factors are transmitted from male to female controlling female body length as well as sexual maturation [12] , [13] . Stimulation of the latter was reported to act locally [14] , [15] , also a tactile impulse was proposed , while sperm or seminal fluid were excluded [12] , [14] . Vague evidence was found for a male-secreted hormone or protein to act on the female , however , none of these leads resulted in the identification of a concrete male stimulus [16]–[20] . Since glucose and other substances like cholesterol were shown to be transferred from the male to the female , also the supply of nutrients by the male was suggested to be the basis for female development [21]–[24] . So far only one glycoprotein ( GCP = gynecophoral canal protein ) was identified [25] and even described to be essential for pairing in S . japonicum [26] . Localization to the male gynecophoral canal as well as on the female surface further indicated the putative importance of GCP for the male-female interaction [27] . Although its function has not been clearly identified yet , evidence was obtained for the regulation of GCP by a TGFβ-dependent pathway in S . mansoni [28] . Other studies indicated that pairing may have an effect on the male as well , by influencing its capacity to stimulate mitosis in the female [29] . This led to the hypothesis that males have to reach a kind of competence before being able to induce developmental processes in the female . Along with the progress of the genome [30]–[33] and large-scale transcriptome sequencing [34] , [35] projects , additional analysis methods became available , which were used among others to compare pairing-experienced ( EM ) and pairing-unexperienced ( UM ) males . While the majority of these studies applied microarrays , several groups used serial analysis of gene expression ( SAGE ) alternatively . One of these studies [36] compared paired adult males and females and their pairing-unexperienced counterparts . For EM and UM the authors found differential regulation for transcripts contributing to developmental processes , metabolism and the redox-system . Already before the genome project was finished , an early microarray-based study compared EM and UM identifying 30 highly expressed genes to be exclusively transcribed in EM and 66 in UM [37] . Their identities indicated RNA metabolic processes to be differentially regulated between EM and UM , which was supported by a subsequent study [38] . Addressing the still unsolved question of male competence , here we investigated the influence of pairing on gene transcription in males . To this end we used two well established transcriptome analysis methods , SuperSAGE and microarray . The combination of both methods aimed at the production of corresponding data sets confirming , but also complementing each other to generate a comprehensive set of differentially transcribed genes in EM and UM that provides new insights into the male-female interaction . Among the most interesting genes identified here was a S . mansoni follistatin homolog ( SmFst ) , a potential inhibitor of TGFβ pathways [39] , [40] . Besides its pairing-dependent transcriptional regulation in males , our first functional analyses demonstrated not only gonad-preferential transcription of SmFst , but also its potential to interact with the TGFβ-receptor agonists SmInAct and SmBMP , which colocalized in the gonads . Thus , first evidence was obtained that TGFβ signaling plays an additional role for schistosome biology being one of probably several elements guiding male competence .
The parasite life cycle was maintained using a Liberian isolate of Schistosoma mansoni [41] , Biomphalaria glabrata as intermediate snail host , and Syrian hamsters ( Mesocricetus auratus ) as final host . To produce EM/EF ( pairing-experienced males/females ) or UM/UF ( pairing-unexperienced males/females ) snails were infected with either several miracidia ( poly-miracidial infection ) , or only one miracidium ( mono-miracidial infection ) . Poly-miracidial snail infections led to populations of male and female cercariae , which were used for bisex hamster infections resulting in EM and EF . Mono-miracidial snail infections led to unisexual populations of cercariae , which upon final-host infection developed into UM or UF . After 42 days ( EM ) or 67 days ( UM ) post infection adult worms were obtained by hepatoportal perfusion . This difference is due to our experience with experimental hamster infections that revealed a positive effect on perfusion efficiency and quality of unisexual worms , when the infection period is elongated to 67 days . This elongation had no influence on further experimental procedures , which concentrated on the comparison of the pairing status of male worms . For EM enrichment , paired males from bisex hamster infections were carefully separated from their partners by feather-weight tweezers , immediately frozen , and stored at −80°C until further use . All experiments with hamsters have been done in accordance with the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes ( ETS No 123; revised Appendix A ) and have been approved by the Regional Council ( Regierungspraesidium ) Giessen ( V54-19 c 20/15 c GI 18/10 ) . Total RNA from adult worms was extracted using TriFast ( PeqLab ) following the manufacturer's instructions . Subsequently extracted RNAs were quality-checked on denaturing formaldehyde gels . Following total RNA extraction , cDNA synthesis was performed with the Quantitect Reverse Transcription Kit ( Qiagen ) following the manufacturer's protocol with 1 µg total RNA from EM or UM as template . Standard PCR reactions were performed in a final volume of 25 µl using primer end concentrations of 800 nM , an annealing temperature of 60°C , elongation at 72°C , and FirePol-Taq ( Solis biodyne ) . Following hamster perfusion , 50 EM or UM were collected , incubated over-night in RNAlater ( Ambion ) and stored at −80°C . For total-RNA isolation approximately 25 worms from each batch were washed twice in 500 µl H2ODEPC , followed by addition of 1 ml TRIzol reagent ( Invitrogen ) . Subsequently , the worms were homogenized mechanically and incubated for 5 min at room temperature ( RT ) before 200 µl chloroform was added , mixed for 15 s and incubated for 2–3 min . Following centrifugation for 15 min at 12 , 000 g and 4°C the upper phase was transferred to a new tube and mixed with 500 µl Isopropanol . After incubation for 10 min at RT the RNA was centrifuged for 10 min at 12 , 000 g and 4°C . The pellet was washed with 1 ml ethanol ( 75% ) and centrifugation repeated for 5 min at 4°C and 7 , 500 g . The supernatant was discarded , the pellet dried and resuspended in 25 µl H2ODEPC . Before determination of the concentration on a spectrophotometer ( Nanodrop ) the RNA was shortly heated to 65°C . RNA purification was done following the animal tissue protocol ( Qiagen RNeasy Mini kit ) with the following modifications: samples were supplemented to a volume of 100 µl , and 350 µl RLT buffer and 250 µl ethanol ( 70% ) were added before the suspension was transferred onto a column ( Qiagen RNeasy Mini kit ) . RNA was eluted with 30 µl H2ODEPC , and the flow-through put on the column for a second elution step . The concentration of RNA was determined again ( see above ) , and its quality checked on a Bioanalyzer ( Agilent 2100 Bioanalyzer , Agilent Technologies ) . RNA was reverse transcribed , in vitro amplified , labeled with Cy3 or Cy5 , and hybridized according to the Agilent technology protocol for “two color microarray based gene expression analysis” . The samples were hybridized on a 4×44 k oligoarray containing 60 mer oligonucleotides that was custom-designed by us [42] , and manufactured by Agilent Technologies; the platform probe sequences are available on Gene Expression Omnibus ( GEO ) under the accession number GPL8606 . This platform was recently re-annotated [43] according to the first draft of the genome project [30] . Three independent biological replicas for EM and UM populations each were used for microarray analyses , each with four technical replicas including dye swaps . Data were extracted using Agilent feature extraction software and raw data are available in NCBI's Gene Expression Omnibus ( GEO ) [44] under the accession number GSE45696 ( subseries number GSE44193 ) . Log2ratios were calculated using LOWESS normalized intensity values of UM and EM ( log2 UM/EM ) with R [45] . Subsequently , a manual filtering process was applied before statistical analysis , keeping only those oligonucleotides that were defined as representative ( unique ) for each gene ( criteria “to be used in analysis” ) according to the information provided together with the re-annotation of the array [43] . Remaining transcripts were submitted to a manual filtering process keeping only those transcripts that were present in all biological replicas , in at least three technical replicas of one biological replica in at least one condition ( EM or UM ) . With these pre-selected data a statistical analysis for microarrays ( SAM ) [46] was performed using a one-class analysis . Significance cutoff was chosen at a FDR ( false discovery rate ) of 0 . 01 , and only average log2ratios <−0 . 585 or >0 . 585 ( which corresponds to a 1 . 5- fold difference in transcript levels ) were defined as relevant . Genes fulfilling these requirements are described in the text as significantly differentially transcribed . The final analysis focused on oligonucleotides representing sense transcripts ( although the array also contained oligonucleotides representing putative antisense RNAs for each corresponding gene locus ) . Sample collection for SuperSAGE equaled that for the microarray approach . Total RNA was extracted from whole worm batches of 50 males ( EM or UM ) . RNA was quality-controlled on a denaturing formaldehyde gel as well as with a bioanalyzer ( Agilent 2100 Bioanalyzer , Agilent Technologies ) . The following experimental procedure to perform SuperSAGE was done as described previously [47] with minor changes [48] . Raw data were deposited at GEO [44] under the accession number GSE45696 ( subseries number GSE45628 ) . Tags were annotated applying the same procedure as for the re-annotation of the microarray [43] and separated or added up according to their annotation to the predicted exon or intron parts of a CDS in either sense or antisense orientation . Counts were normalized to a library size of 1 , 000 , 000 , and a filtering process was applied keeping only those transcripts that were detected in two out of three biological replicas in EM or UM . A program implementing the statistical method of Audic and Claverie [49] using a Bio_Sage script ( pearl ) ( http://search . cpan . org/~scottzed/Bio-SAGE-Comparison1 . 00/lib/Bio/SAGE/Comparison . pm ) was used for significance analysis of the data . The statistical cutoff was p<1−10 . As with the array , only transcripts with average log2ratios <−0 . 585 or >0 . 585 were selected for further analysis . For data comparison only annotated sense transcripts were used , while 7 , 124 transcripts annotated as antisense as well as 19 , 610 tags without annotation were excluded from analysis . Data for sense transcripts from SuperSAGE and microarray were comparatively analyzed using Spotfire [50] and Microsoft-Excel . Smp_numbers without a match in both analyses were manually checked again . The same approach was used for comparative-analysis to other data . An intersection data-set was created , which contained only transcripts detected by both analyses . These were performed on a Rotor Gene Q ( Qiagen ) using SYBR-Green MasterMixes ( PerfeCTa SYBR Green SuperMix ( Quanta ) or RotorGene SYBR Green PCR Kit ( Qiagen ) ) . PCRs were performed in a total volume of 20 µl , with a three-step thermo-profile and a final melting-curve analysis . Primers , their concentrations , annealing temperatures and efficiencies are listed in Supplementary Table S1 . All primers were synthesized by Biolegio ( Netherlands ) . Primer-efficiencies were determined with a standard-curve on diluted gel-eluate with 1∶10 dilution steps [51] . Efficiencies were considered to be optimal between 85–100% . RNAs from EM and UM were evaluated for similar quality on denaturing formaldehyde-agarose ( 1 . 2% ) gels . cDNAs were diluted 80-fold for usage and added 1∶4 to the final reaction . Absolute quantification was achieved by including the standard curve in each run [52] . Fold changes ( EM/UM ) were calculated using UM as calibrators [52] . To facilitate comparisons to microarray and SuperSAGE data log2-values of the fold changes were determined as previously described [53] , [54] . Standard-curves were performed in duplicate after initial tests for primer concentrations , and reactions on cDNA-samples were performed in triplicates . The significance of individual experiments was checked applying the “Exact Wilcoxon rank sum test” using the exactRankTests package for R [45]; [53]–[57] . Correlation of real-time PCR and transcriptome data was checked with the Spearman's correlation coefficient [53] . For stage-specific detection of SmFst transcripts , cDNAs from EM , UM , EF , UF , miracidia , and cercariae were generated and tested in standard PCR-reactions with primers for SmFst ( fwd-5′- TGTTGTAAACGTGGTGGATTC-3′ and rev-5′-CGACATTfTGCATTTTGGTTC-3′ ) and primers for actin ( fwd-5′-GGAAGTTCAAGCCCTTGTTG-3′ and rev-5′-TCATCACCGACGTAGCTGTC-3′ ) as positive control . PCR-products were separated on a 2% agarose gel . To obtain organ-specific RNA a recently established protocol was used [58] . In short , adult schistosomes ( about 50 individuals ) maintained in M199-medium at RT were transferred to reaction vessels containing 500 µl of tegument solubilisation ( TS ) -buffer ( 0 . 5 g Brij35 ( Roth ) , 0 . 5 g Nonidet P40-Substrate ( Fluka ) , 0 . 5 g Tween80 ( Sigma ) , and 0 . 5 g TritonX-405 ( Sigma ) per 100 ml PBS ( 137 mM NaCl , 2 . 6 mM KCl , 10 mM Na2HPO4 , 1 . 5 mM KH2PO4 in DEPC-H2O , pH 7 . 2–7 . 4 ) ) . Following incubation at 37°C in a thermal shaker ( TS-100 , Biosan ) at 1 , 200 rpm for 5 min to solubilise the tegument , the musculature was digested by protease treatment . To this end 500 µl elastase-containing medium ( Sigma , #E0258; freshly dissolved in non-supplemented M199-medium , 5 units/ml ) were used and the worms slightly agitated ( 600 rpm ) in the thermal shaker at 37°C for 30–40 min . Progress of digestion was monitored by microscopic inspections using 20 µl aliquots . Upon the start of tissue fragmentation , reproductive organs such as ovary and testes were liberated . 1 ml non-supplemented M199-medium was added and the content of the vessel decanted to Petri dishes for manual collection of testes and ovaries , which were identified by their characteristic morphologies . If necessary , purification of the gonad tissue from residual parenchyma tissue was achieved by repeatedly collecting and transferring the organs to further Petri dishes containing 2 ml of non-supplemented M199-medium . Finally , the organs were collected using a 10 µl-pipette , transferred to 1 . 5 ml-tubes , and concentrated by centrifugation for 5 min at 1 , 000 g , and 1 min at 8 , 000 g . Following removal of the supernatant , the gonads were immediately frozen in liquid nitrogen and stored at −80°C for further use . Total RNA was extracted from the organs as described before [58] using the PeqGOLD TriFast reagent ( Peqlab; 500 µl TriFast-solution per extraction of 50 testes or 50 ovaries ) , and the resulting RNA pellet was resuspended in 10 µl DEPC-H2O each . RNA quality and quantity were checked by electropherogram analysis ( Bioanalyzer 2100; Agilent Technologies ) . RT-PCRs were basically performed as described above ( standard PCRs ) using the following primer combinations to amplify gene transcripts of SmFST ( fwd-5′-GAACCAAAATGCAAATGTCG-3′; rev-5′-GCCATGATTGTTCATTCCA-3′ ) , SmBMP ( q51- fwd-5′- GTCAAAATGAACAAAATCA-3′; q51- rev-5′- GTTACGTCGAACACTTTG-3′ ) , and SmInAct ( q1b-fwd-5′- CACAATTTGGTAATGTTCAACG-3′; q1b-rev-5′- AACTACAAGCACATCCTAAAACAA-3′ ) . Localization experiments were performed as previously described [59] with the following modifications: hybridization temperature was 42°C , and slides were washed up to 1× SSC . Two different probes were used for detection of SmFst transcripts: probe 1 was 571 bp long ( position 208–778 ) , while probe 2 was 306 bp long ( position 914–1219 ) . The probe for SmInAct was 440 bp long ( position 206–646 ) . Two probes were designed for SmBMP detection: probe 1 had a length of 455 bp ( position 2240–2694 ) and probe 2 was 565 bp long ( position 1041–1605 ) . For SmFst–SmInAct/SmBMP interaction studies , yeast two-hybrid ( Y2H ) assays were performed . To this end full-length SmFst was cloned into the Gal4-BD vector pBridge using the following primers , which were designed according to the sequence information available at SchistoDB 2 . 0 [60] for Smp_123300: fwd-5′-GAATTCATGGAAGAGAGTATATCACAATTAG-3′ ( italics: cleavage site for EcoRI ) , rev - 5′-GTCGACTTAGAATAAATTTGAATATTTTCC-3′ ( italics: cleavage site for SalI ) . Full-length SmInAct was cloned into the Gal4-AD vector pACT2 , using the following primers: fwd-5′-CCCGGGGATGAATAGAATGTTTAAATTAATAAAAC-3′ ( italics: cleavage site for SmaI ) , rev-5′-CTCGAGTTAACTACAAGCACATCCTAAA-3′ ( italics: cleavage site for XhoI ) . Due to its large size , the sequence for SmBMP was split into four sub-fragments , which were separately cloned into pACT2 . The following primers were used: SmBMP-Y2H-Cterm-5′- 5′-CCCGGGGAAACCAAGATCAATTAATTATCCTAAC-3′ , SmBMP-Y2H-ncbi-5′– 5′-CCCGGGGATGAACTCAAATATTTTAACAAAATCAG-3′ , SmBMP-Y2H-ncbi-Nterm-5′ – 5′-CCCGGGGATGGAAACAGAAAAGACAAAAC-3′ , SmBMP-Y2H-overlap-5′ – 5′-CCCGGGTGAAATAAATAGTACATCATTCTACTGG-3′ ( italics: cleavage site for SmaI ) ; SmBMP-Y2H-Cterm-3′ – 5′-CTCGAGTTAACGACAAGCACAACTTTC-3′ , SmBMP-Y2H-db-Nterm-3′ – 5′-CTCGAGAATTGCTTACATTATTATTATTCAGAGG-3′ , SmBMP-Y2H-ncbi-Nterm-3′ – 5′-CTCGAGGTTCTTTAGATGGTTTTCGTATATTATC-3′ , SmBMP-Y2H-overlap-3′ – 5′-CTCGAGGATGATTATTTGTTTGTAATACATTTG-3′ ( italics: cleavage site for XhoI ) . PCR products were separated on 1 . 0% agarose gels . The amplicons were cut out from the gel , and the DNA extracted using the PeqGold Gel Extraction Kit ( Peqlab ) following the manufacturer's protocol . Extracted fragments were cloned into pDrive ( Qiagen ) and later regained by restriction-digestion to be again checked for correct size on a 1 . 0% agarose gel and extracted . Finally , fragments were ligated into pBridge and pACT2 , respectively , using T4 Ligase ( Promega ) . Sequences were checked for integrity and a correct ORF by commercial sequencing ( LGC Genomics , Berlin ) . The SmFst-containing plasmid was transformed in to yeast cells ( AH109 ) together with either one of the other plasmids prepared for the interaction studies . To control successful transformation yeast clones were grown on selection plates ( SD-Trp/-Leu/-His/-Ade ) . β-galactosidase ( β-gal ) liquid- and filter- assays were performed to confirm interactions ( Yeast protocols handbook , Clontech ) . The following public domain tools were used: SchistoDB ( http://www . schistodb . net/schisto/; [60] ) , BLASTx ( http://www . ncbi . nlm . nih . gov/BLAST ) , restriction mapper version 3 ( http://www . restrictionmapper . org/ ) . Data were analyzed for enriched genes of the ontology categories with Ontologizer [61] using contig annotations , which were the basis for the microarray design [41] , [42] . Furthermore , only sense-orientated genes/transcripts were used for this analysis . Network enrichment analyses were done with the Ingenuity Pathway Analysis ( IPA ) tool ( http://www . ingenuity . com; [62] ) . Only those transcripts with homology to a human molecule >60% and an e-value<10−10 were used , as defined during the re-annotation of the microarray [43] . For SuperSAGE-detected transcripts the according information was obtained from the same source as far as available . For allocation of transcripts to predicted S . mansoni metabolic pathways the function “omics viewer” of the software tool SchistoCyc was used ( available at SchistoDB 2 . 0; [59] . The online-tool SMART ( http://smart . embl-heidelberg . de/; [63] , [64] ) was used to predict protein domains . Smp_135230 - dopa decarboxylase; Smp_145140 - wnt5A; Smp_155340 – frizzled; Smp_036470 - oxalate-formate antiporter; Smp_135020 - oxalate-formate antiporter; Smp_169190 - tegument protein; Smp_161500 - rhodopsin-like orphan GPCR; Smp_131110 - p14; Smp_000270 - fs800-like; Smp_000430 - ‘eggshell precursor protein’; Smp_00280 - fs800-like transcript; Smp_123300 ( KC165687 ) – S . mansoni follistatin; Smp_090140 . 2 - Ftz-F1 interacting protein; Smp_090520 - purin nucleoside phosphorylase; Smp_123010 – cationic amino acid transporter; Smp_095360 . x – fatty acid binding protein; Smp_065580 . x – heterogeneous nuclear ribonucleoprotein k; Smp_033950 - Smad4; Smp_144390 - S . mansoni activin receptor; Smp_049760 – TGFβRI; Smp_093540 . 3 – ActRI; Smp_124450 - ActRI/BMPRIa; Smp_080120 . 2 – ActRIIa; Smp_144390 - ActRIIb .
To produce a comprehensive data set of genes differentially transcribed between EM and UM , two methods were chosen . Because both methods have successfully demonstrated their capacities in the past , microarray and SuperSAGE analyses were applied in parallel to generate complementary data sets of genes differentially transcribed between EM and UM . For microarray analyses a S . mansoni-specific 60-mer oligonucleotide microarray platform was used , which represents nearly the complete genome of S . mansoni [42] , [43] . The platform-independent SuperSAGE represents a technically improved modification of ‘serial analysis of gene expression’ ( SAGE ) by generating 26 bp sequence tags of all sample mRNAs containing a NlaIII restriction site by a high-throughput sequencing approach [47] . Combining both methods , we expected to produce data sets complementing each other and providing independent indications for the importance of particular transcripts . To confirm differential transcription of genes from an expected overlap , or from individual data sets outside this overlap , we additionally performed real-time PCR experiments for selected candidates . In order to facilitate the interpretation of the large scale transcriptome data and selection of molecules for first characterization studies , gene ontology ( GO ) analysis as well as two further analyses tools , Ingenuity Pathway Analysis ( IPA ) [62] and the metabolomics tool SchistoCyc [60] were applied . IPA operates with a curated biological knowledge base from the literature to generate molecular networks enriched for proteins encoded by significantly regulated genes from large-scale data sets . Similarly it searches for canonical pathways and predicts the activation of transcription factors . The SchistoCyc function ‘omics viewer’ allocates an uploaded set of genes to predicted S . mansoni metabolic pathways . The results of the microarray analysis were extracted and evaluated with Agilent feature extraction software . Subsequent data-processing included calculation of log2ratios [53] , [65] , data-filtering for consistency of transcript detection , and annotation of detected transcripts . After re-annotation of the 44 k oligo array in 2011 [43] , 19 , 197 oligonucleotides were selected as gene representatives: 11 , 132 detecting RNA in orientation of the predicted transcript , 8 , 065 detecting RNA complementary to the predicted transcript . Following hybridization and data processing , 10 , 115 of these representatives were detected as transcribed , 1 , 966 representing putative antisense and 7 , 494 representing sense messages . Significance analysis of microarrays ( SAM ) [46] was performed for all transcripts , and only sense transcripts were analyzed further . Of these , 526 transcripts were significantly differentially transcribed; 229 were up-regulated in EM and 297 up-regulated in UM . Gene ontology ( GO ) analyses were performed in order to get an overview about the categories these genes could be assigned to . Interestingly , enriched categories were only found for genes up-regulated in UM ( Fig . 1 ) . Again , three samples of EM and UM were collected each , independently from each other as well as from the microarray samples . RNA was extracted from 50 worms per sample and quality-checked . The further procedure was executed by GenXPro according to an internal protocol [47] , [48] . SuperSAGE-tags were annotated in the same way as the revised version of the 44 k oligo array [43] , and counts for different tags were summed up if they had shown the same annotation . Similar to the microarray data-analysis , transcripts were classified into sense- and antisense orientation relative to the protein-coding gene in a given locus . Additionally , SuperSAGE-transcripts were further distinguished between predicted intron- or exon-sequences . Thus four categories of SuperSAGE-detected tags were defined , and accordingly up to four transcripts could represent one gene . This explains the large number of 25 , 597 gene-specific transcripts detected with SuperSAGE , which exceeds the assumed number of S . mansoni genes more than twice [33] . Statistics were performed according to the method of Audic & Claverie [49] . Subsequent to this significance analysis , 5 , 987 transcripts without annotation and 7 , 124 transcripts classified as antisense were excluded from further analyses . Of the remaining 12 , 486 sense transcripts , 8 , 969 were classified as exon and 3 , 517 as intron sequences . Taking out redundancy for genes represented by an exon as well as an intron sequence , sense transcripts were found for a number of 9 , 344 unique genes . Of these , 5 , 601 genes had representatives in both groups exons as well as introns . For 2 , 581 genes only exon-representing sense transcripts were detected , and for 219 genes only intron-representing sense transcripts were detected . Selected from these were candidates showing a normalized detection value of at least 10 tags in at least one library and significantly differential regulation between EM and UM . These criteria were met by 253 transcripts . Of these 218 were up-regulated in EM and 35 in UM . GO analyses showed enriched categories neither for the transcripts up-regulated in EM nor for those up-regulated in UM . Comparing sense transcripts detected by microarray and SuperSAGE , 6 , 326 transcripts were found by both methods , while additional 3 , 018 and 1 , 168 were exclusively detected by SuperSAGE or microarray , respectively ( Fig . 2 ) . The number of counts representing genes being differentially transcribed between EM and UM varied between the methods used . This was influenced by the underlying , method-specific statistics leading to subsets of counts that were significant for only one of the two data sets although the same transcripts were also present in the other data set ( Fig . 2 ) . The stringent analysis criteria for differential regulation were met according to both approaches by 29 transcripts . Among these were genes with various biological functions such as metabolism ( NAD-dependent epimerase/dehydratase , sodium dicarboxylate cotransporter , fatty acid acyl transferase-related , oxalate-formate antiporter ) , neurotransmitter synthesis ( aromatic amino acid decarboxylase ) , enzyme activity ( kunitz-type protease inhibitor ) , microfilament organization ( villin , nebulin ) , membrane dynamics/vesicle formation ( endophilin b1 , snf7-related ) , molecular interaction/communication ( surface protein PspC , cadherin ) , calcium metabolism ( sarcoplasmic calcium-binding protein ) , chromatin organization ( histone h1/h5 ) , signal transduction ( pinch , dock , follistatin ) , and others less well defined ( cancer-associated protein gene , loss heterozygosity 11 chromosomal region 2 gene ) ( Table 1 ) . Differentially transcribed genes from either microarray or SuperSAGE were assigned to predicted S . mansoni metabolic pathways using the “omics viewer” function of SchistoCyc [60] ( Supplementary Table S2 ) . Comparing the results of these analyses , transcripts coding for enzymes involved in carbohydrate metabolic processes , citrate cycle , aerobic respiration and amino acids metabolic processes were mostly down-regulated in EM compared to UM . For base metabolic processes , transcripts for enzymes involved in synthesis were rather down-regulated in EM , while transcripts for enzymes involved in degradation and salvage pathways were up-regulated in EM . Several transcripts with different directions of regulation were found for enzymes participating in lipid and fatty acid metabolic processes . Interestingly , data sets for differentially transcribed genes contained none coding for enzymes related to pentose-phosphate cycle or glycolysis . Only one enzyme within the SuperSAGE/microarray-intersection of differentially transcribed genes was allocated to metabolic pathways , aromatic amino acid decarboxylase ( dopa decarboxylase , DDC , Smp_135230/NP_001076440 . 1 ) . DDC was up-regulated in EM and allocated to phenylethanol and catecholamine biosynthesis . This indicates a pairing-dependent adaptation for usage of neurotransmitters like dopamine can be assumed . IPA was performed for the data-overlap of transcripts differentially regulated according to both transcriptome analyses . While no significantly enriched canonical pathways were detected , one significantly enriched network was found , namely ‘embryonic development , hair and skin development and function , organ development’ . It contained nine differentially transcribed genes , seven up-regulated in EM and two up-regulated in UM , including SmFst ( Supplementary Table S2 , Supplementary Figure S1 ) . Thus IPA additionally highlighted SmFst and a putative differential regulation of neurotransmitter synthesis through DDC . Looking for explanations why transcripts were detected by one method only without complementary counterpart , these specific transcript groups were selectively analyzed in more detail . Out of the 47 transcripts differentially regulated in the SuperSAGE-only group , oligonucleotides existed on the microarray for 25 out of 47 , thus 22 transcripts were newly detected by SuperSAGE ( Table 2 ) . Next , transcripts within the microarray-only group ( 110 ) were analyzed for the presence of NlaIII restriction-enzyme recognition sites . cDNA synthesis and NlaIII restriction are the first two steps during the SuperSAGE procedure . Transcripts lacking the restriction site or with an NlaIII restriction too close to the polyA-tail will not be detected . As a snap sample of 110 candidates of the microarray-only group , we selected 10 which not only had a Smp_number but also a functional annotation other than ‘hypothetical protein’ . Of these 8 had NlaIII restriction sites within their predicted CDS ( Table 3 ) . Thus indeed , the existence of NlaIII restriction may have influenced the detection of transcripts by SuperSAGE . Also other methodological differences may have contributed to the resulting number of transcripts detected by both methods , although not as differentially transcribed in both analyses . While in microarray experiments light intensities are measured , during SuperSAGE the number of transcripts is counted , resulting in different log2ratios ( some below the selected threshold of log2ratio = 0 . 585 ) . Also the number of technical replicas differed . While three biological replicas without technical replicas were used for SuperSAGE , four technical replicas for each biological one were done in case of the microarrays , which is part of the methodological procedure . Furthermore , different statistics had to be used for the analyses of the results from both methods , which influenced the outcome with regard to the significance of the differences detected between EM and UM . For a sub-set of 21 transcripts we performed real-time PCR experiments ( i ) to confirm the results obtained in the combinatory analysis , and ( ii ) to confirm those microarray or SuperSAGE data , for which no complementary results existed . With respect to their putative biological relevance , transcripts for signal transduction molecules , surface molecules , metabolism-associated proteins and transcription factors were chosen for verification ( Supplementary Table S1 ) . Additionally , the regulation of three transcripts for egg-shell precursor proteins was analyzed . A Wilcoxon rank sum test for real-time PCR data detected significant differences between EM and UM transcript levels for most of the differentially transcribed genes present within the overlap of microarray and SuperSAGE data ( Supplementary Table S1 ) . The test also confirmed significant differences between the two male stages for two genes detected by only one of the transcriptome studies . Using the Spearman's rank correlation coefficient , overall results from real-time PCR were compared to those from either one of the two transcriptome approaches . Data correlated significantly ( p<0 . 01 ) between real-time PCR and microarray ( r = 0 . 676 ) or SuperSAGE ( r = 0 . 621 ) each . Representatives for signal transduction-associated transcripts were SmFst , dock , and pinch ( Figure 3; Supplementary Table S1 ) . For all three genes transcription regulation was confirmed by real-time PCR . Comparing the results of all approaches , dock and pinch showed a slight biological variation within only one of the three analyses . In contrast , the transcriptional activity of SmFst was found to be consistently down-regulated without exception in EM ( Figure 3 ) , which substantiated the results of the transcriptome approaches . Since follistatins ( FSTs ) have regulatory function for TGFβ-signaling [39] , [40] , we additionally tested the transcriptional activities of further genes coding for members of TGFβ-pathways ( Supplementary Table S1 ) . In those cases where microarray or SuperSAGE indicated significant regulation for a TGFβ-pathways member , this finding was not confirmed by the complementary method or real time PCR . Furthermore , we included wnt5A ( Smp_145140 ) and one of its potential receptors , the seven transmembrane receptor frizzled ( Smp_155340 ) into this analysis , since wnt-pathways are linked to developmental processes [66] . According to the microarray data Wnt5A and frizzled transcripts were significantly down-regulated in EM . The direction of regulation of both transcripts detected by SuperSAGE varied strongly between the biological replicates , and differences between EM and UM were not significant . This was confirmed by real-time PCR experiments ( Figure 3 , Supplementary Table S1 ) , indicating that the wnt-pathway and/or associated members may not be essential with respect to EM/UM differences and are presumably influenced by additional biological parameters beyond pairing . The AMP-dependent ligase attracted attention due to its uniform differential transcriptional regulation in both approaches and its putative function in DNA synthesis processes . Besides SmFst , it was one of the transcripts for which real-time PCR supported the previous findings , but in contrast to SmFst , transcription of the AMP-dependent ligase was strongly up-regulated in EM ( Figure 3 ) . A similar consistent result was obtained for one of the oxalate-formate antiporters ( OxlT ) ( Smp_036470 ) , whose transcript amount was lower in EM . In contrast , a second OxlT ( Smp_135020 ) was detected as significantly up-regulated in EM by the microarrays only , which was confirmed by real-time PCRs . Further transcripts for membrane proteins analyzed in real-time PCR experiments were a tegument protein ( Smp_169190 ) , significantly up-regulated in EM according to the microarray analysis ( Supplementary Table S1 ) and a rhodopsin-like orphan GPCR ( Smp_161500 ) up-regulate in EM according to the SuperSAGE analysis ( Figure 3 ) . Both transcripts were not detected by the respective other transcriptome analysis . Real-time PCRs confirmed the presence and direction of regulation for both transcripts substantiating that microarray and SuperSAGE experiments can complement each other . Enhanced transcript levels of the rhodopsin-like orphan GPCR in EM were also confirmed by semi-quantitative real-time PCR [data not shown] . With respect to the fact that the physical contact between the genders stimulates differentiation processes in the female , tegumental proteins as such are potentially important for male-female interaction . Since GPCRs represent the biggest receptor class in schistosomes [66] , and since they are known to be involved among others in regulating differentiation processes in diverse organisms [67]–[70] , there may be candidates with roles in male-female interaction as well . Surprisingly , transcript amounts for the egg-shell precursor protein p14 ( Smp_131110 ) were significantly elevated in EM according to the microarray analysis . Our real-time PCR experiments , however , demonstrated strong variations between biological replicas , previously also indicated by SuperSAGE results . Analogous results were obtained for two other egg-shell precursor transcripts , fs800-like ( Smp_000270 ) , and ‘eggshell precursor protein’ ( Smp_000430 ) ( data not shown ) . Thus it seems obvious that egg-shell precursor protein transcripts can be strongly up-regulated in EM compared to UM , but this depends on the worm batch . Indirect support for this interpretation comes from a previous study , which provided evidence for another fs800-like transcript ( Smp_00280 ) to be up-regulated in testicular lobes of paired males [71] . Other transcriptome studies , comparing EM and UM , also found transcripts of supposedly female-specifically expressed genes in males [37] . These findings can be explained by leaky expression control of such gender-associated genes , a phenomenon that can even lead to the development of female reproductive tissue in males as previously observed [11] , [72]–[74] . The follistatin homolog SmFst appeared as one of the most interesting candidates for first functional characterization because of its consistent down-regulation in EM in all analyses , and its putative participation in schistosome TGFβ signaling processes . Based on the gene prediction ( Smp_123300 ) obtained from the Schistosoma genome project , primers were designed to amplify its full-length cDNA . Following cloning and sequencing , minor differences were detected between SmFst ( KC165687 ) and Smp_123300 , revealing parts of predicted intron sequences as exon stretches ( Supplementary Figure S2 ) . While other typical Fsts contain three follistatin-domains ( FstDs ) [75] , SmFst encodes an open reading frame containing two , according to a SMART-domain analysis . FstDs are further subdivided into an EGF ( epidermal growth factor ) - and a Kazal ( a protease inhibitor ) -domain , which according to SMART do not follow each other directly in SmFst ( amino acid positions: EGF-domains: 52–74 , 312–337; Kazal-domains: 264–307 , 398–434 ) , as is the case for other Fsts [75] . The first EGF-domain of SmFst is located upstream of the first EGF-domain of other Fsts , and the second EGF-domain of SmFst is homologous to the third one of other Fsts . The first and second Kazal-domains of SmFst are homologous to the second and third Kazal-domains of other Fsts . To confirm the presence of SmFst in different life cycle stages of S . mansoni RT-PCR reactions were performed with RNAs from EF ( pairing-experienced females ) , UF ( pairing-unexperienced females ) , EM , UM and the free-living larval stages . SmFst transcripts were detected in all adult stages and miracidia , but not in cercariae . As positive control actin was amplified from all tested life cycle stages ( Supplementary Figure S3 ) . Localization studies by in situ-hybridization with couples as well as UM detected SmFst transcripts within the testicular lobes of both male groups and also in the vitellarium and ovary of female worms ( Figure 4 ) . Results concerning the detection of sense or antisense transcripts varied with probe sequence and probe batch . Organ-specific RT-PCR [58] confirmed the presence of transcripts for SmFst within ovary and testes ( Supplementary Figure S4 ) . Fst was shown before to bind activin , which itself is an agonist of TGFβ-pathways [76] , [77] . Although with lower affinity it also binds to a bone morphogenic protein ( BMP ) , another agonist of TGFβ-pathways [78]–[81] . SmInAct , a Schistosoma activin-inhibin , was previously characterized [82] as well as SmBMP [83] . In our study , SmInAct transcripts were detected only by SuperSAGE but without differential regulation between EM and UM , which was also confirmed by real-time PCR results . SmBMP was down-regulated according to the microarray data , however , this was neither confirmed by SuperSAGE nor by real-time PCR . To obtain evidence on possible interactions between SmFst and SmInAct and/or SmBMP , co-localization and Y2H interaction studies were performed . Previous studies had localized SmInAct transcripts in the ovary and vitellarium of EF , without providing information on their presence in testicular lobes [82] . The results obtained in our study using two replicas with a hybridization probe based on the same sequence as used previously [82] showed sense and also antisense transcripts exclusively in the ovary of EF . No signal was detected within the vitellarium or testes of EM or UM ( Figure 5 ) . However , organ-specific RT-PCR on ovary and EM testes detected transcripts not only in the ovary but also in testes of EM ( Supplementary Figure S4 ) . For the localization of SmBMP two different probes were used . Transcripts were detected in all reproductive organs of couples as well as UM , but especially for probe 2 in the area around the ootype ( Figure 6 ) . Again , sense and antisense transcripts were detected , and results for EM-testes and ovary were confirmed by organ-specific RT-PCR ( Supplementary Figure S4 ) . Even though the major interaction partner of Fsts is activin , it can also bind to BMP [79] . For males our in situ-hybridization experiments rather indicated physical proximity of SmFst transcripts to those for SmBMP , as both transcripts were detected within testicular lobes . In females , transcripts for all three molecules were detected within the ovary by in situ-hybridization , though organ-specific RT-PCRs also indicated the presence of SmInAct in male testes . To provide evidence for protein interaction between SmFst and SmBMP or SmInAct , Y2H interaction studies were performed . To this end full-length SmFst was cloned into the Gal4-BD vector pBridge , while full-length SmInAct and four different sequence stretches of SmBMP were cloned into the Gal4-AD vector pACT2 . The SmFst-containing plasmid was transformed into yeast cells ( AH109 ) together with SmInAct or one of the BMP variants each . Following growth on selection plates ( SD-Trp/-Leu/-His ) , β-gal liquid- and filter-assays were performed to relatively quantify interactions . These were shown between SmFst and SmInAct as well as between SmFst and SmBMP . The result of the β-gal liquid assay indicated a stronger binding of SmFst to SmInAct , while no evidence was obtained for an interaction with SmBMP-C-term or negative controls ( Figure 7 ) . This is surprising since this part of SmBMP is most conserved in comparison to human BMPs and contains several residues essential for receptor binding , which are supposedly blocked by FST [76] , [84]–[87] . However , it is possible that within its quaternary structure FST binds to other residues , besides those at the C-terminus , thereby prohibiting the receptor binding of BMP . Together with the results from the localization experiments , evidence is provided that SmFst interacts with SmInAct and/or SmBMP in the testes of EM and UM but also in the female ovary , and in case of SmBMP in the vitellarium .
The results of our study provided conclusive evidence for pairing-influenced transcriptional processes in males . Together with previous findings about pairing-dependent gene transcription in females and first transcriptome analyses for males , all data clearly demonstrate a bidirectional transcriptional influence during male-female interaction [6] , [7] , [36] , [37] , [88] , [89] . Our expectation to find a comprehensive data set of genes differentially transcribed in males upon pairing was met , just as obtaining congruent and complementing results using two independent techniques . Since both are basically different , each generated additional data that were not obtained by the other method . From our point of view , both methods have their advantages and disadvantages , and none seemed superior over the other . Both methods required the application of different statistics since on the one hand signal intensities were determined ( microarray ) including technical replicas and on the other hand transcript counts , both resulting in different log2-ratios . Finally , also biological variability , a well-known phenomenon even within schistosome strains used for such kind of analyses [54] , may have had an influence on the data obtained . Although different methodological and analytical approaches were applied , an overlap of interesting genes was obtained on the basis of stringent analysis criteria . Taking into account that we made use of a second-generation microarray , which contained the majority of genes known from S . mansoni [42] , [43] , as well as SuperSAGE , which can theoretically detect all genes transcribed [47] the most complete data set of genes differentially transcribed between EM and UM was obtained . The credibility of the data was confirmed by the intersection of differentially transcribed genes identified by both techniques , but also by additional real-time PCR experiments . Furthermore , a comparison to data previously generated in other studies [36] , [37] , [90] , [91] supported the reliability of our results . Waisberg et al . [90] found a number of genes for which transcription in males was influenced by final-host sex . Analyzing the transcription of the top 30 of these genes with our data sets revealed no significant differences between EM and UM . Thus an influence of the host sex within our experimental setup can be excluded . Of the transcripts differentially regulated between male stages or strongly up-regulated in at least one male stage found by Fitzpatrick et al . [37] , we found 7 to be differentially regulated by SuperSAGE , including a Ftz-F1 interacting protein ( Smp_090140 . 2 ) , and 14 were differentially regulated according to microarray analysis , including several genes encoding metabolism-associated proteins . From the overlap of differentially transcribed genes between SuperSAGE and microarray only two genes , the AMP-dependent ligase and dock , were found within the data-set obtained by Fitzpatrick et al . [37] . Here , transcriptional regulation showed the same direction as in our study . Our data also confirmed results of one study previously applying SAGE to reveal differences between EM and UM [36] . As far as comparison was possible , a purin nucleoside phosphorylase ( Smp_090520 ) was up-regulated in our microarray analysis as well as in the dataset of Williams et al . [36] . Also a cationic amino acid transporter ( Smp_123010 ) ( down-regulated in EM ) , a fatty acid binding protein ( Smp_095360 . x ) ( down-regulated in EM ) , and a heterogeneous nuclear ribonucleoprotein k ( Smp_065580 . x ) ( up-regulated in EM ) showed the same direction of regulation in our study and were significant within our SuperSAGE data set . Among the transcripts identified in our study to be regulated between EM and UM were also putative antisense RNAs . Some of these may have yet unknown protein-coding function , however , the majority of these RNAs are probably non-coding RNAs ( ncRNAs ) . Their discovery in eukaryotic genomes has significantly influenced research recently , and first evidence for regulatory functions of ncRNAs has been obtained [91] . Also for S . mansoni the existence of antisense RNAs was reported [42] , and it was estimated recently that ≥10% of the transcriptome may represent ncRNAs [43] . Life-stage analyses indicated alterations in the occurrence of specific ncRNAs pointing to diversified functions in biological processes [43] , and we observed antisense RNAs ( microarray: 211; SAGE-exon: 261; SAGE-intron: 107 ) that are differentially transcribed between EM and UM . Their analysis will be subject of future studies , when more knowledge about this class of molecules will be available for schistosomes . This applies also for the number of “hypothetical proteins” identified as being differentially transcribed ( Supplementary Table S4 ) , which could add up to the list of interesting candidates for further analyses of their potential function during pairing-associated processes in males . Applying stringent analysis criteria and focusing on sense-transcripts only , a number of candidate genes were identified , which may be responsible for male competence and/or inducing female maturation . Data interpretation based on a combination of bioinformatics tools permitted first important conclusions ( i–iv ) . According to GO-analyses ( i ) EM seem to loose complexity with regard to functional categories . This interpretation is supported by a previous study , in which less enriched GO categories were found in EM applying a first-generation microarray containing a lower number of gene-representing oligonucleotides [37] . In the same study the authors concluded that worms from mixed-sex populations are transcriptionally less complex than those from unisexual populations . Because in females paired to UM the induction of mitotic activity was found to be delayed compared to the situation in females paired to EM [29] , it was hypothesized that EM and UM differ with respect to their mitosis-inducing capacity , what we like to define as competence that has to be reached before males have the full capacity to govern developmental processes in their pairing partners . The biological variety of transcripts belonging to distinct functional categories found to be differentially regulated indicated that ( ii ) gaining male competence is a process in which different systems are involved . This may also apply to male factors inducing female maturation . Although the involvement of neuronal processes [15] as well as sperm or seminal fluid [12] , [14] as players during male-female interaction was dismissed in the past , ( iii ) we obtained first evidence that neuronal and testes-associated factors nevertheless may be involved . IPA analysis for genes of the intersection highlighted among others DDC , which was further accentuated through the metabolomic data-analysis with DDC as one of two molecules identified . Thus , together with the enhanced DDC transcript level in EM , our data suggest the possibility that neurotransmitters such as dopamine could play a role during male-female interaction . With regard to testes-associated genes , ( 72 genes were found in our study to be differentially transcribed according to microarray or SuperSAGE , which were previously detected as transcriptionally up-regulated in EM testes compared to whole worms [55] . These included dock and the OxlT ( Smp_135020 ) . Our in situ-hybridization experiments for two members of TGFβ-pathways as well as the OxlT ( Smp_036470 ) ( data not shown ) localized transcripts for these molecules to the testicular lobes of EM and UM . The differential regulation of molecules like OxlTs , known for their participation in the indirect proton pump of Oxalobacter formigenes [92] , ( iv ) may indicate their pairing-dependent function in metabolism processes . Previous studies already suggested that EM support females by supplementing their partner with nutrients [21]–[24] , for which they might need a wider functional assembly of molecular processes than UM . In this context it is noteworthy that base metabolic processes seem to differ between the two male stages for anabolic and catabolic pathways , indicating higher nucleic acid synthesis rates in UM . Also , transcription of enzymes involved in carbohydrate metabolic processes , citrate cycle , aerobic respiration , and amino acids metabolic processes was rather up-regulated in UM . Metabolic differences between EM and UM were also found by Williams et al . [36] . The GCP protein was previously reported to be up-regulated in EM as a result of the male-female interaction [25] , [27] , [93] and proposed to be essential for pairing in S . japonicum [26] . However , transcriptional differences between EM and UM for GCP were neither confirmed by our data , nor by previous transcriptome analyses [36] , [37] , [88] , [89] . This discrepancy could be explained by post transcriptional and/or post translational regulations . Interestingly , GCP was proposed to be a downstream target of TGFβ-pathways in schistosome couples [28] . This pathway is well known in schistosomes and was previously pointed out for its possible importance in the female reproductive biology being involved in regulating mitosis and egg production [9] , [10] . Here we provide first evidence for an additional role of TGFβ-pathways during schistosome development as shown by the discovery of SmFst , a potential inhibitor of the TGFβ-pathway . Besides its down-regulation in EM , confirmed by all analyses , SmFst stood out in the IPA network identified for genes within the intersection of microarray and SuperSAGE data . Besides SmFst and its two potential interaction partners SmInAct and SmBMP , two other members of TGFβ-pathways were tested in real-time PCR experiments , Smad4 ( Smp_033950 ) and one S . mansoni activin receptor ( Smp_144390 ) . Apart from SmFst none of the transcripts was differentially transcribed between EM and UM . First functional studies demonstrated that SmFst transcripts were present in male testes and female reproductive organs . Furthermore , Y2H experiments confirmed its potential to interact with SmInAct and SmBMP . In in situ-hybridization experiments SmFst and SmInAct both localized in the female ovary , while SmFst and SmBMP each localized in male testes and the female reproductive organs . In addition , organ-specific RT-PCRs indicated the presence of all three transcripts in EM-testes and the ovary . A previously described transcriptome study [37] did not detect SmFst to be differentially regulated between EM and UM . This may be due to the absence of the corresponding oligonucleotide on the first-generation microarray used ( the respective annotation was not found in the data-set [37] ) , which represented about 50% of the S . mansoni transcriptome . Within the S . mansoni organ-specific transcriptome data [71] SmFst was not found to be up-regulated in EM-testes compared to whole worms using a cutoff for 2-times higher transcription . Compared to our results , this seems not surprising since the EM transcript-level of SmFst was generally low , about two times weaker than in UM according to the calculated log2ratios . With respect to transcript detection in the ovary , results of a previous report on SmInAct [82] corresponded to our localization data . While similar SmInAct transcript levels between EM and UM were detected in the earlier study and our analysis , SmInAct protein was only detected in EM and EF but not in UM or UF before , which suggested that SmInAct expression is linked to the reproductive capacity of the worm [82] . Assuming that SmInAct is the preferential binding partner of SmFst as shown in other organisms [78] , [79] , [94] two possible scenarios exist . First , transcript levels detected for SmFst in this study may not be representative for translation , and thus SmFst interactions would not occur in UM . Secondly , SmFst is translated and may influence TGFβ-pathways in the male testes , however , through other TGFβ-agonists such as SmBMP . Indeed , besides our Y2H results interactions between SmFst-SmBMP have been shown also in other organisms , where they are among others involved in gonad-specific developmental and physiological processes [64] , [65] , [95]–[100] . However , although the presence of SmBMP protein was demonstrated , it was not detected within the testes yet [83] , which may have been caused by protein amounts below the detection limit . From all results available today , we hypothesize a tissue- and stage-dependent interplay of TGFβ-family proteins in schistosomes that also affect the gonads . This view is supported by the presence of various type I and II receptors in the S . mansoni genome [30] , [33] . Most type I receptors belong to the sub-group of activin-like receptors . Among these is an alternatively spliced variant of the S . mansoni TGFβRII [28] , previously described as ActRII [98] . Other members are TGFβRI [100] ( Smp_049760 ) , ActRI ( Smp_093540 . 3 ) , ActRI/BMPRIa ( Smp_124450 ) , ActRIIa ( Smp_080120 . 2 ) or ActRIIb ( Smp_144390 ) . Since multiple receptor-combinations are possible as well as their activation by promiscuously acting agonists [101] , numerous interactions are imaginable with the potential to govern tissue-specific activities . In this scenario , SmFst may regulate TGFβ signaling by binding agonist such as SmBMP in testes of UM , which may impede processes not needed or not intended to occur before pairing . Whether the role of SmFst in schistosomes covers modulating agonist activities in the extracellular environment , or whether it is involved in processing SmInAct or SmBMP pre-pro-peptides to become pro-peptides , a normal part of the activation of these agonists [102] , remains unclear at this stage of its analysis but will be subject of further studies . Although typical furin cleavage sites , being necessary for processing BMP or activin proteins in vertebrates [103] , are present in the schistosome homologs , a role of SmFst in SmInAct or SmBMP protein processing appears unlikely since the main function of FSTs known today is its antagonist activity in the extracellular environment . Here it was shown that FSTs among others play prominent roles in the gonads controlling different testicular and ovarian functions including cell proliferation , apoptosis , folliculogenesis , luteogenesis , hormone release and fertility [104] , [105] . In summary , the presented results demonstrate that processes leading to male competence may be far more complex than hypothesized before by reports suggesting that single molecules from the male and/or nutritional support could be in charge of the fundamental consequences of pairing on female development . From our results we conclude that besides metabolic processes , neuronal processes may be involved in the initial phase of male-female interaction but also TGFβ-signaling , which has been described before to be involved in differentiation processes in fully developed females [9] , [10] and embryogenesis [82] . The meaning of this pathway for schistosome biology appears to go beyond that , since SmFst has leaped into view emerging as a regulatory molecule for TGFβ signal-transduction pathways that is pairing-dependently transcribed in the male gonad probably contributing to processes leading to male competence . | Schistosomiasis is an important infectious disease caused by worm parasites of the genus Schistosoma and directly affects more than 240 million people in 78 tropical and sub-tropical countries but also animals . Pathogenesis is triggered by eggs that are produced by paired females and get trapped in liver and gut causing severe inflammation . While studies have concentrated on the reproductive biology of schistosome females in the past , not much is known about males even though they are indispensable for female sexual development and egg production . Therefore , we studied pairing-dependent processes in S . mansoni males using two independent transcriptomics approaches providing a congruent and most comprehensive data-set on genes being differentially transcribed between pairing-experienced , competent males and pairing-unexperienced , naive males . Besides confirming former studies concerning changes in metabolic processes , our results give new insights into processes leading to male competence indicating among others a potential role of neurotransmitters and TGFβ signal-transduction processes . We especially highlight the follistatin gene SmFst , which codes for an inhibitor of the TGFβ-pathway . SmFst transcription was localized in the testes and found to be down-regulated in pairing-experienced males . This indicates a yet unknown function of pairing on the male gonad and a further role of TGFβ-signaling for schistosome biology . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| []
| 2013 | Combinatory Microarray and SuperSAGE Analyses Identify Pairing-Dependently Transcribed Genes in Schistosoma mansoni Males, Including Follistatin |
Host-to-host transmission of a pathogen ensures its successful propagation and maintenance within a host population . A striking feature of disease transmission is the heterogeneity in host infectiousness . It has been proposed that within a host population , 20% of the infected hosts , termed super-shedders , are responsible for 80% of disease transmission . However , very little is known about the immune state of these super-shedders . In this study , we used the model organism Salmonella enterica serovar Typhimurium , an important cause of disease in humans and animal hosts , to study the immune state of super-shedders . Compared to moderate shedders , super-shedder mice had an active inflammatory response in both the gastrointestinal tract and the spleen but a dampened TH1 response specific to the secondary lymphoid organs . Spleens from super-shedder mice had higher numbers of neutrophils , and a dampened T cell response , characterized by higher levels of regulatory T cells ( Tregs ) , fewer T-bet+ ( TH1 ) T cells as well as blunted cytokine responsiveness . Administration of the cytokine granulocyte colony stimulating factor ( G-CSF ) and subsequent neutrophilia was sufficient to induce the super-shedder immune phenotype in moderate-shedder mice . Similar to super-shedders , these G-CSF-treated moderate-shedders had a dampened TH1 response with fewer T-bet+ T cells and a loss of cytokine responsiveness . Additionally , G-CSF treatment inhibited IL-2-mediated TH1 expansion . Finally , depletion of neutrophils led to an increase in the number of T-bet+ TH1 cells and restored their ability to respond to IL-2 . Taken together , we demonstrate a novel role for neutrophils in blunting IL-2-mediated proliferation of the TH1 immune response in the spleens of mice that are colonized by high levels of S . Typhimurium in the gastrointestinal tract .
Host-adapted pathogens depend on their host for transmission and dissemination within a population . Recent epidemiological studies have uncovered heterogeneities in infection wherein a minority of the infected individuals ( 20% ) are responsible for the majority of the infections ( 80% ) , described as the 80/20 rule [1] . In the case of pathogens transmitted via the fecal-oral route , these individuals are the ones that shed the highest numbers of bacteria . Recent studies on the transmission of Escherichia coli O157 within cattle herds demonstrated that over 95% of the infections were caused by between 8–10% of the most infectious individuals , or super-shedders [2]–[4] . Identification of these individuals is required for control of the infection [1] , [5] , [6] . However , little is known about what distinguishes them from other infected hosts . Salmonella enterica serovar Typhi , the causative agent of typhoid fever in humans , is a human-adapted pathogen and establishes a persistent long-term infection in about 1–6% of the infected hosts [7] , [8] . These individuals are known as typhoid carriers and periodically excrete large amounts of the bacilli in their feces , thereby offering both a reservoir for the pathogen and the opportunity of transmission to new hosts . However , they do not display any of the clinical signs characteristically associated with typhoid fever [7] , [9] . While individuals with acute infections can transmit the pathogen for brief periods of time , for the purposes of this study , we will focus on transmission from persistently infected hosts who play a far larger role in transmission of host-adapted pathogens [1] , [2] , [4] , [10] . Host immune responses to persistent microbial infections must balance between control of the pathogen and minimizing inflammatory damage to the host [11] . To this end , chronic viral infections often result in a contraction of the adaptive immune response , an example of which would be T cell exhaustion [12] , [13] . An ineffective CD4 T cell response , characterized by anergy or apoptosis , has also been observed in persistent bacterial infections such as with Helicobacter pylori , Staphylococcus aureus and Salmonella enterica serovar Typhimurium [14]–[16] . Intriguingly , regulatory T cells have also been shown to lose their suppressive ability during the later stages of persistent S . Typhimurium infection [17] . Characterizing the host immune response in individuals that transmit disease is important for two reasons . First , understanding the mechanistic differences in this subset of hosts might explain the heterogeneity of infectiousness observed in relatively homogenous populations , such as with inbred herds of cattle . Second , such studies could lead to the development of biomarkers that are unique to the identification of individuals with the highest risk of transmitting the pathogen within a population . Modeling the immune state using laboratory animals allows us to dissect the mechanisms behind host-pathogen interactions that lead to transmission . Our lab has established a mouse model of persistent S . Typhimurium infection wherein 30% of the infected mice , termed super-shedders , shed >108 Salmonella and rapidly transmit disease to naïve cage mates [18] . This variation in infectiousness is observed in inbred strains amongst cage mates and siblings , implying predisposition to super-shedder status might not be heritable . Super-shedder mice also develop colitis , displaying moderate to severe inflammation in the colon and ceca . Surprisingly , super-shedder mice do not display outward signs of illness such as ruffled fur , fever or malaise [18] suggesting that they can tolerate , and perhaps control the inflammation . How , then , does the interplay between pathogen and the murine immune system evolve to allow such high levels of gastrointestinal Salmonella in some mice , but not others ? The host immune response to Salmonella infections has been characterized primarily in mice that have increased susceptibility to intracellular pathogens due to the presence of a mutated Nramp1 gene [19]–[21] . In susceptible mice , wild type S . Typhimurium infection results in death within one week of infection . In contrast , these susceptible mice strains survive infection with attenuated S . Typhimurium strains ( e . g , AroA− AroD− mutant ) . In this model the adaptive immune response to persistent Salmonella infection was found to be TH1 biased and was dependent upon expression of transcription factor T-bet [22] . In this study , we used a mouse strain ( 129x1/SvJ ) that carries functional Nramp1 and a non-attenuated Salmonella strain to more closely mimic natural infection [23] . This persistent infection model was used to investigate host immune adaptations to gastrointestinal inflammation that are associated with survival . We hypothesized that the ability of the host to co-exist with large numbers of bacteria in the gastrointestinal tract requires either a dampened systemic inflammatory response or the ability to tolerate inflammation . To test this , we compared the host immune responses in super-shedder and moderate-shedder mice and characterized an immune state specific to super-shedders . Broadly , we found super-shedders have an activate innate inflammatory response in both the gastrointestinal tract and the systemic organs but a dampened adaptive T cell response specific to the systemic sites . Finally , we identify an unexpected host immune mechanism mediated by neutrophils that controls TH1 cell expansion in the super-shedder immune state .
To examine the host immune response to persistent Salmonella infection , we first enumerated the bacterial burden in gastrointestinal and systemic tissues of mice infected with S . Typhimurium for 30 days . Bacterial loads in the spleen and the mesenteric lymph nodes ( MLN ) were tightly clustered across mice , while those in the gastrointestinal sites showed the large variation characteristic of super-shedders and moderate-shedders ( [18] , Figure 1A ) . It is known that super-shedder mice develop colitis that results in an influx of granulocytes to the colon [18] . We identified moderate and super-shedder mice as outlined in material and methods and asked if the neutrophilia extended to secondary lymphoid organs . We observed a notable increase in the frequency of Gr1+ ( Ly6G+Ly6C+ ) granulocytes in the spleen , MLN , and blood ( Figure 1B ) of super-shedder mice compared to moderate-shedders . Super-shedder spleens were composed of 12±2 . 8% neutrophils; in contrast , on average , the moderate-shedder spleen contained only half that amount ( 5 . 3±1 . 1% neutrophils ) . Histological analysis demonstrated that neutrophils constituted over 80% of splenic myeloid cells ( Fig . 1C ) . Neutrophils were also the most numerous cell type in super-shedder blood , with 63 . 6±21 . 6% of the non-red blood cells identified as circulating neutrophils , while moderate-shedder blood contained only 25 . 3±3% neutrophils . When compared to uninfected mice , the levels of neutrophils in the systemic organs of moderate-shedders were significantly higher ( Figure 1B ) . However , no significant difference in colonic neutrophil levels between uninfected and moderate-shedder mice was observed indicating that neutrophilia in the systemic sites was a stronger indicator of shedding status . Taken together , our data suggest that neutrophil levels in the systemic sites are positively correlated with gastrointestinal bacterial load . Having shown that splenic neutrophilia varied with gastrointestinal and not splenic bacterial burden , we examined whether there were associated variations in adaptive immune responses . To exploit the variation in fecal shedding , we asked what aspects of the host immune response varied with fecal bacterial load . Given the importance of CD4 T cells in the host defense against Salmonella infection , [24] we first focused on the frequencies of two CD4 T cell subsets: TH1 and Tregs . The host immune response against Salmonella infection requires the induction of a CD4 type 1 Helper T cell or TH1 immune response , involving CD4 T cells expressing the transcription factor T-bet . TH1 activity can in turn be controlled by regulatory T cells ( Tregs ) expressing the transcription factor FoxP3 . In one representative experiment , we measured these subsets in the spleens and colons of 12 mice of which five were super-shedders . This was confirmed by the levels of colonic inflammation though 2 of the 5 mice were shedding between 107 and 108 cfu/gm . The infected mice clustered into two distinct groups with 5 of the 7 moderate-shedders clustered together with fewer Tregs and more TH1 cells while 4 out of the 5 super-shedders were in a cluster that contained fewer TH1 and more Tregs ( pink box vs . blue box , Figure 2A ) . The percentage of T-bet+ CD4 T cells ( TH1 ) cells in the spleen significantly negatively correlated with fecal bacterial burden ( Spearman's correlation = −0 . 58 ) but was not significantly correlated to the splenic bacterial burden ( data not shown ) . Additionally , this dichotomy between active and suppressive T cells was not observed in the colon ( Figure S1A ) . Importantly , in uninfected mice there was a positive correlation between the frequencies of CD4 T cells expressing T-bet and those expressing FoxP3 , indicating that the skewing in the populations of TH1 and Treg cells is dependent on infection and is not a result of an underlying natural variation in the uninfected mouse population ( Figure S1B ) . We found very few Rorγt-expressing CD4 T cells in persistently infected mice ( data not shown ) . Infection-induced variation was further evidenced by the relationship between TH1 and neutrophil percentages in the spleens of infected mice . All 5 super-shedder mice clustered together ( blue box ) with high levels of neutrophils and correspondingly lower levels of TH1 cells . All moderate-shedders clustered on the opposite end with fewer than 5% splenic neutrophils but higher frequencies of TH1 cells ( Figure 2B ) . Correspondingly , splenic neutrophilia was significantly positively correlated with fecal bacterial load ( spearman's correlation R = 0 . 9 ) . Importantly , splenic bacterial load did not correlate with fecal bacterial load ( Figure S2A ) . CD4 T cell exhaustion is a hallmark of persistent viral infections [12]–[14] so we sought to determine if these TH1 cells maintained antigen-responsiveness . Splenocytes from infected mice were incubated with S . Typhimurium-infected macrophages and the level of intracellular IFNγ levels was measured . Salmonella-specific IFNγ+ Tbet+ CD4 T cells were first detected at 8 days post-infection and expanded continuously through 30 days post-infection ( Figure S3A ) . While culling of flagellin-specific CD4 T cells has been previously reported during Salmonella infection [16] , we saw a steady expansion of total memory effector CD4 T cells over a time course of infection ( Figure S3B ) . Antigen specific IFNγ production was observed in CD4 T cells across all mice , regardless of shedder status ( data not shown ) . These data show that while super-shedder spleens have lower numbers of TH1 T cells relative to moderate shedders , these cells still make IFNγ in response to Salmonella antigen . Finally , we asked if the dampened TH1 response also resulted in reduced antibody production . To determine this , we measured anti-Salmonella antibodies in the serum of persistently infected mice and found no correlation with shedding status ( Figure S2B ) indicating that the blunted TH1 response in super-shedders did not affect antibody production . Having determined that the TH1 cells were antigen-responsive , we further investigated the ability of CD4 T cells to respond to IL-2 , a cytokine that induces T cell proliferation . The high affinity IL-2 receptor is expressed on Tregs and memory effector CD4 T cells; upon binding IL-2 one of the first steps initiated in the intracellular signaling cascade is the phosphorylation of STAT5 protein . Thus , the mean fluorescence intensity ( MFI ) level of phosphorylated STAT5 was measured in the subset of CD4 T cells that responded to IL-2 ( gated on pSTAT5+CD4+ T cells ) . This metric represented the degree of cytokine responsiveness and correlated negatively with the gastrointestinal Salmonella burden ( Figure 2C ) . pSTAT5 response to ex vivo IL-2 stimulation has been previously established as an indicator of T cell expansion in vivo [25] . To determine if this reduction in IL-2 responsiveness in super-shedders coincided with reduced T cell proliferation we measured the expression of Ki-67 , a marker of actively proliferating cells . Consistent with decreased IL-2 responsiveness , super-shedder mice had significantly fewer Ki-67+ CD4 T cells in the spleen ( Figure 2D ) . That STAT5 phosphorylation correlated inversely with the levels of bacterial shedding suggests that a feature of the super-shedder immune response involves blunting of IL-2 responsiveness . We therefore evaluated the extent to which other alterations in persistent immune responses were linked to fecal shedding status . Given the dampened IL-2 responsiveness of CD4 T cells , we asked if the ability to respond to other cytokines was also impaired . Previous work in a mouse model of septicemia showed that naïve splenic CD4 T cells have a dampened response to IL-6 compared to uninfected mice , indicated by a reduced ability to phosphorylate STAT1 in response to ex vivo IL-6 stimulation [26] . In the CD4 T cell compartment , the IL-6 pSTAT1 response is primarily restricted to naïve cells , as memory effector cells express very little IL-6 receptor ( data not shown ) . Naïve CD4 T cell pSTAT1 responsiveness to IL-6 negatively correlated to fecal shedding levels ( Figure 2E ) , reminiscent of the IL-2 response observed earlier . It is important to note that the IL-6 response is dampened in super-shedder mice only with respect to moderate or low shedders . Compared to uninfected mice , Salmonella infection induces increased IL-6 responsiveness in naïve CD4 T cells ( Figure S4 ) but this responsiveness varied with the levels of fecal shedding . Furthermore , circulating IL-6 levels were significantly higher in super-shedder mice compared to moderate-shedders and uninfected mice ( Figure 2F ) . Taken together , these results reveal that across equivalently infected mice , those that developed as super-shedders are characterized by an activated innate inflammatory response with high levels of circulating IL-6 and neutrophils that is associated with a spleen-specific dampened CD4 T cell response . This dampened T cell response is characterized by a partial loss of cytokine responsiveness to IL-2 and IL-6 compared to moderate-shedder mice . Finally , in super-shedder spleens , the balance of CD4 T cell subsets supports a dampened CD4 T cell response , with fewer TH1 cells and more Tregs ( Table 1 ) . Does the gastrointestinal Salmonella burden dictate the systemic immune profile , or does the immune response control the bacterial load ? Since the numbers of bacteria in the gastrointestinal tract are correlated with specific changes in the neutrophil and CD4 T cell immune response in the spleen , we investigated whether altering the levels of S . Typhimurium in the gut was causal to changes in the splenic immune response . Previously it was shown that alterations of the microbiota in moderate-shedders via a single dose of streptomycin resulted in super-shedder levels of Salmonella in the gastrointestinal tract [18] , [27] . Therefore , moderate-shedder mice were treated with an oral dose of streptomycin and their fecal shedding and splenic immune state were monitored ( the Salmonella strain used - SL1344 - is resistant to streptomycin ) . Three days after streptomycin treatment , moderate-shedders shed >108 Salmonella per gram of feces , i . e . super-shedder levels . Their splenic bacterial burden remained unchanged , as compared to untreated moderate-shedder mice ( Figure 3A ) . Moreover , within 3 days streptomycin-treated moderate-shedders developed increased levels of neutrophils in the colon and spleen , comparable to those seen in super-shedders ( Figure 3B ) . This was accompanied by a decrease in splenic TH1 cells in 3 out of the 5 streptomycin-treated mice ( Figure 3C ) , and by one week post-treatment , all of the streptomycin-treated moderate-shedders had fewer splenic TH1 cells ( data not shown ) . Notably , streptomycin-treated moderate-shedders had increased levels of circulating IL-6 and a concomitant decrease in the ability of CD4 T cells to respond to IL-6 as measured by pSTAT1 levels ( Figure 3D , Figure 3E ) . However , the percentage of splenic Tregs did not change ( Figure S7A ) . Therefore , many aspects of the splenic super-shedder immune phenotype are induced by raising gastro-intestinal levels of Salmonella , although the frequency of regulatory T cells in the spleen is independently regulated . We next sought to identify which components of the host immune response control the dampened TH1 response observed in the spleens of super-shedder mice as compared to moderate shedders . Increased neutrophil numbers were seen in the spleen as early as four days post-infection , a time point at which Salmonella was undetectable outside the gastrointestinal tract ( Figure S5 ) . Since increased levels of neutrophils in the colons and spleens of super-shedder mice correlated with the dampened adaptive TH1 immune response , we proposed that neutrophils play a role either directly or indirectly in mediating the immune blunting . We depleted neutrophils using the monoclonal antibody RB6 , which targets cells expressing both Ly6C and Ly6G . Neutrophil depletion increased the levels of splenic TH1 cells from 10 . 2±5 . 8% in control mice to 24 . 1±8 . 8% in RB6-treated mice ( Figure 4A ) . Similar results were obtained with a Ly6G-specific depletion antibody , IA8 ( Figure S6A , B ) . An increase was observed in the frequency of pSTAT5+ CD4 T cells that responded to ex vivo IL-2 stimulation regardless of infection , indicating that neutrophils suppress CD4 T cell responsiveness to IL-2 ( Figure 4B ) . Intriguingly , uninfected mice depleted of neutrophils also showed a similar increase in pSTAT5+ CD4 T cells , but without TH1 expansion . This indicates that the IL-2/pSTAT5 response may be an intermediate step to TH1 expansion and that TH1 biasing occurs only in the context of infection . The percentage of splenic Tregs in infected and uninfected mice did not statistically change upon neutrophil depletion , implying that the increase in IL-2 responsive CD4 cells was not due to Treg reduction ( Figure 4C ) . In addition , RB6-treated mice had significantly higher bacterial burdens in the spleen compared to control mice ( Figure S6C ) . This demonstrates that neutrophils are necessary to control splenic infection , and that the increased TH1 response is unable to compensate for neutrophil depletion . Surprisingly , there was no difference observed in fecal bacterial burden ( Figure 4D ) suggesting that there might be an organ specific function for neutrophils in persistent Salmonella infection . Taken together , the results indicate that high levels of neutrophils in the spleen are necessary for both dampened IL-2 responsiveness and a reduction in the levels of of TH1 cells . Based on the finding that neutrophil depletion induced TH1 expansion , we investigated whether the strong systemic neutrophil induction seen in super-shedder mice was sufficient for limiting the TH1 response . Moderate-shedder and uninfected mice were injected with G-CSF for 3 days to induce granulopoiesis . G-CSF treatment increased splenic neutrophil levels to those observed in the spleen of super-shedder mice ( Figure 5A ) . Importantly , G-CSF treatment of moderate-shedder mice led to a concomitant decrease in the frequency of splenic TH1 cells ( Figure 5B ) . However , G-CSF treatment did not influence the frequency of Tregs in the spleens of moderate-shedder or uninfected mice ( Figure S7B ) . Histological analysis of the spleen and bone marrow of G-CSF-injected mice revealed that the granulocytes induced were primarily neutrophils . Additionally , there was a significant increase in the levels of immature neutrophils in the bone marrow of G-CSF-treated moderate-shedders compared to untreated mice ( Figure S8A–C ) . Since G-CSF treatment caused a reduction in splenic TH1 cell frequencies , we assayed whether cytokine responsiveness of CD4 T cells was also blunted . Compared to untreated moderate-shedders , naïve CD4 T cells from G-CSF-treated moderate-shedders displayed a dampened pSTAT1 response to IL-6 stimulation similar to that observed in super-shedders ( Figure 5C ) . This was finding was in alignment with the trend towards increased levels of serum IL-6 in G-CSF-treated moderate-shedders ( Figure S9 ) . G-CSF treatment also significantly dampened IL-2-mediated induction of pSTAT5 in CD4 T cells that responded to IL-2 ( Figure 5D ) . Furthermore , uninfected mice treated with G-CSF demonstrated dampened responses to IL-2 and IL-6 , indicating that neutrophil-mediated control of IL-2 induced pSTAT5 and IL-6 induced pSTAT1 responses are independent of infection ( Figure 5C , 5D ) . These data suggested that neutrophils might suppress TH1 expansion via the IL-2/pSTAT5 pathway . Treatment with G-CSF did not increase fecal bacterial load ( Figure 5E ) . Importantly , in ex vivo experiments , G-CSF induced STAT5 activation in granulocytes but not CD4 T cells ( data not shown ) , indicating that G-CSF does not act directly on T cells . Having observed that G-CSF mediated neutrophilia dampens IL-2 responsiveness across the CD4 T cell population; we next investigated whether TH1 and Treg CD4 T cell subsets differed in their IL-2 responsiveness with functional consequences . Both Tregs and TH1 cells activated by infection induce phosphorylation of STAT5 in response to IL-2 ( Figure S7C ) . Previous studies have shown that IL-2 antibody complexed with IL-2 cytokine ( hereafter referred to as IL-2 antibody complex ) can induce expansion of Tregs in uninfected mice [28] , [29] . When S . Typhimurium-infected mice were treated with IL-2 antibody complex , we observed expansion of both Tregs and TH1 cells ( Figure 6B ) . This was accompanied by an increase in the number of pSTAT5+ total CD4 T cells both before ( basal ) and after ex vivo IL-2 stimulation ( Figure 6A ) . Furthermore , IL-2 mediated TH1 expansion was significantly greater in moderate-shedders than super-shedders ( p<0 . 05 ) . These results indicate that high levels of gastrointestinal Salmonella burden and neutrophilia may be associated with an impairment in the ability of splenic TH1 cells to undergo IL-2 mediated proliferation . Our findings that both gastrointestinal Salmonella and G-CSF-mediated neutrophilia are associated with dampened IL-2 responsiveness in TH1 cells , suggest that neutrophil levels influence the ability of TH1 cells to expand . To investigate this , moderate-shedders were treated with G-CSF for 3 days , then subsequently administered IL-2 antibody complex for another 2 days , to determine the effect of neutrophilia on T cells expansion . After IL-2 antibody complex treatment , moderate-shedders pretreated with G-SCF had lower levels of both basal and IL-2 responsive pSTAT5+ CD4 T cells compared with mice that were not administered G-CSF ( Figure 6C ) . This loss of IL-2 responsiveness correlated with significantly fewer Ki-67+ CD4 T cells ( Fig . 6D ) , indicating that G-CSF treatment inhibited the ability of CD4 T cells to proliferate in response to IL-2 . However , this inhibition was specific to TH1 cells , as G-CSF did not affect Treg expansion in response to IL-2 antibody complex ( Figure 6E ) . This result is consistent with our previous finding that only TH1 cells and not Tregs expanded upon neutrophil depletion ( Figure 4A , 4C ) . Thus , CD4 T cells in G-CSF treated moderate-shedders recapitulate the super-shedder phenotype . Taken together , we show that treatment of moderate-shedders with G-CSF is sufficient to recapitulate specific aspects of the super-shedder immune response ( Figure 7 ) .
We have described here a unique immune phenotype in the spleen that is linked to the ability of an enteric bacterial pathogen to replicate to high numbers in the gastrointestinal tract and thus transmit to a new host . One interesting aspect of this phenotype is that the immune state in the spleen is associated with the levels of bacteria in a distal site ( the gastrointestinal tract ) rather than local bacterial burden . Previous work has shown that the gastrointestinal commensal microbiota can activate antigen-presenting cells , which go on to drive adaptive immunity in distal sites such as the lung [30] . However , to the best of our knowledge , this is the first report of a splenic immune phenotype specifically associated with gastrointestinal pathogen load and inflammation . The super-shedder immune phenotype is composed of a highly inflammatory response in both gastroinestinal and systemic sites ( evidenced by neutrophilia and serum IL-6 ) but a dampened adaptive T cell immune response specific to the systemic sites . The high levels of circulating IL-6 and moderate to severe colonic inflammation seem at odds with the absence of weight loss or malaise observed in these mice . The blunted splenic CD4 T cell response , dampened cytokine responsiveness and increased levels of regulatory T cells might be instrumental in the tolerance of the inflammatory environment that is the super-shedder gut . In persistent bacterial infections , this might provide an opportunity for the host to suppress the long-term inflammatory effects of the adaptive T cell response while still controlling pathogen load via neutrophil recruitment . The molecular mechanism behind neutrophil recruitment in persistent Salmonella infection is yet to be identified . Gastrointestinal Salmonella is sufficient to induce granulopoiesis and systemic neutrophilia . Neutrophils were observed in the spleen at four days post infection , a time point when Salmonella was not detected in systemic sites but was present in the gut . Previous studies have demonstrated a role for IL-17 mediated neutrophil induction [31] . IL-17 is secreted by multiple cell types including Rorγt expressing CD4 T cells or TH17 cells which have also been shown to play a role in acute Salmonella infection [32] , [33] . However , we found very few Rorγt+ CD4 T cells in the spleen of persistently infected mice ( data not shown ) . Additionally we did not detect IL-17 in serum or spleen supernatant of these mice indicating that the TH17 or IL-17 response does not play a role in persistent Salmonella infection ( data not shown ) . Two interventions shifted moderate shedders towards the super-shedder immune phenotype – streptomycin induction of gastrointestinal bacterial expansion and G-CSF mediated neutrophilia . In both instances , the cytokine signaling profiles and TH1 levels recapitulated those of natural super-shedders . However , a concordant increase in Tregs was not observed . This suggests that the increased Treg levels observed in super-shedders are regulated by a mechanism independent of bacterial load or neutrophil levels . Indeed , the co-expansion of Tregs and TH1 cells is decoupled in super-shedder mice where IL-2 mediated TH1 but not Treg expansion is dampened . It was surprising to find that short-term treatment with G-CSF was sufficient to recapitulate this phenotype . The mechanism behind this and other aspects of the super-shedder immune state remain unknown , as do the bacterial and host effectors required to establish this immune phenotype . The mechanism by which induction of granulopoiesis inhibits the TH1 response , both in T-bet expression and IL-2 mediated expansion is unclear . It will be important to determine whether this mechanism is mediated by cell-cell contact or through cytokine secretion . Anti-inflammatory neutrophil populations which secrete IL-10 in response to Gram-negative bacteria have been previously described [34] . However , we were unable to detect IL-10 in supernatants of cultured neutrophils isolated from moderate and super-shedder mice . Additionally , co-administration of IL-10 receptor-blocking antibody along with G-CSF did not prevent the dampening of the TH1 response of infected mice treated with G-CSF ( data not shown ) . One of the notable findings of this study was that Tregs are not suppressed in a similar manner to TH1 cells . Previous work has shown that Tregs lose potency in the later stages of persistent Salmonella infection [17] . Our data indicated that Tregs from mice infected for 30 days were capable of IL-2-mediated expansion . However , the suppressive potency of these cells was not analyzed and it is possible that the expanded population is not capable of suppressing T cells . While induction of granulopoiesis was sufficient to induce aspects of the super-shedder systemic immune network in moderate-shedder mice , no differences were seen in fecal shedding . We suspect that this was because the persistent Salmonella infection had already been established . The only factors shown to induce super-shedder status so far have been antibiotic-mediated ablation of the gut microbiota or neutralization of IFNγ cytokine [18] , [23] , [27] . The immune processes pivotal in the establishment of the super-shedder state may take place at the onset of infection . It should also be noted that G-CSF induction only mimicked the super-shedder phenotype in the spleen . The colonic and cecal inflammation typically seen in super-shedders did not develop in G-CSF-treated moderate-shedders ( data not shown ) . It is possible that either a longer G-CSF treatment or Salmonella-specific induction of host chemokine responses are required for sustained gastrointestinal neutrophilia associated with gastrointestinal pathology . As noted , the systemic super-shedder immune phenotype is primarily one of an active inflammatory response and a dampened TH1 response . However , these TH1 cells are not anergic , as they retain the ability to recognize Salmonella antigen and secrete IFNγ in response . The serum levels of IFNγ were elevated by four days post-infection and remained consistently high during the first month of infection ( Figure S3A ) . Since the cytokine levels were elevated before the T cell response was initiated , these data suggest that antigen-specific CD4 T cells may not be the primary source of serum IFNγ during infection . It is likely that the source of the early IFNγ is activated macrophages or monocytes . This is further supported by our observation that serum IFNγ did not correlate with fecal shedding and was not a part of the super-shedder immune phenotype . Salmonella-induced blunting of flagellin-specific T cells has been previously reported [16] , however , while we did not investigate the clonality of the CD4 T cell population , we saw a steady expansion of antigen-specific , IFNγ producing , memory effector CD4 T cells ( Figure S3 ) . This study describes a distinct splenic immune signature in the Salmonella carrier state responsible for transmission . These results might have important implications for identification of S . Typhi carriers . For instance , current serological methods do not distinguish between carriers and people who have cleared the infection [35] , [36] . While Typhi infection induces a neutrophillic inflammatory response , clinical reports on typhoid patients have only described a mild transient granulocytosis [37]–[39] and neutrophils are typically not found in stool samples of S . Typhi patients [40] . It remains unknown if any variation in neutrophil frequency was observed in typhoid patients and if this correlates with carrier status . For example , serum IL-6 levels are elevated in patients with typhoid and have been correlated with prolonged fever [41] . However , in these studies and others , immune correlates with fecal shedding are difficult to obtain . A study on 13 long-term typhoid carriers actively shedding Typhi demonstrated a wide variance in antibody titers in accordance with our data in mice infected with S . Typhimurium ( [9] , Figure S2B ) . In other animal models , studies in S . Typhimurium-infected swine uncovered positive correlations between early fecal shedding of salmonella ( within the first week of infection ) with increased neutrophil numbers and high serum IFNγ [42] . Identifying immune correlates that are associated with active shedding of Salmonella by the host would help determine biomarkers for screening of Typhoid carriers . Interestingly , there is evidence for an immune phenotype associated with human S . Typhi carriers . Transcriptional profiling of a cohort of acute , convalescing and recovered typhoid patients uncovered specific neutrophil and lymphocyte gene expression sets associated with each of those stages . Specifically , while 50% of the convalescing patients had a gene expression signature indistinguishable from healthy controls , 25% of the patients showed a distinct gene expression pattern in multiple cell types that were more similar to those of newly admitted patients despite being collected 9 months after treatment . This dataset supports the possibility of a long-term alteration in immune response in a subset of S . Typhi patients [43] . The results presented here are a step toward the definition and practical identification of immune states associated with high levels of Salmonella fecal shedding and transmission . That this is mediated by a novel neutrophil-dependent mechanism of IL-2- mediated TH1 blunting , suggests novel disease management approaches both for individuals and human communities where Salmonella is endemic .
All animal experiments were performed in accordance with Stanford University's Institutional Animal Care and Use Committee and NIH Guidelines for Euthanasia of Rodents Using Carbon Dioxide . All animal experiments were approved by Stanford University's Administrative Panel on Laboratory Animal Care ( A-PLAC ) . Stanford University Animal Welfare Assurance Number: A3213-01 . Protocol ID 12826 . All animals were housed in a centralized research animal facility , fully staffed with trained personnel and accredited by the Association of Assessment and Accreditation of Laboratory Animal Care International ( AAALAC ) . Mice were monitored daily; mice displaying signs of pain , distress ( hunched posture , lethargy , ruffled fur ) and weight loss were euthanized humanely . Salmonella enterica serovar Typhimurium wild type strain SL1344 was used for all infections . This strain is resistant to streptomycin . The bacteria were grown , shaking , at 37°C overnight in Luria-Bertani ( LB ) broth containing 200 µg/mL streptomycin . Bacteria were spun down and washed with phosphate buffered Saline ( PBS ) before being resuspended into the desired concentration for infection . For macrophage infection , bacteria were diluted to the desired concentration and pipetted onto the cells . Female 129x1/SvJ mice were obtained from The Jackson Laboratory and infected when they were 7–9 weeks old . Food and bedding were changed once a week by the Stanford Animal Facility and access to food and water was unlimited . For infections , food but not water was removed 12–16 hours ahead . Mice were infected orally , drinking 108 CFU in 20 µL PBS from a pipette tip . For the streptomycin treatment , the antibiotic was delivered orogastrically with 5 mg streptomycin ( Sigma Aldrich , S6501 ) dissolved in 100 µL PBS . Where indicated , infected mice were identified as super-shedders ( fecal shedding >108 cfu/gm ) and moderate-shedders ( fecal shedding <106 cfu/gm ) at 15 days post infection . To determine fecal bacterial load , fresh fecal pellets were collected by placing individual mice in sterile isolation chambers until 2–4 pellets were excreted . These were weighed and placed in 500 µL PBS . Pellets were resuspended via vortexing and CFUs were determined by plating log dilutions on LB agar plates containing 200 µg/mL streptomycin . Fecal CFU was checked at 2–4 timepoints between 15 and 30 days post infection . Mice that consistently shed <106 cfu/gm at all times were identified as moderate-shedders while mice that shed >108 cfu/gm at all times were identified as super-shedders . Infected mice remained with their original cage mates even after determination of shedding status . Super-shedder status was also confirmed by verifying colonic and cecal inflammation after sacrifice of the animal . Individual organs were collected , weighed and homogenized in 1 mL of PBS and log dilutions plated onto LB agar containing 200 µg/mL streptomycin . Spleens from mice were mechanically dissociated into single cell suspensions in RPMI ( Gibco , 11875 ) media with 10% fetal bovine serum ( FBS ) ( Gibco , 26140 ) using glass slides . Spleens weighing less than 0 . 25 gm were suspended in a total volume of 10 mLs , those between 0 . 25–0 . 4 gm in a total volume of 15 mLs while spleens weighing more than 0 . 4 gm were resuspended in total of 30 mLs . Mesenteric lymph nodes were dissociated into single cell suspensions using a motorized pestle into 1 mL ( Kontes , K749540-0000 ) . All single cell suspensions were then filtered through a 70 µm cell strainer ( BD Falcon , 352350 ) . Blood was collected via cardiac puncture into a syringe containing 100 µL Heparin ( BD , 366480 ) . Samples were spun down for 10 minutes at 8000 rpm and serum removed and stored at −80°C . The cell pellet was then resuspended in 1 mL PBS . The blood sample was then treated with ACK buffer ( 0 . 15 M NH4Cl , 10 mM KHCO3 , 0 . 1 mM Na2-EDTA , pH 7 . 4 ) to lyse red blood cells as described in [44] . Colons were isolated , opened , cleaned and washed four times with PBS . Colonic tissue was subsequently cut into 5–10 mm pieces and added to 10 mLs of RPMI containing 10% FBS , 0 . 5% Hepes ( Sigma , H3375 ) , 0 . 1% β Mercaptoethanol ( Sigma , M3148 ) as well as the following enzymes at 1 mg/mL: Collagenase Type 1A ( Sigma C9891 ) , DNase I ( Roche . 10104159001 ) , Trypsin Inhibitor Type 1-S ( Sigma , T6522 ) . The tissue was gently agitated every 15 minutes and incubated for an hour at 37°C . The cells were then passed through a 70 µm filter and spun down at 1500 rpm . Leukocytes were then isolated from the colonic cells using CD45+ microbeads ( Miltenyi Biotec , 130-052-301 ) as described in the product datasheet . All cells were then fixed with 1 . 6% paraformaldehyde ( Electron Microscopy Sciences , 15710 ) for 10 minutes at room temperature , washed twice with FACS buffer ( PBS containing 0 . 5% BSA and 0 . 02% sodium azide ) . Cells were then either stored in methanol or permeabilized with saponin and stained as described in the flow cytometry section . Cytokine stimulations were conducted as described in [26] . Briefly , single cell suspensions of splenocytes were recovered for 30 minutes at 37°C . Splenocytes were then left unstimulated or stimulated with 40 ng/mL IL-2 or IL-6 ( BD Biosciences ) for 15 minutes at 37°C . Splenocytes were then fixed as described previously , spun down and resuspended in cold methanol . Samples were stored at −80°C until staining for flow cytometry . Samples stored in methanol were washed twice with FACS buffer . Cells were stained for 30 minutes with surface marker antibody cocktail comprised of Gr1-APC-Cy7 , B220-PE-Texas Red , CD4-Alexa Fluor 700 , CD11b-PerCP-Cy5 . 5 , CD44 ( -V500 or in-house conjugated to Pacific Orange ) , CD25-PE , Ki-67-Alexa Fluor 647 and CD62L-biotin ( with streptavidin-quantum dot605 secondary ) and phospho-Stat antibodies pStat5-Alexa Fluor 488 or pStat1-Alexa Fluor 488 . All these antibodies were purchased from BD Biosciences . For measurement of transcription factors , FoxP3-PE and T-bet-Alexa Fluor 647/eFluor 660 were used ( eBioscience ) . After staining , the cells were washed with FACS buffer and run on an LSR II flow cytometer ( Becton Dickinson ) . Cells were acquired with DIVA software ( BD Biosciences ) and analyzed using FlowJo software ( Tree Star ) . Cells were either measured as a percentage of total intact cells ( determined by forward and side scatter measurements ) or as a percentage of a specific cell type ( e . g . total CD4 T cells ) . Alternatively , when measuring phospho-protein expression , median fluorescence intensity ( MFI ) of the cell populations was used . Staining for transcription factors was carried out using saponin permeabilization buffer ( PBS containing 0 . 3% saponin , 0 . 5% BSA and 0 . 02% sodium azide ) . Cell populations were gated as described in Figure S10 . Mice were injected with 1 µg each of one of two different neutrophil depletion antibodies; anti-Ly6G ( clone IA8 , BioXcell , BE0075-1 ) and anti Gr1 ( clone RB6-8C5 , BioXcell , BE0075-1 ) or PBS controls . Antibodies or PBS were administered intraperitoneally every day for 3 days and mice sacrificed on the fourth day . Serum was collected as described above and IFNγ ( eBioscience , 88-8314-22 ) and IL-6 ( BD Bioscience , 550950 ) levels were measured using sandwich ELISA kits . ELISA plates were coated with Salmonella lysate for 1 hour at 37°c for 1 hour , then blocked with 3% Bovine Serum Albumin in PBS for 1 hour . Serum samples were then diluted 10 fold in wash buffer for a minimum of six serial dilutions . Samples were incubated for 2 hours at 37°c , washed , and incubated with Biotin conjugated anti- mouse IgG ( Abcam , ab64255 ) for a further 2 hours at the same temperature . Plates were washed then incubated with Streptavidin conjugated Horse Radish Peroxidase ( R&D , DY998 ) for 30 minutes , washed and incubated with TMB substrate ( BD , 555214 ) for 10 minutes , then stopped with 2N H2SO4 . Titers were estimated by determining the lowest sample dilution with an optical density reading higher than undiluted serum from uninfected mice . All washes were carried out 5 times in between all steps using wash buffer consisting of 0 . 5% Tween in PBS . Spleens were processed as described above and bone marrow was harvested by flushing a single tibia with 1 mL of RPMI containing 10% FBS . Cytologic specimens were prepared from single-cell suspensions of harvested bone marrow and spleen via concentration of the suspensions using a cytocentrifuge ( Shandon Cytospin 4 , Thermo Fisher Scientific , Waltham , MA ) . Slides were fixed and stained with modified Wright-Giemsa ( Accustain , Sigma-Aldrich , St . Louis , MO ) . All slides were reviewed in a blinded fashion by a board-certified veterinary clinical pathologist . 1000-cell differential counts were performed with all myeloid and erythroid cells categorized as either proliferative or maturing ( post-mitotic ) . For example , proliferative neutrophilic cells include myeloblasts , promyelocytes and myelocytes; maturing neutrophilic cells include metamyelocytes , band and segmented neutrophils . Neutrophils , eosinophils , macrophages , lymphocytes and plasma cells were further separated into individual categories . Pegylated G-CSF ( GenScript , Z00393-50 ) was resuspended in PBS at a final concentration of 1 mg/mL and injected intraperitoneally at a dosage of 1 ug/mouse . Control mice were injected with 100 ul of PBS intraperitoneally . Mice were injected for three days consecutively and sacrificed on the fourth . Bone marrow-derived macrophages were prepared as previously described [45] . Five days after thawing , macrophages were plated at 2 . 5×105 cells per well in a 24 well dish ( Corning ) . The cells were then infected at a multiplicity of infection of 5 . Three hours after infection , supernatant was removed and 500 µL of the splenocyte single cell suspension added . After three hours of incubation , the cell suspension was collected . Cells were fixed , permeabilized with saponin and stained for intracellular IFNγ and transcription factors . IL-2 antibody complex was prepared as described previously [29] . IL-2 mouse antibody ( clone JES6-1 , eBiosciences , 16-7022-81 ) was incubated with recombinant mouse IL-2 ( eBiosciences , 14-8021-64 ) for 15 minutes before intraperitoneal injection into mice . Mice were injected for 2 days consecutively and sacrificed on the third . Control mice were injected with an equivalent volume of PBS . Bubble plots in Figures 2 A , B and Supplementary Figure 1A were constructed using JMP software ( SAS software , Cary , NC ) . Supplementary Figure 4C was visualized using MATLAB ( MathWorks , Natick , MA ) . All other Figures were made using Prism ( GraphPad , La Jolla , CA ) . All statistics were calculated using Prism and a two-tailed Mann-Whitney non-parametric test of significance was used unless otherwise mentioned . Spearman's correlations values were deemed significant using a two-tailed calculation based on the number of samples . | Bacteria belonging to the genus Salmonella are capable of causing long-term chronic systemic infections in specific hosts where they are shed in the feces . These persistently infected individuals include typhoid carriers and they serve as a reservoir for disease transmission . Despite the importance of Salmonella as a human pathogen , relatively little is known about the host immune response to persistent bacterial infections in the context of transmission . We had shown previously in a mouse model of Salmonella infection that mice shedding high levels of Salmonella ( >108 bacteria per gram of feces ) , known as super-shedders , transmit disease to naïve mice . We show here that these super-shedder mice have a unique immune state compared to mice that have lower levels of Salmonella in their gut . The super-shedder immune state is characterized by an active inflammatory immune response with elevated serum IL-6 and high levels of neutrophils in both the gastrointestinal tract and the systemic sites but a dampened adaptive CD4 T helper type1 ( TH1 ) cell response specific to the spleen . Importantly , we show that the blunted adaptive response , as characterized by reduced TH1 cell frequencies and ability to respond to IL-2 and IL-6 , is intimately linked to the levels of neutrophils present in the spleen . We go on to show the functional consequences of dampened cytokine responsiveness , as TH1 cells from moderate-shedders are unable to undergo IL-2-mediated expansion when neutrophilia is induced . Additionally , we show that neutrophil control of IL-2 mediated expansion of TH1 cells is independent of infection . In summary , we describe an immune phenotype associated with transmission of a pathogen and a single immune cell type , neutrophils , which control specific aspects of the super-shedder immune state . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"immune",
"cells",
"cytokines",
"immunity",
"to",
"infections",
"immunology",
"immune",
"suppression",
"microbiology",
"lymphoid",
"organs",
"host-pathogen",
"interaction",
"emerging",
"infectious",
"diseases",
"immune",
"defense",
"immunoregulation",
"immunomodulation",
"bacterial",
"pathogens",
"inflammation",
"medical",
"microbiology",
"t",
"cells",
"microbial",
"pathogens",
"biology",
"pathogenesis",
"salmonella",
"immune",
"response",
"immune",
"system",
"immunity",
"bone",
"marrow",
"innate",
"immunity",
"immune",
"tolerance"
]
| 2013 | The Systemic Immune State of Super-shedder Mice Is Characterized by a Unique Neutrophil-dependent Blunting of TH1 Responses |
Axonal degeneration is a key event in the pathogenesis of neurodegenerative conditions . We show here that mec-4d triggered axonal degeneration of Caenorhabditis elegans neurons and mammalian axons share mechanistical similarities , as both are rescued by inhibition of calcium increase , mitochondrial dysfunction , and NMNAT overexpression . We then explore whether reactive oxygen species ( ROS ) participate in axonal degeneration and neuronal demise . C . elegans dauers have enhanced anti-ROS systems , and dauer mec-4d worms are completely protected from axonal degeneration and neuronal loss . Mechanistically , downregulation of the Insulin/IGF-1-like signaling ( IIS ) pathway protects neurons from degenerating in a DAF-16/FOXO–dependent manner and is related to superoxide dismutase and catalase-increased expression . Caloric restriction and systemic antioxidant treatment , which decrease oxidative damage , protect C . elegans axons from mec-4d-mediated degeneration and delay Wallerian degeneration in mice . In summary , we show that the IIS pathway is essential in maintaining neuronal homeostasis under pro-degenerative stimuli and identify ROS as a key intermediate of neuronal degeneration in vivo . Since axonal degeneration represents an early pathological event in neurodegeneration , our work identifies potential targets for therapeutic intervention in several conditions characterized by axonal loss and functional impairment .
Neuronal loss constitutes an irreversible end point of several neurodegenerative conditions triggered by diverse stimuli . Nevertheless , early neuronal dysfunction is associated with degeneration of neuronal processes , including axons and dendrites , causing a progressive loss of neuronal function [1] , [2] . Axons are particularly susceptible to genetic and toxic insults and represent a key therapeutic target for neuroprotection [3] , [4] . Degeneration of axons is non-apoptotic [5] , [6] , and depends on the activation of the mitochondrial permeability transition pore ( mPTP and reference [7] ) , intra-axonal calcium rise , and calpain activation [8] . Much of our current knowledge regarding the mechanisms of axonal degeneration originates from studies in the Wlds mice , which show delayed axonal degeneration caused by mechanical injuries , hypoxia and other toxic stimuli [5] . The Wlds-dependent protection of axons is given by overexpression of a chimera between nuclear NMNAT1 and a non-catalytic sequence from a ubiquitin ligase [9] . Importantly , slowing down axonal degeneration by Wlds expression in mouse models of neurodegenerative conditions delays the disease progression and severity [10]–[12] , pointing out to a crucial role of axonal degeneration in diseases of varied nature . Despite the advancement in the identification of molecular players and pathways implicated in axonal degeneration , how they integrate into a degenerative process triggered by a wide variety of conditions remains inconclusive and needs further examination in vivo . Therefore it is important to characterize axonal degeneration in tractable model systems such as C . elegans , which are suitable for genetic and pharmacological interventions with functional and morphological readouts of neuronal integrity . In C . elegans , constitutively active degenerins cause degeneration of neuronal somas and axons by ultrastructural and macroscopical features reminiscent of excitotoxic neuronal death in mammals [13] . Non-apoptotic death of degenerin-expressing C . elegans neurons involves intracellular calcium rise , activation of intracellular proteases and autophagy [14]–[16] , and is currently considered to occur by necrosis , also responsible for cell death in stroke , brain trauma and other human neuropathological conditions [17] . Key players in the execution of necrosis are mitochondrial dysfunction and oxidative stress [17] , but the role that intracellular ROS generation plays in axonal degeneration has not been addressed in vivo . The cell is equipped with a homeostatic network to cope with oxidative stress in physiological and pathological situations . One of the master regulators of this homeostatic response to oxidative stress is the Insulin/IGF-1-like signaling ( IIS ) pathway , which has been associated with the protection of diverse tissues from chemical and genetic insults [18] . Here , we show for the first time that axonal and neuronal degeneration triggered by the hyperactivated degenerin channel MEC-4d , can be prevented by diapause entry and by increasing the anti-oxidative capacity of the neuron . Mechanistically , this increase is mediated by the IIS pathway in a DAF-16/FOXO dependent manner and relies on the activity of the antioxidant enzymes superoxide dismutases ( SODs ) and catalases . In addition to including the axonal compartment in our analyses , we performed a functional assessment of neuronal integrity to evaluate if prevention of neurodegeneration rendered functional neurons [19] , generating a powerful model for mechanistic enquiries . Neuronal function is a critical aspect to be evaluated in therapeutic applications and a readout missing from commonly used models . Importantly , we demonstrate that mec-4d-dependent degeneration can be prevented morphologically and functionally by mPTP inhibition and NMNAT overexpression , which delay axonal degeneration in mammals [7] , [9] . Finally , we extend these findings to mice and demonstrate that axonal degeneration can be strongly delayed in vivo by systemic manipulation of the cellular antioxidant capacity . In summary , our work shows that prevention of neuronal degeneration and functional loss can be achieved by a systemic shift of metabolic state leading to an increase in cellular anti-oxidative defense systems .
The use of C . elegans to systematically study axonal degeneration has not been explored so far . C . elegans touch receptor neurons ( TRNs ) expressing the constitutively open MEC-4d degenerin channel , constitute an ideal model for axonal degeneration as the pro-degenerating stimuli is endogenously triggered ( mec-4d expression ) and its degeneration and protection can be assessed morphologically and by loss of touch sensitivity . From the six TRNs ( ALML , ALMR , PLML , PLMR , AVM and PVM; see Figure 1A ) , we investigated the AVM neuron because it arises post-embryonically , thus the degenerative process can be observed from the beginning . Additionally , the AVM by itself gives a functional response to anterior touch [19] , as when AVMs arise , the other neurons of the anterior touch circuit , the ALMs , have already degenerated ( data not shown ) . To assess the AVM neuron morphologically , we used a strain that expresses gfp under a TRN specific promoter ( Pmec-17mec-17::gfp ) in wild type or mec-4d ( e1611 ) worms . AVM neurons in mec-4d mutants , appeared and extended axons that reached similar sizes to wild type AVM neurons . Thereafter , AVM somas and axons followed a stereotyped form of degeneration ( Figure 1B ) . First , full length axons become beaded and later become truncated from the most distal end ( Figure 1B ) . Somas become vacuolated and later disappear ( Figure 1B; see Methods for a description of categories ) . This order of events for AVM soma and axon degeneration in mec-4d mutants is consistent with previous qualitative electron microscope analysis [13] . We performed a thorough temporal analysis of the birth and degeneration of AVM somas and axons by using gfp expression as a reporter . In wild type worms , most AVM somas and axons appear 12 hours post hatching; 24 hours after hatching , all axons have reached their full length ( Figure 1C and 1D ) . In mec-4d mutants , 89% of somas have appeared at 12 hours , and around 48 . 9% axons are intact ( Figure 1E and 1F ) . Thereafter , AVM axons degenerate as described above ( Figure 1B ) . By 24 hours after hatching , only 7% of axons remain intact ( Figure 1F ) , and 58% of somas appear healthy ( Figure 1E ) . 72 hours post hatching , both somas and axons have nearly completely degenerated ( Figure 1E and 1F ) . We performed touch tests to mec-4d mutants at different times after hatching . The percentage of anterior-touch responsive worms at each time point correlates well with AVMs in the axon wild type ( AxW ) category ( Figure 1G ) . At 12 hours after hatching , mec-4d animals do not respond because the AVB interneuron , which connects the AVM to motoneurons has not yet arisen [20] . To evaluate our genetic and pharmacological interventions on mec-4d mutants we scored at 72 hours after hatching , when AVM neurons have almost completely degenerated ( Figure 1E and 1F ) , hence an adequate time point to observe unequivocal neuronal protection . Additional relevance comes from the fact that at 72 hours after hatching , worms have reached adulthood and have completed their life cycle . In summary , we have introduced two novel elements in the model of degeneration triggered by mec-4d: we included the axonal compartment and performed functional analyses of neuronal integrity , which allows a behavioral evaluation of treatments that delay or prevent neuronal degeneration . Upon harsh conditions , C . elegans larvae enter a diapause state called dauer , from where they resume development once the environment becomes favorable [21] . Among other stress resistant changes , dauers have enhanced anti-oxidative defense systems [22]–[25] due to downregulation of the IIS pathway , which leads to the activation of transcription factors DAF-16/FOXO [26] and SKN-1/Nrf-2 [27] . DAF-16 targets genes like sod-3 and ctl-1 [28] , while SKN-1/Nrf-2 mobilizes the phase 2 detoxification response [29] , [30] . To examine whether neuronal degeneration is hampered by diapause entry , mec-4d worms were induced to form dauer by starvation and kept in diapause for a week or a month . Strikingly , there was an almost complete prevention of somatic and axonal degeneration by morphological criteria in mec-4d dauer , independently of the diapause length ( Figure 2A–2C ) . Furthermore , AVM neurons of one-month old mec-4d dauers responded to the anterior touch ( Figure 2D ) . In one-month old wild type dauers , 21 . 9% were anterior and posteriorly touch insensitive ( Figure 2D ) . Considering this percent of insensitive wild type dauers , there is a precise correlation between morphologically wild type axons of mec-4d AVMs and their ability to respond to touch . Noticeably , this protection is stronger than most other genetic or pharmacological treatment reported previously for the mec-4d stressor , and additionally includes a functional readout . Dauer mec-4 ( u253 ) loss of function mutants were largely insensitive to touch ( 2 out of 90 responded to two of ten touches ) , demonstrating that touch response during dauer is MEC-4 dependent . We then examined if the protection conferred by dauer was reversed upon exit from this state . For this purpose , dauer worms were placed on food and AVM integrity was scored every 24 hours for 120 hours . Degeneration of somas and axons during dauer exit resumes in a dauer length-independent fashion ( Figure 2E–2H ) . To rule out that AVM protection during dauer was due to a reduction of mec-4d expression in diapause , we followed MEC-4::GFP expression by means of the uIs58 transgene ( Pmec-4mec-4::gfp ) in dauer and non-dauer wild type animals . The MEC-4 channel is patterned in regularly spaced puncta along the TRN axons ( Figure 2I and reference [31] ) , allowing the quantification of both intensity by gfp expression and the interpuncta distance , which reflects the distribution of the mechanosensory channel . The same criteria was used to assess whether in mec-4d dauers the expression of MEC-4 was impaired or diminished by the mec-4d ( e1611 ) dominant mutation . Dauer entry did not affect MEC-4 expression nor did the mec-4d ( e1611 ) mutation , as GFP intensity was similar in all three conditions ( Figure 2I and 2J ) . Importantly , the interpuncta distance was also unchanged in both wild type and mec-4d animals compared to non-dauer wild type worms ( Figure 2I and 2K ) , confirming that the distribution of the mechanosensory channel is not affected neither by the mec-4d mutation nor by dauer entry . In addition , both wild type and mec-4d dauers respond to anterior gentle touch ( Figure 2D ) , strongly suggesting that MEC-4 is expressed and functional . To study whether this phenomenon was extensive to other cells and pro-degenerative stimuli , we tested whether dauer entry prevented degeneration of PVC neurons caused by the deg-1 ( u38 ) mutation . Degeneration of the PVC renders worms insensitive to posterior touch [32] . To observe degeneration of somas and axons caused by the deg-1 ( u38 ) dominant mutation , we created a strain that expresses gfp under the nmr-1 promoter ( Pnmr-1nmr-1::gfp , reference [33] ) in a deg-1 ( u38 ) mutant background . At the time corresponding to dauer entry , 30% of PVC somas and 23% of axons were wild type ( Figure 3A and 3B ) . We induced deg-1 ( u38 ) animals to form dauer for one month and analyzed the morphology of their PVC neurons and performed a posterior touch test . Impressively , 91% of somas and 68% of axons remained wild type in dauer , and the protection of touch sensitivity was also maintained ( Figure 3A–3D ) , and decayed as worms exited this stage by food reposition ( Figure 3D ) . These results demonstrates that protection conferred by dauer to degenerating neurons is independent of cell type and stimuli , and persists as long as animals remain in diapause . The C . elegans insulin receptor DAF-2 negatively controls DAF-16/FOXO and SKN-1/Nrf-2 activity , and daf-2 deficient worms share with dauers similar transcription signatures leading to an increase in expression of antioxidant enzymes ( Figure 4A and references [22] , [27] ) . We wanted to dissect out the DAF-2 dependent components of dauer that protects mec-4d neurons from degeneration . To this end , we generated a strain with a temperature sensitive mutation in daf-2 ( e1370ts ) , together with mec-4d ( e1611 ) and a fluorescent marker in TRNs ( Pmec-17mec-17::gfp ) . At 25°C , DAF-2 is not functional in daf-2 ( e1370ts ) mutants and nuclear translocation of DAF-16 and SKN-1 activates their target genes [30] , [34] . At 25°C , mec-4d AVMs appear and degenerate and by 72 hours almost no axons or somas remain ( Figure 4B and 4C ) . In contrast , raising daf-2 ( e1370ts ) ;mec-4d ( e1611 ) at 25°C from the time of hatching confers an almost complete protection to both axons and somas similar to the provided in dauer ( Figure 4B and 4C ) . Insulin pathway in C . elegans mediates life span extension , dauer formation and stress resistance by systemic and cell autonomous mechanisms [35]–[38] . We wanted to know if cell autonomous downregulation of DAF-2 in TRN prevents mec-4d-dependent TRN degeneration . To this end , we created a systemic RNAi defective strain with TRN-enhanced RNAi also carrying the mec-4d ( e1611 ) mutation and a fluorescent marker in touch neurons ( Pmec-17mec-17::gfp , WCH6; see Methods ) . We fed daf-2 and control unc-22 ( RNAi ) to newly hatched nematodes and scored AVMs 72 hours later . As expected , unc-22 ( RNAi ) treated animals shown no twitching phenotype , since systemic RNAi is impaired in this strain . Impressively , TRN-autonomous daf-2 ( RNAi ) prevented AVM degeneration , rendering about 69% somas and 33% axons in a wild type category , compared to 8% and 3% in control RNAi ( Figure 4D and 4E ) . Taken together our data suggest that repression of the IIS pathway leads to prevention of neuronal demise triggered by mec-4d and that downregulation of the IIS only in TRN is sufficient to provide this protection . To define the contribution of DAF-16 and SKN-1 in the prevention of mec-4d-dependent neuronal degeneration by DAF-2 downregulation , we performed double RNAi of daf-2 together with daf-16 or skn-1 TRN-autonomously . Combinatorial RNAi has been shown extensively to be an efficient tool to target two genes concomitantly [39] . daf-2/skn-1 ( RNAi ) resembled the protection that daf-2 ( RNAi ) alone conferred to axons and somas while daf-2/daf-16 ( RNAi ) reverted this protection ( Figure 5A and 5B ) . This result suggests that elevation in DAF-16 , but not SKN-1 targets in TRN is responsible for neuronal protection after DAF-2 downregulation . No effects in degeneration of somas or axons triggered by mec-4d were observed when skn-1 or daf-16 ( RNAi ) were performed alone ( Figure 5A and 5B ) . To rule out that the effect observed in daf-2/skn-1 ( RNAi ) was due to inefficient skn-1 ( RNAi ) , we fed bacteria expressing skn-1 dsRNA to mec-4d worms to reveal the embryonic lethal phenotype that has been reported for this gene ( www . wormbase . org ) . More than 90% of mec-4d embryos on skn-1 ( RNAi ) did not hatch , confirming the expected phenotype . Additionally , it is unlikely that the lack of effect of daf-2/skn-1 ( RNAi ) on neuronal protection is due to an artifact of double RNAi inefficiency since daf-16 combined with daf-2 ( RNAi ) successfully reverted the daf-2 ( RNAi ) protection . Key enzymes that control oxidative stress downstream of DAF-2 are superoxide dismutases and catalases [40] , which are compartmentalized within the cell to distinct organelles . To directly test whether antioxidant endogenous defenses are needed for the protection conferred by downregulation of the IIS pathway , we performed combinatorial TRN-autonomous RNAi of daf-2 with sod-2 , sod-4 , ctl-1 or ctl-2 . Protection of AVM somas and axons by daf-2 ( RNAi ) was reverted by sod-2 , sod-4 , ctl-1 and ctl-2 ( RNAi ) in somas and axons ( Figure 5C and 5D ) . These results strongly argue that protection of neuronal degeneration by downregulation of DAF-2 relies on the activity of antioxidant enzymes , supporting a role for ROS in neuronal degeneration triggered by MEC-4d . Heightened antioxidant capacity of worms by activation of DAF-16/FOXO after daf-2 downregulation , dramatically prevented somatic and axonal degeneration of AVM touch neurons in mec-4d mutants . Importantly , TRNs in mec-4d animals were positive for a fluorogenic ROS-dependent dye ( Figure 5E ) , demonstrating a heightened oxidative stress state in TRNs during the degenerative process . We then tested whether the ROS scavengers trolox , a vitamin E derivative , and ascorbic acid ( AA ) acutely protect the AVM from mec-4d dependent degeneration . Trolox and AA treatment from the time of hatching greatly protected AVM somas and axons from degeneration at 72 hours ( Figure 5F and 5G ) . Protection of neuronal somas was almost complete , from 8% of worms with wild type somas in controls ( no treatment ) to 88% and 55% in trolox and AA-treated worms , respectively ( Figure 5F ) . Axons were also protected , with 6% of worms with intact axons in controls , rising to 34% and 22% under trolox and AA treatment , respectively ( Figure 5G ) . From these results , we conclude that elevated ROS is a key intermediate pathway for neuronal degeneration triggered by MEC-4d stimuli , and that degeneration of both somas and axons can be prevented by systemic antioxidant treatment . Animals fed with a caloric restricted diet have reduced oxidative stress levels , and this is proposed as one of the underlying causes for life span extension and delay in age-associated conditions [41] . Hence , we wondered if dietary restriction also impacted degeneration of the AVM in mec-4d worms . We designed an intermittent fasting protocol during dauer exit as mec-4d dauers have protected AVMs and are developmentally synchronized . To this end , mec-4d dauers ( Figure 2A–2D ) were exited from diapause by a 3 hour food period followed by 9 hours without food ( see Methods ) . This schedule was repeated until 72 hours and AVM integrity was compared to dauers exiting in an ad-libitum food regime . Importantly , nematodes under intermittent fasting developed normally and were fertile when they reached adulthood ( data not shown ) . Dietary restriction conferred a strong protection of 69% to somas and 35% to axons , compared to 7% and 13% in the ad libitum regime , respectively ( Figure 6A–6C ) . This demonstrates , that dietary restriction , a treatment that lowers oxidative stress , protects the TRN from degenerating . Ideally , the knowledge we gather from a simple model organism as C . elegans should be translated into a mammalian animal model . To this end , we tested if mec-4d triggered neuronal degeneration proceeds by similar mechanisms to degeneration of mammalian neurons . As reported only for somas [14] , supplementing EGTA to mec-4d mutants from the time of hatching also significantly reduces degeneration of AVM axons at 72 hours ( Figure 7A–7C ) , as in mammals [8] . Second , it has been shown that axonal degeneration in mammalian neurons requires mPTP activation [7] , which is pharmacologically inhibited by cyclosporin A ( CsA ) . CsA treatment strongly reduces both soma and axon demise ( Figure 7A and 7B ) . Finally , delay of axonal degeneration by NMNAT overexpression after diverse stimuli constitutes a molecular signature of non-apoptotic neuronal degeneration in several animal models , including mice , rats , flies and zebrafish [9] , [42]–[46] . We asked if overexpression of C . elegans NMNAT protected against mec-4d-triggered neurodegeneration . To this end , we expressed F26H9 . 4 , one of the two C . elegans NMNAT genes , fused to gfp under the TRN specific mec-18 promoter . As expression of the F26H9 . 4 protein is restricted to the cytosol ( Figure 7D ) , like NMNAT-2 in mammalian cells , and is similar in sequence ( data not shown ) , we called it NMAT-2 . Overexpression of NMAT-2::GFP in TRN strongly protected AVM somas and axons against mec-4d mediated degeneration at 72 hours post-hatching ( Figure 7E and 7F ) . Importantly , AVM function was also rescued by NMAT-2 overexpression when the touch response was evaluated ( Figure 7G ) . These results demonstrate that C . elegans neurons degenerate by similar mechanisms to mammalian neurons giving a strong base for extending our findings to a mouse model of neuronal degeneration . We investigated if systemic antioxidant treatment ( AOX ) has an effect in Wallerian degeneration in mice , a well established model of axonal degeneration triggered by axotomy , and delayed by NMNAT overexpression [5] . Wild-type mice ( C57BL/6J strain ) were supplemented with ascorbic acid in the drinking water for 10 days before a nerve injury was performed in the sciatic nerve . After 3 , 5 and 7 days , the degree of axonal degeneration distal to the injury site was quantified . In non-treated mice , axons disconnected from their somas degenerate to a large extent in 3 days as seen by the almost complete disappearance of neurofilament markers ( Figure 8A and 8D ) . AOX treatment strongly protected axons from degeneration over 3 , 5 and even 7 days ( Figure 8B and 8D ) . Ultrastructurally , control sciatic nerves display mixed populations of conserved myelinated and non-myelinated axons . Three days after damage , distal nerves exhibit degenerating axons and collapsed myelin sheaths ( Figure 8E ) . Consistent with the immunofluorescence results , AOX treatment robustly protects axons from axotomy-dependent degeneration when examined at the EM level ( Figure 8F ) . We then assessed if dietary restriction replicate in mice the protective effects obtained in mec-4d worms . To this end , mice were subjected to intermittent fasting by means of ad libitum feeding every other day for 10 days . Subsequently , sciatic nerves were injured and analyzed 3 , 5 and 7 days later as described above . Impressively , dietary restriction in mice delays axonal degeneration at 3 , 5 and 7 days after axotomy as seen both by immunofluorescence ( Figure 8C and 8D ) and ultrastructural analyses ( Figure 8G ) . Ultrastructurally , the reported increase in axonal mitochondria diameter at 3 days post-axotomy ( Figure 8H and reference [7] ) was prevented by both AOX and dietary restriction ( Figure 8H ) . In contralateral , non-damaged nerves , neither AOX nor DR treatments affect axonal parameters as density or ultrastructural characteristics ( Figure 8B–8D , 8F and 8G ) . Due to the robust protection over injury-induced axonal degeneration conferred by AOX , we wanted to establish the dynamics of oxidative stress after nerve damage . To this end , we measured lipid peroxidation , a widely used reporter of oxidative modifications [47] . Injured sciatic nerves kept ex vivo showed a gradual increase in lipid peroxidation over time ( Figure 8I ) , and at 3 days post injury , lipoperoxide levels increased about 20-fold compared to non-injured sciatic nerves ( Figure 8I ) . Importantly , this increase in lipid peroxidation was significantly prevented by AOX treatment 3 days after nerve damage ( Figure 8I ) . In situ observation of oxidative stress using a ROS-sensitive dye shows a clear appearance of positive signal in sciatic nerve axons 24 hours after nerve damage ( Figure 8J ) , consistent with the increase in lipid peroxidation ( Figure 8I ) . Taken together , our data demonstrate that in mice , as in C . elegans , axonal degeneration depends on the anti-oxidative cellular capacity , which can be enhanced directly by antioxidant treatment or indirectly by dietary restriction .
Lack of nutrients signals the downregulation of the IIS pathway in C . elegans and the formation of the dauer larvae , a diapause state that can last for up to four months [18] , [49] , [50] . Entry into dauer prevented neuronal degeneration as long as worms were kept in diapause . Exit from dauer resumes the degeneration program of TRNs . Protection from neuronal degeneration during dauer was not exclusive of the mec-4d triggered death: nematodes with the deg-1 mutation , which kills the PVC cells , also were protected from neuronal demise during diapause . This speaks of a broad mechanism of neuronal protection in the dauer state , which represents a novel characteristic of this diapause condition . Downregulation of the IIS pathway by daf-2 mutation protected TRN from degeneration to a great extent , comparable to the dauer-dependent protection , suggesting that the mechanism responsible for neuronal protection in dauer is related to this pathway . By using neuronal-specific RNAi we demonstrated that protective mechanisms of the IIS pathway are activated in a neuronal autonomous fashion ( Figure 4D–4E ) . Downstream of DAF-2 , the transcription factors DAF-16/FOXO and SKN-1/Nrf2 activate different mechanisms of oxidative stress surveillance . DAF-16 targets SODs and catalases while SKN-1 activates the detoxification phase II genes such as gcs-1 [30] . Our results demonstrate that AVM protection from mec-4d induced degeneration is at least in part due to the transcriptional activation of DAF-16 downstream of DAF-2 in a neuronal autonomous fashion ( Figure 5A–5B ) . Consistently , the protection granted by DAF-2 downregulation requires the expression of the antioxidant enzymes SODs and catalases ( Figure 5C–5D ) . It is likely that reduction of cellular oxidative stress by activation of DAF-16/FOXO represents a neuroprotective mechanism in part by its ability to inhibit axonal degeneration , preventing the consequent demise of the neuronal soma . At the same time , the expression of organelle specific anti-ROS systems ( i . e . SODs and catalases ) as a result of DAF-16 activation may simultaneously prevent soma and axonal demise . There is a good correlation between factors that promote longevity and those that we have found here to prevent neuronal degeneration [51] . Likely these are processes that rely on mechanisms that preserve the cellular integrity through lowering , for example , oxidative stress . Caloric restriction ( CR ) decreases oxidative stress in several tissues [41] and in mammals , reduces insulin levels with a concomitant nuclear translocation of FOXO and activation of its target genes [52] . Several studies have shown that caloric restriction increases life expectancy and is beneficial for healthy aging of the nervous system [52] , [53] . CR also reduces brain damage in models of focal cerebral ischemia [54] and Parkinson's disease [55] . In our work , both adult worms and mice subject to CR showed delayed axonal degeneration . CR and reduced IIS are believed to exert their effects by a decrease in cellular oxidative damage [41] . CR causes a decrease in energy expenditure , ROS production and oxidative damage , while downregulation of the IIS increases cellular antioxidant defenses , which elevates resistance to oxidative stress [56] . CR has the advantage of being an intervention that is readily applicable to humans [57]–[59] . Importantly , these acute treatments can be implemented once organisms reach their fertile age with very short or even no pretreatment in the case of C . elegans and mice , and rendering impressive results both by genetic pro-degenerative triggers ( worms , degenerin activation ) as well as mechanical damage ( mice , nerve damage ) . Although the mechanisms by which CR increases resistance of cells to age-related pathology are not firmly established , two not mutually exclusive possibilities have been suggested: a reduction in mitochondrial oxyradical production by lowering metabolic flux [60] and activation of cellular antioxidant mechanisms [61]–[64] by an hormetic cellular response after a mild-stress associated to intermittent induction of mitochondrial metabolism and sub-lethal ROS production [65] . Degeneration of TRN somas by mec-4d expression has been studied as a model of necrosis [66] . Calcium increase , activation of intracellular proteases ( including calpains and cathepsins ) and autophagy , are intermediary steps leading to vacuolation of TRN somas , which are later engulfed by neighboring cells in a ced-2 , 5 and 10-regulated process [67] . Our results demonstrate a critical role for ROS in this process , which can be controlled by manipulating the IIS pathway . This , together with its nmat-2 dependence , better defines this necroptotic death mechanism of neuronal somas . We have further demonstrated that neuronal degeneration in worms and mice also depends on ROS production , which can be counteracted by genetic or pharmacological means , as well as reduction in the metabolic burden to the organism by CR . Based on our results and published data , we propose a model in which opening of the MEC-4d channel results in an increase in intracellular calcium , which together with the endoplasmic reticulum calcium contribution [14] , leads to mitochondrial dysfunction , ATP depletion and ROS generation , which enters into positive feedback loops , finally leading to activation of proteases in somas and axons . Importantly , the presence of axonal beading in mec-4d worms suggests that axonal transport defects might represent an early event triggering degeneration , in agreement with recent data [68] . In both dauer and daf-2 mutants , anti-ROS systems are elevated , decreasing mitochondrial dysfunction and blocking neuronal degeneration ( Figure 9 ) . Accumulative stress with age just as a consequence of living may resemble the acute effect caused by damage to neurons . Factors such as DAF-16/FOXO integrate and orchestrate a response to oxidative stress provoked by either processes being physiological or pathological . Since neurons are mostly irreplaceable , the nervous system is particularly susceptible to aging and injury and may profit from an intrinsic ability to detoxify itself that can be exploited for therapeutic purposes .
C . elegans wild type ( N2 ) , mutants ( daf-2 ( e1370 ) III , deg-1 ( e38 ) X , mec-4d ( e1611 ) X ) and transgenic strains [TU2773 {uIs31 ( Pmec-17mec-17::gfp ) ; mec-4d ( e1611 ) X} , TU3755 {uIs58 ( Pmec-4mec-4::gfp ) } , VM484 {akIs3 ( Pnmr-1nmr-1::gfp ) }] were grown as previously described [69] . For time course experiments we synchronized worms by collecting the newly hatched L1 two hours after washing off a plate with mixed stage worms . Groups of 30 to 50 worms were gently collected in M9 using a mouth pipette and placed in new plates with food to be examined at the appropriate times . All experiments , unless noted were done at 20°C . A 1 kb fragment 5′ from the start of translation of the F26H9 . 4 ( nmat-2 ) gene was amplified from genomic DNA using the following primers AAAAGGATCCATGAAACGAGTCGCTCTTCTTGC and TTTTGGATCCATTTTCTGATACAGATTATTCTCCC , that introduced 5′ and 3′ BamHI sites . The resulting PCR fragment was cloned between the mec-18 promoter and yfp coding sequence in TU#739 [70] to create WCH#19 . We generated transgenic nematodes by microinjection [71] of 10 ng/µl of WCH#19 , 30 ng/µl of pCW2 . 1 ( a ceh-22::gfp plasmid [72] ) and 60 ng/µl of pBSK ( Stratagene ) as filling DNA . WCH1 contains wchEx1 , an extrachromosomal array containing Pmec-18nmat-2::yfp; Pceh-22gfp expressed in wild-type worms . wchEx1 was crossed into mec-4d ( e1611 ) X to create WCH2 . WCH4 contains uIs31 ( Pmec-17mec-17::gfp ) in a daf-2 ( e1370ts ) III; mec-4d ( e1611 ) double mutant . WCH6 contains uIs71 ( Pmec-18sid-1;Pmyo-2mcherry ) and uIs31 ( Pmec-17mec-17::gfp ) arrays in a sid-1 ( pk3321 ) V , mec-4d ( e1611 ) X double mutant . uIs58 ( Pmec-4mec-4::gfp ) was crossed into mec-4d ( e1611 ) X animals to create WCH32 , and akIs31 ( Pnmr-1nmr-1::gfp ) was crossed into deg-1 ( u38 ) X mutants to create WCH33 . To assess touch neuron morphology , worms were immobilized by placing them on a drop of 20 mM sodium azide directly on a 2% agarose pad . All worms were observed under a 100× objective , in an upright fluorescent microscope ( Olympus ) in a 20 minutes range to avoid damage for long exposure to sodium azide . We used the uIs31 strain ( Pmec-17mec-17::gfp ) as a control for the time of appearance of the postembryonic cells ( AVM and PVM ) . We established the following morphological categories for somas and axons: Somas , SoW: soma wild-type , SoV: soma vacuolated , swollen several times the wild type size; SoØ: soma absent . Axons , AxW: axon wild-type; AxB: beaded axon of wild type length; AxT: axon truncated , of smaller length than wild type and sometimes detached from soma; AxØ: axon absent . When the axon was truncated and beaded it was scored as AxT . Since we used gfp expression as the only morphological measure of neuronal integrity , there is a possibility that neuronal degeneration assessed by this mean follows a different degeneration dynamic than when studied by other techniques . The fact that TRN axons degenerate in mec-4d mutants , has been solidly established by electron microscopy analysis [13] . Worms were touched gently with an eyebrow hair [19] with alternative anterior and posterior touches ( five times each ) to determine an average response . These experiments were done in triplicates for each time point . Because we scored the functionality of the AVM touch neuron , which participates in the anterior touch circuit , only anterior responses ( 5 in total ) are reported here . When AVM is intact ( as in mec-4d dauer ) the maximum anterior response is of three touches . Only worms that moved when prodded with a platinum wire pick or responded to the nose touch , were assayed . To test the function of the PVC neurons , a posterior ( tail ) touch was performed in deg-1 ( u38 ) animals at various times . deg-1 ( u38 ) animals have normal locomotion and are sensitive to anterior touch , which permits unequivocal interpretation of a negative posterior touch response . Mixed-stage worms were grown until all food exhausted to induce the formation of large amounts of dauer [73] . Plates were observed periodically every 24 hours for the appearance of dauer larvae . The first day dauers appeared on the plate was considered the day zero of dauer and from then a week or a month was counted . At the appropriate time , all worms were collected in M9 , centrifuged and incubated in 1% SDS for 30 minutes to eliminate all but dauer larvae [21] . Quantification of AVM integrity was carried out in the same way as non-dauer worms ( see above ) . Touch tests: wild type ( N2 ) , uIs31;mec-4d ( e1611 ) or deg-1 ( u38 ) SDS treated dauers were placed on a plate without food and allowed to disperse before touching them . All strains were touched 10 times in a head to tail fashion . In deg-1 ( e38 ) mutants , which are characterized by tail insensitivity due to death of the PVC neuron , we scored the posterior response [32] . mec-4d ( e1611 ) mutants were assessed for the rescue of the anterior response , as a measure of functionality of the AVM . Responsive mec-4d ( e1611 ) worms were separated then from non-responsive and both groups observed separately in the microscope for AVM integrity . Dauer recovery: One week and one month dauers were treated with SDS , plated on food and scored every 24 hours for 120 hours . Each time point was done in triplicates of 30 worms each . Experiments to restrict the expression of daf-2 in CB1370 [daf-2 ( e1370ts ) III] and WCH4 ( daf-2 ( e1370ts ) III; uIs31 ( Pmec-17gfp ) ; mec-4d ( e1611 ) X were done at 25°C . L3–L4 worms grown at 20°C were passed to 25°C and allowed to grow for 2 days and lay eggs . We synchronized worms by washing off the plates and leaving eggs that remained attached to the bacterial lawn or agar . Two hours later , 50 newly hatched larvae were picked with a mouth pipette to each new plate and kept at 25°C . Each time point was done in triplicates of 30 worms each . Bacteria expressing dsRNA from the Ahringer library [74] , [75] were grown on LB plates supplemented with ampicillin at 37°C overnight . Next morning a large amount of bacteria was inoculated in LB liquid supplemented with ampicillin and grown for 6–8 h . The resulting culture was seeded onto 1-day-old NGM-IPTG-carbenicillin plates and allowed to dry for 24 hours . For combinatorial RNAi , each bacterial culture was grown separately overnight in liquid LB . Before seeding the plates , the contents of each culture were spun down , concentrated in half the volume and mixed in equal parts to reconstitute the same concentration used in single RNAi experiments . We added 30–50 newly hatched L1 with a mouth pipette onto each plate and grew them at 20°C . We designed an intermittent fasting protocol with 12 hour-shifts that consisted of 3 hours with ad libitum food followed by 9 hours without food , for 72 hours at 20°C . We began with 1 month-old dauers , selected by SDS treatment as detailed above . After the 3 hours on food , worms were washed off using M9 medium and spun down for 3 minutes at 2000 rpm . Three washes were performed to eliminate bacteria . For the 9 hours without food , worms were placed on NGM plates carrying ampicillin to avoid any growth of E . coli OP50 . Drugs were administered in the food , by mixing the appropriate concentrations of the chemical with E . coli OP50 . New plates were made every 24 hours to keep the chemical fresh , and worms changed by hand picking on the next morning . The scoring was always done at 72 hours post-hatching . The following drugs and concentrations were used; EGTA ( Sigma , 50 mM ) , cyclosporin A ( CsA , LC Laboratories , 50 µM ) , trolox ( Sigma , 100 µM ) , ascorbic acid ( Sigma , 50 mM ) . Mice of the wild-type ( WT , C57BL/6J ) strain were obtained from the University animal house . Transgenic YFP ( B6 . CG-T6 , THY1-YFP ) mice strain were obtain from Jackson laboratories ( USA ) . Adult mice 11-week old , weighting 20–25 g were used . Experiments with mice followed protocols approved by the Institutional Animal Care and Use Committees and complied with NIH guidelines . Ascorbic acid was diluted in the mice drinking water at a concentration of 0 , 5 g/L . A pretreatment of 10 days was performed prior to the nerve crush . As mice consumes between 4–5 ml of water per day ( as measured daily in our experimental groups ) , the systemic dose of ascorbic acid corresponds to 2 , 5 mg/g per day . We used the body surface area ( BSA ) normalization method [76] to calculate the mice dose administration of ascorbic acid from the human dose , in which the Human Equivalent Dose ( HED , 10 mg/kg ) is multiplied by the ratio between the Km factor of the human ( 37 ) and the Km factor of the mouse ( 3 ) ( Office of New Drugs , 2005 ) , resulting in an animal dose of 123 , 33 mg/kg . Considering a 20 mg mouse the daily dosage of ascorbic acid given was 2 , 47 mg/animal , dissolved in its drinking water according to an intake per day of 5 ml/animal ( data not shown ) . No difference in the intake of water with or without ascorbic acid was observed . For dietary restriction , mice were fasted ad libitum every other day with regular mice food for 10 days before nerve crush . Mice weights were daily measured in all experimental groups with no significant differences between them , except in the dietary restricted group , which shows a regular decrease of 3 g after the non-feeding day , but with a net weight increase comparable to the other experimental groups . For nerve crush , mice were anaesthetized with Avertin ( 0 . 37 mg/g ) and the sciatic nerve was exposed at the mid-thigh and crushed three times for 5 seconds each , using Dumont #5 forceps , the crush site was marked with graphite powder applied in the forceps and the wound was closed using surgical clips . For immunofluorescence analysis , sciatic nerves were fixed by immersion in 4% paraformaldehyde in 0 . 1 M phosphate buffer saline ( 1× PBS , pH 7 . 4 ) for 1 hour . Followed by 3×10 minutes washes in 1× PBS , sucrose gradient ( 5% , 10% , 20% in 1× PBS ) and then embedded in OCT ( Sakura Finetek ) . Cryostat sections from the middle of the explants were cut transversely at 10 µm thickness and mounted on Superfrost Plus slides ( Thermo Scientific ) . Sections were washed in 1× PBS for 10 minutes and then blocked/permeabilized in 0 . 1% Triton X-100 , 2% fish skin gelatin ( Sigma Aldrich ) in 1× PBS for 1 hour at room temperature ( RT ) . Sections were incubated in primary antibodies in blocking/permeabilizing solution overnight at 4°C , washed in 1× PBS 3×10 minutes , and incubated in secondary antibodies for 2 hours at RT . Sections were washed 3×10 minutes in 1× PBS and mounted in Vectashield ( Vector laboratories ) . Number of axons per area of nerve tissue was assessed in images of neurofilament-immunostained explant sections using the particle analysis macro of ImageJ . The following antibodies were used for immunofluorescence analysis: rabbit anti-neurofilament heavy chain ( Sigma-Aldrich , #N4142 ) at 1∶1000; chicken anti-neurofilament medium chain ( Chemicon International , #AB5753 ) at 1∶2000; donkey anti-rabbit TRITC ( Jackson Immunoresearch Lab . Inc . , #711-025-152 ) at 1∶300; and , goat anti-chicken FITC ( Invitrogen , #A-21449 ) at 1∶300 . For EM analyses , nerves were fixed overnight by immersion in 2 . 5% glutaraldehyde , 0 . 01% picric acid , 0 . 1 M cacodylate buffer ( pH 7 . 4 ) . Nerves were rinsed in the same buffer , immersed in 1% OsO4 for 1 hour followed by in block incubation with 2% uranyl acetate for 2 hours . Nerves were dehydrated with a graded series of ethanol , propylene oxide and infiltrated with Epon ( Ted Pella Inc . ) . Ultrathin sections from the middle of the explants were contrasted with 1% uranyl acetate and lead citrate . Grids were examined with a Philips Tecnai 12 electron microscope operated at 80 kV . Negative films were developed and scanned . Sciatic nerve explants were performed as previously described [7] . Briefly , sciatic nerve segment were dissected from adult mice and cultured in Neurobasal medium supplemented with 2% B27 ( Invitrogen ) , 0 . 3% L-glutamine , and 1% streptomycin/penicillin . C57BL/6J nerves were incubated in culture medium alone ( Vehicle ) or with trolox ( Sigma-Aldrich , 1 mM ) . The explant culture medium was daily replaced . Sciatic nerves were frozen in liquid nitrogen and homogenized in 10% trichloroacetic acid ( TCA ) . The homogenate was subjected to centrifugation at 10 . 000 rpm for 10 min at room temperature , the resulting supernatant was used as the lipoperoxide fraction . Oxidized lipids were detected by mixing the supernatant with a reaction buffer ( 0 , 5% thiobarbituric acid and 20% TCA ) . The reaction solution was then incubated at 100°C for 30 min and thereafter the absorbance was measured at 512 nm . Lipoperoxide content was determined using the extinction coefficient of 155 mM−1×cm−1 and normalized to the starting nerve dry weight ( modified from reference [77] ) . Worm embryos were collected after hypochlorite treatment and embedded in OCT ( Sakura Finetek ) without a fixation step . Cryostat sections were cut at 10 µm thickness and mounted on Superfrost Plus slides ( Thermo Scientific ) . Sections were washed in 1× PBS for 10 minutes and then incubated at room temperature in the fluorogenic probe CellROX Deep Red Reagent ( Invitrogen , C10422 ) at a final concentration of 20 µM for 30 min . After washing in 1× PBS , sections were fixed in 4% paraformaldehyde in 1× PBS for 10 minutes followed by 3×10 minutes washes in 1× PBS . Sections were mounted in DAPI-containing Vectashield ( Vector laboratories ) . Sciatic nerves from wild type and YFP expressing mice were dissected at different times post injury and the perineurium was removed under a dissecting microscope . Nerve fiber bundles were attached to 3-aminopropyltriethoxysilane ( Sigma-Aldrich ) coated slides and incubated with the fluorogenic probe CellROX Deep Red Reagent ( Invitrogen , C10422 ) at final concentration of 20 µM for 1 h at 37°C . After washing in 1×PBS , nerve fibers were mounted in Fluoromount-G ( Southern Biotech ) and visualized by confocal microscopy . | Axonal degeneration and neuronal loss are currently considered crucial pathological factors in neurodegenerative diseases . Therefore , delaying or blocking these procesess is key for neuroprotection . In this work , we used an in vivo approach combining invertebrate ( C . elegans ) and vertebrate ( mice ) model systems to identify a novel and unexpected player in the mechanisms of axonal degeneration . Here , we demonstrate that both neuronal somas and axons degenerate through a step dependent on oxidative stress that can be efficiently delayed by genetic downregulation of a pathway controlling oxidative stress resistance . Impressively , we discovered that diapause formation , which is a state related to hibernating conditions , fully prevents neuronal degeneration . We uncovered new players in the degenerative mechanisms of neurons with relevance for several conditions associated to axonal degeneration , such as multiple sclerosis , motoneuron , and Parkinson diseases , offering novel potential targets for neuroprotection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"animal",
"models",
"molecular",
"neuroscience",
"caenorhabditis",
"elegans",
"cellular",
"neuroscience",
"model",
"organisms",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"neuronal",
"morphology",
"signaling",
"pathways",
"biology",
"mouse",
"neuroscience"
]
| 2012 | Diapause Formation and Downregulation of Insulin-Like Signaling via DAF-16/FOXO Delays Axonal Degeneration and Neuronal Loss |
Chromosomal instability , which involves the deletion and duplication of chromosomes or chromosome parts , is a common feature of cancers , and deficiency screens are commonly used to detect genes involved in various biological pathways . However , despite their importance , the effects of deficiencies , duplications , and chromosome losses on the regulation of whole chromosomes and large chromosome domains are largely unknown . Therefore , to explore these effects , we examined expression patterns of genes in several Drosophila deficiency hemizygotes and a duplication hemizygote using microarrays . The results indicate that genes expressed in deficiency hemizygotes are significantly buffered , and that the buffering effect is general rather than being mainly mediated by feedback regulation of individual genes . In addition , differentially expressed genes in haploid condition appear to be generally more strongly buffered than ubiquitously expressed genes in haploid condition , but , among genes present in triploid condition , ubiquitously expressed genes are generally more strongly buffered than differentially expressed genes . Furthermore , we show that the 4th chromosome is compensated in response to dose differences . Our results suggest general mechanisms have evolved that stimulate or repress gene expression of aneuploid regions as appropriate , and on the 4th chromosome of Drosophila this compensation is mediated by Painting of Fourth ( POF ) .
The effects of deficiencies , duplications or chromosome losses ( e . g . somatic elimination ) on the regulation of whole chromosomes and large chromosome domains are largely unknown , although the gene dose at most specific loci generally has little effect on the development of Drosophila . A useful scale for assessing the magnitude of aneuploidies that can be accommodated in the D . melanogaster genome without loss of viability was provided by Bridges , who divided the genome into 102 numbered divisions , based on cytological analysis of polytene chromosomes [1] . Deletions extending over more than one of these 102 divisions ( which have estimated sizes of 800–1500 kb [2] , with a median length of 1114 kb according to Flybase annotation ) are generally lethal [3] . However , there are a few known exceptions of longer , non-lethal deletions , such as Df ( 2L ) H and Df ( 3L ) Vn , which span <2 . 8 Mb and <1 . 7 Mb , respectively [4] . A general rule in Drosophila is that viability and fertility are reduced when having a single copy of ∼1% of the genome , but raising this proportion to ∼3% is lethal [3] . Hence , segmental aneuploidy-induced mortality could be explained by altered levels of gene expression within the aneuploid region , leading to an overall disturbance of gene networks [5] . However , it has been suggested that a reduced dose of any region will cause a general effect on expression of the genome and since most effects are negative in correlation to dose this is sometimes referred to as the “inverse dosage effect” [6] . Intuitively , we may expect transcript levels of genes within an aneuploid region to correlate directly with the gene dosage . However , some reports have suggested that functional autosomal dosage compensation , also known as the “buffering” effect , may occur , e . g . activities of proteins expressed from genes present in three copies , due to segmental trisomy , were found to be very similar to wild type levels in several early dosage studies [7]–[9] . Since these early studies of correlations between expression levels and gene doses relied mainly on enzyme assays ( although transcript levels of single genes were sometimes measured ) , dose responses at the transcription level were unclear , due to the potential effects of post-transcriptional processes . However , indications of buffering effects have also been obtained in recent dose response studies using genome-wide approaches [10]–[12] . For example , ∼1 . 4 fold differences in mRNA levels associated with three-fold differences in gene doses in a Drosophila autosomal region have been found in microarray analyses [10] , [13] , substantially lower than the expected 3-fold differences in the absence of compensation . It should be noted that genome-wide studies inevitably include analyses of non-expressed genes and genes expressed at sub-detectable levels; two groups of genes that will inevitably be scored as fully compensated ( i . e . as being expressed at apparently wild type levels ) and thus influence the mean calculated buffering effect . Convincing reports of chromosomal dosage compensation have hitherto only been observed in the sex chromosomes , leading to the general conclusion that this mechanism exclusively equalizes transcription between the two sexes , and compensates for the difference in the expression of sex chromosomes in relation to autosomes [13]–[17] . However , we have previously demonstrated another chromosome-wide regulatory system in Drosophila [18] , [19] , in which the Painting of fourth ( POF ) protein binds specifically to the 4th chromosome and together with heterochromatin protein 1 fine-tune the expression of genes in this chromosome [18] , [20] . Further , flies with a single 4th chromosome are viable and fertile , like flies that have a single X-chromosome , but in marked contrast to flies that have lost any other autosome . These and other observations have prompted suggestions that a dosage compensation mechanism may act upon the 4th chromosome [21] . To gain insight into the expression consequences upon chromosome 4 aneuploidies and also segmental aneuploidies in general , we have made a detailed genome-wide analysis of gene expression in aneuploidy regions in Drosophila . Using expression microarrays of haplo-4 , diplo-4 and triplo-4 flies , we show that expressed genes are significantly compensated , and that the compensation in haplo-4 flies is dependent on POF . Furthermore , we show that segmental aneuploidy regions are slightly buffered and this buffering is suggested to be at a general level and not mainly caused by a single gene feed-back regulation . Overall , the presented results suggest that general mechanisms exist to stimulate and repress gene expression .
To study the effect of gene dose on gene expression total RNA was prepared from flies with the following genotypes: heterozygous for Df ( 2L ) J-H , Df ( 2L ) ED4470 and Df ( 2L ) ED4651 deletions; heterozygous for the Dp ( 2;2 ) Cam3 duplication; monosomic for chromosome 4 ( 4/0 ) ; trisomic for chromosome 4 ( 4/4/4 ) ; and wild type controls ( where Df and Dp indicate deficiency and duplication , respectively ) . Each of these genotypes , the lengths of the affected sequences , and the respective numbers of affected genes are listed in Table S1 . Three biological replicates representing each genotype were hybridized to Affymetrix Drosophila v2 arrays , and the resultant raw data were normalized and summarized using RMA [22] . Global effects in the genome outside of our used aneuploidies can potentially influence data analysis and normalisation . We therefore analysed the raw data prior to any normalisation and could not detect any major global effects . Global effects are further discussed in Text S1 and Figure S1 . Non-expressed genes and genes with expression levels that are sub-detectable in the micro-array analysis will be scored as fully compensated when the aneuploids are compared to the wild type . Including these genes inevitably shifts the average closer to wild type expression levels , potentially leading to over-estimates of any buffering effect . Therefore , cut-offs for genes with correctly measured expression levels were determined by plotting transcription levels in mutants against wild type expression levels ( Figure S2 ) . The resulting plots showed that aneuploidy effects were only detected for genes with wild type expression levels >6 ( log2-scale ) . In all arrays we then removed the genes with wild type expression values below 6 and renormalized the expression values . In this normalisation , a constant was added to all the mutant array expression values to ensure that the total genomic expression matched that of the wild type . The average expression relative to wild type was then measured for all of the expressed genes within each aneuploid region . Genes within the Df regions were significantly buffered ( one sample Wilcoxon test , p<<0 . 001 ) , since they were expressed at 64% of wild type levels , compared to the 50% expression level expected under the naïve assumption of regulatory independence ( Figure 1B ) . This buffering effect was weaker than those observed in previous studies [10]–[12] , and we hypothesized that this difference was mainly due to our exclusion of non- and weakly-expressed genes . This speculation was confirmed , since the buffering levels in our pre-cut-off data were similar to previously reported values ( data not shown ) . However , it is important to note that it is still not known whether weakly-expressed genes are actually buffered , and if so to what degree . The effects of the aneuploid regions are shown in plots of moving median expression ratios along the chromosome arms in Figure S3 . A significant buffering effect was detected in the 4/0 flies ( one sample Wilcoxon test , p<<0 . 001 ) , of similar strength to that observed in the Df flies ( Mann-Whitney U test , p = 0 . 21 ) . A triploid region ( Dp ) in the 4/4/4 background also showed a buffering effect , with a slight decrease in expression ( 146% compared to the expected 150% ) , although this was not significant ( one sample Wilcoxon test , p = 0 . 079 ) . Df ( 2L ) J-H/+ flies are viable also in 4/0 background and there was no significant difference in the effects of the Df ( 2L ) J-H deficiency in wild type compared to 4/0 backgrounds ( Mann-Whitney U test , p = 0 . 28 ) . However , the entire 4th chromosome was significantly compensated in 4/4/4 flies ( 139% compared to the expected 150% , one sample Wilcoxon test , p = 0 . 015 ) . Chromosome 4 will be discussed in more detail below . The observed buffering effect could have been caused by either the feed-back regulation of individual genes or a more general buffering mechanism . However , if it was mainly caused by the former , the distribution of differences in expression levels between the Df and wild type genotypes would probably be highly skewed , since most genes would be expected to be expressed at close to 50% of wild-type levels , while the expression of a few genes would be buffered to varying degrees . Instead , the expression differences were approximately normally distributed ( Shapiro-Wilk's W test , p = 0 . 20 ) around a mean of 64% wild-type expression ( Figure 2A ) . In contrast , the Dp genotypes showed no significant buffering effects , and the differences between their expression levels and wild-type levels were not normally distributed ( Shapiro-Wilk's W test , p = 0 . 0030 , Figure 2B ) . This could mean that any potential buffering system for genes when they are present in three copies is less evolved than when they are present in one copy . We then asked whether the observed buffering effect correlated with any particular class of genes . No correlations were found between the buffering effect and expression levels , except for a weak relationship in 4/4/4 flies ( Spearman correlation , p = 0 . 032 , Figure S4 ) . Neither were there any correlation between the buffering effect and gene length ( data not shown ) . However , a clear correlation was found between buffering and ubiquitously expressed genes ( UEGs ) ( Figure 3 ) , here defined as genes expressed at levels >6 in all 12 tissues present in the FlyAtlas database [23] . The UEGs were significantly less buffered than non-ubiquitously expressed genes ( NUEG ) in the Df and 4/0 flies ( Mann-Whitney U test , p = 0 . 021 and p = 0 . 00045 , respectively , Figure 3A and 3C ) . Conversely , the NUEGs were significantly less buffered in the Dp and 4/4/4 flies ( Mann-Whitney U test , p = 0 . 038 and p = 0 . 0017 , respectively , Figure 3B and 3D ) . Thus , UEGs appear to be only buffered when present in three copies . As shown in Figure 1 , chromosome 4 is compensated in response to altered dose . Compensation of the 4th is slightly higher but not significantly different from compensation in segmental aneuploidies ( deficiencies ) . We have previously shown that the protein POF specifically stimulates gene expression on the 4th chromosome , and that Pof is essential for the survival of 4/0 flies [18] . Hence , we constructed expression arrays from Pof mutants with two or three copies of the 4th chromosome ( no arrays of mutants with a single copy could be made , since haplo-4 flies do not survive without POF ) . As seen in Figure 4A , POF always stimulated expression , regardless of the 4th chromosome copy number . Strikingly , there was also a clear negative linear correlation between the differences in expression , relative to the wild type , between the 4/0 and Pof mutants ( Figure 4B , Pearson correlation , r = −0 . 48 , p<<0 . 001 ) . This implies that the level of compensation in 4/0 flies is inversely proportional to the level of expression change in Pof mutants . Thus , we conclude that the compensation observed in 4/0 is directly mediated by POF . Moreover , the distributions of the buffering effects in 4/0 and Pof mutants were not normal ( Shapiro-Wilk's W test , p = 0 . 014 and p = 0 . 014 respectively ) , but rather displayed two clear peaks ( Figure 4C ) . Both of these data sets therefore appear to contain data on one group of strongly affected genes and one that is almost unaffected . The unaffected groups consisted mainly of NUEGs in 4/0 and UEGs in Pof mutants , whereas the strongly affected groups were mainly composed of UEGs in 4/0 and NUEGs in Pof mutants ( Figure 4D and 4E ) . The high expression of POF in the testes and the strong relationship between POF and dosage compensation prompted us to examine the role of POF in the testes . In order to understand the role of POF we performed immunostainings for POF and immunofluorescens localisation of a P[Pof . EYFP] transgenic constructs in male testes . The results are presented and discussed in Figure S5 and Text S1 . Expression arrays were then used to assess the influence of POF on transcription in the dissected testes ( Pof mutants and wild type control ) , and the results clearly showed that POF mainly altered the expression of genes in the 4th chromosome ( Figure 5A ) . We then tested whether the expression levels of testes-specific genes were altered in Pof mutants . We did observe a weak effect on these genes , although unexpectedly the expression was higher in Pof mutants ( 104% , one sample Wilcoxon test , p<<0 . 001 ) , which we hypothesise could be caused by delayed spermatogenesis . Average reductions in expression levels were found to be similar in Pof mutant adult female , testes and first instar larvae tissues ( first instar data from [18] , Kruskal-Wallis ANOVA by Ranks , p = 0 . 18 ) . The effect on individual genes was also seen to be linearly correlated ( Figure 5B , three pair-wise Pearson correlations , r = 0 . 51–0 . 68 , p<<0 . 001 ) , and thus we conclude that the effect of POF on chromosome 4 genes is the same in all three of these tissues .
Previous studies on the relation between chromosome dose and transcript level response suggest the existence of buffering effects [10]–[12] . The effect is dramatic , a three-fold difference in gene dosage , obtained using the Df and Dp genotypes examined here , were found to be associated with ∼1 . 4 fold differences in transcript levels , rather than the expected 3-fold differences [10] , [13] . It is important to note that mRNA levels have been measured in most genome-wide expression studies , and thus it is still unclear whether the observed effects are due to transcriptional differences or post-transcriptional effects . Using all our data we found a buffering effect of similar strength to those previously reported ( a 3-fold difference in gene dosage resulted in 1 . 5 fold differences in transcript levels ) . However , we also found that expression can only be reliably measured for genes with relative expression levels >6 , and when we only analyzed these genes we found a less dramatic , but still significant buffering effect of deficiencies . In contrast , when expressed genes were analyzed , no buffering effects in responses to duplication were detected . Hence , gene dosage reductions ( but not apparently increases in dosage ) can be compensated for by buffering , when all the expressed genes are considered . What causes the observed buffering effect ? We can consider two plausible models to explain this . First , the calculated buffering effect may be a consequence of a more or less complete feed-back regulation of a subset of genes . Secondly , the observed buffering is mainly caused by a general increased expression of the genes uncovered by the Df . The obtained expression values for Df-WT were normally distributed and centred on a mean expression value of 0 . 64 ( Figure 2 ) . The normality of the distribution suggests that the observed buffering effect was general , and thus that individual gene feed-back regulatory mechanisms ( which would probably have yielded a skewed distribution ) were not primarily responsible for the calculated mean effect . Hence , the results from the Df indicate that the buffering system is general , and that the variation around the mean is mainly caused by array noise . Two possible general buffering mechanisms could also be envisioned . Firstly , a monosomic region could be “sensed” and actively targeted by compensating protein complexes , similar to those described for the male X-chromosome and the 4th chromosome in Drosophila [15] , [18] . Alternatively , there could be feedback regulation of a few individual genes , and stimulated expression could result from high local concentrations of transcription-stimulating factors and/or “spread” from the nuclear environment of a single region . The mechanism for the suggested general buffering effect is likely to be a mixture of events at different levels which remains to be unravelled . We examined whether the observed buffering was correlated to expression levels . This is a reasonable assumption since in the two known chromosome-wide regulatory systems in Drosophila , the MSL mediated dosage compensation and POF mediated regulation of the 4th chromosome , there is a relation between protein binding to genes and expression levels . In the case of dosage compensation , MSL binding is correlated to expressed genes but not to expression levels [24] , [25] . However , to a large extent MSL binding reflects the expression levels in young embryos and the binding is then for most genes stable throughout development [26] . On the other hand , POF binding to the 4th chromosome is linearly correlated to gene expression levels [20] . Even though POF binding to genes is directly correlated to gene expression levels we find no correlation of buffering effects to gene expression levels . We also examined whether differences in the normal regulatory patterns of genes affect their degree of buffering , by dividing the set of studied genes into ubiquitously expressed genes ( UEGs ) and non-ubiquitously expressed genes ( NUEGs ) , then comparing their buffering levels . The results indicated that UEGs can be repressed , but not stimulated ( as seen in the Dp and Df genotypes , respectively ) . The UEG expression levels are probably primarily limited by their copy numbers , and thus it is not possible to further stimulate their expression when they are present as single copies . In contrast , UEGs in trisomic regions are generally more repressed than NUEGs . It should be stressed that while both UEGs and NUEGs are buffered in Df and 4/0 conditions , the UEGs are buffered to a much smaller extent . However , the NUEGs show no signs of buffering in Dp and 4/4/4 conditions . The observed disparity between the UEGs and NUEGs must , presumably , be mainly due to regulatory differences , i . e . mechanisms have evolved that allow expression of the NUEGs to be responsive to various inducting and silencing signals , while the transcription of UEGs is steady , stable and more resistant to signal variations . The difference is even more pronounced on the 4th chromosome , where the NUEGs are strongly compensated when present in single copies , i . e . in 4/0 . In addition , our data show that POF was responsible for the observed buffering of the 4th chromosome , and the buffering of 4/0 was of similar strength to Df buffering on the major autosome arms . POF shows strong similarities to the dosage-compensating MSL complex in evolutionary terms [19] , [27] , in binding profile [20] and in its function as a chromosome-wide regulator [18] . The mechanism responsible for MSL dosage compensation of the X-chromosomes is MOF-mediated hyperacetylation of H4K16 . It should be noted that recent genome-wide studies suggest that MOF also acts as a more general regulator of gene expression in Drosophila . However , it is not known whether this general function is involved in the general dose response [28] . Nevertheless , it seems reasonable to hypothesize that the buffering effect seen in Df genotypes acts similarly to POF- and MSL-mediated stimulation , i . e . at the transcriptional level . We speculate that the more generally and stably expressed UEGs are less responsive to buffering functions than NUEGs , however the reasons why UEGs are less dose-responsive than NUEGs when present in three copies remains to be elucidated . What causes the lethality in haplo-lethal deficiencies ? It is obvious that genes with a strong influence on viability as exemplified by Minute ( ribosomal protein encoding ) genes will , when uncovered , increase the risk for lethality [29] , [30] . Still , there seem to be a strong link between length of a deficiency and haplo-lethality [3] . Various models can be proposed to explain haplo-lethality caused by deficiencies that delete a large number of genes , one of which suggests that large deficiencies alter the doses of a number of genes involved in one or more genetic networks , thereby inducing lethality through a network collapse rather than alteration of the dose of any single gene [5] . Haplo-lethality could also be a consequence of the inverse dosage effect . In this model a haploid region will cause a general genome-wide stimulation since most effects are negative in correlation to dose [6] . It is difficult to predict the outcome of the inverse dose effect since the magnitude of this effect is not known . It is also unclear whether it will act on the whole genome or will be biased to the aneuploidy region as a consequence of gene clustering . Based on our data we suggest that general buffering mechanisms are present , and although no molecular mechanisms have been ascribed to buffering effects associated with segmental or chromosomal aneuploidies we speculate that increases in the length of deletions increase the pressure on the flies' buffering capacity . Hence , the plasticity of this system could compensate for monosomy up to a certain threshold , at which lethality may occur due to a collapse of buffering properties . Our study indicates the presence of buffering in Df but not as well in Dp , and a model suggesting haplo-lethality to be a consequence of buffering collapse would be consistent with such results . In general , flies tolerate duplications better than deficiencies , and our results are consistent with this general rule , since the pressure on buffering capacity seems to be weaker in the Dp than in the Df genotypes . We have previously shown that POF stimulates 4th chromosome gene expression , and that the absence of Pof results in haplo-4th lethality [18] . The results from the study presented here also show a significant negative linear correlation between the effects of 4/0 and the lack of Pof . This is intriguing , since it demonstrates that compensation of the 4th chromosome is mediated by POF . Thus , we have identified the mechanism responsible for buffering of the 4th chromosome . In addition , POF almost exclusively acts on NUEGs ( Figure 4 ) , although previous ChIP-chip analyses have shown POF targeting of genes to be proportional to their expression levels , regardless of whether they are UEGs or NUEGs [20] . Therefore , we hypothesize that POF binds to all expressed genes on the 4th chromosome , but only the NUEGs respond to POF-mediated stimulation of expression , implying that buffering occurs after transcription initiation . Notably , both the 4th chromosome and the major chromosome arms respond to buffering functions in haplo-conditions . This compensation is mediated by POF in the 4th chromosome , but the mechanisms responsible for buffering of the major autosome arms are still unknown . In contrast to Dp , significant ( repressive ) buffering was also detected in 4/4/4 , possibly mediated by heterochromatin protein 1 . MSL-complex mediated , 2-fold up-regulation of the male X-chromosome is generally agreed to be the dosage compensation mechanism in somatic cells [14]–[16] . However , X-chromosome dosage compensation also occurs in the testes , where the MSL complex is not present , and to date no mechanism has been identified for this germline dosage compensation [31] , [32] . However , POF is highly expressed in testes tissues [19] , which along with the striking similarities between POF- and MSL-mediated chromosome-wide regulation prompted us to examine the importance of POF in the dose compensation of the X-chromosome in the testes . The nuclear localisation of POF in many studied cell types indicates that it is associated with the 4th chromosome , in accordance with results of previous ChIP-chip analyses [20] , [27] . Drawing definitive conclusions about which genes , if any , POF associates with in spermatocytes is difficult ( although our microarray analysis of testes tissue demonstrated the 4th chromosome genes to be the main regulatory targets for POF in the male germline ) due to the intense POF nuclear staining , which may mask more localised association in the spermatocyte nuclei ( Text S1 , Figure S5 ) . However , there were no significant buffering effects of X chromosome genes in Pof mutants , so there was no evidence of POF-mediated dosage compensation in the mutant male germlines . The Pof mutants did show a slight increase in the expression of testes-specific genes , but this effect was minor and could have been a consequence of minor differences in spermatogenesis between our Pof mutant and wild type . We conclude that the average reduction in gene expression on the 4th chromosome of Pof mutants is similar in the three studied tissue stages ( adult females , testes and 1st instar larvae ) , and that the effect on individual genes is linearly correlated . The results shown here have implications . Deficiency screens are commonly used as a method to find genes involved in different biological pathways . Based on our results we anticipate that these screens will find UEGs more efficiently than NUEGs , although it should be stressed that the dose responses of genes with low expression levels are still not understood . The higher dose sensitivity of UEGs is supported by the dramatic effects of reductions in doses of ribosomal protein genes , as manifested in the associated Minute phenotypes [29] , [30] . Notably , our simple categorization of UEGs and NUEGs classified all but one of the 61 annotated Minute ribosomal protein genes as UEGs . The difference in dose response between genes based on their expression also has consequences for our understanding on how chromosomal aberrations and chromosomal aneuploidies influence proper development .
Flies were cultivated and crossed at 25°C in vials containing potato mash-yeast-agar . The Df ( 2L ) J-H/SM5 stock were obtained from the Kyoto Drosophila Stock Center , the Dp ( 2;2 ) Cam3/CyO from Bloomington , and the Df ( 2L ) ED4651/SM6a and Df ( 3L ) 4470/TM6C from Szeged ( Df and Dp indicate deficiency and duplication , respectively ) . y1 w67c23 was used as wild type . Df/+; 4/4 females were generated by crossing Df/Bal flies to wild type Oregon R . Df ( 2L ) J-H/+; 4/0 females were generated by crossing Df ( 2L ) J-H/SM5 to C ( 4 ) RM svspa-pol/0 . The Df ( 2L ) J-H/+; 4/0 offspring were isolated based on their Minute phenotype . +; 4/0 females were generated similarly by crossing wild type to C ( 4 ) RM svspa-pol/0 . Dp ( 2;2 ) Cam3/+; 4/4/4 females were generated by crossing Dp ( 2;2 ) Cam3/CyO to C ( 4 ) RM svspa-pol/0 . The Dp ( 2;2 ) Cam3/+; 4/4/4 offspring were isolated based on non-Minute phenotype . The Pof119; 4/4/4 females were generated similarly by crossing Pof119/CyO; C ( 4 ) RM svspa-pol/0 to y1 w67c23; PofD119/PofD119 and the PofD119; 4/4 females were offspring from the y1 w67c23; PofD119/PofD119 stock . For microarray analysis total RNA was isolated using TRIzol reagent ( Invitrogen ) followed by a purification using RNeasy kit ( Qiagen ) according to the instruction by the suppliers . 10 adult females ( 0–24h ) were used for each of three biological replicates of each genotype . For testes microarrays , 60 testes from 0–24 old males were used for each of three biological replicates of y1 w67c23; PofD119/PofD119 and three replicates of y1 w67c23 as controls . The 33 labelled cDNA probes were then hybridized to an Affymetrix Drosophila gene chip ( version 2 ) and the intensity values were normalised and summarized using robust multi-array analysis ( RMA ) [22] . Other normalisation methods , such as MAS5 , were also tested and they all gave similar results to RMA . All microarray data analyses were done using R ( www . R-project . org ) and the Bioconductor package [33] . The resulting data are available at http://www . ncbi . nlm . nih . gov/geo/ ( Accession: GSE14517 , GSE14516 ) . Based on expression array data in the FlyAtlas database [23] ( Geo accession number: GSE7763 ) , ubiquitously expressed genes ( UEGs ) were defined as genes showing expression levels of at least 6 in all of the 12 examined tissues after RMA normalization , while all other genes were defined as non-ubiquitously expressed genes ( NUEGs ) . Testis-specific genes were defined , using the same dataset , as genes showing an expression level of ≥6 in testes and < 6 in all other tissues . The first instar larvae data from [18] and the testis data were renormalized in the same way as the adult female data after removing all genes expressed below 6 ( after RMA ) in the respective wild type . All statistical analyses were performed on log2-scaled data using Statsoft Statistica 8 . 0 . For whole mount immunostaining , wild type testes were dissected in PBS , fixed for 30 minutes in 4% para-formaldehyde in a solution containing 0 . 1 M Hepes , 2 mM EGTA and 1mM MgSO4 ( pH 6 . 9 ) , then stained essentially according to [34] , using an anti-POF chicken polyclonal primary antibody ( 1∶100 dilution ) followed by a pre-absorbed biotinylated Donkey anti-chicken IgY secondary antibody ( 1∶300 , Jackson ) , which was detected by the brown HRP reaction ( H2O2 , DAB ) . For indirect immunofluorescence staining , testes squashes were fixed according to [35] ( Protocol 5∶5 ) . The slides were then washed in 1×PBT for 30 min , transferred to a blocking solution ( 0 . 1 M maleic acid , 0 . 15 M NaCl , 1% Boehringer blocking reagent ) and incubated for 30 min at room temperature . The slides were incubated overnight at 4°C with a 1∶100 diluted anti-POF chicken polyclonal primary antibody , then washed for 2×10 minutes ( in 0 . 1 M maleic acid , 0 . 15 M NaCl , 0 . 3% Tween 20 ) , and then blocked for 30 minutes . A 1∶300 diluted donkey anti-chicken IgY conjugated with Cy3 ( Jackson ) was then applied as a secondary antibody prior to a further 2 h incubation at room temperature . The squashes were counterstained with DAPI ( 1 µg/ml ) and washed for 2×10 minutes ( in 0 . 1 M maleic acid , 0 . 15 M NaCl , 0 . 3% Tween 20 ) before mounting with Vectashield ( Vector ) . Live testes squashes from young adults carrying the P[w+ Pof . EYFP] construct ( Pof fused to enhanced yellow fluorescent protein-encoding sequence under the control of the endogenous Pof promoter [27] ) were dissected in TB ( 183 mM KCl , 47 mM NaCl , 10 mM TRIS-HCl , 1 mM PMSF , 1 mM EDTA , pH 6 . 8 ) and prepared according to [35] . Preparations were examined by phase contrast , Nomarski and fluorescence microscopy under a Zeiss Axiophot microscope equipped with a KAPPA DX20C charge-coupled device camera . The images obtained were assembled and contrasted using Adobe Photoshop . The microarray data reported in this paper have been deposited at http://www . ncbi . nlm . nih . gov/geo/ ( Accession: GSE14517 , GSE14516 ) . | Although deletion heterozygotes and chromosomal aneuploidies have been used in genetic studies for decades , the relationships between chromosome doses and transcript outputs have been difficult to unravel . In other words , the effects of copy changes on the regulation of entire chromosomes or large chromosomal domains are largely unknown . Hence , we studied these relationships in Drosophila using microarrays prepared from flies with a dosage series of chromosomal domains and a dosage series of the 4th chromosome . We observed significant buffering of expressed genes , i . e . , on average they were expressed at >50% of wild-type levels when present in single copies instead of two copies ( the normal complement of diploids ) . This buffering was also seen to be much stronger for differentially expressed genes than ubiquitously expressed genes . Our findings therefore support the presence of chromosome-wide buffering mechanisms . In addition , we found evidence of a chromosome-specific protein POF-mediated mechanism in the buffering of the 4th chromosome . Overall , our results suggest that a general buffering system acts on most genes present as single copies due to deletions or chromosome losses . Further work on gene buffering effects should make substantial contributions to our understanding of genome-wide gene regulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/genomics",
"molecular",
"biology/chromatin",
"structure",
"genetics",
"and",
"genomics/gene",
"expression",
"molecular",
"biology/chromosome",
"structure",
"genetics",
"and",
"genomics/chromosome",
"biology",
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics/bioinformatics"
]
| 2009 | Buffering of Segmental and Chromosomal Aneuploidies in Drosophila melanogaster |
The clinical syndrome associated with secondary syphilis ( SS ) reflects the propensity of Treponema pallidum ( Tp ) to escape immune recognition while simultaneously inducing inflammation . To better understand the duality of immune evasion and immune recognition in human syphilis , herein we used a combination of flow cytometry , immunohistochemistry ( IHC ) , and transcriptional profiling to study the immune response in the blood and skin of 27 HIV ( - ) SS patients in relation to spirochetal burdens . Ex vivo opsonophagocytosis assays using human syphilitic sera ( HSS ) were performed to model spirochete-monocyte/macrophage interactions in vivo . Despite the presence of low-level spirochetemia , as well as immunophenotypic changes suggestive of monocyte activation , we did not detect systemic cytokine production . SS subjects had substantial decreases in circulating DCs and in IFNγ-producing and cytotoxic NK-cells , along with an emergent CD56−/CD16+ NK-cell subset in blood . Skin lesions , which had visible Tp by IHC and substantial amounts of Tp-DNA , had large numbers of macrophages ( CD68+ ) , a relative increase in CD8+ T-cells over CD4+ T-cells and were enriched for CD56+ NK-cells . Skin lesions contained transcripts for cytokines ( IFN-γ , TNF-α ) , chemokines ( CCL2 , CXCL10 ) , macrophage and DC activation markers ( CD40 , CD86 ) , Fc-mediated phagocytosis receptors ( FcγRI , FcγR3 ) , IFN-β and effector molecules associated with CD8 and NK-cell cytotoxic responses . While HSS promoted uptake of Tp in conjunction with monocyte activation , most spirochetes were not internalized . Our findings support the importance of macrophage driven opsonophagocytosis and cell mediated immunity in treponemal clearance , while suggesting that the balance between phagocytic uptake and evasion is influenced by the relative burdens of bacteria in blood and skin and the presence of Tp subpopulations with differential capacities for binding opsonic antibodies . They also bring to light the extent of the systemic innate and adaptive immunologic abnormalities that define the secondary stage of the disease , which in the skin of patients trends towards a T-cell cytolytic response .
Syphilis is a sexually transmitted multi-stage disease caused by the spirochetal bacterium Treponema pallidum ( Tp ) , subspecies pallidum [1] , [2] . Despite the existence of inexpensive and effective antibiotic treatment regimens , more than 10 . 5 million new syphilis cases are estimated to occur yearly throughout the world [3] . Infection begins when the bacterium comes into contact with skin or mucous membranes , multiplying locally , while simultaneously disseminating through blood vessels and lymphatics [1] , [2] , [4]–[6] . The distinctive painless ulcer ( chancre ) of primary syphilis typically appears 2–4 weeks after the initial contact with the spirochete [2] , [6] , [7] . By this time , organisms that have disseminated from the primary site of infection have invaded various organ tissues , most notably the skin [2] , [6] , setting the stage for what is classically known as secondary syphilis ( SS ) . SS , the principal focus of the current study , characteristically presents with a variety of muco-cutaneous manifestations as well as systemic signs and symptoms within 4–10 weeks of the initial infection [4] , [5] , [8] . Despite the robust nature of the cellular and humoral immune responses associated with SS , weeks to months may elapse before lesions resolve . Infectious relapses are common during the first few years of infection [2] , while approximately one-third of untreated patients develop one of the potentially devastating forms of recrudescent disease collectively referred as tertiary syphilis [9] . The factors that influence the complex and shifting balance between this persistent bacterium and host clearance mechanisms are not well understood . T . pallidum contains abundant lipoproteins which are capable of activating macrophages and DCs via CD14 [10]–[13] and Toll-like receptor 1 ( TLR1 ) and TLR2-dependent signaling pathways [11] , [12] , [14]–[16]; consequently , these pathogen associated molecular patterns ( PAMPs ) are believed to be major pro-inflammatory agonists during spirochetal infection [17] . However , due to the bacterium's unique outer membrane ( OM ) structure , which includes a lack of surface exposed lipoproteins [18]–[22] , these PAMPs are not readily accessible to TLRs or other pattern recognition receptors ( PRRs ) present on monocytes/macrophages or dendritic cells ( DCs ) . As a result , it is believed that spirochetes can replicate in tissues and disseminate without triggering innate pathogen recognition systems . Presumably , as local spirochetal burdens increase , a small number of organisms are taken up by tissue-based DCs; which then traffic to draining lymph nodes to present cognate treponemal antigens to naïve T and B-cells . The emergence of opsonic antibodies would then enhance uptake and degradation of the bacterium in tissues , allowing spirochetal PAMPs to gain access to PRRs lining the phagocytic vacuole and triggering their activation [23] . Because of the bacterium's extraordinarily low density of integral outer membrane proteins ( OMPs ) [1] , [19] , [24] , [25] and the limited antibody responses they elicit in humans [24]–[26] , anti-treponemal antibodies alone are unlikely to be sufficient to control bacterial replication and prevent further dissemination . In support of this idea , ex vivo opsonophagocytosis assays using either rabbit peritoneal macrophages [27] or human PBMCs [28] point out that even in the presence of syphilis immune sera , substantial numbers of spirochetes avoid phagocytosis . Lastly , findings from a recent study provide additional evidence that organisms within Tp populations differ widely with respect to the density of surface antigens recognized by syphilitic sera [25] . T . pallidum is capable of provoking an intense cellular immune response generally believed to be the cause of the tissue damage that gives rise to clinical manifestations [5] . The extent to which the diverse cellular components of syphilitic infiltrates contribute to clearance of spirochetes , however , remains an open question . In the rabbit model , the appearance of Tp reactive lymphocytes correlates with the progression of mononuclear cell infiltration and macrophage activation at the sites of experimental inoculation [29]–[31] . Immunohistochemistry ( IHC ) and RT-PCR analysis of biopsy specimens obtained from patients with primary and secondary syphilis lesions demonstrate that syphilitic skin lesions are also composed of lymphocytes and macrophages capable of expressing mRNA for the Th1 cytokines , IL-2 , IFNγ and IL-12 [32] , [33] . While helper T-cells outnumber cytolytic T-cells in experimentally infected rabbit tissues [34] and in human primary syphilitic lesions [35] , equal or greater numbers of CD8+ T-cells characterize human SS syphilis inflammatory infiltrates [35]–[38] . The finding by Van Voorhis et al [32] that both perforin and granzyme B are expressed in human syphilis lesions supports the idea that in Tp-infected SS skin tissues cytolytic T-cells have a role in bacterial clearance . How CD8+ T-cells are activated in the skin is unclear given that this lymphocyte subset usually responds to antigens presented via the class I Major Histocompatibility Complex ( MHC ) pathway [39] , which is generally not associated with control of extracellular pathogens like Tp . Efforts to understand the duality of immune evasion and immune recognition in syphilis have been hindered by the inability to propagate the bacterium in vitro and the lack of a suitable inbred animal model for performing immunologic studies . To circumvent these problems and obtain information directly relevant to the disease process in humans , we have been studying SS , the stage in which the dichotomous features of syphilitic infection are clearly evident and specimens are readily obtainable . Herein , we used a combination of flow cytometry , IHC and transcriptional profiling to investigate key aspects of the innate and adaptive immune response in the blood and skin of untreated SS patients in relation to the spirochetal burdens present in each of these two immunologically distinct compartments . We then used our previously described ex vivo opsonophagocytosis assay [28] , [40] to model spirochete-monocyte/macrophage interactions in the blood and skin . As a whole , our findings support the importance of opsonophagocytosis as a primary means for clearance of treponemes , while suggesting that the balance between phagocytic uptake and evasion is determined by the relative burdens of bacteria and the presence of Tp subpopulations with differential capacities for binding opsonic antibodies . The findings in the skin demonstrate that in addition to CD4+ and CD8+ T-cells , CD56+ NK-cells are also enriched and are thus likely to participate in activation of dermal macrophages through their ability to secrete IFN-γ . Unexpectedly , we discovered that patients have profound immunophenotypic alterations in circulating monocytes , DCs and NK-cells , including the emergence of a CD56negativeCD16high NK-cell subset that is known to be highly dysfunctional in patients with uncontrolled chronic viral infections [41] , [42] . These findings reveal the extent of the systemic innate and adaptive immunologic abnormalities that define the secondary stage of the disease , which in the skin of patients trends towards a T-cell cytolytic response .
Adult SS patients were identified and referred for enrollment through a previously described network of health care professionals in Cali , Colombia [8] . The diagnosis of SS was based on the medical history and compatible skin or mucosal lesions , reactive non-treponemal test ( RPR , Rapid Plasma Reagin titer ≥1∶8 ) and a positive confirmatory treponemal test ( FTA-ABS , Fluorescent Treponemal Antibody Test Absorbed ) . All serological tests were performed at a reference laboratory in Colombia ( Clínica Colsanitas ) . Patients were excluded if they were known to be HIV-positive , if they had serologic evidence of current or prior infection with hepatitis B or hepatitis C , were receiving anti-inflammatory or immunosuppressive medications , had recently used antibiotics , or had a history of chronic dermatitis or other underlying acute or chronic disease . Peripheral whole blood samples obtained from enrolled patients , along with 4-mm punch skin biopsies from secondary syphilis lesions from a subset of these patients , were processed for immunological and molecular assays as described below . All patients were treated with 2 . 4 million units of intramuscular benzathine penicillin as recommended by Colombian public health standards , which are in accord with available CDC treatment guidelines . Patients were asked to return two months after receiving antibiotic treatment for a clinical and immunological follow-up . Healthy control volunteers ( non-reactive RPR , negative FTA-ABS , non-HIV/HBV/HCV ) , of similar background and socio-economic status , were recruited by the study site in Cali . Healthy volunteers , with no serologic evidence of prior or current syphilis , were recruited at the University of Connecticut Health Center ( UCHC ) to serve as controls for the ex vivo Tp-monocyte stimulation experiments ( described below ) . The Institutional Review Boards of , Centro Internacional de Entrenamiento e Investigaciones Médicas ( CIDEIM ) in Cali , Colombia , the Connecticut Children's Medical Center ( CCMC ) , UCHC and the Center for Diseases Control and Prevention ( CDC ) approved all relevant study protocols . All healthy volunteer and syphilis patients , regardless of whether they were enrolled by the Cali site or at UCHC , gave voluntary written informed consent to participate in the study . A total of 27 HIV-negative SS patients were eligible for participation . Clinical and epidemiologic features for these patients are summarized in Table 1 . Peripheral blood mononuclear cells ( PBMCs ) obtained from whole blood samples from these patients were examined by flow cytometry and RT-PCR at the time of enrollment as described below . Flow cytometric analysis was repeated in a subset ( n = 13 ) of enrolled SS patients approximately 60 days after receiving antibiotic treatment . A total of 12 of the 27 SS patients also had skin biopsies processed for targeted array analysis ( 12/12 ) ; skin biopsies were also studied by IHC in four of the twelve patients . We previously reported quantitative Tp DNA results from whole blood samples obtained from all 27 SS patients studied herein [8] . In the current study we also determined spirochetal burdens in 4 SS skin lesion samples from these same patients and that also were studied by IHC . A total of 26 healthy volunteers were enrolled at the Cali site; 23 controls were included for flow cytometric immunologic studies and three additional subjects provided healthy skin control samples for microarray analysis ( see below ) . Live Tp ( Nichols strain ) was used for the monocyte simulation experiments on the same day of the extraction from rabbit testicles as previously described [28] . All animal experimentation was conducted following the NIH guidelines for housing and care of laboratory animals and was performed in accordance with the UCHC institutional regulations after review and approval by Institutional Animal Care and Use Committee . Highly purified human monocytes were isolated from healthy volunteer PBMCs using a magnetic cell sorting monocyte isolation kit ( Miltenyi Biotech , Auburnas ) as previously described [17] . Cells were plated and incubated with 10% heat inactivated normal human sera ( NHS ) or human syphilitic sera ( HSS ) for 8-hours at 37°C/5% CO2 with fresh Tp at multiplicities of infection ( MOIs ) of 1 , 10 and 30 . In some assays , 100 ng/ml of LPS ( Sigma-Aldrich ) was used as a positive control for cytokine production . At the end of the 8-hr incubation period , cells were harvested for flow cytometry , epifluorescence and confocal microscopy . Supernatants were collected for cytokine analysis and Tp counting . Experiments with HSS were performed using a pool of sera from a group of HIV-seronegative SS patients as previously reported [28] . All culture media and reagents utilized in the stimulation experiments were confirmed to be free of LPS contamination ( <10 pg/ml ) by Limulus amoebocyte lysate assay quantification ( Cambrex , MA ) . Isolated monocytes from healthy US volunteers and PBMCs from SS syphilis patients and controls , where processed for flow cytometry as previously described [43] . The antibody panels used in for flow cytometry are listed in Table 2 . Surface staining procedures were done as previously described [28] . Individual cell populations were selectively gated for analysis based on the expression of corresponding immuno-phenotypes . Multiparameter files were analyzed using WINMDI v2 . 8 software ( Joseph Trotter , Scripps Clinic ) . Human PBMCs obtained from healthy US volunteers were plated and stimulated with freshly extracted Tp Nichols strain at 37°C/5% CO2 . Selected samples were incubated with 10% heat inactivated ( 56°C for 30 min ) NHS or with 10% heat inactivated HSS obtained from individual SS patients or pooled samples from Cali SS patients . Samples were incubated in the presence of LysoTracker Red endosomal dye ( Molecular Probes ) , and harvested after a 4-hr incubation period . Tp–cell associations were visualized by immunofluorescence assay ( IFA ) as previously described [28] . Images were acquired on an Olympus BX41 epifluorescence microscope equipped with a Retiga Exi CCD camera ( QImaging ) and processed with ImageJ 1 . 40 ( NIH , USA ) . To quantitate spirochetal uptake , up to 10 fields were selected sequentially and monocytes containing internalized and degraded spirochetes in the form of fluorescent blebs were counted using images acquired by epifluorescence microscopy . A total of 100 cells were counted for isolated monocyte experiments . After an 8-hr incubation period 10 µl aliquots from Tp-stimulated-monocyte supernatants were enumerated , in triplicate , by dark-field microscopy on a Petroff-Hausser counting chamber . Percentage of bacterial recovery was calculated using a “time zero” spirochetal count . Simultaneous measurements of TNF-α , IL1-β , IL-6 and IL-10 were performed in supernatants from ex vivo experiments and in individual SS patient's serum , using a Human Inflammatory Cytokine Bead Array ( CBA ) per the manufacturer's ( BD ) protocol . Isolated PBMCs ( 2×106 cells ) from SS patients and healthy controls were stored in 300 µl of RNA later at −80°C until processing . RNA was extracted at the Cali site using the RNeasy Mini Kit ( Qiagen ) according to the manufacturer's protocol . Up- or down-regulation of selected transcripts were measured in Complementary DNA ( cDNA ) from ex vivo Tp-monocyte stimulation experiments for selected genes by quantitative RT-PCR ( qRT-PCR ) analyses . RNA was extracted from both stimulated and unstimulated cells using the Paxgene blood RNA kit ( Qiagen , Valencia , CA ) . The quality of the RNA was verified both with the DU 530 Life Science spectrophotometer ( Beckman , Fullerton , CA ) and Agillent Bioanalyzer . cDNA was prepared from both patient and healthy donor extracted RNA samples using a high capacity cDNA RT kit . ( Qiagen , Foster City , CA ) . Commercially available gene expression assays ( Applied Biosystems ) were used for amplification of the following transcripts; TNF-α ( Hs00174128_m1 ) , IL-1β ( Hs00174097_m1 ) , IL-6 ( Hs00985639_m1 ) , IL10 ( Hs00174086_m1Hs ) , IFN-β ( Hs00277188_s1 ) , TLR2 ( Hs00610101_m1 ) , TLR7 ( Hs00152971_m1 ) , TLR8 ( Hs00152972_m1 ) , TLR9 ( Hs00152973_m1 ) , CD40 ( Hs00374176_m1 ) , IL-17 ( Hs99999082_m1 ) and IFN-γ ( Hs00174143_m1 ) . qRT-PCR gene expression assays for the house keeping gene , glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) ( Hs99999905_m1 ) , were performed using identical aliquots of each cDNA as normalization controls . All amplification reactions were performed in triplicate; control reactions without reverse transcriptase also were performed to confirm the absence of contaminating DNA . Expression levels of all transcripts studied were normalized to the GAPDH level and the relative changes in gene expression were calculated using the 2−ΔΔCt method [44] . DNA from tissues ( 15–25 mg ) was extracted using the QIAamp DNA minikit ( QIAGEN Inc . , Valencia , CA ) following procedures recommended by the manufacturer . DNA was eluted from the QIAGEN columns in 100 µl of elution buffer at 70°C and stored at −80°C . The concentration of DNA was determined spectrophotometrically by the 260/280 nm absorbance . The quality and integrity of the DNA were determined by electrophoretic fractionation of 5 µl of extracted DNA through 1 . 2% agarose gels ( E-gels: Invitrogen Corp . , Carlsbad , CA ) at 70 V for 30 min . PCR amplification of the Tp polA gene was performed using forward primer TP-1 ( 5′CAGGATCCGGCATATGTCC3′ ) , reverse primer TP-2 ( 5′AAGTGTGAGCGTCTCATCATTCC3′ ) , and probe TP-3 ( 5′CTGTCATGCACCA GCTTCGACGTCTT3′ ) as previously published [45] , with some exceptions . The probe was labeled with Cyanine ( Cy5 ) at the 5′ end and black-hole quencher 3 ( BHQ3 ) at the 3′ end . Thermocycling was performed in a Rotor-Gene 6000 instrument ( Qiagen , Valencia , CA ) as follows: two hold cycles at 50°C for 2 min and 95°C for 10 min , respectively; and 45 cycles of 95°C for 15 sec and 60°C for 1 min . Each PCR run included positive and negative ( no template ) control reactions . The Tp copy numbers for each skin biopsy specimen were extrapolated from the standard curve generated using ten-fold serial dilutions of purified Tp DNA . The raw data obtained from the amplifications were adjusted for quantity tested to generate the polA DNA concentration , expressed as copies/ml or copies/µg of extracted cellular DNA from tissues . A 4-mm punch biopsy from SS skin lesions was obtained from a group of 12 patients and from normal skin from 3 healthy Colombian controls ( see above ) , snap-frozen and stored in liquid nitrogen in preparation for overnight transportation on dry ice from Cali to UCHC . Upon arrival at UCHC , tissues were homogenized in Trizol ( Invitrogen ) , RNA was isolated , cleaned with Turbo DNase ( Ambion , Applied Biosystems ) followed by cDNA synthesis using High Capacity cDNA Archive Kit ( Applied Biosystems ) according to the manufactures instructions . Gene transcripts were amplified per manufacturer's instructions ( Applied Biosystems ) using two commercially available array kits; TaqMan® Human Immune Array and TaqMan® Human Phagocytosis Array . Briefly , the array was performed in a 2 µL reaction volume containing 62 . 5 pg of cDNA , 1 µL of water and 1 uL of gene expression master mix; and the Phagocytosis Array was processed in a 20 µL reaction volume containing 5 ng of cDNA , 10 µL water and 10 µL Gene Expression Master Mix . Amplification reactions were performed with 7900HT Fast Real Time ( Applied Biosystems ) using the following conditions: 95°C for 20 min , and 40 cycles of 95°C for 1 s and 60°C for 20 sec . Expression levels of all transcripts studied were normalized to the GAPDH level and the relative changes in gene expression generated between 12 SS patients and three healthy controls were calculated with the 2−ΔΔCt method using DataAssist™ v2 . 0 Software ( Applied Biosystems ) . Up- or down-regulation of gene transcripts for this analysis were considered significant if their expression pattern in tissue was at least 2-fold higher or lower than control skin-samples and if the p-value was <0 . 05 . Selected gene transcripts were confirmed by conventional RT-PCR as described above . Paraffin embedded skin biopsies which were available from 4/12 patients studied by microarray ( see above ) , were immunohistochemically labeled with antibodies against CD4 , CD8 , CD56 , CD11c , CD14 and CD68 using an automated IHC staining platform ( Bond Max , Leica-Microsystems , Buffalo Grove , IL ) . IHC staining for Tp was manually performed at room temperature . Slides sections were depparaffinized in xylene ( Allegiance Healthcare Corporation , McGaw Park , IL ) and rehydrated in graded alcohol to water . After quenching endogenous peroxidase activity and a wash in phosphate-buffered saline ( PBS ) , the slides were incubated for 30 minutes at room temperature with a rabbit polyclonal anti-Tp antibody ( dilution 1∶500; Biocare , Concord , CA ) . Following a short wash in PBS , slides were covered with anti-rabbit EnVision+ detection system ( Dako , Carpinteria , CA ) for 30 minutes . Following a final wash in PBS , slides were incubated with the AEC+ ( aminoethylcarbazole ) chromogen ( Dako ) for 10 minutes , rinsed and counterstained in hematoxylin . Positive and negative controls were run in parallel for each of the antibodies used . For evaluation of immunofluorescence results biopsy specimens were read in a blinded fashion by at least one independent investigator . Labeled cells were enumerated per visual field and expressed as a percentage of inflammatory cells per 200 cells counted per high power field ( HPF ) . Immunologic markers of interest were first compared between a group of healthy controls affiliated by employment to the CIDEIM facility and healthy controls of similar socio-economic background to SS patients enrolled in the study . Student t test or the equivalent non-parametric methods ( i . e . Mann-Whitney U ) test results allowed us to conclude that immunologic parameters of interest between the two healthy control populations were not statistically different ( data not shown ) . Results from the combined control group were thus used for all comparisons between healthy volunteers and SS patients . Flow cytometry cell surface expression patterns of immunologic markers of interest and cytokine outputs were compared between patients and controls by using unpaired Student t tests or where indicated the equivalent non-parametric test ( i . e . Wilcoxon ) . A two-tail statistical analysis was performed for all comparisons , except to analyze dose-responses in the Tp-monocyte stimulation assay results . For each analysis , both the standard deviation and the standard error of the mean ( SEM ) were calculated and p values of <0 . 05 were considered significant . Statistical analysis was done using GraphPad prism 4 . 0 ( GraphPad Software , San Diego , CA ) .
We recently reported that a significant proportion of a cohort of untreated SS patients had low-level spirochetemia based on whole blood Tp-DNA quantitation [8] . This finding , together with our earlier demonstration that HSS induces opsonophagocytosis-dependent activation of monocytes and DCs in PBMCs [28] , prompted us to examine whether circulating monocytes and DCs obtained from these same SS patients exhibited evidence of immune activation at the time of initial presentation . Unlike healthy volunteers , at study entry more than half ( 7/13 ) of the patients studied exhibited increases in the size and granularity in the total monocyte population , which were no longer present two months after penicillin treatment ( Figure 1 ) . Expression of the activation marker CD40 and mean fluorescence average values for CD14 ( Figure 2 ) also were increased in monocytes from untreated SS patients; statistically significant post-treatment reductions confirmed that this finding was disease-specific . Despite these immunophenotypic alterations , there were no significant pre-treatment increases in selected cytokines ( TNF , IL-1β , IL-10 or IL-6 ) based on either qRT-PCR analysis of isolated PBMCs or CBA of pre-treatment sera ( data not shown ) . We also studied circulating DCs , using expression of CD11c [46] , [47] to distinguish monocytoid ( CD11c+ ) and plasmacytoid ( CD11c− ) DCs . As depicted in the representative flow cytometry dot plots in Figure 3 , 67% ( 10/15 ) of the patients studied exhibited a selective decrease in the proportion of CD11c+ DCs ( <35% ) , which corrected in all but three patients at the follow-up visit . Neither DC population displayed increased expression of the activation marker CD83 ( Figure 3 ) . NK-cells play a critical role in the immune response to human pathogens by secreting IFN-γ and other immunomodulatory molecules [48] , [49] and by promoting T-cell polarization and DC maturation [50] . The finding that NK-cells are the principal source of IFN-γ in Tp-stimulated PBMCs [28] , together with existing evidence that total NK-cell numbers and function may be altered during SS [51] , [52] , prompted us to study circulating NK-cell subsets in the blood of our patients . NK-cells were classified by flow cytometry using a previously described scheme according to their relative expression of CD16 and CD56 , ( Figure 4A ) [53] , [54] . CD56bright cells are known to be potent cytokine producers with limited cytotoxic activity , while CD56dim cells have strong cytotoxic capacity but a decreased ability to produce cytokines [53] , [54] . As displayed in Figure 4B–C , when compared to healthy controls , a significantly greater percentage of SS patients had total circulating NK-cell values below the 5th percentile of published normal adult NK-cell numbers [55] ( 40% vs . 4 . 3% respectively , p = 0 . 01 ) . Significant decreases in IFN-γ-producing ( CD56bright ) and cytotoxic ( CD56dimCD16bright ) NK-cell subsets were largely responsible for the decline in total NK-cell values ( Figure 5 ) . By contrast the CD56negativeCD16bright NK-cell subset , a recently described NK-cell population which exhibits both poor cytolytic activity and impaired cytokine production [41] , was significantly increased in most untreated SS patients ( Figure 6 ) but returned to near normal values in all patients at the follow-up visit . We previously reported [8] that the routine histology for the patients described in this study was characteristic of typical SS lesions [2] , [5] , [32] , [33] , [56] , [57] . Herein , we used IHC staining techniques to analyze four SS skin biopsies and corroborate that the cellular infiltrates were in agreement with previously published IHC analysis [32] , [38] , [56] , [58] and to explore potential mechanisms for immune recognition of spirochetes within tissues . Substantial numbers of dermal mononuclear cells expressed the macrophage marker CD68 ( Table S3 and Figure 7A and B ) [59] . Staining with two other macrophage markers ( CD11c and CD14 ) revealed very similar patterns ( data not shown ) . Syphilitic lesions were also comprised of CD4+ and CD8+ lymphocytes ( Figure 7B–E ) , with the CD8+ phenotype predominating in three of the four biopsies [36] ( Table S1 ) . Interestingly , 5% of dermal mononuclear cells expressed the NK-cell marker CD56+ , an approximate five-fold increase from normal percentages ( <1% ) of NK-cell values in healthy skin [48] ( Figure 7F and G and Table S1 ) . Because of the lack of specific markers for CD56negativeCD16high NK-cells , we were unable to determine if this unexpected circulating NK-cell phenotype was also present in the skin of SS patients . Several recent studies have also called attention to the sensitivity of IHC for detection of spirochetes in tissues in addition to its well-recognized ability to provide information regarding the spatial relations between Tp and cellular infiltrates in the skin [2] , [5] , [32] , [33] , [57] . In agreement with these prior reports , dense clusters of spirochetes could be seen in a perivascular location within the papillary dermis in close physical proximity to aggregates of lymphocytes and histiocytes ( Figure 8 ) . Spirochetes were also visualized in the mid- and deep-layers of the dermis away from infiltrating cells , straddling the dermal-epidermal interface , and within the lower layers of the epidermis ( data not shown ) . Lastly , we confirmed that all four biopsies studied had Tp DNA by quantitative PCR analyses ( Table S1 ) . Transcriptional analysis of SS skin biopsies has thus far been limited to a small number of gene products [32] , [38] . In this study , we used transcriptional profiling to gain additional insights into the molecular mechanisms underlying the inflammatory responses elicited by spirochetes in skin . Table 3 highlights key transcripts associated with the array , while the complete list is presented in Tables S2 and S3 . Consistent with the finding by IHC that syphilitic lesions contain an abundance of macrophages , the transcript for CD68 [59] was significantly up-regulated . Also upregulated were transcripts for the macrophage activation markers CD40 [60] , CD80 and CD86 [61] , a number of cytokines known to be secreted by human monocytes/macrophages in response to opsonized Tp ( TNFα , IL-6 , IL-1β and IL-10 ) , and numerous other molecules associated with macrophage activation . Interestingly , transcripts for both TLR1 and TLR2 , which are required for recognition of treponemal lipoproteins by monocytes/macrophages [28] , were up-regulated , whereas TLR6 , which recognizes diacylated lipoproteins in association with TLR1 [62] , was not . Transcripts for three different FCγ phagocytic receptors , FCγR1A/C ( CD64 ) , FCγR2A ( CD32 ) , and FCγR3A/B ( CD16 ) also were significantly over-expressed in lesional skin . Transcripts for the T-cell receptors CD3 , CD4 and CD8 , as well as the T-cell activation marker CD38 , were all markedly increased in lesional biopsies ( Table 3 ) . IFNγ , a potent macrophage activator that can be produced by CD4+ and CD8+ memory T-cells as well as NK-cells , also was significantly up-regulated in lesional biopsies . The transcript for IL-17 also was expressed in the lesional biopsies , which is in agreement with a recent report that IL-17+ T-cells are present in the skin of SS patients [38] . Of particular interest , we saw a dramatic increase in expression of transcripts for granulysin , perforin and granzyme B , which can be produced by both NK-cells [63] and CD8+ T-cells [64] , thus , providing evidence for a strong cytotoxic response . Type I IFNs modulate multiple aspects of innate and adaptive immunity in response to bacterial infections [23] , [65] , [66] , including activation of NK-cells , DCs and macrophages . The arrays revealed marked increases in the expression of three endosomal TLRs , TLR-7 , TLR-8 and TLR-9 [67] , all of which are associated with the production of type I IFNs [65] , [68] . TLR7 and TLR9 are expressed predominantly by plasmacytoid DCs , a subset which we previously have shown is enriched in SS skin lesions [69] , whereas TLR8 is expressed by activated human macrophages [70] . Two type I IFN-inducible chemokines , CXCL10 ( IP-10 ) and CXCL11 ( IP-9 ) , were significantly over-expressed in lesional biopsies . In line with these results , we found by RT-PCR that the transcript for IFN-β was also up-regulated in lesional biopsies ( data not shown ) . Syphilitic antibodies are believed to play an essential role in both cellular activation and bacterial clearance by promoting opsonophagocytosis of the syphilis spirochete by macrophages [40] , [57] , [71] , [72] . Herein , we used an ex vivo stimulation assay to model the effect of opsonic antibodies on spirochete-monocyte/macrophage interactions at graded MOIs as it may occur in the skin and blood of patients . In comparison to normal human sera , pooled syphilitic sera significantly enhanced monocyte uptake of Tp in a dose-dependent manner ( Figure 9A ) . Similar results were obtained for serum specimens of 12 different patients confirming that opsonic antibodies are commonly produced in early syphilis ( data not shown ) . Interestingly , even in the presence of HSS a large proportion ( 56% ) of the spirochetes was not phagocytosed ( Figure 9B ) . The observation that the percentage of spirochetes recovered was not significantly different between the MOIs of 1 , 10 and 30 ( Figure 9C ) , argues that the lack of uptake at the higher MOI is not due to FC-receptor saturation . As shown in Figure 10 , opsonized Tp also induced a marked dose-dependent increase in secretion of TNF and IL-1β . Importantly , cytokine production was minimal at the lowest MOI ( 1∶1 ) tested , which is similar to the MOI in spirochetemic SS patients ( Table 1 ) . A similar dose-dependent increase was seen in the production of Il-6 and IL-10 and expression of the activation markers CD40 and CD83 ( data not shown ) .
Venereal syphilis can be considered a contest between the ability of T . pallidum to avoid immune recognition and the adeptness of the host's innate and adaptive immune responses to “track down” and eliminate the spirochetal pathogen . To begin to understand the mechanisms that underlie the dichotomy between immune evasion and immune recognition of the syphilis bacterium , herein we compared key aspects of the innate and adaptive immune response in the blood and skin of SS patients , to spirochetal burdens present in these two immunologically distinct compartments . The evidence suggests that spirochetes circulate through the blood mostly unimpeded by host's immune defenses , while the larger burden of treponemes present in the skin elicit a highly complex inflammatory cellular immune response that paradoxically does not rapidly control spirochetal replication . Our results reinforce the importance of the macrophage in the immune response to Tp and establish that the balance between phagocytic uptake of the spirochete and its ability to evade innate immune recognition is influenced by the number of bacteria present in either the blood or the skin , as well as the emergence of Tp-subpopulations with differential capacities for binding opsonic antibodies . We hypothesize that the striking immunophenotypic alterations found in circulating innate immune cells in SS patients are not the result of their direct interaction with spirochetes in the blood , but instead a manifestation of the systemic effects of the bacterium in other tissues including the bone marrow . Lastly , we demonstrate that in addition to CD4+ and CD8+ T-cells , CD56+ NK-cells are enriched in Tp-infected skin lesions and , thus , could contribute to macrophage activation and bacterial clearance through their ability to secrete IFN-γ . One of the most significant findings in this study is the extent of the immunophenotypic alterations that distinguish monocytes , DCs and NK-cells in the blood of untreated SS patients from those of healthy controls . Despite the demonstration that circulating monocytes were noticeably larger by flow cytometry and expressed higher levels of CD14 and CD40 than monocytes obtained from healthy volunteers , SS patients did not have measurable increases in transcription or secretion of monocyte-derived cytokines in their blood . The ex vivo finding that bacterial uptake and cytokine production was minimal at low spirochete-monocyte ratios ( 1∶1 ) , an MOI that closely mirrors the calculated MOI in the blood of SS patients , may explain why low levels of spirochetes in the blood are unable to induce cytokine production by circulating monocytes . Given that similar morphologic changes were not evident ex vivo Tp-stimulated monocytes at a similar spirochete to cell ratio of 1∶1 ( data not shown ) , leads us to conclude that the immunophenotypic alterations in circulating monocytes could be provoked directly by the spirochete in the blood . Instead they raise the possibility that the syphilis spirochete could be affecting macrophage-DC progenitor cells in the bone marrow , before they differentiate into CD14+ monocytes and mobilize into the blood stream [73] . The striking decrease in circulating CD11c+ DCs suggests that these cells are marshaled from the blood into infected skin tissues [74] . Our prior findings that monocytoid DCs obtained from the blood and skin of SS patients express high levels of the C-type lectin DC-SIGN [43] , an adhesion molecule which is known to regulate DC trafficking from blood into tissues [75] , supports the notion that this DC subset migrates into infected skin . A particularly novel finding in this study was the marked decrease in total circulating NK-cell numbers and the distinct emergence of a highly atypical circulating CD56negativeCD16bright NK-cell population . Mavilio and colleagues [41] , [42] previously reported similar increases in HIV-infected patients with uncontrolled viremia and confirmed that this subset of NK-cells was not only poorly cytolytic but also had an impaired capacity to produce IFN-γ and other cytokines . Our report is the first to show that this abnormal NK-cell subset can also be increased in the course of a bacterial infection . The critical role of the macrophage in the pathogenesis of venereal syphilis was initially ascertained from histological analysis of Tp-infected rabbit tissues [76] , [77] and the finding that rabbit immune sera markedly enhanced spirochetal uptake and clearance by peritoneal macrophages in vitro [71] . Results from prior human studies are generally in line with those in the rabbit model in that they confirm that large numbers of macrophages and T-cells are also present in early syphilis lesions [36] , [37] , [56] , [78]–[80] . Herein , we used a combination of IHC and transcriptional profiling to corroborate that macrophages are indeed the predominant inflammatory cell in the skin . The confirmation that HSS enhances uptake of spirochetes by isolated monocytes ex vivo , inducing their activation in a dose-dependent manner , underscores the importance of opsonophagocytosis in spirochetal recognition and clearance . While low-level spirochetemia seemingly facilitates chronic spread of the bacterium , in Tp-rich skin infiltrates opsonized spirochetes are more likely to be taken up by IFN-γ activated tissue macrophages . Paradoxically , the ex vivo model results also indicate that even at high MOIs , a large subset of the spirochetes avoid phagocytosis by monocytes . This finding is in accord with the observation in the rabbit model by Lukehart and co-workers [27] that a subpopulation of opsonic antibody-resistant spirochetes emerges during active infection . The same group has proposed that antigenic variation in candidate OMP antigenic targets ( i . e . TprK ) helps us understant how Tp evades host antibody responses [81] . Our own data suggests that an additional explanation is that Tp populations differ widely with respect to the density of surface antigens recognized by syphilitic serum [25] , which would then allow populations of spirochetes to escape opsonization and avoid clearance . The role of the adaptive cellular immune response in treponemal clearance has been studied in the rabbit model and in humans with active disease [2] . Replication of treponemes at the site of inoculation in rabbit tissues elicits an intense inflammatory response that histologically resembles a classic delayed type hypersensitivity reaction ( DTH ) [56] , which in addition to macrophages is composed predominantly of CD4+ lymphocytes [34] . IHC and molecular studies in humans confirm that primary and SS lesions are also enriched for Th1-cytokine producing CD4+ lymphocytes [32] , [33] . In contrast to the rabbit , however , and in support of the findings in this study , CD8+ T-cells are often the predominant T-cell immunophenotype in SS lesions [37] , [38] , [78] . Our prior demonstration that CD4+ and CD8+ T-cells in blister fluid elicited over SS lesions are predominantly of the memory and memory effector immunophenotype , expressing the activation marker CD38 [43] , can be interpreted as an indication that populations of T-cell subsets in the skin of patients are antigen-specific . While naïve CD4+ T-cells are primed in the lymph-nodes by treponemal antigens via MHC-class II pathways [82] , CD8+ T-cells will require some form of cross-presentation of spirochetal peptides via MHC-class I molecules [82] , [83] . One plausible explanation for how cross-presentation might occur is that treponemal constituents enter alternative endocytic pathways in DCs or macrophages , allowing bacterial peptides to bind to MHC-class I molecules in the endoplasmic reticulum [82] . It is interesting to note , in this regard , that Bouis et al . [84] showed that DCs ingest virulent treponemes by coiling phagocytosis , an uptake mechanism that has been associated with cross-presentation [85] . An additional reason for this phenomenon is that circulating spirochetes could be internalized and cross-presented directly by lymph node-resident DCs; a highly specialized DC population which in humans can cross-present antigen without activation [86] . Because NK-cells promote the development of adaptive immunity via a bi-directional cross-talk between naïve CD4+ T-cells and DCs [42] , [53] , one could envision a model where alterations in circulating NK-cell populations , as shown herein , could interfere with adequate antigen presentation to CD4+ T-cells in the lymph nodes . Cross-priming of naïve CD8+ T-cells in SS could , thus , serve as a compensatory mechanism for less than optimal CD4+ T-cell priming in the lymph nodes of SS patients . How CD8+ T cells are activated in the skin is not entirely clear since this subset typically responds to intracellular bacterial pathogens [87] , [88] . Perhaps CD8+ T-cells are required to eliminate intracellular reservoirs of the bacterium that may be present in non-phagocytic cells from early syphilis lesions [89]–[91] . Then again , treponemal antigens could be cross-presented to cytolytic T-cells in the skin by tissue based macrophages and/or DCs inducing their activation . Given that cytokine producing CD56+ NK-cells were also enriched in the skin of patients , it is also plausible that this innate immune lymphocyte provides an additional source of IFN-γ in Tp-infected tissues . In support of this idea , we previously demonstrated that in Tp-stimulated PBMCs , NK-cells are a major source of IFN-γ [53] and showed that production of this cytokine is dependent on the presence of accessory cells ( i . e . , DCs ) . Lastly , in agreement with a recent report that IFN-γ and IL-17 producing CD8+ T-cells are present in the skin of SS patients [38] , the transcript for IL-17 also was up-regulated in SS biopsies [38] . It is conceivable that IL-17 producing T-cells play an important compensatory role in SS , particularly in HIV-syphilis co-infected patients with very low CD4+ T-cell counts . Although type I IFNs have generally been associated with antiviral immune responses , there is now compelling evidence that these cytokines also are induced in response to several intracellular [68] , [92]–[95] and extracellular bacteria [23] , [96] , [97] . It was , therefore , not at all surprising that IFN-β and several type I IFN associated transcripts ( Table 3 ) were up regulated in SS lesions . A variety of ligands , including bacterial DNA and RNA , can activate PRRs present in either the cell cytosol or membrane bound TLRs [68] to induce type I IFNs . In this regard , we recently provided evidence that transcription of IFN-β in human monocytes stimulated ex vivo with Borrelia burgdorferi , the Lyme disease spirochete , was dependent on phagocytosis and degradation of the bacterium and required signaling through TLR8 [23] . In support of a similar role for phagosomal signaling in Tp-mediated induction of type I INFs , three endosomal TLRs ( TLR7 , 8 and 9 ) , capable of sensing nucleic acids within phagosomes of macrophages and DCs [98]–[100] , also were markedly up-regulated in syphilis lesions . Type I IFNs are likely to have several important roles in the immune response to the spirochete . Firstly , type I IFNs can induce the differentiation of plasmacytoid DCs into mature antigen presenting DCs through their ability to up-regulate surface expression of co-stimulatory molecules like CD80 , CD86 , and CD40 [68] . Type I IFNs could also help regulate NK-cell function by inducing the production of IL-15 by macrophages , a cytokine which was increased in SS lesions and can promote NK-cell survival and proliferation [101] . Lastly , type I IFNs may facilitate cross-presentation of antigens via MHC class I molecules to CD8+ T-cells [65] . Based upon the findings from this and prior studies [43] , we now propose a revised early syphilis pathogenesis model that integrates innate and adaptive immune responses to the bacterium and also takes into account the spirochete's immunoevasive countermeasures against host defenses . According to this model , spirochetes replicate at the site of initial inoculation unchecked by the innate immune surveillance system and rapidly disseminate to the skin and other tissues . At some point after initial entry of the bacterium increasing local spirochetal burdens allow a small number of organisms to be taken up by resident phagocytes , although this process is inefficient in the absence of opsonic antibodies . APCs containing phagocytosed spirochetes can then migrate into draining lymph nodes where they present treponemal antigens to naïve CD4+ T cells and B-cells . We postulate that neo-sensitized T-helper cells traffic back into the primary lesion , where they recognize their cognate antigens and release IFN-γ . Clearance of organisms by IFN-γ activated tissue macrophages is markedly facilitated by the emergence of high titers of Tp-specific opsonic antibodies . In parallel events , while the chancre resolves , as soon as treponemal loads in the skin of early syphilis patients reach a sufficient density capable of triggering the local inflammatory response , SS skin lesions become clinically apparent . In contrast to the immunologic events that initially take place in the primary chancres , innate and adaptive immune responses in SS skin lesions appear to co-evolve in the presence of both memory and memory effector CD4+ and CD8+ T cells and high titers of opsonic antibodies . One would thus predict that these changes would be sufficient for the immune response to eradicate the bacterium . However , the paucity of OMP antigenic targets on the outer leaflet of the bacterium together with the emergence of Tp-subpopulations resistant to opsonophagocytosis , permits varying numbers of bacteria to avoid opsonization , uptake and clearance by dermal macrophages . The low-level bacteremia which ensues allows the spirochete to avoid recognition by host innate and adaptive immune defenses in the blood compartment . The constant spread of Tp into other tissues during SS , specifically the bone marrow , could affect the development of myeloid and lymphoid progenitors of monocytes/macrophages , DCs and NK-cells . Fortunately for the host , over time , the emergence of greater numbers of activated memory and memory effector CD4+ and CD8+ T-cells , IFN-γ producing CD56+ NK-cells together with increasing titers of Tp-specific opsonic antibodies , allows the host to ultimately gain the upper hand against the bacterium . The complex shifting balance between immune evasion and bacterial persistence to immune recognition and spirochetal clearance , will thus , not only determine the intensity and duration of the clinical manifestations of venereal syphilis but also how long the spirochete can endure in blood and tissues . | Syphilis , a sexually transmitted disease caused by the spirochetal bacterium Treponema pallidum , affects close to 10 million people per year worldwide . Despite the robust nature of the humoral and cellular immune responses associated with the disease , weeks to months may elapse before the host gains control of the infection . Moreover , in the absence of antibiotic treatment , containment is often incomplete and relapses are common . Herein we studied aspects of the immune response in the blood and skin of patients with secondary syphilis to better understand the factors that determine whether the bacterium evades host defenses or is cleared in its natural human host . Our findings support the importance of the macrophage as a primary means of bacterial killing in the skin , while suggesting that the extent of bacterial clearance is determined by the bacterial loads present in either the blood or skin of patients and the appearance of spirochetes which are resistant to uptake ( phagocytosis ) by the macrophages . Study results underscore the extent of the systemic immunologic abnormalities triggered by the bacterium and provide new insights regarding the complexity of the immune response in the skin of untreated patients . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"infectious",
"diseases",
"neglected",
"tropical",
"diseases",
"syphilis"
]
| 2012 | Immune Evasion and Recognition of the Syphilis Spirochete in Blood and Skin of Secondary Syphilis Patients: Two Immunologically Distinct Compartments |
Sumoylation , the covalent attachment of SUMO ( Small Ubiquitin-Like Modifier ) to proteins , differs from other Ubl ( Ubiquitin-like ) pathways . In sumoylation , E2 ligase Ubc9 can function without E3 enzymes , albeit with lower reaction efficiency . Here , we study the mechanism through which E3 ligase RanBP2 triggers target recognition and catalysis by E2 Ubc9 . Two mechanisms were proposed for sumoylation . While in both the first step involves Ubc9 conjugation to SUMO , the subsequent sequence of events differs: in the first E2-SUMO forms a complex with the target and E3 , followed by SUMO transfer to the target . In the second , Ubc9-SUMO binds to the target and facilitates SUMO transfer without E3 . Using dynamic correlations obtained from explicit solvent molecular dynamic simulations we illustrate the key roles played by allostery in both mechanisms . Pre-existence of conformational states explains the experimental observations that sumoylation can occur without E3 , even though at a reduced rate . Furthermore , we propose a mechanism for enhancement of sumoylation by E3 . Analysis of the conformational ensembles of the complex of E2 conjugated to SUMO illustrates that the E2 enzyme is already largely pre-organized for target binding and catalysis; E3 binding shifts the equilibrium and enhances these pre-existing populations . We further observe that E3 binding regulates allosterically the key residues in E2 , Ubc9 Asp100/Lys101 E2 , for the target recognition .
Protein function is regulated by numerous mechanisms , one of which is post-translational modification . Covalent binding of ubiquitin ( Ub ) and ubiquitin-like ( Ubl ) modifiers to target proteins constitute a key step in cellular processes including differentiation , apoptosis , cell cycle , and stress response [1]–[4] . Here , we focus on one member of the Ubl super-family , SUMO , with the aim of figuring out the mechanism through which SUMO is conjugated to its target proteins . SUMO-1 ( Small ubiquitin-like modifier , also known as PIC1 , UBL1 , GMP1 , Sentrin ) , -2 , -3 and -4 exist in mammals [5]–[10] . Sumoylation can change the proteins' intracellular localization , interaction patterns with other proteins and modifications by other post-translational events . It is important in development [11] and is related to cancer drug resistance [12] , [13] . For simplicity , below , SUMO refers to SUMO-1 . At least 100 different proteins have been reported as targets for sumoylation [14]–[17] . Analogous to conjugation mechanisms of Ub/Ubls , SUMO is attached to target proteins following sequential activation by E1 , E2 and in most cases , E3 enzymes [18] . Following activation of the SUMO precursor [4] , the E1 enzyme Aos1/Uba2 and SUMO form a thioester bond . The SUMO thioester is next transferred to the active cysteine of Ubc9 , the single known E2 enzyme of the sumoylation pathway [1] , [4] , [18] . Then SUMO is transferred from E2 to a target protein lysine residue . E3 enzymes that ensure target specificity and increase reaction efficiency usually mediate this step ( Figure 1 ) . Among the sumoylation targets , RanGAP1 , p53 and IκBα are modified without an E3 ligase in vitro , although the reaction rates are slower compared to E3-mediated conjugation [1] . E2 ligase Ubc9 is essential [1] , [19] and conserved [1] . It recognizes a consensus sumoylation motif , “Ψ-K-x-D/E” , where Ψ represents a hydrophobic residue , K is the SUMO acceptor lysine , x is any amino acid and D/E is an acidic residue [4] . The E2 ligase also interacts with E3 enzymes during the transfer of SUMO to targets [20] . In addition to the consensus sumoylation motif , sumoylation target RanGAP1 has a second contact surface with the E2 ligase Ubc9 , which is thought to be responsible for the higher efficiency of modification compared to other substrates [4] . A fragment of the E3 enzyme RanBP2 , consisting of the IR1-M-IR2 domains is sufficient for E3 activity in vivo and in vitro [18] . Moreover , IR1-M and M-IR2 constructs are also functional with IR1-M being the catalytic core domain [20]–[22] . The activity of the E3 fragment indicates that E3 exerts its catalytic effect by altering the structural properties of the E2-SUMO complex , increasing the affinity of the complex for specific protein targets , rather than by forming direct target interactions [21] . The crystal structure of the SUMO-RanGAP1-Ubc9-RanBP2 complex supports this idea [20] . Recent work also shows that E3 ligase RanBP2 prevents dissociation of SUMO from its target RanGAP1 , leading to an increase in the sumoylated RanGAP1 levels [23] . Due to the strong interactions between RanGAP1 and E2 , it has been a debated question whether RanBP2 exerts its E3 activity for RanGAP1 or whether it only maintains the complex at the nuclear pore complex ( NPC ) [4] , [20] . Our aim is to understand the mechanism through which the E3 ligase RanBP2 triggers target recognition and catalysis by E2 in sumoylation . We carried out explicit solvent molecular dynamic simulations for the E2 ligase Ubc9 , SUMO , and the E2-SUMO complex with and without the E3 enzyme RanBP2 . We modeled the conjugated E2-SUMO complex , in RanBP2 bound and unbound forms , based on the SUMO-RanGAP1-Ubc9-RanBP2 crystal structure ( Figure 2 ) . Our results indicate that E3 binding induces a higher population of target binding and catalysis-ready E2 , restricting the conformational space of the E2-SUMO complex . We observe that RanBP2 binding enhances the correlations between the fluctuations of E2 residues involved in catalytic activity and target recognition , which implies that RanBP2 is indeed an E3 ligase for the sumoylation of the target protein RanGAP1 . Our results further lead us to propose that the mechanism through which E3 ligase RanBP2 triggers E2 target recognition and catalysis in sumoylation is allostery: RanBP2 is an allosteric effector of E2 ligase Ubc9 . Below , we refer to the specific proteins simulated ( Ubc9 , RanGAP1 , RanBP2 ) rather than the protein functional class ( E2 , target protein , E3 , respectively ) to which they belong . These were the ones crystallized by Reverter and Lima [20] .
When simulated without RanBP2 , the Ubc9-SUMO complex structure displays a significant deviation from its crystal structure . The two representative conformations from the clustering analysis of the sampled conformational space of the complexes display the change in the quaternary structure of the Ubc9-SUMO complex ( Figure 3 ) . The Ubc9-SUMO rotates and moves away from its position in the crystal structure ( Figure 3 ) . Accompanying the orientation change of SUMO , there is a minor re-organization of the hydrogen bond network in the catalytic area ( Figure S1 ) . The rmsd ( root mean square deviation ) of the Ubc9-SUMO complex shows that the deviation is more dramatic between 5–12 ns and stabilizes at the end of this period ( Figure S2A ) . Yet , the monomers do not show increased deviations from their initial structures ( Figure S2B , Figure S2C ) . This indicates that the rmsd increase of the complex structure originates from a change in the relative positions of the chains with respect to each other . On the other hand , with RanBP2 , SUMO does not move or rotate but fluctuates around its original crystal conformation . For a more quantitative measure of this orientation change , we utilize the representative structures from the clustering analysis . We carried out rmsd calculations for Ubc9 and SUMO , with alignments of Ubc9-SUMO and Ubc9-SUMO-RanBP2 complexes on the individual proteins . The results are listed in Table 1 . As expected , when the proteins are aligned on the corresponding chains in the complex structure , the rmsd values are low; however , they show some increase when the complex is aligned on the other chain . The rmsd values for SUMO and Ubc9 are not significantly different for Ubc9-SUMO and Ubc9-SUMO-RanBP2 simulations when each is aligned on the corresponding chain in the complex . When the structures are aligned with Ubc9 as the pivot , the SUMO rmsd in the Ubc9-SUMO-RanBP2 complex increases up to less than 2 fold ( representative structure 3 ) , while the SUMO rmsd in the Ubc9-SUMO complex increases more than 4 fold ( representative structure 4 ) . This implies an overall quaternary structure change; that is , an orientation change of SUMO in the Ubc9-SUMO complex , which is not observed in the Ubc9-SUMO-RanBP2 complex . An additional set of simulations further validate this major orientation change ( Table S1 ) . Along with the limitation of SUMO orientation , E3 binding restricts the conformational space of Ubc9 . We combined the Ubc9 conformations from the simulations of the Ubc9-SUMO and Ubc9-SUMO-RanBP2 complexes , eliminating the initial 10 ns from each simulation , and clustering the remaining conformations . In the three resulting clusters , nearly all the Ubc9 conformations from the Ubc9-SUMO-RanBP2 complex are in one cluster , and the Ubc9 conformations from the Ubc9-SUMO complex are distributed among all three clusters . The distributions of the Ubc9 conformations are given in Table 2 . The time distribution of cluster members displayed in Figure 4D shows that Ubc9 from the Ubc9-SUMO complex samples conformations from the Ubc9-SUMO-RanBP2 complex throughout the whole simulation time . The distribution of the conformations among the clusters shows that RanBP2 binding restricts the conformational space of Ubc9 . To further validate this restriction , we projected the Ubc9 conformational space from the Ubc9-SUMO and Ubc9-SUMO-RanBP2 complexes on the principal components ( Figure 4A–C , Figure S4 ) . The projections and the clustering analysis demonstrate the restriction of Ubc9 conformational space upon RanBP2 binding . The mean-square fluctuations ( msf ) of the proteins in their unbound states ( Ubc9 and SUMO only ) , and in the complexes are given in Figure 5 . Ubc9 Cys93 is the active cysteine , and residues from Asn124-Pro128 are part of the loop region that is in contact with the tetrapeptide motif of the sumoylation targets [4] . The fluctuations of Cys93 and Asn124-Pro128 are restricted in both Ubc9-SUMO and Ubc9-SUMO-RanBP2 complexes , as compared to the unbound state of Ubc9 ( Figure 5 ) . Thus , the catalytic residue and the residues maintaining the structural integrity around the catalytic residue of Ubc9 already display a reduced mobility in the Ubc9-SUMO complex . In terms of reduced conformational states of the catalytic region , Cys93 and Asn124-Pro128 , the RanBP2 binding does not stabilize further these regions . This reduction in Ubc9 mobility is a direct result of SUMO binding . Ubc9 residues Val27-Met39 comprise the loop between β-sheets that serve as RanBP2 binding sites [20] . These residues already display high fluctuations in the isolated state , but their mobilities are allosterically further enhanced by SUMO binding . RanBP2 binding reduces the fluctuations to the values in the unbound state of Ubc9 ( Figure 5 ) . These residues also show cooperative fluctuations with RanBP2 binding sites on SUMO ( discussed below ) . The mobility of residues Glu98-Asp102 of Ubc9 increase upon SUMO binding , and again decrease upon RanBP2 binding ( Figure 5 ) . Residues Asp100 and Lys101 take part in target recognition , interacting with the approaching RanGAP1 [24] , [25] . RanBP2 does not have direct contacts with this loop . The fluctuations of Asp100 and Lys101 in the Ubc9-SUMO-RanBP2 complex are lower than in the unbound Ubc9 , whereas their fluctuations in the Ubc9-SUMO complex are higher than in the unbound Ubc9 . This suggests that the restricted fluctuations of Asp100 and Lys101 are the outcome of E3 binding . On the whole , the reduced mobility of this region upon RanBP2 binding is consistent with the proposed allosteric effect of RanBP2 on the Ubc9 target recognition [20] , [21] . Although the fluctuations of some SUMO residues demonstrate dramatic changes amongst all three ( the two complex and one isolated ) states , most of these do not coincide with functional residues . The high mobility of the unbound SUMO structure ( see Text S1 ) necessitates further experimental evidence . The time delayed auto-correlations of the backbone bond vectors is a measure of their orientational freedom . They provide information on the time dependent changes in the orientations of the backbone bonds . Here , the backbone bond vector refers to the virtual bond vector between two successive Cα atoms ( see Methods , Text S1 ) . A backbone bond vector will be closely correlated with the same vector calculated after a short time interval . As the time delay increases , the correlation between the vectors will decrease . High auto-correlation of a backbone bond vector indicates a restricted orientational freedom for the backbone bond . In the Ubc9-SUMO-RanBP2 complex , the correlations of backbone bonds are not lost with time delays as high as 30 ns , with almost all virtual bond vectors having auto-correlations above 0 . 9 , unlike the Ubc9-SUMO complex ( Figure 6 ) . This indicates that RanBP2 binding restricts the orientational freedom of Ubc9-SUMO residues . Upon RanBP2 binding , the backbone vectors between Glu99-Asp100 and Asp100-Lys101 of Ubc9 display the most significant restraint in their orientational behavior ( Figure 6 ) , particularly compared to the other loops of Ubc9 . Residues Asp100-Lys101 are not in close vicinity to the RanBP2 binding regions or the catalytic region of Ubc9 . The average distances of Asp100 to Cys93 throughout the Ubc9-SUMO and Ubc9-SUMO-RanBP2 simulations are 13 . 25 Å and 12 . 82 Å , respectively . Similarly , the average distances between Lys101 and Cys93 throughout the Ubc9-SUMO and Ubc9-SUMO-RanBP2 simulations are 11 . 54 Å and 9 . 83 Å , respectively . Asp100 and Lys101 of Ubc9 are known to be important for target recognition and functional defects have been observed for Ubc9 upon mutations of these residues [24] , [25] . Restriction of the mobility and orientational freedom of these residues upon RanBP2 binding can hinder a pre-organization of the target binding site on Ubc9-SUMO leading to a shift of the conformational ensemble [26] of Ubc9 . Since these residues are far from the RanBP2 binding site on Ubc9 , the rigidification of the 99–102 loop on Ubc9 is allosterically induced by RanBP2 binding . The Ubc9 residues Asn121-Ala131 interact with the consensus sumoylation motif and Glu132-Arg141 are important for specific RanGAP1 target binding [4] , [25] , [27] . The orientational freedom of the bond vectors Pro128-Ala129 , Gln130-Ala131 , Glu132-Ala133 and Ala133-Tyr134 are significantly reduced when bound to RanBP2 . Yet , the bond vectors Asp127-Pro128 and Ala129-Gln130 already have a restricted orientational freedom in Ubc9-SUMO , together with the low msf ( Figure 5 ) . Nevertheless , RanBP2 allosterically affects the dynamics of the Ubc9 regions that interact with RanGAP1 and facilitates the RanGAP1 recognition . Other than residues around Asp100 , Figure 6 points to altered orientational freedom in the Ubc9 region around Lys30 . This region is part of the Val27-Met39 loop , which is highly mobile in the Ubc9-SUMO complex and displays a reduced mobility upon RanBP2 binding . The Ubc9-SUMO-RanBP2 complex has correlated fluctuations between Ubc9 regions His83-Ser89 and Asn121-Arg141 . Residues His83-Ser89 include the HPN ( His83-Pro84-Asn85 ) motif that has a structural role , maintaining the hydrogen-bonding networks around the catalytic site of Ubc9 and assisting in orienting the SUMO C-terminal Gly-Gly motif [4] . Tyr87 is in contact distance with the sumoylation motif . Ubc9 residues Asn121-Ala131 interact with the sumoylation motif of the targets , and Glu132-Arg141 play a role in the recognition of the sumoylation target RanGAP1 [4] , [25] , [27] . Mutations of Glu132 and Tyr134 reduce the efficiency of RanGAP1 sumoylation; whereas mutations of Asn85 and Tyr87 reduce the sumoylation efficiency for all targets [4] , [25] , [27] . The correlated fluctuations observed in the Ubc9-SUMO-RanBP2 complex between residues His83-Ser89 and Asn121-Ala131 are still preserved in the Ubc9-SUMO complex . However , in contrast to the stable correlations in Ubc9-SUMO-RanBP2 , in the absence of RanBP2 , the correlations between residues His83-Ser89 and the additional Glu132-Arg141 binding surface , fluctuate ( Figure 7 , Figure S3 ) . Analysis of the Ubc9-SUMO trajectory over the time windows suggested by the clustering ( see Methods , Text S1 , Figure S2A and Table S2 ) , shows that the correlation between His83-Ser89 and the Glu132-Arg141 is lost between 12–24 ns ( 20 . 9 percent of simulation time ) , and still weak between 24–31 ns ( 12 . 2 percent of simulation time ) with respect to the Ubc9-SUMO-RanBP2 ( Figure S3 ) . Coinciding with the orientation change of SUMO , the correlations between the fluctuations of the Ubc9 region including the mobile loop , Val27-Glu42 , and the rest of the Ubc9 residues , become more negative . Furthermore , the same region , Val27-Glu42 , displays cooperative fluctuations with SUMO residues Phe36-Leu47 and Asp73-Ile88 . The two regions of SUMO are either on the β-sheets packed against RanBP2 or in the vicinity of these β-sheets in RanBP2 binding site in SUMO . Upon RanBP2 binding these correlations become more prevalent . The loop in Ubc9 and those SUMO regions are spatially far away . When using the centers of mass , the average distance between Val27-Glu42 loop of Ubc9 and Phe36-Leu47 of SUMO is ∼45 . 0 Å and 47 . 6 Å for Ubc9-SUMO and Ubc9-SUMO-RanBP2 simulations . Similarly , the average distance between the Val27-Glu42 loop and the Asp73-Ile88 of SUMO is 44 . 8 Å for Ubc9-SUMO and 45 . 6 Å for the Ubc9-SUMO-RanBP2 complex simulations . Alterations in the correlated fluctuations of regions that are not listed above are observed from the correlation maps . In many of these cases , although an overall change is observed in a region in the vicinity of a key residue , the correlation of the residue itself is conserved . One such example is the reduction in the correlated fluctuations of Ubc9 residues around position 70 , with both residues around 100 and the first 10 residues of Ubc9 . Lys74 , the key residue in this region which contacts the consensus sumoylation motif in targets [4] , conserves its correlations . To summarize , the correlated fluctuations among the residues responsible for the structural integrity of the complex catalytic region ( His83-Ser89 ) and the residues that interact with the sumoylation motif of the targets ( Asn121-Ala131 ) are stable for both Ubc9-SUMO and Ubc9-SUMO-RanBP2 complexes . On the other hand , the correlated fluctuations between His83-Ser89 and residues playing a role in selective target recognition ( Glu132-Arg141 ) are less stable in the Ubc9-SUMO complex . These correlations are allosterically stabilized upon RanBP2 binding . Further , the correlations between RanBP2 binding sites on Ubc9 and on SUMO pre-exist RanBP2 binding; yet , as expected , they are enhanced upon binding . Our results are summarized in Table 3 .
Two mechanisms have been proposed [1] , [4] for RanGAP1 target sumoylation ( Figure 1 ) . In both , the first step involves Ubc9 conjugation to SUMO . In the first mechanism Ubc9-SUMO binds to the target and an E3 ligase , whereas in the second Ubc9-SUMO can bind and sumoylate the target without an E3 ligase . In order to understand the role of E3 enzymes in the pre-organization of the Ubc9-SUMO complex in the mechanism of sumoylation , we simulated the Ubc9-SUMO complex with and without the E3 ligase RanBP2 . Based on the two conformational ensembles with and without E3 , Ubc9 is already largely pre-organized for target binding and catalysis [28] , [29] , yet the orientation of SUMO differs in Ubc9-SUMO and Ubc9-SUMO-RanBP2 complexes . Analysis of the conformational ensembles of Ubc9 , SUMO , Ubc9-SUMO , and Ubc9-SUMO-RanBP2 revealed that RanBP2 binding allosterically shifts the equilibrium of Ubc9 conformations , restricts SUMO orientation , and enhances the populations of the pre-organized conformational states [26] , [28] , [29] , [30] . RanBP2 binding reduces the conformational space sampled by Ubc9 ( Figure 4 , Figure S4 ) . At the same time , the RanBP2 binding allosterically reduces the mobility and the orientational freedom of the Ubc9 residues , with the effect being particularly dramatic around Asp100 and Lys101 target recognition residues [24] , [25] . RanBP2 binding allosterically enhances the correlated fluctuations of Ubc9 , mainly for residues around the catalytic and specific target recognition sites . The RanBP2 binding sites on Ubc9 and on SUMO which are spatially far apart display correlated fluctuations even in the absence of E3; however upon E3 binding these correlations get stronger . RanBP2 was proposed to limit the available conformations of the Ubc9-SUMO complex and prevent non-productive conformations [20] , [21] . Our simulations demonstrate that upon RanBP2 removal there is a change in the relative position of SUMO with respect to Ubc9 , yet the removal of RanGAP1 does not affect SUMO's position . Furthermore , in the absence of RanBP2 , RanGAP1 is not sufficient to prevent SUMO's position change ( unpublished data ) . This leads us to propose a mechanism where RanBP2 binding to the Ubc9-SUMO complex triggers SUMO's stabilization in a catalytically efficient orientation , with a subsequent target binding . This is consistent with RanBP2 enhanced allosteric effects on Ubc9's Asp100-Lys101 , the specific target recognition regions , and the correlated fluctuations of RanBP2 binding sites in Ubc9 and SUMO in the absence of RanBP2 . A correlation between the fluctuations implicates a network of interacting residues . It is highly plausible to expect an overlap of such a network with functional residues . We observe coupled fluctuations displaying changes between RanBP2 bound and unbound states . Ubc9 residues Lys74 , Tyr87 , Ser89 , Thr91 , Cys93 , Asp127 , Pro128 , Ala129 , Gln130 and Ala131 , interact with the consensus sumoylation tetrapeptide motif in most sumoylation targets [4] . The correlations between Ubc9 residues responsible for the structural integrity of the complex catalytic region ( His83-Ser89 ) and residues Asn121-Ala131 are conserved in both Ubc9-SUMO and Ubc9-SUMO-RanBP2 . Additionally , there is a second contact surface on Ubc9 ( Glu132-Asn140 ) , specific for the sumoylation target RanGAP1 . Because of the strong interactions between Ubc9 and RanGAP1 [4] through this additional binding region , the need for E3 activity for sumoylation of this target has not been apparent [4] , [20] . The enhanced correlations between Ubc9 residues His83-Ser89 and Glu132-Arg141 upon RanPB2 binding indicate that this additional binding surface is linked to the catalytic activity . The stronger correlations ( Figure 7 ) suggest that RanBP2 increases the efficiency of this additional target binding surface on Ubc9 . The mobilities of the catalytic Cys93 , Asp127 and Pro128 of Ubc9 , which contact the consensus sumoylation tetrapeptide motif in sumoylation targets [4] , are reduced with the SUMO binding , and RanBP2 binding does not lead to a further stability of these residues . Additionally , we observe the conservation of the correlations between His83-Ser89 and Asn121-Ala131 of Ubc9 , and the restrictions of the orientational freedom for Asp127-Pro128 and Ala129-Gln130 in the Ubc9-SUMO complex without RanBP2 . The pre-existing tendencies of these residues which have roles in catalysis and target recognition in the absence RanBP2 may indicate why Ubc9 can function without the aid of an E3 enzyme . Nevertheless , the correlations between other functional regions , such as between His83-Ser89 and Glu132-Arg141 , are enhanced with a restriction in the conformational freedom of Ubc9-SUMO with RanBP2 . Indeed , the restriction in the conformational space of Ubc9-SUMO with RanBP2 was suggested as a means of increasing sumoylation efficiency [20] , [21] . The mobility and orientational freedom of the Ubc9 Val27-Met39 loop is affected by RanBP2 binding . This loop displays correlated fluctuations with SUMO residues Phe36-Leu47 and Asp73-Ile88 , which are in close vicinity to the RanBP2 binding sites . Together , these point to a sequence of events in the formation of the complex which translate to SUMO binding to Ubc9 , followed by RanBP2 binding . From a mechanistic point of view , SUMO binds to Ubc9 , allosterically enhancing the mobility of the Val27-Met39 loop and residues Asp100-Lys101 , with the Val27-Met39 loop coordinated with the RanBP2 binding sites on SUMO . Next RanBP2 binds to the Ubc9-SUMO complex , moving SUMO closer to the target binding orientation , restricting the Ubc9 conformations , and at the same time RanBP2 induces allosteric changes in Ubc9 target recognition and catalytic sites . The changes in the network of hydrogen-bonds between Ubc9 and SUMO between the RanBP2 bound and unbound states appears to be the driving force behind the orientation limitation of SUMO induced by RanBP2 ( Figure S1 ) . The limitations imposed on the orientational freedom of Asp100-Lys101 of Ubc9 , and increased correlations between Ubc9 catalytic site residues ( His83-Ser89 ) and specific target recognition residues ( Glu132-Asn140 ) [4] , [20] induced by RanBP2 binding , support the proposed mechanism . Here we propose that the role of E3 ligase RanBP2 in sumoylation is to restrict the conformational freedom of the E2-SUMO complex and to increase the reaction efficiency via allosteric effects [31] on the E2 Ubc9 . RanBP2 binding to the E2-SUMO complex limits the accessible conformations of Ubc9 and the orientational space of the Ubc9 and SUMO monomers . In particular , the positional and orientational freedom of Ubc9 residues Asp100-Lys101 , important for target recognition [24] , [25] is restricted upon RanBP2 binding . RanBP2 binding stabilizes the correlations among Ubc9 residues that are functional in specific target recognition ( Glu132-Asn140 ) and catalytic activity ( His83-Ser89 ) [4] , [20] . Mechanistically , the correlations we observe in the dynamics of the E2-SUMO complex argue for such sequence of events in sumoylation and provide an explanation to the question of why sumoylation can also take place in the absence of an E3 , although with lower efficiency .
Molecular dynamics ( MD ) simulations are run on Ubc9 , SUMO , and the Ubc9-SUMO and Ubc9-SUMO-RanBP2 complexes , using Amber 8 [32] , [33] . Software and simulation parameter details are provided in the Text S1 . The structures of human Ubc9 ( PDB code: 1A3S ) and human SUMO-1 ( PDB code: 1A5R ) provide a base to analyze the effects of RanBP2 binding to the Ubc9-SUMO complex . The modeled structures for the intermediate complexes Ubc9-SUMO and Ubc9-SUMO-RanBP2 are extracted from the crystal structure of Ubc9-SUMO-RanGAP1-RanBP2 complex ( PDB code: 1Z5S ) . In the latter complex , as SUMO is bound to the acceptor lysine of target protein RanGAP1 , the thioester bond between Ubc9 active cysteine and SUMO C-terminal glycine is modeled in the co-crystal complexes of Ubc9-SUMO and Ubc9-SUMO-RanBP2 . The modeling of the thioester bond is detailed in Text S1 . Two sets of simulations are carried out . In the set where detailed time window based analysis is carried out , the simulation lengths for the structures are: unbound Ubc9 , 32 . 5 ns; unbound SUMO , 35 ns; Ubc9-SUMO complex , 58 ns; Ubc9-SUMO-RanBP2 complex , 50 ns . The second set is carried out to validate the major orientation change of SUMO observed in the Ubc9-SUMO structure in the first set of simulations . The simulation lengths for each complex or protein consider the specific characteristics of the structure , details of which are given in the Text S1 . The simulations generate a large number of different conformations of the structures , many of which may be similar . To obtain distinct representative conformations , we clustered the conformations with k-means clustering ( MMTSB Toolset's kclust utility [34] ) , taking the rmsd of residue positions from the cluster centroid as the similarity measure . Rmsd values of 2 Å , 1 . 7 Å and 1 . 5 Å are tested; a smaller number of clusters appear as the rmsd increases . The rmsd is set to 1 . 7 Å for Ubc9-SUMO complex and 2 Å for Ubc9-SUMORanBP2 complex . For the joined conformational space analysis on Ubc9 , the alignment is made on the Ubc9 conformation from the Ubc9-SUMO complex at 10 ns , and 1 . 7 Å cut-off is used . The principal component analysis is carried out using the ptraj module of AMBER 8 . 0 . The alignment of the joined ensemble , Ubc9 conformations from the Ubc9-SUMO and the Ubc9-SUMO-RanBP2 complexes , followed the same procedure as in the clustering analysis . The results are generated by projecting the Ubc9 conformations on the PC of the joined ensemble . The proportion of the eigenvalue of each PC to the sum of all eigenvalues represents the contribution of the PC to the all conformations in the trajectory . The cumulative contribution of all PCs up the PC of interest is given in the axes of Figure 4 and Figure S4 . The equilibration periods of the simulation ( 2 . 5 ns for Ubc9 , 5 ns for SUMO , 0 . 5 ns for Ubc9-SUMO and 0 . 4 ns for Ubc9-SUMO-RanBP2 ) are excluded in the calculations . For the msf calculations of each chain in the complexes , the alignment is carried out on the corresponding chain only , to eliminate the poor alignment that may possibly result from the structural changes in the quaternary structure . In the case of SUMO , the flexible N-terminal tail ( residues −1 to 19 in PDB 1A5R ) of SUMO is eliminated for the same reason . The normalized correlations between the fluctuations of residues , the cross-correlations , are defined by Eq . 1 as: ( 1 ) Where ΔRi and ΔRj are the fluctuation in the position vectors , Ri and Rj of residues i and j , respectively . The cross-correlations vary in the range [−1 , 1] with the lower and upper limit indicating fully anti-correlated and correlated fluctuations , respectively . The correlations are calculated for the total length of the trajectory and for the time windows defined by the clustering analysis . The backbone bond vector is defined as the normalized vector Mi from the α carbon Cαi−1 of residue i−1 to the α carbon Cαi of residue i . The normalized time-delayed auto-correlations of these virtual bond vectors are defined as in Eq . 2: ( 2 ) Where Mi ( t ) and Mi ( t+τ ) are the virtual bond vectors of i at time t and t+τ , respectively . The brackets represent averages over recorded snapshots . The auto-correlations are in the range [−1 , 1] with the lower and upper limit indicating fully anti-correlated and correlated virtual bonds , respectively . τ = 0 gives the equal-time auto-correlations , which is 1 for all virtual bond vectors . The correlations are calculated for several time delays τ , from 0 to 30 ns . The highest value of the time delay ( 30 ns ) is selected to be slightly longer than half simulation times , for both Ubc9-SUMO-RanBP2 and Ubc9-SUMO complexes . | Post-translational modifications constitute key regulatory mechanisms in the cell . One of these modifications is the tagging of the target protein with a smaller molecule . SUMO is such a ubiquitin-like tag protein , and sumoylation is the process of tagging proteins with SUMO . The malfunctioning of sumoylation is linked with diseases such as Alzheimer's , Parkinson's , and cancer . Based on experimental observations , two paths were suggested for sumoylation , the first and more efficient involves the E1 , E2 and E3 enzymes; the second only the E1 and E2 . Here we investigate these alternative paths of sumoylation . Our results offer an explanation for how sumoylation can take place with only the E1 and E2 enzymes , and for the mechanistic role of E3 . They emphasize that E2 bound to SUMO is already pre-organized for the transfer of SUMO to a target protein and E3 binding further stabilizes the conformations , shifting the ensemble and thus increasing the efficiency of the sumoylation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"computational",
"biology/molecular",
"dynamics",
"biophysics/macromolecular",
"assemblies",
"and",
"machines"
]
| 2010 | A Mechanistic View of the Role of E3 in Sumoylation |
The functions of several SOS regulated genes in Escherichia coli are still unknown , including dinQ . In this work we characterize dinQ and two small RNAs , agrA and agrB , with antisense complementarity to dinQ . Northern analysis revealed five dinQ transcripts , but only one transcript ( +44 ) is actively translated . The +44 dinQ transcript translates into a toxic single transmembrane peptide localized in the inner membrane . AgrB regulates dinQ RNA by RNA interference to counteract DinQ toxicity . Thus the dinQ-agr locus shows the classical features of a type I TA system and has many similarities to the tisB-istR locus . DinQ overexpression depolarizes the cell membrane and decreases the intracellular ATP concentration , demonstrating that DinQ can modulate membrane-dependent processes . Augmented DinQ strongly inhibits marker transfer by Hfr conjugation , indicating a role in recombination . Furthermore , DinQ affects transformation of nucleoid morphology in response to UV damage . We hypothesize that DinQ is a transmembrane peptide that modulates membrane-dependent activities such as nucleoid compaction and recombination .
Exposure of E . coli to DNA damaging agents induces the SOS response , which is under control of the RecA and LexA regulatory proteins . The SOS response upregulates gene functions involved in numerous cellular processes such as nucleotide excision repair ( NER ) , UV induced mutagenesis , recombination , inhibition of cell division and replication . The LexA repressor downregulates more than 50 SOS genes by binding to the operator sequence in their promoter regions [1] , [2] . SOS inducers ( e . g . UV ) cause replication blocks and generate RecA/ssDNA nucleoprotein filaments that mediate auto-proteolysis of the LexA repressor . Both NER and recombination are required to maintain DNA integrity . NER repairs numerous lesions introducing helical distortions , in which UvrA , UvrB and UvrC work in sequential steps to recognize and remove the lesion . The RecBCD complex is the major component for initiation of recombinational repair ( RR ) of DNA double strand breaks ( DSBs ) by processing a blunt dsDNA end into a dsDNA molecule possessing a 3′-terminated ssDNA tail . As part of this process RecBCD mediates RecA filamentation required for presynaptic processing of dsDNA ends . Most of the characterized LexA regulated genes play important roles in the physiology of E . coli , but there are still several genes of unknown function . One of these uncharacterized genes is dinQ , which is located in the 823 bp region between arsR and gor , 78 . 58 min on the E . coli chromosome . DinQ is predicted to encode an open reading frame ( ORF ) of 49 aa 139 nt downstream from a LexA operator sequence [1] . Small proteins of less than 50 amino acids are important in cellular processes such as regulation , signalling and antibacterial action [3]–[5] . More than 50 chromosomally encoded small proteins with a validated expression of less than 50 aa have been identified so far in E . coli [6]–[10] . Several of the newly discovered peptides are hydrophobic single transmembrane helices belonging to toxin-antitoxin systems . In this work we characterize the arsR-gor intergenic region in which an endonucleolytic product of dinQ is translated into a small hydrophobic peptide of 27 aa . DinQ is under LexA control and antisense regulation by a novel small RNA , agrB . DinQ is localized in the inner membrane as a single transmembrane peptide that modulates nucleoid compaction and conjugal recombination .
The SOS inducible dinQ gene in E . coli was identified in a search for new LexA regulated genes [1] . The dinQ gene was found in the 823 bp arsR-gor intergenic region ( Figure 1A ) , encoding a ∼330 nt transcript with a putative LexA operator sequence ( heterology index ( HI ) = 4 . 83 ) in the promoter region and two putative ORFs of 18 and 49 amino acids . No biological function , phenotype or significant homologies to proteins with known function were associated with DinQ [1] . Except for the dinQ gene , no other genes have been reported in the arsR-gor intergenic region . We performed a search for promoter and transcriptional terminator sequences in the arsR-gor intergenic region . As expected this search identified the dinQ LexA operator sequence identified earlier in a screen for LexA regulated genes in E . coli [1] . However , a second operator sequence for LexA ( HI = 13 . 82 ) in close proximity to the first was identified ( Figure 1A and 1B ) . Further , we identified putative −10 and −35 sequences corresponding to the dinQ promoter which overlaps both operator sequences , and a putative dinQ terminator sequence a few nucleotides downstream of the translational stop codon of the gor gene ( Figure 1B ) . Finally , the sequence search identified two new small noncoding RNAs , termed agrA and agrB ( arsR-gor region gene A and B , respectively ) , containing consensus like −10 and −35 sequences and rho independent terminator sequences ( Figure 1A and 1B ) . AgrA and agrB are transcribed in the opposite direction of dinQ but encode no putative ORFs . Thirty-one nucleotides at the 5′ end of agrA and agrB show antisense complementarity to dinQ ( Figure 1B and 1C ) . Twenty-five out of 31 nucleotides show complementarity to agrA while 30 out of 31 nucleotides show complementarity to agrB . This putative base pairing is indicative of possible RNA interference with the dinQ transcript . It thus appears that the arsR-gor region contain one protein coding gene , dinQ , and two small non-coding RNAs , agrA and agrB , with antisense complementarity to dinQ . To examine a potential role for the small non-coding RNAs agrA and agrB in regulating dinQ , three single mutants ( dinQ , agrA and agrB ) , one double mutant ( agrAB ) and one triple mutant ( dinQ agrAB ) were generated ( Table S1 ) . To estimate the approximate size of the dinQ , agrA and agrB transcripts , northern blots with total RNA isolated from UV exposed ( and unexposed ) wild type and mutant strains ( dinQ , agrA and agrB ) were hybridized with radiolabeled riboprobes against the respective genes ( Figure 2A ) . The dinQ probe generates five specific signals ( a–e ) , in which the main transcript ( a ) migrates according to the expected size of full-length dinQ , ∼330 nt . DinQ-b , -c , -d and -e migrates as transcripts of about 290 nt , 250 nt , 200 nt and 130 nt , respectively , according to the size marker . All signals are absent in the dinQ mutant demonstrating that all five transcripts are derived from dinQ . The full-length dinQ product is 3- and 4 . 6-fold upregulated in the agrA and agrB mutants respectively under normal growth ( without UV exposure ) . Notably , the dinQ-b signal is 4 . 8- and 3-fold upregulated in the agrB mutant as compared to the wild type and agrA mutant respectively . The dinQ-c product is not detectable in the agrB mutant . Further , the dinQ transcripts are induced in response to UV in wild type and agrA but not in agrB . These data indicate a regulatory mechanism by RNA interference , in which the agrA and agrB interfere differently with the dinQ transcript . Further , primer extension and 3′ mapping of dinQ RNA revealed transcript starts at 0 , +44 and +125 corresponding to the estimated size of dinQ-a , -b and –d , respectively ( Figure 2B ) . In agreement with the northern analysis we find that the +44 primer extension product ( dinQ-b ) is upregulated in the agrB mutant as compared to wild type and agrA mutant . The primer extension could not identify any products corresponding to dinQ-c or -e . The agrAB probes showed that the agrB transcript migrates slightly slower than the agrA transcript , and none of the transcripts were regulated in response to UV irradiation ( Figure 2A , middle panel ) . The agrA transcript is upregulated 9 . 5 times in the unexposed agrB mutant as compared to wild type , indicating that the absence of agrB somehow promotes agrA RNA stability or transcription of agrA . Further , primer extension revealed that the sequence of transcription start was identical for the agrA and agrB genes ( Figure 2C ) . 3′ mapping of agrA and agrB showed different transcription stops around the rho terminator ( data not shown ) , indicating that the transcripts are processed/terminated differently at the 3′end . In summary , these data demonstrate that both agrA and agrB downregulate the level of dinQ full-length transcript whereas agrB is particularly important for down regulation of +44 dinQ ( dinQ-b ) . The 31 nt antisense region in agrB gene ( Figure 1C ) indicate a function in antisense regulation of dinQ via RNA interference . This antisense sequence is partially complementary in agrA ( Figure 1C ) suggesting that both the agrA and agrB transcripts could base pair with the dinQ transcript . To ensure the function of dinQ when generating mutants , the deletions of agrA and agrB were made without destroying the dinQ promoter . DinQ belongs to the LexA regulon in E . coli [1] which regulates the SOS response . Several mutants of the SOS response , which play a direct role in DNA repair , display UV sensitivity . To examine the role of the dinQ-agrAB locus in the SOS response we tested the UV sensitivity of the various mutants ( Figure 3A ) . The agrB single mutant and agrAB double mutant showed a significant increase in UV sensitivity compared to the isogenic wild type . In contrast , the agrA and dinQ single mutants and the dinQ agrAB triple mutant showed no UV sensitivity . Further we examined UV survival of the agrB mutant carrying a plasmid expressing the agrB gene ( Figure S1 ) , demonstrating that the mutant recovered completely . These data indicate a role for agrB in protection against UV exposure , in which the agrB transcript modifies the dinQ transcript . According to the northern analysis ( Figure 2A ) agrB represses accumulation of the +44 dinQ/dinQ-b transcript . It thus appears that the +44 dinQ product mediates the UV sensitivity of the agrB mutant . To further investigate the role of the arsR-gor intergenic region in UV protection , we cloned agrA , agrB and dinQ separately or the entire region containing both small RNAs and the dinQ gene into the cloning vector pKK232-8 . Wild type cells transformed with pKK232-8-dinQ or pKK232-8 ( control plasmid ) showed the same viability during normal growth conditions ( data not shown ) . In contrast , it was not possible to transform pKK232-8-dinQ into the agrB mutant ( Table S2 ) , indicating that the endogenous level of agrB in wild type cells is sufficient to inhibit dinQ expression from the pKK232-8-dinQ construct during normal growth . However , wild type transformed with pKK232-8-dinQ showed increased sensitivity to UV as compared to wild type transformed with the control plasmid ( pKK232-8 ) ( Figure 3B ) . When the wild type was transformed with constructs expressing agrA or agrB they did not increase sensitivity of the cells to UV ( data not shown ) . Interestingly , wild type cells transformed with the plasmid carrying the entire arsR-gor locus , expressing all three genes , showed no UV sensitivity , demonstrating that agrAB can neutralize the UV sensitizing effect of dinQ . The agrB mutant ( Figure 3A ) and the wild type cells transformed with pKK232-8-dinQ ( Figure 3B ) displayed similar sensitivity to UV . These results suggest that the agrB transcript counteracts the UV sensitivity induced by dinQ expression . During construction of the agrB and agrAB mutants we observed that they form small colonies when plated on LB agar . To further investigate this growth phenotype we compared the growth rate in LB medium of the agrB , agrA and dinQ mutants and their isogenic wild type . OD600 was measured during growth and a sample was diluted and plated for viable counts . This experiment showed that only the agrB single mutant and agrAB double mutant grow more slowly than the wild type cells ( Figure S2A ) . Also in glucose-CAA medium agrB mutant cells grew more slowly than wild type cells ( Figure S2B and S2C ) . In another set of experiments we utilized flow cytometry to analyze whether DNA replication was affected in the growth impaired agrB mutant . We found that cells were smaller than normal with a reduced DNA concentration ( Figure S2D ) . The total time for replication from origin to terminus was shorter in the mutant and the number of origins and replication forks per cell were fewer compared to the wild type . There was also a considerable heterogeneity in the observed reduction in cellular DNA concentration . This heterogeneity could be due to cell-to-cell differences in expression of the DinQ peptide . The mechanism underlying the UV sensitive phenotype of dinQ in a multicopy situation or under constitutive upregulation in an agrB mutant is not clear . The dinQ gene contains two putative ORFs in which the second ORF contains three putative start codons ( Figure 3C ) . In this work the corresponding peptides are termed DinQ I-IV . None of the putative DinQ peptides show homology to known peptides . To examine if any of the ORFs mediate the UV sensitivity shown by dinQ , each of the putative DinQ peptides ( I–IV ) were cloned into the expression vector pET28b ( + ) and expressed under control of IPTG . DinQ I displayed no increased sensitivity in absence or presence of UV , suggesting that the putative peptide translated from ORF I ( Figure 3C ) does not induce DinQ toxicity ( Figure 3D and data not shown ) . In contrast , we observed that DinQ peptides II , III and IV showed a strong toxic/growth inhibitory effect even in absence of UV , demonstrating that the C-terminal amino acid sequence translated from start codon IV of the second ORF is sufficient to induce DinQ toxicity ( Figure 3D ) . Next , expression of DinQ II was titrated with increasing concentrations of IPTG in absence or presence of UV ( Figure 3E ) , showing that DinQ is highly toxic to the cells at very low doses of IPTG induction and the toxicity was UV independent . In another set of experiments we used a coupled in vitro transcription/translation E . coli T7 S30 extract to examine translation from pET28b ( + ) constructs encoding the putative peptides predicted from DinQ I–IV . Expression of DinQ I could not be detected whereas DinQ II–IV were highly expressed ( Figure 4A , lanes 2–5 ) . Notably , extracts with the DinQ IV construct produced two peptides of approximately 7 . 0 and 5 . 0 kDa , in which the smallest peptide indicates a fifth start codon . A closer inspection of the dinQ sequence uncovered a Shine Dalgarno motif within the DinQ IV sequence in optimal position to initiate translation at a GTG ( termed codon V in Figure 3C ) , which encodes a putative peptide of 27 aa , termed DinQ V . To examine if the DinQ V peptide induces toxicity we cloned the sequence into pET28b ( + ) , transformed the construct into wild type cells and monitored cell survival under the control of IPTG induction in presence or absence of UV exposure . DinQ V expression induced cell killing independently of UV treatment , suggesting that the sequence of peptide V is sufficient to mediate DinQ toxicity ( Figure 3C and 3D ) . In vitro transcription/translation assays with the pET28b ( + ) DinQ V construct produced a peptide corresponding to the predicted molecular weight of peptide V ( Figure 4A , lane 6 ) . Northern analysis identified five dinQ transcripts ( Figure 2A; dinQ a–e ) , in which the start site were determined for transcript a , b and d . To assess the translational activity of the in vivo dinQ transcripts a , b and d we synthesized the corresponding PCR products carrying the T7 RNA polymerase promoter and added E . coli T7 S30 extract . Only the dinQ-b/+44 transcript was translationally active , generating a peptide with a molecular weight similar to DinQ V , whereas the other transcripts were translationally inert ( Figure 4A , lanes 7–10 ) . Thus , it appears in vivo that the biologically active DinQ peptide ( peptide V ) is translated from the post transcriptionally modified +44 dinQ RNA . To examine endogenous expression of DinQ in vivo , a 3×FLAG tag was inserted chromosomally in frame with the C-terminal of the dinQ gene in the wild type and agrB mutant . Western analysis revealed a faint band for the FLAG tagged DinQ peptide in the agrB mutant while the peptide was barely detectable in wild type ( Figure 4B , lanes 5 and 6 ) . In UV treated cells the DinQ level was about two fold higher in the agrB mutant as compared to wild type ( Figure 4B , lanes 3 and 4 ) . It thus appears that the phenotypes of the agrB mutant are not due to polar effects of extensive overexpression of DinQ . Further , we introduced a chromosomal stop codon in the Lys4 position of dinQ in the agrB mutant and wild type ( Figure 3C , base labeled in red ) . Survival experiments showed that UV resistance was restored to wild type level in the agrB mutant ( Figure 4C ) , indicating that the UV sensitivity of the agrB mutant is caused by translation of a functional DinQ peptide . Next , we introduced three chromosomal point mutations in the dinQ up-stream sequence predicted to be involved in base pairing with agrB ( Figure 3C , bases labeled in red ) . Exposure of these strains to UV in the agrB mutant background showed wild type levels of survival ( Figure 4C ) . All together these results suggest that DinQ is translated into a peptide of 27 aa and the agrB-dinQ RNA interference is important for correct regulation of DinQ translation . To determine the intracellular localization of DinQ , western analysis was performed on extracts after subcellular fractionation . As antibodies against the native DinQ peptides could not be obtained , we introduced a 3×FLAG epitope at the N-terminal of DinQ ( peptide V ) . Spot assays on LB agar containing IPTG showed that the FLAG tagged peptide induced the same toxicity as the native peptide , demonstrating that the N-terminal FLAG tag had no effect on DinQ toxicity ( data not shown ) . Cells were harvested at several time points after IPTG induction to test the level of expression by western analysis of whole cell extracts . The FLAG-DinQ peptide could not be detected before induction , but showed strong signals 5 to 40 min after induction ( data not shown ) . To examine subcellular fractionation , antibodies against Lep and TolC were used as positive markers for inner and outer membrane fractions , respectively . The western blot showed that the inner and outer membranes are completely separated whereas the cytoplasmic fraction contains some contamination from the inner membrane ( Figure 4D , middle panel ) . DinQ localized to the inner membrane but could not be detected in the outer membrane ( Figure 4D , lower panel ) . The faint signal of the FLAG epitope in the cytoplasmic fraction is possibly due to cross contamination from the inner membrane . These data suggest that DinQ localization is confined to the inner membrane of E . coli . Analysis of the DinQ amino acid sequence using the consensus secondary structure prediction tool Jpred3 [11] revealed that DinQ has high propensity to form a single α-helix . All residues except a few on each flanking terminal are predicted with high confidence to belong to the predicted α-helix ( Figure 4E ) . With 20–22 residues in a single α-helix , the DinQ peptide could straightforwardly form a transmembrane helix of 6 full turns spanning more than 30 Å , as shown by modelling of DinQ using a regular α-helical template ( Figure 4F ) . The two positively charged lysine residues ( Lys4 and Lys9 ) are close to the phospholipid head groups , while particularly the charged Glu17 , but also Arg20 and Gln24 may form a polar patch that can interact with other membrane embedded proteins ( Figure 4F ) . The predicted single transmembrane peptide supports the localization of DinQ in the inner membrane ( Figure 4D ) . Previously , we showed that overexpressing another SOS inducible peptide , TisB , which encodes a small toxic inner membrane peptide , inhibits several SOS functions in wild type E . coli [12] . To determine whether DinQ affected induction of the SOS response we measured the level of recA and lexA mRNA in a mutant which constitutively overexpressed dinQ , agrB mutant or a dinQ deletion mutant ( Table S1 ) . Exponentially growing cells were exposed to UV and the amount of recA and lexA mRNA were determined prior to , and 20 min after irradiation by RT-qPCR . The expression levels of recA and lexA were similar in both mutants and wild type indicating that DinQ in contrast to TisB is not affecting regulation of the SOS response ( Figure S3 ) . To investigate a potential role of dinQ in filamentation , we stained cell samples with acridine orange prior to and after UV exposure . All strains showed the same filamentation pattern , in which cells displayed short filaments 1 h after irradiation and long filaments after 2 . 5 h , indicating that DinQ is not involved in the filamentation process of E . coli ( Figure S4 ) . Next , we measured spontaneous and UV induced mutagenesis as the frequency of rifampicin resistant colonies in wild type and the dinQ and agrB single mutants . The results showed no significant differences in mutation frequency in the mutant strains as compared to wild type suggesting that DinQ is not altering spontaneous and SOS induced mutagenesis ( data not shown ) . To examine whether high levels of DinQ induce changes in membrane potential we tested the ability of E . coli cells overexpressing DinQ to take up the dye DiBAC4 ( 3 ) [bis- ( 1 , 3-dibarbituric acid ) -trimethine oxanol] . The quantity of dye entering cells is proportional to membrane polarization , the less polarised the membrane the greater the quantity entering the cells and so increased fluorescence intensity due to binding to the membrane and intracellular components [13] . Cells were analyzed 5 and 20 min after IPTG induction of DinQ expression , incubated with DiBAC4 ( 3 ) for 20 min and analyzed by flow cytometry ( Figure 5A ) . No changes were observed for the plasmid control pET28b ( + ) . However , IPTG induction of DinQ ( peptide V ) showed a rapid increase in DiBAC4 ( 3 ) uptake ( Figure 5A ) , suggesting that elevated levels of DinQ depolarize the cell membrane . This data indicates that DinQ overexpression interferes with membrane polarity and could therefore lead to a loss of viability . Subcellular fractionation of E . coli showed that DinQ is localized to the inner membrane ( Figure 4D ) . DinQ is predicted to be a hydrophobic single transmembrane peptide that might compromise inner membrane integrity ( Figure 4F ) . We speculated that if DinQ affected the proton motive force , it would affect ATP production and intracellular ATP concentration . The intracellular ATP concentration was measured in wild type cells and in the agrB mutant , using a quantitative luciferase-based assay . This experiment showed that the concentration of ATP in the agrB mutant was about 50% of the concentration measured in wild type cells ( Figure 5B ) . Further , UV exposure increased the ATP concentration 0 . 6 fold in both cell types . Thus , it appears that insertion of the DinQ peptide into the inner membrane of E . coli impairs the energy supply in the form of ATP . The agrB mutant displayed sensitivity to UV suggesting that DinQ could have a role in the repair of UV induced DNA damage . Both nucleotide excision repair ( NER ) and recombinational repair ( RR ) counteract the genotoxic effects of UV irradiation . NER is required for the repair of UV induced photoproducts such as thymine dimers and cyclopyrimidine dimers , while RR is required for the repair of strand gaps and double strand breaks . To examine if DinQ is involved in NER we analysed UV sensitivity of the uvrA agrB and uvrA dinQ double mutants as compared to the single mutants . The survival analysis showed an additive effect between agrB and uvrA ( Figure 6A ) , whereas uvrA and dinQ showed no additional effect ( Figure S5 ) . These data indicate that elevated levels of DinQ in the agrB mutant sensitize the cell to UV via a pathway which is independent of NER . To examine the role of DinQ in recombination , mutant strains dinQ , agrB , recB , ( recB agrB ) and ( dinQ recB ) were exposed to UV irradiation . The double mutant ( recB agrB ) was slightly more sensitive to UV than the agrB single mutant ( Figure 6A ) , whereas recB and dinQ showed no additional effect ( Figure S5 ) . To further examine the role of DinQ in recombination we performed Hfr conjugation assays with a donor strain containing Tn10 , which carries the tetracycline resistance gene integrated in its chromosome and agrB , dinQ and uvrA single mutants as recipient strains . We used the uvrA mutant as control strain since UvrA is not involved in recombination ( and carry the kanamycin resistance gene required to detect the recipient ) . Hfr conjugation of Tn10 was at least 400-fold more efficient in dinQ and uvrA mutants as compared to agrB ( Figure 6B ) , suggesting that elevated levels of DinQ inhibit recombination . To examine if the conjugational process itself is affected in an agrB recipient we performed plasmid conjugation assays with a donor strain carrying an F′-plasmid with tetracycline resistance and the same recipient strains as in the Hfr conjugation experiment . Hfr conjugation differs from F′-plasmid conjugation in that transfer of genes after Hfr requires recombination whereas the F′-plasmid does not recombine in the recipient . In these experiments we find no differences in plasmid conjugation frequencies between the dinQ , uvrA and agrB recipient strains ( data not shown ) . In sum , these results suggest that recombination is inhibited in the agrB mutant during Hfr conjugation , but not in the transfer and uptake of DNA or survival of the agrB recipient . It thus appears that elevated levels of DinQ affect the recombination process . In dividing cells , replication forks are stalled by DNA lesions that impair DNA unwinding or block synthesis by the DNA polymerase subunits . In E . coli , UV lesions cause a delay in DNA synthesis for a period of time while stalled forks undergo repair . Fluorescence microscopy of Hoechst stained cells has demonstrated that the DNA often forms a compact structure during this phase , and suggests that the nucleoids undergo a major reorganization after UV exposure [14] . To investigate whether DinQ affects nucleoid organization , we used this technique to examine the shape and size of the nucleoids at different time points after UV exposure . In undamaged cells the nucleoids have characteristic shapes and numbers depending on the growth medium . When grown in glucose-CAA medium most cells have two nucleoids and some ( the largest cells ) have four ( Figure 7A , 0 min ) . Microscopy of cells 15 min after UV irradiation shows that all the wild type cells had lost the normal nucleoid morphology . In approximately 45% of the cells the nucleoids had been rearranged into a highly compact structure , whereas in the rest of the cells the nucleoids were found to be extended throughout the cells ( Figure 7A and 7B ) . Sixty minutes after UV irradiation all cells were found to contain extended nucleoids . After 90 min approximately 30% of the cells had divided and contained nucleoids with normal morphology . In the agrB mutant the degree of nucleoid compaction was similar to that of wild type cells at 15 min after UV exposure ( Figure 7A and 7B ) . However , in the period from 30 to 90 min 25–30% of the nucleoids of the agrB mutant cells were still locked in a compact state whereas a decreasing number of the wild type cells contained compact nucleoids . The result indicates that the transition from compact to extended nucleoid was inhibited in the agrB mutant . We also investigated the dinQ mutant with respect to nucleoid morphology after UV irradiation . At the 15 min time point approximately 35% of dinQ mutant cells contained a compact nucleoid compared to about 45% of wild type cells ( Figure 7B ) . This indicates that the compaction process might be affected in cells without DinQ . At 30 , 60 and 90 min similar numbers of cells with compact and extended nucleoids were found in the dinQ mutant compared to in wild type cells ( Figure 7B ) . The results indicate that cells lacking DinQ have an impaired ability to form a compact nucleoid structure after UV irradiation . Taken together the data reveals that the presence of DinQ is required in order to execute a transformation of nucleoid morphology in response to UV damage , and that overexpression of DinQ leads to a delay in decompaction and extension of the nucleoid during the later stages of the response . In conclusion , DinQ is under the control of the SOS response and the agrB antisense RNA , and expresses a single transmembrane peptide that has an effect on nucleoid compaction and when overexpressed on conjugal recombination ( summarized in Figure 7C ) .
Previous attempts to identify a translation product for dinQ have been unsuccessful , presumably because transcription and translation of dinQ are strictly regulated . First , the promoter region of the dinQ gene contains two LexA operators with different HI , which may suggest differential expression of the transcript early and late in the SOS response . Second , we identified two novel small non coding RNAs agrA and agrB with sequence complementarity to dinQ in the arsR-gor region that regulate dinQ by RNA interference . Notably , only the agrB RNA repressed the translational active +44 dinQ transcript ( dinQ-b ) whereas both sRNA repressed the primary translationally inactive dinQ transcript . Further , only the agrB antisense RNA counteracts DinQ toxicity . As such the dinQ-agrAB system appears to conform to the definition of a classical type I toxin-antitoxin ( TA ) system . It thus appears that the dinQ/agrB complex inhibit the endonucleolytic cleavage producing the active mRNA ( dinQ-b ) . Presumably , these antisense sRNAs have been tandemly duplicated in the genome , in which agrA has partly degenerated and appears to be non-functional due to less antisense sequence complementarity with the dinQ sequence as compared to agrB . The genomic organization and mode of antisense regulation of dinQ in the arsR-gor region resembles regulation of another SOS induced TA system , tisAB [15] , [16] . Similar to the dinQ RNA , endonucleolytic processing of the primary tisAB transcript is required to generate an active mRNA producing the toxic TisB peptide . Further , the tisAB locus contains an antisense RNA , IstR-1 that inactivates the translationally active mRNA by RNaseIII dependent cleavage . It appears that agrB may have a similar role in RNase dependent cleavage of the translationally active dinQ mRNA ( dinQ-b ) . In addition , Darfeuille et al revealed that the antisense RNA IstR-1 inhibits translation of the TisB toxin by competing with standby ribosomes binding upstream of the translation initiation region ( TIR ) . It is proposed that binding to the “standby” site is required for initiation of protein synthesis at the highly structured tisB TIR by ribosome sliding to the transiently open TIR [16] . In a similar manner , we speculate that the agrB antisense RNA regulates/inhibits translation from the active +44 transcript by binding a potential “standby” site upstream of the dinQ TIR . In order to investigate DinQ biochemically , we have attempted to purify DinQ , including fusion peptides . However , all attempts to purify DinQ as well as chemical synthesis of the peptide failed because of the hydrophobic nature of the peptide . Further , a general feature of small hydrophobic peptides including DinQ is the lack of obvious phenotypes associated with their inactivation [17]–[20] . As an alternative strategy we characterized the phenotype of augmented DinQ in an agrB mutant . The DinQ concentration in the agrB mutant after UV treatment is elevated only two fold as compared to the wild type , indicating that the DinQ levels in the agrB mutant is physiologically relevant . Of particular interest was the 400-fold reduction in the recombination frequency in the agrB mutant as compared to wild type cells , suggesting that augmented DinQ inhibits recombination in the agrB mutant . However , genetic data suggests that DinQ may also play a role in UV protective mechanisms independent of recombination . It appears that the large , ordered hyperstructures involved in homologous recombination are associated with the cell membrane [21] . The hyperstructures are dynamic and their size is dependent on the extent of the initial or ongoing DNA damage . The DinQ peptide is localized in the inner membrane of the cell and it is tempting to speculate about a role for DinQ in regulating DNA repair hyperstructures at the inner membrane . In addition , the prolonged period of nucleoid condensation in the agrB mutant may contribute to the impairment of DNA repair processes . Although such a direct role for DinQ is speculative , several small hydrophobic peptides have been demonstrated to modulate membrane dependent processes . The B1500 protein ( 65 aa ) interacts with the PhoQ sensor [18] , the 30 aa protein MgtR ( 30 aa ) interacts with MgtC [19] , the KdpF protein ( 29 aa ) is part of the Kdp complex [9] and the SidA protein ( 29 aa ) interacts directly with FtsW and FtsN [22] . The intracellular concentration of ATP was reduced in the agrB mutant compared to the wild type both before and after UV exposure . In wild type cells UV irradiation induces a two fold increase in ATP concentration during the first 20 to 30 min after exposure , and the increase is RecBC dependent [23] . These data suggest that loss of agrB and thereby excess of DinQ limit the cellular energy supply and may also explain some of the observed phenotypes . Our FLAG tag experiments revealed that DinQ increases only two fold in an agrB mutant and this apparently modest increase is sufficient to mediate dramatic effects on conjugal recombination rates , membrane depolarization , ATP levels and nucleoid reorganization . In a wild type cell population the level of DinQ translation is kept strictly under control by the LexA repressor , antisense agrB RNA and dinQ RNA processing , so for the majority of cells DinQ may never reach a level high enough to mediate the effects observed in an agrB mutant . Heterogeneity in the expression of LexA repressed genes has been observed by studying SOS promoter fusions in combination with imaging techniques and a subpopulation of cells clearly have a stronger SOS induction [24]–[26] . It is tempting to speculate that a higher level of DinQ is reached only in a subpopulation of cells where SOS induction is particularly strong or long lasting leading to a permanent or temporary agrB phenotype . Such an effect has been proposed for some toxin/antitoxin pairs in promoting formation of persister cells [27] , [28] . To gain a more detailed knowledge about the biological function of DinQ the agrB mutant could be an excellent model for studying the effects of DinQ and similar hydrophobic peptides in bacterial subpopulations .
The experiments were carried out in an AB1157 background [29] . Except for chromosomal point mutations and chromosomal 3×FLAG tags all mutants were made in strain BW25113-pKD46 [30] and introduced into AB1157 via T4GT7 transduction [31] . The agrA ( BK4042 ) , agrB ( BK4043 ) and dinQ ( BK4040 ) single mutants were made by deleting each of the genes and introducing a kanr cassette . Next , the agrAB double mutant ( BK4041 ) was generated by deleting both genes and introducing a kanr cassette . To construct a triple mutant the entire arsR-gor intergenic region containing dinQ , agrA and agrB was deleted ( BK4044 ) and replaced with the kanr cassette . Table S1 summarizes all strains used and generated in this work . Vector pKK232-8 ( 10–25 copies pr cell in E . coli ) contains a promoter less cat gene allowing selection of DNA fragments containing promoter activity [32] . pBK440 ( dinQ-agrAB ) /pBK444 ( dinQ ) is based on the vector pKK232-8 ( Pharmacia ) with a 2065/415 bp insert respectively from the intergenic region between arsR-gor in MCS , resulting in a plasmid that expresses E . coli dinQ from its own SOS inducible promoter . Cloning primers are listed in Table S3 . Expression plasmids pET28b ( + ) -DinQ I , pET28b ( + ) -DinQ II , pET28b ( + ) -DinQ III , pET28b ( + ) -DinQ IV and pET28b ( + ) -DinQ V contain the DinQ I to V ORFs inserted in the NcoI-BamHI restriction sites of the pET28b ( + ) vector ( Novagen ) . Chromosomal point mutations in dinQ to either introduce a premature translational stop codon in DinQ ORFV ( K4stop ) or introduce three point mutations in the agrB antisense region of dinQ ( A108T , C112G , A115G ) or to introduce a chromosomal DinQ C-terminal 3×FLAG tag were made by splicing PCR products with overlap extension ( SOEing PCR ) and recombine the final SOEing PCR product into a MG1655 background as described [30] . All SOEing products contained a flanking kanr cassette close to the arsR gene to facilitate selection of recombinants . To avoid unwanted recombination between the kanr cassette and the point mutations or the 3×FLAG tag during strain construction the SOEing products were transformed into strain BK5444-pKD46 which lacks the chromosomal dinQ-agrAB locus and where insertion/recombination of the SOEing products is possible only in the flanking homologous DNA sequences . The final PCR products were transformed into MG1655 containing pKD46 . Cells were cured for pKD46 and insertions verified by PCR and sequencing . Details of strain construction and oligos used are listed in Tables S1 and S3 , respectively . GenScript Corp . gene service constructed DinQ II and V with an N-terminal 3×FLAG tag that was inserted in the NcoI-BamHI restriction sites of the pET28b ( + ) vector ( Novagen ) . Cells were grown in LB- or K-medium [33] with appropriate antibiotics ( 100 µg/ml ampicillin and 50 µg/ml kanamycin ) . For the nucleoid compaction studies cells were grown in AB minimal medium [34] supplemented with 1 µg/ml thiamine , 0 . 2% glucose and 0 . 5% casamino acids . In vitro transcription/translation on circular pET28b ( + ) templates or linear PCR products were performed according to Promegas protocols E . coli T7 S30 Extract System for Circular DNA and E . coli S30 Extract System for Linear Templates , respectively , with [14C]-Leucine as radiolabeled amino acid . The translation products were analysed by SDS-PAGE and visualized on Typhoon 9410 ( Amersham ) . Aliquots of exponentially growing ER2566/pET28b ( + ) -DinQ V were harvested by centrifugation 20 min after IPTG induction ( 1 mM ) . Cells were resuspended in 4 ml 50 mM phosphate buffer pH 7 . 2 and sonicated three times for 15 sec . Further fractionation was performed as described by [20] . Proteins from all fractions were acetone precipitated 1∶1 overnight at -20°C , pellets after centrifugation was resuspended in 4× NuPAGE sample loading buffer ( Invitrogen ) and loaded onto 10% NuPAGE Novex Bis-Tris gels ( Invitrogen ) . Cells were grown to OD600≈0 . 4 in LB and induced with IPTG ( 1 mM ) . At 0 , 5 and 20 min culture samples were diluted 1∶10 in filtered AB minimal medium [34] +10 µg/ml DiBAC4 ( 3 ) ( Sigma-Aldrich ) . After 20 min incubation in the dark at room temperature , cells were analysed in a Flowcytometry LSRII ( Becton Dickinson ) equipped with an argon ion laser and a krypton laser ( both Spectra Physics ) . DiBAC4 ( 3 ) was detected using 488 nm laser . The distribution of DiBAC4 ( 3 ) fluorescence was plotted on a logarithmic scale . The data obtained was analyzed by winMDI software . Cell aliquots were harvested before and 20 min after induction with IPTG ( 1 mM ) and washed once in 50 mM Tris-acetate pH 7 . 75 . ATP was extracted from washed cells by 1% trichloroacetic acid ( TCA ) in 50 mM Tris-acetate pH 7 . 75 for 10 min . Tris-acetate pH 7 . 75 was added 1∶10 to obtain optimal pH of 7 . 75 before mixing with rL/L reagent ( ENLITEN ATP assay , Promega ) at room temperature . The amount of ATP extracted ( RLU value ) was measured with 20/20 Luminometer ( Turner Designs ) and related to the OD600 for each sample . Aliquots of exponentially growing recipient strains dinQ ( BK4040 ) , agrB ( BK4043 ) and uvrA ( BK4180 ) were mixed in equal volumes with donor strains BW7623 ( with the tetracycline resistance gene , Tn10 , integrated in its chromosome ) or ER2738 ( carrying a tetracycline resistance conjugative plasmid ) and incubated at 37°C for 30 minutes . BW7623 was used to examine chromosomal transfer to the recipient strains ( recombination dependent ) whereas ER2738 was used for plasmid conjugation . Cells were vortexed thoroughly and spread on selective LB plates . Hfr recombination rate and plasmid conjugation rate was calculated as number of recombinants/conjugated cells pr 106 cells . Exponentially growing wt ( AB1157 ) , dinQ ( BK4040 ) and agrB ( BK4043 ) cells were UV irradiated with 3 J/m2 while stirring . 1 . 5 ml samples were taken at 0 , 15 , 30 , 60 and 90 min after irradiation . Washed once and resuspended in 100 µl cold , filtered TE buffer . Then 1 ml of cold , filtered 77% ethanol was added for fixation . Fixed cells were mounted on a poly-L-lysine coated microscope slide and the DNA was stained with Hoechst 33258 ( 5 µg/ml , Sigma ) in mounting medium ( 40% glycerol in PBS pH 7 . 5 ) . Visualization of stained cells was performed using a Leica DM6000B phase-contrast/fluorescence microscope equipped with a 63× objective and a BP340-380 excitation filter . Pictures were taken using a Leica DFC350 FX digital camera that was connected to a computerized image analysis system ( LAS AF software , version 2 . 0 . 0 , Leica ) . The fluorescent image was merged with the phase-contrast image . | Exposure of the bacterium Escherichia coli to DNA damaging agents induces the SOS response , which up-regulates gene functions involved in numerous cellular processes such as DNA repair , cell division , and replication . Most of the SOS regulated genes in E . coli have been characterized , but still there are several genes of unknown function . One of these uncharacterized genes is dinQ . In this work we characterize dinQ and two novel small RNAs , agrA and agrB , that regulate expression of dinQ . The DinQ peptide is localized in the inner membrane as a single transmembrane peptide of 27 amino acids . Small proteins of less than 50 amino acids are important in cellular processes such as regulation , signalling , and antibacterial action . Here we demonstrate that DinQ modulates recombination and transformation of nucleoid morphology in response to UV damage . Our results provide new insights into small hydrophobic peptides that could regulate important DNA metabolic processes dependent on the inner membrane of the cell . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genome",
"expression",
"analysis",
"cellular",
"stress",
"responses",
"rna",
"interference",
"gene",
"regulation",
"microbiology",
"gene",
"function",
"escherichia",
"coli",
"cell",
"growth",
"dna",
"recombination",
"molecular",
"genetics",
"dna",
"microbial",
"growth",
"and",
"development",
"bacterial",
"pathogens",
"gene",
"expression",
"membranes",
"and",
"sorting",
"biology",
"molecular",
"biology",
"rna",
"rna",
"processing",
"cell",
"biology",
"nucleic",
"acids",
"protein",
"translation",
"gene",
"identification",
"and",
"analysis",
"genetics",
"dna",
"repair",
"genomics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
]
| 2013 | Single Transmembrane Peptide DinQ Modulates Membrane-Dependent Activities |
Maternal factors initiate the zygotic developmental program in animal embryos . In embryos of the chordate , Ciona intestinalis , three maternal factors—Gata . a , β-catenin , and Zic-r . a—are required to establish three domains of gene expression at the 16-cell stage; the animal hemisphere , vegetal hemisphere , and posterior vegetal domains . Here , we show how the maternal factors establish these domains . First , only β-catenin and its effector transcription factor , Tcf7 , are required to establish the vegetal hemisphere domain . Second , genes specifically expressed in the posterior vegetal domain have additional repressive cis-elements that antagonize the activity of β-catenin/Tcf7 . This antagonizing activity is suppressed by Zic-r . a , which is specifically localized in the posterior vegetal domain and binds to DNA indirectly through the interaction with Tcf7 . Third , Gata . a directs specific gene expression in the animal hemisphere domain , because β-catenin/Tcf7 weakens the Gata . a-binding activity for target sites through a physical interaction in the vegetal cells . Thus , repressive regulation through protein-protein interactions among the maternal transcription factors is essential to establish the first distinct domains of gene expression in the chordate embryo .
In animal embryos , maternal information initiates the zygotic developmental program . Maternal factors are often specifically localized to set up pre-patterns . This mechanism has been extensively studied in embryos of invertebrates including sea urchin and flies [1 , 2] . In syncytium embryos of Drosophila , maternally localized factors define embryonic axes and initiate specific gene expression patterns from the zygotic genome [1] . Several specifically localized maternal factors are also known in vertebrates [3–7] . However , the whole system by which localized maternal factors establish the initial zygotic gene expression patterns is not yet fully understood . In the chordate , Ciona intestinalis , the first zygotic gene expression begins between the 8- and 16-cell stages . A comprehensive study has revealed that a limited number of regulatory genes encoding transcription factors and signaling ligands are expressed at the 16-cell stage [8–10] . The majority of genes activated at the 16-cell stage are expressed either in the entire animal hemisphere domain ( AD ) , entire vegetal hemisphere domain ( VD ) , or the posterior vegetal domain within the VD ( PVD ) ( Figs 1A and S1 ) . Cells in the AD mainly give rise to epidermal and neural cells , while cells in the VD give rise to mesendodermal tissues and the nerve cord . Muscle and mesenchymal cells are mainly derived from cells in the PVD . Three maternal factors , β-catenin , Gata . a , and Zic-r . a ( also called Macho-1 ) are involved in establishing these initial gene expression patterns [11–15] . β-catenin and Gata . a are required for expression in the vegetal and animal blastomeres , respectively [11–13] . The activity of β-catenin is restricted to the VD [11 , 16] , where it is suggested to suppress Gata . a activity [13] . Zic-r . a is localized in the posterior-most cells and required for activation of genes in the PVD [14 , 15 , 17] . In addition to these three maternal factors , Pem-1 is also known to be localized in the posterior-most cells . In the present study , we did not consider the posterior-most vegetal cells or Pem-1 , because Pem-1 is thought to maintain the posterior-most cells , from which germ cells are derived , in a transcriptionally quiescent state by suppressing RNA polymerase II functions [18 , 19] . It is not yet known whether combinatorial regulation by the above three maternal factors , β-catenin , Gata . a , and Zic-r . a , is sufficient to establish the initial zygotic gene expression pattern at the 16-cell stage . Furthermore , is it not understood how these maternal factors interact with each other to define precise regions of gene expression . Here , we describe the whole system that initiates the zygotic developmental program .
To understand how maternally expressed β-catenin , Gata . a , and Zic-r . a regulate gene expression at the 16-cell stage , we re-examined expression of genes that are activated at the 16-cell stage in morphant embryos of either of β-catenin , Gata . a , or Zic-r . a . In normal embryos , Fgf9/16/20 is expressed in the VD , and the same expression pattern was observed in embryos injected with a control morpholino oligonucleotide ( MO ) . The expression of Fgf9/16/20 was lost in β-catenin morphants as a previous study has shown [20] , whereas Fgf9/16/20 expression was scarcely affected in morphants of Gata . a or Zic-r . a ( Fig 1B–1E; S1 Table ) . These results suggest that Fgf9/16/20 is regulated by β-catenin only . This finding is consistent with a previous study indicating similar regulation of Foxd . b , which is also expressed specifically in the VD [13 , 21] ( S2A–S2D Fig; S1 Table ) . As observed in normal embryos , in embryos injected with the control MO , Tbx6 . b was expressed only in the PVD . This expression was lost in β-catenin and Zic-r . a morphants , but it was unaffected in Gata . a morphants ( Fig 1F–1I; S1 Table ) ; thus , β-catenin and Zic-r . a coordinately activate Tbx6 . b in the PVD . In normal embryos and embryos injected with the control MO , Efna . d and Tfap2-r . b ( AP-2-like2 ) are expressed in the AD . The expression of Efna . d and Tfap2-r . b was greatly reduced in Gata . a morphants , and expanded to the vegetal hemisphere in β-catenin morphants ( Figs 1J–1M and S2E–S2H; S1 Table ) . This observation is consistent with a previous study in which a synthetic reporter construct with 12 Gata-binding sites behaved similarly [13] . On the other hand , Efna . d and Tfap2-r . b expression was unaffected in Zic-r . a morphants; thus , Efna . d and Tfap2-r . b are activated by Gata . a in the AD , and repressed by β-catenin in the VD . We examined the distribution of Gata . a , Zic-r . a , and Tcf7 , which employs β-catenin as a co-factor , by immunostaining with specific antibodies . Previous studies have shown that Gata . a , Zic-r . a , and Tcf7 mRNAs are expressed mainly in the endoderm , nervous system , and mesenchyme , respectively , at the tailbud stage [8 , 15] . Indeed , cell nuclei of these tissues were stained specifically with the antibodies ( S3 Fig ) . At the 16-cell stage , Gata . a and Tcf7 proteins were observed equally in all nuclei of both of the AD and VD , while Zic-r . a was observed in all nuclei of the PVD and in the most posterior region where Zic-r . a mRNA is localized ( Fig 2 ) . The above analyses raised three points . The first is how β-catenin and Zic-r . a cooperate to activate Tbx6 . b specifically in the PVD . The second related point is why Tbx6 . b is not activated in the anterior vegetal cells by β-catenin . The third is how β-catenin represses Gata . a activity in the VD . To better understand the regulation of Fgf9/16/20 expression , we prepared reporter gene constructs in which upstream regulatory sequences were fused to the green fluorescent protein ( Gfp ) gene . The reporter constructs were introduced into fertilized eggs by electroporation , and their expression was examined by in situ hybridization . A series of deletion constructs showed that 219 bp upstream sequence of the transcription start site of Fgf9/16/20 was sufficient to drive expression of the reporter gene specifically in the VD ( Figs 3A , 3B and S4A ) . Because β-catenin functions as a cofactor of Tcf7 , we searched for potential Tcf7-binding sites using a position weight matrix , and mutated the site with the highest score . The mutant sequence greatly reduced expression of the reporter [−219+μTcf ( b ) ] . Although mutations introduced into two additional potential Tcf7-binding sites rarely affected expression of the reporter [−219+μTcf7 ( a ) and −219+μTcf7 ( c ) ] , double mutations of these sites in combination with the site of the highest score ( site b ) completely abolished reporter expression ( Fig 3A and 3C ) . Therefore , the Tcf7 sites within the 219 bp region , especially site b , are responsible for the expression of the reporter . We confirmed that Tcf7 bound to the 219 bp region by a chromatin immunoprecipitation ( ChIP ) assay using an antibody against Ciona Tcf7 followed by high-throughput sequencing ( ChIP-seq ) and microarray analysis ( ChIP-Chip ) . As shown in Fig 3D , a clear ChIP peak was seen around the identified region . Indeed , peak caller programs for ChIP-seq and ChIP-chip data both identified peaks in this region ( see Materials and Methods ) . A gel-shift assay also indicated that the TCF7-binding site of Fgf9/16/20 bound Tcf7 specifically ( Fig 3E ) . We also analyzed upstream regions of Foxd . b and Lefty , which are expressed in the VD ( S1 Fig ) . A region upstream of Foxd . b between −1241 and −1041 was required to drive expression of a reporter in the VD ( S4B Fig ) . A previous study has shown that Tcf7 sites are essential for Foxd expression in a closely related species , Ciona savignyi [21] . We confirmed that mutations introduced into five putative Tcf7-binding sites in the upstream sequence of Foxd . b in C . intestinalis abolished reporter expression ( S4B and S4C Fig ) . This region is highly conserved in the upstream region of a paralogous gene , Foxd . a ( S4D Fig ) . This high conservation suggested the importance of this sequence . Although it prevented us from mapping ChIP data confidently to these regions , a gel-shift assay showed that the proximal Tcf7-binding site bound Tcf7 specifically ( S4E Fig ) . We also confirmed that a 769 bp upstream region of Lefty was sufficient for expression in the VD ( S4F Fig ) , and that this region contained a region that bound Tcf7 ( S4G Fig ) . These observations are consistent with a previous study indicating that 12 repeats of the Tcf7-binding site can activate a reporter gene in the VD [13] . Our findings extended the results of previous studies [13 , 21] for the following points . First , Tcf7 sites are essential for the specific expression of Fgf9/16/20 and Foxd . b . Second , Tcf7 binds to the enhancers of genes that are specifically expressed in the VD . Thus , β-catenin and Tcf7 are the only factors required for specific expression in the entire VD . A previous study [22] showed that an upstream region between −862 and −2400 bp is required for the expression of Tbx6 . b at the 32-cell stage . Our analyses using a series of deletion reporter constructs identified the −189 bp upstream of Tbx6 . b as a region sufficient to drive reporter expression in the PVD at the 16-cell stage ( Figs 4A , 4B and S5A ) . When two putative Tcf7-binding sites were mutated , the expression of the reporter was completely abolished ( Fig 4A and 4C ) . ChIP showed that this region bound Tcf7 in vivo ( Fig 4D ) , and a gel-shift assay showed that Tcf7 bound to the Tcf7 site in vitro ( Fig 4G ) . In contrast , no significant peak regions for Zic-r . a were identified in this region by the peak caller programs , although a weak peak was visible ( Fig 4D ) . We previously determined nucleotide sequences that preferentially bind Zic-r . a by an in vitro selection assay ( see S11F Fig ) [17] . However , we did not identify clear binding sites for Zic-r . a in the 189 bp region of Tbx6 . b ( S5A Fig ) . Therefore , the weak peak in this region might represent indirect binding of Zic-r . a ( see below ) . We also examined upstream regulatory sequences of Wnttun5 and Admp , which are also expressed specifically in the PVD at the 16-cell stage ( S1 Fig ) . A 188 bp region within the upstream sequence of Wnttun5 ( −792 to −979 ) was necessary for specific expression in the PVD ( S5B and S5C Fig ) . Mutations introduced into either of two putative Tcf7-binding sites greatly reduced expression ( S5B , S5D and S5E Fig ) , and this region bound Tcf7 in vivo and in vitro ( Figs 4E and S5F ) . The 323 bp upstream region of Admp was sufficient to drive reporter expression specifically in the PVD at the 16-cell stage ( S5G and S5H Fig ) , and this region bound Tcf7 in vivo ( Fig 4F ) . The peak caller programs again did not identify peaks for Zic-r . a in these essential regions of Wnttun5 and Admp ( Fig 4E and 4F ) . Because Zic-r . a is required for the expression of Wnttun5 and Admp , Zic-r . a might indirectly bind to the upstream regulatory regions of Wnttun5 and Admp ( see below ) . Zic can function as a co-factor of transcription factors in vertebrates [23 , 24] . Therefore , we hypothesized that Zic-r . a might function as a co-factor of Tcf7 and bind to the regulatory elements of Tbx6 . b , Wnttun5 , and Admp indirectly through Tcf7 . To test whether Tcf7 interacted with Zic-r . a , 3xmyc-tagged Tcf7 and 3xflag-tagged Zic-r . a were overexpressed under the control of the upstream sequence of Dlx . b in epidermal cells of tailbud embryos [25] . By immunoprecipitation with a specific antibody against the myc-tag , we found that these two proteins can interacted with each other , although these two proteins might be expressed more abundantly in epidermal cells of experimental embryos than in the PVD of normal embryos ( Fig 5A ) . We further confirmed this interaction in vitro using 3xmyc-tagged Tcf7 and 3xflag-tagged Zic-r . a proteins produced in E . coli ( Fig 5B ) . Therefore , Zic-r . a may bind to DNA indirectly through binding to Tcf7 , although our data do not rule out the possibility that Zic-r . a binds directly to regulatory elements of genes not tested in the present study . The above analyses showed that the Tcf7 sites in the upstream sequences of Fgf9/16/20 and Foxd . b direct expression in the VD , and that the Tcf7 sites in the upstream sequences of Tbx6 . b and Wnttun5 do not direct expression in the anterior vegetal domain ( AVD ) . To understand the differences between these two types of cis-regulatory elements , we first examined the possibility that Tcf7 sites in the upstream sequence of Tbx6 . b and Wnttun5 bound Tcf7 weakly , and that Zic-r . a enhanced activity of the β-catenin-Tcf7 complex in the PVD . We replaced two essential Tcf7-binding sites in the construct that contained the 219 bp upstream sequence of Fgf9/16/20 ( sites a and b; see Fig 3A ) with the proximal Tcf7-binding site important for expression of Tbx6 . b . This construct promoted reporter expression in the AVD and PVD ( Fig 5C; n = 60 , 57 . 1% of the anterior vegetal cells , 60 . 0% of the posterior vegetal cells , and no animal hemisphere cells expressed the reporter ) . Thus , the Tcf7-binding site in the upstream region of Tbx6 . b was not likely to be qualitatively different from that of Fgf9/16/20 . Another possibility is that additional cis-regulatory elements are required for the different expression patterns . Because a construct consisting of twelve Tcf7-binding sites induces reporter expression in the VD [13] , it is very unlikely that Fgf9/16/20 and Foxd . b have additional cis-regulatory elements for expression in the AVD . Instead , it is more likely that Tbx6 . b and Wnttun5 have additional cis-regulatory elements that repress the activity of Tcf7-binding sites in the AVD . To investigate this possibility , we further narrowed down the cis-regulatory region of Tbx6 . b . We first prepared a construct in which four repeats of a sequence that included the regulatory element of Tbx6 . b were fused to the basal promoter of Brachyury . While the basal promoter alone cannot drive reporter expression [26] , the fusion construct was specifically expressed in the PVD ( Fig 5D–5F ) . Therefore , these sequences contained sufficient elements to drive the reporter expression in the PVD . Next , we prepared a series of Tbx6 . b constructs in which various 15 bp regions were mutated . The mutant construct μC induced ectopic expression of the reporter in the AVD . The remaining constructs did not induce ectopic expression , although μD weakened reporter expression in the PVD . Therefore , the region mutated in μC ( region C ) is likely to repress expression in the AVD; note that our data do not rule out a possibility that there are additional repressive elements , because the mutations did not cover the entire region required for specific expression of Tbx6 . b . We next inserted four repeats of the 22 bp sequence containing region C into the downstream sequence of the critical Tcf7-binding site of the construct containing the 219 bp upstream sequence of Fgf9/16/20 that was used in Fig 3 . With this insertion , the construct rarely drove reporter expression in the AVD ( Fig 5G ) . The same DNA fragment also reduced the activity of the essential regulatory region of Foxd . b in the anterior vegetal cells , when it was inserted downstream of the essential regulatory region ( Figs 5G , S6A and S6B ) . Although the result does not necessarily indicate that the 22 bp region is the only repressive element in the upstream region of Tbx6 . b , it showed that this 22 bp region has a repressive activity . Expression of the construct with four repeats of the regulatory element of Tbx6 . b and the basal promoter of Brachyury , which was used in Fig 5D , was dependent on Zic-r . a , because concomitant injection of the MO against Zic-r . a abolished reporter expression . However , the μC construct was expressed in both the AVD and PVD , even in Zic-r . a morphants ( S6C Fig ) . Similarly , concomitant injection of the MO against Zic-r . a reduced PVD-specific expression of the construct containing the 219 bp upstream sequence of Fgf9/16/20 and four repeats of the repressive elements , but did not affect expression of the construct containing only the upstream sequence of Fgf9/16/20 ( Fig 5H–5L ) . Therefore , this repressive element functions in both the AVD and PVD , and its repressive activity is overcome by Zic-r . a in the PVD of normal embryos . We examined whether Wnttun5 also had a similar repressive element . We prepared constructs in which four repeats of the sequence that contained the regulatory element of Wnttun5 were fused to the basal promoter of Brachyury . While a construct containing a 297 bp upstream sequence between −994 and −697 drove reporter expression specifically in the PVD , constructs containing shorter upstream sequences ( −994 to −735 , −994 to −793 , −994 to −853 , and −979 to −860 ) drove reporter expression in both the AVD and PVD ( S6D Fig ) . Four repeats of the sequence containing this putative repressive region repressed the activity of the regulatory region of Foxd . b in the AVD when it was inserted downstream of the essential regulatory region ( S6E and S6F Fig ) . Thus , similar to Tbx6 . b , Wnttun5 also has a cis-regulatory element that repressed the reporter expression in the AVD . Our previous study showed that specific expression of Zfpm ( Fog ) in the AD requires two Gata . a-binding sites [13] . Furthermore , a construct , in which twelve GATA-binding sites are placed in front of the Brachyury basal promoter ( G12 construct ) , drives specific expression of a reporter in the AD [13] . In the present study , we also confirmed that Gata . a-binding sites of Efna . d were necessary for specific reporter expression in the AD . As shown in Fig 6A–6C , a series of deletion constructs of the upstream sequence of Efna . d revealed that the region between −419 and −220 was essential for driving reporter expression in the AD . Furthermore , mutations introduced into three or four putative Gata . a-binding sites within this region impaired the activity of this regulatory region . This finding was mostly consistent with a result published recently [27] , except that a 319 bp upstream region , which lacked the critical Gata . a site identified in the previous study , activated the reporter weakly in our study . Indeed , ChIP analysis with an antibody against Gata . a showed that this region bound Gata . a in vivo ( Fig 6D ) . The 1513 bp upstream region of Tfap2-r . b ( AP-2-like2 ) drove reporter expression specifically in the AD ( S7B Fig ) , and a clear peak of Gata . a binding was observed within this region . Previous studies have shown that the 204 , 975 , and 314 bp upstream regions of Gdf1/3-r , Fzd4 , and Zfpm , respectively , are sufficient to drive the expression of reporters in the AD [13 , 27] . Clear peaks of Gata . a binding were observed within these upstream sequences ( S7C–S7F Fig ) . For further confirmation , we injected the β-catenin MO with the construct containing the 419 bp upstream region with intact or mutant Gata . a sites that were used in Fig 6A . While the reporter construct with the intact Gata . a sites was expressed ectopically in the VD of β-catenin morphants ( Fig 6E , 6F and 6G ) , the reporter construct with mutant Gata sites was not expressed in the AD or VD ( Fig 6E , 6H and 6I ) . Thus , ectopic expression of Efna . d in β-catenin morphants was dependent on these Gata . a sites . We next addressed how β-catenin suppresses Efna . d and Tfap2-r . b in the VD . The regulatory regions of Efna . d , Tfap2-r . b , Gdf1/3-r , Fzd4 , and Zfpm did not bind Tcf7 ( Figs 6D and S7C–S7F ) . This observation indicated that binding of the β-catenin and Tcf7 complex to the regulatory regions was not required for suppression of Efna . d expression in the VD . Hence , we reasoned that β-catenin might prevent Gata . a from binding to its target sites . To test this possibility , we performed gel-shift analysis . As shown in Fig 7A , Gata . a bound to the proximal site ( site 1 ) of the three Gata sites tested in the sixth construct ( in Fig 6A ) of Efna . d in vitro , and the shifted band disappeared in the presence of a specific competitor . While this specific binding was not disrupted by co-incubation with either Tcf7 or β-catenin , it was reduced by co-incubation with both Tcf7 and β-catenin ( Figs 7B and S8A ) . Co-incubation of Gata . a with β-catenin and Tcf7 also reduced Gata . a binding to other Gata sites in the upstream sequences of Efna . d and Gdf1/3-r , although different Gata sites showed different rates of reduction ( Figs 7C and S8B ) . Therefore , in the VD of normal embryos , β-catenin and Tcf7 interact with Gata . a and suppress the binding activity of Gata . a . For further confirmation , we used the 314 bp upstream sequence of Zfpm , which contains two critical Gata sites and is sufficient for activating a reporter in the AD ( S9A Fig ) [13] . When β-catenin was overexpressed in the AD under the control of the Zfpm enhancer , expression of a LacZ reporter under Zfpm was reduced markedly ( S9B Fig ) . Similarly , injection of β-catenin mRNA greatly reduced the expression of Efna . d in the AD ( Fig 7D; n = 75 , 1 . 7% of anterior animal cells , 17% of posterior animal cells , and no vegetal hemisphere cells expressed Efna . d ) . In addition , treatment with BIO , which is a specific inhibitor of Gsk3 and thereby stabilizes β-catenin [16] , reduced the expression ( S9D and S9E Fig; n = 25 , 0% of anterior animal cells , 0% of posterior animal cells , and no vegetal hemisphere cells expressed Efna . d ) . We confirmed that Gata-a binding to the Efna . d upstream region was significantly reduced in BIO-treated embryos by ChIP-qPCR ( Fig 7E ) . The above finding suggested that Gata . a physically interacted with β-catenin and Tcf7 . To test this inference , we used embryos in which tagged proteins were misexpressed in the epidermis using the upstream sequence of Dlx . b . As expected , endogenously expressed β-catenin was co-immunoprecipitated with overexpressed myc-tagged Tcf7 but not myc-tagged Gfp ( positive and negative controls , respectively; Fig 8A and 8B ) . We found that β-catenin was also co-immunoprecipitated with myc-tagged Gata . a ( Fig 8C ) . In addition , flag-tagged Gata . a was co-immunoprecipitated with myc-tagged Tcf7 when they were co-expressed under the Dlx . b upstream sequence ( Fig 8D ) . Although Tcf7 and Gata . a might be expressed more abundantly in epidermal cells of experimental embryos than in the AD of normal embryos , these results indicate that Gata . a can physically interact with Tcf7 and β-catenin in vivo . We also prepared recombinant proteins in E . coli , and confirmed that Gata . a interacted with β-catenin and Tcf7 ( Fig 8E ) . The interaction between Tcf7 and Gata . a was not affected by the presence of β-catenin ( Fig 8E ) . Thus , Gata . a can physically interact with Tcf7 and β-catenin .
We showed coordination of four maternal factors to activate the first zygotic gene expression ( Fig 9 ) . In Ciona , it is known that β-catenin activity is restricted to the VD . Therefore , β-catenin and Tcf7 activate gene expression in the VD . Because a complex of Gata . a , β-catenin , and Tcf7 interferes with Gata . a binding to Gata . a-binding sites , and the expression of Efna . d depends on Gata . a sites , Efna . d is not expressed in the VD . An interaction between TCF7L2 and GATA3 has been previously reported in human cell lines . This interaction is thought to tether TCF7L2 to the Gata3-binding site and repress transcription [28] . Drosophila Tcf can bind to sequences containing AGA[T/A]A[T/A] in addition to the canonical binding sequence [29] . However , such direct or indirect binding of Tcf7 to Gata sites was not detected by our ChIP analyses of the essential Gata sites of genes expressed in the animal hemisphere of Ciona embryos . Instead , in early Ciona embryos , formation of a complex of β-catenin , Tcf7 and Gata . a suppressed Gata . a-binding activity . In Xenopus , maternal β-catenin establishes the dorsoventral axis [30 , 31] . Gata-5 is expressed in early embryos under the control of maternal VegT and plays an important role in endoderm formation [32] . Hence , it is possible that the same mechanism operates in Xenopus embryos to establish clear boundaries of gene expression . β-catenin also regulates Tbx6 . b and Wnttun5 that are specifically expressed in the PVD . Tbx6 . b and Wnttun5 have repressive cis-regulatory elements that prevent activation of these genes by β-catenin and Tcf7 . Our data suggest that Zic-r . a can function in regulatory regions to overcome the repressive activity in the upstream sequences of Tbx6 . b and Wnttun5 without direct binding . First , within the essential regulatory region for specific expression of Tbx6 . b , no clear Zic-r . a-binding sites were found . Second , we observed weak ChIP peaks for Zic-r . a , which were not identified as peaks by the peak finding programs , around the essential regulatory regions of Tbx6 . b and Admp . Weak peaks may represent indirect binding of Zic-r . a , although ChIP peaks were hardly visible in the upstream sequence of Wnttun5 . Third , Zic-r . a physically interacted with Tcf7 , suggesting that Zic-r . a binds to the regulatory regions indirectly through Tcf7 . Because Zic functions in Xenopus as a co-factor of the transcription factor Gli [23] , and Zic forms a complex with Tcf4 in Xenopus and Caenorhabditis [33 , 34] , our finding is reasonable . Fourth , when the repressor elements were inserted , reporter constructs of Fgf9/16/20 and Foxd . b , which are usually expressed independently of Zic-r . a , evoked Zic-r . a-dependent expression in the PVD . It is unlikely that the repressive elements bind Zic-r . a , because clear ChIP peaks were not found in this region . Nevertheless , we do not rule out the possibility that Zic-r . a binds directly to the regulatory regions of other genes expressed specifically in the PVD . Even if this were so , Zic-r . a would suppress repressor activity . Indeed , the ChIP assays of Zic-r . a identified more than 3800 peaks over the whole genome , and the Zic-r . a motif was enriched within these peaks , indicating that Zic-r . a binds directly to DNA as shown previously [17] ( S11F Fig ) . The evidence that Tbx6 . b and Wnttun5 have a repressive element is persuasive . The Wnttun5 upstream sequence with the repressive activity contains several regions similar to the 15 bp repressive element of Tbx6 . b , suggesting that a transcription factor commonly binds to these repressive elements . A small number of genes are expressed at the 16-cell stage in patterns different from those of the genes examined in the present study . Foxa . a is expressed in the anterior animal cells and in the entire VD , Soxb1 is expressed in the AD and in the AVD , and Hes-a is expressed in the AD and VD [8] . Hes-a might have two different enhancers that are responsible for expression in the AD and VD . However , the results of the present study cannot explain the expression patterns of Foxa . a and Soxb1 . Thus , there may be additional maternal factors that regulate their expression . In conclusion , we revealed a mechanism in which three maternal factors coordinate to establish three distinct expression domains , and this coordination is mediated through interactions with Tcf7 . Our results show that negative regulatory mechanisms of the animal fate in the vegetal hemisphere and those of the posterior vegetal fate in the anterior vegetal hemisphere are important . Such negative regulatory mechanisms of developmental fates have not been well studied . However , they are likely to be important mechanisms of animal development .
C . intestinalis ( type A ) adults were obtained from the National Bio-Resource Project for Ciona . cDNA clones were obtained from our EST clone collection [35] . Whole-mount in situ hybridization was performed as described previously [8] . Identifiers for genes examined in the present study were as follows: CG . KH2012 . C8 . 396 for Foxd . b , CG . KH2012 . C8 . 890 for Foxd . a , CG . KH2012 . C2 . 125 for Fgf9/16/20 , CG . KH2012 . C3 . 716 for Efna . d , CG . KH2012 . S654 . 3 for Tbx6 . b , CG . KH2012 . L20 . 1 for Gata . a , CG . KH2012 . C1 . 727 for Zic-r . a ( Macho-1 ) , CG . KH2012 . C9 . 53 for β-catenin , CG . KH2012 . C6 . 71 for Tcf7 , CG . KH2012 . C9 . 257 for Wnttun5 , CG . KH2012 . C2 . 421 for Admp , CG . KH2012 . C3 . 411 for Lefty , CG . KH2012 . C4 . 547 for Gdf1/3-r , CG . KH2012 . C7 . 43 for Tfap2-r . b , CG . KH2012 . C6 . 162 for Fzd4 , and CG . KH2012 . C10 . 574 for Zfpm ( Fog ) . To identify potential Tcf7-binding sites , we used a position weight matrix available from the JASPAR database ( MA0523 ) [36] . We calculated scores over candidate regulatory regions and excluded candidates without the core ‘TTT’ sequence . For Gata sites , we manually inspected ‘GATA’ sequences . The MOs ( Gene Tools , LLC ) against Gata . a , β-catenin , and Zic-r . a , which block translation , have been used previously and their specificity evaluated [8 , 12 , 17] . We also used a standard control MO ( 5′-CCTCTTACCTCAGTTACAATTTATA-3′ ) purchased from Gene Tools , LLC . These MOs were introduced by microinjection under a microscope . For synthetic mRNAs , coding sequences of Gata . a , β-catenin , and Zic-r . a were cloned into pBluscript RN3 [37] , and synthetic mRNAs were transcribed using the mMESSAGE mMACHINE T3 Transcription Kit ( Life technologies ) . β-catenin protein encoded by the synthetic mRNA lacked the N-terminal 45 amino acids that included phosphorylation sites for GSK3 [38] . Therefore , the overexpressed protein was detected in nuclei and expected to function in a constitutive active form ( S12 Fig ) . Reporter constructs were introduced into fertilized eggs by electroporation . When they were introduced with MOs or mRNAs , we used microinjection . Chromosomal positions of the upstream sequences for each series of reporter constructs are indicated in S4 , S5 and S7 Figs . Deletions and mutations were introduced using the PrimeSTAR Mutagenesis Basal Kit ( Takara ) . The mutated sequences are indicated in S4 , S5 and S7 Figs . The Brachyury basal promoter consists of the chromosomal region , KhS1404:6203–6275 . We randomly picked up experimental embryos , and scored the number of cells that expressed the reporter mRNA in all retrieved embryos . We did not quantify the reporter mRNA level . All gene knockdown/overexpression experiments and reporter gene assays were performed at least twice with different batches of embryos . Coding sequences of Gata . a , Tcf7 , and Zic-r . a were cloned into pET16b ( Novagen ) , and His-tagged proteins were produced in E . coli . After purification with NiNTA resin ( Qiagen ) , recombinant proteins were used for immunizing rabbits and polyclonal antibodies were obtained . These antibodies recognized Gata . a , Tcf7 , and Zic-r . a as shown in S11A–S11C Fig , and the specific bands recognized by anti-Gata . a , -Tcf7 , and -Zic-r . a antibodies were diminished by pre-incubation with their antigens but not with the control protein , GFP ( S11D Fig ) . To detect protein localization , embryos were fixed with 3 . 7% formaldehyde in PBS , treated with 3% H2O2 for 30 minutes , and then incubated with the antibodies in Can-Get-Signal-Immunostain Solution B ( Toyobo ) . The signal was visualized with a TSA kit ( Invitrogen ) using horseradish peroxidase-conjugated goat anti-rabbit IgG and Alexa Fluor 488 tyramide . Control embryos incubated without primary antibodies yielded no signal . For ChIP assays , we collected 32-cell embryos from multiple batches . We used these multiple replicates for analysis , because a sufficient amount of material could not be obtained from a single batch . ChIP was performed as described previously [39] . Immunoprecipitated DNA was then split into two fractions . The first fraction was analyzed on a microarray as described previously [39] ( GEO accession number: GSE70902 ) . The second portion was subjected to sequence analysis using the Ion Plus Fragment Library Kit and a Ion PGM machine ( SRA accession number: DRA003742 ) . We obtained 1934656 , 2225624 , 3201484 , and 4254462 tags for Gata . a ChIP , Zic-r . a ChIP , Tcf7 ChIP , and the whole cell extract control , respectively . Analysis of the microarray data has been previously described [39] . In brief , peaks were called with two different programs independently [false discovery rate ( FDR ) : 1%] [40 , 41] , and only peaks identified by both programs were adopted . Sequence data were analyzed with the program package Homer [42] . Using this package , immunoprecipitation efficiencies of Gata . a , Zic-r . a , and Tcf7 were estimated as 14% , 5% , and 16% , respectively . We confirmed that Gata , Tcf7 , and Zic-r . a sites were successfully enriched in peaks identified in each immunoprecipitation using the mouse Gata-1 motif ( MA0035 . 2 in the JASPAR database [37] ) , human TCF7L2 motif ( MA0523 . 1 in the JASPAR database ) , and Ciona Zic-r . a motif [17] , respectively ( S11E–S11G Fig ) . Peaks were called without filtering based on local signals , because cis-regulatory elements are often densely encoded in the compact Ciona genome ( FDR: <0 . 1% ) . DNA immunoprecipitated with the anti-Gata . a antibody was subjected to quantitative PCR . An Efna . d upstream region ( KHC3:2 , 807 , 067–2 , 807 , 176 ) was amplified with the following primers: 5′-CAATATTGCACACGGACACAAT-3′ and 5′-GGTCGCTGTTCGCTATCTCTC-3′ . A control region ( KHC13:910 , 574–910 , 630 ) was amplified with the following primers: 5′-TCCTTGTGCAACAAGTCGCT-3′ and 5′-GCGGCACGAGGTGTATGAA-3′ . Gata . a , Zic-r . a , β-catenin , and Tcf7 with and without a 3×myc tag or 3×flag tag at their C-terminus were expressed in epidermal cells under the upstream sequence of Dlx . b as described previously [25] . β-catenin protein encoded by the Dlx . b>β-catenin construct lacked the N-terminal 45 amino acids that included phosphorylation sites for GSK3 [38] . Therefore , the overexpressed protein was expected to function in a constitutively active form . The lysates obtained from the resultant tailbud-stage embryos were used in co-immunoprecipitation assays with an anti-myc antibody ( Abcam , ab9106 ) and the Dynabeads Co-Immunoprecipitation Kit ( Life technologies ) . Immunoprecipitated samples were resolved on a sodium dodecyl sulfate-polyacrylamide gel and then subjected to western blot analysis with anti-FLAG ( Sigma , F1840 ) and anti-β-catenin ( a kind gift from Prof . Hiroki Nishida , Osaka University , Japan ) antibodies . Signal detection was carried out using horseradish-peroxidase-labeled anti-mouse or rabbit IgG and an ECL-kit ( GE Healthcare ) . Recombinant Gata . a , Zic-r . a , β-catenin , and Tcf7 proteins with a 3×myc tag or 3×flag tag at their C-terminus were produced in E . coli using pET16b , and purified with NiNTA agarose ( Qiagen ) . Co-immunoprecipitation assays and detection of immunoprecipitated samples were performed as described above . Tcf7 protein for assays shown in Figs 3 , 4 , S4 and S5 was expressed as a GST-fusion protein in E . coli and purified with glutathione sepharose ( GE healthcare ) . A double-stranded DNA for a Tcf7 site in the upstream sequence of Fgf9/16/20 was prepared from the following oligonucleotides: 5′-AAAGTTCACCGACAAAGATAAGA-3′ and 5′-AAATCTTATCTTTGTCGGTGAAC-3′ . The protruding ends were filled with biotin-11-dUTP ( Thermo Fisher Scientific ) using Taq DNA polymerase . Other probes were similarly prepared and their sequences are indicated in S4 and S5 Figs . Proteins and the biotin-labeled double-stranded DNA or unlabeled double-stranded DNA were mixed in 10 mM Tris ( pH 7 . 5 ) , 50 mM KCl , 1 mM DTT , 2 . 5 mM EDTA , 50 ng/μL poly ( dAdT ) , 0 . 05% NP40 , and 1 μg of the recombinant Tcf7-Gst fusion protein or GST protein . After incubation for 20 min , protein-DNA complexes were resolved by electrophoresis on a 6% native polyacrylamide gel in 0 . 5× TBE and then transferred to a nylon membrane . The Chemiluminescent Nucleic Acid Detection Module Kit ( Thermo Fisher Scientific ) was used to detect protein-DNA complexes . Recombinant proteins for assays shown in Figs 7 and S8 were separately produced with a rabbit reticulocyte lysate system ( TnT T7 Quick Coupled Transcription/Translation System , Promega ) . Probes were prepared using digoxigenin-11-dUTP ( Roche ) . Their sequences are indicated in S7 Fig . Proteins and the digoxigenin-labeled double-stranded DNA or unlabeled double-stranded DNA were mixed in 10 mM Tris ( pH 7 . 5 ) , 50 mM KCl , 1 mM DTT , 5 mM MgCl2 , 50 ng/μL poly ( dIdC ) , and 0 . 05% NP40 . The amount of proteins was determined empirically . Protein-DNA complexes were detected with an alkaline phosphatase-conjugated anti-digoxigenin antibody ( Roche ) and CDP-star ( Roche ) . Bands were quantified as arbitrary units by a molecular imager ( ChemiDoc XRS ) using Quantity One software ( Promega ) . Each experiment was independently performed two or three times . | During animal development , transcription factors and signaling molecules transcriptionally regulate one another and constitute a gene regulatory network . This network is evoked by maternally provided factors . Many maternal factors are localized and thereby activate a set of genes in a specific region . In embryos of the chordate , Ciona intestinalis , three maternal factors with localized activities are known . The present study demonstrated that these localized maternal factors interact with one another through a fourth non-localized transcription factor , Tcf7 , and negatively regulate one another . These repressive interactions are essential to establish the first distinct domains of gene expression and evoke the gene regulatory network properly . The findings indicate that not only activating target genes but also repressing activities of other transcription factors through protein-protein interactions are important to properly initiate the zygotic program . Intriguingly , in one repressive interaction , a transcription factor loses its binding activity for target sites through an interaction with another transcription factor . Thus , this study provides a description of the entire system in which maternal factors initiate the zygotic developmental program of the Ciona embryo . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"sequencing",
"techniques",
"gene",
"regulation",
"reporter",
"genes",
"developmental",
"biology",
"immunoprecipitation",
"gene",
"types",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"embryos",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"embryology",
"animal",
"cells",
"gene",
"expression",
"molecular",
"biology",
"precipitation",
"techniques",
"dna",
"sequence",
"analysis",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"blastomeres"
]
| 2016 | A Maternal System Initiating the Zygotic Developmental Program through Combinatorial Repression in the Ascidian Embryo |
Despite the formidable mutational capacity and sequence diversity of HIV-1 , evidence suggests that viral evolution in response to specific selective pressures follows generally predictable mutational pathways . Population-based analyses of clinically derived HIV sequences may be used to identify immune escape mutations in viral genes; however , prior attempts to identify such mutations have been complicated by the inability to discriminate active immune selection from virus founder effects . Furthermore , the association between mutations arising under in vivo immune selection and disease progression for highly variable pathogens such as HIV-1 remains incompletely understood . We applied a viral lineage-corrected analytical method to investigate HLA class I-associated sequence imprinting in HIV protease , reverse transcriptase ( RT ) , Vpr , and Nef in a large cohort of chronically infected , antiretrovirally naïve individuals . A total of 478 unique HLA-associated polymorphisms were observed and organized into a series of “escape maps , ” which identify known and putative cytotoxic T lymphocyte ( CTL ) epitopes under selection pressure in vivo . Our data indicate that pathways to immune escape are predictable based on host HLA class I profile , and that epitope anchor residues are not the preferred sites of CTL escape . Results reveal differential contributions of immune imprinting to viral gene diversity , with Nef exhibiting far greater evidence for HLA class I-mediated selection compared to other genes . Moreover , these data reveal a significant , dose-dependent inverse correlation between HLA-associated polymorphisms and HIV disease stage as estimated by CD4+ T cell count . Identification of specific sites and patterns of HLA-associated polymorphisms across HIV protease , RT , Vpr , and Nef illuminates regions of the genes encoding these products under active immune selection pressure in vivo . The high density of HLA-associated polymorphisms in Nef compared to other genes investigated indicates differential HLA class I-driven evolution in different viral genes . The relationship between HLA class I-associated polymorphisms and lower CD4+ cell count suggests that immune escape correlates with disease status , supporting an essential role of maintenance of effective CTL responses in immune control of HIV-1 . The design of preventative and therapeutic CTL-based vaccine approaches could incorporate information on predictable escape pathways .
Genetic variation within the highly polymorphic human leukocyte antigen ( HLA ) class I region contributes to diversity of pathogen recognition by cytotoxic T lymphocytes ( CTLs ) [1] , and acts as a selective force shaping viral evolution within an infected host [2–6] through selection of mutations that allow the virus to escape recognition by HLA-restricted CTLs [5 , 7–9] . Immune escape may also represent a significant force shaping viral evolution at the population level through an HLA “imprinting effect , ” in which escape mutations selected in the context of common HLA class I alleles may become predominant in the circulating viral population if they do not revert upon transmission to new hosts [2 , 10 , 11] . One of the major challenges to HIV vaccine design is the extensive worldwide sequence diversity of this pathogen , fueled in part by the extreme mutational capacity of the virus [12] . However , despite this considerable diversity , evidence indicates that there are constraints on viral evolution [2 , 13 , 14] , and that escape in response to specific immune selective pressures ( similar to escape from drug selective pressures [15] ) follows broadly predictable mutational patterns [13 , 14] . A comprehensive identification of specific sites and patterns of immune escape in clinical HIV-1 isolates will further our understanding of how immune selection contributes to viral diversity [2 , 16] and will also identify specific viral regions under active immune selection pressure , thus providing information relevant to the selection of candidate immunogens for an HIV-1 vaccine . Improvements in DNA sequencing technologies and the availability of large cohorts of HIV-1 infected individuals now allow us to employ population-based genetic association approaches to identify viral amino acids ( aa ) under active immune selection pressure in vivo [2]; however , methodological challenges associated with identifying such mutations are now recognized [16] . Moore et al . [2] were the first to identify HLA-class I-associated polymorphisms across codons 20–227 of HIV-1 reverse transcriptase ( RT ) in a large clinically derived dataset using a Chi-squared association approach , thus providing evidence for HLA class I-mediated viral evolution on a population level . However , the application of standard statistical tests is inappropriate for the analysis of viral isolates with a shared phylogenetic history , since descent from a common ancestor means that viral sequences may not be treated as statistically independent entities [17] . Specifically , a cause for concern is the application of standard statistical methods to identify HLA-associated viral polymorphisms in cohorts comprising individuals of diverse genetic backgrounds ( sampled from populations with differential HLA allele distributions ) infected with heterogeneous viral strains . In this case , standard statistical approaches such as the Chi-squared test may identify confounding associations between strain- or lineage-specific viral polymorphisms and specific HLA alleles that are over-represented in subpopulations of individuals harboring infections with those strains . In this case , the observed “HLA-associated polymorphism” is not evidence of active HLA-mediated immune selection . Rather , the association is simply a statistical correlation between possession of a particular HLA allele observed among persons of a particular ethnic background , and a lineage-specific viral polymorphism , arising as a result of descent from a common ancestor ( “founder effect” ) [16] . The use of population-based , viral lineage-corrected analyses , such as those recently developed by Bhattacharya et al . [16] , are therefore essential in order to accurately identify sites of active immune selection in the genomes of sequence-diverse pathogens such as HIV-1 . In addition , although there is clear evidence supporting HIV-1 adaptation to HLA class I-mediated CTL selection pressure from an evolutionary standpoint [2–5] , the relevance of immune escape to clinical HIV disease progression remains unclear , due in part to the fact that many studies have focused small numbers of participants and/or escape within a limited number of HLA-restricted CTL epitopes in the viral genome [3–6 , 18–20] . Furthermore , no studies to date have linked HIV disease progression to HLA-associated polymorphisms corrected for lineage effects . Here we identify lineage-corrected [16] HLA class I-associated polymorphisms across select functional and accessory/regulatory HIV-1 genes in a cross-sectional analysis of a large cohort of chronically infected , treatment-naïve individuals , and investigate the relationship between these polymorphisms and clinical markers of HIV disease .
A large , well-characterized cohort of chronically HIV-1 infected , antiretroviral drug-naïve individuals from British Columbia , Canada [21] , for whom HLA class I typing and HIV RNA genotyping of select functional and accessory/regulatory genes were performed , was used to identify HLA class I allele-associated viral polymorphisms across a 499 aa fragment spanning protease and most of RT p51 ( n = 532 successfully genotyped ) , 96 aa of Vpr ( n = 425 ) , and 206 aa of Nef ( n = 686 ) . HLA class I allele-associated viral polymorphisms were identified using analytical approaches described in [16] , which feature a correction for viral lineage effects by adjusting for phylogenetic relationships between sequences [16] , and a correction for multiple comparisons using a q-value approach [22] , which sets the false-discovery rate ( 20% with q < 0 . 2 ) among significant associations . The level of variation at single residues in protease , RT , Vpr , and Nef ranged from 0% to maxima of 50% , 57% , 73% , and 77% respectively ( Figure 1 ) , while the mean pairwise amino acid identity for these same genes ( calculated as the percentage of codons exhibiting identical amino acids for each pairwise combination of sequences ) was 92 . 9% , 95 . 3% , 89 . 0% , and 83 . 1% , respectively , indicating typical intrasubtype levels of HIV sequence diversity in this cohort of relatively homogeneous subtype distribution ( 97 . 5% HIV-1 subtype B ) . It is important to note that the phylogenetically corrected methods for identification of HLA-associated viral polymorphisms developed by Bhattacharya et al . [16] do more than simply correct for confounding due to HIV intersubtype ( or interclade ) variation . Even among clade-homogeneous datasets , “subclade” lineage-specific effects may yield confounding associations with HLA alleles , especially if the cohort is composed of subpopulations with differential HLA allele distributions . Indeed , there was clear evidence of phylogenetic subclusters within subtype B sequences in this cohort ( Figure S1 ) . We therefore compared phylogenetically corrected methods to a simple uncorrected test ( simple Fisher ) , and found that even in this predominantly subtype B-infected cohort , HLA-associated polymorphisms identified using phylogenetically corrected methods had higher fractions of associations that could be independently validated by immunological data than those defined using a simple Fisher exact test ( unpublished data ) . An example of a case in which an apparent HLA-associated polymorphism identified using a simple Fisher exact test represents an artifact of the phylogenetic tree is illustrated in Figure S1 . In this analysis , therefore , we report only HLA-associated polymorphisms defined using phylogeny-based methods [16] . Application of the viral lineage-corrected method [16] yielded a total of 478 unique HLA allele-associated viral polymorphisms with q < 0 . 2 across the genes investigated ( Table S1 ) . These occurred at nine ( 9% ) , 28 ( 7% ) , 12 ( 12 . 5% ) , and 84 ( 41% ) unique codons in protease , RT , Vpr , and Nef , respectively , highlighting a dramatic variation in HLA-associated imprinting across HIV-1 genes ( Figure 1 ) . HLA-B alleles accounted for half ( 241 of 478; 50 . 4% ) of the total number of HLA-associated polymorphisms , while HLA-A and C alleles accounted for 112 ( 23 . 4% ) and 125 ( 26 . 2% ) , respectively . Previous studies have validated the application of such genetic association analyses of large clinically derived datasets in order to identify HLA-restricted CTL escape mutations selected in vivo [16] . Knowing that with q < 0 . 2 , about 20% of identified HLA-associated polymorphisms will represent false-positive results , we set about classifying the 478 identified polymorphisms into putative true-positive or false-positive results based on the strength of independent biological evidence supporting each polymorphism as an escape-associated mutation . The highest level of biological support was assigned to HLA-associated polymorphisms falling within or proximal to ( ± 3 aa ) a published CTL epitope [23] restricted by that particular HLA allele , thereby supporting these associations as in vivo-selected mutations directly or indirectly affecting MHC binding , T cell receptor recognition and/or intracellular peptide processing [24–26] . A second level of support was assigned to those associations falling within or similarly proximal to putative/novel HLA-restricted epitopes , identified by scanning the cohort consensus sequence for HLA-restricted epitope anchor residue motifs using two independent bioinformatic tools ( MotifScan [Los Alamos National Laboratory] , http://www . hiv . lanl . gov/content/immunology/motif_scan/motif_scan; and Epipred [Microsoft Research] , http://atom . research . microsoft . com/bio/epipred . aspx ) . To provide further biological support for these associations we drew upon an independent cohort of 372 HIV-1 infected individuals screened for in vitro HLA-restricted , CTL-mediated interferon-gamma ( IFN-γ ) responses against a set of overlapping HIV-1 subtype B consensus peptides spanning the entire viral proteome using the IFN-γ enzyme-linked immunosorbent spot assay ( ELISpot ) [27] , in order to identify HLA class I alleles significantly associated with CTL-mediated IFN-γ production in response to stimulation with consensus HIV peptides ( see Methods ) . HLA allele-associated polymorphisms mapping within a significantly reactive HLA allele/HIV consensus peptide pair were identified as potential escape-associated mutations to known or novel HLA-restricted CTL epitopes . Finally , we grouped together HLA allele-specific associations clustering within these epitopes or motifs , and paired together alleles in linkage disequilibrium ( Table S2 ) associated with the same HIV polymorphism ( s ) , to create a series of immune escape maps capturing the minimum number of HLA-restricted epitopes and/or motifs required to explain the data ( Figures 2–5 ) . Associations that did not map within a known epitope or motif , and were not supported by ELISpot data or attributable to HLA allele linkage , were listed in a separate map ( Figure 6 ) . After pairing together linked alleles , approximately 35% of codons in protease , RT , Vpr , and Nef exhibiting HLA-associated polymorphisms mapped inside ( n = 77; 81% ) or within ± 3 aa ( n = 18; 19% ) of a published CTL epitope specific to that HLA allele ( Figures 2 and 3 ) . Significant associations were collapsed into two categories based on the direction of the HLA selection pressure: amino acids enriched in the presence of a specific allele ( positive or “escape” correlations , presumably representing the escape variant for that allele ) , and amino acids depleted in the presence of a specific allele ( negative or “reversion” correlations , presumably representing the immunologically susceptible or “wild-type” form for that allele , and also representing the amino acid to which the residue will likely revert to upon transmission an individual lacking that allele ) . Overall , the majority of HLA-associated polymorphisms ( 58% of epitope-supported associations ) represent negative ( “reversion” ) correlations ( p = 0 . 002 ) . Note that detection of a “reversion” correlation in the absence of an associated “escape” correlation may arise in the case where a specific allele selects for multiple amino acids at a given position , creating a situation where there may be sufficient statistical power to detect the “reversion” correlation but not to identify all possible escape variants . A considerable number of codons exhibit multiple HLA associations , particularly in Nef . A total of 57 multiple associations were observed , with 2 , 6 , 3 , and 46 occurring across protease , RT , Vpr , and Nef , respectively . In 23 of these 57 cases ( for example , Nef codons 81 and 135 ) , the same amino acid represents an escape variant for one HLA allele , but the susceptible form for another , highlighting a “tug-of-war” of differential HLA selective pressures contributing to populational HIV sequence diversity at specific codons . There were dramatic differences in the number of HLA-associated polymorphisms across the genes investigated . Not only did Nef exhibit a much higher density of epitope-supported associations compared to protease/RT and Vpr , but the escape patterns also tended to be more complex in Nef than in other genes . A total of 53% of escaping epitopes in Nef exhibited HLA-associated polymorphisms at multiple positions within the epitope , compared with 12% and 0% in protease/RT and Vpr , respectively ( p = 0 . 004 ) . Similarly , epitope-proximal associations ( occurring within 3 aa of a published epitope ) were also observed more frequently in Nef ( n = 16 [22%] ) while occurring only relatively rarely in protease/RT/Vpr ( total n = 2 [9%] ) , although this did not achieve statistical significance ( p = 0 . 2 ) . Overall , HLA-associated polymorphisms were observed with relatively equal frequency across all positions within published HLA-restricted epitopes . There was no statistically significant enrichment for HLA-associated polymorphisms at anchor residues ( generally defined as epitope residues 2 and C-terminal with some exceptions [28 , 29] ) over other residues in Nef ( p = 0 . 7 ) or protease/RT/Vpr ( p > 0 . 1 ) suggesting that amino acid changes potentially affecting peptide binding to HLA class I molecules are not a favored mechanism of escape . We organized a further ~50% of the identified associations into “motif-support” maps ( Figures 4 and 5 ) that grouped HLA-associated polymorphisms within HLA-restricted epitope anchor residue motifs identified by scanning the cohort consensus sequence . Based on evidence that HLA-associated polymorphisms identified in genetic association studies predict the location of previously uncharacterized epitopes [16] , we would expect that a substantial proportion of motif-supported associations represent escape mutations within novel epitopes , a hypothesis supported by the fact that many motif-supported associations ( 40% , 31% , 22% , and 19% in protease , RT , Vpr , and Nef , respectively ) are substantiated by in vitro IFN-γ ELISpot responses to HIV-specific consensus peptides containing these motifs . Consistent with observations drawn from the epitope-support maps ( Figures 2 and 3 ) , the majority ( 63% ) of associations in the motif-support maps represent “reversion” associations , with a much more complex pattern of escape observed in Nef compared to protease/RT/Vpr . The remaining ~15% of HLA-associated polymorphisms did not map to known epitopes and were unlikely to lie within or proximal to novel epitopes as suggested by in vitro IFN-γ ELISpot responses or bioinformatic motif scans ( Figure 6 ) . Although these proportions are consistent with the false-discovery rate of ~20% ( q < 0 . 2 ) , lack of biological support cannot be used to definitively categorize these as “false-positive” associations in any particular case . In some cases , these may represent processing escape mutations occurring distant from the epitope site , compensatory mutations , unusual epitopes , or other factors . Similarly , HLA-associated polymorphisms mapping within an HLA-matched epitope or motif are likely highly enriched for mutations directly or indirectly conferring immune escape , but likely contain smaller numbers of false-positive associations as well . Although there is clear evidence documenting the selection of escape variants over the course of HIV infection [3–5 , 7–9] , the clinical significance of immune escape remains incompletely understood [18–20] . Moore et al . reported a significant association between HLA-associated polymorphisms and plasma viral load [2]; however , no studies to date have linked lineage-corrected HLA-associated polymorphisms with markers of disease progression on a population basis . We therefore investigated correlations between the presence of HLA-associated polymorphisms and clinical status in chronic untreated infection as measured by pretherapy CD4+ cell number and plasma viral load . In order to adopt the most conservative definition of “escape , ” the primary analysis was restricted to those amino acid associations mapping inside or within ± 3 aa of a known HLA-restricted CTL epitope ( Figures 2 and 3 ) . A significant inverse dose–dependent relationship was observed between the median pretherapy CD4+ cell count and the number of epitope-associated polymorphisms observed in protease/RT ( p = 0 . 006 ) , Vpr ( p = 0 . 01 ) , and Nef ( p = 0 . 008 ) ( Figure 7 ) . A trend was observed between accumulation of epitope-associated polymorphisms in protease/RT ( but not other proteins ) and higher pretherapy viral load ( p = 0 . 06 [unpublished data] ) . The dose-dependent association between epitope-associated polymorphisms and lower CD4+ cell counts supports the ability of large genetic association studies to identify biologically relevant in vivo CTL escape-associated mutations , but more importantly , supports a clinically relevant link between immune escape and HIV disease progression . Note that the observed association between HLA-associated polymorphisms and lower CD4+ cell count is specific to HLA-associated polymorphisms mapping within or near published epitopes , and not simply a general association between viral mutations and HIV clinical status . In a secondary analysis we investigated correlations between the presence of motif-associated ( Figures 4 and 5 ) and unsupported ( Figure 6 ) polymorphisms and clinical parameters . A nonsignificant trend ( p = 0 . 07 ) was observed between accumulation of motif-associated polymorphisms in protease/RT ( but not other proteins ) and lower median CD4+ cell counts , while no significant association was observed between clinical parameters and the presence of biologically unsupported associations , consistent with a stepwise enrichment for false-positives among associations in these categories .
The present study represents to our knowledge the largest population-based investigation of HLA class I-mediated imprinting on HIV sequence to date , as well as the first characterization of HLA-associated polymorphisms in each of a functional , accessory and regulatory gene . Results identify viral polymorphisms selected in vivo in context of a wide array of class I alleles . The confirmation of the B*1501-associated polymorphism at protease codon 93 reported by Bhattacharya et al . [16] and several reported by Moore et al . [2] in RT suggest that immune escape patterns in HIV-1 subtype B are consistent across the globe . The confirmation of several functionally verified CTL escape mutations previously observed in clinically derived isolates ( including escape at residues 2 , 8 , 2 , and 5 of the HLA-B*57 restricted IW9-RT [13 , 30] , B*51-restricted TI8-RT [31] , A*24-restricted RF10-nef [32] , and B*08-restricted FL8-nef [5] epitopes , respectively ) confirm the utility of genetic association studies to identify escape variants commonly selected in vivo . Taken together , results provide proof of principle that population-based approaches could complement smaller functional studies by providing a whole-gene or whole-virus picture of immune escape . Results of this large-scale , multigene analysis reveal dramatically different levels of HLA-associated polymorphisms across HIV proteins , with a previously unreported , extraordinary density and complexity of HLA-associated polymorphisms in Nef . Nef exhibits considerable sequence diversity and thus may exhibit higher levels of mutational plasticity in response to selective pressures compared to genes exhibiting structural ( e . g . , Gag ) or functional ( e . g . , protease/RT ) constraints; however it is important to note that protease ( and to a lesser extent RT ) exhibit extensive mutational capacity under antiretrovirally mediated selection pressure [15] , suggesting that mutational constraints on functional genes are unlikely to fully account for the relative paucity of HLA-associated polymorphisms across these regions . Rather , results are consistent with the density of CTL epitopes across these regions , as well as the relative immunogenicity of these proteins over the course of infection [27 , 33] . Limited data from longitudinal studies suggest that CTL escape mutations in Nef are selected earlier in infection [33 , 34] , and thus , in a population of chronically infected individuals , one may expect a large burden of Nef escape mutations to have already accumulated . Note that , in the current study , Nef sequences were available for a larger number of participants , thus potentially increasing power to detect significant associations . These data are also relevant to CTL-based HIV vaccine design . First and foremost , the analysis of clinically derived datasets identifies viral epitopes under active immune selection pressure , thus identifying in vivo immunogenic viral targets . The fact that we observed such a large number of HLA-associated polymorphisms , including many instances of specific codons apparently under diametrically opposed HLA-selective pressures ( an observation consistent with Iversen et al . [35] ) , provide some evidence against the complete disappearance of all active viral epitopes under the HLA “imprinting hypothesis” ( which states that escape mutations selected in response to the most common HLA alleles may become fixed in the circulating viral population [2 , 10] , thus resulting in a potential loss of CTL responses to these epitopes and rendering them inappropriate as candidate immunogens ) . Taken together with evidence supporting rapid reversion of escape mutations after transmission to a new host [36] , and the fact that escape mutations in one individual may represent the susceptible form in another [16] , the “HLA imprinting effect” is unlikely to result in the creation of an immunologically refractive circulating viral population by eliminating all active CTL epitopes in this population . Rather , selection pressures mediated by diverse HLA class I alleles in HIV-1 infected populations appear to be actively contributing to viral diversity thus preserving a substantial number of immunologically active epitopes in the circulating population . These active epitopes , most notably those which exhibit the “push-and-pull” of diametrically opposed HLA selection pressures , could perhaps be incorporated into a CTL-based HIV-1 vaccine strategy . The locations of HLA-associated polymorphisms relative to known or predicted HLA-appropriate epitopes revealed no statistically significant enrichment for mutations at epitope anchor residues versus other positions . Theoretically , if the predominant mechanism of CTL escapes were abrogation of peptide-MHC binding through anchor residue mutation , a polyvalent vaccine approach may have little merit . However , these observations , combined with previous documentation of de novo T cell responses arising in response to escape variants [37] , strongly support the utility of incorporating viral sequence variation into immunogen design . Given the adaptable nature of the CTL response [37] , combined with the fact that the majority of reports of CTL escape to date have focused on small numbers of individuals and/or a select few epitopes [3–6 , 18–20 , 30–32 , 35] , it is not surprising that the clinical consequences of CTL escape remain incompletely understood . Some studies report an association between selection of escape variants and loss of viremia control [20 , 38] and disease progression [6 , 18]; however , this does not seem to equally apply to all CTL epitopes [35 , 39] . Here we observe a significant , dose-dependent inverse relationship between HLA-associated mutations within published epitopes in functional and accessory/regulatory genes and lower CD4+ cell counts in chronic untreated HIV infection , thus supporting a link between presence of escape mutations and HIV disease status . Although detection of escape mutations indeed preceded a loss of immune control in previous case reports [6 , 18] , it is important to note that the cross-sectional nature of the current study precludes any inferences regarding cause and effect . Likely , a longer duration of infection ( among those with lower CD4+ cell counts in this cohort ) may have facilitated the accumulation of CTL escape variants , a hypothesis we were unable to investigate , because seroconversion dates were generally unknown . Other limitations of this analysis include the inherent limitations associated with the use of a single CD4+ cell measurement in a cross-sectional study design , the lack of longitudinal HIV sequence data , as well as the fact that the cohort represents a group of individuals referred for antiretroviral treatment , and thus may be biased toward more rapid progression to disease . Despite these limitations , our findings support those of Moore et al . [2] who reported that HLA-associated polymorphisms in RT predicted plasma viral load ( CD4+ cell counts were not investigated ) . At first , results appear inconsistent with those of Iversen et al . [35] who reported higher viral loads in patients with efficient CTL selection; however , results may be reconciled by the fact that the previous study [35] investigated clinical correlates of escape to a single HLA-restricted epitope , whereas the current study evaluates HLA-associated polymorphisms across multiple genes . Ideally , however , the relationship between selection of HLA-associated escape mutations and HIV disease progression should be addressed in an unbiased , longitudinal cohort study of untreated HIV-1 seroconverters for whom infection dates , viral loads and CD4+ T cell setpoints , and rates of disease progression are known . Although a systematic in vitro characterization of novel CTL epitopes was beyond the scope of this manuscript , the observation that a substantial number of motif-associated polymorphisms are supported by HLA-restricted , peptide-specific IFN-γ responses in an ELISpot assay suggest that they represent escape mutations within uncharacterized epitopes [16] . As the locations of published epitopes tend to be biased toward conserved regions ( due to the historic use of consensus or reference strains to construct peptide libraries ) , the “motif maps” could complement traditional epitope mapping by identifying epitopes located in more variable regions . After controlling for the potentially confounding effects of viral lineage [16] , strong evidence for HLA class I-mediated selection is observed across functional and accessory/regulatory HIV-1 genes , with up to 40% of residues in some HIV proteins ( Nef , for example ) exhibiting evidence for HLA-restricted immune selection . Our results thus confirm an active and substantial contribution of human immunogenetic selection pressure on viral evolution [2] and underscore the importance of understanding how HLA class I diversity drives HIV diversity . The observed correlation between the presence of HLA-associated CTL escape mutations and lower CD4+ cell counts supports the hypothesis that maintenance of effective CTL responses plays an important role in immune control of HIV infection , although further research in additional cohorts is needed . The observation that epitope anchor residue mutation appears not to be the predominant mechanism of CTL escape supports the incorporation of HIV sequence diversity in the development of preventative and therapeutic CTL-based vaccine approaches .
In British Columbia ( BC ) , antiretroviral drugs are distributed free of charge to HIV-infected individuals through a centralized drug treatment program ( for details , see [21] ) . The HAART Observational Medical Evaluation and Research ( HOMER ) cohort is an open cohort comprising all HIV-infected , antiretroviral-naïve adults who initiated HAART since August 1996 ( n > 2 , 200 individuals enrolled to date ) . A subset of HOMER , comprising all treatment-naïve individuals who initiated HAART in BC between August 1996 and September 1999 ( n = 1 , 191 ) has been described in detail previously [21] . Participants in the current cross-sectional study represent a nonrandom subset ( n = 765; 64% ) of these 1 , 191 individuals at baseline ( prior to initiation of HAART ) included based on the availability of a peripheral blood sample for HLA typing . A comparison of pre-therapy characteristics of those included ( n = 765 ) and excluded ( n = 426 ) reveals no significant differences in pretherapy CD4+ cell count ( 280 cells/mm3 ) ; however , those included had slightly lower pretherapy plasma viral load ( pVL ) ( median 5 . 07 versus 5 . 15 log10 copies HIV RNA/ml , p = 0 . 03 ) , were on average slightly older ( median 37 . 2 versus 36 . 5 y , p = 0 . 02 ) , and were more likely to be male ( median 88% versus 77% male , p < 0 . 0001 ) than those excluded . CD4+ cell count , plasma viral load , and HIV genotype data for each participant represent the latest pre-therapy measurement collected within 180 d prior to HAART initiation . Ethical approval for this study was granted by the Providence Health Care/University of British Columbia Research Ethics Board . HIV RNA was extracted from a single pre-therapy ( “baseline” ) plasma sample using the QIAGEN ( http://www . qiagen . com ) viral RNA kit using a BioRobot 9600/9604 or extracted manually using guanidinium-based buffer followed by isopropanol/ethanol washes . The HIV protease ( codons 1–99 , HXB2 nt 2253–2549 ) , RT ( codons 1–400 , or for ~25% of sequences codons 1–240 only; nt 2550–3749 or 2550–3269 , respectively ) , Vpr ( codons 1–96; nt 5559–5847 ) , and Nef ( codons 1–206; nt 8797–9414 ) were amplified using nested RT-PCR , and “bulk” sequenced in both the 5′ and 3′ directions on an ABI 3700 or 3100 ( http://www . appliedbiosystems . com ) automated DNA sequencer . HIV sequence data were analyzed using the software Sequencher ( Genecodes , http://www . genecodes . com ) . Nucleotide mixtures were called if the height of the secondary peak exceeded 25% of the height of the dominant peak . Sequence data were aligned to HIV-1 subtype B reference strain HXB2 ( Genbank accession number K03455 ) using a modified NAP algorithm [40] . HIV subtyping was performed by comparing HIV sequence data across HIV protease , RT , and Nef to all known subtype reference sequences in the Los Alamos HIV sequence database ( http://hiv-web . lanl . gov/content/hiv-db/mainpage . html ) . Of total participants , 97 . 5% harbored subtype B infections . Consensus sequences reflecting the most common amino acid at each codon were generated from pretherapy sequences: these differed from the 2005 HIV-1 subtype B consensus at 1/99 protease , 7/400 RT , 2/96 Vpr , and 7/206 Nef codons , respectively . Genbank accession numbers of all unique HIV sequences used in this study are listed in Text S1 . Sequence-based typing ( SBT ) for HLA-A , B , and C was performed on DNA extracted from a PBMC-enriched frozen blood sample for each participant ( n = 765 ) . The SBT protocol is a validated “in-house” procedure based on International Histocompatibility Working Group ( IHWG ) protocols and involves independent , locus-specific , nested PCR amplification of exons 2 and 3 of HLA-A , B , and C followed by automated bidirectional DNA sequencing . Allele interpretation was performed by comparing SBT data against all alleles listed in the IMGT/HLA database ( ftp://ftp . ebi . ac . uk/pub/databases/imgt/mhc/hla/ ) as of August 2005 ( Release 2 . 10 ) . This yields intermediate-to-high level resolution of HLA allele combinations . In order to achieve appropriately sized groups for statistical analysis , HLA alleles were summarized to two-digit resolution; note however that this approach may group together alleles which bind slightly different peptides , thus potentially reducing power to detect HLA-associated polymorphisms in some cases . Ambiguous allele combinations were resolved through incorporation of published allele frequencies and/or haplotype data . HLA-A and B typing was completed for all 765 participants , while HLA-C types were determined for 706 individuals . Although complete ethnicity data are unavailable , class I allele frequencies were consistent with those expected in a predominantly North American white population . In order to discriminate between associations likely attributable to viral lineage effects and those that provide evidence for HLA-associated escape or reversion , we adopted the phylogenetically corrected analysis methods described in detail in [16] . Briefly , we used cohort HIV sequences to construct maximum likelihood phylogenetic trees ( one for each gene ) . Since HLA types are available only for the infected individuals sampled , whose sequences form the tips of the tree , we used a maximum likelihood estimate of the sequence at the parental ( interior ) node proximate to each observation , and counted inferred escape or reversion in these last branches as independent events to be correlated with the HLA of the infected person at the terminal sequence by a Fisher exact test ( method 1 ) ; alternatively , we used a likelihood ratio test to evaluate whether a model incorporating the effect of HLA association in addition to the phylogenetic structure was significantly better at explaining the data ( method 2 ) [16] . The final list of identified associations represents the union of associations identified by both methods ( Table S1 ) . In order to adjust for multiple comparisons , a q-value approach [22] , rather than a Bonferroni correction [41] , was employed: whereas a Bonferroni correction attempts to limit the probability of even a single false positive ( and thus increases the rate of false-negative results ) , the q-statistic sets the proportion of false positives among results identified as significant ( the false-discovery rate ) , an approach which we believe to be more appropriate for gene-wide association scans such as the present one . Associations with q < 0 . 2 ( indicating a ~20% false-discovery rate ) are presented; in this dataset this corresponded to unadjusted p-values 0 . 0055 > p > 3 . 3x10−45 for all genes . Note that the results of the lineage-corrected analysis groups associations into two broad categories based on the direction of the HLA selection pressure . Positive correlations , in which the presence of a specific HLA is associated with the presence of a particular amino acid—or , correspondingly , where the absence of the allele is associated with the absence of the amino acid—are termed “escape” associations , as they presumably reflect the escape variant for that specific HLA allele . Negative correlations , in which the presence of a specific HLA allele is associated with the absence of a particular amino acid—or , correspondingly , where a specific amino acid is enriched in the absence of a particular HLA allele—are termed “reversion” associations . In this case , the “reversion” amino acid presumably reflects the immunologically susceptible ( “wild-type” ) form specific for that HLA allele , as well as represents the amino acid most likely to re-emerge following transmission to an individual lacking that HLA allele . Associations were organized into gene-specific “immune escape maps” whose goal was to capture the minimum number of epitopes ( known or putative ) required to explain the data . Three sets of escape maps were generated based on the strength of biological evidence supporting each association . The highest level of support was granted to those associations that fell within or proximal to ( ± 3 aa ) a published HLA-restricted epitope ( defined as all HLA class 1-restricted ≤ 15-mer epitopes listed in the Los Alamos HIV Immunology database as of December 2006 [23] ) . HLA-matched associations that fell within these boundaries were grouped together . Note that the ± 3 aa proximal “window” was chosen to identify putative proteasomal processing escape mutations [24–26] based on evidence indicating that the majority of such mutations occur in the three amino acids immediately flanking the epitope [42] . The secondary level of support was granted to associations which fell within or proximal to a known HLA-restricted epitope anchor residue motif ( using MotifScan , http://hiv-web . lanl . gov/content/immunology/motif_scan/motif_scan ) and/or a putative HLA-restricted epitope identified by an independently validated CTL epitope prediction algorithm ( Epipred , http://atom . research . microsoft . com/bio/epipred . aspx [43] ) based on scanning the cohort consensus sequence . Again , associations falling within the “motif ± 3 aa flanking window” were grouped together . If specific amino acid variants were associated with additional HLA alleles in linkage disequilibrium ( LD ) , these alleles were also grouped together within the epitope or motif . To identify HLA alleles in LD , we investigated all possible pairwise allele combinations using a simple Fisher's exact test and conservatively defined all allele pairs with p < 0 . 05 ( q < 0 . 2 ) as linked ( Table S2 ) . In cases where LD allele pairs were associated with variation at the same codon , the allele exhibiting the strongest association ( as estimated by lowest p-value ) was classified as the allele driving the association . To provide in vitro functional support to identified associations , we drew upon a partially published ELISpot dataset of 372 HIV-1 infected , non-white individuals screened for HLA-restricted , CTL-mediated IFN-γ responses against set of 410 overlapping subtype B consensus peptides ( OLP ) 15 to 20 amino acids in length , spanning the whole expressed HIV-1 subtype B proteome [27] . Associations between possession of individual HLA alleles and responses to specific consensus peptides in the OLP set [27] were assessed by simple Fisher exact test . HLA allele/OLP associations with p < 0 . 05 were considered to be “significantly reactive” and thus indicative that an HLA-restricted CTL epitope lay in the boundaries of that OLP . HLA-associated polymorphisms identified in the present study that mapped directly within an HLA-specific reactive OLP were identified and annotated as “in vitro-supported” on the immune escape maps ( green; Figures 2–5 ) . Note that the differences in ethnic composition of the current and ELISpot-characterized [27] study populations may result in an underestimation of in vitro-supported associations , due to differences in cohort HLA composition and thus power to detect significant associations . | One of the greatest challenges facing HIV-1 vaccine design today is the formidable capacity of the virus for mutation and adaptation , a characteristic that has contributed to the extensive worldwide genetic variability of HIV-1 strains observed today . On an individual basis , evolutionary selective pressures imposed by each infected person's unique immune response results in the selection and outgrowth of viral “escape” mutants capable of evading immune recognition , while on a population basis , complex evolutionary selective pressures imposed by the highly polymorphic genes of the human immune system shape HIV-1 diversity on a global level . Making sense of the seemingly infinite complexity of HIV immune escape is of paramount importance in our goal of developing a successful HIV vaccine . The current study uses cutting-edge statistical methods to identify specific sites and patterns of human leukocyte antigen ( HLA ) class I-restricted escape mutations in various HIV genes . Researchers summarize their findings in the form of “immune escape maps , ” which highlight the differential contribution of immune imprinting to HIV genetic diversity , as well as identify specific sites in the viral genome under active immune selection pressure . Results from the present study contribute to our understanding of how human immune selective pressure contributes to variation in different HIV genes , and could help inform the development of HIV vaccines that take into consideration viral diversity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"viruses",
"immunology",
"virology"
]
| 2007 | Evidence of Differential HLA Class I-Mediated Viral Evolution in Functional and Accessory/Regulatory Genes of HIV-1 |
Body size is a quantitative trait that is closely associated to fitness and under the control of both genetic and environmental factors . While developmental plasticity for this and other traits is heritable and under selection , little is known about the genetic basis for variation in plasticity that can provide the raw material for its evolution . We quantified genetic variation for body size plasticity in Drosophila melanogaster by measuring thorax and abdomen length of females reared at two temperatures from a panel representing naturally segregating alleles , the Drosophila Genetic Reference Panel ( DGRP ) . We found variation between genotypes for the levels and direction of thermal plasticity in size of both body parts . We then used a Genome-Wide Association Study ( GWAS ) approach to unravel the genetic basis of inter-genotype variation in body size plasticity , and used different approaches to validate selected QTLs and to explore potential pleiotropic effects . We found mostly “private QTLs” , with little overlap between the candidate loci underlying variation in plasticity for thorax versus abdomen size , for different properties of the plastic response , and for size versus size plasticity . We also found that the putative functions of plasticity QTLs were diverse and that alleles for higher plasticity were found at lower frequencies in the target population . Importantly , a number of our plasticity QTLs have been targets of selection in other populations . Our data sheds light onto the genetic basis of inter-genotype variation in size plasticity that is necessary for its evolution .
Body size has a great impact on the performance of individuals [1 , 2] , as well as that of species [3] . Diversity in this trait is shaped by the reciprocal interactions between the developmental processes that regulate growth , and the evolutionary forces that determine which phenotypes increase in frequency across generations [4] . Body size varies greatly within and between populations [2 , 5] and is controlled by both genetic and environmental factors [6–9] . Studies in different animal models have provided insight about the selection agents that shape the evolution of body size ( e . g . predators [10 , 11] , mates [12] , thermal regimes [13 , 14] ) , and about the molecular mechanisms that regulate body size and body proportions during development [15–18] . Body size is also a prime example of the environmental regulation of development , or developmental plasticity [19 , 20] , and it is influenced by different factors , including nutrition and temperature . This plasticity can help organisms cope with environmental heterogeneity and , as such , can have major implications for population persistence and adaptation [20–22] . Thermal plasticity in body size is ubiquitous among insects [23–25] , with development under colder temperatures typically resulting in larger bodies , which is presumably advantageous for thermal-regulation [1 , 26] . The environmental dependency of body size , and other plastic traits , is often studied using reaction norms , in which phenotypic variation is plotted as a function of gradation in the environment [27] . The properties of these reaction norms , including their shapes and slopes , can differ between genotypes [28–30] , and the genes underlying such variation can fuel the evolution of plasticity . Little is known about what these genetic variants are and what types of functions they perform ( e . g . perception of environmental cues , conveying external information to developing tissues , or executing actions in developing plastic organs ) . It is also unclear to what extent the loci contributing to variation in thermal plasticity in size are the same for different body parts , and whether the loci contributing to variation in size plasticity are the same that underlie inter-individual variation in body size at a given temperature . Studies in D . melanogaster have provided much insight about the evolution and development of body size , body proportions , and body size plasticity [7 , 31–38] . Size differences among populations , including clinal [31 , 39] and seasonal variation [40] , and among individuals within a population , are due to the effects of genes , environment , as well as genotype-by-environment interactions [33 , 41–43] . While we have increasing detailed knowledge about the genetic basis of adaptation and of natural variation for many adaptive traits in D . melanogaster and other species [6 , 43–45] , little is known about the genetic basis of variation in plasticity . Widely-accessible mapping panels [46 , 47] allowed the dissection of the genetic architecture of various quantitative traits in D . melanogaster [48–50] , including body size [51] . However , with a few recent exceptions [52–55] , the genetic basis of phenotypic variation has been investigated under a single environmental condition , precluding assessment of differences between environments and of the genetic basis of plasticity itself . Series of isogenic lines from these mapping panels can be reared under different conditions to characterize reaction norms and ask about the genes that harbor allelic variation for their properties [53] . Here , we use a panel of D . melanogaster lines representing naturally segregating alleles from one natural population , the DGRP [46 , 48] , to characterize genetic variation for thermal plasticity in thorax and abdomen size , and to identify loci contributing to variation in the slopes of their thermal reaction norms . We document correlations between body size and body size plasticity , as well as correlations with other traits , using published data for the same panel . We also ask about the extent of overlap between QTLs for size and for size plasticity , and between QTLs for size plasticity of the different body parts . We then use different approaches to validate and further characterize the role of selected QTLs , and to ascertain their pleiotropic effects .
To assess the contribution of genotype and temperature to body size variation , we quantified length of abdomens and thoraxes ( Fig 1A , S1A Fig ) of adult females from ~196 isogenic lines reared at either 17°C or 28°C ( S1 Dataset , S2 Table ) . We found significant differences between DGRP genotypes and between developmental temperatures , as well as significant GxE interaction effects ( thermal plasticity ) for the size of both body parts ( Fig 1A , S1C Fig ) . We also found variation between individuals of ( presumably ) the same genotype and same rearing temperature; the coefficients of variation ( CV ) varied between 0 . 6 and 15 for thorax measurements and between 0 . 3 and 23 . 8 for abdomen measurements ( Table 1 ) . Broad-sense heritabilities for body size ( at 17°C and 28°C ) and body size plasticity ( between-environment variation ) varied between 33 and 52% ( Table 1 ) . We computed a correlation matrix to assess relationships between the different components of variation in body size ( Fig 2 ) . We found a significant positive correlation between thorax and abdomen size within each rearing temperature ( Fig 2A and 2B ) . We also found significant positive correlations between our measurements of thorax ( both temperatures ) , but not abdomen size , and the size of several body parts measured in other studies [51] for the same genotypes reared at 25°C ( S2A Fig ) . To address the question of association between body size and fitness , we measured correlations between our measurements of body size and a series of fitness related traits quantified in other studies of the DGRPs [46 , 49 , 50 , 56] . We found significant correlations with chill coma recovery ( thorax size at both temperatures and abdomen size at 28°C ) and with survival upon infection with Metarhizium anisopliae fungi ( abdomen size at 17°C ) , but not with longevity , starvation resistance , and other immune-defense traits ( S2B Fig ) . To address the question of what might explain the extent of inter-individual variation within genotype and temperature , we calculated correlations between our measurements of CV and of size . We found the CV to be: 1 ) positively correlated between body parts for flies reared at 28°C , but not for those reared at 17°C , and 2 ) negatively correlated with mean size for thoraxes , but not for abdomens ( Fig 2B ) . Using reaction norms , we studied the extent and properties of thermal plasticity for body size in the DGRP lines ( Fig 1B ) . We calculated the slope of the regression lines for size across temperatures for each body part and DGRP genotype , and found genetic variation for both the intercept and slope of the reaction norms ( Fig 1B , S2 Table ) . From each reaction norm , we extracted two properties of the thermal plasticity in body size: i ) the absolute value of the slope , describing only the magnitude of the response to temperature , and ii ) the raw value of the slope , which describes magnitude and direction of the response ( Fig 1C , S1D Fig ) . Using the reaction norms for 191 DGRP lines , we identified plastic and non-plastic genotypes . Slopes of the thermal reaction norm were significantly different from zero for 55% and 57% of the lines for the size of thoraxes and abdomens , respectively , with most of the plastic genotypes having smaller sizes when reared at higher temperatures ( Fig 1B ) . However , we also found plasticity in the opposite direction ( i . e . genotypes with smaller flies at lower temperatures ) , corresponding to a positive significant slope for the thermal reaction norms: 8% of the DGRP reaction norms for the thorax and 22% for the abdomen ( Fig 1B ) . Even though the levels of thermal plasticity for thorax and abdomen size were significantly positively correlated ( Fig 2B ) , lines having the highest levels of plasticity for one body part were not necessarily the most plastic for the other body part ( S1B Fig ) . Furthermore , for thorax , but not abdomen , we found that genotypes with larger CVs had steeper negative reaction norms . Finally , we also found no correlation between our size plasticity measurements and various fitness-related traits measured in the DGRPs ( S2B Fig ) , including longevity [49] , starvation resistance , chill coma recovery [46] , and immune-defense traits [50 , 56] . We used a GWAS approach to identify DNA sequence polymorphisms associated with variation in thermal plasticity for thorax and abdomen size in the DGRP . Because the loci carrying allelic variation for the direction and extent of environmental responsiveness are not necessarily the same , we used both the raw and absolute values of the slopes of the DGRP reaction norms as quantitative traits ( Fig 3A , S3 Fig ) . The candidate QTLs significantly associated with variation in plasticity were typically only so in relation to a single property of the reaction norm ( raw or absolute slope ) or body part ( thorax or abdomen; Fig 3C , S3 Table ) . We also found that these allelic variants fell within different genomic regions ( e . g . UTR , intronic , coding ) within or nearby 192 different putative genes ( Table 2 , S3 Table ) . Gene ontology enrichment analysis of the candidate QTLs showed an over-representation of vesicle-mediated processes ( e . g . phagocytosis and endocytosis; S6A Fig ) , while network enrichment analyses ( protein-protein interaction network followed by a KEGG pathways enrichment analysis ) revealed an over-representation for SNARE interactions and Notch pathways ( S6B Fig ) , both of which have been implicated in diverse biological functions [57–60] . We also found that in the vast majority of cases , alleles associated to increased environmental responsiveness were at lower frequencies in the DGRP ( Fig 3D , S5 Fig ) . To explore to what extent loci that contribute to variation in size plasticity also contribute to variation in body size within environments , we also performed GWAS analyses using body size at 17°C and 28°C as quantitative traits ( Fig 3B , S4 Fig , S4 Table ) . This analysis revealed QTLs that were mostly environment- and body part-specific ( Fig 3C , S4 Table ) , including no overlap between our candidate QTLs associated with variation in thorax and abdomen size and those reportedly associated with head size at 25°C [51] . Moreover , we found little overlap between candidate QTLs contributing to variation in size plasticity and those contributing to within-environment size variation ( Fig 3C ) , but some overlap in terms of network enrichment ( S6 Fig ) . None of our size traits or plasticity therein was affected by chromosomal inversions ( p-value > 0 . 01 ) , or by the genetic relatedness among DGRP lines ( low and non-significant coefficients of phylogenetic signal Blomberg’s K and Pagel’s λ; S2C Fig ) . We selected a number of significant QTLs for validation via different approaches ( Fig 4 , Table 2 , S2 Dataset ) . To test selected candidate genes , we used available null mutants and inducible gene knock-downs ( with RNA interference using the Gal4/UAS system ) . If a candidate gene affects plasticity , we expected to see a difference in the slope of thermal reaction norms between genotypes with abolished ( mutant ) or reduced ( knock-down ) gene function in relation to the corresponding control genotypes ( with “wildtype” gene function ) . To test specific significant SNPs/Indels , we used a SNP-based validation approach , hereafter called Mendelian Randomization ( see Materials and methods ) , that allowed comparisons between genotypes fixed for the candidate SNP ( minor versus major alleles ) but not for any other significant SNP for the same trait . If a candidate SNP affects plasticity , we expected a difference in slope of the reaction norms between newly-established genotypes carrying the minor versus the major allele at the target SNP . Using these methods , we confirmed a role in thermal plasticity for six out of seven candidate QTLs ( Fig 4 , S7 Fig , Table 2 ) . For genes Hsp60 ( Fig 4A ) , btv ( Fig 4B ) , Men ( Fig 4E ) , and Eip75B ( Fig 4F ) , plasticity was different between genotypes with impaired gene function ( knock-out or knock-down ) versus controls . For SNPs in genes CG43902 ( Fig 4C ) and ACC ( Fig 4D ) , plasticity was different between new genotypes with minor versus major allele . We did not find a difference in abdomen size plasticity between genotypes for candidate QTL CG43117 ( S7G Fig , Table 2 , S2 Dataset ) . For all the confirmed candidates for plasticity , the genotypes with impaired gene function were more plastic than their respective controls , and the DGRP genotypes harboring the minor allele were more plastic than those harboring the major allele ( S7 Fig ) . We also validated three out of four candidate QTLs for within-temperature variation in body size ( Fig 4 , S7 Fig , Table 2 , S2 Dataset ) . Genes Nmdmc ( Fig 4G ) and Optix ( Fig 4I ) affected thorax size at 28°C and at 17°C , respectively , while a SNP in gene CG14688 ( Fig 4H ) affected thorax size at 28°C . We did not find a difference in abdomen size at 28°C between genotypes for candidate QTL Wapl ( S7K Fig , Table 2 , S2 Dataset ) . To explore the pleiotropic effect of validated plasticity QTLs , we investigated whether the plastic response was also seen in the body part for which the SNP/gene had not been significantly associated to in the GWAS analysis ( S7 Fig , Table 2 ) . Out of the six validated plasticity QTLs , we only found cross-body part effects for gene btv; initially implicated in variation in plasticity of thorax size , this gene was also found to influence variation in plasticity of abdomen size ( S7B Fig , Table 2 ) .
Studying a panel of populations representing naturally-segregating alleles , the DGRP , we quantified effects of G , E , and GxE interactions on the size of two body parts ( thorax and abdomen ) . As is well documented for various species of insects [32 , 67 , 68] , most D . melanogaster genotypes we analyzed showed larger bodies when flies were reared at our lower temperature . However , we also documented cases of genotypes showing no plasticity ( robustness ) or showing size plasticity in the opposite direction . We also found a positive correlation in the levels of plasticity for the two body parts . The strong associations between the sizes of different body parts and plasticity therein is likely to reflect the tight regulation of body proportions , which is key for organismal performance [7 , 69] . It is unclear to what extent a genotype’s responsiveness to environmental conditions ( i . e . its plasticity ) is associated with inter-individual differences found within a given environment for that same genotype ( quantified with the coefficient of variation ) . While the latter is presumably un-accountable for by the effects of G , E , or GxE , it could reflect small genetic differences between individuals ( e . g . derived from somatic mutation ) , micro-environmental variation ( e . g . differences within a vial ) , or stochasticity in phenotype expression ( e . g . developmental noise ) . We found that genotypes that were more plastic for thorax size ( but not abdomen ) also had higher levels of intra-genotype , intra-environment variation ( i . e . higher coefficient of variation ) . Whether this component of phenotypic variance is assignable to micro-environmental variation and whether it has its own genetic basis has started to be investigated [53 , 70 , 71] and will , undoubtedly , be a topic of targeted future research . By using the raw and absolute values of the slopes of reaction norms as quantitative traits , we identified loci associated with variation in size plasticity . Genetic variation for environmental responsiveness could , in principle , involve different types of molecular players and could affect multiple traits in a similar or different manner . We described QTLs influencing size plasticity corresponding to different functions in terms of putative SNP effects ( e . g . missense , regulatory , or synonymous mutations; Table 2 , S3 Table ) , and described molecular function and biological process for corresponding genes . These genes could potentially be mediating environmental effects at different levels , from the perception of the environmental cue ( e . g . gene btv , which has been implicated in sensory perception [72] ) , to the transmission of that information to developing tissues ( e . g . gene Eip75B , coding for an ecdysone receptor [73] ) , or the execution of the information on those tissues ( e . g . genes Men , ACC , and Hsp60A , coding for two metabolic enzymes and a chaperone [73] , respectively ) . Presumably , genes higher up in the process of responding to the environment ( e . g . those involved in the perception of external conditions versus those responding in specific tissues ) would be more likely to affect multiple plastic traits in a concerted manner . With the exception of btv , none of our validated plasticity QTLs affected plasticity for other than the body part they had been identified as QTL for . Even for the complete set of candidate loci , we found very little overlap between QTLs for plasticity of different body parts ( thorax versus abdomen ) , as well as for different properties of the reaction norms ( raw versus absolute value of slopes ) . Furthermore , we also documented mostly private QTLs influencing variation in size at any given environment ( i . e . body part and temperature-specific ) . Sex- , body part- , and environment-specific QTL effects had been previously documented for various traits in different models [51 , 74–76] . In D . melanogaster , for example , different loci have been associated to variation in bristle number in different body parts [76] , and to variation in size in different environments [77] . Such private QTLs can potentially facilitate independent evolution of the traits . Previous works exploring the genetic basis of environmentally sensitive variation have mostly focused on investigating QTLs whose effect vary across environments ( QTL-by-environment interactions ) for a variety of traits in different species [66 , 78–80] . Much less attention has been paid to unraveling the allelic variants contributing to differences in plasticity itself . Exceptions include mapping of the genetic basis of thermal plasticity in life-history traits in Caenorhabditis elegans [81] , of photoperiodic plasticity in multiple traits in Arabidopsis thaliana [82] and , more recently , of thermal plasticity in cold tolerance in D . melanogaster [53] . Our results revealed little overlap between the QTLs that contribute to variation in trait within environments and the QTLs that contribute to variation in trait plasticity , assessed from the slope of reaction norms . We documented loci underlying variation in size plasticity ( i . e . properties of reaction norms ) that are different from those underlying variation in size at any temperature ( i . e . at 17°C and at 28°C ) . These results shed light onto a long-standing discussion about the genetic underpinnings of plasticity , which argue that either the genetic control of phenotypic plasticity happens via specific loci determining plastic responses or via the same loci that control trait values at a given environment [83–86] . Our data show that the genetic basis for trait plasticity , to a large extent , differs from the genetic basis for phenotypic variation in the trait itself . Plasticity can be adaptive in that it helps populations cope with environmental heterogeneity , and it has even been argued that it can promote phenotypic and taxonomic diversification [20–22 , 27] . Theoretical models highlight the ecological conditions that should influence the evolution of plasticity , such as the predictability of environmental fluctuations [87] and costs of plasticity [88 , 89] . Plasticity is generally presumed to be costly and only selected for in predictably heterogeneous environments , such as seasons [90] . The absence of a correlation between our thermal plasticity measurements and various fitness-related traits measured for the same genotypes could not identify any such cost . These potential costs might involve traits that have not been considered here , or these same traits but under ( environmental ) conditions that were not those assayed . The ability to respond or resist environmental perturbation , and the balance between both processes , can be crucial for fitness in variable environments . In the DGRP , even though some degree of environmental responsiveness is maintained , we found that the alleles contributing to increased levels of plasticity occur nearly always at lower frequencies ( i . e . the genotypes with the minor allelic variant having steeper reaction norms than those with the major allele ) . It is unclear to what extent this is the result of natural selection by the thermal regime that the natural population from which the DGRP was derived was exposed to , and/or the result of the process of deriving the mapping panel in the laboratory . It is also unclear to what extent QTLs for size plasticity in the DGRPs are those under selection in other populations . While it is conceivable , if not likely , that different QTLs contribute to variation in plasticity ( or other quantitative traits ) in different populations , we did find that a number of our plasticity QTLs have been targets of selection in other populations . Specifically , some of our candidate genes for size plasticity appear to have been selected in experimental populations of D . melanogaster evolving under different fluctuating thermal regimes [91] . Among our 192 candidate QTLs for thermal plasticity , eight genes ( including the validated btv ) had changes in the populations evolving under hot and cold temperatures fluctuations [91] , nine genes ( including Men ) had changes in populations evolving under hot fluctuations , and 25 genes had changes in the populations evolving under cold temperatures fluctuations ( see all overlaps in S3 Table ) . Gene Eip75B has also been previously implicated in the response to artificial selection for body size [92] and in differentiation between clinal populations [93 , 94] , which typically represent different thermal environments . Altogether , our results shed light onto the nature of inter-genotype variation in plasticity , necessary for the evolution of plasticity under heterogeneous environments . We showed that QTLs for size plasticity: 1 ) bear alleles for increased plasticity at low frequencies , 2 ) correspond to polymorphisms in different genomic regions and within genes of a multitude of functional classes , and 3 ) are mostly “private QTLs” , with little overlap between our various GWAS analysis . The latter underscores the potential for independent evolution of trait and trait plasticity ( different QTLs for size plasticity and for within-environment size variation ) , plasticity of different body parts ( different QTLs for size plasticity of thorax and of abdomen ) , and even properties of the environmental response ( different QTLs for raw and absolute slopes of thermal reaction norms ) .
Data for the GWAS was collected from adult female flies of the Drosophila Genetic Reference Panel ( DGRP ) obtained from the Bloomington Stock Center . The DGRP is a set of fully sequenced inbred lines collected from a single population in Raleigh , NC , USA [46 , 48] . The number and the details of the lines included in the GWAS for each trait can be found in S2 Table . Mutant stocks for the functional validations were: Hsp60A ( stock 4689 from Bloomington ) , MenEx3 and MenEx55 ( obtained from the T . Merritt lab ) . Control genetic backgrounds were w1118 ( stock 5905 , from Bloomington ) and Canton-S ( obtained from C . Mirth lab ) . UAS-Gal4 and UAS-RNAi lines used for validations were: stocks 6803 for bab-Gal4 , 5138 for tub-Gal4 , 28737 for btv-RNAi , and 35785 for mCherry-RNAi ( all obtained from the Bloomington stock center ) , and stocks v108399 and v110813 for Eip75B and Optix-RNAi , respectively ( obtained from the VDRC stock center ) . Fly stocks were maintained in molasses food ( 45 gr . molasses , 75gr sugar , 70gr cornmeal , 20 gr . Yeast extract , 10 gr . Agar , 1100 ml water and 25 ml of Niapagin 10% ) in incubators at 25°C , 12:12 light cycles and 65% humidity until used in this study . For the experiments , we performed over-night egg laying from ~20 females of each stock in vials with ad libitum molasses food . Eggs were then placed at either 17°C or 28°C throughout development . We controlled population density by keeping between 20 and 40 eggs per vial . We reared 200 DGRP lines and quantified thorax and abdomen size of 5 to 20 females per line , temperature and replicate . For 135 DGRP lines , we ran two replicates and for 33 lines we ran three replicates . The total number of flies used varied among DGRP lines due to differences in mortality at one or both of the temperatures . For some specimens , we could only quantify size of one body part if , for example , the individual was not properly positioned in the image or was damaged . Full details on the stocks used and the number of flies used per stock and temperature can be found in S1 Dataset and S2 Table . Rearing conditions for the validations of candidate QTLs were similar to those used for the DGRP lines . Adult female flies ( 8–10 days after eclosion ) were placed in 2ml Eppendorf and killed in liquid nitrogen followed by manual shaking to remove wings , legs and bristles . Bodies were mounted on Petri dishes with 3% Agarose , dorsal side up , and covered with water to avoid light reflections . Images containing 10 to 20 flies were collected with a LeicaDMLB2 stereoscope and a Nikon E400 color camera under controlled imaging conditions of light and white-balance . Images were later processed with a customized Mathematica macro to extract size measurements . For this purpose , we drew two transects per fly , one in the thorax and one in the abdomen , using body landmarks ( as shown in S1A Fig ) . Size of each body part was initially quantified as the number of pixels in the transect and later converted to millimeters . For abdominal transects , when necessary , we performed an additional step that involved the removal of pixels corresponding to the membranous tissue that is sometimes visible between segments . For each body part ( thorax and abdomen ) , we performed four independent Genome-Wide Analyses ( GWAS ) : two for thermal plasticity ( raw and absolute values of the slopes of the reaction norms ) , and two for within-environment variation ( length at 17°C and length at 28°C ) . Slopes of the reaction norm were calculated as the slope of the linear model lm ( Size ~ Temperature ) for each body part and DGRP line . We used linear mixed-effects models ( lmer ) in lme4 R package to test which polymorphisms explained variation in size plasticity or in size . The GWAS analyses for variation in thermal plasticity tested for effects of fixed factor Alelle ( corresponding to each polymorphic site ) , and random factors Wolb ( corresponding to Wolbachia status ( 0 or 1 ) of each DGRP line [7 , 8] ) and DGRP ( i . e . genetic background ) to variation in the dependent variable Slope ( i . e . raw or absolute values of the reaction norms ) . This corresponds to notation lmer ( Slope ~ Allele + ( 1|Wolb/DGRP ) ) in lme4 R package [95] . The GWAS analyses for within-environment variation tested the model lmer ( Size ~ Allele + ( 1|Wolb/DGRP ) , where Size is the dependent variable ( thorax or abdomen length from flies reared at 17°C or at 28°C ) and all the other terms are the same as described above for the plasticity GWAS . All the GWAS were performed by using polymorphisms where we had information for at least ten DGRP lines per allele . We did not find an effect of Wolbachia in any of our GWAS analyses . For each of the GWAS , we annotated the SNPs with an arbitrary and commonly used p-value < 10e-5 using the FlyBase annotation ( FlyBase release FB2017_05; [73] ) . For the same SNPs , we performed first , a gene ontology enrichment analysis using the publicly available GOrilla Software [96 , 97] and second , a network enrichment analysis using gene-enrichment and pathway-enrichment analyses were done using the publicly available NetworkAnalyst Software [98 , 99]; using all nodes from first order network generated with IrefIndex Interactome settings . Analyses of the overlap in significant QTLs ( p-value < 10e-5 ) from different GWAS analyses was done for individual polymorphic sites ( considered to be the same based on genomic position and type ) and for the genes those sites were annotated to ( using FlyBase release FB2017_05; [73] ) . We tested for the effect of the chromosomal inversions ( In_3R_K , In_3R_P , In_2L_t , In_2R_NS and In_3R_Mo ) on our thorax and abdomen traits by using linear models ( lm ( Mean Size ~ Inversion ) for within-environment size variation and lm ( Slope ~ Inversion ) for size plasticity variation ) . Genetic distance matrix for the DGRPs was obtained from http://dgrp2 . gnets . ncsu . edu/data . html and was used to perform a cluster hierarchical dendogram using ape and phylobase R packages . We estimated the phylogenetic signal and statistical significance for each of our traits using Blomberg’s K [100] and Pagel’s λ [101] metrics with the phylosig function in the phytools R package [102] . The subsample of significant QTLs to be validated was taken from a first list of candidates ( S3 and S4 Figs ) selected based on p-value and corresponding peaks in the Manhattan plots ( clear peaks prioritized ) , putative effect ( missense and regulatory variants prioritized over intergenic variants ) , associated genes ( annotated and known function prioritized ) . We used three methods for validation , depending on QTL properties: null mutants and RNAi ( Gal4/UAS system ) for genes containing several significant SNPs and/or containing SNPs corresponding to missense variants , and Mendelian Randomization ( MR ) for SNPs in genes with little or no information available . Mutant and RNAi test that no or low levels of peptide affect variation in the quantitative trait for which the gene was identified as a candidate QTL while MR tests for sufficiency and independence from genetic background of the specific allele . Following these criteria we tested a total of 11 candidate SNPs/genes . Validations by null mutants were done by comparing the phenotype in the heterozygous mutant stock with its respective genetic background . Validations by RNAi were done by comparing , for each Gal4 driver line , the phenotype of the gene of interest knockdown with the corresponding control cross using UAS-mCherryRNAi as well as with the corresponding control genetic background for the UAS line . We always used two different driver lines for our validations by RNAi: tub-Gal4 and bab-Gal4 . Gene tub is ubiquitously expressed throughout development and gene bab is expressed in multiple tissues during different stages of development . Both genes include expression in developing thoraxes and abdomens ( see FlyBase reports FBgn0003884 and FBgn0004870 , respectively ) . However , for all candidate genes selected for RNAi validation , except Nmdmc , the crosses between RNAi line and tub-Gal4 were lethal . The identity of the SNPs tested by MR is given by their annotation with Genome Release v6 . For each candidate SNP , we first selected 10 DGRP lines to make a population with the minor allele fixed and 10 others to make a population with the major allele fixed . The 10 DGRP lines used to create each population were checked for having only one of the significant QTLs ( p-value < 10e-5 ) fixed and not the others . These lines were used to generate four populations , two fixed for the major allele and two for the minor allele ( S2 Table ) . Each population was established by crossing 8 virgin females from each of 5 of the same-allele lines to 8 males of the other 5 lines . Reciprocal crosses were used to set two independent populations per allele . These populations were allowed to cross for eight generations to randomize genetic backgrounds . We confirmed by Sanger sequencing that those populations had our candidate allele fixed . Primer sequences used to confirm the allele in each population were: All statistical analyses were performed with R Statistical Package version 3 . 3 . 1 [103] . For each body part and temperature , we used linear mixed-effects model ( lmer ) in lme4 R package [95] to test for the effect of replicate on size ( model lmer ( Size ~ Replicate + ( 1|DGRP/Replicate ) ) ) , that were found to be non-significant ( p-value > 0 . 05 ) . Individuals from different replicates were pooled for all other analyses . We then used linear models ( lm ) to test for the within-environment effect of genotype ( DGRP ) on size ( model lm ( Size ~ DGRP ) ) . We also used linear models to test for plasticity in size ( i . e . the interaction between genotype ( DGRP ) and temperature ( model lm ( Size ~ DGRP*Temperature ) ) ) . Reaction norms for each DGRP line were calculated by using the linear model lm ( Size ~ Temperature ) . We extracted two properties of the reaction norms per DGRP line and body part: the absolute value of the slope as a measurement of thermal sensitivity , describing only the magnitude of the response to temperature , and the raw value of the slope as a measurement which describes also the direction of that response . We defined plastic genotypes as those DGRP lines whose reaction norm slope was significantly different from zero ( p-value < 0 . 05 ) . We used the total number of phenotyped individuals for each DGRP and temperature to perform the calculations of the summary statistics in S2 Table . We used Pearson correlations ( alpha = 0 . 99 ) to test for linear correlation in size between body parts , controlling for DGRP lines . We also used Pearson correlations to test for linear correlations among our measured traits and between those and other available datasets for the DGRPs . For this , we used the mean value per DGRP line for each trait and the corrplot R package [104] . We report both correlation coefficient and significance levels . Available DGRP phenotypes that were used to correlate with our traits were: size measurements at 25°C [51] , longevity [49] , starvation resistance , chill coma recovery [46] , tolerance to infection with Providencia rettgeri bacteria [56] and resistance to infection with Metarhizium anisopliae fungi or with Pseudomonas aeruginosa bacteria [50] . Broad sense heritability for size at each temperature was estimated as H2 = σ2A/ ( σ2A + σ2W ) where σ2A and σ2W are the among-line and within-line variance components , respectively . Heritability of plasticity was calculated as H2 = σ2G*E/σ2TOTAL where σ2G*E and σ2TOTAL are the variance associated with the genotype by environment interaction and total variance components , respectively , as proposed in Scheider and Lyman ( 1989 ) . Variance components were extracted using varcomp R package . For the functional validations of within-environment SNPs and genes we used the linear models lm ( Size ~ Allele ) and lm ( Size ~ Genotype ) , respectively . For the validations of plasticity SNPs and genes we used the models lm ( Size ~ Genotype*Temperature ) and lm ( Size ~ Allele*Temperature ) , respectively . In all cases , significant differences among groups were estimated by post hoc comparisons ( Tukey’s honest significant differences ) . | Environmental conditions can influence development and lead to the production of phenotypes adjusted to the conditions adults will live in . This developmental plasticity , which can help organisms cope with environmental heterogeneity , is heritable and under selection . Its evolution will depend on available genetic variation . Using a panel of D . melanogaster flies representing naturally segregating alleles , we identified DNA sequence variants associated to variation in thermal plasticity for body size . We found that these variants correspond to a diverse set of functions and that their effects differ between body parts and properties of the thermal response . Our results shed new light onto the long discussed genes for plasticity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"genome-wide",
"association",
"studies",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"abdomen",
"quantitative",
"trait",
"loci",
"population",
"genetics",
"animals",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"physiological",
"parameters",
"genome",
"analysis",
"experimental",
"organism",
"systems",
"molecular",
"genetics",
"population",
"biology",
"drosophila",
"research",
"and",
"analysis",
"methods",
"genetic",
"polymorphism",
"animal",
"studies",
"molecular",
"biology",
"genetic",
"loci",
"insects",
"arthropoda",
"eukaryota",
"anatomy",
"physiology",
"thorax",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"computational",
"biology",
"organisms",
"human",
"genetics"
]
| 2018 | Genetic basis of thermal plasticity variation in Drosophila melanogaster body size |
Drosophila segmentation is a well-established paradigm for developmental pattern formation . However , the later stages of segment patterning , regulated by the “pair-rule” genes , are still not well understood at the system level . Building on established genetic interactions , I construct a logical model of the Drosophila pair-rule system that takes into account the demonstrated stage-specific architecture of the pair-rule gene network . Simulation of this model can accurately recapitulate the observed spatiotemporal expression of the pair-rule genes , but only when the system is provided with dynamic “gap” inputs . This result suggests that dynamic shifts of pair-rule stripes are essential for segment patterning in the trunk and provides a functional role for observed posterior-to-anterior gap domain shifts that occur during cellularisation . The model also suggests revised patterning mechanisms for the parasegment boundaries and explains the aetiology of the even-skipped null mutant phenotype . Strikingly , a slightly modified version of the model is able to pattern segments in either simultaneous or sequential modes , depending only on initial conditions . This suggests that fundamentally similar mechanisms may underlie segmentation in short-germ and long-germ arthropods .
Like other arthropods , the fruit fly Drosophila melanogaster has a segmented body plan . This segmental pattern is laid down in the embryo during the first 3 hours of development . During this time , the anteroposterior ( AP ) axis of the blastoderm is progressively patterned down to cellular-level resolution by an elaborate , multi-tiered network of genes and their encoded transcription factors [1 , 2] . These genes were first identified in a landmark genetic screen [3 , 4] , and their regulatory interactions have subsequently been dissected by 3 decades of genetic experiments . Along the way , this body of research has revealed many fundamental principles of transcriptional regulation [5] , and Drosophila segmentation remains a central model for developmental systems biology today . Much of the “heavy lifting” of segment patterning is carried out by the so-called “pair-rule” genes , which make up the penultimate tier of the Drosophila segmentation cascade . The pair-rule genes are the first genes to be expressed in spatially periodic patterns in the Drosophila embryo and are collectively responsible for patterning the expression of the “segment-polarity” genes , which organise and maintain segmentally reiterated compartment boundaries termed “parasegment boundaries” . Notably , this involves transducing a double segment pattern of early pair-rule gene expression , in which each set of stripes is offset slightly from the others , into a single-segment pattern of segment-polarity gene expression , in which most genes are expressed in discrete , non-overlapping domains [6–8] . There are 7 canonical pair-rule genes: hairy [9] , even-skipped ( eve ) [10] , runt [11] , fushi tarazu ( ftz ) [12] , odd-skipped ( odd ) [13] , paired ( prd ) [14] , and sloppy-paired ( slp ) [15] . 5 of these genes ( hairy , eve , runt , ftz , and odd ) are known as the “primary” pair-rule genes because they are expressed earlier than the 2 “secondary” pair-rule genes , prd and slp [16] . ( Note that these terms have a somewhat tortuous history , and older literature will classify the genes differently . ) Each of the primary pair-rule genes is initially patterned by spatial inputs from the upstream tier of transcription factors , encoded by the “gap” genes , which are expressed in broad , overlapping AP domains during cellularisation [17] . This patterning occurs in an ad hoc manner , with specific combinations of gap factors regulating the expression of particular pair-rule stripes through discrete “stripe-specific” enhancer elements [18–21] , which act additively with one another . For certain pair-rule genes , such as eve , this regulation is sufficient to generate an overall pattern of 7 equally spaced stripes along the future trunk of the embryo [16 , 22–24] . For other pair-rule genes , such as odd , the gap-driven pattern is irregular and may have missing stripes [16] . In these cases , the initial patterns are regularised by cross-regulatory “zebra” enhancer elements [25–27] , which take periodic inputs from other pair-rule factors and yield periodic outputs . Similar zebra elements are responsible for driving the periodic expression of the secondary pair-rule genes , which turn on after the primary pair-rule patterns have refined [16 , 28] . At gastrulation , the segment-polarity genes turn on , activated by a broadly expressed transcription factor , Odd-paired ( Opa ) [29] , and spatially regulated by the pair-rule genes [6 , 7 , 30 , 31] . Opa activity also “rewires” the regulatory interactions between the pair-rule genes , causing several of their expression patterns to transition dynamically from double- to single-segment periodicity ( i . e . , from 7 stripes to 14 stripes ) [32] . These pair-rule factors ( and/or their paralogs ) then play roles in the segment-polarity network , which also contains several components of the Wingless ( Wg ) and Hedgehog signalling pathways [33–37] . The Drosophila gap gene network has been used frequently in recent years as a case study for the application of dynamical systems [38–40] and information theory [41–43] approaches to developmental patterning , but the pair-rule network has received little system-level attention . Indeed , the most recent models of pair-rule patterning [8 , 44] date from more than 10 years ago . Since these were published , 3 important discoveries have been made about segment patterning , all of which challenge established assumptions about the Drosophila segmentation cascade and all of which concern the pair-rule genes in some way . So long as the pair-rule network remains poorly understood , key questions raised by these findings will go unanswered . The first discovery is from comparative studies in other arthropod embryos . Drosophila is a “long-germ” embryo , patterning almost all of its segments simultaneously in the blastoderm prior to germ-band extension [45] . However , the ancestral mode of arthropod development is “short-germ” embryogenesis , in which segmentation is sequential and coordinated with germ-band elongation [46–48] . Orthologs of the pair-rule genes play a key role in segment patterning in all arthropods studied thus far ( for example , [49–52] ) , but in short-germ embryos , their expression has been shown to oscillate in a posterior “segment addition zone” throughout germ-band extension [53–55] . This periodic dynamic expression indicates that in these organisms , they are either components of or entrained by a segmentation “clock” [56] . How the expression of the pair-rule genes in long-germ embryos such as Drosophila relates to their expression in short-germ embryos ( for example , the model beetle , Tribolium castaneum ) is unclear . It is thus not understood how long-germ segmentation was derived from short-germ segmentation , an important evolutionary transition that has occurred multiple times independently within the higher insects [57] . The second discovery stems from quantitative studies of Drosophila segmentation gene dynamics . These studies have revealed that the domains of gap gene expression in the trunk of the embryo shift anteriorly across the blastoderm over the course of nuclear division cycle 14 ( cellularisation ) [58–60] . The shifts are mirrored downstream in similarly shifting expression of the pair-rule stripes [59 , 61] , a finding that is at odds with existing models of segment patterning , which assume these stripes to be static domains [6 , 8 , 44 , 62–64] . While we know that these subtle shifts are ultimately driven by feedback interactions within the gap gene network [38 , 39 , 65–67] , their functional role ( if any ) remains unclear . The final key finding relates to the structure of the pair-rule network itself . In a recent paper on the pair-rule network [32] , Michael Akam and I showed that many of the regulatory interactions between the pair-rule genes are temporally regulated ( by Opa , as described above ) . We argued that the pair-rule network is best viewed as 2 distinct networks , 1 operating during cellularisation and 1 during gastrulation , each with a specific topology and resultant dynamics . Analysing the “early” ( cellularisation-stage ) and “late” ( gastrulation-stage ) pair-rule networks separately should lead to a better understanding of pair-rule patterning and might also reveal why the network shows this bipartite organisation in the first place . In this paper , I present a new model of the pair-rule system , which incorporates the stage-specific architecture of the pair-rule network . I take the set of identified genetic interactions between the pair-rule genes as a starting assumption , formalise them in a Boolean logical model , and use dynamical simulations to analyse how they collectively lead to complex pattern formation . I find that gap-mediated kinematic shifts are required for correctly phasing the pair-rule stripes , something that proves crucial for downstream segment patterning . I also find that graded Eve activity is not strictly necessary for pair-rule patterning , and I explain the aetiology of the surprisingly severe eve null mutant phenotype . Finally , I show that a slightly modified version of the Drosophila pair-rule network gains the capacity to pattern in both simultaneous and sequential modes , conceptually reconciling long- and short-germ segmentation .
Fig 1A summarises the inferred regulatory interactions between the pair-rule genes . Following Clark and Akam ( 2016 ) [32] , individual interactions are assigned to distinct “early” and “late” networks , which operate during mid-cellularisation or late cellularisation/gastrulation , respectively . Note that a few regulatory interactions ( e . g . , repression of ftz by Eve ) are common to both networks , but the majority are restricted to a single phase of patterning . Gene regulatory network models represent “intellectual syntheses” of diverse experimental data [70] . I arrived at the topologies in Fig 1A by carefully analysing relative expression data in tightly staged wild-type embryos and cross-referencing these observations with the large number of mutant and misexpression experiments recorded in the genetic literature ( for example , [8 , 30 , 31 , 64 , 71–78] ) . In places where the data were particularly ambiguous , I also re-characterised pair-rule gene expression in select pair-rule mutants in order to pick apart direct versus indirect regulatory interactions . In almost all cases , the interactions in the network diagrams have been previously inferred by multiple sets of researchers; my contribution has been ( 1 ) to bring this body of work together into something consistent and relatively complete and ( 2 ) to recognise the distinction between the early and late phases of regulation , rather than pooling all interactions into a single network . Most of the evidence and reasoning behind the inferred interactions ( and interactions inferred to be absent ) in Fig 1A are described in Appendix 1 of Clark and Akam ( 2016 ) [32] . Additional evidence in favour of the “early” cross-regulatory interactions between the primary pair-rule genes ( boxed yellow area in Fig 1A , left ) is presented in S1 Text , based on patterns of pair-rule gene expression in hairy , eve , and runt mutants . Two things are immediately clear from the network diagrams . First , the direct regulatory interactions between the pair-rule genes are overwhelmingly repressive . This is consistent with a mode of patterning consisting of spatially ubiquitous activation ( by maternally provided factors , for example ) combined with precisely positioned repression from other segmentation genes [76 , 79–81] . While certain of the pair-rule factors ( e . g . , Ftz and Prd ) have been shown to quantitatively up-regulate the expression of other pair-rule genes and thus contribute to this background activation in a spatially modulated way [82–84] , these effects do not , for the most part , seem to be important for qualitatively determining the spatial pattern of pair-rule gene expression and so have been omitted from the diagram . Most described incidences of one pair-rule gene genetically activating another pair-rule gene are instead indirect ( i . e . , mediated by the direct repression of another repressor ) . Second , the 2 networks have very different structures , presumably reflecting the different patterning function each must perform . During mid-cellularisation , pair-rule gene cross-regulation is responsible for refining many of the pair-rule stripes and standardising their phasing relative to other pair-rule stripes , resulting in a regular repeating pattern of double-segment periodicity . This is carried out by the early network , which is sparse , composed of unidirectional regulatory interactions , and has no feedback loops . Two of the pair-rule genes , eve and hairy , are patterned by gap factors rather than other pair-rule factors and so represent “input-only” factors to the network [16] . The remaining primary pair-rule genes ( runt , ftz , and odd ) do receive extensive gap inputs at early stages of cellularisation , but , by mid-cellularisation , their patterns are largely specified by other pair-rule genes ( see S1 Text ) . ( Note , however , that some aspects of the ftz pattern cannot be explained by pair-rule inputs alone , see Appendix 2 of [32] . ) The secondary pair-rule genes turn on later ( prd at mid-cellularisation and slp towards the end of cellularisation ) and are patterned by primary pair-rule genes . Overall , the early network has a hierarchical structure , in which Eve and Hairy convey positional information derived from the gap factors to the remaining primary pair-rule genes and eventually to the secondary pair-rule genes . The late network , on the other hand , is extremely dense and consists largely of mutually repressive pairs of interactions . It is responsible for converting a double-segmental pattern of overlapping stripes into a segmental pattern of discrete segment-polarity fates . This is the final step in the Drosophila “segmentation cascade” and completes the transition from the analog ( graded ) positional information carried by the maternal and gap gene products to the essentially digital positional information carried by the segment-polarity genes [64] . The numerous positive ( i . e . , double-negative ) feedback loops within the late network are consistent with it acting like a multi-stable switch , individual segment-polarity fates representing attractor states towards which the system will rapidly converge . As described above , gap inputs and the early pair-rule network combine to establish a repeating double-segmental pattern of pair-rule gene expression . The positional information within this pattern is then converted into a stable output pattern of segment-polarity states by the late network . Each initial double-segment repeat is about 7–8 nuclei wide , and each specified segment will consist of at least 3 distinct states characterised by the expression of engrailed ( en ) , odd , and slp , respectively ( Fig 1B ) . The en and odd stripes are about 1 nucleus wide , while the slp stripes are about 1–2 nuclei wide . Parasegment boundaries form wherever En and Slp domains abut , while Odd provides a buffer zone that preserves the AP polarity of each segment . ( This tripartite segment pattern conforms to prescient theoretical predictions made by Hans Meinhardt in the early 1980s [62 , 85 , 86] . ) It is crucial that all 3 domains are specified within each segment—and that they are in the correct order—because patterning defects such as boundary losses , ectopic boundaries , and/or polarity reversals arise when the pattern is perturbed [8 , 31 , 33 , 76 , 87] . The extremely high resolution of the final segmental output pattern implies that the initial double segment pattern established by the early pair-rule network must contain sufficient positional information to allow almost every nucleus to be distinguished from its immediate neighbours . We are thus left with 2 questions . First , how does the early network establish a situation in which the different nuclei within a double-segment repeat each expresses a unique combination of pair-rule factors ? Second , how exactly is this code “read” by the late network ? ( Or , in other words , which sets of initial conditions will result in a cell following an expression trajectory that ends at , for example , stable en expression , rather than stable odd or stable slp ? ) In later sections , I address these questions by simulating and analysing the networks shown in Fig 1A . However , before getting into specifics of how particular genes are regulated and expressed , it is worth considering a more fundamental question: where is the positional information coming from in the first place ? The topology of the early network ( Fig 1A , left ) implies that , to a first approximation , all the positional information in the final pattern must trace back to the expression patterns of just 2 factors , Eve and Hairy . Boolean ( ON/OFF ) combinations of Eve and Hairy would only be sufficient to specify 4 different domains within each double-segment repeat ( Fig 1C , top ) , whereas the real output pattern consists of at least 6 distinct domains ( i . e . , En , Odd , Slp , En , Odd , Slp ) . How is it possible that just 2 independent spatial signals are able to give rise to such a high-resolution final output ? One potential answer is that stripes of Eve and/or Hairy might carry quantitative information that permits them to convey more than 2 “states” within the positional code . Since the early 1990s , this idea has been applied to the graded margins of the early Eve stripes [30 , 75 , 88] . These stripes have been proposed to act as local morphogen gradients , repressing different target genes at different concentration thresholds and thus differentially positioning their respective expression boundaries . Current models of pair-rule patterning rely on the assumption that there are 4 functionally distinct levels of Eve activity across an Eve stripe ( from the centre to the edge: HIGH , MEDIUM , LOW , and OFF ) [8 , 44] . These different levels would provide cellular-level resolution within each double-segment repeat and , combined with information from the Hairy stripe , allow each nucleus to be uniquely specified ( Fig 1C , bottom ) . While a given concentration of Eve protein may well repress its various targets with different efficacies , it is unlikely that segment patterning relies significantly upon this mechanism , for 3 main reasons . First , for the model to be viable , the Eve stripes would need to provide an accurate and precise set of positional signals within each double-parasegment repeat , i . e . , the Eve stripes would have to be extremely regular and all share the same shape and amplitude . However , more posterior Eve stripes show significantly lower expression levels than more anterior Eve stripes throughout most of cellularisation [59] . Furthermore , pair-rule transcripts are apically localised , and therefore pair-rule gene expression becomes effectively cell autonomous soon after membrane invagination begins [89 , 90] . This means that , unlike for the gap genes ( whose transcripts remain free to diffuse between neighbouring nuclei ) , for eve , there is little or no spatial averaging to buffer the high intrinsic noise of transcription [91 , 92] . This reduces the precision of the Eve signal and thus its capacity to reliably convey analog information . Second , the morphogen model also requires the readout of the Eve signal to be very sensitive; i . e . , eve target genes would have to reliably discriminate between different Eve expression levels and pattern their expression boundaries accordingly . However , it is not clear that this actually occurs within the embryo—for example , the model proposes that graded Eve stripes result in offset boundaries of the Eve targets odd and ftz , but recent observations indicate that these offsets are in fact produced by other mechanisms [32] . Third , the morphogen model does not explain the full severity of the eve null mutant phenotype , in which aberrant expression patterns are seen even in regions that would be outside the Eve stripes in wild-type embryos . Neither does the morphogen model explain the patterning robustness of eve heterozygotes , in which halving Eve expression levels fails to perturb the overall pattern of segment-polarity domains . How , then , might the spatial resolution of the segment pattern be explained if not by an Eve morphogen gradient ? Traditional models of Drosophila segmentation are essentially static: each tier of segmentation gene expression provides a single set of spatial signals , which is transduced into a new set of spatial signals by the tier below . This simplifies the real situation in the embryo , in which both gap and pair-rule expression domains shift subtly from posterior to anterior over time [59 , 93] . Explicitly considering these temporal aspects of segmentation gene expression suggests an alternative segment patterning mechanism: using the temporal dynamics of a relatively coarse pair-rule signal to provide high-resolution spatial information across each pattern repeat . A signal that varies over time can be used to convey an arbitrary quantity of information , even if each reading of that signal provides very little ( think of Morse code or binary storage ) . The eve and hairy stripes continue to be regulated by gap inputs throughout most of cellularisation and therefore shift across nuclei in concert with the gap domains . This means that , rather than each nucleus having to deduce its position from a single level of , e . g . , Eve protein ( as in the morphogen model ) , the nucleus actually experiences a temporal sequence of Eve protein levels . Strikingly , an overall shift of just 2 nuclei would be theoretically sufficient for a Boolean Eve stripe to , on its own , specify the positions of all 6 segment-polarity domains within a double-segment repeat ( Fig 1D ) . This kind of mechanism would , however , rely on the downstream targets of Eve and Hairy being able to decode a temporal sequence of Eve/Hairy expression and convert it into an appropriate segment-polarity fate . In the following sections , I carry out simulations and analysis of the network shown in Fig 1A and , based on the results , argue that the cross-regulatory interactions between the pair-rule genes function to achieve exactly this task . In order to investigate how pair-rule patterning works , I used the networks shown in Fig 1A to create a toy model of the pair-rule system and then simulated pair-rule gene expression across an idealised 1-dimensional tissue . In this section , I briefly describe the structure and assumptions of the modelling approach; a full description of the model plus details of all simulations are given in S2 Text . ( Source code for running the simulations is available in S1 and S2 Files , while pair-rule networks in SBML-qual format are available in S3 and S4 Files ) . The genes whose regulation I model explicitly are the 7 pair-rule genes , plus en , whose product plays a key role in regulating late pair-rule gene expression . I have also included 4 inputs that are extrinsic to the system: 2 temporal signals , Caudal ( Cad ) [94] and Opa , and 2 signals to represent the positional information provided by the gap system , “G1” and “G2” . Cad represses the secondary pair-rule genes during early stages of patterning [87] , while Opa turns on midway through patterning and triggers the switch from the early network to the late network [32] . G1 is responsible for patterning the hairy pair-rule stripes while G2 is responsible for patterning the eve pair-rule stripes . G1 and G2 do not represent specific gap factors but are instead an abstraction of the spatial inputs ( i . e . , stripe boundary locations ) provided by the gap system as a whole . Each gene in the system is represented by a Boolean variable , and its control logic is formalised using logical rules ( essentially equivalent to the “logical equations” used in Sanchez and Thieffry [2003] [44] or the “vector equations” used in Peter et al . [2012] [70] ) . For example , if Opa is OFF ( early network ) , odd is expressed only if both Hairy and Eve are also OFF , while if Opa is ON ( late network ) , odd expression relies on all of Runt , En , and Slp being OFF ( compare Fig 1A ) . In most cases ( apart from , e . g . , activation of en by Ftz or Prd ) , gene activation is assumed to be driven by some ubiquitous background factor ( s ) and is not explicitly included in the model . The network simulation proceeds by discrete time steps , with expression output at time point t + 1 calculated from the state of the system at time point t . Because of the speed and dynamicity of segment patterning , time delays associated with protein synthesis and protein decay imply that protein and transcript expression domains for a given gene will often be non-congruent within the Drosophila embryo . This is likely to be significant for patterning , and I therefore approximate this effect by adding simple time delay rules into the simulation . Each gene has associated “synthesis delay” and “decay delay” parameter values s and d , both of which are integers representing a certain number of time steps . ( Once a gene turns on , transcript will be present immediately , but protein will only appear s time steps later . Similarly , once a gene turns off , transcript will disappear immediately , but protein will only disappear after another d timesteps have elapsed . ) For parsimony ( and consistent with real expression kinetics [82] ) all pair-rule genes and en are assigned the same delay value , which applies to both s and d . The specific value of this delay is fairly arbitrary , because the ratio between different delays is what affects how the system behaves , but I have chosen this value to be 6 . Given that the half-life of ftz RNA during cellularisation is 7 minutes [79] , this means that each time step in the simulation can be thought of as representing on the order of 1 minute of real developmental time . The time delays of the other components in the network ( Cad , Opa , G1 , and G2 ) are assigned appropriate values relative to this timescale , so that their simulated behaviour roughly approximates their spatiotemporal expression in a real embryo . The simulation is set up to occur across a row of 20 “cells” , an idealised representation of the AP axis . This row of cells is not meant to correspond to a specific region of the Drosophila embryo but rather to be generally representative of patterning within the main trunk ( i . e . , pair-rule stripes 3–6 ) , in which pair-rule genes are not additionally affected by cephalic or terminal factors . Each cell within this “tissue” is simulated independently , starting from a specific set of initial conditions . ( As mentioned above , pair-rule transcripts are apically localised , and therefore the cross-regulation between the pair-rule genes is likely to be effectively cell autonomous from roughly mid-cellularisation onwards . ) The starting conditions for each cell usually involve specifying the appropriate expression of G1 and G2 , setting Cad to ON , and setting all other genes to OFF . G1 and G2 are initialised with patterns that are offset by 2 cells and repeat every 8 cells , meaning that the hairy and eve stripes specified by these inputs will partially overlap and exhibit a double-segment periodicity , as in real embryos . Gap inputs into runt , ftz , and odd are omitted , meaning that their early expression is organised entirely by the spatial inputs from Hairy and Eve . As the simulation proceeds , Cad protein will disappear , allowing prd to turn on [87] , followed by slp . ( Note that we do not currently know how exactly the timing of slp expression is controlled , so in order to reproduce the timing observed in the embryo , slp expression in the simulation requires Prd expression to already be present . ) Shortly afterwards , Opa protein will turn on , switching the control logic of pair-rule gene expression to the late network and causing pair-rule gene expression to eventually reach a final , stable state . After the switch to the late network , the gap factors and Hairy cease to regulate the pair-rule genes and then fade away , as in real embryos . Note that the model just described , which is Boolean , deterministic , and uses discrete time steps , is not designed to capture the full complexity of the embryo ( in which gene expression is , of course , quantitative , stochastic , and continuous ) . Rather , it represents a tool to expose the key mechanisms of patterning—and to delineate how much of what is observed in the embryo follows simply from the qualitative structure of the regulatory network . It also provides an important sanity check of the inferences that led to that structure being proposed in the first place . Using the model described above , I first simulated a scenario in which gap gene inputs and hence the pair-rule stripes of Hairy and Eve are completely static . The results are shown in S9 Movie and are summarised in Fig 2A . Under these conditions , the positional information provided by Hairy and Eve is essentially equivalent to the situation diagrammed in Fig 1C ( bottom ) and thus has no possibility of generating the correct segmental output . Unsurprisingly , the simulation does a bad job of recapitulating the patterns of pair-rule gene expression seen in real embryos ( see below ) . In particular , at the end of the simulation , there is no en expression anywhere at all , and neither odd nor slp is expressed in a segmental pattern . I then simulated a scenario in which the gap gene inputs and hence the Hairy and Eve stripes shift anteriorly over time . Given that the shift rate in real embryos is of the same order as the synthesis and decay rates of the segmentation gene products , I set the rate of these shifts to be such that the time taken for an expression domain to shift anteriorly by 1 cell is equal to the synthesis/decay delay parameter value of the pair-rule genes , i . e . , 6 time steps ( see S2 Text ) . The simulation output for this scenario is shown in S10 Movie and summarised in Fig 2B . Even though the pair-rule network is unchanged and the Hairy and Eve stripes retain the same pattern and relative phasing as for the static simulation , the model now performs completely differently . Qualitative aspects of actual pair-rule gene expression ( i . e . , whether the expression domains of each pair of genes are congruent , overlapping , abutting , or separate , and the way this changes over time ) are recapitulated remarkably well . For all pair-rule genes except prd , the match between the model output and the real spatiotemporal dynamics of gene expression is about as close as could be achieved by a simple , Boolean model—a few examples are highlighted in Fig 3 , and the full set of comparisons is shown in S1 Fig . For prd , the real spatiotemporal expression profile is only partially recovered: the early pair-rule stripes are positioned correctly but do not refine correctly at later stages—they narrow rather than split , meaning that alternate segmental stripes are missing from the final pattern ( S3 Fig ) . However , the prd domains missing from the simulation are not actually required for segment boundary patterning in real embryos ( they are not reflected in the larval cuticles of prd mutants , although they do have minor effects on wg expression [6 , 31 , 95] ) . Accordingly , the simulation still generates the correct final segmental output: a repeating pattern of En , Odd , Slp x2 , En , Odd , Slp x2 . These results tell us a number of things . First , it is not strictly necessary to invoke morphogen gradients in order to account for Drosophila pair-rule patterning . Second , posterior-to-anterior shifts of the Hairy and/or Eve stripes appear to be crucial for properly patterning the other pair-rule genes , and analysing the different behaviour of the static and shifting simulations should reveal exactly why . Third , the model as formulated is too simple to explain important aspects of the prd expression profile . Additional complexities that influence prd expression in real embryos could include ( 1 ) additional spatial or temporal regulatory inputs missing from the model , ( 2 ) quantitative information from existing spatial or temporal inputs that is not captured by the use of Boolean variables , or ( 3 ) differential synthesis/degradation rates of particular segmentation gene products not accounted for by the equal time delays assumed by the model . At least the first option seems to apply , as I have recently discovered that the Sox transcription factor Dichaete [96 , 97] also affects prd regulation [87] . Above , I described how the final segmental output consists of the pattern [En , Odd , Slp , En , Odd , Slp] across each double-parasegment repeat . I then used a dynamical model of the pair-rule system to show how the Eve stripes are directly or indirectly responsible for patterning most of the expression boundaries in this pattern , including both sets of parasegment boundaries . In this section , I use the same model to simulate and dissect the eve mutant phenotype , which has proved hard to account for using traditional patterning models . Although eve was originally identified as a pair-rule gene on the basis of a pair-rule cuticle phenotype [3] , it turned out that this particular mutant allele was an eve hypomorph , while eve null mutants yield an aperiodic denticle lawn phenotype instead [98] . Both odd-numbered and even-numbered en stripes are absent from eve null mutant embryos [71] , indicating severe mispatterning of upstream pair-rule gene expression . To investigate the aetiology of these effects , I characterised pair-rule gene expression patterns in precisely staged eve mutant embryos using double fluorescent in situ hybridisation ( FISH ) ( Fig 7 ) and then cross-referenced these observations with the patterning output of an “in silico” eve mutant ( Fig 2C; S11 Movie ) , simulated by starting with the dynamic model and then setting eve transcription to remain off ( see S2 Text ) . The experimental results are in accordance with earlier , more fragmentary characterisations of eve mutants [30 , 64 , 73 , 74 , 99–101] and reveal a number of significant changes to pair-rule gene expression patterns . The odd and ftz primary stripes are broader than usual ( Fig 7A ) , and the early expression of prd and slp is largely aperiodic rather than pair rule ( Fig 7A and 7C ) . Then , at gastrulation , segmental patterns of pair-rule gene expression fail to emerge; instead , prd resolves into broad pair-rule stripes ( Fig 7D ) , runt and slp become expressed fairly ubiquitously ( Fig 7F ) , and odd and ftz expression largely disappears ( Fig 7E ) . These changes are also seen in the eve mutant simulation ( Fig 2C ) and , as described below , follow logically from the structure of the pair-rule network . Because Eve expression plays a relatively minor role in late patterning , most of the expression changes just described result from the loss of Eve activity during cellularisation ( time points 0–36 in Fig 2C and S11 Movie ) . First , odd , ftz , and prd , all of which are direct targets of Eve repression ( Fig 7A′ ) , are expressed ectopically: the odd and ftz primary stripes expand anteriorly ( judged relative to hairy , Fig 7A , top/middle rows ) , while the prd interstripes ( i . e . , the gaps between the early broad stripes ) are derepressed ( Fig 7A , bottom row ) . The ectopic Odd expression has a knock-on effect on runt expression , which becomes down-regulated ( Fig 7B and 7B′ ) . ( Note that while runt expression is almost entirely lost at this stage of the simulation , in real embryos runt remains fairly widely expressed , although at significantly lower levels—see S5 Fig . This difference may be partially due to expression from the runt stripe-specific elements , which are not included in the model . ) The loss of Runt activity then contributes to the misexpression of the slp primary stripes , which turn on at the end of cellularisation and would normally be patterned by both Eve and Runt ( Fig 7C′ ) . Given the absence of Eve activity and the loss/weakening of Runt activity , slp becomes expressed almost ubiquitously within the trunk of the mutant embryos ( Fig 7C ) . Finally , the downstream effects of these aberrant patterns play out over the course of gastrulation , after the switch to the late network ( time points 37–60 in Fig 2C / S11 Movie ) . Odd represses prd , causing the aperiodic domain of prd to resolve into a pair-rule pattern ( Fig 7D and 7D′ ) . Odd and Ftz also repress slp , causing some small gaps to appear in the slp pattern ( Fig 7E and 7F ) . However , odd and ftz are themselves repressed by Slp very strongly ( Fig 7E′ ) , and the ectopic Slp expression in the embryo causes their expression to be almost completely lost ( Fig 7E ) . In , contrast , new runt expression emerges throughout most of the trunk , due to the absence of its repressors , Eve and Odd ( Fig 7F and 7F′ ) . As a consequence of all this mispatterning , en expression is completely repressed , and parasegment boundaries never form . The odd-numbered en stripes are specifically blocked by the ectopic Slp expression that replaces the Eve stripes . Above , we saw that these en domains are specified by the short regulatory chain [Eve––| Slp––| En] , thus explaining why they have been observed to reappear in eve , slp double mutants [8 , 101] . On the other hand , the even-numbered en stripes are redundantly repressed in eve mutants , by both ectopic Odd and ectopic Slp , as a result of the regulatory chains [Eve––| Odd––| En] and [Eve––| Odd––| Runt––| Slp––| En] , respectively . The model therefore explains why these stripes reappear in eve , odd double mutants [102] but not in eve , slp double mutants [8] . By simulating the Drosophila pair-rule network , I have shown that dynamic spatial inputs from just 2 factors , Hairy and Eve , are sufficient to organise the expression of the system as a whole . In the Drosophila blastoderm , these inputs are driven by the dynamic output of the posterior gap system . However , the elaborate control of pair-rule gene expression by gap factors appears to be a relatively recent novelty in arthropod segment patterning , originating during the evolutionary transition from short-germ to long-germ embryogenesis [103 , 104] . It is currently not clear how much of the cross-regulation between the pair-rule genes seen in Drosophila is a new adaptation to long-germ development and how much is retained from a short-germ ancestral state . To explore this question , I determined how many changes to the “wild-type” Drosophila simulation it would take in order to produce something resembling a sequential , “clock-and-wave-front” mode of segmentation . I managed to achieve this transition by way of a few plausible alterations , leaving the bulk of the network untouched ( Fig 8C ) . The only explicit regulatory changes that I made to the simulation were to the control logic of eve , hairy , and opa . First , I removed the regulation of eve by gap inputs ( G2 in the model ) and replaced it with early repression from Runt and Odd , both of which repress the eve late element in wild-type embryos ( Fig 1A , right ) . Second , I removed the regulation of hairy by gap inputs ( G1 in the model ) and replaced it with autorepression , which is a common phenomenon for her/hes family genes [107 , 108] . Third , I tied the onset of opa expression to the decay of Cad by making Cad repress opa . These changes allowed the gap inputs G1 and G2 to be removed from the model and simplified the temporal control of the patterning process , putting everything downstream of Cad and Hairy . I took this modified version of the Drosophila network and explored how it behaved at the tissue level , given various sets of initial conditions . I found that if I set up a repeating initial phase gradient of Hairy expression ( i . e . , for each cell along a double-segment repeat , Hairy expression begins at a slightly different point in its feedback cycle ) and left initial Cad expression unchanged , the simulation output was essentially identical to before ( Fig 8A and S12 Movie ) . This is because the waves of oscillating Hairy expression that sweep through the tissue while the early network is active are sufficient to correctly organise the pair-rule stripes of all the remaining primary pair-rule genes ( including Eve ) , and therefore downstream patterning proceeds unperturbed . However , I also discovered that if I removed all spatial patterning of Hairy from the initial conditions and instead established an anterior-to-posterior decay gradient of Cad ( mimicking the retracting Cad domain seen in short-germ insects [87 , 109] ) , the simulation still produced the correct segmental output except that , now , this pattern emerged sequentially instead of simultaneously ( Fig 8B and S13 Movie ) . As Cad decays and disappears from progressively more posterior cells over time , Opa turns on in an anterior-to-posterior wave , switching cells from the early network over to the late network . Cells in the posterior portion of the tissue synchronously passage through a repeating sequence of pair-rule gene expression ( Hairy → Eve → Runt → Ftz/Odd → Hairy ) until this point and subsequently differentiate into particular segment-polarity fates . The modified network ( Fig 8C , right ) is thus compatible both with Drosophila-like pair-rule gene expression and with “clock-and-wave-front” pair-rule gene expression similar to that seen in , e . g . , Tribolium . This finding suggests ( 1 ) that many of the regulatory interactions within the Drosophila pair-rule network might be conserved in short-germ arthropod species and ( 2 ) that long-germ and short-germ segmentation involve essentially similar patterning mechanisms and dynamics . Specifically , segment-polarity fate appears to be determined in both cases by the intersection of 2 sets of temporal signals , those carried by the pair-rule genes and those carried by broadly expressed extrinsic timing factors . In long-germ embryos , a periodic pattern of segment-polarity fates is achieved by directly imparting spatial information to the pair-rule genes using the gap system and stripe-specific elements . In short-germ embryos , however , retracting wave fronts of the timing factors provide this spatial information , and the gap genes play other roles .
In this manuscript , I have analysed the structure and dynamics of the Drosophila pair-rule network using a combination of simulated and experimental data to reveal how segment patterning is achieved . I have discovered a functional role for dynamic gap inputs in correctly phasing the pair-rule stripes and propose revised mechanisms for the patterning of the odd-numbered and even-numbered parasegment boundaries . In contrast to previous models based around the principle of static morphogen gradients , these mechanisms involve a coordinated interplay between intrinsic network dynamics and extrinsic spatiotemporal signals and do not necessarily require graded pair-rule activity . These findings contribute to the evolving view of the role of Even-skipped , perhaps the best known of the pair-rule factors . Eve has long been known to be required for the expression of both sets of en stripes and hence both sets of parasegment boundaries [71 , 98] . Originally , Eve was thought to achieve this directly by activating en . Later , it was recognised that Eve does not regulate en directly but instead represses several other pair-rule factors that themselves repress en [75 , 76]; however , quantitative information inherent within the Eve stripes was still believed to establish the template for the en stripes [8 , 30] . This conclusion is challenged by the new model presented here , which suggests that static domains of Eve expression would cause a similar degree of pattern loss to that seen in eve mutant embryos . Instead , I propose that Eve conveys positional information largely via its expression dynamics , which are decoded downstream by the rest of the pair-rule gene network . ( Interestingly , while “French flag” type morphogen gradient mechanisms have been proposed to underpin many developmental patterning systems , several modern studies have found that the reality often involves complex dynamics [58 , 110–113] . ) The findings also clarify the relationship between long-germ and short-germ segmentation . I have shown that pair-rule patterning is fundamentally a temporal process and that it is therefore straightforward to convert the Drosophila pair-rule network into a clock-and-wave-front system . Under this scenario , the dynamic “early” network regulates oscillatory expression in the posterior of the tissue , while the switch-like “late” network stabilises these outputs into segmental patterns in the anterior . Given that long-germ insects such as Drosophila are derived from short-germ ancestors , it seems unlikely that this fluidity between simultaneous and sequential patterning modes is a coincidence . I therefore propose that output from the Drosophila early pair-rule network is developmentally homologous to the oscillatory pair-rule gene expression seen in the segment addition zone of short-germ insects such as Tribolium , while the output from the late pair-rule network is developmentally homologous to the stripe refinement and frequency doubling that occurs just anterior to the segment addition zone ( Fig 8D ) . This hypothesis is strongly supported by a recent comparative study carried out by Andrew Peel and myself , which found that different phases of pair-rule gene expression in Tribolium exhibit the same correlations with cad and opa expression as seen in Drosophila [87] . Strikingly , the hypothesis implies that the “zebra” elements of runt and odd ( which drive pair-rule patterns in the Drosophila blastoderm ) represent the relictual components of some ancestral segmentation clock that drove oscillations of these genes with double-segment periodicity [50] . The evolutionary transition from short-germ segmentation to long-germ segmentation is thought to have been driven largely by selection for faster development . A segmentation clock can only tick so fast , and therefore represents a tight production bottleneck when generating segments: a longer body requires a longer time ( S7A Fig ) . ( Selection for quicker development may also underlie the evolutionary success of “pair-rule” patterning , seen in centipedes [114] as well as insects [115 , 116] , which effectively halves the number of clock cycles required to produce a body of a given length . ) In contrast , segmentation can be accomplished remarkably quickly when pair-rule gene expression is initialised with a periodic spatial pattern , as in Drosophila . Mechanistically , the transition to long-germ segmentation seems to have involved 2 main changes: ( 1 ) a heterochronic shift in the deployment of the pair-rule network from posterior segment addition zone to blastoderm and ( 2 ) the elaboration of the posterior gap network and associated stripe-specific elements . These latter processes are likely to have occurred progressively along the AP axis , with intermediate forms exhibiting a composite mode of development: simultaneous for more anterior segments , sequential for more posterior segments [103] . The process also seems to have been facilitated by the re-use of anterior stripe-specific elements to also drive more posterior pair-rule stripes [23 , 117 , 118] . However , given the complexity of Drosophila pair-rule patterning , it has not been clear how this process could have evolved gradually and seamlessly , all the while maintaining a perfect segmental output pattern . The findings in this paper significantly mitigate this problem . In my model of the Drosophila pair-rule network , 2 of the primary pair-rule genes are patterned by gap inputs while 3 are patterned by cross-regulation . However , the system works equally well if only 1 of these genes is explicitly patterned and the rest are cross-regulated ( S12 Movie ) or if all 5 are patterned by gap inputs and there is no cross-regulation ( S8 Movie ) . Therefore , direct stripe patterning through stripe-specific elements and indirect stripe patterning through zebra elements can stand in for one another without affecting events downstream . ( Similar conclusions , including a potential role for stripe shifts , were reached by a recent in silico evolution study focused on the Drosophila pair-rule genes [119] . ) Consequently , there is no need for multiple stripe-specific elements to have evolved simultaneously during the transition to long-germ segmentation: new stripe-specific elements could have evolved one by one , each time slotting into or replacing parts of the existing patterning machinery . Retention of an intact segmentation clock could also have buttressed gap-driven segment patterning during the transition , and such a scenario may apply today to insects such as Nasonia and Bombyx , which exhibit some evidence of both patterning modes [120–122] . The corollary of such flexibility between gap-driven and cross-regulated pair-rule gene expression is , however , that the dynamics of the gap network need to be highly constrained and should resemble those of the original segmentation clock . This appears to be the case [67] . Above , I stated that stripe-specific elements and cross-regulation are theoretically interchangeable for positioning pair-rule stripes . In the Drosophila blastoderm , the regulatory situation appears to be partially redundant: at least 3 of the primary pair-rule genes possess stripe-specific elements in any given pair-rule repeat , but only 2 of these genes lack zebra elements [16] . Why then have multiple pair-rule genes evolved stripe-specific elements , when patterning just a single gene in this way is theoretically sufficient for making segments ? Additionally , why do genes such as runt retain a zebra element when they already have a full set of stripe-specific elements ? The first question has 2 likely answers . Using gap inputs to directly pattern multiple genes speeds up the first emergence of a correctly phased pair-rule repeat and hence reduces the total time required for segmentation as well as the minimum magnitude of the stripe shifts ( S7B Fig; S6–S8 Movies ) . ( The extreme scenario is that gap shifts are dispensed with entirely , as seems to have occurred in the anterior of the Drosophila trunk , based on comparisons with the scuttle fly Megaselia abdita [123 , 124] . ) The acquisition of extra stripe-specific elements could therefore have been driven by selection for marginal decreases to the length of development . Alternatively , these elements could have been selected for in order to mitigate irregularities in the patterns of upstream pair-rule genes . For example , runt , ftz , and odd all possess a stripe-specific element for stripe 3 [16] , corresponding to a region of the blastoderm in which their repressor , Hairy , is inappropriately expressed early on [59] . While the stripe-specific elements of runt , ftz , and odd may therefore increase the speed of patterning , their output patterns are not as regular as those of hairy and eve , probably answering the second question . Irregular gap-driven pair-rule stripes , if not later refined by cross-regulation , would result in aberrant segment-polarity patterns . For example , cross-regulation of the runt pair-rule stripes appears to be absent in Dichaete mutant embryos , and this leads to severe and variable segmentation defects [87] . However , even if these runt stripes were as regular as those of eve , cross-regulation might still be required . The eve stripes are patterned with a precision of about 1% AP length , i . e . , differing in position from embryo to embryo by roughly 1 nucleus [41] . This positional noise is equivalent to the spatial resolution of the final segment-polarity pattern , in which most stripes are only 1 cell wide ( Fig 1B ) . Therefore , if the eve and runt stripes both showed positional variation of this magnitude , and this noise was uncorrelated , segmentation defects would likely occur . The benefit of the dynamic patterning carried out by the pair-rule network is that the influence of Eve activity on gene expression is not restricted to the cells currently within each Eve stripe but also extends to cells posterior to these stripes , like wakes stretching out behind moving ships . This means that entire pair-rule repeats become coordinated with the Eve stripes , regardless of where exactly these stripes are located in the blastoderm . Thus , spending extra time to cross-regulate is likely to minimise the consequences of noise in the original spatial signals , a trade-off that is characteristic of many developmental patterning systems [125] . I have presented theoretical evidence that posterior-to-anterior shifts of pair-rule stripes , driven by dynamic gap inputs , play a functional role in segment patterning . The plausibility of my model therefore rests on these shifts existing and being of an appropriate speed and magnitude . The basic model ( only hairy and eve patterned by gap inputs ) requires a minimum shift of 3 nuclei , while versions including greater control of pair-rule gene expression by gap inputs reduce this minimum to 1 or 2 nuclei ( S6–S8 Movies ) . As yet , these shifts have only been measured in real embryos by comparing population estimates of absolute stripe position ( in percent egg length ) in fixed material of different ages [59 , 61] . Interpreting these measurements is not straightforward , but adjustments for nuclear migration suggest shifts of at least 2–3 nuclei for eve stripes 3–7 , consistent with the requirements of the model . Recent developments in live imaging [126–128] mean that it should now be possible to test these estimates directly in individual embryos . While the new model clarifies many aspects of segment patterning , it also highlights certain regulatory phenomena that are currently unexplained and require further study . For example , it is not clear how the timing of slp transcription is controlled , how exactly prd is spatially regulated , or which factors are responsible for activating the pair-rule genes . Finally , the model is obviously simplistic and represents only the first step towards a modern , system-level understanding of the pair-rule network . For example , while the model demonstrates that morphogen-like effects are not necessary to account for pair-rule patterning , it does not rule them out . ( Indeed , anterior-to-posterior expression shifts might well contribute to concentration-dependent spatial patterning of target genes , by helping shape the contours of the pair-rule stripes at the protein level . ) Careful experimental characterisation of the quantitative relationship between the protein concentrations of input factors and the transcription rate of output factors could form the basis for future , more sophisticated models of Drosophila pair-rule patterning , which explore the roles of quantitative and stochastic effects . Future models of short-germ segmentation , on the other hand , could account for morphogenetic processes such as cell division and rearrangement [129–131] and/or incorporate known intercellular communication processes that occur both upstream ( e . g . , Notch-Delta signalling [132–134] ) and downstream ( e . g . , Toll receptor codes [135 , 136] ) of the pair-rule genes . In the introduction to this paper , I highlighted 3 recent findings related to pair-rule patterning: ( 1 ) that pair-rule orthologs exhibit oscillating expression in the segment addition zones of short-germ arthropods , ( 2 ) that pair-rule gene expression in long-germ insects is patterned by dynamic gap gene expression , and ( 3 ) that the pair-rule network in Drosophila undergoes an extensive topology change at gastrulation . I then presented evidence that dynamic Eve expression is integral to segment patterning and also that the Drosophila pair-rule network is broadly compatible with both simultaneous and sequential modes of segmentation . I therefore propose the following evolutionary hypothesis , which ties together and potentially explains the 3 original findings . ( 1 ) Long-germ and short-germ segmentation are not dichotomous modes of patterning but rather alternative behaviours of a largely conserved pair-rule network . ( 2 ) In long-germ insects , ad hoc patterning of stripes by gap inputs stands in for the expression of particular pair-rule gene “clock enhancers” , meaning that gap gene networks are under strong selection to preserve dynamic behaviour . ( 3 ) In Drosophila , the “early” pair-rule network is derived from the “clock” part of an ancestral short-germ segmentation mechanism , while the “late” network is derived from the stabilising genetic interactions that would have originally established parasegment boundaries anterior to the segment addition zone . These predictions can be tested in the future by comparative studies in emerging arthropod model species .
Wild-type in situ images are from a previously published data set deposited in the Dryad repository: http://dx . doi . org/10 . 5061/dryad . cg35k [137] . The eve mutation used was eve3 ( gift of Bénédicte Sanson ) and was balanced over CyO hb::lacZ ( Boomington stock no . 6650 ) in order to easily distinguish homozygous mutant embryos . Whole mount double FISH , microscopy , and image analysis were carried out as described previously [32] . The eve/Eve doubles in Fig 4D were generated by incubating embryos with 1:1 , 000 rabbit anti-Eve [138] and 1:500 anti-rabbit 488 following in situ hybridisation to eve . Images were analysed using Fiji [139 , 140]: intensity profiles were produced using the “multi plot” function , while thresholded images were produced using the “make binary” tool . Simulations were coded in Python ( www . python . org ) using the libraries NumPy [141] and Matplotlib [142] . SBML-qual files were generated using the GINsim software ( http://ginsim . org ) [143] . All models and simulations are described in detail in S2 Text . | Segmentation in insects involves the division of the body into several repetitive units . In Drosophila embryos , all segments are patterned rapidly and simultaneously during early development , in a process known as “long-germ” embryogenesis . In contrast , many insect embryos retain an ancestral or “short-germ” mode of development , in which segments are patterned sequentially , from head to tail , over a period of time . In both types of embryo , the patterning of segment boundaries is regulated by a network of so-called “pair-rule” genes . These networks are thought to be quite divergent due to the different expression patterns observed for the pair-rule genes in each case: regularly spaced arrays of transient stripes in Drosophila , and dynamic expression within a posterior “segment addition zone” in short-germ insects . However , even in Drosophila , a clear understanding of pair-rule patterning has been lacking . Here , I make a computational model of the Drosophila pair-rule network and use simulations to explore how segmentation works . Surprisingly , I find that Drosophila segment patterning relies on pair-rule gene expression moving across cells over time . This conclusion differs from older models of pair-rule patterning but is consistent with the subtly dynamic nature of pair-rule stripes in real embryos , previously described in quantitative studies . I conclude that long-germ and short-germ segmentation involve similar expression dynamics at the level of individual cells , even though they look very different at the level of whole tissues . This suggests that the gene networks involved may be much more conserved than previously thought . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"invertebrates",
"genetic",
"networks",
"morphogenic",
"segmentation",
"molecular",
"probe",
"techniques",
"gene",
"regulation",
"animals",
"simulation",
"and",
"modeling",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"network",
"analysis",
"experimental",
"organism",
"systems",
"molecular",
"biology",
"techniques",
"embryos",
"morphogenesis",
"drosophila",
"research",
"and",
"analysis",
"methods",
"embryology",
"computer",
"and",
"information",
"sciences",
"probe",
"hybridization",
"gene",
"expression",
"molecular",
"biology",
"insects",
"arthropoda",
"fluorescent",
"in",
"situ",
"hybridization",
"eukaryota",
"cytogenetic",
"techniques",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
]
| 2017 | Dynamic patterning by the Drosophila pair-rule network reconciles long-germ and short-germ segmentation |
The genetic basis of the development and variation of adult form of vertebrates is not well understood . To address this problem , we performed a mutant screen to identify genes essential for the formation of adult skeletal structures of the zebrafish . Here , we describe the phenotypic and molecular characterization of a set of mutants showing loss of adult structures of the dermal skeleton , such as the rays of the fins and the scales , as well as the pharyngeal teeth . The mutations represent adult-viable , loss of function alleles in the ectodysplasin ( eda ) and ectodysplasin receptor ( edar ) genes . These genes are frequently mutated in the human hereditary disease hypohidrotic ectodermal dysplasia ( HED; OMIM 224900 , 305100 ) that affects the development of integumentary appendages such as hair and teeth . We find mutations in zebrafish edar that affect similar residues as mutated in human cases of HED and show similar phenotypic consequences . eda and edar are not required for early zebrafish development , but are rather specific for the development of adult skeletal and dental structures . We find that the defects of the fins and scales are due to the role of Eda signaling in organizing epidermal cells into discrete signaling centers of the scale epidermal placode and fin fold . Our genetic analysis demonstrates dose-sensitive and organ-specific response to alteration in levels of Eda signaling . In addition , we show substantial buffering of the effect of loss of edar function in different genetic backgrounds , suggesting canalization of this developmental system . We uncover a previously unknown role of Eda signaling in teleosts and show conservation of the developmental mechanisms involved in the formation and variation of both integumentary appendages and limbs . Lastly , our findings point to the utility of adult genetic screens in the zebrafish in identifying essential developmental processes involved in human disease and in morphological evolution .
The genetic and developmental basis of the formation of organismal shape and form is a long-standing question in biology . The analysis of mutations has been essential in identifying the genes and regulatory networks underlying development . However , while the genetic basis of embryonic development has been extensively studied by systematic mutagenesis screens , we know little of the genes involved in the development of adult morphology . Yet , it is the heritable variation in adult form that natural selection primarily acts on during evolution . In order to understand the basis of variation , we need to know more about the genetic control of the development of adult form: which genes are involved , what are their function , and when are they required in development [1] , [2] . To identify genes important for development of adult structures , we initiated a large-scale mutagenesis screen in zebrafish and scored for mutants affected in the shape and pattern of adult structures . We isolated only adult viable mutants , therefore we selected for genes that have an increased probability to be involved in morphological change during evolution . Identification of zebrafish genes homologous to human genes associated with disease that arise during postnatal development into adulthood is also likely in this screen . We focused on mutants that exhibit defects in the dermal skeleton of the adult zebrafish . The dermal skeleton encompasses the external form of the adult fish . The most prominent dermal skeletal elements are the dermocranium of the skull and lateral bones of the opercular series , the scales , and the fin rays ( or lepidotrichia ) . Additionally , the teeth and gill rakers ( bones that support the gills in teleosts ) are elements of the dermal skeleton [3] , [4] . Unlike the ossification process that occurs during endochondrial bone development in which organic matrix is deposited by osteoblasts over a chondrogenic scaffold , dermal skeletal elements originate as direct mineralization of a collagenous matrix deposited by dermal fibroblasts . This process occurs in close association with the epidermis . The initiation and patterning of dermal elements are thought to be similar to epidermal appendages ( e . g . hairs and feathers ) and is controlled by reciprocal signaling between an epithelium and mesenchyme ( see [5] , [6] ) . Importantly , in zebrafish , as in most teleosts , the majority of dermal skeletal elements are not formed during larval development , rather through juvenile metamorphosis and development of the adult pattern . Those that begin to form in late larval development such as the teeth and gill rakers , do not fully attain their shape and pattern until juvenile metamorphosis . Variations in the shape of dermal skeletal elements of the fins , scales , cranium , and teeth play a significant role in adaptations of fish populations to new environments ( e . g . dermal plate development and stickleback radiation [7] ) . Additionally , integumentary appendages , such as hair and feathers , have been essential and defining traits of vertebrate classes . Early vertebrates , the conodonts , ostracoderms and placoderms , possessed a pronounced dermal skeleton , often in the absence of an ossified axial skeleton [8] . Through vertebrate evolution from fish to tetrapods , dermal structures such as lateral bones of the opercular series , scales , dermal plates and fin rays were either reduced or lost . This evolutionary transition was paralleled with the elaboration of the cartilaginous endoskeleton of the limbs and the evolution of specialized keratinized appendages of the integument such as epidermal scales , feathers and hairs . In contrast , the diversity of form in extant bony fishes involves modification in size , shape and number of the scales/dermal plates , fin rays , cranial dermal bones and teeth . Here , we describe a collection of mutants that have shared defects in the formation of the dermal skeletal elements of the skull , fins , scales and teeth of the adult zebrafish . The mutations disrupt the genes ectodysplasin ( eda ) and edar encoding the eda receptor . In mammals the EDA signaling pathway is involved in hair and teeth formation [9] and mutations affecting this pathway cause the human hereditary disease hypohidrotic ectodermal dysplasia ( HED ) . Loss of Eda signaling in the zebrafish causes a spectrum of phenotypes corresponding to those described for HED in humans , and therefore the zebrafish mutants may serve as a genetic model of this disease . We describe the requirement of Eda signaling in the zebrafish epidermis for the formation of a structure resembling an epidermal placode seen in the early development of other vertebrate integumentary appendages . The mutations also result in defects of skeletal elements unique to fish and suggest an ancestral role of Eda signaling in the formation and patterning of the dermal skeleton . Lastly , whereas loss of function of Eda signaling causes a severe phenotype , the expressivity of dominant alleles is sensitive to background modifiers that buffer the phenotypic consequences of loss of Eda signaling . Additionally , we find that the response to reduction of Eda signaling is dose sensitive and organ specific . We suggest that such alleles may provide a basis for morphological variation in evolution .
In a mutagenesis screen for mutations affecting adult zebrafish structures , we identified three mutants that showed nearly identical defects in the formation of scales , lepidotrichia , and shape of the skull of homozygous fish . These mutants fell into two complementation groups . The first is allelic to the finless ( flste370f ) mutant that was previously isolated in the background of the Tübingen wild type stock ( Tü ) on the basis of the loss of fins in adults [10] . We isolated two new alleles of fls in the screen and found another in the background of the TLF wild type stock . The majority of the fls alleles isolated are recessive and have a strong phenotype ( see below ) . However , the flsdt3Tpl allele is dominant with a partial scale loss phenotype in heterozygotes ( Figure 1G , J ) . One further fls allele was isolated in a screen for mutations that failed to complement the flste370f mutant ( Figure 1M ) . We named this allele fang ( flstfng ) after its unique dental phenotype in homozygotes of having only one tooth on the fifth ceratobranchial ( Figure 1N ) . The flstfng allele shows no effect on fin development and has a slight increase in the number of scales than the other fls alleles isolated . The second complementation group was comprised of a single gene , which we called Nackt ( Nkt ) . This allele is dominant causing a slight defect in the patterning and shape of scales as heterozygotes ( Figure 2D ) . The homozygous phenotype is more severe than that of strong fls alleles ( Figure 2A ) . Phenotypic defects of fls and Nkt mutants become apparent in juvenile fish; as larvae , homozygous fls and Nkt mutants are visibly unaffected . Homozygous adults are viable , and of normal size . With the exception of the flstfng allele , the lepidotrichia that form during juvenile metamorphosis are defective , leading to fin loss in the adult ( Figure 1D , 2A ) . The dermal bones of the pectoral girdle are present and patterned appropriately in both the fls and Nkt mutants . By close examination of the visceral skeleton we found that neither the pharyngeal teeth , nor the bony substrates of the gills , the gill rakers , are formed ( Figure 1E and F; Figure 2B , C ) . In addition , scales are largely absent with infrequent formation of inappropriately shaped scales near the dorsal , anal and pectoral fins ( Figure 1D , 2A ) . Nkt homozygous fish exhibit more severe defects in the formation of the dermal skeleton than fls alleles in the extent of lepidotrichial growth and number of residual scales formed ( compare Figure 1D and Figure 2A ) . The skull of mutants has a normal appearance with all the bones being present , although the size , shape and relative proportion of the various bones differ compared to wild type individuals ( Figure S1 ) ; no change in cranial shape was apparent in larvae . Nkt heterozygous fish exhibit a dominant phenotype as they lack several scales on the flank and those present at the flank are elongated dorso-ventrally . The number of teeth and gill rakers is reduced , however lepidotrichia formation and growth of the fins are not affected ( Figure 2D–F ) . The skulls of Nkt heterozygotes do not show increased size , but retain altered shape and proportion as seen in homozygotes ( Figure S1 ) . We isolated a dominant allele of fls that exhibits a distinct phenotype in heterozygotes that we named Topless ( flsdt3Tpl ) . Heterozygous flsdt3Tpl have a reduction in the number of scales , teeth and gill rakers , but show little to no effect on lepidotrichia development ( Figure 1G–I ) . Mutant flsdt3Tpl fish exhibit the strong fls phenotype when homozygous or heterozygous with other fls alleles . Similar to Nkt , flsdt3Tpl exhibits a dominant effect on skull shape as well ( Figure S1 ) . The expressivity of the dominant flsdt3Tpl phenotype depends on the genetic background . Fish heterozygous for flsdt3Tpl exhibited either a “strong” or a “weak” phenotype in the Tü background ( Figure 1G and J , respectively ) . The “strong” phenotype shows loss of scales regionally in the midflank , loss of medial pharyngeal teeth along the fifth ceratobranchial and loss of posterior gill rakers of the anterior arches ( Figure 1G–I ) . In contrast , the “weak” phenotype displays only subtle variation in scale patterning and no effect on the teeth or gill rakers could be detected ( Figure 1J–L ) . The segregation pattern of the two phenotypic classes of flsdt3Tpl suggests the presence of separate , unlinked , modifier loci in the Tü background affecting the number of scales ( Table 1 and data not shown ) . Additionally , we found that the flsdt3Tpl “strong” phenotype was partially suppressed when crossed with the polymorphic WIK mapping strain indicating the presence of dominant modifier ( s ) in the WIK line ( Table 1 ) . The resulting heterozygous progeny showed reduced scale-loss compared to flsdt3Tpl heterozygotes in a Tü background , but had similar defects in scale shape ( Table 1 ) . Therefore , dominant modifier loci are present in the WIK strain that buffer the expressivity of the flsdt3Tpl dominant phenotype . None of the other fls alleles showed any dominance in the Tü , TLF , or WIK strains . We identified the affected loci of the fls mutants by positional cloning . The flste370f mutation was linked to SSLP markers on linkage group 9 ( LG9 ) . Due to similarity of the fls phenotype to ectodermal dysplasia phenotypes in mammals , we mapped several genes of the ectodysplasin pathway to the zebrafish radiation hybrid map to see if any of these genes were linked to fls . The edar gene is located on LG9 within the determined linkage interval for fls ( see Methods ) . We cloned the full-length wild type cDNA of edar and found several polymorphisms in the Tü edar cDNA when compared to the WIK mapping strain; these polymorphisms were tightly linked with the fls mutation and did not show recombination in 238 meioses ( Figure S2 ) . The edar gene encodes a transmembrane protein with similarity to tumor necrosis factor receptor ( TNFR ) . The Edar protein contains a conserved TNFR extracellular ligand binding domain and a cytoplasmic terminal death domain essential for protein interactions with signaling adaptor complexes . The flste370f mutation is an A to T transversion at a splice acceptor site , resulting in missplicing of the mRNA leading to a frame shift in translation and the generation of a premature stop codon ( Figure 3B and Figure S2 ) . This allele is a likely molecular null mutation as only a fragment of the ligand-binding domain is present while the transmembrane and cytoplasmic death domains , which are essential for function of this protein , are both absent . The spontaneous mutation flst0sp212 was found to have a splicing defect leading to the inclusion of intronic sequence . This is predicted to form a protein with incorrect amino acid sequence after residue 212 , at the end of the transmembrane domain leading to a premature termination codon ( Figure 3B , Figure S2 ) . The two alleles generated by the ENU mutagen both have missense mutations resulting in amino acid changes in the death domain ( flst3R367W , R367W ( C-T ) ; flsdt3Tpl , I428F ( A-T ) ) . These mutations were found at identical positions as seen in familial cases of HED in humans ( Figure 3B , E; [11] , [12] . The fang allele of fls was isolated in an allele screen for mutants that failed to complement flste370f ( Figure 1P ) . flstfng homozygotes do not show any observable effect on lepidotrichia development yet have a reduction of scales and teeth/rakers as seen in other fls alleles ( Figure 1M–O ) . The fang allele in trans to the te370f putative null allele shows an intermediate phenotype affecting lepidotrichial growth and a further reduction of teeth and scales suggesting that the fang allele is a hypomorph ( Figure 1P–R ) ; flstfng heterozygotes do not show any differences compared to wild type . The shape and number of the scales in fang is similar to the other homozygous fls alleles ( Table 1 ) . Analysis of edar RNA from homozygous flstfng showed the presence of two distinct transcripts with an additional larger isoform than seen in wildtype . Analysis of the sequence of the novel isoform showed the addition of intronic sequence leading to a premature termination codon ( Figure 3C ) . The predicted protein would be similar to the flst0sp212 allele having truncation just after the transmembrane domain at amino acid 218 ( Figure 3B and Figure S2 ) . Analysis of the genomic sequence in the mutant revealed that the altered splicing is due to an A to G transition leading to the creation of a new splice donor site in the intron ( Figure 3C ) . Given the presence of both isoforms in homozygous individuals , this novel splice site is used in addition to the normal splice junction . Using quantitative real time PCR we found that the fang-specific edar transcript represents 74% of the total pool of edar transcripts in homozygous mutants ( Figure 3D ) . The dilution of wild type transcripts can explain the observed hypomorphic effect of the allele . From this unique allele of fls , it is clear that the phenotypic effect of loss of Eda signaling is dose dependent and that scales and teeth are more sensitive to alterations in the level of Eda signaling than fins . EDAR and its orthologue XEDAR recognize specific EDA isoforms that vary by two amino acids [13]–[16] . The receptor-ligand complex signals though NF-κB using several adaptor proteins that are generally specific to each receptor . Together , mutations in Edar and Eda lead to the majority of cases of human HED in which the development of integumentary appendages ( hairs , glands and teeth ) are affected ( OMIM 300451 , [17]; OMIM 604095 [18] ) . We reasoned that , because of the similarity in phenotype to fls , the Nkt gene could be eda , encoding the ligand for Edar . We isolated the entire coding region for zebrafish eda by RACE ( Figure S3 ) . The eda transcript from the Nkt mutant shows a precocious stop codon predicting a truncation of the protein at the beginning of the TNF domain , which is involved in ligand-receptor binding ( S243X ( C-A ) ) ; Figure 3G and Figure S3 ) . An analysis of the location of eda in the zebrafish radiation hybrid map placed eda on LG5 . Subsequent linkage analysis of the Nkt mutation and markers indicated by radiation hybrid analysis demonstrated tight linkage of the mutant to this region ( Figure 3F ) ; the S243X mutation was always found in fish with the Nkt phenotype and served as a consistent genotypic marker . In fish , scales are bony elements that develop in the dermis underlying the epidermis . In amniotes , most integumentary organs affected by loss of Eda signaling have structural derivatives stemming from the epidermis ( e . g . specific keratins of hair , feather and nail , secretory cells of glands ) . These integumentary organs develop from reciprocal signaling interactions between the basal epidermis and subjacent mesenchyme often controlled by a regional epithelial thickening called the epidermal placode . Eda signaling is necessary for the development and patterning of epithelial placodes of many integumentary organs in both the mouse and chick [19]–[22] . Expression of Eda and Edar is found predominantly in the basal epidermal cells , but in the case of feathers Eda is detected in the subjacent mesenchyme as well [21] , [22] . Whereas expression of developmental signaling genes such as sonic hedgehog ( shh ) in the development of integumentary appendages are comparable between vertebrates [23] , evidence for an early developmental role of the epidermis in induction or patterning of the teleost scale is lacking . The formation of an epithelial placode and signaling center in the development of amniote integumentary appendages is associated with histological changes in the basal cells of the epidermis; a similar structure has not been described in fish epidermis [24] . As early teleost scale development is quite different to that of other vertebrate integumentary organs , such as hairs and feathers , we addressed the question whether Eda signaling had a similar function in the epidermis of zebrafish during scale formation . We detected the expression of both edar and eda in the skin of juvenile fish by whole mount in situ analysis ( WMISH ) . The expression of both genes presaged the formation of the initial scale row along the flank just ventral from the midline mysoseptum in the caudal peduncle ( arrowheads , Figure 4A and C; [24] ) . During scale formation , the expression of edar becomes progressively restricted to the posterior margin ( Figure 4B ) while eda expression persists throughout the scale primordia ( Figure 4D ) . Developmental genes shh and bone morphogenic protein 2b ( bmp2b ) , whose orthologues are known to be essential for placode development in the mouse and chick , show similar placodal expression as seen with edar ( Figure 4E and G ) . Analysis of shh and bmp2b expression in flste370f indicated the necessity of edar function for their expression ( Figure 4F and H ) . We investigated the development of scale primordia in wild type and mutant flste370f fish by light and transmission electron microscopy . Previous detailed histological work found evidence for raised signaling activity in the epidermis as measured by increased endoplasmic reticulum ( ER ) , and secretory activity of the basal epidermal cells prior to scale formation [23] . However these changes in the basal epidermal cells were not associated regionally with sites of scale formation nor was there any indication of histological changes in basal cell morphology that are associated with placode formation in other vertebrates . To our surprise , in wild type juvenile fish , we discovered the formation of histologically defined , localized assemblies of cells of the basal epidermis that resemble early stages of the formation of hair and feather placodes . Prior to the development of the scale , the dermis consists of compact collagen layers ( stratum compactum ) and scattered dermal fibroblasts [24] . The epidermal basal cells have a uniform elongate morphology ( black arrows , Figure 5A , D ) with high levels of basally located intermediate fibrils ( Figure 5G ) . At the initiation of scale development , there is an accumulation of fibroblasts subjacent to the basal epidermis , associated with a reworking of the collagen strata [24] . We find a specific alteration in the morphology of the basal epidermal cells in wild type juvenile fish that coincides with the initial accumulation of fibroblasts at the sites of scale development ( Figure 5B , E ) . These basal cells become cuboidal and have decreased width , such that they form a unit of closely packed cells ( black arrows , Figure 5B , E ) . This is observed above the localized accumulation of fibroblasts in the dermis ( white arrows , Figure 5B , E ) . In addition , in these placodal-like cells , the ER appears less prominent ( data not shown ) , and hemidesmosomes , the cellular junctions involved in the attachment of the basal epidermal cells to the basal lamina , are almost completely absent ( brackets , Figure 5H and I ) . In contrast , the adjacent lateral epidermal cells show high levels of both ER and hemidesmosomes ( data not shown and Figure 5G brackets , respectively ) . In flste370f juvenile fish , at a corresponding site on the flank as in wild type , we detected the formation of similar aggregations of basal epidermal cells ( black arrows . Figure 5C , F ) . However , unlike the structures found in the wild type zebrafish , the epidermal cells of the placode were disorganized and showed histological evidence of cell death ( Figure 5F ) . As is the case in wild type , the epidermal basal cells in the placode of fls display a reduced ER , however hemidesmosomes are present in the same high numbers as in adjacent cells in wildtype ( brackets , Figure 5H compared to brackets Fig . 5I ) . Lateral basal epidermal cells in flste370f/edar showed elongate morphology similar to those of their wild type siblings ( data not shown ) . The expression of edar is seen in the basal cells of forming scale placodes ( Figure 5J–L; stages 1–3 Figure 5M ) arising during early specification of the scale placode ( s1; arrowhead Figure 5K ) . We were unable to detect eda expression in sections due to the weak hybridization signal . These data support the notion that an epidermal placode is involved in dermal scale formation . Further we find that the epithelial organization and function of the developing scale placode is dependent on edar . The phenotype of both Nkt and fls demonstrate that Eda signaling is necessary for fin development . The growth and patterning of lepidotrichia are affected in all fins . Lepidotrichia are specified , however fail to maintain growth and elaboration of the fin rays ( Figure 6A–C , H–P ) . Unpaired fins showed no defects in patterning of the endochondrial bones of the proximal and distal radials ( Figure 6K–P ) ; the dorsal pitch of the caudal fin is an indirect effect of the mutation on swimming without fin rays ( amputated fins that fail to regenerate show similar morphology ) . In fls adults , fusions of the distal radials of the pectoral fin are seen at a low penetrance ( Figure 6G , data not shown ) . In Nkt , there is an increase in the frequency of patterning and growth defects of the endochondrial components of the fin ( Figure 6G ) . These alterations include the loss of the fourth proximal radial , altered growth patterns of anterior proximal radials 1 and 2 , as well as lack of articulation of the distal radials ( Figure 6 D–F ) . Nkt causes a strong effect on lepidotrichial growth of both the pectoral and pelvic fins ( Figure 6E–F , J ) . In contrast , a specific effect on the growth of anterior lepidotrichia of the pelvic fin is seen in fls where the dermal rays of the anterior ( e . g . 1 , 2 ) are significantly shorter than rays at equivalent positions in wild type ( Figure 6I ) . The asymmetry of lepidotrichial development suggests that , like the proximal endochondrial fin skeleton , the fin rays have a specific regional identity to provide the shape and form of the fin . Early limb development is driven by a localized organization of epithelial cells at the distal tip of the forming limb , termed the apical ectodermal ridge ( AER ) [25] . In zebrafish , the AER is involved in larval patterning of the paired fins , while the later stages of fin development are organized by an analogous epidermal formation of the fin fold in both paired and unpaired fins [26] , [27] . In tetrapod limb development anterior-posterior specification is controlled by posterior mesenchyme expressing Shh . The function of this zone of polarizing activity ( ZPA ) expressing Shh is maintained by reciprocal signaling interactions between the ZPA and the AER . This interaction is necessary for proper patterning and growth of the tetrapod limb . In the zebrafish , shh and signals from the AER also orchestrate patterning and outgrowth of the early fin buds [28]–[30] . In addition , genes functioning early in fin development , such as shh , play important roles during late fin development regulating growth and branching of lepidotrichia growth [31] . We investigated the regulation of edar in mid to late fin development focusing on the development of the paired fins . In early fin fold stage of pectoral fin development ( 8 mm ) , we detected edar expression in both the distal margin of the endochondrial radials ( black arrowhead , Figure 7A ) as well as more distally in the forming lepidotrichial rays ( Figure 7A ) . The expression of edar in the fin fold had a posterior bias in wild type fins ( Figure 7A , white arrow ) . The pelvic fin showed similar expression of edar in forming lepidotrichial rays ( Figure 7E ) . shh and bmp2b expression was observed in the forming lepidotrichia of both the pectoral and pelvic fins of wild type juveniles , having a similar distal bias in the leading margin of all rays ( Figure 7C , G and I , K , respectively ) . Expression of all three genes in fls was decreased in the anterior portion of the pectoral fins ( Figure . 7B , D , J ) . However , residual expression of all three genes was found in the posterior margin of the fin ( Figure 7B , D , J arrow ) . In the pelvic fins , similar loss of anterior expression of edar ( Figure 7F ) and shh ( Figure 7H ) was seen in the fls mutant . We did not detect any difference in bmp2b expression in the pelvic fins even though obvious morphological differences in the developing rays of the samples could be seen ( Figure 7L ) . We asked if the alteration of polarity of gene expression in the fls mutant was associated with regional cell death . Using acridine orange uptake as an assay for cell death ( e . g . [32] ) , we analyzed fins of fls and siblings at size matched stages ( 7–9 mm ) for regional patterns of cell death . In 7 mm juveniles , we detect a differential retention of acridine orange between fls and siblings in the anterior , and anterior-distal margin of the developing pectoral fin ( Figure 7M–N ) . Similarly , in the pelvic fin of 8–9 mm juveniles , retention of the label was seen in the anterior distal margin of the fin ( Figure 7O–P ) . Consistent with these data suggesting asymmetrical loss of the fin fold epidermis , we find that msxa , a marker of the distal epithelium , is differentially expressed in the mutant ( Figure 7Q and R ) . We next analyzed gene expression during late development of the lepidotrichia . The expression of edar during late fin development was observed in forming lepidotrichia of all fins with a distal bias in its expression ( Figure 8C ) . The expression in the forming dermal ray was similar in both location and timing to that of bmp2b and shh ( Figure 8A–B , asterisk ) . In addition , edar was found expressed proximally between forming rays and at the distal margin ( Figure 8C arrow ) . We were unable to resolve a clear signal for eda in the forming fins using WMISH . The expression of edar in the distal lepidotrichial tips suggests a late developmental role of Eda signaling in regulating formation of the lepidotrichia in concert with shh and bmp2b . Histological analysis of fls mutant fins at an early stage of lepidotrichial formation reveal a general deficiency of the development of the entire mesodermal component of the fin such as cartilage and muscle in both the paired ( pectoral fin , Figure 8D , E ) and unpaired fins ( anal and caudal fins , Figure 8F–G; H–I , respectively ) . In contrast , the epidermis of the fin is formed and is similar to that of size matched siblings . However , close inspection of the distal tip of the fins showed disorganization of the epidermis and degeneration of distal epidermal nuclei ( insets Figure 8F , J ) . From these analyses , we hypothesize that loss of edar-mediated signaling leads to a defect in mesenchymal cell proliferation , muscle cell migration and defective lepidotrichial growth in the fin that correlates with degenerative defects seen in the distal epidermal fin fold .
We show that Eda signaling is necessary for the development and patterning of the dermal bones of the skull , scales , fin rays as well as teeth of the adult zebrafish . The correlated effect in these zebrafish structures is due to a developmental role of Eda signaling in organizing epithelial cells into signaling centers . In the case of scale development , Eda signaling is necessary for the basal epidermal cells to form a functional placode . Epidermal placodes are involved in the formation of integumentary appendages of other vertebrates such as hair , glands , feathers and teeth . These structures have been shown to act as signaling centers to orchestrate appendage development . We speculate that a primary function of Eda signaling in scale development is to promote cell-cell adhesion within the placode and that the coordinated signaling of the placode induces fibroblast assembly in the underlying dermis , an early step in scale formation . Schmidt-Ullrich et al . documents the formation of the hair placode and outline a stage series of placode formation in the mouse [35] . They report that the downless mouse mutant , which has a mutation that disrupts the mouse Edar gene [36] , causes arrest of placode formation at a pre-placode stage of development ( P0–P1 ) . This stage closely resembles the stage of scale placode formation that is affected in fls shown here . In agreement with our findings , Schmidt-Ullrich et al . further note a reduction of cell-to-cell adhesion within the placode and find increased apoptosis in the absence of Edar function . This suggests that there is a conserved developmental role of Edar between dermal scales of fish and mammalian hairs . During normal hair development , the hair placode invaginates to form the hair bulb . By contrast , the post-placodal events of scale formation in fish do not involve morphogenetic changes of the epidermis , rather the accumulation of mesenchymal cells subjacent to the epidermal placodal cells to form the scale pocket . Thus , Eda signaling in mammals and teleosts is conserved in the early phases of placode formation in controlling the functional continuity and signaling of the epidermal placode to orchestrate appendage formation . However , the downstream interpretation of the epithelial-mesenchymal signaling differs beyond this point leading to altered morphogenetic responses and histological differentiation to form diverse appendages such as scales and hair . In the fin , Eda signaling directs late stages of fin development such as the formation and growth of the dermal rays . The effect of loss of edar function on fin development uncovers an intrinsic developmental polarity of the late developing fish fin . This is seen both in the development of the proximal endochondrial bones as well as in the formation of the fin rays . We find that the change in patterning in the mutants is correlated with asymmetrical cell death of the distal marginal fin fold as well as a reduction of shh expression . This finding is similar to the effect of loss of AER function resulting in anterior-distal cell death and reduction of Shh activity in tetrapod limbs [37]–[39] . While there has not been any previous indication of a role of Eda signaling in tetrapod limb development , both the expression of Edar and related receptor , Troy , have been detected in the AER of mice [40] , [41] . A second developmental role of Eda signaling in the developing fin is observed in the outgrowth and patterning of the individual lepidotrichial rays evidenced by expression of edar in the distal tip of the forming rays and distal epidermis . The expression of edar is again associated with that of shh and bmp2b . The expression of shh and bmp2b has been shown to be within the basal epidermis overlying the forming lepidotrichia [31] . Given the expression of edar during fin development and the defects observed in the distal epidermis in the mutant , it is likely that the function of Eda signaling is to maintain the growth permissive function of the fin fold through its regulation of a distal signaling center of individual rays . The concomitant expression of edar , shh and bmp2b in both distal lepidotrichia development and during placode specification suggest that they work in concert to mediate the inductive and/or permissive effects of the epidermis – thus organizing signaling centers for the development of the dermal skeleton . While the nature of the defect in tooth formation or dermal bone patterning of the skull in the fls and Nkt mutants has not been characterized in detail , there is evidence that inductive signaling from the pharyngeal epithelium or cranial epidermis is necessary for appropriate development of both tooth [42] and skull [6] , respectively; Eda signaling likely shares a common role in inductive signaling in each of these diverse organs . Mutations in the EDAR and EDA genes underlie a large percentage of autosomal and X-linked HED in humans , respectively [18] , [43] . In the case of EDAR , both recessive and dominant mutations are associated with the HED phenotype in humans , however dominant mutations are found only within the death domain of the protein . These mutations are believed to act in a dominant negative fashion , although by unknown mechanisms [44] . We see similar dominance of a fls allele that affects the death domain of edar while all fls mutations outside this region do not show a dominant phenotype . Autosomal dominant HED in humans caused by mutation of EDAR within the death domain displays a large degree of phenotypic variability [12] . For example , the I418T mutation in human , which affects the same amino acid as flsdt3Tpl ( I327F ) , shows distinct phenotypic variability depending on genetic background [18] . Interestingly , the flsdt3Tpl zebrafish mutant displays similar dominance and variation as the human allele affecting the same residue . These findings suggest that the molecular mechanisms of Edar function are similar between fish and humans . X-linked HED caused by mutations in the human EDA gene represents the majority of cases of this disease [43] , [45] . The zebrafish Nkt mutation described here is affected in the TNF domain and shows a mild dominant phenotype ( S243X ) . As the EDA gene is sex linked in humans the molecular nature of different alleles can not be analyzed since the allele will be hemizygous in males and mosaic in female carriers . The zebrafish eda gene is autosomal in the zebrafish . Thus , Nkt exposes previously unknown dominant function of mutations in this gene since a true heterozygous condition is formed . Hemizygous wildtype condition in humans indicates that the dominance we see in Eda is probably not due to haploinsufficiency . Since EDA functions as a homotrimeric protein [46] , a plausible mechanistic explanation for the observed dominance of Nkt is that the C-terminal truncation inhibits the function of the wild type protein in binding to Edar . Mutations affecting Eda signaling lead to impaired development of integumentary appendages of fish , birds , and man . These changes lead to viable changes in adult morphology . Mutations disrupting Eda signaling have been described for another teleost species . The spontaneous rs-3 mutant in medaka ( Oryzias latipes ) , is shown to have a transposon insertion in the 5′ UTR of edar resulting in the reduction of scales but no effect on fin or teeth development [47] . The zebrafish mutations described here show a previously undescribed role of Eda signaling in the development of the fins , teeth , as well as dermal bones of the skull – phenotypic traits observed in human alleles but not reported in the medaka mutant . As the phenotype of the rs-3 mutant is similar to the flsdfang allele in the zebrafish , it is likely that the more subtle phenotypes observed in the medaka mutant is due to partial loss of function of edar caused by a hypomorphic rs-3 allele [47] . The graded effects seen in the expressivity of mutations affecting eda and edar points to a dose sensitive readout of the Eda signaling pathway that affects different organ systems with varied expressivity . In the dominant flsdt3Tpl or Nkt heterozygotes , the shape and number of scales and teeth as well as patterning of the skull are affected , however there is no change in fin ray development . Similarly , the fang allele of fls clearly demonstrates this dose sensitivity as functional copies of Edar are titrated by the concomitant use of a new splice site in the mutant leading to the reduction in the amount of wild type transcript made ( Figure 3D ) . This reduction in the amount of edar transcripts cause defects in scale and tooth development , however fins are normal . fang/te370f , in which the fang allele is in trans to a presumed null , further reduces the relative levels of wild type edar transcripts leading to further reduction of both teeth as well as fin lepidotrichia . Similar dose sensitive responses to levels of EDA signaling are seen in tooth development of the mouse regulating the number and shape of teeth [48] , [49] . There are several reports of hypodontia in humans resulting from altered EDA function that do not show other phenotypes such as hypothrichosis or nail defects [50]–[52] . Given our findings , it is likely that these particular alleles are hypomorphic and this is sufficient to explain the differential organ sensitivity to levels of EDA signaling during development . These data indicate that control of the level of Eda signaling in post-embryonic development is an essential component for the determination of the number and form of many different organ systems of the adult . Supporting this finding , we observed significant modification of expressivity of flsdt3Tpl in different genetic backgrounds indicating the existence of genetic modifiers of Eda signaling . This sensitivity of Eda signaling to genetic modifiers occurs in other teleost fish as well . In our analysis of the medaka rs-3/edar mutant , we find a high degree of variability in the extent of scale formation ( Figure S4 ) suggesting the existence of background modifiers of Edar function in this species . Additionally , evidence from the stickleback , Gasterosteus aculeatus , suggests that genetic variance at the eda locus underlies differences in the extent of dermal plate formation in diverged populations of this species [33] . A quantitative trait analysis ( QTL ) of lateral plate formation in a low-plated form of the stickleback indicates a significant modification of the reduced plate phenotype ( eda locus ) with modifying effects within and between loci affecting plate number and size [53] , [54] . Interestingly , recent evidence also shows a significant association between the edar locus and dermal plate number in sticklebacks in addition to the predominant eda locus [55] . Thus variation at these gene loci may act in concert to regulate number of dermal plates/scales . Thus , while loss of Eda signaling can lead to severe phenotypes , the phenotypic consequences of variation in Eda signaling are graded and canalization of Eda signaling is prevalent . Therefore , buffering of the phenotypic outcome that results from defective Eda signaling could be a common mechanism that permits viable and diverse phenotypes . These viable phenotypic variations then could serve as a basis for selection . The lack of a coding change at the eda locus in sticklebacks that is associated with the loss of dermal plates has lead to the argument that , in this case , evolution of this trait is due to changes at cis-regulatory elements controlling eda expression [33] . Our findings on the dose and organ specific sensitivity of Eda signaling in different structures of the zebrafish argues that evolution of this trait could result from a regulation of absolute levels of expression . Interestingly , recent analysis of single nucleotide polymorphism ( SNP ) frequency in human populations supports the role of Eda signaling in causing phenotypic variation . Analysis of SNP variation between diverse human populations shows evidence of selection of the EDAR locus in East Asian and American populations [56] , [57] . A defined allelic variant of EDAR within these populations leads to a coding change in the death domain of EDAR and is a candidate allele for altered gene function that could have lead to the region being fixed in these populations [34] . There is evidence from association data that this allele is associated with thick hair in these populations [58] , however the full extent of phenotypes that are affected in these populations that are related to EDA signaling has not been analyzed . It is interesting to note that recent work has identified this allele of EDAR as having an enhanced effect on Eda signaling in mouse models containing the altered human residue [59] . Given that variation in the number and shape of integumentary derivatives of the dermal skeleton are a common morphological change in teleost evolution e . g . [60] , it will be important to further investigate the prevalence and type of genetic changes in Eda signaling genes in cases of natural variation of these adult characters . Zebrafish mutagenesis was performed following [61] with 5 treatments of 3 . 3–3 . 5 mM ethylnitrosourea . Screen design was similar to that described [62] . Allele designation was determined using standard nomenclature with the addition of the molecular lesion or phenotypic description ( when appropriate ) to the designation . The serial numbers of the mutants found in the ZF models screen are as follows: flst3R367W ( #0621 ) ; flsdt3Tpl ( #1248 ) ; Nktdt3S243X ( #1261 ) . Information on the screen can be found at http://www . zf-models . org/ . The screen for additional fls alleles used mutagenized TLF founder males treated similarly as Tü males used in the screen . Rough mapping of F2 progeny against a reference panel of SSLP markers [63] indicated that fls was located on linkage group ( chromosome ) 9 ( LG9 ) with loose linkage to z20031 ( 61 . 3cR; Figure 3A ) . We found fls to be closely linked to markers z7001 and z11672 . Results from radiation hybrid screening indicated linkage of zebrafish edar to markers positioned on LG9 in the region predicted by initial mapping analysis . Analysis of flanking markers and internal polymorphisms in edar showed tight linkage of the fls mutation to the edar gene . Using the defined molecular differences between WIK and Tü strains , we did not find recombination in 238 meioses indicating that the mutation was located less than 0 . 4 cM away . We isolated the full-length cDNA of zebrafish edar and eda by reverse transcription ( RT ) PCR using sequences provided from genomic alignments and subsequent amplification of the cDNA ends by rapid amplification of DNA ends ( RACE ) . cDNA was generated from RNA from blastemas of amputated caudal fins that had been allowed to regenerate for two days . cDNA sequences of zebrafish edar and eda genbank accession numbers are EF137867 and EF137866 , respectively . Protein alignment of Edar and Eda were generated by ClustalW alignment ( http://www . ebi . ac . uk/clustalw/ ) and Box Shade software ( http://www . ch . embnet . org/software/BOX_form . html ) using a 0 . 4 identity threshold . Edar and eda sequences of other species were obtained from genomic databases at NCBI ( http://www . ncbi . nlm . nih . gov/ ) , Sanger ( http://www . ensembl . org/index . html ) , and Tigr ( http://www . tigr . org/tdb/tgi/ ) . Real Time PCR was performed on cDNA obtained from blastemas from two day old caudal fin regenerates . Calculations were made from three biological replicates and three technical replicates according to [64] . Primers were designed for wild type specific transcripts by using sequence from the neighbouring exon borders that are not adjacent to each other in Fang-transcripts . Fang specific primers were designed against the fang specific transcript sequence , which is spliced out in wild type . Crossing points of the control reaction , which were higher than in the water control , were set to the value for water . Normalization was done against the efficiency of primers to β-actin . Adult bones were stained with alizarin red . Embryos were fixed in formalin ( 3 . 7% formaldehyde ) , briefly dehydrated in 70% ethanol , and placed in 1 g/l alizarin red; 0 . 5% KOH until bones suitably stained . Fish were destained in 1% KOH until background stain was lost and subsequently cleared in glycerol for analysis . For analysis of forming cartilage , fish were prestained in alcian blue from 4–24 hours . The fish were then destained , lightly trypsinized ( 3 g/l; 37°C ) and processed for alzarin red staining . Skeletal measurements were made using digitizing software from Zeiss using a dissecting microscope . Measurements were made from fixed landmarks on each axis of the skull that did not vary depending on position of the suspensorium: the premaxilla was used for the distal most point on the length ( L ) axis , while the quadrate-anguloarticular joint was used as a ventral landmark for the height ( H ) axis . Raw measurement values are represented as normalized ratios of the distance along each axis in relation to the position of the center of the eye; values are normalized for standard length of the fish . Probes for whole mount in situ hybridisation were generated by reverse transcription from cDNA made from regenerating caudal fin tissue . Digoxigenin labeled RNA probes were purified using P-30 micro bio-spin columns before use ( BioRad ) . WMISH protocol was performed as described [65] , at 70°C and with the addition of 0 . 1% CHAPS to hybridization and post hybridization wash buffers . Reactions were stopped in PBS , post-fixed and placed in methanol overnight to reduce non-specific staining . Acridine orange ( Sigma ) was used as a marker of apoptosis in developing tissue [32] . Juveniles were immersed in fish water containing 5 μg/ml acridine orange for 5 minutes and then washed with fish water , anesthetized and post-fixed in formalin to assist visualization of staining . Analysis of cranial measurements were performed using Hotelling's T squared test for two dependent variables . For scale counts and size dimensions , a t-statistic for differential means was used to assess significance . Calculations and probability assessment were calculated using Biosoft 200 software ( www . biosoft . com ) and Excel statistical package . Specimens of 8–9 mm juvenile fish were fixed with a mixture of 4% formaldehyde in PBS and 1–2 . 5% glutaraldehyde at room temperature and subsequently placed at 4°C . After post-fixation with 1% osmium tetroxide in 100 mM PBS for 1 h on ice , samples were washed with H2O , treated with 1% aqueous uranyl acetate for 1 h at 4°C , dehydrated through a graded series of ethanol and embedded in Epon . Ultrathin sections were stained with uranyl acetate and lead citrate and viewed in a Philips CM10 electron microscope . In addition , toluidine blue stained Epon sections of 0 . 5 or 3 μm thickness were prepared for light microscopy . | A major goal of the study of developmental genetics is to understand the genes and developmental mechanisms underlying the formation of organismal complexity and diversity . Here , we focus on genes controlling postembryonic development and describe mutations in genes of the ectodysplasin ( Eda ) pathway in regulating the formation of the scales , skull , fins , and teeth . Mutations in genes of this signaling pathway are common in humans with defects in ectodermal structures such as hair , glands , and teeth . We show that the similar phenotypes of loss of Eda signaling in fish and human are due to a conserved early developmental stage in the development of mammalian hair and fish scales; subsequent development of these two structures diverge . Our findings show that the Eda signaling pathway has an ancestral role in regulating the developmental interactions involved in patterning and growth of the dermal skeleton of fish . Recent work has shown that these genes are associated with morphological variation between humans and evolution within fish populations , suggesting that alteration in the function of these genes permits viable morphological change . Our data support the value of forward genetic studies on postembryonic development to reveal the genetic and developmental basis of both human disease and morphological evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
]
| [
"genetics",
"and",
"genomics/animal",
"genetics",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics",
"dermatology/hair",
"and",
"nail",
"diseases",
"developmental",
"biology/developmental",
"evolution",
"genetics",
"and",
"genomics/disease",
"models",
"developmental",
"biology/organogenesis",
"developmental",
"biology/developmental",
"molecular",
"mechanisms",
"evolutionary",
"biology/developmental",
"evolution"
]
| 2008 | Zebrafish eda and edar Mutants Reveal Conserved and Ancestral Roles of Ectodysplasin Signaling in Vertebrates |
The eukaryotic XPD helicase is an essential subunit of TFIIH involved in both transcription and nucleotide excision repair ( NER ) . Mutations in human XPD are associated with several inherited diseases such as xeroderma pigmentosum , Cockayne syndrome , and trichothiodystrophy . We performed a comparative analysis of XPD from Homo sapiens and Chaetomium thermophilum ( a closely related thermostable fungal orthologue ) to decipher the different molecular prerequisites necessary for either transcription or DNA repair . In vitro and in vivo assays demonstrate that mutations in the 4Fe4S cluster domain of XPD abrogate the NER function of TFIIH and do not affect its transcriptional activity . We show that the p44-dependent activation of XPD is promoted by the stimulation of its ATPase activity . Furthermore , we clearly demonstrate that XPD requires DNA binding , ATPase , and helicase activity to function in NER . In contrast , these enzymatic properties are dispensable for transcription initiation . XPD helicase is thus exclusively devoted to NER and merely acts as a structural scaffold to maintain TFIIH integrity during transcription .
The human XPD protein is a helicase with 5′–3′ polarity [1] , [2] that contains a 4Fe4S cluster ( FeS ) [3]–[6] and is organized within the context of the general transcription factor IIH ( TFIIH ) . TFIIH consists of a total of 10 subunits of which XPB , p62 , p52 , p44 , p34 , and p8 build the core and cdk7 , MAT1 , and cyclin H comprise an additional regulatory unit called the CAK complex; both subcomplexes are bridged by XPD [7]–[10] . The first assigned function of TFIIH was its role as a basal transcription factor promoting RNA polymerase II ( RNAPII ) –based transcription [10]–[13] . Within this process , TFIIH participates in initiation , promoter escape , and the early elongation steps [9] , [14]–[18] . Later it has been shown that TFIIH is also involved in RNA polymerase I– and probably III–based transcription [8] , [19] . While the TFIIH core assumes responsibility for promoter opening , the CAK subcomplex promotes initiation and elongation through phosphorylation of the C-terminal domain ( CTD ) of the Rbp1 subunit of RNAPII . In addition , it has been shown that CAK is also responsible for the transactivation of nuclear receptors . Next to its function in transcription , TFIIH plays a major role in eukaryotic nucleotide excision repair ( NER ) [10] , [20]–[22] . NER is the most versatile DNA repair pathway due to its ability to remove bulky lesions , as well as smaller lesions such as cisplatin adducts or photoproducts caused by ultraviolet light [23]–[26] . In NER , TFIIH is recruited to the site of damage by the XPC–HR23B complex after initial damage recognition has been achieved [23] , [27]–[29] . At this point , TFIIH is of vital importance for damage verification and the procession of the NER cascade resulting in the excision of the damaged DNA fragment . The significance of TFIIH within the repair cascade is reflected in its association with three severe hereditary human diseases: xeroderma pigmentosum ( XP ) , Cockayne syndrome ( CS ) , and trichothiodystrophy ( TTD ) . It is interesting to note that TTD patients seem to be also affected partly in aspects of transcription [26] , [30] , [31] . In the context of human diseases , it is particularly noteworthy that mutations causing XP , CS , or TTD can be located on a single gene , namely XPD [1] , [32] , that also harbors most of the described mutations that are associated with TFIIH . It has been enigmatic for a long time how these very different phenotypes can be associated with a single gene . Recent crystal structures of archaeal XPD homologues shed some light on this puzzling matter . The structures revealed that XPD contains four domains ( Figure 1 ) , of which two are canonical RecA-like domains ( helicase domains 1 and 2 , HD1 and HD2 ) , and two additional domains that are inserted in HD1 , a so-called Arch domain and a domain harboring the FeS cluster [3] , [5] , [6] , [23] . XPD as well as XPB belong to superfamily 2 ( SF2 ) helicases but operate with opposite polarities [7] , [9] , [10] , [33] . In addition , the roles of XPD and XPB in transcription and repair seem to differ significantly [10] , [34] . The helicase activity of XPB is required for promoter opening and promoter escape after the preinitiation complex ( PIC ) has been assembled [9] , [16] , [17] , [26] , [35] . In NER only the ATPase activity of XPB is necessary to anchor TFIIH to the lesion site [19] , [26] , [32] , whereas its helicase activity seems to be dispensable [10] , [19] , [25] . In contrast , XPD is a vital factor for NER , including its helicase activity [23] , [25] , [26] , [36]; XPD is interacting ( Figure 1 ) with the p44 subunit within the TFIIH core and with the MAT1 subunit of the CAK complex [23] , [28] , [37] . Whereas p44 binds to the very C terminus of XPD [4] , [6] , [26] , [37] , [38] , MAT1 is recruited via the Arch domain [23] , [32] . It has been shown that p44 is able to stimulate the helicase activity of XPD [5] , [6] , [23] , [37] and that the interaction with CAK negatively regulates the helicase activity of XPD [33] , [37] . For transcription , however , it is less clear which of the XPD activities are required . Early work in yeast has shown that a temperature-sensitive mutant of Rad3 , the XPD ortholog in yeast , is affected in transcription [34] , [39] . It has also been shown that TTD-inducing XPD mutations that abolish the interaction with p44 have an impact on the transcriptional properties of TFIIH [9] , [22] , [26] . Furthermore , mutations in the Arch and FeS domains cause a TTD phenotype and show decreased transcriptional activity [26] , [32] . In contrast , the Walker A motif variant K48R , which displays no ATPase and helicase activities , supports transcription [19] , [25] , [26] . In this study , we have used a combinatorial approach of characterizing two eukaryotic XPD proteins to delineate which of its properties and domains are dedicated to transcription and NER . We show that the FeS domain of XPD is crucial for NER . In addition , our results clearly demonstrate that the repair process depends on XPD's biochemical and enzymatic properties for successful NER , thus requiring DNA binding , ATPase activity , helicase activity , as well as the correct interaction with other TFIIH subunits . In stark contrast , all enzymatic properties including DNA binding are not essential for the different steps in transcription; here , XPD only acts as a scaffold within TFIIH with the sole requirement to support protein–protein interactions .
Proteins from the fungus Chaetomium thermophilum provide an ideal basis to perform biochemical studies , as they are closely related to their human homologues but are significantly more stable and thus amenable to purification [36] , [40] . We employed a comparative functional mutagenesis study of C . thermophilum XPD ( ctXPD ) and human XPD ( hsXPD ) . The sequence alignment ( Figure S1a ) shows that ctXPD and hsXPD are closely related with a relative homology of 74% and an identity of 55% ( as defined by Blast ) . Based on the sequence alignment , homology modeling , and previous studies on Thermoplasma acidophilum XPD ( taXPD ) [37] , [41]–[43] , we chose the following variants to be generated in hsXPD and ctXPD: hsC134S/ctC133S , hsY158A/ctY156A , hsF193A/ctF192A , hsR196A/ctR195A , hsR196E/ctR195E , and hsR722W/ctK719W , of which five residues are identical and the remaining is type-conserved . The variants include a TTD mutation ( hsR722W/ctK719W ) and point mutations that have been shown to be of functional significance in archaeal homologs of XPD ( Figure S1a , b and Table 1 ) . In addition , we generated the ctK48R variant , which disrupts the Walker A motif and thus ATP hydrolysis of XPD in C . thermophilum , to serve as a negative control . We focused on variants located in the N-terminal domain of XPD—that is , the FeS domain—as it has been implicated as being highly important for archaeal XPD function [4] , [6] , [28] , [37] , [38] . To further analyze hsXPD , we introduced three additional variants . HsL372A was generated to test whether a mutation in the Arch domain , which is most likely not involved in DNA binding , ATPase activity , and helicase activity , would affect the function of hsXPD . In addition to hsC134S/ctC133S , the variant hsC155S was designed to further probe the FeS cluster of hsXPD ( both Cys134 and Cys155 coordinate the FeS ) , and hsF161A as an alternative residue for the corresponding ctY156A variant ( see Figure S1a ) . None of the ctXPD variants affected its overall fold , as they could be expressed and purified to at least 95% homogeneity , with the exception of the ctC133S variant ( Figure S2a ) and the analysis by CD spectroscopy showed that the wild-type protein and all variants display similar CD spectra ( Figure S2b ) . To fully exploit our ctXPD model system , we also cloned and expressed C . thermophilum p44 ( ctp44 ) comprising residues 1–285 . This construct is analogous to an hsp44 construct that is fully capable of activating hsXPD [8] , [23] but lacks the C-terminal Ring and Zn-finger domains of p44 . Size exclusion chromatography ( SEC ) experiments revealed that equimolar ratios of ctXPD and ctp44 form a stable complex exemplified by a significant shift in the elution volume of a single peak representing the ctXPD–ctp44 complex , which can be clearly distinguished from the peaks of the single proteins ( Figure 2a ) . In contrast , the ctR719W variant , which corresponds to the human R722W variant that abrogates the p44 interaction with hsXPD , displays no shift , thus clearly indicating the impairment of complex formation ( Figure 2b ) . We then investigated all ctXPD variants for complex formation with ctp44 and observed ctXPD/ctp44 wild-type–like complex formation for all other variants , thus indicating that none of the other mutations interfered with p44 interaction ( Figure S2c ) . We assessed the ssDNA binding properties of ctXPD and its variants ( Figure 2c and Table 2 ) . The wild-type protein displayed a dissociation constant ( KD ) of 118 nM , which was not altered in the presence of p44 ( 95 nM ) ; hence , we omitted p44 from the analysis of the variants . The affinity of ctXPD to ssDNA is very comparable to the affinity of the archaeal homologues [5] , [6] , [37] , [38] , [44] . The ctK48R and ctK719W variants display no major alteration in affinity for ssDNA with KD values of 159 nM and 192 nM , respectively . Because the ctK48R variant contains a mutation in the helicase Walker A motif that is solely responsible for ATP hydrolysis , the results are in line with the role of this motif . The ssDNA affinity of ctK719W clearly shows that DNA binding is not affected , implying that the deficiency to interact with p44 impairs XPDs helicase activity by other means; this is in line with the fact that p44 does not influence the affinity of ctXPD for ssDNA . All other ctXPD variants are significantly impaired with respect to their ability to bind to ssDNA , with KD values ranging from 561 nM to 1 , 035 nM ( Table 2 ) , which is in agreement with their proposed proximity to the DNA path along the XPD molecule [37] , [38] . Due to the possibility to express functional fungal XPD and p44 separately , we first analyzed whether the ctXPD ATPase activity is dependent on p44 activation . Surprisingly , we observed a clear p44-related effect . The activity increased from 0 . 12 mol ATP·mol XPD−1·s−1 to 0 . 7 mol ATP·mol XPD−1·s−1 after adding ctp44 to ctXPD in a 2∶1 molecular ratio ( Figure 2d and Table 2 ) . Because the ssDNA affinity of XPD itself is not altered by p44 binding , p44 must directly activate the ATPase function of XPD . As expected , the ctK48R Walker A variant is ATPase deficient . The decreased affinity for ssDNA in the ctY156A variant leads to a reduced activation of the ATPase activity , with 61% of wild-type activity . Surprisingly , the ctF192A variant that bound to ssDNA with a 6-fold increase in KD ( Figure 2c ) only displayed 27% of wild-type ATPase activity . Inferred from the structures of archaeal XPDs [37] , this residue is not involved in ATP binding; thus , F192 may participate in an ATP hydrolysis-mediated helicase step on the translocated ssDNA using base stacking interactions [4]–[6] , [39] . The impaired translocation on ssDNA could disturb the concerted action of HD1 and HD2 during ssDNA-dependent ATP hydrolysis and provides a likely explanation for its strongly reduced ATPase activity . The two variants R195E/A displayed an approximately 10-fold decrease in affinity for ssDNA . Whereas the ATPase activity of R195A was only moderately affected with a decrease to 54% compared to wild-type ctXPD , ctR195E is even further reduced and displayed only 20% residual activity . The p44 interaction-deficient ctK719W variant displayed a highly decreased ATPase activity reflecting the basal ctXPD ATPase level in the absence of p44 , thus further supporting the notion that p44 directly stimulates XPD's ATPase activity . All other variants displayed a significant decrease in p44-dependent ATPase activity , which can be explained by their reduced DNA binding ability , as ssDNA is necessary to trigger XPD's ATPase activity . To investigate the influence of the XPD mutations on the helicase activity of XPD , we performed in vitro helicase assays with ctXPD and hsXPD in the presence and in the absence of p44 ( see Materials and Methods , Figure 3a , b , and Table 2 ) . In the absence of ctp44 , no significant unwinding by ctXPD could be detected . In the presence of ctp44 , wild-type ctXPD was readily unwinding the 5′ overhang substrate and yielded an activity of 1 , 906 . 3 ΔFl . ·s−1 , indicating a robust 5′–3′ polarity ( Figure 3 ) . The walker A motif mutant ctK48R , which is unable to hydrolyze ATP , also failed to separate dsDNA ( 7 . 8 ΔFl . ·s−1 ) . The ctK719W variant was also highly affected in its helicase activity ( 4 . 8 ΔFl . ·s−1 ) , due to its loss of p44 interaction , resulting in a highly decreased ATPase activity . The ctF192A and ctR195A/E variants that were impaired in DNA binding and ATPase activity were highly deficient with respect to their p44-dependent helicase activity , with values of 8 . 6 ΔFl . ·s−1 , 24 . 1 ΔFl . ·s−1 , 9 . 3 ΔFl . ·s−1 , and 11 . 2 ΔFl . ·s−1 , respectively . Because the loss of helicase activity cannot be explained by the lack of p44 interactions , these data substantiate the role of these residues in DNA binding , which subsequently affects the ATPase and helicase activities of XPD . The only variant displaying notable p44-dependent helicase activity was ctY156A , with an approximately 5-fold reduction in activity , thus still being significantly impaired . We analyzed the helicase activity of the hsXPD variants with an in vitro assay that detects the displacement of a 32P-labeled DNA fragment previously annealed to a single-stranded circular DNA [22] , [26] . Upon interaction with p44 , hsXPD exhibited a high and significant 5′–3′ helicase activity ( Figure 3b , lanes 2–3 and the histogram below ) , whereas the hsR722W variant was not stimulated by p44 ( lane 12 ) due to the loss of p44 interaction ( Figure S3c ) [23] , [26] , as also observed for the corresponding proteins from C . thermophilum ( Figure 2b ) . All other variants displayed a strongly diminished helicase activity . Notably , hsF161A , which was chosen as a possible alternative for ctY156A , showed highly impaired helicase activity , indicating its functional importance . Curiously , hsC134S , the counterpart to ctC133S , seems to behave more stably than its fungal homologue , which exhibited an altered expression profile ( Figure S2a ) . However , even though the expression behavior of hsC134S was comparable to wild-type hsXPD , it displayed no helicase activity , which is most likely caused by affecting the overall stability of the FeS domain of hsXPD . A similar explanation can be assumed for the other variant targeting the FeS cluster , hsC155S , which also abolished helicase activity . The only exception to the loss-of-function variants was hsL372A , which displayed wild-type–like helicase activity , confirming that this residue is not directly involved in helicase activity . Importantly , the helicase phenotypes observed for the hsXPD variants are highly comparable to the phenotypes of the ctXPD variants . Due to the strong correlation of the helicase phenotypes in the human and the fungal system as well as the interaction studies with p44 and the CAK subunits ( see below ) , it is valid to assume that the defects causing the deficiency in helicase activity by the lack of ssDNA binding , ATPase activity , or p44 interaction can be directly translated from ctXPD to hsXPD . To assess the ability to perform NER in vitro , a dual incision assay was used in which hsXPD or its variants were added to purified recombinant human core-TFIIH ( rIIH6 ) ( Figure 4a ) , together with XPC-HR23B , XPA , RPA , XPG , ERCC1-XPF , and a closed circular plasmid with a single 1 , 3-intrastrand d ( GpTpG ) cisplatin-DNA crosslink [25] , [26] , [34] . In the absence of hsXPD , the XPG and ERCC1-XPF endonucleases were unable to excise the damaged DNA and liberate the damaged oligonucleotide with a length of 23–25 nt ( Figure 4b , c , compare lanes 1 to lanes 2 and 3 ) . All investigated hsXPD variants , with the exception of the L372A variant , lacked the ability to catalyze successful NER within the reconstituted rIIH complexes ( Figure 4b , c ) . The activity of the variants is comparable to that of the hsR722W mutation ( Figure 4c , lanes 12 and 13 and histogram ) . HsL372A behaves indistinguishably from wild-type XPD , further supporting that this residue is not relevant for NER activity . A host cell reactivation assay was used to investigate the in vivo repair capabilities of selected XPD variants . In this assay , the capacity of XPD variants to repair and express a damaged gene in a cellular context utilizing HD2 cells is measured . HD2 cells are NER-deficient , resulting from the fusion between human fibroblasts harboring the XPD/R683W point mutation and HeLa cells [32] , [40] . A plasmid encoding the firefly luciferase gene was exposed to UVC radiation and then transfected in combination with a second vector expressing Renilla luciferase to normalize transfection efficiencies . A third vector expressing the XPD variants was added to the cells ( Figure 4d , e ) . The firefly luciferase activity in cell lysates was measured 48 h posttransfection and normalized using wild-type activity as a reference . Transfection of hsC134S , which exhibited almost no in vitro double incision activity , did not permit the expression of the luciferase gene after UV irradiation and exhibited an activity comparable to that of the control transfected with an empty vector ( Figure 4d , lanes 2 and 3 ) . However , hsR196A/E , which also showed reduced in vitro double incision activity , partially permitted the expression of luciferase . The nonirradiated control experiments showed a comparably high expression activity in all cases ( Figure 4e ) . These data demonstrate that the XPD variants are defective in DNA repair both in vitro and in vivo . The expression of the luciferase in the control experiment without UV irradiation was not affected , indicating regular transcription and expression promoted by the XPD variants . To exclude the possibility that the negative effects in repair are due to a disturbed CAK interaction , we performed interaction studies of wild-type ctXPD and its variants with ctMAT1 ( residues 1–248 ) employing native PAGE experiments . Figure S3a shows that all variants form a complex with ctMAT1 in a wild-type–like manner , indicating no impairment in CAK interaction ( Figure S3a ) . In a parallel approach , we probed the interaction of p44 and CAK with hsXPD , hsC134S , hsR722W , and hsK48R using pull-down experiments ( Figure S3b , c ) . Our data show that hsC134S and hsK48R still interact with p44 and the CAK complex . This is not the case when looking at hsR722W , which shows an impairment in p44 interaction but not in CAK interaction . Our data thus exclude the possibility that an interruption between XPD and the CAK subcomplex leads to the altered NER phenotype of the investigated XPD variants and the pull-down assays with p44 further signify the similarity between hsXPD and ctXPD We next investigated the transcriptional properties of the rIIH6 complexes in an in vitro reconstitution assay in which XPD variants were added in increasing amounts to purified rIIH6 core and CAK in addition to the recombinant basal transcription factors TBP , TFIIA , TFIIB , TFIIE , TFIIF , as well as purified endogenous RNAP II ( Figure 5a ) . Surprisingly , most hsXPD variants led to a transcriptional activity comparable to wild-type XPD in terms of transcript length and amount , regardless of their enzymatic impairment . The exceptions are the hsR722W ( activity of 20%±3% ) and , to a much lesser extent , the hsC134S ( activity of 65%±18% ) variants . The transcriptional defect of hsR722W ( Figure 5a , lanes 33–35 and Figure 5b ) can readily be explained due to the lack of an interaction with p44 ( Figure S3c ) . This mutant also displayed the highest impairment of all variants . HsC134S most likely constitutes a protein with a compromised FeS domain that could affect its overall stability . Titration experiments show that the effect on transcriptional activity of hsC134S is only moderate as compared to other variants or hsXPD ( Figure 5b ) . However , although the FeS domain is compromised , the protein clearly supported transcription , indicating that the FeS domain is not vital for the transcription function of XPD . In order to exclude that residual helicase activity of some XPD variants is responsible for transcriptional activity , we performed time course transcription experiments with selected hsXPD variants . The data show that wild type and K48R show a similar increase over time , hsC134S shows a slightly affected activity , whereas the hsR722W variant displays strongly reduced activity ( Figure S4 ) . Because the Walker A mutant K48R and the hsC134S variant only display background levels of helicase activity but remain active in transcription , an effect of residual helicase activity on transcription can be excluded . To substantiate our results , we further analyzed the hyperphosphorylation state of the carboxy terminal domain ( CTD ) of the largest subunit ( Rbp1 ) of RNAP II by the CAK subcomplex of TFIIH that is supposed to be anchored to the core through an interaction with XPD ( Figure 5c ) , an important requirement for the transition between initiation and promoter escape [3] , [5] , [6] , [41]–[43] . The hyperphosphorylated form of RNAP II ( IIO ) was prevalent in all variants and comparable to wild-type hsXPD , demonstrating that RNAP II was capable of elongating normally . The only exception here is the hsR722W variant that displayed nearly no hyperphosphorylation activity ( Figure 5c , lanes 10 and 11 ) . Lastly , we analyzed the effect of increasing concentrations of the CAK complex on the in vitro transcription activity ( Figure 5d ) . To this end , we selected the variants hsF193A and hsR722W . While increasing amounts of the CAK complex stimulated the activity of wild-type hsXPD and hsF193A , this stimulation was substantially reduced when no XPD was present ( Figure 5d , lanes 1 and 2 ) . A similar phenotype was observed for hsR722W , which also showed a significantly reduced transcription activity and increasing amounts of CAK led to similar results , as observed in the absence of XPD ( Figure 5d , lanes 7 and 8 ) . Because the hsR722W variant is not impaired in the CAK interaction ( Figure S3b ) [28] , [32] , the recruitment of CAK by XPD must be disturbed due to the lack of p44 interaction ( Figure S3c ) , which impairs the incorporation of XPD into the core TFIIH . This effect cannot be rescued by addition of CAK in excess .
Several studies have underlined the dual function of TFIIH in both transcription and DNA repair ( reviewed in [8] ) . However , these very distinct tasks may require different roles/functionalities for each of the 10 subunits and , in particular , for the two ATP-dependent helicases XPB and XPD . We here focus on XPD and its involvement in both processes , and demonstrate its specific and exclusive role as an enzyme in DNA repair . We chose a combinatorial approach that allowed us to dissect functions of XPD by assessing the influence of the introduced variants on the primary enzymatic properties such as DNA binding , ATPase activity , helicase activity , and , in addition , on protein–protein interactions that are important for enzymatic function . In a second step , these variants were investigated in NER and transcription , to establish a relationship between a phenotype and its underlying molecular cause in a TFIIH-dependent context . We introduced mutations that especially cover the FeS domain of XPD ( Figure 1 and Figure S1 ) , as this domain has been proposed to be of importance for XPD function [37] , [38] , [44] . In addition , we have chosen the disease-related variant R722W located at the C terminus of XPD that disrupts the interaction with p44 and lastly the Walker A motif variant K48R . Our analysis shows that ctXPD and hsXPD are highly comparable with respect to their particular enzymatic properties and that results obtained with ctXPD can be readily extended to hsXPD . Therefore and for reasons of simplicity , we will only mention the human XPD numbers of all analyzed variants from here on . All variants located in the FeS domain of XPD are impaired in helicase activity , which is caused by different defects in singular properties such as reduced DNA binding , ATPase activity , or the inability to separate dsDNA . The highly altered DNA binding ability observed in the R196E variant therefore leads subsequently to diminished ATPase and helicase activity . An interesting phenotype is , however , constituted by the Y158A variant . Although it retains 61% ATPase activity compared to wild-type XPD , the helicase activity is affected more than one would anticipate based on the ATPase data . The position of this variant in the structural model ( Figure S1b ) suggests that it could be involved in duplex separation , which is predicted to take place at the outer rim of the FeS domain [37] , [38] . The observed lack of duplex unwinding would be a direct result of the inability to separate dsDNA , thereby targeting the helicase function of XPD without a major effect on ATPase activity . Because F161A is also highly diminished in helicase activity and is located in close proximity to Y158 , it is tempting to speculate that a similar phenotype for these variants is the likely explanation . Consequently , the C134S and C155S variants directly targeting the coordination of the FeS cluster in the FeS domain are also impaired in their helicase activity , which is most likely due to unfolding of the entire FeS domain [4]–[6] . Since the structural part responsible for the separation of dsDNA is located in the FeS domain , the unfolded FeS domain cannot mediate the separation of dsDNA , resulting in an inactive helicase . In contrast , the Walker A K48R variant interacts with DNA like the wild-type protein but lacks ATPase activity and thereby causes an impaired helicase activity . The p44 interaction-deficient R722W is helicase-deficient due to an impaired interaction with p44 . So far , however , it has not been known how p44 activates XPD . We show here that p44 strongly activates the ATPase activity ( and consequently the unwinding activity ) of XPD . This effect is not mediated by an increase in affinity for ssDNA but through a direct up-regulation of the ATPase motor activity of XPD via a yet unknown mechanism . However , the lack of p44 interaction impacts not only XPD's helicase activity but also the recruitment of XPD to TFIIH , thereby intensifying its phenotype [26] . All variants that fail to separate dsDNA for the reasons explained above are also impaired in NER . Our data thus show that it is essential for XPD to maintain its capability to act as a classical helicase in NER and that it is involved in unwinding the duplex around the lesion promoting downstream processes that lead to a successful excision ( Figure 6 ) . It is thus clear that the two helicase domains ( HD1 and HD2 ) are essential , but our data demonstrate that the investigated FeS domain of XPD is also crucial for its enzymatic function and subsequently for NER . We utilized the available “molecular toolkit” of variants to further dissect the involvement of XPD in transcription initiation in a TFIIH and CAK context . To our great surprise , most of the variants ( C134S , C155S , Y158A , F161A , F193A , and R196A/E ) supported transcription in a wild-type–like manner ( Figure 5 ) . However , it remains possible that transcription could be affected in cells , perhaps at a subset of genes , because of auxiliary factors whose activity or association with TFIIH is compromised by an XPD mutation . Our data demonstrate that none of the enzymatic properties of XPD—namely , DNA binding , ATPase activity , and helicase activity—are relevant for this process . Even the R196E/A variant , which is highly impaired in DNA binding , has no impact on in vitro transcription or CTD phosphorylation ( Figures 2 and 5 ) , indicating that XPD does not interact with the DNA during transcription . This observation is further supported by recent EM structures from the PIC [45] , [46] in which XPD was modeled in positions where it is not in close proximity to DNA . However , this leads to the question of which function XPD is required to fulfill during transcription , as its presence is undoubtedly important [25] , [26] , [34] . This can be addressed with the phenotype of the hsR722W variant , which does not support transcription . We show that this variant is neither impaired in its tertiary structure nor in DNA binding; hence , the lack of ATPase and helicase activity of this variant must be explained by the loss of p44 interaction . Taken together with the fact that the enzymatic activities of XPD are obsolete in transcription , these observations suggest that the decisive phenotype for transcription in this variant is exclusively the lack of an interaction with p44 . This lack of interaction most likely results in a misplaced XPD regarding the TFIIH composition and subsequently also the displacement of the CAK complex , as XPD not only interacts with the core-TFIIH subunit p44 but also provides the bridge to the MAT1 subunit of the CAK complex via its Arch domain ( Figure 1 ) [32] . It has been shown that the interaction with CAK is not interrupted in the hsR722W variant ( Figure S3b ) [26] , [32]; thus , the lack of CTD phosphorylation and transcriptional activity in our studies must be attributed to the wrong positioning of the CAK complex caused by the impaired XPD/p44 interaction . This is underlined by the observation that the hsR722W variant displays a phenotype comparable to a TFIIH lacking XPD in its entirety and cannot be rescued by increasing amounts of CAK ( Figure 5d ) . The data therefore suggest that the essential role of XPD during transcription is to anchor the CAK to the core TFIIH and consequently support its function within the PIC . The phenotype of other TTD mutants strongly supports this hypothesis . From the X-ray structures of XPD , it was speculated that mutations , which alter the overall stability of XPD , exhibit a TTD phenotype . This includes R112H , C259Y , R592P , and D673G of human XPD [3] , [5] , [6] . C259Y is located in the Arch domain of XPD , and it was recently shown that it influences the CAK interaction [32] , thus being the probable cause for the transcriptional defect of this variant . We therefore propose that XPD only acts as a scaffold for the structural integrity within the core TFIIH and is required for the recruitment of other enzymatic factors like the CAK . When the XPD/p44 interactions , on the one hand , and the XPD/MAT1 interactions , on the other hand , are permitted , XPD promotes transcription , despite being completely inactive as an enzyme ( Figure 6 ) . This is further supported by the phenotypes of the C134S and C155S variants that presumably comprise an unfolded FeS domain . Because the two major protein–protein interaction sites in XPD reside in the C-terminal HD2 and in the Arch domain ( Figure 1 ) , the FeS domain does not seem to be vital for either of these interactions as these two variants still support transcription . The observed reduced transcriptional activity is most likely due to an overall effect on protein stability originating from the loss of a folded FeS domain , as also proposed for the R112 TTD mutant [2] , [4] . Thus , our data indicate that the FeS domain of XPD seems to be uniquely dedicated to the helicase function in NER and is dispensable for transcription . Intriguingly , the special role of the FeS cluster for NER is further supported by other studies that link this domain specifically to damage recognition [4] , [44] , [47] , [48] . In summary , our data demonstrate that the tasks of XPD within transcription and repair are strikingly different . In NER , XPD has to fulfill several roles by promoting protein–protein interactions and acting as a helicase , whereas for transcription only the protein–protein interactions are vital . In contrast to XPB , which acts as an enzyme during NER and transcription , XPD does not assume a dual role as an enzyme ( Figure 6 ) . The enzyme XPD is vital for NER , whereas for transcription only the intact “shell” of XPD is sufficient . Moreover , the present data allow the design of a TFIIH variant that is solely active in transcription , with repair being abrogated . In this aspect , it will be tempting to design an animal model with such a mutation that would certainly represent a pure form of XP and would be greatly useful for a better analysis of the disease itself . Moreover , in cancer research , this represents an ideal model for drug design .
The gene encoding full-length ctXPD , ctp44 ( 1–285 ) , and ctMAT1 ( 1–248 ) were cloned from a cDNA library from C . thermophilum ( provided by Ed Hurt ) . ctXPD and ctp44 were cloned in pETM11 ( EMBL-Heidelberg ) . CtXPD mutants were generated using the Quick-Change site-directed mutagenesis kit ( Stratagene ) . The reactions were carried out as suggested by the manufacturer's instructions . All mutants were verified by double-stranded sequencing . CtXPD wild type , the variants , and p44 were expressed as N-terminally His-tagged proteins in Escherichia coli BL21-CodonPlus ( DE3 ) -RIL cells ( Stratagene ) by induction with 0 . 1 mM isopropyl-β-thiogalactoside at 14°C for 18 h . The proteins were purified to homogeneity by metal affinity chromatography ( Ni-NTA , Invitrogen ) followed by anion exchange chromatography ( AEC ) in the case of ctXPD or by SEC where indicated . AEC was performed using a MonoQ 5/50 GL column ( GE Healthcare ) with 50 mM NaPO4 pH 7 . 5 , 1 M NaCl , and 1 mM TCEP as elution buffer . The final buffer after AEC was NaPO4 pH 7 . 5 , 200 mM NaCl , and 1 mM TCEP . SEC was performed using a HiLoad 26/60 Superdex 200 prep grade column ( GE Healthcare ) in 20 mM Tris-HCl or HEPES ( pH 8 . 0 ) , 200 mM NaCl , and 1 mM TCEP . The proteins were concentrated to 5 mg/ml based on their calculated extinction coefficient using ProtParam ( SwissProt ) and then flash frozen for storage . In the case of ctMAT1 , recombinant expression of the N-terminally thioredoxin-His-tagged fusion protein was carried out in E . coli BL21 ( DE3 ) star ( Invitrogen ) pRARE2 cells ( Novagen ) in TB-medium , supplemented with 50 µg/ml Zinc acetate , by addition of 0 . 3 mM IPTG for 16 h at 18°C . The protein was purified by IMAC ( Ni-IDA , Macherey-Nagel ) and dialyzed overnight in the presence of HRV-14 3C protease against 20 mM Hepes pH 8 . 0 , 0 . 8 M NaCl , and 7 mM ß-Mercaptoethanol . SEC was performed as outlined above . Real-time binding assays between ssDNA ( 5′-gactacgtactgttacggctccatctctaccgcaatcaggccagatctgc-3′ ) and purified ctXPD wild type , variants , and ctp44 were performed using biolayer interferometry on an Octet RED system ( Fortebio , Menlo Park , CA ) . This system monitors interference of light reflected from the surface of a fiber optic sensor to measure the thickness of molecules bound to the sensor surface . The 3′-biotinolyted DNA was obtained from Biomers and coupled to kinetics grade streptavidin biosensors ( Fortebio , Menlo Park , CA ) at a concentration of 100 nM . Sensors coated with ssDNA were allowed to bind ctXPD in reaction buffer ( 20 mM Tris-HCl pH 8 . 0 , 150 mM NaCl , 10 mM MgCl2 , 1 mM DTT , and 1 mg/ml BSA ) at different ctXPD concentrations ranging from 0 . 5 to 5 µM . If ctp44 was added , it was always added using a molar ratio of 1∶1 with respect to ctXPD . Measurements were carried out in triplicates and with different protein batches . Binding kinetics were calculated using the Octet Data Analysis Software 6 . 3 , with a 1∶1 binding model , to calculate the association rate constants . Binding affinities were calculated as the ratio of dissociation and association rate constants . CtXPD ATPase activity was measured with an in vitro ATPase assay in which ATP consumption is coupled to the oxidation of NADH via pyruvate kinase and lactate dehydrogenase activities . Activities were measured at 37°C in 100 µl solution , containing 1 . 5 U pyruvate kinase , 1 . 9 U lactate dehydrogenase , 2 mM phosphoenolpyruvate , and 0 . 15 mM NADH , 10 mM KCl , 1 mM MgCl2 , 1 mM TCEP , and 20 mM Tris-HCl ( pH 8 . 0 ) . ssDNA ( 5′-gctcgagtctagactgcagttgagagcttgctaggacggatccctcgagg-3′ ) was added at a final concentration of 2 µM . The assay was carried out under saturating concentrations of ATP ( 2 mM ) using ctXPD wild type and variants at a concentration of 500 nM with ctp44 in a 1∶2 stoichiometric ratio as indicated . Prior to the analysis , ctXPD was heated to 50°C . For catalytic measurements , the mix of all reagents , with the exception of ATP , was preincubated at 37°C until a stable base line was achieved . Enzyme catalysis was initiated by the addition of ATP . The activity profiles were measured at 340 nm using a Floustar Optima plate reader . Initial velocities were recorded and ATP consumption was determined using the molar extinction coefficient of NADH . The measurements were carried out in triplicates and with at least two different protein batches . Helicase activity was analyzed utilizing a fluorescence-based helicase assay [8] , [37] . We used a 5′ overhang substrate with a cy3 label at the 3′ end of the translocated strand ( 5′-agctaccatgcctgcacgaattaagcaattcgtaatcatggtcatagct-3′-cy3 ) and a dabcyl modification on the 5′ end of the opposite strand ( Dabcyl-5′-agctatgaccatgattacgaatt-3′ ) . This results in a quenching of the cy3 fluorescence that is removed upon unwinding of the substrate . Assays were carried out in 20 mM Tris-HCl pH 8 . 0 , 10 mM KCl , 1 mM MgCl2 , and 1 mM TCEP . ctXPD wild type and variants were used at a concentration of 500 nM , with concentrations of ctp44 in a 1∶2 molar ratio where indicated . Prior to the analysis , ctXPD was heated to 50°C . The proteins were mixed with 40 nM open fork substrate and 250 nM capture oligonucleotide ( 5′-caattcgtaatcatggtc-3′ ) . The reaction was subsequently started with the addition of 2 mM ATP . Kinetics were recorded with a Flouromax4 fluorescence spectrometer ( Horiba Jobin Yvon ) and monitored until the reaction was completed , where possible . Fluorescence was detected at an excitation wavelength of 550 nm ( slid width , 2 nm ) and an emission wavelength of 570 nm ( slid width , 2 nm ) . Initial velocities were fitted with Origin8 and represent the averages of at least two different reactions and two independent protein batches . The cDNAs encoding full-length hsXPD wild type and variants ( C134S , C155S , Y158A , F161A , F193A , R196A , R196E , L372A , and R722 ) were cloned into pAC8F [11] , [12] , [49] . Resulting transfer vectors were recombined with baculovirus DNA ( BaculoGold DNA , Pharmingen ) in Sf9 cells to generate viruses for production of hsXPD in fusion with the Flag peptide ( DYKDDDDK ) . For production of proteins and complexes , Sf21 cells were infected/coinfected with the appropriate viruses/combinations of viruses , collected 48 h postinfection , and purified as described [32] , Briefly , cells resuspended in buffer A ( 20 mM Tris–HCl , pH 8 . 0 , 250 mM KCl , and 1 mM DTT ) containing complete protease inhibitor cocktail ( Roche ) were disrupted by sonication , and after clarification , the lysate was incubated with protein A sepharose beads cross-linked to the M2 anti-Flag antibody ( SIGMA™ ) for purification of XPD or CAK ( both harbor an epitope FLAG ) and to the 1H5 anti-p44 antibody for purification of core-IIH . After extensive washing in buffer A and equilibration in buffer B ( 50 mM Tris-HCl pH 8 . 0 , 75 mM KCl , 20% glycerol , 0 . 1% NP-40 , and 1 mM DTT ) , proteins were eluted by competition with 2 CV of buffer B containing the appropriate synthetic peptide at 0 . 5 mg/ml . Protein concentrations were estimated by quantitative WB analysis and adjusted to ∼100 ng/µl . Monoclonal antibodies against the human TFIIH subunits XPB ( 1B3 ) , XPD ( 2F6 ) , p52 ( 1D11 ) , p44 ( 1H5 ) , CDK7 ( 2F8 ) , MAT1 ( 2D3 ) , and p8 ( 1D1 ) were obtained from IGBMC's facilities . The 5′-3′ hsXPD helicase activity was analyzed using a mono-directional strand displacement assay ( for review , see [8] , [23] ) . The probe was prepared by mixing an oligonucleotide ( 5 ng ) corresponding to nucleotides 6228–6251 of the single-stranded M13mp18 ( + ) DNA with single-stranded M13mp18 ( − ) phage ( 1 mg ) in the presence of NaCl ( 25 mM ) and MgCl2 ( 2 . 5 mM ) . The mixture was heated for 2 min at 100°C and cooled slowly to RT to allow annealing of the DNA heteroduplex . Probe labeling was performed using the Klenow fragment in the presence of dTTP ( 50 mM ) and [α32P]dATP ( 70 µCi; 3 , 000 Ci/mmol; Amersham ) . After phenol/chloroform extraction , the labeled probe was purified using Micro Bio-Spin clean-up columns ( Biorad™ ) . Equivalent amounts of purified hsXPD wild type and variants ( ∼200 ng ) were added to a 5′-strand extension probe in the presence of a 1 . 5 molar excess of p44 ( ∼150 ng ) to evaluate their 5′–3′ helicase activity . The reaction was performed for 45 min at 37°C by adding the immunopurified helicase to the DNA probe at 10 nM in 20 mM Tris-HCl ( pH 8 . 0 ) , 75 mM KCl , 4 mM MgCl2 , 1 mM DTT , 4 mM ATP , and 0 . 1 mg/ml BSA with a total reaction volume of 25 µl . The reaction was stopped by adding 20 mM EDTA , 14% glycerol , 0 . 2% SDS , and 0 . 028% bromophenol to the reaction mixture . Analyses were performed by migration in a 14% polyacrylamide gel ( acrylamide/bis-acrylamide ratio , 33/1 ) and autoradiography . NER dual incision assays were performed as described [20]–[22] , [26] using a plasmid with a single 1 , 3-intrastrand d ( GpTpG ) ( 30 ng ) in a buffer containing 50 mM Hepes-KOH ( pH 7 . 8 ) , 5 mM MgCl2 , 1 mM DTT , 0 . 3 mM EDTA , 10% glycerol , 2 . 5 µg BSA , 50 mM KCl , and 2 mM ATP . Reaction mixes ( 25 µl ) containing human XPG ( 5 ng ) , XPF/ERCC1 ( 15 ng ) , XPC/hHR23B ( 10 ng ) , RPA ( 50 ng ) , XPA ( 25 ng ) , as well as a mixture of purified core-IIH ( 250 ng ) and XPD ( 100 and 200 ng ) were assembled to reconstitute TFIIH activity . Mixes were incubated at 30°C for 90 min and analyzed as previously described . The pGL3-SV40 vector expressing the FireFly luciferase under the control of the SV40 promoter was UV irradiated ( 254 nm , 1 , 000 J/m2 ) at a concentration of 100 ng/µl in 10 mM Tris-HCl ( pH 8 . 0 ) and 1 mM EDTA . HD2 fibroblasts were seeded in DMEM/HAM-F10 containing 10% FCS and 10 µg/ml gentamcin in a 24-well plate ( 50 , 000 cells/well ) . After 20 h , each well was co-transfected using JetPEI ( Polyplus ) with 450 ng pGL3-SV40 ( UV+/ ) , 100 ng of pRL-TK expressing Renilla luciferase under the control of the thymidine kinase promoter to normalize transfection efficiencies , and 25 ng of pIRES2-EGFP derivatives for expression of hsXPD wild type or variants under the control of the CMV promoter and 450 ng of pBS . The medium was exchanged after 3 h and further incubated for 36 h . Cell lysates were prepared and activities were assayed using the Dual-Glo Luciferase assay system ( Promega ) . Run-off transcription assays were performed as described [12] , [24] , [25] except that TFIIH was substituted by a mixture of purified core-IIH , CAK , and XPD , which allowed us to prepare a pre-mix containing all components with the exception of the XPD variants . Reaction mixes contained the adenovirus major late promoter sequence ( AdMLP ) EcoRI–SalI DNA template ( 75 ng ) , TFIIB ( 15 ng ) , TFIIE ( 160 ng ) , TFIIF ( 500 ng of the phenyl fraction from Pol II and the GTF purification scheme ) , TBP ( 30 ng ) , endogenous RNAP II ( 10 µg of the 1 M DEAE fraction ) , and a mixture of the purified core-IIH , CAK , and XPD . RNAP II phosphorylation was carried out as a classical runoff transcription except that ATP was added to a final concentration of 5 mM , and the amount of purified RNAP II polymerase was adjusted ( typically reduced by a factor of 3 ) . Hypo ( IIA ) and hyper ( IIO ) phosphorylated forms of RNAP II were resolved on a 6% SDS-PAGE and detected by Western blot using the monoclonal antibody ( 7C2 ) directed against the CTD . | The multiprotein complex TFIIH is crucially involved in two fundamental cellular processes—the transcription of genes by RNA polymerase II and the repair of UV-induced DNA damage by a mechanism called nucleotide excision repair ( NER ) . The xeroderma pigmentosum complementation group D ( XPD ) helicase , which is mutated in the eponymous human photosensitivity and cancer syndrome , is an important subunit of TFIIH , where it is assumed to act as a helicase , unwinding the DNA double helix . In our study , we show that XPD assumes an entirely different role in transcription and NER . In the case of repair , this protein works as an enzyme , requiring all its known functional properties . For transcription , however , none of the enzymatic functions is essential and XPD switches from an enzyme to a structural protein whose job is merely to preserve the integrity of the TFIIH complex . We have shown thus that the helicase activity of the XPD protein is exclusively devoted to repair processes and could be targeted by drugs without affecting transcription . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"proteins",
"gene",
"expression",
"genetics",
"biology",
"and",
"life",
"sciences",
"dna",
"repair",
"dna",
"dna-binding",
"proteins",
"dna",
"transcription"
]
| 2014 | In TFIIH, XPD Helicase Is Exclusively Devoted to DNA Repair |
There are no oral drugs for human African trypanosomiasis ( HAT , sleeping sickness ) . A successful oral drug would have the potential to reduce or eliminate the need for patient hospitalization , thus reducing healthcare costs of HAT . The development of oral medications is a key objective of the Consortium for Parasitic Drug Development ( CPDD ) . In this study , we investigated the safety , pharmacokinetics , and efficacy of a new orally administered CPDD diamidine prodrug , 2 , 5-bis[5- ( N-methoxyamidino ) -2-pyridyl]furan ( DB868; CPD-007-10 ) , in the vervet monkey model of first stage HAT . DB868 was well tolerated at a dose up to 30 mg/kg/day for 10 days , a cumulative dose of 300 mg/kg . Mean plasma levels of biomarkers indicative of liver injury ( alanine aminotransferase , aspartate aminotransferase ) were not significantly altered by drug administration . In addition , no kidney-mediated alterations in creatinine and urea concentrations were detected . Pharmacokinetic analysis of plasma confirmed that DB868 was orally available and was converted to the active compound DB829 in both uninfected and infected monkeys . Treatment of infected monkeys with DB868 began 7 days post-infection . In the infected monkeys , DB829 attained a median Cmax ( dosing regimen ) that was 12-fold ( 3 mg/kg/day for 7 days ) , 15-fold ( 10 mg/kg/day for 7 days ) , and 31-fold ( 20 mg/kg/day for 5 days ) greater than the IC50 ( 14 nmol/L ) against T . b . rhodesiense STIB900 . DB868 cured all infected monkeys , even at the lowest dose tested . In conclusion , oral DB868 cured monkeys with first stage HAT at a cumulative dose 14-fold lower than the maximum tolerated dose and should be considered a lead preclinical candidate in efforts to develop a safe , short course ( 5–7 days ) , oral regimen for first stage HAT .
Human African trypanosomiasis ( HAT , sleeping sickness ) is caused by two trypanosome species that are transmitted through the bite of blood-sucking tsetse flies ( Glossina spp ) . Trypanosoma brucei ( T . b . ) gambiense is endemic to West and Central Africa , while T . b . rhodesiense is endemic to East and Southern Africa [1] . The disease is focal in distribution and is marked by wide temporal and spatial variations in incidence and prevalence [2]–[4] . HAT is characterised by two clinical stages . During the first ( early , haemolymphatic ) stage , trypanosomes proliferate at the site of the fly bite , travel to local lymph nodes and bloodstream , and progressively invade other tissues [5] . Approximately 3–4 weeks post-infection with T . b . rhodesiense , or months to years with T . b . gambiense , trypanosomes invade the central nervous system ( CNS ) , initiating the second ( late , meningo-encephalitic ) stage of HAT [5] . Second stage HAT is marked by neurological and endocrine disorders . If patients are not treated , they lapse into coma and die [6] . HAT chemotherapy is stage specific . Only two drugs have been approved for the treatment of first stage HAT , pentamidine and suramin . Pentamidine , a diamidine first used clinically in 1941 , is used to treat first stage T . b . gambiense infections . Suramin , a naphthylurea first introduced for clinical use in 1921 , is effective against both trypanosome species but is mainly used against first stage T . b . rhodesiense HAT [7] , [8] . Pentamidine is associated with hypoglycaemia , pain at the injection site , diarrhoea , nausea and vomiting , while suramin is associated with hypersensitivity reactions , albuminuria , haematuria and peripheral neuropathy [7] . In addition , these drugs must be administered via intramuscular injection or intravenous infusion in well-equipped hospitals , which are not readily available or accessible in rural areas where HAT typically occurs . To overcome these limitations , two orally active compounds , fexinidazole and the oxaborole SCYX-7158 , have recently entered clinical development to treat both stages of the disease [9] . In addition , efforts by the Consortium for Parasitic Drug Development ( CPDD ) to address these limitations have resulted in the synthesis of a collection of diamidines with promising pharmacologic properties [10] , [11] . One of the new aza diamidines , 2 , 5-bis ( 5-amidino-2-pyridyl ) furan ( DB829; Figure 1 ) , exhibited an IC50 of 14 nmol/L against T . b . rhodesiense STIB900 in vitro [12] . In addition , it was shown to be 100% curative in both the acute ( T . b . rhodesiense STIB900 ) and chronic CNS ( T . b . brucei GVR35 ) mouse models of HAT after intraperitoneal administration [12] . However , the dicationic nature of DB829 and other diamidines ( e . g . , pentamidine and furamidine ) contributes to poor permeation through biologic membranes and in turn , poor systemic exposure after oral administration [13] . As such , a prodrug of DB829 was designed , 2 , 5-bis[5- ( N-methoxyamidino ) -2-pyridyl]furan ( DB868; Figure 1 ) , by masking the cationic functionalities of the active compound with methoxy groups [14] . Oral administration of DB868 was 100% curative in both the acute and chronic CNS mouse models of HAT [12] . Based on these desirable properties , DB868 progressed into our vervet monkey model to assess its potential as a new lead compound for oral treatment of first stage HAT . The purpose of this study was to evaluate DB868 metabolism in monkey liver microsomes , safety in uninfected monkeys , pharmacokinetics in both uninfected and infected monkeys , and efficacy in infected monkeys .
This study was conducted in accordance with experimental guidelines and procedures ( Ref: C/TR/4/325/116 ) approved by the Institutional Animal Care and Use Committee ( IACUC ) at the Kenya Agricultural Research Institute's Trypanosomiasis Research Centre ( KARI-TRC ) . These IACUC regulations conformed to national guidelines provided by the Kenya Veterinary Association . The test drug , 2 , 5-bis[5- ( N-methoxyamidino ) -2-pyridyl]furan diacetate ( DB868; CPD-007-10; Lot #2-JXS-28; Base MW = 366 . 37; FW = 564 . 37 ) , was supplied by the University of North Carolina-led CPDD as a yellow powder stored in opaque , water tight bottles . In the laboratory , the drug-containing vials were wrapped in aluminium foil as further protection from light and stored at room temperature . The drug was dissolved fresh daily in distilled de-ionised water ( pH 4 . 5±0 . 2 ) at concentrations permitting administration of 2 mL/kg body weight per oral administration . For example , a 15 mg/mL dose solution was prepared for animals receiving a dose of 30 mg/kg and a 10 mg/mL dose solution for a dose of 20 mg/kg . Pentamidine isethionate , supplied by the World Health Organization ( WHO ) , was used as the comparator drug . Pentamidine was dissolved in sterile distilled water and administered intramuscularly at 0 . 5 mL/kg body weight . Eighteen vervet monkeys , also known as African green monkeys or Chlorocebus ( Cercopithecus ) aethiops , weighing from 2 . 0 to 4 . 5 kg , were acquired from the Institute of Primate Research in Kenya . To ensure animal welfare and ameliorate suffering , upon arrival at KARI-TRC , the monkeys were subjected to standard quarantine procedures , including screening for zoonotic and non-zoonotic diseases/infections and treatment for both endo- and ectoparasites , for a minimum of 90 days prior to study commencement as previously described [15] , [16] . They became accustomed to staying in individual squeeze-back stainless steel cages during this time . The monkeys were maintained on a diet of fresh fruits and vegetables ( bananas , tomatoes , carrots and green maize ) and commercial monkey cubes ( Unga Feeds , Nakuru , Kenya ) fed twice daily , and were given water ad libitum . The commercial monkey cubes were manufactured to have the following nutrient composition: crude protein , 19 . 4% ( w/w ) ; crude fiber , 5 . 6% ( w/w ) ; ether extracts that include fats and lipids , 4 . 2% ( w/w ) ; and nitrogen-free extracts , 66 . 5% ( w/w ) . DB868 metabolism was studied in male vervet monkey liver microsomes ( custom-prepared by XenoTech , LLC , Lenexa , KS , USA ) by adapting a previously published method [17] . Briefly , incubation mixtures contained 10 µM DB868 , 0 . 5 mg/mL monkey liver microsomes , and 3 . 3 mM MgCl2 in 100 mM phosphate buffer ( pH 7 . 4 ) . Reactions were initiated by the addition of NADPH ( 1 mM final concentration ) . Control incubations were carried out without NADPH , DB868 , or liver microsomes . Aliquots ( 100 µL ) of the reaction mixtures were removed at 0 , 5 , 10 , 15 , 30 , 60 and 120 min and mixed with 100 µL of ice-cold acetonitrile . After centrifugation to pellet precipitated proteins , the supernatants were analyzed by HPLC/UV and fluorescence using the method previously described for pafuramidine ( DB289 ) and furamidine ( DB75 ) [18] . Metabolite identification was performed by comparing retention times to those of synthetic standards for M1 ( DB1679 ) , M2 ( DB840 ) , M3 ( DB1712 ) , and DB829 . DB868 and its metabolites were quantified using a calibration curve ( 0 . 1–10 µM ) generated using synthetic standards . Eight uninfected vervet monkeys , divided into two dose-groups of four monkeys ( two females and two males ) each , were used . Baseline weight and clinical and haematological data were collected over a 14-day period , after which the monkeys were orally administered DB868 at 10 mg/kg/day ( group 1 ) or 30 mg/kg/day ( group 2 ) for 10 days ( Table 1 ) . Care was taken to avoid spillage and to minimise the time between drug preparation and dosing . Drug administration utilized a dose volume of 2 mL/kg body weight . The animals were monitored for indicators of overt toxicity , including changes in feed intake , weight , demeanour , posture and stool composition and consistency . Feed intake was assessed by scoring the proportion of the daily ration consumed by each monkey on a scale of 1 ( full ration eaten ) , ¾ , ½ , ¼ , and 0 ( no feed eaten ) as previously described [15] . To increase chances of detecting potential drug-related gastrointestinal toxicity , stool samples were collected and examined visually and by faecal occult blood tests conducted according to the modified guaiac method [19] . Post-last drug dose ( LDD ) monitoring extended to a minimum of 60 days . During pre- and post-dose monitoring , monkeys were anaesthetized by intramuscular injection of ketamine HCl ( 10–15 mg/kg ) to facilitate physical examination , body weight measurements , and sample collection . Blood was collected from the femoral vein via inguinal venipuncture as described previously [16] and divided into aliquots: 1 mL blood into EDTA-containing tubes ( 1 . 5 mg EDTA/mL blood ) for full haemogram determination and 2 mL blood into EDTA-containing tubes for plasma separation . Plasma was separated using a cool spin centrifuge ( 1500 rpm for 10 min at 4°C ) . The harvested plasma was divided into aliquots for clinical chemistry determinations ( 500 µL ) , prodrug ( DB868 ) and active compound ( DB829 ) concentration measurement ( 150 µL ) , and preservation as a stock sample ( approximately 250 µL ) . All plasma aliquots were frozen at −20°C before analysis . The efficacy of DB868 administered orally at 20 mg/kg/day for 5 days was compared to that of pentamidine administered intramuscularly at 4 mg/kg/day for 7 days . To obtain an indication of dose response , DB868 also was evaluated at 10 and 3 mg/kg/day , administered orally for 7 days . The 10 and 3 mg/kg dose regimens were evaluated at a time when a regulator-imposed freeze on the acquisition of non-human primates was in force , hence only 2 animals were available per dose group . In each experiment , a 14-day baseline data collection period was observed , after which the monkeys were infected by intravenous injection of 104 T . b . rhodesiense KETRI 2537 trypanosomes diluted from the infected blood of immuno-suppressed donor Swiss white mice [16] . Post-infection monitoring for the development of parasitaemia was initiated at three days post-infection ( DPI ) while treatment began 7 DPI , subsequent to confirmation of first stage HAT ( defined as trypanosomes detectable in blood and not cerebrospinal fluid [CSF] , and CSF white cell counts less than 5 cells/mm3 ) [16] . Ear-prick blood samples to determine parasitaemia were collected prior to daily drug administration . Clinical and parasitological cure was evaluated for at least six months as previously described [16] . In our studies with the related prodrug DB844 , plasma samples collected out to 28 days post-LDD ( 6 mg/kg ) were insufficient to recover a robust estimate of the elimination half-life of the active compound DB820 [20]; therefore , in the current study , plasma samples were collected for at least 60 days post-LDD for pharmacokinetic analysis . Haematology samples were analysed using an AC3diffT Coulter Counter ( Miami , FL , USA ) . Clinical chemistry determinations were performed using a Humalyzer analyser system . Plasma was analyzed for prodrug and active compound concentrations using high performance liquid chromatography-tandem mass spectrometry ( HPLC-MS/MS ) as described below . Monkeys that were deemed as treatment failures/relapses or developed severe adverse clinical signs as defined in the protocol ( e . g . , inability or reluctance to perch , less than ¼ of normal daily feed intake for 2–3 consecutive days ) were immediately withdrawn from the study and humanely euthanized for post-mortem examination . These monkeys were euthanized by intravenous administration of 20% ( w/v ) pentobarbitone sodium solution ( 150 mg/kg body weight; Euthatal; Rhône-Mérieux , United Kingdom ) . Monkey plasma samples were processed for quantification of DB868 and DB829 using previously described methods [21] , [22] with modifications made to the transition and mass spectrometer parameters . DB868-d6P ( DB868 with deuterated pyridyl rings; 30 nM ) and DB829-d6 ( DB829 with deuterated pyridyl rings; 30 nM ) were used as internal standards and were supplied by the CPDD . Prodrug and active compound were separated on an Aquasil C18 HPLC column ( 50×2 . 1 mm , 5 µm; Thermo Fisher Scientific , Waltham , MA , USA ) and quantified using an Applied Biosystems API 4000 triple quadrupole mass spectrometer equipped with a Turbo V source and electrospray probe ( Foster City , CA , USA ) . The following transitions ( in positive ion mode ) were used in multiple reaction monitoring scans: 367 . 1→320 . 2 ( DB868 ) , 373 . 7→323 . 2 ( DB868-d6P ) , 307 . 1→290 . 1 ( DB829 ) , and 313 . 2→296 . 2 ( DB829-d6 ) . Calibration standards and quality controls were prepared in blank monkey plasma to mimic the matrix of the unknown test samples . Analyte concentrations were reported only for those samples that were between the standards and controls that had an accuracy and precision within 100%±20% . If samples were below this range , data are reported as below the limit of quantification . Data below the limit of quantification were not used for the pharmacokinetic analysis . Data were analysed statistically using StatView for Windows Version 5 . 0 . 1 ( SAS Institute Inc . , Cary , NC , USA ) as previously published [20] . Repeated measures ANOVA , with Fisher's PLSD post hoc test , was used to test the effects of trypanosomal infection , as well as DB868 , on haematological and clinical chemistry parameters in comparison with respective baseline values ( α = 0 . 05 ) . Confidence intervals ( 95% ) were derived to further test the significance of observed findings . The clinical data arising from the efficacy study are presented descriptively since the group sizes were too small for statistical analysis . Pharmacokinetic outcomes were determined with standard non-compartmental methods using Phoenix WinNonlin ( version 6 . 2; Pharsight , Mountain View , CA , USA ) .
The prodrug DB868 was metabolized in male vervet monkey liver microsomes to M1 ( DB1679 ) , M2 ( DB840 ) , M3 ( DB1712 ) , M4 , and the active compound DB829 ( Figure 2 ) . These metabolites were similar to those observed when DB868 was incubated with human liver microsomes [23] . Detection of these metabolites indicated that , like pafuramidine ( DB289 ) and DB844 , DB868 undergoes sequential O-demethylation and N-dehydroxylation reactions to form the active compound DB829 in monkey liver microsomes . M2 ( DB840 ) , a bis-amidoxime metabolite , had the highest concentration at the end of the 120-min incubation . DB829 was at the limit of detection in UV mode; however , formation was confirmed by subsequent parallel fluorescence and mass spectrometric detection ( data not shown ) . Uninfected monkeys administered DB868 orally at 10 mg/kg/day for 10 days ( n = 4 ) did not exhibit any adverse clinical signs throughout the study . In addition , two of the four monkeys in the 30 mg/kg/day group did not display overt toxicity . The remaining two monkeys exhibited mild signs of toxicity , including excess mucous in the stool ( monkey 567 ) and transient inappetance 1–2 days post-LDD ( monkeys 567 and 546 ) . In general , stool texture and consistency was unchanged and faecal occult blood tests revealed nothing significant in any study subject , suggesting that no notable gastrointestinal toxicity occurred . The mean body weight of monkeys in the 30 mg/kg group exhibited minimal variation ( Figure 3 ) , with a maximum decline of 6 . 5% from baseline ( 3 . 1 kg±0 . 3 ) . A maximum decline of 6 . 2% from baseline ( 3 . 2 kg±0 . 5 ) was observed in the 10 mg/kg group ( data not shown ) . Overall , the two oral DB868 dose regimens were well tolerated . Haematological parameters of the two treatment groups did not vary significantly from baseline throughout the study . For the 30 mg/kg group , the baseline mean red blood cell ( RBC ) and platelet counts ( ± SE ) were 5 . 5 ( ±0 . 1 ) ×106 and 3 . 7 ( ±0 . 2 ) ×105 cells/µL of blood , respectively , and showed little change throughout the study ( p = 0 . 10 and 0 . 06 , respectively; Figure 3 ) . Similarly , no significant variations were seen in the RBC or platelet counts for the 10 mg/kg group ( data not shown ) . The mean white blood cell ( WBC ) count exhibited a minor transient increase post-LDD ( 1 . 5-fold over the baseline count ( ± SE ) of 4 . 8 ( ±0 . 5 ) ×103 cells/µL of blood; p = 0 . 05; Figure 3 ) , which returned to baseline after 24 h . A comparable trend was seen in the WBC count of the 10 mg/kg group ( data not shown ) , as well as in a separate monkey that was not administered drug or vehicle but had blood sampled at the same time as monkeys in the current experiment , suggesting that the change in the WBC count was not drug-related . Plasma biomarkers of liver injury , alanine aminotransferase ( ALT ) , aspartate aminotransferase ( AST ) and total and direct bilirubin , were monitored in uninfected monkeys prior to ( baseline ) , during , and following completion of the 10-day DB868 dosing regimens . At the two baseline time points ( −16 and −10 days post-LDD ) , mean ( ± SE ) ALT levels were 14 . 6 ( ±3 . 1 ) and 11 . 1 ( ±2 . 3 ) IU/L , respectively , for the 10 mg/kg group and 27 . 3 ( ±13 . 0 ) and 16 . 0 ( ±7 . 2 ) IU/L for the 30 mg/kg group . During and following the completion of dosing , mean ALT levels varied considerably in the 10 mg/kg group ( Figure 4A ) . The highest post-treatment mean ALT values observed were 2 . 8-fold ( 31 . 3/11 . 1 ) greater than the second baseline sample for the 10 mg/kg group and 1 . 3-fold ( 20 . 5/16 . 0 ) for the 30 mg/kg group . Neither variation was statistically significant ( p = 0 . 71 , 10 mg/kg group; p = 0 . 37 , 30 mg/kg group ) . Mean AST and total and direct bilirubin levels also exhibited variability prior to , during , and following dosing ( Figure 4 , B and C ) but overall , were not statistically different from baseline values ( p>0 . 05 ) . Two biomarkers of kidney injury , creatinine and urea , were also evaluated in plasma samples . Mean ( ± SE ) creatinine concentrations at the two baseline time points were 63 . 6 ( ±4 . 0 ) and 54 . 1 ( ±5 . 5 ) µmol/L , respectively , for the 10 mg/kg group and 61 . 5 ( ±14 . 5 ) and 57 . 3 ( ±6 . 0 ) µmol/L for the 30 mg/kg group . Post-dosing variations were minimal ( Figure 5A ) and not statistically significant ( p>0 . 05 ) . However , mean urea levels exhibited a transient increase following the completion of each dosing regimen ( Figure 5B; Figure S1 ) . Plasma urea peaked 1–2 days post-LDD and was 2 . 1- and 2 . 7-fold greater than baseline for the 10 mg/kg and 30 mg/kg groups , respectively ( both p<0 . 05 ) . To determine whether increases in plasma urea levels were due to kidney dysfunction , the blood urea nitrogen ( BUN ) ∶creatinine ratio was calculated for the 1–2 days post-LDD period . The peak mean urea concentration for the 10 mg/kg group was 12 . 8 mmol/L , which is equivalent to 35 . 9 mg/dL of BUN . The mean creatinine concentration was 72 . 3 µmol/L ( equivalent to 0 . 8 mg/dL ) . Therefore , the BUN∶creatinine ratio was 45∶1 ( 35 . 9∶0 . 8 ) . The BUN∶creatinine ratio for the 30 mg/kg group was higher at 48∶1 . Following oral administration of the prodrug DB868 to uninfected monkeys , DB868 was detected in plasma at 4 h post-LDD ( Figure 6; Table 1 ) . DB868 concentrations declined to below the limit of detection ( BLD ) within 1–2 days post-LDD for the 30 mg/kg group and in less than 1 day post-LDD for the 10 mg/kg group ( data not shown ) . Accurate recovery of pharmacokinetic outcomes for DB868 was precluded for the 10 mg/kg dose group ( Table 1 ) . Greater inter-individual variability was observed for the 4 h post-LDD concentration ( C4 h ) of DB868 than for DB829 ( Table 1 ) . The geometric mean DB868 C4 h for the 30 mg/kg group ( 466 nmol/L ) was 5 . 2-fold greater than that for the 10 mg/kg group ( 89 nmol/L ) . The geometric mean DB829 C4 h for the 30 mg/kg group ( 320 nmol/L ) was 1 . 6-fold greater than that for the 10 mg/kg group ( 185 nmol/L ) . The geometric mean AUClast and AUC0-∞ for DB829 in the 30 mg/kg group were 2 . 2-fold greater than that in the 10 mg/kg group . The geometric mean terminal elimination half-life for DB829 was comparable between the two dose groups ( 29 and 31 days , respectively ) . Following inoculation , the median ( range ) prepatent period of T . b . rhodesiense infection in the monkeys was 4 . 5 ( 3–6 ) days ( Table 2; Figure 7 ) . The bloodstream form of T . b . rhodesiense KETRI 2537 trypanosomes multiplied rapidly , reaching a peak mean count of 1 . 1×107 trypanosomes/mL blood; in some monkeys , the count peaked as high as 1 . 3×108 trypanosomes/mL ( antilog 8 . 1; Table 2 ) . Classical signs of T . b . rhodesiense infection were observed , including rough hair coat , dullness , marked loss of appetite , and marginal declines in body weight ( 4% of pre-infection weight ) and RBC count ( 7% of pre-infection value ) . Rectal body temperature increased from a pre-infection mean ( ± SE ) of 38 . 3 ( ±0 . 2 ) °C to a high of 38 . 7 ( ±0 . 2 ) °C at 7 DPI; however , the increase was not statistically significant ( p = 0 . 06 ) . Trypanosomes were not detected , nor were white cell counts elevated in the CSF ( data not shown ) , confirming that the monkeys were in the first stage of disease when treatment was initiated at 7 DPI . The prodrug DB868 was administered orally to three groups of monkeys: 20 mg/kg/day for 5 days ( n = 3 ) , 10 mg/kg/day for 7 days ( n = 2 ) , or 3 mg/kg/day for 7 days ( n = 2 ) . A fourth group of monkeys ( n = 3 ) was treated intramuscularly with the comparator drug , pentamidine , at 4 mg/kg/day for 7 days ( Table 2 ) . Both oral DB868 and intramuscular pentamidine demonstrated efficacy against first stage infection as discussed below . Trypanosome-associated waves of parasitaemaia were not observed ( Figure 7 ) , likely because all infections were treated during the first wave of parasitaemia . The prodrug DB868 was detected in the plasma of all monkeys with first stage HAT , regardless of the dosing regimen , at 0 . 04 days ( 1 h ) post-LDD ( data not shown ) . The Tmax varied between individuals , with the majority ( 4/6 monkeys ) occurring at 0 . 04 days ( 1 h ) post-LDD . DB868 concentrations declined rapidly . Only one monkey ( monkey 675; 10 mg/kg/day for 7 days ) had detectable levels at 8 h post-LDD , precluding accurate recovery of pharmacokinetic outcomes for DB868 . The median Cmax for DB829 for the 20 and 10 mg/kg groups ( 435 and 205 nmol/L , respectively ) were 2 . 6- and 1 . 2-fold higher , respectively , than that for the 3 mg/kg group ( 170 nmol/L ) ( Figure 8; Table 3 ) . The median Tmax for the 20 and 3 mg/kg groups were similar ( 4 h ) , whereas that for the 10 mg/kg group was longer ( 1 day ) . The median Tmax for DB829 in each dose group was longer than that for DB868 . The mean AUClast for DB829 in the 20 and 10 mg/kg groups was 10- and 7-fold greater , respectively , than that in the 3 mg/kg group ( Table 3 ) . Accurate AUC0-∞ and terminal elimination half-life were only recoverable for the 20 mg/kg group and one monkey in the 10 mg/kg group , precluding between-dose comparisons . The median AUC0-∞ and terminal elimination half-life for the 20 mg/kg group were twice those of the one ( monkey 691 ) in the 10 mg/kg group ( Table 3 ) .
Consistent with observations involving human liver microsomes [23] , vervet monkey liver microsomes metabolized the prodrug DB868 to four intermediate metabolites and the active compound DB829 . Similar metabolic pathways have been reported for the related prodrugs pafuramidine and DB844 in rat , monkey and human liver microsomes [20] , [25]–[27] , demonstrating that these alkoxy-type diamidine prodrugs [28] are converted to active compounds in different animal species . The low DB829 concentrations in the microsomal samples were not unexpected , as similar results were observed with the active compound generated from the related prodrug pafuramidine [17] . The final metabolic steps in the formation of DB829 , the N-hydroxylation of M2 and subsequently M4 , are analogous to those in the conversion of pafuramidine to furamidine . During microsomal pafuramidine metabolism , these steps are catalyzed by cytochrome b5/b5 reductase [29] . This enzyme is also abundant in mitochondria and Golgi [29] , [30] , explaining why DB868 is more efficiently converted to DB829 in intact hepatocytes compared to the isolated microsomal system [27] , [31] . Collectively , these results provided justification for in vivo testing of the prodrug DB868 in uninfected and infected vervet monkeys . No significant overt toxicity was seen with up to 30 mg/kg/day DB868 orally for 10 days ( cumulative dose [CD] = 300 mg/kg ) in the vervet monkey safety study , suggesting that this dose was below the maximum tolerated dose , but slightly above the no observed adverse effect level ( NOAEL ) . Pharmacokinetic analysis of plasma from the uninfected monkeys showed that the geometric mean C4 h of the active compound DB829 in the 30 and 10 mg/kg groups were 23- and 13-fold greater than the IC50 ( 14 nmol/L ) against T . b . rhodesiense STIB900 , respectively . These results confirmed that DB868/DB829 are available systemically following oral administration , similar to pafuramidine/furamidine and DB844/DB820 [20] , [32] , prompting further evaluation in the monkey model of first stage HAT . Based on the above observations , DB868 efficacy was evaluated in vervet monkeys with first stage HAT using doses below 30 mg/kg/day in order to minimise the risk of unfavourable clinical outcomes . Dosing durations of 5–7 days were chosen based on the hypothesis that first stage disease can be cured using short treatment durations . DB868 , administered orally , cured all monkeys of their experimentally introduced T . b . rhodesiense infection ( Table 2 ) . Post-treatment monitoring must be at least 180 days in order to declare a cure in the monkey model of first stage HAT [16] . All three DB868 dosing regimens , including the lowest evaluated ( 3 mg/kg/day for 7 days; CD = 21 mg/kg ) , effectively cleared the monkeys of their considerably high parasitaemia , which in some cases was as high as 108 trypanosomes/mL of blood . Elimination of the pathogens allowed the monkeys to return to their clinical and haematological baselines within one month post-LDD ( data not shown ) , similar to what was observed in pafuramidine and DB844 efficacy studies conducted in this monkey model [16] , [20] . These results highlight the ability of diamidines to eliminate injurious trypanosomes , allowing the body to repair/heal itself . Furthermore , oral DB868 appears to be superior to oral pafuramidine in this first stage HAT monkey model , as a higher dose of pafuramidine than DB868 was required to achieve complete cure ( 10 mg/kg for 5 days vs . 3 mg/kg for 7 days , respectively ) [16] . The longer dose regimen for DB868 compared to pafuramidine is consistent with the in vitro observation that DB829 required longer exposure periods than furamidine to kill T . b . brucei s427 trypanosomes [33] . For example , a 24-h exposure to DB829 ( 2 . 7 µM ) was required , whereas a 1-h exposure to furamidine ( 3 . 2 µM ) was required to kill these trypanosomes in culture . Based on a mg dose basis , DB868 has a larger therapeutic window than pafuramidine in vervet monkeys . No notable drug-induced overt toxicity was observed in either uninfected or infected monkeys administered DB868 , except for mild excess mucous in the stool ( n = 1 ) and transient inappetance ( n = 2 ) in the group receiving 30 mg/kg/day for 10 days . In comparison , similar mild adverse events were observed when pafuramidine was administered at 10 mg/kg/day for 10 days to vervet monkeys ( unpublished data; JK Thuita ) . DB868 , at all doses tested , did not cause significant elevations in plasma biomarkers of liver ( ALT , AST , total and direct bilirubin; Figure 4 ) and kidney ( creatinine; Figure 5 ) injury . These results contrasted with those of pafuramidine , which caused transient liver injury during an extended phase I clinical trial in humans [11] . In addition , only a slight increase in ALT ( less than 2-fold ) was observed in female Sprague-Dawley rats administered DB868 orally ( 25 mg/kg/day for 3 weeks ) compared to untreated rats , whereas an 18-fold increase was observed in rats administered pafuramidine ( 12 mg/kg/day for 4 weeks ) [34] . In the current study , plasma urea concentrations , and therefore BUN levels , were transiently increased ( 2–3-fold; Figure 5B ) shortly ( 1–2 days ) after the last drug dose . However , the BUN∶creatinine ratio was above the critical 20∶1 ratio , suggesting that the elevations were likely due to pre-renal causes such as dehydration . Direct comparison of plasma liver and kidney injury biomarkers between DB868 and pafuramidine in vervet monkeys are not possible due to insufficient data on pafuramidine . Nevertheless , the safety profile of DB868 is improved over that of the prodrug DB844 , which caused significant liver injury when administered to monkeys at doses above 10 mg/kg/day [20] , necessitating withdrawal of DB844 from further development . As discussed above , our study has demonstrated that oral DB868 has excellent efficacy and an improved therapeutic window in the first stage HAT monkey model , making it a promising lead candidate for further preclinical development . However , based on the previous lessons learned from the development of pafuramidine [11] , several issues warrant mention . First , the kidney safety liability of DB868 needs to be further examined using more predictive models and biomarkers . Pafuramidine development was terminated due to an unexpected severe kidney injury that occurred in five patients ( ∼6% ) , a liability not predicted by traditional preclinical safety testing in rodents [11] . Recently , Harrill et al . [21] showed , using a mouse diversity panel comprised of 34 genetically diverse inbred mouse strains , marked elevations of urinary kidney injury molecule-1 ( KIM-1 ) in sensitive mouse strains following oral administration of pafuramidine , while classical kidney injury biomarkers , BUN and serum creatinine , remained unchanged . Hence , it may be prudent to screen DB868 for kidney injury liability using the sensitive mouse strains therein identified and KIM-1 . Encouraging results from Sprague Dawley rats administered pafuramidine or DB868 orally ( 12 mg/kg/day×28 days ) showed that DB868 had no effect on KIM-1 during the entire 4-month observation period ( 28 days of drug administration and 92 days of recovery period ) , whereas pafuramidine caused a 13-fold increase in KIM-1 one week post-LDD [34] . Second , treatment regimens should be optimized with the pharmacokinetics taken into consideration . The active compound DB829 was readily detected in the plasma following oral administration of the prodrug DB868 ( Figures 6 and 8 ) . Afterwards , DB829 was slowly eliminated from the blood with a terminal elimination half-life ranging from days to nearly three months depending on the dosing regimen ( Tables 1 and 3 ) . This is similar to suramin [7] , the only other first stage HAT drug besides pentamidine . Plasma concentrations of DB829 remained >100 nmol/L for long periods following the last DB868 dose , in some monkeys up to 7 days post-LDD . This finding was comparable to that reported for DB844 [20] and provides additional evidence that 1 ) prolonged treatment durations may not be necessary , especially for first stage HAT , and 2 ) daily dosing of DB868 and other diamidine prodrugs may not be necessary . However , since trypanosomes are tissue invasive , a follow-up pharmacokinetic study is needed to determine if plasma active drug concentrations are predictive of tissue concentrations . Third , combined treatment of DB868 with a fast-acting trypanocide may accelerate recovery , improve efficacy and clinical outcomes , and prevent resistance . The time to clearance of trypanosomes from peripheral blood was shorter in monkeys treated with pentamidine intramuscularly ( 2 days after the 1st 4 mg/kg dose ) than with oral DB868 ( 2–14 days after the 1st dose depending on the dose; Figure 7 ) . It took longer for the lower DB868 dose regimen groups ( 3 and 10 mg/kg ) to clear parasites from the blood than the 30 mg/kg group ( 6–14 days vs . 2–5 days after the 1st drug dose; Figure 7 ) . The difference in parasite clearance between pentamidine and DB868 ( or the active compound DB829 ) is consistent with observations in mouse models of HAT ( Wenzler et al . , Antimicrob Agents Chemother . under review ) . However , the slower parasite clearance by DB868 did not seem to compromise efficacy in the monkey model . Nevertheless , combining oral DB868 with another fast-acting trypanocidal agent , such as the oral drugs currently in clinical trials , may offer fast elimination of parasitaemia , the ease of oral pills , and a low probability of developing resistance . In conclusion , oral DB868 demonstrated improved efficacy and safety profiles in the vervet monkey model of first stage HAT , in comparison to the previous clinical candidate pafuramidine . As such , DB868 should be considered a preclinical candidate for oral treatment of first stage HAT , supplementing the current drug development pipeline . | Development of orally administered medicines for human African trypanosomiasis ( HAT ) would potentially reduce the need for patient hospitalization , thus lowering healthcare costs . In this study , we investigated the potential of a novel diamidine prodrug , DB868 ( CPD-007-10 ) , as an oral treatment for first stage HAT . When administered to uninfected monkeys by oral gavage , DB868 was well tolerated up to a maximum dose of 30 mg/kg/day for 10 days ( cumulative dose [CD] = 300 mg/kg ) . DB868 was absorbed into the systemic circulation and was converted to the active compound DB829 in concentrations that were potentially therapeutic for blood trypanosomes . Subsequently , DB868 was evaluated for efficacy in the first stage vervet monkey model of HAT in which treatment was initiated at 7 days post-infection with T . b . rhodesiense KETRI 2537 . All infected monkeys were cured , even at the lowest of the three dose regimens tested: 3 mg/kg/day for 7 days ( CD = 21 mg/kg ) , 10 mg/kg/day for 7 days ( CD = 70 mg/kg ) and 20 mg/kg/day for 5 days ( CD = 100 mg/kg ) . DB868 conversion to DB829 was comparable between uninfected and infected monkeys . In view of its favourable safety and oral efficacy profile , we conclude that DB868 is a suitable candidate for development as a new treatment for first stage HAT . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"infectious",
"diseases",
"drug",
"research",
"and",
"development",
"drugs",
"and",
"devices",
"african",
"trypanosomiasis",
"neglected",
"tropical",
"diseases",
"parasitic",
"diseases",
"drug",
"discovery"
]
| 2013 | Safety, Pharmacokinetic, and Efficacy Studies of Oral DB868 in a First Stage Vervet Monkey Model of Human African Trypanosomiasis |
Bunyaviruses are a large family of segmented RNA viruses which , like influenza virus , use a cap-snatching mechanism for transcription whereby short capped primers derived by endonucleolytic cleavage of host mRNAs are used by the viral RNA-dependent RNA polymerase ( L-protein ) to transcribe viral mRNAs . It was recently shown that the cap-snatching endonuclease of influenza virus resides in a discrete N-terminal domain of the PA polymerase subunit . Here we structurally and functionally characterize a similar endonuclease in La Crosse orthobunyavirus ( LACV ) L-protein . We expressed N-terminal fragments of the LACV L-protein and found that residues 1-180 have metal binding and divalent cation dependent nuclease activity analogous to that of influenza virus endonuclease . The 2 . 2 Å resolution X-ray crystal structure of the domain confirms that LACV and influenza endonucleases have similar overall folds and identical two metal binding active sites . The in vitro activity of the LACV endonuclease could be abolished by point mutations in the active site or by binding 2 , 4-dioxo-4-phenylbutanoic acid ( DPBA ) , a known influenza virus endonuclease inhibitor . A crystal structure with bound DPBA shows the inhibitor chelating two active site manganese ions . The essential role of this endonuclease in cap-dependent transcription was demonstrated by the loss of transcriptional activity in a RNP reconstitution system in cells upon making the same point mutations in the context of the full-length LACV L-protein . Using structure based sequence alignments we show that a similar endonuclease almost certainly exists at the N-terminus of L-proteins or PA polymerase subunits of essentially all known negative strand and cap-snatching segmented RNA viruses including arenaviruses ( 2 segments ) , bunyaviruses ( 3 segments ) , tenuiviruses ( 4–6 segments ) , and orthomyxoviruses ( 6–8 segments ) . This correspondence , together with the well-known mapping of the conserved polymerase motifs to the central regions of the L-protein and influenza PB1 subunit , suggests that L-proteins might be architecturally , and functionally equivalent to a concatemer of the three orthomyxovirus polymerase subunits in the order PA-PB1-PB2 . Furthermore , our structure of a known influenza endonuclease inhibitor bound to LACV endonuclease suggests that compounds targeting a potentially broad spectrum of segmented RNA viruses , several of which are serious or emerging human , animal and plant pathogens , could be developed using structure-based optimisation .
Bunyaviridae is the largest single family of mostly animal viruses comprising more than 300 species , divided into five genera: Orthobunyavirus , Phlebovirus , Nairovirus , Hantavirus and Tospovirus , the latter infecting plants . The viruses are mainly insect transmitted except Hantaviruses which are rodent borne . They possess a tri-partite negative sense RNA genome , the segments being designated according to size as L , M and S . The L segment encodes a single protein , the RNA-dependent RNA polymerase ( polymerase or L protein ) which ranges according to genus from 240–460 KDa; the M segment encodes two glycoproteins ( Gn , Gc ) and in some cases a non-structural protein ( NSm ) and the S segment encodes the nucleocapsid protein ( N ) and generally a non-structural protein ( NSs ) . In common with other negative strand RNA viruses , the RNA genome is coated with N protein forming ribonucleoprotein complexes ( RNPs ) which also contain the polymerase . Bunyavirus particles are generally spherical with the glycoproteins embedded in a membrane envelope which surrounds the RNPs . Replication occurs in the cytoplasm , unlike influenza virus , a negative strand segmented RNA virus of the orthomyxovirus family , which replicates in the nucleus . Bunyaviruses are globally widespread although individual species may be locally restricted by the specificity for particular insect species . Several bunyaviruses are important or emerging human or plant pathogens including La Crosse orthobunyavirus ( childhood encephalitis ) , Hantaan virus ( hemorrhagic fever with renal syndrome ) , Rift Valley fever phlebovirus , tomato spotted wilt tospovirus and Crimean-Congo ( hemorrhagic fever ) nairovirus . Bunyaviridae polymerases share with those of Orthomyxoviridae ( e . g . influenza viruses ) use of the mechanism of ‘cap-snatching’ for viral mRNA transcription , since , unlike the polymerases from non-segmented negative strand RNA viruses , they do not possess a capping activity . Cap-snatching involves binding of host capped mRNAs to the RNPs , cleavage of these RNAs close to the 5′ cap by a viral endonuclease activity and use of the short capped fragments as primers for viral mRNA transcription . This mechanism was first demonstrated for influenza virus polymerase [1] . An additional 11–15 nucleotides , heterogeneous in sequence , at the 5′ end of the viral mRNA prior to the start of the viral transcribed sequence was observed for snowshoe hare virus [2] and subsequently it was shown that La Crosse virions contain a primer-stimulated RNA polymerase and a methylated cap-dependent endonuclease [3] , analogous to the situation found for influenza virus . Subsequently it has been shown that cap-snatching is employed by representative viruses of all five genera of Bunyaviridae [4] , [5] , [6] . Arenaviridae , another family of segmented RNA viruses , are also proposed to have a cap-snatching activity [7] . Although it is well known that the bunya- and arenavirus L-proteins contain in their central region the six polymerase motifs ( designated preA , A–E ) characteristic of negative-strand RNA viruses [8] , [9] , [10] , the rest of the large protein is completely uncharacterised functionally and structurally , partly due to its lack of sequence homology with other proteins . Recently , crystallographic studies of functional domains of influenza virus polymerase , which is likely to be evolutionary related to the bunyavirus and arenavirus L-protein [8] , have precisely defined the location and atomic structure of the two key domains for cap-snatching [11] . The mRNA cap-binding domain is located in the central region of the PB2 subunit [12] , whereas the endonuclease activity resides in the N-terminal region of the PA subunit [13] , [14] . We therefore asked the question whether this new structural information could aid localisation of domains relevant to cap-snatching in the bunyavirus L-protein ? The influenza virus endonuclease domain has a core fold and divalent cation binding residues characteristic of the PD- ( D/E ) xK nuclease superfamily [15] . Unusually , it has a histidine as one of the metal ligands , which leads to a strong manganese preference for activity [13] . Surprisingly , a highly conserved motif ( H . . . . PD . . . D/E . . . K ) at the extreme N-terminal region of diverse bunyavirus L-proteins with very similar features as now recognised to be important in the influenza PA N-terminal domain , was reported some time ago [8] , [16] ( Figure 1 ) . This strongly suggested the presence of an endonuclease at the N-terminal of bunyavirus L-proteins . To investigate this further , we used the fact that the influenza endonuclease domain is about 200 residues [13] and made a synthetic gene comprising the first 250 residues of the La Crosse orthobunyavirus ( LACV ) L protein . Here we present biochemical and structural data that clearly show that the LACV L protein has a functional , manganese-dependent N-terminal endonuclease domain that indeed has very similar characteristics to that of influenza virus endonuclease . We also show that single point mutations that disable the nuclease activity in vitro , when introduced into the full-length L-protein , eliminate cap-dependent transcription in a LACV RNP reconstitution assay in cells . By sequence analysis we extend our results to show that all Bunyaviridae most likely possess such an endonuclease as well as members of other segmented RNA virus families , such as the bi-segmented Arenaviridae and four to six segmented Tenuiviruses . Implications for the evolution of segmented RNA viral polymerases are discussed as well as the prospects for a broad spectrum anti-viral targeting this endonuclease .
The original 1–250 residue construct of LACV L-protein ( LC250 ) was truncated on the basis of partial proteolysis with papain in order to identify a minimal active and stable fragment that was well expressed and soluble . Papain resistant constructs with C-terminal residue 176 , 180 , 183 , 186 and 190 were produced . The protein encompassing residues 1–180 ( LC180 ) was biochemically characterised and found to be active as a nuclease ( see below ) . The protein encompassing residues 1–183 ( LC183 ) yielded hexagonal crystals which diffract to 2 . 1 Å resolution , with four molecules per asymmetric unit . The crystal structure was solved by the single anomalous dispersion ( SAD ) method using seleno-methionine substituted protein . A native data set was refined to an R-factor/R-free of 0 . 185/0 . 223 at 2 . 2 Å . This structure shows clearly one manganese ion bound with octahedral co-ordination in the active site cavity ( designated site 1 ) ( Supplementary Figure S1 ) . A second structure , at 2 . 3 Å resolution ( R-factor/R-free = 0 . 177/0 . 216 ) , was obtained after soaking the crystals with the diketo acid inhibitor 2 , 4-dioxo-4-phenylbutanoic acid ( DPBA ) . This is a member of the family of 4-substituted 2 , 4-dioxobutanoic acids which are known inhibitors of influenza virus endonuclease ( [13] , [17] ) . This structure clearly shows in addition to the manganese ion in site 1 , a second in an adjacent site 2 , with the inhibitor co-ordinating the two ions . The two ions are separated by 3 . 8 Å and have overlapping octahedral co-ordination ( Supplementary Figure S1 ) . In both cases , the identity of the manganese ions was indicated by anomalous scattering ( Supplementary Figure S1 ) . The crystal structure of LC183 and its comparison with the N-terminal endonuclease domain of influenza virus polymerase PA subunit ( PA-Nter , PDB entry 2W69 [13] ) is shown in Figure 2 . The secondary structure of LC183 , together with a structural alignment of the N-terminal regions of selected orthobunya and tospoviruses L-proteins , is displayed in Figure 1 . Comparison of Figures 2a and 2b shows that LC183 has a very similar alpha-beta topology to PA-Nter , although the helices are of significantly different lengths . Notably , the different position of PA-Nter helix αa and the increased length of helix αb gives LC183 a more slender , elongated shape with a more exposed active site that actually lies in a groove between two lobes ( Supplementary Figure S2 ) . Focussing in on the active site region , based around a four-stranded anti-parallel beta sheet , the similarity in structure is even more striking ( Figure 3 ) , despite essentially no sequence homology . As expected from the initial sequence analysis , LC183 has exactly the same core , cation-binding fold as found in PA-Nter and more generally in the PD- ( D/E ) xK nuclease superfamily [15] . This core region comprises 55 residues which can be superposed with a root-mean-square deviation of carbon alpha positions of 1 . 36 Å . Indeed there is a one-to one mapping between the ligands of the two metal binding sites: site 1 has ligands His34 , Asp79 , Asp92 and the carbonyl-oxygen of Tyr93 in LC183 corresponding to His41 , Asp108 , Glu119 and Ile120 in PA-Nter; site 2 has ligands Asp52 and Asp79 in LC183 corresponding to Glu80 and Asp108 in PA-Nter . Interestingly the putative catalytic lysine , characteristic of the PD- ( D/E ) xK nuclease superfamily , is likely to be Lys94 in LC183 ( for confirmation , see below ) . As in EcoRV restriction enzyme ( see [13] for a comparison of PA-Nter with EcoRV ) , this residue emerges from the central β-strand ( βb ) of the core fold rather than from helix αd as in the case of PA-Nter ( Figure 3 ) . Finally there is a clear correspondence between Lys108 and Lys137 in respectively LC183 and PA-Nter , both emerging from helix αd; in both cases this basic residue is in a position to potentially interact with a nucleic acid substrate . These similarities strongly suggest that LC183 will have a similar two-metal dependent nuclease activity to that of PA-Nter [13] . The inhibitor DPBA binds tightly to the two metal ions in the active site with three of its oxygen atoms replacing three water molecules in the two metal ion co-ordination ( Figure 3c , Supplementary Figure S1 ) . The phenyl group of the inhibitor is less well-defined in the electron density indicative of some residual rotational flexibility . This is indicative of the fact that no direct interactions are made between the DPBA and residues of the protein . Despite the overall high degree of structural similarity of LC183 and PA-Nter , there are some significant differences . In the case of LC183 , Asp52 , one of the acidic ligands of cation site 2 , is on a flexible loop . Indeed in the native structure , this loop is in an open conformation with Asp52 turned away from the active site and consequently only the manganese ion bound in site 1 is present ( Figure 3c ) . In the inhibitor bound structure , the loop is in a closed conformation and Asp52 contributes to binding the second manganese . This suggests that there is preferential and tighter metal binding to site 1 , consistent with it having four protein and two water ligands , and weaker metal binding to site 2 , which has only two protein oxygen ligands . Furthermore metal binding to site 2 requires closure of the Asp52 loop and may be co-operatively dependent on binding of a nucleic acid substrate or metal binding inhibitor such as DPBA . In contrast , in PA-Nter there is no evidence of flexibility of the corresponding residue Glu80 and PA-Nter can in fact bind a single magnesium atom in site 2 only in the absence of manganese ions [14] , [18] . Also , as mentioned above , in PA-Nter , helix α2 and the following loop are positioned to restrict substrate access to the active site cavity , whereas in LC183 the active site opens into a channel which could allow larger , more structured substrates to be cleaved ( Supplementary Figure S2 ) . These differences might account for some of the small discrepancies observed in nuclease activity between the two enzymes ( see below ) . Biochemical characterisation of the nuclease activity of LC180 was investigated using RNA and DNA digestion assays in the presence of a variety of divalent metal ions . Because of the structural similarity of the active sites of LC180 and PA-Nter , the experiments were guided by our previous work on influenza PA-Nter and used two of the same RNAs , a single-stranded , unstructured 51 nucleotide ( nt ) U-rich RNA and a highly structured 110 nt RNA , SRP Alu RNA as well as ssDNA [13] . Using 2 mM metal ions , Figure 4a shows that LC180 fully digests the U-rich RNA and partially digests the Alu RNA only in the presence of manganese , cobalt , zinc and nickel ions , with the preference Mn>Co≫Zn>Ni and not in the presence of magnesium , calcium or iron . Manganese dependent ssDNA endonuclease activity was observed for LC180 using a circular ssDNA , as for PA-Nter ( Figure 4c ) . Since no digestion of RNAs was observed with 2 mM magnesium ions , increasing amounts of magnesium were tested . Weak nuclease digestion of the U-rich RNA was only observed at very high magnesium concentrations above 12 . 5 mM ( Supplementary Figure S3 ) . We next tested inhibition of the manganese dependent nuclease activity by the diketo acid DPBA . As in the case of PA-Nter , DPBA inhibits digestion of both test RNAs with an estimated IC50 of between 25–50 µM ( Figure 5 ) . In parallel with these nuclease activity tests we measured the metal ion and inhibitor dependent thermal stability of LC180 by a Thermofluor assay [19] , again in analogy to previous experiments described for PA-Nter [13] . The metal-free domain has an apparent melting temperature of 52 . 2 ( ±0 . 7 ) °C . Thermal stability is enhanced by 9°C upon addition of 2 mM manganese , presumably due to metal binding , with smaller increases with magnesium , calcium and cobalt ( Figure 4b ) . A supershift of about 5°C in thermal stability is observed when DPBA is added to LC180 that is pre-bound with manganese and this supershift correlates with the nuclease inhibition affect of DPBA ( Figure 5b ) , strongly suggesting that DPBA binds to the metal ions in the active site , as indeed observed in the crystal structure ( Figure 3c ) . The relative activity of influenza virus endonuclease ( PA-Nter ) and LC180 was compared under the same experimental conditions for both test RNAs with 2 mM of manganese , magnesium and calcium . Both enzymes are inactive with calcium , LC180 is more active with manganese and PA-Nter is active with 2 mM magnesium , whereas LC180 is not ( Supplementary Figure S4 ) . These experiments highlight three differences between LACV and influenza endonucleases . Firstly , influenza endonuclease is active in the presence of magnesium whereas LACV is not , secondly LACV is more active against the largely double stranded SRP Alu RNA and thirdly , LC180 is intrinsically more thermally stable with an apparent melting temperature of 52°C compared to 44°C for PA-Nter [13] . The more efficient activity against structured RNA could be due to the greater accessibility of the active site for LC183 as mentioned above ( Supplementary Figure S2 ) . It remains to be investigated whether there are any sequence preferences in the cleavage site favoured by the LACV endonuclease . We made a series of alanine point mutants of key conserved residues in the active site of LC180 in order to assess their importance for activity . These included the ligands of metal 1 ( His34 , Asp92 and Asp79 ) and of metal 2 ( Asp79 and Asp52 ) , the putative catalytic lysine 94 and a second lysine ( Lys108 ) close to the active site that is also highly conserved in bunya viruses and in PA-Nter . As a negative control , we also mutated Glu48 , again conserved in all orthobunya viruses , which was not predicted from the structure to be directly involved in the nuclease activity . All mutant proteins were well expressed as for wild-type and purified as folded proteins as judged by behaviour on gel-filtration and in thermal stability assays ( Figure 6 ) . Only the H34A mutated protein was found to be somewhat less temperature stable , probably due to a loss of charge complementation in the highly acidic active site . A H34K mutant was made and assayed instead . The results of nuclease assays with these mutant proteins with the two RNAs and of thermal stability assays are shown in Figure 6a and 6b respectively . Mutations of any single of the four metal binding ligands ( H34K , D52A , D79A and D92A ) leads to elimination of nuclease activity , as does mutation of the catalytic lysine ( K94A ) . The mutation E48A has no effect on activity and the mutation K108A leads to a reduction in activity , possibly because loss of the positively charged side-chain reduces substrate RNA binding . As described above , the wild-type protein shows significantly higher temperature stability upon binding of 2 mM manganese which is reinforced by subsequent binding of the inhibitor DPBA ( Figure 4 , Figure 5 ) . Essentially the same pattern is shown by the mutants E48A ( negative control , which however has somewhat reduced protein stability ) , K108A and K94A . This is consistent with the structural information which shows that none of these residues are directly involved in metal ligation or inhibitor binding . The D52A binding retains the manganese effect but has no inhibitor effect . Since D52 only binds metal 2 , and as our structures show , a single manganese can bind strongly to metal site 1 ( with the D52 loop in an open conformation , Figure 3c ) , we interpret this to imply that a single manganese ion bound in site 1 is sufficient to give the enhanced stability effect , whereas one ion is not sufficient to bind the inhibitor DPBA . The mutant D79A has no enhanced stability ( in fact slightly reduced stability ) in either the presence of manganese ions , with or without inhibitor , consistent with the fact that it binds simultaneously both metal ions . The H34K mutant has slightly increased stability compared to wild-type , probably because the lysine side-chain charge compensates the acidic active site better than the histidine , but there is no effect of manganese or inhibitor . This mutation thus almost certainly prevents any metal binding . Finally the mutation D92A shows a very modest manganese and inhibitor effect . It is thus possible that this mutant has still some low affinity for both metals . In summary these mutational studies show that the nuclease activity of LC180 depends critically on an intact binding site for two metal ions , preferably manganese , as well as the presence of the catalytic Lys94 . Furthermore , binding of only one manganese ion is sufficient to lead to enhanced thermal stability , whereas both metal ions are required for DPBA binding . These results are fully consistent with the crystal structures described above . To quantify the thermodynamics of binding of manganese to LC180 we used isothermal titration calorimetry in which manganese ions were titrated into wild-type or D52A mutant LC180 ( see methods and Supplementary Figure S6 ) . For wild-type protein the ITC data were fitted with a model comprising two independent sites yielding Kd's of 7 . 20 ( ±1 . 73 ) and 159 . 0 ( ±42 . 9 ) µM , although in the experiment saturation of the weaker binding site was not achieved . For the D52A mutant the ITC data were satisfactorily fitted with a model comprising a single site giving a Kd of 21 . 0 ( ±2 . 3 ) µM , with saturation of the single site being achieved . More complete results for the thermodynamic parameters of manganese binding are given in Supplementary Table S1 . Once again these results are fully consistent with our structural and thermal stability experiments with the interpretation that the strongly bound ion for the wild-type and the single site for the D52A mutant ( which have comparable affinities ) corresponds to metal site 1 and the more weakly bound site for the wild-type corresponds to metal site 2 . When magnesium was substituted for manganese no binding was detected by ITC . An analogous mutational and quantitative metal binding analysis has recently been performed for influenza virus endonuclease [20] , with slight differences in behaviour being observed , as mentioned above . To test the effect of the nuclease inactivating mutants in the context of the full-length LACV L-protein we used a previously described in vivo RNP reconstitution system in which a Renilla Luciferase ( REN-Luc ) reporter gene is used as a readout of cap-dependent transcription by the viral polymerase [21] ( For a schematic outline of this assay see Supplementary Figure S7 ) . From the in vitro work we know that the mutations do not disrupt the folding of the endonuclease domain and therefore presumably not of the full-length L-protein . Moreover , expression levels of full-length wild-type and mutant L constructs are comparable as detected by immunofluoresence ( Supplementary Figure S8a ) . The transcription assay results with the various mutants ( Figure 6c ) parallel very closely the in vitro nuclease activity of the isolated LC180 domain mutants . Only the wild-type , negative control ( E48A ) and K108A ( slightly reduced activity ) L proteins give rise to significant REN-Luc production . To detect whether these active mutants are indeed producing capped mRNAs , we co-expressed them with the polio virus 2Apro protein . This protease specifically abrogates cap-dependent mRNA translation by cleaving eukaryotic initiation factor ( eIF ) 4G [22] . The T7-driven expression constructs for the LACV L and N proteins , as well as the firefly luciferase ( FF-Luc ) transfection control escape this inhibition , since their translation is mediated by a viral internal ribosome entry site ( IRES ) . As shown in Supplementary Figure S8b the Ren reporter activity of all active LACV L variants is drastically reduced upon co-expression of 2Apro , whereas the 2Apro mutant G60A , which has lost eIF4G cleaving activity [22] , had no such effect . Moreover , IRES-driven FF-Luc expression was not affected by 2Apro , as expected ( Supplementary Figure S8c ) . Thus , the specific sensitivity of L-protein driven Ren activity to the polio virus 2Apro indicates that wt L and both the E48A and the K108A mutant transcribe capped mRNAs . Taken together , these results show that cap-dependent transcription is absolutely dependent on a functional two manganese-dependent nuclease activity at the N-terminus of the LACV L-protein , strongly suggesting that this domain is the cap-snatching endonuclease of the viral polymerase .
Cap-snatching as a method of priming transcription is uniquely restricted to segmented negative strand viruses , notably orthomyxoviruses ( influenza ) , bunyaviruses and arenaviruses . The recent structural characterisation of two functional domains relevant for cap-snatching by influenza polymerase , the cap-binding domain and the endonuclease , in respectively the PB2 and PA polymerase subunits , raise the question as to whether similar domains exist in the L-protein ( polymerase ) of bunya- and arenaviruses . The work presented here shows unequivocally that the extreme N-terminal 200 residues of LACV has a cap-snatching endonuclease activity with very close structural and biochemical features to that of the N-terminal domain of the influenza virus polymerase PA subunit . We do not yet know the context of the bunyavirus N-terminal endonuclease within the 3-dimensional structure of the complete polymerase . However it is likely that there is a cap-binding domain and probably other RNA binding domains within the polymerase ( this is certainly true for influenza polymerase ) that enhance affinity and provide specificity for capped cellular mRNAs . Also it is possible that , as with influenza virus , there are allosteric effects that activate or make accessible the endonuclease active site only upon cap-binding . We next examined whether the endonuclease signature could be identified in the L-protein of other segmented RNA viruses . Sequence analysis gives strong evidence that a homologous endonuclease domain exists at the N-terminus of the L-protein of four Bunyaviridae genera , orthobunya- , tospo , phlebo and hantaviruses , as well as tenuiviruses ( which have four to six genome segments , [23] , http://www . ncbi . nlm . nih . gov/ICTVdb/ICTVdB/00 . 069 . 0 . 01 . Tenuivirus ) and orthomyxovirus ( Figure 7 ) . In each case , the key metal binding and catalytic lysine residues can be identified . The sequence analysis shows that there are two sub-groups of these enzymes , with slightly different endonuclease signatures . Orthobunya- and Tospoviruses have the motif H . . . . D . . . PD . . . . DxK . . . . . T , whereas Phlebo- and Hantaviruses have the motif H . . . . E . . . PD . . . . ExT . . . . . K ( although in Phleboviruses the first E is replaced by a D ) . The Hantavirus motif is identical to that found in orthomyxoviruses ( Figure 7 ) . The first version has a preference for aspartates and the catalytic lysine emerges from beta-strand βb , whereas the second version has a preference for glutamates and the catalytic lysine emerges from alpha helix αd ( see Figure 3ab ) . Interestingly , the catalytic lysine interchanges with an absolutely conserved threonine at the two alternative positions ( Figure 7 ) . Nairoviruses are not included in this alignment as the location of the endonuclease is less certain . This genus of Bunyaviridae , which includes Crimean-Congo hemorrhagic fever virus , has an unusually long L-protein ( about 4000 residues , compared to 2100–2900 for most other bunyaviruses ) . The N-terminal half of nairovirus L-proteins ( i . e . prior to the polymerase motifs which start around residue 2050 ) contains a putative ovarian tumour ( OTU ) -like cysteine protease at the beginning [24] , [25] as well as other predicted motifs and domains [10] . A putative endonuclease motif of the Phlebo/Hanta/Orthomyxo type exists in the residue range 630–710 ( H ( 632 ) . . . PD ( 672 ) . . . . E ( 686 ) F . . . . K ( 699 ) , numbering for Crimean-Congo virus ) [10] , but this needs to be confirmed by structural and functional data . It is interesting to note that the rice stripe tenuivirus also contains a predicted N-terminal OTU-like protease before the endonuclease motif [26] . It has been suggested that the protease might release the viral polymerase and one or more additional proteins by autoproteolytic cleavage and/or have de-ubiquitination activity [26] . Indeed de-ubiquitination activity of Crimean-Congo virus OTU domain has been shown to inhibit Ub- and ISG15-dependent antiviral pathways [27] . Arenavirus L-proteins have a highly conserved N-terminal region of about 200 residues that contains the absolutely conserved sequence of residues PD ( 89 ) . . . E ( 102 ) xF . . . . K ( 122 ) L ( alignment not shown , numbering for Lassa virus ) . This closely resembles the Phlebo/Hanta/Orthomyxo endonuclease motif , although the histidine is clearly lacking . Very recently , systematic alanine mutation of conserved charged residues in Lassa virus L-protein outside the polymerase motifs have been performed and the effect on transcription and replication have been tested in a RNP reconstitution system [28] . Seven charged residues in the N-terminal region , including Asp89 , Glu102 and Lys122 and Asp129 , were selectively important for mRNA synthesis but did not affect genome replication . The authors concluded from these results , combined with sequence similarities to type II endonucleases and influenza virus endonuclease , that this region of the L-protein was likely to be the cap-snatching endonuclease of arenaviruses , in full agreement with our analysis . Finally , the endonuclease signature is also clearly present in the L-proteins of two related but unclassified bunyaviruses ( proposed to be called emaraviruses ) which have four rather than the usual three genome segments , European mountain ash ringspot disease ( Acc . No . YP003104764 , [29] ) and fig mosaic virus ( Acc . No . CAQ03479 , [30] ) . Both have the motif RH ( 105 ) D . . . PD ( 144 ) . . . E ( 158 ) xK ( 160 ) ( numbering for mountain ash ringspot disease virus ) and are thus most closely related to the Orthobunya and Tospoviruses , All these observations are summarised in Figure 8 which shows a schematic diagram of the architecture of polymerases from negative strand segmented RNA viruses . It is well known that the 6 motifs characteristic of negative strand RNA-dependent RNA polymerases ( pre-motif A and motifs A–E ) are present in the central region of bunya and arenavirus L-proteins and in the PB1 subunits of orthomyxoviruses [8] , [9] , [10] . The current work shows that the extreme N-terminal region of bunya- , tenui- and arenavirus L-proteins functionally corresponds to the N-terminal region of the PA subunit of orthomyxoviruses . Given that the three influenza A polymerase subunits total 2252 residues , very similar to the size of many bunyavirus complete L-proteins and all these viral enzymes have common mechanisms of transcription ( cap-snatching ) and replication , a natural hypothesis that follows is that the L-proteins might be architecturally , structurally and functionally equivalent to a concatemer of the three influenza polymerase subunits in the order PA-PB1-PB2 ( Figure 8 ) . Some indirect support for the functional concatenation of the influenza polymerase subunits comes from the fact that the inter-subunit interactions are dominated by contacts between the C and N-terminal extremities of respectively PA and PB1 and PB1 and PB2 as visualised by recent crystal structures ( reviewed in [11] ) . The most significant implication of this hypothesis is that the C-terminal third of the L-protein might be structurally and functionally equivalent to PB2 , which contains the cap-binding domain required for cap-snatching . Unfortunately , this region of the L-protein is the least well conserved and there are no obvious cross-genera conserved motifs that could point to a putative cap-binding site similar to that described for influenza A PB2 subunit [12] . This is perhaps not surprising as the PB2-like subunits of , for instance , salmon anaemia and Quaranfil viruses , two non-influenza orthomyxoviruses , are highly diverged from influenza [31] , [32] , even though both these viruses appear to possess an endonuclease at the N-terminus of the PA subunit ( Figure 7 ) . Furthermore the fact that the distance of endonucleolytic cleavage from the 5′ cap is rather variable amongst cap-snatching viruses [6] suggests that the location of the cap-binding domain might vary . In fact , there is no clear proof that any L-protein directly binds capped RNAs and even some evidence that in hantaviruses the viral N-protein may play this role [33] . Clearly more experimental work is required to elucidate the complete mechanism of cap-snatching in bunya- , tenui- and arenaviruses and to validate or otherwise the hypothesis that L-proteins are architecturally equivalent to the concatenation of PA-PB1-PB2 . Finally it is important to note that for nearly two decades , influenza virus endonuclease has been targeted for anti-viral drug discovery and a number of specific endonuclease inhibitors have been described [17] , [34] , [35] , [36] , [37] . Most of these compounds implicitly target the two metal binding site of the endonuclease , which is also the target for many HIV integrase inhibitors [38] including the currently approved raltegravir [39] , [40] . The recent structure determination of the endonuclease of influenza virus polymerase [13] , [14] gives new impetus to structure-based optimisation of these inhibitors . The results described here show that bunyaviruses and arenaviruses , amongst which are several dangerous and emerging pathogens , contain a very similar endonuclease to influenza virus , which is also therefore a good target for anti-viral drug design . Indeed , the close similarities between influenza and bunyavirus endonucleases suggests that compounds targeting a broad spectrum of segmented negative strand RNA viruses could be envisaged . Our structure of DPBA bound to LACV endonuclease shows that this is indeed the case , although this compound is of low potency [17] . In addition this structure provides the first concrete proof that these compounds do indeed chelate the two divalent cations in the endonuclease active site .
The coding sequence of the N-terminal 250 residues of LACV-L ( LC250 ) ( UniProt accession code A5HC98 ) was optimised for expression in E . coli and synthesized ( Geneart ) . A histidine tag followed by a linker and a TEV cleavage site ( MGHHHHHHDYDIPTTENLYFQG ) was added to the amino terminus of all protein constructs . All protein variants were amplified by PCR and cloned into a pET9a ( Novagen ) modified vector between NdeI 5′ and NotI 3′ sites for expression in E . coli . Mutagenesis of the proteins expressed in E . coli was performed on LC180 . Mutant constructs were obtained by site directed mutagenesis using overlapping oligonucleotides and Pfu or KOD ( Novagen ) DNA Polymerases . The constructs pTM-LACV-L , pTM-LACV-N , pLACV-vRen , pCAGGs-T7 and pTM-FF-Luc used in the RNP reconstitution have been described previously [21] . The pTM1-based expression constructs for poliovirus 2APro wt and G60R mutant were kindly provided by Luis Carrasco , Universidad Autónoma de Madrid , Spain [22] . Mutagenesis of the cDNA for the RNP reconstitution experiments was performed by generating mutant DNA fragments by PCR and insertion into the KpnI/BmtI restriction sites of the pTM-LACV-L vector . In all cases the correctness of the DNA constructs were confirmed by DNA sequencing . Proteins were expressed in Escherichia coli strain BL21 ( DE3 ) in LB media with 25 µM kanamycin at 18°C overnight after induction with 0 . 2 mM of IPTG . Labelled protein was obtained by expressing LC183 protein in E . coli with M9 minimal medium and 50 mg/L of seleno-methionine . The cells were disrupted by sonication on ice for 3 minutes in lysis buffer ( 20 mM Tris-HCl pH 7 . 6 , 150 mM NaCl , 2 . 5 mM β-mercapto-ethanol ) with EDTA-free protease inhibitor cocktail ( Roche ) . The protein from the soluble fraction was loaded onto a 5 ml Nickel column , washed with 10 volumes of lysis buffer with 50 mM imidazol and eluted with 5 volumes of 400 mM imidazol . The eluted protein was cleaved with histidine tagged TEV protease overnight at 4°C in dialysis against lysis buffer . After TEV cleavage all proteins have an additional glycine at the N-terminus . A second nickel column step was performed to remove unwanted material . The resulting untagged proteins were concentrated and purified by gel filtration chromatography using a SD75 column ( Pharmacia ) with lysis buffer for in vitro experiments or 20 mM HEPES pH 7 . 6 , 150 mM NaCl , 2 . 5 mM β-mercapto-ethanol for crystallization trials . Influenza A/H3N2 endonuclease ( PA 1–209 ) was obtained as described [13] . Purified LC proteins are contaminated with a small percentage of a degradation fragment of size LC163 . The length of the proteolytically stable amino terminal domain was defined from the LC250 purified protein by limited papain digestion with 1∶500 ( w∶w ) papain: protein ratio . Products were characterized by N-terminal sequencing and mass spectrometry . The resulting papain resistant fragments had molecular weights between 20 . 7 and 21 . 2 KDa corresponding to the first 175–178 residues of the LACV-L protein . Proteins LC176 , 180 , 183 , 186 and 190 were subsequently produced . Finally , the protein construct LC180 was used for all in vitro biochemical experiments and LC183 for structural studies . The influence of metal ion and DPBA binding on protein stability was measured by Thermofluor assays [19] at a protein concentration of 25 µM in lysis buffer and 2 mM concentration of various metal ions . For nuclease activity experiments , 12 µM of LC180 wild type and mutant proteins were incubated with 12 µM of Alu RNA ( 110 nucleotides of the Alu domain of Pyrococcus horikoshii SRP RNA ) or 15 µM of 51 nucleotides U-rich RNA ( 5′-GGCCAUCCUGU7CCCU11CU19-3′ ) [13] at 37°C in the same buffer . The reaction was stopped by adding EGTA at a final concentration of 12 mM . Divalent cations were added to 2 mM final concentration . The reaction products were loaded onto 8 M urea , 15% acrylamide , Tris-borate gels and stained with methylene blue . ITC experiments were performed using a high-precision VP-ITC titration calorimetric system ( Microcal Inc . , Northampton , MA ) . Binding experiments were performed with 60 µM of freshly purified LC180 protein at 25 C in the same buffer used for the nuclease activity assays . Titrations were made by injecting 15 µl of 1 . 8 mM or 3 mM MnCl2 into the LC180 D52A or wt respectively . For data analysis the heat produced by the metal ion dilution into the buffer was subtracted from the heat obtained in the presence of protein . The same procedure was performed with up to 12 mM of MgCl2 but gave no interaction signal . The binding isotherms were analyzed by non-linear least squares fitting ( Microcal Origin software ) using models corresponding to a single site or two independent sites for the D52A and the wt respectively . Thermodynamic values given are the average and standard deviation of at least two experiments . Proteins LC176 , 180 , 183 , 186 and 190 were expressed and tested for crystallization using a Cartesian nanovolume robotic system for screening . Only LC183 and LC186 crystallised and LC183 was used for all subsequent work . Crystals were obtained by mixing 1∶1 ratio protein: reservoir solution of 15–20 mg/ml LC183 protein in 20 mM HEPES pH 7 . 5 , 150 mM NaCl , 5 mM MnCl2 and 2 . 5 mM β-mercapto-ethanol , and a reservoir composition of 3 . 4 M sodium formate , 0 . 1 M Tris-HCl at pH 8 . The seleno-methionine LC183 crystals were obtained with a reservoir composition of 3 . 6 M Na-formate , 0 . 1 M HEPES pH 7 . The dataset of the inhibitor-endonuclease complex was obtained after an overnight soaking of native crystals into reservoir buffer with 5 mM MnCl2 , 10 mM MgCl2 and 5 mM of DPBA . The crystals were frozen in liquid nitrogen in the reservoir buffer with 30% glycerol for the selenomethionine labelled protein and with 30% glycerol , 5 mM MnCl2 , 10 mM MgCl2 and 5 mM of DPBA for the inhibitor complex . Crystals are of space-group P6122 with four molecules in the asymmetric unit . Selenomethionine derivative data were collected on a 180×160×140 µm3 crystal to 2 . 1 Å resolution on beamline ID29 at the European Synchrotron Radiation Facility ( ESRF ) at the selenium edge ( X-ray wavelength 0 . 979 Å ) for experimental phasing . Native and DPBA data were collected to 2 . 2 Å resolution on ID29 with wavelengths of 0 . 954 Å and 0 . 976 Å respectively . Data were processed and scaled with the XDS package [41] and subsequent analysis performed with the CCP4i package . Statistics of data collection and refinement are given in Table 1 . The structure solution was obtained by the SAD method using autoSHARP [42] which found 16 anomalous sites , four ( including a manganese site ) for each of the four chains in the asymmetric unit . The resultant map was excellent and could be largely built automatically by ARP/wARP [43] . Refinement was performed with REFMAC [44] without applying non-crystallographic symmetry restraints . Extra density was observed for a single Mn2+ ion in the active site of each of the four molecules in the asymmetric unit as confirmed by strong anomalous scattering , even though the X-ray energy was well away from any manganese edge ( Supplementary Figure S1 ) . The loop containing Asp52 is either in the open position or partially open and intermediate . The structure of the complex with the inhibitor was solved by molecular replacement using PHASER [45] and the previously obtained model . Extra density was observed for a second Mn2+ and for the DPBA ( Supplementary Figure S1 ) . The loop containing Asp52 is in the closed position . Sub-confluent monolayers of Huh7 cells in 12-well plates were transfected with 0 . 25 µg each of pLACV-vREN and pCAGGs-T7 , 0 . 4 µg of pTM-LACV L ( wild type or mutants ) and pTM-LACV N , and 0 . 1 µg of pTM-FF-Luc using Nanofectin transfection reagent ( PAA ) . In the negative control , the LACV-L expression plasmid was omitted from the transfection mix . An additional 0 . 2 µg of empty vector pTM1 , or expression constructs pTM1-2APro or pTM1-2APro ( G60R ) were transfected in some experiments , as indicated . After transfection , cells were incubated for 24 h and lysed in 100 µl Dual Luciferase Passive Lysis Buffer ( Promega ) . An aliquot of 20 µl of the lysate was assayed for FF-Luc and Ren-Luc activities as described by the manufacturer ( Promega ) . Structure figures were drawn with Molscript [46] or Bobscript [47] and rendered with Raster3d [48] . Sequence alignments were performed with ClustalW [49] and drawn with ESPript ( http://espript . ibcp . fr/ESPript/cgi-bin/ESPript . cgi ) [50] . Molprobity was used to analyse the quality of the structures ( http://molprobity . biochem . duke . edu/ ) . The native structure of LC183 has wwPDB ID 2xi5 for the coordinate entry and r2xi5sf for the structure factors . The DPBA-bound form of LC183 has wwPDB ID 2xi7 for the coordinate entry and r2xi7sf for the structure factors . | Bunyaviruses are a large family of RNA viruses that include serious human , animal and plant pathogens . The viral RNA-dependent RNA polymerase ( L-protein ) is responsible for replication and transcription of the viral RNA , but apart from its central polymerase domain , it is poorly characterized . Like influenza virus polymerase , bunyavirus L-proteins employ a cap-snatching mechanism to transcribe viral mRNAs , by which host mRNAs are endonucleolytically cleaved as a source of short capped primers . Influenza polymerase endonuclease has recently been located at the PA subunit N-terminus . Here we show biochemically and by crystal structure determination that a similar two-manganese dependent nuclease exists at the N-terminus of La Crosse orthobunyavirus L-protein , whose function is required for cap-dependent transcription . By sequence analysis we show that similar endonuclease signature motifs exist in almost all known segmented RNA , cap-snatching viruses including arenaviruses , bunyaviruses , tenuiviruses and orthomyxoviruses . This suggests that the polymerases of these viruses might share a conserved global architecture with the L-protein being equivalent to a concatenation of the orthomxyovirus PA-PB1-PB2 subunits . We also propose that broad spectrum drugs targeting the endonuclease domain of such viruses could be developed , as exemplified by our structure of the LACV endonuclease complexed with a known influenza endonuclease inhibitor . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"biochemistry/biomacromolecule-ligand",
"interactions",
"biochemistry/biocatalysis",
"virology/viral",
"replication",
"and",
"gene",
"regulation"
]
| 2010 | Bunyaviridae RNA Polymerases (L-Protein) Have an N-Terminal, Influenza-Like Endonuclease Domain, Essential for Viral Cap-Dependent Transcription |
Glioblastomas are deadly cancers that display a functional cellular hierarchy maintained by self-renewing glioblastoma stem cells ( GSCs ) . GSCs are regulated by molecular pathways distinct from the bulk tumor that may be useful therapeutic targets . We determined that A20 ( TNFAIP3 ) , a regulator of cell survival and the NF-κB pathway , is overexpressed in GSCs relative to non-stem glioblastoma cells at both the mRNA and protein levels . To determine the functional significance of A20 in GSCs , we targeted A20 expression with lentiviral-mediated delivery of short hairpin RNA ( shRNA ) . Inhibiting A20 expression decreased GSC growth and survival through mechanisms associated with decreased cell-cycle progression and decreased phosphorylation of p65/RelA . Elevated levels of A20 in GSCs contributed to apoptotic resistance: GSCs were less susceptible to TNFα-induced cell death than matched non-stem glioma cells , but A20 knockdown sensitized GSCs to TNFα-mediated apoptosis . The decreased survival of GSCs upon A20 knockdown contributed to the reduced ability of these cells to self-renew in primary and secondary neurosphere formation assays . The tumorigenic potential of GSCs was decreased with A20 targeting , resulting in increased survival of mice bearing human glioma xenografts . In silico analysis of a glioma patient genomic database indicates that A20 overexpression and amplification is inversely correlated with survival . Together these data indicate that A20 contributes to glioma maintenance through effects on the glioma stem cell subpopulation . Although inactivating mutations in A20 in lymphoma suggest A20 can act as a tumor suppressor , similar point mutations have not been identified through glioma genomic sequencing: in fact , our data suggest A20 may function as a tumor enhancer in glioma through promotion of GSC survival . A20 anticancer therapies should therefore be viewed with caution as effects will likely differ depending on the tumor type .
Tumors are aberrant organ systems that display a complex interplay between neoplastic cells and recruited vascular , inflammatory , and stromal elements [1] . Cellular heterogeneity within the neoplastic compartment has been modeled with complementary stochastic and hierarchical paradigms . Molecular signals that drive tumor formation and maintenance frequently are shared with normal development and wound responses , processes in which normal stem and progenitor cells function [1]–[4] . Stem cell–like cancer cells ( or cancer stem cells ) need not be derived from normal stem cells but may be subjected to evolutionary pressures that select for the capacity to self-renew extensively or differentiate depending on conditions [1]–[5] . Cancer stem cells have been derived from several primary brain tumors , but both their derivation and characterization are incomplete and rapidly expanding [6]–[35] . Glioblastoma ( World Heath Organization grade IV astrocytoma ) is the most common primary brain tumor in adults and one of the most aggressive and deadly cancers [36] , [37] . Current glioblastoma therapies , including radiotherapy and chemotherapy , are highly toxic , offering only palliation [36] , [37] . Although brain tumor stem cells remain controversial due to the evolving understanding of their nature , a number of reports have demonstrated that glioblastomas contain cancer stem cells and that these cells contribute to therapeutic resistance and tumor angiogenesis [1]–[35] . Significant effort has been undertaken to identify potential targets in cancer stem cells that promote tumor maintenance and that might be amenable to disruption [35] . To identify molecular targets in cancer , the majority of analyses completed to date compare bulk tumor to normal tissues and may therefore underestimate the importance of genes and proteins expressed within the cancer stem cell subpopulation . For example , comparison of GSCs to non-stem glioma cells or bulk tumor has resulted in greater understanding of the importance of HIF2α [10] , L1CAM [21] , and Bmi-1 [22] for GSC tumorigenic capacity . These proteins are all relatively overexpressed in glioblastoma stem cells ( GSCs ) and now known to regulate GSC growth , survival , and self-renewal [10] , [21] , [22] . These biological processes are also regulated in GSCs by more well-established molecular mediators of cancer such as c-myc [23] and AKT [24]–[26] . Together , these studies demonstrate that isolation and characterization of GSCs can define new molecular targets for cancer therapy and determine novel roles for established signaling pathways in cancer stem cell biology . We speculated that the cell survival and NF-κB regulator A20 , or Tumor Necrosis Factor α inducible protein 3 ( TNFAIP3 ) , was one molecular target with a greater role in glioma than currently understood [38] , [39] . An oncogenic role for A20 is suggested by the increased A20 expression in some cancers: A20 is elevated in undifferentiated nasopharyngeal carcinoma [40] , poorly differentiated head and neck cancers [40] , gliomas [41] , and inflammatory breast cancer [42] . Increased A20 expression in breast cancer cells confers resistance to TNFα [43] , [44] and tamoxifen [45] , suggesting A20 mediates survival and chemoresistance . Although these data imply a protumorigenic role for A20 , gene expression and functional studies in other cancer types suggest A20 is a tumor suppressor . Deletions and inactivating mutations have been identified in B-cell lymphoma , Hodgkin lymphomas , and non-Hodgkin lymphomas [46]–[51] . In addition , decreased A20 was associated with resistance to DNA-damaging agents in glioma cells [52] . Together , these data indicate that the role of A20 in cancer biology may be context and tissue-type dependent and is an important area for further investigation . Evidence from the A20 null mice also suggested a link to stem cell biology: the epidermal and dermal layers of A20−/− mice are significantly thicker than that of wild-type controls , demonstrating A20 regulates skin cell fate [53] . Based on these data , we examined the role of A20 in cancer stem cell biology in gliomas . We determined A20 is elevated in GSCs where it is required to maintain growth and survival . Importantly , we find increased A20 mRNA expression or copy number is associated with poor glioma patient survival .
To determine whether A20 could differentially contribute to glioma biology through the recently identified glioma subfractions , we utilized several complementary methods to evaluate A20 expression in freshly isolated GSC-enriched and -depleted cultures derived using our previously published methodology [11] , [12] , [15] , [21] , [23] , [25] . To determine whether A20 was differentially expressed at the transcript level , total mRNA was collected and analyzed by quantitative real-time PCR . We found GSC-enriched cells isolated from short-term xenografts expressed elevated A20 mRNA levels in comparison to matched non-stem glioma cells ( Figure 1A ) . We had previously confirmed that these cells self-renew and propagate tumors in an immunocompromised host [11] , [12] , [21] , [23] . As in our past studies , the GSCs also express high levels of the glioma stem cell marker encoded by Olig2 ( Figure 1B ) . Consistent with these results , GSCs isolated directly from patient specimens also expressed elevated levels of A20 and Olig2 ( Figure 1C ) . To determine whether elevated A20 mRNA expression correlated with increased A20 protein levels , we next visualized A20 expression using immunofluorescence ( Figure 1D and 1E ) . Increased expression of A20 was observed in GSCs in comparison to matched non-stem glioma cells ( Figure 1D ) . GSCs form neurospheres when cultured in serum-free media . When single neurospheres were sectioned , we determined A20 was coexpressed with the stem cell transcription factor Sox2 ( Figure 1E ) . To confirm these results , we enriched or depleted GSCs from either short-term xenografts or patient tumor specimens and isolated lysates for immunoblotting ( Figure 1F; Figure S1 ) . In every tumor tested , GSCs displayed strikingly elevated A20 levels compared to matched non-stem cells . Similar to results with mRNA , the differential expression of A20 was true whether subfractions were isolated from patient specimens passaged short term in immunocompromised mice or directly from patient specimens ( Figure 1F; Figure S1 ) . To further evaluate whether cells expressing a GSC marker also highly express A20 at the single-cell level in a quantitative manner , we performed flow cytometric analysis with cells double labeled for A20 and CD133 . We confirmed that A20 is highly coexpressed with the glioma stem cell marker CD133 ( Prominin-1 ) when cells were isolated from a patient specimen passaged short term in immunocompromised mice ( Figure 2A ) or directly isolated from patient specimens ( Figure 2B; Figure S2A and S2B ) . When bulk tumor cells were analyzed for expression of CD133 and A20 , greater than 75% of CD133+ cells were also A20+ , whereas less than 10% of CD133− cells were A20+ ( Figure 2C; Figure S2A and 2SB ) . The percentage of CD133+ cells is also higher in the A20+ subpopulation: greater than 50% of A20+ cells were CD133+ , whereas less than 8% of A20− cells were CD133+ ( Figure 2D; Figure S2C and S2D ) . Coexpression of CD133 and A20 also occurred when cells were cultured short term in vitro after enrichment or depletion of GSCs from a xenografted patient specimen ( Figure S2C and S2D ) . Together , these data strongly support elevated expression levels of A20 in GSCs . Although our findings demonstrated A20 was consistently up-regulated in GSCs , no studies to date have suggested a functional role for A20 in cancer stem cells . As A20 was linked to cell survival in some reports [53]–[57] , we first assessed the ability of A20 to regulate GSC cell growth and apoptosis by targeting A20 expression using lentiviral transduced short hairpin RNAs ( shRNAs ) ( Sigma Mission shRNA ) . To control for potential off-target shRNA effects , two different sequences of shRNA directed against A20 and a nontargeting shRNA were used when cell numbers permitted . Transduction with A20 shRNA reduced A20 protein levels in GSCs in comparison to the nontargeting control , but did not alter Olig2 expression ( Figure 3A and 3B ) . A20 targeting profoundly impacted GSC growth as demonstrated by a marked reduction in cell numbers over time ( Figure 3C and 3D ) . In contrast , A20 knockdown minimally altered the growth patterns of non-stem glioma cells ( Figure S3A and S3B ) . The differential dependence of A20 in GSCs and non-stem cells can be further demonstrated by analysis of the relative effect of A20 shRNA on cell growth ( Figure S3C–S3F ) . To determine whether the decreased growth of GSCs with A20 knockdown was associated with changes in the cell cycle , we performed flow cytometric analysis with DNA content determination . Cellular entry into S phase was decreased with A20 targeting ( Figure 3E ) , supporting a role for A20 in proliferation . We also observed an increase in the percentage of cells in the SubG0 phase of the cell cycle ( Figure 3F ) and a consistent , but relatively modest , 1 . 1–1 . 3-fold increase in G1 phase cell cycle arrest ( unpublished data ) . Thus , knockdown of A20 inhibits cell growth due in part to decreased proliferation associated with increased cell death and cell-cycle arrest . As changes in the cell cycle suggest that A20 regulates cell survival , we evaluated apoptosis with complementary assays . Annexin V assays detect phosphatidylserine expression on the cell surface , a process that occurs during apoptosis and other forms of cell death [5] , [58] . In GSCs isolated from two different human glioma xenografts ( Figure 4A ) and directly from a patient specimen ( Figure 4B ) , introduction of A20-directed shRNA increased the percentage of Annexin V–positive cells when compared to nontargeting control shRNA . Caspases , including caspase 3 and caspase 7 , are cysteine-aspartic acid proteases that are activated during apoptosis [5] , [58] . In GSCs isolated directly from a patient specimen ( Figure 4C ) or from a human patient specimen passaged short term in immunocompromised mice ( Figure 4D ) , caspase 3/7 activity normalized to cell number increased with A20 targeting compared to control . Terminal deoxynucleotidyl transferase dUTP nick end labeling ( TUNEL ) staining detects DNA fragments that occur in the last phase of apoptosis [5] , [58] . Increased TUNEL staining was observed in isolated GSCs with knockdown of A20 ( Figure 4E ) . The potent induction of apoptosis with A20 targeting was restricted to GSCs as minimal effects were observed with decreased A20 expression in matched non-stem glioma cells ( Figure 4B–4D ) . Together , these data demonstrate that targeting A20 in GSCs results in increased apoptosis and suggest that A20 is a prosurvival factor for GSCs . To evaluate the potential mechanism for the decreased survival of GSCs with A20 targeting , we determined activation of p65/RelA , a central component of NF-κB signaling [59]–[63] . Phosphorylation of RelA at Serine 536 is known to enhance the transactivation potential of the NF-κB complex [62] , [63] . NF-κB signals have been shown to be elevated in glioma [64] , and targeting RelA in glioma has been shown to decrease cell growth [65] . We found that knockdown of A20 decreased the activating phosphorylation of RelA without changes in total RelA levels ( Figure 4F ) . These data suggest that failure to maintain active RelA in GSCs contributes , at least in part , to the decreased growth and survival of GSCs with A20 knockdown . Cancer stem cells are functionally defined through their capacity for sustained self-renewal . As the growth and survival of GSCs was affected by A20 knockdown , we next examined whether A20 was important for self-renewal . To more definitively evaluate this possibility , we utilized an in vitro indicator of self-renewal in normal and cancer stem cells: the neurosphere assay [1]–[9] , [66]–[68] . We found that targeting A20 in GSCs decreased neurospheres formation in comparison to cells transduced with nontargeting control shRNA ( Figure 5 ) . A20 loss decreased the percentage of wells with neurospheres in both primary ( Figure 5A–5C ) and secondary ( Figure 5C ) passages . Neurospheres that did form from A20-targeted GSCs were smaller than those forming from nontargeted GSCs ( Figure 5D–5E ) , suggesting decreased proliferation . Thus , the formation of neurospheres is significantly hampered by the loss of A20 , indicating a role for A20 in GSC self-renewal . According to the cancer stem cell hypothesis , GSCs and other cancer stem cells are responsible for tumor maintenance and recurrence after therapy due to the ability of these cells to survive cellular assaults that would typically result in cell death [1]–[5] . For example , GSCs have been shown to preferentially survive radiotherapy [15] , [16] , chemotherapy [17] , [18] , treatment with TNF-related apoptosis-inducing ligand ( TRAIL ) [19] , and Fas-induced apoptosis [20] . These data demonstrate GSCs are resistant to a wide variety of prodeath signals . Prior studies suggested that TNFα therapy may be beneficial for the treatment of glioma due to the ability to promote apoptosis [69] , but the sensitivity of cancer stem cells to this effect has not been determined . As A20 can inhibit TNF-induced apoptosis in some cell types [43] , [44] , [53]–[57] , and A20 levels are elevated in GSCs , we hypothesized that GSCs are resistant to TNFα-induced apoptosis . We therefore investigated the effect of TNFα on the survival of GSC-enriched and -depleted cultures . TNFα increased apoptosis in non-stem glioma cells ( Figure 6A and 6B ) , consistent with prior results in glioma cell lines [70] , [71] . In contrast , GSCs isolated from multiple human glioma xenografts were resistant to TNFα-induced cell death ( Figure 6A and 6B and unpublished data ) . To determine the role of A20 in GSC TNFα apoptotic resistance , GSCs were transduced with either shRNA directed against A20 or nontargeting shRNA and subsequently treated with TNFα . Consistent with our prior results ( Figure 4 ) , GSC apoptosis was increased with targeting of A20 as measured with caspase activity ( Figure 6C; Figure S4A ) . This increase in apoptosis was significantly enhanced when A20 knockdown cells were additionally treated with TNFα ( Figure 6C; Figure S4A ) . To determine whether the TNFα effects could differentially regulate self-renewal in A20 knockdown cells , we evaluated neurosphere formation in the presence and absence of A20 targeting and TNFα treatments . We found that addition of TNFα in the presence of A20 shRNA , but not nontargeting shRNA , significantly decreased neurosphere formation ( Figure 6D; Figure S4B ) . Together , these data indicate that GSCs are resistant to TNFα-induced apoptosis in an A20-dependent manner and further demonstrate A20 is an important prosurvival factor in GSCs . Our results thus far determined an important role for A20 in GSC growth , survival , and self-renewal in vitro , but the ultimate goal of any cancer stem cell–directed therapy is to provide therapeutic benefit in vivo . We therefore evaluated the ability of A20 targeting to increase the survival of immunocompromised mice bearing intracranially implanted human glioma cells . For initial experiments , we performed an in vivo limiting dilution assay with GSCs transduced with nontargeting control shRNA or shRNA directed against A20 . Tumor-bearing mice were allowed to survive until the development of neurologic signs in each animal ( including lethargy , ataxia , paralysis , or seizure ) in accordance with Institutional Animal Care and Use Committee–approved protocols . For all cell numbers transplanted , the median survival of mice injected with GSCs derived from either T3359 ( Figure 7A ) or TB-08-0118 ( Figure 7B ) was increased with A20 knockdown . When 300–1 , 000 cells were injected , the tumor incidence also decreased when A20 was targeted ( Figure 7A and 7B ) . Kaplan-Meier curves further demonstrate significant increases in survival with introduction of A20 shRNA when T3359 ( Figure 7C ) or TB-08-0118 ( Figure 7D ) GSCs were injected . Although cells used in the in vivo studies displayed successful targeting of A20 expression prior to implantation , immunohistochemical analysis of A20 shRNA tumors that grew out showed that these tumors escaped A20 knockdown ( Figure S5 ) . In a separate experiment with T4105 cells in which all animals were sacrificed simultaneously at the onset of the first neurologic sign in any animal , infiltrating glioma cells were observed in animals injected with GSCs infected with nontargeting shRNA , but not A20-targeting shRNA ( Figure 7E ) . These in vivo data indicate that targeting A20 in GSCs can increase survival in mouse models of human brain tumors and suggest A20 could be a useful therapeutic target in glioma . Our data in human glioma patient specimens and xenografts in vitro and in vivo suggested that elevated A20 levels in GSCs are protumorigenic . To extend these results into a clinical analysis , we utilized two databases: the National Cancer Institute's Repository for Molecular Brain Neoplasia Data ( REMBRANDT ) and The Cancer Genome Atlas ( TCGA ) , which respectively contain information from multiple brain tumor types or glioblastoma only . Using REMBRANDT , we found that up-regulation of A20 mRNA 2-fold or greater in all glioma patients correlated with a significant decrease in survival ( Figure 8A ) . When the analysis was restricted to Grade II or Grade III astrocytoma , A20 mRNA up-regulation 2-fold or greater also significantly correlated with decreased survival ( Figure 8B ) . However , A20 mRNA up-regulation did not correlate with survival in glioblastoma patients in either REMBRANDT or TCGA ( Figure 8C and unpublished data ) . These data demonstrate that elevation of A20 in early clinical stages of glioma correlates with poor survival , but differences in A20 expression in the bulk tumor at the time of glioblastoma diagnosis cannot predict survival . As global gene expression analyses currently available in the public databases cannot consider the contribution of the GSC subfraction , these data may underestimate the importance of A20 in glioblastoma . However , it remained possible that an overall elevation of A20 in glioblastoma patients compared to lower grade astrocytomas contributed to the poorer survival of glioblastoma patients . To evaluate this possibility , we compared the median expression intensity of A20 across tumor types in REMBRANDT . We found elevated levels of A20 in glioblastoma compared to all other types of brain tumors as well as control ( nontumor ) tissue ( Figure 8D ) . As elevated levels of mRNA may reflect an increase in gene copy number , we sought to determine whether genomic changes in A20 occurred in glioma patients . Copy number REMBRANDT analysis for A20 indicated that a 3-fold or greater amplification of the 6q23 . 3 chromosomal region correlated with poor survival , although the number of patients in this group was small ( n = 4; Figure 8E ) . It is also important to note that , in contrast to lymphoma [46]–[51] , inactivating point mutations in A20 were not identified in recently completed genetic screens of the glioma genomes [72] , [73] . These data further suggest that A20 activity is important for glioma development and biology . Overall , our data demonstrate that A20 promotes the tumor initiating capacity of GSCs and strongly suggests that increased A20 expression contributes to poor glioma patient outcome . Understanding the genetic and/or protein expression profile of patients in which A20 is elevated may be important for elucidating the mechanisms through which A20 is induced or the pathways through which A20-expressing cells regulate survival . We therefore utilized the TCGA glioblastoma database to evaluate the presence of common glioblastoma mutations ( Figure S6A and S6B ) and expression of TNFα signaling mediators ( Figure S6C ) in patients with different levels of A20 expression . Tumors with TP53 and NF1 mutations were enriched in tumors with high A20 expression in comparison to tumors with intermediate or low A20 levels ( Figure S6A and S6B ) . In contrast , the percentage of samples with EGFR mutations was lowest in tumors with high A20 expression ( Figure S6A and S6B ) . Further evaluation of expression of TNFα signaling mediators in patients with differential A20 levels demonstrated higher TNF receptor and RelB mRNA expression in tumors with elevated A20 expression . These data suggest that A20 expression may be indicative of elevated TNFα or NF-κB signals . As the effects of these two pathways on GSC pro-tumorigenic behaviors are relatively unknown , more thorough evaluation of their biologies in the context of cancer stem cells may be warranted .
Our appreciation of the complex interactions between cell types during tumor initiation , progression , and recurrence continues to grow with our increasing understanding of cancer biology . Whereas researchers once considered tumors to be masses of clonal cancer cells , the involvement of immune cells , endothelial cells , and neighboring fibroblasts to tumor growth is now well recognized [1] . We believe that the identification of cancer stem cells with a probable concurrent stochastic clonality builds upon this model to recognize a previously underestimated contribution of the heterogeneity of tumor cells themselves [1]–[4] . By profiling the molecular and biological properties of cancer stem cells , we may therefore identify genes and proteins whose importance in cancer was poorly recognized . We have now determined that the inhibitor of apoptosis A20 is a cancer stem cell target . A20 was elevated in GSCs in comparison to non-stem glioma cells at both the mRNA and protein levels in cells isolated directly from glioma patient specimens and human glioma xenografts . Targeting A20 expression with shRNA in GSCs significantly impaired their growth and survival in vitro and increased tumor latency in mice bearing human glioma xenografts . The importance of A20 to human glioma patients is further demonstrated by the association of elevated A20 levels with poor outcome . Although current methods for cancer stem cell enrichment from solid cancers have been sufficient to differentiate tumor subpopulations , prospective identification of cancer stem cells has limitations that contribute to the controversy surrounding their existence [1]–[35] . Due to the restricted amount of tissue often available after pathologic review , it is difficult to generate enough patient-derived GSCs for the majority of experiments without culture or amplification as a xenograft . As we believe that microenvironmental conditions within the tumor contribute to GSC maintenance [74] , we utilize a xenograft isolation system to obtain sufficient GSCs , but validated key studies with direct analysis of patient specimens . Experiments with cells directly derived from primary glioma specimens would be optimal , but results in GSCs from xenograft and patient-derived specimens have , thus far , been similar [11] , [15] . Once tissue or xenografts are obtained , enrichment or depletion of cancer stem cells from the bulk tumor can be useful to respectively isolate glioma stem cell and non-stem glioma cell fractions with the stipulation that both GSC and non-stem glioma populations are heterogeneous and unlikely to be mutually exclusive based on current sorting protocols [1]–[4] , [27] . Fluorescence-activated cell sorting ( FACS ) sorting to enrich for glioma stem cells has relied on the presence of the glycosylated form of the cell surface marker CD133 ( Prominin-1 ) : a protein whose role in tumorigenesis and glioma biology remains unclear [27] . However , CD133 is not the only marker useful for prospective identification of glioma stem cells , not all CD133+ cells are GSCs , and CD133 cannot exclusively segregate for tumorigenic potential and self-renewal in all glioma patient samples and cell lines studied [27]–[33] . The side population associated with ABCG2 transporter activity has been shown to be an important segregator for tumorigenic potential in mouse and human gliomas [24]–[33] . In gliomas without CD133 expression , the carbohydrate antigen SSEA-1/CD15/LeX can enrich for tumor-initiating cells [32] . CD133 and SSEA-1 separations are based on the use of antibodies against unspecified carbohydrate modified epitopes , adding further complexity by suggesting that posttranslational modifications to cell surface proteins are important for cancer stem cell biology [27] . These data suggest that , similar to leukemias , no one cell surface marker will be sufficient to isolate a homogeneous population of cancer stem cells from solid tumors . It is therefore enticing to suggest that molecular targets , such as A20 , that are elevated in GSC-enriched fractions may segregate for a cancer stem cell subpopulation . Validating our findings in a model without dependence on CD133 , we characterized A20 expression in lines with tumor enrichment in the SSEA-1+ fraction and found increased A20 expression ( Figure S1 ) . These results strongly suggest that A20 segregates with tumor initiating potential . Our data with A20 expression suggest that , regardless of the cell surface markers used for isolation , the tumor-maintaining glioma subfractions may have common intracellular molecular targets . Determining the direct contribution of A20 ( or cells expressing other identified nuclear targets such as HIF2α[11] and Bmi-1 [22] ) is limited by our inability to sort for live A20-positive cells due to its intracellular localization . However , one recent study was able to utilize an indirect reporter based method to further elucidate the role of the transcription factor Oct4 in cancer stem cell biology . When the Oct4 promoter was used to drive green fluorescent protein ( GFP ) expression in osteosarcoma cells , reporter activity could identify tumor-initiating cells [75] . Application of similar methodologies to other cancer stem cell targets could permit the development of non-cell surface marker based sorting techniques and lead to further confirmation of the existence of cancer stem cells . Although cancer stem cells can be isolated using different methodologies , a common theme in cancer stem cell biology is the ability of this tumor subpopulation to survive cellular assaults and repopulate the tumor . To date , only a few molecular mechanisms for GSC resistance to apoptotic signals have been identified . GSC radioresistance is linked to elevated checkpoint activation and DNA repair [15] , whereas chemoresistance is associated with improved drug efflux due to the presence of the ABCG2 transporter [24] . Resistance to TRAIL-induced apoptosis in GSCs may be due to reduced levels of caspase 8 [19] , and GSCs appear to be less sensitive to Fas-induced apoptosis due to decreased levels of oligomeric Fas [20] . Our data now add A20 as one of this growing list of prosurvival mediators in GSCs . We find that knockdown of A20 induces apoptosis in GSCs and sensitizes GSCs to TNFα-induced apoptosis in cell culture , although we have not measured apoptosis specifically in the GSC compartment in vivo . Whether elevated levels of A20 in GSCs could also regulate other forms of therapeutic resistance remains to be investigated , but it is interesting to note that A20 was one of a set of genes identified as mediators of resistance to O6-alkylating agents [52] . Cell lines were derived from primary and recurrent tumors selected for resistance to 1 , 3-bis ( 2-chloroethyl ) -1-nitrosourea ( BCNU ) or temozolomide in media with 20% fetal calf serum [52] . Under this prodifferentiating condition , down-regulation of A20 and the established pluripotency gene leukemia inhibitory factor ( LIF ) were both associated with chemoresistance [52] . Roles for A20 ( our data ) and LIF [34] in GSC self-renewal imply that elevation of these proteins would be more likely to facilitate rather than impede GSC mediated chemoresistance . These data reinforce the notion that the differentiation state of the glioma cells may differentially impact the mechanisms through which tumor cells survive cellular stresses ( i . e . , conditional essentiality ) . Our examination of A20 in the context of glioma heterogeneity revealed A20 contributes to GSC protumorigenic behaviors , but recent evidence in the literature suggests a tumor suppressive role for A20 in other cancers . Although we find A20 is highly expressed in GSCs and point mutations in A20 have not been identified in glioma [72] , [73] , A20 deletion and mutation is prevalent in lymphoma [46]–[51] . A20 knockdown in GSCs promotes apoptosis and reduces tumorigenic potential in mouse models of human cancer , but overexpression of A20 in A20-deficient lymphoma cells produces similar results [46] , [47] . Together , these data suggest that A20 may be a tumor suppressor or a tumor enhancer depending on the cancer type . Many molecular pathways , including the NF-κB pathway which A20 can inhibit , may be either pro- or antitumorigenic depending on the cellular context and tumor stage [59] . For example , inhibition of NF-κB signals in mouse epidermis resulted in squamous cell carcinomas [76] , whereas a similar transgenic strategy in transformed hepatocytes prevented tumor progression to hepatocellular carcinoma [77] . A20 may similarly have differential effects on tumor development or progression depending on the biological requirements for A20 in specific tissues . Indeed , unique roles for A20 in lymphoma and glioma tumor biology may be anticipated based on differences in basal expression between lymphoid tissues and the brain . In the majority of tissues , including brain , A20 expression is induced by a variety of stimuli ( TNFα , lipopolysaccharide , interleukin-1 ) , but unexpressed or expressed at very low levels under basal conditions [53] . However , A20 is constitutively expressed in lymphoid tissues , particularly the thymus and lymph nodes where A20 is critical for suppression of inflammatory responses mediated by NF-κB [53] , [57] . Mutation of A20 may therefore be more beneficial during the development of lymphoma . In contrast , our analysis suggests A20 levels increase with brain tumor grade , suggesting a benefit for A20 elevation in astrocytoma growth and linking A20 to glioma tumor progression . Thus , the precise biological and molecular outcomes of targeting A20 in each tumor type must be further defined , particularly before broadly applying A20-based therapies for cancer treatment . Our data implicate A20 as an important mediator of cancer stem cell biology by demonstrating that A20 is involved in glioma maintenance through the regulation of GSC growth and survival . The increased survival of mice bearing intracranial tumors upon A20 targeting , and the decreased survival of glioma patients with elevated mRNA , both indicate that inhibition of A20 ( or its downstream mediators ) may be beneficial for glioma therapy . However , the increased survival of mice upon A20 restoration to A20-deficient lymphoma cells demonstrates that targeting A20 may be harmful for other tumor types . As we do not fully understand the mechanisms that cause A20 to have differential effects on tumor growth and cancer cell behaviors , further elucidation of A20 molecular and biological signals is warranted .
Primary human brain tumor patient specimens were obtained from patients providing informed consent under protocols approved by the Duke University or Cleveland Clinic Institutional Review Boards . All animal experiments were performed in accordance with a Duke University or Cleveland Clinic Institutional Animal Care and Use Committee–approved protocol . As previously described [11] , [12] , [15] , [21] , [23] , [25] , matched cultures enriched or depleted for GSCs were isolated from primary human brain tumor patient specimens directly or those passaged short term in immunocompromised mice . A Papain Dissociation System ( Worthington Biochemical ) was used to dissociate tumors according to the manufacturer's instructions ( detailed protocol: http://www . worthington-biochem . com/PDS/default . html ) . Cells were then cultured in Neurobasal medium supplemented with B27 without vitamin A , l-glutamine , sodium pyruvate ( Invitrogen ) , 10 ng/ml basic fibroblast growth factor ( bFGF ) , and 10 ng/ml epidermal growth factor ( EGF ) ( R&D Systems ) for at least 6 h to recover surface antigens . Cells were then labeled with an allophycocyanin ( APC ) -conjugated CD133 antibody ( Miltenyi Biotec ) , and sorted by fluorescence-activated cell sorting ( FACS ) . Alternatively , cells were separated microbead-conjugated CD133 antibodies and magnetic columns ( Miltenyi Biotec ) . CD133-positive cells were designated as GSCs whereas CD133-negative cells were designated as non-stem glioma cells . Consistent with previously defined methods for GSC and non-GSC cell culture [10] , GSCs were cultured in the earlier-defined medium: matched non-stem glioma cells were cultured for at least 24 h in 10% serum containing DMEM to allow cell survival . After recovery , DMEM medium was removed and the cells cultured in supplemented Neurobasal medium for at least 12 h before experiments were performed in identical medium . The cancer stem cell nature of the CD133-positive cells was confirmed by fluorescent in situ hybridization ( FISH ) analysis , serial neurosphere assays , and tumor formation assays , but cultures depleted of cancer stem cells did not self-renew and or initiate tumors ( unpublished data ) . Total RNA was prepared using the RNeasy kit ( Qiagen ) , and reverse transcribed into cDNA using a SuperScript III First-Strand Synthesis Kit ( Invitrogen ) . To investigate expression of A20 and Olig2 , individual gene primers were ordered from Integrated DNA technologies and Master Mixes were purchased from SuperArray Bioscience Corporation . mRNA levels were measured using an ABI-7900 system ( Applied Biosystems ) . Sequences for primer sets were as follows: A20: Forward 5′-AGT GTT CCC AGG TGG CCT TAG AAA-3′; Reverse 5′-TCT CAG CCA AGA CGA TGA AGC AGT-3′ . Olig2: Forward 5′-GGT AAG TGC GCA ATG GTA AGC TGT-3′; Reverse 5′-TAC AAA GCC CAG TTT GCA ACG CAG-3′ . Cells , neurospheres , or tumor sections were fixed with 4% paraformaldehyde , washed with Tris-buffered saline , and incubated with polyclonal mouse anti-A20 ( Santa Cruz Biotechnology ) and goat anti-Sox2 ( Santa Cruz Biotechnology ) where indicated . Primary antibodies were incubated for 16 h at 4°C followed by detection with Alexa 488 donkey anti-mouse ( Invitrogen ) and Alexa 568 donkey anti-rabbit ( Invitrogen ) secondary antibodies . Nuclei were stained with Hoechst 33342 ( Invitrogen ) , and slides were mounted using Fluoromount ( Calbiochem ) . Images were taken with a Leica SP-5 confocal microscope . Equal amounts of cell lysate were resolved by SDS-PAGE , transferred to polyvinylidene difluoride membranes ( Millipore ) , and detected using an enhanced chemiluminescence system ( Pierce Biotechnology ) with antibodies against A20 ( Abcam or Santa Cruz Biotechnology ) , Olig2 ( R&D Systems ) , and Tubulin ( Sigma ) . Tumors were dissociated as described earlier and bulk or isolated GSCs , and non-stem glioma cells were fixed in 4% paraformaldehyde and subjected to FACS analysis . FACS analysis was performed on a FACS Aria with 100-µm nozzle and low sheath pressure . Human-specific anti-CD133 ( 293C3 ) conjugated to allophycocyanin5 ( APC ) ( Miltenyi ) was used with anti–A20-PE generated using the Lightning-Link PE kit ( Innova Biosciences ) in combination with an A20 antibody ( Abcam ) . Lentiviral shRNA clones ( Sigma Mission RNAi ) targeting A20 and a scrambled nontargeting control ( SHC002 ) were purchased from Sigma . These vectors were cotransfected with the packaging vectors psPAX2 and pCI-VSVG ( Addgene ) or the ViraPower Lentiviral Expression System packaging mix ( Invitrogen ) into 293FT cells by Lipofectamine 2000 ( Invitrogen ) to produce the virus . Efficiency of different lentiviral shRNA clones in cells was determined by Western blot analysis and real-time PCR . The sequence of the shRNAs utilized for shRNA1 ( NM_006290 . 2-635s1c1 ) , shRNA2 ( NM_006290 . 2-2104s1c1 ) , and shNRA3 ( NM_006290 . 2-957s21c1 ) was as follows: 5′-CCGGCACTGGAAGAAATACACATATCTCGAGATATGTGTATTTCTTCCAGTGTTTTTG-3′ and 5′-CCGGGAAGCTCAGAATCAGAGATTTCTCGAGAAATCTCTGATTCTGAGCTTCTTTTTG-3′; and 5′-GTACCGGGATGAAGGAGAAGCTCTTAAACTCGAGTTTAAGAGCTTCTCCTTCATCTTTTTTG-3′ . GSCs infected with lentivirus expressing the indicated shRNAs for 24 h were plated in 96-well plates at 1 , 000 cells per well . Cell titers were determined after the indicated number of days after plating using the CellTiter-Glo Luminescent Cell Viability Assay kit ( Promega ) . GSCs plated in six-well plates at 100 , 000 cells per well were infected with lentivirus expressing the indicated shRNAs for 48 hours . To determine the percentage of cells in each phase of the cell cycle , cells were fixed with ethanol and stained with propidium iodide followed by cell-cycle analysis . To detect apoptotic cells , Annexin V-FITC staining was performed with the Annexin V-FITC Apoptosis Detection Kit ( BD Pharmingen ) according to the manufacturer's instructions . For experiments in which GSCs and matched non-stem glioma cells were treated with TNFα , cells were plated at a density of 100 , 000 cells per well in a six-well plate and treated for 72 h with 5 ng/ml TNFα . GSCs infected with lentivirus expressing the indicated shRNAs for 24 h were plated at 1 , 000 cells per well and caspase 3/7 activity measured with a commercially available kit ( Promega ) after an additional 24 h . Relative caspase activity was then determined by correcting for cell titers determined as indicated above . When TNFα treatment was combined with shRNA treatments , cells were infected with lentivirus expressing the indicated shRNAs for 24 h followed by TNFα treatment for 24 h . GSCs infected with lentivirus expressing the indicated shRNAs for 36 h were stained for TUNEL using an Apo-BrdU-Red In Situ DNA Fragmentation Assay Kit ( Biovision ) according to the manufacturer's instructions . GSCs infected with lentivirus expressing the indicated shRNAs for 24 h were plated in 24-well plates at 10 cells per well and the percentage of wells containing neurospheres quantified at indicated times . For secondary sphere formation , neurosphere forming cells from the first plating were trypsinized and plated at 10 cells per well in 24-well plates . Neurospheres were imaged with an Olympus CK40 digital camera mounted to a light microscope and neurosphere size was calculated using ImageJ software . When TNFα treatment was combined with shRNA treatments , cells were infected with lentivirus expressing the indicated shRNAs for 24 h followed by treatment with 5 ng/ml TNFα treatment . Intracranial transplantation of GSCs into nude mice was performed as described [11] , [12] , [15] , [23] , [25] in accordance with a Duke University or Cleveland Clinic Institutional Animal Care and Use Committee approved protocol . Briefly , 36 h after lentiviral infection , cells were counted and the indicated number of live cells implanted into the right frontal lobes of athymic nude mice . Mice were maintained until the development of neurological signs . Significance was tested by t-test or ANOVA using GraphPad InStat 3 . 0 software . For repeated measures ANOVA and in vivo studies where Kaplan-Meier curves and log-rank analysis were performed , MedCalc software was used . | Glioblastomas are the most common and aggressive primary brain tumors in adults , with a median survival of only 12–15 months . Glioblastomas display a cellular hierarchy with a subset of cells having stem cell–like properties , including the capacity to self-renew and propagate tumors . Specific ablation of cancer stem cells is widely thought to be critical for effective and long-lasting treatment of cancers . We report the identification of the antiapoptotic protein A20 ( which is also known as TNFAIP3 ) as a novel regulator of glioma stem cell survival . Glioma stem cells overexpress A20 relative to non-stem glioma cells , and this protects them from cell death , whereas depletion of A20 attenuates glioma stem cell survival and tumor growth . Interrogation of a molecular glioma database reveals that A20 levels correlate with decreased survival in patients . These data indicate that A20 is a tumor enhancer in the context of glioma , which importantly contrasts with its known function as a tumor suppressor in the context of lymphoma . Therefore , A20 may be a context-specific regulator of cancer stem cell survival and growth . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"neurological",
"disorders/neuro-oncology"
]
| 2010 | Targeting A20 Decreases Glioma Stem Cell Survival and Tumor Growth |
Ryanodine receptor type 1 ( RyR1 ) produces spatially and temporally defined Ca2+ signals in several cell types . How signals received in the cytoplasmic domain are transmitted to the ion gate and how the channel gates are unknown . We used EGTA or neuroactive PCB 95 to stabilize the full closed or open states of RyR1 . Single-channel measurements in the presence of FKBP12 indicate that PCB 95 inverts the thermodynamic stability of RyR1 and locks it in a long-lived open state whose unitary current is indistinguishable from the native open state . We analyzed two datasets of 15 , 625 and 18 , 527 frozen-hydrated RyR1-FKBP12 particles in the closed and open conformations , respectively , by cryo-electron microscopy . Their corresponding three-dimensional structures at 10 . 2 Å resolution refine the structure surrounding the ion pathway previously identified in the closed conformation: two right-handed bundles emerging from the putative ion gate ( the cytoplasmic “inner branches” and the transmembrane “inner helices” ) . Furthermore , six of the identifiable transmembrane segments of RyR1 have similar organization to those of the mammalian Kv1 . 2 potassium channel . Upon gating , the distal cytoplasmic domains move towards the transmembrane domain while the central cytoplasmic domains move away from it , and also away from the 4-fold axis . Along the ion pathway , precise relocation of the inner helices and inner branches results in an approximately 4 Å diameter increase of the ion gate . Whereas the inner helices of the K+ channels and of the RyR1 channel cross-correlate best with their corresponding open/closed states , the cytoplasmic inner branches , which are not observed in the K+ channels , appear to have at least as important a role as the inner helices for RyR1 gating . We propose a theoretical model whereby the inner helices , the inner branches , and the h1 densities together create an efficient novel gating mechanism for channel opening by relaxing two right-handed bundle structures along a common 4-fold axis .
Maintaining a precise intracellular Ca2+ concentration that is 10 , 000-fold lower than the surrounding environment of the cell , and the ability to dramatically increase intracellular calcium to trigger downstream events in response to specific stimulus are key for cell survival [1] . Ryanodine receptors ( RyRs ) are high-conductance intracellular Ca2+ channels regulated by both exogenous and intracellular mediators , which release Ca2+ stored in the endoplasmic reticulum . RyRs are the largest ion channels known , with an average molecular weight of 2 . 26 MDa , with most of its mass ( ∼4/5 ) forming the cytoplasmic domain . The skeletal muscle isoform , RyR1 , has a bidirectional interaction with the slow voltage-gated calcium channel in the cell membrane , or dihydropyridine receptor ( DHPR ) , which acts as RyR1′s voltage sensor for cell membrane depolarization [2] . Two key questions to understand RyR1′s function are how are signals transmitted from peripheral cytoplasmic domains to the ion gate , and what is the gating mechanism itself . Cryo-electron microscopy ( cryoEM ) and single-particle image analysis of frozen-hydrated RyR1 revealed the 3D structure of RyR1 at approximately 25–30 Å resolution [3–5] . Its cytoplasmic domain is shaped like a flat square prism of 290 Å side and 120 Å high , with at least 12 reproducible domains that have been assigned numerals 1–12 [6 , 7] . Using the same technique , it has also been possible to map the binding sites for several ligands: the FK506-binding protein 12 kDa ( FKBP12 ) , calmodulin ( Ca2+-CaM , apoCaM ) , and imperatoxin A ( IpTxA ) [8–10] . All these interactions , which are known to modulate RyR1 gating , take place at distances at least 130 Å away from RyR1′s putative ion gate and suggest that RyR1 makes use of long-range allosteric pathways between the cytoplasmic sensing domains and the ion gate . The 3D reconstructions of RyR1 in the open conformation indicated several conformational changes involving both the cytoplasmic and transmembrane domains with respect to the closed conformation [11 , 12]; however , the resolution of these reconstructions ( ∼30 Å ) is insufficient to understand the connection between the two domains or to distinguish the substructure within the transmembrane domain itself . The wealth of atomic structures of K+ channels solved by X-ray crystallography obtained in the last decade has allowed extensive study of the structural rearrangements underlying ion gating for this channel family . In the prevalent model for the ion gating of the K+ channel , the inner helices bend outwards around their midpoint ( through a Gly or a Pro-X-Pro hinge ) to increase the diameter of the ion gate so that it becomes permeable to ion flow . These inner helices are connected to their sensing domains using a plethora of structural arrangements to respond to a variety of effectors ( voltage , ion concentration , pH , redox state , small molecules , and ligands ) . However , with one exception [13] , models of K+ channel gating have been deduced from comparison of unrelated K+ channels . The only other case in which structural data at near-atomic detail is available for both the open and closed states in the same channel is for the nicotinic acetylcholine receptor ( nAChR ) , as determined by electron crystallography [14] . Unlike other K+ channels , the nAChR is a pentamer with its ion gate formed by a hydrophobic girdle in the middle of the membrane . Binding of acetylcholine induces a rotation in protein chains that communicates to the inner helices of the pore , resulting in modulation of the ion gate diameter . To date , nothing is known about the ion gating mechanism of RyR1 . Using cryoEM , we previously defined the architecture of RyR1′s transmembrane domain in the closed state at higher detail [7] . RyR1′s closed ion pore is defined by an axial structure formed by two sets of four rods each forming a right-handed bundle , which we defined as the inner helices , and the inner branches . The inner helices shape the core of the transmembrane assembly . The inner branches are in the center of the cytoplasmic assembly and are directly connected with the peripheral cytoplasmic domains . The two bundles converge into a ring of high density , which we presumed to be the ion gate . A second constriction , which would correspond to the selectivity filter , is on the sarcoplasmic reticulum ( SR ) luminal side of the transmembrane assembly . These two constrictions define a central cavity . The structure formed by RyR1′s inner helices in the closed state appears to be parallel to the canonic structure of the inner helices of closed K+ channels [15 , 16] . A second group of investigators has also reported the presence of the inner helices in the core of the transmembrane assembly of RyR1 in the closed state , achieving similar resolution using the same method and almost identical biochemical conditions [17] . Intriguingly , the conformation that they reported for the inner helices corresponded best to that of an open K+ channel , and suggested that the ion gating mechanism used by RyR1 must be radically different than that used by K+ channels . To better understand the basis for RyR1′s gating and to solve the controversy on the conformation of the inner helices in the closed state , we sought to obtain the open and closed conformations of RyR1 in their ( frozen ) hydrated state using single-particle cryoEM . Furthermore , we used single-channel biophysical characterization of the two states in bilayer lipid membranes ( BLMs ) using identical samples and conditions , to have a more direct correspondence between conformation and biophysical state of the channel . Here , we present the first demonstration , to our knowledge , of a midlevel resolution 3D model for the open state of RyR1 bound to its accessory protein FKBP12 . A 3D reconstruction of RyR1-FKBP12 in the closed state was obtained in parallel for comparison . Thus , in this study , we are able to directly compare both conformations of the same protein , rather than comparing related proteins . Furthermore , both structures correspond to the protein in its fully hydrated state , and both are correlated directly to a functionally characterized biophysical state . We found that upon opening , the cytoplasmic domain undergoes an overall conformational change that involves the connections with the transmembrane domain . In the transmembrane assembly , we find that the inner helices corresponding to the open and closed states of RyR1 have a high cross-correlation with parallel structures of K+ channels in the corresponding state . Nevertheless , the ion pathway of RyR1 has features not present in K+ channels , which has allowed us to create a novel heuristic model for RyR1′s ion gating .
To obtain the resolution necessary for the visualization of secondary structure ( ∼9 Å ) , it is critical to obtain a highly homogeneous dataset . Obtaining a homogeneous population of RyR1 in the closed state is relatively easy . By contrast , the typical flickering behavior of RyR1 under physiologic activating conditions represents a significant limitation , since it produces a mixed population of open and closed states , e . g . , under maximum Ca2+ activating conditions ( 50 μM Ca2+ on the cytoplasmic side ) , the channel open probability ( Po ) of reconstituted purified RyR1-FKBP12 channels is less than 30% ( unpublished data ) . Our previous studies using vesicles demonstrated that the neuroactive noncoplanar polychlorinated biphenyl 2 , 2′ , 3 , 5′ , 6-pentachlorobiphenyl ( PCB 95 ) had an unprecedented activating effect on RyR1 [18 , 19] , suggesting that it would be a candidate small molecule to stabilize RyR1′s open state . The BLM studies of reconstituted purified RyR1-FKBP12 channels indicate that PCB 95 stabilizes the full open ( conducting ) state in ten out of ten reconstituted channels , resulting in extremely long-lived openings interspersed with rare short-lived transitions to the closed state . This results in a mean Po of 0 . 96 and thus produces a highly homogeneous dataset ( Figure 1C–1E ) . By contrast , addition of 2 mM EGTA to the cis chamber ( pCa2+ < 108 ) after fusion of an actively gating channel completely stabilized the fully closed state of the channel with no gating transitions observed for the entire recording period ( Po = 0 ) ( Figure 1A , 1B , and 1E ) . High-affinity [3H]ryanodine binding experiments query the conformational state of a large number of RyR1s [20] . The presence of EGTA ( 2 mM ) in the assay buffer negates specific binding of [3H]ryanodine because the channels are in a closed conformation . By contrast , the presence of PCB 95 and optimal Ca2+ produced nearly 10 pmol of binding sites per milligram of SR protein ( ∼35 , 000 disintegrations per minute [dpm]/25 μM of SR protein ) at steady state ( Figure 1F ) , indicative of the fact that the channels are stabilized in the open state . These biophysical and biochemical data provide two independent measures of the ability of PCB 95 to stabilize the open state of the RyR1 channel having a unitary current level indistinguishable from a native channel in the full open state . The unitary current is a fundamental parameter for any given channel [21] , thus it is safe to assume that the PCB 95–stabilized RyR1 has a pore structure representative of the native open state ( in which only the kinetic/thermodynamic parameters have been altered ) . To exert its effect , PCB 95 requires that RyR1′s FKBP12 accessory subunit be bound [22] . In vivo , FKBP12 is constitutively bound to RyR1 and is known to stabilize its fully closed state and minimize subconductance states [23 , 24] . Both the position and orientation of FKBP12′s atomic coordinates with respect to RyR1 have been mapped and have been shown not to alter RyR1′s closed-state conformation at 16 Å resolution [9] . Our RyR1 purification method [7] produced a single band on PAGE ( Figure 2A ) indicative of a pure RyR1 preparation , and RyR1s with well-preserved structure when viewed with cryoEM ( Figure 2B ) . The relatively high concentration of RyR1 , approximately 2 mg/ml , enabled the successful cryo-preparation of RyR1 suspended over holes instead of lying on a carbon support , a method that allows increased resolution of the 3D reconstruction because it considerably increases the randomness of orientations [7] . CryoEM and image processing of two frozen-hydrated RyR1-FKBP12 datasets corresponding to the open and closed states , with approximately 17 , 000 particles each , yielded two 3D reconstructions . The homogeneous angular distribution for both datasets ( Figure 2C ) indicates that all orientations are equally represented in both datasets; thus the two 3D reconstructions have isotropic resolution and are free of the missing-cone artifact [25] . The nominal resolution of the two reconstructions , 10 . 2 Å , was determined by Fourier shell correlation ( FSC ) using a cutoff criterion of 0 . 143 [26] ( Figure 2D ) , which in this study was a conservative value relative to the five times noise-correlation cutoff . The resolution value of 10 . 2 Å appears reasonable , taking into account the fact that in general , positive identification of secondary structure is indicative of 9 Å or better resolution . We have focused our analysis on only those structures readily visible in the cryoEM density map without any further manipulation . Specifically , we have centered our study on structures with densities at least 2 . 8 σ levels above the mean value . When comparing the 3D reconstructions corresponding to closed and open RyR1 , they look rather similar ( Figure 3 ) . However , careful analysis reveals that they are different conformomers of the same molecule . The coarse conformational changes may be better appreciated when the two 3D reconstructions are filtered to lower resolution and directly superimposed ( Figure 4A ) , or when the two 3D structures alternate between the closed and open states ( Video S1 ) . Whereas most of the domains appear to move , the largest conformational changes take place in the distal regions of the cytoplasmic domains . The larger conformational changes are also evident in the 3D difference maps ( Figure 4B ) . The difference was performed in both directions ( closed minus open , and open minus closed ) , providing the regions of mass that were exclusive for the closed and open states , respectively . Because the open- and closed-state datasets were processed in parallel , starting from a common low-resolution structure , and result in clearly different conformations , we believe that these are genuine representations of the two physiological states . Furthermore , given the large dimensions of the RyR1 ( e . g . , 30× larger than the K+ channel KcsA ) , domains separated by more than 100 Å may be regarded as resolved independently from each other . Yet , these domains follow the same direction of movement when they are connected by intervening density . Finally , for each domain that moved , there is a pair of complementary differences ( see Figure 4B ) , which is also indicative of high data quality and actual movement . The largest-magnitude conformational change occurs in the cytoplasmic domain , whereby each of the quadrants swivels outwards . The corners or clamp domains ( domains 9 and 10 ) together with the structure formed by domains 7 , 8 , and 8a move away from the T-tubule and towards the SR membrane by approximately 8 Å . Concomitantly , domain 2 , more central and facing the T-tubule , moves approximately 4 Å towards the T-tubule , and outwards away from the 4-fold axis ( Figure 4A ) . We do not see an opening of the clamp domains in the open state as was suggested previously ( see Discussion ) . Domain 6 , protruding towards the T-tubule , moves approximately 5 Å outwards when the channel is in the open state , and a similar magnitude of outward movement takes place at domain 11 , facing the SR membrane . The main effect of this swiveling movement is that the mass moves from the center to the outside , making the 4-fold axis less crowded . This movement in the cytoplasmic regions is clearly conveyed to the inner branches ( Figure 3A and 3C , Video S1 ) . In the closed state , the overall structure of the inner helices and inner branches of RyR1-FKBP12 display a structure almost identical to the structure of the closed state of RyR1 that we determined previously in the absence of FKBP12 [7] ( compare Figures 3B , 3D , and 5A ) . As in our earlier report , the inner helices have a tepee-like arrangement that overlaps directly with the tepee structure described for the ion pathway in the atomic models of K+ channels [15 , 16 , 27–29] ( e . g . , see Figure 6 ) . Although a resolution of 9 Å or better is needed to visualize α helices [30 , 31] , it has been described that resolution of 10 Å or even less may suffice to identify α helices , if they are separated from surrounding structures [32] . Another report of the closed state of RyR1 at 10 Å [17] also indicated four inner helices in the same location—although in a different configuration—supporting this finding ( Figure 5B ) . When compared to our closed-state reconstruction , the inner branches in the open state are clearly recognizable but in a different conformation , and the central passage has significantly lower density than in the closed state ( stereo pairs shown in Figures 3B and 5A ) . The inner branches and inner helices define three main constrictions along the 4-fold axis , represented in Figure 7 . The upper , or cytosolic , constriction is defined by the distal enlargement of the inner branches ( Figure 7A ) . The meeting point between the inner branches and the inner helices defines the ion gate ( Figure 7B ) . The lowest constriction defines the opening to the SR lumen ( Figure 7C ) , and is formed by the pore helices in a region that corresponds to the selectivity filter of the K+ channels . The inner helices , the ion gate , and the putative selectivity filter surround the central cavity ( see Figures 3 , 8B-c , and 8D-c ) . In agreement with our previous 3D structure of RyR1 [7] , the transmembrane assembly of both new 3D reconstructions reveals at least six distinct regions of high density per subunit that can be attributed to α helices ( Figure 8 ) . These rod-like structures have a density >3 σ above the mean and are clearly differentiated from their surroundings; they are identified as red contoured regions in Figure 8 . There is a remarkable similarity between the arrangement of all six α helices of the mammalian voltage-dependent shaker channel Kv1 . 2 in the open state and the putative RyR1 transmembrane α helices in the same condition ( see Results and Figure 9 ) . For the purpose of comparison , we designate the putative α helices of RyR1 as R1–R6 , where R6 is the inner helix . These are named according to the comparably positioned α helices of the K+ channel ( S1–S6 ) . We compared the diameter of the ion gate of our open/closed RyR1-FKBP12 with that of previous 3D determinations of RyR1 in the closed state [7 , 17] . Furthermore , taking into account that the diameter of the K+ and Ca2+ ions is very similar , around 4 Å , we compared the diameter of the ion gate of RyR1 with that of the K+ channels and nAChR in open/closed conformations that have been determined at atomic resolution . To accomplish this for RyR1 , we measured the diameter of the ion gate at a density threshold corresponding to the secondary structure ( Figure 11 ) . The diameter of the ion gate of our closed RyR1-FKBP12 and closed RyR1 [7] 3D reconstructions is 8 Å , whereas the diameter of the ion gate of the closed RyR1 obtained at similar resolution in the same conditions by another group is 15 Å [17] . The diameter we find here for the RyR1-FKBP12 ion gate in the open state is 12 Å . From the known atomic structures of K+ channels and nAChR , we took the equivalent measurement , defined by the inner edge of the inner helices , and find that in the closed state , their ion gate diameters range between 7–8 Å ( closed K+ channels , which are tetramers ) and 10 Å ( closed nAChR , which is a pentamer ) . For all of the open channels , the diameter is 12–13 Å [29 , 33–35] . Taken into this general context , our measurement of 8 Å for RyR1′s closed ion gate falls within the values found for the closed conformations , and the measurement of 12 Å for RyR1′s open ion gate corresponds to that found for the open conformations of the known K+ channels ( Figure 11 ) . When the side chains of the K+ channel's atomic model are taken into consideration , the actual diameter of the closed pore is 4 Å [15] . Thus , it is likely that when atomic resolution of RyR1′s structure is obtained , our 8 Å diameter will result in similar pore dimensions , which is an appropriate conformation for a closed Ca2+ channel . Likewise , the observed increase to 12 Å diameter in the open state should be sufficient to enable Ca2+ flow . When the atomic structure of open Kv1 . 2 [35] is superimposed on the open RyR1 density map , the positions of the α helices of Kv1 . 2 , S1–S6 , correlate well with high-density regions of RyR1 ( Figure 9 ) . Starting from the 4-fold axis , we assign S6 , the four inner α helices of the K+ channel , to the four central rod-like structures ( inner helices ) in RyR1 ( R6 , see Figures 8D-b through 8D-d and 9B–9D ) . The tips of the pore helices in Kv1 . 2 also overlay those of RyR1 ( p in Figures 8D-d and 9D ) . Four rod-like structures that are in the same position as the outer helices of the K+ channel ( S5 ) can be identified in the region of RyR1′s transmembrane domain proximal to the lumen ( R5 ) ( see Figures 8B-d , 8D-d , and 9D ) . We suggest that they are the putative outer helices , or R5 . The S1–S4 helices form the voltage sensor of Kv1 . 2 . Although RyR1 does not have known voltage-sensing activity , we observe that S1–S4 , which form the voltage sensor in Kv1 . 2 , overlap with the corners of the transmembrane assembly of RyR1 . Two densities in RyR1 , R1 and R3 , are in a similar configuration to S1 and S3 , although slightly farther away from the 4-fold axis ( Figures 8D-c , 8D-d , 9B , and 9D ) . R2 , a weaker density , matches with S2 , and the intervening density between R3 and R5 , indicated as R4 in Figure 8D-d , could correspond to S4 . At the level of the ion gate , R6 continues to overlap with S6 , and the horizontal rod-like density 1 ( h1 ) of RyR1 overlaps with the S4–S5 linker structure ( Figures 7B , 8D-b , and 9B ) . The h2 structure coincides with an α helical structure of unknown sequence in the Kv1 . 2 atomic model [35] ( see u in Figure 9A superimposed on the open RyR1 density map ) . Despite the structural similarity , we could not find sequence homology between the transmembrane segments of Kv1 . 2 and the aliphatic segments of RyR1 . In contrast with Kv1 . 2 , two other atomic models of K+ channels with six transmembrane α helices per subunit [29 , 34] do not match well with our cryoEM density map . The region of discordance in these atomic models is the S1–S4 formation; however , this could well be the result of the presence of the Fab/Fv fragments against the voltage sensors that were needed for crystallization .
A previous 30 Å resolution reconstruction of RyR1 was prepared in conditions designed to represent the open conformation ( 100 μM Ca2+ , 100 nM ryanodine ) [11] . This reconstruction indicated that in going from the closed to the open state , the protein undergoes several conformational changes: a counterclockwise rotation of the transmembrane domain with respect to the cytoplasmic assembly , an elongation of approximately 10 Å of the overall structure in the 4-fold axis direction , an opening of the clamp domains between domains 9 and 10 , and an increase in pore diameter from 0 to approximately 18 Å . Two other 3D reconstructions of RyR1 at similar resolution were prepared to represent fully and transiently open states ( 100 μM Ca2+ , and 100 μM Ca2+ plus 1 mM AMP-PCP , respectively ) [12] also indicated an opening of the clamp domains between domains 9 and 10 . In these cases , no elongation along the 4-fold axis was observed . Due to the limited resolution , the pore diameter was highly threshold-dependent and in a range between 0–7 Å diameter . Many of these features are not compatible with our observations , and because of the low resolution of these reconstructions , some genuine structural differences were likely to have been confounded by effects resulting from low or anisotropic resolution . In our current reconstructions and a previous closed-state reconstruction [7] of RyR1 , all at around 10 Å resolution , the only connection from domain 10 to the rest of the structure is domain 9 , making it impossible for the clamp domains to “open” during gating by separating domains 10 and 9 [11 , 12] . Their observed gap is likely to be a consequence of the lower resolution and the threshold nonequivalence between the open state and closed states as it is known that the choice of threshold in low-resolution reconstructions dramatically affects the surface representations . Second , the elongation in the z direction that they observed , but was not observed in our reconstruction , is likely due to averaging of the central domains moving away from the transmembrane assembly and the peripheral domains moving toward it . In addition , the missing-cone artifact [25] , whereby a large proportion of 4-fold views with respect to side views , could provoke an artifactual elongation along the 4-fold axis . Last , the diversity of dimensions for the open pore and the much lower resolution in the previous open-state 3D reconstructions [11 , 12] do not warrant a comparison of pore dimensions . In trying to elucidate the molecular mechanism for ion gating using cryoEM , we have found that upon channel opening , structural changes in the cytoplasmic domains are coordinated with structural changes in the ion gate . All domains appear to move in an orchestrated manner , resulting in a significant lowering of density along the 4-fold axis of the protein and an increase of the ion gate diameter . The most obvious connection that we can see between changes at the cytoplasmic domains and ion gate opening is how the upward and outwards movement of the cytoplasmic domains pulls the inner branches in that same direction . Because the inner branches are directly connected to the ion gate , it is straightforward to see how their being pulled apart increases the diameter of the ion gate . RyR1′s large cytoplasmic domain interacts with several proteins such as the voltage sensor ( DHPR ) , FKBP12 , CaM , and IpTxa , and all four affect RyR1′s gating . In intact skeletal muscle , RyR1 appears to open exclusively under the control of the DHPR . Removal of FKBP12 or addition of IpTxa is known to induce subconductance states , whereas CaM modulates the Ca2+ dependence of RyR1′s probability to open . The binding sites for FKBP12 , CaM , and IpTxa have been mapped by cryoEM and 3D difference mapping [8–10 , 49 , 50] , and in all cases , they bind at least 130 Å away from the ion gate ( positions of FKBP12 and Ca2+-CaM binding sites are indicated in Figure 4A ) . We suggest that the conformational changes associated with gating that we have found here are very likely to be the same as the long-range allosteric pathways that convert remote signals sensed through protein/peptide/small molecule–protein interaction in RyR1′s cytoplasmic domains into the appropriate response ( e . g . , the probability of RyR1′s ion gate to open ) . By superimposing the open/closed 3D reconstructions , one can observe regions of density displacement near regions that remain almost stationary . This indicates the presence of structural hinges , i . e . , boundaries between regions of RyR1 that move with different breadths . The two more noticeable regions where this takes place are the crevice near domain 4 , and the one between domains 5 , 9 , and 3 ( Figure 4A ) . Interestingly , these hinges correspond to previously mapped binding sites . The crevice near domain 4 is the target for IpTxa and Ca2+-CaM [10 , 50] . Likewise , the intersection between domains 5 , 9 , and 3 , constitutes the FKBP12 binding site [9 , 10] . Thus , it appears that the hinges may constitute regulatory sites where binding of a relative small effector could produce optimal effect . It has been previously reported that the dimension of the closed ion gate in a 9 . 6 Å reconstruction of RyR1 is 15 Å [17] . This is surprising because it is almost twice the size of the pore we have found in our 10 . 2 and 10 . 3 Å resolution reconstructions of RyR1 in the closed state and 20% larger than the open ion gate reported here ( Figure 11 ) . A pore of 15 Å would leave a large gap that , based on the dimensions of open ion gates for other known cation channels , should not be impermeable to Ca2+ ions . The inner branches were not observed in this 9 . 6 Å 3D reconstruction , and the density in several portions of their putative inner helices is discontinuous ( Figure 5B ) , which raises the possibility that this reconstruction was obtained from a preparation that contained a mixture of open and closed conformations . Such heterogeneity would give a low signal-to-noise ratio in those parts of the structure that change conformation during ion gating . The presence of a low signal-to-noise ratio in their reconstruction required the assistance of helix hunter [51] to identify the putative helices rather than being able to see them directly by increasing the threshold as was done here . There are several portions of their putative inner helices that do not overlap with either our closed- or open-state 3D maps ( compare Figure 5B with Figures 3B , 3D , and 5A ) . Ludtke et al . interpreted their results as meaning that the inner helices of RyR1 in the closed state are more similar to an open than a closed K+ channel conformation . This contradicts our report here in which cross-correlation measurements between our open and closed states and all K+ channels indicated a direct equivalence of physiological state and inner helix conformation ( Figure 6 ) . Finally , the fact that we provide three independent 3D reconstructions supports further the ion gate dimensions and inner helix conformation of open/closed RyR1 , and that they are in a similar range of these reported for the K+ channels . Based on our results , we propose that three structures , the inner branches , the inner helices , and h1 densities , by forming a mobile axial structure , are the three main gating effectors . In the closed state , the two right-handed bundles ( inner helices , inner branches ) form the high-density constriction ( ion gate ) at their meeting point . In going from the closed to the open state , both sets of bundles relax and appear to contribute equally to lowering the density of the ion gate ( see arrows in Figure 3B ) . The h1 densities also contribute to the constricting effect in the closed state and move outwards as the gate opens . The resulting profile of the pore along the ion pathway looks dramatically different as it transitions from the closed to the open state . Such a three-way mechanism appears to constitute a very efficient mechanism to open and close the ion gate and is compatible with the complex regulation of RyR1 through its interaction with the DHPR and other exogenous or intracellular modulators [52] . In summary , we have obtained the 3D reconstructions of the hydrated RyR1-FKBP12 complex both in open and closed conformations . The use of neuroactive PCB 95 [18 , 53] to favor the stability of the full open conformation of the RyR1 channel enabled 3D reconstructions of the ion pathway with high detail . The conformational change of the peripheral cytoplasmic domains is directly related to conformational changes in the transmembrane domain . The architecture of the RyR1 appears to be designed to support precise long-range allosteric pathways such as these involved in efficient coupling with the voltage sensor and in the modulation by ligands such as FKBP12 and CaM . Finally , we have shown that there is a striking similarity between the architectural organization of the transmembrane α helices of the K+ channel family and those of RyR1 . Beyond this similarity , we find that the inner branches , a structure that connects the cytoplasmic domains of RyR1 to the ion gate , appear to play a direct role in ion gating .
[3H]Ryanodine ( [3H]Ry; 60–90 Ci/mmol; >99% pure ) was purchased from Perkin-Elmer New England Nuclear . PCB 95 ( >99% pure ) was purchased from Ultra Scientific . All other reagents were of the highest purity commercially available . RyR1 was purified from rabbit skeletal muscle to concentrations of 2 mg/ml as previously described [7] . Prior to freezing , all RyR1s were incubated with FKBP12 ( Sigma ) at a molar ratio of 8× for 20–40 min . Final buffer conditions to lock the RyR1 into the closed state were 20 mM Na-MOPS ( pH 7 . 4 ) , 0 . 9 M NaCl , 0 . 5% CHAPS , 2 mM DTT , 2 mM EGTA , 5 μg/ml aprotinin , 5 μg/ml leupeptin , and 2 . 5 μg/ml Pefabloc . To set RyR1 in the open state , the same buffer was used except that 10 μM PCB 95 and 50 μM Ca2+ were added and EGTA was excluded . Bilayers were made of phosphatidylethanolamine: phosphatidylserine: phosphatidylcholine ( 5:3:2 w/w , Avanti Polar Lipids ) dissolved in decane at a final concentration of 30 mg/ml . The bilayer partitioned two chambers ( cis and trans ) containing buffer solution ( in mM ) 500 CsCl , 50 μM Ca2+ , and 20 Hepes-Tris ( pH 7 . 4 ) on cis and 50 CsCl , 7 μM Ca2+ , and 20 Hepes-Tris ( pH 7 . 4 ) on trans . The addition of protein was made to the cis solution that was held at the virtual ground , whereas the trans solution was connected to the head stage input of an amplifier ( Bilayer Clamp BC 525C; Warner Instruments ) . Purified RyR1-FKBP12 complexes preincubated for 20–40 min were introduced to cis solution . Upon the incorporation of a single RyR1 channel into BLM , the cis chamber was perfused with cis solution to prevent additional channel incorporation . Single-channel gating was monitored and recorded at a holding potential of −40 mV ( applied to the trans side ) . The sidedness ( cytosolic ) of the channel was verified by the positive response to addition of micromolar Ca2+ and response to 2 μM ryanodine and 5 μM Ruthenium Red ( at the end of the experiment ) . The amplified current signals , filtered at 1 kHz ( Low-Pass Bessel Filter 8 Pole; Warner Instrument , ) were digitized and acquired at a sampling rate of 10 kHz ( Digidata 1320A; Axon-Molecular Devices ) . All the recordings were made for a duration between 12 s and 6 min under each experimental condition . The channel open probability ( Po ) was calculated using Clampfit , pClamp software 9 . 0 ( Axon-Molecular Devices ) without further filtration . Equilibrium measurements of specific high-affinity [3H]Ry binding were determined as previously indicated [20 , 54] . Junctional SR vesicles of rabbit skeletal muscle ( 50 μg protein/ml ) were incubated with or without PCB 95 in buffer containing 20 mM HEPES ( pH 7 . 4 ) , 250 mM KCl , 15 mM NaCl , defined concentration of CaCl2 , and 2 nM [3H]Ry for 3 h at 37 °C . The reactions were quenched by filtration through GF/B glass fiber filters and washed twice with ice-cold harvest buffer ( 20 mM Tris-HCl , or 20 mM Hepes , 250 mM KCl , 15 mM NaCl , 50 μM CaCl2 [pH7 . 4] ) . Nonspecific binding was determined by incubating JSR vesicles with 1 , 000-fold excess unlabeled ryanodine . Each of the conditions was replicated four times in two separate junctional SR preparations , and each of the readings was performed in triplicate or quadruplicate . A 5-μl aliquot of the 2–4 mg/ml RyR1-FKBP12 complex incubation mixture was adsorbed onto a glow-discharged quantifoil holey grid and the excess blotted off with Whatman 540 filter paper . The sample was vitrified by plunging the grid into liquid ethane . CryoEM was performed on a FEI Tecnai F20 FEG microscope operated at 200 kV . Untilted images with defoci between 2 . 5 and 4 . 0 μm were recorded on Kodak SO-163 film under standard low-dose conditions ( dose <10 e−/Å2 ) at a nominal magnification of 50 , 000× . A total of 257 and 233 micrographs for the closed and open states , respectively , were digitized on a Zeiss SCAI scanner at a step size of 7 μm , and subsequently binned down to a final pixel size of 2 . 8 Å . A total of 15 , 625 and 18 , 527 particles for the closed and open states , respectively , were selected interactively using the program WEB . The defocus parameters were determined for every particle using CTFTILT [55] . Individual particles were subjected to a reference-based algorithm starting from an initial 3D model of RyR1 [7] filtered to 40 Å resolution where no substructure is detectable , thus avoiding model bias . Fifty percent of the particles from each dataset with the lowest cross-correlation with the 3D model were discarded . This was followed by several iterations of refinement until the shifts and rotations stabilized . The final number of particles was 9 , 331 and 8 , 133 particles for the closed and open states , respectively . Reference alignment and 3D reconstruction enforcing 4-fold symmetry were performed using the program FREALIGN [56] , which takes account of phase and amplitude contrast transfer function ( CTF ) correction for every particle . This program has implemented a weighting scheme to correct for noise bias , an artifact that could result in an artificial overestimation of the resolution [57] . Resolution values were calculated according to the Fourier shell correlation ( FSC ) curve between two half datasets . The 0 . 143 cutoff [26] was chosen because it was optimistic with respect to the 5 σ noise correction calculated taking into account the 4-fold symmetry ( and thus data redundancy ) of the RyR1 . The final 3D structure of RyR-FKBP12 was normalized and filtered to a resolution of 10 . 2 Å using a B factor of −300 Å3 . The mean and standard deviation values of the volume were calculated within a spherical mask of the same diameter as that used in the iterative alignments . For 3D difference mapping , both 3D reconstructions were filtered to 18 Å resolution and normalized by adjusting the average and standard deviation of densities in both reconstructions to the same level as previously done [9] . Then the open-state RyR1 3D reconstruction was directly subtracted from the closed-state RyR1 3D reconstruction and vice versa . No further data manipulation such as postsubtraction filtering or masking was performed . SPIDER software [58] was used for preparation of the initial volumes , normalization , 3D difference mapping , filtration of the Protein Data Bank ( pdb ) files for comparison with the cryoEM density maps , and calculation of cross-correlation values . Image rendering , docking of atomic structures , and alignment of the other RyR1 3D reconstruction from the database were performed in Chimera [59] ( http://www . ebi . ac . uk/pdbe/emdb/ ) . Both closed-state RyR1 3D reconstructions that have been previously published [7 , 17] are available in the Electron Microscopy Database ( http://www . ebi . ac . uk/msd-srv/docs/emdb/ ) . Hydropathicity , transmembrane propensity , and α helical prediction analyses were performed using several packages available on public servers . The different packages for α helical prediction provided reasonable overlapping sequence segments . The proposed secondary structure is based on the PSIPRED prediction method [60] ( http://bioinf . cs . ucl . ac . uk/psipred/ ) . Electron Microscopy Data Bank ( http://www . ebi . ac . uk/pdbe/emdb/ ) accession numbers for the structures of the RyR1 in closed and open conformations are 1606 and 1607 respectively . | Maintaining a precise intracellular calcium concentration is key for cell survival . In skeletal muscle , ryanodine receptor type 1 ( RyR1 ) is an intracellular calcium-release channel that is critical for contraction . Here , we used single-channel techniques to demonstrate the presence of functionally homogenous populations of RyR1 in either the closed or open state and then applied cryo-electron microscopy and image processing to determine the 3D structure of each state . The 3D structures show that RyR1′s ion pathway is formed by two sets of bundles , each containing four rods along a common axis . One set ( inner helices ) stretches from the lumen to the ion gate , whereas the second ( inner branches ) stretches from the ion gate to the peripheral cytoplasmic domains . The configuration of the two bundles is clearly different in the two physiological states , allowing a 4 Å increase in diameter of the ion gate upon opening . This diameter increase is sufficient to ensure flow of calcium ions . Upon gating , the cytoplasmic domains undergo a conformational change that converges on the inner branches , revealing a long-range allosteric mechanism that directly connects effectors acting on the cytoplasmic moiety with the ion gate . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"computational",
"biology",
"biophysics"
]
| 2009 | Coordinated Movement of Cytoplasmic and Transmembrane Domains of RyR1 upon Gating |
The cytoplasmic polyadenylation element binding protein CPEB1 ( CPEB ) regulates germ cell development , synaptic plasticity , and cellular senescence . A microarray analysis of mRNAs regulated by CPEB unexpectedly showed that several encoded proteins are involved in insulin signaling . An investigation of Cpeb1 knockout mice revealed that the expression of two particular negative regulators of insulin action , PTEN and Stat3 , were aberrantly increased . Insulin signaling to Akt was attenuated in livers of CPEB–deficient mice , suggesting that they might be defective in regulating glucose homeostasis . Indeed , when the Cpeb1 knockout mice were fed a high-fat diet , their livers became insulin-resistant . Analysis of HepG2 cells , a human liver cell line , depleted of CPEB demonstrated that this protein directly regulates the translation of PTEN and Stat3 mRNAs . Our results show that CPEB regulated translation is a key process involved in insulin signaling .
The maintenance of glucose homeostasis requires that organisms respond to changing environmental conditions by balancing pancreatic insulin secretion with the ability of tissues , particularly liver , muscle , and fat , to respond to hormone-induced signaling by importing or secreting glucose [1] , [2] . Obesity and other insults that promote diabetes do so by causing inflammation and insulin resistance , which is characterized by the inability of tissues to transduce insulin/insulin receptor interactions into elevated glucose uptake in muscle and adipose tissue and/or efficient insulin-mediated suppression of gluconeogenesis in liver . In peripheral tissues such as those noted above , insulin association with the insulin receptor ( InsR; NM_010568 . 2 ) induces insulin receptor substrates ( IRS1/2; NM_010570/NM_001081212 . 1 ) tyrosine phosphorylation , which through activated PI3-kinase , promotes the phosphorylation of phosphatidylinositol 4 , 5-bisphosphate ( PIP2 ) to yield phosphatidylinositol ( 3 , 4 , 5 ) -trisphosphate ( PIP3 ) . This process is inhibited by the phosphatase PTEN ( NM_008960 . 2 ) . PIP3 promotes the membrane localization of PDK1 ( NM_008960 . 2 ) and Akt1 ( NM_011062 . 4 ) , where PDK1 and PDK2 phosphorylate Akt1 , resulting in glucose uptake and metabolism . In addition to PTEN , a number of factors can regulate these events . For example , inflammatory pathways have a profound negative impact on insulin signaling; in one instance , Stat3 activation induces Socs3 transcription , which in turn indirectly represses InsR and IRS1/2 activities [3] , [4] . CPEB ( NM_007755 . 4 ) is an mRNA binding protein that controls cytoplasmic polyadenylation-induced translation by interacting with the 3′ UTR cis-acting cytoplasmic polyadenylation element ( CPE ) and three regulatory proteins including Gld2 ( NM_001094423 . 1 ) , a poly ( A ) polymerase; PARN ( NM_001087674 . 1 ) , a deadenylating enzyme , and Maskin ( NM_001088495 . 1 ) , which also associates with eIF4E ( NM_001090548 . 1 ) , the cap-binding factor [5] . Initial CPEB-repression occurs because Maskin binding to eIF4E precludes the eIF4E-eIF4G interaction that is necessary to recruit the 40S ribosomal unit to the 5′ end of the mRNA . Because Maskin is tethered to CPE , only CPE-containing RNAs are repressed . CPEB-induced stimulated ( i . e . , deprepressed ) translation is initiated when CPEB is phosphorylated on S174 or T171 ( species-dependent ) , which expels PARN from the ribonucleoprotein complex . This event allows Gld2 to polyadenylate the RNA by default . The newly elongated poly ( A ) tail is then bound by poly ( A ) binding protein ( PABP ) , which also binds eIF4G; the PABP/eIF4G complex displaces Maskin from eIF4E , allowing for formation of the initiation complex on the 5′ cap structure . By regulating mRNA-specific translation , CPEB profoundly influences gametogenesis and early development [6] , [7] , [8] , neuronal synaptic plasticity [9] , [10] , and cell-cycle progression [11] . CPEB also regulates translation during cellular senescence; fibroblasts derived from Cpeb1 knockout ( KO ) mice do not senesce as do wild type ( WT ) cells , but are immortal [12] . Similarly , human cells depleted of CPEB bypass senescence and have a three-fold extended lifespan [13] . In the mouse cells , aberrant over-expression of myc ( NM_001177353 . 1 ) mRNA is at least one event that mediates the senescence response [12] , while in human cells , reduced p53 ( NM_000546 . 4 ) mRNA translation accounts for the senescence bypass [13] . To identify additional mRNAs that are translationally regulated by CPEB during senescence , we used sucrose density gradient centrifugation to prepare polysomes from WT and Cpeb1 KO mouse embryo fibroblasts ( MEFs ) followed by microarray analysis . Unexpectedly , several mRNAs encoding insulin-signaling molecules ( see below ) were found to be excessively polysomal in the KO versus WT MEFs . Consequently , we investigated the involvement of CPEB in insulin signaling and glucose homeostasis . In Cpeb1 KO liver , muscle , and fat , there was a dramatic and widespread mis-expression of insulin-signaling proteins . Increased expression of two major negative regulators of insulin signaling , PTEN and Stat3 ( NM_213659 . 2 ) , was associated with reduced Akt phosphorylation in both Cpeb1 KO liver and in CPEB-depleted HepG2 human liver cells . When fed on a high fat diet , WT and Cpeb1 KO mice were both obese , but only the KO animals displayed liver insulin resistance . These and other data demonstrate that CPEB control of PTEN and Stat3 mRNA translation is essential for liver insulin signaling and glucose homeostasis .
MEFs derived from CPEB knockout mice do not senesce as do WT MEFs , but instead are immortal [12] . Similarly , primary human skin fibroblasts in which CPEB is depleted also bypass senescence [13] , [31] . Senescence bypass is caused by altered translation of myc and p53 mRNAs , which are directly bound by CPEB . We suspected that additional mRNAs whose translation is controlled by CPEB would also be involved in senescence; to identify them , we performed microarray analysis of mRNAs associated with polysomes in wild type ( WT ) and Cpeb1 KO MEFs ( Figure 1; Table S1 ) . Several of these mistranslated mRNAs control p53-related processes , which given the importance of p53 in senescence , was not unexpected result . However , we also noticed that several aberrantly translated mRNAs encoded proteins involved in insulin signaling , which surprisingly suggested that this signal transduction pathway and perhaps glucose metabolism might be compromised in Cpeb1 KO mice . Consequently , we probed western blots for insulin signaling proteins derived from WT and Cpeb1 KO liver , muscle , and fat . Figure 2A shows dramatic and widespread mis-expression of many insulin-signaling molecules ( reference 8 shows knockout of CPEB gene; Figure S1A shows a western blot of CPEB in WT and KO liver ) . Critical proteins such as PTEN , Stat3 , and Socs3 ( NM_007707 . 3 ) were all elevated in KO liver , muscle , and fat , usually by 2–3 fold . With the exception of Socs3 and IRS2 mRNAs , the levels for the mRNAs encoding these proteins were rarely changed in liver . ( Figure 2B ) . One central insulin-signaling molecule is Akt , which is phosphorylated on S473 and T308 following insulin stimulation; the phosphorylation of these residues is necessary to maintain glucose homeostasis . In fasting WT mice injected with insulin , there was a dramatic increase in liver phospho-S473 and phospho-T308 Akt , but not total Akt , as expected . In Cpeb1 KO mice , however , there was virtually no phosphorylation of Akt irrespective of insulin treatment ( Figure 3A ) . In both fat ( Figure 3B ) and muscle ( Figure 3C ) , AKT was phosphorylated in the KO in response to insulin , indicating that liver is uniquely affected by the lack of CPEB . In the KO liver , phospho-GSK-3β is elevated upon insulin stimulation , albeit not as much as WT ( Figure 3D ) . Phospho-FOXO-1 was also somewhat elevated in KO liver in response to insulin , but significantly less than in WT . In muscle and fat , these and other signaling molecules were largely unaltered by insulin treatment ( Figure S1B ) . It is curious why these two tissues , but not liver , had normal levels of signaling molecules following application of insulin . Perhaps tissue-specific regulatory molecules such as kinases or miRNAs ameliorate or compensate for the lack of CPEB . Taken together , these data indicate that insulin signaling , particularly in the liver , is compromised by CPEB depletion . The data in Figure 2 suggest that the absence of CPEB may may affect the translation of mRNAs that encode insulin-signaling proteins . To assess this possibility , we employed HepG2 human hepatocarcinoma cells that were infected with lentivirus expressing a control shRNA or one against CPEB ( shCPEB ) . The knockdown was confirmed by RT-PCR of CPEB RNA and by down-regulation of ectopically expressed FLAG-CPEB ( Figure 4A , 4B ) . As noted above , Akt phosphorylation was strongly reduced in Cpeb1 KO liver following insulin injection . Depletion of CPEB from HepG2 cells also resulted in reduced insulin-stimulated Akt S473 and T308 phosphorylation ( Figure 4C ) , although not to the extent observed in Cpeb1 KO liver . Ectopic expression of CPEB following shRNA-mediated depletion mostly restored Akt S473 phosphorylation ( Figure 4D; phospho-T308 was not examined ) , which further demonstrates the importance of CPEB for Akt activation . Thus , insulin-treated HepG2 cells depleted of CPEB mimic insulin-injected Cpeb1 KO mouse liver . Finally , we expressed FLAG-tagged CPEB and CPEBΔZF , which lack two zinc fingers that are necessary for RNA binding , in HepG2 cells followed by FLAG immunoprecipitation ( Figure S1C ) and RT-PCR to detect putative target mRNAs . Figure 4E shows that of seven CPE-containing RNAs tested , those encoding Stat3 , PTEN , PDK1 , IRS1 , and Pik3C ( subunit of PI3-kinase ) were co-immunoprecipitated with CPEB but not CPEBΔZF , suggesting that they might be direct targets of CPEB regulation . Because of their central role in insulin signaling and their robust change in KO liver , we focused on PTEN and Stat3 as possible direct targets of CPEB regulation . The 3′ UTRs of both RNAs contain conserved CPEs ( Figure 5A , 5B ) , indicating that CPEB association with these transcripts as demonstrated in Figure 4E is probably direct . Depletion of CPEB from HepG2 cells resulted in elevated amounts of Stat3 and PTEN when examined by western blotting ( Figure 5C , 5D; Figure 5E and 5F show that CPEB depletion had no effect on Stat3 or PTEN mRNA levels ) . However , because western blots reflect steady state protein levels and not necessarily protein synthesis , control and CPEB-depleted cells were pulse-labeled with 35S-methionine for 15 minutes , followed by Stat3 and PTEN immunoprecipitation and SDS-PAGE analysis . Figure 5G and 5H show that de novo labeling of Stat3 and PTEN mimics the western analysis of these proteins , and indicates that their mRNAs are under direct translational control by CPEB . As a control , Figure 5G shows that 35S-methionine labeled tubulin was unaffected by CPEB depletion . To buttress the conclusion that CPEB regulates Stat3 and PTEN at the translational level , the 3′ UTRs of these mRNAs , containing or lacking the CPEs , were appended to firefly luciferase RNA ( Figure 5I ) . These reporters , which together with Renilla luciferase RNA that served as an internal reference standard , were electroporated into control or CPEB-depleted HepG2 cells . The depletion of CPEB resulted in ∼50% increase of firefly luciferase RNA translation when the reporter contained either the Stat3 or PTEN 3′ UTRs . Moreover , nearly identical expression levels were observed when the CPEs were mutated , irrespective of whether the cells contained CPEB ( Figure 5J , 5K , upper graph ) . There was no detectable difference in the stabilities of any of the electroporated RNAs ( Figure 5J , 5K , lower panel ) . Finally , the polysome sedimentation profiles in Figure S1C show that CPEB represses translation of Stat3 and PTEN mRNAs , but not IRS1 or PDK1 mRNAs . Taken together , the data in Figure 5 as well as in Figure S1C demonstrate that CPEB acts directly on Stat3 and PTEN mRNAs to repress their translation . Although the data in Figure 2 demonstrate that the Cpeb1 KO mice have substantial alterations in the levels of insulin signaling proteins , the animals are normal in size and mass . However , because neither of these observations speaks directly to possible changes in glucose metabolism , we subjected the animals to a glucose tolerance test ( GTT ) , which measures glucose clearance from the blood . In this regard , the WT and KO animals were indistinguishable ( Figure 6A ) . On the other hand , serum insulin levels measured during the GTT were significantly higher in the KO mice , which would indicate insulin resistance . To examine this possibility , we performed an insulin tolerance test ( ITT ) . After a 5 hour fast , animals were injected with insulin , which was followed by serum glucose determination 0–60 minutes later . Compared to WT , the KO animals had significantly higher levels of glucose ( Figure 6A ) ; these data , together with those showing that insulin-injected KO animals have very low levels of phospho-Akt , indicate that the absence of CPEB induces insulin resistance . We also examined insulin secretion from isolated pancreatic islets from WT and CPEB KO mice; when challenged with glucose , we detected no statistical difference between the two groups of animals ( data not shown ) . We measured the levels of a number of cytokines and hormones and found that IL6 ( NM_031168 . 1 ) was elevated by ∼5 fold in KO mouse serum ( Figure S2A ) . Increased amounts of this cytokine often correlate with insulin resistance and diabetes [14] . This observation supports the notion that the KO animals have an activated inflammatory Jak-Stat signaling pathway , which is further indicated by enhanced Stat3 phosphorylation and Socs3 expression in the liver ( Figure 2 ) . Because inflammation can exacerbate insulin resistance , CPEB might control the expression of a number of molecules in several tissues to ensure that proper glucose levels are maintained . To determine whether the insulin resistance can be magnified by metabolic insult , WT and Cpeb1 KO mice were fed a high fat diet ( HFD ) for 7 weeks , which elicited obesity in both groups of animals ( whole body mass ) ; there was also no difference in fat or lean mass between the genotypes ( Figure 6B ) . To investigate organ-specific effects of CPEB deletion on insulin action , we performed a 2-hr hyperinsulinemic-euglycemic clamp in conscious WT and KO mice . The steady-state glucose infusion rate to maintain euglycemia during clamps was reduced by ∼50% in the KO animals , although this was not statistically significant ( p = 0 . 197 ) . In addition , insulin-stimulated whole body glucose turnover , glycolysis , and glycogen plus lipid synthesis were unaffected in the Cpeb1 KO mice ( Figure 6C ) . However , hepatic insulin action , expressed as insulin-mediated percent suppression of hepatic glucose production ( HGP ) , was significantly compromised in the Cpeb1 KO mice ( Figure 6C ) . The cause of this decrease in basal HGP is not clear , but because fasting levels of glucose and insulin are not affected by genotype , it may indicate , for example , aberrant gluconeogenesis . Finally , analysis of WT and KO liver following the euglycemic clamp demonstrates that although Akt was strongly phosphorylated in WT animals following insulin administration as expected , this was not the case with the Cpeb1 KO liver ( Figure 6D ) . We also measured fasting glucose and insulin levels in WT and KO mice fed normal chow and a high fat diet ( Figure S2B ) . As expected , glucose levels for both genotypes increased on a high fat diet . Relative to WT , insulin levels in the CPEB KO animals were elevated on normal chow , but did not change further on a high fat diet . These data demonstrate that Cpeb1 KO mice fed a HFD developed defects in insulin signaling and hepatic insulin resistance .
Although the inflammatory response is known to be regulated by 3′UTR binding proteins that affect translation and stability [15] , there is a paucity of information regarding such proteins in insulin resistance or diabetes . Our data indicate that CPEB integrates the expression of several mRNAs involved in insulin signaling . Coordinate posttranscriptional regulation of factors in a given signal transduction pathway , or coordinate regulation of multiple processing steps ( e . g . , splicing , export , localization , and translation ) of a given RNA are two examples of a ribonome regulation ( also sometimes referred to as a regulon ) [16] , [17]; see also [18] ) . Post-transcriptional regulons mediate the inflammatory response [15] , and neurologic disease [19] , [20]; they may do so by , for example , controlling RNA decay [15] , among other events . In addition to networks mediated by RNA binding proteins , others are controlled by miRNAs . For example , miR143/145 ( MI0000257/MI0000169 ) control the expression of several mRNAs to ensure proper smooth muscle development [21] , [22] . Similarly , miR126 ( MI0000153 ) acts on a number of RNAs to mediate angiogenesis [23] . Finally , RNA binding proteins and miRNAs can act together or oppose one another to control the expression of multiple RNAs that define complex cellular states such as neuronal differentiation or epithelial to mesenchyme transition [24] . In the examples noted above , however , no single protein or miRNA controls mRNA expression in a particular signal transduction pathway , rather , a given physiology is modulated at multiple levels ( e . g . , by controlling multiple signaling events ) by certain proteins or miRNAs . In contrast , results reported here show that CPEB regulates the expression of several components of a single signal transduction pathway . Consequently , we propose that an insulin-signaling regulon is controlled at least in part by CPEB . It is remarkable that of the 25 components noted in Figure 7 that influence insulin signaling , mRNAs encoding 21 of them contain CPEs conserved across species . While not all of these mRNAs are likely to be directly regulated by CPEB ( indeed , two of them , IRS2 and Socs3 , do not co-IP with CPEB ) , the preponderance of CPEs suggests that CPEB influences the expression of many of them . Although CPEB was first described as an mRNA stimulatory factor by way of inducing poly ( A ) tail length [25] , [26] , recent evidence shows that it can also repress translation [12] . Presumably , the factors with which CPEB associates in any given cell determines whether it will stimulate or repress translation . In the liver , at least for the messages we examined , CPEB seems to repress translation; when it is not present in KO mice , the synthesis of certain proteins such as PTEN and Stat3 are elevated . However , it should be borne in mind that translational repression is often reversible . It is possible that under some conditions , CPEB would be released from the RNA , which would elicit enhanced translation . Irrespective of how CPEB regulates translation , why would the insulin signaling cascade be controlled by this protein ? We propose that CPEB acts as a rheostat to modulate the levels of insulin signaling proteins in response to particular environmental cues . For example , CPEB activity might be turned up or down in response to a high fat diet , and thereby modulate the degree of insulin sensitivity . If such is the case , then CPEB performance could play a central role in glucose homeostasis .
Male C57BL/6 Cpeb1 KO mice ( 12 weeks old ) ( Tay and Richter , 2001 ) were fed normal chow diet or a high fat diet ( HFD; 55% fat by calories , Harlan Teklad ) for 7 weeks . Mostly littermates were used for all experiments , in some cases mice with +/−1 week of birth date were used . The animals , which were housed in the UMass Medical School animal facility , were used according to guidelines approved the Institutional Animal Care and Use Committee and fully comply with all applicable Federal and State requirements . Mouse CPEB ( WT and ΔZF ) was cloned into a FLAG containing vector ( Nagaoka and Richter , submitted ) . PTEN and Stat3 3′ UTRs ( nucleotides 1–90 and 1–195 , respectively ) were cloned into EcoRI-XhoI sites of pcDNA3 . 1+ vector containing firefly luciferase . In some cases , the CPEs were mutated to C or G in place of T . Renilla luciferase ( pRL-TK; Promega ) was used as a control vector in the luciferase experiments . Antibodies to PTEN ( Cell Signaling ) , Akt1 , and pAkt473 ( Cell Signaling ) were a generous gift from M . Sherman . Antibodies to Socs3 ( Cell Signaling ) and pAkt308 ( Cell Signaling ) were a generous gift from R . Davis . IRS2 and IR antibody was a generous gift from M . White . Antibodies to PDK1 ( Genetex ) , IRS1 ( Upstate ) , tubulin ( Sigma ) . were purchased from the indicated commercial sources . Antibodies to Stat3 , pStat3 , GSK-3 , pGSK-3 , FOXO , pFOXO , pIRS-1 were obtained from Cell Signaling . See Protocol S1 for additional information . Mouse tissues ( liver , fat- white adipose tissue only , muscle ) and HepG2 cells were lysed in buffer ( 50 mM Tris-HCl ( pH 7 . 4 ) , 0 . 25 M NaCl , 1 mM MgCl2 , 0 . 1 mM CaCl2 , 1% NP-40 , 0 . 5% deoxycholate , 0 . 1% SDS ) with protease inhibitor ( Complete , Roche ) and phosphatase inhibitors ( vanadate and fluoride , NEB ) . Western blots were performed with these samples using ECL ( Perkin-Elmer ) , or Femto-ECL ( Pierce ) detection systems . RNP co-immunoprecipitation using Dynabeads ( Invitrogen ) was performed in the same buffer with RNase inhibitor ( RNAseOut , Invitrogen ) and antibody against the FLAG epitope ( Sigma ) . The RNA was extracted from the final precipitate ( or from total RNA ) with Trizol ( Invitrogen ) and subjected to quantitative RT-PCR ( Quantitect RT kit , Qiagen ) . Firefly luciferase RNAs appended with Stat3 or PTEN 3′ UTRs containing or lacking CPEs were synthesized in vitro with T7 RNA polymerase ( mMessage Machine kit; Ambion ) . Control Renilla luciferase RNA containing an irrelevant 3′ UTR was polyadenylated with E . coli polyA polymerase ( NEB ) . Each of the firefly luciferase RNAs together with the Renilla luciferase RNA was used to transfect ( with lipofectamine 2000 ) or nucleofect ( with Amaxa nucleofector ) HepG2 human hepatocarcinoma cells ( ATCC #CRL-10741 ) that were grown to 50% confluency . Some of these cells were also infected with lentivirus harboring control or CPEB shRNA as described by Udagawa et al . ( submitted ) . Infectivity was monitored by GFP , which was encoded by the virus . After infection , fresh DMEM with 10% FBS was added . Luciferase activity was determined 12 hr after cell transduction with a Dual-Luciferase Reporter Assay System ( Promega ) and normalized to the Renilla control . Luminescence was detected with a SAFIRE multimode microplate reader ( Tecan ) . Glucose tolerance ( GTT ) and insulin tolerance tests ( ITT ) were performed using methods described previously [27] . Serum insulin concentrations during the GTT were determined using an insulin ELISA ( Crystal Chem ) . Blood glucose was measured using an Ascensia Breeze 2 glucometer ( Bayer ) . Statistical analysis was performed by ANOVA . Serum adipokines , cytokines , and insulin were measured by ELISA using a Luminex 200 luminometer ( Millipore ) . Hyperinsulinemic-euglycemic clamps were performed at the UMass Mouse Phenotyping Center . Mice were fed normal chow or a HFD for 7 weeks . Whole body fat and lean mass were non-invasively measured using proton magnetic resonance spectroscopy ( 1H-MRS ) ( Echo Medical Systems ) . Following an overnight fast , a 2 hr hyperinsulinemic-euglycemic clamp was performed with a primed and continuous infusion of human insulin ( 150 mU/kg body; 2 . 5 mU/kg/min; Humulin; Eli Lilly ) , and 20% glucose was infused to maintain euglycemia ( Kim et al . , 2004 ) . Whole body glucose turnover and glucose uptake in individual organs were determined by infusion of 3H-glucose and a bolus injection of 14C-2-deoxyglucose during clamps . Following the clamp , tissues were harvested for biochemical analysis [28] . Fibroblasts ( MEFs ) isolated from WT and CPEB KO mice E12 . 5–E14 . 5 embryos ( MEFs ) were cultured in Dulbecco's Modified Eagle's Medium with 10% fetal bovine serum according to a 3T3 protocol [12] . The cells were harvested in the presence of 100 µg/ml cycloheximide ( Sigma ) and subjected to polysome fractionation [30] . The lysate was layered onto a continuous sucrose gradient ( 15%–50% ) and centrifuged for 2 h at 40 000 RPM in a SW41 rotor . The RNA was isolated from polysomal fractions , pooled , and subjected to microarray analysis by UMass Medical School Genomics Core facility with Affymetrix GeneChip Mouse Genome 430A 2 . 0 array . The gradients were performed with biologic triplicates . The data were analyzed with DAVID Bioinformatics Resources 6 . 7 , Panther pathway annotation [29] , [32] , and deposited at http://www . ncbi . nlm . nih . gov/geo/ ( series number GSE28106 ) . Results are presented as mean ± SEM . The error bars in all figures are SEM . A p<0 . 05 was considered significant . We used two-way ANOVA to assess significance ( p<0 . 05 ) . We further performed a pair-wise comparison ( t-test ) between groups . | One major hallmark of diabetes is insulin resistance in peripheral tissues that is controlled at the posttranslational level . For example , insulin activates a kinase cascade that leads to the phosphorylation of Akt , a centrally important molecule that regulates glucose metabolism . In this study , we define a translational regulatory pathway that mediates insulin action in the liver . The Cytoplasmic Polyadenylation Binding Protein ( CPEB ) interacts with mRNA to control translation; knockout mice that lack CPEB exhibit high-fat-diet-induced liver insulin resistance and do so by having aberrant expression of major insulin signaling molecules , in particular PTEN 3 and Stat3 . Our data further suggest that CPEB modulates , in a manner similar to a rheostat , functionally related mRNAs that encode proteins involved in insulin signaling . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"gene",
"expression",
"biology",
"molecular",
"cell",
"biology",
"molecular",
"biology"
]
| 2012 | Cytoplasmic Polyadenylation Element Binding Protein Deficiency Stimulates PTEN and Stat3 mRNA Translation and Induces Hepatic Insulin Resistance |
Mitochondrial DNA is a valuable taxonomic marker due to its relatively fast rate of evolution . In Trypanosoma cruzi , the causative agent of Chagas disease , the mitochondrial genome has a unique structural organization consisting of 20–50 maxicircles ( ∼20 kb ) and thousands of minicircles ( 0 . 5–10 kb ) . T . cruzi is an early diverging protist displaying remarkable genetic heterogeneity and is recognized as a complex of six discrete typing units ( DTUs ) . The majority of infected humans are asymptomatic for life while 30–35% develop potentially fatal cardiac and/or digestive syndromes . However , the relationship between specific clinical outcomes and T . cruzi genotype remains elusive . The availability of whole genome sequences has driven advances in high resolution genotyping techniques and re-invigorated interest in exploring the diversity present within the various DTUs . To describe intra-DTU diversity , we developed a highly resolutive maxicircle multilocus sequence typing ( mtMLST ) scheme based on ten gene fragments . A panel of 32 TcI isolates was genotyped using the mtMLST scheme , GPI , mini-exon and 25 microsatellite loci . Comparison of nuclear and mitochondrial data revealed clearly incongruent phylogenetic histories among different geographical populations as well as major DTUs . In parallel , we exploited read depth data , generated by Illumina sequencing of the maxicircle genome from the TcI reference strain Sylvio X10/1 , to provide the first evidence of mitochondrial heteroplasmy ( heterogeneous mitochondrial genomes in an individual cell ) in T . cruzi . mtMLST provides a powerful approach to genotyping at the sub-DTU level . This strategy will facilitate attempts to resolve phenotypic variation in T . cruzi and to address epidemiologically important hypotheses in conjunction with intensive spatio-temporal sampling . The observations of both general and specific incidences of nuclear-mitochondrial phylogenetic incongruence indicate that genetic recombination is geographically widespread and continues to influence the natural population structure of TcI , a conclusion which challenges the traditional paradigm of clonality in T . cruzi .
Mitochondrial genes are among the most popular markers for the reconstruction of evolutionary ancestries and resolution of phylogeographic relationships [1] . Their pervasive use in population genetics can be attributed to several intrinsic characteristics , notably , their high copy number , small size ( ∼15–20 kb ) and faster mutation rate ( compared with nuclear DNA ) . In addition , their widespread application is founded on the assumptions that mitochondrial genomes are homoplasmic , uniparentally inherited and lack homologous recombination [2] . However , with technological advances affording increased sensitivity and greater sample throughput , a growing number of reports of heteroplasmy ( heterogeneous mitochondrial genomes in an individual cell ) , introgression and inter-molecular recombination are challenging what was previously regarded as a strict set of rules for eukaryotic mitochondrial inheritance . Chagas disease remains the most important parasitic infection in Latin America , where an estimated 10–12 million individuals are infected , with a further 80 million at risk [3] . The aetiological agent , Trypanosoma cruzi , displays remarkable genetic diversity and is currently recognized as a complex of six lineages or discrete typing units ( DTUs ) , each broadly associated with disparate ecologies and geographical distributions [4] . T . cruzi infection is life-long and can lead to debilitation and death by irreversible cardiac and/or gastrointestinal complications [5] . It has been suggested that the geographical heterogeneity in Chagas disease pathology is related to the genetic variation among T . cruzi DTUs [6] , [7] . However , the relationship between parasite genotype and clinical outcome remains enigmatic . DTU nomenclature has recently been revised by international consensus to reflect the current understanding of T . cruzi genetic diversity [8] . Several evolutionary scenarios have been proposed to account for the emergence of two hybrid lineages ( TcV and TcVI ) and their parental progenitors ( TcII and TcIII ) . However , the number of ancestral nuclear clades ( two or three ) remains controversial [9] , [10] . TcI is the most abundant and widely dispersed of all T . cruzi lineages , with an ancient parental origin estimated at ∼0 . 5–0 . 9 MYA [11] . The distribution of domestic TcI , propagated by domiciliated triatomine vector species , principally extends from the Amazon Basin northwards , where it is implicated as the main cause of Chagas disease in endemic areas such as Venezuela and Colombia [12] , [13] . TcI is also ubiquitous in sylvatic transmission cycles throughout South America and extends into North and Central America [14] , [15] . Recent advances in new high resolution genotyping techniques have seen a resurgence of interest in unravelling TcI intra-lineage diversity . In Colombia , sequencing of the mini-exon spliced leader intergenic region ( SL-IR ) has subdivided TcI isolates from domestic and sylvatic transmission cycles , irrespective of geographical origin [16]–[18] . Other studies have demonstrated geographical clustering of TcI strains and an ecological association between specific genotypes and Didelphis hosts [19] . Higher resolution studies exploiting multiple microsatellite markers ( MLMT ) also report limited gene flow between sylvatic and domestic transmission cycles manifesting as genetic diversity between TcI isolates from sympatric sites [20] , [21] . In addition , unexpectedly high levels of homozygosity in multiple clones from single hosts may be indicative of recombination between similar genotypes ( inbreeding ) or recurrent , genome wide , and dispersed gene conversion [20] , [22] . The frequency and mechanism of natural intra-TcI genetic exchange are thus unknown , largely due to inappropriate or inadequate sampling . Evidence for such recombination is increasing and has already been documented among strains isolated from sylvatic Didelphis and Rhodnius in the Amazon Basin [23] and within a domestic/peridomestic TcI population in Ecuador [21] . Furthermore , the generation of intra-lineage TcI hybrids in vitro indicates that this ancestral lineage has an extant capacity for genetic exchange [24] . In kinetoplastids , the mitochondrial genome is represented by 20–50 maxicircles ( 20–40 kb ) which , together with thousands of minicircles ( 0 . 5–10 kb ) , form a catenated network or kinetoplast ( kDNA ) , comprising 20–25% of total cellular DNA [25] . Maxicircles are the functional equivalent of eukaryotic mitochondrial DNA , encoding genes for mitochondrial rRNAs and hydrophobic proteins involved in energy transduction by oxidative phosphorylation [26] . Previously , phylogenetic analyses of T . cruzi maxicircle fragments classified isolates into three mitochondrial clades A ( TcI ) , B ( TcIII , TcIV , TcV and TcVI ) and C ( TcII ) [10] , [27] . To date , maxicircle typing has been principally used to examine T . cruzi inter-lineage diversity , with sequencing efforts reliant on a limited number of genes [28] and often in the absence of any comparative nuclear targets [29] , [30] . However , the inherent features of mitochondrial markers argue for their inclusion as principal but not solitary components of phylogenetic studies . Indeed , the caveats highlighted by other eukaryotes are especially pertinent with respect to T . cruzi . Mitochondrial introgression has been reported in North America where identical maxicircles circulate in sympatric TcI and TcIV from sylvatic reservoirs [27] and in South America where maxicircle haplotypes are shared between TcIII and TcIV strains with highly divergent nuclear genomes [11] . However , this phenomenon has not been described among South American TcI isolates . In addition , mitochondrial heteroplasmy , a possible confounder of phylogenetic studies , has not been examined in the coding region of the T . cruzi maxicircle but is not unexpected considering the presence of up to fifty maxicircle copies within an individual parasite . The potential for mitochondrial DNA to reveal diversity hidden at the sub-DTU level in T . cruzi has been largely overlooked . To address this deficit , we first employed a whole genome approach to investigate the existence of maxicircle heteroplasmy and to resolve its role as a source of genotyping error . Secondly , we exploited the online availability of three complete T . cruzi maxicircle genomes [31] , [32] to develop a high resolution mitochondrial multilocus typing scheme ( mtMLST ) in order to describe TcI intra-lineage diversity . Lastly , we investigated the extent of incongruence between mitochondrial and nuclear loci ( SL-IR , GPI and 25 short tandem repeat ( STR ) loci ) to detect incidences of genetic exchange .
The maxicircle genome from Sylvio X10/1 ( TcI ) was sequenced at 183X coverage using Illumina HiSeq 2000 technology as part of the Sylvio X10/1 Whole Genome Shotgun project [33] . A total of 66 , 882 reads were generated which covered the maxicircle coding region ( 15 , 185 bp ) . The consensus maxicircle genome sequence was derived from the predominant nucleotide present across multiple read alignments at each position . However , this criterion masks minor maxicircle haplotypes ( evidence of heteroplasmy ) by disregarding low abundance single nucleotide polymorphisms ( SNPs ) . To assess the presence/absence of true minor SNPs , all 66 , 882 reads were re-aligned to the Sylvio X10/1 maxicircle genome using the alignment software SAMtools [34] and SNPs were called using the SAMtools mpileup commands . A SNP was defined as a nucleotide variant present in at least 5 independent reads ( with parameters: 20X coverage; and mapping quality , 30 ) . The final alignment was manually inspected using Tablet [35] . In parallel , ten maxicircle gene fragments , described below , were amplified by PCR and Sanger sequenced from Sylvio X10/1 . A panel of 32 TcI isolates was assembled for analysis ( Table 1 ) . Parasites ( epimastigotes ) were cultured at 28°C in RPMI-1640 liquid medium supplemented with 0 . 5% ( w/v ) tryptone , 20 mM HEPES buffer pH 7 . 2 , 30 mM haemin , 10% ( v/v ) heat-inactivated fetal calf serum , 2 mM sodium glutamate , 2 mM sodium pyruvate and 25 µg/ml gentamycin ( Sigma , UK ) [23] . Genomic DNA was extracted using the Gentra PureGene Tissue Kit ( Qiagen , UK ) , according to the manufacturer's protocol . Isolates were previously characterized to DTU level using a triple-marker assay [36] and classified into seven genetic populations by microsatellite profiling [20]: North and Central America ( AMNorth/Cen ) , Venezuelan sylvatic ( VENsilv ) , North-Eastern Brazil ( BRAZNorth-East ) , Northern Bolivia ( BOLNorth ) , Northern Argentina ( ARGNorth ) , Bolivian and Chilean Andes ( ANDESBol/Chile ) and Venezuelan domestic ( VENdom ) . Genotypes for additional TcI–TcVI strains were included for comparison in selected analyses as indicated ( Tables S1 and S2 ) . Ten maxicircle gene fragments were amplified: ND4 ( NADH dehydrogenase subunit 4 ) , ND1 ( NADH dehydrogenase subunit 1 ) , COII ( cytochrome c oxidase subunit II ) , MURF1 ( Maxicircle unidentified reading frame 1 , two fragments ) , CYT b ( cytochrome b ) , 12S rRNA , 9S rRNA , and ND5 ( NADH dehydrogenase subunit 5 , two fragments ) coding regions . Degenerate primers were designed in primaclade [37] using complete maxicircle reference sequences from CL Brener ( TcVI ) , Sylvio X10/1 ( TcI ) , and Esm cl3 ( TcII ) available online at www . tritrypdb . org [38] . Primer sequences and annealing temperatures for PCR amplifications are given in Table 2 . Robust amplification was first confirmed across a reference panel of all six T . cruzi DTUs ( see Table S1 and Figure 1 ) . Amplifications for all targets were achieved in a final volume of 20 µl containing: 1× NH4 reaction buffer , 1 . 5 mM MgCl2 ( Bioline , UK ) , 0 . 2 mM dNTPs ( New England Biolabs , UK ) , 10 pmol of each primer , 1 U Taq polymerase ( Bioline , UK ) and 10–100 ng of genomic DNA . PCR reactions were performed with an initial denaturation step of 3 minutes at 94°C , followed by 30 amplification cycles ( 94°C for 30 seconds , 50°C for 30 seconds , 72°C for 30 seconds ) and a final elongation step at 72°C for ten minutes . PCR products were purified using QIAquick PCR extraction kits ( Qiagen , UK ) according to the manufacturer's protocol . The mini-exon spliced leader intergenic region ( SL-IR ) and glucose-6-phosphate isomerase ( GPI ) were amplified as previously described by Souto et al . ( 1996 ) [39] and Lewis et al . ( 2009 ) [36] , respectively . PCR products were visualized in 1 . 5% agarose gels and if necessary purified using QIAquick PCR and gel extraction kits ( Qiagen , UK ) to remove non-specific products . Bi-directional sequencing was performed for both nuclear and maxicircle targets using the BigDye® Terminator v3 . 1 Cycle Sequencing Kit ( Applied Biosystems , UK ) according to the manufacturer's protocol . Maxicircle PCR products were sequenced using the relevant PCR primers described in Table 2 . Nuclear amplicons were sequenced using their respective PCR primers . When ambiguous sequences were obtained , PCR products were cloned into the pGEM® - T Easy Vector System I ( Promega , UK ) , according to the manufacturer's instructions , and transformed into XL1-Blue E . coli ( Agilent Technologies , UK ) , prior to colony PCR and re-sequencing . For strains that produced incongruent nuclear and maxicircle phylogenetic signals , PCR and sequencing reactions were replicated twice using DNA derived from two independent genomic DNA extractions . Data from 25 previously described microsatellite loci [20] , distributed among ten chromosomes [40] , were included for analysis . Loci were selected from a wider panel of 48 microsatellite loci based on their level of TcI intra-lineage resolution . In addition , these 25 microsatellite loci were amplified across eight new unpublished biological clones ( M16 cl4 , SJM22 cl1 , SJM39 cl3 , USAARMA cl3 , USAOPOSSUM cl2 , 92090802P cl1 , 93070103P cl1 and DAVIS 9 . 90 cl1 ) . Primers and binding sites are listed in Table S3 . The following reaction conditions were implemented across all loci: a denaturation step of 4 minutes at 95°C , then 30 amplification cycles ( 95°C for 20 seconds , 57°C for 20 seconds , 72°C for 20 seconds ) and a final elongation step at 72°C for 20 minutes . Amplifications were achieved in a final volume of 10 µl containing: 1× ThermoPol Reaction Buffer ( New England Biolabs , UK ) , 4 mM MgCl2 , 34 µM dNTPs , 0 . 75 pmol of each primer , 1 U Taq polymerase ( New England Biolabs , UK ) and 1 ng of genomic DNA . Five fluorescent dyes were used to label the forward primers: 6-FAM and TET ( Proligo , Germany ) and NED , PET and VIC ( Applied Biosystems , UK ) . Allele sizes were determined using an automated capillary sequencer ( AB3730 , Applied Biosystems , UK ) , in conjunction with a fluorescently tagged size standard , and were manually checked for errors . All isolates were typed “blind” to control for user bias . Pair-wise distances ( DAS ) between microsatellite genotypes for individual samples were calculated in MICROSAT v1 . 5d [41] under the infinite-alleles model ( IAM ) . To accommodate multi-allelic genotypes ( ≥3 alleles per locus ) , a script was written in Microsoft Visual Basic to generate random multiple diploid re-samplings of each multilocus profile ( software available on request ) . A final pair-wise distance matrix was derived from the mean of each re-sampled dataset and used to construct a Neighbour-Joining phylogenetic tree in PHYLIP v3 . 67 [42] . Majority rule consensus analysis of 10 , 000 bootstrap trees was performed in PHYLIP v3 . 67 by combining 100 bootstraps created in MICROSAT v1 . 5d , each drawn from 100 respective randomly re-sampled datasets . Nucleotide sequences were assembled manually in BioEdit v7 . 0 . 9 . 0 sequence alignment editor software ( Ibis Biosciences , USA ) [43] and unambiguous consensus sequences were produced for each isolate . Heterozygous SNPs were identified by the presence of two coincident peaks at the same locus ( ‘split peaks’ ) , verified in forward and reverse sequences and scored according to the one-letter nomenclature for nucleotides from the International Union of Pure and Applied Chemistry ( IUPAC ) . For both nuclear genes ( SL-IR and GPI ) , edited sequences were used to generate Neighbour-Joining trees based on the Kimura-2 parameter model in MEGA v5 [44] . Bootstrap support for clade topologies was estimated following the generation of 1000 pseudo-replicate datasets . Once both trees were visualized independently to confirm congruent topologies , nuclear SNPs were re-coded numerically and concatenated with microsatellite data ( see Dataset S1 ) . DAS values were calculated for the concatenated dataset as described above and used to generate a single Neighbour-Joining phylogenetic tree encompassing all nuclear genetic diversity . Nucleotide sequences for GPI and the SL-IR are available from GenBank under the accession numbers JQ581371–JQ581402 and JQ581481–JQ581512 , respectively . Sequence data were assembled manually as described for nuclear loci . For each isolate , maxicircle sequences were concatenated according to their structural arrangement ( 12S rRNA , 9S rRNA , CYT b , MURF1 , ND1 , COII , ND4 and ND5 ) and in the correct coding direction ( alignment available on request ) . Nucleotide sequences for all ten gene fragments are available from GenBank under the accession numbers listed in Table 2 . Phylogenies were inferred using Maximum-Likelihood ( ML ) implemented in PhyML ( 4 substitution rate categories ) [45] . The best-fit model of nucleotide substitution was selected from 88 models and its significance evaluated according to the Akaike Information Criterion ( AIC ) in jMODELTEST 1 . 0 . [46] . The best model selected for this dataset was GTR+I+G . Bootstrap support for clade topologies was estimated following the generation of 1000 pseudo-replicate datasets . Bayesian phylogenetic analysis was performed using MrBAYES v3 . 1 [47] ( settings according to jMODELTEST 1 . 0 ) . Five independent analyses were run using a random starting tree with three heated chains and one cold chain over 10 million generations with sampling every 10 simulations ( 25% burn-in ) . Shimodaira-Hasegawa likelihood tests ( SH tests ) [48] were implemented in PAML v . 4 [49] to statistically evaluate incongruencies between alternative tree topologies derived from the mitochondrial and nuclear data .
Across the 15 , 185 bp of the Sylvio X10/1 maxicircle coding region a total of 74 SNPs were identified among eight genes ( 12S rRNA , 9S rRNA , MURF5 , CYT b , MURF1 , MURF2 , CR4 and ND4 ) and three intergenic regions ( between 12S rRNA and 9S rRNA , between 9S rRNA and ND8 and between CR4 and ND4 , respectively ) ( Figure 2 and Table S4 ) . Average read depth for each SNP site was 163 . At heterozygous sites , the minor nucleotide was present among an average of 12 . 2% ( ±9 . 1% ) of sequence reads . In each gene , SNPs were clustered often <5 bp apart in pairs and triplets . The most common mutations were transversions from A→T ( 14/74 ) , T→A ( 10/74 ) , T→G ( 7/74 ) and G→T ( 6/74 ) and transitions from A→G ( 13/74 ) . SNPs were bi-variable at all sites . The presence of different contiguous SNPs distributed across separate sequencing reads at overlapping positions suggests the occurrence of at least two minor maxicircle templates within the same sample . However , the short average length of Illumina reads ( ∼100 bp ) prohibits the full reconstruction of minor maxicircle sequence types . No evidence of heterozygosity was observed in any of the ten maxicircle Sanger sequences ( from the mtMLST scheme ) that covered the corresponding areas of heteroplasmy identified in Sylvio X10/1 , which is consistent with the low sensitivity of this method . Degenerate primers were designed by reference to complete TcI , TcII and TcVI maxicircle genomes . Ten gene fragments from eight maxicircle coding regions were selected in order to sample genetic diversity present across the whole T . cruzi maxicircle . For two genes ( MURF1 and ND5 ) two fragments were selected from each coding region to examine intra-gene variation . Reliable PCR amplification of all ten maxicircle fragments was first confirmed using a panel of T . cruzi reference strains from each DTU ( see Figure 1 ) . The maxicircle gene targets were then sequenced across the TcI panel ( Table 1 ) and seven additional TcIII/TcIV strains ( Table S2 ) . Relatively uniform substitution rates were observed among all genes ( gamma shape parameter α = 0 . 8121 , based on the GTR+I+G model ) . For each TcI isolate , gene fragments were concatenated according to their structural position and assembled into a 3686 bp alignment . Twenty-two unique haplotypes were identified from a total of 355 variable sites ( ∼9 . 6% sequence diversity ) . No evidence of heterozygosity ( ‘split peaks’ ) was observed . Maximum-Likelihood ( Figure 3 , right ) and Bayesian phylogenies were both constructed from the concatenated maxicircle data . No statistically-supported incongruence was observed between the two topologies ( Bayesian tree L = −6770 . 21 , ML tree L = −6768 . 85 , P = 0 . 428 ) . The presence of at least three incongruent haplotypes ( see below ) precludes the accurate clustering of their respective populations ( AMNorth/Cen , VENdom and BRAZNorth-East ) . However , phylogenetic analysis does resolve two well-supported clades corresponding to VENsilv and ANDESBol/Chile ( 90 . 8%/1 . 0 and 100%/1 . 0 , respectively ) . Once the two TcIV-type maxicircles were excluded from analysis , the mtMLST was re-evaluated with respect to intra-TcI discriminatory power . One hundred SNPs were identified among 3681 bp ( ∼2 . 7% sequence diversity ) , corresponding to twenty maxicircle haplotypes . Both Bayesian and Maximum-Likelihood topologies were congruent with those constructed previously for the entire TcI isolate panel . The resolutive power of the mtMLST scheme was evaluated by comparison to current markers used to investigate TcI intra-DTU nuclear diversity , specifically , a housekeeping gene ( GPI ) , a non-coding multi-copy intergenic region ( SL-IR ) and a MLMT panel of 25 loci . Sequences for GPI were obtained for 32 T . cruzi isolates ( Table 1 ) and assembled into a gap-free alignment of 921 nucleotides . Of the 921 bp , a total of 911 invariable sites and 10 polymorphic sites were identified ( ∼1 . 1% sequence diversity ) . A 350 bp alignment corresponding to the SL-IR was generated for the same panel of samples . Strains from two populations ( 5/6 BOLNorth and 4/4 ANDESBol/Chile ) presented sequences with multiple ambiguous base calls due to the presence of a GTn microsatellite at positions 14–24 . For these nine isolates , haplotypes were determined by sequencing four cloned PCR products to derive a consensus sequence . In the 350 bp alignment , 323 conserved sites and 36 polymorphic sites were observed ( ∼10 . 3% sequence diversity ) . All samples were also typed at 25 polymorphic microsatellite loci yielding a total of 1612 alleles . The majority of strains presented one or two alleles at each locus . Multiple alleles ( ≥3 ) were observed at a small proportion of loci ( 1 . 5% ) . Individual Neighbour-Joining trees were re-constructed for GPI , SL-IR and the MLMT data . No well-supported sub-DTU level clades were recovered using GPI sequences . The SL-IR phylogeny resolved two populations ( VENsilv and ARGNorth ) with strong statistical support ( 85% and 99% , respectively; data not shown ) . Three major clades were identified by MLMT ( VENdom , ARGNorth and ANDESBol/Chile ) with good bootstrap support ( 72 . 6% , 99 . 3% and 98 . 4% , respectively; data not shown ) . There was no bootstrap-supported incongruence between the three nuclear tree topologies . This justified their concatenation and these data were re-coded and analyzed in a single distance-based phylogeny ( independent of mutation rate heterogeneity ) ( Figure 3 , left and Dataset S1 ) . The concatenated nuclear tree recovered three well supported clades corresponding to TcI populations ( VENsilv , ARGNorth and ANDESBol/Chile ) ( 96% , 100% and 77 . 9% , respectively , Figure 3 ) . Isolates belonging to the VENdom population remained grouped together but with a minor reduction in bootstrap values ( 64 . 8% ) , compared to the MLMT tree . In addition , the concatenated tree also subdivided BOLNorth into two well defined sympatric clades each containing three isolates ( 99 . 8% and 82 . 2% ) . No nuclear targets ( either individually or concatenated ) were able to reliably identify AMNorth/Cen , or BRAZNorth-East as discrete clusters . However , AMNorth/Cen was more closely related to VENdom than any other population by MLMT ( 90 . 2% ) , the SL-IR ( 99% ) and the concatenated nuclear tree ( 100% ) . Comparison of the mitochondrial and nuclear phylogenies revealed clear incongruence at multiple scales . The nuclear topology was a significantly worse model to fit the maxicircle data ( nuclear tree L = −7008 . 72 , mtMLST ML tree L = −6554 . 50 , P<0 . 001 ) . Three individual isolates had unambiguously different phylogenetic positions between the nuclear and mitochondrial datasets: 9307 , 9354 and IM48 ( Figure 3 ) . The maxicircle sequences from 9307 , a sylvatic TcI AMNorth/Cen strain , and 9354 , a human TcI strain from VENdom , were divergent from all other TcI strains . Comparison with sequences from other DTUs indicates that the maxicircle from 9307 was most closely related to those found in TcIV samples from North America ( 92122 ) ( 100%/1 . 0 ) while 9354 shared its mitochondrial haplotype with TcIV and TcIII strains from neighbouring areas of Venezuela , Bolivia and Colombia ( ERA , 10R26 , X106 , Sairi3 and CM17 ) ( 97 . 8%/0 . 9 ) . IM48 from BRAZNorth-East also had a distinct maxicircle haplotype that formed a long branch separated from the other members of this population whereas for nuclear data all BRAZNorth-East isolates , including IM48 , clearly grouped together . To test whether inclusion of these isolates could explain the overall incongruence , the SH analysis was repeated for alternative nuclear vs . mitochondrial topologies with each of these strains excluded individually and then collectively . In all cases , statistically significant incongruence persisted ( no 9307 P = 0 . 004 , no 9354 P = 0 . 002 , no IM48 P<0 . 001 and without all three P = 0 . 008 ) . This indicated that mitochondrial introgression was generally pervasive in the TcI panel beyond these three isolates . For example , ARGNorth samples , which formed a homogeneous monophyletic clade that was most closely related to ANDESBol/Chile by nuclear data , grouped paraphyletically amongst subsets of BOLNorth strains in the maxicircle tree . In addition , BRAZNorth-East is grouped with one of the BOLNorth clades in the nuclear tree , but receives a basally diverging position in the maxicircle phylogeny . In agreement with the nuclear data , AMNorth/Cen was most closely related to VENdom . However , two isolates from AMNorth/Cen ( ARMA and OPOS ) displayed an unexpected level of maxicircle diversity and are grouped separately with strong bootstrap support ( 96 . 6%/1 . 0 ) .
Elucidating the complex epidemiology , phylogeography and taxonomy of T . cruzi requires a clear understanding of the parasite's genetic diversity [4] . One objective of this study was to develop the first mitochondrial ( maxicircle ) multilocus sequence typing scheme ( mtMLST ) to investigate T . cruzi intra-lineage diversity and to critically assess its resolutive power compared to the current repertoire of phylogenetic markers . The presence of intra-strain maxicircle diversity within Sylvio X10/1 is the first demonstration of heteroplasmy in the coding region of a T . cruzi maxicircle genome . Seventy-four variable sites were identified by read depth analysis of Illumina sequence data but undetected by conventional Sanger sequencing . These SNPs indicate the occurrence of at least two additional maxicircle genomes , present at a ∼10-fold lower abundance compared to the consensus published Sylvio X10 maxicircle genome [32] . Most heteroplasmic SNPs were linked . This may indicate an older most recent common ancestor ( MRCA ) between the major and minor maxicircles than that expected to have emerged in culture post-cloning . Thus these minor maxicircle classes more likely represent heteroplasmy within a single parasite than within a subpopulation of cells . Furthermore , the presence of SNPs <3 bp apart on contiguous sequence reads may have non-synonymous coding implications , although their relative rarity , and a lack of indels suggest that minority and majority maxicircle variants would not differ phenotypically . Finally , the presence of heteroplasmy at less than 0 . 5% of sites indicates it is unlikely to represent a major source of typing error when using maxicircle Sanger sequencing to characterize isolates . Several factors are likely to contribute to mitochondrial heteroplasmy . Mutation in length or nucleotide composition and/or bi-parental inheritance in genetic exchange events are both exacerbated by differential replication rates and inequitable cytoplasmic segregation of mitochondrial genomes during mitosis [50] , [51] . In kinetoplastids , maxicircle intra-clone diversity in the non-coding region was previously reported in both T . cruzi [31] and Leishmania major [52] , [53] . In addition , an earlier study attributed a change in T . cruzi maxicircle gene repertoire ( elimination of one of two heteroplasmic ND7 amplicons ) to sub-culture [54] . However , biologically cloned samples were not used and the possibility of a mixed infection was excluded on the basis of only four microsatellite loci . Sylvio X10/1 ( a biological clone produced by micromanipulation ) was first isolated from a Brazilian patient in 1979 [55] and has been in intermittent sub-culture ever since . The retention of minor maxicircle classes in Sylvio X10/1 for over thirty years suggests that a heteroplasmic state in T . cruzi is naturally sustained . The observations that T . cruzi mitochondrial heteroplasmy is not present at sufficient levels to adversely disrupt phylogenetic reconstructions stimulated the development of the mtMLST scheme and its assessment against traditional nuclear targets . Initially , three types of nuclear marker were evaluated , each characterized by different rates of evolution . Unsurprisingly GPI was highly conserved across TcI and lacked sufficient resolution to discriminate between isolates . The slow accumulation of point mutations at housekeeping loci , which are generally under purifying selection , renders these targets more appropriate to describe inter-DTU variation . Thus they are valuable candidates for inclusion in traditional nuclear MLST schemes [56] . The mini-exon SL-IR is widely used as a TcI taxonomic marker in view of its heterogeneity and ease of amplification [57] . In this study , SL-IR variability manifested as a ten-fold increase in sequence diversity as compared to that of GPI , and supported the robust delineation of two nuclear populations ( VENsilv and ARGNorth ) . However , there are several caveats associated with the SL-IR , notably the presence of multiple tandemly-repeated copies with undefined chromosomal orthology between strains [58] . Previous attempts to estimate the level of intra-isolate SL-IR diversity have reported >96% homology between copies [19] . However , only ten clones were sequenced from each sample , representing less than 10% of the ∼200 copies present per genome . Recent observations of substantial variation in gene copy number and chromosomal arrangement between T . cruzi strains further discourage the use of such targets for taxonomy [59] . In addition , numerous indels in the SL-IR prevent the sequencing of a suitable outgroup [39] and multiple ambiguous alignments , introduced by the microsatellite region , can disrupt phylogenetic signals [60] . Ultimately both GPI and the SL-IR suffer from the same fundamental criticism that single genes are inadequate to infer the overall phylogeny of an entire species [61] . Recombination , gene conversion and concerted evolution have all contributed to the genealogical history of T . cruzi [62] but remain undetectable using single loci . The 25 microsatellite loci afforded the highest level of resolution from an individual set of markers , defining three statistically-supported groupings ( VENdom , ARGNorth and ANDESBol/Chile ) . Their superior performance compared to GPI and the SL-IR is expected considering microsatellites are neutrally-evolving , co-dominant and hypervariable with mutation rates several orders of magnitude higher than protein-coding genes [63] . However , the use of these markers is not devoid of limitations . Most importantly , microsatellites are particularly sensitive to homoplasy , a situation where two alleles are identical in sequence but not descent , and thus fail to discriminate between closely related but evolutionarily distinct strains [64] . The three nuclear markers ( GPI , SL-IR and microsatellites ) were concatenated based on the assumption that no robust incongruence was observed between individual phylogenetic trees . However , concatenating these data did not have a significant additive effect on the level of resolution , with just three populations ( VENsilv , ARGNorth and ANDESBol/Chile ) emerging as well-supported groups . Importantly this dataset did reveal a subdivision in the BOLNorth group , which went undetected by all individual nuclear markers . Gross incongruence between the mtMLST and nuclear phylogenies revealed two incidences of inter-DTU mitochondrial introgression , indicative of multiple genetic exchange events in T . cruzi . Introgression was detected in North America , where identical maxicircles were observed in sylvatic TcI and TcIV isolates . A 1 . 25 kb fragment ( COII-ND1 ) of this TcIV maxicircle haplotype has been previously described in other TcI samples from the US states of Georgia and Florida [11] , [27] . On the basis of the limited nuclear loci examined , and in line with previous work [27] , only TcI derived nuclear genetic material appears to have been retained in these hybrids . The genetic disparity between North and South American TcIV isolates , coupled with their geographical and ecological isolation [65] , implies that this event most likely occurred in North/Central America . A second , independent novel mitochondrial introgression event was identified in a Venezuelan clinical isolate . This TcI strain ( 9354 ) shares its maxicircle haplotype with a subset of human and sylvatic TcIV and TcIII isolates from Bolivia , Venezuela and Colombia , consistent with a local and possibly recent origin . Presumably TcIV , a known secondary agent of human Chagas disease in Venezuela , is a more likely donor candidate than TcIII , which is largely absent from domestic transmission cycles [4] . Nonetheless , evidence of homogeneous maxicircle sequences in multiple , geographically dispersed isolates from different transmission cycles implies the occurrence of several genetic exchange events . It is conceivable that the TcIV/TcIII-type maxicircle sampled in this study is a relic from a TcI antecedent , supporting a common ancestry between TcI , TcIII and TcIV [9] . Alternatively , this haplotype may have originated from a TcIV or TcIII strain and its distribution reflects a recent unidirectional backcrossing event into TcI . Introgression is a more parsimonious explanation than the retention of ancestral polymorphisms through incomplete lineage sorting , particularly in areas of sympatry or parapatry among DTUs [66] . However , the historical diversification of TcI [67] and TcIII [68]–[70] , driven by disparate ecological niches [71] , and the current separation between most arboreal and terrestrial transmission cycles of TcIV and TcIII , respectively , challenge the likelihood of secondary contact between these lineages , a prerequisite of introgressive hybridization . Resolving the donor DTU of this event is complicated by the presence of indistinguishable mitochondrial sequences and paradoxically divergent nuclear genes in TcIII and TcIV isolates . It is unclear whether this results from a mechanism acting to homogenize maxicircles while allowing nuclear genes to slowly deviate [11] ( unlikely ) , repeated and recurrent backcrossing ( more likely ) , or merely reflects the relative paucity of available TcIV and TcIII genotypes for comparison ( a certainty ) . Regardless of the underlying mechanisms , it is clear that genetic exchange continues to influence the natural population structure of T . cruzi TcI . In this study , the failure to detect reciprocal transfer of nuclear DNA using an array of loci readily demonstrates the importance of adopting an integrative approach , complementing traditional nuclear markers with multiple mitochondrial targets . In the absence of comparative genomics , it is impossible to establish whether mitochondrial introgression is entirely independent of nuclear recombination . Another advantage of the mtMLST scheme is its ability to reveal cryptic sub-DTU diversity . The significantly different evolutionary histories of the nuclear and maxicircle genes from members of BOLNorth and ARGNorth are consistent with intra-lineage recombination . The low levels of diversity observed within this incongruent maxicircle clade are indicative of recent and possibly multiple exchange events . In addition , two divergent maxicircles from AMNorth/Cen have also exposed a level of diversity that conflicts with earlier reports of reduced genetic differentiation in this group resulting from their recent biogeographical expansion [18] , [72] . Furthermore , the incongruent basal phylogenetic position of most of BRAZNorth-East in the maxicircle tree as well as the presence of another divergent maxicircle in one isolate ( IM48 ) from this population highlights the extent to which intra-lineage diversity can be neglected by other genotyping methods . The phylogenetic placement of IM48 suggests it may be the product of an intra-TcI introgression event . However , IM48 is also a geographical outlier within the BRAZNorth-East population and it is difficult to determine the origin of this maxicircle haplotype in the absence of additional isolates from West-Central Amazonia . The mechanisms governing maxicircle genetic exchange and the origins of heteroplasmy observed in Sylvio X10/1 are debatable . Currently , all reported maxicircle inheritance in natural [11] and experimental T . cruzi hybrids [24] is uniparental . However , the demonstration of heteroplasmy in this study suggests that , following genetic exchange , any minor maxicircle genotypes may be undetectable using conventional sequencing techniques . In addition , evidence of bi-parental transmission of both maxicircles [73] , [74] and minicircles [75] in experimentally-derived T . brucei hybrids indicates that this phenomenon can occur in kinetoplastids as a result of recombination . The mechanism of genetic exchange in T . cruzi [24] differs from meiosis , which is observed in T . brucei [73] , [76] . Current data suggest in vitro recombination in T . cruzi may be analogous to the parasexual cycle of Candida albicans where nuclear fusion creates a tetraploid intermediate , followed by genome erosion and reversion to aneuploidy [24] , [77] , [78] . It is not implausible to suggest that the process of cell fusion and nuclear re-assortment may be accompanied by asymmetrical kinetoplast distribution to progeny cells . Furthermore , the sequence redundancy observed among minicircle guide RNAs has been postulated to allow biparental inheritance to occur with no detrimental consequences to mitochondrial RNA editing and hybrid viability [79] . Most importantly , the phenotypic implications of mitochondrial heteroplasmy and introgression in T . cruzi are unknown . Maxicircles play a fundamental role in parasite metabolism and development in the triatomine bug vector . Therefore the relationship between genetic recombination and phenotypic heterogeneity may have important implications for disease epidemiology . mtMLST presents a valuable new strategy to detect directional gene flow and examine the dispersal history of T . cruzi at the transmission cycle level . Furthermore , mtMLST is an excellent tool to identify genetic exchange between closely related isolates in conjunction with nuclear MLMT data . By adopting a combined nuclear and mitochondrial approach , one can simultaneously address local , epidemiologically important hypotheses as well as robustly identify parasite mating systems . Thus in combination with adequate spatio-temporal sampling , we strongly recommend this methodology as an alternative to exclusively nuclear or mitochondrial population genetic studies in future work with medically important trypanosomes . Finally , the level of resolution that the mtMLST method provides should greatly facilitate attempts to elucidate the relationship between specific parasite genotypes and phenotypic traits relating to Chagas disease pathology . | Chagas disease , caused by the protozoan parasite Trypanosoma cruzi , is an important public health problem in Latin America . While molecular techniques can differentiate the major T . cruzi genetic lineages , few have sufficient resolution to describe diversity among closely related strains . The online availability of three mitochondrial genomes allowed us to design a multilocus sequence typing ( mtMLST ) scheme to exploit these rapidly evolving markers . We compared mtMLST with current nuclear typing tools using isolates belonging to the oldest and most widely occurring lineage TcI . T . cruzi is generally believed to reproduce clonally . However , in this study , distinct branching patterns between mitochondrial and nuclear phylogenetic trees revealed multiple incidences of genetic exchange within different geographical populations and major lineages . We also examined Illumina sequencing data from the TcI genome strain which revealed multiple different mitochondrial genomes within an individual parasite ( heteroplasmy ) that were , however , not sufficiently divergent to represent a major source of typing error . We strongly recommend this combined nuclear and mitochondrial genotyping methodology to reveal cryptic diversity and genetic exchange in T . cruzi . The level of resolution that this mtMLST provides should greatly assist attempts to elucidate the complex interactions between parasite genotype , clinical outcome and disease distribution . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"genome",
"evolution",
"parasite",
"evolution",
"biology",
"genomics",
"microbiology",
"genetics",
"and",
"genomics",
"parasitology"
]
| 2012 | Multiple Mitochondrial Introgression Events and Heteroplasmy in Trypanosoma cruzi Revealed by Maxicircle MLST and Next Generation Sequencing |
Epstein Barr virus ( EBV ) infection expands CD8+ T cells specific for lytic antigens to high frequencies during symptomatic primary infection , and maintains these at significant numbers during persistence . Despite this , the protective function of these lytic EBV antigen-specific cytotoxic CD8+ T cells remains unclear . Here we demonstrate that lytic EBV replication does not significantly contribute to virus-induced B cell proliferation in vitro and in vivo in a mouse model with reconstituted human immune system components ( huNSG mice ) . However , we report a trend to reduction of EBV-induced lymphoproliferation outside of lymphoid organs upon diminished lytic replication . Moreover , we could demonstrate that CD8+ T cells against the lytic EBV antigen BMLF1 can eliminate lytically replicating EBV-transformed B cells from lymphoblastoid cell lines ( LCLs ) and in vivo , thereby transiently controlling high viremia after adoptive transfer into EBV infected huNSG mice . These findings suggest a protective function for lytic EBV antigen-specific CD8+ T cells against EBV infection and against virus-associated tumors in extra-lymphoid organs . These specificities should be explored for EBV-specific vaccine development .
Epstein Barr virus ( EBV ) is a ubiquitous human γ-herpesvirus that establishes persistent infection in more than 90% of the human adult population [1] . Like other herpesviruses , EBV can undergo two infection programs . Latent infection leads to the expression of eight viral proteins and more than forty non-translated RNAs . This program is able to immortalize human B cells and is found in EBV associated tumors [2] . In contrast , lytic EBV infection leads to the expression of more than eighty lytic EBV proteins and the production of infectious viral particles [3] . Human T cells recognize both latent and lytic EBV antigens with distinct hierarchies [4] . While the latent nuclear antigen 1 ( EBNA1 ) is a consistently recognized CD4+ T cell antigen [5] , latent membrane protein 2 ( LMP2 ) and the EBNA3 proteins are prominent CD8+ T cell antigens [6] . Furthermore , CD4+ T cells often recognize late lytic EBV antigens [7] , while CD8+ T cells arise in response to early lytic antigens , including BMLF1 [8] . While EBV transformed B cells ( Lymphoblastoid cell lines ( LCLs ) ) can be readily recognized by CD8+ T cells specific for lytic EBV antigens , the protective value of lytic antigen-specific CD8+ T cells has not been readily demonstrated so far . Moreover , the expansion of these lytic EBV antigen-specific CD8+ T cells seems to follow viral load during symptomatic primary infection , called infectious mononucleosis , while latent EBV antigen-specific T cells peak during convalescence consistent with their involvement in viral immune control [4] . Thus , we aimed to evaluate the protective role of lytic EBV antigen specific CD8+ T cells in vitro and in vivo . Due to its exclusive tropism for human cells , this oncogenic γ-herpesvirus is difficult to study in vivo . However , recent advances have led to the development of two informative in vivo models . One model examines the infection of rhesus macaques with lymphocryptoviruses ( LCV ) , a subgroup of γ-herpesviruses that includes EBV [9] , and the other model examines EBV infection in mice with reconstituted human immune system components [10] . In both systems , T cell targeted immunosuppression leads to loss of viral immune control and virus-associated tumor formation [11] , [12] , [13] . We have explored EBV infection of non-obese diabetic mice with a severe combined immunodeficiency mutation and with complete loss of the interleukin 2 receptor gamma chain locus ( NOD-scid IL2Rγnull or NSG ) . These mice were reconstituted with human immune system components ( huNSG mice ) . In this model , both CD4+ and CD8+ T cells contribute to adaptive immune control of EBV [13] , [14] . Furthermore , it allows the assessment of innate immune responses by natural killer ( NK ) cells in response to EBV infection [14] , [15] and the exploration of EBV-specific vaccine candidates targeted to dendritic cells [16] , [17] . Finally , the infection of huNSG mice with EBV isolates and mutants with enhanced tumorigenesis replicate clinical features of EBV infection [18] , [19] . Thus , this in vivo model of EBV infection recapitulates main features of EBV infection in humans and should allow us to interrogate the protective value of T cell responses against latent and lytic EBV antigens . In this study , we demonstrated that wild-type ( WT EBV ) and BZLF1 deficient EBV ( ZKO EBV ) , which lacks with BZLF1 one of the immediate early transactivators of lytic replication , replicate to similar viral titers in huNSG mice . However , BZLF1 deficient virus establishes B cell lymphomas less efficiently outside of secondary lymphoid tissues . Furthermore , CD8+ T cells specific for the lytic EBV antigen BMLF1 eliminate lytically EBV replicating B cells efficiently in LCL cultures in vitro and in huNSG mice in vivo , thereby transiently controlling EBV infection in huNSG mice . Thus , lytic EBV antigen-specific CD8+ T cells are able to target cells infected with lytically replicating virus . These antigens should be considered for EBV-specific vaccine formulations , and in particular for patients with uncontrolled primary lytic EBV replication such as infectious mononucleosis .
NOD-scid IL2Rγnull HLA-A2 transgenic ( NSG-A2tg ) mice were obtained from the Jackson Laboratory , and bred and raised under specific pathogen-free conditions at the Institute of Experimental Immunology , University of Zürich , Switzerland . Newborn NSG-A2tg mice ( 1 to 5 days old ) were irradiated with 1 Gy and injected intrahepatically 5–7 hours later with 1–2×105 HLA-A*02 positive CD34+ human hematopoietic progenitor cells . CD34+ cells were isolated as described previously from human fetal liver tissue ( Advanced Bioscience Resources , Alameda , CA , USA ) [13] , [18] . The reconstitution of human immune system components in the peripheral blood of humanized NSG-A2tg mice ( huNSG-A2tg ) was analyzed for each cohort 12 weeks after engraftment and prior to the beginning of experiments . All animal protocols were approved by the cantonal veterinary office of Zurich , Switzerland ( protocol nos . 116/2008 and 148/2011 ) . All studies involving human samples were reviewed and approved by the ethical committee of Zurich , Switzerland ( protocol no KEK-St-Nr 19/08 ) . These protocols follow the European Convention for the Protection of Vertebrate Animals used for Experimental and Other Scientific Purposes as well as the Swiss Animal Welfare Act ( TSchG; 455 ) ( Amendment of 15 June 2012 ) and the Swiss Animal Welfare Ordinance ( TSchV; 455 . 1 ) . Epstein-Barr virus B95-8 wild-type ( WT EBV ) and BZLF1 knock-out ( ZKO EBV ) virus were produced in 293 HEK cells ( kindly provided by Regina Feederle and Henri Jacques Delecluse , Heidelberg , Germany ) . Titration of viral concentrates was performed on Raji cells using serial dilution and calculated as Raji infection units ( RIU ) using flow cytometric analysis of GFP-positive Raji cells 2 days after infection . Bulk peripheral blood mononuclear cells ( PBMCs ) were isolated from buffy coats by centrifugation using Ficoll-Paque , stained with carboxyfluorescein succinimidyl ester ( CFSE ) and incubated with mock , WT or ZKO EBV at a multiplicity of infection of 0 . 01 in the presence of cyclosporine A ( CsA , 1 µg/ml ) . Cultures were harvested and counted every 2–3 days . Absolute numbers of cells were counted using trypan blue exclusion and assessed by flow cytometry . In order to maintain cultures in the logarithmic growth phase , they were maintained at a maximal density of 1×106 cells/ml and supplemented with fresh media every 5 days . Assays were performed on PBMCs from two EBV-seropositive donors in duplicate . Fourteen to sixteen weeks after engraftment , huNSG-A2tg mice were injected with 105 RIU of WT , ZKO EBV or with PBS intraperitoneally , bled weekly starting from week 2 and euthanized 6 weeks after infection . For adoptive transfer studies , mice received 1×106 of LMP2- or BMLF1-specific CD8+ T cell clones intravenously one day prior to infection , and they were euthanized 4 or 6 weeks post-infection . Clones for adoptive transfer were selected based on their level of IFNγ secretion . All cell transfer experiments were performed with either of two independently obtained LMP2- or BMLF1-specific T cell clones . After termination of the experiments , body and spleen weight were determined . Abdominal organs were analyzed macroscopically for the presence of visible tumors . Tissue was fixed using 4% formalin and then paraffin embedded . For immunohistochemistry , anti-CD20 ( L26 , Cell Marque Lifescreen ) , anti-EBNA2 ( PE2 , Novocastra Laboratories Ltd ) , anti-NKp46 ( BAF 2225 , R&D Systems ) , anti-human CD68 ( 514H12 , Novocastra Laboratories Ltd ) , anti-neutrophil elastase ( NE , ab21595 , Abcam ) , anti-gp350 ( OT6 , kindly provided by Dr . J . M . Middledorp , VU University Medical Center , The Netherlands ) or anti-ZEBRA antibodies ( AZ-69 , Argene ) were used and 4 µm sections were processed using standard procedures on a BOND-MAX automated immunohistochemistry system ( Leica Microsystems ) . Isotype staining served as negative control and EBV-positive specimen of human tonsil or control or M81-infected huNSG-A2tg mouse tissue as positive controls . The absolute number of ZEBRA+ ( BZLF1+ , BZ . 1+ ) cells per spleen section was counted in a blinded fashion and normalized to the total area of spleen section . The total area of spleen sections was determined using ImageJ Software . The number of CD20+ , EBNA2+ , NKp46+ , CD68+ and NE+ cells was counted for 5 and 10 fields per spleen at ×200 and ×400 magnification , respectively . Total DNA from the whole blood was extracted using NucliSENS ( bioMérieux ) according to manufacturer's instructions . Splenic tissue was processed using QIAmp tissue Kit ( QIAGEN ) using the manufacturer's protocol . Quantitative analysis of EBV DNA was performed by TaqMan ( Applied Biosystems ) real-time PCR technique as described [20] with modified primers for the BamH1 W fragment ( 5′-CTTCTCAGTCCAGCGCGTTT-3′ and 5′-CAGTGGTCCCCCTCCCTAGA-3′ ) and a fluorogenic probe ( 5′- ( FAM ) -CGTAAGCCAGACAGCAGCCAATTGTCAG- ( TAMRA ) -3′ ) . All PCRs were run on an ABI Prism 7700 Sequence Detector ( Applied Biosystems ) and samples were analyzed in duplicates . No EBV DNA was detected in the blood of mock-infected animals for the duration of the experiment . Mice were considered uninfected if EBV DNA was not detected in the blood and spleen during the experiment . To analyze B cell outgrowth in in vitro assays , cells were labeled using anti-CD19 ( HIB19 , Biolegend ) , anti-CD3 ( UCHT1 , Biolegend ) , and anti-CD23 ( M-L233 , BD Biosciences ) antibodies . The composition of blood and spleen samples from humanized mice was analyzed using anti-human CD45 ( HI30 , Biolegend ) , anti-CD3 ( MHCD1918 , Invitrogen ) , anti-CD4 ( RPA-T4 , Biolegend ) , anti-CD8 ( SK . 1 , Biolegend ) , and anti-CD19 ( MHCD1917 , Invitrogen ) antibodies . Spleens were mechanically disrupted and filtered through a 70 µm cell strainer . Erythrocyte lysis in whole blood or in spleen suspensions was done using NH4Cl . Cell suspensions were stained with antibodies for 20 min at 4°C and washed . To sort EBV-specific CD8+ T cells , PBMCs were isolated using Ficoll-Paque , washed and stained using PE-labeled HLA-A*0201 dextramers complexed with the following peptides: HIV gag77–85 ( SLYNTVATL ) , LMP2426–434 ( CLGGLLTMV ) and BMLF1259–267 ( GLCTLVAML ) ( Immudex MHC Dextramer ) . For staining HLA-A*0201 dextramers loaded with the BRLF1109–117 ( YVLDHLIVV ) or with the LMP1159–167 ( YLQQNWWTL ) were used in addition . Cells were incubated with dextramers for 10 min at room temperature followed by antibody labeling with anti-CD3 ( UCHT1 , Biolegend ) , and anti-CD8 antibodies . To assess the purity of the clones , cells were stained using the antibody combination described in the assessment of the reconstitution above together with relevant dextramer or irrelevant dextramer as negative control . To assess the expression of homing and activation molecules , cells were labeled with anti-CD62L ( DREG56 , BD Pharmigen ) , anti-CCR7 ( clone 150503 , R&D Systems ) , anti-HLA-DR ( L243 , Biolegend ) and anti-CD25 ( M-A251 and 2A3 , BD ) antibodies . To detect cytotoxic granules in T cell clones , cells were stained as described , fixed , permeabilized and stained intracellularly using anti-perforin ( δG9 , BD Pharmigen ) and anti-granzyme B ( GB11 , BD Pharmigen ) with corresponding isotype controls . To detect lytically replicating B cells in killing assays , cells were labeled with anti-CD19 ( HIB19 , Biolegend ) and anti-CD3 ( MHCD0328 , Invitrogen ) antibodies , fixed , permeabilized and stained intracellularly using mouse anti-ZEBRA ( BZ . 1 , Santa Cruz ) primary antibodies ( or isotype control ) and secondary goat anti-mouse PE antibodies ( BD Biosciences ) . Dead cells were excluded based on LIVE/DEAD fixable Aqua labeling ( Invitrogen ) . Fixation and permeabilization steps prior to intracellular stainings were performed using BD Cytofix/Cytoperm Kit . Fluorescently labeled cell suspensions were analyzed on a BD FACS Canto II or BD LSR Fortessa flow cytometer ( BD Biosciences ) . Flow cytometric data analysis was performed using FlowJo software . EBV-specific T cells were sorted from the blood of a healthy HLA-A*0201 positive EBV carrier using LMP2- or BMLF1-specific dextramers . The donor was genotyped as HLA-A*02 , A*68 , B*44 , and B*07 . Dextramer positive populations live CD3+CD8+ cells were single-cell sorted and plated onto irradiated PBMCs and LCLs as feeders in the presence of 1 µg/ml phytohemagglutinin L ( PHA-L ) and 150 U/ml hrIL-2 as previously described [21] . After 2 weeks , T cells were tested in a split well assay for IFNγ secretion upon re-stimulation with 1 µM of relevant peptides . T cells specifically recognizing the tested peptide were re-stimulated with the same protocol . They were tested on day 13 for phenotypic characteristics by flow cytometry and for functional T cell avidity in peptide titration assays . The TRAV and TRBV chain repertoire was assessed using a primer set as previously described [22] . Primers were obtained from Biomers ( Ulm , Germany ) . PCR amplification was performed in a 25 µl reaction volume containing Pfu polymerase Buffer , 200 mM deoxynucleotide triphosphate , 0 . 5 µM C3 primer , 0 . 5 U Pfu DNA Polymerase ( all reagents were provided by Thermo Fisher Scientific , Reinach , Switzerland ) , 0 . 5 µM forward primer , and 100 ng cDNA . The cycling conditions were as follows: initial denaturation for 4 min at 95°C and 35 cycles of 95°C for 30 s , primer annealing at 60°C for 20 s , and primer extension at 72°C for 60 s , terminated by a final extension at 72°C for 10 min . The PCR product was validated by electrophoresis in a 2% agarose gel . Nucleotide sequencing of PCR products was performed at Microsynth ( Balgach , Switzerland ) with 30 pmol of reverse C1 primer . TCR gene designations follow the ImMunoGeneTics ( IMGT ) nomenclature ( http://www . IMGT . org ) . For peptide titration assays , 1×104 cells of EBV-specific clones were incubated with synthetic cognate peptides ( Shanghai Biochem ) for 18 h . LMP2 peptides and BMLF1 peptides at 5 µM concentration served as an irrelevant control for BMLF1- and LMP2-specific CD8+ T cell clones , respectively . For co-culture assays , clones were incubated with HLA-A*02 mismatched or autologous lymphoblastoid cell lines ( LCLs ) in an effector to target ( E:T ) ratio of 1∶5 . LCLs were optionally pulsed with 1 µM cognate peptide for 1 h at 37°C and washed extensively . Supernatants were used for IFNγ ELISA ( Mabtech ) . For degranulation assays , clones were incubated with cognate or control peptide for 8 h and then stained intracellularly with perforin and granzyme B antibodies using the BD Cytofix/Cytoperm Kit . To evaluate the cytotoxic activity of T cell clones against the autologous LCLs , we performed functional in vitro killing assays . Autologous LCLs were generated by incubating bulk PBMCs from the T cell clone donor with B95-8 supernatants and used as targets . AKBM cells , genotyped as A*24 , A*31; B*35 , B*51; C*09 , and C*14 , were used as a mismatched control target . EBV positive AKBM cells that express GFP upon switching from latent to lytic EBV replication were induced to enter lytic cycle by ligation with F ( ab′ ) 2 IgG and used directly for co-culture experiments [23] . AKBM or autologous LCLs were incubated with BMLF1- or LMP2-specific CD8+T cell clones for 18 h at an E:T ratio of 1∶1 and stained intracellularly using ZEBRA ( BZ . 1 , Santa Cruz ) antibodies . Specific lysis was determined by using the formula: % lysis = 100× ( 1−[ ( experimental ZEBRA+ with effectors − mean isotype ctrl ) / ( mean ZEBRA+ without effectors − mean isotype ) ] . Negative value for specific killing indicates an increase in number of ZEBRA+ B cells . Two to four replicates were used for each condition . Genomic DNA was isolated from cell type of interest using Qiagen DNeasy Kit and genotyped by PCR-SSP method using HLA-A*/-B*/-DRB1* und HLA-DQB1* Kit ( Ptotrans ) . All data were analyzed with the Mann Whitney test , unless otherwise stated . A p value of <0 . 05 was considered statistically significant . Statistical analysis and the generation of graphs was performed using Prism software ( GraphPad Software ) . The GenBank accession number for the complete B95-8 wild type sequence is AJ507799 . 2 ( Table S3 ) . The UniProtKB/Swiss-Prot identification of BMLF1-protein is Q04360 . The UniProtKB/Swiss-Prot identification of LMP2-protein sequence is P13285 . The Immune Epitope Database identification numbers of HLA-A*02-restricted BMLF1259–267 and LMP2426–434 epitopes are 20788 and 6568 , respectively .
To understand the relevance of lytic infection and EBV lytic antigen-specific T cell immune responses in the control of EBV infection and virus-mediated B cell transformation , we used a recombinant EBV virus devoid of the lytic immediate early transactivator BZLF1 that is severely compromised to enter the lytic replication [24] . We first performed in vitro assays to assess if lytic replication itself has an effect on the initial transformation of B cells . Bulk peripheral blood mononuclear cells ( PBMCs ) from two EBV seropositive donors were infected with either WT or ZKO EBV with equal multiplicities of infection . This was done in the presence of cyclosporine A , an immunosuppressive drug that inhibits T cell responses , so that B cell outgrowth would not be affected by the presence of endogenous EBV-specific memory T cells in the assay . Three weeks post-infection , the absolute number of cells increased in the EBV-containing wells ( Figure 1A ) , as previously reported for another EBV strain [25] . Flow cytometry analysis of the infected cells revealed that elevated cell numbers were due to the outgrowth of EBV transformed B cells ( Figure 1B ) . Transformed B cells up-regulated expression of the cellular activation marker CD23 early after infection as compared to the non-infected control ( Figure S1 ) . Both viruses had similar initial transformation kinetics , which was confirmed by comparable CFSE dilution in both WT and ZKO EBV transformed B cells ( Figure 1C , Figure S1 ) . Additionally , these transformed cells revealed similar proliferation rates few months after transformation as reported previously ( data not shown ) [26] . Thus , lytic replication , as demonstrated using ZKO EBV , does not significantly contribute to the transformation of B cells and the subsequent proliferation of EBV-transformed B cell lines , referred to as lymphoblastoid cell lines ( LCLs ) , in vitro . To address whether lytic EBV replication has an effect on in vivo pathogenesis , we used a mouse model with reconstituted human immune system components ( huNSG-A2tg mice ) . Three independent cohorts of humanized NSG-A2tg mice were infected with WT or ZKO EBV and monitored for 6 weeks after infection ( Figure 2A ) . Whereas infrequent lytically replicating cells , which express BZLF1 protein ( ZEBRA , BZ . 1 ) , were detected in the spleen sections of the majority of the WT EBV infected animals , no lytically replicating ZEBRA+ cells were detected in the spleen sections from ZKO EBV-infected mice ( Figure S2 ) . However , expression of the late lytic EBV antigen gp350 persisted in ZKO EBV-infected animals , which might indicate a low level of weak lytic replication in the absence of BZLF1 ( Figure S2 ) . EBV DNA was detected in the blood starting at week 2 . The EBV virus titer rose rapidly until week 4 and moderately from week 4 to week 6 post-infection ( Figure 2B ) . The detection of EBV DNA in whole blood in the first and second week post-infection , however , is limited by method sensitivity ( <83 EBV copies/ml ) . WT and ZKO EBV replicated in the blood of infected animals with similar titers at all the time points , except for week three after infection . Combined data from seven independent cohorts of animals reveals a significantly lower viremia in the ZKO EBV infected animals at that time point ( Figure 2B ) . Notably , while blood viremia in all animals infected with ZKO EBV rarely exceeded 3×103 EBV copies/ml in the third week post-infection , half of the WT EBV infected animals reached up to 1×104 EBV copies/ml in the blood ( Figure 2B ) . This transient difference was , however , no longer observed at later time points . These findings suggest a low frequency of spontaneous EBV lytic reactivation , which transiently elevated viral loads in half of the WT EBV-infected humanized NSG-A2tg mice . We observed an expansion of the CD8+ T cell compartment in the blood of infected animals , which is a hallmark of acute EBV-specific immune responses . Healthy EBV carriers as well as PBS-treated humanized mice maintain CD4+ cells as a majority of the CD3+ T cell population , whereas viral infection led to expansion of a subset of cytotoxic CD8+ T cells and the inversion of the CD8:CD4 ratio , which was more pronounced in WT animals compared to ZKO EBV infected animals ( Figure 2C ) . Expanded T cells may be specific for the antigens expressed during the lytic phase of the EBV life cycle , which is reduced in ZKO EBV infected animals . While EBV-specific T cells could be detected by MHC dextramer staining in the blood of healthy EBV carriers ( Figure 3A ) , the staining intensity for expanded blood and splenic CD8+ T cell populations in EBV-infected huNSG-A2tg mice was low ( Figure S3 ) , as previously reported [13] , [27] . Interestingly , dextramer staining indicated also lytic EBV antigen specific T cell populations in the spleen of ZKO EBV infected animals , which along with the low level of gp350 staining in splenic sections could indicate a low level of weak lytic EBV replication in the absence of BZLF1 . Despite this , the observed CD8+ T cell expansion suggests that adaptive cell-mediated responses are mounted against EBV antigens , similar to what has been observed in primary EBV infection . Similar levels of EBV DNA were detected in the spleens of WT and ZKO EBV infected animals after six weeks of infection , while spleens of PBS-treated animals were free of EBV DNA ( Figure 2D , data not shown ) . We also examined mice for the presence of EBV-associated lesions in the abdominal organs , such as spleen , liver and kidneys , 6 weeks after infection . As previously reported , ZKO EBV infection seemed to be less aggressive than WT EBV infection and leads to fewer lymphomas during six weeks post infection [28] . However , the composite data from four and six weeks experiments demonstrated no significant difference in total number of lymphomas between WT and ZKO EBV infected mice: 12 out of 24 WT EBV infected ( 50% ) and 8 out of 21 ZKO EBV infected ( 38% ) animals developed lymphomas in one or more abdominal organs . At least one macroscopically visible B cell tumor was observed in the spleen of 58% of WT EBV infected animals , while ZKO EBV caused splenic tumors in 45% of the infected animals 6 weeks post-infection . Despite this fairly similar frequency in tumorigenesis in the spleen , a detailed analysis of abdominal non-lymphoid organs revealed that ZKO EBV infected animals showed a trend to reduced numbers of lymphomas in the liver ( p = 0 . 08 , Figure 2E ) . A single mouse developed a kidney lymphoma six weeks after WT EBV infection . When combining four and six weeks experiments from 7 independent experimental cohorts with 21–24 animals per group the trend for loss of lymphomagenesis in the liver upon ZKO EBV infection was confirmed ( p = 0 . 06 ) . Histological analysis of the tumors confirmed the B cell origin of the lymphoproliferative lesions . This was concomitant with expression of virus-encoded nuclear antigen EBNA2 , which is essential for the establishment of latent EBV infection ( Figure 2F , Figure S4 ) . Up to 60% of B cells of WT and ZKO EBV infected animals expressed EBNA2 , revealing no difference between those viruses ( Figure S4 ) . Furthermore , the number of NK cells , macrophages and neutrophils were comparable in the spleens of WT and ZKO EBV infected animals ( Figure S5 ) . Macrophage and less often NK infiltration was a predominant feature of hepatic B cell lesions in WT EBV infected animals . Most of the infected animals had splenomegaly , and this affected WT EBV infected animals more frequently ( Figure 2G ) . The absolute number of B cells in the spleens remained unchanged upon infection with WT or ZKO EBV as compared to uninfected mice ( data not shown ) . Splenomegaly in WT EBV infected animals resulted from the expansion of CD3+ T cells , while the expansion of T cells in the spleen was less pronounced in animals infected with ZKO EBV ( Figure 2H ) . Despite moderate T cell expansion in the spleens of ZKO EBV infected animals , the CD8:CD4 ratio was inverted for most of the infected animals independent of lytic replication , indicating the enrichment of cytotoxic CD8+ T cells irrespective of lytic replication ( Figure 2I ) . These results indicate that lytic EBV replication affects viremia in the blood only transiently . Despite this , high viral loads lead to the subsequent expansion of CD8+ T cells . The increased occurrence of tumors in organs other than the spleen indicates that lytic replication might contribute to the establishment of lymphoproliferative disease outside of secondary lymphoid organs in huNSG-A2tg mice . As we have previously reported , depletion of CD8+ T cells leads to poorly controlled WT EBV infection in the humanized NSG mouse model [13] , [14] . In order to assess the role of EBV latent and lytic protein-specific CD8+ T cells in the control of EBV infection , we established numerous CD8+ T cell clones specific to representative immunodominant HLA-A*02 epitopes such as LMP2426–434 ( latent ) and BMLF1259–267 ( lytic ) from the blood of an HLA-A*0201 positive healthy EBV carrier ( Figure 3A ) . After two rounds of re-expansion using the unspecific T cell mitogen phytohemagglutinin plus interleukin 2 in vitro , the cells were confirmed by dextramer straining to be pure clonal CD8+ T cell populations . Residual activation-driven CD8 down-regulation was observed for some clones early after re-stimulation ( Figure 3B ) . For further characterization we sequenced T cell receptor variable genes of obtained several clones ( Table S1 ) . We observed frequent TRBV20-1 usage by BMLF1-specific CD8+ T cell . Whereas , each of BMLF1-specific CD8+ T cell clones had a unique CDR3β and CDR3α sequence , two out of three LMP2-specific clones shared the same CDR3β and CDR3α sequences , indicating their common clonal origin and explaining their similar behavior in vitro . We also observed frequent TRVB5-1 usage by LMP2-specific T cell clones . In peptide titration assays , the functional avidity for LMP2-specific clones was greater than for BMLF1-specific clones ( 13 . 8±3 . 2×10−7 M and 4 . 4±2 . 7×10−6 M , respectively , Figure 3C , Table S2 ) . To assess if these CD8+ T clones were able to recognize cognate antigen in the context of MHC class I molecules , we performed co-culture experiments with autologous or allogenic HLA-A*02 negative EBV-transformed LCLs . BMLF1-specific clones secreted limited amounts ( less than 260 pg/ml ) of IFNγ upon co-culture with autologous WT EBV-transformed LCLs ( WT LCLs ) . In contrast , no IFNγ was detected upon co-culture with control HLA-A*02 negative WT LCLs ( nA2 WT LCLs ) or LCLs deficient in lytic replication ( ZKO LCLs ) ( Figure 3D , upper panels ) . LMP2-specific CD8+ T cell clones were capable of recognizing both autologous LCLs ( WT LCLs and ZKO LCLs ) expressing endogenously processed LMP2 protein , indicating sufficient presentation of LMP2 by cell lines . Interestingly , in two independent experiments the amount of IFNγ secreted by three LMP2-specific T cell clones in response to autologous ZKO LCLs accounted for 49±2% and 20±10% of IFNγ secreted towards WT LCLs ( Figure 3D , lower panel ) . As expected , recognition of autologous WT LCLs by either BMLF1- or LMP2-specific clones was improved dramatically when targets were pre-incubated with cognate but not irrelevant peptide ( Figure 3D ) . In order to address whether the BMLF1- and LMP2-specific clones were cytotoxic , we first assessed the presence of the pore-forming protein perforin and the cell death inducing serine protease granzyme B by intracellular staining . While incubation with irrelevant peptide resulted in high levels of perforin and granzyme B in these clones , stimulation with relevant peptide decreased the intensity of the staining , indicating the release of cytotoxic granules upon binding of the TCR with the cognate antigen ( Figure 3E ) . To demonstrate that BMLF1-specific CD8+ T cell clones were able to eliminate autologous lytically EBV replicating LCLs in an MHC class I restricted manner , we co-cultured BMLF1-specific clones with autologous LCLs that spontaneously enter lytic replication ( B95-8 EBV strain ) , or with HLA-mismatched AKBM cells that require the induction of lytic replication ( Akata EBV strain ) at a 1∶1 effector:target ratio . After overnight co-culture , the presence of lytically active LCLs was assessed using intracellular BZLF1 ( ZEBRA , Zta , BZ . 1 ) staining . The rate of spontaneous lytic replication in autologous LCLs was very low and reached a maximum of 2 . 3% of total B cells at high-density growth conditions ( Figure 4A ) , while induced lytically active AKBM cells accounted for 30% of total B cells in culture ( data not shown ) . Several independent BMLF1-specific CD8+ T clones efficiently eliminated ZEBRA+ cells from autologous co-cultures ( 74–90% of ZEBRA+ LCLs ) , while unspecific killing of mismatched AKBMs did not exceed 24% ( Figure 4B ) . As a control , co-cultures of autologous LCLs with LMP2-specific clones led to increased frequencies of ZEBRA+ cells as compared to control wells without T cells , which is likely due the killing of latently infected B cells and the subsequent increase in the relative number of ZEBRA+ cells ( Figure 4C ) . Thus , BMLF1-specific CD8+ T cell clones can eliminate LCLs with lytic EBV replication in vitro and might restrict the horizontal transmission of EBV through release of newly synthesized viral infectious particles . In order to assess if BMLF1- and LMP2-specific CD8+ T cell clones were able to ameliorate EBV induced pathogenesis in vivo , we performed adoptive transfer experiments in huNSG-A2tg mice . Five independent cohorts of humanized NSG-A2tg mice were injected intravenously with BMLF1- or LMP2-specific CD8+ T cells in order to mimic endogenous circulating effector T cell populations , which could for example be induced by vaccination and indeed the transferred clones carried an effector memory phenotype after expansion ( Figure S6 ) . On the next day , mice were infected with either WT or ZKO EBV and the development of lymphoproliferative disease was monitored over 4 or 6 weeks ( Figure 5A , Table S2 ) . Adoptive transfer of a BMLF1-specific CD8+ T cell clones tended to reduce blood viremia in WT EBV infected animals 3 weeks post-infection , a time-point when EBV lytic replication results in elevated blood viremia in humanized NSG-A2tg mice ( Figure 2B and 5B ) . Furthermore , ZEBRA+ B cells , containing lytically replicating WT EBV , were diminished in splenic sections of the majority of WT EBV infected animals upon transfer of BMLF-1 specific CD8+ T cell clones ( Figure 5C ) . No ZEBRA+ cells were found in splenic sections from ZKO EBV infected animals ( Figure S2 ) . However , the amelioration of the EBV viremia in the blood of WT EBV infected animals by adoptive transfer of BMLF1-specific CD8+ T cell clones varied in the experimental cohorts of humanized NSG-A2tg mice ( Figure 5D ) . Taken together , these five adoptive transfer experiments demonstrated that animals which received BMLF1-specific CD8+ T cells prior to WT EBV infection were less likely to develop high blood viremia ( defined as greater than 3×103 EBV/ml ) at three weeks post WT EBV infection: only 3 animals out of 14 ( 21% ) developed high viremia after BMLF1-specific T cell transfer as compared to 6 out of 14 ( 43% ) after LMP2-specific T cell transfer or 7 out of 16 ( 44% ) animals in the non-treated group ( Figure 5E ) . Adoptive transfer of LMP2-specific T cells did not seem to have an ameliorating effect in WT EBV infected animals , however ZKO EBV infected animals might have benefitted from the T cell infusion on the third and fourth week after infection , possibly due to elimination of latently infected B cells from blood circulation . Adoptive transfer of partially mismatched T cells in humans is considered relatively safe [29] , despite rare cases of mild graft-versus-host disease [30] . Humanized mice , which received partly matched cytotoxic T cell clones , had no signs of graft-versus-host disease during the experiment . We attempted to identify the defining characteristics for protective EBV specific T cell clones by examining peptide epitope avidity , functional potency and surface phenotype , but we were unable to identify any in vitro predictors for in vivo protection . Additionally , attempts to find the transferred cells by specific amplification of mismatched MHC class I alleles of the donor using DNA from the blood at weeks 2 , 3 and 4 post-infection , as well as in spleens and tumor samples upon termination of the experiments were limited by the sensitivity of our PCR-based method ( 1% of CD8+ T cells ) and did not detect persistence of the transferred clonal T cell populations ( data not shown ) . We also assessed the effect of adoptively transferring lytic and latent antigen specific T cell clones on the formation of lymphoproliferative disease induced by WT and ZKO EBV . Consistent with our earlier experiments , EBV infected animals developed splenic EBV-associated B cell lymphomas with comparable frequency irrespective of lytic replication , whereas extra-lymphoid lymphomas were less frequent when EBV lytic replication was impaired ( Figure 5F ) . Lymphomas from both WT and ZKO EBV infected animals contained EBNA2+ B cells , indicating the latency III program of the EBV life cycle . This form of latency is characterized by expression of all latent EBV antigens , namely EBNA proteins 1 , 2 , 3A , B , C , -LP , and LMP proteins 1 and 2 . Latency III is commonly found in the lymphoproliferative lesions of immune-compromised patients . Our study demonstrates subtle effects of EBV lytic replication on blood viremia and suggests that endogenous EBV lytic antigen-specific T cells might be primed in the huNSG-A2tg mice . Furthermore , we observed that EBV lytic replication may be required for the establishment of extra-lymphoid lymphoproliferative lesions in our mouse model . In healthy EBV carriers , memory CD8+ T cell specific for lytic and latent antigens of EBV are found in significant numbers . In this study , we provide the first evidence that BMLF1-specific CD8+ T cells eliminate EBV-transformed B cells undergoing lytic replication in vitro and in vivo , and are able to transiently control EBV viremia in the blood of infected humanized mice .
One form of uncontrolled lytic EBV replication is infectious mononucleosis . This symptomatic primary EBV infection occurs more frequently when primary viral infection is delayed into adolescence . It is characterized by high viral titers , particularly in saliva , and by CD8+ T cell lymphocytosis [31] . Otherwise , pathology by lytic replication of EBV is rare in healthy human carriers and EBV transformed lymphoblastoid cell lines switch only at low frequencies into lytic replication in vitro . Despite these findings , the role of lytic virus replication during EBV infection is only poorly understood . This is in part because the number of experimental systems that are permissive for EBV infection limits studies on EBV biology in vivo . We therefore examined EBV lytic replication in humanized NSG-A2tg mice and the functional role of BMLF1-specific T cells in the control of EBV infection in vivo and in vitro . This in vivo model of EBV infection recapitulates hallmarks of innate and adaptive cell-mediated immune control of EBV , but lacks the development of humoral immune responses and efficient mucosal homing of human immune cells [10] . Our longitudinal in vivo study suggests that EBV lytic replication is limited in our humanized mouse model and plays a transient role in dissemination of the EBV early after infection and prior to innate and adaptive immune control of lytically EBV replicating cells: only mice infected with lytic replication competent EBV , but not lytic replication compromised EBV , developed high viremia in the blood on the third week post-infection . The massively expanding CD8+ T cells in infectious mononucleosis patients are mostly directed against lytic EBV antigens [32] , suggesting that primarily lytic viral replication is ill-controlled during infectious mononucleosis . Particularly , CD8+ T cells that are specific for immediate early and early EBV antigens are observed at high frequencies [33] . These events correlate with more efficient LCL recognition by early lytic EBV antigen specific CD8+ T cells , while late lytic EBV antigen recognition by specific CD8+ T cells is presumably compromised by viral immunoevasins targeting MHC class I antigen presentation [34] . In our humanized mouse model we observed that high blood viral load frequently correlated with the number of EBV-specific T cells found in the spleens of the WT and ZKO EBV infected animals as detected by IFNγ ELISPOT assays ( data not shown ) . Also , we report the inversion of the CD8:CD4 ratio in blood of infected animals , which is less pronounced in animals infected with EBV that is inhibited in its lytic replication . Our lab previously demonstrated that a significant proportion of EBV-specific T cells are primed specifically to lytic antigens rather than latent antigens in humanized NSG-A2tg mice [13] . These data suggest that lytic EBV antigens contribute to CD8+ T cell expansion after infection . However , in our current study , we observe more EBV-specific T cells among splenocytes from both WT and ZKO EBV infected mice , if we stimulated the splenocytes with LCLs capable of lytic EBV replication ( data not shown ) . Together with the observation in vitro that LMP2-specific CD8+ T cell clones consistently secreted more IFNγ in response to autologous WT LCL than to ZKO LCL stimulation , this might indicate decreased antigen presentation of LCLs deficient in lytic replication . Additionally , detection of EBV-specific T cells using HLA-A*02 dextramers bound to human immunodominant epitopes of EBV lytic and latent proteins was difficult due to low staining intensity ( Figure S3 ) . The process of T cell education of the T cells in humanized mice remains unclear . In these mice human T cells are educated on mouse thymic epithelial cells , human bone marrow derived cells or transgenic human MHC class I molecules stabilized by murine β2-microglobulin . We can speculate that this may result in T cells with TCRs , that are suboptimally recognized by human HLA-A*02 dextramers . Alternatively , maintenance of high TCR density on T cells in the mouse host might be compromised by lack of human cytokines and further compromised due to the high antigenic load after EBV infection . Pudney and colleagues demonstrated direct recognition of lytically infected cells by CD8+ specific cells obtained from infectious mononucleosis patients [34] . However , in these studies , the authors could only demonstrate cytokine production by lytic EBV antigen specific CD8+ T cell clones in response to wild-type , but not BZLF1-deficient LCLs . Extending these findings , we can now document that BMLF1 specific CD8+ T cell clones eliminate ZEBRA expressing EBV infected B cells in LCL cultures . BMLF1-specific clones eliminated EBV lytically replicating autologous LCLs , but not MHC-class I mismatched control LCLs . Thus , lytic EBV antigen specific T cell clones contribute to the immune control of EBV infection by eliminating lytically virus replicating cells . Another EBV-associated pathological condition in solid-organ and hematopoietic stem cell transplant patients is post-transplant lymphoproliferative disease ( PTLD ) . This disease almost exclusively involves EBV-associated B cell proliferations characterized by expression of all latent proteins of EBV ( latency III ) . In immune-compromised PTLD patients , mostly polyclonal B cell lesions could be found in secondary lymphoid organs and extra-lymphoid tissues [2] . EBV infection of humanized mice leads to the formation of lymphoproliferative B cell lesions resembling PTLD in histology and expression pattern of viral proteins . EBER+ EBV-transformed B cells were found in similar multifocal sites independently of lytic replication in humanized NSG mice , which were implanted with human fetal thymus and liver fragments in addition to the fetal liver derived CD34+ stem cell transfer of our model [28] . Our study for the first time demonstrates that EBV may require lytic replication to efficiently establish lymphomas in non-lymphoid tissues like kidneys and liver . This could result from decreased expression of growth factors in ZKO-derived LCLs , which may be more critical for the establishment of lymphoproliferative lesions outside of secondary lymphoid organs [26] . One therapy for patients with PTLD is adoptive T cell transfer . Both CD8+ and CD4+ EBV-specific T cells are being explored for adoptive transfer into PTLD patients [35] , [36] , [37] , [38] , [39] . Here we report , that adoptive transfer of BMLF1-specific CD8+ T cell clones into WT EBV infected humanized NSG mice transiently controlled EBV infection and was able to eliminate BZLF1 expressing cells from huNSG-A2tg spleens . However , given the limitation of detecting lytically replicating cells indirectly by examining blood viremia , this transfer did not result in significant and reproducible long-term control of EBV infection . We addressed whether fluctuations in the control of lytic replication in vivo was due to the characteristics of the transferred clones ( epitope avidity , phenotype etc ) . However , no clear correlations could be found ( Table S2 ) . Alternatively , the inefficiency of clonal T cell transfer may result from the overall HLA mismatch – except matched HLA-A2 expression – between reconstituted human immune system components and transferred T cell clones . This might have led to their alloreactive rejection by the host and the limited persistence of the transferred cells in vivo . Clinical studies suggest that a positive outcome of cytotoxic T cell transfer into PTLD patients positively correlates with the matching of donor T cells to the recipient [38] . Despite this , an ameliorating effect of adoptive transfer in PTLD patients could also be observed in partly matched settings [30] , [39] . Within five cohorts of huNSG-A2tg mice reconstituted with one ( Cohorts 2 , 3 ) or more ( Cohorts 1 , 4 , 5 ) matched HLA alleles , the most pronounced effect of BMLF1-specific clone on viremia in blood was observed irrespective of the number of matched MHC class I alleles ( Cohorts 1 and 2 at week 3 , cohort 3 at week 4 ) . Likewise , adoptive transfer of partly matched T cell clones in primates revealed no correlation between persistence of the infused cells and genetic backgrounds of donor and recipient [40] . Taken into account that WT EBV infection caused high viremia in half of the infected humanized NSG-A2tg mice , our study indicates that BMLF1-specific CD8+ T cells seem to maintain some immune control and can eliminate B cells undergoing lytic replication . Our data provide a rationale for reconsidering lytic EBV antigens for EBV specific vaccination , particularly during uncontrolled lytic replication and to limit EBV associated tumorigenesis in non-lymphoid tissues , which seems to be supported by lytic replication per se . Accordingly , lytic EBV antigens should be explored as vaccine components to mitigate infectious mononucleosis . Currently , several EBV specific vaccine candidates are being explored . These include recombinant gp350 protein formulations that aim to elicit neutralizing antibodies against the main attachment protein for complement receptor 1 and 2 ( CD35 and CD21 ) dependent B cell infection by EBV [41] , [42] , [43] . Furthermore , recombinant viral vaccines encoding latent EBV antigens , for example modified vaccinia virus Ankara ( MVA ) , encoding a EBNA1-LMP2 fusion protein , are being explored [44] , [45] . EBV derived virus like particles devoid of viral DNA have been investigated as vaccine candidates [46] , [47] , [48] , [49] . Finally , we have been characterizing EBNA1 targeting to dendritic cells as a vaccine formulation [16] , [17] , [50] . However , all of these vaccination schemes include either latent or late lytic EBV antigens . They do not provide any stimulation for the immediate early and early lytic EBV antigen specific CD8+ T cells , which constitute the most frequent EBV specificities in both symptomatic primary infection and healthy EBV carriers . Since these specificities , nevertheless , efficiently recognize and eliminate lytically EBV replicating B cells ( [34] and our data ) , they should be included into vaccine formulations against infectious mononucleosis and possibly also against EBV associated malignancies in extra-lymphoid tissues . | Epstein Barr virus persistently infects more than 90% of the human adult population . While fortunately carried as an asymptomatic chronic infection in most individuals , it causes B cell lymphomas and carcinomas in some patients . Symptomatic primary EBV infection , called infectious mononucleosis , predisposes for some of these malignancies and is characterized by massive expansions of cytotoxic T cells , which are mostly directed against lytic EBV antigens that are expressed during virus particle production . Therefore , we investigated the protective role of lytic EBV antigen specific T cells during EBV infection and the contribution of lytic EBV infection to virus-associated tumor formation . We found that lytic EBV antigen specific T cells kill B cells with lytic virus replication and might thereby transiently control EBV infection in mice with human immune system components . Furthermore , we observed that EBV associated B cell tumors outside secondary lymphoid organs may require lytic replication for efficient formation . Thus , we suggest that lytic EBV antigens should be explored for vaccination against symptomatic EBV infection and EBV associated extra-lymphoid tumors . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"viruses",
"and",
"cancer",
"immunology",
"microbiology",
"tumor",
"immunology",
"epstein-barr",
"virus",
"infectious",
"mononucleosis",
"infectious",
"disease",
"immunology",
"vaccination",
"and",
"immunization",
"animal",
"models",
"of",
"infection",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"cell",
"biology",
"clinical",
"immunology",
"virology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases"
]
| 2014 | Adoptive Transfer of EBV Specific CD8+ T Cell Clones Can Transiently Control EBV Infection in Humanized Mice |
During immature capsid assembly , HIV-1 genome packaging is initiated when Gag first associates with unspliced HIV-1 RNA by a poorly understood process . Previously , we defined a pathway of sequential intracellular HIV-1 capsid assembly intermediates; here we sought to identify the intermediate in which HIV-1 Gag first associates with unspliced HIV-1 RNA . In provirus-expressing cells , unspliced HIV-1 RNA was not found in the soluble fraction of the cytosol , but instead was largely in complexes ≥30S . We did not detect unspliced HIV-1 RNA associated with Gag in the first assembly intermediate , which consists of soluble Gag . Instead , the earliest assembly intermediate in which we detected Gag associated with unspliced HIV-1 RNA was the second assembly intermediate ( ~80S intermediate ) , which is derived from a host RNA granule containing two cellular facilitators of assembly , ABCE1 and the RNA granule protein DDX6 . At steady-state , this RNA-granule-derived ~80S complex was the smallest assembly intermediate that contained Gag associated with unspliced viral RNA , regardless of whether lysates contained intact or disrupted ribosomes , or expressed WT or assembly-defective Gag . A similar complex was identified in HIV-1-infected T cells . RNA-granule-derived assembly intermediates were detected in situ as sites of Gag colocalization with ABCE1 and DDX6; moreover these granules were far more numerous and smaller than well-studied RNA granules termed P bodies . Finally , we identified two steps that lead to association of assembling Gag with unspliced HIV-1 RNA . Independent of viral-RNA-binding , Gag associates with a broad class of RNA granules that largely lacks unspliced viral RNA ( step 1 ) . If a viral-RNA-binding domain is present , Gag further localizes to a subset of these granules that contains unspliced viral RNA ( step 2 ) . Thus , our data raise the possibility that HIV-1 packaging is initiated not by soluble Gag , but by Gag targeted to a subset of host RNA granules containing unspliced HIV-1 RNA .
For released HIV-1 particles to be infectious , they must contain two copies of unspliced ( full-length ) HIV-1 RNA that are packaged during assembly of the immature HIV-1 capsid . Each immature capsid is composed of ~3000 copies of the HIV-1 structural protein Gag , which initially oligomerize in the cytoplasm and subsequently target to the plasma membrane ( PM ) , where Gag multimerization is completed . Packaging of the viral genome is initiated when Gag first associates with unspliced viral RNA during assembly , and requires the nucleocapsid domain ( NC ) of Gag as well as specific encapsidation signals in unspliced HIV-1 RNA ( reviewed in [1] ) . Immature capsids subsequently undergo budding , resulting in release of immature virus particles that contain the encapsidated genome and undergo maturation ( reviewed in [2] ) . In the absence of unspliced HIV-1 RNA , Gag proteins assemble and release properly but the resulting virus-like particles are non-infectious [3] . In addition to being packaged , unspliced HIV-1 RNA is used for translation of Gag and GagPol ( reviewed in [1] ) . It is generally agreed that translation and packaging are unlikely to occur concurrently , given that translation requires melting of secondary structures that are utilized during packaging; therefore translation and packaging are likely to be mutually exclusive ( reviewed in [4 , 5] ) . However , the determinants that govern whether an unspliced HIV-1 RNA is utilized for translation or for packaging remain unclear . Mechanisms that have been proposed to explain how an unspliced HIV-1 RNA is directed towards packaging instead of translation include alternate RNA conformations that mask the translation start site and expose elements that favor packaging ( reviewed in [5 , 6] ) ; alternate 5' mRNA cap sequences that promote conformations favorable for packaging [7]; and inhibition of translation by accumulated Gag [8 , 9] . Additionally , it has been proposed that prior translation could lead to preferential packaging of an unspliced HIV-1 RNA [10] . However , other studies argue that trans packaging is the dominant packaging mechanism [11] , and it is very clear that prior Gag translation is not necessary for packaging given that RNA generated from a lentiviral vector provided in trans can be packaged by Gag proteins synthesized from a different transcript ( reviewed in [1 , 4] ) . Thus , in aggregate , the data argue that non-translating unspliced HIV-1 RNA undergoes packaging , and that one or more regulatory mechanisms play a role in determining whether a particular unspliced HIV-1 RNA is translated or packaged . The process of packaging likely involves multiple steps that are closely coordinated with immature capsid assembly , culminating in complete encapsidation of the unspliced HIV-1 RNA within the fully assembled capsid . One would expect the first step of this process to involve association of Gag with unspliced HIV-1 RNA . Indeed , biochemical and imaging studies have supported the idea that association of Gag with unspliced viral RNA initiates the packaging process [12 , 13] . These studies demonstrated that Gag is likely a dimer or small oligomer when it first associates with unspliced HIV-1 RNA to initiate packaging; additionally , studies of an assembly-defective Gag mutant ( Gag G2A ) indicated that this initial association likely occurs in the cytoplasm [13] . The complex in which Gag first associates with unspliced HIV-1 genomic RNA to initiate packaging , here termed the packaging initiation complex , has not been identified , nor is it known how this complex is formed; additionally it is not known whether this complex contains only Gag or also contains cellular proteins . Answering these questions could lead to novel strategies for inhibiting genome encapsidation in infected cells . Despite this , to date , no study has even identified a candidate packaging initiation complex ( let alone a definitive packaging initiation complex ) , leaving a large gap in our understanding of early events in the packaging process . Because packaging occurs simultaneously with immature capsid assembly , insights into early events in packaging could be gained by studying the association of HIV-1 Gag with unspliced HIV-1 RNA during early events in immature capsid assembly . Here we leveraged our understanding of the sequence of events in HIV-1 immature capsid assembly to identify the earliest assembly itermediate in which HIV-1 Gag is associated with unspliced HIV-1 RNA . Previously , we showed that Gag progresses through a pathway of intracellular assembly intermediates consisting of sequential complexes of increasing size ( referred to as the ~10S , ~80S , ~150S , and ~500S assembly intermediates ) and culminating in formation of the ~750S completed immature capsid ( reviewed in [14] ) . The sequential nature of these complexes was initially demonstrated by pulse-chase experiments , in cells and in cell extracts , which showed that over time newly synthesized Gag moves from an ~10S complex to an ~80S/150S complex , and then to an ~500S complex , before forming a completed ~750S immature capsid [15 , 16] . Two additional approaches confirmed the sequential order of these intracellular assembly intermediates that was predicted by pulse-chase studies . The first utilized an analysis commonly employed in the study of signaling pathways , in which the temporal order of events in a signaling pathway is determined by blockade of specific steps in that pathway . In an analogous manner , use of mutational blockade revealed that every assembly-defective Gag mutant studied to date is arrested at a specific point along the proposed capsid assembly pathway and forms only the intermediate at which it is arrested and the smaller intermediates that precede the point of arrest [15 , 17–20] . The second approach that confirmed the predicted temporal order of intermediates in this pathway involved defining the subcellular localization of these intermediates using biochemical as well as imaging approaches . These studies showed that , at steady state , the earliest intermediate ( ~10S ) is largely cytosolic , the next intermediate ( ~80S/150S ) is both in the cytosol and at the PM , and the final intermediate ( ~500S ) is exclusively at the PM [19] , thereby demonstrating the expected cytosol to PM progression of assembling Gag . Further support for the assembly intermediates being precursors to immature capsids came from the finding that the ~80S/150S and ~500S intermediates contain two other viral proteins found in the completed immature capsid: HIV-1 GagPol , which is present in a 1:20 ratio relative to Gag in assembly intermediates as is the case in released virus , and HIV-1 Vif [15 , 21] . Thus , together the pulse-chase , mutational , compositional , and subcellular localization analyses strongly support a temporal progression of assembling Gag from ~10S to ~80S/150S to 500S assembly intermediates , before formation of the ~750S completed immature capsids which subsequently undergo budding and release ( reviewed in [14] ) . Notably , studies of the assembly pathway had also revealed that the ~80S/150S , and ~500S intermediates contain host RNA granule proteins , such as DEAD-box RNA helicase 6 ( DDX6 ) [22] . RNA granules are host ribonucleoprotein complexes that contain non-translating mRNA , in contrast to ribosomes , which contain translating mRNA ( reviewed in [23] ) . Different classes of RNA granules exist , distinguished by their sizes and marker proteins , and functioning in silencing , storage , degradation , stress , and other events in RNA metabolism ( reviewed in [23 , 24] ) . Some RNA granules , such as P bodies and stress granules , form relatively large foci that are easily visible by light microscopy , while others are smaller and poorly understood . Given that RNA granule proteins are not associated with HIV-1 Gag in the ~10S early assembly intermediate [19 , 20 , 22] , these studies suggested that the ~80S HIV-1 assembly intermediate is formed when assembling Gag co-opts a poorly understood , small host RNA granule that contains canonical RNA granule proteins , such as DDX6 . These Gag-containing granules also contain the host ATP-binding cassette protein E1 ( ABCE1 ) [15 , 17–21] , which has not been reported to be present in larger RNA granules . ABCE1 and DDX6 remain with Gag until immature capsid assembly is completed , at which point these host proteins dissociate [15 , 21 , 22] and the completed immature capsid undergoes budding and release . Previous studies had also shown that , like the completed capsid , the ~80S/150S and ~500S assembly intermediates also contain unspliced HIV-1 RNA , as would be expected if they are precursors to the completed immature capsid . In these studies , an antibody directed against ABCE1 , a cellular marker of assembly intermediates , was found to coimmunoprecipitate Gag as well as unspliced HIV-1 RNA from fractions containing the ~80S and ~500S assembly intermediates [19] . These findings are consistent with the assembly intermediates containing ABCE1 , HIV-1 Gag , and unspliced HIV-1 RNA . However , they do not answer the key question of which assembly intermediate is the first to contain Gag associated with unspliced HIV-1 RNA , since the ~10S assembly intermediate does not contain ABCE1 and was therefore not immunoprecipitated in that study . Determining whether Gag associates with unspliced HIV-1 RNA in the earliest assembly intermediate ( ~10S intermediate ) would require asking whether antibodies directed against Gag in the ~10S complex coimmunoprecipitate unspliced HIV-1 RNA . Importantly , this approach would also test between two fundamentally different models for the initial association of Gag with unspliced HIV-1 RNA during assembly . If Gag in the ~10S assembly intermediate is associated with unspliced HIV-1 RNA , that would suggest that unspliced HIV-1 RNA first associates with soluble Gag , given the small size of the ~10S complex ( Model 1 ) . Alternatively , if the ~80S and ~500S intermediates are associated with unspliced HIV-1 RNA but ~10S Gag is not , that would suggest that the first association of Gag with unspliced HIV-1 RNA occurs not in the soluble fraction , but within an assembly intermediate that is derived from a host RNA granule [22] ( Model 2 ) . To date , studies have not tested between these two models for when and where assembling Gag first associates with unspliced HIV-1 RNA , each of which has compelling features . The soluble Gag model ( Model 1 ) is appealing in its simplicity , and is consistent with longstanding studies showing that Gag associates specifically with unspliced HIV-1 RNA ( reviewed in [1 , 25] ) . The RNA granule model ( Model 2 ) is also consistent with these prior studies , but builds on the concept that in cells unspliced HIV-1 RNA likely behaves like cellular mRNA , which is found almost exclusively in ribonucleoprotein complexes . Cellular ribonucleoprotein complexes are generated during transcription , and undergo successive rounds of remodeling ( reviewed in [26] ) , with their changing protein components dictating their changing fates ( reviewed in [27] ) . In the RNA granule model , some of the ribonucleoprotein complexes that contain unspliced HIV-1 RNA would become translating complexes upon entering the cytoplasm , while others would form cytoplasmic non-translating complexes , including RNA granules . This model also suggests that Gag would have to localize to such non-translating host RNA granules in order to associate with the pool of unspliced HIV-1 RNA that is not occupied by translation machinery and to initiate interactions with that RNA . Having the first association between assembling Gag and unspliced HIV-1 RNA occur within RNA granules would provide numerous advantages to the nascent virus—for example , it could sequester unspliced HIV-1 RNA away from the host innate immune system , concentrate assembling Gag at a site rich in unspliced HIV-1 RNA , and place the Gag-RNA association and assembly in proximity with host enzymes that could facilitate those events . In keeping with the latter possibility , ABCE1 and DDX6 , two of the host proteins present in both the assembly intermediates and host RNA granules , have been shown to facilitate immature HIV-1 capsid assembly [21 , 22] by mechanisms that remain to be determined . To determine whether Gag first associates with unspliced HIV-1 RNA in the earliest assembly intermediate ( ~10S Gag-containing complex ) or in an RNA-granule-derived intermediate ( ~80S or 500S Gag-containing complex ) , here we sought to identify all non-nuclear cellular complexes that contain unspliced HIV-1 RNA , as well as the subset of these complexes that contains Gag in association with unspliced HIV-1 RNA . First we found that , in the presence or absence of assembling Gag , essentially all non-nuclear unspliced HIV-1 RNA , whether translating or non-translating , is in complexes ≥30S . Next we showed that in lysates of cells expressing proviral Gag or Gag expressed with a genomic construct in trans , the ~80S assembly intermediate was the smallest previously described assembly intermediate in which Gag was associated with unspliced viral RNA at steady state . Unspliced HIV-1 RNA was found in the ~80S assembly intermediate regardless of whether the experiment was performed in the absence or presence of PuroHS , which was used to disrupt translating ribosomes . Additionally , in cells expressing Gag mutants that associate with unspliced viral RNA but are arrested at early stages of assembly , the ~80S , but not the ~10S assembly intermediate , contained unspliced HIV-1 RNA . Notably , we found no evidence for unspliced HIV-1 RNA in association with ~10S Gag at steady state under any experimental condition , despite the presence of abundant ~10S Gag in all our experiments . Thus , our data favor Model 2 , which proposes that the RNA granule derived ~80S assembly intermediate is the complex in which Gag first associates with unspliced HIV-1 RNA . Consistent with Model 2 , we also found that DDX6 and ABCE1 , which facilitate assembly and are present in RNA-granule-derived assembly intermediates , are associated with unspliced HIV-1 RNA in fractions containing the ~80S assembly intermediate , as would be expected . Also consistent with Model 2 , in chronically infected human T cells , unspliced HIV-1 RNA was found associated with ABCE1 in fractions containing the ~80S assembly intermediate . In situ studies confirmed the association of RNA granule proteins with unspliced HIV-1 RNA at PM sites of budding , and the colocalization of assembly-competent Gag with RNA granule proteins . Our in situ experiments also revealed that complexes containing the RNA granule protein DDX6 colocalized with Gag are far more numerous than P bodies , and likely correspond to small RNA granules that are visible as fluorescent foci . Finally , we demonstrated that assembling Gag uses a two-step process to localize to a subset of RNA granules that contains unspliced HIV-1 RNA , leading to formation of the ~80S assembly intermediate that contains Gag associated with unspliced HIV-1 RNA . One of these steps is dependent on binding to HIV-1 RNA , as would be expected . However , the other step that Gag uses to target to RNA granules is independent of HIV-1 RNA binding , suggesting a novel and poorly understood mechanism for localizing assembling Gag to the subset of RNA granules that contains unspliced HIV-1 RNA . Together , our data support a model in which Gag associates with unspliced HIV-1 RNA within a poorly understood subclass of host RNA granules . These findings advance our understanding of RNA packaging by identifying a candidate complex in which packaging may be initiated and by raising the possibility that packaging is initiated within host RNA granules .
To study the association of HIV-1 Gag with unspliced HIV-1 RNA in HIV-1 capsid assembly intermediates , we used a variety of previously published HIV-1 expression systems ( Fig 1A , sets I—IV ) ; these produce WT Gag or Gag mutants with known phenotypes for production of virus-like particles ( VLPs ) that either do or do not contain the genome ( Fig 1B ) . Given these known VLP phenotypes and an understanding of the specific Gag mutant defect , one can also predict phenotypes for intracellular association of Gag with unspliced HIV-1 RNA . Thus , because WT Gag expression results in release of VLPs that contain unspliced HIV-1 RNA , one would also expect to find intracellular complexes containing Gag associated with unspliced HIV-1 RNA ( Fig 1B ) . In contrast , assembly-defective Gag mutants do not produce VLPs , but whether these Gag mutants would be expected to associate with intracellular unspliced HIV-1 RNA depends on their exact defect ( Fig 1B ) . Specifically , the assembly-incompetent truncated Gag MACA mutant , which is arrested as a soluble ~10S assembly intermediate , would not be expected to associate with unspliced HIV-1 RNA because it lacks the RNA-binding NC domain ( reviewed in [14 , 25 , 28] ) . In contrast , Gag G2A , which is arrested in the cytoplasm due to a point mutation that prevents the myristoylation required for PM targeting and VLP production [29–31] , would be expected to associate with intracellular unspliced HIV-1 RNA ( Fig 1B ) , as shown previously [13] . We also examined the HIV-1 Gag Zip chimera , in which the RNA-binding NC domain of Gag is replaced with a dimerizing leucine zipper ( LZ ) that allows for capsid assembly but not RNA association . Gag Zip produces VLPs that lack unspliced HIV-1 RNA [18 , 32–34] and would not be expected to associate with unspliced HIV-1 RNA ( Fig 1B ) . Gag Zip is of interest because it forms assembly intermediates even though it fails to package unspliced viral RNA [18]; thus , Gag Zip could provide insights into how association with unspliced viral RNA can fail to occur during Gag assembly . To confirm the VLP production phenotypes of these Gag proteins and define their association with intracellular unspliced HIV-1 RNA , COS-1 cells were transfected with proviral constructs expressing WT Gag or these Gag mutants ( Set I constructs in Fig 1A ) . VLP phenotypes were determined by analyzing transfected cell lysates and VLPs for Gag levels by Western blot ( WB ) , and for copies of unspliced HIV-1 RNA by reverse transcription followed by quantitative PCR ( RT-qPCR; Fig 1C , top row ) . Association of WT Gag and Gag mutants with intracellular unspliced HIV-1 RNA was assessed by immunoprecipitation ( IP ) with antibody directed against Gag ( αGag ) followed by RT-qPCR to quantify the number of unspliced HIV-1 RNA copies associated with Gag in cell lysates ( Fig 1C , bottom row ) . Results of these assays confirmed the known phenotypes for VLP production and the expected phenotypes for association with intracellular unspliced HIV-1 RNA . Thus , these four constructs display a range of phenotypes , with assembly-competent WT Gag and assembly-defective Gag G2A associating with intracellular unspliced HIV-1 RNA , but assembly-incompetent MACA and assembly-competent Gag Zip not associating with intracellular unspliced HIV-1 RNA , as expected ( summarized in Fig 1B ) . Next we defined the spectrum of cytoplasmic complexes that contain unspliced HIV-1 RNA in the absence of assembling Gag . For this purpose , we transfected cells with MACA provirus ( Fig 2A; Set I constructs in Fig 1A ) , which expresses an otherwise full-length viral RNA encoding a truncated MACA Gag protein that is assembly-incompetent [17–19 , 35–38] , is arrested at the first assembly intermediate ( the ~10S complex; [19] ) , and does not associate with unspliced viral RNA ( Fig 1C , bottom row ) . We chose not to use a provirus that contains a premature stop codon in Gag because stop codons early in Gag typically result in synthesis of a poorly-studied N-terminally truncated Gag protein that starts from a downstream internal AUG codon [10 , 39] . Instead , we used a provirus that produces the well-studied assembly-defective MACA Gag mutant that does not associate with unspliced HIV-1 RNA ( Fig 1C ) . Note that in all our RNA quantification experiments , we removed nuclei by centrifugation , allowing us to focus on non-nuclear RNA , including cytoplasmic RNA as well as membrane-associated RNA that was solubilized with non-ionic detergent during the harvest . In our initial experiments , these lysates contained both translating and non-translating complexes , and were analyzed using velocity sedimentation on gradients that resolved complexes in the ~5S to ~150S range , followed by RT-qPCR to determine the approximate S values of complexes containing specific types of RNA . Previously , we have used a combination of mathematical approaches ( McEwen method; [40] ) and analysis of complexes with known S values to define migrations in these gradients [41] . In all our gradients , we name a complex by the S value of the fraction in which it peaks; thus a complex that spans the ~40-80S region but peaks in the ~80S region is called an ~80S complex . Wide peaks can result when particular methods are used ( e . g . when fewer fractions are taken or when harvest conditions partially disrupt the integrity of protein complexes , as discussed below ) . Here , we confirmed these expected migrations by analyzing gradient fractions for types of cellular RNA with known S values ( Fig 2B , open symbols ) . To identify small ribonucleoprotein complexes , we analyzed 7SL RNA , a component of signal recognition particle , which contains six proteins and one RNA and migrates at ~11S [42] . To identify the expected position of larger complexes , we quantified 28S ribosomal RNA ( rRNA ) , which marks the position of the large ribosomal subunit ( 60S ) , monosomes ( 80S ) , and polysomes ( 160S and larger ) . When lysates expressing the MACA provirus were analyzed under conditions that leave ribosomes intact , 7SL RNA was found almost entirely in the 10-20S region ( fractions 3/4; Fig 2B , open orange circles ) , while 28S rRNA was found in two peaks , one centered around the mathematically predicted ~80S region ( fractions 11–14 ) , representing monosomes , and one in the predicted ≥ 150S region ( fractions 19/20 ) , representing polysomes ( Fig 2B , open green circles ) . Thus , under standard harvest conditions , most ribosomes were in the form of intact monosomes and polysomes . Having confirmed the migration of particles of diverse sizes , we then examined the same fractions for unspliced HIV-1 RNA ( Fig 2C , open blue circles ) and found that it was broadly distributed in fractions ≥40S ( fractions 7 through 20 ) , with a large peak in the region corresponding to polysomes ( ≥150S , fractions 19/20 ) . Notably , unspliced HIV-1 RNA was not observed in fractions 1–6 , which contain particles of <40S and soluble complexes . As an additional marker for the soluble fraction , we also examined these gradient fractions for MACA Gag by WB ( Fig 2C , grey triangles ) ; as expected , the truncated , assembly-incompetent MACA Gag protein was found entirely in fractions 1–4 ( with its peak in fractions 1/2 ) , further confirming that this represents the soluble fraction . Thus , it appears that all detectable unspliced HIV-1 RNA is in complexes >40S ( fractions 7–20 ) that include both translating complexes ( in ~80S monosomes and a prominent ≥150S polysome peak ) as well as non-translating complexes of diverse sizes . To assess whether unspliced HIV-1 RNA is in the same fractions as mRNA , we also examined the migration of one subgenomic viral mRNA ( Tat mRNA ) and the cellular mRNA for GAPDH ( Fig 2D and 2E , respectively , open symbols ) . We found that , like unspliced HIV-1 RNA , viral and cellular mRNA migrated in the ≥ 40S region ( fractions 7–20 ) , with a prominent peak in the position of polysomes ( fractions 19/20 ) , and no significant signal in the <40S region ( fractions 1–6 ) . Thus , these findings indicated that cellular and viral mRNA ( e . g . GAPDH and Tat mRNA , respectively ) are found largely in translating and non-translating complexes of diverse sizes and are not found in the soluble fraction of the cytoplasm . Moreover , we concluded that complexes containing unspliced HIV-1 RNA have the same size distribution as complexes containing cellular mRNA ( or subgenomic viral mRNA ) when assembling Gag is absent and ribosomes are intact; additionally , to the level of our detection , unspliced HIV-1 RNA is either in translating complexes ( which are mainly polysomes ) or in non-translating ribonucleoprotein complexes ≥ 40S , but is largely absent from the soluble fraction . Our findings are consistent with a previous study that found no HIV-1 RNA in the soluble fraction of cell lysates [43] . Our data also demonstrated that following standard cell lysis and velocity sedimentation , ribonucleoprotein complexes of different sizes ( monosomes , polysomes , and SRP ) remain intact and retain their expected S values . Our next goal was to extend this analysis to lysates of cells transfected with proviruses expressing WT Gag or Gag mutants that associate with unspliced HIV-1 RNA , so that we could identify the smallest HIV-1 capsid assembly intermediate containing Gag that is associated with unspliced HIV-1 RNA by αGag IP . However , first we needed to address the problem that αGag would be expected to immunoprecipitate both non-translating complexes containing unspliced HIV-1 RNA ( e . g . assembly intermediates ) and actively translating complexes in which Gag epitopes are exposed while Gag is still being synthesized from the unspliced mRNA template . These translating and non-translating complexes would be indistinguishable in our IP-RT-qPCR analyses . Thus , to identify all non-translating complexes containing Gag associated with unspliced HIV-1 RNA in provirus expressing cells , we needed a way to eliminate complexes involved in Gag translation from our analyses since these complexes are unlikely to be involved in assembly or packaging for reasons described above . We reasoned that the best way to remove translating Gag associated with unspliced HIV-1 RNA would be to efficiently disrupt ribosomes and leave intact the complexes that contain non-translating unspliced HIV-1 RNA . Treatment of cell lysates with puromycin and high salt ( PuroHS ) has long been used to disrupt translating ribosomes , resulting in release of translating polypeptides ( e . g . nascent Gag ) from ribosomes , dissociation of functional ribosomal subunits , and release of free mRNA [44 , 45] , which then likely shifts into non-translating ribonucleoprotein complexes [46] . Thus , complexes that contain unspliced HIV-1 RNA and remain intact after PuroHS treatment would be expected to be mainly non-translating complexes . Moreover , a subset of those non-translating complexes likely corresponds to assembly intermediates containing Gag associated with unspliced HIV-1 RNA . Before utilizing PuroHS to identify all assembly intermediates containing Gag associated with unspliced HIV-1 RNA , we first determined the efficacy of this treatment for ribosome disruption by analyzing the effect of PuroHS on the HIV provirus in absence of assembling Gag . Gradient fractions of cell lysates expressing the MACA provirus ( diagrammed in Fig 2A ) and harvested after PuroHS treatment were analyzed for each RNA of interest ( Fig 2B–2E , solid symbols; note that PuroHS treatment was performed and analyzed in parallel with lysates harvested in the absence of PuroHS treatment , which are shown in Fig 2 with open symbols and described above . ) As described above , in the absence of PuroHS treatment , 28S rRNA ( a marker for the 60S large ribosomal subunit; Fig 2B , open green circles ) migrated almost entirely in the position of monosomes ( ~80S ) and polysomes ( >150S ) . In contrast , following PuroHS treatment , 28S rRNA migrated almost entirely in a ~60S peak representing the dissociated large ribosomal subunit ( Fig 2B , solid green squares ) , with almost no 28S rRNA remaining in the polysome region ( >150S ) . The near complete absence of 28S rRNA ( i . e . the large ribosomal subunit rRNA ) in the polysome region after PuroHS treatment indicated highly effective ribosome disassembly . Thus , it appears that PuroHS disassembles most monosomes and polysomes into ribosomal subunits . PuroHS treatment did not affect the migration of 7SL RNA ( Fig 2B , compare open orange circles vs . solid orange squares ) , as expected given that 7SL RNA is a component of the ribosome-independent ~11S signal recognition particle [42] . From these data , we conclude that treatment of cell lysates with PuroHS disrupts translating ribonucleoprotein complexes very effectively but has little effect on non-translating , ribosome-independent complexes , such as signal recognition particle . We also examined unspliced HIV-1 RNA in these same MACA-containing gradient fractions following PuroHS treatment , and found that > 95% of this RNA was in diverse ribonucleoprotein complexes of ≥ 30S ( fractions 5–20 ) that formed a broad peak centered at ~60S ( Fig 2C , solid blue squares ) . Overall , more unspliced HIV-1 RNA was found in the ~40S to ~60S region following PuroHS treatment , consistent with PuroHS treatment causing disassembly of ribosomes into ribosomal subunits . Comparison of the distribution of RNA obtained with and without PuroHS treatment suggested that at least 30% of total unspliced HIV-1 RNA was initially in translating polysomes , defined as complexes ≥150S that are lost upon PuroHS treatment . Given that PuroHS treatment almost completely dissociated ribosomes ( as measured by 28S rRNA ) in the ≥ 150S polysome fraction ( Fig 2B , compare solid green squares to open green circles ) , these data indicate that , following PuroHS treatment , unspliced HIV-1 RNA shifts from polysomes ( ≥150S ) into non-translating ribonucleoprotein complexes of ~30-150S , some of which are present even in the absence of PuroHS treatment and overlap in size with translating complexes ( Fig 2C , compare solid blue squares to open blue circles ) . A similar shift out of polysomes and into the ~30-150S size range of non-translating ribonucleoprotein complexes was observed for subgenomic HIV-1 Tat mRNA and cellular GAPDH mRNA ( Fig 2D and 2E , compare solid squares to open circles ) . Note that because PuroHS treatment was performed after cells were lysed and lysates were clarified to remove nuclei and large organelles , it is unlikely that PuroHS resulted in formation of stress granules or P bodies . Moreover , our finding that unspliced and spliced HIV-1 RNA as well as a cellular GAPDH mRNA shifted to smaller complexes after PuroHS treatment rather than larger complexes ( Fig 2 ) is consistent with this conclusion , given that stress granules and P bodies are very large . Together , these studies reveal that non-translating unspliced HIV-1 RNA , like cellular mRNA and subgenomic viral mRNA , are found in complexes ≥ 30S under standard harvest conditions , and that these non-translating complexes become more abundant following disruption of ribosomes in lysates with PuroHS . Our finding that , in the absence of assembling Gag , translating and non-translating unspliced HIV-1 RNA are found almost entirely in ≥ 30S complexes indicates that like cellular mRNA , HIV-1 RNA is sequestered within either translating or non-translating host ribonucleoprotein complexes . Next , we assessed the association of unspliced HIV-1 RNA with WT Gag or assembly-defective Gag in each assembly intermediate . When proviruses encoding WT Gag are expressed , much of the steady state Gag is found in the ~10S assembly intermediate , which likely contains a monomer or dimer of Gag , while the remainder is found in the RNA-granule-derived ~80S/150S and ~500S assembly intermediates . We had previously showed that because assembly-defective Gag mutants are arrested at key steps in the assembly pathway , Gag mutants can effectively trap Gag in a subset of assembly intermediates [15 , 18–20 , 47] . We began these analyses with the well-studied , targeting-defective Gag G2A mutant , described above . Others have used Gag G2A to demonstrate that the initial association of Gag with unspliced HIV-1 RNA occurs in the cytoplasm [13] . Given those previous data and our confirmation that Gag G2A associates with unspliced HIV-1 RNA in the cytoplasm by αGag IP ( Fig 1C ) , we expected Gag G2A to form at least one assembly intermediate that contains Gag associated with unspliced HIV-1 RNA . We had previously observed that Gag G2A forms the first two intermediates in the pathway , soluble ~10S Gag and the RNA-granule-derived ~80S intermediate; thus , we reasoned that either the ~10S or ~80S assembly intermediate is likely to be the intermediate in which Gag first associates with unspliced HIV-1 RNA . While the arrested ~80S Gag G2A assembly intermediate is clearly defective , it nevertheless closely resembles the WT ~80S assembly intermediate , both in its size and composition [19 , 20 , 22] and would therefore be useful for identifying the earliest assembly intermediate in which Gag first associates with unspliced HIV-1 RNA . To allow us to focus only on assembly intermediates ( which are non-translating complexes ) , we treated lysates with PuroHS during harvest to eliminate translating complexes that could contain Gag associated with translating unspliced HIV-1 RNA ( as shown in Fig 2 ) . We first confirmed the known distributions of MACA and Gag G2A protein by analyzing PurosHS-treated lysates of cells transfected with proviruses to express MACA or Gag G2A ( Set I constructs in Fig 1A ) to similar steady state levels ( Fig 3A ) . As expected , when these lysates were analyzed by velocity sedimentation followed by WB of gradient fractions , the distribution of the MACA vs . Gag G2A protein across the gradient differed dramatically , with MACA protein forming only the ~10S intermediate ( Fig 3B , MACA WB , fractions 1–4 , with a trail in 5–7 ) , while Gag G2A formed both the soluble ~10S intermediate and a complex that spans the ~60-80S region and corresponds to the ~80S assembly intermediate ( Fig 3B , G2A WB , fractions 7–13 , peak in fraction 10 ) . In contrast , both MACA and Gag G2A lysates displayed the same distribution of non-translating unspliced HIV-1 RNA across the gradient ( Fig 3B , graph ) , with unspliced HIV-1 RNA from both lysates mainly in the ~40-80S fractions ( and peaking at ~60S ) and almost no unspliced HIV-1 RNA in the soluble fractions ( fractions 1–4 ) . Thus , the only population of Gag protein that co-migrated with the non-translating unspliced viral RNA peak was Gag G2A in the ~60-80S fractions , with neither soluble Gag G2A nor soluble MACA co-migrating with non-translating unspliced HIV-1 RNA . Having demonstrated that Gag G2A in the ~60-80S region of the gradient co-migrates with a population of unspliced viral RNA , we next asked whether Gag G2A in this region is actually associated with unspliced HIV-1 RNA , or simply co-migrates in a separate complex . For this purpose , gradient fractions from Fig 3B were subjected to αGag IP , followed by quantitation of viral RNA in IP eluates ( Fig 3C ) . Note that all IP analyses in this study were performed under native conditions , and are thus expected to pull down the protein targeted by the antibody as well as any other components that are stably associated with the target protein through direct or indirect interactions . Even though αGag coimmunoprecipitated MACA effectively ( Fig 1C ) , no unspliced HIV-1 RNA was associated with MACA by αGag IP in any fraction ( Fig 3C ) , as expected given that the MACA peak does not co-migrate with unspliced HIV-1 RNA ( compare Fig 3B graph to Fig 3B blot ) . Similarly , soluble Gag G2A protein ( in the ≤ 20S region ) was associated with little or no unspliced HIV-1 RNA by coimmunoprecipitation ( coIP ) ( Fig 3C ) . Indeed , αGag IP from fractions containing ≤ 20S Gag G2A ( fractions 1–4 in Fig 3C ) contained 2 . 2 copies of unspliced HIV-1 RNA per fraction per cell . Given that the limit of detection in gradient IP samples is 1 copy per cell ( as indicated by standard curves run with every assay ) and given that our samples are loaded at the top of the gradient and would be expected to leave behind some contamination in the top fractions ( fractions 1 and 2 ) , these data indicate that the substantial pool of soluble Gag G2A proteins in fractions 1–4 is not associated with unspliced HIV-1 RNA . In contrast , the αGag IP of Gag G2A from ~80S fractions ( fractions 11–14 in Fig 3C ) contained an average of 285 copies of unspliced HIV-1 RNA per fraction per cell . Thus , to the limit of detection of our very sensitive assay , we were unable to identify a significant amount of unspliced HIV-1 RNA associated with soluble Gag G2A; instead , we found that unspliced HIV-1 RNA was strongly associated with a pool of Gag G2A that peaks at ~80S and likely corresponds to the ~80S intermediate ( Fig 3C , graph ) , which is the second intermediate in the assembly pathway . These findings suggest that the ~80S assembly intermediate is the first intermediate in the previously described assembly pathway that contains Gag associated with unspliced HIV-1 RNA . Previously , we had demonstrated that the ~80S assembly intermediate formed by Gag G2A and WT Gag contains host proteins , including the RNA granule protein DDX6 and the cellular ATPase ABCE1 , as shown by coIP of both Gag proteins with antibodies to DDX6 ( αDDX6; [22] ) and ABCE1 ( αABCE1; [21] ) . DDX6 , a cellular RNA helicase found in P bodies , is involved in mRNA silencing and mRNA storage , but is not typically associated with actively translating mRNA [48] . The association of HIV-1 Gag with DDX6 is RNase-sensitive [22] , thus DDX6 does not bind directly to Gag . Importantly , the presence of DDX6 in the ~80S assembly intermediate suggests that Gag co-opts a host RNA granule to generate the ~80S intermediate [22] . Thus , if the ~80S complex containing Gag G2A associated with unspliced HIV-1 RNA in Fig 3C corresponds to the ~80S assembly intermediate , then unspliced viral RNA in the ~80S fractions should be associated with DDX6 . In PuroHS-treated lysates of cells expressing G2A provirus ( Fig 3D ) , unspliced HIV-1 RNA was again observed almost exclusively in the ~40-80S region of gradients ( Fig 3E , compare to Fig 3B ) . Additionally , IP with αDDX6 revealed that unspliced HIV-1 RNA is associated with DDX6 in a broad complex that peaks at a point slightly larger than ~80S , confirming that unspliced HIV-1 RNA is associated with DDX6 in fractions containing the ~80S assembly intermediate ( Fig 3F ) . While we would not expect DDX6 to be associated with ribosomes , our use of PuroHS-treated lysates allows us to confirm that the ~80S complex containing DDX6 associated with unspliced HIV-1 RNA in Fig 3F is not a translating complex . Taken together with our previous coIP analyses showing that Gag G2A is associated with DDX6 [22] , these data suggest that the Gag G2A ~80S assembly intermediate likely contains non-translating unspliced HIV-1 RNA in association with Gag G2A and the RNA granule protein DDX6 . These data are consistent with the ~80S assembly intermediate , not the ~10S intermediate , being the first assembly intermediate that contains unspliced HIV-1 RNA in association with Gag G2A . Although the Gag G2A mutant has been used by others to study the initial association of Gag G2A with unspliced HIV-1 RNA [13] , it is formally possible that Gag G2A , being a mutant , forms an ~10S complex containing Gag G2A associated with unspliced HIV-1 RNA that is unstable and was therefore not detected at steady state in our experiments; alternatively , the ~80S intermediate that contained Gag G2A associated with unspliced HIV-1 RNA could be an abnormal complex unique to the G2A mutant . For these reasons , we next asked which assembly intermediates formed by WT Gag contain unspliced HIV-1 RNA associated with Gag . Like Gag G2A , WT Gag forms an ~80S assembly intermediate; unlike Gag G2A , WT Gag also forms an ~500S assembly intermediate after membrane targeting [19 , 20 , 22] . However , to ask whether Gag is associated with unspliced HIV-1 RNA in all of these assembly intermediates , we needed a method for immunoprecipitating multimerized ~500S Gag because diverse Gag antibodies fail to immunoprecipitate multimerized Gag due to epitope masking , even though they successfully immunoprecipitate soluble Gag and oligomerized Gag [49] . For this reason , we utilized antibodies to GFP ( αGFP ) to immunoprecipitate Gag GFP from lysates of cells co-transfected with GFP-tagged , codon-optimized Gag ( Gag GFP ) and VIB , a modified proviral construct that expresses a viral RNA that contains all the signals needed for packaging [12 , 13] and has been successfully coimmunoprecipitated with αGFP [13] . Notably , the V1B viral genome encodes a truncated assembly-incompetent Gag that does not interfere with assembly of Gag GFP expressed in trans . Co-transfection of Gag GFP and the V1B genome ( Set II constructs in Fig 1A ) has been well vetted in live imaging and biochemical studies [12 , 13] . Moreover , use of this in trans system would also allow us to express a single well-studied V1B genomic construct with different Gag constructs ( using qPCR oligos that detect cDNA made from unspliced VIB RNA but not cDNA made from Gag GFP mRNA ) . As expected , we found that co-transfection of Gag GFP and V1B plasmids resulted in the same phenotypes for VLP production and Gag association with intracellular unspliced viral RNA that were observed for proviral constructs in Fig 1B ( S1A–S1C Fig ) . To identify complexes that contain WT Gag in association with unspliced viral RNA , we analyzed cell lysates co-expressing either GFP-tagged WT Gag or Gag G2A along with the V1B genomic construct . To allow a direct comparison with experiments in Fig 3 , we treated lysates with PuroHS to disrupt monosomes and polysomes . In this experiment , unlike in Fig 3 experiments , we used a velocity sedimentation gradient that separately resolves complexes of ~80S , ~500S , and ~750S , which are formed by WT Gag ( reviewed in [14] ) . Unlike the gradients shown in Fig 3 , the gradients used here do not resolve soluble proteins and small complexes well , since complexes ranging from ~10S to ~80S are distributed across only a few fractions . Levels of intracellular Gag protein were similar at steady state for both WT Gag and Gag G2A , as were levels of unspliced viral RNA ( Fig 4A ) . Gag G2A protein formed only the ~10S and ~80S intermediates ( as observed in Fig 3B ) ; in contrast , at steady state , WT Gag protein formed both of these intermediates as well as the expected ~500S late assembly intermediate and ~750S completed immature capsid ( Fig 4B , WB ) , as shown previously [15 , 19 , 20] . Notably , in the WT Gag and Gag G2A lysates , nearly all the unspliced viral RNA was in the ~80S assembly intermediate ( Fig 4B , graph ) . Additionally , IP with αGFP revealed that both WT and Gag G2A were associated with unspliced viral RNA in the ~80S assembly intermediate ( Fig 4C , graph ) . Moreover , IP with αGFP showed that WT Gag was also associated with unspliced viral RNA in ~500S and ~750S complexes ( Fig 4C , graph ) , both of which are formed by WT Gag but not targeting-defective Gag G2A , and likely represent the ~500S assembly intermediate and the ~750S completed capsid , respectively . Most importantly , although WT Gag forms a prominent ~10S assembly intermediate , αGFP IP of this complex revealed no unspliced viral RNA associated with this intermediate ( Fig 4C; 0 . 4 copies of unspliced viral RNA per cell in the fraction 1 IP ) , as was the case for Gag G2A ( Figs 3C and 4C ) . Note that since our lysates are loaded with sample at the top of the gradients , this very low value includes potential contamination from the top-loaded lysate . We also observed that the amount of WT Gag in the ~80S region is smaller following PuroHS treatment than in previous studies in which we did not treat with PuroHS [19 , 20 , 22] . This suggested that PuroHS , while having the advantage of disrupting translating complexes , may also cause some disruption of the ~80S assembly intermediate , leading to Gag-containing complexes that peak in the ~80S region but are broader , and less uniform in size and composition , than in our previous studies . Thus , for technical reasons the size of the ~80S assembly intermediate appears to be more approximate here than in our previous studies . Given this caveat ( which is addressed by harvest in the absence of PuroHS below ) , our data are consistent with the ~80S assembly intermediate being the smallest assembly intermediate that is detected at steady state and contains WT Gag associated with unspliced HIV-1 RNA . Additionally , we found that WT Gag is also associated with unspliced HIV-1 RNA in a ~500S late assembly intermediate and the fully assembled ~750S completed immature capsid ( Fig 4C ) , as would be expected . Thus far in this study , we had shown that 1 ) both Gag G2A and WT Gag associate with unspliced viral RNA in an ~80S complex ( Figs 3C and 4C ) ; 2 ) that this ~80S complex likely contains the RNA granule protein and assembly facilitator DDX6 ( Fig 3F ) and therefore likely corresponds to the previously described RNA-granule-derived ~80S capsid assembly intermediate; and 3 ) as expected , WT Gag is associated with unspliced viral RNA in the 80S assembly intermediate as well as the ~500S assembly intermediate and the fully assembled ~750S completed capsid ( Fig 4C ) . If the ~80S and ~500S complexes that contain WT Gag associated with unspliced viral RNA correspond to assembly intermediates , we would also expect unspliced viral RNA in these complexes to be associated with another host protein marker of assembly intermediates , the cellular facilitator of assembly ABCE1 [19 , 20 , 22] , even after PuroHS treatment . Our previous data suggested that ABCE1 is also found in a subclass of small host RNA granules that are co-opted by HIV-1 [22] and in ~80S and ~500S assembly intermediates that are derived from co-opted host RNA granules and contain Gag in association with ABCE1 by coIP [15 , 19 , 21 , 22] . Additionally , we had shown that unspliced viral RNA is associated with ABCE1 in ~80S/150S and ~500S assembly intermediates by αABCE1 IP of lysates that were not treated with PuroHS [19] . However , because ABCE1 is known to be critical for translation termination ( reviewed in [50] ) , here we asked whether ABCE1 is associated with unspliced viral RNA following treatment of lysates with PuroHS , which fully disassembles polysomes and monosomes ( Fig 2 ) . Indeed , analysis of PuroHS-treated lysates from cells expressing WT Gag GFP and the V1B genomic construct in trans revealed that αABCE1 immunoprecipitated unspliced viral RNA from the ~80S/150S and ~500S regions of the gradient ( S2A–S2C Fig ) , which are known to contain the ~80S and ~500S assembly intermediates . Thus , PuroHS treatment allows us to conclude that the ~80S and ~500S complexes containing ABCE1 associated with unspliced HIV-1 RNA are not monosomes and polysomes , and instead likely represent ABCE1-containing assembly intermediates . However , a disadvantage of using PuroHS is that the high salt concentration appears to cause broader peaks due to partial disruption of assembly intermediates . Such partial disruption of the ~80S Gag-containing complex could explain why only a small amount of Gag is observed in the ~80S region following PuroHS treatment , compared to the large and distinct ~80S Gag peak observed in earlier studies where PuroHS was not used [19 , 20 , 22] . We also observed that the peak of unspliced HIV-1 RNA immunoprecipitated by αABCE1 lies slightly to the right of the main ~80S Gag peak ( compare S2C Fig , graph to S2B Fig , WB ) . This could be due to heterogeneity of complexes in this region or variability in the accessibility of epitopes in this region . While these migration issues will need to be explored further in the future , the data suggest that unspliced viral RNA is associated with both WT Gag ( Fig 4C ) and ABCE1 ( S2C Fig ) in the ~80S/150S and the ~500S fractions . So far , our data supported a model in which HIV-1 Gag first associates with HIV-1 RNA within RNA granule-derived complexes found in the ~80S and ~500S regions of the gradient ( Model 2 ) . Moreover , while we repeatedly identified a large pool of soluble Gag , we did not find it associated with unspliced HIV-1 RNA , arguing against a model in which soluble ~10S Gag is associated with unspliced HIV-1 RNA ( Model 1 ) . However , it is possible that our approaches prevented us from detecting a soluble complex containing Gag associated with unspliced HIV-1 RNA . Approaches that could have been problematic include 1 ) our use of only the Gag G2A mutant , rather than other mutants , to trap a small Gag-containing complex associated with unspliced viral RNA , 2 ) transfection of non-human primate COS-1 cells , which we used to minimize problems resulting from endocytosis of released virus as previously described [19] , and 3 ) treatment with PuroHS to eliminate translating complexes from our analyses . We addressed the first of these concerns by examining a second assembly-defective Gag construct that associates with intracellular unspliced viral RNA . For this purpose , we analyzed the W184A/M185A Gag mutant , which fails to complete multimerization , but targets to the PM [18 , 19 , 49 , 51–54] unlike the Gag G2A mutant . The W184/M185 residues are known to be important for interhexameric CA-CA dimer interface contacts , which are necessary for subsequent steps in assembly [19 , 52–54] . We had previously shown that when expressed from an HIV-1 provirus , Gag W184A/M185A is arrested in the form of an ~80S intermediate that is associated with the PM [19] . Here we analyzed PuroHS treated lysates of COS-1 cells transfected with Gag GFP W184A/M185A and the V1B genomic constructs in trans ( Set II constructs in Fig 1A; also in S3A Fig ) . As expected , under these conditions Gag W184A/M185A is assembly-defective but associates with intracellular unspliced viral RNA ( S3B and S3C Fig ) . We also confirmed that Gag W184A/M185A is arrested as an ~80S assembly intermediate ( S3D Fig ) . We then used RT-qPCR to identify all complexes in lysates that contain unspliced viral RNA , and the subset of those complexes containing Gag W184A/M185A in association with unspliced viral RNA by αGFP coIP ( Fig 4D–4F ) . To maximize our ability to detect small complexes , we used a velocity sedimentation gradient that provided high resolution in the ~10S - 150S range ( similar to Fig 2B–2E , Fig 3B and 3C and Fig 3E and 3F; and unlike Fig 4B and 4C ) . We observed essentially no unspliced viral RNA in fractions <30S; instead all unspliced viral RNA was found in fractions ≥30S ( Fig 4E ) . Moreover , αGFP IP of gradient fractions revealed that Gag GFP W184A/M185A was associated with unspliced viral RNA in a complex that peaks at ~80S and likely corresponds to the ~80S assembly intermediate , as observed for WT Gag GFP analyzed in parallel ( Fig 4F; also compare to Fig 4C ) . Note that both the complex containing WT Gag associated with unpsliced viral RNA and the complex containing Gag WM184A/M185A associated with unspliced viral RNA fit the definition of ~80S complexes given that both clearly peak in the 80S region in Fig 4F . However , since these peaks are relatively wide and could therefore include smaller complexes within them , we also quantified 18S rRNA , which is found in the 40S small ribosomal subunit , to define the smallest complex that could be hidden within this peak . We found that the 40S marker is first found in fraction 4 and peaks in fraction 5 ( Fig 4F ) ; thus , the smallest complex that could be associated with unspliced viral RNA but hidden within the broad ~80S peak is ≥~30-40S . Given that the 40S ribosomal subunit contains one RNA and 33 cellular proteins , complexes in this size range are likely to be ribonucleoprotein complexes that contain numerous host proteins rather than a solely a dimer of Gag associated with unspliced viral RNA . From this analysis , which maximizes resolution of soluble complexes present at steady state , we conclude that most likely the smallest complex containing unspliced viral RNA in association with WT Gag or assembly-defective viral-RNA-binding Gag mutants ( Gag G2A and Gag W184A/M185A ) is either the ~80S assembly intermediate or a complex of ~30-40S , at the smallest , if a buried peak is present . Although we cannot exclude the possibility that a buried complex of ~30-40S is present , we favor the hypothesis that the broad ~80S peak represents an ~80S complex containing Gag associated with unspliced viral RNA that has been partially disrupted due to exposure to high salt during PuroHS treatment; this possibility is further supported by experiments described below . Notably , once again we did not detect a discrete ~10-20S peak of Gag associated with unspliced HIV-1 RNA , arguing against soluble Gag being associated with unspliced HIV-1 RNA to the limit of our detection . To address the possibility that use of non-human primate COS-1 cells prevented us from identifying smaller complexes containing Gag-associated with unspliced viral RNA , we repeated our analyses in human 293T cells that were transfected to express WT Gag GFP and the V1B genomic construct ( Fig 5A ) and were harvested , using PuroHS treatment of lysates , at an early time point to avoid virus endocytosis . We observed that the amount of Gag in the ~80S peak by WB following harvest with PuroHS treatment was smaller than in a previous analysis of 293T cells that were harvested without PuroHS treatment ( compare Fig 5B WB in the current study to Fig 11B WB in [20] ) . We hypothesize that this is because high salt partially disrupts the ~80S assembly intermediate , as described above . Despite the possible partial disruption of the ~80S intermediate , αGFP IP revealed that WT Gag is associated with unspliced viral RNA in an ~80S complex that likely corresponds to the ~80S assembly intermediate in 293T cells ( Fig 5C ) ; additionally , Gag in the ~500S assembly intermediate from these cells is also associated with unspliced viral RNA ( Fig 5C ) . Notably , the large amount of soluble Gag observed in ~10S fractions of 293T lysates ( Fig 5B , WB ) was not associated with unspliced viral RNA by αGFP IP ( Fig 5C ) . Two other observations are worthy of comment . First , once again the ~80S complex that contains Gag associated with unspliced HIV-1 RNA is a relatively wide complex ( Fig 5C ) ; this is consistent with the disruptive effect of PuroHS , especially given results described below . However , further studies will be needed to show this definitively . Second , a ~750S completed capsid was not observed in 293T lysates by WB or by αGFP IP of unspliced viral RNA unlike in COS-1 cells ( compare Fig 5C to Fig 4C ) . This was not surprising since completed immature capsids undergo rapid budding and release from 293T cells and therefore are typically not observed in cell lysates , especially when harvested at early time points ( see Fig 10 in [19] ) . Overall , the results we obtained in 293T cells were similar to those obtained in COS-1 cells ( compare Fig 5B and 5C to WT Gag in Fig 4B and 4C ) . Additionally , we addressed the possibility that PuroHS treatment of cell lysates could lead to mis-identification of the earliest assembly intermediate containing Gag associated with unspliced viral RNA , for example by causing broadening of the ~80S peak with masking of a complex smaller than ~80S that contains Gag associated with unspliced viral RNA ( as alluded to in the description of Fig 4F above ) . Cells transfected to express WT Gag GFP and the V1B genomic construct in trans ( Fig 5D ) were harvested without PuroHS treatment , leaving monosomes and polysomes intact as shown above ( Fig 2B–2E , open symbols ) . Analysis by velocity sedimentation once again demonstrated that almost no unspliced viral RNA was present in the <40S fractions ( Fig 5E ) , and WT Gag was only associated with unspliced viral RNA by αGFP IP in a large peak that corresponds to the ~80S assembly intermediate and in a small peak that likely represents the ~500S late assembly intermediate ( fractions 12–14 , Fig 5F ) . Note that when Gag GFP is expressed with the V1B genomic construct in trans , αGFP would be expected to immunoprecipitate both translating Gag GFP associated with Gag GFP mRNA and non-translating Gag GFP associated with unspliced V1B RNA; however RT-qPCR analysis of αGFP IP samples should not detect the translating complexes since our qPCR oligos only detect V1B genomic RNA and not the mRNA from Gag GFP supplied in trans . Therefore , the ~80S complex observed in Fig 5F does not represent monosomes , and should instead represent a non-translating complex containing assembling WT Gag GFP associated with unspliced HIV-1 RNA . Importantly , in this experiment , the ~80S assembly intermediate that was observed in the absence of PuroHS , and contains Gag associated with nontranslating unspliced viral RNA Gag , is defined by a much narrower and more homogenous peak than the corresponding complex detected after PuroHS treatment of comparable lysates ( compare ~80S complexes in Fig 5F vs . Fig 4C ) . This is consistent with the hypothesis that PuroHS treatment causes the ~80S assembly intermediate to partially dissociate . Thus , our studies of WT Gag expressed with the V1B genome in trans harvested with ribosomes intact once again identify the ~80S assembly intermediate as the smallest complex containing Gag and unspliced viral RNA; moreover , these data support a model in which the smallest complex that contains Gag associated with unspliced viral RNA is an assembly intermediate of relatively uniform ~80S size when cells are harvested in a more physiological buffer without high salt treatment . Previously , we showed by coIP that , in human H9 T cells chronically infected with HIV , Gag associates with ABCE1 and DDX6 in complexes that correspond to the ~80S and ~500S assembly intermediates [22] . Here we analyzed lysates of these cells to determine whether the ~80S intermediate is the smallest assembly intermediate that contains HIV-1 Gag associated with unspliced HIV-1 RNA in HIV-1-infected human T cells . PuroHS treatment was used in these analyses because these cells express proviruses ( with Gag produced in cis ) ; thus , in these cells , unspliced HIV-1 RNA would be found in both non-translating and translating complexes that contain Gag , and PuroHS would allow us to examine only the non-translating complexes . Because these chronically infected cells express native Gag , not Gag GFP , we did not immunoprecipitate with αGFP; nor did we immunoprecipitate with αGag since Gag epitopes are masked during multimerization , as described above . Instead , these IP analyses were performed with αABCE1 , which we had shown to be effective in immunoprecipitating unspliced viral RNA in the ~80S and ~500S assembly intermediates ( S2C Fig ) . Our analyses revealed that , in chronically infected H9 cells , the vast majority of the non-nuclear non-translating unspliced HIV-1 RNA is in ~80S and ~500S complexes , and unspliced HIV-1 RNA in those complexes is associated with ABCE1 by IP ( Fig 6A–6C ) . Gag was also present in the ~80S and ~500S regions of the gradient by WB ( Fig 6B ) , as shown previously ( Fig 3 in [22] ) ; moreover those previous data had demonstrated that αABCE1 coimmunoprecipitated Gag in ~80S and ~500S assembly intermediates from chronically infected H9 T cells ( Fig 3 in [22] ) . Thus , these data suggest that , in chronically infected human T cells , ABCE1 in ~80S and ~500S assembly intermediates is associated with unspliced HIV-1 RNA . Two additional observations are of interest in the analysis of chronically infected human T cells . First , some unspliced HIV-1 RNA was observed in fractions 2–5 of this gradient ( Fig 6B ) ; however , additional experiments suggested that the unspliced HIV-1 RNA in these fractions was not associated with Gag ( S4 Fig ) . In those experiments , lysates from chronically infected H9 T cells were subjected to IP with αGag , which detects Gag multimers poorly as noted above , but detects soluble Gag and small Gag oligomers effectively . In gradients of infected T cells , αGag immunoprecipitated fewer than 1 copy of unspliced HIV-1 RNA per cell from the ~10-40S region and immunoprecipitated 100-fold more copies of unspliced HIV-1 RNA from both the ~80S region and the ~500S region ( S4 Fig ) . Thus , these additional experiments suggest that unspliced HIV-1 RNA in the first few fractions of Fig 6B is unlikely to be in small Gag-associated complexes , and is more likely to have resulted from high-salt-mediated disruption of ~80S assembly intermediates . Separate from that issue , it was interesting that at steady state in chronically infected H9 T cells , the ratio of non-translating unspliced HIV-1 RNA in 80S vs . 500S complexes , as determined by RNA peak height , was shifted towards the ~500S peak , compared to transfected COS-1 or 293T cells ( e . g . compare ~80S to ~500S RNA peak height in Fig 6B vs Figs 4B and 5B ) . Such variability in the ~500S to ~80S ratio could reflect kinetic differences in efficiency of assembly or budding in different cell types and/or between experiments . For example , human T cells could form the ~500S assembly intermediate more efficiently or could undergo virus budding less efficiently; alternatively there could be experiment-to-experiment differences in formation of these transient , highly dynamic assembly intermediates and in budding kinetics . While further studies of these cells will be needed , our initial studies of these chronically infected H9 T cells support a model in which the majority of HIV-1 RNA in HIV-1 infected human T cells is present in high molecular weight complexes . To generate an RNA that is packaging defective , others have utilized a genomic construct in which stem loops 3 and 4 of the RNA packaging element psi ( Ψ; which contains four stem loops ) , are deleted ( ΔΨ; [12 , 13] ) . Previous studies in 293T cells revealed only a 3–4 fold reduction in the intracellular association of WT Gag GFP with unspliced ΔΨ V1B viral RNA , compared to WT unspliced viral RNA; the overlapping error bars in these data emphasize the modest nature of the defect ( Fig 1 in [13] ) . If the ~80S complex is the first assembly intermediate in which Gag associates with unspliced HIV-1 RNA , we would predict that cells expressing WT Gag and an unspliced HIV-1 RNA that is profoundly packaging-defective would generate an ~80S assembly intermediate that contains WT Gag but little or no packaging-incompetent unspliced HIV-1 RNA . However , if the packaging defect of a co-transfected HIV-1 RNA is minimal , a less dramatic ( and possibly insignificant ) defect in association of ~80S Gag and unspliced HIV-1 RNA would be expected , making such a construct less useful for testing the RNA granule model of packaging . Therefore , we first sought to define the magnitude of the packaging defect of the ΔΨ V1B genomic construct . Cells were co-transfected to express WT Gag GFP and either the ΔΨ or WT V1B genome ( Set II constructs in Fig 1A ) . Cell lysates and VLPs were analyzed by WB for Gag , and by RT-qPCR for unspliced V1B viral RNA ( S5A Fig ) . Intracellular WT Gag GFP and intracellular unspliced viral RNA were expressed to similar steady state levels for both groups , but the cells expressing ΔΨ viral RNA produced somewhat fewer VLPs by Gag WB in two independent experiments . When the results of these two experiments were averaged and normalized to intracellular unspliced viral RNA levels , we observed a reduction in ΔΨ unspliced viral RNA in VLPs to 29% relative to WT RNA in VLPs analyzed in parallel ( S5A Fig , Efficiency of ΔΨ packaging graph ) . When the results were normalized to both intracellular unspliced viral RNA levels and VLP Gag levels , a reduction in ΔΨ unspliced viral RNA in VLPs to 46% was observed . This two- to three-fold defect in packaging of ΔΨ observed in our VLP analyses is consistent with the similar modest defect observed by others for WT Gag association with intracellular ΔΨ unspliced viral RNA [13] . Since our analyses detected >20 , 000 copies of unspliced HIV-1 RNA in the ~80S peak per 1000 cells ( Fig 4C ) , a two- to three-fold reduction in this peak is unlikely to display the significance needed to unambiguously detect a reduction in candidate packaging complex formation by the ΔΨ construct . For this reason , we concluded that the ΔΨ3/4 packaging defect is too modest to be useful in testing the RNA granule model of packaging; instead , we opted to analyze a Gag construct that displays more impressive packaging defects . In contrast to the modest VLP packaging defect displayed by ΔΨ viral RNA ( S5A Fig ) , the assembly-competent Gag Zip construct releases VLPs that closely resemble WT VLPs morphologically [18 , 32 , 34 , 55] but display a profound defect in HIV-1 RNA packaging , as observed by others [34] and by us ( Fig 1C ) . Therefore this GagZip construct , which contains LZ in place of NC , provides a statistically significant packaging defect that should allow us to test whether HIV-1 Gag has evolved one or more mechanisms for targeting to RNA granules that contain unspliced HIV-1 RNA . Given the longstanding observation that the NC domain is required for association of Gag with unspliced HIV-1 RNA , we hypothesized that NC would be critical for targeting Gag to the subset of RNA granules that contains unspliced HIV-1 RNA . Additionally , other domains could also be involved in localizing Gag to a broader class of RNA granules , as suggested by our earlier studies of Gag Zip [22] . Previously , Gag Zip was used to demonstrate that NC has two functions during immature capsid assembly—NC binds specifically to unspliced HIV-1 RNA and also promotes oligomerization of Gag via non-specific RNA association [33 , 34] . Because LZ promotes direct protein-protein interactions , it substitutes for the oligomerization function of NC; thus , Gag Zip is assembly-competent and produces VLPs [18 , 34 , 56] . However , because LZ does not bind to RNA , these Gag Zip VLPs lack viral RNA or other RNA [34 , 56] . Interestingly , we found previously that , despite its inability to interact with RNA ( Fig 1C ) , Gag Zip forms the ~80S and ~500S ABCE1- and DDX6-containing assembly intermediates [18 , 22] . These data raised the possibility that Gag Zip contains a determinant that allows it to localize to a broad class of ABCE1- and DDX6-containing RNA granules , but not to the subset of these granules that contains unspliced viral RNA because it lacks the viral-RNA-binding NC domain . Before testing this hypothesis , we first confirmed that Gag Zip GFP , like Gag Zip , produces VLPs that lack unspliced viral RNA when co-transfected with the V1B plasmid in trans ( S5B Fig ) . We next analyzed PuroHS-treated lysates of cells transfected with WT Gag or Gag Zip GFP , and the V1B genomic construct in trans ( Set II constructs in Fig 1A ) . Both Gag proteins were expressed at similar steady state levels , as was the V1B RNA ( Fig 7A ) , and unspliced viral RNA was primarily in an ~80S complex in both cases ( Fig 7B graph ) . WB confirmed that Gag Zip GFP forms a ~500S complex ( Fig 7B WB ) , consistent with the previously described ~500S Gag Zip assembly intermediate [18] . In addition , previously we confirmed that Gag Zip also forms the ~80S assembly intermediate , albeit at lower levels than for WT Gag as observed previously ( see dark exposures of Fig 4 and Fig 5 in [18] ) . Notably , αGFP coimmunoprecipitated unspliced viral RNA from ~80S and ~500S fractions of cells expressing WT Gag GFP , but failed to coIP unspliced viral RNA from any fraction of the Gag Zip GFP gradient ( Fig 7C , compare to WT in Fig 4C ) . Controls using IP followed by WB showed that αGFP immunoprecipitated Gag Zip GFP protein as effectively as WT Gag GFP protein from ~80S and ~500S fractions ( S5C Fig ) , so the failure to coimmunoprecipitate unspliced viral RNA with Gag Zip GFP cannot be attributed to reduced αGFP IP efficiency . These data argue that both WT Gag and Gag Zip localize to ~80S RNA granules to form ~80S assembly intermediates , but WT Gag stably associates with a subset of these RNA granules that contains unspliced viral RNA , while Gag Zip does not stably associate with this unspliced-viral-RNA-containing subset even though it associates with a related but broader class of RNA granules . From these data , we conclude that while GagZip is capable of forming an RNA-granule-derived ~80S assembly intermediate , in the absence of NC this ~80S assembly intermediate will contain Gag but not unspliced HIV-1 RNA . Thus , these data suggest that NC directs Gag to the subset of the granules that contains unspliced viral RNA . Importantly , our data also suggest that a region of Gag outside of NC ( present in Gag Zip but not in MACA ) is responsible for bringing Gag to a broader class of RNA granules , most of which lack viral RNA . Finally , we asked whether a heterologous domain that confers binding to unspliced viral RNA ( here termed a heterologous viral-RNA-binding domain ) is sufficient to restore targeting of Gag Zip to the subset of the granules that contains unspliced viral RNA . Because the V1B genomic RNA contains binding sites for the bacteriophage MS2 ( referred to as MS2 stem loops ) , we reasoned that the MS2 coat protein ( MCP ) , which binds with high affinity to MS2 stem loops ( reviewed in [57] ) , could serve as a heterologous viral-RNA-binding domain if inserted into Gag Zip . Therefore , we generated a construct , here called Gag Zip MCP , in which MCP is fused to the C-terminus of Gag Zip . Additionally , we showed that Gag Zip MCP forms VLPs that contain the V1B genome , like WT Gag and unlike Gag Zip ( S5B Fig ) . When the VIB genomic construct was co-transfected either with Gag Zip or Gag Zip MCP to similar steady state levels ( Fig 7D ) and analyzed on gradients , unspliced viral RNA was found largely in the ~80S fractions in both cases ( Fig 7E ) . However , αGFP coimmunoprecipitated unspliced viral RNA from the ~80S and ~500S assembly intermediates formed by Gag Zip MCP GFP , but not from the corresponding complexes formed by Gag Zip , analyzed in parallel ( Fig 7F ) . Thus , fusion to MCP , a heterologous viral-RNA-binding domain , redirected Gag Zip to the specific subset of RNA granules that contains unspliced viral RNA , resulting in formation of ~80S and ~500S assembly intermediates containing unspliced viral RNA associated with Gag Zip MCP . In keeping with this , the MCP fusion also restored unspliced viral RNA in released VLPs ( Fig 7F and S5B Fig ) . These data confirmed that either a native or heterologous viral-RNA-binding domain is required to target Gag to the subset of granules containing unspliced viral RNA . Previously , to confirm coIP studies suggesting that assembling Gag associates with a host RNA granule , we had used in situ double labeling immunoelectron microscopy ( IEM ) to show colocalization of WT Gag ( as well as some assembly-competent Gag mutants ) with ABCE1 , DDX6 , and AGO2 ( another RNA granule protein ) at PM sites of assembly and budding in intact cells [15 , 19 , 20 , 22] . Here we asked whether we could use a second in situ technique to confirm that Gag associates with RNA granule proteins in intact cells . To do this , we sought to identify sites where Gag colocalizes with DDX6 or ABCE1 using the proximity ligation assay ( PLA ) . PLA produces fluorescent spots at sites where two proteins are within 40 nm of each other in situ . Briefly , this method uses species-specific secondary antibodies conjugated to complementary oligonucleotide probes to detect primary antibodies bound to two proteins of interest; when a linker and other reagents are added , the probes on the secondary antibodies anneal if the two proteins of interest are ≤ 40 nm apart , thereby forming the template for a rolling circle amplification product that is recognized by a fluorophore-conjugated oligonucleotide [58] ( Fig 8A ) . We hypothesized that if assembling Gag associates with RNA granules containing ABCE1 and DDX6 as indicated by our biochemical studies , then the assembling Gag constructs ( WT Gag , Gag G2A , and Gag Zip ) should produce abundant Gag-DDX6 and Gag-ABCE1 PLA spots relative to assembly-incompetent MACA Gag , which fails to associate with granules ( Fig 3A–3C ) and does not coIP with DDX6 or ABCE1 [19 , 22] . To test this hypothesis , 293T cells were transfected with WT vs . mutant provirus ( Set I constructs in Fig 1A ) and analyzed for Gag-DDX6 or Gag-ABCE1 colocalization by PLA . Concurrent Gag IF allowed us to confirm that the vast majority of PLA spots were observed in Gag-expressing cells , and to choose fields for quantitation with comparable Gag levels . For cells expressing WT Gag , Gag G2A , or Gag Zip , these quantified fields contained ~50 Gag-DDX6 PLA spots per cell , three-fold more than for cells expressing MACA ( Fig 8 ) . Similar results were observed for Gag-ABCE1 PLA spots ( Fig 9 ) . Some PLA background was expected in cells transfected with MACA provirus , given that much of the non-nuclear DDX6 and ABCE1 is found in the soluble fraction of lysates ( S5D Fig ) and could therefore be in proximity outside of granules along with MACA , which is entirely localized to the soluble fraction ( Figs 2C and 3B ) . As specificity controls for our PLA experiments , we also demonstrated that almost no DDX6 PLA signal was observed in Gag-expressing cells when either αGag or αDDX6 used for PLA was replaced with isotype-specific , non-immune control antibodies ( S6A–S6C Fig , Negative controls 1 and 2 respectively ) . Additionally , mock-transfected cells displayed no PLA signal when assayed with Gag and DDX6 antibodies ( S6C Fig ) . Similar results were obtained for ABCE1 negative controls but are not presented here . Thus , PLA appears to identify DDX6- and ABCE1-containing RNA granules that contain assembling Gag . Moreover , PLA confirms biochemical studies above , in which we showed that WT Gag and Gag G2A target to ABCE1- and DDX6-containing RNA granules ( Fig 3F , Fig 6C , and S2C Fig ) , as well as previous quantitative IEM studies showing the colocalization of ABCE1 and DDX6 with assembling Gag proteins [15 , 19 , 20 , 22] . DDX6 is known to be a marker of P bodies , as shown by high intensity labeling of P bodies by DDX6 IF [24]; however , our studies suggested that DDX6 is also found in smaller RNA granules ( Fig 8 and [22] ) . Additionally , because cells typically contain fewer than ten P bodies [59] , our finding that each Z stack image contains ~50 Gag-DDX6 PLA spots ( Fig 10 ) suggested that granules containing Gag and DDX6 are far more numerous than P bodies . While PLA spots are likely not an exact measure of the number of DDX6-containing granules , the relative ratio of DDX6-Gag PLA spots vs . P body spots , and the extent of overlap between the two , should provide insight into whether the putative ~80S assembly intermediates containing Gag and DDX6 correspond to P bodies . To determine this ratio , we analyzed 293T cells for Gag G2A-DDX6 PLA spots with concurrent DDX6 IF , to allow detection of PLA spots and P bodies in the same fields ( Fig 10A ) . G2A provirus was used here because Gag G2A is arrested as an ~80S assembly intermediate that likely contains Gag G2A and DDX6 in association with unspliced HIV-1 RNA ( Fig 3B , 3C , 3E and 3F ) ; thus , most Gag G2A-DDX6 PLA spots are likely to represent ~80S assembly intermediates containing Gag G2A , DDX6 , and unspliced HIV-1 RNA . We found that each Z stack image from G2A-expressing cells displayed an average of 56 Gag-DDX6 PLA spots , but only one P body by DDX6 IF ( Fig 10B and 10C ) . Interestingly , DDX6 IF images with high gain revealed an abundant diffuse , low-intensity , granular DDX6 signal ( e . g . in dotted circle in inset in Fig 10C top row , far left panel ) , in addition to the bright , DDX6-positive P body spots found in both G2A-expressing and mock cells ( Fig 10C , arrows in multiple panels ) . While Gag-DDX6 PLA spots occasionally overlapped with P body spots ( Fig 10C top row , far right panel ) , most of the Gag-DDX6 PLA spots overlapped with the diffuse , low-intensity , granular DDX6 signal , which includes small foci ( Fig 10C top row , merge ) and also likely includes soluble DDX6 observed in gradients ( S5D Fig ) . Since we have previously shown by coIP that Gag and DDX6 in the soluble fraction are not associated [22] , most likely soluble pools of DDX6 contribute to the DDX6 and Gag PLA colocalization signal in Fig 8 at only a low background level . Instead , most of the PLA signal likely represents colocalization of Gag with DDX6 in small low-intensity foci observed by DDX6 IF ( Fig 10C ) . Thus , we hypothesized that these small low-intensity foci likely correspond to DDX6-containing ~80S RNA granules that are co-opted to form ~80S assembly intermediates . To determine whether the low intensity DDX6 IF signal represents more than just background fluorescence , we compared total DDX6 IF signal in Gag-expressing 293T cells obtained by labeling with αDDX6 followed by fluorophore-conjugated secondary antibody ( Total signal ) to background signal obtained with secondary antibody alone ( S7A–S7D Fig ) . P body signal ( high intensity DDX6 IF signal ) was also quantified . These comparisons revealed that the low intensity DDX6 IF signal accounts for 66% of total signal and is distinct from and significantly greater than background ( S7B–S7D Fig; p value < 0 . 01 , with background signal accounting for 32% of total ) . Interestingly , the low intensity DDX6 IF signal accounts for the majority of overall signal , with high intensity P body signal accounting for only ~2% of overall DDX6 IF signal ( S7B–S7D Fig ) . Overall , the PLA and IF data indicate that the ~80S assembly intermediates that likely contain Gag and DDX6 in association with unspliced HIV-1 RNA are far more numerous than P bodies; moreover , most of the Gag-DDX6 PLA signal overlaps with low intensity DDX6 IF signal , which includes a population of small DDX6-positive foci that are visible by light microscopy and therefore likely represent small RNA granules . Our biochemical studies showed that WT Gag remains associated with viral-RNA-containing RNA granules during late stages of assembly ( Fig 4A–4C ) , suggesting that assembling Gag takes the co-opted viral-RNA-containing granule to PM sites of budding and assembly . In contrast , we found that Gag Zip associates with RNA granule proteins at late stages of assembly , but not with the subset of RNA granules that contains unspliced viral RNA ( Fig 7 ) . Thus , we would expect in situ approaches to reveal DDX6-containing assembly intermediates to be present at WT Gag or Gag Zip PM sites of assembly; however , the DDX6-containing granules formed by WT Gag should be colocalized with unspliced viral RNA , while the DDX6-containing granules formed by Gag Zip should not be colocalized with unspliced viral RNA . Previously , we used quantitative IEM to demonstrate that RNA granule proteins ( DDX6 , AGO2 , and ABCE1 ) are recruited to sites of WT Gag and Gag Zip assembly at the PM [15 , 18–20 , 22] . However , these earlier studies did not assess whether unspliced viral RNA is also associated with these RNA granules . Here we used quantitative IEM with double labeling for viral RNA and DDX6 to ask whether viral RNA is associated with DDX6-containing assembly intermediates at PM sites of assembly for WT Gag vs . Gag Zip in situ ( Fig 11 ) . Notably , to preserve epitopes for IEM , reduced levels of fixatives are required compared to standard EM; this in turn preserves fewer ultrastructural details ( as shown previously; compare Fig 2 to Fig 6 in Klein , 2011 #961 ) . While electron dense capsid structures at PM sites of budding and assembly can be seen in many fields , definitive identification of true budding sites is more difficult in such images if Gag is not immunolabeled . However , our previous quantitative IEM studies showed that immunolabeled DDX6 specifically colocalizes with immunolabeled Gag at PM sites of assembly , relative to low background levels of DDX6 at the PM in cells expressing targeting-defective or assembly-incompetent Gag mutants [19 , 22] . Thus , our previous studies established that PM sites displaying distinctive budding structures in association with DDX6 typically represent sites of Gag assembly . To test whether V1B RNA colocalizes with PM sites of WT Gag and Gag Zip assembly , we transfected HeLa cells stably expressing MCP fused to GFP ( HeLa-MCP-GFP cells ) with V1B genomic constructs encoding MS2 binding sites and Gag in cis ( Fig 11A; Set IV constructs in Fig 1A; phenotypes confirmed in S8 Fig ) . These cells were utilized because they had been validated previously in live imaging studies of HIV-1 packaging [12] . Sections were labeled with αDDX6 ( large gold ) to mark RNA granules , and with αGFP to mark MCP-GFP-tagged unspliced viral RNA ( small gold ) . PM assembly sites , defined by the presence of membrane deformation consistent with budding , were quantified and scored for viral RNA labeling , DDX6 labeling , and double labeling ( Fig 11B and 11C; S1 Table ) . Many of these sites displayed distinctive electron dense capsids , as expected at sites of Gag assembly ( Fig 11C ) . When all DDX6 labeling events at PM assembly sites were quantified , similar high levels of DDX6 labeling were observed at both WT and Gag Zip PM assembly sites ( 56% vs . 40% of all WT vs . Gag Zip PM assembly sites displayed DDX6 labeling , respectively , p>0 . 01; shown as total DDX6+ in Fig 11B; shown as D+ tot in S1 Table ) . Notably , quantitation of unspliced viral RNA labeling at all PM sites revealed that unspliced viral RNA was significantly more common at WT assembly sites relative to Gag Zip PM assembly sites ( 64% vs . 20% of all WT vs . Gag Zip PM assembly sites displayed viral RNA labeling , respectively , p<0 . 005; shown as total viral RNA+ in Fig 11B; shown as g+ Tot in S1 Table ) . Our most striking results were obtained upon quantitation of PM assembly sites that were double labeled for DDX6 and viral RNA . Abundant double labeling of unspliced viral RNA and DDX6 was observed at WT PM assembly sites , but not at Gag Zip PM assembly sites ( 33% vs . 5% of all WT vs . Gag Zip PM assembly sites , respectively , displayed double labeling , p<0 . 005; shown as viral RNA+/DDX6+ in Fig 11B; shown as g+D+ in S1 Table ) . As expected , the assembly-incompetent MACA formed very few early or late PM assembly sites , unlike WT and Gag Zip . The same patterns were observed when early and late assembly sites were analyzed separately ( S1 Table ) . Thus , IEM analysis of PM assembly sites supports our conclusion that both WT Gag and Gag Zip co-opt DDX6-containing RNA granules during packaging and assembly , but only the RNA granules co-opted by WT Gag also contain unspliced viral RNA . Moreover , these quantitative IEM studies ( Fig 11 ) along with our PLA studies ( Figs 8–10 ) , which were all performed without PuroHS treatment , demonstrate that HIV-1 Gag and unspliced HIV-1 RNA are associated with RNA granules in situ .
Previously , we identified a pathway of sequential ( early , intermediate , and late ) HIV-1 capsid assembly intermediates and used this temporal pathway to understand intracellular events in the assembly of the immature HIV-1 capsid . Since the process of HIV-1 genome packaging occurs at the same time as capsid assembly , we reasoned that assembly intermediates are likely to be packaging intermediates; thus , identification of the assembly intermediate in which HIV-1 Gag first associates with unspliced HIV-1 RNA could shed light on the nature of the earliest HIV-1 packaging intermediate . If the first assembly intermediate ( ~10S Gag ) is associated with unspliced HIV-1 RNA , that would support a model in which a Gag monomer or dimer alone initates this association in cells . In contrast , if the initial association of Gag with unspliced HIV-1 RNA occurs in any other assembly intermediate besides ~10S Gag , that would support a model in which Gag first associates with HIV-1 RNA in host ribonucleoprotein complexes , since all the intermediates except the first assembly intermediate contain host RNA granule proteins . Here , using a variety of approaches , we were unable to identify a population of soluble ~10S Gag associated with unspliced HIV-1 RNA; indeed , to the limit of our detection , almost no unspliced HIV-1 RNA was detected in the soluble fraction when gradients that fully resolve the soluble region were analyzed . Instead , the first assembly intermediate in which Gag was associated with unspliced HIV-1 RNA was the second assembly intermediate , an RNA-granule-derived ~80S complex . Consistent with these findings , ABCE1 and DDX6 , which are associated with Gag in the ~80S intermediate , were also associated with unspliced HIV-1 RNA in ~80S fractions . Based on our findings , we hypothesize that the first association of HIV-1 Gag with unspliced HIV-1 RNA occurs only after Gag localizes to an RNA-granule-derived complex containing unspliced HIV-1 RNA ( the RNA granule model , Fig 12 ) . If this is the case , then the ~80S assembly intermediate could be the first complex in which Gag associates with the HIV-1 RNA that ultimately gets packaged , here termed the packaging initiation complex . While more studies will be needed to determine whether the ~80S RNA-granule-derived assembly intermediate actually is the packaging initiation complex , our findings advance the field by identifying , for the first time , a candidate packaging initiation complex that can be analyzed further . While we did not identify a complex in which soluble Gag associates with unspliced HIV-1 RNA , our search for such a complex was extensive . We examined lysates generated from two different transfected cell types ( COS-1 and 293T cells ) ; lysates expressing WT Gag or two well-studied assembly-defective Gag mutants that are arrested in association with unspliced HIV-1 RNA ( one arrested in the cytosol and one arrested at the PM ) ; conditions that largely dissociated ribosomes ( through use of PuroHS ) versus conditions that preserve ribosomes intact; and cells expressing the HIV-1 provirus or a Gag construct with the viral genome provided in trans . Under all these conditions , the smallest complex containing WT or mutant Gag associated with unspliced viral RNA that we could detect at steady state by αGag IP formed a peak in the ~80S region ( Figs 3C , 4C , 4F , 5C , 5F , 7C and 7F ) . When lysates were harvested under gentle conditions , this complex formed a sharp peak in the ~80S region of our gradients ( Fig 5E and 5F ) and closely corresponded to the previously described ~80S assembly intermediate , the second assembly intermediate in the HIV-1 capsid assembly pathway . However , when the same cell type expressing the same plasmids was harvested using PuroHS treatment , the peak appeared considerably broader , and encompassed the ~40S region in addition to the ~80S region ( e . g . Fig 4B and 4C , green lines ) . There are two possible explanations for the smallest complex containing Gag and unspliced HIV-1 RNA sometimes migrating as a narrow ~80S peak and sometimes migrating as a broader ~40S - ~80S peak . One possibility is that the smallest complex containing Gag associated with unspliced HIV-1 RNA is a partially hidden ~40S complex that is not well resolved; an alternate possibility is that the smallest complex containing Gag associated with unspliced HIV-1 RNA is an ~80S intermediate whose integrity has been partially disrupted by PuroHS , resulting in its migration in an ~40S – 80S region rather than in the sharper ~80S peak we have previously observed . We favor the latter explanation because we have found previously that intracellular assembly intermediates formed by HIV-1 and other retroviral Gag proteins are quite labile , and can be partially disrupted by solutions of high or even modest ionic strength [19 , 60] . Indeed , when isolated ~80S assembly intermediates formed by some HIV-1 Gag mutants are subjected to an ionic stress considerably less harsh than what was used here , the ~80S Gag is found in a trail that extends into the ~10S-~40S region ( e . g . Fig 7D in [19] ) . Nevertheless , definitively distinguishing between these two possibilities is difficult here because most of the biochemical experiments shown here were performed with PuroHS treatment to ensure that we were not studying ribosome-associated complexes ( e . g . Figs 3A–3F , 4A–4F , 5A–5C , 6A–6C and 7A–7F and S2A–S2C Fig ) . Thus , while the smallest complex containing Gag associated with unspliced HIV-1 RNA that we detected forms a peak in the ~80S region of gradients , more detailed studies using gentle harvest conditions or cross-linkers will be needed in the future to fully resolve the question of whether this represents a single ~80S complex or multiple complexes of ~40S - ~80S . A second issue worthy of discussion involves our identification of a complex in chronically HIV-infected human T cells that was immunoprecipitated using αABCE1 , contains unspliced HIV-1 RNA , and appears to correspond to the ~80S assembly intermediate ( with formation of a sharp ~80S peak ) ( Fig 6B and 6C ) . In these T cells , our gradient analysis did reveal small pools of unspliced HIV-1 RNA in the 10S - 40S region ( Fig 6B ) . Those pools of unspliced HIV-1 RNA were not apparent in other experiments ( e . g . ( compare Figs 3B and 5E to Fig 6B ) . For three reasons , we do not think these pools of unspliced HIV-1 RNA in Fig 6B correspond to Gag-associated complexes of ~10-40S . First , these minor pools of RNA in the 10S - 40S region were observed when cells were harvested in harsh salts as part of PuroHS treatments; thus , the minor pools of RNA likely came from ~80S complexes that were disrupted . Second , unspliced HIV-1 RNA in the 10S - 40S region was only observed in gradients that do not resolve this region well; whenever we have utilized gradients that resolve the 10S - 40S region well ( e . g . Fig 3 , S4 Fig ) , we failed to detect unspliced HIV-1 RNA in the ≤40S region of the gradient . Third , αGag IPs failed to immunoprecipitate unspliced HIV-1 RNA from 10S - 40S fractions in chronically infected human T cells ( S4 Fig ) . Thus , to date we have been unable to demonstrate that minor pools of unspliced HIV-1 RNA in the 10S - 40S region are Gag-associated . Nevertheless , we plan to study small Gag and RNA complexes , especially in chronically infected human T cells , in more detail . Ultimately , we cannot exclude the possibility that our experiments failed to identify or selectively lost a soluble complex containing Gag associated with unspliced HIV-1 RNA , or that such a complex is extremely transient and not detectable at steady state by the approaches we used , or that a buried peak is present in the ~40S region of our gradients . However , given that all our lysates contained a large pool of soluble Gag , and that soluble Gag was not associated with unspliced viral RNA to an extent detectable by our RT-qPCR analyses in any of our experiments , our data support a model in which the first association between Gag and unspliced HIV-1 RNA occurs not in the soluble fraction but in a host ribonucleoprotein complex , with more work being needed to determine if this is this host ribonucleoprotein complex is ~40S or ~80S . Moreover , to our knowledge , a soluble intracellular complex that contains Gag in association with unspliced viral RNA has not been identified and reported to date . Importantly , a model in which HIV-1 RNA packaging is initiated in a host ribonucleoprotein complex ( Fig 12 ) , supported by our studies , is also in keeping with important concepts in cell biology . Specifically , cell biologists argue that , within cells , the fate of a particular cellular mRNA is determined in large part by its associated cellular proteins ( reviewed in [27] ) , which first interact with cellular mRNA during transcription and subsequently undergo successive rounds of remodeling ( reviewed in [26] ) . Consistent with these cell biological studies of host mRNA , our model proposes that unspliced HIV-1 RNA first associates with host ribonucleoproteins in the nucleus , and upon entering the cytoplasm is found either in translating or non-translating host ribonucleoprotein complexes . In the absence of assembling Gag , complexes that contain unspliced HIV-1 RNA are of diverse sizes >~30S . The components and identities of these complexes remain to be determined . Based on studies presented here along with our previous studies , we hypothesize that WT Gag first interacts with unspliced HIV-1 RNA to initiate packaging when it co-opts a subpopulation of the non-translating complexes containing unspliced HIV-1 RNA to form the ~80S assembly intermediate , which would then become the packaging initiation complex ( Fig 12A ) . Targeting , packaging , and late stages of WT Gag multimerization continue at the PM in association with this RNA granule , leading to formation of the ~500S packaging/assembly intermediate and the ~750S completely assembled capsid , both of which contain unspliced HIV-1 RNA . The ~80S assembly intermediate/putative packaging initiation complex and the ~500S late packaging/assembly intermediate contain ABCE1 and DDX6 , two host enzymes that facilitate virus assembly [21 , 22] and may distinguish this subclass of granules from other host ribonucleoprotein complexes . Upon completion of immature capsid assembly , the RNA granule proteins dissociate from the ~750S capsid [15 , 17 , 21 , 22] , which then undergoes budding and release . Like WT Gag , Gag G2A forms the ~80S putative packaging initiation complex , but is arrested at that step . In contrast , MACA , which lacks NC and is oligomerization-incompetent , fails to associate with RNA granules of any kind . Interestingly , Gag Zip , which also lacks NC but is assembly-competent because it contains an oligomerization domain in place of NC , targets to granules that lack unspliced viral RNA via a mechanism that is independent of NC ( Fig 7C; [18 , 22] ) . Our additional studies of Gag Zip led us to define two steps that direct Gag proteins to ~80S granules containing unspliced HIV-1 RNA ( Fig 12B ) . While Gag Zip targets to RNA granules that lack unspliced viral RNA , addition of a heterologous viral-RNA-binding domain redirects Gag Zip to RNA granules that contain unspliced viral RNA ( Fig 7F ) . Therefore , we hypothesize that most ~80S RNA granules likely contain cellular RNA , with only a small subset containing unspliced viral RNA . Together , our data argue that two targeting events are required for stable association of Gag with granules containing unspliced HIV-1 RNA . One step is a poorly understood , oligomerization-dependent step that requires NC or a heterologous oligomerization domain and allows Gag to target to a large class of ~80S ABCE1- and DDX6-containing RNA granules , most of which lack unspliced HIV-1 RNA . The other step involves HIV-1-RNA-binding , which requires NC or a heterologous viral-RNA-binding domain and enables stable association of Gag with a subset of these granules that contains unspliced viral RNA . While MACA lacks the ability to complete either step in this two-step process , Gag Zip is capable of completing the oligomerization-dependent step , but not the step dependent on viral RNA binding ( Fig 12B ) . Thus , a poorly understood feature of Gag that is present in Gag Zip but lacking in MACA is responsible for targeting Gag to a broader subclass of ABCE1- and DDX6-containing host RNA granules that lack unspliced HIV-1 RNA and likely contain cellular mRNA . Importantly , we do not know the mechanism by which unspliced HIV-1 RNA associates with host RNA granules . Most likely , ribonucleoprotein complexes containing unspliced HIV-1 RNA are generated during transcription , as is the case for cellular mRNA , and undergo successive rounds of remodeling , both during nuclear export and in the cytoplasm , to form cytoplasmic >30S non-translating RNA granules . Additionally , while our coIP studies demonstrate that Gag is associated with unspliced HIV-1 RNA in the packaging/assembly intermediates , we do not know whether Gag and unspliced HIV-1 RNA make direct contact with each other in these complexes . Based on studies by others [61] , we speculate that Gag proteins make direct contact with only a few regions of unspliced HIV-1 RNA in the ~80S putative packaging initiation complex , but contact many more regions of the unspliced HIV-1 RNA in the ~500S late packaging intermediate . Also , because we did not analyze nuclear complexes , we cannot exclude the possibility that Gag first forms the ~80S putative packaging initiation complex in the nucleus; however , like others [13] , we have not observed enough Gag in the nucleus of provirus-expressing cells to test this possibility . Some other aspects of our study deserve further comment . First , PuroHS treatment was used in many of our experiments because it allows us to exclude translating complexes , which migrate at similar sizes as our assembly/packaging intermediates and therefore prevent us from studying non-translating RNA that undergoes packaging . Excluding translating complexes is particularly important when analyzing provirus-expressing cells , in which Gag would be expected to associate with unspliced HIV-1 RNA during its translation and also during packaging . However , our use of Gag GFP constructs that package a modified viral genome in trans allowed us to show that the same results were obtained when we repeated key experiments without PuroHS treatment ( compare Fig 5D–5F performed without PuroHS treatment to Fig 4A–4C performed with PuroHS treatment ) . In the experiment performed without PuroHS treatment , the in trans system allowed us to still exclude translating Gag from our analyses of complexes containing Gag and unspliced viral RNA because Gag in that system is translated from an mRNA with a different nucleotide sequence that is not detected by our RT-qPCR oligo . Our demonstration that the ~80S complex containing Gag associated with unspliced viral RNA is present even in cells harvested with ribosomes intact but with translating Gag mRNA excluded from the analysis through oligo design ( Fig 5F ) indicates that the ~80S candidate packaging initiation complex is not an artifact of PuroHS treatment . We should also note that differences in accessibility of epitopes during IP could account for differences in migration of immunoprecipitated complexes relative to complexes shown in Gag blots and profiles of total unspliced viral RNA . For example , the difference in the exact migration of the ~80S peak in Fig 4C ( in which αGag was used to IP the ~80S complex ) and S2C Fig ( in which αABCE1 was used to IP the ~80S complex ) could be explained by ABCE1 in the ~80S complex being less accessible than ABCE1 in the ~500S complex ( while conversely Gag in the ~80S complex is known to be more accessible than Gag in the ~500S complex ) . Performing more detailed studies to address whether the ~80S peak is heterogeneous and define the basis of differences in epitope accessibility are of high priority in the future . Additionally , we should also emphasize that the pool of all complexes that contain unspliced HIV-1 RNA in the absence of Gag is very diverse , with complexes ranging in size from 30S to >150S ( Fig 2C ) . In the presence of assembling Gag , it is likely that only a small percentage of these complexes become complexes containing Gag associated with unspliced HIV-1 RNA . Finally , with respect to the putative ~500S late packaging intermediate , as mentioned earlier , the ratio of this complex relative to the ~80S complex also varied considerably in our experiments . This is not surprising since the amount of the late packaging intermediate present at steady state could be highly dependent on the kinetics of assembly and budding , both of which could vary with cell type , stage of assembly , cell viability , etc . Interestingly , peaks corresponding to both the ~80S and ~500S putative packaging intermediates were particularly prominent in human 293T cells and infected human T cells ( Figs 5C and 6C ) . Notably , the ~80S assembly intermediate in which we first detected Gag associated with unspliced HIV-1 RNA ( the candidate packaging initiation complex ) has an S value similar to the eukaryotic ribosome , which is a large ribonucleoprotein complex containing four species of RNA and 79 cellular proteins . Interestingly , studies by others are consistent with this finding . Specifically , the reported diffusion coefficient of ~70S bacterial ribosomes ( 0 . 04 μm2 sec-1; [62] ) is similar to the ~0 . 07 and 0 . 014 μm2 sec-1 diffusion coefficients reported for cellular subpopulations of HIV-1 RNA [63] and Gag [64] , respectively . Our finding that the packaging initiation complex is ~80S suggests that this complex contains numerous host components , since Gag is likely a dimer or small oligomer at this stage [13] , and a Gag dimer on its own would be ~5S . Although we know that two other viral proteins , HIV-1 GagPol [15] and Vif [21] , are present in assembly intermediates , it is likely that host components account for most of the molecular mass of the ~80S packaging initiation complex . Here we showed that the candidate packaging initiation complex likely contains at least two host enzymes that are known to facilitate HIV-1 capsid assembly , ABCE1 and DDX6 [21 , 22] . In addition , our earlier studies demonstrated that the RNA granule proteins AGO2 and DCP2 are also present in the ~80S and ~500S RNA-granule-derived assembly intermediates [22] . Interestingly , we recently showed that the small ribosomal S6 protein was not detected in IP analyses of RNA-granule-derived ~80S and ~500S capsid assembly intermediates formed by a non-human lentiviral Gag ( feline immunodeficiency virus; [60] ) . The lack of ribosomal proteins , along with the presence of RNA granule proteins , in capsid assembly intermediates further supports a model in which assembling retroviral Gag proteins co-opt a host RNA granule during assembly/packaging . Importantly , our in situ studies demonstrate that the RNA-granule-derived putative HIV-1 packaging intermediates that we identified are not simply artifacts formed during cell lysis . Previously , we had used quantitative double-label IEM to show that DDX6 is recruited to PM sites of WT Gag assembly , but not to PM sites containing assembly-incompetent MACA Gag [19 , 22] . Here we showed that these PM sites of assembly , detected because they contain DDX6 at sites of PM deformation typical of HIV-1 budding , also contain unspliced viral RNA , thus demonstrating colocalization of unspliced viral RNA with an RNA granule protein at likely PM assembly sites in situ ( Fig 11 ) . Moreover , because some investigators define RNA granules as complexes containing non-translating RNA that are detectable as discrete foci by light microscopy , we also developed approaches for asking whether complexes that contain Gag colocalized with DDX6 correspond to fluorescent foci visible by light microscopy . First we showed that we can detect Gag colocalized with DDX6 ( or ABCE1 ) in situ by light microscopy using PLA ( Figs 8 and 9 ) . Notably , our biochemical studies suggested that the ~80S putative packaging initiation complex should be similar in size to the ~80S ribosome , which is ~25 nm in diameter [65] , i . e . roughly one-quarter or one-twelfth the size of a DDX6-containing P body ( which ranges from 100–300 nm in size [66] ) . For this reason , we hypothesized that the ~80S assembly intermediate/putative packaging initiation complexes might be detectable as small fluorescent foci that are distinct from large P bodies but nevertheless still visible by light microscopy . Consistent with this hypothesis , we found that indeed the vast majority of ~80S Gag G2A-DDX6 PLA spots overlapped not with P bodies , but with a lower intensity granular DDX6 signal ( Fig 10 ) that accounts for the majority of DDX6 IF signal . Moreover , this low intensity DDX6 signal includes discrete small cytoplasmic foci that are distinguishable from background even though they are considerably smaller than P bodies ( S7 Fig ) . Together , our PLA and IF data suggest that the ~80S assembly intermediate/putative packaging initiation complexes likely correspond to small cytoplasmic foci; the finding that these foci are visible by light microscopy and contain the RNA granule protein DDX6 supports our description of them as small RNA granules that are distinct from ( but possibly related to ) P bodies . Our demonstration that PLA spots containing Gag colocalized with DDX6 are far more numerous than P bodies is consistent with an earlier study showing that Gag and viral RNA are not found in P bodies [67] . However , the small RNA granules co-opted by assembling Gag could represent subunits that exchange with P bodies or other RNA granules , perhaps explaining an earlier report of HIV-1 RNA colocalizing with P bodies [68] . In keeping with this possibility , we observed that Gag-DDX6 PLA spots occasionally overlap with P bodies ( Fig 10 ) . A model in which P bodies consist of smaller functional subunits , such as the small ~80S RNA granules co-opted by HIV-1 , is appealing given numerous studies showing that when large P body foci are dissociated and no longer visible , their functions are not lost [69 , 70] , consistent with P bodies being composed of smaller , independent functional units . Thus , for three reasons , it seems appropriate to call these DDX6-containing complexes , including those co-opted by HIV-1 , small RNA granules: 1 ) in IF images they appear to be discrete low intensity foci ( Fig 10 and S7 Fig ) , albeit with sizes that are close to the limit of light microscopic detection; 2 ) they contain canonical RNA granule proteins such as DDX6 ( Fig 8 , Fig 10 and S6 Fig ) and AGO2 [22]; and 3 ) they could correspond to the smaller DDX6-containing subunits that others have found to be functional when P bodies are disrupted . Further studies will be required to define the exact proteome and RNAome of these small RNA granules and the ~80S packaging initiation complex . Additionally , crosslinking followed by multiple purification steps will be required to test definitively whether Gag , DDX6 , ABCE1 , and unspliced viral RNA are all together in the same complex ( rather than only two or three of these components being together in two or more separate complexes of similar size ) . More studies will also be required to determine whether these complexes contain other RNA binding proteins such as Staufen1 , which plays a role in HIV-1 packaging and assembly [71–73] , or MOV10 , which is packaged by HIV-1 and modulates virus infectivity [74 , 75] . It is also worth noting the similarities between the RNA granule model of HIV-1 packaging proposed here , and what has been observed for the Ty3 yeast retrotransposon , which is distantly related to HIV-1 . Assembling Ty3 Gag and its packaged Ty3 RNA are found in large clusters that contain the yeast DDX6 homologue dhh1 [76 , 77]; moreover knockdown and mutational analyses indicate a role for RNA granule proteins in formation of functional Ty3 retrotransposons in yeast [78 , 79] . Similarly , we had previously shown that DDX6 acts enzymatically to promote immature HIV-1 capsid assembly , since siRNA knockdown of DDX6 decreased virus production without affecting steady-state Gag levels and virus production was rescued by a siRNA resistant WT DDX6 but not a ATPase-defective DDX6 mutant [22] . Additionally , DDX6 knockdown in primary human T cells reduced production of infectious HIV-1 in that study [22] . While another study did not find DDX6 to be required for HIV-1 capsid assembly [67] , this could be because of differences in the extent of DDX6 depletion or because other helicases in the co-opted RNA granule can substitute for DDX6 in some cell types or when given enough time to be upregulated in shRNA-expressing cell lines . Interestingly , human foamy virus was also reported to require DDX6 , with DDX6 being necessary for packaging rather than virus particle assembly [80] . Thus , the overall similarity between our data and data obtained for Ty3 and human foamy virus raises the possibility that co-opting of host RNA granules for RNA packaging and assembly could be an ancient mechanism that has been conserved across retroelement evolution . Together our data argue for a new view of packaging . To date , packaging studies have not specified how or in what complex Gag associates with viral RNA; here we fill this gap by identifying a candidate packaging initiation complex in provirus-expressing cells , that corresponds to a poorly understood host RNA granule . Sequestration of unspliced HIV-1 RNA within a subset of host RNA granules suggests that HIV-1 RNA mimics host RNAs , which also localize to RNA granules when they are not translating . Interestingly , this localization of HIV-1 RNA to host RNA granules could be both beneficial and problematic for the virus . On the one hand , sequestration would allow viral RNA to evade detection by the host immune system , create a site where assembling Gag can be concentrated , and provide Gag access to host RNA helicases that could facilitate displacement of host RNA binding proteins from unspliced HIV-1 RNA , allowing them to be replaced with Gag . On the other hand sequestration of unspliced HIV-1 RNA creates a dilemma for the virus in that it puts the unspliced HIV-1 RNA that must be encapsidated into a different subcellular compartment ( the RNA granule ) than newly translated Gag , which is initially found either with translating ribosomes or in the soluble compartment . To solve this problem , Gag may have evolved an oligomerization-dependent mechanism for localizing to a subclass of mRNA-containing host RNA granules; association with RNA granules could allow efficient binding of Gag to unspliced HIV-1 RNA , which would in turn localize Gag to the small subset of these granules that contains unspliced HIV-1 RNA . Future studies will be needed to further test this model and better understand how the virus co-opts this subclass of host RNA granules .
Four types of expression systems were utilized in this study . Proviruses ( Set I constructs in Fig 1A ) are from the LAI strain and were described previously [15 , 18 , 81] . These have native HIV-1 sequences and the HIV-1 LTRs . Proviruses were transfected into COS-1 or 293T cells ( American Type Culture Collection ( ATCC ) , Manassas , VA ) as indicated for some biochemical studies and all PLA studies . For other biochemical studies , we used an in trans expression system in which the genome is provided by V1B , a modified proviral plasmid that expresses an RNA encoding an assembly-incompetent truncated gag gene , all cis-acting packaging signals , full-length tat , rev , and vpu genes , and 24 MS2 stem loops that bind to MCP [12] . V1B was transfected into COS-1 or 293T cells with WT and mutant SynGag GFP constructs ( here referred to as Gag GFP constructs ) provided in trans , where indicated . V1B and the codon optimized SynGag GFP WT and G2A constructs ( Set II constructs in Fig 1A ) were provided by P . Bieniasz ( Rockefeller University , New York , N . Y . ) and were utilized in previous live imaging studies [12] and coIP studies [13] . Additional Gag GFP constructs ( MACA and Gag Zip ) were generated from the WT SynGag GFP construct . For G2A , the glycine in position 2 of Gag was converted to an alanine by via site-directed mutagenesis , as described previously [15]; for Gag Zip , LZ was inserted in place of NC using Gibson assembly , as described previously [18] . For IEM experiments , HeLa cells that express MCP-NLS-GFP as described previously [12] were obtained from P . Bieniasz ( Rockefeller University , New York , N . Y . ) . To ensure that all HeLa-MCP-GFP cells expressed both Gag and genome following transfection , V1B constructs expressing WT and mutant Gag in cis were generated by inserting relevant Gag coding regions from HIV-1 proviruses ( Set I constructs in Fig 1A ) into V1B constructs containing MS2 binding sites via Gibson assembly to generate Set IV constructs in Fig 1A . Oligos used for site-directed mutagenesis and Gibson assemblies are available upon request . H9 cells ( ATCC ) were used to generate chronically HIV-infected H9 T cells expressing a pro-genome in cells ( Set III constructs in Fig 1A ) by infection with virus . Plasmid used for virus production was generated by inserting three protease inactivation mutations into a previously described HIV-1 provirus , LAI strain , that encodes deletions in env and vif , a frameshift in vpr , and substitution of nef with a puromycin resistance gene , as described previously [22] . 293T cells were transfected with this plasmid to produce virus , and H9 cells were infected with this virus and maintained under puromycin selection as described previously [17] . COS-1 and 293T cells were maintained in DMEM ( Life Technologies ) with 10% FBS . H9 cells were maintained in RPMI ( Life Technologies ) with 10% FBS under puromycin selection . HeLa-MCP-GFP cells were obtained from P . Bieniasz and were maintained in DMEM with 10% FBS , and periodically subjected to blastocidin selection . COS-1 , 293T , or HeLa-MCP-GFP cells were transfected with 1–5μg DNA using polyethylenimine ( Polysciences , Warrington , PA ) . Cell lysates were harvested in 1X Mg+2-containing NP40 buffer ( 10 mM Tris-HCl , pH 7 . 9 , 100 mM NaCl , 50 mM KCl , 1 mM MgCl , 0 . 625% NP40 ) in the presence of freshly prepared protease inhibitor cocktail ( Sigma , St Louis , MO ) and RNaseOUT ( Invitrogen ) . Where indicated , lysates were treated with 1 mM puromycin HCl ( Invitrogen ) for 10 min at 26°C followed by 0 . 5M KCl for 10 min at 26°C before further analysis; otherwise cells were harvested in 1X NP40 buffer ( 10 mM Tris-HCl , pH 7 . 9 , 100 mM NaCl , 50 mM KCl , 0 . 625% NP40 ) in the presence of freshly prepared protease inhibitor cocktail ( Sigma , St Louis , MO ) and RNaseOUT ( Invitrogen ) and analyzed immediately . Lysates were analyzed by WB or RT-qPCR , or subjected to IP as described below . Alternatively , lysates were analyzed by velocity sedimentation , as described below , and gradient fractions were then analyzed by WB , RT-qPCR , or IP . Except where indicated , lysates or gradient fractions were subjected to IP with affinity purified αABCE1 [21] , HIV immunoglobulin NIH AIDS Reagents Catalog #3957 , from NABI and NHLBI ) , a monoclonal to GFP ( Roche ) , or αDDX6 ( #461 , Bethyl Laboratories , Montgomery , TX ) , using protein G-coupled Dynabeads ( Life Technologies ) . IP eluates were analyzed by SDS-PAGE , followed by western blot ( WB ) using the primary antibodies described above or a monoclonal antibody directed against HIV-1 Gag p24 ( HIV-1 hybridoma 183-H12-5C obtained from Bruce Chesebro through the AIDS Reagent Program Division of AIDS , NIAID , NIH ) , followed by an HRP-conjugated anti-human IgG secondary antibody ( Bethyl Laboratories , Montgomery , TX ) , or an HRP-conjugated anti-mouse-IgG1 ( Bethyl Laboratories ) or anti-mouse-IgG or anti-rabbit secondary antibody ( Santa Cruz Biotechnology , Dallas , TX ) . In S5C Fig , IP eluates were subjected to WB with HIV immune globulin ( provided by NABI and NHLBI , catalog no . 3957 in the AIDS Reagent Program Division of AIDS , NIAID , NIH ) for detection of Gag . WB signals from IP eluates were detected using Pierce ECL substrate ( Thermo Fisher Scientific ) with Carestream Kodak Biomax Light film . For detection of Gag in total cell lysates , velocity sedimentation fractions , and membrane flotation fractions , WBs were performed as described above , or using antibodies conjugated to infrared dyes ( LI-COR Biosciences , Lincoln , NE ) . Quantification of Gag bands on film was performed using Image J software or LI-COR Odyssey software . Transfected cells or VLPs were harvested as described above . Where indicated lysates were analyzed by velocity sedimentation and/or IP , as described above except that IP samples that were prepared for viral RNA quantification were washed four times in detergent buffer and once in non-detergent buffer . For total cell lysate analysis , aliquots corresponding to ~5 x103 COS-1 cells , 2 x104 293T cells and 8 . 5 x103 HeLa-MCP-GFP cells were used , and results of quantification were normalized to 1 x103 cells . For VLP analysis , aliquots corresponding to VLPs from 1 x105 293T cells and 1 . 5 x104 HeLa-MCP-GFP cells were used , and results of quantification were normalized to 1 x103 cells . For gradient analysis , ~4 x105 COS-1 cells , 2 x 106 203T cells , or 6 x 105 H9 cells were analyzed on a single 5 ml gradient . Aliquots of gradient fractions or gradient IP fractions were treated with proteinase K ( Sigma ) at a final concentration of 150 μg/ml in 0 . 1% SDS , followed by total RNA extraction by using Trizol ( Ambion ) . RNA was precipitated with isopropanol , extracted with BCP ( Molecular Research Center ) , pelleted at 12 , 000xg for 15 min at 4°C , and the RNA pellet was subjected to DNase I ( Invitrogen ) treatment ( 2 u per 50 μl reaction ) . The iScript Advanced cDNA synthesis kit ( Bio-Rad ) was used to generate cDNA from 10% of the RNA using random hexamer primers at 42°C for 30 min , followed by heat inactivation . An aliquot of cDNA ( 2 . 9% ) was used for qPCR using SYBR Green ( Bio-Rad ) to determine RNA copy number . For HIV-1 viral RNA qPCR , we used the following oligos that target bp 162 to 269 within the Gag open reading frame in HIV-1 LAI ( at the end of MA and start of CA ) and result in a 108 bp amplicon: 5’-AGAAGGCTGTAGACAAATACTGGG-3’ ( forward ) ; 5’-TGATGCACACAATAGAGGGTTG-3’ ( reverse ) . These oligos detect the full length HIV-1 provirus as well as the V1B genomic construct , but do not recognize the Gag GFP constructs that were transfected in trans , as expected since the Gag GFP constructs were codon-optimized . To determine the copy number of other RNA species , we used the following qPCR oligos: Tat mRNA , 5’-TCT ATC AAA GCA ACC CAC CTC-3’ ( forward ) and 5’-CGT CCC AGA TAA GTG CTA AGG-3’ ( reverse ) ; 28S rRNA , 5’-CCC AGT GCT CTG AAT GTC AA-3’ ( forward ) and 5’-AGT GGG AAT CTC GTT CAT CC-3’ ( reverse ) ; 18S rRNA , 5’-GCA ATT ATT CCC CAT GAA CG-3’ ( forward ) and 5’-GGC CTC ACT AAA CCA TCC AA-3’ ( reverse ) ; GAPDH mRNA , 5’-AGG TCA TCC CTG AGC TGA AC-3’ ( forward ) and 5’-GCA ATG CCA GCC CCA GCG TC-3’ ( reverse ) ; and 7SL RNA 5’-GCT ATG CCG ATC GGG TGT CCG-3’ ( forward ) and 5’-TGC AGT GGC TAT TCA CAG GCG-3’ ( reverse ) . All qPCR samples were analyzed in duplicate using the MyiQ RT-PCR detection system and iQ5 software ( Bio-Rad ) . Amplicons corresponding to regions amplified by qPCR were used to generate standard curves . Duplicate nine-point standard curves were included on every qPCR plate , ranging from 101 copies to 108 copies , with a typical efficiency of ~90% or greater and an R2 of 0 . 99 . Standard curves were able to detect 10 copies per reaction but not 1 copy , thereby setting the detection threshold at 10 copies per reaction , which was equivalent to ~1000 copies per 1000 cells for inputs and total viral RNA from gradient fractions , ~100 copies per 1000 cells for IP from total cell lysates or gradient fractions , and ~50 copies per 1000 cells for VLPs . Minus RNA controls were included in each experiment and were always zero . RT minus controls were also included in each experiment and ranged from 0–100 copies per reaction . Mock transfected VLP controls were used to set the baseline in graphs and ranged from 1–1000 copies per 1000 cells . For IP , nonimmune viral RNA copy number was analyzed in parallel and was typically 1–2 logs lower than in immune IP samples . Quantification of the total cell number used in each experiment allowed us to represent all qPCR data as number of viral RNA copies per 1000 cells , except for Fig 2 data which is presented as % of total viral RNA to allow comparison across different RNA species . Log scales were used to display all VLP , which exhibit large differences . Linear scales were used for IP data , which exhibit smaller differences . Note that differences in transfection efficiency resulted in a range of total viral RNA copies per 1000 cells between experiments; likewise , IP efficiency also varied between experiments . Thus the exact number of viral RNA copies immunoprecipitated in fractions from different experiments varied considerably , but the pattern did not . Protocols for VLP and gradient RT-qPCR are available at: dx . doi . org/10 . 17504/protocols . io . k73czqn dx . doi . org/10 . 17504/protocols . io . k74czqw Supernatants of COS-1 or HeLa-MCP-GFP cells , transfected as described above , were centrifuged at 2000 rpm ( 910 x g ) for 10 min at 4°C , filtered ( 0 . 45 μm ) to remove remaining cells , and purified through a 30% sucrose cushion in an SW60Ti rotor at 60 , 000 rpm ( 370 , 000 x g ) for 30 min at 4°C , as described previously [18] . Transfected COS-1 cells or 293T cells were harvested at 36 h or 15 h posttransfection , respectively , as described above and diluted into 1X NP40 buffer ( 10 mM Tris acetate pH 7 . 4 , 50 mM KCl , 100 mM NaCl , 0 . 625% NP-40 ) . For each sample , 120 μl lysate was layered on a step gradient . To resolve complexes of ~10S to ~150S , step gradients were prepared from 5% , 10% 15% , 20% , 25% , and 30% sucrose in NP40 buffer without MgCl ( 10 mM Tris-HCl , pH 7 . 9 , 100 mM NaCl , 50 mM KCl , 0 . 625% NP40 ) and subjected to velocity sedimentation in a 5 ml Beckman MLS50 rotor at 45 , 000 rpm ( 162 , 500 x g ) for 90 min at 4°C . To resolve from ~10S to ~750S , step gradients were prepared from 10% , 15% , 40% , 50% , 60% , 70% , and 80% sucrose in NP40 buffer without MgCl , and subjected to velocity sedimentation in a 5 ml Beckman MLS50 rotor at 45 , 000 rpm ( 162 , 500 x g ) , for 45 min at 4°C . Gradients were fractioned from top to bottom , and aliquots were analyzed by WB , IP , and/or RT-qPCR as described above . Expected S value migrations were determined using a published equation [40] and confirmed using S value markers , as described previously [16] . These expected S value migrations are shown as black bars above gradient fractions , but were also confirmed by RT-qPCR for 7SL RNA ( ~11S ) , 18S rRNA ( 40S small ribosomal subunit ) , and 28S rRNA ( 60S large ribosomal subunit ) . 293T cells were plated into 6-well dishes containing coverslips with Grace Biolabs CultureWell silicone chambers ( Sigma-Aldrich ) attached to create four chambers on each coverslip . Cells were transfected with 3 μg of plasmid per well and 16 . 5 h later were fixed for 15 min in 4% paraformaldehyde in PBS pH 7 . 4 , permeabilized in 0 . 5% saponin in PBS , pH 7 . 4 for 10 min , and blocked in Duolink blocking solution ( Sigma-Aldrich ) at 37°C for 30 min . Cells were incubated in primary antibody ( described under IP methods above ) , followed by Duolink reagents ( Sigma-Aldrich ) : oligo-linked secondary antibody , ligation mix , and red or green amplification/detection mix , with washes in between , as per the Duolink protocol . For concurrent IF , cells were incubated following the final PLA washes for 15 min at RT with 1:1000 secondary antibody , either Alexafluor 594 conjugated to anti-mouse IgG or Alexafluor 488 conjugated to anti-rabbit IgG . Cover slips were mounted using Duolink In Situ Mounting Media with DAPI , sealed to the glass slides with clear nail polish , allowed to dry for 24 h at RT , and stored at -20°C . Imaging was performed with a Zeiss Axiovert 200M deconvolution microscope using Zeiss Plan-Apochromat 63X/ aperture 1 . 4 objective with oil immersion , using AxioVision Rel . 4 . 8 software . For quantification , exposure times were set so that all measured PLA spots and Gag IF signal in test fields fell below saturation . Once a non-saturating exposure time was identified , five fields containing at least three IF-positive cells ( for Figs 8 and 9 ) or PLA-positive cells ( for Fig 10 ) were chosen at random and imaged using identical exposure times for the red channel , and identical exposure times for the green channel ( red/green exposures were 1 sec/1 . 5 sec for Fig 8; 2 sec/1 sec for Fig 9; and 40 msec/250 msec for Fig 10 ) . Images were captured as ten 1-μm Z-stacks centered on the focal point for the PLA . Images were deconvolved using the AxioVision software , then exported as . tif files , and Image J was used to outline Gag-positive cells in each field . Within those positive cells , the central Z-stack image was used to count PLA “spots” , and quantify IF intensity where indicated , using Image J . PLA spot number for each field was then normalized to the average IF intensity within that field , and the results were plotted with error bars representing the SEM for five fields . For Fig 10 , Gag-DDX6 PLA was performed either with concurrent Gag IF or concurrent DDX6 IF , and the PLA fields with Gag IF were used for PLA spot quantitation to allow exclusion of background spots in Gag-negative cells , while the PLA fields with DDX6 IF were used for P body quantitation . Because spots in fields used for quantitation were difficult to see in figures , the gain was increased proportionally in quantified images solely for the purpose of display . Specifically , after imaging and quantification , the red channel gain in representative images was increased proportionally ( from 1 to 3 in Fig 8 , to 11 in Fig 9 , and to 7 in Fig 10 ) using the AxioVision Rel . 4 . 8 software for all conditions , to allow better display of red spots in final figures . The same was done for the green channel gain ( increased from 1 to 7 ) to display smaller DDX6 granules in the Fig 8D insets . Images were imported in 8-bit color into Adobe Illustrator to create the final figure layout , without further adjustments to color balance or gamma correction . Data shown are from one experiment that is representative of two independent replicate experiments . PLA experiments in S6 Fig were carried out as above using the red PLA reagents , with the following modifications . First , where indicated , NI antibodies were used in place of αGag or αDDX6 primary antibodies in PLA negative controls . Secondly , for Gag-Non-immune ( NI ) PLA , as well as DDX6-Gag PLA in both mock- and Gag-transfected cells , concurrent IF was performed as described above using 1:1000 Alexafluor 488 anti-mouse secondary antibody; however , for NI-DDX6 PLA , a sequential IF was performed , in which , following the final PLA washes , samples were incubated with anti-Gag primary antibody for 30 min at RT . Samples were then washed and incubated for 15 min at RT with 1:1000 Alexafluor 488 anti-mouse secondary antibody . This allowed PLA spots to be counted in Gag-positive cells for both NI conditions . For NI quantification , five fields were chosen at random as above and imaged using identical exposure times for the red channel . The green channel was imaged using identical exposure times for Gag-DDX6 and Gag-NI fields , but exposure time was increased for the NI-DDX6 Gag IF condition to match Gag-DDX6 and Gag-NI signal saturation , since the sequential IF method gave lower overall signal than the concurrent IF method ( red/green exposures were 1 sec/1 sec for Gag-DDX6 and Gag-NI; 1 sec/2 sec for NI-DDX6 ) . Also , as above , after imaging and quantitation , the red channel gain was increased proportionally to 3 in the AxioVision Rel . 4 . 8 software for all conditions for display purposes only , without further adjustments in the final layout . For IF alone in S7 Fig , 293T cells were plated , transfected as indicated , and harvested as for PLA . However , after blocking , cells were incubated with primary antibody ( described under IP methods above ) in PLA antibody diluent at 37°C for 1 h ( or with buffer for the secondary antibody alone group ) , followed by four Buffer A and two Buffer B washes . After these washes , cells were incubated with 1:1000 anti-mouse IgG conjugated to Alexafluor 594 and anti-rabbit IgG conjugated to Alexafluor 488 ( secondary antibodies ) for 30 min at RT . Finally , cells were washed three times with Buffer B and once with 0 . 01X Buffer B , and then mounted as for PLA . Imaging was also performed as for PLA , with red/green exposures of 1 . 25 sec/0 . 5 sec–exposure time for the green channel was lowered from the automatic threshold determined by the imaging software to avoid saturating P body signal . Five fields were imaged for each condition , with each field containing at least two Gag-positive cells . For quantification , two saturation thresholds were chosen in the ImageJ software , one to capture all the DDX6 IF signal in the entirety of all Gag-positive cells , with Gag-positive cells defined by the red channel , and one to capture only the signal from the P bodies within these cells . The total area and mean signal over this area were then determined for the green channel ( DDX6 IF or secondary antibody only ) using both thresholds in the DDX6 IF condition , and using only the threshold encompassing the entire Gag-positive cell in the secondary-antibody-only condition . These calculations were performed for all five fields for each condition . The mean signal within all Gag positive cells in the DDX6 IF green channel was termed the “P bodies + low intensity signal + background” value ( Total signal ) , and in the secondary antibody only condition was termed the “Background” value ( Background signal ) . To subtract P body signal , for each field the mean total signal and mean P body signal were multiplied by their total area area to get a value for each representing the total DDX6 IF signal intensity in the saturated area , and the P body signal value was subtracted from the total signal value . The area of the P body regions was then subtracted from the area of the Gag-positive regions , and the new P-body-subtracted signal value was divided by the new P-body-subtracted area to calculate a “Low intensity signal + Background” fluorescence intensity value . The means for each of these three values ( “P bodies + Low intensity signal + Background” ( Total signal ) , “Low intensity signal + Background” , and “Background” ) were then plotted for the five imaged fields as percent of Total signal , with error bars representing the SEM , and p values calculated with n = 5 fields using the Student's t-test ( two-tailed ) . Protocol for PLA with concurrent IF is available at: dx . doi . org/10 . 17504/protocols . io . k7yczpw HeLa-MCP-GFP cells were transfected with the indicated constructs . Cells were harvested at 24 h posttransfection in fixative ( 3% paraformaldehyde , 0 . 025% glutaraldehyde in 0 . 1 M phosphate buffer , pH 7 . 4 ) , pelleted , and subjected to high pressure freezing using the Leica EMPACT2 , followed by freeze substitution . Samples were infiltrated overnight with LR White embedding resin ( London Resin Company Ltd , Reading , Berkshire , England ) in ethanol , changed to straight LR White , embedded in gelatin capsules ( Electron Microscopy Sciences ( EMS ) , Hatfield , PA , USA ) , and cured overnight in a UV light cryo-chamber at 4o C . Sections ( ~50 nm ) were placed on grids , treated with 0 . 05 M glycine for 20 min at RT , rinsed in PBS , blocked for 45 min with 1% bovine serum albumin ( EMS ) , and washed in PBS with 0 . 1% bovine serum albumin-C ( BSA-C ) ( EMS ) . For immunogold double labeling , a previously described peptide-specific antiserum directed against DDX6 was affinity purified , desalted , and concentrated [22] . Grids were blocked in 0 . 5% BSA-C , then incubated with rabbit αDDX6 ( 0 . 1 mg/ml in 0 . 5% BSA-C ) , followed by goat anti-rabbit F ( ab’ ) 2 fragment secondary antibody conjugated to 15nm gold particles ( EMS ) , with washes after each step . Grids were then labeled with the second primary , mouse αGFP ( Roche ) at 0 . 2 mg/ml in 0 . 1% BSAC with 0 . 002% Tween , followed by goat anti-mouse F ( ab’ ) 2 fragment conjugated to 6 nm gold particles ( EMS ) . Fixation , negative staining , imaging with the JEOL-1400 transmission electron microscope , and image acquisition have been described previously [22] . DDX6 antibody was validated previously for EM , including using DDX6 knockdown cells [22] . GFP antibody was validated by showing absence of labeling in non-transfected control cells . For quantification , images were acquired for ten cells from each of the three groups , with the goal being to analyze similar total PM lengths in each group . Cells were chosen randomly , but excluded for the WT and Gag Zip groups if they had fewer than ten particles at the PM visible at low power . Images encompassed the area of each cell that contained PM assembly sites , with images obtained for ~250 μm of PM total per group . The number of assembly sites analyzed within this ~250 μm of PM are not equivalent since the number of assembly sites depends on VLP phenotype and kinetics . A total of 760 WT events and 409 Gag Zip events were analyzed , but are shown as number of sites per 25 μm PM per cell in S1 Table . Each PM assembly site was scored as genome positive ( g+ ) , DDX6+ ( D+ ) , or double labeled ( g+D+ ) . The following definitions were used for image analysis: early PM assembly sites were defined as displaying curvature at the membrane but with < 50% of a complete bud; late assembly sites at the PM were defined as displaying curvature but with ≥ 50% of a complete bud . If early or late sites contained two or more small gold particles within the full circle defined by the bud , they were scored as g+ . If these sites contained one or more large gold particles within a 150 nm perimeter outside the full circle defined by the bud ( roughly the size of an RNA granule plus space to account for the antibodies and gold particle bound to an antigen at the periphery of such a granule ) , then they were scored as D+ . S1 Table shows the average number of early , late , and early+late PM assembly events per 25 μm of PM per cell ( n = 10 cells +/- SEM ) , along with the breakdown of how many of these events were g+ ( total vs . single-labeled ) , D+ ( total vs . single-labeled ) , or g+D+ ( double-labeled ) . In italics are g+ , D+ , and g+D+ per 25 μm of PM per cell as a percentage of the total for each group . Significance was determined on percentage data using a two tailed t-test; not significant was defined as p > 0 . 01 . Labeling of early+late events as a percent of total early+late events is also shown in graphical form in Fig 11B . As described previously [18] , the sensitivity of IEM for capturing colocalization is limited by a number of factors including the fact that the 50 nm sections only capture ≤ 50% of a single capsid , which has a diameter of ~100–150 nm . | During HIV-1 immature capsid assembly , packaging of the viral genome is initiated when the HIV-1 capsid protein , Gag , first associates with unspliced HIV-1 RNA . Although the complex in which this association initially occurs is critical for formation of infectious virus , the identity , composition , and the mechanism by which this complex forms remain unknown . To address this question , we utilized a previously described temporal pathway of intermediates in HIV-1 immature capsid assembly . The late intermediates in this pathway are derived from host RNA granules , which are diverse complexes utilized for cellular RNA storage and degradation . Here we sought to identify the intracellular capsid assembly intermediate in which HIV-1 Gag initially associates with unspliced HIV-1 RNA . We failed to detect an association between the first assembly intermediate , which contains soluble Gag , and unspliced HIV-1 RNA . Instead , the association between Gag and unspliced HIV-1 RNA was observed only in complexes corresponding to the RNA-granule-derived assembly intermediates . We also showed that Gag uses two determinants to form RNA-granule-derived intermediates that contain unspliced HIV-1 RNA . Together , these studies support a novel model for HIV-1 genome packaging , in which the first association between HIV-1 Gag and unspliced HIV-1 RNA occurs within a host RNA granule . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"293t",
"cells",
"pathogens",
"biological",
"cultures",
"messenger",
"rna",
"microbiology",
"immunology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"rna",
"viruses",
"materials",
"science",
"materials",
"physics",
"sedimentation",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"t",
"cells",
"viral",
"packaging",
"hiv-1",
"viral",
"replication",
"cell",
"lines",
"ribosomes",
"physics",
"biochemistry",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"virology",
"viral",
"pathogens",
"ribonucleoproteins",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"lentivirus",
"organisms"
]
| 2018 | Identifying the assembly intermediate in which Gag first associates with unspliced HIV-1 RNA suggests a novel model for HIV-1 RNA packaging |
Finding where transcription factors ( TFs ) bind to the DNA is of key importance to decipher gene regulation at a transcriptional level . Classically , computational prediction of TF binding sites ( TFBSs ) is based on basic position weight matrices ( PWMs ) which quantitatively score binding motifs based on the observed nucleotide patterns in a set of TFBSs for the corresponding TF . Such models make the strong assumption that each nucleotide participates independently in the corresponding DNA-protein interaction and do not account for flexible length motifs . We introduce transcription factor flexible models ( TFFMs ) to represent TF binding properties . Based on hidden Markov models , TFFMs are flexible , and can model both position interdependence within TFBSs and variable length motifs within a single dedicated framework . The availability of thousands of experimentally validated DNA-TF interaction sequences from ChIP-seq allows for the generation of models that perform as well as PWMs for stereotypical TFs and can improve performance for TFs with flexible binding characteristics . We present a new graphical representation of the motifs that convey properties of position interdependence . TFFMs have been assessed on ChIP-seq data sets coming from the ENCODE project , revealing that they can perform better than both PWMs and the dinucleotide weight matrix extension in discriminating ChIP-seq from background sequences . Under the assumption that ChIP-seq signal values are correlated with the affinity of the TF-DNA binding , we find that TFFM scores correlate with ChIP-seq peak signals . Moreover , using available TF-DNA affinity measurements for the Max TF , we demonstrate that TFFMs constructed from ChIP-seq data correlate with published experimentally measured DNA-binding affinities . Finally , TFFMs allow for the straightforward computation of an integrated TF occupancy score across a sequence . These results demonstrate the capacity of TFFMs to accurately model DNA-protein interactions , while providing a single unified framework suitable for the next generation of TFBS prediction .
Transcription factors ( TFs ) and their specific binding sites act to modulate the rate of gene transcription . They are central to key biological processes , such as organ development and tissue differentiation , nutrient and environmental stress responses and physiological signals . Delineating specific positions at which TFs bind to DNA is of high importance in deciphering gene regulation at the transcriptional level . Each TF binds a variety of DNA sites with sequence-specific affinity [1] . As TFs bind to DNA in a sequence specific manner , computational methods for motif discrimination have been critically important for the prediction of transcription factor binding sites ( TFBSs ) . Unfortunately , TFBSs are usually short and in most cases TFs are tolerant of sequence variations at many positions of the TFBS . This tolerance for variation impacts the accuracy of genome-scale prediction of TFBSs , as suitable TFBS sequences are found frequently . The accurate prediction of TFBSs is an enduring challenge [2] , with ongoing approaches introduced to decrease the high rate of false predictions . In a human cell , it appears that most computationally predicted TFBS are not available for binding , presumably due to chromatin structure and/or local epigenetic properties . Approaches to address the false prediction issue have varied: phylogenetic methods to focus on sequences conserved during evolution [3] , using experimentally mapped transcription start site data to focus on promoter proximal regions [4] , using histone modification or DNA accessibility data to highlight likely regulatory sequences [5] , or focusing on locally dense combinations of motifs [6] , [7] defined from TFBS enrichment analysis of co-expressed genes . Classically , computational prediction of TFBSs is based on models called position weight matrices ( PWMs ) that reflect the preferred binding motifs associated to corresponding TFs by providing an additive score for any sequence . They approximate the true specificity of a TF and their parameters can be estimated through different methods ( see [8] for a review ) . The basic PWMs make the assumption that each nucleotide within a TFBS participates independently in the corresponding DNA-protein interaction . Basic PWMs can perform well in modeling TFBS properties , but do not account for position interdependencies , that have repeatedly been observed to exist within TFBS motifs using crystal structure analyzes [9] , biochemical studies [10]–[12] , statistical analyzes of large collections of TFBS [13]–[15] , or quantitative analyzes of protein binding microarray ( PBM ) data [16]–[18] . The latter study based on PBM data has demonstrated that position dependencies are stronger between neighboring positions than others . Incorporating these dinucleotide dependencies into the binding models has been shown to improve predictive accuracy [18] , albeit with modest impact in most cases . The recent comparison of several algorithms using PBM binding assays data in [17] also included models considering dinucleotide composition . Recently , a new approach called dinucleotide weight matrices ( DWMs ) has been developed to extend the basic PWMs by considering the dinucleotide interactions between all pairs of positions within the TFBSs [19] . Moreover , basic PWMs are restricted to the detection of motifs with a fixed length . This constraint has previously led to alternative heuristic approaches for the modeling of TFBS for TFs tolerant of variable widths , such as nuclear receptors [20] and p53 [21] , [22] . The analysis of variable spacing in the context of Escherichia coli promoter prediction was an early advance in the field , with the spacing between elements addressed with either PFMs [23] or PWMs using logarithms of the probabilities [24] ( see [8] for a review ) . With the growth in TFBS data , early findings of variable TFBS configurations [25] , [26] are not unique [27] . Several efforts have explored more flexible models for the prediction of TFBSs . Bayesian hierarchical hidden Markov models ( HMMs ) have been used to model cis-regulatory modules ( CRMs ) in [28] but TFBSs included in the CRMs are predicted using basic PWMs . In [29] , the authors present a Boltzmann chain ( i . e . an HMM generalization ) to model the competition between DNA-binding factors as TFs which TFBSs are predicted using energy PWMs computed from PBM data . HMMs , which have been widely used in computationally biology for the prediction of protein motifs , have also been applied to the identification of TFBS . Implementations based on HMMs have been made for the detailed study of specific TFs or classes of TFs [20] , [25] , [30] , and in a few cases the methods were generalized as a framework to theoritically analyze TFBS features [11] , [31] , [32] . The MAPPER software was implemented using HMMs [33] , [34] , but the approach retained the classic focus on positional independence consistent with the preponderance of data available at the time . The HMMs approach used in [35] allows for flexible length motifs through the use of profile HMMs , with the unconstrained potential for gaps to be incorporated at any position , as previously introduced for protein families [36] . To accommodate dependencies between TFBS positions , [37] introduces a variable-order Bayesian network to model TFBSs , but suggested that greater training data would allow HMM-based approaches to model dependencies at all positions . The growing community interest in the use of HMMs and more advanced models to discriminate TFBS reflects an underlying expectation that emerging data can lead to more effective models . Recently , a new experimental technique has been developed to study sequences where proteins interact with DNA . This procedure is a combination of chromatin immunoprecipitation and massively parallel sequencing technologies - the well-known ChIP-seq procedure [38] . It gives , with good sensitivity and specificity , DNA sequences to which proteins of interest bind , providing the opportunity to precisely map those binding sites within the genome . Using such data , we can analyze in depth transcription regulation by focusing on DNA sequences that are bound by specific TFs . The availability of thousands of experimentally validated DNA-TF interaction sequences coming from ChIP-seq data and stored in databases such as produced by the ENCODE project [39] allows researchers to develop new approaches for the prediction of specific locations of TFBSs with greater confidence than was previously possible . We introduce here a novel TFBS model and prediction system based on HMMs , hereafter referred to as TF Flexible Model ( TFFM ) . Building upon previously developed models capturing dinucleotide dependencies and flexible lengths as described previously ( see Table S1 for comparison of TFFMs with previous HMM-based TFBS predictions ) , our approach allows for the capture of these different features within a unique framework . The availability of thousands of ChIP-seq regions for a TF , potentially representing the full diversity of TFBS configurations , motivates the effort to transition to an HMM-based approach to TFBS prediction . The new HMM framework is flexible , supports dinucleotide composition analysis and variable lengths to predict TFBSs . The performance of TFFM was compared to established TFBS prediction methods through analysis of numerous ChIP-seq data sets . Most methods are comparable for TFs exhibiting classic TFBS structures , but TFFMs show distinct performance advantage for the subset of TFs with more diverse binding characteristics . By evaluating the correlation between TFFM scoring with ChIP-seq peak signals and experimentally measured DNA-binding affinities , we found that TFFM scores reflect TF-DNA interactions . Moreover , the probabilistic scheme of the TFFMs allows for a straightforward calculation of a total occupancy score for a DNA region . Researchers may construct and apply the TFFMs through open-source code via an application programming interface at http://cisreg . cmmt . ubc . ca/TFFM/doc/ or directly through our web-based application at http://cisreg . cmmt . ubc . ca/TFFM/ . TFFMs build upon the best properties of the established methods , while offering novel capacities within a unified framework . Every TF has its own DNA binding characteristics ( position dependencies , spacing , variable flanking regions , occupancy ) which can be captured within the unified TFFM-framework .
We present a new HMM-based framework to model and predict TFBSs . HMMs have been extensively used in computational biology to model DNA sequences [36] . They offer a flexible probabilistic method that gives us the opportunity to model TFBSs with their dinucleotide characteristics and that can be extended to take into consideration flexible motifs . In the context of modeling DNA sequences , an HMM is composed of a set of hidden states emitting nucleotides with defined probabilities ( corresponding to the set of emission probabilities ) and a set of transition probabilities from state to state . HMMs conveniently accommodate large data , deriving an optimal ( at least locally ) set of probabilities . HMMs can be trained using different well-established algorithms such as the Baum-Welch algorithm [36] or the Viterbi [36] algorithm . We chose the widely used Baum-Welch algorithm since it converges to a local optimum depending on a set of initialized probabilities . Recent advances in the prediction of TFBSs have incorporated inter-positional properties through the analysis of dinucleotide properties across the sites . To construct models capturing the dinucleotide compositional properties of TFBSs , we implemented two HMM-based approaches . Initially , we constructed standard first-order HMMs as TFFMs ( denoted later as 1st-order TFFMs ) . In such models , each position within a TFBS is represented by a state emitting a nucleotide with probabilities dependent on the nucleotide found at the prior position ( see Figure 1A ) . In 1st-order TFFMs , we considered nucleotides surrounding the TFBSs ( i . e . nucleotides located before and after a TFBS ) by using a specific state modeling the background . The models developed here are similar to the probabilistic dinucleotide PWMs used in [40] to model nucleosomes . However , the 1st-order TFFMs add the capability to capture the properties of the surrounding sequences through the background state , introduce motif length flexibility , and are built upon HMMs as a more dedicated probabilistic framework to model stochastic sequences . In 1st-order TFFMs , starting a TFBS is given by a unique probability ( representing the transition from the background to the TFBS ) whatever the nucleotide found in the surrounding sequence . To allow for starting a TFBS depending on the nucleotide emitted in the background state , we implemented a more detailed and descriptive HMM template as TFFMs ( denoted later as detailed TFFMs ) mimicking the theoretical analysis made in [11] ( see Figure 1B ) . The intrinsic different structures of the 1st-order and detailed TFFMs lead to different probabilities for going from the background to the foreground . This is emphasized by the training through the Baum-Welch algorithm which reaches local maximum . By constructing models taking into consideration local dinucleotide dependencies , we aim to better model , characterize , and understand TFBS properties . When trying to analyze and understand a model , a visual representation provides insight into the underlying properties . Basic PWMs for instance can be graphically represented using sequence logos [41] where each position gives the information content obtained for each nucleotide . The greater the height of the letter corresponding to a nucleotide , the higher the information content and higher the probability of getting it at this position . Using HMMs , we can derive the probability of obtaining each nucleotide at each position , allowing for the generation of sequence logos representing the TFBSs modeled using what we call a summary TFFM logo ( see Figure 2B ) . In summary TFFM logos , we use the probability of each nucleotide at each position to compute the corresponding information content represented in the logo . But this graphical representation fails to convey the local dinucleotide dependencies that motivate the work . To tackle this issue , we introduce a new graphical representation ( which we call a dense TFFM logo ) allowing researchers to perceive the dinucleotide dependencies captured by the model . As the emission of a nucleotide at each position depends on the nucleotide emitted at the previous position , we represent the nucleotide probabilities at position for each possible nucleotide at position . Hence , each column represents a position within a TFBS and each row the nucleotide probabilities found at that position . Each row assumes a specific nucleotide has been emitted by the previous hidden state . The intersection between a column corresponding to position and row corresponding to nucleotide gives the probabilities of getting each nucleotide at position if has been seen at position ( see Figure 2A ) . For instance , in Figure 2C , we can observe that a “C” is more likely to appear at position 12 if nucleotide “T” was found at position 11 ( green box and arrow ) whereas a “T” is more likely to appear at position 12 if nucleotide “G” was present at position 11 ( orange box and arrow ) . In order to highlight the most probable row to be used by the TFFM , we vary the opacity to represent the sequence logo . The higher the probability of getting a nucleotide at position , the higher the opacity of the row corresponding to this nucleotide at position . Unfortunately , the current graphical representation does not allow for variable length or spacing of the motif modelled . One could envision introducing the variable length HMMs graphical representation by following the tool specifically developed in [42] for protein families . We provide the TFFM-framework to construct TFFMs from ChIP-seq data sets and to predict TFBSs within DNA sequences . When constructing a TFFM from ChIP-seq data , we extract ( using MEME [43] , [44] ) the most over-represented motif out of the top scoring ChIP-seq sequences and use it to initialize the HMM probabilities ( see Materials and Methods ) . Then , the final probabilities are learned using the Baum-Welch algorithm . When predicting TFBSs within DNA sequences using a TFFM , the software gives , at each position within the sequence , the probability of being in a final matching state ( corresponding to the last position of a TFBS ) in the underlying HMM . These probabilities correspond to posterior probabilities given an HMM and are computed using the well-known forward and backward algorithms [36] . When assessing the predictive power of the models , one can vary a threshold through these output probabilities to compute values of sensitivity and specificity . 1st-order and detailed TFFMs have been constructed using ENCODE [39] ChIP-seq training data sets . The trained TFFMs were used to predict TFBSs within test ChIP-seq data sets by following a 10-fold cross-validation methodology . Finally , the results obtained with TFFMs were compared to the ones obtained from PWMs and DWMs constructed from and applied to the same data . All ChIP-seq ENCODE data sets from human and mouse ( with at least 1800 peaks and a peak max position indicated , i . e . 206 data sets ) were used to compare the two types of TFFMs with PWMs and DWMs . Sequences around the peak max positions ( 50 nucleotides on both sides ) were extracted to construct the models and make predictions . The rationale for this is that ChIP-seq peak max positions represent where the maximum amount of ChIP-seq reads map on the genome of reference and TFBSs are expected to be strongly enriched in close proximity to the peak max position [45] . For each data set , the 600 peaks with the highest signal were used to extract the most over-represented motif within the sequences and to initialize the model probabilities . To avoid over-fitting the data when assessing the predictive power of the different models , we used the remaining sequences to construct training and testing data sets following a 10-fold cross-validation approach . Assuming that high quality ChIP-seq data contain at least one true TFBS within each peak region , we considered the hit ( matching sequence ) with the best score per peak as a TFBS prediction . To assess the level of specificity for each method , we generated background data sets by randomly drawing sequences from mappable regions of the human and mouse genomes by keeping the same %GC composition distribution as in the ChIP-seq testing sequences ( see Materials and Methods ) . Another set of background sequences has been generated from a first-order HMM reflecting the background dinucleotide composition of each ChIP-seq testing data set . Across varying thresholds for TFBS prediction scores for the four different models , we calculate predictive sensitivity and specificity at each threshold value and trace the corresponding receiver operating characteristic ( ROC ) curves . For each ENCODE ChIP-seq data set , the area under the curves ( AUC ) for the corresponding ROC curves ( for all predictive methods ) have been computed . To compare the predictive powers of the different methods , we focus on ChIP-seq data sets for which at least one predictive method achieves an when discriminating ChIP-seq from genomic background sequences ( i . e . 96 ChIP-seq data sets ) . We plot the ratios of performance between the best model and the others on the set of ChIP-seq data for which at least one predictive method is of high quality . In Figure 3 , we show the ratio of performance to the best model for ChIP-seq data giving a high quality discriminative performance . When considering a similar performance between models when the ratio of the AUCs is above 95% , we show that the specificity of the binding proteins are captured similarly by using weight matrices ( WMs ) or TFFMs . Where the performance ratio is below 95% , we can observe an increase in discriminative power in favour of the TFFMs when compared to the WMs ( compare the right part of Figure 3 to the left part ) . When considering the strict difference between respective performance ( i . e . when getting strictly higher AUC values ) , the results indicate that the TFFMs are performing strictly better than both the PWMs and the DWMs in discriminating ChIP-seq peak sequences from background sequences in two thirds of the data sets . Namely , the 1st-order and the detailed TFFMs are performing strictly better than both the PWMs and the DWMs for 63 and 65 data sets , respectively , over the 96 ChIP-seq data sets considered ( see Figure 3 ) . Taken together , the TFFMs perform strictly better than WMs in 67 data sets over 96 ( 70% ) . Similar results ( 65 over 94 , 69% ) are obtained when considering background sequences generated from a first-order HMM ( see Figure S1 ) . Explicit AUC values are plotted in Figure S2 . We can observe that the TFFMs perform as well as or better than WMs overall on the sets of ChIP-seq data for which at least one predictive method is of high quality when discriminating ChIP-seq sequences from genomic background sequences . See similar results in Figure S3 when considering HMM-generated background sequences . Statistical significances of the differences in terms of discriminative power between the different methods has been computed for each pair of methods using a Wilcoxon signed rank test [46] assuming the null hypothesis of a symmetric distribution of AUC differences around 0 when two methods perform similarly . Table 1 contains the Benjamini-Hochberg [47] corrected statistical significance of the differences between each one of the predictive methods when considering data sets where at least one method obtains an . It shows that the performance difference between the two TFFMs is not significant with a -value of 0 . 528 . On the contrary , the difference of performance when comparing both of the TFFMs with the PWMs and the DWMs is statistically significant ( see Table 1 where the maximal -value is equal to ) . These -values indicate that the null hypothesis can be rejected . PWMs and DWMs have more comparable behavior since the difference in performance is less statistically significant ( , see Table 1 ) . Again , similar results are obtained when considering HMM-generated background sequences ( see Table S2 ) . To understand whether the TFFMs perform better than the PWMs because of the model or because of the training method ( as both differ ) , we introduce a 0-order TFFM which is basically a PFM modeled by an HMM and trained using the Baum-Welch algorithm ( see Figure S14 ) . We used the 0-order TFFMs to discriminate ChIP-seq sequences from background sequences on the same data sets as previously and computed the corresponding AUCs . In Figure 4A , we observe that the 1st-order and detailed TFFMs outperform the 0-order TFFMs ( best AUC of the 1st-order or detailed TFFMs in 90 data sets out of 96 , i . e . 94% ) emphasizing the need to consider dinucleotide dependencies in the models . In Figure 4B , we compare the 0-order TFFMs to the WMs . The discriminative power of the WMs is higher than the 0-order TFFMs ( WMs obtain strictly better AUCs than 0-order TFFMs in 67 data sets out of 96 , i . e . 70% , and PWMs perform strictly better than 0-order TFFMs in 63 out of 96 data sets , i . e . 66% ) showing that the good performance of the 1st-order and detailed TFFMs does not directly arise from the training methodology since PWMs are performing better , overall , than 0-order TFFMs . Similar results are obtained when using HMM-generated background sequences ( see Figure S4 ) . This analysis shows that the TFFMs perform better than PWMs and DWMs more often , with a statistically significant difference , and lead us to hypothesize that the TFFMs are , overall , better at capturing TFBS features found in the experimental data . To further evaluate this property , we analyzed how TFFM scoring correlates with the biological signal found in ChIP-seq data . An attractive feature of PWMs is that they can produce scores that are correlated with the energetic binding affinity between a protein and a DNA sequence [48]–[50] . We sought to confirm this property using ChIP-seq data and estimate whether the TFFMs similarly exhibit this capacity . From ChIP-seq experiments , we will assume that the maximum number of reads mapping a peak is representative of the DNA-binding affinity of the corresponding TF to the corresponding DNA sequence ( while recognizing the limitations of this assumption ) . We assess whether the different tested scoring models correlate with the peak signal ( corresponding to the level of enrichment for TF-binding within the region ) when ranking the peaks with respect to their ChIP-seq signal values compared to ranking with respect to the scores of their best hit . For each data set used in the previous AUC analysis ( note that the 600 best peaks are not considered in the 10-fold cross-validation ) , we extracted the predicted scores associated with each ChIP-seq peak on the foreground testing sets for each of the four predictive methods . ENCODE ChIP-seq data also contain a ChIP-seq signal value associated with each one of the peaks . A potential relationship between prediction scores and peak signal values has been evaluated by separating the ChIP-seq peaks into twenty 5-percentile groups using ChIP-seq peak signal values . Spearman's rank correlation coefficients [51] were computed using the percentiles as the x-axis and the median of prediction scores as the y-axis . The density distribution of Spearman's rank correlation coefficients for each predictive method is given in Figure 5A . The correlation between the prediction scores ( for all of the four methods ) and the signal value of the corresponding ChIP-seq peaks is mainly located around 1 ( with WMs giving slightly better results , see Figure 5A ) , corresponding to a correlation between the scoring of the methods and ChIP-seq peak signal values . Spearman's rank correlation coefficients indicate that the higher the ChIP-seq peak score , the higher the score we expect to get from the different predictive methods , which is in agreement with the models reflecting the energetic binding affinity between a protein and a DNA sequence . We used 5-percentile groups of sequences ( using ChIP-seq peak signal values ) to compute Spearman's rank correlation coefficients . The Spearman's rank correlation coefficients computed using all the ChIP-seq peak signal values would result in low correlation values due to the variation of the values from the average trend . Hence , using percentiles gives the general trend that shows the average behaviour of the correlation . We can see from Figure 5A that the trends show a positive correlation . The statistical significance of the trends is assessed by computing the -values of getting a regression line ( corresponding to the general trend of correlation ) for the observed slope . Figure S5 plots the distribution of the computed -values . We observe that the -values are , overall , close to 0 and indicate that the average behaviour of the correlation given by a linear regression is far from a horizontal line . These results show that all of the predictive methods have the tendency to correlate with ChIP-seq peak signal values . To further analyze the property of correlation between TFFM scores and binding affinities , we considered experimentally measured binding affinities between a TF and DNA sequences and compared these values to the predictive scores obtained from the different models . In the previous section , we hypothesized that the signal values from ENCODE ChIP-seq peaks reflected the affinity of the TF protein to bind to DNA sequences . In [52] , Maerkl and Quake measured experimentally the binding affinities of two isoforms of the human Max TF to DNA sequences by testing sequence permutations . The Max TF binds to a core consensus motif of the form CACGTG . As the motif is palindromic , they measured the binding affinities of the Max proteins for all the possible mutated sequences of the first half of the core motif ( namely nucleotides to from the consensus sequence , see Figure S6 ) and conserving the second half of the consensus core binding sites ( GTG ) . We compared these experimental measures to the predictive scores obtained on the Max K562 ChIP-seq data , selected from the available Max ChIP-seq data as it exhibits the highest AUCs in the 10-fold cross-validation experiments for all tested methods . Five different models ( 0-order TFFM , 1st-order TFFM , detailed TFFM , PWM , and DWM ) were constructed from the top 600 peaks for the initialization step and all ChIP-seq sequences were used to train the models . The sequence logo obtained using MEME on the top 600 peaks is given in Figure S6 for reference . We can see that the CACGTG palindromic motif is captured by MEME using the ENCODE ChIP-seq data set . The trained models were then applied to the mutated sequences to get their corresponding predicted scores . Note that the models are trained here on a data set that is independent of the testing data set used in [52] . For each predictive model , we computed the correlation between the predicted scores and the DNA-binding affinity values measured experimentally . Since some mutated sequences can no longer be bound by the Max TF ( or with very weak affinity ) , it is interesting to focus on the sequences to which the TF can actually bind . Hence , we analyzed the correlation between predicted and experimentally measured DNA-binding affinity values by first focusing on the sequences lying in the top 10-percentile affinity values , then the top 20-percentile , and up to including all the sequences using 10-percentile steps . The results of higher interest , corresponding to stronger DNA-binding affinity values , are the top percentiles but all percentiles were computed for completeness . Using such a methodology , we expect the predicted scores obtained from the models to better correlate with high DNA-binding affinity values than with low values . Figure 5B gives the Pearson correlation coefficients with respect to the top percentiles of DNA-binding affinity values for both isoforms of the Max TF . We observe that all the TFFMs ( whatever the order ) correlate well with the experimental data with a Pearson correlation coefficient over 0 . 6 up to the top 80-percentiles . Coefficients obtained when considering the top 10- and 20-percentiles ( when TF-DNA interaction is the strongest ) are around 0 . 7 , showing the high correlation between TFFM scoring and experimental values . In contrast , scores obtained from PWMs and DWMs give Pearson correlation coefficients in the range of 0 . 45–0 . 55 . Analysis of the Max data indicates that the TFFMs reflect the DNA-binding affinities measured experimentally for the Max TF . To understand what characteristic ( s ) the TFFMs are capturing that is not represented by either the PWMs or the DWMs , we examined the DNA sequences obtaining the highest DNA-binding affinity values . We looked at the motifs for which the DNA-binding affinities are the highest by considering the top-scored 25 , 50 , 75 , and 100 sequences ( see Figure S7 ) . Figure S8 contains the sequence logos corresponding to the TFFMs and the PWM ( no such representation has been made available for the DWM in [19] ) . One can observe that the TFFMs better capture the C/T pattern at position 7 that is found strongly at position 4 of the top DNA-binding affinity sequences ( Figure S7 ) . To a lesser extent , the same can be observed at position 6 with the A/G captured by the TFFMs . These patterns are not captured by the PWM since the CAC is strictly expected at positions 5–7 , this coming from the construction and training of the PWMs made from the over-represented motif with a strict CAC . Hence , the PWM does not reflect the needed flexibility captured by the TFFMs through the training step , which is captured by the 0-order TFFM ( conceptually the same as a PWM but with a different method of training ) . In [52] , binding affinity differences between the optimal sequence ( with a core CACGTG motif ) and the mutated sequences were computed . We performed the same analysis using the predictive scores for the different models in order to compare the results . Table 2 summarizes the comparison in which we observe that the TFFMs perform better than both the PWM and the DWM . Pearson correlation coefficients for TFFMs are whereas the PWM and the DWM obtain at most a coefficient of 0 . 66 . As a conclusion , these results emphasize the capacity of the TFFM scores to correlate with DNA-binding affinity . In the previous sections , the TFFMs were used to model TFBSs with fixed length by taking into consideration the dinucleotide composition of the sequences . Another feature of the TFBSs that can be accommodated by TFFMs is flexible length . A subset of TFs bind to the DNA with different structural conformations , leading to TFBSs of different lengths [25]–[27] . In order to model a binding site with a flexible length , we can use the transition probabilities of the underlying HMMs of the TFFMs . For instance , a 1st-order HMM state corresponding to position of a TFBS can transition to state at position and to state at position to allow to be omitted for some TFBSs of smaller length . The same applies to HMMs of detailed TFFMs where the probabilities are decomposed for each nucleotide in positions to . To assess the capacity of the TFFMs to model flexible length TFBSs , we applied the 1st-order and detailed TFFMs to ChIP-seq data corresponding to three TFs with potentially flexible DNA-binding motifs ( JunD , STAT4 , and STAT6 ) and compared their discriminative power with fixed length TFFMs , PWMs and DWMs following the cross-validation methodology used previously . We compared the methods to GLAM2 which was first developed to find motifs in proteins with arbitrary insertions and deletions but which can also be applied to DNA sequences [53] . In the previous sections , TFFMs have been used to predict specific TFBS positions . We extended the TFFM-framework to compute an integrated TF occupancy score across a DNA sequence using the TFFM scores . Using the TFFMs , the probability of occupancy ( Pocc ) of a TF within a defined DNA sequence is obtained by multiplying the TFBS probabilities at each position ( see Material and Methods section for details ) . This is a simpler approach than the physico-chemical models used in tools like GOMER [55] and TRAP [56] , and somewhat similar in concept to the approach of OHMM [57] which uses HMMs to predict occupancy of TFs with self-overlapping binding motifs . To observe whether Pocc can improve the discriminative power of the TFFMs , Pocc have been computed from TFFMs and assessed for their capacity to discriminate between ChIP-seq data and background sequences and compared to original TFFMs using the best site per ChIP-seq peak to discriminate between ChIP-seq data and background sequences . When analyzing data sets for which at least one method obtains an ( 97 with a genomic background and 100 with an HMM-generated background ) , we found that using Pocc values improves discrimination as measured by AUC for 62 out of 97 ( i . e . 64% ) data sets when considering genomic background and 63 out of 100 ( i . e . 63% ) when considering HMM-generated background ( see Figure S14 ) .
In this report , we have introduced a flexible HMM-based framework for TFBS prediction . The new models are demonstrated to perform as well as classic methods for most data , while exhibiting improved performance for a subset of TFs . The new approach retains the desirable attribute of producing scores correlated with the binding energy of TF-DNA interactions . A new graphical representation is introduced to illustrate the properties of the models , complementing the classic and widely used sequence logos . In applications , the TFFM models have been shown to handle variable spacing between half sites , and to allow for the incorporation of flanking sequence properties into TFBS analysis . With a convenient software package and a breadth of opportunities for improvement , TFFMs are a suitable foundation for the next generation of TFBS prediction . The new TFFM-framework provides an opportunity for researchers to analyze more deeply the features of TF-DNA binding interaction by looking at local dinucleotide dependencies captured by the TFFMs and represented by the new logos . One can see the TFFMs as the probabilistic analog of the energetic BEEML models developed in [18] . Unfortunately , the two modeling approaches cannot be directly compared since they are elaborated from two different types of data sets ( PBM data for the BEEML tool [18] and ChIP-seq data for the TFFMs ) . BEEML software cannot use ChIP-seq data ( personal communication with the authors ) and the TFFMs have not been developed to consider PBM data information in their current form but both models are able to capture TFBS features with good specificity . For TFFMs , the greatest utility is in handling the growing subset of TFs with complex binding properties . Such complex binding characteristics of TFs may be decomposed into four categories [1] based on the structure of their corresponding motifs: position interdependence ( i . e . the probability of observing a nucleotide at one position is informed by the nucleotide observed at another position ) , variable width , multiple effects where we can observe a combination of position interdependence and variable length , and alternate recognition interface where bound DNA segments cannot be accounted for using models of either variable length or position interdependence . The models we used in these analyzes aim at addressing the three first categories of TFBS characteristics using only one framework , while the framework provides sufficient flexibility to incorporate the fourth , such as subtle flanking sequence properties . The TFFM-framework creates new opportunities for innovation in TFBS bioinformatics analysis . Drawing from the initial studies here , it is apparent that refined approaches can be pursued for the identification of TFs capable of binding to motifs of variable width and the analysis of the role of TFBS flanking sequence on TF binding . While the number of cases of TFs tolerant of variable width binding sites has grown with access to high-throughput TFBS data , the TFFM-framework could be extended to enable a comprehensive survey of ChIP-seq data collections to identify additional cases . As observed in the analysis of MafK TFBS flanking sequences , TFFMs are sufficiently flexible to incorporate additional information represented in TFBS proximal sequences . There have been some indications that such sequences may specify interactions with co-factors [58] . TFFMs offer advantages over past methods for the detection of such weak signals with variable positions . It is our plan to expand the TFFM-framework to automatically look for variable-length motifs . Beyond the analysis of non-canonical TF binding motifs , there is a significant scientific opportunity to develop a new computational approach for the prediction of functionally significant DNA variations within cis-regulatory sequences . The global relationship of TFBSs and nucleotide variations is largely unknown [59] . Recent studies have shown extensive genetic variations on human TFBSs often correlated with differences in gene expression [60] and identified TFBSs as genetic determinants of retroviral integration in the human genome [61] . TFFMs have the capacity of modeling the impact of mutations on the TF-DNA binding affinity , as demonstrated for the Max TF . These early results show the promise for using TFFMs to score the impact of nucleotide variations on TF-DNA interactions . A key to the long-term development and adoption of TFFMs is the access of researchers to both the binding models and the software for their generation . It is our plan to generate a collection of TFFMs trained on ChIP-seq data sets from ENCODE , as well as other sources compiled into the PAZAR repository [62] , [63] . Models shall also be incorporated in the next release of the JASPAR collection [64] , with a parallel release of PWMs constructed from the same data . Analysis of DNA sequences will be supported through both a web application and a standalone version using already trained TFFMs directly downloadable from JASPAR . For bioinformatics research , we have provided the code of the TFFM package and its documentation ( accessible at http://cisreg . cmmt . ubc . ca/TFFM/doc/ ) to allow others to refine the approaches and make further innovations to broaden the use of TFFMs . The new TFFMs described in this report are designed to address the confounding properties of position inter-dependencies in site composition and variable lengths observed in experimental data . These two challenges have emerged as an increasing issue with the availability of large-scale ChIP-seq data , which reveals greater complexity of TFBSs than could be observed in the past . The TFFM graphical motif representation conveys properties of position inter-dependence , allowing researchers to visually analyze the features captured by the model . TFFMs have been assessed on human and mouse ChIP-seq data sets coming from ENCODE , revealing a higher discriminative power than established methods . TFFMs produce scores consistent with observed protein-DNA affinities measured experimentally and have the capacity to predict the impact of TF binding site mutations on TF-DNA binding affinities . The analysis of TFBS is a central challenge in bioinformatics . TFFMs provide a powerful and flexible framework within which a broad range of problems can be addressed . While many motif discrimination methods are available , it is our perception that TFFMs will emerge as a preferred approach for TFBS analysis .
Comparisons between the different predictive methods were done using ChIP-seq data sets from the ENCODE project [39] . We used ChIP-seq data sets from human ( hg19 assembly ) and mouse ( mm9 assembly ) containing at least 1800 peaks for which the peak max position is known ( i . e . narrowPeak-formatted data ) , representing 206 ChIP-seq experiments . In limiting to sets with at least 1800 peaks , we ensure that at least two times the number of peaks used to search for an over-represented motif by MEME ( top 600 peaks used , see below ) will be used during the 10-fold cross-validation procedure as explained below . As DNA-binding events should be located around the peak max area ( corresponding to the highest coverage of ChIP-seq reads ) of the ChIP-seq peaks [45] , we extracted 50 nt on each side of the peak max position . Hence , each peak is composed by 101 nt centered at the peak max position and is associated to a signal value corresponding to the enrichment for TF binding in the region of the peak . For assessing the performance of TFFMs allowing for flexible length motifs , we used the following ChIP-seq data sets: human ENCODE JunD TF from K562 cells by the University of Chicago , mouse ENCODE MafK TF from Ch12 cells by Stanford University , and STAT4 and STAT6 TFs from [54] . 1st-order HMMs used in 1st-order TFFMs are composed of a state modeling the background sequences surrounding TFBSs and one state per position within the TFBSs . The use of a 1st-order HMM allows the model to capture the dinucleotide dependencies through emission probabilities at position dependent on the nucleotide found at position ( see Figure 1A ) . One can move from the background state to the first “matching state” ( i . e . the first position within a TFBS ) with a defined probability , whatever the nucleotide generated by the background state . Figure 1A gives a representation of the 1st-order HMM template where the first state corresponds to the background state representing the nucleotides surrounding TFBSs . Following states correspond to the matching states where each one corresponds to a position within a TFBS . Each state emits a nucleotide with probabilities dependent on the nucleotide emitted by the previous state . Within matching states , moving from one TFBS position to the next is given by transition probabilities equal to 1 . Probabilities are learned using the Baum-Welch algorithm on ChIP-seq sequences , starting from initialized values . HMMs used in detailed TFFMs decompose each state of the 1st-order HMM with four corresponding states in the detailed HMM , each one emitting a nucleotide ( A , C , G , or T ) with a probability equal to 1 ( see Figure 1B ) . Dinucleotide dependencies are modeled by four transition probabilities getting out of each state at position and directing to each state at position ( see Figure 1B ) . HMMs used in 0-order TFFMs are constructed with the same set of states as the ones used for the HMMs of the 1st-order TFFMs . The emission probabilities are different since no dependency between positions is captured . Hence , each state is associated to only four emission probabilities for the four nucleotides ( see Figure S14 ) . The TFFMs provide , at each position within a DNA sequence , the probability of being in a final matching state ( corresponding to the last position of a TFBS ) . Directly following the spirit of [55] , the probability of occupancy ( Pocc ) of a TF , which TFBSs are modeled by a TFFM , on a DNA sequence of length can be computed as:where represents the probability of a TF not occupying the DNA sequence at position . The different model predictive powers were compared using a 10-fold cross-validation methodology on human and mouse ChIP-seq ENCODE data sets . The summarized features captured by the TFFMs are represented through a sequence logo similar to the ones used for basic PWMs . To construct the sequence logos , the probability of getting each one of the four nucleotides is computed at each position starting from an equiprobability of A , C , G , and T in the background . Let be the probability of letter at position . We compute as equal to where corresponds to the emission probability of the nucleotide at position when nucleotide was found at position . When considering the 1st-order TFFMs , we use the emission probability values whereas the transition probability values give the information for the detailed TFFMs . The classic sequence logos do not give any information about the dinucleotide dependencies captured by the TFFMs . We introduce a new graphical representation of the TFBSs modelled by the TFFMs that is able to capture this feature ( see Figure 2 ) . As for a regular sequence logo , each column corresponds to a position within a TFBS . Each row captures the probabilities of each nucleotide knowing the nucleotide at the previous position ( one row per nucleotide A , C , G , and T ) . It follows the same computation as explained above but considering a specific nucleotide found at the previous position . For instance , the probability of emitting at position for row A ( so A was found at position ) is equal to . As for the summary logo , the emission probabilities are used for the 1st-order TFFMs whereas the transition probabilities are used for the detailed TFFMs . The height of the letters reflect their probability ( the greater the height , the higher the probability ) . In order to highlight the preferred rows , the opacity of a case ( intersection of a row and a column ) represents the probability of finding the nucleotide corresponding to this specific row at the previous position of the TFBS ( the higher the opacity , the higher the probability ) . Given the probabilities of finding each nucleotide at each TFBS position , we compute the information content ( IC ) of a TFFM by summing the IC of all the positions computed as . DNA-binding affinities between human Max transcription factor ( isoforms A and B ) and DNA sequences have been obtained experimentally by using the MITOMI method and reported in [52] . Absolute affinity measures were calculated by varying four nucleotides in the first half of the core binding-motif with the preferred GTG second half-site kept . The changes in the energy between the optimal sequence and mutated ones were also computed in [52] by subtracting the energy associated with the mutated sequences to the energy of the optimal sequence . Both absolute affinity measures and changes in the binding energy were compared to predicted values obtained with the different models . The TFFM-framework is available at http://cisreg . cmmt . ubc . ca/TFFM/doc with a web-based application available at http://cisreg . cmmt . ubc . ca/TFFM/ . The software have been implemented in Python using the Biopython tools [67] and the General Hidden Markov Model library [68] . The web-based application has been developed using the Python cgi_app ( https://pypi . python . org/pypi/cgi_app/1 . 3 ) and the Python port of the Template Toolkit ( http://template-toolkit . org/python/index . html ) . | Transcription factors are critical proteins for sequence-specific control of transcriptional regulation . Finding where these proteins bind to DNA is of key importance for global efforts to decipher the complex mechanisms of gene regulation . Greater understanding of the regulation of transcription promises to improve human genetic analysis by specifying critical gene components that have eluded investigators . Classically , computational prediction of transcription factor binding sites ( TFBS ) is based on models giving weights to each nucleotide at each position . We introduce a novel statistical model for the prediction of TFBS tolerant of a broader range of TFBS configurations than can be conveniently accommodated by existing methods . The new models are designed to address the confounding properties of nucleotide composition , inter-positional sequence dependence and variable lengths ( e . g . variable spacing between half-sites ) observed in the more comprehensive experimental data now emerging . The new models generate scores consistent with DNA-protein affinities measured experimentally and can be represented graphically , retaining desirable attributes of past methods . It demonstrates the capacity of the new approach to accurately assess DNA-protein interactions . With the rich experimental data generated from chromatin immunoprecipitation experiments , a greater diversity of TFBS properties has emerged that can now be accommodated within a single predictive approach . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | The Next Generation of Transcription Factor Binding Site Prediction |
Dengue fever is reemerging on the island of Martinique and is a serious threat for the human population . During dengue epidemics , adult Aedes aegypti control with pyrethroid space sprays is implemented in order to rapidly reduce transmission . Unfortunately , vector control programs are facing operational challenges with the emergence of pyrethroid resistant Ae . aegypti populations . To assess the impact of pyrethroid resistance on the efficacy of treatments , applications of deltamethrin and natural pyrethrins were performed with vehicle-mounted thermal foggers in 9 localities of Martinique , where Ae . aegypti populations are strongly resistant to pyrethroids . Efficacy was assessed by monitoring mortality rates of naturally resistant and laboratory susceptible mosquitoes placed in sentinel cages . Before , during and after spraying , larval and adult densities were estimated . Results showed high mortality rates of susceptible sentinel mosquitoes treated with deltamethrin while resistant mosquitoes exhibited very low mortality . There was no reduction of either larval or adult Ae . aegypti population densities after treatments . This is the first documented evidence that pyrethroid resistance impedes dengue vector control using pyrethroid-based treatments . These results emphasize the need for alternative tools and strategies for dengue control programs .
Control of the vector Aedes aegypti remains the primary approach to reducing transmission of dengue and dengue hemorrhagic fever in human populations [1] , [2] . Adult control is generally implemented by spraying chemicals into the air to reduce adult populations after larval control has failed to limit their density or virus transmission has reached outbreak levels [3] . However there has been considerable debate as to the efficacy of space spray applications for the control of dengue epidemics in the tropics [4] . Several studies found space spraying effectiveness as variable and often poor due to limited penetration of insecticides into dwellings [5] , [6] . In contrast , Gratz [7] reported successful control of adult Ae . aegypti populations in south-east Asia where spatial applications were applied in a well organized manner ( e . g . using adapted materials and formulations , effective application procedures and planning ) . Furthermore dengue control programs rely on just two of the four major classes of insecticides available for use in public health; pyrethroids and organophosphates . The spread of resistance to these two chemical classes is a major concern for dengue control in the tropics [8] . In Martinique ( French Caribbean ) 5 major dengue outbreaks have occurred in the last 15 years [9] , and each only involved the vector Ae . aegypti . DDT and several organophosphates ( e . g . malathion , fenitrothion ) were used to control adults for decades , but in the early 1990s there was a switch to pyrethroids ( i . e . deltamethrin and permethrin ) because of their strong insecticidal properties at low application rates and for their safety in the environment and towards humans [10] , [11] . Since then local populations of Ae . aegypti have developed high levels of resistance to pyrethroids [12] . This is due to high frequencies ( >80% ) of the “knock-down resistance” ( kdr ) mutation ( V1016I ) , and elevated activity of detoxification enzymes ( i . e . cytochrome P450 mono-oxygenases , glutathione-S-transferases and , carboxylesterases ) [12] . Detox chip microarray and RT-qPCR validation showed an over-transcription of multiple detoxification genes , essentially P450s , in pyrethroid-resistant adults compared to their susceptible counterparts [12] , [13] . This multiple resistance had a strong negative effect on the efficacy of pyrethroid and pyrethrins space sprays against natural Ae . aegypti mosquitoes in semi-field conditions ( based on data from mosquito sentinel cages ) [14] . This finding is all the more worrying as these two adulticides ( deltamethrin and natural pyrethrins ) remain the only insecticides available for the control of adult mosquitoes in Martinique due to European legislation regarding pesticide use and its application [15] . We conducted a medium-scale field trial in Martinique to assess the operational impact of pyrethroid-resistance on the efficacy of deltamethrin ( 1 g/Ha ) and pyrethrins ( 10 g/Ha ) ULV space sprays following WHO protocols [16] , [17] . Our results demonstrate that the efficacy of pyrethroid and pyrethrins space sprays was strongly reduced when applied against natural populations of Ae . aegypti resistant to pyrethroids . This emphasizes the urgent need to develop alternative tools and strategies for the control of dengue vectors .
The two districts of Lamentin and Ducos located in the western part of Martinique were selected for the field trial . Both have a dry tropical climate with a rainy season occurring between May and November and an annual precipitation of 2 , 500 mm . Five localities were selected in the districts of Ducos ( 14°34′0″N , 60°58′60″W ) : Bac , Durivage , Bois Neuf , Canal and Morne Carette . Four localities were chosen in Lamentin ( 14°36′0″N , 60°0′0″W ) : Grand Case , La Favorite , Place d'Armes and Long Pré . The localities were separated from each other by 1 to 3 km , thus minimizing interference between or among localities . The 9 localities were housing estates and hamlets , more or less urbanized , composed of ∼60–150 houses ( Figure 1 ) . The island of Martinique is an overseas department of France and subject to European law , including laws related to the use , application and handling of insecticides . One month before the field evaluation , larvae from the 9 localities were sampled and reared in the laboratory . Adult females of the F1 progeny were used in tarsal contact tests with treated filter paper and compared with the susceptible Bora strain . Tests were run using filter papers treated with technical grade deltamethrin ( 100% [w/w]; AgrEVO , Herts , United Kingdom ) and pyrethrum ( mixture of 6 pyrethrins [pyrethrin I , pyrethrin II , cinerin I , cinerin II , jasmolin I , jasmolin II]; 25 . 44% [w/w]; Pyrethrum Board of Kenya , Nakuru , Kenya ) following WHO guidelines [18] . The resistance status of Ae . aegypti in each locality was determined by using discriminating dosages of deltamethrin ( 0 . 05% ) and pyrethrum ( 1% ) [19] . For each strain , five batches of 20 non-blood fed females ( 2–5 days old; n = 100 ) were exposed to the insecticides for 60 minutes and mortality recorded 24 hours later . The efficacy of synergized pyrethrins and deltamethrin was evaluated in field conditions by conducting space spray applications using three 4x4 vehicles mounted with a Curtis Dinafog MaxiPro4 thermal fogger . In each site , 3 rounds of treatment were made at 2 day intervals according to the treatment procedure used by the mosquito control unit of Martinique . Formulations of deltamethrin ( K-Othrine ULV 15/5 , 15% [w/v] +0 . 5% esbiothrine [w/v]; Bayer Environmental Science , Lyon , France ) and synergized pyrethrins ( AquaPy , 3% EW [w/v] +13 . 5% piperonyl butoxide ( PBO ) [w/v] ) were used . These two insecticides are routinely used in Martinique and their use is authorized by European legislation [15] . The formulation of pyrethrins was mixed with water as a carrier for the thermal fogging , as recommended by the manufacturer , and deltamethrin was mixed with mineral oil ( Banole W ) , which was also used alone as a control treatment . K-Othrine ULV 15/5 is the reference formulation that has been used for many years in Martinique for the control of Ae . aegypti . Each insecticide and their formulations were notified in the European Directive 98/8EC of 16 February 1998 [15] . The choice of localities was based on Ae . aegypti density ( larval indices and electric traps ) and resistance status . Three groups of 3 localities having similar entomological indices were formed and one locality of each group was allocated to an insecticide or control treatment; these groups acted as a blocking factor during data analysis . Following this procedure , the localities of Canal , Morne Carette and Grand Case were treated with deltamethrin; Bac , Bois Neuf and La Favorite were treated with pyrethrins; and Durivage , Long Pré and Place d'Armes were control sites and sprayed with Banole W . A single vehicle was allocated to each insecticide throughout the study . Treatments were carried out in each locality at the period corresponding to the peak of mosquito flight activity ( between 5:00 PM and 8:00 PM ) . Before treatment , the spraying apparatus was calibrated ( i . e . flow rates were 450 mL water/min and 380 mL oil/min ) and blank trials were made in each locality in order to estimate the length of routes and the quantity of formulation needed ( Figure 1 ) . During application , the speed of vehicles was between 8 and 10 km/hr , and the volume of mixture applied between 760 and 980 mL/Ha . With an efficient pulverization swath of 40 meters , the 9 areas treated were between 5 . 6 and 10 . 6 Ha . During application a hand-held anemometer ( TFA ) recorded temperature , wind speed and its direction . Temperatures and relative humidity were recorded constantly over the period of experiment at each locality using a Hobo Pro v2 probe . Precipitation was recorded daily with a meteorological unit ( Auria 4 ) placed in Lamentin and provided by the General Council of Martinique . Operators and people involved in the evaluation were informed as how to handle insecticides and the corresponding safety procedures . Operators used protective clothing , shoes and facemasks to reduce the risk of exposure to insecticides . The sprays performed in our study were part of routine spraying campaigns in Martinique organized by French governmental authorities . The spraying procedures used in this study ( i . e . trucks , foggers , asking people to open their doors and windows , etc . ) were the same as those used by French governmental authorities in dengue control programs in Martinique . The experiments were validated by the health departments , the mayors of the two municipalities ( Ducos and Lamentin ) , and by the French Agency for Environmental Health and Safety ( AFSSET ) . Before the experiments , inhabitants of each locality received a brochure with information explaining the treatments and when they would be applied . People living in individual residences where cages and traps were installed were asked for their permission to do so and the details of each location were recorded by the vector control staff of Martinique . All the inhabitants gave their oral consent prior to treatments; their names , addresses and phone numbers were recorded by the vector control staff and are held at the offices of the Conseil Général de la Martinique , Fort-de-France , France . Insecticidal activity of pyrethroid space sprays was evaluated against a susceptible reference strain ( Bora ) on 3 separate occasions according to the WHO cage bioassay method [16] . Cylindrical steel frame cages ( 90 mm diameter x 153 mm height ) covered on all three walls with a mosquito net ( 1 mm mesh ) were used to house groups of 20 adult female mosquitoes . In each locality , 5 houses located on the path of the vehicle were chosen to place the cages ( Figure 1 ) . Twenty minutes before ULV sprays , 5 cages were placed outside in the gardens and 5 were placed inside the houses ( 10 cages per locality ) . For each day of treatment and for each insecticide , 30 cages containing 20 Bora females were used for measuring insecticide penetration rate ( n = 600 females per insecticide and per replicate ) . Inhabitants were asked to open their doors and windows during treatments to enable maximum penetration of the aerosol into houses . After application , cages were brought back to the laboratory for assessment of post-treatment mortality 24 hours later . During the observation period , mosquitoes were fed with sugar-soaked cotton and maintained in the laboratory at 27±2°C with a relative humidity of 80±10% . On each occasion that spraying took place , data on the mortality of caged females of the Bora strain was complemented by data involving resistant females sampled from the population at the locality “Place d'Armes” . In each case a total of 10 cages harboring 20 females were used , with one cage on the inside and outside of the same 5 houses used to assess mortality of the Bora strain . On the 1st spray these additional cages were in the locality “Place d'Armes” and exposed to the control treatment ( Banole ) ; for the 2nd spray they were in “La Favorite” and exposed to pyrethrins ( AquaPy ) , and for the 3rd spray they were exposed to deltamethrin ( K-Othrine ) in the locality “Canal” . The density of adult Ae . aegypti population , before and after spraying , was measured by using 72 BG-Sentinel-Traps ( BGS-Trap , BioGents GmbH , Regensburg , Germany ) in the 9 selected localities . Traps were equipped with a mosquito attractant which was given off by the BG-Lure , a dispenser which releases a combination of lactic acid , ammonia and caproic acid , substances found in human skin [20] . Eight traps were allocated to each locality and were positioned in the same 4 houses where cage-mortality was evaluated . Four traps were placed inside houses and distributed in rooms occupied by the inhabitants , whereas the four other traps were placed outside houses on terraces where they were protected from sun and rain . Traps were deployed at 11:00 AM and collected 24 hours later . In each locality , abundance was estimated weekly starting 2 weeks before and ending 3 weeks after treatments . Abundance estimates were also made between the 1st and 2nd treatments . The density of Ae . aegypti larvae in each locality was estimated from 7 entomological surveys . A weighted Breteau Index ( WBI ) was used to determine the “productivity” of female adult mosquitoes [17] . The Breteau Index is defined as the number of positive containers per 100 inspected houses . To estimate the productivity , a coefficient is attributed to each type of positive container [17] , [21] to take into account the number of larvae produced in different habitats ( e . g . drums [coefficient of 5 . 5] vs . flower pots [coefficient of 1 . 5] ) . Estimates of WBI were determined weekly , starting 2 weeks before treatments and ending 3 weeks after . Monitoring was also performed between each treatment . Mortality recorded in laboratory bioassays was corrected for control mortality by Abbott's formula [22] in case of control mortality >5% . Data were analyzed using the Analysis of Overdispersed Data ( AOD ) package of R which is suitable for the analysis of data based on proportions ( e . g . percentage mortality ) [23] . Bioassay , trap and productivity data were each analyzed using a split-plot analysis of variance ( ANOVA ) . The bioassay and trap data showed heteroscedasticity among individual houses and so weighted means of data collected from cages either on the outside or inside of houses were calculated for each locality and day of sampling . The bioassay data was analyzed as a split-plot ANOVA with whole plots consisting of the 3 treatments blocked into 3 groups according to the entomological indices of each locality , the 1st sub-plot was defined by the 3 spray treatments , with the 2nd sub-plot depending on whether data came from cages on the inside or outside of houses . The trap and productivity data were analyzed as repeated measures split-plot ANOVA with the 6 days of sampling acting as the repeated measure . Whole plots were defined as above , with localization ( inside vs . outside ) forming the sub-plot for the bioassay data; these data were log ( x+1 ) transformed prior to analysis , where x was the mean weighted value from houses within a locality . ANOVA were performed by the software JMP ( v7 . 1 ) using the restricted maximum likelihood method ( REML ) [24] . The additional cages holding resistant females in the bioassay of cage mortality were not included in analysis above as there was no replication for each treatment on a spatial ( locality ) or temporal ( spray ) scale . Estimates for the mortality of these females are presented separately .
Results obtained from WHO cylinder tests are shown in the Figure 2 . The laboratory Bora strain was susceptible to pyrethrins ( 1% ) and deltamethrin ( 0 . 5% ) with 99% mortality . Mosquitoes from the 9 localities selected for the field trial showed high levels of resistance to pyrethrins ( from 1 to 10% mortality ) and deltamethrin ( from 6 to 47% mortality ) . ULV sprays were performed on May 13 , 18 and 20 , 2009 . During treatments , temperatures ranged from 25 to 31°C , relative humidity ranged from 60 to 90% , and wind speed ranged from 0 to 4 meters/second . Mortality of caged susceptible females of the Bora strain 24 hours after treatment strongly depended on the insecticide used and on whether the cages were on the outside or inside of houses ( Table 1 , Figure 3 ) . Mortality in control cages was low ( <1% ) , showing Banole W was not toxic when used alone . The mortality of females exposed to pyrethrins was 10% and 17% for those placed inside and outside the houses , respectively , and significantly lower than mortality recorded for the deltamethrin treatment . The mortality of females exposed to deltamethrin on the outside of houses was significantly higher than for those placed inside houses . These results demonstrate the efficacy of deltamethrin ULV space sprays against susceptible mosquitoes and that K-Othrine penetrates to the inside of houses . Mortality of caged resistant females did not exceed 10% for either insecticide treatment ( inside or outside ) . Monitoring of Ae . aegypti populations levels in each district were made from May through June 2009 . Temperatures recorded varied from 22 to 35°C and relative humidity ranged from 55 to 98% . Daily precipitation varied from 0 to 105 mm during the evaluation with an exceptional precipitation of 300 mm on May 5th , 2009 one week before the 1st treatment . Estimates from the ANOVA for the numbers of females trapped on the inside or outside of houses are shown in Figures 4 and 5 , respectively , grouped by day of sampling and treatment . The only significant effect found was for day of sampling ( Table 2 ) . Whereas the average number of females trapped was fairly constant over the different sampling days , it was low on the 2nd day of sampling . This day was 24 h after the application of the first spray treatments . However , it was also one week after the study area had experienced heavy rainfall . This event may have reduced recruitment into the adult population a week later due to larvae being washed out of breeding sites . Support for this argument comes from the observation that the number of females trapped on the 2nd day was also lower for locations treated with Banole ( Figures 4 and 5 ) . The application of insecticides had no detectable effect on mosquito productivity as the larval surveys found no significant differences among treatments or days of sampling ( Table 3 , Figure 6 ) .
This study provides the first evidence that resistance in Aedes aegypti can seriously reduce the efficacy of pyrethroid space sprays and their ability to control populations of the Dengue vector . Cage bioassay experiments showed that space sprays with deltamethrin at the recommended dose ( i . e . 1 g/Ha ) failed to eliminate natural populations of pyrethroid resistant Ae . aegypti ( <10% mortality ) , whereas a relatively high proportion of susceptible mosquitoes ( from 47 to 63% ) were killed in the same experimental conditions . This lack of effect was confirmed by the trap data showing that 3 rounds of applying deltamethrin had no effect on the number of females mosquitoes caught in different locations ( Figures 4–6 ) . In addition , no reduction in larval indices was noted in localities treated with insecticides compared to those of the control group . Previous studies have found the penetration of insecticides to the inside of houses following aerial spraying can be limited [5] , [6] . The mortality of susceptible females held in cages within houses shows the insecticide penetrated houses but these results have to be taken with caution as in natural conditions part of the mosquito population will rest in confined locations ( e . g . under beds ) where the insecticide may not always reach . These results suggest deltamethrin space sprays might remain effective against adult mosquitoes coming into contact with this insecticide in areas where pyrethroid-susceptible mosquitoes are present , especially if treatments are implemented during the activity period of the mosquito . However this treatment is unlikely to reduce the overall density of adult mosquitoes and virus transmission in areas where pyrethroid resistance is already present . Synergized pyrethrins are considered as potential alternatives to synthetic pyrethroids for the control of resistant mosquitoes [19] . However the impact of this treatment against both susceptible and resistant mosquitoes was worse than for deltamethrin: it only killed relatively few caged females and caused no reduction in adult densities or larval indices after 3 rounds of application . The absence of penetration with this formulated product may be explained by an inappropriate target dose , the use of a sub-optimal spraying method ( i . e . thermal fogging ) and/or equipment , or a combination of these factors [7] , [25] . For example , several studies have shown the efficacy of pyrethrins in reducing the incidence of West Nile virus in the USA when aerially applied ( ULV ) against Culex tarsalis and Cx . pipiens [26] , [27] . This indicates the necessity of testing new formulations in real conditions and appropriately evaluating their performance , before regular control operations can be implemented for reducing mosquito abundance and infection rates [25] . Our laboratory bioassay showed that Ae . aegypti populations collected in Martinique were strongly resistant to pyrethrins and deltamethrin , confirming previous results from this area [14] . The present study showed insecticide resistance reduced the efficacy of space sprays which are the only tools available for controlling adult mosquitoes during an outbreak of dengue . Control of Ae . aegypti adults remains crucial to reduce the basic reproduction number ( R0 ) to below a threshold were the disease cannot invade or persist in a human population [28] , [29] . Early detection , with appropriate entomological indices , of localities at high risk for dengue transmission is essential for public health operators to implement early effective control especially during the extrinsic incubation of the virus [30] , [31] , [32] . Our study emphasizes the need to reconsider the use of pyrethroids for controlling adults in the strategy currently being implemented in Martinique . More generally , further studies should address the relationship between insecticide resistance and the efficacy of insecticide-based interventions to increase our knowledge in this poorly documented domain [8] . In addition , resistance monitoring must routinely be undertaken in all dengue control programs to help authorities and public health sector workers select the best insecticide for control of Ae . aegypti . Given the fact that increasing number of countries have reported alarming increases in the level of pyrethroid resistance [8] , the replacement of pyrethroids by alternative molecules should be urgently addressed . The use of new chemicals with different modes of action than pyrethroids remains a feasible option . However the lack of insecticides that can be used in public health programs , due to legitimate concerns for the environment and human health , limits the options available [15] . In this context , there is an urgent need to develop strong partnerships between the public health sector and private companies to encourage investment in the research and development of new molecules , or formulations , for mosquito control [33] . However , this process is long and expensive [34] . In the short term , research on adult control should focus on the use of existing molecules . For example , the association of pyrethroid in mixtures with other insecticides with different modes of action may act synergistically and restore the efficacy of pyrethroids against resistant mosquito populations . The use of adulticides combined with larvicides ( e . g . insect growth regulators with long-lasting residual activity ) for spatial treatments may be a promising option during epidemics , as it can reduce simultaneously larval and adult mosquito densities [35] , [36] , [37] , thus also making this approach cost effective [38] . | The mosquito Aedes aegypti is the major vector of the Dengue virus in human populations and is responsible of serious outbreaks worldwide . In most countries , vector control is implemented by the use of insecticides to reduce mosquito populations . During epidemics , insecticides of the pyrethroid family ( blocking the voltage gated sodium channel protein in the nerve sheath ) are used by space spraying with vehicle mounted thermal foggers to kill adult mosquitoes . Unfortunately some populations of Ae . aegypti have become resistant to these insecticides , leading to operational challenges for public health services . In Martinique ( French West Indies ) , resistance to pyrethroids was detected in the 1990s . The present study assessed the impact of this resistance on the efficacy of vector control operations in 9 localities of Martinique . Here we showed that the resistance strongly reduces the efficacy of pyrethroid-based treatments , thus emphasizing the urgent need for alternative insecticides or tools to reduce dengue transmission . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"infectious",
"diseases",
"mosquitoes",
"pest",
"control",
"dengue",
"fever",
"neglected",
"tropical",
"diseases",
"biology",
"arboviral",
"infections",
"vectors",
"and",
"hosts",
"infectious",
"disease",
"control",
"agriculture"
]
| 2011 | Pyrethroid Resistance Reduces the Efficacy of Space Sprays for Dengue Control on the Island of Martinique (Caribbean) |
Epigenetic modifications such as DNA methylation have large effects on gene expression and genome maintenance . Helicobacter pylori , a human gastric pathogen , has a large number of DNA methyltransferase genes , with different strains having unique repertoires . Previous genome comparisons suggested that these methyltransferases often change DNA sequence specificity through domain movement—the movement between and within genes of coding sequences of target recognition domains . Using single-molecule real-time sequencing technology , which detects N6-methyladenines and N4-methylcytosines with single-base resolution , we studied methylated DNA sites throughout the H . pylori genome for several closely related strains . Overall , the methylome was highly variable among closely related strains . Hypermethylated regions were found , for example , in rpoB gene for RNA polymerase . We identified DNA sequence motifs for methylation and then assigned each of them to a specific homology group of the target recognition domains in the specificity-determining genes for Type I and other restriction-modification systems . These results supported proposed mechanisms for sequence-specificity changes in DNA methyltransferases . Knocking out one of the Type I specificity genes led to transcriptome changes , which suggested its role in gene expression . These results are consistent with the concept of evolution driven by DNA methylation , in which changes in the methylome lead to changes in the transcriptome and potentially to changes in phenotype , providing targets for natural or artificial selection .
Epigenetic modifications affect gene regulation and genome maintenance [1] , [2] . DNA methylation is an important epigenetic modification in bacteria with functions in gene expression regulation , genome replication initiation , cell cycle regulation , anti-mutagenesis , and genome maintenance [3] , [4] . Eukaryotes use a few DNA methyltransferases to mainly methylate DNA at CpG and other low-specificity sequences . In bacteria , DNA is methylated by a variety of DNA methyltransferases , most of which have high sequence specificity . Methylations in both promoter and coding regions affect gene expression [2] , [5] , [6] . Methyltransferases are often members of restriction-modification ( RM ) systems [7] and are called modification ( M ) enzymes . Three types of methyltransferases , corresponding to RM systems Type I through III , are known [8] . A Type II methyltransferase such as M . EcoRI methylates a base within a recognition sequence that is often palindromic [7] . Several classes of Type II methyltransferases recognize nonpalindromic sequences [9] . In living cells , DNA methylation protects the genome from cleavage by a cognate restriction enzyme such as R . EcoRI , which recognizes the same sequence as its cognate methyltransferase [10] . Type III methyltransferases recognize nonpalindromic sequences and methylate only one of the two strands of the recognition sequence [11] . In a Type I RM system , an M gene product forms a complex with its specificity ( S ) gene product to define the recognition sequence for methyltransferase activity ( Figure 1 ) [12] . RM systems are often mobile and vary among bacterial species and strains [13] , [14] . Helicobacter pylori , a gastric bacterium that is pathogenic in humans , has one of the highest numbers of identified M genes [15] . Its genome is highly diverse among strains [16] , [17] . Each strain has a unique set of M genes , which suggests variable genomic methylation states or methylomes [15] . In addition to the divergent repertoire of M genes , H . pylori genome comparisons suggest that Type I and Type III target recognition domain ( TRD ) sequences are themselves mobile , leading to variation in the DNA sequences recognized by these RM systems . [18]–[20] . TRDs of Type III systems even move between nonorthologous genes by recombination at weakly similar DNA sequences encoding conserved amino acid motifs of DNA methyltransferases; this mechanism spreads TRDs beyond species boundaries [19] . The Type I S protein carries two TRDs , TRD1 and TRD2 ( Figure 1 ) , with each domain recognizing half of a bipartite recognition sequence [12] . Not all but some of the Type I S genes show tandem repeat sequences flanked by the two TRDs [18] , and their copy number correlates with the length of the central nonspecific region ( Ns ) in the recognition sequence [21] . TRD sequences of Type I S genes can be shuffled at each domain site ( TRD1 and TRD2 ) , leading to diversity in methylation sequence specificity [22]–[27] . Similar amino acid sequences that recognize similar DNA sequences were found in TRD1 of one Type I RM system and in TRD2 of another Type I RM system from a different bacterial species [27] . For H . pylori and several other bacteria , a genome comparison revealed that TRD sequences likely move between TRD1 and TRD2 by recombination at their flanking repeat sequences , in a process called Domain Movement ( DoMo ) [18] . TRD amino acid sequences in Type I S genes and Type III M genes in H . pylori fall into distinct homology groups . Amino acid sequences are nearly identical within each group and are expected to correspond to a unique set of recognition sequences . Therefore , the movements of TRD sequences will lead to changes in their recognition sequences . TRD sequence movements along with allelic recombination events , point mutations , and changes in the copy number of tandem repeats between TRD1 and TRD2 are expected to be sources of methylome diversity in H . pylori [14] . We hypothesized that such methylome changes might lead to changes in the transcriptome and cell phenotypes and contribute to adaptive evolution [14] , [20] . Recognition sequences are known only for some H . pylori RM systems , mostly for Type II methyltransferases [28] , [29] . For Type II and Type III systems , recognition sequences can be determined by cleaving a DNA molecule of a known sequence and identifying the sequence common to every cut site . However , this method cannot be applied to Type I systems because the corresponding restriction enzyme complex cleaves DNA at unpredictable positions outside the recognition sequence [12] . Their recognition sequences have been determined by transfer of labeled methyl groups by the methyltransferase [30] , [31] and transformation by plasmids with or without their candidates [32] . The recent development of single-molecule real-time ( SMRT ) sequencing technology has facilitated detection of methylated DNA bases [33] . This technology uses a single DNA polymerase to incorporate one of four fluorescent analogs for dATP , dTTP , dGTP , and dCTP onto a DNA template , and monitors the incorporation to decode the template sequence . This technology allows detection of several base modifications in the template DNA because the modifications delay incorporation of the dNTP analog . This method has been established as reliable for accurately detecting methylation motifs in plasmids and bacterial genomes [34]–[39] , but no study has been reported that uses this method on more than three closely related bacterial strains . In this study , we decoded the methylome of closely related H . pylori strains and compared their methylomes . We verified a correspondence between genes , TRD sequences and recognition sequences for Type I , II and III systems and found that DoMo in Type I S genes indeed changed recognition sequences and the methylome . Furthermore , transcriptome analysis revealed that methylation by a Type I S protein affected gene expression .
We analyzed five H . pylori strains ( P12 , F16 , F30 , F32 and F57 ) and two isogenic P12 derivatives ( HPYF1 and HPYF2 , see below ) . The complete genome sequences of the first five strains were obtained by the Sanger method [17] , [40] . P12 was isolated in Germany [40] and belongs to hpEurope cluster in the current population assignment based on STRUCTURE analysis of 7 housekeeping genes [17] . F16 , F30 , F32 and F57 , isolated from the same hospital in Japan [17] , [40] , fall into the hspEAsia of the hpEastAsia cluster [17] , [41] . Their genome sequences are closely related [17] , [41] , although their synteny has changed through multiple inversion events [42] . A PacBio RS ( Pacific Biosciences ) was used for SMRT sequencing for methylome analysis . Two biological replicates were analyzed for each strain ( DRA accession no . DRA001084 ) . Results were reproducible for output read numbers and for detection of methylated motifs ( Table S1 ) . Around 45 , 000 to 80 , 000 bases per genome were detected as methylated ( N6-methyladenine and N4-methylcytosine , Figure S1 ) . The genome of all strains is around 1 . 6 Mbp [17] , [40] , so 1 . 4 to 2 . 6% of bases were methylated ( with consideration of both strands ) . Thus , these genomes represent the most heavily methylated genomes analyzed so far by SMRT technology [35]–[37] . Motif search between 20 bp upstream and downstream of each methylated base identified 15 to 24 unique methylation motifs ( with variation among strains ) , with methylation of more than 20% of the copies of each methylation motif in a genome in both biological replicates ( Figure 2 , Table 1 , Table 2 , Table S2 ) . In addition to simple 4 bp , 5 bp and 6 bp palindromic methylation motifs , we found nonpalindromic methylation motifs . We also found bipartite methylation motifs that include a long tract of Ns ( N = A , T , G or C ) , which are typical of recognition sequences of Type I RM systems and some subclasses of Type II systems . Many of the methylation motifs were successfully assigned to M/S genes using previous knowledge about recognition sequences [10] , presence or absence of apparently intact and untruncated M/S gene orthologs in genomes , and combinations of TRD amino acid sequences in Type I specificity genes [18] . To identify hypermethylated or hypomethylated genomic regions , the number of detected methylated bases per 1 kb was calculated for each strand . The five most and least methylated regions for each strain were determined ( Table 3 , Table 4 , Figure 3 , Figure S2 ) . A region within the rpoB gene , encoding the RNA polymerase beta subunit , was identified as a densely methylated region in three of the five strains ( P12 , F16 , F30 ) . A region within the groEL gene , encoding a chaperonin , was also identified as densely methylated and was in the top 5 densely methylated regions in four of the five strains ( P12 , F30 , F32 , F57 ) . Both regions were especially heavily methylated at 5′-CATG ( methylated base is underlined , Figure S3 ) . The M gene with this recognition sequence ( M . hpyAI ) is highly conserved among the five strains ( Table 1 , Table S2 ) [43] and likely responsible for the conservation of the hypermethylated state . Other genes that were hotspots were central to translation ( fusA , 16S ribosomal RNA ) , cell shape determination ( mreB ) , or related to host interaction/virulence ( flgE , ureC , cagY ) ( Table 3 ) . Relationships were not clear between the hypermethylation in cagY gene and its unusual DNA structure with frequent rearrangements leading to gain/loss of function in the Cag type IV secretion system [44] . Hypermethylation in a DNA methyltransferase gene ( HPP12_0447 ) suggested some interaction , such as one in gene expression regulation , between multiple DNA methyltransferases . The bisC gene , hypermethylated in an H . pylori strain isolated in Japan , appears to be truncated in all hspEAsia strains [17] . The biological signficance of these cases of hypermethylation is not yet clear . In three strains , P12 , F32 and F57 , several regions with the lowest methylation ( Table 4 ) were in conjugative transposons , or TnPZs [40] , [45] . Fewer methylated sites imply fewer methylation motifs and , therefore , fewer targets for the cognate restriction enzymes . The paucity of methylated motifs and bases might have resulted from selection by restriction attack during horizontal transfer of these elements [46]–[48] . This observation is in contrast to the results from eukaryote genomes , where mobile elements tend to be silenced by methylation [49] . The presence of a hypomethylated region in a Type II modification enzyme gene ( HPF32_0484 , M . hpyAXII homolog ) suggests another type of interaction between multiple DNA methylation systems . We do not know whether its own hypomethylation is related to the mobility found in this family [50] and serves as a means to avoid restriction . Its target sequence ( 5′-GTAC ) is abundant in 16S rRNA gene [51] and contributes to its hypermethylation . Other hypomethylated genes encoded outer membrane proteins ( homC , HPF30_0293 ) , a virulence factor ( mviN ) , and a cell wall synthesis enzyme ( murC ) ( Table 4 ) . For each of the five studied strains , we compared the list of present or absent M/S genes as annotated in the REBASE database [10] with the list of detected methylation motifs for assignment . The two lists matched well with only few exceptions . For most of the Type II M genes with a known recognition sequence for N6-methyladenine or N4-methylcytosine , the studied genomes had a high fraction of the copies of the recognition sequence methylated ( Table 1 ) . For many of the recognition sequences , around 90% of the copies were methylated . Orthologs of HPP12_0488 ( 5′-ATTAAT ) , HPP12_1052 ( 5′-TCNNGA ) , and HPP12_1173 ( 5′-CATG ) ( Table 1 ) were highly conserved in all five strains , suggesting a conserved function in this species . As mentioned above , HPP12_1173 orthologs responible for the conserved methylation hotspot in rpoB methylation hotspot are highly conserved in H . pylori [43] . No Type II-like methylation motifs were newly assigned to M genes with a hitherto unknown motif . H . pylori has three groups of Type I RM systems with a total of at most five Type I S genes at different genetic loci . We defined these as Group 1 through 3 [18] . Type I S genes encode two TRDs , TRD1 and TRD2 , in tandem , each binding to half of a bipartite recognition sequence ( Figure 1A ) . For example , in Figure 1A , TRD1 corresponds to the left half 5′-TAG-3′/3′-ATC-5′ whereas TRD2 corresponds to the right half 5′-TAC-3′/3′-ATG-5′ . A TRD recognizes half of a target sequence in an inverted configuration when it is at different TRD sites [27] . TRDs in H . pylori are classified into homology groups and members of the same homology group have identical or nearly identical amino acid sequences [18] . In the five strains analyzed , TRD homology groups TRD a , TRD b , TRD c , TRD d , TRD e , TRD f , and TRD h were identified for Group 2 ( Figure 2A ) ; TRD homology groups TRD k , TRD m , TRD n , and TRD o were identified for Group 3 ( Figure 2B ) . The combination of TRD homology groups varies among strains [18] . For example , strain P12 carries the combination TRD c-TRD a ( Figure 2A ) , while strain F16 carries the combination TRD d-TRD h and the combination TRD a-TRD h ( Figure 2A ) . We identified 18 or fewer hemimethylated nonpalindromic methylation motifs that were not Type I-like but similar to the methylation motifs of Type III and subclasses of Type II RM systems [9] , [55] ( Table 6 ) . H . pylori strains carry up to five loci for Type III M genes , each of which contains a TRD [19] , [20] . One Type III TRD , in HPF16_0033 and HPF30_0034 , was assigned as recognizing 5′-GGCAA . This is because this methylation motif was detected only in F16 and F30 among the five strains and because this TRD represents the only one TRD with such distribution . Many Type III M genes and Type IIG genes were truncated by mutations [13] so that we could not assign them hemimetylated nonpalindromic motifs . Only few of the TRD sequences identified in the apparently intact Type III M gene ORFs were shared by more than two strains used in the present work [13] . Four 7 bp methylation motifs were found in F30 ( Table 6 ) . These methylation motifs , 5′-CAAGWAG , 5′-CRTGHAG , 5′-CTNGNAG and 5′-CCDGNAG , were included in a single degenerated methylation motif , 5′-CNNGNAG . However , the other methylation motifs represented as 5′-CNNGNAG were rarely methylated . A single M gene might recognize these four sequences or four different M genes might recognize each sequence . As the first step to examine the biological significance of the variety in methylation specificity , we examined the effect of a Type I S gene , HPP12_0797 , on the transcriptome ( DRA accession no . DRA001073 ) . We replaced this gene of strain P12 by a kanamycin-resistance gene . Its derivative with insertion of this gene downstream of the S gene was constructed for control . Loss of methylation at the recognition sequence deduced above , 5′-GAAN8TAG , was confirmed by comparison of their decoded methylomes ( Table S2 ) . Transcriptomes were analyzed for differentially expressed genes ( Table 7 ) . The results were confirmed by quantitative real-time PCR . Transcripts of the S gene itself were absent in the knockout strain as expected . The gene cluster HPP12_0959 through HPP12_0962 ( Figure 6A ) showed increased transcript accumulation in the knockout strain . This cluster appears to be an operon whose transcription start site is upstream of an HPP12_0963 ortholog in another strain [56] . These results showed that methylation by Type I M and S gene products can significantly repress expression of genes . Three copies of the recognition sequence for the S gene product were found in the body of those genes with significant expression changes ( Figure 6B ) . The first of them was within a 22 bp palindromic sequence . We do not know whether this is a binding site for a symmetrical protein and we do not yet know the relationship , if any , between the long palindromic sequence and the transcriptional changes .
Assignment of methylation motifs to specificity-determining genes , that is Type I S , Type II M , and Type III M genes , is usually difficult for species with many of these genes . Most methylome analyses using SMRT sequencing used bacterial species with a small number of specificity-determining genes [36] , [37] . Recognition sequence assignment in species and strains with a larger number of specificity-determining genes required cloning of each gene into well-defined Escherichia coli laboratory strains that lack DNA methyltransferase genes and methylation-specific nuclease genes [35] . The activities of specificity-determining genes can be different under intrinsic and cloned conditions and their expression level is critical for methylome analysis [34] , [35] . H . pylori has a large number of specificity-determining genes . Because H . pylori strains also show large diversity in genome sequence [57] , changes in methylation level and sequence specificity through subtle changes in amino acid sequence of specificity-determining genes was expected . To analyze the H . pylori methylome , strain differences in the repertoire of methylation motifs and in specificity-determining genes were used to assign methylation motifs to specificity-determining genes . This method also revealed a high diversity in target methylation level and sequence specificity between orthologous genes . For Type I systems , methylation sequence assignment was carried out for two TRDs within each S gene product . Our analysis assigned a methylation motif to each Type I S gene and a half-recognition sequence to each TRD sequence ( homology group ) . Of three groups of Type I S genes , two ( Group 2 and Group 3 ) showed well-conserved methylation activity from apparently intact ORFs . These assignments combined together with the rules about the central repeats now allow prediction of target sequence of many S genes from their sequence . For the other group of Type I S genes ( Group 1 ) and Type IIG S genes , however , sufficient information on methylation motif was not obtained . One reason for the difficulty was the very low expression level of the Group 1 S gene , as revealed by transcriptome analysis of the P12-derived strains HPYF1 and HPYF2 ( data not shown ) . Another reason was the absence of DoMo ( Figure 1B ) in Group 1 . Some unassigned Type I-like methylation motifs were identified as candidates for recognition sequences of Type I S gene products , but these methylation motifs did not match the domain combinations for other groups of Type I S genes and Type IIG S genes . This result suggested that only a few combinations of TRD homology groups had methylation activity or that other unannotated S genes were present . We also cannot exclude the possibility of inactivation by simple mutation . Our detection of examples of strain-to-strain diversity in methylation activity by a specific methyltransferase suggested new mechanisms for inactivating a methylation system in addition to truncation by an insertion or deletion [20] . An untruncated gene that could not be assigned to a methylation motif in one strain could have an activity in another strain . In addition to detecting complete inactivation of DNA methyltransferase genes with untruncated ORFs , we also observed intermediate methylation activity for some methylation motifs , such as methylation detection in 50–60% copies of methylation motif . This did not fit the hypothesis that the methylation level switches digitally between two states: 0% and 100% . Variation occurred even among members of the same ortholog group . A simple explanation for this variation is strain-specific mutations at residues important for activity . Indeed , many strain-specific amino acid changes were found within genes associated with the variable methylation levels . Another mechanism for the variation is competition for the recognition sequence among M or S genes with overlapping recognition sequences . We indeed detected many examples of recognition sequence overlap that had a substantial effect on the methylome . Methylation motifs that we did not detect by SMRT sequencing , such as those for 5-methylcytosine , are candidates for such competition because H . pylori has many strain-specific 5-methylcytosine methyltransferases [28] . A related issue to the variation in methylation level is the subtle strain-to-strain variation in recognition sequence . Type I S gene TRDs in the same homology group might recognize different sequences in different strains: for example , 5′-CCA is recognized by one member of one TRD homology group in one strain and 5′-CTA is recognized by another member in another strain . Amino acid changes likely responsible for the changes were identified . Methylome comparisons of more strains might reveal more examples of such microevolution in sequence recognition , which would help our understanding of DNA sequence recognition by Type I S proteins . Another type of microevolution we observed extended a recognition sequence from 5′-CCGG to 5′-CCGGH ( H = A , T or C ) for a Type II system . We do not yet know whether amino acid changes in the corresponding DNA methyltransferase or another factor was responsible for this change . We also noticed presence of several unassigned motifs within a strain that were similar to each other ( Table 6CD ) . An example is a group of four sequences related to 5′-CNNGNAG . These cases might be explained by intermediate stages in the switching of sequence specificity . Recent work on Type IIL restriction enzymes indicates their target specificity can be easily changed [9] , [55] . The results of this study revealed diversity in the methylome within H . pylori and demonstrated a built-in mechanism , DoMo , for generating this diversity . For Group 2 Type I S genes , we identified 5 active TRD homology groups . These groups generated diversity in S genes and their recognition sequences by allelic homologous recombination at TRDs and by DoMo . The copy number of central tandem repeats flanked by TRDs might change , resulting in variation in the number of Ns in a recognition sequence ( Figure 1 ) . For example , 10 TRD homology groups with 7 or 8 Ns at the center of methylation motif could generate 10×2×10×1/2 = 102 structural variants of a single S gene . If one homology group correspsonds to one methylation motif , these structural variants correspond to sequence variants in the methylation motifs . Combined methylation sequence specificities could be even larger . Four such S loci would result in 102×102×102×102 = 108 combined structural variants and a corresponding number of combined sequence specificities . Even greater diversity is possible when other types of sequence-specific DNA methyltransferases are considered . What could be the biological significance of this enormous diversity ? We earlier proposed , in an epigenetics-driven adaptive evolution model , that diverse methylomes serve as units of natural selection , with each unique gene experssion pattern and a unique set of phenotypes [14] . We observed that a Type I specificity gene affected the transcriptome . In eukaryotes , gene regulation frequently occurs through changes in protein binding affinity , for example of transcription factors , that is caused by methylation near promoters [1] , [58] . Recent work shows that methylations within gene body might also regulate gene expression [6] . In prokaryotes , gene expression changes resulting from methylation distribution have been well studied [4] , [20] , [59] . Methylation of 5′-GATC , which was found here partly responsible for hypermethylation of an RNA polymerase gene , is important for gene expression regulation . Further work is necessary to elucidate the biological significance of the methylome data from this study . We earlier proposed the hypothesis that specificity changes in methyltransferases might lead to changes in phenotype [14] . These changes might not only detract but also might have potential to enable adaptive evolution . Under this hypothesis , the roles of changes in methylation specificity could be similar to the roles of genome rearrangements in adaptive evolution , for example , the antigenic variation that results from gene conversion to adapt to host immunity [60] . We need to learn more about methylation specificity changes and their effects , if any , on phenotypes and genotypes to evaluate the functions of these changes . In this work , we provided evidence that a bacterial species has an enormous diversity in methylome status through various mechanisms including point mutations and DoMo ( movement of target recognition domain sequences between sites within a gene ) . Deletion of a methylation specificity-determining gene affected the transcriptome . These findings are consistent with our hypothesis that methylome changes might lead to changes in cell physiology through transcriptome changes , and might contribute to adaptive evolution . Epigenetic changes in DNA methylation might be a potential source of variation for adaptive evolution , similar to DNA sequence changes . During the reviewing process of this manuscript , a paper decoding methylome of two other H . pylori strains appeared [39] . We have not noticed any inconsistency between these two works . In particular , their assignments of Type I S genes to the full target sequences and our assignments of the TRDs within S genes to the half target sequences are consistent .
H . pylori strain P12 [40] was kindly provided by Rainer Haas ( Ludwig-Maximilians-University of Munich , Germany ) . H . pylori strains F16 , F30 , F32 and F57 were previously described [17] . According to multilocus sequence typing based on seven housekeeping genes , P12 belongs to the hpEurope group and F16 , F30 , F32 and F57 belong to the hspEAsia group [17] . Strains were inoculated from 50% glycerol stocks onto tripticase soy agar ( TSA ) -II/5% sheep blood plates ( Becton Dickinson , NJ ) and incubated under microaerobic conditions ( O2 , 5%; CO2 , 15%; N2 , 80% ) at 37°C for 3 days . Colonies were collected by resuspending in 1 ml Brucella ( Becton Dickinson , NJ ) broth , and transferred to 99 ml of Brucella broth with 10% fetal calf serum . After growth for one day under microaerobic conditions , cells were centrifuged at 8000×g for 5 min and the supernatant discarded . Genomic DNA was extracted from the pellet by a protease/phenol method as described elsewhere [17] and resuspended in 300 μl of TE buffer ( 10 mM Tris HCl , pH 7 . 8 ) . A region covering the HPP12_0797 ORF and 1 kb flanking sequences on both sides was amplified by PCR with KOD FX Neo ( TOYOBO , Japan ) using primers P12_group2S_EcoRI_for ( GGGGAATTCGGAATTACAAGGGTTTCAGCATTCAGCC ) and P12_group2S_BamHI_rev ( GGGGGATCCGCTTACCCAAGCTAAAAGCATCGC ) . Amplified fragments were cleaved with EcoRI and BamHI , followed by ligation with Ligation high Ver . 2 ( TOYOBO , Japan ) to pBR322 cleaved with the same enzymes . Ligation products were transformed into E . coli DH10B competent cells by electroporation , resulting in plasmid pYF166 . For replacement of HPP12_0797 on pYF166 with a kanamycin-resistance gene , pYF166 other than HPP12_0797 ORF region was amplified using primers P12_group2S_sub_ClaI_for ( GGGATCGATGCCCTTCTTCTAAATGGCTAATG ) and P12_group2S_sub_KpnI_rev ( GGGGGTACCCAAAATACCCCCCTATCCCC ) . For preparation of a control strain with a kanamycin-resistance gene at the downstream of HPP12_0797 , almost whole the pYF166 was amplified using primers P12_group2S_sub_ClaI_confrol_for ( GGGATCGATCCCCCTTAACCCCCAACTAG ) and P12_group2S_sub_KpnI_rev . The kanamycin-resistance gene was amplified by PCR from pHel3 [61] , kindly provided by Rainer Haas ( Ludwig-Maximilians-University of Munich , Germany ) , using primers P3_ClaI ( GGGATCGATAAAATTGGAACCGGTACGCTTA ) and P4_KpnI ( GGGGGTACCAGACATCTAAATCTAGGTAC ) [62] . Amplified fragments with the kanamycin-resistance gene were cleaved with ClaI and KpnI and ligated with Ligation high Ver . 2 ( TOYOBO , Japan ) to fragments from pYF166 ( described above ) cleaved with the same enzymes . Ligation products were transformed into DH10B by electroporation to obtain pYF171 ( knockout ) and pYF173 ( control ) . For transformation into H . pylori P12 , inserts in pYF171 and pYF173 were amplified by PCR with primers P12_group2S_EcoRI_for and P12_group2S_BamHI_rev . A P12 culture was prepared as described above and 1 μg of amplified DNA fragment was added . After one day growth under microaerobic conditions at 37°C , 200 μl culture was plated on TSA-II/5% sheep blood plates with 8 mg/L kanamycin and incubated at 37°C for 3 days under microaerobic conditions . Single-colony isolation was carried out under the same conditions , resulting in the HPP12_0797-knockout strain , HPYF1 , and the control strain , HPYF2 . Cultures were prepared as described above and stored at -80°C as 50% glycerol stocks . Genomic DNA samples were sheared to ∼500 bp using a S2 Focused-ultrasonicator ( Covaris , MA ) . SMRT bell libraries for SMRT sequencing were prepared with DNA Template Prep Kit 2 . 0 ( Pacific Biosciences , CA ) ( 250 bp <3 kb ) . SMRT sequencing was performed using a DNA Sequencing Kit 2 . 0 with C2 polymerase ( Pacific Biosciences , CA ) , following standard instructions for a PacBio RS ( Pacific Biosciences , CA ) . Two biological replicates were performed for each strain . The read depth was approximately ×100 ( Table S2 ) . SMRT sequencing data were analyzed by the RS_Modification_and_Motif_Analysis . 1 protocol in SMRT Analysis version 1 . 4 . 0 through the SMRT Portal . In brief , reads were mapped to the genome sequences ( Accession numbers: P12 , NC_011498; F16 , AP011940; F30 chromosome , AP011941; F30 plasmid , AP011942; F32 chromosome , AP011943; F32 plasmid , AP011944; F57 , AP011945 ) . Interpulse durations were measured for all nucleotide positions in the genomes and compared with expected durations in a kinetic model of the polymerase [63] for significant associations . To analyze methylation distribution , the number of methylated bases in a 1 kb window was counted with sliding by 500 bp for each strand . All 20 bp sequences upstream and downstream of a methylated nucleotide that were not in a methylation motif detected by the above protocol were collected and searched for methylation motifs by MEME-ChIP [64] . Score thresholds were chosen to fulfill the condition that methylation positions without detectable methylation motifs after MEME-ChIP analysis were less than 5% of the number of detected methylated positions . A methylation motif was assumed to be methylated if more than 20% of copies of the methylation motif in the genome were detected as methylated in both biological replicates . HPYF1 and HPYF2 were grown to OD600 of 0 . 3–0 . 4 and cell pellets were prepared as described above . Whole RNA was extracted by PureLink RNA Mini Kit ( Life Technologies , MD ) . RNA samples were prepared with mRNA-Seq Sample Prep Kit ( Illumina , CA ) for construction of libraries for strand-specific RNA-seq . Libraries were sequenced by HiSeq 2000 ( Illumina , CA ) . Resulting sequences were mapped to protein coding sequences in the P12 genome ( Accession number: NC_011498 ) . Mapped read counts for each coding sequence were compared by the DESeq package to detect significant differences in expressed genes using a threshold of P<0 . 001 [65] . Quantitative real-time PCR used KAPA SYBR FAST One-Step qRT-PCR Kit ABI Prism ( KAPA Biosystems , MA ) . Each sample was analyzed as triplicate technical replicates . HPP12_0008 , encoding GroEL , was used as an internal control [56] . An Applied Biosystems 7300 Real-Time PCR System ( Life Technologies , MD ) was used for detection and analysis . Primers are in Table S4 . | Living organisms are affected by epigenetic variation in addition to DNA sequence variation . DNA methylation is one of the most studied epigenetic modifications in both prokaryotes and eukaryotes . In prokaryotes , most DNA methylation is by DNA methyltransferases with high sequence specificity . Helicobacter pylori , a human stomach pathogen responsible for stomach cancer and other diseases , carries a large number of DNA methyltransferase genes that vary among strains . In this work , we examined the distribution of DNA methylation in multiple H . pylori genomes using single-molecule real-time sequencing technology , which detects DNA methylation with single-base resolution . Comparison of methylation motifs between closely related genomes allowed assignment of a recognition sequence to each DNA methylation specificity-determining gene . Highly methylated genes were detected , although the general DNA methylation pattern varied among strains . Knockout of a methylation specificity-determining gene led to changes in the transcriptome . These findings are consistent with our hypothesis that changes in the methylome lead to changes in the transcriptome and to changes in phenotypes , providing potential targets for natural and artificial selection in adaptive evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"genomics",
"dna",
"modification",
"genome",
"evolution",
"gene",
"expression",
"genetics",
"biology",
"and",
"life",
"sciences",
"dna",
"comparative",
"genomics",
"computational",
"biology",
"epigenetics"
]
| 2014 | Methylome Diversification through Changes in DNA Methyltransferase Sequence Specificity |
Most tissues in metazoans undergo continuous turnover due to cell death or epithelial shedding . Since cellular replication is associated with an inherent risk of mutagenesis , tissues are maintained by a small group of stem cells ( SCs ) that replicate slowly to maintain their own population and that give rise to differentiated cells . There is increasing evidence that many tumors are also maintained by a small population of cancer stem cells that may arise by mutations from normal SCs . SC replication can be either symmetric or asymmetric . The former can lead to expansion of the SC pool . We describe a simple model to evaluate the impact of ( a ) symmetric SC replication on the expansion of mutant SCs and to show that mutations that increase the probability of asymmetric replication can lead to rapid mutant SC expansion in the absence of a selective fitness advantage . Mutations in several genes can lead to this process and may be at the root of the carcinogenic process .
Most tissues of the body experience continuous cell turnover due to either apoptosis or loss of epithelial cells due to shedding as occurs in the gastrointestinal and respiratory tract as well as the ductal epithelium of the breast . This cell turnover also gives tissues the ability to repair after injury . Cell replication is associated with an inherent risk of mutation due to the imperfect replication machinery [1] . Cancer is due to the progressive accumulation of mutations [2] , and to minimize the risk of malignant transformation , higher vertebrates have evolved tissue architecture so as to limit the retention of cells that acquire mutations [3–5] . In general , tissues are maintained by a small group of slowly replicating cells that have the dual capacity of self-renewal and differentiation into the more mature cells required by a given tissue . Cells that exhibit these two capabilities are known as stem cells ( SCs ) . The SC concept was initially proposed for cells responsible for the maintenance of hematopoiesis [6–8] . Hematopoietic SCs are located within the bone marrow , and their self-renewal capacity has classically been demonstrated by an ability to reconstitute hematopoiesis after serial transplantation in sublethally irradiated mice [9 , 10] . Recent studies have shown that SCs are present in most tissues of the body including the colonic mucosa , the breast , and the central nervous system [11] . Moreover , it is increasingly recognized that many tumors have a population of cancer stem cells ( CSCs ) that maintains the growth of the neoplasm [7 , 11–16] . Eradication of the CSCs is necessary for therapeutic cure of the neoplasm [17] . A critical aspect of SC replication is the diverse fate of the two daughter cells that result from each division . Conceptually , such divisions are considered to be symmetric if the two daughter cells are identical to each other ( whether they both retain SC properties or become committed cells ) , or asymmetric if one daughter cell retains the SC properties of the parent while the other cell commits to a more differentiated stage ( Figure 1 ) [18] . In principle , asymmetric replication within the SC pool can maintain cellular homeostasis since one daughter cell is retained as an SC while the other enters a committed pathway . However , SCs have to be able to replicate symmetrically to generate two daughter SCs , since this is the only way the SC pool may expand during ontogeny or after injury such as SC transplantation [18 , 19] . An SC may divide symmetrically into two committed daughter cells , and therefore an SC is “lost” from the pool , whereby a symmetric division in which the daughter cells remain as SCs again becomes necessary . Studies on the methylation patterns of DNA in daughter cells provide experimental evidence for all three types of SC division in colonic crypts [20] , and recent experiments in animal models with serial hematopoietic SC transplantation also support the existence of all three types of division [19] . The mechanisms regulating the ( a ) symmetry of SC divisions are not completely understood . However , it appears that both interactions with the SC niche as well as intracellular mechanisms relating to the mitotic apparatus and differential partitioning of intracellular organelles contribute to the different modes of replication of these cells [18] . Studies in model organisms such as Drosophila melanogaster and Caenorhabditis elegans are beginning to provide a mechanistic understanding of the various determinants of symmetric versus asymmetric SC division [18] . The dynamic implications of symmetric versus asymmetric SC replication have been modeled only indirectly [21 , 22] , even though this is an important area of investigation since mutations in tissue-specific SCs may be the source of many neoplasms [11 , 23 , 24] . Here we develop a simple mathematical model of SC dynamics that only considers the probability of symmetric versus asymmetric replication within the SC compartment to illustrate the potential consequences of disrupted SC division .
Consider a microenvironment composed of a limited number of SCs; an example is the colonic crypt that houses intestinal SCs [25] . SCs divide to maintain themselves or differentiate into the colonic epithelial cells that migrate up the crypt where they undergo apoptosis and are either shed into the lumen or engulfed by stromal cells [25–28] . The number of SCs remains approximately constant ( approximately one to ten SCs per crypt ) . Therefore , SC dynamics in the crypt can be modeled by a Moran process [29] , which is a stochastic process assuming that the total population size remains strictly constant over time ( Figure 2 ) . Denote the total number of SCs in a given crypt by n . Consider the fate of a single mutant SC: it can either go extinct or reach fixation in the population . The evolutionary dynamics of the SC population as a function of symmetric and asymmetric replication is modeled as follows . Denote by i . the number of mutant SCs; therefore , the number of normal SCs is given by n − i . The relative reproductive fitness of normal SCs is 1 while that of the mutant SC is given by r . If r < 1 , mutant cells have a lower fitness as compared with normal SCs , while if r = 1 the mutant SCs have the same fitness as normal SCs . If r > 1 , the mutant cells have a higher reproductive fitness compared with the normal SC . During each replication cycle , SCs are randomly chosen for reproduction proportional to their fitness . If an SC divides symmetrically to produce two daughter SC , then one SC from the whole pool is selected at random to be eliminated so that the total number of SC , n , remains strictly constant . If the SC selected for reproduction gives rise to two differentiated daughter cells , one SC is effectively lost from the compartment . To maintain SC homeostasis in that population , another SC is selected ( again according to fitness ) for reproduction and gives rise to two daughter SCs so that the total number of SCs is restored . Finally , the selected SC may divide asymmetrically to give rise to one differentiated cell and another SC , whereby the number of SCs in the compartment remains constant . Therefore , with each replication event there are three potential scenarios for both mutated and normal SCs ( Figure 2 ) . The number of mutants can increase by one , stay the same , or decrease by one . We denote by pa the probability that the mutant SC divides asymmetrically , while qa is the probability of a symmetric replication that produces two differentiated cells . Finally , the probability that the mutant SC divides symmetrically to give two daughter SCs is given by 1 − pa − qa . The same nomenclature with subscript b denotes the respective division probabilities for the normal SC ( Figure 2 ) . Let us consider the transition probabilities of the stochastic process . For the mutant SC population to increase by one , a mutant SC must be selected for reproduction , and a normal SC selected for death . Alternatively , a normal SC may divide symmetrically to produce two differentiated daughter cells , and a mutant SC is selected for reproduction where the cell has to reproduce itself . The transition probability therefore is given by Similarly , two separate paths decrease the mutant SC number by 1 in a unit time step . A normal SC may reproduce itself , and a mutant SC is selected for death . Otherwise , a mutant SC may reproduce to give rise to two differentiated cells , and a normal SC is selected for self-renewal . The transition probability for this process is given by Finally , the probability that the number of mutant SCs remains constant is given by To calculate the probability that i mutant SCs ultimately take over the population , we observe that this fixation probability Φi follows the equation Iterating this equation leads to the solution See , e . g . , Antal and Scheuring [30] for details of the derivation . Here Pk , k + 1 and Pk , k−1 are as defined in Equations 1 and 2 , respectively . The fixation probability of a single mutant in the SC population is given by It can be shown that if pa = qa = pb = qb = 0 , the process is identical to the classical Moran process [29] with and . The conditional average time τi until a population of i mutant CSs reaches fixation follows the equation Starting with zero mutant cells , fixation is never reached , and hence τ0 = 0 . If we start from n mutant cells , fixation has already occurred , and therefore τn = 0 . With these boundary conditions , we can solve Equation 7 . The average conditional time required for a single mutated cell to take over the SC pool is given by where
We performed exact stochastic computer simulations for different parameter regimes . We varied both the fitness r of the mutant SC population as well as the probabilities for symmetric versus asymmetric reproduction to show the time development of the process . Initially , we explored the model for n = 1 , 000 . We chose this value for n since it is unlikely that any SC compartment is composed of a larger number of active SCs [31] . In the absence of a mutated SC , the number of normal SCs remains constant ( Figure 3A ) . We set the baseline transition probabilities for the normal SC as pb = qb = 1/3 , so that the probability for each type of mitotic division is equal . A mutant SC may appear by alteration of a gene that regulates the symmetry of cellular replication . Examples of such genes include adenomatous polyposis coli ( APC ) [32 , 33] or HUGL [34]; both are frequently mutated in colon cancer . A mutation that changes the probability of symmetric versus asymmetric division may or may not alter the reproductive fitness of the cell . In the following , we considered these possibilities separately or in combination . We assumed that one mutant cell appears and followed the evolutionary dynamics of that mutant and its progeny; we neglected continuous production of mutants from wild-type SCs . This assumption holds if the time to fixation of such a mutant is smaller than the waiting time for production of another mutant . The process has two absorbing states: either all SCs are normal or all SCs are mutated . We performed 1 , 000 simulations for each set of parameter values and plotted the fraction of simulations in which the mutant reaches fixation . A mutant without a selective advantage may take over the population by neutral drift; however , the fraction of compartments in which the mutant will dominate is small . If the mutant SC has a higher probability of self-renewal compared with the normal SC ( i . e . , 1 − pa − qa > 1 − pb − qb ) , a higher fraction of compartments is dominated by the mutant clone even in the absence of a fitness advantage ( Figure 3B ) . As expected , mutations that increase the probability of self-renewal and improve the fitness of the SCs result in a higher frequency of fixation and in a shorter average fixation time ( Figure 3C ) . The average time required for fixation of mutant cells decreases as the probability of self-renewal increases . For a given set of transition probabilities , the average time for fixation decreases as the population size gets smaller ( Figure 4A ) . The relative fitness of the mutant SCs as compared with normal cells has a determining role in the evolutionary dynamics of the population . Mutant SCs with a reduced fitness ( r < 1 ) will rarely take over the population , if the cell division properties of the mutant are the same as the wild-type SCs; as the fitness of the mutant increases , the fraction of SC compartments that will be taken over by the mutated cells increases ( Figure 4B ) . Let us now determine how the average fixation time depends on the probability of SC self-renewal , the size of the pool , and the fitness associated with such a mutation . A mutation that increases SC fitness requires a shorter average time for fixation in the population ( Figure 5A ) . As the SC population size increases , the average fixation time increases nonlinearly ( Figure 5B ) . Although mutant SCs with a reduced fitness can reach fixation in a short time , this is a highly improbable event ( see Figure 4 ) . The average fixation time for the neutral mutant is always the longest and consistent with random drift . In a similar fashion , we compute the average time to fixation , as the probability of mutant SC renewal varies from zero to two-thirds for a fixed fitness . Figure 5C shows that , as the probability of self-renewal increases ( qa decreases for a fixed value of pa = 1/3 ) , the average time for fixation of the mutant decreases . This is also true as qa increases , and therefore the probability of self-renewal decreases ( Figure 5C ) ; however , this is again an improbable event ( Figure 3 ) .
The field of cancer research is undergoing a transformation with the recognition that at the root of many tumors is a relatively small group of CSCs that maintains the bulk of the tumor population . The presence of these cells was initially demonstrated for hematopoietic neoplasms [35] and has been used to explain drug resistance and regrowth of the tumor when therapy is stopped [36] . However , there is increasing evidence that CSCs exist for solid tumors as well , including breast carcinoma [13] , brain tumors [15] , and colon carcinoma [12 , 16] . Studies of cells isolated from primary tumors or metastases have shown that only a fraction of the tumor population has a long-term colony forming ability or can induce tumor xenograft growth when implanted into immuno-compromised mice [12 , 16] . At present , it is not clear whether CSCs result from the accumulation of mutations in normal SCs or whether more committed cells acquire SC-like properties as a result of mutations . There is data to support both possibilities at least for some neoplasms , and these hypotheses are not mutually exclusive [37–39] . In this study , we explored a simple model of SC replication to illustrate the impact of changes in the probability for symmetric versus asymmetric replication on SC dynamics within a constant population . One might assume that the natural tendency of a SC is to self-renew and generate two daughter SCs . However , there is a limit on the total number of SCs in a given environment , perhaps due to a finite number of SC niches that can be occupied . Growth factor stimulation and the microenvironment within the SC niche together impose differentiation on one or both daughter cells that result from an SC division . However , the tendency of SCs to self-renew will ensure that if the SC pool is depleted ( e . g . , with high-dose chemotherapy ) , the remaining SCs can expand to occupy the available niches [40] . In evolutionary biology , reproductive fitness is defined as reproductive success within a given environment that exerts a selection pressure . In the context of cancer evolution , cells with a higher reproductive fitness either replicate faster or produce more progeny in the same time interval compared with other cells . This is often modeled as the mutant cells being selected for reproduction with a higher probability compared with the normal cells . Our model shows that mutations that increase the probability of SC self-renewal ( larger 1 − pa − qa ) give the mutant cell an advantage even though the mutant may not be selected for reproduction more often than a normal SC . This can be deduced from Equations 1 and 2 since whenever 1 − pa − qa > 1 − pb − qb , the probability of fixation of the mutant increases . If we consider a mutation that does not confer a selective advantage to the cell ( r = 1 ) , the ratio of the transition probabilities reduces to: Thus , as the probability to generate two daughter cells , qa , decreases , while all the other parameters remain fixed , the ratio of the transition probabilities decreases and makes fixation of the mutant more likely . Moreover , mutations such as those in APC that increase both the probability of self-renewal and the probability that the mutant is selected for reproduction ( i . e . , increase r ) give an even higher advantage to the cell , and the fixation time decreases while the probability of fixation increases . The risk of cellular transformation is in part dependent on the population of cells at risk as well as on the mutation rate . While the number of normal SCs appears to be tightly regulated , mutations in critical genes may reduce the responsiveness of SCs to environmental controls , and their population can expand . Alternatively , mutations in genes such as Partner of Inscuteable ( PINS ) [41] , lethal giant larvae ( LGL ) [42] , and HUGL-1 that regulate the symmetry of SC replication may lead to expansion of this cell pool . Mutations in these genes are associated with the development of tumor-like tissue in model organisms [42 , 43] . In colonic crypts , APC normally controls SC reproduction by inducing the degradation of β-catenin [44] . Therefore , mutations in APC might lead to an expansion of the SC pool at the base of the colonic crypt . In addition to inhibiting β-catenin , APC also regulates the ( a ) symmetry of stem cell division in Drosophila [45] . It is not known whether APC mutations within colonic crypt SCs increase the probability of symmetric division with SC expansion . However , patients with familial adenomatous polyposis develop aberrant crypt foci that are associated with an increased risk of colonic adenomas as well as colon cancer [46] . With the identification of CD133 as a marker for colon CSCs [12 , 16] , it would be interesting to determine whether aberrant crypt foci are enriched for crypt SCs and what additional mutations are present in these cells apart from APC . Further , there is evidence that prostaglandins may be necessary for SC growth and asymmetric replication [47] . Perhaps this may in part explain the reduced risk of colon cancer associated with the use of nonsteroidal anti-inflammatory drugs [48] and other COX-2 inhibitors . In summary , SCs may in principle replicate via three pathways to expand their own population or produce differentiated daughter cells . Mutations in genes that increase the probability of self-renewal provide an additional facet for the selective advantage of tumor SCs . The identification of additional genes that regulate the ( a ) symmetry of SC replication will expand our understanding of the growth of the tumor SC clone and possibly identify novel targets for cancer therapy . | In multicellular organisms , tissues such as skin , the gut , and blood undergo continuous cell turnover . These tissues are maintained by a small group of tightly regulated cells known as stem cells ( SCs ) that have two defining properties: they can renew themselves and give rise to more specialized cells that perform tissue specific tasks . Somatic SCs live for many years and replicate slowly to minimize the risk of acquiring mutations in their DNA . When a SC divides , the two daughter cells may have similar properties ( symmetric division ) or may have different fates ( asymmetric division ) . Symmetric division may allow SCs to expand , and mutations that alter the probability of symmetric versus asymmetric division might increase the risk of tumor growth . This property is important since there is increasing evidence that even tumors have their own SCs . Mutations can transform wild-type SCs into tumor SCs with modified cell division properties , which have decisive impact on cancer progression . Here we develop a mathematical model to illustrate the impact of mutations that regulate the symmetry of SC division on the development of tumors . Our results provide novel insights on the pathway to cancer by mutations within SCs . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"homo",
"sapiens",
"oncology",
"developmental",
"biology",
"mammals"
]
| 2007 | (A)Symmetric Stem Cell Replication and Cancer |
MicroRNAs ( miRNAs ) are thought to exert their functions by modulating the expression of hundreds of target genes and each to a small degree , but it remains unclear how small changes in hundreds of target genes are translated into the specific function of a miRNA . Here , we conducted an integrated analysis of transcriptome and translatome of primary B cells from mutant mice expressing miR-17~92 at three different levels to address this issue . We found that target genes exhibit differential sensitivity to miRNA suppression and that only a small fraction of target genes are actually suppressed by a given concentration of miRNA under physiological conditions . Transgenic expression and deletion of the same miRNA gene regulate largely distinct sets of target genes . miR-17~92 controls target gene expression mainly through translational repression and 5’UTR plays an important role in regulating target gene sensitivity to miRNA suppression . These findings provide molecular insights into a model in which miRNAs exert their specific functions through a small number of key target genes .
MicroRNAs ( miRNAs ) are endogenously encoded single stranded RNAs of about 22 nucleotides ( nts ) in length . They suppress target gene expression by translational repression and promoting mRNA degradation . The relative contribution of these two modes of action to miRNA regulation of its target gene expression is a matter of ongoing debate [1–3] . It was initially thought that animal miRNAs repress the protein output of target genes without significantly effecting mRNA levels [4 , 5] . Subsequent genetic studies in C . elegans and zebrafish showed that miRNAs also promote the degradation of their target mRNAs [6 , 7] . To reveal the global effect of miRNA on target gene mRNA and protein levels , a series of genome-wide studies applied microarray , RNA-seq , proteomics , and ribosome profiling to mammalian cell lines transiently transfected with miRNA mimics or inhibitors or primary cells from miRNA mutant mice . Two early studies showed significant correlations between the mRNA and protein levels of miRNA target genes , as well as widespread target mRNA degradation [8 , 9] . This was followed up by a study concluding that mammalian miRNAs predominantly act to decrease target mRNA levels [10] . However , other studies that employed the same experimental approach , namely transient transfection of miRNA mimics or inhibitors into in vitro cultured mammalian cell lines , came to an opposite conclusion . These studies showed that miRNAs affect the expression of most target genes through translational inhibition [11 , 12] . Subsequent studies employing temporal dissection of miRNA action seemed to have resolved this discrepancy by showing that translational repression precedes target mRNA deadenylation and decay [13–18] . This order of events can be interpreted either as evidence that mRNA decay is a consequence of translational repression [17 , 19] , or as reflection of the kinetic differences between these two mechanisms that operate independently from each other [20] . In line with the latter interpretation , analyses performed either in cultured cells or in vitro extracts showed that miRNA-mediated translational repression can occur in the absence of target mRNA deadenylation and decay [19 , 21–27] . Therefore , it remains an unanswered question whether mRNA degradation is always the end result of miRNA targeting and whether miRNA-mediated translational repression and target mRNA degradation are molecularly coupled under physiological conditions [1 , 28 , 29] . In contrast to the efforts to search for a unified mechanism of miRNA action , studies of individual miRNA-target mRNA interactions in miRNA mutant mice are painting a rather different picture . A recent survey of literature focused on studies in which target gene mRNA and protein levels were measured concurrently in primary cells and tissues from mutant mice with genetic ablation or transgenic expression of individual miRNA genes [2] . This survey analyzed a total of 159 miRNA-target mRNA interactions in 77 strains of miRNA mutant mice . Among them , 48% target genes are predominantly regulated by translational repression , 29% are regulated mainly by mRNA degradation , and 23% are regulated by both . This heterogeneity in miRNA mechanisms of action has been increasingly recognized as more and more miRNA mutant mice are generated and analyzed , but what determines the dominant mode of miRNA action remains unclear . An interesting finding of this survey is that most target genes identified in developing cells or tissues are regulated by mRNA degradation , whereas target genes identified in terminally differentiated cells tend to be regulated at the translational level . It is conceivable that mRNA degradation gets rid of target mRNA in a non-reversible manner and provides an efficient way for cell fate determination , while translational repression is immediate , transient and reversible , which is more suitable for differentiated cells to respond to environmental stimuli [2] . Indeed , previous studies have shown that miRNA regulation of target gene translation can occur in a rapid and reversible manner under various stress conditions [30 , 31] . These studies highlight the importance of cellular context in determining the dominant mode of miRNA action . The mode of action can also be miRNA-dependent . Transcriptome analysis of mouse liver showed that miR-122 and let-7 cause significant target mRNA degradation , whereas miR-21 has little impact on its target gene mRNA levels [32] . Another study of primary cells from miRNA mutant mice showed that miR-155 in B cells and miR-223 in neutrophils cause significant target mRNA degradation , while miR-150 in B cells and miR-21 in neutrophils have absolutely no effect on their target mRNA abundance [14] . Considering the cellular context- and miRNA-dependency , it is essential to investigate miRNA mechanisms of action in the cellular contexts where miRNA of interest performs its physiological or pathological functions . Another controversial issue in miRNA research is about how miRNAs achieve their specific functions . On one hand , bioinformatic analysis and experimental target gene identification using the recently established PAR-CLIP and HITS-CLIP methods often find hundreds of target genes for a miRNA [33–36] . Proteomic analysis of mammalian cell lines transiently transfected with miRNA mimics showed that a miRNA regulates the protein output of hundreds of target genes , and that the effect on each target gene is often moderate [8 , 9] . These studies led to the conclusion that miRNAs exert their functions by modulating the expression of hundreds of target genes and each to a small degree [37 , 38] . However , when the hundreds of target genes regulated by a miRNA are closely examined , they often fall into a broad spectrum of functional categories [36 , 39 , 40] . How small changes in hundreds of target genes with diverse functions are translated into specific phenotypic outcomes has been a conceptual conundrum . On the other hand , recent genetic studies demonstrated that mutation of miRNA binding sites in a single target gene can phenocopy miRNA deficiency in a cell context-dependent manner in both mice and worms [41–43] . These results provide strong support to the key target gene model , which postulates that the function of a miRNA is often mediated by a small number of key target genes in a given cellular context [44] . We speculated that the discrepancy between these two types of studies regarding how miRNAs exert their specific functions stems from the transient transfection approach , which may not recapitulate the actions of endogenous miRNAs under physiological conditions [2] . Recent studies showed that transient transfection of miRNA mimics into in vitro cultured cell lines led to increase of mature miRNAs to supraphysiological levels , appearance of high molecular weight RNA species , frequent mutation of guide strands of miRNA mimics , accumulation of unnatural passenger strands of miRNA mimics , and non-specific alterations in gene expression [45–47] . These findings call into question the physiological relevance of previous studies employing the transient transfection approach to investigate the functions and mechanisms of miRNAs . As increasing numbers of animals harboring gain- and loss-of function mutations for individual miRNA genes are being generated [2 , 48] , primary cells from these miRNA mutant animals are better systems for studying miRNA mechanisms of action under physiological conditions . In this study , we investigated miRNA mechanism of action in lymphocytes by conducting an integrated analysis of the transcriptomes and translatomes of primary B cells from miR-17~92 transgenic and knockout mice . The miR-17~92 family consists of three miRNA clusters: miR-17~92 , miR-106a~363 , and miR-106b~25 ( S1 Fig ) . Together , these three clusters contain 15 miRNA stem-loops that give rise to 13 distinct mature miRNAs . They fall into four miRNA subfamilies ( miR-17 , miR-18 , miR-19 , and miR-92 subfamilies ) , with members in each subfamily sharing the same seed sequence . Germline knockout of miR-17~92 family in mice is incompatible with life [49] . These miRNAs are essential for the development of lung , heart , central nervous system , fetal liver , and B lymphocytes [49] . B cell-specific deletion of the miR-17~92 family ( CD19-Cre;miR-17~92fl/fl;miR-106a~363-/-;miR-106b~25-/- , termed TKO mice ) severely impaired antibody responses , while B cell-specific miR-17~92 transgenic ( TG ) mice develop lymphomas with high penetrance [40] . This conditional transgene and knock-out strategy bypasses developmental defects caused by dysregulated miR-17~92 expression during the early stages of B cell development [50 , 51] . We have now performed a comprehensive molecular analysis of primary B cells expressing miR-17~92 miRNAs at three different levels ( TKO , WT and TG ) . In this cellular context , we found that target genes exhibit differential sensitivity to miRNA suppression , and that only a small fraction of target genes are actually suppressed by a given concentration of miRNA . Absolute quantification of miRNA and miRNA binding site revealed there are more miRNA binding sites than miRNA molecules so that only a small fraction of binding sites are occupied by miRNA molecules at a given time . Moreover , miR-17~92 controls key target gene expression mainly through translational repression and 5’UTR plays an important role in regulating target gene sensitivity to miRNA suppression . These findings provide mechanistic insights into the functional specificity of miRNAs .
We have previously identified 868 target genes harboring a total of 1139 miR-17~92 binding sites conserved in human and mouse ( termed miR-17~92 targets ) by PAR-CLIP analysis of B cells [40] . This list contains most of miR-17~92 target genes validated in previous studies . We investigated the effect of transgenic miR-17~92 expression and complete deletion of the miR-17~92 family on the mRNA levels of these target genes during B cell activation . We first generated a complete list of significantly expressed mRNAs and their absolute molecule numbers by RNA-seq analysis of WT B cells spiked with a known quantity of ERCC control ( ERCC-RNA-seq , S2A and S3 Figs ) [52] . This analysis showed that 8 , 000 ( naïve B cells ) to 11 , 000 ( B cells activated for 25 . 5h ) genes are transcribed in B cells at greater than 0 . 5 copy per cell ( termed transcribed genes ) , with median copy numbers of 2 . 6 ( naïve ) , 10 ( 13 . 5h ) , and 31 ( 25 . 5h ) ( Fig 1A and S1 Table ) . This general transcriptional upregulation is essential for activation-induced cell growth and proliferation . Consistent with previous reports , the abundance of significantly expressed mRNAs spans three to four orders of magnitude ( S3A Fig ) , with 1 RPKM roughly corresponding to 1 copy per cell ( S3B Fig ) [53 , 54] . The transcribed genes included 85% ( 743 in naïve B cells ) to 90% ( 780 in 25 . 5h activated B cells ) of miR-17~92 target genes ( termed transcribed targets ) . We next performed microarray analysis of TKO , WT and TG B cells before and after activation ( S2B Fig ) , focusing on the transcribed targets ( Fig 1B and 1C ) . The time frame used in this study covered both the induction and termination phases of major signaling pathways involved in B cell activation ( S4A Fig ) . We confirmed that miR-17~92 expression in TG B cells was 3 fold higher than in WT B cells , and was completely absent in TKO B cells ( S4B and S4C Fig ) . When miR-17~92 target genes were analyzed , neither transgenic miR-17~92 expression nor deletion of miR-17~92 family caused significant global changes in their mRNA levels throughout B cell activation ( Fig 1B and 1C and S2 Table ) . Analysis of target genes regulated by individual members of the miR-17~92 cluster came to the same conclusion ( S5A and S5B Fig ) . We next examined the effect of miR-17~92 on predicted target genes with the highest context++ scores based on the most recent TargetScan 7 . 0 algorithm ( S3 Table ) [55] . We selected 128 top target genes for each miRNA subfamily in this cluster , and analyzed the mRNA levels of these transcribed in B cells at greater than 0 . 5 copy per cell ( S5C and S5D Fig ) . In a previous study , transfection of chemically synthesized miRNA mimics into HCT116 cells led to an average reduction of 19% in the mRNA levels of target genes with the same context scores [55] . During B cell activation , there was indeed an inverse correlation between the expression levels of miR-17~92 and these target gene mRNAs at all time points examined , but the average change in target mRNA levels was only 3 . 7% in TG and 6 . 6% in TKO B cells ( S5C and S5D Fig ) . This rather modest global effect of miR-17~92 on the mRNA levels of its target genes is consistent with the results from previous studies performing transcriptome analysis of T cells , B lymphoma cells , and embryonic heart and tail bud with genetic ablation of either the whole miR-17~92 cluster or its individual members [51 , 56–58] . We speculate that these subtle changes in target gene mRNA levels may not explain the dramatic phenotypes observed in TG and TKO mice . Moreover , most of the small number of target genes that show greater than 1 . 4-fold changes in mRNA levels have not been previously implicated in lymphoma development , cell survival and proliferation , and are unlikely to mediate the functions of miR-17~92 in B cells ( S2 Table ) . Therefore , we investigated the possibility that miR-17~92 regulates the expression of functionally relevant target genes mainly at the protein level . We compiled a list of 63 miR-17~92 target genes , which were either validated in previous studies [59–64] , or are novel but functionally relevant to B cell lymphoma development or B cell immune responses ( S4 Table ) . Among these 63 targets , we were able to detect and quantify 47 by immunoblot , while the other 16 were discarded due to poor antibody quality ( S6 Fig ) . Only 13 of the 47 target genes showed significant reduction in protein levels in TG B cells ( S7A Fig ) , including several inhibitors of the PI3K ( Pten and Phlpp2 ) and NF-κB ( Tnfaip3/A20 , Itch , Rnf11 , Tax1bp1 , Cyld , and Traf3 ) pathways previously implicated in miR-17~92-driven B cell lymphoma development [40 , 65] , and five additional tumor suppressor genes ( Hbp1 , Stk38 , Arid4b , Rbbp8 and Ikzf1 ) [66–69] . Among the other 34 targets , there were no significant changes in protein levels for 25 targets , time- or isoform-dependent changes for 3 targets , and increased protein levels for 6 targets in TG B cells ( S7B and S7C Fig ) . We further examined the 13 target genes that exhibited reduced protein levels in TG B cells , focusing on the relative contribution of translational repression and mRNA degradation ( Fig 1D ) . We validated the microarray data by qRT-PCR . All of them are regulated either exclusively or significantly at the protein level , except Hbp1 , which is regulated mainly at the mRNA level ( Fig 1D ) . We also measured the protein levels of 16 genes that control translation initiation and elongation in a global manner and completely lack miR-17~92 binding sites in their mRNAs [70] . As shown in S8 Fig , none of them was significantly downregulated in TG B cells . Taken together , these results demonstrated that the global impact of miR-17~92 on its target gene mRNA levels is subtle , that only a subset of functionally relevant target genes are suppressed by transgenic miR-17~92 expression , that miR-17~92-mediated suppression occurs predominantly at the protein level , and that this suppression is not caused by an altered translational environment in TG B cells . We next assessed the impact of miR-17~92 expression on target genes using ribosome profiling ( S2C Fig ) . This technology directly captures genome-wide maps of protein synthesis ( the translatome ) by quantifying ribosome density on each mRNA with high resolution and depth , but does not measure post-translational changes in gene expression [71 , 72] . We quantified ribosome footprints of 8 , 271 mRNAs ( termed translated genes ) in TKO , WT and TG B cells after 25 . 5h of activation , corresponding to more than 70% of transcribed genes in these cells ( S5 Table ) . The ribosome footprint abundance spans six orders of magnitude ( S9A Fig ) , which is at least two orders of magnitude broader than that of mRNA abundance ( S3A Fig ) , in agreement with previous global gene expression analysis in mammalian cells [73] . We confirmed that ribosome footprint abundance changes highly correlated with protein abundance changes as determined by immunoblot , therefore excluding significant contribution from post-translational regulatory mechanisms such as miRNA-dependent nascent polypeptide destruction ( S9B–S9E Fig ) [74] . Among the 780 transcribed miR-17~92 targets , 641 were detected by significant numbers of ribosome footprints ( termed translated targets ) ( S10A Fig ) . Notably , only 123 ( 19 . 2% of translated targets ) showed greater than 1 . 4-fold reduction in ribosome footprint abundance in TG B cells ( termed ribo-downregulated TG targets ) , while only 80 ( 12 . 5% of translated targets ) were de-repressed by 1 . 4 fold or more in TKO B cells ( termed ribo-upregulated TKO targets ) ( S10B and S10D Fig ) . When the median values of translation changes were compared with these of mRNA changes , translation changes were dominant at the global level , in both TG and TKO B cells ( S10C and S10E Fig ) . Therefore , only a small fraction of target genes respond to changes in miR-17~92 expression levels and miR-17~92 regulates its target gene expression mainly at the translational level . We compared the list of target genes de-repressed by miR-17~92 family miRNA deletion ( ribo-upregulated TKO targets ) with those suppressed by transgenic miR-17~92 overexpression ( ribo-downregulated TG targets ) . To our surprise , these two lists overlapped by only 8 genes , including four previously validated targets ( CD69 , Fbxw7 , Egr2 , and Caprin2 ) ( Fig 2A ) [75–80] . When ribosome profiling data of TKO , WT , and TG B cells were analyzed together , it became clear that ribo-upregulated TKO targets as a group showed significant reduction in ribosome density when miR-17~92 family miRNA expression increased from almost zero in TKO B cells to WT levels , but did not show further reduction when miRNA expression increased from WT to TG levels ( Fig 2B and S6 Table ) . In the same analysis , the ribosome density of ribo-downregulated TG targets did not exhibit any significant changes between TKO and WT B cells , but showed significant reduction when miR-17~92 expression increased from WT to TG levels ( Fig 2C and S6 Table ) . Translated genes lacking miR-17~92 binding sites were used as negative control , whose ribosome density showed no significant alterations in B cells expressing miR-17~92 at three different levels ( Fig 2D ) . The differential responses of target genes to three different levels of miR-17~92 expression in TKO , WT and TG B cells were confirmed by immunoblot analysis of individual target genes ( S11 Fig ) . We first examined TKO B cells for their expression of the 13 targets suppressed in TG B cells ( S7A Fig ) . Six of them ( Phlpp2 , Rnf11 , Arid4b , Tax1bp1 , Cyld and Pten ) showed significant de-repression in protein levels , but the degree of de-repression was relatively small ( 1 . 2–1 . 5 fold ) , while the expression of the other seven was not altered ( S11A and S11B Fig ) . In contrast , among the 34 targets that were not suppressed in TG B cells ( S7B and S7C Fig ) , 10 showed significant increase in ribosome footprint abundance in TKO B cells and belonged to ribo-upregulated TKO targets ( Mink1 , Phlda3 , Fbxw7 , Map3k3 , Tmem127 , Rapgef2 , Dusp2 , Rb1 , Sos1 and Lats2 ) ( S6 Table ) . This was further confirmed by immunoblot analysis of TKO B cells , which showed up to 3 . 3-fold increases in protein levels of these genes ( S11C and S11D Fig ) . When the relative protein levels of these 23 targets in TKO , WT , and TG B cells were plotted together , it became obvious that different targets exhibit different sensitivity to changes in miR-17~92 expression levels ( Fig 2E ) . Ten targets ( termed group 1 targets ) were suppressed when miR-17~92 expression increased from TKO to WT levels , but showed little suppression in TG B cells . Seven targets ( termed group 3 targets ) were suppressed when miR-17~92 expression increased from WT to TG levels , but showed only marginal de-repression in TKO B cells . The other six targets ( termed group 2 targets ) showed suppression when miR-17~92 expression increased from WT to TG levels , and were de-repressed in TKO B cells ( Fig 2E ) . In contrast to the significant changes in their protein levels , the mRNA levels for most of them remain the same in TKO , WT , and TG B cells , regardless of target groups ( Fig 2F ) . We next performed reporter assays in wild type B cells to investigate whether miR-17~92 exerts its effect on these target genes through its cognate binding sites on target mRNAs . As shown in S12 Fig , mutation of miR-17~92 binding sites led to increased activity of a luciferase gene fused to target gene 3’UTRs , therefore demonstrating direct regulation of these target genes by miR-17~92 in B cells . Based on these results , we propose the following model of differential sensitivity of target genes to miRNA suppression . For a miRNA-target mRNA interaction , there is a threshold level and a saturation level of miRNA concentration ( Fig 2G ) . Target gene expression is suppressed by miRNA in a dose-dependent manner when miRNA concentration is between these two levels . Below the threshold level , target gene expression cannot be suppressed by miRNA . Above the saturation level , target gene expression cannot be further suppressed by increasing concentration of miRNA . The maximal difference in target gene protein levels ( termed amplitude ) is reached when miRNA concentration increases from the threshold level to the saturation level . Different target genes exhibit different threshold level , saturation level , and amplitude in their responses to the same miRNA ( or miRNA cluster ) ( Fig 2H ) . The differences in threshold and saturation levels underlie the different sensitivity of group 1 , 2 , 3 target genes to changes in miR-17~92 expression levels , while the differences in amplitude explain the various degrees of suppression or de-repression in TG and TKO B cells , respectively ( Fig 2E ) . A prediction of this model is that not all miRNA binding sites are occupied by miRNA . Therefore , it is likely that there are less miRNA molecules than miRNA binding sites in WT B cells . To test this , we determined the copy numbers of miR-17~92 miRNA molecules and miR-17~92 binding sites present in B cells during activation . The miRNA molecule numbers in WT B cells were determined by quantitative Northern blot analysis of WT B cells and TKO B cells spiked with graded amounts of chemically synthesized mature miR-17~92 family miRNAs ( Fig 3A–3C and S7 Table ) . By combining the mRNA molecule numbers determined by ERCC-RNA-seq ( S1A Fig ) and conserved miR-17~92 binding sites determined by PAR-CLIP [40] , we calculated the number of conserved miR-17~92 binding sites in a B cell ( Fig 3D and S8 Table ) . Our calculation showed that each naïve B cell expresses 900–1 , 800 molecules of miR-17 , miR-19 , and miR-92 subfamily miRNAs , and 80 molecules of miR-18 subfamily miRNAs ( Fig 3B and 3C and S7 Table ) . The ratios between conserved miR-17~92 binding sites and miRNA molecules range from 0 . 5 ( miR-92 subfamily ) to 4 . 6 ( miR-18 subfamily ) in naïve B cells ( Fig 3E ) . Upon activation , both miR-17~92 miRNAs and their target mRNAs are up-regulated ( Fig 3C and 3D ) , but the fold increase of the latter outpaces the former , thereby increasing the ratios between conserved binding sites and miRNA molecules to 2 . 8 ( miR-92 family ) and 8 . 7 ( miR-18 family ) in 25 . 5h activated B cells ( Fig 3E ) . Moreover , the PAR-CLIP analysis identified 2 . 4-fold more non-conserved binding sites than conserved ones [40] . Previous studies showed that non-conserved binding sites can also be occupied by RISC [36] . Taking non-conserved binding sites into account , potential miR-17~92 binding sites outnumber miRNA molecules even further , by as much as 20-fold . These estimations are consistent with results from previous studies measuring the molecule numbers of miRNAs and their binding site numbers on target mRNAs in hepatocytes and stem cells [81 , 82] . Therefore , we conclude that only a fraction of potential binding sites are occupied by miR-17~92 miRNAs at any given time . We next investigated how miRNAs regulate target gene translation using polysome profiling ( S2D Fig ) [23 , 83] . While ribosome profiling measures ribosome footprint abundance , which is a sum of mRNA abundance and translation rate [72] , polysome profiling directly measures the number of ribosome associated with a mRNA molecule , independent of mRNA expression level ( S13 Fig ) [84] . We first confirmed that miRNA gene mutations had little impact on the global polysome profile ( S14 Fig ) . We then investigated the distribution of individual miRNAs and mRNAs in the sucrose gradient . miR-21 , one of a few miRNAs enriched in monosome fractions [32 , 85] , and highly abundant in B cells [86] , was used as control . In contrast to miR-21 , miR-17~92 miRNAs were mainly associated with light polysomes ( Fig 4A and 4B ) . This suggests that miR-17~92 miRNAs are predominantly associated with target mRNAs undergoing slow translation . Next we measured the distribution of target mRNAs in the sucrose gradient . While the β-Actin mRNA ( Actb ) was enriched in heavy polysome fractions , mRNAs of all validated miR-17~92 target genes exhibited a bimodal distribution ( Fig 4C ) . The first peak was located at fractions 10–11 , corresponding to mRNAs associated with 3–4 ribosomes , while the second peak was located at fractions 14–16 , corresponding to mRNAs associated with more than 7 ribosomes ( Fig 4C ) . Our quantification of miR-17~92 family miRNA molecules and their potential binding sites on target mRNAs in B cells suggested that only a fraction of target mRNA molecules are occupied by these miRNAs ( Fig 3E ) . In addition , the distribution of miR-17~92 family miRNAs largely overlapped with the first peak of their target mRNAs ( Fig 4B and 4C ) . Taken together , these results suggest that target mRNAs are compartmentalized: target mRNAs in the first peak are associated with miR-17~92 family miRNAs and undergo slow translation , while target mRNAs in the second peak are largely free of miR-17~92 family miRNAs and undergo more active translation . Transgenic miR-17~92 expression shifted a fraction of target mRNAs from the second peak into the first peak ( Fig 4C ) , thereby reducing the overall translation rate and protein output ( S7A Fig ) . Consistent with the previous observations that miR-17~92 regulation of Hbp1 occurs mainly at the mRNA level ( Fig 1D and S9E Fig ) , the distribution of the Hbp1 mRNA in the sucrose gradient showed little change ( Fig 4C ) . We conducted the same analyses for another well-studied lymphocyte-specific miRNA , miR-155 [87] , to see whether our observation is a general phenomenon . This included absolute quantification of miR-155 and its binding sites in WT B cells and ribosome profiling analysis of miR-155-deficient ( 155KO ) B cells . Our results showed that there are 7-fold more conserved miR-155 binding sites than miR-155 molecules ( S15A Fig ) , that miR-155 was enriched in light polysome fractions ( S15B Fig ) , and that deletion of miR-155 caused a significant shift of mRNAs of previously validated target genes ( Aicda , Sfpi1 , Jarid2 and Peli1 ) from light to heavy polysomes ( S15C Fig ) [88–91] . To independently confirm these results , we took an un-biased approach to assess changes of target mRNA distribution in the sucrose gradient ( Poly-RNA-seq , S16A Fig ) . We performed RNA-seq analysis of total RNA purified from polysome fractions 10–11 and 14–16 , corresponding to the first and second peaks of target mRNA distribution in the sucrose gradient , and calculated the relative abundance of target mRNA of interest in these two peaks in B cells from mice of different genotypes . This analysis produced results consistent with polysome profiling analysis , showing that miR-17~92 target mRNAs were enriched in fractions 10–11 while depleted in fractions 14–16 in TG B cells , and miR-155 target mRNAs were depleted in fraction 10–11 while enriched in fractions 14–16 in miR-155 KO B cells ( S16B and S16C Fig ) . Taken together , our polysome profiling analysis of individual target gene mRNAs demonstrated that miRNAs suppress target gene expression by reducing ribosome occupancy on a fraction of target mRNA molecules . We sought to understand what determines target gene sensitivity to miRNA-mediated translational repression . While the contribution of seed types and other cis-factors has been extensively investigated in the cellular contexts in which miRNAs predominantly act to decrease target mRNA levels [8 , 92] , the factors that regulate miRNA-mediated translational repression remain largely unknown . We systematically investigated the length of 5’UTR , coding region ( CDS ) and 3’UTR , numbers of conserved miR-17~92 binding sites , enrichment of specific seed types , and locations of binding sites . We found mRNAs with miR-17~92 binding sites tend to have longer 5’UTR , CDS , and 3’UTR , but their length did not predict target gene sensitivity . None of the other features correlates with target gene sensitivity globally ( S17 Fig ) . As our results showed that miR-17~92 suppresses target gene expression mainly through translational repression , we then focused on molecular features implicated in translational regulation [93] . Ribosome footprint distribution analysis showed that there were ribosome footprints in 5’UTRs of miR-17~92 target genes , though their abundance was lower than ribosome footprint abundance in CDS ( S18 Fig ) . A close examination revealed a significant accumulation of ribosome footprints in 5’UTRs of ribo-upregulated TKO targets in WT B cells as compared to TKO B cells ( Fig 5A ) . This suggested that miR-17~92 represses translation initiation of these target genes through their 5’UTRs ( See discussion ) . Consistently , ribosome footprint abundance in 5’UTRs of ribo-downregulated TG targets was increased when miR-17~92 expression increased from WT to TG levels ( Fig 5B ) , while other non-responsive target genes did not show significant changes in ribosome footprint abundance in their 5’UTRs in TKO , WT or TG B cells ( Fig 5C ) . Moreover , local ribosome occupancy in 5’UTRs inversely correlated with overall ribosome density , which is a good indicator of translation rate and protein output . This suggests a role of ribosome hindrance in 5’UTR in suppressing translation initiation ( Fig 5D ) . We searched the 5’UTRs of ribo-upregulated TKO targets for potential enrichment of specific sequence motifs but did not find any . Instead , we found high GC content in these 5’UTRs , and the position of the GC content peak correlated with the position of the ribosome footprint peak ( Fig 5A and 5E ) . The translation efficiency of mammalian mRNAs is highly sensitive to GC content of 5’UTR , as high GC content often indicates the presence of secondary structures . A previous study showed that an increase in 5’UTR GC content from 52% to 62% led to a 2-fold decrease in translation efficiency [94] . In line with this , recent bioinformatic analyses implied that local structures in 5’UTRs contribute to efficient miRNA-mediated gene regulation via translational repression [19 , 95] . Moreover , our reporter assay experiments confirmed direct regulation of group 1 targets by miR-17~92 in wild type B cells , but the degree of de-repression in reporter activity caused by binding site mutation was often less than the degree of de-repression in target gene protein levels in TKO B cells ( Fig 2E and S12 Fig ) . This suggests that cis-elements beyond miRNA binding sites in 3’UTRs contribute to the amplitude of target gene regulation . Taken together , we surmised that ribosome hindrance mediated by secondary structures in 5’UTRs contribute to target gene sensitivity to miRNA suppression at the translation initiation stage . We explored this idea further by focusing on CD69 , the most sensitive target gene among the 24 validated by immunoblot ( Fig 2E ) . CD69 has a relatively short 5’UTR ( 84nt ) , which harbors no internal ribosome entry sites ( IRES ) or 18s rRNA binding regions that may enhance cap-independent translation [97–99] . Instead , there are two sub-optimal start codons ( AUC and GUG ) and a potential hairpin [100 , 101] ( Fig 6A and S19 Fig ) . Consistent with the global analysis of ribo-upregulated TKO targets ( Fig 5A ) , there was an accumulation of ribosome footprints in CD69 5’UTR in WT B cells ( Fig 6B ) . The ribosome footprint is 31 nt long , corresponding to the width of a single ribosome . Interestingly , the ribosome footprint overlaps with the two sub-optimal start codons and the 5’ arm of the putative hairpin , while its abundance shows positive correlation with miR-17~92 expression levels and negative correlation with CD69 expression ( Figs 6B and 2E ) . We hypothesized that the two sub-optimal start codons and the hairpin work together to slow down translational initiation , thereby rendering CD69 mRNA sensitive to translational repression . Indeed , polysome profiling analysis confirmed that miR-17~92 represses CD69 expression at the translation level ( Fig 6C ) , and deletion of the miR-17~92 family miRNAs led to a 4 . 5-fold increase in cell surface expression of CD69 in TKO B cells , with only marginal effect on its mRNA level ( Fig 6D ) . We investigated the functional contribution of CD69 5’UTR to its regulation by miR-17~92 using a modified form of the dual luciferase reporter psiCheck-2 ( Fig 7A ) [102] . In this plasmid ( termed psiCheck-2-pd ) , the firefly luciferase gene ( Fluc ) controlled by the human thymidine kinase ( TK ) gene promoter is used as an internal reference for transfection efficiency . We placed the CD69 3’UTR downstream of the renilla luciferase gene ( hRluc ) . The wild type CD69 3’UTR ( wt ) contains three binding sites for miR-17~92 miRNAs ( one for miR-17 subfamily and two for miR-92 subfamily ) . We introduced 3nt mutations into these binding sites to abolish their interactions with miR-17~92 miRNAs to generate a mutated form of CD69 3’UTR ( mut ) . A comparison of the renilla/firefly luciferase activity ratio ( hRluc/Fluc ) between psiCheck-2-pd containing wt and mut CD69 3’UTR should reveal the sensitivity of the renilla luciferase mRNA to miR-17~92-mediated suppression . To examine the role of CD69 5’UTR in regulating translation rate and sensitivity to miRNA suppression , we inserted CD69 5’UTR between the transcription start site of ( TSS ) of the SV40 promoter and the Kozak sequence of the renilla luciferase gene ( Fig 7A ) . The β-Actin 5’UTR contains no obvious secondary structures , exhibited minimal accumulation of ribosome footprints in TKO , WT , and TG B cells ( Fig 6B ) , and was used as a control . We performed dual-luciferase reporter assays in in vitro activated WT B cells to closely imitate the experimental conditions of ribosome profiling and polysome profiling ( Fig 7B ) . As expected , the firefly luciferase activity remained as a constant ( Fig 7C ) . The renilla luciferase reporter containing CD69 5’UTR showed a 4 . 4 fold de-repression when miR-17~92 binding sites in its 3’UTR were mutated , very similar to the fold de-repression of the endogenous CD69 gene in TKO B cells ( Figs 6D and 7C ) . Replacing CD69 5’UTR with β-Actin 5’UTR significantly reduced the sensitivity of the renilla luciferase reporter gene to miR-17~92 suppression ( Fig 7C ) . qRT-PCR analysis of renilla and firefly luciferase mRNAs showed that the ratio between these two mRNAs was not affected by changes in 5’UTR or 3’UTR , excluding any substantial contributions from mRNA changes ( Fig 7D ) . We next performed similar reporter assays in TG , WT , and TKO B cells , using the psiCheck-2-pd reporter with wild type CD69 5’UTR and 3’UTR . Consistent with CD69 expression in B cells of these three genotypes ( Fig 2E ) , the expression of renilla luciferase was more sensitive to miR-17~92 depletion than overexpression ( S20A Fig ) . To understand the functional contribution of the putative hairpin and two sub-optimal start codons in CD69 5’UTR to the sensitivity of CD69 mRNA to miR-17~92 suppression , we deleted the left arm of the hairpin ( ΔHP ) or mutated these two sub-optimal start codons ( Mut-uORF ) , and performed reporter assays in WT B cells . Deletion of the left arm of the hairpin reduced the sensitivity of renilla luciferase to miR-17~92 suppression , but no significant effect was observed for mutating the two sub-optimal start codons ( S20B Fig ) . Taken together , these results demonstrate that structural components in 5’UTR play an important role in regulating the sensitivity of target mRNA to translational repression by miRNAs .
This study provides mechanistic insights into the functional specificity of miRNAs and the key target gene model , which postulates that miRNAs exert their specific functions by suppressing the expression of a small number of key target genes [44] . Our findings , together with previously published studies [41–43] , suggest that key target genes emerge from a pool of hundreds of target genes via multiple mechanisms . That is , there are mechanisms that regulate miRNA binding to target mRNAs , the consequences of miRNA binding , and cellular responses to reduced target gene protein levels ( S21 Fig ) . First , there are more binding sites than miRNA molecules and only a fraction of binding sites are occupied by miRNA-containing RISC complexes at any given time . Target mRNAs often associate with RNA-binding proteins ( RBPs ) and exhibit certain secondary and tertiary structures , which interfere with the recruitment of RISC and result in differential accessibility and affinity to miRNA [30 , 103–106] . When hundreds of target mRNA species compete for a limited amount of miRNA molecules , binding sites with easy accessibility and high affinity are preferentially occupied . Increasing the cellular concentration of miRNA molecules leads to saturation of the most favorable binding sites and occupation of additional binding sites with lower accessibility and affinity . Consistent with our view , a previous study demonstrated that target accessibility is a critical determinant of miRNA-mediated translational repression in the cellular context where miRNAs do not cause target mRNA degradation [107] . Therefore , the accessibility and affinity of binding sites , as well as the presence of competing target mRNA species , establish the threshold and saturation levels of miRNA for a given target mRNA ( Fig 2G ) . Second , miRNA binding does not necessarily warrant functional consequence . There are mechanisms that determine whether miRNA binding leads to changes in target gene protein levels . This study shows that 5’UTR is a part of the mechanisms regulating target gene sensitivity to miRNA suppression at the translation initiation stage . Third , there are mechanisms that regulate cellular responses to changes in target gene protein levels . We speculate that changes in the protein levels of many target genes brought about by a miRNA are functionally inconsequential , as shown by many examples of genetic mutant mice with no observable phenotypes [108] . Nevertheless , there are a small number of target genes that are functionally sensitive to reduced protein levels in a given cellular context , as documented by the pathologies arising from haploinsufficiency [109–111] . These dosage sensitive target genes likely serve as critical mediators of miRNA functions and are the key target genes in that particular cellular context ( S21 Fig ) . How many key targets are there to mediate the function of a miRNA in a given cellular context ? Our global analysis of miR-17~92 target genes in primary B cells provide insights into this question . Among the 868 experimentally identified targets with conserved miR-17~92 binding sites [40] , 780 are significantly transcribed and 641 are significantly translated . When the cutoff is set at 1 . 4 fold change in ribosome footprint abundance , only 80 of them are suppressed by the WT levels of miR-17~92 and qualify as responsive targets , amounting to 9% of experimentally identified targets . As discussed above , it is likely that only a fraction of these 80 target genes are relevant for the function of miR-17~92 in B cells . Therefore , the number of key target genes is further reduced to a few percent of the 868 targets . For miRNA genes encoding a single mature miRNA , which often has 100–200 putative target genes , this would translate into only a few key targets for a given cellular context ( S21 Fig ) . Consistent with this estimation , recent genetic studies showed that mutation of miRNA binding sites in a single target gene phenocopied defects caused by miRNA deficiency in a cell context-dependent manner , demonstrating that individual miRNA-target mRNA interactions can play critical roles in mediating the function of miRNAs in animals [41–43] . For most mRNAs , translation initiation occurs by a cap-dependent scanning mechanism , which requires the binding of the trimeric complex eIF4F ( comprised of eIF4E , eIF4A , and eIF4G ) to the m7G cap structure , followed by recruitment of the preinitiation complex ( PIC ) and scanning of PIC to the first AUG codon positioned within a good context [112] ( S22A Fig ) . The secondary structures in their 5’UTRs play important roles in regulating translation initiation [113] . Scanning through these secondary structures require additional factors and ATP , and this requirement depends on the position and stability of secondary structures [114 , 115] . RNA helicases such as eIF4A are required for unwinding these secondary structures and for facilitating the scanning of PIC [116] . Recent studies suggested that miRNAs require eIF4As to regulate translation of their target mRNAs [19 , 117 , 118] . While two studies demonstrated miRNAs repress target gene translation by facilitating dissociation of eIF4As from target mRNAs [117 , 118] , a third one proposed they repress target mRNAs by recruiting eIF4AII [19] . Even though the detailed molecular mechanisms by which eIF4As mediate miRNA function are contradictory in these reports , the requirement of eIF4As in miRNA function during PIC scanning is consistent with other studies that utilized reporter constructs whose translational initiation bypasses the PIC scanning process . These reporter genes were immune to miRNA-mediated repression , suggesting that miRNA repression takes place during PIC scanning [23 , 119] . It should also be noted that other studies demonstrated that miRISC and the CCR4-NOT complex can silence target mRNA in an eIF4A-independent manner , suggesting the eIF4A dependency can be context-specific [120] . Our study suggests that the most sensitive targets ( such as ribo-upregulated TKO targets ) contain more structured 5’UTRs . In the absence of miR-17~92 family miRNAs ( in TKO B cells ) , eIF4As or other RNA helicases facilitate the unwinding of these secondary structures , allowing PIC to scan through and to initiate translation . In the presence of WT levels of miR-17~92 family miRNAs , RISC complexes are recruited to these target mRNAs through their cognate binding sites in the 3’UTRs , and dissociate RNA helicases from the 5’UTRs . This results in stabilization of secondary structures and accumulation of PIC ( and ribosome footprints in ribosome profiling experiments ) in the 5’UTRs , repression of translation initiation , and a reduction in protein output ( S22B Fig ) . When miR-17~92 expression is further increased to the TG levels , less sensitive targets ( such as ribo-downregulated TG targets ) that do not respond to WT levels of miR-17~92 become responsive at this higher level . Our reporter assays demonstrate that specific structural components in 5’UTR indeed regulate miRNA-mediated translational repression , but the detailed molecular interactions between miRNA , 5’UTR , and the translation initiation machinery warrant future investigation . Our findings also provide a straightforward explanation for the recent observations that deletion and overexpression of the same miRNA gene can lead to unrelated phenotypes [121 , 122] . As a representative example , early studies have shown that overexpression of members of the miR-34 family miRNAs has potent tumor suppressor function downstream of p53 [121] . However , mice carrying target deletion of all miR-34 genes display normal p53 responses to a variety of cellular insults , including ionizing radiation and oncogenic stress [123] . Another study reported that mice deficient of all the six miRNAs in the miR-34/449 family exhibited postnatal mortality , infertility and strong respiratory dysfunction caused by defective mucociliary clearance , resulting from a significant decrease in cilia length and number [122] . Our study suggests that different functions of miR-34 family miRNAs in these overexpression and deletion studies can be explained by different sensitivity of target genes to miR-34 suppression . When these miRNAs are expressed at WT levels , target genes regulating cilia assembly ( i . e . Cp110 ) are among the most sensitive and their expression is suppressed . Deletion of all miR-34/449 family genes results in de-repression of these genes and impaired cilia assembly [122] . When miR-34 family miRNAs are overexpressed at levels much higher than WT levels , another group of target genes , which are less sensitive and only respond to higher than WT levels of miR-34 , are suppressed . This group contains positive regulators of cell cycle and DNA-damage responses ( i . e . Cdk4 , Ccne2 , and Met ) , whose suppression bestows anti-tumor functions to miR-34 family miRNAs [121] . Therefore , different sensitivity of these two groups of target genes , one regulating cilia assembly and the other regulating cell cycle and DNA damage response , to miR-34 suppression underlies the different phenotypic consequences brought about by overexpression and deletion of this family of miRNAs . A recent study investigating real-time translation of single mRNA molecules in live mammalian cells revealed surprising heterogeneity in the translation of individual mRNAs from the same gene within the same cell , including rapid and reversible transitions between translationally active and inactive states [124] . The same study showed that the long form 5’UTR of the Emi1 gene , when placed upstream of a GFP reporter gene , caused a 40-fold reduction in the GFP protein level . While a great majority of GFP mRNAs containing the long form 5’UTR of the Emi1 gene were strongly translationally repressed , a small subset of these mRNAs still escaped repression and underwent robust translation . These results suggest that cis-elements in the long form 5’UTR of the Emi1 gene drastically shifted , but did not completely shut off , the GFP mRNAs from translationally active states into inactive states . This is quite similar to our polysome profiling analysis of miR-17~92 target genes in WT and TG B cells , which showed that transgenic miR-17~92 expression shifted only a fraction of its target mRNAs from rapid translation states into slow translation states . A previous study investigating the effect of endogenous Let-7 miRNA on a reporter target gene in HeLa cells came to similar conclusions [23] . The authors further proposed that the translationally repressed reporter mRNAs , as well as Let-7 miRNAs , are localized in processing bodies , a subcellular structure for mRNA storage or degradation . Considering our study together with these other studies , it is likely that miRNAs repress target gene expression by tipping the dynamic balance between translationally active and inactive states . Similar to the heterogeneity in the translation of individual mRNAs from the same gene within the same cell , emerging evidence suggests that mechanisms of miRNA action are also heterogeneous . A recent survey of studies investigating miRNA effect on functionally important target genes in 77 strains of miRNA mutant mice found that miRNA-target interaction can lead to translational repression , target mRNA degradation , or both [2] . It remains unclear what determines the relative contribution of these two modes of miRNA action to target gene suppression . Previous studies suggest that this could be both cellular context- and miRNA-dependent [1 , 2 , 125] . In this study , we showed that most functional target genes of miR-17~92 are suppressed at the translational level , but some target genes are suppressed by mRNA degradation , either completely or partially . This target gene-dependency became more clear when we applied the same transcriptome and translatome analyses to miR-155-deficient B cells . Our unpublished results showed that in the same cellular context , miR-155 suppresses its target gene expression by translational repression , mRNA degradation , or both , and this is completely target-gene dependent . Future investigation is warranted to identify cellular factors , as well as cis-elements in both miRNAs and target mRNAs , that determine the molecular consequence of individual miRNA-target mRNA interactions . In summary , we conducted an integrated analysis of miR-17~92 family miRNAs , their target genes , and the functional consequences of these miRNA-target gene interactions in primary B cells expressing miR-17~92 family miRNAs at three different physiological levels . We present evidence showing that there are more binding sites than miRNA molecules , that target genes exhibit differential sensitivity to miRNA suppression , and that only a small fraction of target genes are actually suppressed by a given concentration of miRNA . Transgenic expression and deletion of the same miRNA gene regulate largely distinct sets of target genes . miR-17~92 regulates functional target gene expression mainly through translational repression in activated B cells and 5’UTR plays an important role in regulating target gene sensitivity to miRNA suppression . These findings provide mechanistic insights into the key target gene model in which the specific function of a miRNA is achieved by regulating a small number of key target genes .
All mice were used in accordance with guidelines from the Institutional Animal Care and Use Committees of The Scripps Research Institute and Xiamen University . The generation of miR-17~92 Tg ( Jax stock 008517 ) , miR-17~92fl/fl ( Jax stock 008458 ) , miR-106a~363-/- mice ( Jax stock 008461 ) , miR-106b~25-/- mice ( Jax stock 008460 ) , CD19-Cre ( Jax stock 006785 ) was previously reported [49 , 126 , 127] . MiR-17~92 Tg mice were crossed with CD19-Cre mice to generate miR-17~92 Tg/Tg;CD19Cre ( TG ) mice [40] . miR-17~92fl/fl mice were crossed with miR-106a~363-/- mice , miR-106b~25-/- mice and CD19-Cre mice to generate miR-17~92fl/fl;miR-106a~363-/-;miR-106b~25-/-;CD19-Cre ( TKO ) mice . miR-155-/- were obtained from The Jackson Laboratory ( Jax stock 007745 ) [87] . Spleen and peripheral lymph nodes were collected from 2~3 month old TG , TKO and wild type ( WT ) mice . WT and TG B cells were purified by depleting cells positive for AA4 . 1 ( CD93 ) , CD43 and CD5 , while TKO B cells were purified by depleting cells positive for AA4 . 1 ( CD93 ) , CD43 and CD9 using MACS LD columns ( Miltenyi Biotec ) following manufacturer’s instructions . Purified B cells were cultured at a density of 5x106 cells/ml for indicated amounts of time in B cell medium plus LPS ( 25μg/ml ) and IL-4 ( 5ng/ml ) in 37°C incubator , unless indicated otherwise . At the time of harvest , live cells were purified by Ficoll ( GE Healthcare , 17-1440-02 ) to achieve a purity of >90% live cells and >98% B220+CD19+ B cells before further analysis . B cell medium was made of DMEM GlutaMAX ( Gibco 10569 ) plus 10%v/v FCS , 1x non-essential amino acids ( Corning , 25-025-CI ) , 10mM HEPES ( Gibco , 15630 ) , 50μM β-ME ( Gibco , 21985 ) , 1x Penn/Strep . P values were determined by using two-tailed Student’s t-test . Statistical significance is displayed as *P < 0 . 05 , **P < 0 . 01 and ***P < 0 . 001 . Detailed procedures and analysis methods are present as a supplementary material . Please see “S1 Methods” Microarray , RNA-Seq , and ribosome profiling data are available at NCBI Gene Expression Omnibus through the accession numbers GSE56379 , GSE83734 , and GSE83684 . | MicroRNAs ( miRNAs ) are small RNAs encoded by our genome . Each miRNA binds hundreds of target mRNAs and performs specific functions . It is thought that miRNAs exert their function by reducing the expression of all these target genes and each to a small degree . However , these target genes often have very diverse functions . It has been unclear how small changes in hundreds of target genes with diverse functions are translated into the specific function of a miRNA . Here we take advantage of recent technical advances to globally examine the mRNA and protein levels of 868 target genes regulated by miR-17~92 , the first oncogenic miRNA , in mutant mice with transgenic overexpression or deletion of this miRNA gene . We show that miR-17~92 regulates target gene expression mainly at the protein level , with little effect on mRNA . Surprisingly , only a small fraction of target genes respond to miR-17~92 expression changes . Further studies show that the sensitivity of target genes to miR-17~92 is determined by a non-coding region of target mRNA . Our findings demonstrate that not every target gene is equal , and suggest that the function of a miRNA is mediated by a small number of key target genes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"blood",
"cells",
"cell",
"binding",
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"gene",
"regulation",
"immunology",
"messenger",
"rna",
"micrornas",
"cellular",
"structures",
"and",
"organelles",
"white",
"blood",
"cells",
"animal",
"cells",
"gene",
"expression",
"ribosomes",
"biochemistry",
"rna",
"antibody-producing",
"cells",
"cell",
"biology",
"nucleic",
"acids",
"b",
"cells",
"protein",
"translation",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"non-coding",
"rna"
]
| 2017 | Differential Sensitivity of Target Genes to Translational Repression by miR-17~92 |
Chromatin regulation underlies a variety of DNA metabolism processes , including transcription , recombination , repair , and replication . To perform a quantitative genetic analysis of chromatin accessibility , we obtained open chromatin profiles across 96 genetically different yeast strains by FAIRE ( formaldehyde-assisted isolation of regulatory elements ) assay followed by sequencing . While 5∼10% of open chromatin region ( OCRs ) were significantly affected by variations in their underlying DNA sequences , subtelomeric areas as well as gene-rich and gene-poor regions displayed high levels of sequence-independent variation . We performed quantitative trait loci ( QTL ) mapping using the FAIRE signal for each OCR as a quantitative trait . While individual OCRs were associated with a handful of specific genetic markers , gene expression levels were associated with many regulatory loci . We found multi-target trans-loci responsible for a very large number of OCRs , which seemed to reflect the widespread influence of certain chromatin regulators . Such regulatory hotspots were enriched for known regulatory functions , such as recombinational DNA repair , telomere replication , and general transcription control . The OCRs associated with these multi-target trans-loci coincided with recombination hotspots , telomeres , and gene-rich regions according to the function of the associated regulators . Our findings provide a global quantitative picture of the genetic architecture of chromatin regulation .
The genetic basis of gene expression has been studied in various organisms [1]–[5] . For example , two different strains of Saccharomyces cerevisiae ( BY and RM ) were crossed to produce a number of different genetic recombinants , and their expression levels and genotypes were analyzed [1] , [6] . We previously utilized this system to separate the cis- and trans-components of variation in gene expression [7] . Tirosh et al . [8] profiled nucleosome patterns in the inter-specific hybrids of two yeast species to dissect cis- and trans-effects on nucleosome positioning . Recently , variations in the binding patterns of transcription factors ( TFs ) have begun to be studied [9]–[11] . Chromatin structure controls the access of a wide spectrum of DNA binding proteins involved in not only transcription but also DNA repair , recombination , and replication . Therefore , open chromatin areas can indicate DNA regions accessible to such regulators and thus have been used to identify regulatory regions or elements in the genome . In addition to the well-known DNaseI hypersensitivity assay , the FAIRE technique has been used to capture open chromatin sites in the genome with the aid of massively parallel sequencing ( FAIRE-seq ) [12]–[14] . In a recent study , the FAIRE DNA was analyzed by genotyping arrays to identify functional regulatory polymorphisms [15] . FAIRE-seq , however , is capable of providing a quantitative measure of chromatin accessibility along with sequence polymorphisms so that the direct effects of DNA sequences on chromatin accessibility can be examined . For example , it has been shown that SNPs located within open chromatin can influence chromatin accessibility , thus demonstrating that chromatin structure can be a heritable feature [11] . As chromatin is a genetically regulated material , a genetic association approach could be used to understand the genetic architecture of chromatin regulation by examining open chromatin in multiple genetically different individuals . A recent study [16] used this approach for chromatin accessibility across 70 human individuals . Because of the large size of the human genome , open chromatin sites were analyzed only in association with local genetic markers to identify cis-associations . Transcription factor binding was shown to be one of the main mechanisms by which DNA polymorphisms affect chromatin structure . In this work , we took advantage of the compact size and comprehensive annotation of the yeast genome to dissect the entire genetic architecture of chromatin regulation , including both cis- and trans-associations , to better interpret the functional association of trans-acting factors . To this end , we generated open chromatin maps of 100 yeast samples , including the parental strains ( BY and RM , two replicates of each ) and their descendants [6] by means of the FAIRE-seq technique .
Open chromatin peaks were first identified for each sample . We then obtained a total of 7 , 527 OCRs by combining the peak signals of the 96 genetically different yeast strains . For each OCR , the density of the corresponding peak in each strain was calculated and normalized across the strains . The normalized peak density measures showed high reproducibility ( R = 0 . 95∼0 . 99 ) between the replicates from different FAIRE batches and sequencing libraries ( Figure S1 ) . More than half of the OCRs were located at promoters , and 18 . 6% and 16 . 4% of the peaks fell near transcription termination sites and within ORFs , respectively ( Figure S2 ) . The OCRs mostly coincided with nucleosome-free regions at promoters or transcription termination sites ( Figure S3 ) . Approximately 57% of yeast genes contained an OCR at their promoter , and 40% of replication origins overlapped with 14 . 3% of the OCRs ( Figure S2 ) . The average size of the OCRs in BY and RM was 159 bp , while the average size of the OCRs combined across all the strains was 236 bp ( Figure S4 ) . We sought to estimate the direct influence of underlying DNA sequences on chromatin configuration by quantitatively comparing sequence-dependent ( cis ) variation and sequence-independent ( trans ) variation in chromatin accessibility . Cis-variation indicates variation in chromatin accessibility among individuals in which the DNA sequences of the given open-chromatin locus are different , while trans-variation indicates variation in chromatin accessibility among individuals with an identical genotype at the given locus . To measure cis-variation as the magnitude of chromatin variation caused primarily by cis-acting elements residing directly beneath open chromatin , we sought to determine the genotype of each OCR based on the SNP profiles generated from our sequence data . This enabled the classification of OCRs into either BY or RM groups according to each strain's inheritance of the locus ( Figure 1A ) . The cis-variation of each OCR was defined as the variance of peak density among the strains with the same genotype at that OCR . The two cis-variation measures ( each from the BY and RM group ) were highly consistent ( Figure 1B ) . Approximately 23% ( 1 , 738 OCRs ) had more than ten individuals in each group . We assessed the statistical significance of trans-variation by considering the within-group variance ( cis-variation ) : 11 . 8% ( P<0 . 05 ) or 4 . 8% ( P<0 . 01 ) of the 1 , 738 OCRs were called significant ( Figure 1C ) . QTL mapping was performed by interrogating the 7 , 527 OCRs against the genetic markers selected and processed based on the previous genotype data [6] ( see Materials and Methods ) . A total of 11 , 048 associations were identified at a false discovery rate ( FDR ) of 0 . 01 by our chromatin QTL mapping . Approximately 7 . 9% of the associations involved cis-acting loci within 100 kb ( 12 . 66% within 1 Mb ) , whereas the majority of chromatin traits were linked to trans-regulatory loci . The OCRs associated in trans tended to display a higher trans-variation ( P<2×10−16 ) , while those associated in cis had a higher cis-variation ( P = 1 . 1×10−4 ) , indicating consistency between sequence-based genotyping and microarray-based genotyping . We employed the gene expression data for the 96 strains [6] and carried out expression QTL mapping by repeating the procedures used for the chromatin QTL mapping ( see Materials and Methods ) . At an FDR of 0 . 01 , 12 , 317 associations between genotypes and expression levels were identified . We identified a total of 2 , 234 OCRs in which there was a TF-binding motif that contained a polymorphism and found that these OCRs were twice as likely to be associated in cis than other OCRs ( P = 4 . 6×10−7 ) . However , there was no difference with respect to trans-association . This implies that the effect of DNA sequence variation on chromatin structure is often manifested through underlying TF-binding motifs independently of trans-acting regulators . To determine whether cis-associations can also be explained by differential nucleosome formation , we searched for cis-QTL SNPs in the well-known poly A/T tract nucleosome depletion signature . We extracted the reference genome sequences surrounding the SNP locations within the OCRs from our FAIRE-seq data and then looked for the presence of a poly A/T tract . Even with a very loose threshold ( five consecutive A/Ts ) , we could only identify five such instances . This is contradictory to the major role of the AT-rich sequences in the divergence of nucleosome positioning between different species [8] . We propose that poly A/T tracts residing in open chromatin may be under strong selective pressure and thus resistant to sequence changes because of their importance in regulatory function . Because the cis-associations between DNA sequences and chromatin accessibility are likely to be mediated by TF binding , a sequence polymorphism that affects chromatin accessibility in cis should also affect gene expression in the neighborhood . Indeed , a sizeable fraction ( 45% ) of the chromatin-associated SNPs were associated with the expression of nearby genes . By contrast , only 15% of the expression-associated SNPs turned out to influence the accessibility of nearby chromatin , indicating that there are mechanisms by which sequence polymorphisms can affect the expression of nearby genes without affecting chromatin accessibility . Reciprocal regulation of two chromatin loci by DNA sequences could be observed in OCR #464 and OCR #465 . These two OCRs were associated with multiple cis-markers encompassing 100 kb upstream to 15 kb downstream of the loci . Sequence analysis detected two underlying SNPs that were associated with the peak density of OCR #464 ( Figure 1A and 1D ) . Interestingly , the density of the adjacent peak ( OCR #465 ) was negatively correlated with that of OCR #464 across the strains ( Figure 1E ) , demonstrating a reciprocal regulation of the two chromatin loci . In line with our sequencing-based genotypes , all the cis-markers indicated that the RM genotype increases the peak density of OCR #464 . The sum of trans-variation in the trans-associated OCRs was divided by the sum of trans-variation across all the OCRs , revealing that 45 . 2% of the total trans-variation across the OCRs could be explained by genetic factors . To examine how much of the trans-variation of each OCR is explained by trans-acting genetic factors , we computed the explanatory power of the linear regression ( R2 ) for each OCR and its associated trans-loci . The average R2 of the trans-associated OCRs was 33% . Enrichment of high trans-variation OCRs was observed in the vicinity of telomeres ( Figure 2A and green marks in Figure S5 ) . This pattern was not observed for cis-variation ( Figure 2A and Figure S5 ) . High trans-variation OCRs also coincided with gene-rich regions ( Figure 2B and blue ticks in Figure S5 ) and gene-poor regions ( Figure 2B and light-blue ticks in Figure S5 ) . Approximately 50% of chromatin QTLs were gene expression QTLs and vice versa , indicating that the trans-associations we identified are technically robust and biologically meaningful . However , only 17 . 6% of these dual QTLs were associated with chromatin and expression traits at the same locus . In other words , many of the dual QTLs were responsible for chromatin traits and gene expression traits that are distantly located ( e . g . , in different chromosomes ) . It is possible that regulatory SNPs affect chromatin accessibility for DNA regulation other than transcription ( e . g . , DNA repair , recombination , etc . ) , which in turn leads to secondary gene expression changes , and that regulatory loci affect the expression of downstream regulators in trans , which in turn causes secondary changes in the accessibility of the target chromatin regions . We examined the number of trans-linkages for each OCR . Most OCRs were responsive to a small number of regulatory loci . Only a few ( 6 . 8% ) had more than five linkages with the average number being three times lower than for gene expression traits ( 2 . 1 versus 5 . 9 ) ( Figure 3A ) . This implies that chromatin traits are rather specifically governed by a handful of trans-regulators , whereas gene expression processes are responsive to more regulatory inputs . An opposite trend was observed for regulatory loci ( Figure 3B ) . There were regulatory loci responsible for an extremely large number of chromatin traits , with a few cases in which >200 OCRs were linked to a single promiscuous chromatin QTL ( Figure 3B ) . The horizontal dots observed in the chromatin association map ( Figure 3C ) illustrate ‘extensive’ regulation by chromatin regulatory loci ( Figure 3D ) , as opposed to the ‘intensive’ regulation of gene expression traits ( Figure 3E ) . To investigate the multi-target chromatin regulatory loci , or hotspot QTLs , we first selected those with >65 trans-associated OCRs . We annotated each locus by searching for known DNA or chromatin regulators flanking the marker within 10 kb [17] and merged the adjacent markers covering the same regulator . A total of 32 initial hotspot loci were merged into 17 hotspots , 14 of which flanked at least one known regulator ( master regulators listed in Figure 3C ) . The annotated ( regulator-containing ) loci tended to influence more chromatin traits than the unannotated loci ( P = 5×10−4 ) ( Figure 3B ) . By contrast , no enrichment of known regulators near multi-target expression regulatory loci was observed ( Figure 3B ) . Among the master regulators ( Figure 3C ) were three TFs with sequence-specific DNA binding activity: DAL82 , TEC1 , and NRG2 . Position weight matrices were available for the DNA-binding motif of Dal82p and Tec1p . Remarkably , 62% of the 71 DAL82-associated OCRs contained the Dal82p-binding motif . However , no Tec1p-binding motif enrichment was observed in the associated OCRs . The influence of Tec1p might be exerted not through direct binding but via interaction with other factors under normal growth conditions . Data for Nrg2p binding sites are not available . SET2 and MED2 are involved in the transcription of many genes in a non-sequence-specific manner . Set2p is a histone methyltransferase that plays a role in general transcription elongation , and Med2p is a subunit of the mediator complex that forms the RNA polymerase II holoenzyme . Their target OCRs were identified in gene-rich regions ( Figure 4A and Figure S6 ) . Rdh54p is a Swi2/Snf2-like factor that plays a role in recombinational repair of DNA double-strand breaks ( DSBs ) during mitosis and meiosis by interacting with Rad51p and Rad54p [18]–[20] . DSBs occurring at recombination hotspots in yeast are found near open chromatin [21] . We employed a measure of “recombination hotness” that was globally obtained based on DSB distribution [22] . The RDH54 OCRs showed the highest recombination hotness among the master regulators ( Figure 4B ) , with a P value of 9×10−25 ( Figure S7 ) , and tended to fall near the recombination hotspots ( Figure 4C ) . Cdc13p is a multi-functional telomere-binding protein that participates in telomere replication and maintenance especially by mediating telomerase access to telomeric chromatin [23]–[25] . Among the hotspot loci , the CDC13 locus had the largest number of associated OCRs in close proximity to telomeres ( seven OCRs within 1 kb from telomeres ) . The enrichment of CDC13-associated OCRs near telomeres is shown in Figure 4D . Telomeres are associated with recombination coldspots [22] . Indeed , the recombination hotness of the CDC13 OCRs was very low ( Figure 4B ) .
In this work , we sought to dissect the genetic architecture of chromatin regulation . The multi-target regulatory structure reflects the wide-ranging nature of certain chromatin regulators . Surprisingly , however , many chromatin QTLs were found to govern only a few target traits . It is conceivable that the chromatin structures at particular loci are not susceptible to genetic perturbations or that the technical limitations of our method for detecting subtle changes in chromatin traits may prevent the identification of weakly associated targets . In this case , there may be numerous potential regulatory targets that have not passed our statistical threshold . On the other hand , the chromatin traits that were responsive to certain genetic perturbations had only a few regulatory inputs , in contrast to the high responsiveness of gene expression traits to multiple regulatory signals . Therefore , chromatin states alone may not be sufficient to explain the precise level of transcription . Once upstream regulators set the stage by priming the chromatin structure , various downstream regulatory inputs may add additional layers of complexity to gene expression control . This is also reflected in the lack of common targets between chromatin QTLs and expression QTLs . Only 18% of the dual QTLs ( i . e . , SNPs that are both chromatin QTLs and expression QTLs ) were associated with chromatin accessibility and gene expression at the same locus simultaneously . However , the identification of many dual QTLs was encouraging itself because it suggests that the detected QTLs are likely to contain functional regulators . We successfully annotated chromatin QTLs , particularly those responsible for a large number of target chromatin traits . The identification of functionally relevant trans-regulators from expression QTL mapping has been reported to be difficult [26] . Sequence-specific TF binding appears to be very important in cis-associations . We observed an enrichment of cis-associations for TF-motif-containing OCRs and common QTLs linking chromatin accessibility and nearby gene expression . This is consistent with the finding that human SNPs associated with chromatin in cis are frequently found in TF-binding sites [16] . Moreover , consistency in allele frequencies were observed between the sequence reads for open chromatin and those for TF binding . In contrast to the previous study [16] in which only cis-regulation was thoroughly examined , here we took advantage of the compact size and comprehensive annotation of the yeast genome to dissect the architecture of trans-regulatory mechanisms as well . In conclusion , our work provides insight into the genetic basis of chromatin regulation and its relationship with transcription control . Genetic variation in open chromatin in the human genome can underlie disease phenotypes , and thus , the current study has medical implications . For example , previous studies [13] , [15] identified regulatory polymorphisms in open chromatin that were previously linked through genome-wide association studies with diabetes and HDL cholesterol levels .
We obtained the BY-RM cross strains from the original authors [1] , [6] . FAIRE experiments were performed based on the published protocol [12] . We selected 94 yeast segregants and subjected them and the BY and RM strains to 100-bp sequencing on Illumina HiSeq2000 . To identify the FAIRE-seq read peaks , we ran F-Seq [27] as previously suggested for FAIRE-seq data analysis [13] . Small-sized peaks ( <15 bp ) were extended in both directions such that all the peaks were at least15 bp long . To identify all possible OCRs , we combined the extended peaks of the 96 yeast strains ( exclusive of the replicates ) and merged overlapping peaks into a single peak using BEDTools [28] , resulting in 7 , 527 unique OCRs . The number of FAIRE-seq reads that mapped uniquely to each OCR was counted in each yeast strain . The read count of each OCR was normalized by taking into account the size of the peak and the total number of tags produced from each FAIRE library as . After the log2 transformation , the negative values were set to zero ( ceiling ) . This normalization scheme was used in our previous work [29] . We further normalized the final matrix of the 7 , 527 OCRs and 96 strains by scaling the 96 sample vectors to zero mean and unit variance . To assess reproducibility , the FAIRE-seq reads of the parental replicates were mapped to the predefined OCRs and the same normalization scheme was repeated for the four independent samples . SNPs were detected from the FAIRE-seq reads using the Illumina's CASAVA suite . SNP calls with fewer than five reads were discarded . For heterogeneous calls , only the major polymorphism with a certain frequency ( >80% ) was taken . The genotype of each OCR was determined based on its SNP profile . The OCR in the given strain was considered to have inherited the BY ( or RM ) allele if its genotype perfectly matched with the genotype of the OCR in the BY ( or RM ) strain . For genotyping at a less stringent threshold , the OCRs whose SNP profile matched with either the BY or RM profile for >50% of the SNPs were also classified as BY or RM . To compute trans-variation , the standard deviations of the normalized peak density measures within the BY and RM groups was measured . We identified a total of 1 , 738 OCRs for which at least ten individuals inherited either a BY or RM allele; we then re-grouped the yeast strains according to the genotype of the given OCR . To assess the statistical significance of cis-variation , we used the two-sample t test to measure the difference in the means of the BY and RM groups . The genetic markers from the original study [6] were used for QTL mapping . As suggested by Lee et al . [17] , adjacent markers with no more than two genotypic mismatches across the 96 strains were merged into one average genotype profile , resulting in 1 , 533 markers . As suggested previously [17] , we identified the genes located within 10 kb upstream or downstream of the genomic region covered by the merged genetic marker . To identify potential regulators , we used Gene Ontology to identify 495 genes involved in “DNA binding” , and 508 genes known to be involved in transcription and chromatin regulation , resulting in a total of 752 unique genes . For QTL mapping , we measured associations by means of the correlation coefficient or the linear regression between the genotypes represented as a categorical variable ( 0: RM , 0 . 5: missing , 1: BY ) and the chromatin traits represented as the normalized peak-density measure . False discovery rates ( FDRs ) were computed based on the permutation test , as follows . The matrix of peak density was shuffled by resampling the sample vectors ( yeast strains ) to generate randomized matrices , . The P value was determined by comparing the observed association with the expected associations from the permuted data as , where is an interpretation function . was used . The P values were adjusted for multiple testing to yield FDRs , as suggested by Benjamini and Hochberg [30] . An FDR of 0 . 01 was used . A distance of 100 kb between the marker and the trait was used to differentiate cis- and trans-associations . We employed the gene expression data for the 96 strains [6] and performed expression QTL mapping by repeating the same procedures . A total of 11 , 048 marker-trait associations involving 3 , 522 OCRs were identified at an FDR of 0 . 01 when the correlation coefficient was used . To evaluate the consistency between the FDR-based non-parametric approach and the parametric method , we obtained a P value for each marker-trait pair based on the linear regression . At parametric P values<10−3 and <10−5 , 91 . 1% and 81 . 1% of the identified associations were called significant , respectively . Adjacent genetic markers ( <10 kb ) associated with a common trait in QTL mapping were combined . Trans-loci were examined to determine whether the corresponding genetic marker covered at least one of the 752 regulators . According to this criterion , all trans-loci were classified into annotated loci or unannotated loci . We defined hotspot chromatin loci as having more than 65 genetic linkages . Adjacent genetic markers covering the same regulator were manually merged . To calculate the density of the genes surrounding each OCR , the number of genes located within 50 kb upstream and 50 kb downstream of the OCR peak boundaries was determined . This number was divided by the size of the peak for normalization . The microarray data for the recombination hotspots of the yeast genome [22] were downloaded from http://derisilab . ucsf . edu/hotspots/ . The cy5/cy3 ratios from seven ORF arrays were averaged and log2 transformed . The positions of replication origins were downloaded from http://cerevisiae . oridb . org . For TF motif analysis , we used position weight matrices [31] based on in vivo binding assays by chromatin immunoprecipitation for 203 yeast TFs [32] and another set of position weight matrices based on systematic in vitro assays of 112 yeast TFs [33] . TF motifs occurring in OCRs were identified by means of the HOMER package [34] using the two position weight matrix sets . The FAIRE-seq data for the 96 yeast strains are available at the GEO database with accession number GSE33466 . The following link has been created to allow review of the record GSE33466: http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? token=zvyznqwickewmto&acc=GSE33466 | Quantitative trait loci ( QTL ) mapping is a genetic approach that allows the identification of genetic factors underlying a phenotype of interest . Genomic technologies such as DNA microarray and next-generation sequencing provide data that can be used for the analysis of multiple molecular phenotypes . For example , the expression levels of thousands of genes can be associated with subject-specific genome-wide genetic information in expression QTL mapping . Similarly , the genetic regulation of transcription factor binding or epigenetic mechanisms such as DNA methylation or chromatin structure has begun to be investigated . In particular , the mechanisms controlling chromatin accessibility have attracted special interest due to their importance in a variety of DNA regulation processes including recombination , repair , replication , and transcription . In this work , we sought to dissect the genetic architecture of chromatin accessibility regulation by harnessing the power of genetic and genomic techniques . By analyzing open ( accessible ) chromatin maps of multiple yeast individuals in association with their genetic backgrounds , we were able to characterize the regulatory structure of chromatin traits versus that of gene expression . Importantly , we observed that the genetic loci responsible for multiple open chromatin regions were enriched for known regulatory factors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"systems",
"biology",
"genome",
"expression",
"analysis",
"genomics",
"genetic",
"polymorphism",
"gene",
"expression",
"genetics",
"epigenetics",
"biology",
"population",
"genetics",
"chromatin",
"genetics",
"and",
"genomics"
]
| 2013 | Genetic Landscape of Open Chromatin in Yeast |
The G protein-coupled receptor ( GPCR ) Smoothened ( Smo ) is the requisite signal transducer of the evolutionarily conserved Hedgehog ( Hh ) pathway . Although aspects of Smo signaling are conserved from Drosophila to vertebrates , significant differences have evolved . These include changes in its active sub-cellular localization , and the ability of vertebrate Smo to induce distinct G protein-dependent and independent signals in response to ligand . Whereas the canonical Smo signal to Gli transcriptional effectors occurs in a G protein-independent manner , its non-canonical signal employs Gαi . Whether vertebrate Smo can selectively bias its signal between these routes is not yet known . N-linked glycosylation is a post-translational modification that can influence GPCR trafficking , ligand responsiveness and signal output . Smo proteins in Drosophila and vertebrate systems harbor N-linked glycans , but their role in Smo signaling has not been established . Herein , we present a comprehensive analysis of Drosophila and murine Smo glycosylation that supports a functional divergence in the contribution of N-linked glycans to signaling . Of the seven predicted glycan acceptor sites in Drosophila Smo , one is essential . Loss of N-glycosylation at this site disrupted Smo trafficking and attenuated its signaling capability . In stark contrast , we found that all four predicted N-glycosylation sites on murine Smo were dispensable for proper trafficking , agonist binding and canonical signal induction . However , the under-glycosylated protein was compromised in its ability to induce a non-canonical signal through Gαi , providing for the first time evidence that Smo can bias its signal and that a post-translational modification can impact this process . As such , we postulate a profound shift in N-glycan function from affecting Smo ER exit in flies to influencing its signal output in mice .
The Hedgehog ( Hh ) signaling pathway contributes to developmental patterning and adult tissue homeostasis , and when corrupted , can contribute to tumor initiation and maintenance [1–3] . Originally identified in Drosophila , the Smoothened ( Smo ) protein is an evolutionarily conserved G protein-coupled receptor ( GPCR ) that functions as the signal transducer of the Hh cascade [4–6] . Its activity is indirectly regulated by the Hh ligand through the receptor Patched ( Ptc ) , which in the absence of Hh , blocks Smo signaling by inhibiting its accumulation on the plasma membrane in Drosophila or in the primary cilium in vertebrates [7 , 8] . As such , in the absence of Hh , Smo is localized predominantly to recycling endosomes where it does not signal [9–11] . The mechanism by which Ptc controls Smo localization and signaling has not yet been defined , however a predominant model postulates that Ptc governs the availability of an unknown Smo ligand [12 , 13] . In vertebrate systems this ligand is likely to be a sterol-like compound because 20 ( S ) -hydroxy cholesterol ( 20 ( S ) -OHC ) and the steroidal alkaloid cyclopamine are both modulators of Smo activity [14 , 15] . Naturally occurring compounds modulating insect Smo activity have yet to be discovered . However , it is well established that Hh controls pathway activation in both vertebrates and invertebrates by binding to Ptc and its associated co-receptors [12 , 16–19] . This binding induces a Ptc conformation shift that triggers its internalization and lysosomal degradation , thereby allowing Smo to translocate to its active signaling location [9 , 20] . From there , Smo communicates with its downstream effectors to induce intracellular signaling . In flies and canonical vertebrate signaling , this culminates in activation Hh target gene expression through the Gli/Ci family of transcription factors [11 , 21 , 22] . Although both Drosophila and vertebrate Smo proteins are capable of activating Gαi heterotrimeric G proteins in response to Hh , the role of Gαi in Smo signaling has evolved [6 , 23] . In vertebrates , Smo-mediated Gαi activation appears to be dedicated to induction of a distinct , non-canonical Hh signal that can alter intracellular Ca2+ levels to modulate phospholipase C activity or induce RhoA and Rac to govern cell migration [23–25] . This suggests that , in some contexts , vertebrate Smo might display biased signaling whereby one effector route is favored over the other [26 , 27] . Although it is not yet possible to predict how signal bias is controlled for a given GPCR , it is possible that post-translational receptor modifications affecting ligand sensitivity and responsiveness may contribute [28 , 29] . A common post-translational modification occurring on GPCRs is N-linked glycosylation , which can affect a multitude of processes including receptor folding , trafficking , stability , cell-surface localization , ligand binding and ligand responsiveness [30–34] . Although N-linked glycosylation occurs in both insect and vertebrate systems , clear differences exist in oligosaccharide processing and glycan complexity [35] . In this report , we interrogate N-linked glycosylation of Smo proteins in fly and murine systems to determine whether differential glycosylation patterns between the two have distinct effects on signaling activity . We describe a clear evolutionary divergence in the role of N-glycans for Smo activity and postulate that with the emergence of non-canonical Smo signaling in vertebrates , the role of glycosylation in its activity evolved from assisting in protein folding and ER exit in flies to a novel role of influencing signal output in vertebrates .
To identify the predicted N-linked glycosylation sites in Smo that are conserved across phyla , Smo protein sequences from human , mouse , rat , chicken , zebrafish and fly were analyzed using NetNGlyc prediction software ( Fig 1A ) . This identified four high confidence motifs ( NxS/T ) , one intracellular and three extracellular , that are conserved across the vertebrate Smo proteins ( Fig 1A , N1-N4 ) . These are localized to the extracellular amino-terminal region flanking the cysteine rich domain ( CRD ) , in extracellular loop EC3 and in intracellular loop IC3 ( Fig 1A and S1B Fig ) . The predicted N-glycosylation pattern differs significantly for Drosophila Smo ( dSmo ) , which contains seven predicted consensus sites: five within its amino-terminal CRD , one in EC1 and one in EC2 ( Fig 1A and S1A Fig ) . Multiple sequence alignment of all sequenced Drosophila Smo proteins using Clustal W software revealed that of these seven predicted sites , five are conserved across Drosophila species: N95 , N184 , N195 , N213 and N336 ( S1C Fig ) . To determine whether the identified motifs harbored N-glycosylation moieties , compound Smo mutants harboring N to Q alterations of the conserved acceptor sites in dSmo ( N95Q , N184Q , N195Q , N213Q and N336Q , hereafter referred to as dSmoNQ5 ) and mouse Smo ( N38Q , N192Q , N450Q and N497Q , mSmoNQ4 ) were generated and tested for EndoH and PNGase sensitivity ( Fig 1B and 1C ) . PNGase cleaves all N-linked glycans including mannose rich , hybrid and complex oligosaccharide species . EndoH recognizes and cleaves N-linked mannose rich oligosaccharides that are present on ER-resident proteins , but does not cleave the highly processed complex oligosaccharides that are added in post-ER compartments . Cell lysates prepared from Drosophila Clone 8 ( Cl8 ) cells expressing Hh with wild type or NQ5 dSmo proteins were incubated with the indicated deglycosylating enzymes . Wild type dSmo was sensitive to both of these deglycosylating enzymes , revealing an EndoH-resistant post-ER pool ( Fig 1B , arrowhead ) and an EndoH-sensitive ER pool ( arrow ) . The residual shift evident following PNGase treatment is likely due to Hh-induced phosphorylation ( lane 3 and [36] ) . dSmoNQ5 migrated similarly to the fully deglycosylated , PNGase treated species under all conditions ( Fig 1B lanes 4–6 compared to 3 ) , supporting that the mutant is N-glycan deficient . Consistent with the established role of N-linked glycosylation in protein turnover [37] , dSmoNQ5 protein levels appeared higher than wild type ( Fig 1B , compare lanes 1 and 4 ) . Half-life analysis confirmed that NQ5 was indeed stabilized , demonstrating a half-life of ~4 hours compared to ~1 hour for wild type ( S1D Fig ) . For mSmo studies , Smo-/- cells were stably transfected with vector control , wild type or NQ4 mSmo expression plasmids in order to minimize transfection and expression variability across experiments . To minimize over-expression effects , stable lines showing the lowest equivalent levels of wild type and NQ4 mSmo proteins were selected ( Fig 1C ) . Deglycosylating assays were performed on lysates prepared from Smo-/- cells stably transfected with plasmids encoding wild type or NQ4 mutant mSmo ( Fig 1C ) . mSmoNQ4 demonstrated a faster migration pattern than wild type , consistent with it having decreased glycosylation ( Fig 1C compare lanes 2 and 5 ) . Whereas wild type mSmo possessed both simple N-linked glycans ( left arrow ) and complex post-ER N-linked glycans ( left arrowhead ) , the NQ4 mutant did not ( Fig 1C compare lanes 2–4 with 5–7 ) . NQ4 demonstrated a banding pattern similar to PNGase-treated wild type Smo under its basal conditions , and did not collapse upon EndoH or PNGase treatments ( lane 4 compared to 5–7 ) . In addition to possessing N-linked glycosylation , mSmo also harbors O-linked glycan modifications [36] . Accordingly , the upper band evident for mSmoNQ4 ( Fig 1C , lane 7 right arrowhead ) was unaffected by lambda phosphatase treatment , but collapsed upon O-glycosidase/neuraminidase treatment ( Fig 1C’ ) . Thus , stripping mSmoNQ4 of its N-glycans does not impact O-linked glycan addition . To narrow down which of the predicted N-glycan acceptor sites in the two proteins harbor modifications , single N to Q mutations were generated for dSmo and mSmo proteins , and their migration assessed by SDS-PAGE ( Fig 1D and 1E ) . Each of the dSmo mutants exhibited a modest migration shift , consistent with each of the predicted acceptor sites harboring glycan modifications ( Fig 1D , lane 2 compared to 3–6 ) . For mSmo , 3 of the 4 predicted sites displayed altered migration ( Fig 1E ) . Consistent with N-glycosylation machinery in the ER lumen not having access to the cytoplasmic domains of mSmo , the predicted site in IC3 ( N3 ) did not shift upon N to Q mutation ( compare lanes 2 and 5 ) . Conversely , each of the three remaining sites demonstrated size collapse and faster gel migration upon N to Q mutation ( Fig 1E compare lane 2 with 3 , 4 and 6 ) . Having established that both fly and mouse Smo proteins harbor N-linked glycans , we next wanted to determine the contribution of glycans to Smo function . To interrogate the fly protein , the ability of dSmoNQ5 to rescue Smo-dependent activation of a Hh reporter construct ( ptcΔ136-luciferase ) in Cl8 cells following knockdown of endogenous smo was tested ( Fig 2A ) . Whereas cells transfected with control dsRNA facilitated robust Hh-induced reporter gene activation , cells transfected with smo 5’UTR dsRNA did not . Expression of a wild type smo cDNA lacking UTR sequence into the knockdown background fully rescued Hh-mediated reporter induction in this assay . dSmoNQ5 failed to rescue the Hh response , suggesting that loss of glycosylation renders dSmo incapable of signaling to induce Hh target gene expression . In response to Hh stimulation , dSmo translocates from intracellular endosomes to the plasma membrane where it signals to its downstream effectors [9 , 11] . To determine whether dSmoNQ5 could traffic to an active location , Myc epitope-tagged WT or NQ5 dSmo proteins were co-expressed with the Calreticulin-EGFP-KDEL ER marker ( GFP-ER ) in Cl8 cells in the absence and presence of Hh ( Fig 2B and [10 , 38] ) . In the absence of Hh , wild type dSmo localized to intracellular puncta that did not overlap with the GFP-ER marker and transitioned to the plasma membrane in the presence of Hh ( Fig 2B , upper panels ) . dSmoNQ5 demonstrated a different localization pattern , overlapping almost completely with the ER marker in the absence and presence of Hh ( Fig 2B , lower panels ) . Co-localization of GFP-ER with V5 epitope tagged ER resident proteins dBiP , dCnx and dCrc confirmed specificity of the ER marker protein ( Fig 2B’ ) . ER retention suggests that dSmoNQ5 might experience protein folding defects , an outcome that is consistent with the established role of glycosylation in protein folding in the ER [37] . Hh signaling plays an essential role in patterning vein structure and size of the adult Drosophila wing [39] . To assess whether glycosylation-deficient dSmo would function in vivo , the ability of dSmoNQ5 to rescue the phenotype induced by expression of dsRNA targeting the 3’ UTR of the endogenous smo gene was tested ( smo3’UTR , [40] ) . Dicer was co-expressed with the UTR dsRNA transgene to enhance dsRNA processing to yield a stronger phenotype [40] . When expressed with UAS-dicer under control of the nubbin-GAL4 driver , UAS-smo3’UTR triggered disruption of vein patterning in the central region of the wing , indicative of attenuated Hh signaling ( Fig 2C and 2E ) . Co-expression of a transgene encoding EGFP did not trigger a wing phenotype on its own and could not rescue the smo knockdown vein phenotype , thereby confirming that introduction of a third transgene into the nub>Dcr; smo3’UTRdsRNA background does not impact GAL4 activity ( Fig 2F ) . Conversely , expression of wild type dSmo fully rescued the phenotype , and triggered modest ectopic vein formation anterior to longitudinal vein 3 ( Fig 2G ) . Like EGFP , Co-expression of dSmoNQ5 in the dsRNA background failed to rescue the patterning defect ( Fig 2H ) . To confirm that failure to rescue was not the result of lower expression of the NQ5 protein , lysates were prepared from wing imaginal discs and Smo protein was analyzed by western blot . dSmo levels were similar between wild type and NQ5-expressing flies ( Fig 2I , compare lanes 2–3 ) . Thus , N-linked glycans are required for dSmo function in vivo . To identify the essential sites of N-linked glycosylation contributing to dSmo ER exit and signaling , rescue reporter assays and sub-cellular localization analyses were performed for the single site N to Q mutants ( Fig 3 ) . Each of the single glycosylation site mutants was able to rescue reporter gene induction in the smo knockdown background similarly to wild type , with the exception of dSmoN336Q . This mutant demonstrated a ~50% reduction in activity , despite it expressing at higher protein levels ( Figs 1D and 3A ) . Furthermore , dSmoN336Q was the only single glycan mutant that failed to accumulate on the plasma membrane in response to Hh , and instead , accumulated predominantly in the ER ( Fig 3B and 3C ) . Enrichment of N336Q in the ER is also supported by western blot; wild type and N213Q , which reach the plasma membrane , were equally distributed between ER ( black arrowhead ) and post-ER fractions ( white arrowhead ) . N336Q enriched in the EndoH sensitive ER-resident fraction ( Fig 3D , lane 8 , black arrowhead ) . The dSmoN336Q signaling defect was less pronounced than the signal loss observed for dSmoNQ5 ( Fig 2A compared to Fig 3A ) , suggesting that a glycan modification on another residue may partially compensate for N336 glycan loss . To determine whether we could exacerbate the signaling defect observed for N336Q and identify the compensatory glycosylated residue , we generated double mutants harboring N336Q in combination with N to Q alterations for each of the additional conserved sites ( N95 , N184 , N195 and N213 , Fig 4A ) . Although altered migration of single N to Q mutations was modest ( Fig 1D ) , each of the double mutants migrated more quickly through SDS-PAGE gels than wild type dSmo , but not as rapidly as dSmoNQ4 ( N95Q , N184Q , N195Q , N213Q ) or dSmoNQ5 ( Fig 4A ) . Functional assays revealed that N95Q , N336Q , N184Q , N336Q and N195Q , N336Q induced an attenuated signal to downstream effectors that was similar to the dSmoN336Q response ( Fig 4B ) . Conversely , dSmoN213Q , N336Q failed to effectively rescue reporter gene induction , demonstrating a significantly reduced activity level that was closer to that of dSmoNQ5 ( Fig 4B ) . Like dSmoN336Q , each of the N336Q-containing double mutants failed to efficiently reach the cell surface in response to Hh , and instead enriched in the EndoH-sensitive ER fraction ( Fig 4C and 4D , arrowhead ) . N-linked glycans can impact disulfide bond formation during protein folding by recruiting ER chaperones and disulfide bond machinery [37 , 41] . It is therefore possible that ER retention of N-glycan dSmo mutants results from protein folding defects triggered by disulfide bond loss . To test whether disulfide bond formation was affected by N336Q or N213Q , N336Q mutation , lysates were prepared from cells expressing wild type or mutant dSmo proteins , and unpaired cysteines were detected by tagging free thiol groups with biotin-maleimide . Tagged proteins were captured on NeutrAvidin agarose and bound proteins were assessed by western blot and densitometry analysis [10] . Whereas dSmoWT was not effectively captured on NeutrAvadin agarose , both N336Q and N336Q , N213Q dSmo proteins were ( Fig 4E , input vs . bound lanes 2–4 ) . Averaged densitometry of two independent binding assays showed that N-glycan mutants were captured on the beads similarly to a dSmo protein harboring a C320A mutation , which disrupts the conserved disulfide bridge between the CRD and EC1 [10 , 42] ( Fig 4E’ ) . This suggests at least one disulfide bond is disrupted by loss of the sugar modification on N336 , which is consistent with the glycan mutant failing to fold properly . Accordingly , UAS-dSmoN213Q , N336Q was unable to rescue the smo3’UTR-induced phenotype in vivo ( Fig 4F compared to Fig 2F ) . This failure was not the result of diminished protein levels , as evidenced by the abundance of dSmoN213Q , N336Q in imaginal disc lysates ( Fig 2I , lane 3 ) . Taken together , these results suggest that N336 is likely the site of an essential glycan modification that contributes to dSmo folding and ER exit , and that the N-linked glycan on N213 can partially compensate for its loss . To determine whether the essential role of N-glycans in Smo ER exit was conserved in vertebrates , Smo-/- cells stably expressing wild type or NQ4 mutant protein were used to examine Smo cell surface and ciliary localization ( Fig 5A and 5B ) . To assess mSmoNQ4 plasma membrane localization , cell surface biotinylation was performed , and the ratio of surface-labeled to intracellular mSmo was determined by western blot and densitometry analysis . The ratios for WT and NQ4 mSmo were comparable , suggesting that N-glycosylation is not essential for mSmo trafficking to the plasma membrane ( Fig 5A and quantified in A’ ) . Upon agonist stimulation , mSmo enters the primary cilium as a requisite step in vertebrate pathway induction [7 , 43] . This trafficking capability was tested by treating mSmoWT or mSmoNQ4 stable cells with the Smo agonist SAG . Like the wild type protein , mSmoNQ4 enriched in the primary cilium in response to SAG treatment , indicating that the ligand-induced trafficking response is also intact ( Fig 5B and quantified in B’ ) . Smo functions as an obligate dimer that assembles into higher-order oligomers upon Hh stimulation [44 , 45] . To determine whether the NQ4 mutant existed in a conformation that was permissive for dimer formation , wild type and NQ4 mSmo proteins harboring YFP or V5 epitope tags were generated . Differentially tagged mSmo proteins were co-expressed in NIH3T3 cells , immunopurified from cell lysates using GFP antibody and immunoblotted using anti-V5 ( Fig 5C ) . mSmoNQ4-V5 immunoprecipitated with mSmoWT-YFP ( right panel , lanes 5 and 7 ) and mSmoNQ4-YFP ( right panel , lane 9 ) , indicating that loss of N-linked glycosylation does not impact mSmo dimerization . Taken together , these results suggest that , contrary to what was observed for the Drosophila protein , loss of glycosylation does not negatively impact mSmo folding or trafficking . Smo has been reported to possess at least two distinct ligand binding pockets [15 , 46 , 47] . These include the 7-TM bundle , which binds SAG and cyclopamine , and an allosteric site in the CRD , which binds 20 ( S ) -OHC [46 , 48 , 49] . Based upon the ability of SAG and cyclopamine to compete with each other for binding to Smo , their binding pockets are predicted to overlap [14 , 50] . As such , we used fluorescently labeled bodipy cyclopamine as a SAG surrogate to test functionality of the 7-TM binding pocket of mSmoNQ4 ( Fig 6A and 6C ) . Like wild type mSmo , mSmoNQ4 bound and concentrated bodipy cyclopamine in the primary cilium ( Fig 6A and 6B ) . Oncogenic SmoM2 signals independently of ligand and is refractory to cyclopamine inhibition [14] . Accordingly , mSmoM2 did not enrich bodipy cyclopamine in the primary cilium ( Fig 6C ) . This result confirms the specificity of ciliary enrichment of the compound through binding to wild type and NQ4 mSmo proteins . To determine whether mSmoNQ4 could bind the allosteric regulator 20 ( S ) -OHC through its CRD , we expressed wild type or NQ4 mSmo proteins in 293T cells , and tested their ability to be captured on 20 ( S ) -OHC conjugated sepharose beads ( Fig 6D ) . Like the wild type protein , mSmoNQ4 was captured on 20 ( S ) -OHC beads , indicating that the CRD ligand binding pocket is competent to associate with this small-molecule modulator ( Fig 6D , lanes 2 and 5 ) . Binding was largely restricted to the post-ER pool of mSmo , as the fully glycosylated form of wild type mSmo and the O-glycosylated form of NQ4 were preferentially enriched on 20 ( S ) -OHC beads ( Fig 6D , lanes 2 and 5 upper bands ) . Binding specificity was confirmed by competition with 50 μM 20 ( S ) -OHC , which fully competed both mSmo proteins from the sterol beads ( lanes 3 and 6 ) . Similar to what we observed with bodipy cyclopamine , neither mSmoM2 nor mSmoM2NQ4 could be captured on the sterol beads ( Fig 6D , lanes 8 and 11 ) . Taken together with the above results , these binding studies suggest that N-glycosylation is dispensable for mSmo folding , trafficking , dimerization and ligand binding . Smo has been reported to signal through G protein-dependent and -independent mechanisms to induce a range of cellular responses [23 , 51 , 52] . Although a subject of some debate , Smo is thought to signal primarily through G protein-independent mechanisms to induce target gene expression by Gli transcriptional effectors [23] . To determine whether signaling to Gli was compromised by loss of N-glycosylation , the ability of mSmoNQ4 to induce reporter gene activation in response to 20 ( S ) -OHC , SAG and Shh conditioned media was tested . The 8Xglibs-luciferase [53] reporter gene and TK-renilla control were transfected into mSmoWT , mSmoNQ4 and control stable lines , and reporter gene induction in response to Shh conditioned media , SAG and 20 ( S ) -OHC was measured ( Fig 6E ) . Both wild type and NQ4 mSmo cells induced reporter gene expression following exposure to Shh conditioned media , SAG or 20 ( S ) -OHC ( Fig 6E ) . Intriguingly , while both mSmo stable lines showed a similar Gli response following 20- ( S ) -OHC treatment , NQ4 appeared significantly more responsive to Shh and SAG than the wild type protein . Although we do not know the reason for the differential agonist responses by NQ4 , we do not think poor binding is the cause of the blunted 20 ( S ) -OHC response , as mSmoNQ4 can be captured on 20 ( S ) -OHC beads ( Fig 6D ) . Despite this , the robust Shh and SAG-induced responses suggest that NQ4 is highly sensitive for induction of canonical signaling to Gli . To assess whether NQ4 could induce the non-canonical signal through Gαi , we performed label-free dynamic mass redistribution ( DMR ) experiments using mSmoWT- or mSmoNQ4-expressing HEK293T cells . This cell line has previously been demonstrated to be competent for the Smo-Gαi signaling axis , and in our hands expresses WT and mutant mSmo proteins similarly ( Fig 7A and [54–56] ) . DMR employs a biosensor to track real time ligand-induced changes in cell biomass resulting from cell shape alteration and/or redistribution of intracellular material . Optimized protocols for monitoring GPCR activity by DMR have been established , and can be used to examine direct agonist and antagonist-induced responses through all classes of heterotrimeric G proteins ( Please see [54 , 57] for a comprehensive description ) . Because SAG induced a more robust reporter gene response by mSmoNQ4 , we chose to use SAG in DMR assays . Consistent with its capacity to activate Gαi [58] , mSmoWT induced a robust positive DMR response that was blunted by pretreating cells with Gαi-inactivating pertussis toxin ( PTX , Fig 7B , red vs . green ) . The SAG-induced mSmoWT DMR was not affected by pretreating with Gαs-targeting cholera toxin ( CTX , Fig 7B , blue ) , confirming that the response is specific to Gαi heterotrimeric G proteins . The positive SAG-induced DMR could also be attenuated by treatment with 2 μM cyclopamine ( Fig 7B’ , green ) , which is consistent with higher concentrations of cyclopamine being needed to compete with SAG for Smo binding [59] . Cyclopamine has been reported to function as a partial agonist for non-canonical Smo signaling in adipocytes to influence Warburg-like metabolism [60] . Cyclopamine effects on mSmoWT were therefore examined in the absence of SAG ( Fig 7B’ ) . Cyclopamine triggered a modest negative DMR response , indicating a change in basal mSmo signaling distinct from that induced by SAG ( Fig 7B’ blue and S2A Fig ) . The shift was reversed by PTX treatment , suggesting that it is Gαi dependent , and may represent the PTX-sensitive partial agonism that has previously been reported ( S2A Fig , blue vs . red ) [60] . Examination of mSmoNQ4 Gαi signaling activity revealed a significant difference in the capacity of wild type and glycan-deficient proteins in mounting a non-canonical response . NQ4 failed to induce a measurable DMR in response to SAG or cyclopamine , suggesting a compromised Gαi signaling arm ( Fig 7B” and S2B Fig ) . To confirm that attenuation of the DMR signature was not due to altered G protein engagement by mutation of the IC3 asparagine , we performed DMR using mSmoNQ3 in which the IC3 asparagine is not altered . Like mSmoNQ4 , mSmoNQ3 failed to induce a robust DMR in response to SAG or cyclopamine , which is consistent with loss of extracellular glycosylation compromising non-canonical signal output ( Fig 7B* and S2C Fig ) . To further test this , ligand-induced cellular cAMP modulation was measured in mSmo-expressing HEK293T cells . Cells were pretreated with forskolin to raise basal cAMP , and the ability of SAG or 20 ( S ) -OHC to decrease cAMP was monitored ( Fig 7C and 7D ) . Cells expressing mSmoWT triggered a dose-dependent reduction in intracellular cAMP following SAG or 20 ( S ) -OHC stimulation ( Fig 7C and 7D , black bars ) . Conversely , cAMP levels were not efficiently decreased in cells expressing mSmoNQ4 in response to these agonists ( gray bars ) further supporting that SmoNQ4 is compromised in its ability to signal through Gαi . A documented non-canonical cell biological Smo response is its ability to influence cell migration via activation of RhoA and Rac GTPases [24] . To determine whether this non-canonical activity would be impacted by glycan loss , Smo-/- cells stably expressing mSmoWT , mSmoNQ4 or stably transfected with vector control were subjected to a scratch test , and their ability to migrate through the scratch following Shh ligand treatment was quantified ( Fig 7E and 7F ) . Consistent with Smo being required for Shh-induced cell migration , vector transfected Smo-/- cells demonstrated a modest 8 hour migration that was not significantly affected by cyclopamine treatment ( Fig 7E and 7F ) . Conversely , Shh-treated Smo-/- cells stably expressing WT mSmo showed an obvious migration into the scratch zone , traveling on average 4-fold further than vector control cells ( Fig 7E and 7F ) . This migration was attenuated by cyclopamine treatment , indicating that the Gαi-mediated response is sensitive to cyclopamine inhibition , as was suggested by our DMR analysis ( Fig 7B’ and 7F ) . Conversely , mSmoNQ4-expressing cells failed to efficiently migrate into the scratch zone over the 8 hour assay window ( Fig 7E and 7F ) . The inability of SmoNQ4 to mount a strong non-canonical Gαi response , as evidenced by attenuated DMR , cAMP reduction and migration capacity , are in direct contrast to the robust SAG- and Shh-mediated Gli responses elicited by mSmoNQ4 ( Fig 6E ) . Taken together , these results suggest an intrinsic bias of the under-glycosylated Smo protein toward the canonical signaling arm .
Drosophila Smo Signaling was impacted by loss of N-linked glycosylation at N336 in EC1 , and was further compromised by introducing a second mutation at N213 in the CRD . The combined loss of these two N-linked glycosylation acceptor sites recapitulated the loss of activity observed for dSmoNQ5 , which harbors mutations of the five sites that are conserved across sequenced Drosophila Smo proteins . Disruption of glycosylation at these two sites triggered disulfide bond loss and ER retention , suggesting an essential role for these post-translational modifications during dSmo protein folding . Glycosylation plays a range of roles in membrane and secretory protein biology including trafficking , stability , structural rigidity , functionality and most commonly , protein folding in the ER [37] . Glycan modifications are frequently localized to regions where protein secondary structure changes and stabilizing influences are needed [37 , 61] . Accordingly , N-linked glycans can impact the formation of stabilizing disulfide bonds through binding to the ER chaperones Calnexin and Calreticulin . Glycan-bound chaperones then recruit ERp57 to the client protein to drive disulfide bond formation [37 , 41] . Consistent with this essential functionality , N336 of Smo is adjacent to a conserved disulfide bond that is formed between TM3 residue C339 and EC2 residue C413 [10 , 42] . The functional effects of disrupting this disulfide bridge in flies ranges from robust ligand-independent signaling activity when the C339 bond partner is targeted , to weak hypomorphic activity when the C413 partner is affected [10 , 62] . SmoN336Q signaling activity is reduced by ~50% , which is consistent with the loss of signal observed with a C413A mutation [10 , 62] . As such , we speculate that glycosylation at N336 serves as a linchpin for dSmo protein folding by contributing to formation of this essential disulfide bridge . Accordingly , N336Q mutation resulted in an increase in free thiols , which is consistent with disruption of at least one disulfide bond . A theme that is common to glycobiology that was recapitulated with dSmo is the observation that a single glycan modification may not be necessary when targeted individually , but when disrupted along with a second modification , becomes essential [37] . We observed this for N213 , a glycan acceptor site localized to the amino-terminal CRD . When mutated alone , this residue failed to impact dSmo function , but when targeted along with N336 , ablated Smo signaling . Like N336 , N213 is localized adjacent to a cysteine engaged in a disulfide bond that is required for Smo functionality [42 , 47] . This disulfide bond occurs between C218 and C238 of the D . melanogaster Smo protein and assists in positioning the amino-terminal CRD toward the EC loop domains [42] . This conformation has been postulated to affect the binding of small-molecule Smo modulators , as its disruption severely compromises Smo signaling [47] . Despite some under-glycosylated dSmo mutants being retained in the ER , their protein levels appeared significantly higher than wild type , suggesting that the mutants are not misfolded enough to be actively targeted for ER associated degradation ( ERAD ) . We have observed similar behavior by other dSmo mutants affecting the TM3-EC2 disulfide bond; C339A and C413A dSmo mutants also accumulate in the ER and escape ERAD under basal conditions [10 , 36] . A potential explanation for the accumulation of ER-retained dSmo mutants is that their failure to efficiently exit the ER removes them from the normal recycling/degradation circuit [11 , 63 , 64] . In the absence of Hh , dSmo that reaches the plasma membrane is subject to ubiquitination and Shibiri-mediated endocytosis , resulting in its eventual degradation [63 , 64] . Mutant protein that lingers in the ER is removed from this regulatory cycle , thereby allowing for its accumulation . Consistent with this hypothesis , we found that ER-retained SmoNQ5 demonstrated a significantly longer half-life than the wild type protein . Combined , these results support an essential role for two N-linked glycosylation sites in dSmo , N213 and N336 , thereby broadening the knowledge about post-translational modifications contributing to Drosophila Smo functionality . Contrary to the essential role we uncovered for N-linked glycans in dSmo ER exit and signaling , the mSmo glycosylation mutant demonstrated normal trafficking , dimerization , ligand binding and canonical signaling to Gli transcriptional effectors . However , a clear difference was evident in the ability of mSmoNQ4 to induce the non-canonical signal via Gαi . Although biased agonism was initially thought to result only from synthetic ligands , it is now believed that signal bias can be facilitated through natural ligands , and may serve as a mechanism to diversify GPCR signaling capability [26 , 28 , 29] . Consistent with this model , a recent study demonstrated that distinct chemokines targeting the same chemokine receptor could elicit clear differences in G protein-dependent and -independent signals [29] . The exact mechanism ( s ) by which individual ligands induce differential signal output from the same receptor are not yet clear . However , ligand binding to allosteric sites on a given receptor likely stabilizes different active receptor conformations that may result in varying signal efficiencies to distinct routes [28 , 29 , 65] . We therefore hypothesize that the Gli signal bias induced by stripping mSmo of its N-linked glycosylation results from a change in its agonist-stabilized conformation . Rather than being in a conformation that is equally competent for both signals , the glycan-stripped protein is stabilized in a conformation more permissive for the canonical signal route ( Fig 8 ) . Consistent with this , mSmoNQ4 showed significantly enhanced SAG- and Shh-induced Gli responses compared to those of the wild type protein . Intriguingly , both WT and NQ4 showed a similar 20 ( S ) -OHC Gli response . One explanation for this could be that the NQ4 conformation shifts induced by SAG and 20 ( S ) -OHC binding differ , with the SAG shift being more permissive for Gli activation . However , SAG and Shh induced a similar enhanced NQ4 Gli response , leading us to favor an alternate interpretation . The endogenous Smo ligand controlled by Ptch has not yet been established , but is suggested to be an oxysterol-like molecule [15 , 46 , 49] . It is therefore possible that availability of 20 ( S ) -OHC to mSmo may be influenced by Ptch , which would account for the similar 20 ( S ) -OHC responses observed . The predicted N-linked glycosylation sites are localized near to the known mSmo ligand binding pockets; N1 and N2 flank the CRD , which binds 20 ( S ) -OHC and N4 localizes to EC3 , which is believed to assist in coordinating small molecule binding to the 7-TM core [42 , 46 , 48 , 49 , 66] . As such , context-specific changes in modification of these sites could alter ligand-induced responses , and may be one mechanism by which signal axis specification occurs in nature . Although we did not observe changes in the Smo glycosylation pattern in response to pathway activation in our assays , it is possible that chronic Hh stimulation might lead to changes in glycosylation that impact signal output over time . To our knowledge , there are only two examples of N-linked glycans influencing signal bias; 1 ) glycosylation of Protease-activated receptor 1 governs bias between Gq vs . G12/13 signal output [67] , and 2 ) differential glycosylation of follicle stimulating hormone ( FSH ) dictates signal bias of the FSH receptor [68 , 69] . We therefore propose that signal bias is a new functionality to be added to the list of GPCR processes affected by N-linked glycosylation , and that vertebrate Smo may have evolved to exploit this functionality as it acquired non-canonical signaling capabilities .
The Clustal W2 program was used to align human ( UniProt Q99835 ) , mouse ( UniProt P56726 ) , rat ( UniProt P97698 ) , chicken ( UniProt O42224 ) , zebrafish ( UniProt Q90X26 ) and the following Drosophila Smo protein sequences: D . melanogaster ( FBgn0003444 ) , D . simulans ( FBgn0194405 ) , D . sechellia ( FBgn0171642 ) , D . erecta ( FBgn0116840 ) D . yakuba ( FBgn0234200 ) , D . ananassae ( FBgn0097636 ) , D . persimilis ( FBgn0163503 ) , D . pseudoobscura ( FBgn0243541 ) , D . mojavensis ( FBgn0139780 ) , D . willistoni ( FBgn0217303 ) , D . virilis ( FBgn0211743 ) and D . grimshawi ( FBgn0119026 ) . Glycosylation sites were predicted using NetNGlc prediction software at www . cbs . dtu . dk/services/NetNGlyc/ . The Smo snake plots were generated with the transmembrane topology prediction tool Protter ( http://wlab . ethz . ch/protter/start ) . pCDNA-V5-mSmo was generated by introducing DNA sequence encoding the V5 epitope tag ( GKPIPNPLLGLDST ) after amino acid 51 in pCDNA-mSmo [36] using Phusion site directed mutagenesis kit ( Thermo Scientific ) . Asn ( N ) to Gln ( Q ) mutagenesis was performed on Drosophila and mouse Smo expression vectors using QuikChange II Mutagenesis Kit ( Stratagene ) . dBip , dCnx and dCrc expression plasmids were generated by cloning the respective cDNAs obtained from the DGRC into pAc5 . 1 ( Life technologies ) . pCS2-YFP-mSmo , the pAc-Cal-KDEL ER marker and pAc-Myc-dSmo were described previously [10 , 38 , 70] . Antibodies used for western blot analyses include anti-mSmo ( SCBT ) , anti-GFP ( Cell Signaling ) , anti-V5 ( Life technologies ) , anti-kinesin ( Cytoskeleton , Inc . ) , anti-alpha tubulin ( cell signaling ) and anti-Myc ( Sigma ) and anti-SmoC [15] . All cells were grown at 37°C in an atmosphere of 5% CO2 in DMEM supplemented with 10% heat inactivated FCS or FBS ( HEK293T ) , 0 . 1 mM nonessential amino acids , 2 mM L-glutamine , 1% Pen-Strep and 1mM sodium pyruvate . For Smo-/- cells FCS was not heat inactivated . To generate stable lines from Smo-/- cells , ecotropic retroviruses were produced by cotransfection of retroviral expression plasmids ( pMSCV-puro , pMSCV-puro-mSmo or pMSCV-puro-mSmoNQ4 ) with packaging plasmids ( pMD-old-Gag-Pol and pCAG4-Eco ) into HEK293T cells using FuGene 6 ( Roche Applied Bioscience ) . HEK293T cells were maintained in DMEM ( Invitrogen , Grand Island , NY ) supplemented with 10% FBS , Pen/Strep , 1mM Sodium Pyruvate and 1mM L-Glutamine . Approximately 3x106 cells were plated the day before transfection on 10cm dishes . The next day cells were transfected with 12μg vector DNA , 6μg pMD-old-gag-pol and 2μg CAG4-Eco using 60μl of Fugene6 . Twenty-four hours post-transfection media was removed and replaced . Six hours later viral supernatant was harvested , replaced and incubated overnight . In the morning viral supernatant was harvested and the procedure was repeated . Viral supernatants were filtered using a 0 . 45 micron filter prior to use . For viral transduction 1 x 106 Smo-/- cells were seeded ~24 hours prior to viral transduction . Cells were then incubated in media containing 20ug/ml polybrene ( American Bioanalytical , Inc . , Natick , MA ) plus viral supernatants . The selection process was started ~48 hours post incubation using Puromycin ( Invitrogen , Grand Island , NY ) . For biochemical assays in NIH3T3 cells , the indicated cell types were seeded at a density of 1 x106 cells/60 mm dish , then transfected the following day with 3 μg of the indicated mSmo expression vectors using Lipofectamine 2000 or 3000 ( Invitrogen ) or Fugene6 ( Promega ) . Cell lysates were prepared ~48 hours post transfection in RIPA buffer ( 50 mM Tris-HCl ( pH 7 . 4 ) , 150 mM NaCl , 0 . 25% deoxycholic acid , 1% NP-40 , 1 mM EDTA , 0 . 1% SDS , 0 . 5 mM DTT , and 1× PIC ( Roche ) ) as described [36] . For Drosophila biochemical analyses , ~3X106 Cl8 cells were plated and transfected the following day with 3 μg of pAc-hh and 5μg of the indicated dSmo expression vectors using Lipofectamine 2000 ( Invitrogen ) . DNA content was normalized with pAc5 . 1A empty vector . Cellular lysates were prepared ~48 hours post-transfection using NP-40 lysis buffer ( 1% NP-40 , 150 mM NaCl , 50 mM Tris , 50 mM NaF , 0 . 5 mM DTT , and 1X PIC ( Roche ) , pH 8 . 0 ) and centrifuged at 2000 x g for 10 minutes . Resulting supernatants were used in assays . Cellular lysates were treated with 1000U peptide-N-glycosidase F ( PNGase ) , 1000U endoglycosidase H ( EndoH ) or 800U O-glycosidase + 100U Neuraminidase for two hours at room temperature prior to SDS-PAGE and western blot . All enzymes were obtained from New England Biolabs ( NEB ) . Lambda phosphatase treatment was preformed exactly as previously described [36] . Experiments were performed a minimum of three times and representative blots are shown . Cl8 cells transfected with WT or NQ5 dSmo expression vectors were treated with 25 μg/ml cycloheximide ( Calbiochem ) and incubated for the indicated time prior to harvesting and lysis in NP-40 buffer . Protein samples were analyzed by western blot and Myc-Smo protein levels relative to Kin were determined by densitometry analysis using Photoshop CS4 . Data were plotted as percent signal relative to the 0 time point to determine half-life . The experiment was performed three times and representative results are shown . Maleimide labeling of free thiols was performed on lysates from Cl8 cells expressing the indicated dSmo protein as we have previously described [10] . Labeling experiments were repeated twice without Hh and once with Hh . Similar results were obtained and a representative western blot is shown . Densitometry analysis was performed on the two Hh ( - ) experiments using Photoshop CS4 . The graph represents the pixel intensity ratio of bound to unbound protein normalized to kinesin . Error bars indicate standard deviation , and were provided to illustrate the minimal variation between the two experiments . RIPA lysates were prepared from NIH3T3 cells expressing the indicated mSmo proteins . Approximately 200 μg of total protein was pre-cleared against 20 μL of equilibrated protein A/G plus agarose beads ( 50% slurry , SCBT ) . Cleared lysates were incubated with 5 μg anti-GFP or rabbit IgG control with rocking at 4°C for 2 hours . Immune complexes were collected on protein A/G beads for 45 minutes with gentle rocking at 4°C . Beads were washed twice in lysis buffer and associated proteins were extracted in 2x sample buffer ( 2% w/v SDS , 2 mM DTT , 4% v/v glycerol , 0 . 04 M Tris-HCL , pH 6 . 8 and 0 . 01% w/v Bromphenol blue ) and analyzed by SDS-PAGE and western blot . Smo-/- cells stably expressing mSmoWT or mSmoNQ4 were seeded at a density of ~2x106 cells/100 mm dish and allowed to grow for 24 hours . Cells were then incubated overnight in culture media supplemented with 0 . 5% FCS plus 100 nM SAG . Forty-eight hours post seeding , live cells were washed three times with cold 1x phosphate buffered saline ( PBS ) pH 7 . 4 and then incubated with gentle shaking for 30 minutes at 4°C in 2 mL of 1x PBS containing 0 . 5 mg/ml EZ-Link Sulfo-NHS-Biotin ( Pierce ) . Biotinylation was quenched by washing cells twice with cold 1x PBS containing 50 mM Tris . Cells were harvested and lysed in RIPA buffer . Lysates were incubated with 50 μl of Streptavidin agarose beads ( Thermo Scientific ) for 1 hour at 4°C . Beads were spun down and supernatants were collected . Beads were washed three times with RIPA buffer and bead purified proteins were extracted in sample buffer . Proteins from the supernatant and Streptavidin-purified beads were analyzed by SDS-PAGE and western blot . Densitometry analysis was performed using Photoshop CS4 and the ratio of signal densities for Streptavidin-bound cell surface vs . intracellular Smo was determined . The experiment was repeated three times and all data pooled . Error bars indicate s . e . m . The western blot is representative . NIH3T3 cells were transiently transfected with the indicated mSmo expression vectors . Forty-eight hours post transfection , live cells were incubated for 4 hours at 37°C in serum-free media containing 5 nM bodipy cyclopamine ( Toronto Research Chemicals Inc . ) . After three 10 minute washes in 1x PBS , cells were fixed and immuno-stained for mSmo as described [36] . Purification on 20 ( S ) -OHC beads was performed as described [15 , 49] . For Drosophila reporter assays , ~3 . 75X105 Cl8 cells were plated in 24-well dishes and transfected the following morning with 50 ng ptcΔ136-luciferase , 10 ng pAc-renilla normalization control , 50 ng pAc-hh and 50 ng pAc-myc-smo . DNA content was normalized with the empty vector pAc5 . 1A . Cell lysates were prepared in passive lysis buffer ~48 hours post-transfection , and luciferase activity was measured using Dual-Luciferase Reporter Assay System ( Promega ) . All assays were performed two or three times in duplicate or triplicate , and all data pooled . Reporter activity is shown as the percent activity relative to the control Hh response , which was set to 100% . Error bars indicate standard error of mean ( s . e . m . ) . Statistical significance was determined using the two tailed Student’s t-test . For murine reporter assays approximately 300 , 000 pMSCV-puro , pMSCV-mSmoWT or pMSCV-mSmoNQ4 stable cells were transfected with 200 ng 8Xglibs-luciferase and 50 ng pRL-TK as previously described [15 , 49] . Approximately 16 hours post transfection , cells were incubated in serum free media for ~ 2 hours and then switched to 0 . 5% serum media containing SAG ( 100 nM ) , Shh conditioned media ( 300μl/ml ) or 20 ( S ) -OHC ( 30 μM ) and incubated for another 24 hours before measuring reporter activity . Assays were performed two times in triplicate , and all data pooled . Error bars indicate s . e . m . Shh conditioned media was generated as described [14] . The DMR signature was determined using an EnSpire Multimode Plate Reader ( PerkinElmer , Waltham , MA , USA ) by label-free technology optimized for G-protein response as previously described [54 , 55] . Briefly , 24 hours before the assay , cells were seeded at a density of 7000 cells per well in 384-well sensor microplates . Cells were treated with vehicle , CTX or PTX ( 10 ng/mL ) in cell culture medium overnight . Prior to taking readings , cells were washed twice with assay buffer ( Hank’s Balanced Salt Solution ( Life Technologies ) , 20 mM HEPES , 0 . 1% BSA , pH 7 . 5 ) , then incubated for 2 hours at 24°C in assay buffer with 0 . 1% DMSO ( vehicle ) prior to scanning for a baseline optical signature . SAG was added to 200 nM and cyclopamine to 2 μM final concentration and DMR responses were monitored for 8000–15 , 000 seconds . Kinetic results were analyzed using EnSpire Workstation Software v 4 . 10 . Approximately 3500 HEK293T cells expressing the indicated mSmo proteins were plated per well in 384 well plates and pretreated for 10 minutes with 0 . 5 μM forskolin ( Sigma ) prior to adding SAG or 20 ( S ) -OHC ( 1x = 100nM ) . cAMP levels were determined by using the LANCE Ultra cAMP kit ( PerkinElmer ) and a Pherastar FS Microplate Reader ( BMG Labtech , Ortenberg , Germany ) . Drosophila immunofluorescence analyses were performed exactly as previously described [36] . For detection of mSmo in the primary cilium , Smo-/- cells stably transfected with vector control , mSmoWT or mSmoNQ4 plasmids were incubated overnight in culture media supplemented with 0 . 5% FCS plus SAG ( 100 nM ) . The following morning cells were fixed in 4% paraformaldehyde for 15 minutes at room temperature followed by three 10 minute washes in 1xPBS . Fixed cells were permeabilized and blocked in PBGT ( 1xPBS+0 . 1% TritonX-100+5% Normal goat serum ) for 1 hour , then incubated overnight with anti-Smo ( SCBT ) and anti- γ-tubulin ( Sigma ) antibodies . AlexaFluor 488 or 555 conjugated secondary antibodies ( 1:1000; Life technologies ) were used . Slides were mounted using Vectashield with DAPI ( Vector Labs ) . Immunofluorescence data were collected using Zeiss LSM 510 or 710 and images were processed using Zen2009 and Photoshop CS6 . For all immunofluorescence experiments , multiple cells were examined over a minimum of three experiments , and representative images are shown . For quantification of ciliary localization , 96–126 cells were counted over three experiments and the percent of cells showing Smo in the primary cilium was determined . The paired t-test was used to determine statistical significance . Cell migration was assessed by performing scratch assays as previously described [24 , 71] . Smo-/- cells stably transfected with mSmoWT , mSmoNQ4 or empty pMSCV-puro vector control were seeded in a 12 well dish at a density of 3x105 cells/well and allowed to grow in complete media until fully confluent . Once cells reached confluency , they were incubated in media supplemented with 0 . 5% FCS overnight . The following morning ( time T = 0 ) , cells were washed with PBS and a scratch was made across the well using a p20 pipette tip . Scratched wells were washed twice in PBS to remove scratched cells , then incubated for 8 hours ( T = 8 ) in low serum media supplemented with Shh conditioned media ( 1:4 ratio ) , 0 . 5 μM cyclopamine or vehicle control . For Shh + cyclopamine conditions , cyclopamine was added to the overnight incubation prior to scratching and adding Shh media . Images were taken at T = 0 and T = 8 using an EVOS cell imaging system ( Life Technologies ) . In each experiment , three different zones along the scratch were imaged , and in each zone 10 equally spaced points were used for analysis . The experiment was performed three times and all data pooled for quantification . Distance migration was calculated and analyzed as previously described [24] . UAS-myc-smo transgenic flies were generated by targeting the respective transgenes to the landing site 3L-68E using the PhiC31 system [72] . UAS-smo3′UTR-dsRNA and UAS-Dicer; nubGAL4 fly lines were gift from David Hipfner [40] . Genotypes are indicated in figure labels . All crosses were performed at 29°C . Wings from adult females were mounted on glass slide using DPX mounting media and imaged using a Ziess LSM200-C microscope with a Ziess AxioCam ICc3 camera . Wings from multiple progeny over two independent crosses were analyzed and representative wings are shown . To test protein expression levels , ten wing imaginal discs for each genotype were isolated and lysed in 2x SDS sample buffer ( 4% SDS , 4 mM DTT , 8% glycerol , 0 . 08 M Tris-HCL , pH 6 . 8 , 0 . 02% bromphenol blue ) . Lysates were analyzed by SDS-PAGE and western blot . | N-linked glycosylation is a post-translational modification occurring on membrane proteins such as G protein-coupled receptors ( GPCR ) . Smoothened ( Smo ) is a GPCR that functions as the signal transducer of the Hedgehog ( Hh ) pathway . We used a mutagenesis approach to assess the role of N-glycans in Smo signaling in two genetic models for Hh pathway activity , Drosophila and mouse . In doing so , we discovered a divergence in glycan function between them . We mapped an essential N-glycan acceptor site that when lost in Drosophila , triggered ER retention , altered Smo protein stability and blunted its signaling capacity . Conversely , ER exit of the murine protein was unaffected by glycan loss , as was its ability to traffic and induce a G protein-independent signal to activate Hh target genes . However , the ability of vertebrate Smo to induce a distinct G protein-dependent signal was lost . This suggests that N-linked glycosylation may influence signal bias of vertebrate Smo to favor one signal output over the other . We therefore propose that the role of this conserved post-translational modification may have been repurposed from governing Smo ER exit in the fly to influencing effector route selection in vertebrates . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Functional Divergence in the Role of N-Linked Glycosylation in Smoothened Signaling |
Virtual compound screening using molecular docking is widely used in the discovery of new lead compounds for drug design . However , this method is not completely reliable and therefore unsatisfactory . In this study , we used massive molecular dynamics simulations of protein-ligand conformations obtained by molecular docking in order to improve the enrichment performance of molecular docking . Our screening approach employed the molecular mechanics/Poisson-Boltzmann and surface area method to estimate the binding free energies . For the top-ranking 1 , 000 compounds obtained by docking to a target protein , approximately 6 , 000 molecular dynamics simulations were performed using multiple docking poses in about a week . As a result , the enrichment performance of the top 100 compounds by our approach was improved by 1 . 6–4 . 0 times that of the enrichment performance of molecular dockings . This result indicates that the application of molecular dynamics simulations to virtual screening for lead discovery is both effective and practical . However , further optimization of the computational protocols is required for screening various target proteins .
In our screening approach , we adopted molecular docking and the MM/PB-SA method based on MD simulations as the first and second filters , respectively . First , we performed molecular docking by using the conformations of a target protein and the compounds contained in the compound library . Additionally , the results of molecular docking were applied to the post-processing for the selection of successfully docked compounds and the classification of multiple binding poses ( see Materials and Methods ) . Next , all of the conformations obtained from the molecular docking were energy-minimized using molecular mechanics ( MM ) force-field ( hereafter we call this MM calculations ) . MD simulations were then applied to multiple conformations of the protein-ligand complexes . The binding free energies were calculated by the MM/PB-SA method using the coordinate sets obtained from the MM calculations and MD simulations . Finally , we assessed the enrichment of active compounds by using ranked lists of compounds graded on the basis of their binding free energies .
To evaluate the ability of the MM/PB-SA method to act as a filter after molecular docking , we performed MD-based compound screening for four target proteins ( trypsin , HIV-1 protease ( HIV PR ) , acetylcholine esterase ( AChE ) , and cyclin-dependent kinase 2 ( CDK2 ) ) . These targets have been widely evaluated in structure-based computer-aided drug design [26] , [29]–[34] . For each target protein , we first assessed the enrichment of 12 types of binding free energies ( Table 1 ) . These 12 types of binding free energies were classified into four categories . G01–G03 in category 1 were the energies calculated from the MM calculations . The other categories 2–4 , which contained the energies calculated from the MD simulations , were classified according to the combination of coordinate sets used for the enthalpy calculations; G04–G06 , G07–G09 , and G10–G12 belonged to categories 2 , 3 , and 4 , respectively ( a detailed explanation is given in the Materials and Methods section . ) . Analyses of the Receiver Operating Characteristic ( ROC ) curves [35] are given in Table 2 . An ROC curve is closely related to an enrichment curve but is not exactly equivalent to it . This curve describes the tradeoff between sensitivity and specificity . Sensitivity is defined as the ability of the classifier to detect true positives , while specificity is the ability to avoid false positives . The area under an ROC curve , i . e . , the ROC value , indicates the quality of enrichment . The ROC value of a random classifier is 0 . 5 , while that of an excellent classifier is greater than 0 . 9 . Table 2 shows the ROC values for all of the target proteins . From these values , we can observe three common features for three of the target proteins ( trypsin , HIV PR , and AChE ) , excluding CDK2 . It is obvious that the ROC values for all of the binding free energies ( G01–G12 ) of multiple poses are higher than those of a single pose , suggesting that docking and its post-processing can sample potentially correct docking poses of active compounds . This implies that the potentially correct binding mode is contained within the top 10 highest-scored docking poses but is not always the highest-scored docking pose . In our study , after docking and post-processing , MD simulations were applied to an average of 5–6 docking poses for each compound in order to increase the efficiency of the sampling of a ligand's conformations . Although MD simulations of multiple poses are expensive , they are necessary for improving enrichment . The second common feature is that the highest ROC value for each target protein was obtained for the energies calculated from the MD simulations , rather than for those calculated from the MM calculations . This implies that the introduction of protein flexibility and the effect of water molecules facilitated the refinement of the protein-ligand interactions and that the MD-based MM/PB-SA method provided a more reliable binding free energy . These two common features were clearly seen in the results for the active compounds . A typical successful example of MD simulations using multiple poses is shown in Figure 1 . In the crystal structure of trypsin complexed with an inhibitor [36] , the amidine group of the inhibitor ( Figures 1 and S1; active compound ( 13 ) of trypsin ) formed hydrogen bonds with the important residue Asp-180 in the binding pocket . Further , the highest-scored docking pose was so inaccurate that no important interactions were observed at all , but the 7th ranked docking pose was similar to that of the crystal structure . In addition , the application of the MD simulation to the 7th ranked docking pose appropriately improved the key hydrogen bonds and the position of the naphthalene group and G06 value of the 7th ranked docking pose was the lowest in all the poses . The last common feature was that the binding free energies with no entropy terms ( i . e . , G06 , G09 , and G12 ) , which were obtained by using the trajectories of the MD simulations , showed the highest ROC values in the respective energy categories ( 2–4 ) . Thus , the introduction of entropy terms tended to reduce enrichment . This is probably due to the difficulty of computing entropy values for the MM-PB/SA energy function . We will further discuss this problem in the Discussion section . Our MD simulations encouraged conformational relaxation , and the binding enthalpy from the MM-PB/SA method could satisfactorily increase the enrichment performance . However , the treatment of binding entropy terms involves certain unsolved problems . Here , we performed a statistical analysis using data on the ROC values to evaluate the differences between key classifiers , G01 ( multiple poses ) , G06 ( multiple poses ) , and molecular docking ( Table 3 ) . The program DBM MRMC version 2 . 1 was used in this analysis [37]–[41] . From this analysis , it was obvious that the differences in the ROC values between G06 and docking , and those between G06 and rescoring ( docking ) , were statistically significant for trypsin and HIV PR , but the difference in the ROC values between G06 and docking for AChE was not statistically significant . On the other hand , the differences in the ROC values between G01 and docking were not statistically significant for trypsin , HIV PR , and AChE . An examination of the entire data set indicated that the binding free energies of multiple poses , especially G06 , which was obtained from the MD trajectories of just the protein-ligand complexes with no entropies , showed a high and stable ability to enrich the active compounds . This paper provides a detailed account of the ability of our approach to discriminate active compounds from inactive ones . Figure 2 shows the ROC curves for the respective target proteins . The ROC curves of the binding free energies , with the highest ROC values in the respective categories , were observed for trypsin , HIV PR , and AChE . Table 4 shows the information on the enrichment factors to allow the abilities of the classifiers to be understood clearly . For trypsin , 10 active compounds ( out of 21 ) were ranked in the top 1 , 000 compounds ( Figure S1 ) . No significant difference was observed in the results of molecular docking and random screening for these top 1 , 000 compounds . Considerable improvement was observed in the results of the MM calculations and MD simulations . G01 and G06 , in particular , showed high enrichment performances , and the enrichment factors for the top 100 compounds were 5 . 00 and 4 . 00 , respectively ( see Table 4 ) . Furthermore , G06 detected no less than nine active compounds in the top 300 . For HIV PR , 6 active compounds ( out of 8 ) were ranked in the top 1 , 000 compounds ( Figure S2 ) . A slightly better enrichment was achieved by docking than by random screening . As seen in the curves , we found drastically improved enrichment by G06 . It detected 6 active compounds in the top 100 compounds , and the enrichment factor for the top 100 was 10 . 0 ( Table 4 ) . For AChE , 7 active compounds ( out of 14 ) were detected in the top 1 , 000 compounds ( Figure S3 ) . We found that the enrichments of G06 and G09 were considerably better than that of molecular docking , although the difference in the ROC values between G06 ( or G09 ) and docking was not statistically significant . Because there was only a slight difference between G06 and G09 , both of them detected five active compounds in the top 100 . For CDK2 , the ROC curves of the following representative binding free energies were drawn: G01 ( single and multiple poses ) , G04 ( single pose ) , G07 ( multiple poses ) , and G10 ( multiple poses ) . Seventeen active compounds ( out of 26 ) were ranked in the top-scoring 1 , 000 compounds ( Figure S4 ) . The G01 of single and multiple poses as obtained from the MM calculations showed higher enrichment than random screening; however , there was no statistically significant difference between the results of G01 and molecular docking ( see Table 3 ) . The G01 of single and multiple poses detected 10 active compounds in the top 300 and showed only slightly higher enrichment factors than molecular docking ( Table 4 ) . In contrast , G04 , G07 , and G10 , which were obtained from the MD simulations , remained unchanged or worsened as compared to docking , although they identified 6 or 7 active compounds in the top 200 . Over all , the enrichments for CDK2 were not at all improved as mentioned above . The ROC values for CDK2 showed a different tendency as compared to those for the other three proteins ( Table 2 ) . Among the 12 types of energies , the G01 of single and multiple poses showed the highest values ( 0 . 685 and 0 . 719 , respectively ) , which implies that the enrichments of the MM calculations were better than those of the MD simulations . Moreover , for 7 types of energies ( out of 12 ) , the single pose results showed higher enrichment than those of multiple poses . In addition , the binding free energies with entropy terms showed slightly high enrichment performances in the respective categories ( 2–4 ) , which were calculated from MD simulations . In particular , G04 , G07 , and G10 , which included the binding entropy effects of the ligands , showed the highest ROC values in their respective categories . We monitored the mobility of ligand molecules in MD simulations of CDK2 . Figure 3 shows the cumulative percentages of positional displacements of ligand molecules relative to each protein between the docking and final MD structures . From this figure , it is clear that the docked ligands for CDK2 did not move very much in the MD simulations , as compared to the other target proteins , which implies that the protein-ligand interactions were not fully relaxed . Such insufficiency in conformational relaxation/refinement directly influences protein-ligand interactions . Particularly for active compounds , the binding modes obtained by MD simulations were some different from those of experimental structures ( refer to Figures S5 and S6 ) . These data suggested that the use of MD simulations for CDK2 led to structural uncertainties for active compounds . In addition , the interactions of the inactive ( decoy ) compounds would not be refined fully in MD simulations . We think that such low mobility for ligand molecules and the improper conformational dynamics are due to an improper MD setup . This would be the reason why the enrichments of the MD simulations using multiple poses were worse than those of the MM calculations for CDK2 .
We evaluated the ability to enrich active compounds for four target proteins: trypsin , HIV PR , AChE , and CDK2 . Our screening approach could improve the molecular docking results for all of the proteins except CDK2 . For trypsin , HIV PR , and AChE , our results indicated that the use of multiple poses improved the enrichments of all the MM calculations and MD simulations . In addition , the binding free energies calculated from the MD simulations showed higher and more stable enrichments than those of the docking and MM calculations . In particular , the G06 using multiple poses was considered to be effective . This energy contained no entropy components . Further , the enthalpy components were calculated using the coordinate sets extracted from the MD simulation of a complex . Kuhn and coworkers [25] reported that for the MM/PB-SA values of MM calculations , the strategy of using multiple poses could only show a high enrichment when the correct binding mode was contained within the higher-scored docking conformations , but was not captured with a single pose . In our study , we carefully selected multiple docking poses by the post-processing of docking results and used an average of five to six docking poses for each compound . As a result , the correct binding modes or potentially correct modes that could be refined by the MD simulations were sampled within the selected multiple docking poses , which did not often correspond to the top-scored pose . Therefore , the results using multiple poses showed a higher enrichment than those obtained using a single pose . In addition , Kuhn et al . [25] showed that the use of MD simulations often leads to structural uncertainties and an inaccurate estimation of the binding free energy . The MM/PB-SA energies of the MM calculations and MD simulations in their study corresponded to G01 and G04 in our study . A comparison between G01 and G04 indicated that the enrichment of G04 was lower than that of G01 , which is consistent with the results of Kuhn and coworkers [25] , although there were large differences in the MD setup , MM/PB-SA setup , and target proteins . G01 contained only the enthalpy components that were calculated using the coordinate sets derived from the MM calculation of a complex . G04 contained the binding entropy effect of the ligand . Further , the enthalpy components were calculated using the coordinate sets from the MD calculation of a complex . The difference between G04 and G06 was the presence of the entropy effect . Therefore , we consider that for trypsin , HIV PR , and AChE , the structural refinement/relaxation by longer and higher time resolution MD simulations and the relatively accurate estimation of binding free energy ( enthalpy ) by the MM/PB-SA method led to increased enrichment , but the introduction of the entropy values induced an uncertainty in the binding free energies . On the other hand , we think that the use of MD simulations for CDK2 led to structural uncertainties and then an inaccurate estimation of the binding free energy ( Figures S5 and S6 ) . This would be due to an improper MD setup , as Kuhn and co-workers suggested in their paper [25] . Basically , it is well-known that it is difficult to calculate entropy values properly . In our work , the entropy values were calculated by principal component analysis ( PCA ) . These values are sensitive to the data sampling frequency [42] , [43] and are likely to be overestimated [44] . Therefore , we believe that the entropy values were slightly unstable and not completely reliable . An alternative computational method is normal mode analysis . This may be stable to some extent , but it is known that conformations at different local energy minima provide rather similar entropy values even though there are differences in the finite temperature [42] . Moreover , the computational cost is significantly high to use for the calculation of many structures . Thus , even if we were to use normal mode analysis , the entropy values would induce an uncertainty in the binding free energies . Therefore , in order to achieve further improved enrichment , it is necessary to improve the calculations for the entropy terms . Our strategy could not significantly improve the molecular docking results for CDK2 . It is well known that , as compared to the binding pockets of the other three proteins , the binding pocket of CDK2 is more flexible and hydrophobic . We compared the binding pockets in two different X-ray crystal structures of CDK2 [45] , [46] ( Figure 4 ) . This figure indicates that the shape of the binding pocket is very flexible and that the hydrophobic region covers the surface of the binding pocket . In addition , a study on molecular docking using different CDK2 crystal structures reported that the volume ( flexibility ) of the binding site is a key factor for predicting docking poses [29] . Although only one CDK2 structure was used in this study , we applied MD simulations to protein-ligand structures obtained from molecular docking to facilitate the relaxation of protein-ligand interactions . Unfortunately , our MD simulations were insufficient to relax the protein-ligand conformations in the binding pockets ( see Figure 3 ) . Such insufficiency is believed to be due to the MD setup . To improve the insufficient relaxation , we applied the MD simulations for 1 . 4 ns to each configuration; this simulation time was twice that of the initial MD simulation time . These MD simulations effected some relaxation/refinement of the ligand conformations ( Figure 3 ) ; the enrichments of the multiple poses were found to be higher than those of the single pose ( Table 2 ) . Despite this , G04–G09 showed only small improvements in the enrichment performance . These results suggest that further improvement of the MD setup was necessary . To obtain information about how to improve the MD setup , we attempted to maximize the ROC values by using an approach based on the linear response ( LR ) [47] and MM/PB-SA methods ( LR-MM/PB-SA approach [48] ) The LR-MM/PB-SA equation was derived from equations 2–4 ( see Materials and Methods ) : ( 1 ) where a , b , c , d , e , and f are weighting factors ranging from 0 . 5 to 1 . 5 . The terms on the right side of equation 1 represent the energy difference between the complex and protein plus ligand . This approach is usually used for estimating the binding affinity by combining an empirical MM/PB-SA energy calculation with an LR optimization of coefficients against the experimental binding affinities of several compounds . The optimized free energy model is used for interpreting the binding model and predicting the binding affinity of unknown molecules . In our study , we optimized the weighting factors to maximize the ROC value , that is , the enrichment performance , using a genetic algorithm ( GA ) . We applied the LR-MM/PB-SA approach to the G10 of multiple poses obtained from the initial MD simulations , because G10 showed the highest ROC value among those of the binding free energies calculated from the MD simulations ( Table 2 ) . As a result , when the weighting factors of a–f were 1 . 12 , 0 . 91 , 1 . 47 , 1 . 01 , 0 . 87 , and 1 . 49 , respectively , a maximum ROC value of 0 . 812 was obtained ( Figure 5 ) . This result suggested that the LR-MM/PB-SA approach was effective at improving the enrichment performance , and these weighting factors indicated an improvement plan for the MD setup . The weighting factor of the entropy term , 1 . 49 , would contribute to dilute the percentage of inactive ( decoy ) compounds . This would be related to the fact that the ligands in the binding pocket of CDK2 could not move largely ( Figure 3 ) . In addition , it is conceivable that the weighting factor of ΔEvdW , i . e . , 1 . 47 , enriched the active compounds because they include a hydrophobic region and formed comparatively strong hydrophobic interactions with the binding pocket ( See Figure S4 ) . As the binding modes obtained through MD simulations were some different from the experimentally observed binding modes , the conformational refinement was considered insufficient or improper to accurately predict the binding free energy . This information also suggests that fully conformational relaxation/refinement is required to improve the enrichment performance . In this study , the ligand , water molecules , and protein residues around the binding pocket were allowed to move , but other protein residues were restrained to the X-ray structure in all of the MM calculations and MD simulations ( a detailed explanation is given in the Materials and Methods section ) . Hence , to achieve the conformational relaxation of the binding pocket of CDK2 , allowing wider protein residues to move in MD simulations and longer MD simulations are required . The former is an especially important parameter for improving the enrichment performance , although it would increase the computational cost . In our next study , which will focus on the extent of the mobility of protein residues , along with simulation time and force-field parameters for organic small molecules , we will attempt to optimize the MD setup using a recent widely used dataset of decoy compounds [49] . The computational screening of large compound libraries involves the use of hierarchical multiple filters , such as ligand- and structure-based approaches . Molecular docking plays the primary role in these filters . With advancements in computer performance and computational chemistry , docking programs have become more accurate , but their ability to enrich hit compounds remains unsatisfactory . In order to improve the enrichment performance of molecular docking , we attempted to use the MM/PB-SA method [50] as a post-molecular docking filter . The basis of our approach was to perform massive MD simulations of protein-ligand conformations obtained from molecular docking , aim at the refinement/relaxation of protein-ligand conformations after docking , and predict more accurate binding free energies using the MM/PB-SA method in a practical time for lead discovery . Combining molecular docking and MD simulations basically allows each of them to neutralize the other's defects , but certain problems remain even with MD simulations , particularly with regard to compound screening applications . The major drawback of MD simulations is insufficient sampling due to the significant computational cost involved . To solve this problem , we performed MD simulations using various docking conformations obtained by molecular docking . However , the computational cost of this technique was approximately five to six times that of MD simulations using single docking conformations , such as the top-scored docking conformation . The enormous computational time needed for MD simulations is a serious problem . Here , we solved this problem by accelerating most of the time-consuming operations of the MD simulation using a high-performance special-purpose computer for MD simulations , “MDGRAPE-3” [27] , [28] . Accordingly , our approach could be performed in a practical time ( about a week ) for lead discovery . The evaluation in this study provides valuable information on in-silico drug design . Further , a more rigorous MD-based filter is under consideration for further improving the enrichment performance . This technique will also be applied to the lead optimization stage of drug development research . In conclusion , our approach could improve the enrichment of virtual screening by molecular docking . Among the 12 types of binding free energies , G06 , which was obtained from the MD simulations using multiple poses , showed the highest and most stable ability to enrich the active compounds . The strategy of multiple poses can be used to sample the potentially correct poses of active compounds; thus , it increases the enrichment performance . Since the G06 enrichment factors for the top 100 compounds ranged from 4 to 10 ( see Table 4 ) , which indicates approximately 1 . 6–4 . 0 times higher values than the enrichment performance of molecular docking , with the exception of CDK2 , it is obvious that a stable and high enrichment can be achieved after molecular docking . In addition , G06 is suitable for compound screening because its computational cost is the least among those of the other MM/PB-SA energies obtained from the MD simulations . We also confirmed that G01 , which was obtained from the MM calculations , showed good enrichment ability despite its low computational cost . This result agreed with that of the previous study [25] . The ability of G01 to enrich active compounds was lower and less stable than that of G06 , but we believe that G01 acted as an effective filter between molecular docking and the MD-based MM/PB-SA method . From this study , we conclude that the application of MD simulations to virtual screening for lead discovery is effective and practical , but that further optimization of the MD simulation protocols is required for the screening of various target proteins , including kinases .
We applied our approach to four target proteins: trypsin , HIV PR , AChE , and CDK2 . These structures with crystallographic resolutions of less than 3 . 0 Å , were retrieved from the Protein Data Bank ( PDB ) because the conformations of residues in the binding pocket affect the molecular docking results ( PDB Id: 1C5S ( trypsin ) [51] , 1HWR ( HIV PR ) [52] , 1E66 ( AChE ) [53] , and 1FVV ( CDK2 ) [46] ) . All of the bound crystal water molecules , ligands , and other organic compounds were removed from each protein . Hydrogen atoms were added , and energy minimizations on the hydrogen atoms were performed using the Molecular Operating Environment ( MOE ) program ( Chemical Computing Group Inc . [54] ) . For each target protein , we prepared a test set of compounds that included 10 , 000 randomly selected compounds , or decoys , from the Maybridge library of compounds and experimentally known active compounds . It was confirmed that 95 . 5% of the selected decoy compounds obeyed the Lipinski rule of 5 [55] . The active compounds , which had binding affinities ( Ki , Kd , or IC50 ) below 30 µm , were selected from the PDBbind database [56] , [57] and by referring to the literatures [26] , [58] . Most of the active compounds also obeyed the Lipinski rule of 5 . The numbers of active compounds selected for each of the respective target proteins was as follows: 21 ( trypsin ) , 8 ( HIV PR ) , 14 ( AChE ) , and 26 ( CDK2 ) ( see Figure S1 , S2 , S3 , S4 ) . For each compound of the test set , a 3D conformation was generated , ionized , and energy minimized using LigPrep ( Schrödinger Inc . [59] ) , assuming a pH of 7 . 0 . Molecular dockings were performed using the Genetic Optimisation of Ligand Docking ( GOLD ) version 3 . 1 [9] , [10] . This program employs a GA to explore the possible binding modes . The standard default settings for the GA parameters were used . The binding site radius was 12 Å . We performed the docking run three or four times using the GoldScore or ChemScore function for each target protein and selected the result that showed the best enrichment . GoldScore ( default settings ) was used as the scoring function for trypsin and HIV PR . In contrast , ChemScore ( default settings ) was used for AChE and CDK2 because docking runs using GoldScore can detect few of the successfully docked active compounds for AChE and CDK2 . For AChE alone , the torsional rotations of Phe-330 ( chi1 and chi2 ) were treated as flexible in the docking process . For each docking run , the 10 highest-scoring docking poses were saved to obtain a variety of binding modes . First , among the 10 highest-scoring docking poses saved for each compound , those in which the compound did not occupy the binding pocket or did not interact with the important residues were removed . The latter was used only for trypsin and HIV PR . The important residues were Asp180 for trypsin and Asp24 in each monomer for HIV PR . These treatments had the effect of reducing the false positives for molecular docking . The docked compounds were then arranged in descending order from the highest score with respect to the multiple docking poses , and the top 1 , 000 compounds were selected from the test set . Finally , for the top 1 , 000 compounds , the docking poses of each compound were clustered using the root mean square deviation of 0 . 9 Å ( complete link method [60] ) . After post-processing , approximately 6 , 000 docking poses were selected for the 1 , 000 compounds , which were then used as the initial conformations for MD simulations . Some active compounds were not ranked in the top 1 , 000 . The numbers of active compounds in the top-scoring 1 , 000 were 10 , 6 , 7 , and 17 for trypsin , HIV PR , AChE , and CDK2 , respectively . In addition , the compounds in the top-scoring 1 , 000 were rescored with ChemScore ( trypsin and HIV PR ) or GoldScore ( AChE and CDK2 ) because it is known that the rescoring approach increases the enrichment performance [61] . Furthermore , we analyzed ROC curves using molecular weight as classifier ( Figure S7 ) . From statistical analysis , it is obvious that the differences in the ROC values between G06 and molecular weight were statistically significant for trypsin , HIV PR , AChE . We performed MD simulations of each complex ( ligand-bound protein ) , protein , and ligand to obtain various types of binding free energies ( see the following subsection ) . The active sites of the protein-ligand complexes were immersed in an approximately 28–30 Å sphere of transferable intermolecular potential 3 point ( TIP3P ) water [62] molecules . The radius of the water droplet was selected such that the distance of the atoms of all the docked compounds from the water wall was greater than 15 Å ( see Figure 6 ) . The total number of atoms in the respective systems was approximately 8 , 000–12 , 000 . On the solvent boundary , a half-harmonic potential ( 1 . 5 kcal/mol-Å2 ) was applied to prevent the evaporation of the water molecules . The ligand , water molecules , and protein residues that were approximately 12 Å of the active center were allowed to move , but other protein residues were restrained to the X-ray structure by the harmonic energy term ( 1 . 5 kcal/mol-Å2 ) in all of the MM calculations , namely the MM energy-minimization , and MD simulations . For the simulations of the ligands , each ligand was immersed in a water droplet , and this structure was used as the initial structure for the MD simulation of the ligand . In addition , the simulation of each protein ( trypsin , HIV PR , AChE , and CDK2 ) was performed in the same manner as that of the complex . All of the simulations were performed using AMBER 8 . 0 [63] modified for MDGRAPE-3 [27] , [28] . The ff03 force field [64] was adopted , and the time step was set at 0 . 5 fs . To carefully consider the motion of hydrogen atoms in the interactions between the ligands and protein residues , no bond length constraint was applied to solute atoms . The temperature of each system was gradually increased to 300 K during the first 25 ps , and additional MD simulations were performed for 700 ps for equilibration . The temperature was maintained at 300 K by using the method described by Berendsen et al . [65] , and the system was coupled to a temperature bath with coupling constants of 0 . 2 ps . The parameters and charges for the ligands were determined using the antechamber module version 1 . 27 of AMBER 8 . 0 [63] by utilizing the general atom force field ( GAFF ) [66] and the AM1-BCC charge method [67] , [68] . Although the computational cost of the AM1-BCC charge method is low , a some difference between the charge and that of ff03 was noticeable . Since the original GAFF parameters were insufficient to cope with the parameters of all the ligands , we filled the missing parameters on the basis of the information on regarding atom types , bonds , valences , angles , and dihedrals by using an in-house program ( see Text S1 ) . ( Note: these parameters for proteins and small organic molecules are very important to calculate the binding free energies between proteins and ligands ) Our MDGRAPE-3 system is a cluster of personal computers , each equipped with two MDGRAPE-3 boards . Each board contains 12 MDGRAPE-3 chips and has a peak speed of approximately 2 Tflops . The computations of non-bonded forces and energies for MD simulations were accelerated by MDGRAPE-3 , and the other calculations were performed by the host central processing unit ( CPU ) . In this study , we used 50 host computers equipped with 100 MDGRAPE-3 boards . The calculations for an MD simulation and the estimation of the binding free energies by the MM/PB-SA method were performed simultaneously . The average computational time for a single protein-ligand complex was 2 . 5 h , and the computations for approximately 6 , 000 protein-ligand conformations obtained by docking for each protein were completed in a week . The total simulation time for each protein was 4 µs , which corresponded to an 8-µs MD simulation with a standard time step of 1 fs . A single MD simulation for the system ( Figure 6 ) , without using MDGRAPE-3 , requires more than 10 times the abovementioned computational time . Thus , in the current state , it would be quite difficult to use our screening approach without the MDGRAPE-3 system in a practically appropriate time for lead discovery . Therefore , our study can provide important information for MD-based screening . The production MD trajectory was collected for the last period of 210 ps . In the calculation of the binding free energies by the MM-PB/SA method , the water molecules were replaced with implicit solvation models . The binding free energy was calculated by the following equations . ( 2 ) ( 3 ) ( 4 ) ( 5 ) In the above equations , < > denotes the average for a set of 30 conformations along an MD trajectory . Eint includes the bond , angle , and torsional angle energies; Eele and EvdW represent the intermolecular electrostatic and van der Waals energies , respectively . GPB was calculated by solving the PB equation with the DelPhi program [69] , [70] , using the PARSE radii [71] , [72] and AMBER charges . The grid spacing used was 0 . 5 Å . The dielectric constants inside and outside the molecule were 1 . 0 and 80 . 0 , respectively . In equation 5 , which calculates the nonpolar solvation contribution , A is the solvent-accessible surface area that was calculated using the Michael Sanner's Molecular Surface ( MSMS ) program [73] , and γ and b are 0 . 00542 kcal/mol-Å2 and 0 . 92 kcal/mol , respectively . The probe radius was 1 . 4 Å . The conformational entropy term of the solute , TS , was approximated by a combination of a classical statistics expression and PCA [74] , using the PTRAJ module of AMBER 8 . 0 [63] . In the PCA calculation , the last 210 ps ( 3 , 000 conformations ) of each production trajectory were used . The analysis of the binding free energy involved the calculation of the energies for conformations obtained from the MM ( namely , energy-minimized ) coordinates or MD trajectories . When the MM calculations or MD simulations of a complex , protein , and ligand were performed , we could obtain various types of binding free energies by combining the respective coordinate sets . The enthalpy contributions of Gprotein and Gligand in equation 2 were calculated in the following 2 ways: ( 1 ) by using the coordinate sets of a protein ( or ligand ) obtained from the MD simulations ( or MM calculations ) of the protein ( or ligand ) and ( 2 ) by using the coordinate sets extracted from the MD simulation of a complex . Similar to the enthalpy contribution , the entropy contribution was calculated by using the MD trajectories . When the entropy contributions of Gcomplex , Gprotein , and Gligand were calculated by using the MD trajectory of only the complex , we considered the entropy contribution of ΔGbind to be zero because the energy components were almost cancelled . In this study , in order to thoroughly investigate which MM/PB-SA energies were suitable for compound screening , we adopted 12 binding free energies , G01–G12 , to manage the entropy contributions independently of the enthalpy contributions ( see Table 1 ) . It should be noted that the coordinate sets for calculating the entropy contributions were not always consistent with those for calculating enthalpy contributions . Table 1 shows the enthalpy and entropy terms for computing of Gcomplex , Gprotein , and Gligand in equation 2 . We classified the 12 binding free energies into four categories . Category 1 contained the energies obtained by the MM calculations , and categories 2 , 3 , and 4 contained those obtained by MD calculations . These categories were classified according to the combination of coordinate sets used for enthalpy calculations: G01–G03 , G04–G06 , G07–G09 , and G10–G12 belonged to categories 1 , 2 , 3 , and 4 , respectively . Each binding free energy of a ligand adopts the minimum energies from among the energies of multiple poses . Thus , by gathering and arranging their energies , we were able to assess the enrichment performance of the screening approach . | Lead discovery is one of the most important processes in rational drug design . To improve the rate of the detection of lead compounds , various technologies such as high-throughput screening and combinatorial chemistry have been introduced into the pharmaceutical industry . However , since these technologies alone may not improve lead productivity , computational screening has become important . A central method for computational screening is molecular docking . This method generally docks many flexible ligands to a rigid protein and predicts the binding affinity for each ligand in a practical time . However , its ability to detect lead compounds is less reliable . In contrast , molecular dynamics simulations can treat both proteins and ligands in a flexible manner , directly estimate the effect of explicit water molecules , and provide more accurate binding affinity , although their computational costs and times are significantly greater than those of molecular docking . Therefore , we developed a special purpose computer “MDGRAPE-3” for molecular dynamics simulations and applied it to computational screening . In this paper , we report an effective method for computational screening; this method is a combination of molecular docking and massive-scale molecular dynamics simulations . The proposed method showed a higher and more stable enrichment performance than the molecular docking method used alone . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"computational",
"biology/molecular",
"dynamics",
"pharmacology/drug",
"development"
]
| 2009 | High-Performance Drug Discovery: Computational Screening by Combining Docking and Molecular Dynamics Simulations |
Lysozymes are ancient and important components of the innate immune system of animals that hydrolyze peptidoglycan , the major bacterial cell wall polymer . Bacteria engaging in commensal or pathogenic interactions with an animal host have evolved various strategies to evade this bactericidal enzyme , one recently proposed strategy being the production of lysozyme inhibitors . We here report the discovery of a novel family of bacterial lysozyme inhibitors with widespread homologs in gram-negative bacteria . First , a lysozyme inhibitor was isolated by affinity chromatography from a periplasmic extract of Salmonella Enteritidis , identified by mass spectrometry and correspondingly designated as PliC ( periplasmic lysozyme inhibitor of c-type lysozyme ) . A pliC knock-out mutant no longer produced lysozyme inhibitory activity and showed increased lysozyme sensitivity in the presence of the outer membrane permeabilizing protein lactoferrin . PliC lacks similarity with the previously described Escherichia coli lysozyme inhibitor Ivy , but is related to a group of proteins with a common conserved COG3895 domain , some of them predicted to be lipoproteins . No function has yet been assigned to these proteins , although they are widely spread among the Proteobacteria . We demonstrate that at least two representatives of this group , MliC ( membrane bound lysozyme inhibitor of c-type lysozyme ) of E . coli and Pseudomonas aeruginosa , also possess lysozyme inhibitory activity and confer increased lysozyme tolerance upon expression in E . coli . Interestingly , mliC of Salmonella Typhi was picked up earlier in a screen for genes induced during residence in macrophages , and knockout of mliC was shown to reduce macrophage survival of S . Typhi . Based on these observations , we suggest that the COG3895 domain is a common feature of a novel and widespread family of bacterial lysozyme inhibitors in gram-negative bacteria that may function as colonization or virulence factors in bacteria interacting with an animal host .
Lysozymes ( EC 3 . 2 . 1 . 17 ) hydrolyse the β- ( 1 , 4 ) glycosidic bond between N-acetylmuramic acid and N-acetylglucosamine in peptidoglycan , the major cell wall polymer in the Bacteria . Peptidoglycan forms a network that surrounds the entire bacterial cell , and its hydrolysis by lysozyme renders bacteria sensitive to lysis driven by turgor pressure . Lysozymes are implicated in defensive and offensive bactericidal systems in a wide range of taxonomically diverse organisms including fungi , protozoa , plants , invertebrate and vertebrate animals and even bacteriophages , indicating their evolutionary success as bactericidal tools . Most gram-negative bacteria are not susceptible to the action of lysozyme alone because their outer membrane prevents access of the enzyme to the peptidoglycan layer . However , this barrier has been overcome in the innate immune systems of animals by the production of accessory antibacterial proteins which permeabilize the outer membrane , such as lactoferrin . In addition , some natural lysozymes as well as chemically or genetically modified hen egg white lysozyme ( HEWL ) have been reported to be active against gram-negative bacteria even in the absence of such permeabilizers [1]–[4] . In view of the widespread occurrence and effectiveness of lysozymes as antibacterial agents , it is not surprising that bacteria have in turn evolved mechanisms to evade or subvert this threat . A bacterial lysozyme resistance mechanism that has been known for long is peptidoglycan modification . Examples are the de-N-acetylation of N-acetylglucosamine in Bacillus subtilis vegetative cells [5] , and O-acetylation of the C-6 hydroxyl group of N-acetylglucosamine residues in Staphylococcus aureus and several other bacteria [6] . In S . aureus , this modification is carried out by a peptidoglycan-specific O-acetyltransferase encoded by oatA , and is believed to contribute greatly to the persistence of pathogenic S . aureus strains on the skin and mucosal surfaces [7] . A different bacterial strategy to evade the bactericidal action of lysozyme that has more recently emerged is the production of lysozyme inhibitors . In group A streptococci , a protein first identified as an inhibitor of the complement system and therefore designated as SIC ( streptococcal inhibitor of complement ) , was later also shown to inhibit lysozyme [8] . However , since SIC does not have a very high affinity for lysozyme ( dissociation constant Kd = 85 . 4 µM ) , and also binds to and inhibits several other components of the innate immune system such as secretory leukocyte proteinase inhibitor and β-defensins at higher affinity [8] , [9] , it can not be considered as a highly specific lysozyme inhibitor . A different lysozyme inhibitor , showing high affinity ( Kd = 1 nM ) , was inadvertently identified during a systematic study of orphan gene products in Escherichia coli [10] . The product of ykfE was shown to strongly bind to and inhibit c-type lysozymes , which include HEWL and human lysozymes , and was accordingly renamed Ivy ( Inhibitor of vertebrate lysozyme ) . Using Ivy-deficient and Ivy-overexpressing E . coli strains , we demonstrated that Ivy contributes to lysozyme resistance of E . coli when the bacteria are simultaneously challenged with lactoferrin or with high hydrostatic pressure to permeabilize their outer membrane [11] , and these findings fed speculations about a possible role for lysozyme inhibitors in bacterial interactions with vertebrate hosts . Pleading against such a role in a wide range of bacteria is the limited distribution of Ivy homologs ( only in a few proteobacterial species ) and in particular their apparent absence in the majority of gram-negative pathogens . However , until now no dedicated function-based screenings for lysozyme inhibitors in bacteria have been reported , and thus the existence of bacterial lysozyme inhibitors different from Ivy can not be excluded . This possibility is supported by our recent observation of lysozyme inhibitory activity in crude cell extracts of Salmonella Typhimurium and S . Enteritidis which do not contain an ivy homolog in their genome ( [12] and unpublished observation ) . In the current paper , we report the identification of this component as a novel type of periplasmic proteinaceous lysozyme inhibitor unrelated to Ivy and we demonstrate that this inhibitor contributes to lysozyme resistance in S . Enteritidis . Furthermore , two other members of the large but cryptic family of proteins with which this novel inhibitor shares a common structural motif are demonstrated to inhibit lysozyme , supporting the functional annotation of this protein family as bacterial lysozyme inhibitors .
In previous work we tested the sensitivity of cell walls of different gram-negative bacteria against several lysozymes [12] . To remove the outer membranes from these cells and make their cell walls accessible to lysozyme , we applied an extraction with chloroform-saturated buffer . A side observation in this work was that this procedure also allowed efficient extraction of the periplasmic lysozyme inhibitor Ivy from E . coli cells since extracts from the wildtype strain showed inhibitory activity against HEWL , while those from the Ivy− strain did not . Interestingly , extracts from S . Typhimurium also showed HEWL inhibition , although S . Typhimurium does not contain an Ivy homolog , nor do any of the other Salmonella serotypes from which a genome sequence is available . This observation was extended to extracts of S . Enteritidis ( data not shown ) . Since we previously purified Ivy by a single HEWL affinity chromatography step to more than 95% purity starting from a periplasmic extract of E . coli overexpressing Ivy from a plasmid [13] , we used the same approach and the same matrix ( HEWL coupled to N-hydroxysuccinimide-activated Sepharose 4 Fast Flow resin ) to isolate the putative lysozyme inhibitor from wildtype S . Enteritidis . When the periplasmic extract obtained from S . Enteritidis ( inhibitory activity of 11 . 6 IU/ml ) was passed over the affinity column , the flow-through fraction did no longer show HEWL inhibitory activity . The elution of the bound proteins , with their corresponding inhibitory activity , is shown in Figure 1 . Two peaks of 27 and 20 milli absorption units were detected at elution volumes of respectively 19 ml and 27 ml , the latter coinciding with a single peak of HEWL inhibitory activity ( 67 IU/ml ) . SDS-PAGE analysis of this active fraction showed only a single band after Coomassie or silver staining ( Figure 1 ) . Material recovered from a Coomassie band was subjected to trypsin digestion and tandem mass spectrometry analysis allowing to identify with high confidence peptides ( MASGANYEAIDK , MASGANYEAIDKNYTYK , TAELVEGDDK and TAELVEGDDKPVLSNCSLAN ) corresponding to fragments of the predicted product of the SEN1802 open reading frame in the genome sequence of S . Enteritidis PT4 ( Wellcome Trust Sanger Institute , Cambridge UK; http://www . sanger . ac . uk/ ) . A SEN1802 homolog is present in S . Typhimurium LT2 and all other sequenced Salmonella genomes ( National Centre for Biotechnology Information; http://www . ncbi . nlm . nih . gov/ ) . The function of this gene product is unknown but it carries a predicted N-terminal signal peptide of 24 amino acids for Sec dependent transport to the periplasm . This prediction is in good agreement with our isolation of the protein from the periplasmic cell fraction and with its supposed activity as a lysozyme inhibitor . SEN1802 has two cysteines in its amino acid sequence for possible disulfide bridge formation , a calculated pI of 4 . 76 and a predicted molecular weight of 9981 Da ( for 90 amino acid residues ) after cleavage of the signal peptide . This is less than our molecular weight estimation from gel migration ( 14 . 4 kDa ) , but such a deviation is not uncommon for acidic proteins and has been ascribed to poor binding of SDS [14] . Because of its HEWL inhibitory activity , we named the protein as PliC ( periplasmic lysozyme inhibitor of c-type lysozyme ) . To investigate the function of PliC in S . Enteritidis , a PliC knock-out ( S . Enteritidis pliC ) and PliC overexpression strain ( S . Enteritidis pliC ( pAA510 ) ) were constructed . The level of PliC production by these strains in comparison to the wildtype strain was evaluated by analyzing the lysozyme inhibitory activity of crude periplasmic protein extracts ( Figure 2 ) . Knock-out of PliC resulted in a strong reduction of inhibitory activity in extracts of S . Enteritidis pliC ( 4 . 3 IU/ml ) compared to wildtype extracts ( 29 . 0 IU/ml ) . Since the open reading frame downstream of pliC has an opposite orientation , this loss of inhibitory activity cannot be due to a polar effect of the knock-out . Introduction of the pAA510 plasmid in S . Enteritidis pliC rescued lysozyme inhibitory activity ( 176 . 5 IU/ml when grown in the presence of 0 . 2% arabinose to induce the cloned pliC gene . These results confirm that the lysozyme inhibitory activity in the periplasmic extracts can be ascribed to the PliC protein . It should be remarked that the inhibitory activity of the wildtype extract in this experiment was higher than in the extract used for chromatographical purification ( 29 . 0 IU/ml versus 11 . 6 IU/ml ) . This is due to variability of yield between different osmotic shock treatments ( data not shown ) . However , the yields of samples that were simultaneously processed in a single osmotic shock treatment were reproducible for a particular strain . Suspensions of late exponential phase wildtype , pliC knock-out and pliC overexpression cells induced with arabinose were treated with 3 . 0 mg/ml lactoferrin , 100 µg/ml lysozyme , or a combination of both , and survivors were enumerated after 24 h ( Figure 3 ) . Most cells survived these treatments very well ( inactivation levels not exceeding twofold ) , except for S . Enteritidis pliC cells in the presence of the lactoferrin - lysozyme mixture , which showed almost 15-fold inactivation . Lactoferrin is known to sensitize gram-negative bacteria to lysozyme and other antibacterial peptides by assisting their penetration through the outer membrane . Although the sensitizing action did not suffice to kill the wildtype S . Enteritidis under the conditions of our experiment , the fact that the pliC knock-out was sensitized demonstrates that natural levels of PliC were sufficient to protect S . Enteritidis cells against lysozyme . An iterative search for sequences similar to the mature PliC protein using Psi-Blast [15] revealed besides the homologs in other Salmonella serotypes , similarity to proteins containing the conserved domain COG3895 ( Clusters of Orthologous Groups , [16] http://www . ncbi . nlm . nih . gov/COG/ ) . Proteins harboring this domain are widespread among members of the Proteobacteria , except the ε-Proteobacteria . Representatives are found in at least 52 different genera of the 155 completely sequenced genomes of all Proteobacteria available as to date ( December 2007 ) and additionally occur in the Acidobacteria , Cyanobacteria and Bacteroides groups . The vast majority of COG3895 proteins are small proteins not containing other conserved protein domains and are predicted to be either periplasmic proteins ( like PliC ) or lipoproteins ( [17] , using the lipoprotein prediction tool available at http://www . mrc-lmb . cam . ac . uk/genomes/dolop/ ) , but their function remains unknown . Also E . coli and Pseudomonas aeruginosa , which already have an active Ivy type lysozyme inhibitor [10] , [18] , encode a COG3895 protein , respectively YdhA and PA0867 . These two proteins are predicted to be anchored to the periplasmic side of the outer membrane [19] , [20] . Because of their homology with PliC of Salmonella and their cellular localization in the bacterial cell , these proteins were renamed as MliC ( membrane-bound lysozyme inhibitor of c-type lysozyme ) . This designation already anticipates on the functionality of these proteins as lysozyme inhibitors which will be demonstrated below . Although both bacteria , like Salmonella , belong to the γ-Proteobacteria , the two predicted MliC proteins share only 32% ( over 53 amino acids ) and 27% ( over 65 amino acids ) identity with PliC , and 38% identity ( over 70 amino acids ) with each other ( Figure 4 ) . Because of this relatively large distance and because a 3-D structure is available for MliC of E . coli ( YdhA , [21] ) , MliC from E . coli and P . aeruginosa were chosen as representatives to further investigate the lysozyme inhibitory activity of the lipoprotein subgroup within the COG3895 group of proteins . mliC from P . aeruginosa and E . coli were cloned under control of an arabinose inducible promoter ( pAA520 and pAA530 respectively ) in an E . coli ivy mliC background , to avoid interference from endogenous E . coli inhibitors . Lysozyme inhibitory activity was measured in the periplasmic extracts and membrane fractions of the overexpression strains after induction and compared to that of the control strain E . coli ivy mliC without overexpression plasmid . No significant differences in lysozyme inhibitory activity were found in the periplasmic protein extracts ( data not shown ) . On the other hand , while only 6 . 3 IU/ml inhibitory activity was detected in the membrane fraction of E . coli ivy mliC , much higher levels of inhibitory activity were measured in the extracts upon induction of MliC expression from P . aeruginosa ( 67 . 6 IU/ml ) or MliC from E . coli ( 40 . 7 IU/ml ) ( Figure 5 ) . Therefore , we can conclude that both MliC of P . aeruginosa and MliC of E . coli are HEWL-inhibitors . It can also be seen in Figure 5 , that knock-out of mliC in E . coli had almost no influence on the level of inhibitory activity of the membrane extracts ( 6 . 7 versus 6 . 3 IU/ml , for an ivy and an ivy mliC strain respectively ) . This is in line with earlier reports that mliC ( previously ydhA ) transcripts of E . coli are not detected under normal laboratory growth conditions [22] . To investigate the actual contribution of the inhibitors to bacterial HEWL resilience , E . coli ivy mliC was rendered sensitive to HEWL by introducing a tolA mutation that increases its outer membrane permeability . The resulting triple mutant was subsequently transformed with different plasmids that enable arabinose induced expression of either Ivy from E . coli ( pAA410 ) , PliC from S . Enteritidis ( pAA510 ) , MliC from P . aeruginosa ( pAA520 ) , and MliC from E . coli ( pAA530 ) . Next , we compared the growth inhibition by HEWL of these strains in the absence and in the presence of arabinose in the medium . At a HEWL concentration of 25 µg/ml , significant differences in optical density ( OD600 ) and in plate counts ( CFU/ml ) of the cultures were observed upon induction of each inhibitor ( Figure 6 ) . Overexpression of Ivy , PliC of S . Enteritidis , MliC of P . aeruginosa or MliC from E . coli increased bacterial growth after 8 hours respectively 9 , 7 , 7 and 5-fold . A control construct ( pAA100 ) containing the gene for green fluorescent protein ( gfp ) in the same vector and E . coli background , showed no significant differences in optical density or plate counts upon induction ( data not shown ) . These results demonstrate that besides Ivy , also at least three members of the newly identified family of lysozyme inhibitors can effectively protect bacterial cells against lysozyme when expressed at appropriate levels .
In this work , we have identified a novel class of lysozyme inhibitors different from Ivy , the lysozyme inhibitor discovered earlier in E . coli [10] . These novel inhibitors belong to a large family of proteobacterial predicted periplasmic proteins or lipoproteins which share a common COG3895 structural motif with unknown function . We demonstrated lysozyme inhibitory activity for one periplasmic ( PliC from S . Enteritidis ) , as well as for two lipoprotein members of this family ( MliC from P . aeruginosa and from E . coli ) . Although no function had hitherto been assigned to any of the COG3895 proteins the 3-D solution structure of MliC from E . coli has been recently resolved , featuring an 8-stranded β-barrel , stabilized by a disulfide bond [21] . At the 3-D level , there is no resemblance with Ivy , which adopts a central β-sheet made of 5 antiparallel β-strands flanked on the convex side by two short helices and on the concave side by an amphipathic helix [18] . The Cys residues engaging in the disulfide bond in MliC from E . coli are conserved in both PliC from S . Enteritidis and MliC from P . aeruginosa , and in the majority of COG3895 proteins , suggesting that they may be important for preserving conformational stability . The existence and possible function of lysozyme inhibitors in bacteria has not received much attention thus far . To our knowledge , a systematic screen for bacterial lysozyme inhibitors has not yet been conducted . This is surprising , given the important role of lysozymes in antibacterial defense in all major eukaryotic lineages , and the extensively documented existence of inhibitors of various other glycosyl hydrolases . Particularly plants produce a wide range of such inhibitors , for example against polygalacturonases , xylanases , α-amylases and β-glucanases , to thwart microbial attack . Therefore , the discovery in this work of a novel class of bacterial lysozyme inhibitors and the wide distribution of homologs of these inhibitors in the Proteobacteria may be indicative for their functional importance , for example in bacteria-host interactions . The location of the bacterial lysozyme inhibitors either in the periplasm ( Ivy and PliC from S . Enteritidis ) , or anchored to the luminal face of the outer membrane ( MliC from E . coli and P . aeruginosa ) is also consistent with a role in protecting peptidoglycan from hydrolysis by exogenous lysozymes . In at least one instance more direct evidence for a role in host interaction exists . In Salmonella Typhi , expression of the mliC homolog was induced in cells residing within macrophages and knockout of mliC reduced macrophage survival [23] . Macrophages are known to produce a battery of antibacterial peptides including lysozyme and membrane permeabilizers , and hence the production of one or more lysozyme inhibitors by intracellular pathogens like S . Typhi makes sense from this point of view . The observed increased lysozyme sensitivity of an S . Enteritidis pliC knockout in the presence of 3 . 0 mg/ml of the outer membrane permeabilizing protein lactoferrin ( Figure 3 ) provides a relevant indication in this context . Lactoferrin concentrations in this range occur in secretions like tears , airway mucus or colostrum [24] , [25] , [26] . Moreover , Ivy and all three new HEWL-inhibitors identified in this study suppressed growth inhibition by HEWL when overexpressed in an E . coli MG1655 tolA ivy mliC strain ( Figure 6 ) . The genomic context of the newly identified lysozyme inhibitor genes also provides some interesting clues about their possible function . Immediately upstream of pliC of S . Typhimurium are the genes pagC , pagD and msgA , which play a role in macrophage survival of S . Typhimurium . Furthermore , transcriptome analysis has revealed that expression of pliC is controlled by SlyA , the same transcriptional activator that controls expression of pagC and pagD and that is necessary for virulence [27] . Based on its low GC content , the region encompassing pagC and a number of its immediate upstream genes was suggested to be acquired by lateral gene transfer , as is often the case for virulence genes [28] . The pliC gene , which is immediately downstream of pagC , also has a markedly lower GC content ( 42 . 0% ) than the average of the LT2 chromosome ( 52 . 2% ) , and thus probably is an integral part of this acquired genome fragment . Interestingly , the mliC gene is located downstream of slyA in all sequenced Salmonella strains . Furthermore , both in E . coli and in Salmonella , mliC or its homolog are adjacent to ydhH , an open reading frame recently renamed to anmK because it encodes an anhydro N-acetyl muramic acid kinase involved in recycling of murein [29] . This allows speculation on a possible role of MliC in murein recycling , for example by controlling excessive hydrolysis of the murein backbone by lytic transglycosylases . However , at present we do not know whether the latter enzymes are inhibited by MliC or any of the other COG3895 proteins . C-type lysozymes ( e . g . HEWL or human lysozyme ) are the major lysozymes produced by most vertebrates . In addition , all vertebrates have genes encoding g-type lysozyme . While the importance of the latter is not clear in man , it is the dominant type in some birds and it also occurs in fish species . A third type of lysozyme , called i-type , is characteristic for invertebrate animals such as arthropods , molluscs , nematodes etc . [30] . Neither PliC from S . Enteritidis , nor MliC from E . coli or P . aeruginosa have inhibitory activity against g-type lysozyme from goose egg white ( data not shown ) . Ivy , in contrast , is active against goose egg white lysozyme [13] but not against g-type lysozyme from the urochordate Oikopleura dioica and i-type lysozyme from the scallop Chlamys islandica [31] . Given the existence and widespread occurrence of two types of c-type-specific lysozyme inhibitors in Proteobacteria , we anticipate that additional inhibitor classes specific against other types of lysozymes are also likely to be produced in bacteria . Screening of crude periplasmic extracts of a diverse range of bacteria for inhibitory activity against these g- and i-type lysozymes seems to corroborate this assumption ( unpublished results ) , but definitive confirmation will have to await isolation and identification of the putative inhibitors . The possible effect of bacterial lysozyme inhibitors in bacterial pathogenesis may even extend beyond neutralizing the direct antibacterial effect of lysozyme . Peptidoglycan has recently emerged as a powerful effector of the innate immune system through interaction with specific host receptors . The actual elicitor molecules are specific muropeptide fragments derived from peptidoglycan by bacterial and/or host lytic enzymes [32] , [33] . This system of pattern recognition is believed to allow the host to distinguish pathogenic from non-pathogenic bacteria and to maintain its immune functions at an appropriate level . Malfunctioning of this system has been linked to chronic immune-related diseases such as inflammatory bowel disease and Crohn's disease . By interfering with the fragmentation of peptidoglycan by host lysozymes , bacterial lysozyme inhibitors can be anticipated to influence this system , and thus to play a potential role in these immune related pathologies . Provided that their role in bacterial pathogenesis can be further substantiated , bacterial lysozyme inhibitors may constitute an attractive new target for the development of anti-inflammatory and/or immunomodulating drugs . In conclusion , we have identified a novel family of bacterial lysozyme inhibitors that contribute to bacterial lysozyme resistance and that have widespread homologs in gram-negative bacteria . Further study of these inhibitors will not only improve our understanding of bacteria-host interactions , lysozyme inhibitors may also turn out to be interesting novel targets for drug development .
Bacterial strains and plasmids used in this study are listed in Table 1 . Construction of mutants and plasmids is discussed in Text S1 . Where appropriate , plasmids were transformed to bacteria by electroporation . All strains were originally cultured on Luria Bertani ( LB; 10 g/l trypton , 5 g/l yeast extract , 5 g/l NaCl ) agar plates and incubated at 37°C for 21 h . Overnight broth cultures were obtained by inoculating a single colony into LB broth containing appropriate antibiotics and incubating at 37°C for 21 h with aeration . Antibiotics ( Sigma-Aldrich , Bornem , Belgium ) were added when necessary to obtain the following final concentrations: 100 µg/ml ampicillin , 50 µg/ml kanamycin or 20 µg/ml chloramphenicol . For the purification of PliC , 500 ml cultures of S . Enteritidis ATCC 13076 were grown on a rotary shaker to stationary phase ( 21 h , shaking at 200 rpm ) in LB at 37°C . Periplasmic cell extracts were then prepared by a gentle cold osmotic shock procedure as described earlier [13] , and stored at −20°C until further analysis . Lysozyme binding inhibitors were isolated from this periplasmic cell fraction on an ÄKTA-FPLC platform ( Amersham Pharmacia Biotech , Upsalla , Sweden ) by affinity chromatography using immobilized HEWL as a ligand as described earlier for the Ivy protein [13] , except that 100 ml of crude extract was loaded rather than 25 ml , and fractions of 5 . 0 ml rather than 2 . 0 ml were collected . The fractions were collected in tubes containing 300 µl of 1 . 0 M Tris-HCl pH 8 . 0 to neutralize the high pH of the elution buffer ( pH 12 . 0 ) , and bovine serum albumin ( BSA , Sigma-Aldrich ) was added to a final concentration of 0 . 5 mg/ml to stabilize the purified protein unless the samples were used for SDS-polyacrylamide gel electrophoresis ( SDS-PAGE ) . Fractions were then desalted by dialysis against 10 mM potassium phosphate buffer pH 7 . 0 ( 12 kDa cut off , Sigma-Aldrich ) and stored at −20°C until further use . After purification , protein purity was checked with SDS-PAGE , conducted according to [34] with a 15% separating gel and a 4% stacking gel . Samples were prepared by boiling for 3 min in the presence of 1% ß-mercaptoethanol and 1% SDS . Gels were stained with Coomassie blue R 250 ( Sigma-Aldrich ) , and , if higher sensitivity was desired , destained and subsequently silver-stained following the procedure of [35] . For the isolation of MliC of P . aeruginosa or E . coli , cultures of E . coli ivy mliC harboring plasmid pAA520 or pAA530 respectively , were grown overnight at 37°C in LB with ampicillin ( 100 µg/ml , Sigma-Aldrich ) , diluted 1/100 in fresh LB without antibiotics , induced with 0 . 2% ( w/v ) L− ( + ) -arabinose after 4 hours of growth , and further incubated at 37°C until stationary phase . Portions of 200 ml were subsequently harvested , resuspended in 10 ml 10 mM Tris-HCl pH 8 and lysed by three cycles of freezing and thawing followed by sonication ( 3×3 min , amplitude 40% , pulses 5 s on/5 s off ) . These suspensions were centrifuged for 1 hour at 100 . 000×g ( 4°C ) . The resulting pellet was resuspended in 10 ml 10 mM Tris-HCl buffer ( pH 6 . 8 ) supplemented with 1 . 0 M NaCl , and sedimented again as described above . The membrane-bound proteins were then extracted using 2% Triton X-100 in a 10 mM Tris-HCl buffer ( pH 6 . 8 ) supplemented with 10 mM MgCl2 and 150 mM NaCl and separated from insoluble material by centrifugation ( 1 hour at 100 . 000×g , 4°C ) . Active fractions containing the purified inhibitor protein were lyophilized , redissolved and subjected to SDS-PAGE and Coomassie staining . A gel fragment from the band corresponding to the inhibitor was trypsin-digested according to the method of [36] , and the digests were then analyzed by electrospray tandem mass spectrometry on a LCQ Classic ( ThermoFinnigan , San Jose , California ) ion trap mass spectrometer equipped with a nano-liquid chromatography column switching system and a nanoelectrospray device . Tandem mass spectrometry data were searched using MASCOT ( Matrix Sciences , London , U . K . ) and SEQUEST ( ThermoFinnigan ) against the GenBank non-redundant protein database . Freeze-dried M . lysodeikticus ATCC4698 cells ( Sigma-Aldrich ) were resuspended at 0 . 5 mg/ml either in appropriate dilutions of the bacterial crude extracts , purified column fractions or in potassium phosphate buffer ( 10 mM , pH 7 . 0 ) with 0 . 5 mg/ml Bovine Serum Albumine ( BSA ) for the controls . Thirty µl of 66 U/ml HEWL ( Hen Egg White Lysozyme; Fluka , 66000 U/mg protein ) in potassium phosphate buffer ( 10 mM , pH 7 . 0 ) was then added to 270 µl of these suspensions and cell lysis was followed during 2 h at 25°C as the decrease in optical density ( OD600 ) using a Bioscreen C Microbiology Reader ( Labsystems Oy , Helsinki , Finland ) . In the absence of inhibitor , this procedure resulted in a linear OD600 decrease of 0 . 27 ± 0 . 04 over 2 h . The percentage inhibition ( I ) for each column fraction was calculated as:with L0 − L , R0 − R and B0 − B representing the OD600 decrease over a period of 2 h of the M . lysodeikticus suspensions respectively in the presence of lysozyme but with buffer instead of a bacterial extract/column fraction , in the presence of the bacterial extract/column fraction and lysozyme , and in the presence of the bacterial extract/column fraction but with buffer instead of lysozyme . Inhibitory activity was expressed in inhibitory units , with one unit being the amount of inhibitor that is needed to decrease the lysozyme activity by 50% under the above assay conditions . S . Enteritidis , S . Enteritidis pliC and S . Enteritidis pliC ( pAA510 ) cultures were grown overnight in LB with ampicillin and/or chloramphenicol when appropriate , diluted 1/100 in fresh LB without antibiotics , induced with 0 . 2% ( w/v ) L− ( + ) -arabinose ( Fluka , Buchs , Switzerland ) , and incubated further . Arabinose served only to induce pliC expression from plasmid pAA510 , but was also added to cultures of strains not carrying this plasmid to ensure identical culture conditions for all strains in the experiment . At an optical density ( OD600 ) of 0 . 6 ( 5 . 108 ± 1 . 108 CFU/ml ) , cells were harvested by centrifugation ( 3800×g , 5 min ) and subsequently resuspended in the same volume of Tris-HCl buffer ( 10 mM; pH 7 . 0 ) without and with lactoferrin ( gift from Morinaga Milk Industries , Kanagawa , Japan; 3 . 0 mg/ml final concentration ) and/or HEWL ( Fluka , 66000 U/mg protein; 100 µg/ml final concentration ) . Samples were serially diluted in sterile Tris-HCl buffer ( 10 mM; pH 7 . 0 ) at the beginning and after 24 hours of treatment , and plated on LB agar plates to determine the degree of inactivation . Inactivation was expressed as a viability reduction factor , No/N , with No and N respectively the colony counts at the beginning and after 24 hours of treatment . Precultures of E . coli MG1655 tolA ivy mliC harboring plasmid pAA410 , pAA510 , pAA520 , or pAA530 were grown overnight in LB broth containing ampicillin , kanamycin and chloramphenicol . Subsequently , cultures were diluted ( 1/100 ) in duplicate in fresh LB containing ampicillin , and after three hours of growth ( exponential phase ) , either H2O or 0 . 02% L− ( + ) -arabinose was added , resulting in control and induced precultures respectively . These cultures were further grown to stationary phase to allow inhibitor expression . Subsequently , test tubes containing 4 ml LB with ampicillin , and either water or 0 . 02% L− ( + ) -arabinose and 25 µg/ml HEWL were inoculated ( 1/100 ) with the control and induced E . coli precultures respectively . These cultures were grown at 37°C during 10 hours . Each hour the OD600 was determined using a Multiscan RC ( Thermo Scientific , Zellik , Belgium ) . After 8 hours the viable cell number was enumerated by plating on LB agar . From E . coli MG1655: ivy ( before ykfE ) : 946530 ( Gene Entrez ) , mliC ( before ydhA ) : 946811 ( Gene Entrez ) , tolA: 946625 ( Gene Entrez ) ; From P . aeruginosa: mliC ( before PA0867 ) : 882238 ( Gene Entrez ) ; From Salmonella Enteritidis: pliC: SEN1802 ( http://www . sanger . ac . uk/ ) . | Lysozyme is an ancient bactericidal enzyme that is part of the antibacterial defense system of vertebrate and invertebrate animals . Bacteria colonizing or infecting an animal host have developed various ways to overcome lysozyme action , a recently proposed mechanism being the production of lysozyme inhibitors . However , the only high affinity bacterial lysozyme inhibitor known thus far is produced only in few bacteria , and this raised questions about their wider relevance in bacteria–host interactions . We here report the discovery of a novel and distinct family of bacterial lysozyme inhibitors that is widely distributed among the Proteobacteria , including several major pathogens . The family comprises periplasmic as well as membrane-bound inhibitors , and both types contribute to lysozyme tolerance of bacterial cells , as we experimentally demonstrate for the periplasmic inhibitor from Salmonella Typhimurium and the membrane-bound inhibitors from Escherichia coli and Pseudomonas aeruginosa . Interestingly , a gene encoding one of the newly identified inhibitors has been previously found to promote macrophage survival of Salmonella Typhi . The widespread occurrence of lysozyme inhibitors in bacteria is likely to reflect their functional importance in a wide range of bacteria–host interactions . As such , they are also attractive novel targets for antibacterial drug development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis"
]
| 2008 | A New Family of Lysozyme Inhibitors Contributing to Lysozyme Tolerance in Gram-Negative Bacteria |
Tuberculosis is a global health problem and at least one-third of the world’s population is infected with Mycobacterium tuberculosis ( MTB ) . MTB is a successful pathogen that enhances its own intracellular survival by inhibiting inflammation and arresting phago-lysosomal fusion . We previously demonstrated that Toxoplasma gondii ( T . gondii ) dense granule antigen ( GRA ) 7 interacts with TNF receptor-associated factor 6 via Myeloid differentiation primary response gene 88 , enabling innate immune responses in macrophages . To extend these studies , we found that GRA7 interacts with host proteins involved in antimicrobial host defense mechanisms as a therapeutic strategy for tuberculosis . Here , we show that protein kinase C ( PKC ) α-mediated phosphorylation of T . gondii GRA7-I ( Ser52 ) regulates the interaction of GRA7 with PYD domain of apoptosis-associated speck-like protein containing a carboxy-terminal CARD , which is capable of oligomerization and inflammasome activation can lead to antimicrobial defense against MTB . Furthermore , GRA7-III interacted with the PX domain of phospholipase D1 , facilitating its enzyme activity , phago-lysosomal maturation , and subsequent antimicrobial activity in a GRA7-III ( Ser135 ) phosphorylation-dependent manner via PKCα . Taken together , these results underscore a previously unrecognized role of GRA7 in modulating antimicrobial host defense mechanism during mycobacterial infection .
Tuberculosis ( TB ) is an infectious disease caused by Mycobacterium tuberculosis ( MTB ) [1] . The World Health Organization reported that in 2014 , 9 . 6 million cases and 1 . 5 million deaths were globally [2] . Recent developments in TB drug-development strategies ( including new and repurposed antimicrobials and host-directed drugs ) have produced new regimens to shorten treatment duration , improve outcomes of TB treatment such as , prevent resistance , reduce lung injury by promoting autophagy , antimicrobial peptide production , and other macrophage effector mechanisms , as well as inhibiting mechanisms causing lung inflammation and matrix destruction [1 , 3–5] . A wide range of candidate host-directed therapies ( HDTs ) -including new and repurposed drugs , biologics , and cellular therapies-have been proposed to accelerate eradication of infection and overcome the problems associated with current treatment regimens . Recent studies have revealed the intracellular signaling pathways that govern the outcome of the innate immune response to mycobacteria infection and antibacterial defense [6–11] . First , the NLRP3 inflammasome complex , an intracellular protein complex consisting of the sensor NACHT , LRR and PYD domains-containing protein 3 ( NLRP3 ) , the adaptor apoptosis-associated speck-like protein containing a carboxy-terminal CARD ( ASC ) , and pro-caspase-1 regulates IL-1β and IL-18 processing [10–12] . Jayaraman et al . showed that IL-1β directly promotes antimicrobial immunity in murine and human macrophages by regulating TNFR signaling and caspase-3 activation against MTB infection [10] . Verway et al . showed that 1-25-Dihydroxyvitamin D ( 1 , 25D ) enhances IL-1β signaling from MTB-infected macrophages , inducing antimicrobial peptide DEFB4/HBD2 in primary lung epithelial cells , which in turn helps control MTB [11] . Second , host phospholipids play a critical role in the activation of the antimicrobial innate immune response [13] . Phospholipase D ( PLD ) , which has two isoforms ( PLD1 and PLD2 ) catalyzes the hydrolysis of the membrane phospholipid , phosphatidylcholine , to generate the metabolically active phosphatidic acid ( PA ) [14] . PLD1 is activated by arf- , ral- , and rho-family GTPases , and protein kinase C ( PKC ) α , while PLD2 activity is elevated by fatty acids [15] . Interestingly , MTB , unlike the nonpathogenic M . smegmatis , inhibits PLD activation during phagocytosis , a process that is associated with intracellular survival of the pathogen [6] . Garg et al . showed that Natural lysophospholipids promote MTB-induced in vitro PLD-dependent phagolysosome maturation and PLD-dependent intracellular killing of MTB in human macrophages [8] and the type II alveolar epithelial cell line A549 [9] . Third , recent studies have highlighted the role of protein kinases in the biology and pathogenesis of mycobacteria . The members of the PKC-family of proteins are classified into three groups , based on the mechanisms regulating their activation in response to different stimuli [7 , 16] . Holm et al . showed that PKCα regulates phagocytosis and the biogenesis of phagolysosomes by promoting the interaction of phagosomes with late endosomes and lysosomes [16] . Furthermore , PKCα also plays an important role in the killing of MTB in human macrophages [7] . Collectively , these infection-induced signaling pathways suggest possibilities for the development of novel therapeutic modalities for tuberculosis that target the intracellular signaling pathways permitting the replication of this nefarious pathogen . However , the roles of MTB-infection signal-dependent HDTs involved in host innate immune responses and their regulatory mechanisms have not yet been fully elucidated . In a previous study , we demonstrated that T . gondii GRA7/MyD88-dependent NF-κB activation is essential for the activation of TNF receptor-associated factor 6 ( TRAF6 ) and ROS generation , and enhances the release of inflammatory mediators . We also found that GRA7 stimulation led to physical and functional associations between GRA7 and TRAF6 , resulting in crucial protective efficacy against T . gondii infection in vivo [17] . It remains to be seen whether GRA7 targeting can be used as a therapeutic strategy for infectious diseases . In this study , we further investigated the intracellular regulatory network of T . gondii GRA7-induced ASC , PLD1 , and PKCα signaling pathways to help identify novel therapeutic modalities for tuberculosis . We found that the PKCα-mediated phosphorylation of GRA7 was essential for interaction between GRA7 and ASC or PLD1 , which contributes to antimicrobial defense against MTB in vitro and in vivo . Our findings demonstrate that GRA7-I and -III play fine-tuning roles in the activation of HDTs and innate immune machineries through direct binding with ASC or PLD1 and may provide a unique opportunity for urgently needed therapeutic interventions against tuberculosis .
All animal experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee of Hanyang University ( protocol 2014–0207 ) and Bioleaders Corporation ( Daejeon , Korea , protocol BLS-ABSL3-13-11 ) . All animal experiments were performed in accordance with Korean Food and Drug Administration ( KFDA ) guidelines . Cultures of MTB H37Rv ( provided by Dr . R . L . Friedman , University of Arizona , Tucson , AZ ) were prepared as described previously [1] . The effective concentration of lipopolysaccharide was <50 pg/ml in those experiments , with a bacterium-to-cell ratio of 10:1 . For all assays , mid-log phase bacteria ( absorbance 0 . 4 ) were used . Bacterial strains were divided into 1-ml aliquots and stored at -70°C . Wild-type C57BL/6 mice were purchased from Orient Bio ( Gyeonggi-do , Korea ) . PKCα-/- ( B6;129-Prkcatm1Jmk/J , 009068 ) and PLD1-/- ( B6 . Cg-Pld1tm1 . 1Gbp/J , 028665 ) mice were obtained from Jackson Laboratory . All animals were maintained in a specific pathogen-free environment . HEK293T cells ( ATCC-11268; American Type Culture Collection ) were maintained in DMEM ( Invitrogen ) containing 10% FBS ( Invitrogen ) , sodium pyruvate , nonessential amino acids , penicillin G ( 100 IU/ml ) , and streptomycin ( 100 μg/ml ) . Human monocytic THP-1 ( ATCC TIB-202 ) cells were grown in RPMI 1640/glutamax supplemented with 10% FBS and treated with 20nM PMA ( Sigma-Aldrich ) for 24 h to induce their differentiation into macrophage-like cells , followed by washing three times with PBS . Primary bone marrow–derived macrophages ( BMDMs ) were isolated from C57BL/6 mice and cultured in DMEM for 3–5 d in the presence of M-CSF ( R&D Systems , 416-ML ) , as described previously [12] . For in vitro experiments , cells were infected with MTB for 2–4 h . Then , cells were washed with PBS to remove extracellular bacteria , supplied with fresh medium , and incubated at 37°C for indicated time points . For in vivo experiments , C57BL/6 mice were i . v . injected with MTB ( 1×106 CFU/mouse ) . After 3 wks of infection , mice injected intraperitoneally with rGRA7 proteins for 7 consecutive days . After 1 wk of treatment , mice were sacrificed for harvesting of the lungs , spleens , and livers . Mice were maintained in biosafety level 3 laboratory facilities . CIP ( P4978 ) and DMSO were purchased from Sigma-Aldrich . PKCα ( C2-4 ) inhibitor peptide ( 17478 ) was purchased from Cayman Chemical . Flag-PKCα , -β , -δ , and -ξ plasmids were a generous gift from Dr . D . Zhou ( Xiamen University , China ) . The GST-tagged GRA7 and truncated mutant genes were described previously [17] . V5-tagged AC or AU1-PLD1 and truncated mutant genes were cloned into the XbaI and BamHI sites in pcDNA3 . 0 . All constructs were sequenced using an ABI PRISM 377 automatic DNA sequencer to verify 100% correspondence with the original sequence . Specific antibodies against phospho- ( Thr147 ) -PLD1 ( 3831 ) , phospho- ( Ser561 ) -PLD2 ( 3834 ) , PLD1 ( 3832 ) , PLD2 ( 13904 ) , PKCα ( 2056 ) , PKCγ ( 43806 ) , and NLRP4 ( 12421 ) were purchased from Cell Signaling Technology . Antibodies specific for actin ( I-19 ) , ASC ( N-15-R ) , IL-18 ( H-173-Y ) , TRAF6 ( H-274 ) , caspase-1 p10 ( M-20 ) , Rab5 ( D-11 ) , Rab7 ( H-50 ) , LAMP1 ( E-5 ) , LAMP2 ( H4B4 ) , Tubulin ( B-5-1-2 ) , Calnexin ( H-70 ) , FACL4 ( N-18 ) , VDAC ( B-6 ) , His ( His17 ) , V5 ( C-9 ) , Flag ( D-8 ) , and GST ( B-14 ) were purchased from Santa Cruz Biotechnology . AU1 ( GTX23402 ) and PKCβI ( A10-F ) were purchased from GenenTex and Antibodies-online Inc . , respectively . IL-1β ( AF-401-NA ) and NLRP3 ( AG-20B-0014 ) were from R&D Systems and Adipogen , respectively . THP-1 , 293T , and BMDMs were treated as indicated and processed for analysis by Western blotting , co-immunoprecipitation , and GST pulldown as previously described [17 , 18] . Immunofluorescence analysis was performed as described previously [1] . The cells were fixed on coverslips with 4% ( w/v ) paraformaldehyde in PBS and then permeabilized for 10 min using 0 . 25% ( v/v ) Triton X-100 in PBS at 25°C . PLD1 or His was detected using a 1/100 dilution of the primary Ab for 1 h at 25°C . After washing , the appropriate fluorescently labeled secondary Abs were incubated for 1 h at 25°C . Slides were examined using laser-scanning confocal microscopy ( model LSM 800; Zeiss ) . For colocalization analysis , the co-distribution of the PLD1 and GRA7 were quantified and validated statistically by Pearson coefficient , as specified by the ZEN 2009 software ( version 5 . 5 SP1; Zeiss ) . Peptide arrays were synthesized using the SPOTs synthesis method and spotted onto a derivatized cellulose membrane ( Intavis ) in the presence of [γ-32P]ATP and calcium as described previously [19 , 20] . Peptide spot phosphorylation was quantified using phosphoimaging . For immunohistochemistry of tissue sections , mouse lungs were fixed in 10% formalin and embedded in paraffin . Paraffin sections ( 4 μm ) were cut and stained with hematoxylin and eosin ( H&E ) [21] . PLD activity was measured using the Amplex Red PLD assay kit ( Molecular Probes , A12219 ) according to the manufacturer’s protocol . The resulting fluorescence was detected using a fluorescence microplate reader at an excitation of 530 nm and an emission of 590 nm . The recombinant GRA7 protein was described previously [17] . GRA7s from amino acid residues 26–80 , 26-80S52A , 26-80S52D , 120-150S135A and 120-150S135D were cloned with an N-terminal 6xHis-tag into the pRSFDuet-1 Vector ( Novagen ) and induced , harvested , and purified from E . coli expression strain BL-21 DE-3 pLysS , as described previously [17 , 22] , following standard protocols recommended by Novagen . Supplemental experimental procedures and supplemental references . All data were analyzed by Student’s t-test with Bonferroni adjustment or ANOVA for multiple comparisons , and are presented as mean ± SD . Grubbs’ test was used for evaluating the outliers . Differences were considered significant at p <0 . 05 .
To establish a role for GRA7 in intracellular signaling pathways as a therapeutic strategy for infectious diseases in macrophages , we investigated whether GRA7 interacts with molecules involved in innate immunity . GRA7 complexes were subjected to co-immunoprecipitation ( co-IP ) of recombinant GRA7 protein with THP-1 lysates . The purified GRA7 complexes retrieved several endogenous proteins selectively , as identified by mass spectrometry analysis , including PLD1 ( 124 K ) , PKCα ( 76 K ) , ASC ( 21 K ) , and TRAF6 ( 60 K ) ( Fig 1A and S1 Fig ) . Endogenous co-IP showed that GRA7 interacted strongly , although temporarily ( from 15 to 60 min ) , with endogenous PLD1 , TRAF6 , and ASC but not with PLD2 , NLRP3 , or NLRC4 after stimulation with rGRA7 in THP-1 cells , and vice versa ( Fig 1B , S2A and S2B Fig ) . As previously reported [17] , GRA7 associated with TRAF6 , based on their molecular weights and co-IP ( Fig 1A–1C , S1 and S2A Figs ) . To determine the mechanism by which rGRA7 interact with the intracellular protein , THP-1 cells were preincubated with cytochalasin D , which inhibits actin polymerization . Pretreatment with cytochalasin D completely blocked the phagocytic activities of rGRA7 and it binding with intracellular proteins ( S2C and S2D Fig ) . Structurally , GRA7 contains a signal sequence , N-terminal domains ( I–IV ) , a transmembrane , and a C-terminal domain ( V ) ( Fig 1C ) [17] . In 293T cells , detailed mapping using various mammalian glutathionine S-transferase ( GST ) -GRA7 fusions and truncated mutants of V5-ASC indicated that the N-terminal I-domain ( aa26-80 ) of GRA7 exhibited only minimal binding affinity to ASC and ASC carrying the N-terminal PYD domain ( aa1-91 ) bound GRA7 as strongly as ASC WT ( Fig 1C and 1D ) . GST pull-down assays using truncated mutants of GST-GRA7 mammalian fusions and AU1-PLD1 showed that the N-terminal III-domain ( aa120-150 ) of GRA7 is required for its interaction with PX ( aa81-212 ) of PLD1 ( Fig 1C and 1E and S2E Fig ) , indicating that the interactions of GRA7 with ASC , PLD1 , and TRAF6 are genetically separable ( Fig 1 ) . These results show that GRA7 interacts with ASC and PLD1 through its N-terminal I- and III-domains in macrophages , respectively . In addition to ASC and PLD1 binding , GRA7 also interacted with PKCα . Endogenous co-IP revealed a robust interaction between GRA7 and PKCα , but not PKCβI or PKCγ , after stimulation with rGRA7 in THP-1 cells , and vice versa ( Fig 2A ) . A large-scale proteomics analysis of the human kinome [23] and computational sequence analysis [24] predicted five PKC phosphorylation residues ( S52LR , T121DR , S135FK , T204TR , S209PR ) within the GRA7 N-terminal I , III domains , and C-terminal V-domain . To confirm that GRA7 was phosphorylated by PKCα , we used several strategies . First , we performed Phos-tag gel electrophoresis , which involves the use of a Phos-tag biomolecule that specifically binds phosphorylated proteins and retards their migration in the gel [25] . The results showed that GRA7 of wild-type , I- , and III-domains migrated more slowly and produced an ‘up-shifted’ band ( as visualized by the Phos-tag labeling system for the analysis of phosphorylation , followed by SDS-PAGE ) when co-expressed with PKCα , but when co-expressed with PKCβ , PKCδ , or PKCξ ( Fig 2B and S3A Fig ) . Furthermore , we performed an in vitro phosphorylation assay using purified recombinant PKCα and a non-biased overlapping peptide array covering the entire GRA7 sequence [19 , 20] . From GRA7 , two peptides ( 49PVDSLRPTNAGVDSK73 and 121TDRKVVPRKSEGKRS135 ) showed a phosphorylation signal >200 PSL/mm2 ( Fig 2C ) . In contrast , none of the peptides spanning the C-terminal of GRA7 showed a significant phosphorylation signal . GRA7 has serine/threonine residues , and two peptides of GRA7 that were phosphorylated contained three potential phosphorylation sites ( Fig 2C and 2D ) . Interestingly , the specific point mutation forms ( IS52A and IIIS135A ) of GRA7 markedly decreased phosphorylation in Phos-tag gel electrophoresis , whereas the mutant IIIT121A of GRA7 did not ( Fig 2B ) . These results indicate that PKCα can specifically phosphorylate S52 and S135 residues of GRA7 , demonstrating that GRA7 is a substrate of PKCα . We next investigated whether phosphorylation of S52 and S135 of GRA7 was necessary for binding with ASC and PLD1 , respectively . The point mutation ( IS52A and IIIS135A ) of GRA7 markedly abolished its interaction with ASC and PLD1 , suggesting that this interaction is S52- and S135-phosphorylation dependent ( Fig 2D and S3B Fig ) . Furthermore , because phosphomimetic residues ( aspartic acid or glutamic acid ) do not fully approximate the electronegativity produced by phosphorylation , we employed the strategy of mutating amino acids to overcome the charge differential [20 , 26] . The GST pull-down assay showed that phosphomimetic mutants of GRA7 strongly bound to ASC and PLD1 compared to GRA7 WT , indicating that mimicking constitutively phosphorylated GRA7 overrode the need for PKCα function in the innate immune pathway . Consistent with the findings shown in Fig 2A–2D , GRA7 interaction with ASC and PLD1 was markedly decreased in BMDMs from PKCα-/- mice , THP-1 from knock down with shRNA specific for PKCα ( Fig 2E and S3C Fig ) , and BMDMs treated with a pharmacological inhibitor of PKCα upon rGRA7 stimulation ( S3D Fig ) . Taken together , these data indicate that PKCα-mediated phosphorylation of GRA7 at Ser52 or Ser135 is essential for interactions between GRA7 and ASC or PLD1 , respectively . To examine the role of T . gondii GRA7-I in innate immune responses of macrophages , we generated bacterially purified His-tagged GRA7-I and its mutant proteins , as described previously [17 , 22] . The purified rGRA7-I ( 10 kDa ) was confirmed through SDS-PAGE and immunoblotting analysis ( Fig 3A ) . No significant difference compared to vector control observed for rGRA7-induced cytotoxicity in macrophages [17] . We showed previously that rGRA7-induced expression of pro-inflammatory cytokine genes and proteins including IL-1β , in macrophages [17] and NLRP3 inflammasomes involves a multimeric protein complex containing NLRP3 interacting the adaptor ASC and caspase-1 to induce the maturation of IL-1β and IL-18 [10 , 12] . To investigate the role of GRA7-I in the regulation of inflammasome activation , BMDMs from PKCα+/+ and PKCα-/- mice were stimulated with rGRA7-I and its mutant proteins . In response to rGRA7-I and -WT , PKCα-deficient BMDMs showed significantly attenuated IL-1β and IL-18 production than WT BMDMs , but the phosphomimetic mutant ( IS52D ) -induced markedly increased secretion of IL-1β and IL-18 ( Fig 3B and S4A Fig ) . Consistent with these results , the caspase-1 activation and IL-1β and IL-18 maturation observed in response to rGRA7-I and -WT proteins were significantly decreased in PKCα-deficient BMDMs , and the constitutively active form ( IS52D ) of GRA7 ‘rescued’ the PKCα deficiency ( Fig 3C ) . Notably , the PKCα non-phosphorylatable mutant ( IS52A ) and shRNA-mediated reduction of endogenous ASC expression led to significant attenuation of IL-1β and IL-18 production ( Fig 3B and S4B Fig ) in an ASC-binding dependent manner . Next , we determined whether ASC is substantially oligomerized and if the intracellular formation of ASC specks is dependent on GRA7-I interaction . In correlation with secretion of active caspase-1 and IL-1β , PKCα-deficient BMDMs showed markedly attenuated ASC oligomerization and speck formation compared to WT BMDMs , but the phosphomimetic mutant ( IS52D ) markedly increased both ( Fig 3D and 3E and S4C Fig ) . Further , the intracellular interaction of GRA7 and ASC was confirmed by their co-localization after stimulation with rGRA7 . Subcellular fractionation and co-IP analysis showed that GRA7-I associated with ASC and PKCα in the mitochondrial fraction in PKCα+/+ BMDMs . Notably , these binding patterns were increased by the phosphomimetic mutant ( IS52D ) in BMDMs from PKCα+/+ and PKCα-/- mice ( S4D Fig ) . These data suggest that GRA7-I acts as a positive regulator of ASC-dependent inflammasome activation via PKCα in mitochondria . IL-1β and IL-18 are cytokines that play crucial roles in host defense and inflammation [11 , 12] . We first measured caspase-1 activation and maturation of IL-1β and IL-18 induced by rGRA7 and its mutants in MTB-infected macrophages . rGRA7-I treatment increased inflammasome activity in MTB-infected macrophages in a dose , ASC-binding and PKCα phosphorylation-dependent manner . Importantly , treatment with the phosphomimetic mutant ( IS52D ) of GRA7 markedly amplified inflammasome activity in MTB-infected conditions in BMDMs from PKCα+/+ and PKCα-/- mice ( Fig 4A ) . The PKCα non-phosphorylatable mutant ( IS52A ) and the shRNA-mediated reduction of endogenous ASC expression led to significant attenuation of caspase-1 activation and maturation of IL-1β and IL-18 ( Fig 4A and 4B ) in an ASC-binding dependent manner . IL-1β directly activates MTB–infected macrophages to restrict intracellular bacterial replication [10 , 27] . We examined whether rGRA7-induced antimicrobial activity was dependent on ASC-dependent inflammasome activation via PKCα in macrophages . The rGRA7-WT and -I-induced antimicrobial responses against MTB were significantly downregulated in BMDMs from PKCα-/- mice and cells transduced with shASC in a dose-dependent manner ( Fig 4C and 4D ) . Notably , the PKCα non-phosphorylatable mutant ( IS52A ) of GRA7 did not induced antimicrobial responses of MTB , compared with the WT- and rGRA7-I treatment , in BMDMs from PKCα+/+ mice . The phosphomimetic mutant ( IS52D ) of GRA7 markedly increased antimicrobial responses to MTB in dose-dependent manner , indicating that the constitutively active form ( IS52D ) of GRA7 partially ‘rescued’ the PKCα deficiency . No significant difference was observed for MTB growth in 7H9 broth with or without rGRA7 ( S5 Fig ) , indicating that ASC-dependent inflammasome-derived IL-1β controls the outcome of MTB infection and is functionally linked via PKCα in macrophages . To examine the role of T . gondii GRA7-III in innate immune responses by macrophages , we generated bacterially purified His-tagged GRA7-III and its mutant proteins , as described previously [17 , 22] . The purified rGRA7-III ( 5 kDa ) was confirmed through SDS-PAGE and immunoblotting analysis ( Fig 5A ) . As GRA7 associates with PLD1 but not PLD2 ( Fig 1 and S1 Fig ) , we sought to determine whether PLD1 activation by GRA7-III was regulated by phosphorylation events in many cellular processes [15 , 28] . PLD1 activity is regulated by phosphorylation of Thr147 in the PX domain and Ser561 in the negative regulatory loop region of PLD1 by PKCα [15 , 28] . We first measured rGRA7-III-induced phosphorylation of PLD1 at Thr147 and Ser561 but not the PKCα non-phosphorylatable mutant ( IIIS135A ) of GRA7 in macrophages . Importantly , the phosphomimetic mutant ( IIIS135D ) of GRA7 treatment markedly amplified PLD1 activation in BMDMs from PKCα+/+ and PKCα-/- mice ( Fig 5B ) . Consistent with these results , PLD activity was significantly decreased by the PKCα non-phosphorylatable mutant ( IIIS135A ) and increased by the phosphomimetic mutant ( IIIS135D ) of GRA7 in BMDMs from PKCα+/+ and PKCα-/- mice ( Fig 5C ) . However , PLD activity was at the basal level in PLD1-/- macrophages with the phosphomimetic mutant , indicating that phosphorylated GRA7-III interacted with activated PLD1 by PKCα and stimulated its enzymatic activity through the phosphorylation of PLD1 Thr147 and Ser561 . Further , the intracellular interaction of GRA7 and PLD1 was confirmed by their co-localization after stimulation with rGRA7 , as documented by immunostaining and image overlay ( Fig 5D and S6 Fig ) . GRA7-III localized with PLD1 and PKCα in the cytoplasm , appearing as small speckles and punctate spots . Notably , these co-localization patterns were increased by the phosphomimetic mutant ( IS135D ) in BMDMs from PKCα+/+ and PKCα-/- mice , but not PLD1-/- mice . These data suggest that GRA7-III acts as a positive regulator of PLD1 activation via PKCα in macrophages . PLD1 activity regulates the actin cytoskeleton , vesicle trafficking for secretion and endocytosis , and receptor signaling . With the emerging concept of dynamic cycling of PLD1 inside the cell , some of the varying reports of localization may be due to differential rates and numbers of vesicles cycling in the cell lines used and thus differential regulation of PLD1 localization [14 , 15 , 28] . MTB preferentially infects alveolar macrophages , although mycobacteria allow only early endosome membrane fusion and induce phagosome arrest by selective Rab GTPase recruitment to avoid fusion with late endosomes and lysosomes [29 , 30] . To investigate the subcellular fractionation of PLD1 , we treated MTB-infected BMDMs with rGRA7 and its mutants , and then examined the induction of protein levels of Rab5 , Rab7 , LAMP1 , and LAMP2 regulators of phagosomal maturation in mycobacteria-containing phagosome fractions ( phagosome and phago-lysosome ) subsequently purified by sucrose-step-gradient-ultra-centrifugations . Interestingly , GRA7-induced MTB-containing phagosomes were recruited to late endosome and lysosome marker Rab7 , LAMP1 , and LAMP2 , indicating that GRA7 facilitates mycobacterial phagosome-lysosome fusion in macrophages in a PKCα- and PLD1-dependent manner ( Fig 6A ) . Furthermore , GRA7 associated with PLD1 in phagosomal fractions in a binding-dependent manner , and the phosphomimetic mutant ( IS135D ) of GRA7 markedly increased phagosomal trafficking and binding to PLD1 , indicating that the constitutively active form ( IS135D ) of GRA7 ‘rescued’ PKCα deficiency ( Fig 6B ) . Consistently , the viability and growth rate of intracellular MTB decreased following treatment with the phosphomimetic mutant ( IS135D ) in BMDMs from PKCα+/+ and PKCα-/- mice , but not PLD1-/- mice in dose-dependent manner ( Fig 6C and 6D ) . These results collectively indicate that GRA7 facilitates phagosomal maturation through interactions with PLD1 and thereby , exerts marked control of bacterial killing activity against intracellular mycobacteria in a binding-dependent manner via PKCα . Drawing on the observation that GRA7-I and -III associate with ASC and PLD1 , respectively , and which contributes to antimicrobial defense against MTB in macrophages ( S7A Fig ) , we next evaluated the in vivo efficacy of rGRA7 and its binding mutants in a mouse model of established tuberculosis [31] . MTB-infected mice were given rGRA7 and its mutants , starting three weeks after infection . Mice treated with rGRA7-WT alone , rGRA7-IWT+IIIWT , or rGRA7-IS53D+IIIS135D , but not the binding deficient mutant ( rGRA7-IS53A+IIIS135A ) had significantly reduced bacillary load in the lung , liver , and spleen , and reduced formation of lung granulomatous lesions in size and number of foci , compared with vector-treated mice ( Fig 7A and 7B ) . Notably , the phosphomimetic mutant ( IS53D+IIIS135D ) of rGRA7 drastically reduced bacillary load , at a level similar to rGRA7-WT in PKCα+/+ and PKCα-/- mice . We further investigated the effect of GRA7-III on in vivo host responses to MTB infection . As shown in Fig 7C and 7D , treatment with PLD1-binding domain ( IIIWT ) and the phosphomimetic mutant ( IIIS135D ) of rGRA7 markedly increased bacterial killing effects and number of granulomatous foci in PLD1+/+ mice , but not PLD1-/- mice . Treatment with the PLD1-binding deficient mutant ( IIIS135A ) of rGRA7 had no significant effect on bacterial killing or granulomatous lesions in either PLD1+/+ or PLD1-/- mice , indicated that the anti-mycobacterial effect of GRA7-III acts in a PLD1-binding dependent manner via PKCα in vivo . However , No significant difference was observed for inflammation score in lung ( S7B and S7C Fig ) . The pharmacokinetics of therapeutic rGRA7 proteins were localized in alveolar macrophages was maintained for up to 7 days and gradually cleared until 25 days was studied by the fluorescence of the fluorophore Alexa 488-conjugated with the proteins ( S8 Fig ) . These results unambiguously show that host defenses against MTB infection are substantially affected by GRA7-I and GRA7-III .
The central finding of this study is that the PKCα-mediated phosphorylation of T . gondii GRA7 is essential for the interaction between GRA7 and ASC or PLD1 , which contributes to antimicrobial defense against MTB ( S9 Fig ) . Specifically , we found that ( 1 ) PKCα specific phosphorylation of Ser52 and Ser135 of GRA7 in vitro and in vivo was functionally required for ASC and PLD1 interactions with GRA7 , respectively , ( 2 ) GRA7 was a novel substrate of PKCα , ( 3 ) the N-terminal of GRA7 ( GRA7-I ) was sufficient for interaction with the PYD domain of ASC in mitochondria , leading to ASC oligomerization and inflammasome activation , and subsequent antimicrobial activity , ( 4 ) GRA7-III interacted with the PX domain of PLD1 in cytosol , facilitating its enzyme activity , phago-lysosomal biogenesis , and subsequent antimicrobial activity , ( 5 ) GRA7-I and -III-dependent host protective effects against MTB infection were demonstrated in vivo , and ( 6 ) a phosphomimetic mutant that constitutively activated GRA7 ‘rescued’ PKCα deficiency both in vitro and in vivo . Collectively , these observations indicate that T . gondii GRA7-mediated HDTs leading to an antimicrobial response , as a novel host defense mechanism may provide a unique opportunity for urgently needed therapeutic intervention strategies for TB and other infectious diseases . Although it is well established that dense granule protein GRA7 is important for immunodiagnosis of toxoplasmosis in patients [32 , 33] , new candidates for further effective vaccine development against T . gondii infection is the need [17 , 34 , 35] . Recent reports showed that GRA7 is associated with T . gondii ROP5 was required for efficient phosphorylation of Irga6 and additional component of the ROP5/ROP18 kinase complex [22 , 36] and binding of ROP2 and ROP4 was shown [37] in T . gondii . However , the modulation of host innate immunity by GRA7 in the early phases of infection is critical for the establishment of both the initial invasion and the subsequent maintenance of latent infection is have not been fully elucidated . Growing evidence suggests that host-pathogen interactions have led to the coevolution of toxoplasmosis-causing T . gondii with its host [17 , 22 , 38] . GRA7 binds to poly ( rC ) binding protein 1/PCBP1 along with PCBP2 and hnRNPK , corresponding to the principal cellular poly ( rC ) binding proteins according to yeast two-hybrid analysis . PCBP1 plays a part in the formation of a sequence-specific α-globin mRNP complex that is associated with the stability of α-globin mRNA [38] . Additionally , GRA7 directly binds to the active dimer of Irga6 in a GTP-dependent manner . The binding of GRA7 to Irga6 led to enhanced polymerization , rapid turnover , and eventual disassembly , which contributed to acute virulence in the mouse [22] . We recently showed that the GRA7-V ( aa 201–236 ) domain led to physical and functional associations with TRAF6 . Furthermore , GRA7-V-induced Th1 immune responses and protective efficacy were crucial for T . gondii infection in vivo [17] . In this study , we showed that host cell ASC , PLD1 , and PKCα bind to GRA7 . The GRA7 protein interacted with a number of host cell proteins including enzymes , and a broad spectrum of structural and functional subcellular organellar proteins revealing a new facet of the role of GRA7 in the regulation of innate host immune responses . Our results correlate with those of previous studies showing that T . gondii is a novel activator of NLRP1 and NLRP3 inflammasomes by activating caspase-1 , an enzyme that mediates cleavage and release of the proinflammatory cytokines IL-1β and IL-18 in vitro and in vivo , thereby establishing a role for these sensors in host resistance to toxoplasmosis [39–41] . Furthermore , Millholland et al . showed that a Gα subunit ( Gα ) q-coupled host-signaling cascade is required for the egress of T . gondii . Gαq-coupled signaling results in PKC-mediated loss of the host cytoskeletal protein adducin and weakening of the cellular cytoskeleton . This cytoskeletal compromise induces catastrophic Ca2+ influx mediated by the mechanosensitive cation channel TRPC6 , which activates host calpain that in turn proteolyzes the host cytoskeleton allowing parasite release [42] . T . gondii induces prostaglandin E2 biosynthesis in macrophages by regulating arachidonic acid production through a Ca2+-dependent pathway and induction of cyclooxygenase-2 expression by a PKC-dependent pathway [43 , 44] . Reinforcing the feasibility of targeting host proteins as an antiparasitic strategy , mammalian PKC inhibitors demonstrate activity in murine models of toxoplasmosis . In this study , we focused on the role of GRA7-I and -III-dependent innate immunity . Future studies will aim to clarify the precise molecular mechanisms of GRA7 and GRA7-II and -IV-related signaling pathways in inflammatory responses and host defense . HDTs aim to modulate immune responses in the TB lung [45 , 46] . Neutralization of pro-inflammatory cytokines such as IL-6 , TNF-α , VEGF , and IFN-α/β , as well as anti-inflammatory IL-4 , during severe pulmonary disease may help reduce ongoing parenchymal damage in the MTB-infected lung [27 , 45–47] . Alternatively , suboptimal activation of anti-TB immune responses due to regulatory T cell activity can be reversed by the use of the anti-cancer drug cyclophosphamide . Drugs with anti-TB potential , such as metformin , imatinib , ibuprofen , zileuton , valproic acid , and vorinostat as well as nutraceuticals such as 1 , 25D , may not only abate the bacterial burden via host-dependent mechanisms , but also fine-tune the immune response to MTB . These drugs increase phagocytosis of extracellular bacteria , improve emergency myeloid response , and increase autophagic and apoptotic killing of bacteria , subsequently editing the T cell response in favor of the host . Immune checkpoint inhibition with blockade of the PD-1/PD-1 ligand 1 , CTLA-4/cytotoxic T lymphocyte-associated antigen 4 , LAG3/lymphocyte-activation gene 3 , and TIM3/T cell immunoglobulin pathways may improve the quality of the cellular immune response to MTB epitopes , as seen in cancer immunotherapy [4 , 5 , 45–47] . Our results partially correlate with those of previous studies showing that host-directed immunotherapy with clinically approved drugs that augment prostaglandin E2 level prevents acute mortality of MTB-infected mice . Thus , IL-1 and type I IFNs represent two major counter-regulatory classes of inflammatory cytokines that control the outcome of MTB infection and are functionally linked via eicosanoids [27] , and IL-1β either directly or via enhancement by 1 , 25D promotes antimicrobial immunity against MTB infection [10 , 11] . Greco et al . showed that PKC-mediated Ca2+ mobilization , PLD activity , and ( auto ) phagolysosome maturation represent effector processes induced by apoptotic body-like liposomes carrying PA that concur with the intracellular killing of MTB [14] . The MTB-containing phagosomes is involved in arresting phagosome maturation and inhibiting phagolysosome biogenesis [6 , 8 , 9] , however , rGRA7-induced PKCα regulates phagocytosis , PLD-dependent the biogenesis of phagolysosomes ( Rab5 conversion to Rab7 ) by promoting the interaction of phagosomes with late endosomes and lysosomes , and Rab7 regulated phagosomal acidification , which is important for the killing of MTB in human macrophages [7 , 16] . Our current observations based on the study of GRA7-III co-localized with PLD1 and PKCα in the cytoplasm ( Fig 5D and S6 Fig ) have the proposal the localized on phagolysosomes , appearing as speckles and punctate spots , because of an artifact of rGRA7 overexpression . Further studies are needed to localization organelle population . The rGRA7 have a function of biologicals as potential therapeutics . However , these rGRA7 do not fulfil the requirements of direct anti-mycobacterial agent , which represent feasible alternatives to conventional chemotherapy to TB , due to the still unclear specificity and selectivity does not enable linking the effects of rGRA7s to host immune systems , as well as limitation of animal experimental model , unknown off-target effects , pharmacokinetics , safety data , and their potential feasibility for in vivo proof-of-concept studies . Further analyses are required to find out whether rGRA7s can be translated to the in vivo situation or be observed in the presence of physiological condition to patient with TB . In conclusion , we provide evidence of a critical role of PKCα-mediated phosphorylation of T . gondii GRA7 in the interaction between GRA7 and ASC or PLD1 , which contributes to antimicrobial defense against MTB ( S9 Fig ) . GRA7-I and -III-dependent host protective effects worked against MTB infection in vivo , and a phosphomimetic mutant that constitutively activated GRA7 ‘rescued’ PKCα deficiency . These observations reveal a new role for GRA7 in regulating innate immune responses in host protective immunity . Our findings establish proof of concept for HDT strategies that manipulate host GRA7-mediated immune networks . Further studies are needed to develop more effective GRA7-based potential therapeutic targets and to understand how GRA7 regulates host defense strategies against TB and other infectious diseases . | We previously demonstrated that Toxoplasma gondii ( T . gondii ) dense granule antigen ( GRA ) 7 interacts with TRAF6 via MyD88 , enabling innate immune responses in macrophages and effective protection against T . gondii infection in vivo . However , its exact role and how it regulates host innate immune responses have not been fully explained . Herein , we show that PKCα-mediated phosphorylation of GRA7 is essential for the interaction between GRA7 and ASC or PLD1 , which can promote antimicrobial defense against Mycobacterium tuberculosis ( MTB ) . Notably , PKCα specifically phosphorylated Ser52 and Ser135 of GRA7 in vitro and in vivo , indicating that GRA7 is a substrate of PKCα . The N-terminal of GRA7 ( GRA7-I ) was sufficient for interaction with the PYD domain of ASC , which is capable of ASC oligomerization and inflammasome activation . Furthermore , GRA7-III interacted with the PX domain of PLD1 , facilitating its enzyme activity , phago-lysosomal maturation , and subsequent antimicrobial activity in a GRA7 phosphorylation-dependent manner . Interestingly , phosphomimetic mutation in GRA7 overcame the need for PKCα . Collectively , these results provide novel insight into how GRA7 can promote ASC and PLD1 activation in a PKCα-dependent manner as an antimicrobial host defense mechanism . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"blood",
"cells",
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"intracellular",
"pathogens",
"toxoplasma",
"gondii",
"pathogens",
"immunology",
"parasitic",
"protozoans",
"protozoans",
"toxoplasma",
"bacteria",
"contractile",
"proteins",
"actins",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"actinobacteria",
"pathogenesis",
"immune",
"response",
"biochemistry",
"cytoskeletal",
"proteins",
"cell",
"biology",
"post-translational",
"modification",
"mycobacterium",
"tuberculosis",
"host-pathogen",
"interactions",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"organisms"
]
| 2017 | Toxoplasma gondii GRA7-Targeted ASC and PLD1 Promote Antibacterial Host Defense via PKCα |
West Nile virus ( WNV ) and Rift Valley fever virus ( RVFV ) are two emerging arboviruses transmitted by Culex pipiens species that includes two biotypes: pipiens and molestus . In Lebanon , human cases caused by WNV and RVFV have never been reported . However , the introduction of these viruses in the country is likely to occur through the migratory birds and animal trades . In this study , we evaluated the ability of Cx . pipiens , a predominant mosquito species in urban and rural regions in Lebanon , to transmit WNV and RVFV . Culex egg rafts were collected in the West Bekaa district , east of Lebanon and adult females of Cx . pipiens were experimentally infected with WNV and RVFV Clone 13 strain at titers of 1 . 6×108 and 1 . 33×107 plaque forming units ( PFU ) /mL , respectively . We estimated viral infection , dissemination and transmission at 3 , 7 , 14 and 19 days post infection ( dpi ) . Results showed that infection was higher for WNV than for RVFV from 3 dpi to 19 dpi . Viral dissemination and transmission started from 3 dpi for WNV; and only from 19 dpi for RVFV . Moreover , Cx . pipiens were able to excrete in saliva a higher number of viral particles of WNV ( 1028 ± 405 PFU/saliva at 19 dpi ) than RVFV ( 42 PFU/saliva at 19 dpi ) . Cx . pipiens from Lebanon are efficient experimental vectors of WNV and to a lower extent , RVFV . These findings should stimulate local authorities to establish an active entomological surveillance in addition to animal surveys for both viruses in the country .
West Nile virus ( WNV ) and Rift Valley fever virus ( RVFV ) are two important emerging mosquito-borne zoonotic agents transmitted by Culex pipiens , a complex of sibling species that includes Cx . pipiens s . s . , Cx . quinquefasciatus and possibly Cx . australicus [1 , 2] . Cx . pipiens s . s . includes two biotypes or subspecies: Cx . pipiens pipiens and Cx . pipiens molestus [1 , 3] . The first biotype is primarily a bird-feeding mosquito present in temperate areas while Cx . pipiens biotype molestus feeds on mammals ( mainly human ) and thrives in sewers in temperate and sub-tropical regions [3 , 4] . Because morphological identification of these biotypes is not possible , they can only be distinguished using molecular techniques [3–5] . WNV is a member of the Flavivirus genus ( Flaviviridae family ) and was first isolated in Uganda in 1937 [6] . Usually , only 20% of infected individuals develop symptoms and less than 1% of infected people develop serious and potentially fatal neurological illnesses such as meningitis and encephalitis . Birds are considered the main reservoir of the virus and Cx . pipiens is recognized as one of the primary enzootic vector [7] . WNV infections have been reported in many tropical and temperate countries in Africa , Europe , Asia and America . The Middle East and North Africa ( MENA ) region has been long considered as a WNV-endemic area [8 , 9] . Locally acquired cases have been recently reported in Israel [10] , Greece [11] , Turkey [12] , and Italy [13] . In addition , evidence of WNV circulation has been reported in Jordan [14] and Egypt [15] . The introduction of WNV into the United States in 1999 , which constitutes a turning point in WNV epidemiology , is thought to have originated from Israel following introduction from Africa [16 , 17] . RVFV belongs to Phlebovirus genus ( Bunyaviridae family ) . It was first identified in Kenya in 1931 [18] . This virus usually affects livestock and causes abortion . The main enzootic vectors belong to the Aedes genus [19] . However , several Culex species , including Cx . pipiens are considered secondary vectors and contribute to the transmission of RVFV to humans [19] . Infected people can be asymptomatic or develop a mild febrile disease . In less than 10% of cases , people may develop more severe symptoms such as encephalitis and hemorrhagic fever . RVFV was responsible for numerous outbreaks among animals and humans in Sub-Saharan Africa [20] up to Mauritania [21] but also in Egypt [22] . In the Middle East , epizootics were reported in Saudi Arabia and Yemen [23] . In Lebanon , Cx . pipiens is a predominant mosquito species besides another vector of arboviruses , Aedes albopictus [24 , 25] . Cx . pipiens colonizes urban and rural habitats whereas Ae . albopictus is mostly present in the densely populated coastal fringe . Local Ae . albopictus are competent to transmit Chikungunya virus and to a lesser extent , Dengue virus [26] . The vector competence of local populations of Cx . pipiens to transmit WNV and RVFV has never been evaluated . Diseases caused by these two viruses have never been reported in Lebanon . Nevertheless , a serological study conducted in a main hospital in the capital city of Beirut , confirmed the presence of neutralizing WNV antibodies in blood donors [27] . In fact , Lebanon is situated in a WNV-endemic area and located on the flyways of migratory birds with potential introduction of the virus into the country . Moreover , Lebanon is geographically close to Yemen and Saudi Arabia , regions where RVFV had circulated actively . Intensive livestock trade between Lebanon and these countries increases the risk of RVFV introduction . Here , we assess the vector competence of local populations of Cx . pipiens towards WNV and RVFV . We estimate viral infection , dissemination and transmission at different days after experimental infections .
Culex egg rafts were sampled in June 2015 in Bab Mareh , in the West Bekaa district , a sub-humid , agricultural area in east of Lebanon with large stagnant water systems ( Fig 1 ) . Egg rafts were collected on the water surface in an artificial basin and placed individually in a tube containing 30 mL of water collected from the breeding site . Collected egg rafts were shipped to the Laboratory of Arboviruses and Insect Vectors ( AIV ) at the Institut Pasteur , Paris . They were reared until the adult stage . Adults emerging from each raft were morphologically identified and only Cx . pipiens species were retained for this study . Two viruses were used in this study: WNV lineage 1 strain isolated from a horse in Camargue ( France ) in 2000 [9] and an avirulent RVFV strain Clone 13 isolated from a human case in Bangui ( Central African Republic ) in 1974 [28] . After passages on Vero ( E6 ) cells ( ATCC cell lines ) , both viruses were produced on C6/36 mosquito cells . Viral stocks were stored at -80°C until use . The infectious blood meal was composed of a viral suspension ( 1:3 ) diluted in washed rabbit erythrocytes ( New Zealand White rabbit , Charles River ) collected at the day of mosquitoes infection . A phagostimulant ( ATP ) was added at a final concentration of 5 mM . Virus titer in the blood meal was 1 . 6×108 plaque forming units ( PFU ) /mL for WNV and 1 . 33×107 PFU/mL for RVFV . The susceptibility of Lebanese Cx . pipiens mosquitoes to WNV and RVFV was tested on F0 and F1 generation respectively . Ten-to-twelve day-old female Cx . pipiens mosquitoes were left to starve for 48 h in Biosafety Level 3 ( BSL3 ) insectary at 28±1°C with 80% relative humidity and a 16h:8h photoperiod . Females were then allowed to feed for one hour through a chicken skin membrane ( obtained from a commercially purchased chicken ) covering the base of a capsule of the feeding system ( Hemotek ) containing the blood-virus mixture maintained at 37°C . Fully engorged females were sorted , then transferred in cardboard containers and maintained with 10% sucrose at 28±1°C until examination . Around 20 female mosquitoes were tested at 3 , 7 , 14 and 19 days post-infection ( dpi ) . For each mosquito , saliva was collected using the forced salivation technique [29] . Briefly , mosquitoes were chilled , their legs and wings removed and the proboscis was inserted into 20 μL tip filled with 5 μL of Fetal Bovine Serum ( FBS ) . After 45 min , medium containing the saliva was expelled into 0 . 2 mL tube containing 45 μL of Dulbecco’s MEM ( DMEM ) medium . Collected saliva and the remaining mosquito bodies were conserved at -80°C for further analysis . In order to assess the ability of both viruses to invade and cross the midgut barrier , the infection rate ( IR ) and the dissemination efficiency ( DE ) were determined . IR reflects the proportion of female mosquitoes with infected bodies ( thorax and abdomen including the midgut ) among tested specimens while DE is the proportion of female mosquitoes with infected head ( detection of the virus having succeeded to reach the mosquito general cavity ) among tested ones . Thus , heads and bodies were separated and ground each in 300 μL DMEM supplemented with 3% FBS . After centrifugation , the supernatant of each homogenate was conserved at -80°C . Then , 20 μL of each sample were diluted in 180 μL DMEM supplemented with 2% FBS and distributed in serial dilutions from 10−1 to 10−3 in duplicates on Vero cell monolayers ( 3 . 105 cells/well ) in 96-well plates . After incubation at 37°C for 6 days , inoculum was removed and the cells were fixed and stained using a crystal violet solution ( 0 . 2% in 10% formaldehyde and 20% ethanol ) . After washing , the presence or absence of cytopathic effect was noted . The capacity of the WNV and RVFV to cross the salivary glands barrier was evaluated by determining the transmission efficiency ( TE ) which corresponds to the proportion of female mosquitoes that secrete infectious saliva among tested specimens . The number of infectious particles within collected saliva samples was estimated on Vero cell culture and expressed as PFU/saliva . Briefly , 20 μL of each saliva were diluted in 280 μL DMEM 2% FBS . The total volume was inoculated on a monolayer of Vero cells ( 8 . 105cells/well ) in six-well plates . Cells were incubated at 37°C for 6 days under an overlay consisting of DMEM , 2% FBS , 1% antibiotic-antimycotic mix and 1% agarose . The lytic plaques were counted after staining with a crystal violet solution . Proportions ( IR , DE and TE ) were compared using Fisher’s exact test and sample distributions ( number of viral particles ) with the Kruskal-Wallis test . Statistical analyses were conducted using the Stata software ( StataCorp LP , Texas , and USA ) . P-values<0 . 05 were considered significant .
Collected egg rafts were hatched in laboratory conditions and provided 480 adult females of Culex pipiens ( F0 generation ) ; of those only 174 ( 36 . 25% ) had successfully fed on a WNV-infected blood . A batch of 600 F1 female mosquitoes was used for the RVFV infection assay . Of those , only 91 ( 15 . 16% ) had successfully fed on infected blood . Infection rate ( IR ) for each virus was estimated by determining the number of infected bodies ( abdomen and thorax ) among all engorged female mosquitoes examined at each dpi ( 3 , 7 , 14 and 19 ) ( Fig 2A ) . For WNV , IRs were very high ( 94 . 7–100% ) from 3 to 19 dpi . For RVFV , IRs were much lower: at 3 dpi , the IR was 44 . 0% and increased gradually to reach 65 . 0% at 14 dpi and 64 . 3 at 19 dpi . The detection of viral particles in mosquito heads allowed estimating the ability of the virus to disseminate from the midgut to internal organs . For WNV , dissemination efficiency ( DE ) increased from 31 . 6% ( 3 dpi ) to 94 . 7% ( 19 dpi ) ( Fig 2B ) . For RVFV , virus was only detected at 19 dpi with a DE of 21 . 4% ( Fig 2B ) . The ability of mosquitoes to transmit the virus was measured by detecting viral particles in saliva expectorated by mosquitoes . With WNV , transmission efficiencies were much higher than with RVFV ( Fig 2C ) . TE increased gradually from 10 . 5% at 3 dpi to 68 . 4% at 19 dpi ( Fisher’s exact test: p < 10−4 ) . To note , TE decreased slightly but not significantly from 86 . 4% at 14 dpi to 68 . 4% at 19 dpi ( Fisher’s exact test: p = 0 . 17 ) . For RVFV , viral particles were only detected in saliva at 19 dpi with a TE of 7 . 1% ( Fig 2C ) . Mosquitoes were able to deliver an average of 550 ( ±450 ) PFU/saliva at 3 dpi with WNV , which increased to reach 1004 ( ±442 ) PFU/saliva at 7 dpi ( Fig 3 ) . Despite a decrease at 14 dpi , the viral load remained high at 19 dpi with 1028 ( ±405 ) PFU/saliva . For RVFV , only one female had infectious particles in saliva at 19 dpi with a viral load of 42 PFU ( Fig 3 ) .
Culex pipiens is the most widely distributed mosquito species in Lebanon and is suspected to transmit WNV and RVFV in several countries [7 , 19] . Using experimental infections , we showed that Cx . pipiens populations collected from West Bekaa , Lebanon were susceptible to infection by these two viruses and ensured efficient transmission of WNV and to a lesser extent , RVFV . Cx . pipiens was capable to ensure viral infection , dissemination and transmission starting from 3 dpi . Most mosquitoes exposed to the infectious blood-meal were infected as IRs reached 100% at the 4 dpi examined ( 3 , 7 , 14 and 19 dpi ) . Dissemination and transmission were slightly lower suggesting that not all infected mosquitoes were able to transmit WNV . Mosquitoes delivered more than 500 viral particles in saliva from 3 dpi . On the other side , infections with RVFV present different patterns: lower IR , DE and TE . Only 21% of mosquitoes were able to ensure viral dissemination at 19 dpi and 7% were able to transmit at 19 dpi . This suggests a significant role of the midgut and the salivary glands as respective barriers to the release of viruses into the body cavity and their excretion in saliva [30] . In this manner , Cx . pipiens was less susceptible to RVFV than to WNV . Overall , Cx . pipiens can transmit experimentally both viruses but the time interval between the ingestion of the viremic blood-meal and the ability to transmit the virus termed the extrinsic incubation period ( EIP ) was 3 days for WNV . For RVFV , we only found one mosquito able to transmit the virus 19 days after ingestion . It is likely that more females would have been able to transmit the virus if more mosquitoes were able to feed on RVFV-infected blood . Cx . pipiens from Tunisia showed similar EIP of 3 days with WNV and a much shorter EIP of 3 days with RVFV [31] underlining the significant role of mosquito genotype in specific interactions between mosquito and virus genotypes; these interactions promoting adaptation of viral lineages to specific mosquito vector genotypes influence the outcome of transmission [32] . In addition , when viral dose increases in blood meals , transmission efficiency also increases suggesting that hosts presenting a high viremia may infect more mosquitoes [33] . Animals susceptible to RVFV can develop very high viremia , higher than 1010 . 1 MIPLD50 ( mouse intraperitoneal 50% lethal dose/mL ) in lambs [34] . Then , at higher titers of blood meal , RVFV may infect more mosquitoes . In conclusion , the predominant Cx . pipiens mosquito in Lebanon is susceptible to both viruses , WNV and RVFV . As Lebanon is located in a region where WNV and RVFV can be potentially introduced ( respectively through migratory birds and animal trades ) , local health authorities should establish an active surveillance to detect any new human cases in addition to reinforce the entomological surveillance allowing an early viral detection in field-collected mosquitoes . | West Nile virus ( WNV ) and Rift Valley fever virus ( RVFV ) are two emerging mosquito-borne arboviruses mainly transmitted by Culex mosquitoes . WNV considered one of the most important causative agent of viral encephalitis has a wide distribution in many tropical and temperate countries including the Middle East . RVFV is mainly distributed in Sub-Saharan Africa but epizootics were also reported in Egypt , Saudi Arabia and Yemen . The mosquito vector belongs to the Culex pipiens species which includes two biotypes: pipiens and molestus . Both biotypes are the most widely distributed mosquitoes in Lebanon . Using experimental infections of mosquitoes , our study showed that Cx . pipiens populations collected in West Bekaa were susceptible to infection by these two viruses and ensured efficient transmission of WNV and to a lesser extent , RVFV . Our findings may help to prepare a control strategy more adapted to these mosquito vectors . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"rift",
"valley",
"fever",
"virus",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"geographical",
"locations",
"microbiology",
"viral",
"structure",
"saliva",
"animals",
"viruses",
"rna",
"viruses",
"insect",
"vectors",
"bunyaviruses",
"lebanon",
"infectious",
"diseases",
"medical",
"microbiology",
"microbial",
"pathogens",
"disease",
"vectors",
"insects",
"virions",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"eukaryota",
"west",
"nile",
"virus",
"asia",
"flaviviruses",
"anatomy",
"blood",
"virology",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"organisms"
]
| 2018 | Experimental transmission of West Nile Virus and Rift Valley Fever Virus by Culex pipiens from Lebanon |
Segregation of chromosomes during the first meiotic division relies on crossovers established during prophase . Although crossovers are strictly regulated so that at least one occurs per chromosome , individual variation in crossover levels is not uncommon . In an analysis of different inbred strains of male mice , we identified among-strain variation in the number of foci for the crossover-associated protein MLH1 . We report studies of strains with “low” ( CAST/EiJ ) , “medium” ( C3H/HeJ ) , and “high” ( C57BL/6J ) genome-wide MLH1 values to define factors responsible for this variation . We utilized immunofluorescence to analyze the number and distribution of proteins that function at different stages in the recombination pathway: RAD51 and DMC1 , strand invasion proteins acting shortly after double-strand break ( DSB ) formation , MSH4 , part of the complex stabilizing double Holliday junctions , and the Bloom helicase BLM , thought to have anti-crossover activity . For each protein , we identified strain-specific differences that mirrored the results for MLH1; i . e . , CAST/EiJ mice had the lowest values , C3H/HeJ mice intermediate values , and C57BL/6J mice the highest values . This indicates that differences in the numbers of DSBs ( as identified by RAD51 and DMC1 ) are translated into differences in the number of crossovers , suggesting that variation in crossover levels is established by the time of DSB formation . However , DSBs per se are unlikely to be the primary determinant , since allelic variation for the DSB-inducing locus Spo11 resulted in differences in the numbers of DSBs but not the number of MLH1 foci . Instead , chromatin conformation appears to be a more important contributor , since analysis of synaptonemal complex length and DNA loop size also identified consistent strain-specific differences; i . e . , crossover frequency increased with synaptonemal complex length and was inversely related to chromatin loop size . This indicates a relationship between recombination and chromatin compaction that may develop as DSBs form or earlier during establishment of the meiotic axis .
Recombination is a defining event of meiosis , resulting in the physical exchange of DNA between homologous chromosomes . It is generally thought that this is essential for proper alignment and subsequent segregation of homologs during meiosis I and , indeed , evidence from yeast [1] , [2] , Caenorhabditis elegans [3] , Drosophila melanogaster [4] , and mammals [5] indicates that alterations in the number or positioning of recombination events increase the likelihood of meiotic nondisjunction . For this reason , it might be expected that recombination levels are strictly regulated but , surprisingly , substantial inter-individual variation in recombination is observed in most mammalian species . For example , linkage studies have demonstrated extensive variation in recombination rates and/or recombination hotspot usage in human males and females ( e . g . , [6] , [7] , [8] ) , and cytological studies of recombination indicate 15–25% individual differences in genome-wide recombination levels in rhesus and human males [9] , [10] , and similar levels of variation in different inbred strains of mice [11] , [12] . Further , the frequency and location of exchanges vary between the sexes; e . g . , in humans the female genetic map is approximately 1 . 6 fold longer than that of males , with interstitial exchanges being more common in females than males ( e . g . , [13] , [14] , [15] ) . These differences beg an obvious question: what is responsible for the variation in recombination levels observed among individuals or inbred strains ? Arguably , there are at least three different time-points in the recombination process at which variation in recombination levels could arise . First , variation could be induced at the beginning of the recombination pathway , when double-strand breaks ( DSBs ) are formed . For example , some individuals might have a greater number of DSBs than others and , assuming that similar proportions of DSBs are converted into crossovers , the end result would be variation in the number of crossovers . On the surface , recent studies in mice would seem to eliminate this possibility: as in other model organisms [16] , crossover homeostasis operates in male mice to ensure the presence of a minimum number of crossovers over a wide range of DSBs , suggesting that variation in DSBs does not translate into variation in crossovers [17] . However , these analyses were intended to assess variation in recombination among mice with different numbers of functional alleles for the DSB-inducing locus Spo11 , and do not preclude the possibility that genetic background differences may influence the overall number of crossovers . Second , variation could arise at some stage in the processing of recombination intermediates . For example , genotypic differences at RNF212 , encoding a protein with homology to crossover promoting proteins in S . cerevisiae and C . elegans , is the best characterized determinant of individual variation in genome-wide recombination rates in humans ( e . g . , [18] ) . Recently , analyses of mice deficient for RNF212 indicate that it acts downstream of the initiation of DSBs , stabilizing joint DNA molecules and promoting the resolution of DSBs as crossovers [19] . Taken together , these observations provide evidence that events occurring after the formation DSBs can , indeed , affect the eventual number of recombination events , although it is not clear that this accounts for all of the among-individual or among-strain differences that have been reported in mammals . Third , differences might arise at the end of the recombination pathway , owing to individual variation in proteins such as MLH1 and MLH3 that are important in the resolution of double Holliday junctions into crossovers ( e . g . , [20] ) . To discriminate among these possibilities in a mammalian system , we took advantage of strain-specific differences in genome-wide rates of meiotic recombination in the male mouse [11] . Specifically , we used immunofluorescence to examine the localization patterns of early- , mid- , and late-acting meiotic recombination proteins , asking whether the patterns were the same or different among inbred strains known to have “low” , “mid” , or “high” genome-wide rates of recombination . For each strain , we analyzed the number of foci for RAD51 and DMC1 ( strand invasion proteins acting shortly after double-strand break formation; [21] ) , MSH4 ( part of the complex stabilizing double Holliday junctions; [22] ) , and BLM ( thought to have anti-crossover activity; e . g . , see [23] ) , and compared these observations with results of analyses of the CO-associated protein MLH1 [20] . Our results demonstrate that inter-strain differences in crossovers ( MLH1 foci ) are preceded by proportionally similar differences in early-acting recombination proteins , indicating that the variation is established at , or before , the time of DSB formation . Subsequent analyses of males heterozygous for the gene encoding the DSB-inducing protein SPO11 [24] , allowed us to eliminate DSBs per se as the source of the variation , since Spo11 heterozygotes exhibited a decrease in DSBs , but not in MLH1 foci . In analyses of chromatin loop size and synaptonemal complex ( SC ) length , we detected striking differences among the three inbred strains , but not between Spo11 heterozygotes and their wildtype littermates . Taken together with the observations on recombination proteins , our results suggest that strain-specific differences in chromatin architecture , presumably established prior to the initiation of recombination , are important determinants of variation in crossover frequency .
In previous studies of recombination in male mice [11] , we identified strain-specific differences in the number of foci per cell of the DNA mismatch repair protein MLH1 , known to mark the vast majority of sites of crossing-over [9] , [26] , [27] . We decided to exploit these differences to investigate the basis of the variation . Accordingly , we examined three inbred strains –C57BL/6J ( “B6” ) , CAST/Ei ( “CAST” ) and C3H/HeJ ( “C3H” ) – assaying a minimum of 15 pachytene stage cells per mouse , and at least five mice per strain , scoring the number of autosomal MLH1 foci per cell ( Figure 1A ) . Two of the inbred strains , CAST and B6 , had previously been found to have “low” and “high” genome-wide MLH1 values , respectively [11] . Our re-analysis produced virtually identical results: the mean number +/− S . D . of autosomal MLH1 foci per cell was 21 . 3+/−1 . 6 for CAST ( n = 105 cells ) and 25 . 0+/−2 . 2 for B6 ( n = 102 cells ) ( Figure 1B; Table S1 ) . Subsequently , we analyzed C3H males , and observed that this strain had mean MLH1 values that were intermediate to the other two strains: i . e . , 22 . 7+/−1 . 9 ( n = 209 cells ) ( Figure 1B; Table S1 ) . Because the number of MLH1 foci per cell is not normally distributed ( i . e . , typically each bivalent has at least one focus , thus constraining the autosomal foci per cell to 19 or more ) , inter-strain differences have to be interpreted with caution . Nevertheless , ANOVA analyses are typically robust in the face of modest departures from normality , and the magnitude of the differences we observed ( F = 105 . 1; p<0 . 0001 ) , make it likely that the variation is real . Thus , these observations confirm our previous conclusion of variation in genome-wide MLH1 values – and presumably MLH1-driven crossovers – among males of different mouse strains . The variation in overall MLH1 frequency was reflected by highly significant strain-specific differences in the proportion of chromosomes with zero , one , two , or three MLH1 foci ( χ2 = 292 . 0 , p<0 . 00001; Table S1 ) . In large part , the difference was attributable to differing ratios of chromosomes with one vs . two MLH1 foci in the three strains; i . e . , 6 . 8∶1 , 3 . 9∶1 and 2 . 2∶1 for CAST , C3H and B6 , respectively ( Table S1 ) . However , intriguingly , the strain with the highest MLH1 average value ( B6 ) , also had the highest proportion of bivalents lacking MLH1 foci; indeed , this value was significantly increased over that for CAST ( χ2 = 12 . 1 , p<0 . 001 ) and for C3H ( χ2 = 13 . 9 , p<0 . 001 ) males . Subsequently , we analyzed the placement of MLH1 foci among the three strains , asking whether variation in MLH1 levels might be linked to differences in the location of the foci . Initially , we simply pooled results from all homologs and found no obvious differences among the strains; i . e . , consistent with previous results , distally located MLH1 foci predominated among all three strains [11] . However , because the strain-specific differences in the proportion of homologs with one , two and three MLH1 foci complicate interpretations of these data , we conducted a second set of studies in which we analyzed individual chromosomes . Specifically , for each strain , we analyzed the placement of MLH1 foci on chromosomes 19 and 1 on SCs that exhibited one and two foci , respectively ( Figure 1C , D ) . No obvious differences were observed among the strains . The MLH1 data demonstrate inter-strain differences in recombination , but provide no information on when the variation originates . To address this question , we examined the abundance and distribution of signals for proteins involved at different stages of DSB repair , as follows: Single end invasion is the earliest stage of recombination that can be consistently assayed using immunofluorescence . The ubiquitous RAD51 protein forms a complex around the resected ends of DSBs and facilitates invasion of the intact homologous chromosome [28] . We assayed localization patterns of RAD51 foci on zygotene stage chromosomes ( Figure 2A ) when RAD51 activity is high [29]; at this stage most RAD51 foci localize to synapsed regions , although in a few instances they ( and also DMC1 foci , see below ) are associated with unsynapsed axial elements . From Figure 2A , it is clear that there was considerable among-animal variation , particularly in the C3H and B6 strains . Nevertheless , considering the pooled data , the mean numbers of RAD51 foci per cell were highly significantly different among the three strains ( F = 113 . 7; p<0 . 0001 ) , with the variation mirroring that observed for MLH1; i . e . , CAST had the lowest mean value ( 163 . 0+/−18 . 6 ) , C3H an intermediate value ( 179 . 9+/−28 . 0 ) , and B6 the highest value ( 222 . 1+/−33 . 8 ) ( Figure 2A; Table S2 ) . Further , considering those animals for which we had both RAD51 and MLH1 values ( see Tables S1 and S2 ) , the ratios of RAD51:MLH1 foci were similar among the three strains , with values of 7 . 7∶1 for CAST , 7 . 3∶1 for C3H and 8 . 8∶1 for B6 . Other than RAD51 , relatively few animals were scored for both MLH1 and another recombination protein ( i . e . , either DMC1 , MSH4 or BLM ) in each of the three strains; thus , we were not able to directly compare the ratios of MLH1:DMC1 , MLH1:MSH4 or MLH1:BLM among the three strains . Nevertheless , the pooled data were consistent with the results from MLH1 and RAD51 . For example , similar strain-specific differences were observed for zygotene stage cells scored for the meiosis specific strand invasion protein DMC1 , which attaches to DSB sites shortly after RAD51 [29] , [30] . Specifically , the mean values rose from 119 . 5+/−16 . 4 for CAST to 149 . 3+/−18 . 1 for C3H and 181 . 8+/−21 . 1 for B6 ( F = 214 . 3; p<0 . 0001 ) ( Figure 2B; Table S3 ) . Consistent with previous reports on DMC1 localization in mice [29] , all three strains exhibited fewer DMC1 than RAD51 foci . While the reason for this variation in numbers of DMC1 and RAD51 foci is not entirely clear , it presumably reflects the different roles the two proteins play in the early stages of the recombination pathway [28] , [31] , [32] or possibly , differences in the time that the proteins remain complexed to the recombination nodules . We investigated a later stage of recombination by assaying MSH4 , a member of the MSH4-MSH5 complex that is thought to stabilize recombination intermediates ( e . g . , [19] , [22] , [33] . Since MSH4 localizes only to synapsed chromosome regions ( Figure 2C ) , we counted foci in cells at the zygotene/pachytene boundary . We observed substantial variation in the values among the different individual mice but , similar to the results for RAD51 and DMC1 , we saw an increase in the mean number of MSH4 foci from CAST ( 105 . 3+/−12 . 8 ) to C3H ( 109 . 0+/−14 . 8 ) and B6 ( 112 . 7+/−17 . 5 ) males ( Figure 2C; Table S4 ) . These differences were more modest than those identified for RAD51 and DMC1 and because there was considerable overlap in individual values from strain to strain , they need to be interpreted with caution . Nevertheless , the mean values for the “low” ( CAST ) and “high” ( B6 ) strains were still highly significantly different from one another ( t = 2 . 9; p<0 . 01 ) ; for CAST vs C3H and C3H vs B6 the values were non-significantly different ( t = 1 . 5 , p>0 . 13 and t = 1 . 5 , p>0 . 14 , respectively ) Finally , we analyzed zygotene cells for the helicase BLM , a regulator of recombination intermediates suggested to have anti-crossover activity ( [23]; Figure 2D; Table S5 ) . Strain specific differences in average numbers of BLM foci were similar to those observed for the positive regulators of crossovers; i . e . , the mean value was lowest for CAST ( 114 . 4+/−15 . 9 ) , intermediate for C3H ( 120 . 0+/−12 . 4 ) , and highest for B6 ( 146 . 4+/−26 . 3 ) ( F = 62 . 3; p<0 . 0001 ) . In summary , our observations on strain-specific variation for five different recombination pathway proteins ( RAD51 , DMC1 , MSH4 , BLM and MLH1 ) are consistent with one another; i . e . , in each instance the mean number of foci per cell was lowest for CAST males , intermediate for C3H males , and highest for B6 males . Unfortunately , we were not able to collect data on each protein from each animal , limiting our ability to directly compare values among the different strains and , accordingly , to address other obvious questions , such as: which protein is the best predictor of MLH1 values , are the ratios of foci for different proteins the same among all three strains , and does the ratio of recombinogenic∶anti-recombinogenic proteins ( e . g . , RAD51:BLM ) vary among strains ? Nevertheless , taken together , the data provide strong evidence that , at least in males of these strains , a similar proportion of DSBs are translated into MLH1-associated crossovers . In subsequent studies , we were interested in determining whether the strain differences were accompanied by variation in the configuration of the meiotic axis and/or DNA loops . Accordingly , for each strain we assayed total autosomal SC lengths in pachytene stage cells . We observed highly significant differences in mean SC length among the strains; i . e . , for CAST 156 . 7+/−2 . 0 µm , for C3H 161 . 5+/−1 . 8 µm , and for B6 170 . 7+/−1 . 9 µm ( F = 13 . 8; p<0 . 0001 ) . These results are similar to the strain-specific observations for MLH1 , suggesting that the strain differences in crossover levels are linked to variation in length of the meiotic axis ( Figure 3A ) . As a surrogate for DNA loop size , we examined the width of the signal from whole chromosome paint probes on representative large , medium-sized and small chromosomes; i . e . , chromosomes , 1 , 12 and 19 , respectively ( see Figure 3B for an example of a chromosome 12 paint probe ) . DNA loop sizes differed significantly among the strains for each of the three chromosomes: i . e . , for chromosome 1 , CAST = 6 . 5+/−1 . 3 µm , C3H = 4 . 5+/−0 . 6 µm and B6 = 3 . 8+/−0 . 7 µm ( F = 183 . 5; p<0 . 0001 ) ; for chromosome 12 , CAST = 6 . 2+/−1 . 2 µm , C3H = 4 . 3+/−0 . 6 µm and B6 = 3 . 6+/−0 . 7 µm ( F = 176 . 5; p<0 . 0001 ) ; and for chromosome 19 , CAST = 4 . 4+/−1 . 0 µm , C3H = 3 . 0+/−0 . 5 µm and B6 = 2 . 7+/−0 . 5 µm ( F = 116 . 5; p<0 . 0001 ) ( Figure 3C ) . Together with the observations on SC length , these analyses indicate that increasing MLH1 values are associated with smaller DNA loops and longer SCs . The correlation between the number of foci for “early” , “mid” , and “late” recombination proteins and strain-specific recombination levels raises the possibility that the number of DSBs per se regulates recombination . We tested this by comparing meiotic profiles of males on the same genetic background but with different rates of DSB formation . Specifically , we compared wildtype B6 males with siblings heterozygous for a null allele of Spo11 , the type II topoisomerase-like protein responsible for programmed DSB formation in meiocytes [24] . Notably , the results from the wildtype Spo11 males were somewhat different than those of the B6 male described above , presumably due to variation from maintenance of the stocks at different facilities . For the Spo11 animals , consistent with previous reports [34] , [35] , zygotene spermatocytes from Spo11+/− males displayed significantly fewer DSBs ( estimated by the number of RAD51 foci ) than spermatocytes from wildtype littermates ( mean values = 152 . 2+/−20 . 6 and 200 . 5+/−21 . 5 , respectively; t = 5 . 8 , p<0 . 0001 ) ( Figure 4A;Table S6 ) . However , this did not translate into a difference in MLH1 values , with mean values of 23 . 9+/−1 . 9 and 23 . 8+/−1 . 6 for heterozygotes and wildtype littermates , respectively ( Figure 4B; Table S7 ) . Although it altered RAD51 values , Spo11 heterozygosity had no obvious effect on chromatin morphology . Specifically , total SC lengths per cell were virtually identical for Spo11+/+ and Spo11+/− littermates ( 167 . 1+/−3 . 3 µm and 169 . 7+/−4 . 8 µm , respectively; t = 0 . 46 p>0 . 65 ) ( Figure 4C ) . Further , there were no consistent differences in DNA loop size on individual chromosomes between Spo11+/+ and Spo11+/− males [i . e , 3 . 3+/−0 . 5 µm and 3 . 2+/−0 . 5 µm , respectively , for chromosome 1 ( t = 1 . 01 , p = 0 . 31 ) ; 2 . 9+/−0 . 5 µm and 2 . 9+/−0 . 5 µm , respectively , for chromosome 12 ( t = 0 . 27 , p = 0 . 78 ) , and 2 . 8+/−0 . 3 µm and 2 . 7+/−0 . 4 µm , respectively , for chromosome 19 ( t = 1 . 00 , p = 0 . 32 ) ] ( Figure 4D ) . Taken together with the results from the RAD51 and MLH1 assays , this indicates that the variation in crossover level among strains is not simply due to variation in the number of DSBs but more likely reflects differences in chromatin morphology .
The purpose of this study was to investigate the basis for individual variation in recombination rates . Over the past 10–15 years a number of tools have become available to investigate the biology of meiotic recombination in mammals; e . g . , knockout mice have been used to identify and characterize the functions of numerous meiotic genes ( e . g . , [20] , [36] , [37] , [38] , [39] ) ; genotypic analysis of individual and pooled sperm samples has led to the identification of small , discrete regions of high recombination activity [40] , [41] , [42] , [43]; linkage and linkage disequilibrium studies have revealed the presence of thousands of recombination hotspots in mammalian genomes [7] , [44] , [45]; and genome-wide analyses of genetic polymorphisms have led to the identification of genes involved in modulating hotspot activity [46] , [47] , [48] or genome-wide recombination levels [13] , [18] . Nevertheless , our understanding of the origin of among-individual variation in recombination levels in mammals remains rudimentary . In recent studies , genotypic differences at three loci have been linked to variation in the recombination phenotype . Specifically , allelic variation in the gene encoding the meiosis-specific histone methytransferase PRDM9 has been associated with hotspot activity in both mice and humans [46] , [47] , [48]; allelic variation in RNF212 , a homolog of the C . elegans synapsis/cross-over associated gene zhp-3 , has been shown to affect genome-wide recombination levels in human males and females [13] , [14] , [18]; and the presence of an inversion at 17q21 . 31 affects recombination rates in human females [13] , [14] , [49] . However , several lines of evidence indicate that these individual loci may be relatively unimportant , or at least not the only , determinants of variation in CO levels . First , in studies of PRDM9 in humans , allelic variation has been shown to influence hotspot usage but appears to have relatively little effect on genome-wide recombination levels [8] , [46] , [47] , [50] . Second , the magnitude of the effects attributable to the other two loci ( RNF212 and inv17q21 . 31 ) is insufficient to account for the level of variation in genome-wide rates identified in humans or in the present study [13] , [18] . Finally , in recent analyses of inbred strains of mice with differing levels of genome-wide recombination conducted by us [51] and others [52] , putative QTL-containing chromosome regions did not include either Prdm9 or Rnf212 . Thus , the available evidence suggests that other , as yet unidentified , loci , are responsible for generating most of the variation in recombination rates among individuals or inbred strains . To address this problem in a somewhat different way , we utilized inbred strains of mice with varying levels of genome-wide recombination to identify the temporal window during which the variation is generated . Surprisingly , our results provide strong evidence that the variation is attributable to processes acting at , or upstream of , DSB formation and repair . Specifically , our observations on foci number for each of three recombination-promoting proteins ( RAD51 , DMC1 , and MSH4 ) and for the anti-recombination protein BLM were similar in finding a direct correlation between foci numbers and levels of recombination . In each of the three strains , the number of DSBs ( measured as RAD51 foci ) exceeded the number of crossovers ( MLH1 foci ) by approximately ten-fold . This is consistent with observations from previous studies of male mice ( e . g , [53] ) and indicates that as in other organisms , the vast majority of DSBs are repaired as non-crossovers . Further , it suggests that , at least for the inbred strains we examined , events occurring downstream of DSBs are relatively unimportant in mediating genetic background-dependent variation in recombination rates . This does not mean that these processes are irrelevant – e . g . , RNF212 , which is known to affect recombination levels in humans , has recently been show to stabilize proteins important in processing recombination intermediates in male mice [19] – only that there are other , more important determinants . While our observations indicate that the variation in recombination levels is established by the time of DSB formation , our analyses of Spo11 deficient animals suggest that this is not attributable to DSBs per se . That is , despite having significantly different numbers of RAD51 foci , wildtype and heterozygous animals had virtually identical mean MLH1 values . These results are similar to those recently reported by Cole and colleagues [17] , in which they demonstrated a positive correlation between the number of functional Spo11 genes and the number of DMC1 and RAD51 foci , but no accompanying change in the number of MLH1 foci . Thus , as in other model organisms ( e . g . , [16] , [54] ) , mammalian meiosis appears to have homeostatic controls that operate to maintain the number of crossovers in the face of variation in the number of DSBs . Taken together , our analyses of different inbred strains of mice and of mice varying at a single locus but on an otherwise isogenic background demonstrate that DSB number itself is not the driver of variation in CO levels . What other processes might be responsible ? One obvious possibility is the way in which the chromatin is packaged , an idea consistent with the observations of the present report . For example , the best predictor of the number of MLH1 foci across the five different categories of mice that we examined ( i . e . , CAST , C3H , B6 , Spo11+/+ and Spo11+/− ) was SC length , since in each instance there was an average of approximately 7 µm SC per MLH1 focus . Similarly , the relationship between DNA loop sizes on individual chromosomes and the number of MLH1 foci was comparable in animals of different genetic background or Spo11 genotype . An obvious caveat to this interpretation is the fact that the observations on SC length and DNA loop size were based on pachytene cells while , presumably , the determination of the relative level of COs occurs much earlier . However , in unrelated studies of human males and females , we found that sex-specific differences in chromatin morphology in pachytene stage meiocytes were mirrored in leptotene preparations [55] . Thus , we think it unlikely that the observations of the present report simply reflect pachytene cells . Accordingly , we suggest that modification of chromatin morphology – but not DSB numbers – is a primary determinant of CO levels . This is consistent with recent analyses of Petukhova and Camerini-Otero and colleagues [56] , [57] , as they found that trimethylation of H3K4 at potential PRDM9 recombination hotspots is not dependent on SPO11; i . e . , the establishment of hotspot-associated marks occurs regardless of the presence of DSBs . What genetic loci might be at work to influence chromatin packaging and , ultimately , variation in CO levels ? In two recent QTL analyses of recombination in different inbred strains of [51] , [52] , several QTLs were shared between the studies; i . e . , overlapping regions were identified on chromosomes 4 , 15 and 17 , and nearly overlapping regions on chromosome 3 and the X chromosome . Notably , in both studies the highest lod scores were observed on the X chromosome , consistent with previous studies linking X-linked loci with variation in genome-wide recombination rates [12] , [58] . Because inactivation of the sex chromosomes ( MSCI ) appears to be essential for normal male meiosis [59] , it is generally thought that few , if any , X-linked loci are necessary for spermatogenesis . The results of the analyses of Murdoch et al [51] and Dumont and Payseur [52] , as well as studies identifying spermatogenic functions for X-linked genes ( e . g . , Tex11; [60] ) , suggests that this is not an absolute rule , and that high resolution analyses of the X chromosome may uncover important recombination-associated loci .
All animal experiments and procedures conducted at Washington State University were performed using protocols approved by the Institutional Animal Care and Use Committee ( WSU IACUC number 03737-014 ) . Breeding stocks of three inbred strains , C57BL/6J , C3H/HeJ , and CAST/EiJ , were obtained from The Jackson Laboratory and maintained by brother-sister mating . Spo11+/+ and Spo11+/− animals ( on a C57BL/6J background ) were kindly provided by Drs . Maria Jasin and Scott Keeney , Memorial Sloan Kettering Cancer Center . All animals were housed in polysulfone cages on ventilated racks or static cages . Experiments were approved by the Washington State University ( WSU ) Institutional Animal Care and Use Committee . WSU is fully accredited by the American Association for Accreditation of Laboratory Animal Care and all investigations were conducted in accordance with the Guide for the Care and Use of Laboratory Animals . Adult male mice between the ages of 6 and 40 weeks were killed by cervical dislocation and the testes removed . Seminiferous tubules were removed from the testes and surface spread preparations of spermatocytes for immunofluorescence studies prepared as described previously [61] . Slides were immunostained using similar methodology to that of Anderson et al . [26] . Antibodies were diluted in sterile filtered 1×ADB consisting of 10 ml normal donkey serum ( Jackson ImmunoResearch ) , 3 g BSA ( Sigma-Aldrich ) , 50 µl Triton X-100 ( Alfa Aesar ) , and 990 ml PBS . Incubations were performed in a dark humid chamber at 37°C . Slides were incubated overnight in a dark humid chamber at 37°C with one of the following primary antibodies: RAD51 ( rabbit anti-human , Santa Cruz Biotechnology ) diluted 1∶75 , DMC1 ( rabbit anti-human , Santa Cruz Biotechnology ) diluted 1∶200 , BLM ( rabbit anti-human , Santa Cruz Biotechnology ) diluted 1∶100 , MSH4 ( rabbit anti-human , provided by Dr . Chengtao Her ) diluted 1∶75 , or MLH1 ( rabbit anti-human , Calbiochem ) diluted 1∶75 . All slides were incubated with a primary antibody for SYCP3 ( goat anti-mouse , Santa Cruz Biotechnology ) diluted 1∶100 for 2 hours . Fluorescein-labeled donkey anti-rabbit ( Jackson ImmunoResearch ) secondary antibody was diluted 1∶75 and incubated overnight , followed by an incubation for 45 minutes with rhodamine-labeled donkey anti-goat secondary antibodies ( Jackson ImmunoResearch ) , diluted 1∶200 . Slides were counterstained and fixed using Prolong Gold antifade reagent with DAPI ( Invitrogen ) and sealed with rubber cement . StarFish whole chromosome paint probes ( Cambio ) were used according to the manufacturer's instructions to estimate DNA loops of specific chromosomes . Briefly , previously immunostained slides were dehydrated in a series of brief ethanol washes ( 75% , 90% , 100% ) , denatured in a 70% dionized formamide/2×SSC solution for 5 minutes at 65°C , quenched in 70% ethanol at −20°C , and dehydrated again with a series of ethanol washes . Paint probes for chromosomes 1 , 12 , and 19 were denatured for 10 minutes at 65°C and applied to the slides overnight in a dark humid chamber at 37°C . Following incubation , slides were soaked twice for 5 minutes in 50% dionized formamide/1×SSC solution at 45°C , washed twice in 1×SSC at 45°C for 5 minutes , and soaked for three cycles in 25% Detergent DT/4×SSC ( Cambio ) for 4 minutes at 45°C . Slides were allowed to drain and mounted with Reagent MD ( Cambio ) and sealed with rubber cement . For immunofluorescence analysis , cells were identified using DAPI and the sub-stage of meiotic prophase determined on the basis of the morphology of the SC ( visualized with SYCP3 antibody ) . For each cell and each protein analyzed , we scored the number of foci localizing to the axial element or SC at the appropriate sub-stage of meiotic prophase; i . e . , for RAD51 , DMC1 , and BLM prior to complete synapsis during zygotene , MSH4 at the zygotene-pachytene transition , and MLH1 during pachytene prior to desynapsis of the XY bivalent . For analysis of MLH1 foci we restricted our scoring to autosomal chromosomes , since the appearance and disappearance of foci on the XY bivalent and on the autosomes are temporally uncoupled . Position data for MLH1 foci was collected by measuring along the fully synapsed SC from the centromere to each MLH1 focus . The placement of MLH1 was calculated by dividing the centromere-MLH1 distances by the full length of each individual SC . Values are represented as a percentage of total SC length , with 0% being the centromeric end and 100% the telomeric end of the SC . Genome-wide SC length was measured in fully synapsed pachytene cells by manually tracing the length of the SYCP3 signal for all SCs except the XY bivalent . The genome-wide SC length per cell was calculated as the sum of the SYCP3 signal lengths for the 19 autosomes . DNA loop sizes were assayed on three representative chromosomes , 1 , 12 , and 19 , identified by FISH paint probes , using an approach similar to that previously described by Novak et al [62] FISH images were then overlaid on immunofluorescence images of the SC and loop . Loop size was assayed by taking linear measurements of the width of the FISH signal perpendicular to the SC at the centromere , telomere , and mid-point of the SC . The three lengths for each chromosome were averaged and compared on a chromosome-specific basis among the strains . All slides were imaged on a Zeiss Axio Imager epifluorescence microscope and analyzed by blinded observers using Zeiss Axiovision software ( version 4 . 7 ) . Comparisons in the average numbers of foci between different strains or genotypes of mice were tested by analysis of variance or Student t-test analysis , pooling the results from multiple mice in a strain or genotype . For post-hoc comparisons of mean values between specific strains , we used the Bonferroni correction to account for the multiple tests . Mean values +/− SD are provided in the text and tables and , for illustrative purposes , as mean values +/− SE in the figures . Differences in the numbers of E0 bivalents were compared by chi-square tests . All statistical analyses were performed with JMP software , version 7 . 0 . 1 ( SAS Institute Inc . ) . | During prophase of meiosis , homologous chromosomes exchange genetic material , in a process known as crossing-over . Crossovers are thought to be essential for proper separation of chromosomes during meiosis but , surprisingly , most mammalian species exhibit substantial individual variation in the number of crossovers per cell . We investigated the basis for this variation by examining localization patterns of crossover-associated proteins in inbred strains of male mice with differing average numbers of crossovers per spermatocyte . Our results indicate that the strain-specific variation is established early in meiotic prophase , possibly even before the DNA is broken in advance of subsequent exchanges between homologous chromosomes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"molecular",
"cell",
"biology",
"meiosis",
"chromosome",
"biology",
"biology"
]
| 2014 | Variation in Genome-Wide Levels of Meiotic Recombination Is Established at the Onset of Prophase in Mammalian Males |
The complement cascade is crucial for clearance and control of invading pathogens , and as such is a key target for pathogen mediated host modulation . C3 is the central molecule of the complement cascade , and plays a vital role in opsonization of bacteria and recruitment of neutrophils to the site of infection . Streptococcal species have evolved multiple mechanisms to disrupt complement-mediated innate immunity , among which ScpA ( C5a peptidase ) , a C5a inactivating enzyme , is widely conserved . Here we demonstrate for the first time that pyogenic streptococcal species are capable of cleaving C3 , and identify C3 and C3a as novel substrates for the streptococcal ScpA , which are functionally inactivated as a result of cleavage 7 amino acids upstream of the natural C3 convertase . Cleavage of C3a by ScpA resulted in disruption of human neutrophil activation , phagocytosis and chemotaxis , while cleavage of C3 generated abnormally-sized C3a and C3b moieties with impaired function , in particular reducing C3 deposition on the bacterial surface . Despite clear effects on human complement , expression of ScpA reduced clearance of group A streptococci in vivo in wildtype and C5 deficient mice , and promoted systemic bacterial dissemination in mice that lacked both C3 and C5 , suggesting an additional complement-independent role for ScpA in streptococcal pathogenesis . ScpA was shown to mediate streptococcal adhesion to both human epithelial and endothelial cells , consistent with a role in promoting bacterial invasion within the host . Taken together , these data show that ScpA is a multi-functional virulence factor with both complement-dependent and independent roles in streptococcal pathogenesis .
The complement cascade is crucial for clearance of invading pathogens and thus represents a key target for disruption by such organisms . Individual bacterial species use multiple strategies to escape the complement system , highlighting the importance of this pathway in bacterial immunity [1–7] . The human pathogen Group A Streptococcus ( GAS ) , the causative agent of over half a million infections globally each year which range from benign to life-threatening , is no exception [8] . The ability of GAS to successfully colonize the host and resist clearance is mediated by an array of virulence factors , a number of which interfere with and inactivate the complement cascade [2 , 4 , 9–11] . The complement pathway comprises a tightly regulated , self-perpetuating proteolytic cascade that results in clearance of pathogens by a combination of opsonization , anaphylatoxin release and formation of the lytic membrane attack complex ( MAC ) . Activation occurs via three routes; classical , alternative and lectin pathways , all of which converge at the central complement component C3 [12] , a 186 kDa member of the α-macroglobulin family [13] . C3 is comprised of an α ( 111 kDa ) and β ( 75 kDa ) chain which are linked by multiple disulfide bonds . C3 function is modulated by a sequence of proteolytic events , the first of which is mediated by the C3 convertase , and results in the release of the 9 kDa anaphylatoxin C3a from the N-terminus of the C3 α-chain . Activity of C3a is dependent on the carboxy-terminus of the protein , and is quenched following cleavage of the C-terminal arginine residue by the carboxypeptidase B enzyme [13] . The remaining 177 kDa protein , C3b , comprises the residual 102 kDa α–chain and 75 kDa β–chain , and is the activated form of C3 . Conformational changes following cleavage result in exposure of a reactive thioester residue permitting covalent deposition on the bacterial surface . Bound C3b interacts with complement receptors expressed by circulating phagocytes , mediating bacterial uptake and killing . C3b also binds to the pro-enzyme Factor B , cleavage of which by Factor D results in formation of the enzyme complex C3bBb . C3bBb catalyzes cleavage of C3 to C3b and C3a thus amplifying the complement response . [12] This activity in turn induces formation of the C3b2Bb complex which cleaves the complement component C5 . C5 is also a member of the α-macroglobulin family and as such shares a similar structure to C3 . C5 is cleaved by C3b2Bb which results in the release of the 10 . 4 kDa anaphylatoxin C5a from the N-terminus similar to the release of C3a from C3 . The remaining , larger C5b protein is necessary for downstream complement activation [13] . While deposition of antibody on the surface of GAS is important in promoting opsonophagocytosis [14] , complement deposition plays a pivotal role in the control of GAS infection . A strong selective pressure to resist complement immunity has resulted in the evolution of numerous evasion strategies in GAS [2 , 4 , 9–11] . These include expression of virulence factors that bind to and inactivate key components of the complement cascade and/or sequester negative regulators of complement at the bacterial surface . GAS also produces enzymes that cleave and inactivate complement components . These include the promiscuous secreted protease SpeB , which degrades numerous host factors including C3b [4 , 11] , and the C5a-inactivating serine protease ScpA , that is common to many pathogenic streptococci , which specifically cleaves C5a [10 , 15 , 16 , 17] . ScpA is a member of the subtilisin-like serine protease family , containing a highly conserved catalytic triad motif ( Asp130 , His193 , Ser512 ) that is critical for enzymatic activity [16] . Additionally ScpA has an LPXTG motif at the C-terminus which permits anchoring of the protein to the bacterial cell wall [15] . ScpA is processed from a pre- to pro-peptide by autocatalytic cleavage of the N-terminal 31 amino acids , resulting in a catalytically active protein [10] . The reported substrate for ScpA is the human anaphylatoxin C5a , which plays a key role in neutrophil activation and recruitment to the site of infection . ScpA cleaves C5a at the His67 residue , releasing the C-terminus and rendering the protein inactive [10] . Thus , ScpA activity significantly impedes neutrophil activation and recruitment to the site of infection [17] , and has been shown to promote bacterial persistence and dissemination in murine models of infection [17–19] . C5a is the only reported substrate for ScpA , which is surprising given the specificity of other members of this enzyme family [20] . Although GAS have been reported to bind to [2 , 6 , 7 , 9] or inactivate [4 , 11] numerous effectors of the complement cascade , they have not been reported to cleave the central molecule C3 , inactivation of which would dampen all anti-bacterial effectors of this pathway . In a previous proteomic screen we identified a putative uncharacterized protein annotated as a “C3-degrading protease ( CppA ) ” [21] which is conserved amongst several streptococal species [21] . This prompted us to question whether GAS and other pyogenic streptococci could actually specifically cleave the central complement component C3 . Here , we report that Groups A , C and G streptococci are able to cleave C3 , but that the phenotype was not mediated by CppA . Using GAS as a model organism , we went on to demonstrate that the complement-cleaving activity was mediated by ScpA . As such , we identified C3a and C3 as novel substrates for ScpA , cleavage of which is associated with reduced human neutrophil activation and chemotaxis , and subsequent reduced bacterial opsonophagocytosis and killing . Expression of ScpA has previously been implicated in streptococcal pathogenesis in murine infection models [17–19] . Here , we demonstrate that ScpA confers resistance to bacterial clearance in a streptococcal soft-tissue infection model that was manifest even in mice lacking both C3 and C5 . Thus , notwithstanding any inactivation of the human complement cascade , ScpA confers virulence to GAS in mice independently of C3 or C5 cleavage , a phenomenon that , we suggest , is likely to be dependent on ScpA-mediated attachment of GAS to endothelial and epithelial cells , as demonstrated in vitro . Overall , our study characterizes ScpA as a multi-functional , complement inactivating protein , with properties that confer pathogenicity at multiple stages of infection .
In a previous study we identified a putative “C3-degrading protease” CppA [21] , for which homologues exist in other streptococcal species [22] , and showed that it is upregulated in GAS lacking a functional version of the regulator RocA [21] . In an effort to determine whether streptococcal species could cleave C3 , as has been demonstrated for S . pneumoniae [22] , whole cell suspensions of streptococcal species representing Lancefield groups A , C and G were co-incubated with human C3 . Human C3 is a 186 kDa protein comprised of a 111 kDa α chain and a 75 kDa β chain [13] . These fragments are linked by multiple disulphide bonds and as such , under reducing conditions , C3 migrates as two separate bands of corresponding size . Incubation of different streptococcal species with C3 generated an additional 100 kDa α-chain ( C3αscpA ) absent from controls incubated with buffer alone . This demonstrated that the experimental setting did not induce physiological activation of C3 to iC3b , which would result in release of a smaller α-chain , consistent with streptococcal-mediated cleavage of the alpha chain of C3 ( Fig 1A ) . Initially we carried out experiments to determine whether the CppA protein , identified in our previous study , was capable of cleaving C3 . We were , however , unable to demonstrate any role for CppA in the observed cleavage of human C3 , since recombinant CppA did not reproduce the activity of whole streptococcal preparations ( S1 and S2A Figs ) . Furthermore , there was no evidence that the streptococcal cysteine protease SpeB , previously reported to degrade C3b [11] , was responsible for cleavage of full-length C3 . Whole bacterial cells from isogenic strains that differed only in production of active SpeB ( GAS-M49 and GAS-M49ΔscpA ) were both equally able to cleave C3 , while supernatant fractions of the same strains were unable to cleave full-length C3 at all ( S2B Fig ) . In order to broadly characterize the type of enzyme mediating C3 cleavage , we tested a panel of 10 protease inhibitors to determine if any were sufficient to impair cleavage . Only the serine protease inhibitor Pefabloc was sufficient to inhibit cleavage of C3 by GAS-M1 ( Fig 1B ) . GAS produce two serine proteases known to interact with the host immune response , the CXC-chemokine-cleaving enzyme SpyCEP [23] and ScpA . Isogenic strains that differed only in production of SpyCEP ( GAS-M81 and GAS-M81ΔspyCEP ) , previously described by our laboratory [24] were equally able to cleave C3 , ruling out SpyCEP as the C3-ase ( S2C Fig ) . Thus in order to ascertain whether cleavage of C3 was mediated by ScpA , ScpA-negative isogenic strains were generated using two major GAS serotypes , M1 and M89 , and extensively screened to ensure no polar effects on expression of adjacent Mga-regulated , or CovR/S regulated genes had been introduced ( S3A–S3F Fig ) . Wildtype strains GAS-M1 and GAS-M89 and the corresponding isogenic ScpA deletion mutants were investigated for C3 cleavage activity . As expected , 16 hour incubation of either wildtype strain with C3 yielded the additional 100 kDa C3αscpA fragment consistent with cleavage; importantly , this band was absent following incubation with each of the mutants ( Fig 2A ) . Cleavage of C3 was restored in GAS-M89ΔscpA following complementation with plasmid pOriscpA ( Fig 2A ) . To verify that ScpA alone was mediating cleavage , the activity of recombinantly expressed ScpA ( rScpA ) ( S1 Fig ) was also evaluated . Similar to wildtype GAS , incubation of C3 with rScpA resulted in the release of a 100 kDa C3αscpA fragment , confirming that C3 is a novel substrate for ScpA ( Fig 2A ) . The structural and functional homology shared between C3a and C5a are well reported in the literature [25] . It therefore seemed plausible that the C3a moiety may represent an additional substrate for ScpA , such that cleavage of both the inert N-terminus of full-length C3 , comprising C3a , and the activated C3a fragment ( released by host C3 convertase from C3 ) could occur . Incubation of wildtype GAS-M1 and GAS-M89 with human C3a resulted in a clear reduction in size when visualized by SDS-PAGE ( Fig 2B ) , consistent with the size difference observed between the C3α and the C3αscpA fragment of C3 ( Fig 2A ) , and similar to the cleavage pattern observed for C5a ( Fig 2C ) . Disruption of the scpA gene and complementation with plasmid pOriscpA demonstrated that ScpA can cleave the C3a moiety , a finding that was confirmed by use of purified rScpA ( Fig 2B ) . The cleavage of C3a we observed was not fully reversed by disruption of the scpA locus , suggesting that an additional as yet unknown GAS virulence factor may contribute to this effect ( Fig 2B ) . Intriguingly , the ScpA-independent cleavage appeared more pronounced following incubation with GAS-M89ΔscpA compared with GAS-M1ΔscpA , suggesting serotype variation in expression levels . The pattern of cleavage provided evidence that the C3α-chain was being cleaved at the N terminus within C3a , and thus we carried out N-terminal sequencing of the C3αscpA moiety following 16 hour incubation with rScpA to identify the ScpA cleavage site . The sequence obtained ( 742-SHLGL-746 ) indicated that ScpA cleaved C3α seven amino acids N-terminal to the site of physiological C3 convertase ( Fig 2D and 2E ) . The ScpA cleavage site was two amino acids N-terminal to the cleavage site of NalP ( S4 Fig ) , a C3-ase of Neisseria meningitidis that impairs complement-mediated immunity [26] . The disparity in size of ScpA-cleaved C3a ( C3ascpA ) and C3α ( C3αscpA ) compared with physiological counterparts C3a and C3b was greater than that reported for NalP , and thus pointed to a likely functional role for ScpA-mediated cleavage on C3 activity . Importantly other complement factors , C1q , C2 and the structurally similar C4 , were not cleaved by GAS-M1 under the same conditions ( S5 Fig ) , demonstrating the restricted substrate specificity of this enzyme . To determine the functional relevance of C3-ase activity of ScpA , cleavage of human complement factors C3 , C3a and C5a by rScpA protein were investigated over a time course ( Fig 3 ) . Cleavage of C3 by rScpA was tested at a 1:1 molar ratio over four hours , and release of C3αscpA fragment was visualized by SDS-PAGE ( Fig 3 ) . Although cleavage of C3 required 30 minutes incubation with rScpA ( Fig 3A ) , complete cleavage of C3a occurred within 30 seconds ( Fig 3B ) . Cleavage of C3a occurred more rapidly than C5a ( Fig 3C ) , suggesting that C3a inactivation may contribute to the phenotype previously attributed to C5a inactivation alone . The staphylococcal cysteine protease Aureolysin ( Aur ) has previously been reported to cleave C3 when purified from S . aureus supernatants , with a recognized impact on complement function [12] . In order to directly compare rScpA protein function with Aur , we amplified aur from S . aureus strain N315 in order to express and purify rAur ( S1 Fig ) . As described previously , rAur was only able to cleave C3 in HEPES++ reaction buffer [12] , so comparative experiments with rScpA were carried out under these conditions rather than THB , as for earlier rScpA cleavage analyses . The rate of C3 cleavage was comparable for rScpA and rAur , and was visible from 30 minutes incubation ( S6A Fig ) . Cleavage of C3α-chain by ScpA results in the release of a longer C3b-like molecule ( C3αscpA ) compared with that produced by physiological C3 convertase ( Fig 2 ) . We hypothesized that this would impact on protein function , specifically C3 deposition on the bacterial surface , and thus went on to explore this in an experimental setting . Following incubation of live GAS with fresh human serum as a source of complement , the level of C3 deposition on the bacterial surface was quantified by flow cytometry . Initially , experiments were carried out using GAS-M1 and the isogenic GAS-M1ΔscpA strain at a range of serum concentrations . Incubation of GAS with serum at a concentration of 25% or more resulted in the detection of significantly less C3 deposition on the surface of GAS-M1 compared with GAS-M1ΔscpA ( Fig 4A ) . Deposition of C3 fragments on GAS required active complement and was barely detectable when cognate heat-inactivated serum samples were used ( Fig 4A ) , despite similar amounts of C3 being present in the heat-inactivated serum ( S7 Fig ) . Serotype M1 GAS are known to inhibit complement deposition via recruitment of host fibrinogen to the bacterial surface by M1 protein [9] . In order to determine if our findings could be reproduced in GAS serotypes where M protein does not bind fibrinogen [27] , we carried out C3 deposition experiments in isogenic M89 strains GAS-M89 , GAS-M89ΔscpA , GAS-M89ΔscpApC and GAS-M89ΔpCscpA ( Fig 4B ) . As observed for GAS-M1 , loss of expression of ScpA resulted in enhanced C3 deposition on the bacterial surface , which could be reversed following complementation of GAS-M89ΔscpA with ScpA expressed in trans ( Fig 4B ) . These data suggest that ScpA contributes to inhibition of C3 deposition by multiple GAS serotypes and may allow escape from complement-mediated immunity . The physiological relevance of the reduction in C3 deposition on the bacterial surface as a result of rScpA activity was further assessed by comparison with rAur . Both proteins led to comparable reduction in C3 deposition on the surface of GAS-M1ΔscpA ( S6B Fig ) . To better understand the mechanism by which C3 cleavage by GAS might contribute to a reduction in C3 deposition , we investigated the stability of the abnormally long C3αscpA moiety , compared with regular C3bα-chain . C3αscpA generated by rScpA cleavage was subject to rapid degradation by serum host factors when compared with physiological C3bα chain , indicating reduced stability ( Fig 4C and 4D ) . Together , these data demonstrated that the C3αscpA molecule generated by ScpA cleavage is functionally impaired in its ability to resist degradation by host serum factors and opsonize invading bacteria . We hypothesized that the observed effects of C3 and C3a cleavage would affect bacterial clearance , as cleaved C3a might lack activity and C3bscpA fragments would be less likely to promote enduring phagocytic uptake and killing by the host immune response . ScpA has previously been reported to promote streptococcal resistance to neutrophil-mediated opsonophagocytosis and survival in whole human blood , however the mechanism underlying this was not characterized [17] . While this phenotype could be attributed to loss of neutrophil activation by C5a , we also considered the possibility that the reduction in opsonization of GAS by C3 following cleavage by ScpA might impact on bacterial uptake and clearance as well . Consistent with a central role of ScpA in interference with phagocytic killing , bacterial survival in whole human blood , quantified by Lancefield Assay , was significantly reduced for isogenic mutants no longer expressing ScpA compared with wildtype parent M1 strains ( Fig 5A ) . Importantly the same phenotype was also observed following blockade of CD88 ( C5aR1 ) with the chemical inhibitor PMX205 ( Fig 5B ) [28] , suggesting at least a partial C5-independent role for ScpA in inhibition of bacterial killing and growth in whole human blood . To confirm that neutrophil uptake was affected by ScpA , we compared the ability of purified human neutrophils to phagocytoze GAS-M1 and GAS-M1ΔscpA following opsonization with human serum . Expression of ScpA enhanced bacterial resistance to neutrophil uptake ( Fig 5C ) and , while we believe the reduction in function of C3αscpA compared with physiological C3bα contributes to this phenotype , we hypothesized that inactivation of C3a and C5a would also impact on neutrophil function . C3a and C5a each display anaphylatoxin and chemoattractive properties , and as such play a key role in the activation and recruitment of neutrophils to the site of infection [25] . Using surface expression of CD11b ( CR3 ) as a marker of activation we observed a significant reduction in neutrophil activation in the presence of ScpA following comparison of wildtype GAS-M1 with isogenic GAS-M1ΔscpA ( Fig 5D ) . This dampening of neutrophil activation suggested a significant loss of function of cleaved C3a and C5a , with potential far-reaching effects on neutrophil-mediated immunity . We hypothesized that neutrophil recruitment via C5a and C3a chemotactic gradients , and activation by calcium flux , would also be impaired . ScpA peptidase activity resulted in a significant reduction in neutrophil migration along both C3a and C5a gradients ( Fig 5E ) . In line with these findings , the cytosolic calcium flux mediated by cleaved C3ascpA and C5ascpA fragments was significantly impaired compared with full-length molecules ( Fig 5F and 5G ) . Importantly , exposure to recombinant ScpA alone did not induce a detectable increase in neutrophil intracellular calcium levels ( Fig 5F and 5G ) . In conclusion , ScpA peptidase activity destroyed the complement chemotactic gradients required for migration of human neutrophils at the local site of infection and significantly impaired functional activation . The structure of complement components is strongly conserved between species so we predicted that the peptidase activity of ScpA would be sufficient to mediate functional cleavage of murine ( m ) C3 , C3a and C5a . Incubation of wildtype GAS-M1 or recombinant ScpA with mC3a and mC5a over 16 hours resulted in a clear reduction in size of both proteins when visualized by SDS-PAGE ( S8A and S8B Fig respectively ) , similar to ScpA-mediated cleavage of human C3a and C5a ( Fig 2 ) . Genetic deletion of scpA in isogenic strain GAS-M1ΔscpA removed the ability of GAS to cleave either murine C3a or C5a ( S8A and S8B Fig ) . Importantly however , the rate of cleavage of C3a and C5a was significantly slower than observed for human counterparts ( S8C and S8D Fig ) . A small reduction in C3 deposition on the surface of wildtype GAS-M1 compared with GAS-M1ΔscpA was observed following incubation with fresh murine serum ( Fig 6A ) , demonstrating that ScpA peptidase activity can influence murine C3-opsonization of GAS . We thus went on to determine whether ScpA might play a C5a-independent role in GAS disease outcome in a mouse model . Soft tissue infections represent an important clinical manifestation of GAS , and can be readily modelled in mice to study the initiation and early events in clearance of infection that were previously implicated in ScpA activity [19] . Accordingly , we utilized a previously characterized model of soft tissue infection to ascertain the role of ScpA in the early stages of dissemination during infection [29] , comparing pathogenesis of isogenic GAS-M1 and GAS-M1ΔscpA in wildtype mice . Importantly we could detect ScpA expression at the site of infection in mice infected with GAS-M1 , demonstrating the pathophysiological relevance of this virulence factor ( S9 Fig ) . Local spread of GAS-M1 from the site of infection was significantly reduced following genetic deletion of scpA ( Fig 6B ) . Three hours after intra-muscular infection , the bacterial burden at the site of infection was comparable between mice infected with either GAS-M1 strain , however dissemination to the locally draining inguinal lymph node was significantly reduced in mice infected with GAS-M1ΔscpA ( Fig 6B ) . At this early timepoint , systemic spread of GAS-M1 was not detectable . By 24 hours however , any difference between strains was no longer detectable , despite systemic spread to spleen , liver and blood ( S10 Fig ) . To investigate these early pathogenic events , histological sections of thigh were examined from C57Bl/6 mice infected with GAS-M1 or GAS-M1ΔscpA ( n = 4/group ) . At this early , 3 hour time-point , bacteria could only be detected in some sections ( GAS-M1: 3/4 , M1ΔscpA: 1/4 ) ( S11 Fig ) ; any neutrophil infiltrate was scant and , under high magnification , appeared necrotic in mice infected by either GAS-M1 strain ( S11B Fig ) . Taken together the quantitative data suggested that the early clearance and local containment of infecting GAS was significantly impaired by ScpA , however , we were unable to detect a specific effect on neutrophil recruitment at this time point histologically , despite our results with human cells . In order to ascertain whether the observed effects of ScpA were independent of any effect on C5 or C5a cleavage , the same experiment was carried out in age and weight-matched C5-/- mice ( Fig 6C ) . Following infection via the intra-muscular route , dissemination of GAS-M1 and GAS-M1ΔscpA was determined by quantitative culture . While no difference was observed at the site of infection , we again found that considerably fewer bacteria were recovered from the draining inguinal lymph node of mice infected with GAS-M1ΔscpA . This suggested that ScpA might be affecting local GAS spread to lymph nodes in a C5a independent manner . To determine whether these results were indeed due to C3-ase activity of ScpA , we carried out the same experiment in age and weight-matched C3-/-/C5-/- knockout mice that are essentially complement deficient ( Fig 6D and 6E ) . No difference was observed between GAS-M1 and GAS-M1ΔscpA infected groups in clearance from the site of infection to the locally draining lymph node ( Fig 6D ) , suggesting that C3-mediated clearance may be more important than C5 in the lymphatic system . However , surprisingly , expression of ScpA significantly enhanced GAS systemic dissemination to blood , spleen and liver ( Fig 6E ) . These unexpected results pointed to a complement-independent role for ScpA in systemic bacterial clearance in settings where there is no C3 and C5 . ScpA has been implicated as an adhesin for both GAS and GBS , however an adhesin effect in live bacteria interacting with live host cells has not been directly demonstrated [30 , 31 , 32] . We hypothesized that the enhanced systemic dissemination of GAS-M1 compared with GAS-M1ΔScpA in the absence of complement might be the result of adhesion to and subsequent invasion through endothelial tissue . However , in natural infection , interactions with epithelial cells may be important in initial colonization events . In order to test this hypothesis we went on to characterize the ability of the A549 human lung epithelial cell line and primary human umbilical vein endothelial cells ( HUVEC ) to support attachment of GAS-M1 and GAS-M89 ( Fig 7A+B and 7C+D respectively ) . After demonstrating that both GAS serotypes could adhere to these cell monolayers , we went on to compare and quantify adherence of isogenic ScpA positive and negative strains . Expression of ScpA enhanced attachment of GAS-M1 and GAS-M89 to A549 ( Fig 7A and 7B ) and HUVEC ( Fig 7C and 7D ) , but had no effect on passive internalization at this 30 minute time-point ( S12 Fig ) . rScpA protein was also able to bind to fixed HUVEC in a dose-dependent manner ( Fig 7E ) . Interestingly , fixed HUVEC monolayers were able to support greater binding of rScpA in the presence of serum ( 10% FCS ) ( Fig 7E ) , an effect which was also seen for attachment of GAS-M1 , but not GAS-M1ΔScpA to live HUVEC monolayers ( Fig 7C ) . Expression of ScpA was also sufficient to promote GAS-M1 adhesion to murine cardiac endothelial cells ( MCEC-1 ) ( Fig 7F ) , suggesting that this phenotype could explain the enhanced virulence displayed by GAS-M1 in the murine model ( Fig 6 ) . These data support a role for ScpA as an adhesin for attachment of live GAS to epithelial and endothelial cells , which may contribute to the invasive potential of this bacterium in vivo . In order to better define the role of ScpA during infection , we generated strain GAS-M89ΔpCcat2 which over-expresses enzymatically inactive ScpA due to a mutation in the second amino acid of the catalytic triad His193 to alanine [16] . The catalytically inactive ScpA included an additional SNP not predicted to affect structure or expression . We went on to confirm that this mutant was no longer able to cleave C3a ( Fig 8A ) or C5a ( Fig 8B ) , in contrast to plasmid-transformed GAS-M89ΔpCscpA that over-expressed intact ScpA , consistent with previously reported ScpA structure function studies [16] . To ascertain whether the observed ScpA-enhanced dissemination of GAS ( Fig 6 ) was dependent on catalytic activity , we compared the virulence of strains GAS-M89ΔpC , GAS-M89ΔpCscpA and GAS-M89ΔpCcat2 during soft-tissue infections in age and weight-matched C57BL/6 mice . Following infection via the intra-muscular route , bacterial dissemination was determined by quantitative culture . As we found previously for GAS-M1 and GAS-M1ΔScpA ( Fig 6B ) , no difference was observed in bacterial counts recovered from the site of infection ( Fig 8C ) . Intriguingly , greater numbers of GAS were recovered from the draining inguinal lymph node of mice infected with GAS expressing ScpA , regardless of whether the ScpA was catalytically active ( Fig 8D ) . This demonstrated that the properties of ScpA that were critical for pathogenesis in the mouse model were independent of any enzymatic activity , pointing to a potentially important role for adhesion in disease progression through interactions with endothelial or epithelial cells . It was not possible to deduce the relative contribution of ScpA adhesion to pathogenesis in a human system , albeit that ScpA conferred adhesion to human and murine cells . The discrepancy in rates of ScpA-mediated cleavage of human and murine complement factors ( Fig 3 , S8A and S8B Fig ) and the smaller reduction in murine C3 ( Fig 6A ) compared with human C3 ( Fig 4A ) deposited on the GAS-M1 surface raised the possibility that , in human disease , the impact of complement cleavage may be of greater importance than adhesion . Our study highlights the multi-functional mechanisms by which ScpA is able to promote bacterial invasion , through a combination of factors including adhesion and active inactivation of the innate immune response .
Streptococcal C5a peptidase is expressed by a variety of streptococcal species . It thus is believed to play a pivotal role in the success of the streptococci , however to date its role in immune evasion has largely been attributed to its ability to cleave C5a . In this work we have identified multifunctional effects of ScpA which include rapid cleavage and inactivation of the anaphylatoxin C3a as well as cleavage of the central complement protein C3 . In addition , we have demonstrated complement-independent activity of ScpA as an adhesin , that may contribute to systemic spread during infection in vivo . The complement cascade plays a crucial role in the innate immune response to invading bacteria and as such represents a key target for pathogen immune evasion [1 , 3 , 4 , 10 , 22 , 33 , 34] . The inactivation of both anaphylatoxins C3a and C5a as well as the central complement component C3 we report here has the potential not only to reduce inflammation and cellular activation and recruitment at the site of infection , but also to dampen amplification of the complement system in response to infection , a phenomenon that has not previously been reported for the Group A Streptococci . ScpA is expressed by all GAS serotypes , and homologous enzymes are encoded by multiple other Lancefield groups [35–38] . Conservation of enzyme function both within and between species points to a clear functional evolutionary pressure to systematically disrupt neutrophil responses [36] , despite ScpA being a primary target for opsonizing or neutralizing antibodies produced during infection [38–43] . ScpA is a member of the subtilisin-like family of serine proteases which are reported to cleave multiple enzymatic substrates , and , while the streptococcal protease SpyCEP cleaves numerous ELR+ CXC chemokines [24] , C5a was hitherto the only reported substrate for ScpA . Our investigation into C3 cleavage by streptococci began following the identification of a putative “C3 degrading protease” ( CppA ) [21] although no such activity could be ascribed to CppA . Initial analyses established that groups A , C and G streptococci were , however , capable of specifically cleaving full-length C3 , a function that has not previously been reported for these species . Furthermore , cleavage activity could be inhibited by the serine protease inhibitor Pefabloc , which led us to hypothesize that ScpA could be mediating C3 cleavage , plausible given the structural similarity between C5a and C3a [25] . The cleavage of C3 that we observe begins within 30 minutes incubation with rScpA protein . Whilst we believe this to be physiologically relevant during infection , it may be that large numbers of bacteria , such as at the site of infection , are required for ScpA C3-ase activity to play a significant role in pathogenesis . The C3 molecule is an obvious target for pathogen immune evasion , and , as such , other bacterial species also express inactivating enzymes . Cleavage of C3 generating a longer C3bscpA and a shorter C3ascpA molecule by GAS reproduces the mechanism described for the N . meningitidis C3-ase NalP [26] . The reported C3-cleaving enzymes of other Gram positive pathogens , Staphylococcus aureus ( aureolysin ) [12] and Enterococcus faecalis ( GelE ) [44] , which both cleave 2 amino acids upstream of the physiological C3 convertase site , generate a smaller C3b and longer C3a moiety , in contrast to our findings for ScpA . Cleavage of C3a is thus a function uniquely shared by ScpA and NalP [26] , however any effect of such cleavage on C3a activity has not previously been reported . The GAS cysteine protease SpeB has previously been reported to degrade C3b in human serum , where ScpA could not [11] . In our study , we examined full-length C3 , which we demonstrate could be cleaved by ScpA , but not SpeB . This suggests that the various C3-ase functions of these enzymes are distinct , yet may act synergistically to overcome the host innate immune response . C3 is present in human serum at approximately 1 mg/ml , therefore multiple strategies for effective inactivation may be necessary to overcome this highly abundant molecule . Although ScpA is known to be regulated by Mga , our preliminary experiments using available isogenic strains deficient in functional CovR/S or RocA ( Regulator of Cov ) , showed that ScpA was also a member of the CovR regulon ( S13 Fig ) . Both SpeB and ScpA are regulated by CovR/S , however they display reciprocal expression profiles . It is therefore possible that each protein acts at a different phase of infection to modulate the host immune response . SpeB expression is significantly reduced in naturally occurring CovR/S mutants , which are frequently associated with invasive infection . As such , ScpA , which is expressed by invasive and non-invasive disease isolates , may represent a common mechanism of action shared amongst GAS isolates . The significance of ScpA as a standalone virulence factor was assessed by the generation of isogenic mutants in two GAS serotypes . Loss of function had a significant effect on C3 deposition and stability , and on neutrophil activation by the anaphylatoxins C3a and C5a . Indeed , cleavage of these two proteins reduced neutrophil recruitment along chemotactic gradients and impaired activation , specifically upregulation of CD11b and induction of intracellular calcium release which normally play an important role in pathogen phagocytosis . Fibrinogen binding by the M1 protein is known to have a major influence on GAS-M1 clearance and may render other virulence factors redundant [9] . Nonetheless , ScpA had a measurable effect on C3 deposition on GAS-M89 ( where fibrinogen binding is not thought to be dominant ) [27] and also in whole blood assays that contain human fibrinogen , both in the presence and absence of a C5aR1 antagonist , suggesting that the C5-independent activities we describe are detectable even in the presence of fibrinogen . In order to determine the functional relevance of the novel C3 and C3a cleavage activity , in vivo experiments were initially performed in both wildtype and C5-/-mice . Interestingly , similar results were obtained in both genetic backgrounds , suggesting that ScpA plays a role in streptococcal pathogenesis that is independent of C5a cleavage . Importantly , cleavage of C3 not only impairs C3a and C3b function , but also reduces generation of C3b and C5a . Generation of C3b via the alternative pathway is amplified by activity of the C3 convertase , which is comprised of multiple subunits including C3b . Reduction in activity of the alternative pathway C3 convertase will subsequently reduce activation of terminal complement , one output of which is functional C5a . Moreover , C3b is essential for formation of the C5 convertase , and , as such , inactivation of C3 will significantly impair C5a generation and reduce the activity of the entire complement cascade [13] . Thus , in combination with other pan-streptococcal cell envelope serine proteases such as SpyCEP , which cleaves the potent chemoattractant IL-8/CXCL8 and other ELR+ CXC chemokines [23] , GAS is capable of systematic inhibition of neutrophil migration and activation , possibly explaining the dearth of functional neutrophils recruited to the site of infection during severe clinical invasive disease . To validate a clear role for the C3-ase activity of ScpA in vivo , utilizing the same soft-tissue infection model , we compared the pathogenesis of GAS-M1 and GAS-M1ΔscpA in complement deficient ( C3-/-/C5-/- ) mice . The difference in clearance of bacteria from the site of infection to the locally draining inguinal lymph node observed in both wildtype and C5 deficient mice was diminished in C3-/-/C5-/- mice , suggesting that cleavage of C3 may be responsible for this phenotype . However , systemic dissemination via the blood was significantly impaired in the absence of ScpA , suggesting an additional complement-independent role for this protein in augmenting systemic spread . Using isogenic GAS strains and recombinant protein , we were able to show that ScpA is a potent adhesin , important for bacterial attachment to both human epithelial and endothelial cells . We went on to demonstrate that ScpA could also mediate attachment of GAS-M1 to murine endothelial cells , providing support for the hypothesis that adhesion was playing a key role in GAS pathogenesis in our in vivo model . Using bacteria expressing ScpA constructs that differed only in catalytic activity , we found that the enzymatic activity of ScpA was not necessary for pathogenicity in a murine model of soft tissue infection , noting the presence of an additional mutation in the enzymatically inactive construct that is not predicted to affect structure or expression . While extrapolating the relevance of our findings to human infection is beyond the scope of this study , the data strongly implicate complement-independent functions of ScpA in streptococcal virulence and disease progression in human infection . We speculate that the failure to control GAS infection in humans can be attributed in part to the complement-independent ability of ScpA to act as an adhesin for endothelial and other cell types , which may play a key role in infection where complement is absent . Here we have identified complement factors C3 and C3a as two novel substrates of the streptococcal enzyme C5a peptidase and demonstrated that the cleavage of these factors has significant functional effects on the host immune system , most explicitly on human neutrophil activation and function . We have also characterized the role of ScpA as an adhesin , important for attachment to epithelial and endothelial cells . The multiple functions of ScpA described demonstrate its fundamental role in streptococcal disease , and underline the importance of assigning novel functions to known virulence factors as a means to provide new insight into disease progression and pathogenesis .
Human blood from anonymized consenting healthy donors was obtained from an approved subcollection of the Imperial College NHS Trust Tissue Bank ( ICHTB reference R12023 ) . In vivo experiments were performed in accordance with the Animals ( Scientific Procedures ) Act 1986 , and were approved by the Imperial College Ethical Review Process ( ERP ) panel and the UK Home Office ( Project Licence 70/7379 ) . The strains used for molecular manipulation were invasive disease isolates GAS-M1 ( H598 ) [14] and GAS-M89 ( H293 ) [45] . All strains used are listed in Table 1 . GAS were cultured on Columbia horse blood agar plates ( OXOID ) Todd-Hewitt ( TH ) agar or in TH broth ( OXOID ) at 37°C , 5% CO2 for 16 hours . E . coli XL-10 gold ( Stratagene ) , TOP10 ( ThermoFisher Scientific ) BL21 ( DE3 ) ( ThermoFisher Scientific ) were grown in LB broth . Growth media were supplemented with antibiotics where appropriate at the following concentrations; E . coli erythromycin 500 μg/ml , kanamycin 50 μg/ml , ampicillin 100 μg/ml; GAS erythromycin 1 μg/ml , kanamycin 400 μg/ml . Genomic DNA was extracted from GAS cultures grown to late logarithmic growth phase ( OD600 0 . 7–0 . 9 ) as described previously [21] . PCR was carried out using a MyCycler ( Bio-Rad ) thermal cycler with Bio-X-Act proof reading Taq ( Bioline ) . Automated-fluorescent sequencing of PCR products and plasmids was performed by the MRC CSC Core Genomics Laboratory , Hammersmith Hospital . Deletion of scpA in GAS-M1 from nucleotide 333–1777 and GAS-M89 from nucleotide 333–2017 , resulting in removal of the characterized catalytic domain ( S3 Fig ) . A 500 bp fragment of the scpA gene was amplified ( GAS-M1: forward primer: 5’- GGAATTCGCTTTAATAATCGTCCATGG -3’ , reverse primer: 5’- GGGGTACCCCTCAAGCAAGGTTCACCTG -3’; GAS-M89: forward primer: 5’- GGAATTCTTCGATATCCTCTATGTTTTC -3’ , reverse primer: 5’- GGGGTACCGAGTTGTATTACCAAGCAAC -3’ ) incorporating EcoRI and KpnI restriction sites into the 5’ and 3’ ends respectively , and cloned into the suicide vector pUCMUT to produce pUCMUTGAS-M1 and pUCMUTGAS-M89 . A 3’ 500 bp fragment of the scpA gene was amplified ( GAS-M1: forward primer: 5’- ACGCGTCGACGACCTGAGAAGGGTCGTTC -3’ , reverse primer: 5’- AACTGCAGCTGTATCATATGCAAATAACC -3’; GAS-M89: forward primer: 5’- ACGCGTCGACCTGAGAAGGGTCGTTCAAATC -3’ , reverse primer: 5’- AACTGCAGCTGTATCATATGCAAATAACC -3’ ) incorporating PstI and SalI restriction sites into the 5’ and 3’ ends respectively , and cloned into PstI/SalI digested pUCMUTGAS-M1 and pUCMUTGAS-M89 respectively . The constructs were introduced into GAS-M1 and GAS-M89 by electroporation and incorporated chromosomally by homologous recombination as described previously [21] . PCR analysis demonstrated that only a single recombination event had occurred . For complementation scpA was cloned into replicative vector pOri23 ( scpApOri F: 5’- CCGGATCCCATCAGGAAAGGACGACACATTGC-3’ , scpApOri R: 5’- CCGGATCCGATCAGTTGTACTAATCTTCAGTGC-3’ ) incorporating in BamHI sites at both 5’ and 3’ termini . pOriscpA was propagated in One Shot® TOP10 Chemically Competent E . coli ( Invitrogen ) and transformed into electrocompetent GAS as described previously [21] . For generation of a complementation vector expressing catalytically inactive ScpA , the second amino acid of the catalytic triad His193 , was mutated to alanine in vector pOriScpA by site directed mutagenesis ( QuikChange XL-II Site-Directed Mutagenesis Kit , Stratagene ) ( forward primer: 5′- GCTGTCGATCAAGAGGCCGGCACACACGTGTC -3′ , reverse primer: 5′- GACACGTGTGTGCCGGCCTCTTGATCGACAGC -3′ ) to produce vector pOriscpAcat2 . Vector pOriscpAcat2 was sequenced and the desired mutation at residue 193 ( His-Ala ) confirmed . An additional non-synonymous SNP at position 243 ( Ala to Thr ) was not deemed to be relevant as it was not within any domains predicted to mediate adhesion . Full-length ScpA protein of size ~120 kDa was confirmed by western blotting and equivalent to that produced by pOriscpA . The vector was transformed into electrocompetent GAS-M89ΔscpA as described previously [21] . Recombinant proteins were produced using the following primer sequences ( CppA: cppA_pET-F: 5’- CCGGATCCAATGACTTTAATGGAAAATATTAC -3’ cppA_pET-R: 5’ CCGGATCCTTCATTTCGTAAACCATACTTC 3’ , HVR: emm_pET-F: 5’- CCGGATCCAGGTTTTGCGAATCAAACAGAGGTTAAG -3’ , emm_pET-R: 5’- CCGGATCCTAGCTCTCTTAAAATCTCTTCCTGCAACTTCC -3’ , Aur: aur_pET-F: 5’- CGGGATCCGATTGATTCAAAAAATAAACC -3’ aur_pET-R: 5’- CGGGATCCTTACTCCACGCCTACTTCATTC ) and the pET-19b overexpression system as previously described [49] . rCppA , rEmm 1 . 0 hypervariable region ( HVR ) , rAur and the previously described rScpA fragment [49] , truncated at amino acid 720 to prevent pro-peptide removal , were purified using the Ni-NTA purification system ( Invitrogen ) according to the manufacturer's instructions . Purified proteins were buffer-exchanged into PBS . Full-length rScpA was expressed following amplification from gDNA ( scpA_pET-F: 5’- CCGGATCCAACCAAAACCCCACAAACTC-3’ , scpA_pET-R: 5’- CCGGATCCTAGAGTGGCCCTCCAATAGC-3’ ) , and purified from BL21 cell lysates by size exclusion chromatography using a 1 × 50 cm Econo-Column® ( Bio-Rad ) packed with Sephadex G-100 ( Pharmacia ) . PBS fractions containing ScpA were identified by SDS-PAGE , pooled and concentrated using a 30 , 000 MWCO spin column ( Vivaspin , GE Healthcare ) . 16 hour timepoint: Recombinant ScpA ( 100 ng ) or stationary phase GAS pellets ( 4x106 cfu ) washed twice in sterile PBS and concentrated 4x , were incubated with C3 ( 1 μg ) ( MD Millipore ) , C3a ( 0 . 5 μg ) ( R&D ) or C5a ( 0 . 5 μg ) ( R&D ) in THB in a final volume 10 μl . Protease inhibitor kit ( Roche ) was used , and each inhibitor used at concentrations described previously [23] . All samples were incubated for 16 hours , 37°C , 5% CO2 . Pellets were centrifuged and resulting supernatants analyzed . Timecourse: 0 . 5 μM rScpA or rAur was incubated with C3 ( 0 . 5 μM ) ( MD Millipore ) , C3a ( 5 mM ) ( R&D ) or C5a ( 5mM ) ( R&D ) in THB or HEPES++ buffer ( 20 mM HEPES , 140 mM NaCl , 5 mM CaCl2 , 2 . 5 mM MgCl2 ) . Samples were fractionated on 3–8% Tris-acetate gels ( C3 and C3b ) or 4–12% Bis-Tris plus gels ( C3a and C5a ) ( ThermoFisher Scientific ) and stained with Instant Blue ( Expedeon ) . To determine the ScpA cleavage site , 8 μg purified C3 ( EMD Millipore ) were cleaved with recombinant ScpA in THB for 16 hours , 37°C . Following cleavage , fragments were subjected to SDS-PAGE and transferred onto Hybond-LFP ( Amersham ) . Membranes were stained with Amido Black ( Sigma ) and destained ( 25% v/v isopropanol , 10% v/v acetic acid ) . Cleaved C3αscpA fragment was excised and subjected to sequencing by Edman degradation ( AltaBioscience , Birmingham UK ) . Primary antibodies: For western blot , chicken anti-C3 ( Abcam ab14232 ) ( 1:10 , 000 ) , goat anti-his ( Invitrogen ) ( 1:50 , 000 ) , rabbit anti-SIC [14] and rabbit anti-SpyCEP [24] serum ( 1:1000 ) . Mouse antisera against Emm 1 . 0 HVR and the truncated ScpA molecule were raised by vaccination as described previously [49] and used at 1:1000 dilution . Secondary antibodies used were HRP-conjugated goat anti-chicken IgY , anti-mouse IgG and anti-rabbit IgG ( Abcam ) ( all used at 1:80 , 000 ) . For C3 deposition , FITC conjugated goat anti-human C3 and goat anti-mouse C3 antibodies ( MP Biomedicals ) were used ( 1:300 ) . Cell wall extracts and bacterial supernatants were prepared as described previously [50] with minor modifications . Briefly , stationary phase cultures were pelleted and supernatants were 0 . 2 μm filtered . Cell pellets were incubated in cell wall extraction buffer ( 30% raffinose , 1 μg/ml lysozyme , 10 mM Tris-HCl ( pH 8 ) ) at 37°C , 3 hours and cell wall fraction was isolated by centrifugation . Raffinose was removed following dialysis into PBS using Slide-A-lyzer cassettes ( ThermoFisher Scientific ) , and concentrated to 100 μg/ml . 1 μg cell wall extract was loaded for each sample to allow direct comparison between strains . Samples were separated on 4–12% Bis-Tris plus gels ( ThermoFisher Scientific ) and transferred onto Hybond-LFP membrane ( Amersham ) . Membranes were blocked ( 5% skimmed milk powder ( Sigma ) , 0 . 05% Tween ( Sigma ) and probed and detected with relevant primary and secondary antibodies . Membranes were developed using ECL Advance western blotting detection system ( GE Healthcare ) . Following cleavage of C3 with recombinant ScpA , the resulting fragments , or purified C3b ( MD Millipore ) were incubated with 1% C3-depleted human serum ( MD Millpore ) for 60 minutes . Samples were taken at 5 , 10 , 15 , 30 and 60 minute timepoints . The reaction was stopped following addition of LDS sample buffer ( ThermoFisher Scientific ) and 0 . 1mM DTT , and degradation visualized by western blot using chicken anti-C3 ( Abcam ) . GAS strains were cultured in THB and capsule quantified as described previously [21] using the hyaluronan DuoSet ( R&D ) . Stationary phase cultures of GAS ( 1x107 cfu ) were washed and resuspended in fresh human serum ( 0–50% ) or mouse serum ( 50% ) and incubated for 30 minutes , 37°C with end-over-end rotation . Control samples were incubated with fresh heat-inactivated human serum and human or mouse C3-depleted serum ( EMD Millipore ) . Cells were stained with FITC-conjugated anti-C3 antibody ( MP Biomedicals ) for 30 minutes on ice [51] and analyzed using a FACSCalibur cell analyzer ( BD Biosciences ) and FlowJo Software ( Treestar , OR ) . In order to combine the percentage of C3-positive bacteria and the C3-binding intensity , the results are presented as fluorescence index ( FI ) , which is calculated as the proportion of positive bacteria expressed as a percentage multiplied by the gMFI ) [52] . Lancefield assays were performed to assess GAS resistance to human phagocytic killing . GAS were cultured to OD600 0 . 15 in THB , and diluted in sterile PBS . Approximately 50 GAS cfu were inoculated into heparinized whole human blood obtained from healthy volunteers , and incubated for 3 hours at 37°C with end-over-end rotation . Quantitative culture at this time point was carried out to determine the number of surviving bacteria . To ascertain the role of ScpA-mediated C5a cleavage , C5a receptors were blocked by pre-treatment of whole blood with 1 μM PMX205 [28] , 37°C , 30 minutes , end-over-end rotation . Bacterial survival was quantified as multiplication factor of number of surviving colonies relative to the starting inoculum . Each strain was cultured in blood from at least three donors and tested in triplicate . Neutrophils were purified from fresh human blood using the MACSxpress® Neutrophil Isolation Kit ( Miltenyi Biotec ) according to manufacturer’s guidelines . Red cells were water-lysed and neutrophils used immediately . All cytometry was carried out on a FACSCalibur cell analyzer ( BD Biosciences ) and analyzed using FlowJo Software ( Treestar , OR ) . For neutrophil uptake studies , GAS were pre-labelled with FITC-conjugated group A carbohydrate antibody ( Abcam ) and uptake by purified human neutrophils was assayed as described previously [29] . Breifly , GAS were opsonized with fresh human serum ( +/- pre-treatment with 0 . 5 μM rScpA or rAur for 30 minutes at 37°C with end-over-end rotation ) . Opsonophagocytosis was carried out following incubation of 5x105 neutrophils with opsonized bacteria for 30 minutes at 37°C with end-over-end rotation . The reaction was stopped by addition of stop solution ( 0 . 02% EDTA in 0 . 9% saline ) and phagocytosis was quantified as percentage of FITC-positive neutrophils , indicative of bacterial uptake . For surface staining , 5x105 fresh neutrophils were incubated with unlabeled GAS for 30 minutes , 37°C with end-to-end rotation and blocked with Human TruStain FcXTM ( Biolegend ) for 10 minutes on ice . Surface staining for CD11b was carried out using AF488-conjugated anti-CD11-b ( Biolegend 301318 ) or isotype control antibody ( Biolegend 400129 ) to assess non-specific binding ( S1 Table ) for 20 minutes on ice , and gMFI ascertained for all samples . 24-well transwell plates ( Corning 3421 ) were blocked with RPMI ( 10% FCS ) for 30 minutes prior to addition of rScpA-cleaved and uncleaved C3a or C5a ( R&D ) ( +/- pre-treatment with 100 ng rScpA , 37°C , 16 hours ) . 1x106 purified neutrophils were added to the inner chamber , and incubated for 30 minutes , 37°C , 5% CO2 . The number of neutrophils which had migrated into the lower chamber were quantified using CountBright™ Absolute Counting Beads ( ThermoFisher Scientific ) . A flow cytometric method for calcium mobilization was carried out as described previously [53] with minor modifications . Briefly , neutrophils ( 1x107/ml ) were labeled with fluo-4-AM ( 0 . 9 μM ) in assay buffer ( 103 mM NaCl , 4 . 6 mM KCl , 1 mM CaCl2 , 5 mM glucose , 20 mM HEPES , pH7 . 4 ) . rScpA-cleaved ( 100 ng ) and uncleaved C3a or C5a ( 100 ng/ml ) ( R&D ) were added directly to neutrophils , and change in mean fluorescence intensity was measured immediately and recorded over 120 seconds . 6–8 week old C57BL/6 ( Charles River , Margate , UK ) , and age and weight-matched C57BL/6 C5-/- and C57BL/6 C3-/-/C5-/- female mice were challenged intra-muscularly with 1×108 GAS , and quantitative endpoints compared at 3 hours post infection . Mice were euthanized , blood taken by cardiac puncture and infected thigh muscle , spleen , liver , and draining inguinal lymph nodes dissected . All organs were plated to quantify bacterial cfu and systemic dissemination . Additional C57BL/6 mice were challenged intra-muscularly with 1×108 GAS . Whole thighs were formalin-fixed , and paraffin-embedded for histopathology imaging as described previously [24] , or homogenized and prepared for western blotting as described above . No animals were excluded from the study and randomization was not necessary as mice were genetically identical and age , weight and sex matched . No blinding was carried out as mice were caged separately to prevent contamination with different GAS strains . The human lung epithelial cell line A549 was cultured in RPMI supplemented with 10% FCS , L-glutamine ( 2 mM ) , 37°C , 5% CO2 . Primary human umbilical vein endothelial cells ( HUVEC ) were obtained from fresh human umbilical cords by collagenase treatment as previously reported [54] . Monolayers were cultured in Medium 199 supplemented with 20% FCS , Heparin ( 10 U/ml ) , endothelial cell growth supplement ( 30 μg/ml , Millipore ) , L-glutamine ( 2 mM ) , Pen/strep ( 1X ) 37°C , 5% CO2 . Murine cardiac endothelial cell line ( MCEC-1 ) [55] was cultured in DMEM supplemented with 20% FCS , Heparin ( 10 U/ml ) , endothelial cell growth supplement ( 30 μg/ml , Millipore ) , L-glutamine ( 2 mM ) , Pen/strep ( 1X ) at 37°C with 5% CO2 . Human and murine endothelial cells were cultured in flasks or plates coated with 1% gelatin ( Sigma ) . A549 and HUVEC cells were seeded into 24-well plates and cultured to confluence . GAS were cultured in THB and washed and resuspended in PBS before being inoculated into each well at an MOI of 10 . Adhesion assays were carried out following incubation for 30 minutes at 37°C , 5% CO2 . Cells were washed 3 times with 1x HBSS ( Gibco ) to remove non-adherent bacteria and cell-associated bacteria were released following lysis of the cell monolayer in sterile water with trypsin-EDTA ( Gibco ) and quantified by serial dilution and plating onto CBA . To ensure any differences observed in bacterial recovery were not due to variation in cellular internalization , replica experimental wells were set up . Instead of cell lysis , monolayers were treated with gentamicin ( 100 μg/ml ) in RPMI+FCS+L-glut for 2 hours , 37°C , 5% CO2 . Cells were subsequently washed and lysed as described above and internalized bacteria were quantified as for adhesion assays . HUVEC were cultured to confluence in 96-well plates and fixed in paraformaldehyde ( 2% ) , 15 minutes , room temperature . Cells were washed 3 times in PBS and blocked in 1% BSA , 1 hour , room temperature . rScpA was added to cells in PBS or PBS-FCS ( 10% ) , in doubling dilutions from 250 ng . Bound protein was detected following incubation with mouse anti-ScpA serum ( 1:1000 ) for 2 hours at room temperature and goat-anti mouse-HRP ( 1:1000 ) ( Abcam ) for 1 hour , room temperature . Samples were incubated with 3 , 3' , 5 , 5'-Tetramethylbenzidine ( TMB ) substrate ( Sigma ) for 20 minutes , room temperature and reaction stopped following addition of 1 M H2S04 . Relative levels of rScpA binding were quantified as A450 values . All statistical analyses were performed with GraphPad Prism 5 . 0 . Comparison of two datasets was carried out using unpaired Mann-Whitney U or T-test . A p-value of ≤0 . 05 was considered significant . | The complement pathway is critical in the innate immune response to bacterial pathogens . It consists of a self-perpetuating proteolytic cascade initiated via three distinct pathways that converge at the central complement protein , C3 . Pathogens must evade complement-mediated immunity to cause disease , and inactivation of the C3 protein can dampen all effectors of this pathway . Streptococcal species are the causative agents of an array of infections ranging from the benign to lethal . Using the human pathogen Group A Streptococcus as a representative species , we show that the enzyme ScpA , which is conserved amongst the pyogenic streptococci , cleaves human C3a and also C3 , releasing abnormally sized and functionally-impaired fragments . As a result , invading streptococci were less well opsonized and host immune cells not properly activated , reducing bacterial phagocytosis and clearance . Despite manifest in vitro activity against complement factors and human neutrophils , ScpA was still able to contribute to systemic bacterial spread in mice lacking C3 and C5 . ScpA was also demonstrated to mediate streptococcal adhesion to both epithelial and endothelial cells , which may enhance bacterial systemic spread . Our study highlights the likely importance of both complement-independent and complement-dependent roles for ScpA in streptococcal pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"blood",
"cells",
"complement",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"body",
"fluids",
"pathogens",
"immunology",
"microbiology",
"animal",
"models",
"bacterial",
"diseases",
"model",
"organisms",
"experimental",
"organism",
"systems",
"bacteria",
"neutrophils",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"streptococcus",
"microbial",
"pathogens",
"mouse",
"models",
"pathogenesis",
"streptococcal",
"infections",
"immune",
"system",
"biochemistry",
"blood",
"cell",
"biology",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
]
| 2017 | Multi-functional mechanisms of immune evasion by the streptococcal complement inhibitor C5a peptidase |
The IRG system of IFNγ-inducible GTPases constitutes a powerful resistance mechanism in mice against Toxoplasma gondii and two Chlamydia strains but not against many other bacteria and protozoa . Why only T . gondii and Chlamydia ? We hypothesized that unusual features of the entry mechanisms and intracellular replicative niches of these two organisms , neither of which resembles a phagosome , might hint at a common principle . We examined another unicellular parasitic organism of mammals , member of an early-diverging group of Fungi , that bypasses the phagocytic mechanism when it enters the host cell: the microsporidian Encephalitozoon cuniculi . Consistent with the known susceptibility of IFNγ-deficient mice to E . cuniculi infection , we found that IFNγ treatment suppresses meront development and spore formation in mouse fibroblasts in vitro , and that this effect is mediated by IRG proteins . The process resembles that previously described in T . gondii and Chlamydia resistance . Effector ( GKS subfamily ) IRG proteins accumulate at the parasitophorous vacuole of E . cuniculi and the meronts are eliminated . The suppression of E . cuniculi growth by IFNγ is completely reversed in cells lacking regulatory ( GMS subfamily ) IRG proteins , cells that effectively lack all IRG function . In addition IFNγ-induced cells infected with E . cuniculi die by necrosis as previously shown for IFNγ-induced cells resisting T . gondii infection . Thus the IRG resistance system provides cell-autonomous immunity to specific parasites from three kingdoms of life: protozoa , bacteria and fungi . The phylogenetic divergence of the three organisms whose vacuoles are now known to be involved in IRG-mediated immunity and the non-phagosomal character of the vacuoles themselves strongly suggests that the IRG system is triggered not by the presence of specific parasite components but rather by absence of specific host components on the vacuolar membrane .
The IFNγ-inducible , immunity-related GTPases ( IRG proteins ) are a family of proteins essential for innate resistance of mice against certain intracellular pathogens . Until now , the protozoon Toxoplasma gondii [1] , [2] , [3] , [4] , [5] and its very close relative , Neospora caninum [6] , [7] , as well as two strains of the bacterium Chlamydia [8] , [9] , [10] , [11] , [12] have been shown to be controlled by the IRG system . However , the IRG resistance system does not engage many other highly diverse organisms , including Salmonella , Listeria , Mycobacteria , Trypanosoma , Rhodococcus or Plasmodium , and the murine-specific strain of Chlamydia [reviewed in [13]] . From extended work in the Toxoplasma system , we and others have demonstrated that effector IRG proteins such as Irga6 , Irgb6 , Irgb10 , Irgb2-b1 and Irgd relocalise from their cytosolic compartments to the cytosolic face of the parasitophorous vacuolar membrane ( PVM ) [4] , [14] , [15] in a GTP-dependent [16] , [17] and cooperative manner [18] . In electron microscopy images , the PVM appears ruffled and vesiculated [3] , [4] , [19] . It is proposed that this action reduces the effective surface-to-volume ratio , putting the PVM under tension against the elastic cytoskeleton of the parasite and leading ultimately to its rupture [13] . Once exposed to the cytosol the parasite dies for unexplained reasons [3] , [4] , [20] . IRG proteins can be divided into the effector GKS subfamily , including the IRGA and IRGB proteins and Irgd ( all carrying a canonical GxxxxGKS motif in the P-loop of the GTP-binding site ) and the regulatory GMS subfamily , namely Irgm1 , Irgm2 and Irgm3 ( with a non-canonical GxxxxGMS motif ) [21] , [22] . GMS proteins populate the vacuoles of T . gondii or inclusions of Chlamydia either not at all ( Irgm1 ) or only to a limited extent ( Irgm2 , Irgm3 ) [1] , [4] , [18] , [23]; their function is to inhibit inappropriate GTP-dependent activation of the GKS proteins on host vesicular compartments in IFNγ-induced cells before the parasite enters [16] . It is not known why the action of IRG proteins is restricted to so few and so dissimilar parasitic organisms . We hypothesized that unusual features of the intracellular replicative niches of T . gondii and Chlamydia strains , neither of which resembles a phagosome , might hint at a common principle . To test this hypothesis we decided to examine cell-autonomous resistance in vitro to the microsporidian , Encephalitozoon cuniculi . The Microsporidia , recently re-classified as the earliest divergent group of the Fungi [24] , [25] , are abundant , obligate intracellular eukaryotic parasites of many diverse animal groups including mammals . E . cuniculi is a convenient representative since it is easily cultivated in vitro and its genome is fully sequenced , which is at 2 . 9 Mb one of the smallest known eukaryotic genomes [26] . E . cuniculi and its relatives E . hellem and E . intestinalis are common opportunistic pathogens for immunocompromised humans [27] . Microsporidia , including E . cuniculi , have a peculiar entry mechanism into host cells utterly unlike conventional phagocytosis . The thick-walled unicellular spore of E . cuniculi contains the organism itself ( the sporoplasm ) and a coiled proteinaceous tube , the polar tube , which can be suddenly extruded as a result of an osmotic stimulus and pushes a deep and narrow invagination in any adjacent host cell plasma membrane . The sporoplasm is then expelled through the polar tube and can be found deep in the host cytoplasm in an intracellular parasitophorous vacuole bounded by a membrane mainly derived from the host plasma membrane [reviewed in [28]] . The intracellular sporoplasm , now termed meront , divides repeatedly in its vacuole and eventually differentiates into large numbers of spores , finally lysing the host cell to release the mature environmentally-resistant spores [29] . By virtue of its remarkable non-phagocytic entry mechanism , this natural parasite of rodents and rabbits was an interesting potential target for the IRG resistance system , and it has been reported that IFNγ induces strong cell-autonomous immunity against this organism [30] , [31] , [32] . In the present study we show that the IRG system is indeed required for cell-autonomous resistance to E . cuniculi . We confirm the development of resistance following IFNγ treatment of fibroblasts and show that several IRG proteins localise to the PVM of intracellular E . cuniculi . Moreover , E . cuniculi infection triggered IFNγ-dependent reactive cell death , as seen earlier in T . gondii resistance [20] . To demonstrate the importance of IRG proteins in IFNγ-dependent growth restriction of E . cuniculi , we show that IFNγ-mediated resistance was completely lost in mouse cells that are deficient in the regulatory GMS proteins , Irgm1 and Irgm3 , a double deficiency that inactivates the whole IRG system . The phylogenetic range of the three classes of target organism of the IRG resistance system - bacteria , a fungus and protozoa - and their vastly dissimilar biology , strongly suggests that the specificity with which IRG proteins localise to parasitophorous vacuoles relates to a common characteristic of the host-derived membrane of such vacuoles rather than to a common ligand derived from the parasites themselves .
It was first reported in 1995 that IFNγ induction restricts microsporidial growth in mammalian cells in vitro using E . cuniculi infection of murine peritoneal macrophages [30] . Subsequent studies confirmed the suppressive effect of IFNγ on E . cuniculi growth as well as on E . intestinalis using murine peritoneal macrophages [31] , [32] , the murine enterocyte cell line CMT-93 and human enterocyte cell line Caco-2 [33] as well as primary human monocyte-derived macrophages [34] . Furthermore , IFNγ-deficient mice are susceptible to E . cuniculi and E . intestinalis infection [32] , [35] , [36] , [37] . It is characteristic of the IRG resistance system that it is efficient in non-myeloid cells such as fibroblasts [1] , [4] , [38] . We therefore tested IFNγ-dependence of cell-autonomous resistance against E . cuniculi in primary mouse embryonic fibroblasts . IFNγ-induced and uninduced C57BL/6 mouse embryonic fibroblast ( MEF ) monolayers were infected with E . cuniculi spores and the replication of the parasites followed by immunofluorescence microscopy and Western Blot analysis . E . cuniculi was detected with an antibody directed against a cytoplasmic protein ( mAB 6G2 ) of the earliest infectious stages ( meronts ) , or an antiserum against the spore wall protein 1 ( pAS anti-SWP1 ) , which is synthesized later in infection [39] . A time series of 0 . 5 to 24 hours post infection showed continuous IFNγ-dependent loss of E . cuniculi meronts ( determined by counting of 6G2-positive meronts per host nuclei ) . Meront numbers at the earliest time point measured ( 0 . 5 hours ) were equivalent in IFNγ-treated and untreated cells showing that E . cuniculi invasion into the host cells was not significantly affected by prior IFNγ induction ( Figure 1A ) . Next , we quantified not only single meronts , but also meronts that had replicated by binary fission ( double meronts ) and compared uninduced to IFNγ-induced MEF cells 24 hours post infection . Of the few surviving meronts at 24 h , very few had successfully divided in IFNγ-induced cells ( Figure 1B ) . In addition , Western Blot analysis of whole cell lysates from infected MEF cells showed that meront development as well as the formation of new spores was blocked by IFNγ ( Figure 1C ) . In uninduced MEF cells , E . cuniculi-dependent protein bands were detected with the meront-specific antibody , indicative of replication , and with the spore-specific antiserum , indicative of maturation at 2 days post infection . The intensity of these parasite-specific bands further increased at 5 days post infection . In contrast , these bands could not be detected in E . cuniculi-infected IFNγ-induced cells either 2 or 5 days after infection . Taken together , IFNγ kills meronts , inhibits meront replication , and blocks spore formation of E . cuniculi in primary mouse embryonic fibroblasts . When T . gondii infects IFNγ-treated mouse fibroblasts , the induced IRG proteins , especially the effector GKS proteins , accumulate on the PVM and lead to disruption of the vacuole [20] . To examine whether similar IRG-related processes might also occur on the microsporidian vacuole , we co-stained IFNγ-treated , E . cuniculi-infected , MEF cells with immunological reagents against individual GKS effector proteins ( Irga6 , Irgb6 and Irgd ) as well against the GMS regulator proteins ( Irgm1 and Irgm2 ) 24 h post infection ( Figure 2 ) . Some meronts were indeed coated with IRG proteins , but most were IRG-negative . Both Irga6-coated and uncoated vacuoles were found together in multiply infected host cells ( Figure 2A ) . Irga6 and Irgb6 were found on vacuoles at higher frequencies , while Irgd and Irgm2 were found at lower but consistent frequencies ( below 5% ) ( Figure 2B–E ) . Irgm1 was never found at the E . cuniculi PVM ( more than 1000 meronts in three independent experiments were analysed ) . In IFNγ-induced cells , cytoplasmic Irga6 is predominantly in the GDP-bound form , but accumulates on the T . gondii PVM in the GTP-bound activated form , which can be specifically detected with the mouse antibody10D7 [17] . Because we could not conduct a co-staining with the mouse anti-meront antibody , 6G2 , in combination with 10D7 , we identified the meront via its enhanced DAPI signal to show that indeed Irga6 was accumulating on E . cuniculi vacuoles in the GTP-bound state ( Figure 2F ) as in T . gondii immunity . The number of vacuoles positive for Irga6 or Irgb6 was examined in more detail at different time points after infection ( Figure 2G ) . The frequency of Irga6- and Irgb6-positive vacuoles varied between experiments ( 1–20% ) , but did not significantly increase or decrease between 0 . 5–24 hours post infection ( Figure 2G ) as the number of meronts progressively dropped , suggesting relatively fast clearance of the IRG-positive vacuoles . A detailed view of IRG loading onto the T . gondii PVM has been established . IRG proteins accumulate on the PVM in a hierarchical order with Irgb6 , Irgb10 and Irga6 as pioneers and demonstrate cooperative behaviour by stabilizing each other at the PVM [18] . In order to investigate cooperative loading on the E . cuniculi PVM , we conducted triple immunofluorescent stainings to identify the meront and two GKS proteins , Irga6 and Irgb6 . Individual vacuoles positive for both IRG proteins were observed at early and late time points , such as 12 h post infection ( Figure 3A ) and 24 h post infection ( Figure 3B , C ) . In most cases , Irga6 and Irgb6 co-localised to single vacuoles ( Figure 3B ) , although often without accurate spatial coincidence on the PVM ( Figure 3C ) . Notably , the number of E . cuniculi vacuoles accumulating both IRG proteins was higher than single-coated ones ( Figure 3D ) . In view of the low frequencies of accumulation of individual IRG proteins , it is clear that the frequency of double-loaded vacuoles is highly non-random , suggesting that loading is cooperative between different IRG proteins , as seen on the T . gondii PVM . To assess the importance of IRG proteins in the IFNγ-dependent restriction of E . cuniculi , we investigated the development of the parasite in cells derived from IRG knock-out mice . First , we examined E . cuniculi infection in IFNγ-induced primary wildtype and Irgm1/Irgm3−/− MEF cells , which lack the two regulator GMS proteins , Irgm1 and Irgm3 , and also express reduced levels of GKS proteins [40] . The number of meronts observed in Irgm1/Irgm3−/− MEF cells 24 h after infection was the same whether the cells were induced with IFNγ or not ( Figure 4A , B ) . In contrast , and as observed in Figure 1 , the number of meronts in IFNγ-induced wildtype cells was drastically reduced at 24 h after infection compared with uninduced controls ( Figure 4A , B ) . We next assayed IFNγ-inducible resistance to E . cuniculi in transformed fibroblasts from mice deficient in single IRG genes as well as in the Irgm1/Irgm3−/− double knock-out cells . Parasite growth was assessed with the anti-meront antibody by Western blot , while the expression of Irgb6 ( or Irga6 ) confirmed successful IFNγ induction ( Figure 4C–D ) . In wildtype cells , IFNγ-induction resulted in complete loss of the meront marker at 2 and 5 days after infection , while in the Irgm1/Irgm3−/− double knock-out cells IFNγ-induction caused no inhibition of meront growth . Single mutants for either Irga6 or Irgd , two members of the GKS effector subfamily , showed no loss of resistance relative to wildtype cells . However , cells lacking one GMS protein , Irgm1 or Irgm3 , both showed clear susceptibility phenotypes . Susceptibility of the Irgm1-deficient cells was incomplete , while Irgm3-deficient cells were apparently as susceptible as Irgm1/Irgm3 double-deficient cells . Interestingly , Irgm3 deficiency also has a stronger susceptibility phenotype than Irgm1 deficiency for T . gondii in IFNγ-induced MEFs ( unpublished observations ) . The stronger phenotype from the GMS knock-outs is expected , because these deficiencies deregulate all the GKS effectors [41] . The undetectable effects of the two GKS effector knock-outs is consistent with the much weaker in vivo phenotypes of single Irga6 and Irgd deficiencies in T . gondii infection [2] , [42] . A deficiency of several GKS effectors would be expected to show a stronger phenotype . We and others have documented the direct disruption of the IRG protein-coated T . gondii PVM , followed by necrotic host cell death in mouse cells induced by IFNγ [4] , [15] , [20] , [43] . It was of interest to find out whether this consequence of IRG protein action could also be observed in IFNγ-induced mouse cells infected with E . cuniculi . In the first experiments we stained infected primary MEF wildtype cells under live-cell conditions with the membrane-impermeable dye propidium iodide in order to stain nuclei of necrotic cells , and with the membrane–permeable dye Hoechst , which stains all nuclei . We found a significant excess of propidium iodide-positive nuclei in E . cuniculi infected and IFNγ-treated cells compared to untreated or single-treated control samples ( Figure 5A , B ) . Next , we used a standard formazan-based colorimetric assay to measure viability of MEF wildtype cells with increasing multiplicity of infection with E . cuniculi . At one day post infection , viability of infected cells was significantly reduced in dependence of IFNγ-induction and this was even more pronounced at 2 days post infection ( Figure 5C ) . Thus , in the presence of IFNγ , E . cuniculi infection seems to induce the same response as T . gondii , namely reactive death of the host cell itself . Restricting nutrient acquisition is a common defence mechanism against intracellular parasites . Deprivation of tryptophan by the IFN-inducible indoleamine 2 , 3-dioxygenase ( IDO ) is often claimed to be the main inhibitor of T . gondii replication in IFNγ-induced human fibroblasts [reviewed in [44] , [45]] , following reports that replication can be rescued by supplementation of the medium with tryptophan [46] , [47] . IDO-mediated growth restriction of E . intestinalis has been proposed following observations in a mouse enterocytic cell line CMT-93 [33] . However , another study in activated mouse peritoneal macrophages showed that L-tryptophan supplementation failed to rescue the infection [48] . In view of these apparently inconsistent results , we analysed E . cuniculi growth in IFNγ-induced mouse cells following tryptophan supplementation ( Figure 6 ) . In wildtype MEF cells , as well as in CMT-93 cells , IFNγ-mediated growth restriction on E . cuniculi could not be reversed by supplementation with excess tryptophan , arguing strongly against mediation of the inhibition via IDO . Taken together with the complete loss of resistance caused by IRG protein deficiencies , we conclude that the IFNγ-mediated restriction of E . cuniculi in non-myeloid cells is mediated exclusively by the IRG system in mice .
The IFN-inducible IRG proteins of the mouse are essential for resistance against some strains of the intracellular bacterium , Chlamydia , and against the intracellular protozoon , Toxoplasma gondii , but seem to play no role in resistance against a multitude of other intracellular bacterial and protozoal infections . The combination of selectivity and lack of phylogenetic consistency in IRG protein action calls for a mechanistic explanation that unifies the two widely disparate target species while excluding organisms that do not engage the IRG system . The purpose of this study was to test the hypothesis that the key lies in how different organisms enter the host cell . Most of the organisms that are ignored by the IRG system , such as Salmonella , Listeria , Leishmania , Mycobacteria , and Rhodococcus , engage the phagocytic mechanism and are taken up into and reside , whether temporarily or permanently , in more or less modified phagosomes . In contrast T . gondii typically enters actively using force generated by its own cortical cytoskeleton without engaging the phagocytic mechanism [reviewed in [49] , [50]] . However , it has recently been described that T . gondii may enter macrophages , but not fibroblasts , by phagocytic uptake , but this is followed by active exit from the phagosome into a conventional parasitophorous vacuole and the loading of Irgb6 appears to be unaffected [51] . Chlamydia is taken up by an unknown process with some features of clathrin-mediated endocytosis but none of phagocytosis [52] . With this disparity in mind , the working hypothesis behind this paper was that organisms that enter cells without engaging the phagocytic mechanism may become preferential targets for IRG protein-mediated resistance , regardless of their taxonomic status . To generalise this idea , we tested another intracellular organism with an anomalous , non-phagocytic mode of cellular invasion and a wide taxonomic divergence from the other two known IRG protein targets: the microsporidian , Encephalitozoon cuniculi . The unambiguous conclusion from our experiments is that infection by E . cuniculi is resisted in IFNγ-induced mouse fibroblasts by the action of the IRG proteins , and several aspects of the process closely resemble features that have been studied in detail in T . gondii infection . Resistance is associated with accumulation of IRG proteins onto at least a proportion of the intracellular organisms within the first 30 minutes after infection , a time at which IRG protein accumulation onto the T . gondii vacuolar membrane is already well-advanced [18] . At the light microscopical level it is not possible to say precisely where the IRG proteins are localised , but images from later time points after infection , when the vacuole is enlarged and the PVM is separated from the organism by an intravacuolar space , suggest that the IRG proteins are loaded onto the parasitophorous vacuole membrane . The cooperative pattern of loading of the different IRG proteins is also familiar from T . gondii . The frequency of vacuoles loaded at any time is low , but the majority of vacuoles carry more than one IRG protein ( Figure 3 ) . This result could of course also arise if only a few vacuoles are receptive to IRG proteins at any time . However , data from T . gondii showed that the loading of Irgb6 was stabilised and enhanced by the loading of Irga6 , and thus clearly cooperative [18] . There is also a tendency in the E . cuniculi infection , perhaps not so well marked as in T . gondii infection , for Irgb6 to load more vacuoles than Irga6 , and Irgd to load fewer . Also , as in T . gondii [1] , [4] and in C . trachomatis infection [8] , [10] , [23] , the IRG regulatory protein , Irgm1 , does not load onto any E . cuniculi vacuoles , while Irgm2 can be found on some . In another respect , however , the loading of E . cuniculi vacuoles with IRG proteins appears to be different from the loading of T . gondii vacuoles . With avirulent T . gondii , the number of vacuoles loaded with IRG proteins rises to as much as 90% of all vacuoles within 2 h after infection . With E . cuniculi , the number of vacuoles loaded reaches a plateau between 5 and 15% within 30 minutes of infection , and persists at that level for many hours while the number of live meronts progressively falls ( Figure 2 ) . These different loading behaviours can be reconciled with qualitatively similar processes operating on the vacuoles of both organisms , if the initiation of IRG protein loading onto individual E . cuniculi vacuoles takes on average longer than onto T . gondii vacuoles , and if E . cuniculi vacuoles subsequently disintegrate and are cleared with faster kinetics than T . gondii vacuoles . With increasing time after infection more and more parasites are cleared from the cells , accounting for the long , slow loss of detectable meronts ( Figure 1 ) reaching about 90% only 24 h after infection ( Figure 7 ) . Light microscopy does not allow us to see exactly what happens to E . cuniculi vacuoles after loading with IRG proteins . Clear-cut disruption typical of the IRG-loaded T . gondii vacuole [20] is not easy to register . Nevertheless the vacuoles and their included parasites disappear ( see Figure S1 ) . In T . gondii , the disruption of the vacuole is followed within about 20 minutes by the death of the parasite and after an hour or two by the necrotic death of the infected cell . We also observed an excess of dead , presumably necrotic , cells in IFNγ-induced fibroblasts infected with E . cuniculi . Lastly , as with T . gondii resistance , the IRG system appears to be the only mechanism in IFNγ-induced mouse fibroblasts that is capable of restriction of E . cuniculi . In fibroblasts from mice double deficient for the regulator IRG proteins , Irgm1 and Irgm3 , in which the whole IRG system is largely disabled , all IFNγ-inducible resistance against the growth and development of the parasite was lost , and the IFNγ-inducible catabolic enzyme for tryptophan played no role in resistance against E . cuniculi . In summary , every property of the IRG-dependent resistance mechanism that has been analysed for T . gondii is probably also valid against E . cuniculi , and , to the extent that it is known , also against Chlamydia . Since effective resistance dependent on IRG proteins seems to be perfectly correlated with the accumulation of IRG proteins on the parasitophorous vacuole , the challenge is to determine the common factor that enables IRG proteins to accumulate on the vacuoles of these three organisms but not on the vacuoles of other organisms . These three organisms cover three kingdoms of life: protozoa , bacteria , and fungi . The broad phylogenetic distribution makes it unlikely a priori that the IRG proteins target a common ligand expressed by all restricted pathogens on their vacuolar membranes . Our preferred view builds on a hypothesis first formulated by Martens [53] to account for the targeting of IRG proteins to the T . gondii vacuole rather than to other cellular organelles . Martens proposed the existence of a self-derived factor X expressed on the membranes of cellular organelles that inhibits the accumulation and activation of IRG proteins on these sites , thereby protecting these organelles from IRG protein mediated damage . Parasitophorous vacuoles , lacking factor X , would be exposed to IRG accumulation and activation . This elegant “missing self” model was confirmed some time later and “factor X” was revealed to be the three GMS proteins , Irgm1 , Irgm2 and Irgm3 , which are bound to distinct subsets of organellar membranes and act as guanine nucleotide dissociation inhibitors of the effector GKS proteins at these sites [16] . In the absence of one or more GMS proteins , GKS proteins form activated , GTP-bound assemblies in the cytoplasm , probably associated with “unprotected” organellar membranes [23] , [41] . The GMS IRG proteins seem to fulfil exactly the role of Martens' Factor X for the distinction between intracellular organelles and a parasitophorous vacuole . However , GKS effector IRG proteins do not accumulate or activate on the plasma membrane , which to the best of our knowledge is not protected by any GMS protein . We are therefore forced to introduce a new hypothetical inhibitor associated with the plasma membrane that inhibits GKS activation at that location ( Figure 8 ) . Parasitophorous vacuoles are formed by invagination of the plasma membrane , and as we know with some precision from experiments with T . gondii , the vacuoles are receptive to IRG loading and activation immediately after parasite entry [18] . Thus entry of the parasite and formation of the parasitophorous vacuole must entail loss of the hypothetical plasma membrane-bound inhibitor . We propose that this is the essential distinction between those organisms that do , and those that do not , engage the IRG system , and that loss of the plasma membrane inhibitor is due to the unusual , non-phagocytic entry mechanisms of all three parasites . Recently , the Atg5-dependent module of the ubiquitin-like conjugation system of autophagy has been proposed to mediate IRG targeting to the vacuole of T . gondii [19] , [54] , [55] . But since IRG and GBP proteins in Atg5−/− cells form cytosolic GTP-bound aggregates [18] , [19] , [55] which are unable to target pathogens , loss of resistance appears rather to be caused by deregulated effector protein homeostasis . It has also been difficult to localise any autophagic component to T . gondii parasitophorous vacuoles [4] . Recent data from Choi et al . ( 2014 ) suggest occasional localisation of native LC3II at the PVM , but on only a small minority of vacuoles , uncorrelated with the presence of IRG proteins [54] . The entry of T . gondii into cells has been studied in considerable detail and probably provides the best system for establishing the identity of the plasma membrane inhibitor . Several categories of protein are depleted from the developing vacuolar membrane , presumably as a result of the sieving action of the parasite-derived RON protein complex formed at the moving junction through which the parasite enters the cell [56] , [57] , [58] , [59] . Apart from transient activation of host actin at the moving junction [60] , [61] , there is also no evidence that components of the cortical cytoskeleton remain associated with the nascent vacuole . In the case of Encephalitozoon cuniculi , electron microscopic studies suggested that the PV membrane is derived from the host cell [29] . Subsequent studies from Bohne and colleagues established that the early PVM is non-fusogenic and devoid of any endolysosomal markers immediately after invasion [39] , and moreover that the lipids of the PVM are indeed host cell-derived and that the PVM also forms simultaneously with the extrusion of the sporoplasm . This is all consistent with the suggestion that the early PVM is an invagination of the host cell plasma membrane [28] , [62] . Due to the high speed of host cell entry , we suggest that physical forces may determine the presence and composition of host cell surface proteins on the newly formed PVM , which would be a prerequisite of IRG protein recognition . Although E . cuniculi spores are actively phagocytosed by macrophages , this is unlikely to be a biologically significant entry route , because it does not contribute to the intracellular meront population in mouse macrophages [63] or mouse embryonic fibroblasts ( unpublished results Springer-Frauenhoff ) . This study was designed based on knowledge of the interaction of the mouse IRG protein system and T . gondii . We found four major similarities in the action of the IRG system in defense against E . cuniculi: ( 1 ) the relocalisation of multiple IRG proteins to the cytosolic face of the PVM; ( 2 ) cooperativity by double-loading; ( 3 ) IFNγ- and infection- dependent host cell death and ( 4 ) IDO-independent IFNγ-mediated restriction in mouse cells . While the IRG system clearly plays an essential role in defending mice and probably other small rodents against certain infections , there is every reason to believe that the system is effectively completely absent from humans and higher primates , birds , cats and doubtless many other mammalian species too [21] . The sporadic occurrence of a developed IRG system among vertebrate groups suggests that its possession is costly and only justified when certain classes of parasite exert intense selection pressures [14] .
All animal experiments were conducted under the regulations and protocols for animal experimentation according to the German “Tierschutzgesetz” ( Animal Experimentation Law ) . The local government authorities , Landesamt für Natur- und Umweltschutz Nordrhein-Westfalen , and its ethics committee approved the work ( LANUV Permit No . 84-02 . 05 . 40 . 14 . 004 ) . Primary C57BL/6 mouse embryonic fibroblasts ( MEFs ) were prepared from mice at day 14 post coitum . Irgm1−/− , Irgm3−/− , Irgm1/Irgm3−/− , Irgd−/− MEFs ( kindly provided by Greg Taylor ) or Irga6−/−MEFs [42] were immortalized by transfection of pSV3-neo plasmid [64] and pPur ( Clontech , Saint-Germain-en-Laye , France ) in a ratio 9∶1 using FuGENE HD ( Roche , Mannheim , Germany ) according to the manufacturer's instructions . After 24 h , cells were put under selection with 3 µg/ml puromycin ( Clontech ) . Primary and transformed MEFs as well as mouse rectal carcinoma CMT-93 cells ( ATCC CCL-223 ) were cultured in DMEM , high glucose ( Invitrogen Life Technologies , Darmstadt , Germany ) supplemented with 10% fetal calf serum ( FCS , Biochrom AG , Berlin , Germany ) , 2 mM L-glutamine , 1 mM sodium pyruvate , 1× MEM non-essential amino acids , 100 U/ml penicillin and 100 mg/ml streptomycin ( all PAA , Pasching , Austria ) . Human foreskin fibroblasts ( Hs27; ATCC CRL-1634 ) were cultured in IMDM , high glucose ( Invitrogen Life Technologies ) supplemented with 5% FCS , 100 U/ml penicillin and 100 mg/ml streptomycin ( PAA ) . Cells were stimulated with 200 U/ml of mouse IFNγ ( PeproTech , Rocky Hill , NJ , USA ) for 24 h . For IDO-inhibition , L-tryptophan ( W ) ( Sigma-Aldrich Co . , Saint Louis , MO , USA ) was added 15 min prior to infection . E . cuniculi spores were a generous gift from Prof . Peter Deplazes ( University of Zürich , Switzerland ) . Spores were routinely propagated in Hs27 cells as described in [62] . Briefly , infected monolayers were scraped 7–12 days post infection and the suspension was passed through a 26G needle . The first centrifugation ( 10 min at 500 rpm ) removed the host cell debris , whereas the second centrifugation ( 20 min at 2500 rpm ) sedimented the spores . A stock solution with 4×107 spores/ml PBS was stored at 4°C for max . 3 month . For infection assays , 8–12×104 host cells were seeded in 6-well plates 48 h prior infection , optionally stimulated , and infected with a multiplicity of infection ( MOI ) of 10 parasites per host cell for microscopic assays and MOI 5 for Western Blot analysis . In order to obtain synchronous infection , spores were allowed to infect the cells for 2–4 h followed by one careful washing step with PBS and addition of fresh medium . Cells were fixed or harvested at the indicated time points post infection . For E . cuniculi genotyping , 60×107 E . cuniculi spores were centrifuged for 20 min at 2500 rpm , resuspended in 200 µl PBS and 20 µl of Proteinase K , DNA was isolated with the DNeasy Blood & Tissue DNA purification kit ( Qiagen , Hilden , Germany ) according to manufactures instructions . The rRNA gene region of large ( LSU rRNA ) and small ribosomal subunit ( SSU rRNA ) and ITS region were amplified as described in [65] . The E . cuniculi stain used in this study had the genotype I . The following immunoreagents were used: rabbit anti-Irgm1 polyclonal antiserum ( pAS ) rbMAE15 [66] , rabbit anti-Irgm2 pAS H53/3 [4] , [18] , rabbit anti-Irga6 pAS 165/3 [67] , anti-Irga6 mouse monoclonal antibody ( mAB ) 10D7/10E7 [17] , anti-Irgb6 mouse mAB B34 [68] , anti-Irgb6 goat polyclonal antibody ( pAB ) A20 ( sc-11079 , Santa Cruz Biotechnology , Inc . , Santa Cruz , CA , USA ) , anti-meront mouse mAB 6G2 [39] , anti-SWP1 [69] , anti-Calnexin rabbit ( pAB ) ( Calbiochem Merck KGaA , Darmstadt , Germany ) . Secondary antibodies were Alexa Fluor 488/555/647-labeled donkey anti-mouse , -rabbit , and -goat antisera ( all Molecular Probes , Invitrogen Life Technology ) , donkey anti-rabbit- ( GE Healthcare , Freiburg , Germany ) , and goat anti-mouse-HRP ( horseradish peroxidase ) ( Pierce , Thermo Fisher Scientific , Bonn , Germany ) antisera . 4′ , 6-Diamidino-2-phenylindole ( DAPI , Roche , Mannheim , Germany ) was used for nuclear staining at a final concentration of 0 . 5 mg/ml . Immunocytochemistry was carried out on paraformaldehyde-fixed cells grown on glass cover slips as described earlier [4] . In brief , cells were permeabilized and blocked with 3% BSA and 0 . 1% saponin ( both Roth , Karlsruhe , Germany ) in PBS , stained with the primary antibodies diluted in blocking buffer for 1 h at room temperature or overnight at 4°C following incubation with the secondary antibody diluted in blocking buffer for 30 min at room temperature . Between all steps cells were triple washed with 0 . 1% saponin in PBS and then mounted on glass microscopic slides in ProLong Gold anti-fade reagent ( Invitrogen Life Technology ) . The images were taken with an Axioplan II fluorescence microscope and AxioCam MRm camera and processed by Axiovision 4 . 7 ( all Zeiss , Oberkochen , Germany ) . All samples were counted blind . Live cell imaging was performed in μ-slide I chambers ( Ibidi , Munich , Germany ) as described earlier [20] . For live cell experiments , wt MEF cells were transiently transfected with pEGFP-N3-Irga6-ctag1 [20] using FuGENE HD ( Roche ) according to the manufacturer's instructions and induced with 200 U/ml IFNγ . After 24 hours cells were infected with E . cuniculi spores at a MOI 50 in phenol red-free RPMI 1640 ( PAA ) . After infection with E . cuniculi , the cells were observed with a Zeiss Axiovert 200 M motorized microscope fitted with a wrap-around temperature-controlled chamber ( Zeiss ) . The time-lapse images were obtained and processed by Axiovision 4 . 6 software ( Zeiss ) . At 2 or 5 days post infection , MEF or CMT-93 cells were washed with PBS once and directly lysed in 200 µl 2× SDS sample buffer ( 2% SDS , 100 mM Tris/HCl ( pH 6 . 8 ) , 10% Glycerol , 0 . 005% bromophenol blue , 1 . 4% β-mercaptoethanol ) . The lysates were transferred into Eppendorf tubes and boiled 5–10 min at 95°C . 15–20 µl were subjected to 10% SDS-PAGE and Western blot . Protein transfer was confirmed by staining the nitrocellulose membranes with Ponceau S solution [0 . 2% Ponceau S ( Roth ) and 3% acetic acid in dH2O] . Membranes were blocked in 5% non-fat dry milk in PBS and probed for the proteins of interest with the indicated primary and HRP-coupled secondary antibodies . Primary MEF cells ( 5000 cells/96-well ) were seeded and induced with IFNγ for 24 h or left untreated . The cells were then infected with E . cuniculi spores at the indicated MOI for 24 h or 48 h . Thereafter , viable cells were quantified by the CellTiter 96 AQueous non-radioactive cell proliferation assay ( Promega , Mannheim , Germany ) according to the manufacturer's instructions . Infection with avirulent T . gondii Me49 served as positive control [20] . MEF cells grown in 6 cm-dishes were induced with IFNγ for 24 h and infected with E . cuniculi spores at a MOI 10 . At 24 h post infection , Bisbenzimide Hoechst 33342 and Propidium Iodide ( both Sigma-Aldrich ) were added to the medium ( 1 µg/ml final concentration for both ) and incubated at 37°C for 15 min . 10 fluorescent pictures per sample were photographed with the Zeiss Axiovert 200 M microscope with a 10 fold magnification . Total cell number ( Hoechst-positive nuclei ) and dead cells ( PI-positive nuclei ) were automatically enumerated using the Volocity software ( PerkinElmer , Santa Clara , CA , USA ) . At least 500 cells were counted per sample and percentage of dead cells per total cell number was calculated . In five independent experiments , a total of 10 . 000 cells or more was counted per sample . | For some time we have studied an intracellular resistance system essential for mice to survive infection with the intracellular protozoan , Toxoplasma gondii , that is based on a family of proteins , immunity-related GTPases or IRG proteins . Immediately after the parasite enters a cell , IRG proteins accumulate on the membrane of the vacuole in which the organism resides . Within a few hours the vacuole membrane breaks down and the parasite dies . A puzzle is why this mechanism works on Toxoplasma , but only on one other organism among the many tested , namely the bacterial species , Chlamydia . What do these widely different parasites have in common that so many other bacteria and protozoa lack ? Neither Toxoplasma nor Chlamydia is taken up by conventional phagocytosis . In this paper we suggest that this is an important clue by showing that a microsporidian , Encephalitozoon cuniculi , a highly-divergent fungal parasite , which also invades cells bypassing phagocytosis , is resisted by the IRG system . Therefore , we propose here the “missing self” principle: IRG proteins bind to vacuolar membranes only in the absence of a host derived inhibitor that is present on phagosomal membranes but excluded from the plasma membrane invaginated by IRG target organisms during non-phagosomal entry . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"infectious",
"disease",
"immunology",
"medicine",
"and",
"health",
"sciences",
"clinical",
"immunology",
"medical",
"microbiology",
"microsporidia",
"microbial",
"pathogens",
"biology",
"and",
"life",
"sciences",
"immunology",
"microbiology",
"fungal",
"pathogens"
]
| 2014 | Identification of the Microsporidian Encephalitozoon cuniculi as a New Target of the IFNγ-Inducible IRG Resistance System |
Copy-number variations ( CNVs ) are widespread in the human genome , but comprehensive assignments of integer locus copy-numbers ( i . e . , copy-number genotypes ) that , for example , enable discrimination of homozygous from heterozygous CNVs , have remained challenging . Here we present CopySeq , a novel computational approach with an underlying statistical framework that analyzes the depth-of-coverage of high-throughput DNA sequencing reads , and can incorporate paired-end and breakpoint junction analysis based CNV-analysis approaches , to infer locus copy-number genotypes . We benchmarked CopySeq by genotyping 500 chromosome 1 CNV regions in 150 personal genomes sequenced at low-coverage . The assessed copy-number genotypes were highly concordant with our performed qPCR experiments ( Pearson correlation coefficient 0 . 94 ) , and with the published results of two microarray platforms ( 95–99% concordance ) . We further demonstrated the utility of CopySeq for analyzing gene regions enriched for segmental duplications by comprehensively inferring copy-number genotypes in the CNV-enriched >800 olfactory receptor ( OR ) human gene and pseudogene loci . CopySeq revealed that OR loci display an extensive range of locus copy-numbers across individuals , with zero to two copies in some OR loci , and two to nine copies in others . Among genetic variants affecting OR loci we identified deleterious variants including CNVs and SNPs affecting ∼15% and ∼20% of the human OR gene repertoire , respectively , implying that genetic variants with a possible impact on smell perception are widespread . Finally , we found that for several OR loci the reference genome appears to represent a minor-frequency variant , implying a necessary revision of the OR repertoire for future functional studies . CopySeq can ascertain genomic structural variation in specific gene families as well as at a genome-wide scale , where it may enable the quantitative evaluation of CNVs in genome-wide association studies involving high-throughput sequencing .
Structural variants in the human genome , such as CNVs or balanced inversions , represent a major form of variation with widespread functional consequences [1] . Numerous surveys mapping CNVs at varying levels of resolution [2] , [3] , [4] , [5] , [6] have created a comprehensive CNV inventory , with the latest survey reporting 1 , 098 CNVs on average between two individuals spanning nearly 0 . 8% of the genome [6] . Collectively , the list of reported CNVs presently involves 8 , 410 loci ( Database of Genomic Variants [3] , DGV ) when applying the frequently used operational definition for CNVs , i . e . , gains and losses of segments 1 kb or larger in size [7] . Recent studies have associated CNVs with various phenotypes , including benign and disease-related phenotypes such as cancer , HIV-1/AIDS susceptibility , autoimmunity , and complex disorders ( [1] and references therein ) . Yet , while different conceptual approaches for CNV-discovery have been developed [8] , [9] , [10] , [11] , [12] , [13] most CNV analysis approaches presently do not distinguish CNVs based on the copy-number of the underlying DNA segment , i . e . , its copy-number genotype , a distinction that is crucial for leveraging CNV assignments for studies focusing on genome evolution and genotype-phenotype associations [14] . For example , copy-number genotypes enable distinguishing bi-allelic loci ( i . e . , loci at which in addition to the reference allele either a single duplication or a single deletion allele is observed ) from multi-allelic loci ( i . e . , loci with more than one variant , such as deletion and duplication , or multiple duplications ) . Furthermore , in bi-allelic loci copy-number genotypes allow discriminating heterozygous from homozygous CNVs . Such information is crucial in association studies , where the failure to assign locus copy-numbers or to discriminate heterozygotes from homozygotes limits the statistical power . Recently , improvements in microarray technology have led to advances in CNV analysis by facilitating the ascertainment of copy-number genotypes in genomic regions amenable to hybridization by high-resolution comparative genome hybridization ( array-CGH ) or state-of-the-art SNP/CNV hybrid array platforms [6] , [14] . While microarrays have advantages in enabling CNV ascertainment at high-throughput and low-cost , their resolution can be limited in CNV-rich regions involving segmental duplications [15] ( SDs ) . This might be because of probe cross-hybridization issues , which may reduce the number of effective oligonucleotide probes that can be designed for these regions [16] . Indeed , commercial microarray-based approaches for copy-number genotyping are restricted to genomic loci for which probes are available at sufficient densities [6] , [14] , while custom array designs may compensate for probe densities with the remaining limitation of relying on regions for which effective probes can be designed . Recent breakthroughs in ‘Next Generation Sequencing’ ( NGS ) technologies have stimulated the development of computational approaches that enable the discovery of CNVs with excellent quantitative and spatial resolution [8] , [9] , [10] , [11] , [13] . In this regard , several studies have demonstrated that the sequencing depth-of-coverage of NGS reads can be employed for CNV-discovery [8] , [9] , [17] , [18] , [19] . For example , Xie et al . [18] and Chiang et al . [8] described CNV-discovery approaches conceptually related to array-CGH analysis , whereby the read-depth in genomic intervals is compared between pairs of samples to detect CNVs as relational changes in studies involving case/reference setups ( e . g . , cancer tissue vs . healthy tissue ) . Furthermore , Alkan et al . recently reported an elegant read-count based approach for mapping locus copy-number differences in large ( ≥20 kb ) SDs using high-coverage ( 6 to 20-fold coverage ) NGS data , by equating averaged and rounded read-counts in individual samples with integer locus copy-numbers [19] . However , with recent advances enabling sequencing hundreds of genomes in studies focused on population genetics or genotype-phenotype correlations , a statistical framework for copy-number genotyping will soon become a prerequisite to enable the probabilistic ascertainment of CNV sets in NGS-based association studies . To be useful for genome-wide association studies , a NGS-based copy-number genotyping approach needs to provide absolute locus copy-number estimates in a sample-specific manner and needs to be able to determine confidence values for each copy-number genotype ( to maximize statistical power ) . Furthermore , it should enable accurate ascertainment of a wide range of CNVs , including rare and common ones , and including those at the 1–20 kb size-range , a highly abundant CNV size-class [4] . Lastly , the ability to utilize low-coverage ( i . e . , ≤4× coverage ) NGS datasets , i . e . , datasets such as the ones generated by the ‘1000 Genomes Project’ ( 1000GP; see http://1000genomes . org ) , will be a crucial asset for such a copy-number genotyping approach , given that sample number and sequencing coverage will be at a constant tradeoff in future association studies . Here , we present CopySeq , a statistical framework for copy-number genotype inference from low-coverage genomes , which is available at http://embl . de/~korbel/copyseq/ . As a benchmark we used CopySeq to genotype a set of CNVs previously analyzed with microarrays and obtained excellent genotyping concordances for CNVs across a wide size-range . In addition , as a proof-of-principle we used the approach to infer copy-number genotypes in the largest human gene family , with many genes and pseudogenes embedded within SDs: i . e . , we analyzed the >800 olfactory receptor ( OR ) genes and pseudogenes in the human genome . OR genes form one of the most genetically variable and rapidly evolving protein-coding gene families and display a strong enrichment for CNVs [4] , [20] , [21] , [22] compared to most other gene families . Thus , the OR gene family represents an appealing model for assessing copy-number genotype ascertainment using low-coverage sequencing and for studying the effect of CNVs on protein coding loci . Owing to the comparative nature of earlier studies , CNVs in ORs were thus far mostly reported as gains and losses relative to an arbitrarily chosen reference sample , and for most ORs no absolute locus copy-number assignments have been reported so far . Thus , the full nature and extent of copy-number variation in ORs remained unknown . Notably , it is presently unclear to what degree single deletions or duplications ( i . e . , bi-allelic ) or multiple recurrent CNV-formation events ( i . e . , multi-allelic ) affect particular OR loci , an information that is crucial for functional analyses as multiple alleles can reduce signals in association studies . Our analysis of ORs using CopySeq revealed a widespread diversity in integer locus copy-numbers in human OR loci in the 150 individuals assessed . We report a segregation of copy-number variable OR loci into bi-allelic , multi-allelic , and non-variable CNVs , with notable population differences in some OR loci . In addition , our analysis enabled us to address and further dissect genomic biases that may influence the extent of CNVs affecting ORs , including functional ( genes vs . pseudogenes ) , DNA sequence context ( non-repetitive vs . repetitive DNA ) , and evolutionary ( ‘young’ vs . ‘ancient’ ORs ) biases .
CopySeq enables the inference of copy-number genotypes in genomic loci suspected to differ in copy-number ( see Materials and Methods for details , and Figure 1 ) . The first step undertaken by CopySeq , termed locus selection , involves the definition of putative CNV loci . Loci may be selected based on biological considerations , e . g . , to enable copy-number genotyping comprehensively in previously published CNV sets , or in a more focused manner in candidate loci of an association study . Following locus selection , the mappability assessment step assesses the mappability of all k-mer subsequences of the selected loci to identify sequence stretches that are unambiguously ( uniquely ) mappable with short reads , i . e . , such with no exact duplicate sequence in the reference genome . This step assures exclusion of ( 1 ) unspecific reads and ( 2 ) reads originating from paralogous sequences . Then , in the read-mapping step DNA reads are aligned onto the reference genome , and only unique matches retained , using a fast read-mapper such as the MAQ or BWA algorithm [23] , [24] . Lastly , the copy-number genotyping step measures the locus-specific read-depth in the mappable sequence fraction and infers copy-number genotypes for each sample by relating the resulting read-depth value to the expected locus read-depth after correction for GC-content biases [25] using a smoothing spline-based approach . In particular , CopySeq uses a Gaussian classifier that regards discrete locus copy-numbers as probability distributions and infers copy-number genotypes with confidence scores ( see Materials and Methods ) . The copy-number genotype thereby indicates the diploid copy-number of a locus of interest in a given genome . In the case of bi-allelic CNVs , this enables distinguishing homozygous from heterozygous CNVs ( e . g . , heterozygous deletion = ‘1’ copy; homozygous deletion = ‘0’ copies; homozygous reference allele , or ‘no deletion’ = ‘2’ copies ) . Optionally , CopySeq incorporates redefined boundaries ( or breakpoints ) of CNVs , available for confined CNV subsets [26] , prior to the copy-number genotyping step by applying different conceptual approaches: i . e . , ‘paired-end mapping’ ( PEM ) , which identifies CNVs from paired reads that map abnormally onto the reference genome [4]; or ‘breakpoint-junction sequence analysis’ ( BJA ) , which detects CNVs by aligning sequence reads onto CNV breakpoint-junctions [26] . The rationale for applying such boundary-redefinition approaches is that accurate ( i . e . , redefined ) CNV-boundaries facilitate the proper interpretation of read-depth data , thus enabling more accurate copy-number genotype inference ( see below ) . To assess the performance of CopySeq on low-coverage genome sequences we acquired NGS data from 150 individuals with different ancestries , i . e . , genomes that were recently sequenced at low coverage in the 1000GP pilot phase 1 ( Table S1 and Materials and Methods ) : 52 unrelated African individuals with ancestry from Nigeria ( Yoruba from Ibadan; YRI ) ; 53 Asians , including 29 unrelated Chinese individuals from Beijing ( CHB ) and 24 unrelated Japanese from Tokyo ( JPT; we analyzed all 53 Asian individuals together as the “CHB+JPT” group [27] ) ; and 45 individuals of European ancestry from Utah ( CEU ) , USA , including 42 unrelated individuals and 3 members of a parent-offspring trio . The analyzed genomes were sequenced at 3–4× fold coverage on average; most reads had a read-length of 36 nt . To evaluate CopySeq , we first assembled known CNVs from human chromosome 1 , for which copy-number genotypes were previously inferred with Affymetrix SNP 6 . 0 microarrays [14] ( a SNP/CNV hybrid microarray platform ) . Namely , we compared CopySeq with copy-number genotypes from McCarroll et al . , who analyzed 270 individuals out of which 118 overlapped with our study , to initially evaluate CopySeq ( see Materials and Methods ) . Out of 100 CNVs [14] with a median size of 6 . 8 kb ( mean = 11 . 2 kb ) , only one CNV displayed less than 500 mappable 36-mers and thus was excluded ( Materials and Methods ) . We found that most CNV loci were covered by an appreciable number of sequencing reads , with a mean of 685 reads ( median = 418 ) . CopySeq was used to generate 14 , 850 ( 99 loci times 150 samples ) copy-number genotypes in this CNV set , with inferred locus copy-numbers ranging from ‘0’ copies up to ‘5’ copies ( Table S5 and Figures S8 , S9 ) . The copy-number genotypes displayed an excellent genotyping concordance of 98 . 9% with the Affymetrix-array based results ( Tables S3 , S6 , S7; Figures 2ABC , S18A ) . Note that by assuming that array-based genotypes are correct , genotyping concordances achieved with CopySeq can be considered as lower bound estimates for genotyping accuracies ( note that discordances in specific genotypes could obviously be either due to errors in the array-based genotypes or due to errors in CopySeq's genotypes ) . In general , deletion genotypes inferred by CopySeq yielded higher concordances than duplication genotypes . We quantified this by calculating the positive predictive value ( PPV ) for deletions ( 99 . 6% ) and duplications ( 89 . 1% ) ( see Table S8 ) , suggesting that while both deletions and duplications are identified at high accuracy , duplications are more difficult to ascertain than deletions by CopySeq , microarray-based genotyping , or both methods . We also estimated genotyping concordances for various different CNV size cutoffs ( Figure 2A ) , and , as expected , observed an increase in concordance between CopySeq and Affymetrix-arrays with CNV size . However , high concordances were obtained also for relatively small CNVs ( e . g . , 97 . 5% for CNVs 1–2 kb in size ) , suggesting the applicability of CopySeq across a wide CNV size-range in 3–4× coverage sequence data . Furthermore , we combined CopySeq with PEM and BJA ( Materials and Methods ) , and found that CNV-boundary redefinition with PEM or BJA leads to improved genotyping concordances . Specifically , when applying PEM we measured a genotyping concordance with Affymetrix-arrays of 99 . 6% , and when applying BJA we measured a concordance of 99 . 7% ( Tables S2 , S9 , S10 ) – although the numbers of CNVs ascertained was comparably low for BJA and PEM , as for many loci the CNV-boundaries were unknown . A further advantage of CNV-boundary redefinition is that it can help untangle complex CNV loci , as we exemplified below . We also compared CopySeq's copy-number genotypes to genotypes recently inferred with a high-resolution Agilent oligonucleotide array-CGH platform [6] through analyzing 401 CNV regions on chromosome 1 with a median size of 3 . 1 kb . Conrad et al . [6] analyzed 450 individuals out of which 149 overlapped with our study . Our analysis resulted in an excellent , albeit slightly weaker genotyping concordance of 89 . 1% – with small CNVs displaying higher concordances than large CNVs and some CNVs displaying 0% concordance ( Tables S15; Figure S18B ) . When looking for the source of the discordance we found that many of the disagreeing copy-number genotypes occurred in large ( >10 kb ) CNVs embedded in SDs ( see Text S1 ) . Altogether , four of these large CNVs were ascertained both by the Agilent and the Affymetrix platform . We found that whereas in all four regions the Agilent CGH arrays tended to agree with CopySeq in a relative sense ( mean Pearson correlation coefficient 0 . 78 ) , Agilent array-CGH was for all four CNVs discordant with both the Affymetrix SNP arrays and CopySeq in terms of the absolute copy-number reported ( note that absolute copy-numbers agreed between CopySeq and Affymetrix arrays in these regions; see Table S17 ) . To improve the comparability between the platforms , we thus reasonably excluded CNVs intersecting with SDs when analyzing the Agilent array-CGH based results . This led to an overall improved copy-number genotyping concordance of 94 . 8% between CopySeq and Agilent array-CGH ( Table S16 ) . Furthermore , we compared CopySeq to Alkan et al . 's approach [19] , which interprets averaged read-depths as locus copy-numbers ( i . e . , depth-of-coverage analysis without probabilistic genotyping model ) . We used CopySeq to analyze published short sequence reads from a single African male individual [28] which previously had been analyzed with regard to copy-number variation [19] , and obtained a better concordance with Affymetrix SNP array-based copy-numbers for CopySeq ( 97 . 2% ) than for depth-of-coverage analysis without genotyping model ( 80 . 2% ) . As suggested in [19] these concordance estimations excluded SD regions , as Alkan et al . 's approach infers copy-numbers in SDs as a genome-wide sum across all paralogous loci , rather than separately for each paralog ( see Text S1 and Table S18 ) . Next , we randomly picked eighteen common CNVs and subjected them to quantitative PCR ( qPCR ) validation in three individuals each . The CNV sizes ranged from 707 bp to 127 kb ( median size 3 . 7 kb ) and most copy-number genotypes ( 65% ) in these eighteen regions differed from the homozygous reference copy-number of ‘2’ . In total , 49 out of 54 ( 91% ) copy-number genotypes were supported by the qPCR experiments , results that were in good agreement with the genotyping concordances determined based on the microarray platforms ( see Text S1 and Table S19 ) . We further compared CopySeq's results to a set of loci that had previously been analyzed by fluorescent in situ hybridization ( FISH ) [19] , with the FISH results validating CopySeq's copy-number genotypes in four out of five assessed loci ( see Text S1 and Table S21 ) . Furthermore , we tested the effect of sequencing coverage on CopySeq's performance by generating sub-coverage datasets ( 0 . 5 , 1 , 2 , 3 , 4 , 5 , 10 , 20 , and 30× ) of the high-coverage ( ∼40× ) NA18507 genome [28] and assessed to what extent the genotype concordance with two complementary microarray platforms changed with coverage ( see Text S1 ) . Although , unsurprisingly , genotyping accuracies improved with increasing coverage , low-coverage ( 3–4× ) genomes displayed only a minor decrease in accuracy compared to a 30–40× genome ( 0 . 8–1 . 6%; see Figure S16 ) , suggesting low-coverage sequencing offers an excellent tradeoff between cost , throughput , and sensitivity in variant detection . We also assessed the effects sequencing errors may have on CopySeq's genotypes in low-coverage data , and found their influence to be minor ( see Text S1 ) . Our initial assessment of CopySeq suggested an excellent accuracy in genomic regions ascertained with microarrays . Given that in principle any genomic region mappable by unique sequence reads can be analyzed with CopySeq , we next specifically assessed CopySeq's performance in a set of relatively hard-to-ascertain regions . Namely , as a proof-of-concept , we assessed genomic loci associated with the largest human gene family – i . e . , the 388 OR gene and 463 OR pseudogene loci , most of which are not ascertained by state-of-the-art commercial genotyping array platforms [14] . In particular , we reasoned that CopySeq may enable the first comprehensive assessment of the extent of variation in terms of integer locus-copy numbers in the OR gene family . We analyzed the OR loci as ∼3 kb regions encompassing the single-exon open reading frames ( ORFs ) and downstream as well as upstream sequence stretches ( Materials and Methods ) . We assessed the mappability of reads onto OR loci and excluded only ∼5% ( 22 genes and 21 pseudogenes ) from our analysis , as they displayed less than 500 mappable 36-mers per locus ( Figure S1 ) . Following read mapping we found that OR loci were covered on average by 209 ( median = 190; see e . g . , Figure 3BCD ) uniquely aligning reads per individual . We analyzed 808 mappable OR loci in 150 humans using CopySeq to construct a comprehensive and accurate OR locus copy-number map . Eleven of these loci are on chromosome X and thus naturally differ in copy-number between females and males . We thus focused in our analyses , described below , on CNVs in the 797 remaining autosomal regions , and made use of the eleven X-chromosomal regions for optimally setting the parameters of the method ( i . e . , the Q-value; see Materials and Methods ) . Altogether , CopySeq inferred 4 , 573 loci with a copy-number different from the homozygous reference allele , which fell into 313 copy-number variable OR loci ( Table S4 ) . These involved 2 , 137 deletions ( autosomal locus copies of ‘0’ to ‘1’ ) and 2 , 436 duplications ( ‘3’ and up to ‘9’ locus copies; Figure 3E ) . We excluded six out of the 313 autosomal copy-number variable OR regions , on the basis of previous reports that these loci , or their closest paralogs in the genome , likely represent extremely rare CNVs in the reference genome , or alternatively mis-assemblies [22] ( these six loci displayed no , or very few , reference alleles ) . The remaining 307 loci were classified into 265 bi-allelic CNVs ( i . e . , 135 bi-allelic deletions and 130 bi-allelic duplications ) and 42 multi-allelic CNVs based on their inferred locus copy-numbers in 150 individuals ( Materials and Methods ) . The fraction of variable OR genes and pseudogenes is 33% ( 130/387 ) and 38% ( 137/464 ) , respectively . On average , we detected 25 copy-number variable OR loci per individual , i . e . , 8 OR genes ( 2 . 3% ) and 17 OR pseudogenes ( 3 . 9% ) ( Figure S17 ) . These correspond to , on average , 43 quantitative inter-individual copy-differences ( Figure 4 and Table S11 ) . We next assessed the accuracy of CopySeq in OR loci using three distinct approaches . First , we examined a parent-offspring trio with European ancestry sequenced at low-coverage for the segregation of 772 OR loci that appeared bi-allelic , or displayed no CNV , across the examined European individuals ( Materials and Methods ) . We found that all 16 copy-number genotypes inferred in the daughter , which included 7 heterozygous and 9 homozygous deletions , were consistent with Mendelian segregation ( Figures 5 , S10 ) suggesting high genotyping accuracy . Second , we compared CopySeq with microarrays , i . e . , copy-number genotypes inferred with Affymetrix SNP 6 . 0 arrays [14] and Agilent CGH arrays [6] ( Figures 6A , S11 ) . The Affymetrix arrays ascertain ∼5% ( 46 ) of the autosomal 3 kb OR loci , allowing us to compare ∼5 , 400 copy-number genotypes in OR loci across the 118 overlapping samples . Indeed , copy-number genotypes reported in McCarroll et al . [14] , ranging from 0 to 6 , show a strong correlation with our genotype calls ( Pearson correlation = 0 . 91; P<2 . 2e-16 ) . We estimated a CNV false discovery rate of 1 . 7% ( 26/1561 ) and a sensitivity of 75% ( 1535/2061 ) for CopySeq ( Materials and Methods ) , under the conservative assumption that the microarray-based calls [14] contain no false positives as well as no false negatives . Furthermore , under the same conservative assumption we estimated positive predictive values ( or PPV ) of 97% ( 683/704 ) for deletions and 99% ( 852/857 ) for duplications in the OR loci . We also compared our OR copy-number genotypes with genotypes inferred with Agilent CGH arrays [6] which ascertained ∼6% ( 51 ) of the OR loci , enabling us to compare >7 , 000 genotypes in the 149 overlapping samples . This comparison also revealed highly significant , albeit slightly weaker correlations ( Pearson correlation 0 . 73; P<2 . 2e-16; Figure S11 ) – similar to the results we obtained for chromosome 1 CNVs . Third , we used qPCR to obtain independent validation results for our copy-number genotypes in 10 individuals across five loci , by assessing 50 copy-number genotypes through experimental validation ( Materials and Methods ) . Three of the five loci were randomly picked and two specifically selected since they displayed particular wide ranges in copy-number genotype ( i . e . , up to ‘9’ copies ) . The loci further included regions that can reasonably be regarded as particularly hard to ascertain: i . e . , >90% of the nucleotides overlapped with SDs in four loci; four loci displayed multi-allelic CNVs; and ∼50% ( 24/50 ) of the assessed copy-number genotypes corresponded to duplications . Nonetheless , measured correlations with CopySeq's copy-number genotypes were excellent for the qPCR measurements ( Pearson correlation = 0 . 96 , P<2 . 2e-16; Table S13 ) . Among these , we found that correlations were of high magnitude for CopySeq's genotypes including the extreme copy-number of ‘9’ ( Figure 6B ) , which suggests that CopySeq enables generating copy-number genotypes accurately over a wide range of locus copy-numbers . Furthermore , we picked the four loci displaying the strongest discrepancies between CopySeq and microarray studies [6] , [14] ( i . e . , genotype concordance <30% ) for further qPCR validation in eight individuals each; three of these four loci intersected with recently duplicated SDs . The qPCR results were consistent with CopySeq in three out of the four assessed loci ( Figure S14 , Table S14 , and Text S1 ) , suggesting that NGS-based genotypes are at least similarly accurate , and may possibly be more accurate , than array-based genotypes in such regions . Having established the accuracy of CopySeq in OR loci we next performed a global analysis of our OR copy-number map . First , we related deletions and duplications to previously published CNVs , i . e . , to CNVs reported in the DGV ( version from December 2009 ) and in a recent microarray-based analysis of OR loci [21] currently not included in DGV . We found that 199 of the identified 307 copy-number variable autosomal OR loci overlap with already published CNVs . 50 out of 52 commonly variable OR loci ( i . e . , such with reference allele frequency <95%; Table S3; see Text S1 ) had previously been reported . The remaining 108 OR loci were previously not reported to vary in copy-number; this included 99/108 ( 92% ) rare CNVs ( allele frequency <1 . 0% ) . Obviously , future surveys examining larger numbers of individuals are likely to report further rare CNVs affecting ORs . Our genotype frequency analysis further revealed several instances in which the majority of individuals displayed a non-reference allele . Given the importance of establishing a common and comprehensive OR repertoire for functional studies we analyzed these cases in detail , first by calculating CNV allele frequencies in all bi-allelic loci . This analysis suggested that in eight OR loci ( including , for instance , OR2BH1P ) the reference allele represented a minor allele ( i . e . , reference allele frequency <50% ) ; two of these loci involved genes ( see below ) . Furthermore , we estimated reference allele frequencies in multi-allelic loci by assuming the presence of the homozygous reference allele if a locus copy-number of ‘2’ was inferred ( see Text S1 ) . This analysis revealed that in one multi-allelic OR pseudogene locus ( OR11J2P ) the reference sequence appears to represent a minor frequency allele ( Table S3 ) . Moreover , we estimated confidence intervals ( 95% ) for reference allele frequencies and identified five additional loci ( e . g . , OR4A45P ) that are situated in transition between minor and major alleles , i . e . , with an alternative allele frequency close to 50% ( see Text S1 ) . We next analyzed in further detail OR loci that displayed unusual ( i . e . , non-reference ) copy-number genotypes in the vast majority of samples . In particular , our results indicated that OR4C3 and OR4C5 genes as well as the OR4C4P pseudogene ( all located in one genomic interval on chromosome 11 ) are duplicated in most ( >95% ) individuals ( Table S3 ) . We mined an alternative assembly of the human genome and found a close duplication ( 95% identity ) of this genomic interval at another , distinct location on chromosome 11 , suggesting that the reference genome version at the original interval may either be based on a rare deletion in the region ( see Text S1 ) , or may potentially represent a mis-assembly of the reference ( as recently discussed in Young et al . , 2008 ) . Whichever the case might be , absence of the common allele sequence from the reference genome and the high sequence identity between the duplicated segments , resulted in mapping of all reads originating from both loci onto one locus . Notably , we also identified a segment on chromosome 12 which contained three OR pseudogenes ( OR7E140P , OR7E148P , OR7E149P ) that were homozygously deleted in all European and Asian individuals , but were present in the Yoruba individuals with ∼36% allele frequency . The absence of orthologs of these pseudogenes in chimpanzee and orangutan suggests that they likely represent a recent human-specific insertion . In addition , we assessed whether common , rather than rare CNVs are responsible for the majority of measured inter-individual OR copy-number differences . We thus ranked copy-number variable OR loci by the frequency at which they displayed an alternative ( CNV ) allele and recomputed the inter-individual OR copy-number differences . We found that a small number of relatively common variants affecting OR loci were responsible for most of the ascertained variation: i . e . , the ∼15% most commonly variable loci captured approximately ∼80% of the inter-individual copy-number differences , and the ∼50% most common loci captured ∼95% of the differences ( Figure 4A ) . Thus , common CNVs lead to most inter-individual differences in OR copy-number and thus may have a relatively strong impact on variation in smell perception in humans; these common variants thus represent attractive candidate regions for future association studies . We mostly analyzed CNVs in an OR locus-by-locus basis . It is evident from Figure 3E , however , that CNVs present at high frequency ( i . e . , OR loci that frequently display CNGs other than ‘0’ ) tend to cluster , suggesting that they may form CNV hotspots or correspond to large CNVs spanning multiple loci [21] , [22] . We thus assessed consecutive CNV calls in annotated genomic OR clusters , assuming that adjacent OR loci that are both involved in a duplication or are both involved in a deletion , respectively , may potentially be explained by a single large CNV . This analysis revealed that ∼36% of CNV events involve single-OR-locus CNVs , whereas the remaining ( potentially large ) CNVs may span at least two adjacent OR loci ( Table S4 ) . Deletions are particularly likely to have an impact on smell perception , as they may abrogate OR function . In our set , we found that 14 . 5% ( 56/387 ) of the OR genes harbored at least one deletion allele , and in 5 . 9% ( 23/387 ) of the OR loci , deletions were observed with an allele frequency >1 . 0% . Homozygous deletions are of particular interest due to their potential phenotypic effects . We found that these are widespread with 25% of the analyzed individuals displaying at least one homozygous OR gene deletion and some individuals displaying up to four such ‘holes’ in their functional OR content . To obtain an inclusive list of alleles responsible for holes in the human OR repertoire we also mined the 150 individuals for SNPs associated with OR gene inactivation ( i . e . , those causing segregating pseudogenes [29] ) and identified 24 previously known and 49 novel SNPs resulting in altered OR gene start and stop codons ( Materials and Methods and Table S12 ) . The list of inactivating genetic variants , including locus deletions and segregating pseudogenes , covers ∼15% and ∼20% of the OR gene repertoire , respectively . These genetic variants represent excellent candidates for future association studies on olfaction . We next used CopySeq to obtain insights into how region-specific genomic biases may have shaped the genomic distribution of CNVs affecting ORs . First , we compared the relative CNV abundance between OR pseudogenes and OR genes . In this regard , both random drift [20] and selective constraints [21] have been implicated in influencing the distribution of CNVs in OR genes and pseudogenes . Our analysis of copy-number genotypes indicated that pseudogenes generally show more variance in locus copy-number than genes ( Figure 4B ) . Selective constraints can be assessed by examining deletions , i . e . , the removal of functional genes , as deletions are more often deleterious than duplications and thus are more biased away from functionally relevant genomic regions ( [1] and references therein ) . We found that on average less genes ( 3 . 8 per individual , i . e . , 1 . 0% of the OR gene set ) than pseudogenes ( 9 . 5 per individual , i . e . , 2 . 0% ) were deleted per individual , a trend that was significant ( 2-fold relative depletion; P = 0 . 009 based on a permutation test; see Text S1 ) . Furthermore , 0 . 4 genes ( 0 . 1% of the OR gene set ) and 3 . 9 pseudogenes ( 0 . 8% of the set ) were homozygously deleted in each individual ( 8-fold relative depletion; P = 0 . 004 , permutation test ) . Both trends persisted , but lost significance when excluding the 7E subfamily , a rapidly evolving OR subfamily with 85 members [30] ( see Text S1 ) . Thus , while selective constraints acting on OR genes may in part explain the distribution of CNVs in OR genes and pseudogenes , these constraints are not extensive for the OR family , and formational biases presumably contributed to the observed genomic distribution of CNVs affecting ORs ( for example , we note that the proportion of multi-allelic loci among CNV loci is similar between OR gene and OR pseudogene loci , i . e . , in each case about 5% ) . In contrast to previous surveys , CopySeq enabled us to comprehensively assess whether multi-allelic CNVs and bi-allelic CNVs affect OR loci in different sequence contexts . In particular , we assessed to what extent bi-allelic and multi-allelic CNVs occur in SDs . The enrichment of copy-variable ORs in SDs ( 90/307 ) was significantly higher compared to non-variable ORs ( 86/464 ) ( ∼1 . 6-fold; P<1e-4 , permutation test ) . Our results revealed that particularly multi-allelic OR loci were strongly enriched in SDs ( 3 . 5-fold over non-variable OR loci; P = 0 , chi-square = 44 . 9 , chi-square test; Figure S12 ) . Furthermore , we observed a 4 . 3-fold enrichment for loci displaying high copy-number genotypes ( ‘5’ and more copies; Figure S15 ) and a 2 . 7-fold enrichment for loci displaying both deletions and duplications ( Figure S12 ) . This association is possibly due to the predisposition of regions rich in SDs to show recurrent CNV formation by non-allelic homologous recombination [15] . In addition , CopySeq enabled us to dissect the contribution of evolutionarily young and more ancient ORs to copy-number variation in OR loci . It was reported that young ORs ( some of which correspond to SDs ) are particularly prone to be affected by CNVs , with young loci defined both based on the presence of paralogs sharing high sequence identity and based on the lack of one-to-one orthologs in the chimpanzee [21] . Our analysis revealed that young ORs affected by CNVs mainly lie in multi-allelic loci . As shown in Figure 7 multi-allelic loci displayed a significant enrichment for ORs with high sequence identity paralogs compared to both bi-allelic loci and non-variable loci ( i . e . , the average sequence identity to the closest paralog was 84 . 5% in the multi-allelic loci and ≤73% in both the non-variable or bi-allelic loci , respectively; the differences are significant with P<0 . 0001; t-test ) . Furthermore , multi-allelic ORs displayed a >2-fold enrichment for ORs lacking a one-to-one ortholog in chimp compared to each other group ( the differences were significant with P<0 . 0001; Chi-square test ) . Possible explanations for the differences include selective constraints and formational biases , both of which likely vary among different genomic regions . We next assessed whether CNVs affecting OR loci display differences among individuals from diverse ancestries . Even when excluding rare CNVs that were identified only once amongst all individuals we observed 19 CNVs only in the analyzed Africans , 18 only in the Asians , and 10 only in the Europeans ( Figure 8A ) . Furthermore , we carried out principal component analysis ( PCA ) of the copy-number genotypes generated across all 265 bi-allelic autosomal CNV loci . The PCA yielded a visible separation of the African group from the combined group of Europeans and Asians by the first two principal components ( Figure S13A ) , and a separation of all three ethnic groups when analyzing the second and third component ( Figure S13B ) . Note that the better distinction of Africans from European and Asian groups is in line with the well documented bottleneck effect , as evident from multiple large scale SNP studies [27] . When examining the failure of the first component to separate the three ethnic groups we identified a common bi-allelic deletion spanning three OR genes ( OR4C11 , OR4P4 , OR4S2 ) and two OR pseudogenes , which drove the separation into three visible clusters by the first component ( Figure S13 ) . The three clusters represent the average OR locus copy-number genotype , with the left cluster representing the homozygous reference allele , the central cluster the heterozygous deletion , and the right cluster the homozygous deletion , respectively . The PCA and further analysis showed that all African individuals analyzed in our survey have at least one copy of the allele , whereas 7–10% of Europeans and Asians have all three functional OR genes homozygously deleted . In this regard , for example , a deletion , which encompassed the OR52E8 gene , was observed with appreciable allele-frequency ( 18% ) in the Africans , whereas the allele was not observed in the other populations . Overall our PCA analysis of bi-allelic loci reflects findings from previously published SNP results [27] , even though we used only 265 bi-allelic loci in our PCA analysis as opposed to hundreds of thousands of SNPs . Lastly , CopySeq enabled us to examine population differences in the distribution of bi-allelic and multi-allelic OR loci: indeed , in at least 11 OR loci CNVs were observed as multi-allelic in one population and bi-allelic in another ( see Text S1 ) . We reasoned that the spatial resolution of NGS data may enable us to further dissect complex multi-allelic OR loci , i . e . , loci in which different CNV alleles coincide in the same genomic segment . Dissecting multi-allelic loci represents a crucial step to inform future association studies that examine the functional impact of each CNV allele separately . As a proof-of-principle we applied CNV-boundary redefinition to analyze a genomic interval containing the adjacent genes OR51A4 and OR51A2 , which in some individuals form a fusion gene [4] ( Figure 3BCD ) . In particular , since the sequenced breakpoints [4] of the deletion leading to the gene fusion fall into the respective OR coding regions , we inferred copy-number genotypes with CNV-boundary redefinition based on breakpoint-junction analysis ( Figure 3BCD ) . Our analysis with CopySeq revealed that while the deletion is a variant with ∼32% allele frequency , an additional duplication comprising only the OR51A2 gene is also frequently present , i . e . , was genotyped in 6 individuals ( Table S4 ) . Thus , applying CopySeq with CNV-boundary redefinition can help facilitate the dissection of multi-allelic CNV loci .
We have developed a computational approach , CopySeq , that discovers copy-number variable loci and subsequently assesses their locus copy-number in NGS data , using a rationale based on formal hypothesis testing . As such , CopySeq may facilitate analyzing CNVs in NGS-based genome–wide association studies . Our analyses revealed an excellent concordance of CopySeq with microarray platforms , qPCR experiments , and FISH experiments , suggesting high genotyping accuracy . We note that one possible source for discrepancies between array-CGH and CopySeq in CNVs intersecting with SDs might be the heuristic transformation of microarray intensity data in SDs into genotypes by population-wise clustering [6] . CopySeq does not apply population-wise clustering nor do its calls depend on comparing read-depths with reference samples for normalization . This makes CopySeq particularly suitable for genotyping CNVs in single individuals or for genotyping rare alleles ( i . e . , cases where too few data exist for population-wise clustering ) . While arrays are presently widely applied for CNV analysis [6] , [14] , [31] we foresee that in the near future with the completion of the 1000GP and other large-scale NGS projects there will be more genomes sequenced than such for which comprehensive array-based genotyping data will be available . Consequently , we anticipate that in the future , NGS-based genotyping of CNVs is likely to be widely applied . NGS data are generated in a genome-wide fashion and sequencing data can be re-interpreted without requiring experimental re-design to enable accurate copy-number genotyping , once new high-confidence CNV sets are becoming available ( e . g . , following the assembly of new sequence insertions [32] ) . We expect that in the future , CopySeq will be applied along with CNV discovery approaches such as paired-end mapping [4] to combine the advantages of copy-number genotype ascertainment with accurate CNV discovery and CNV-boundary redefinition . Furthermore , we demonstrated that CopySeq accurately infers copy-number genotypes in SD regions ( regions known to be hard to ascertain for genetic variants ) , i . e . , in the OR gene family that is rich in highly identical paralogs . Our analysis of 150 individuals afforded the first comprehensive ascertainment of locus copy-numbers in the OR gene family . We found that more than a third of the human reference OR repertoire varies in copy-number across individuals and described many novel CNVs . Our first comprehensive report of copy-number genotypes in these regions provides a valuable resource for the community , since genotypes are an important prerequisite in associating CNVs with odor perception . While previous reports demonstrated an enrichment of variable ORs in SDs [22] , our analysis revealed that multi-allelic and bi-allelic OR loci are differentially affected by SDs . In some cases , we furthermore observed distinct OR gene counts in different populations . We note that while many more individuals need to be genotyped before population-specificity of these variants can be confirmed , allele frequency differences are likely to contribute to population differences in smell-perception [33] . Also , while other studies hypothesized that OR genes and pseudogenes evolved in a neutral fashion by genomic drift [20] , [22] , our data suggest weak evolutionary pressures acting on OR genes ( Figures 4B , S17 ) . We further observed an abundance of OR genes that are dysfunctional in a subset of the individuals analyzed . In this regard , OR deletion alleles and SNPs leading to gene pseudogenization are widespread , i . e . , about 15% and 20% of the functional OR repertoire harbor such variants , respectively . These inactivating variants represent attractive candidates for future association studies focusing on odorant perception [29] , [34] . While CopySeq enables probabilistic copy-number genotyping in NGS data , it still has its limitations . One limitation of CopySeq is that it is confined to sequences already present in the reference genome – a limitation that will likely diminish soon , when more alternative human genome assemblies will become available . Furthermore , only unambiguously mappable sequences are considered by CopySeq . In this regard , ∼1% of the human genome is in very recently segmentally duplicated regions with >99 . 5% identity [35] – a fraction in which most short DNA reads will be non-unique . These regions are presently excluded by CopySeq . However , we reasonable expect that this limitation will diminish soon , as longer and more easily mappable reads ( 150 bp , or longer ) are presently becoming the standard in NGS . In fact , in the upcoming main phase of the 1000GP human genomes will be sequenced mostly with paired-ends , with each end 100–150 bp in size or longer , which will facilitate the application of CopySeq in recently duplicated regions . Also , longer reads will enable the fine-mapping of CNV breakpoints and consequently will enable CNV-boundary redefinition ( e . g . , by BJA ) for a larger fraction of CNVs than is presently possible . Very recently duplicated regions ( >99 . 5% identity ) can already be analyzed with Alkan et al . 's approach , which considers non-unique genomic mapping positions . Nevertheless , in non-SD regions we found that CopySeq displayed higher concordances than Alkan et al . 's approach with Affymetrix array-based locus copy-numbers ( 97 . 2% vs . 80 . 2% ) . Possible reasons for the improved concordance of CopySeq may be an increased accuracy of a statistical copy-number genotyping framework compared to depth-of-coverage analysis without a probabilistic genotyping model . In addition , CopySeq's genomic k-mer filtering scheme may have contributed to its improved concordance by removing read-depth specific noise originating from distant paralogs . Finally , our inference and validation of genetic variants may guide the way to similar analyses for other difficult-to-ascertain CNV regions , such as the medically relevant [1] CCL3L1 , β-defensin , and FCGR loci ( see , for example , our analysis in the Text S1 with regard to the FCGRB locus on chromosome 1 ) . Furthermore , CopySeq can be easily adapted to genome-wide scale analyses . As thousands of human genomes are becoming sequenced in the context of biomedical research studies ( e . g . , cancer genomes or constitutional abnormalities ) , there is a strong need for accurate copy-number genotyping approaches operating on NGS data .
Illumina sequencing data were obtained from the 1000 Genomes Project ( 1000GP; ftp://ftp . 1000genomes . ebi . ac . uk/; July 2009 release ) . Those reads have been aligned against the reference genome ( hg18; Build 36 . 1 ) with the MAQ [24] aligner ( default parameters ) . The DNA reads were mostly sequenced as paired-end fragments with a read length of 36 nt . For each sample we recorded the total coverage of uniquely mapped reads ( ‘ends’ ) , and kept unambiguous read-alignments onto the following regions: sets of previously defined CNVs on chromosome 1; a set of genomic regions comprising ∼1% of the reference genome that were analyzed to correct for the G+C content in a sample-specific manner ( see below ) ; a set of 5 Mb genomic intervals for variance model parameter estimation ( see below ) ; and all human OR loci ( see below ) . The identification of unambiguous ( unique ) read alignments benefitted from the MAQ feature to infer unambiguous alignments even if only one end of a paired read aligns uniquely to the genome , by combining information from the mapped end and the paired-end insert size distribution [24] . Instances of duplicated fragments ( i . e . , PCR artifacts of the NGS library ) were removed during the read mapping process ( using the rmdup function of the MAQ toolkit ) . To assess the performance of CopySeq we obtained 100 CNV loci <50 kb from chromosome 1 for which copy-number genotype measurements based on microarrays were available [14] . Out of these one CNV locus was excluded due to low mappability with 36-mers ( i . e . , less than 500 mappable 36-mer subsequences within the CNV locus ) . CNV sizes in the resulting set of 99 CNVs range from 1–49 kb with a median CNV size of 6 . 9 kb ( mean = 11 kb ) . The G+C correction step of CopySeq required the analysis of regions that ideally should be invariable with regard to locus copy-number . Therefore , we sampled 30 Mb in 10 kb bins ( i . e . , ∼1% of the human reference genome ) and excluded regions annotated as copy-number variable in the Database of Genomic Variants ( DGV ) . We randomly sampled one hundred 50 kb loci from all autosomes that are invariable in copy-number as assessed by CNV entries in the DGV database ( v9 , March 2010 ) . We further controlled that the number of sampled loci within an isochore family is proportional to the genome-wide amount of DNA in isochore families . To model the dependency of locus size and read-depth ratio variance within a locus class ( e . g . , 1 kb or 5 kb ) , we generated in total 15 datasets by subdividing each 50 kb locus into non-overlapping segments of various length ( i . e . , 1 , 1 . 25 , 1 . 5 , 1 . 75 , 2 , 2 . 25 , 2 . 5 , 3 , 5 , 7 . 5 , 10 , 20 , 30 , and 40 kb ) . We obtained the genomic coordinates of 851 annotated human olfactory receptor ( OR ) genes and pseudogenes from the HORDE database ( Build 42; http://genome . weizmann . ac . il/horde/ ) . CopySeq requires at least 500 bp of sequence to which reads can be mapped uniquely . ∼19% of the OR open reading frame ( ORF ) sequence display less than 500 bp of mappable sequence , explaining the necessity to extend OR loci by flanking sequences . OR loci included the ∼1 kb intron-less coding region as well as non-coding segments up- and downstream , i . e . , 100 bp of 5′-sequence and 2 kb of 3′-sequence . We regarded such ‘extension’ of the loci of interest to 3 kb as reasonable , since previously described [4] , [21] CNVs affecting ORs were several kb in size . Indeed , mining CNVs by long insert size paired-end mapping [4] in an individual studied by the 1000GP ( NA12878 ) confirmed that only very few CNVs ( i . e . , three CNVs in NA12878 including the known OR51A2—OR51A4 fusion in Figure 3 ) harbor breakpoints in the OR territories . In the few cases where CNV breakpoints do fall into these territories , we recommend application of CopySeq with CNV-boundary redefinition . Throughout the manuscript the phrase “intact olfactory receptor ORFs” is used synonymously with “OR genes” , “genes” , or “OR gene repertoire” , whereas “disrupted olfactory receptor ORFs” are used synonymously with “OR pseudogenes” or “pseudogenes” . We used Rozowsky et al . 's approach to generate mappability maps of the human reference genome using k-mer lengths of 36 , 51 , and 76nt , respectively [36] . The mappability maps contain information about the frequency of each genomic k-mer sub-sequence , i . e . , how many times the k-mer occurs exactly on the Watson and Crick strand in the reference genome . The 36 , 51 , and 76 k-mers account for the three different Illumina read length sizes that were used in the 1000GP . CopySeq infers copy-number genotypes by assessing reads aligned against the mappable part of the genome , defined as k-mers subsequences that result in a genome-wide k-mer frequency of one ( i . e . , k-mer sequences that remapped against the reference exactly once ) . Before inferring copy-number genotypes , CopySeq measures the locus read-depth ratio for each predefined locus in question . The observed locus read-depth is defined as the sum of reads from a sample that unambiguously map within the boundaries of the predefined locus . The locus read-depth ratio is defined as the ratio between the observed locus read depth D and the expected locus read-depth E , an estimate generated by evaluating the locus-specific G+C-content ( see below ) , the mappability map , and the genome-wide sequencing coverage . In particular , for a predefined locus i , in individual j , and using the mappability map k ( i . e . , k-mer size k ) the expected locus read-depth Eijk for invariant locus i is estimated with , where uik is the number of k-mers within locus i that are unique in the genome , Nj is the number of uniquely aligned sequence reads against the reference genome in individual j , and G is the size of the genome ( 2 , 858 , 018 , 193 nucleotides in hg18 , excluding the mitochondrial DNA ) . In order to infer copy-number genotypes ( copy-number genotypes ) for locus i the read-depth ratio was calculated with , where Dijk is the observed locus read-depth ( i . e . , number of reads mapped onto unique k-mer positions in the locus ) and Eijk is the expected locus read-depth . Thus , for example , a ‘normal’ copy-number genotype ( e . g . , copy-number genotype = ‘2’ in autosomal DNA , which may be considered as the ‘baseline’ for locus copy-number measurements ) will result in a read-depth ratio , a copy-number genotype of ‘1’ ( heterozygous deletion ) will result in , and a copy-number genotype of ‘3’ ( heterozygous duplication ) in ( see Figure S19 ) . Earlier reports observed a correlation between Illumina sequence read coverage and G+C-content [9] . To correct for this confounding factor for read-depth analysis , CopySeq makes use of a set of 3 , 000 normalization loci ( see above ) to construct a G+C-normalization curve separately for each individual . This normalization curve is used for each locus to adjust its locus read-depth ratio according to the G+C content . Outlier loci ( representing for example de novo CNVs ) were identified by conservative criteria ( i . e . , 1st-Quartile ( read-depth ratio ) -3*IQR ( read-depth ratio ) and 3rd-Quartile ( read-depth ratio ) +3*IQR ( read-depth ratio ) ) . For each normalization locus the G+C content of its mappable nucleotides was calculated and the sample-specific relationship between locus G+C content and locus read-depth ratio assessed . A sample-specific cubic smoothing spline function ( using the R function smooth . spline ) was fitted into the distribution of zero-centered data in order to model the underlying trend in the data by controlling for the smoothing parameter via the generalized cross-validation criteria . The spline fit was later on used as a normalization function to predict the read-depth ratio accounting for the locus G+C-content . The resulting fit explained about 67% of the Poisson variance over-dispersion ( Figure S5 ) ( i . e . , the n-fold variance as compared to the theoretical variance expected from random sequencing ) and accounts for the observed reduced locus read-depth ratio at both sides of the G+C-distribution [25] ( see Figure S2 ) . Using the obtained fit we calculated the expected RDR that is solely explained by the locus G+C content with the R function predict . smooth . spline and corrected the raw read-depth ratio with ; as above , refers to normal locus copy-number , duplications result in greater than 1 and deletions in smaller than 1 . Of note , after G+C-content normalization we observed a strong relation between sequencing coverage and read-depth variance ( i . e . , decrease of variance with increase in coverage ) that was not evident before normalization ( see Figure S4 ) . Based on the normalized read-depth ratio variance of the normalization loci and a cutoff value ( 0 . 01 ) we excluded 20 out of 170 initially assessed individuals from our analysis that were sequenced at low-coverage ( <1× ) ( Figure S6 ) . CopySeq initiates the copy-number genotyping step with a CNV identification module , in which two distinct hypotheses are tested at each locus: the null hypothesis H0: ( i . e . , ‘normal’ locus copy-number ) and the alternative hypothesis H1: ( i . e . , presence of a CNV ) . As values of follow approximately a normal distribution in locus-copy number invariant normalization loci ( Figure S3 ) , we reasonably applied the z-statistic for assessing whether a given locus read-depth ratio deviates from the null hypothesis H0 ( i . e . , whether the read-depth ratio is unexpectedly high , or low , indicating a CNV ) . The alternative hypothesis H1 is that is drawn from a different distribution ( duplication or deletion ) . Thus , the z-statistic for locus i in individual j was obtained with , where is the expected read-depth ratio given a normal ( ‘2’ ) locus-copy number , i . e . , , and is the read-depth ratio standard deviation defined as , with as the locus-length dependent read-depth ratio variance scaled by factor alpha ( explained below ) assuming an invariant locus and an additive read-depth ratio variance component representing additional experimental background noise ( see below ) . Each z-score zijk was transformed into a two-sided p-value pijk using the cumulative distribution function of the standard normal distribution . P-values were corrected for multiple testing , yielding Q-values , using Benjamini & Hochberg FDR correction . A global Q-value threshold was empirically determined in the following way . We estimated the CNV recall rate by calling deletion genotypes ( i . e . , copy-number genotype = ‘1’ ) for 11 OR loci on chromosome X in 57 male samples , assuming that they display no CNVs ( Figure S7 ) . Using this setup , a Q-value threshold of 5% resulted in an inferred CNV recall rate of 96 . 5% for ∼3 kb loci , i . e . , in 605 out of 627 cases CopySeq predicted a deletion , as expected in the male samples . We furthermore estimated the CNV false discovery using the female samples , by conservatively assuming that all OR loci on chromosome X display a normal locus copy-number of ‘2’ . Using this setup , in none of 1 , 023 cases CopySeq predicted ‘CNVs’ , suggesting a very low CNV false-discovery rate . To generate genotype calls , CopySeq models copy-number genotypes as probability distributions accounting for locus-length , copy-number genotype , and sample-dependent noise . Specifically , copy-number genotype probability distributions are modeled as Gaussian distributions that incorporate both a Poisson variance term that depends on the read-depth of a locus , as well as an additional global ( background ) variance term that does not depend on locus length and copy-number genotype . The mean of each Gaussian is set according to expected values of the locus read-depth ratio , where the expected value for the locus read-depth ratio is , with m being a specific copy-number genotype among c possible copy-number genotypes C0…Cc . For example , for m = 2 ( no CNV ) μ2 = 1; for m = 1 ( heterozygous deletion ) μ1 = 0 . 5; and for m = 3 ( heterozygous duplication ) μ3 = 1 . 5 . These theoretical means are in excellent agreement with experimental data ( see Figure S19 ) . The probability density function for copy-number genotype m is calculated as:where is the Normal probability density function , is our read-depth ratio variance estimate , is the scaled locus-length- and copy-number genotype-dependent variance term , and is an additive global background noise variance term ( explained below ) . We calculated the locus-length- and copy-number genotype-dependent read-depth ratio variance for copy-number genotype m as , where is the theoretical Poisson read-depth variance that increases linearly with copy-number genotype m , is the expected locus read-depth given an invariant locus copy-number of ‘2’ , and is the read-depth for a locus copy number ‘1’ . The expression is also known as the squared coefficient of variation and can be viewed as the scaled variance of the Poisson distribution . The a priori knowledge about the model and parameters is summarized in the background information I . The most plausible copy-number genotype is inferred using classical Bayes' theorem:The prior density p ( Cm ) is modeled as a uniform density function with p ( Cm ) = 1/c , where c is the total number of possible copy-number genotype values . We reasonably chose to use a uniform prior as a neutral prior for classification on a genome-wide scale . Although we could foresee alternative approaches for estimating non-uniform priors , such as expectation maximization ( EM ) , this would require extensive training data to reflect the underlying copy-number distribution for genomic locus of interest , including regions with high and low allele frequencies . Using Bayes' theorem , each locus was labeled with a copy-number genotype m that maximizes the posterior probability using the maximum a posteriori ( MAP ) estimate . CopySeq calculates a confidence score ( similar to a logarithm of odds , or LOD , score ) for each locus , expressing the uncertainty of assigning the correct copy-number genotype , withwhere and are estimated probabilities for the most plausible and the second most plausible copy-number genotype , respectively . A LOD score of 2 , e . g . , means that the probability for the most plausible copy-number genotype is 100 ( i . e . , 102 ) times higher than for the next plausible copy-number genotype . We observe that the theoretical minimal variance predicted by a Poisson sampling model is insufficient to explain the observed locus read-depth variance in loci of various length ( see Figure S3D ) . We thus consider a variance model that approximates the observed locus length-dependent and length-independent variance in order to account for the over-dispersion . We assume that the read-depth ratios behave independently and are drawn as random samples from a normal distribution for loci with the same size . Normal quantile-quantile ( Q-Q ) plots of loci with different sizes ( e . g . , 1 kb , 5 kb , and 10 kb ) support the assumption of normality for read-depth ratios ( see Figure S3ABC ) . We observe that the total read-depth ratio variance for loci of a specific length class follows a linear combination of a global ( non-locus dependent ) background variance term and a scaled length-dependent Poisson variance term , i . e . , ( see Figure S3E ) . The model parameters alpha and background variance were estimated via linear regression with the variance model dataset as described above . The average background variance among the 150 samples is approximately similar across samples ( ∼0 . 002 ) and the scaling factor alpha ranges between ∼1–4 . CopySeq has the ability to use paired-end mapping [4] ( PEM ) and the recently published breakpoint-junction data [26] ( BJA ) , i . e . , to redefine the boundaries ( i . e . , breakpoints ) of CNVs for copy-number genotyping on a subset of CNV loci ( i . e . , such for which PEM or BJA information are available ) . Thereby , CopySeq defines CNV-boundaries as previously described in publications on PEM and BJA . We assessed the utility of integrating PEM and BJA into CopySeq to enable CNV boundary-redefinition . Specifically , we focused on the chromosome 1 CNV set ( see above ) . In particular , in the case of PEM we fine-mapped the breakpoints of CNVs using high-confidence ( long-read and high-coverage ) paired-end reads from the sample NA18505 , which was sequenced at >8 . 5× physical coverage with 3 kb insert size paired-end reads generated with the 454 sequencing technology; the long reads generated by 454 sequencing allow for high-confidence placement of paired-ends [4] . Furthermore , the paired-end library insert size was ∼3 kb , a reasonable insert size for analyzing OR loci , which may recombine by non-allelic homologous recombination involving ∼900 bp ORFs [4] , [21] . To initiate the CNV-boundary redefinition for CNV-loci of interest , we required a reciprocal overlap of >51% between microarray-based CNV coordinates of the chromosome 1 test set and the CNV-breakpoints as inferred by PEM . BJA was carried out using a library of ∼1 , 800 SVs with sequenced breakpoints [26] , and applying the same overlap criteria ( >51% ) as for the PEM data . When applying PEM or BJA we reasonably assumed , based on previously published observations [4] , [26] , that in bi-allelic CNV loci breakpoints are identical across analyzed individuals; thus , CopySeq was able to use redefined CNV-boundaries in all individuals when assessing bi-allelic CNVs ( note that BJA and PEM cannot be applied in the case of multi-allelic CNV loci , as CNV boundaries may differ in recurrent CNV formation events ) . To assess the concordance of CopySeq-based copy-number genotypes with array-based copy-number genotypes we obtained data from two previous array-based surveys [6] , [14] . The sample overlap between our study and the array-based studies is high with 118 ( McCarroll et al . ) and 149 ( Conrad et al . ) , respectively . Genotyping concordance is defined as the number of copy-number genotypes that display exactly identical values ( e . g . , copy-number genotype = ‘4’ ) between two studies , divided by the total number of copy-number genotypes that have been inferred . Each locus was tested in all individuals where data was available . In a small number of cases ( see Tables S4 and S5 , and data submitted with the array-based studies [6] , [14] ) no copy-number genotypes were inferred at a given confidence score threshold ( ‘NA’ ) ; these specific tests were obviously not considered when estimating genotyping concordance . When estimating genotyping concordance for OR loci , corresponding array-based CNVs [6] , [14] were considered if they fully spanned the respective OR locus ( i . e . , we required the OR locus to be fully contained in the respective CNV region previously assessed with arrays ) . We used standard terminology for statistical measures such as sensitivity , specificity , and precision rate ( or positive predictive value , PPV; see e . g . , Table S6 ) . Real-time quantitative PCR ( qPCR ) was carried out as described in [21] . Each experiment was carried out with 40 cycles and ended with a melting curve step to verify product specificity . Reactions with more than one peak in the melting curves were removed from further analysis . To address experimental variability due to primer differences and fluctuation in DNA concentration , we used a copy number-invariable gene ( DSCR1 ) for normalization [21] and repeated each reaction at four DNA concentrations of 1 , 2 , 4 , and 8 ng , in duplicates; the four concentrations allowed us to estimate the primer efficiency . Importantly , qPCR does not provide a measure of an absolute amount of material , but rather a comparison between samples . Thus , when analyzing a qPCR experiment we used copy-number genotypes inferred by CopySeq as an anchor point for calculations . Correlations were obtained from measuring copy-number differences across individuals at a given locus . Subsequent analyses were carried out as follows: ( 1 ) we calculated a matrix of pairwise Ct differences between all sample pairs for each OR and for DSCR1 , at every DNA concentration separately . ( 2 ) To account for fluctuations in Ct arising from experimental procedures ( e . g . , fluctuation in DNA concentrations or pipetting ) , we subtracted the DSCR1 based difference matrix from the OR difference matrix . We refer to the resulting matrix as the corrected difference matrix . ( 3 ) We combined data from all four concentrations by averaging the true pairwise Ct difference for each sample pair at every locus . ( 4 ) For each sample and locus , we used the predicted values from the other 9 samples as anchor , and thus computed 9 estimates for each genotype . For the transformation we used the experimentally estimated primer efficiency values . We then averaged the 9 values for each sample at each locus before calculating correlations . Primer sequences are listed in Tables S14 , S20 . We followed the segregation pattern of OR copy-numbers in the parent offspring trio of European ancestry ( NA12878 , NA12891 , and NA12892 ) , by classifying 797 autosomal OR loci into 772 mono- and bi-allelic CNVs according to copy-number genotypes in CEU samples only . These 772 regions were analyzed to assess the segregation of copy-number genotype assignments . Coordinates for segmental duplications ( SDs ) were obtained from the UCSC genome browser ( ‘Segmental Dups’ track; hg18 ) . Autosomal OR loci were classified as overlapping a SD if ≥51% of the ∼3 kb locus sequence overlapped . A permutation test was used to assess the significance of enrichment of certain copy-number genotypes among genes and pseudogenes . 1 , 000 permutations of possible values of a test statistic under random rearrangements of the gene and pseudogene labels were calculated to construct an exact ( null ) distribution . The test statistic was defined as the difference of the average number of predefined copy-number genotypes per group with two groups A and B , i . e . , , whereby group A is the collection of genes and group B is the collection of pseudogenes . The difference between the group means without permutation was calculated and referred to the observed value of the test statistic t . The permutation test was designed to determine whether the observed difference between the group means is large enough to reject the null hypothesis that the two groups have an identical probability distribution . To assess significance , a two-sided p-value of the test was calculated as the proportion of sampled permutations where the absolute difference of T was greater than or equal to the absolute value of t . The null hypothesis was rejected at a significance level C = 0 . 05 . We applied principal component analysis ( PCA ) on 150 individuals and on copy-number genotype data of 265 bi-allelic OR loci as implemented in the R function prcomp ( www . r-project . org ) . We extracted SNP data from the April 2009 SNP release of the 1000GP; a release encompassing SNP calls from 59 CHB+JPT , 56 YRI , and 56 CEU individuals , respectively . We identified all SNPs intersecting with OR coding regions and assessed their predicted effect on coding sequence using Perl scripts . Frequency and quality information relating to SNP data are available at the 1000GP website ( http://www . 1000genomes . org ) . CopySeq is implemented in Java and utilizes the SAM-SDK ( http://picard . sourceforge . net/ ) for fast sequence alignment access . It can be obtained from http://embl . de/~korbel/copyseq/ . CopySeq computes a typical genome-wide CNV dataset ( with up to 30 , 000 CNV loci ) in ∼1h . | Human individual genome sequencing has recently become affordable , enabling highly detailed genetic sequence comparisons . While the identification and genotyping of single-nucleotide polymorphisms has already been successfully established for different sequencing platforms , the detection , quantification and genotyping of large-scale copy-number variants ( CNVs ) , i . e . , losses or gains of long genomic segments , has remained challenging . We present a computational approach that enables detecting CNVs in sequencing data and accurately identifies the actual copy-number at which DNA segments of interest occur in an individual genome . This approach enabled us to obtain novel insights into the largest human gene family – the olfactory receptors ( ORs ) – involved in smell perception . While previous studies reported an abundance of CNVs in ORs , our approach enabled us to globally identify absolute differences in OR gene counts that exist between humans . While several OR genes have very high gene counts , other ORs are found only once or are missing entirely in some individuals . The latter have a particularly high probability of influencing individual differences in the perception of smell , a question that future experimental efforts can now address . Furthermore , we observed differences in OR gene counts between populations , pointing at ORs that might contribute to population-specific differences in smell . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/genomics",
"computational",
"biology/genomics"
]
| 2010 | Systematic Inference of Copy-Number Genotypes from Personal Genome Sequencing Data Reveals Extensive Olfactory Receptor Gene Content Diversity |
Identification of the selective forces contributing to the origin and maintenance of sex is a fundamental problem in biology . The Fisher–Muller model proposes that sex is advantageous because it allows beneficial mutations that arise in different lineages to recombine , thereby reducing clonal interference and speeding adaptation . I used the F plasmid to mediate recombination in the bacterium Escherichia coli and measured its effect on adaptation at high and low mutation rates . Recombination increased the rate of adaptation ∼3-fold more in the high mutation rate treatment , where beneficial mutations had to compete for fixation . Sequencing of candidate loci revealed the presence of a beneficial mutation in six high mutation rate lines . In the absence of recombination , this mutation took longer to fix and , over the course of its substitution , conferred a reduced competitive advantage , indicating interference between competing beneficial mutations . Together , these results provide experimental support for the Fisher–Muller model and demonstrate that plasmid-mediated gene transfer can accelerate bacterial adaptation .
Understanding the factors that contribute to the origin and maintenance of sex is an important unanswered problem in evolutionary biology [1–4] . A large body of research has developed several theories predicting potential advantages to sex ( reviewed in [3] ) . However , critical support for these theories remains elusive , in large part because of the difficulty of devising suitable tests using traditional comparative approaches . New experimental approaches have begun to address this problem by allowing experiments to be designed to determine the effect of sex in defined environments [5–12] . The Fisher–Muller ( FM ) model proposes that the advantage of sex results from recombining competing beneficial mutations into one lineage [13 , 14] . In the absence of recombination , beneficial mutations that occur in the same population , but in different lineages , must compete with one another for fixation . This competition , known as clonal interference , slows the spread of each mutation and can reduce the overall rate of fitness increase [15–18] . Subsequent theoretical analysis of the FM model predicts that recombination can increase the rate of adaptation over a range of population sizes and recombination and mutation rates [19–21] . Despite extensive theoretical support for the FM model , the ability of the model to predict the effect of recombination in biological systems is subject to at least two caveats . First , an advantage of recombination depends on competition between beneficial mutations arising in different lineages being strong enough that the fixation of the higher fitness mutation is appreciably slowed . If that is not the case , competition will not effectively limit the rate of adaptation , and recombination will have little effect [22] . Results of experimental studies have been consistent with the presence of widespread competition between beneficial mutations [18 , 23–29] . However , the extent to which this competition slows adaptation has not been well-characterised independently of potentially confounding factors such as population size differences between treatments [17 , 22 , 30] . Second , the extent and type of interactions occurring between beneficial mutations remains largely unknown . Even in the presence of strong competition between mutations , recombination may not provide a substantial advantage if interactions between mutations cause the advantage conferred by a given mutation to depend on a particular genetic background [31] . This might be the case if beneficial mutations tend to occur as part of co-adapted gene complexes . An ideal experimental test of the FM model would compare the effect of recombination on the rate of adaptation in treatments that differ solely in the extent of competition between beneficial mutations . In practice this ideal has been hard to achieve . A number of experimental studies have examined aspects of the relationship between recombination and environment [9–12] . These studies have shown an advantage of recombination; however , the treatment regimes also involved differences in environmental [10 , 12] or life-history [9 , 11] factors , which can cause differences in adaptive opportunities that are independent of the effect of recombination . One study that minimised these differences found a strong interaction between the effect of recombination and mutation supply but required extensive introgression of genes from non-evolved individuals , complicating interpretation of the mechanism underlying this interaction [11] . Most importantly , none of these studies were able to follow , or even identify , any of the mutational changes underlying adaptation . Without this information it is difficult to determine the relative roles of recombination providing a benefit by ( i ) bringing beneficial mutations together versus ( ii ) separating newly arising beneficial mutations from linked deleterious mutations . To circumvent the problems outlined above and isolate the net effect of recombination across mutation-supply treatments , the study presented here takes advantage of results published by de Visser et al . [18] . In that study , conditions were identified in which the absence of a DNA repair gene , mutS , was shown to control the rate of adaptation in evolving populations of the bacterium Escherichia coli by increasing the rate at which new mutations were produced by ∼30-fold . By using the same strains and environmental conditions , I define high and low mutation rate treatments that differ only in the supply of new mutations and thus the degree of competition between beneficial mutations . All other aspects of the environment , including population size , remain constant . To permit recombination I use recombination proficient ( rec+ ) strains made by introducing the F plasmid into the ancestral strains used by de Visser et al . [18] . F mediates recombination by integrating into the bacterial chromosome , where it can facilitate transfer of chromosomal DNA into a recipient cell via conjugation . F is found in many isolates of E . coli and related bacteria , and may play an important role in gene transfer in natural populations [32] . Control recombination deficient ( rec− ) lines were made by deleting a plasmid gene necessary for transfer . To test the FM model I evolved eight replicate lines in each of four treatments: rec+ , high mutation rate; rec+ , low mutation rate; rec− , high mutation rate; and rec− , low mutation rate . All lines were evolved for 1 , 000 generations in the same constant environment . Following this period of evolution the overall adaptation of each line was estimated . In agreement with the FM model , recombination caused a greater increase in fitness in the high mutation rate lines . Subsequent experiments tracked the dynamics of one of the beneficial mutations underlying this adaptation and found that this increase was associated with a reduction in competition between co-occurring beneficial mutations .
Following 1 , 000 generations of evolution , fitness increased significantly in all experimental lines; however , the magnitude of this increase varied across treatments . At the high mutation rate , average fitness increased by ∼43% in rec+ lines , and by ∼32% in rec− lines ( Figure 1 ) . At the low mutation rate , average fitness rose by ∼32% and ∼29% in rec+ and rec− lines , respectively ( Figure 1 ) . A two-way ANOVA found a significant interaction between mutation rate and recombination ( F1 , 28 . 72 = 4 . 7426 , p = 0 . 0378 ) ( Table 1 ) . Therefore , recombination provided a greater advantage at the higher mutation rate . The results presented above are consistent with the FM model , whereby recombination provides an advantage when multiple beneficial mutations compete for fixation by combining them in the same lineage . However , a possible alternative explanation is that recombination was advantageous because it separated beneficial mutations from linked deleterious mutations [10 , 33–37] . Because the evolving populations were large ( Ne ∼ 1 . 66 × 105 ) , were evolved for only 1 , 000 generations , and were started from a single clone , deleterious mutations are extremely unlikely to fix independently of other mutations . However , they can rise in frequency if a beneficial mutation of sufficiently large effect subsequently arises on the same background [34 , 35] . This linkage is more likely to occur in the high mutation rate treatments , where deleterious and beneficial mutations are more common [34] . Thus , this mutation load model also predicts an increased advantage to recombination in high mutation rate populations . A third possible explanation , the mutational deterministic model , is unlikely to explain my results because the genomic mutation rate in the ancestral strains was substantially lower than that required by the model to produce a benefit to recombination . Previous work has calculated a best estimate of the mutation rate in the low mutation rate ancestral strain as 1 . 44 × 10−10 per basepair per generation [38] . Multiplying this rate by the increase in mutation rate caused by the mutS allele and by the genome size gives a genomic mutation rate of only 0 . 023 ( = [1 . 44 × 10−10] × 34 . 9 × [4 . 64 × 106 bp] ) —well short of the genomic mutation rate of ∼1 required by the mutational deterministic model to explain an advantage to recombination . Also , there is no general tendency for deleterious mutations to interact synergistically in this strain , a second prerequisite of the mutational deterministic model [39] . In the light of the alternative mutation load explanation for the observed benefit of recombination , I sought to directly examine the prediction made by the FM model that recombination reduces competition between beneficial mutations . To do this , I compared the dynamics of a focal beneficial mutation as it rose in frequency and ultimately fixed in rec+ and rec− populations . A decrease in the fitness conferred by a focal beneficial mutation , relative to contemporary clones that did not have this mutation , would be consistent with the spread of alternative beneficial mutations in competing lineages . A previous study found a mutation in a regulatory gene , spoT , that contributed to adaptation of E . coli to an environment nearly identical to the one used here [40] . This gene was thus a candidate for harbouring beneficial mutations in the present study . I sequenced spoT in three randomly chosen clones isolated from each of the 16 populations in the high mutation rate treatment and found mutations in four rec+ and two rec− lines . The temporal dynamics of each mutation were examined by screening clones at regular intervals from stored samples of the evolved lines . Figure 2 shows that the dynamics of the evolved spoT ( spoTEv ) alleles were very different in the rec+ and rec− lines . The time between detection and fixation averaged 300 generations in rec+ lines and 900 generations in rec− lines ( t4 = 3 . 098 , one-tailed p = 0 . 018; Mann-Whitney test , one-tailed p = 0 . 066 ) . Thus , a beneficial mutation in the same gene took longer to fix in the absence of recombination . To identify whether the difference in time taken for spoTEv alleles to fix in rec+ and rec− lines was due to an increase in competition , the selective advantage conferred by each allele was measured at different time points during its substitution . I isolated two clones that had spoTEv alleles from three of the rec+ and two rec− lines that fixed a mutation in this gene ( the focal beneficial mutation ) , and estimated their fitness relative to five contemporary clones that were isolated from the same line and time point , but that did not have the focal mutation ( Figure 3 ) . It is important to note that these measurements were made relative to clones that did not have the focal spoTEv beneficial mutation , not against the population as a whole; therefore , a decrease in the relative advantage conferred by the focal mutation indicates an increase in fitness amongst competing clones . At early time points in their respective selective sweeps ( frequency < 0 . 1 except sample point 9 , for which frequency = 0 . 24 ) , the average advantage conferred by focal spoTEv alleles , relative to contemporary clones not having this mutation , did not differ significantly between rec+ and rec− lines , being 3 . 3% and 3 . 9% , respectively ( F1 , 3 . 013 = 0 . 028 , p = 0 . 878 ) ( Table S1 ) . To test the effect of competition from competing mutations I measured the fitness of spoTEv clones isolated from the two rec− lines at later time points , when their spread was much slower ( Figure 2 ) . Here , the relative fitness conferred by the spoTEv alleles decreased significantly , from 3 . 4% to −2 . 6% in one line and from 4 . 6% to −0 . 1% in the other ( Figure 3; Table 2 ) . There was no corresponding change in fitness relative to the ancestor between these time points , so this decrease can only be explained by the spread of one or more beneficial mutations amongst the competing clones that did not have the focal mutation ( F1 , 35 = 1 . 007 , p = 0 . 322 ) ( Table S2 ) . No significant difference was found when the fitness of spoTEv clones isolated from rec+ lines at early and late time points was compared to contemporary clones that did not have the focal mutation ( Table 3 ) . Therefore , the effect of increased competition faced by focal beneficial mutations was specific to the absence of recombination . The findings presented above indicate that multiple beneficial mutations were present in the evolving populations and that , as required by the FM model , in the absence of recombination , competition between these mutations was associated with slower fixation times . Indeed , this competition was so strong that in the rec− lines , the relative effect of the focal spoTEv mutations became negative , such that additional beneficial mutations must have occurred on the same background in order for the focal mutation to fix . The possible existence of linked deleterious mutations cannot explain this result by itself , because the deleterious mutations would impose a constant cost to the focal mutation . By contrast , in the rec+ lines , the relative advantage conferred by the spoTEv mutations showed no overall change , indicating that recombination effectively reduced the effect of competition between beneficial mutations . This reduction in competition is consistent with recombination bringing competing beneficial mutations together into one lineage , the mechanism proposed by the FM model . Three additional lines of evidence support the interpretation that competition between beneficial mutations was a significant factor in the adaptation of the high mutation rate lines . First , there was a significant difference in relative fitness between the two clones containing a spoTEv beneficial mutation in one rec− line , consistent with at least one additional beneficial mutation having arisen and reached an appreciable frequency within this lineage ( t8 = −4 . 489 , two-tailed p = 0 . 002 ) ( sample 8 in Figure 3 ) . Second , sequencing of spoT in the 16 low mutation rate lines found only one mutation . This frequency is marginally non-significantly lower than that of the five mutations found in the high mutation rate lines , consistent with there being a higher probability of fixing large effect mutations—which are generally expected to be less common—when beneficial mutations must compete for fixation [28] ( Fisher's exact test , one-tailed p = 0 . 086 ) . Third , the advantage conferred by spoTEv alleles in the competitions carried out against contemporary clones is substantially lower than the advantage of ∼9 . 4% seen in competition with the ancestral strain ( cf . [40]; Figure 3 ) . This difference provides strong evidence that contemporary clones not having a focal mutation did contain alternative beneficial mutations . It is also worth noting that the relative fitness reductions seen in the rec− lines are substantially greater than the median fitness cost incurred by gene disruption mutations in the same strain ( median effect of gene deletion , s = −0 . 014 ) [41] . It is important to note that the FM and mutation load models are not mutually exclusive . The results presented above support the FM model , but they do not rule out the possibility that removal of linked deleterious mutations may also have played a role in increasing the speed with which the focal beneficial mutation fixed in rec+ lines . One way to assess the potential importance of this process is to estimate the amount of within-population genetic variance in fitness , w ( gen ) , prevailing in the rec+ lines at the time when the focal mutations arose . For deleterious mutations to play an influential role in limiting the spread of these mutations , the benefit they confer must be small relative to this variance ( s ≪ 6σw ( gen ) [35] ) . Removing linked deleterious mutations will only provide a substantial advantage if this condition is met . Using the fitness data reported above I estimated w ( gen ) in the rec+ lines , based on those clones that did not have the focal spoTEv mutation , when this mutation was at a low frequency . This measure is conservative because it combines differences between clones due to beneficial as well as deleterious mutations . I found that the genetic variance in fitness was clearly sufficient to influence the spread of the focal mutation in only one of the three rec+ populations ( Table S3 ) . Therefore , while deleterious mutations may have played some role in slowing adaptation in rec+ populations , they are not sufficient to explain the overall advantage to recombination . The results reported in this study have important implications for bacterial adaptation . Competition between different lineages of the same species may be common in certain environments , for example , in clinical settings where adaptation to novel hosts can select for strains with high mutation rates [42 , 43] . In the absence of recombination , clonal interference means that increasing mutation rates will not generally cause a proportional increase in the rate of adaptation [16–18] . Conjugative plasmids are commonly found in clinical isolates and may reduce this interference by recombining competing beneficial mutations [44] . Plasmid transfer between lineages with different beneficial mutations may also contribute to continued selection for the plasmids themselves [45] . In summary , I found that recombination and mutation rate interact with each other in determining the speed of adaptive evolution . This finding supports the FM model for the evolution of sex , demonstrating that recombination increases the rate of adaptation only when competing beneficial mutations are present in the population . Also , I was able to identify a beneficial mutation that contributed to adaptation in a number of evolved populations . Comparing the dynamics of this mutation across recombination treatments allowed me to demonstrate directly that recombination reduced the amount of interference between a focal beneficial mutation and other competing beneficial mutations . This reduction in competition shortened the time needed for the focal mutation to fix in the population , providing an explanation for the observed benefit of recombination .
An F plasmid was obtained from the Coli Genetic Stock Center ( Yale University , New Haven , Connecticut , United States ) , as an isolate from strain K603 ( CGSC#6451 ) . This plasmid , designated F1–10 , encodes resistance to tetracycline and contains a portion of the E . coli genome encompassing the lac operon . This plasmid was chosen because this region of homology increases the rate at which the plasmid recombines into the host chromosome and therefore the frequency of chromosomal gene transfer . To make a rec− derivative of this plasmid , a non-polar traD deletion was introduced using a PCR-based approach [46] . This mutation decreased the frequency of plasmid transfer by more than 104-fold and reduced the frequency of chromosomal gene transfer to undetectable levels [47] . This gene was chosen because its deletion does not affect plasmid pili production [48] and has been shown not to affect plasmid mutability [49] . The F plasmid imposes a fitness cost of ∼7% on the host cell during growth in minimal glucose medium [50] . To test if the traD deletion affected this cost , I performed a competition assay ( see below ) to measure the relative fitness of identical host cells carrying either the F plasmid or the F ΔtraD derivative . This assay found that the cost of carriage was reduced slightly , but not significantly , by the traD mutation ( difference in relative fitness of ancestral strain containing F − F ΔtraD = −0 . 033 , t16 = −1 . 430 , one-tailed p = 0 . 086 ) . A previously described strain of E . coli B , REL606 , was used as the host bacterium in the low mutation rate lines [51] . This strain carries no known plasmids or bacteriophage , and is therefore strictly asexual . Construction of a mutator derivative of this strain was carried out by P1 transduction of a disrupted allele of mutS , mutS::Tn5 , into REL606 [52] . Disruption of the mutS allele does not have any measurable effect on fitness [53] . Rec+ and control rec− derivatives of REL606 and REL606 mutS::Tn5 were made by using standard methods to separately introduce F and FΔ traD plasmids into both backgrounds to create the four ancestral strains used to found the evolution experiment . REL607 , a spontaneous mutant of REL606 that is able to utilise arabinose , and a spontaneous nalidixic acid ( Nx ) resistant mutant of REL606 were obtained and used to allow selection of different strains in control assays designed to measure recombination rates ( see below ) . Control experiments found that the rate of recombination was not affected by the mutS::Tn5 allele . All incubations were carried out at 37 °C in 96 × 2-ml blocks ( Qiagen; http://www . qiagen . com/ ) with shaking at 150 rpm . Each well was filled with 1 ml of medium . LB broth was used to grow cells from −80 °C freezer stocks . Davis minimal medium supplemented with glucose at 25 mg/l ( DM25 ) was used at all other times . To measure the number of “gene equivalents” transferred between bacteria during each 1-d cycle of the evolution experiment , I introduced the F plasmid into two derivatives of REL606 that were distinguishable on the basis of unique markers ( one strain was Nx resistant and unable to utilise the sugar arabinose [Ara–]; the other strain was sensitive to Nx but was able to utilise arabinose [Ara+] ) . These strains were inoculated from frozen stocks into LB medium and incubated for 24 h . Strains were then diluted 100-fold and 5 μl inoculated separately into DM25 . Following a further 24 h of incubation , the two strains were mixed 1:1 and diluted 1:100 into fresh DM25 medium . After 24 h of co-incubation , cells were plated on minimal arabinose plates supplemented with Nx . Only cells that had recombined the Nx resistant and Ara+ markers could grow on this plate . To estimate the total amount of horizontal gene transfer , the number of recombinant cells was first multiplied by the number of genes separating the two markers . Assuming that only half of these transmitted genes become incorporated into the recipient cell's chromosome [54] and that genes outside these markers were not transferred , this number was divided by two to arrive at a conservative estimate for the total number of genes incorporated into the recipient cell chromosome . This calculation gives an estimate of chromosomal gene transfer between rec+ cells of ∼1 × 10−4 genes/cell/generation in the evolution environment . Although this rate might seem low , it is nevertheless several orders of magnitude higher than the spontaneous mutation rate , even in the presence of the mutS mutation [18 , 38] . It is important to note that this assay was performed between donor and recipient cells that both contained the F plasmid; therefore , this estimate of gene transfer includes any effect of plasmid surface exclusion in reducing the frequency of conjugation between plasmid-containing cells . The evolution experiment consisted of eight replicate lines started with each of four ancestral strains: REL606 ( F ) ( low mutation rate , rec+ ) , REL606 ( F ΔtraD ) ( low mutation rate , rec− ) , REL606 mutS::Tn5 ( F ) ( high mutation rate , rec+ ) , and REL606 mutS::Tn5 ( F ΔtraD ) ( high mutation rate , rec− ) . All lines were propagated in 1 ml of DM25 medium in 96 × 2-ml blocks . Each day 5 μl of culture was transferred to 1 ml of fresh medium , allowing ∼7 . 64 generations per day ( = log2 200 ) . This environment and strain combination corresponds to a relative mutation supply ( calculated as the product of effective population size and mutation rate ) of ∼2 . 9 in the low mutation rate lines and ∼101 . 1 in the high mutation rate lines with respect to Figure 2 of de Visser et al . [18] . Propagation was continued for 130 d , to give a total of ∼1 , 000 generations of evolution . Every 13 d ( ∼100 generations ) , following transfer to a fresh block , glycerol was added to all lines , which were then stored at −80 °C . All lines were initially genetically uniform; therefore , all adaptation arose through de novo mutation . Several precautions were taken to eliminate the possibility of external contamination or cross-contamination during the evolution experiment . The ancestral strain contained several characteristic genetic markers that were checked every ∼100 generations throughout the experiment [51] . At no time were any bacteria observed to differ from the ancestral marker profile; therefore , it is very unlikely that there was any external contamination . I also checked for the continued presence of F and F ΔtraD plasmids by screening evolved bacteria for the plasmid-encoded tetracycline resistance marker . In no case did the frequency of plasmid carriage drop below ∼90% . Monitoring for cross-contamination between lines was facilitated by the arrangement of the evolving populations in a checker-board pattern in the propagation block . In this arrangement the four wells nearest to each population contained uninoculated medium . Observation of bacterial growth in these wells provided a sensitive means by which to observe any splash that could potentially contaminate adjacent wells . On several occasions such contamination was observed; in these cases the experiment was restarted from the previous day's block , which had been kept overnight at 4 °C . As an additional precaution , mutation rate lines that differed by the presence or absence of the kanamycin resistance marker were grown adjacent to one another . Lines were periodically plated to kanamycin-supplemented medium to detect cross-contamination . Following 1 , 000 generations of evolution , several assays were carried out to determine if mutation and recombination rates remained at ancestral values . Recombination rates were assayed as described above , except evolved strains were co-incubated with a reference strain that carried a non-synonymous mutation in the gene galK , rendering them unable to use galactose as a sole carbon source . Recombinants were selected on Davis minimal medium supplemented with Nx and galactose . Mutation rates were assayed as described previously [18] . Mutation rates to Nx resistance and arabinose utilisation were calculated and averaged to estimate the overall mutation rate . No significant changes in either recombination or mutation rates were observed in any of the evolved lines . The fitness of evolved strains relative to the ancestor was assayed by competitions carried out in the same conditions prevailing during the evolution experiment . All evolved strains were Ara− , to allow these strains to be distinguished from their ancestors; derivatives of the ancestral strains were selected that were Ara+ . These two marker types can be distinguished by plating on tetrazolium arabinose indicator medium . On this medium , cells that can utilise arabinose form white colonies , whereas cells that cannot use arabinose form red colonies . The arabinose marker is selectively neutral in the evolution environment [51] . Before each fitness assay , the two competitors were acclimated to the competition environment by growing them separately under the same environmental conditions to be used in the competition . Competitors were then mixed by diluting 400-fold into fresh DM25 medium and a sample immediately plated on tetrazolium arabinose agar to estimate the initial densities of the competing strains . At the end of 1 d of competition ( i . e . , one propagation cycle ) , a further sample was plated on tetrazolium arabinose agar to obtain the final density of each competitor . The fitness of the evolved strain relative to the ancestor was calculated as ln ( NE ( 1 ) /NE ( 0 ) ) /ln ( NA ( 1 ) /NA ( 0 ) ) , where NE ( 0 ) and NA ( 0 ) represent the initial densities of the evolved and ancestral strains , respectively , and NE ( 1 ) and NA ( 1 ) represent corresponding densities at the end of the competition . All competitions between ancestral and evolved clones were carried out with 10-fold replication . Competitions involving evolved clones that did and did not have the spoTEv allele were performed similarly . All clones were initially Ara− . To allow the two types to be distinguished from one another , spontaneous Ara+ revertants were selected from those clones that had the spoTEv allele and used in competitions . These competitions were carried out over two or four transfer cycles to increase the precision of fitness estimates . Competitions were carried out with 3-fold replication . Primers were designed to amplify overlapping fragments of the gene spoT including upstream regulatory regions . Purified PCR products were sequenced on an Applied Biosystems ( http://www . appliedbiosystems . com/ ) 3130XL capillary sequencer . Three clones were assayed at the final time point in each high mutation rate line . I sequenced spoT in three randomly chosen clones isolated from each of the 32 lines in the high and low mutation rate treatments and found mutations in four rec+ and two rec− lines in the high mutation rate treatment and one rec− line in the low mutation rate treatment . In one high mutation rate treatment rec− line , the spoTEv allele had not fixed by 1 , 000 generations . This line was continued for a further 300 generations , after which time the line was screened again and the allele was found to have fixed . The mutations found all caused amino acid substitutions in a region of SpoT involved in the synthesis of the “alarmone” molecule ( p ) ppGpp [55] , but no two lines shared the same mutation . Dynamics of mutations were tracked using a combination of RFLP- and PCR-based approaches on clones isolated from those lines in which mutations were found . For mutations that changed a restriction site I amplified a region around this site , digested this fragment , and assayed for ancestral or evolved restriction patterns . In cases where no restriction site was introduced I developed PCR approaches taking advantage of the reduced binding and extension efficiency that occurs if there is a mismatch at the 3′ end of a PCR primer . All assays were performed at least twice for each clone , and positive and negative controls were included in every block . At least 100 clones were screened at the time point immediately preceding the first detected occurrence of a mutation and at the point at which the mutation had apparently fixed . Absence of mutant and progenitor types in this sample size indicates they are unlikely to be present in the population at greater than 3 . 1% ( Wald test , 95% confidence interval 0%–3 . 1% ) . At least 45 clones were screened at each intermediate time point . Mixed models were run to test the interaction between recombination and mutation rate , with replicate evolved line nested within recombination and mutation rate treatments . Mixed models were also used to test the null hypotheses that ( i ) there was no effect of recombination treatment on the relative fitness of clones having a spoTEv allele isolated from early time points and ( ii ) there was no effect of sample time ( early versus late ) on the relative fitness of clones having a spoTEv allele in the absence of recombination . In all cases denominator degrees of freedom was estimated using a Satterthwaite approximation . | Why have sex ? One explanation is that sex is good because it allows beneficial mutations from different lineages to recombine . This reduces competition between mutations in a population and can increase the speed with which the population can adapt to environmental change . This explanation , known as the Fisher–Muller model , has extensive theoretical support; however , it is difficult to test experimentally . Using a simple microbial system I showed that recombination increased the rate of fitness improvement when beneficial mutations were common in the population and had to compete for fixation , but had little effect when mutations occurred rarely . Sequencing of candidate genes revealed the presence of the same beneficial mutation in a number of replicate populations . In the absence of recombination , this mutation took longer to spread and conferred a lower overall competitive advantage , indicating interference between competing beneficial mutations . Together , these results provide direct experimental support for the Fisher–Muller model . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"microbiology",
"evolutionary",
"biology",
"eubacteria"
]
| 2007 | Recombination Speeds Adaptation by Reducing Competition between Beneficial Mutations in Populations of Escherichia coli |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.