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
|
|---|---|---|---|---|---|
Bacterial outer membrane vesicle ( OMV ) -mediated delivery of proteins to host cells is an important mechanism of host-pathogen communication . Emerging evidence suggests that OMVs contain differentially packaged short RNAs ( sRNAs ) with the potential to target host mRNA function and/or stability . In this study , we used RNA-Seq to characterize differentially packaged sRNAs in Pseudomonas aeruginosa OMVs , and to show transfer of OMV sRNAs to human airway cells . We selected one sRNA for further study based on its stable secondary structure and predicted mRNA targets . Our candidate sRNA ( sRNA52320 ) , a fragment of a P . aeruginosa methionine tRNA , was abundant in OMVs and reduced LPS-induced as well as OMV-induced IL-8 secretion by cultured primary human airway epithelial cells . We also showed that sRNA52320 attenuated OMV-induced KC cytokine secretion and neutrophil infiltration in mouse lung . Collectively , these findings are consistent with the hypothesis that sRNA52320 in OMVs is a novel mechanism of host-pathogen interaction whereby P . aeruginosa reduces the host immune response .
Pseudomonas aeruginosa is a gram-negative , opportunistic pathogen that primarily infects immunocompromised hosts including cancer and AIDS patients , burn victims , patients on ventilators and individuals with chronic obstructive pulmonary disease and cystic fibrosis . According to the Centers for Disease Control and Prevention ( CDC ) 51 , 000 people each year contract a hospital-acquired P . aeruginosa infection in the U . S . , accounting for about 10% of all nosocomial infections [1] . An estimated 6 , 700 of these infections are caused by multidrug-resistant P . aeruginosa strains , which have a very high mortality rate and have therefore been rated as a serious threat by the CDC ( CDC , 2013 ) . Like other gram-negative bacteria , P . aeruginosa produces outer membrane vesicles ( OMVs ) , which constitute an important mechanism of interaction with hosts and competing bacterial strains in their natural environment [2 , 3] . OMVs are 50–250 nm spheroid particles derived from the outer membrane that are constitutively secreted and consist of lipids , proteins and lipopolysaccharide ( LPS ) [3] . OMVs of many gram-negative bacteria including P . aeruginosa also contain DNA [4 , 5] . OMVs are involved in quorum sensing and enable bacteria to establish a colonization niche by selectively killing or promoting the growth of other bacteria , and they transmit virulence factors and toxins to host cells , thereby modulating the host immune response [2 , 6–10] . Distinct P . aeruginosa virulence factors like alkaline phosphatase , phospholipase Cs , β-lactamase and Cif ( CFTR Inhibitory Factor ) are differentially packaged and enriched in OMVs [11 , 12] . OMVs diffuse across mucus and fuse with airway epithelial cells releasing their cargo into host cells [11 , 13] . OMVs elicit a pro-inflammatory host immune response through pathogen-associated molecular patterns ( PAMPs ) [14 , 15] . PAMPs , including LPS , peptidoglycan , flagellin , porins and lipoproteins interact with Toll-like receptors ( TLR ) in host cells , which signal through mitogen-activated protein kinases ( MAPK ) leading to increased secretion of pro-inflammatory cytokines , notably IL-8 in the case of human airway epithelial cells [14 , 16] . Cytokine secretion rapidly attracts neutrophils and macrophages to the site of infection [17] , leading to bacterial clearance in most cases . To establish a chronic infection , P . aeruginosa deploys several strategies that mediate host immune system evasion in the various stages of the colonization process . These strategies include up-regulation of the production of polysaccharide and alginate , down-regulation of virulence factor expression , reduced phagocytic uptake of P . aeruginosa by immune cells and elimination of flagellar motility and conversion to a mucoid , sessile lifestyle [18] . In addition , OMV-mediated mechanisms also counteract the host immune response to P . aeruginosa . The OMV virulence factor Cif ( PA2934 ) , for example , dampens the airway innate immune response by promoting lysosomal degradation of CFTR , which reduces chloride secretion and thereby the airway hydration that is essential for mucocilliary clearance of pathogens [19] . Furthermore , Cif increases degradation of the transporter associated with antigen processing 1 ( TAP1 ) and decreases major histocompatibility complex ( MHC ) class 1 antigen presentation in airway epithelial cells , thus attenuating the adaptive immune response to viral infection [20] . OMVs may also act as decoys that absorb anti-microbial compounds produced by the host [8] , and allow bacteria to evade immune detection during colonization , as has been shown for Neisseria gonorrhoeae [21] . Porphyromonas gingivalis packages gingipains into OMVs , which degrade cytokines , hence down-regulating the host immune response [22] . It is well established that intracellular small RNAs ( sRNAs ) have regulatory functions in P . aeruginosa [23–25] and other bacterial species [26] . sRNAs regulate cell envelope structure , metabolism , bacterial communication , quorum sensing , biofilm formation and virulence [23 , 25 , 27] . The regulatory mechanism involves either binding to target bacterial mRNAs , thereby affecting translation , or direct interaction with protein targets [28] . A particular group of regulatory sRNAs , transfer RNA ( tRNA ) fragments , has gained recognition as important biological regulators in many prokaryotic and eukaryotic species [29 , 30] . In contrast to classic bacterial sRNAs , which are very target-specific , tRNA fragments are thought to repress translation in a manner similar to microRNAs , which often regulate multiple mRNA targets [31 , 32] . Studies on bacterial regulatory sRNAs have focused on describing their endogenous effects , while their inter-species effects have remained largely unknown . Three recent reports describe the RNA content of OMVs secreted by E . coli , P . gingivalis and V . cholerae [33–35] , but no biological effects on host cells were reported . Accordingly , the primary aim of this study was to test the hypothesis that P . aeruginosa OMVs contain sRNAs that impact human cells in biologically important ways . Here , we provide a first characterization of P . aeruginosa sRNAs in OMVs and demonstrate that a specific bacterial sRNA ( sRNA52320 ) is transferred from OMVs to host cells , where it attenuates OMV-stimulated IL-8 secretion by human airway epithelial cells , and KC cytokine secretion and neutrophil recruitment in the lungs of a mouse model .
RNA-Seq analysis of P . aeruginosa and OMVs was conducted to determine if sRNAs are packaged in OMVs . We identified 481 , 480 unique sRNA sequences in OMVs purified from the supernatants of three planktonic cultures of P . aeruginosa strain PA14 ( Fig 1A ) . The median sequence length of the sRNAs was 24 nucleotides , with a minimum of 15 and a maximum of 45 nucleotides . Replicate samples from three separate cultures were highly reproducible and correlated well with each other ( r = 0 . 97 ) . The 1733 most abundant sRNA sequences in OMVs ( i . e . , at least 64 sequence reads ) mapped to 68 loci in the P . aeruginosa genome , many of which represented by similar reads of different lengths . Fifty-two loci were at least twofold enriched in OMVs compared to P . aeruginosa whole cells ( Fig 1B ) , while eight loci were at least twofold less abundant in OMVs compared to P . aeruginosa , and eight loci were about equally abundant . Relative abundance of loci read counts in OMVs compared to P . aeruginosa was very consistent across replicate samples , with eight of the ten most abundant loci being enriched in OMVs ( Fig 1B and Table 1 ) . We used bioinformatic approaches to determine if the ten most abundant sRNAs in OMVs were likely to form stable secondary structures , and to identify potential interactions between these sRNAs and human mRNAs ( Table 1 ) . The complementary sequences of the ten most abundant OMV sRNAs were aligned with NCBI human reference mRNA sequences using the algorithm BLASTN 2 . 2 . 31 [36] to identify perfect matches with human mRNAs . Matches with an E value < 12 and a connection to the innate immune response were considered potential host immune targets of the sRNAs ( Table 1 ) . We chose sRNA52320 ( #7 ) as a candidate for further analysis because of its stable secondary structure , predicted targets , and sufficient length for specific detection with PCR primers . sRNA52320 is a tRNA fragment of the first 24 nucleotides of a tRNA coding for methionine ( tRNA-Met ) . P . aeruginosa has three additional loci coding for tRNA-Met with identical anticodon loops , but distinct sequences in the 5’ and 3’ regions . S1 Fig presents RNA-Seq read alignments to the PA14_52320 locus for sRNAs isolated from OMVs ( S1A ) and from P . aeruginosa ( S1B ) . To determine if sRNAs were inside OMVs , and not adherent to the outside of the OMV , studies were conducted with RNase A , a membrane-impermeable enzyme that will only degrade sRNAs adherent to the outside of OMVs ( Fig 2A ) . In the absence of RNase A , RNA was associated with OMVs in various lengths , including 23S and 16S rRNA as well as many smaller RNAs 15–50 nt long ( Fig 2B , lane 1 ) . The RNA recovered from OMVs after RNase A digestion consisted predominantly of small RNAs around 15–50 nt ( Fig 2B lane 2 ) . As a control , RNase A completely digested RNA extracted from lysed OMVs ( Fig 2B , lane 3 ) . To determine if sRNA52320 was packaged inside OMVs , qPCR with primers specific for sRNA52320 was performed with RNA isolated from RNase-treated and untreated OMVs . RNase digestion increased the relative abundance of sRNA52320 in OMVs treated with RNase compared to OMVs exposed to vehicle ( Fig 2C ) , confirming that sRNA52320 was inside the OMVs . To determine if OMVs deliver their sRNA cargo to host cells , we performed RNA-Seq on primary human bronchial epithelial ( HBE ) cells that had been exposed to OMVs and on HBE cells that had not been exposed to OMVs . In samples from OMV-exposed HBE cells a detectable number of reads aligned to the PA14 reference genome , while RNA from unexposed HBE cells did not yield a signal above background levels ( Fig 3A ) . sRNA52320 was one of the most abundant sRNAs in exposed host cells ( Fig 3B ) . In addition , six other sRNAs were reliably detected in HBE cells exposed to OMVs . We speculated that bacterial sRNAs might function like eukaryotic microRNAs , which repress translation by imperfectly binding to many mRNA targets . To test this hypothesis , a miRanda microRNA target scan [37] was conducted and revealed that sRNA52320 is predicted to target mRNAs encoding multiple kinases in the LPS-simulated MAPK signaling pathway , including MAP2K2 , MAP2K3 , MAP2K4 , MAP3K7 and PIK3R2 . To identify proteins whose abundances were altered upon exposure to sRNA52320 , a proteomic analysis was conducted on LPS-stimulated primary HBE cells transfected with synthesized sRNA52320 or a negative control RNA ( siNC ) . Efficient transfection with sRNA52320 was verified for all samples by PCR as shown in S2A Fig Proteomic experiments with three biological replicates yielded 3902 quantifiable proteins . To select candidate proteins whose abundance were modified by sRNA52320 , we chose the top 320 differentially expressed proteins by p-value . Differential expression ranged from 52% to 173% of control , with a majority of proteins down-regulated in the presence of sRNA52320 ( Fig 4A ) . This broadly down-regulated protein expression profile is consistent with the hypothesis that sRNA52320 directly and/or indirectly represses many targets . Using Ingenuity Pathway Analysis ( IPA ) to analyze the list of differentially expressed proteins , we identified ten pathways regulated by sRNA52320 , including: Integrin Signaling , Rac Signaling , Signaling by Rho Family GTPases , Agrin Interactions at Neuromuscular Junction , Paxillin Signaling , Cdc42 Signaling , CXCR4 Signaling , GNRH Signaling , LPS-stimulated MAPK Signaling and HGF Signaling . Eight of these pathways are directly connected to the host immune response to pathogens and/or epithelial barrier function . As illustrated in Fig 4B , sRNA52320 down-regulated all detectable proteins in the LPS-stimulated MAPK signaling pathway , which IPA predicts to result in decreased IL-8 ( CXCL8 ) levels , a reduced innate immune response , and consequently increased bacterial infection . sRNA52320 was predicted to down-regulate five kinases in the LPS-stimulated MAPK pathway in the bioinformatics analysis described above ( highlighted in purple in Fig 4 ) , and two of these kinases ( MAP3K7 and MAP2K4 ) were significantly reduced by sRNA52320 ( Fig 4B ) . Ingenuity pathway analysis of the data suggest that sRNA52320 will reduce IL-8 ( CXCL8 ) levels in HBE cells ( Fig 4B ) . To test the hypothesis that sRNA52320 reduces LPS-stimulated induction of IL-8 mRNA and IL-8 cytokine secretion , experiments were conducted on primary HBE cells transfected with sRNA52320 or siNC . As described above , efficient transfection with sRNA52320 was verified for all samples by PCR as shown in S2A Fig RT-PCR analysis of IL-8 revealed that sRNA52320 reduced the LPS-mediated induction of IL-8 mRNA compared to siNC ( Fig 5A ) . Likewise , sRNA52320 reduced LPS-induced IL-8 protein secretion compared to siNC ( Fig 5B ) . To assess whether the mechanism of action of sRNA52320 requires sRNA52320 to be taken up into the host cell ( rather than acting upon RNA-sensitive Toll-like receptors on the outside of the cell ) [38] , host cells were exposed to sRNA52320 in the presence and absence of transfection reagent . sRNA52320 could be detected in lysed HBE cells only when cells were exposed to sRNA52320 and the transfection reagent ( S2A Fig ) . Importantly , sRNA52320 reduced LPS-stimulated IL-8 secretion only when it was transfected into HBE cells , but not when it was present exclusively outside of host cells ( S2B Fig ) . These results support the hypothesis that sRNA52320 suppresses LPS-induced IL-8 secretion by interfering with translation of host mRNAs inside HBE cells rather than acting upon RNA-binding cell membrane receptors , such as TLR7/8 . To provide additional support for the observation that sRNA52320 delivered into host cells by OMVs reduces OMV-stimulated IL-8 secretion , we generated a P . aeruginosa deletion mutant for sRNA52320 as well as a re-complemented mutant that stably expressed sRNA52320 from a plasmid with an arabinose-inducible promoter ( hereafter called ΔsRNA+sRNA ) . Deletion mutants were viable despite deletion of the tRNA-Met encoded by the PA14_52320 locus due to the presence of three other redundant loci encoding tRNA-Met with the same anticodon . To allow for a direct isogenic comparison and account for potential effects of the vector or selective antibiotic , OMVs isolated from the re-complemented strain ΔsRNA+sRNA were compared to OMVs isolated from the deletion mutant transformed with an empty vector ( hereafter called ΔsRNA+vector ) . Deletion and re-complementation of sRNA52320 in P . aeruginosa were verified by PCR ( S3A Fig ) . sRNA52320 levels in re-complemented ΔsRNA+sRNA OMVs were similar to the amount of sRNA52320 naturally occurring in wt OMVs ( S3B Fig ) . When comparing ΔsRNA+vector and ΔsRNA+sRNA OMVs there was no significant difference in the amount of LPS ( S3C Fig ) or protein content ( S3D Fig ) . OMV-induced IL-8 secretion was 34% lower in HBE cells exposed to ΔsRNA+sRNA OMVs compared to HBE cells exposed to ΔsRNA+vector OMVs ( Fig 6 ) . This is consistent with the conclusion that sRNA52320 reduces OMV-induced IL-8 secretion by human airway epithelial cells . In addition , OMV-induced IL-8 secretion was 40% lower in HBE cells exposed to wt OMVs compared to HBE cells exposed to ΔsRNA OMVs ( S4 Fig ) . To determine if sRNA52320 suppresses cytokine secretion in vivo , mice were exposed to OMVs isolated from ΔsRNA+vector or ΔsRNA+sRNA P . aeruginosa . Cytokines were measured in bronchoalveolar lavage fluid ( BALF ) recovered after a 6 h exposure to OMVs . Vehicle-treated mice served as a negative control . As expected , both types of OMVs induced a cytokine response compared to control mice exposed to vehicle only ( Fig 7A and S1 Table ) . When comparing the effect of ΔsRNA+vector OMVs versus ΔsRNA+sRNA OMVs , KC , a murine functional homolog of IL-8 , was the only cytokine that was significantly , and differentially expressed in mouse BALF out of the 31 cytokines measured ( S1 Table ) . KC levels were almost twice as high in BALF from mice exposed to ΔsRNA+vector OMVs compared to mice exposed to ΔsRNA+sRNA OMVs ( Fig 7B ) . This observation is consistent with the conclusion that sRNA52320 selectively reduces OMV-induced KC secretion in mouse lung . Moreover , when mice were exposed to OMVs isolated from ΔsRNA or wt PA14 for 6 h , the number of neutrophils in the BALF of mice exposed to ΔsRNA OMVs was 3-fold higher compared to mice exposed to wt OMVs ( Fig 7C ) , suggesting that OMVs lacking sRNA52320 induced a more robust neutrophil response .
We describe the first example of trans-kingdom biological activity of a regulatory sRNA contained in bacterial OMVs . Our studies demonstrate that sRNA52320 in OMVs secreted by P . aeruginosa is a novel mechanism of pathogen-host communication that regulates the immune response in human airway epithelial cells and in mouse lung ( Fig 8 ) . In this report we demonstrate for the first time that OMVs produced by an extracellular pathogen deliver sRNAs to host cells . However , the relative abundance of sRNAs in host cells following transfer does not exactly mirror their relative abundance in OMVs . This discrepancy may be explained by differences in RNA transfer efficiency or stability following transfer to the host . Several recent studies have described the presence of sRNAs in OMVs [33–35] , or characterized sRNA produced by intracellular bacteria [39–42] , but none to date have demonstrated a role of sRNAs in host cell biology . Furuse et al . identified a 22 nt sRNA in M . marinum , but intracellular levels of the sRNA were too low to repress target mRNA in cultured cells [39] . Moreover , a high-throughput bioinformatics approach predicted possible targets of bacterial sRNAs in the human transcriptome , and in vitro transfection of host cells with a selection of these putative sRNAs decreased target mRNA abundance [43] . However , the actual expression levels of the predicted sRNAs or their effect on a biological response , such as cytokine secretion , were not reported . OMVs isolated from P . aeruginosa elicit IL-8 secretion by lung and bronchial epithelial cell lines [14] and increase multiple pro-inflammatory cytokines in a murine macrophage cell line [44] as well as in mouse BALF [15] . In this study we showed that sRNA52320 , which was transferred from OMVs to host cells , reduced OMV-stimulated IL-8/KC secretion by human airway epithelia cells and mouse lung and attenuated neutrophil infiltration in a murine model of OMV exposure . Thus , the present study provides the first evidence that sRNAs contained in OMVs target host cell mRNA and result in a reduction of OMV-induced IL-8 ( KC ) secretion and attenuated recruitment of neutrophils , professional immune cells that phagocytose and kill bacteria . We propose that P . aeruginosa uses sRNAs , such as sRNA52320 , to reduce the ability of the innate immune system to clear P . aeruginosa from the lungs of infected individuals . Although the sRNA content of P . aeruginosa OMVs has not been described previously , multiple groups have used RNA-Seq to characterize the intracellular sRNA content of P . aeruginosa and found that sRNA expression is highly variable among different strains and growth conditions [23 , 25 , 45–47] . Ghosal et al . recently found that OMVs produced by E . coli were enriched in short RNAs ( 15–40 nt ) , and that there was a difference in the profiles of intracellular , OMV-associated and OMV-free extracellular RNA [33] . Moreover , they reported that specific cleavage products of functionally important non-coding RNAs , including tRNAs , constituted a significant portion of the OMV-associated RNA . We also observed differences in the relative expression of sRNAs between OMVs and P . aeruginosa . Among the top ten most abundant sRNAs in OMVs , four tRNA fragments were significantly more abundant in OMVs compared to P . aeruginosa , suggesting selective packaging , although nothing is known about the mechanism for differential packaging of sRNAs into OMVs . tRNA-derived RNA fragments are evolutionarily conserved specific cleavage products found in all domains of life [29 , 31] . Similar to eukaryotic microRNAs , tRNA fragments have been shown to silence mRNA targets in mammalian cells [29 , 32] . tRNA fragments are also selectively packaged into human exosomes and may regulate targets in recipient cells [48] . Here we show that sRNA52320 , a tRNA-derived fragment that is predicted to target kinases in the LPS-stimulated MAPK signaling pathway , reduces the LPS-mediated induction of IL-8 mRNA and protein secretion in primary human airway epithelial cells and KC production in mouse lung . Given that the classic TLR-mediated innate immune response to LPS involves the up-regulation of multiple pro-inflammatory cytokines , it is surprising that only IL-8/KC is repressed by sRNA52320 . A possible explanation for this observation could be that sRNA52320 predominantly targets mRNA in the TLR2/4-induced innate immune response pathway , while cytokine secretion mediated by other receptors and pathways remains unaffected . For example , knockout of TLR4 and TLR2 blocks the OMV-induced secretion of KC/CXCL-1 more than any of the other cytokines [15] . Likewise , inhibition of TLR2 and TLR4 decreases Mycobacterium bovis-induced ERK1/2 activation and subsequent IL-8 secretion in human epithelial cells [49] . Hence , the inhibition of IL-8 and KC secretion by sRNA52320 that we observed in this study is consistent with the hypothesis that sRNA52320 primarily attenuates TLR4 signalling . In contrast to many other cytokines , KC is secreted in the earliest stages of infection , up to 12 h post-exposure [15] , making it one of the key players in the early innate immune response . It has been demonstrated that an effective immune response requires early , KC/IL-8 mediated recruitment of neutrophils to respond to an acute bacterial infection [17] . P . aeruginosa-mediated attenuation of IL-8 secretion by sRNAs and the resulting reduction in neutrophil infiltration may tip the balance from bacterial clearance towards persistent colonization in individuals with compromised immunity or barrier function . Further investigation beyond the scope of this study is needed to determine if deletion of sRNA52320 does in fact improve bacterial clearance and reduce mortality or chronic infection rates . The OMV-mediated delivery of sRNAs to host cells might be common to all gram-negative bacteria . It is tempting to speculate that other clinically relevant gram-negative bacteria like Salmonella , E . coli , Yersinia pestis , Klebsiella , Shigella , Moraxella , Helicobacter , Acinetobacter , Campylobacter , Legionella , Neisseria and Hemophilus might also use OMV-sRNA based mechanisms to their advantage in the course of infection . Looking beyond the host-pathogen interaction , OMV sRNAs may also be a mechanism by which gram-negative bacteria compete with other microbes that inhabit the same ecological niche . For example , we identified several stretches of 12–16 nucleotides of sRNA52320 that are antisense to functional genes in other soil bacteria including Nitrosomonas , Nitrobacter , Rhizobium , Clostridium , Methylobacterium , and Variovorax paradoxus . sRNA52320 is also predicted to target genes that affect metabolism , enzymes , transporters and transcription factors in Streptococcus , Rothia , Prevotella , and Burkholderia , which often co-colonize susceptible hosts [50] . For example , a target of sRNA52320 in multiple other species of bacteria is the TonB-dependent siderophore receptor , which is important for iron uptake , an essential element for bacterial growth and survival [51] . Future work is needed to elucidate fully the role of OMV sRNAs in microbe-host as well as microbe-microbe interactions and to potentially identify new druggable targets for control of bacterial infections . Once these mechanisms are better understood , the targeted design of sRNA antagonists or the inhibition of OMV production or fusion with host cells might open up new avenues of treatment and the prevention of infections in the face of increasing antibiotic resistance .
This study was conducted in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The Dartmouth Institutional Animal Care and Use Committee approved all work with mice ( protocol #hoga . da . 1 ) . Euthanasia was performed in accordance with the 2013 AVMA Guidelines for the Euthanasia of Animals . Mice were anesthetized with isoflurane during instillation of OMVs into the lung . 6 h after OMV exposure , mice were euthanized with a combination of anesthesia until respiration ceased , followed by cervical dislocation to confirm death . P . aeruginosa ( strain PA14 ) was grown in lysogeny broth ( LB ) as described [52] . For RNA isolation from whole bacteria , 1 ml of an overnight culture was pelleted by centrifugation at 3 , 300 g for 3 min and washed twice with phosphate-buffered saline ( PBS , Thermo Fisher Scientific Inc . , Waltham , MA , USA ) . P . aeruginosa RNA was isolated with the miRNeasy kit ( Qiagen ) , which retains the small RNA fraction . OMVs were isolated as described in [14] . Briefly , for OMV RNA isolation , 35 ml of a PA14 overnight culture was centrifuged for 1 h at 2800 g and 4°C to pellet the bacteria . The OMV-containing supernatant was filtered twice through a 0 . 45 μm PVDF membrane filter ( Millipore , Billerica , MA , USA ) followed by ultracentrifugation for 3 h at 200 , 000 g and 4°C to pellet OMVs . The OMV pellet was washed with OMV buffer ( 20 mM HEPES , 500 mM NaCl , pH 7 . 4 ) and re-pelleted by centrifugation at 200 , 000 g for 2 h at 4°C . The supernatant was removed and the OMV pellet lysed with Qiazol reagent . OMV RNA was isolated with the miRNeasy kit ( Qiagen ) , which retains the small RNA fraction . For exposure of mice to OMVs , 50–100 ml of a PA14 overnight culture were pre-processed as described above . OMVs in filtered supernatants were concentrated with 30K Amicon Ultra Centrifugal Filter Units ( Millipore , Billerica , MA , USA ) . Concentrated OMVs were pelleted by ultracentrifugation for 2 h at 46 , 000 g and 4°C , washed with OMV buffer and re-pelleted . OMV pellets were re-suspended in 60% OptiPrep Density Gradient Medium ( Sigma ) in OMV buffer and layered with 0 . 8 ml 40% Optiprep , 0 . 8 ml 35% Optiprep , 1 . 6 ml 30% Optiprep and 0 . 8 ml 20% Optiprep . Samples were centrifuged for 16 h at 100 , 000 g and 4°C . 500 μl fractions were removed from the top of the gradient , with OMVs residing in fractions 2 and 3 , corresponding to 25% Optiprep , as previously shown [14] . Matched samples of RNA isolated from whole bacteria and the corresponding OMVs were sequenced in three individual preparations . RNA-samples were digested with DNase ( DNA-free , Thermo Fisher Scientific ) and RNA quality was assessed with a Bioanalyzer ( Agilent Technologies , Santa Clara , CA , USA ) . For each sample , 1 μg DNase-treated total RNA was used for preparation of cDNA libraries with the TruSeq Small RNA Library Preparation Kit ( Illumina , San Diego , CA , USA ) . Libraries were sequenced as 50 bp single-end reads on an Illumina Genome Analyzer . Reads were trimmed and aligned to the PA14 reference genome ( NC_008463 . 1 ) using CLC Genomics Workbench ( CLC-Bio/Qiagen ) . The RNA-Seq analysis was run with the following modifications from the standard parameters: a ) use of 50 additional bases up- and downstream of annotated genes to capture sRNAs that align to intergenic regions , b ) maximum number of mismatches = zero to eliminate unspecific alignment of yeast sequences from the LB medium and c ) maximum number of hits for a read = 4 . Pileups of uniquely mapped reads as well as frequency tables for each unique sequence ( generated with the CLC Small RNA Analysis tool ) were exported for normalization and further analysis with the R software environment for statistical computing and graphics [53] . The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus [54] and are accessible through GEO Series accession number GSE71598 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE71598 ) . RNA secondary structure predictions were obtained for the ten most abundant OMV sRNAs using the RNAfold WebServer [55] . The complementary sequences of the ten most abundant OMV sRNAs were aligned with NCBI human reference sequences using the algorithm BLASTN 2 . 2 . 31 [36] to identify perfect matches with human mRNAs . In addition , a miRanda microRNA target prediction scan with a pairing score cutoff of 140 [37] was run for the three most promising candidates for follow-up analysis . Intact OMVs or OMV RNA ( 90 ng ) were incubated with 10 pg/μl RNase A ( Thermo Fisher Scientific ) for 1 h at 37°C . Control OMVs were incubated for 1 h at 37°C in the absence of RNase A . To remove RNase , OMVs were washed 3 times with PBS on 30K Amicon Ultra Centrifugal Filter Units ( Millipore ) . OMVs were pelleted by ultracentrifugation at 120 , 000 g for 70 min . RNA was isolated with the miRNeasy kit ( Qiagen ) and separated on a 2% agarose gel . RNA was visualized by staining with SYBR Safe ( Thermo Fisher Scientific ) . The presence or absence of sRNA52320 in whole bacteria , OMVs and transfected HBE cells was detected by RT-PCR using the miRCURY LNA Universal RT microRNA PCR system ( Exiqon , Woburn , MA , USA ) . cDNA was synthesized with the Universal cDNA synthesis kit II ( Exiqon ) according to manufacturer’s instructions . PCR amplification of sRNA52320 was performed using the ExiLENT SYBR Green master mix and custom primers design to specifically target sRNA52320 ( Exiqon ) . Human bronchial epithelial ( HBE ) cells from 6 donors were obtained from Dr . Scott Randell ( University of North Carolina , Chapel Hill , NC , USA ) and cultured as described previously [56] . Briefly , cells were grown in BronchiaLife basal medium ( Lifeline Cell Technology , Frederick , MD , USA ) supplemented with the BronchiaLife B/T LifeFactors Kit ( Lifeline ) as well as 10 , 000 U/ml Penicillin and 10 , 000 μg/ml Streptomycin . Culture plates were coated with PureCol Bovine Collagen Solution ( Advanced Bio Matrix , Carlsbad , CA , USA ) . For OMV exposures , HBE cells were polarized on PureCol-coated permeable supports ( #3801 Corning Inc . , Corning , NY , USA ) at an air-liquid interface for 3–4 weeks . During polarization , HBE cells were supplemented on the basolateral side with Air Liquid Interface ( ALI ) medium [57] . Polarized HBE cells from two donors were exposed to OMVs containing 1 . 5 μg of RNA or 20% Optiprep ( vehicle ctrl ) . After incubating for 1 h at 37°C and 5% CO2 , cells were vigorously washed 5x with PBS and RNA was isolated from HBE cells with the miRNeasy kit ( Qiagen ) . For each sample , 1 μg total RNA was used for preparation of cDNA libraries with the TruSeq Small RNA Library Preparation Kit ( Illumina , San Diego , CA , USA ) . Libraries were sequenced as 50 bp single-end reads on an Illumina HiSeq2500 . Reads were trimmed and aligned to the PA14 reference genome ( NC_008463 . 1 ) using CLC Genomics Workbench ( CLC-Bio/Qiagen ) . The RNA-Seq analysis was run with the following modifications from the standard parameters: a ) Also map to inter-genic regions , b ) Mismatch/Insertion/Deletion cost = 3 , c ) Length/Similarity fraction = 1 . 0 and d ) Maximum number of hits for a read = 1 . Pileups of uniquely mapped reads as well as frequency tables for each unique sequence were exported for normalization to library size and further analysis . The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus [54] and are accessible through GEO Series accession number GSE80421 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE80421 ) . HBE cells were seeded on PureCol-coated 6-well plates ( Corning Inc . ) at 300 , 000 cells per well . Two days after seeding ( at 60–70% confluence ) , cells were switched to antibiotic-free medium and transfected with 10 nM sRNA52320 ( Invitrogen custom siRNA , Thermo Fisher Scientific ) or 10 nM AllStars Negative Control siRNA ( siNC ) using HiPerFect transfection reagent ( both from Qiagen ) . Two days after transfection , cells were exposed to 10 μg/ml P . aeruginosa lipopolysaccharides ( LPS , Sigma L8643 ) for 5 h to induce the release of pro-inflammatory cytokines . HBE cells from 3 donors that had been transfected with sRNA52320 or siNC and exposed to LPS were subjected to proteomic analysis . Cells were trypsinized , washed twice with PBS and counted with a cell counter ( Bio-Rad Laboratories , Hercules , CA , USA ) to ensure that matched samples from each donor had the same number of cells . Each HBE cell pellet was lysed in 8 . 5 M urea/50 mM Tris pH 8 . 2 buffer by sonication . Protein content of each lysate was determined by BCA assay ( Thermo Fisher Scientific ) . Proteins were reduced by addition of 5 mM dithiothreitol ( DTT ) at 50°C for 20 minutes , followed by cooling to room temperature and alkylation with 12 . 5 mM iodoacetamide for 1 hour in the dark . This reaction was quenched by further addition of 5 mM DTT for 15 minutes at room temperature . The protein samples were diluted 1:5 ( vol:vol ) in 20 mM Tris , pH 8 . 2/50 mM NaCl and digested with sequencing grade trypsin ( Promega ) overnight ( 1:75 , w:w ) . Protein digests were desalted over C18 reverse-phase extraction cartridges ( Grace-Vydac ) , and amounts equivalent to 100 micrograms of peptides were dried in separate tubes by vacuum centrifugation . The dried peptide aliquots were labeled by reductive demethylation as described [58] and fractionated by basic pH reverse-phase fractionation into 12 fractions as described [59] . Each fraction was analyzed by UPLC-MS/MS using a Proxeon LC-1000 UPLC system fitted with an in-house fabricated microcapillary column ( 100 micron ID x 40 centimeter long ) packed with reverse-phase material ( Maisch GMBH; Reprosil-Pur C18-AQ , 120 Å , 3 micron beads ) directly into an Orbitrap Fusion tribrid mass spectrometer ( Thermo Fisher Scientific ) [60] . The Fusion was operated in data-dependent mode ( Orbitrap MS1: R = 120 K , AGC = 2 . 5e5 , max ion injection = 20 ms , scan range = 350–1500 m/z; Orbitrap MS2: R = 15 K , AGC = 5e4 , max ion injection = 50 ms , minimum signal = 5e5 , charge states = 2–4 , normalized HCD energy = 29% , top speed mode cycle time = 2 s ) . The resultant MS2 scans were data searched using Comet [61] , filtered to a 1% peptide false discovery rate ( FDR ) using the target-decoy strategy [62] and either assigned to unique protein sequences or discarded . Quantification was performed using a modified version of the MassCroQ algorithm [63] . Mean fold changes comparing sRNA52320 with siNC were calculated for each protein that was detected in all 3 replicate samples and p-values were obtained using a one sample t-test . Proteins were ranked by p-value and network analysis of the top 320 proteins was performed with QIAGEN’s Ingenuity Pathway Analysis ( IPA , QIAGEN Redwood City , www . qiagen . com/ingenuity ) . Selection criteria for top canonical pathways were p < 0 . 05 and absolute activation z-score > 2 . For the quantification of HBE IL-8 mRNA levels , 2 μg of total RNA were converted to cDNA using the RETROscript Kit ( Thermo Fisher Scientific ) . 50 ng of template cDNA were used in a TaqMan gene expression assay for IL-8 ( #Hs00174103_m1 , Thermo Fisher Scientific ) . GUSB ( #Hs99999908_m1 ) served as endogenous control . The P . aeruginosa in-frame sRNA52320 deletion mutant was generated using a construct created via the previously described Saccharomyces cerevisiae recombination technique with pMQ30 allelic replacement vector as described in [64] . The following primers were used: PA14_52320_sRNA_delete_1: cgcttctgcgttctgatttaatctgtatcaggctgaGTCCGGCCGATAACTGCCATCCAG PA14_52320_sRNA_delete_2: GTCCGTAGAATGCGCCCACACAGATCGTCGGGCTCATAACCCGAAGGTC PA14_52320_sRNA_delete_3: GACCTTCGGGTTATGAGCCCGACGATCTGTGTGGGCGCATTCTACGGAC PA14_52320_sRNA_delete_4: gcggataacaatttcacacaggaaacagctatgGAAGACCGCCGGGTTTTTCAGGAGTTG In primer sequences , upper case letters indicate P . aeruginosa-specific sequence . Lower case letters indicate homology with the cloning vector DNA . For re-complementation of the deletion mutant sRNA52320 was cloned into the arabinose-inducible expression vector pMQ70 [64] using EcoRI and SmaI restriction sites . The cloning was performed by GenScript ( GenScript USA Inc . , Piscataway , NJ , USA ) . P . aeruginosa was transformed with the sRNA52320 expression vector via electroporation , as described previously [65] . P . aeruginosa transformed with the arabinose-inducible pMQ70 vector and its derivatives were grown in LB with 100 mM arabinose and 300 μg/ml carbenicillin ( both from Sigma-Aldrich ) . LPS content of ΔsRNA+vector OMVs and ΔsRNA+sRNA OMVs was determined using the Pierce LAL Chromogenic Endotoxin Quantitation Kit and total OMV protein content was measured with the Pierce BCA Protein Assay Kit ( both Thermo Fisher Scientific ) according to manufacturer’s instructions . HBE cells were seeded on PureCol-coated permeable supports ( Corning #3801 ) at 250 , 000–500 , 000 cells/filter and polarized at a liquid-air interface for at least 3 weeks . Equal amounts of ΔsRNA+vector OMVs and ΔsRNA+sRNA OMVs ( 7 μg total protein ) were applied to the apical side . After a 6 h exposure , basolateral medium was collected for cytokine measurements . 8–9 weeks old male C57BL/6J mice ( The Jackson Laboratory , Bar Harbor , ME , USA ) were inoculated by oropharyngeal aspiration [66] with OMVs ( 7 μg total protein ) or vehicle following brief anesthesia with isoflurane . OMV protein concentrations were adjusted to keep the total inoculation volume the same for all mice in a given experiment ( either 1x or 2x 50 μl ) . 6 h after exposure , mice were euthanized using a combination of anesthesia until respiration ceased , followed by cervical dislocation to confirm death . In preparation for broncho-alveolar lavage trachea were surgically exposed and catheter tubing ( BD #427411 , Becton , Dickinson and Company , Franklin Lakes , NJ , USA ) fit to a 23 gauge needle ( BD #305145 ) was inserted into the trachea and stabilized with a single suture ( #100–5000 , Henry Schein Inc . , Melville , NY , USA ) . BAL fluid ( BALF ) was collected by flushing 1 ml of sterile PBS into the lungs and recovered from the lungs with a syringe ( BD #309659 ) . This process was repeated once . Cytokine secretion from HBE cells was measured with a PromoKine Human IL-8 ELISA Development Kit ( PromoCell GmbH , Heidelberg , Germany ) . Cytokines in mouse BALF were detected with the Millipore mouse cytokine 32-plex kit ( EMD Millipore Corporation , Billerica , MA ) . The total number of cells in each BALF sample was counted and samples were adjusted to 750 , 000 cells per ml . 200 , 000 cells per sample were added to the cytospin apparatus and centrifuged onto glass slides at 700 rpm for 5 minutes at room temperature . Slides were dried and stained with the Differential Quik Stain Kit ( Polysciences , Warrington , PA ) according to the included protocol . Neutrophils were enumerated under 100x magnification using an Olympus IX-73 microscope . Original neutrophil numbers per ml of BALF were calculated by accounting for the dilution factors used to adjust cells to 750 , 000/ml . Statistical analysis was performed using the R software environment for statistical computing and graphics [53] and GraphPad Prism 6 for Mac OS X software version 6 . 0h . Differences between experimental and control groups were evaluated using a two-tailed unpaired Student’s t-test where appropriate and are reported as differences in means ± SEM . For experiments using primary HBE cells paired t-tests or mixed effect linear models with donor as a random effect were used to account for donor-to-donor variability . Statistical significance of mouse cytokine data was determined with ANOVA followed by Tukey’s post-hoc test . sRNA52320 ( NCBI locus tag PA14_RS21305 ) , IL-8 ( P10145 ) , KC/CXCL-1 ( P12850 ) , MAP2K2 ( P36507 ) , MAP2K3 ( P46734 ) , MAP2K4 ( P45985 ) , MAP3K7 ( O43318 ) and PIK3R2 ( O00459 ) . | Pseudomonas aeruginosa is a gram-negative , opportunistic pathogen that accounts for about 10% of all hospital-acquired infections in the US and primarily infects immunocompromised hosts , including patients with chronic obstructive pulmonary disease and cystic fibrosis . Gram-negative bacteria like P . aeruginosa produce outer membrane vesicles ( OMVs ) , which constitute an important mechanism for host colonization . In this study we demonstrate a novel mechanism of pathogen-host interaction that attenuates the innate immune response in human airway epithelial cells and in mouse lung through a regulatory sRNA contained inside OMVs secreted by P . aeruginosa . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"immunology",
"rna",
"extraction",
"microbiology",
"pseudomonas",
"aeruginosa",
"physiological",
"processes",
"developmental",
"biology",
"molecular",
"development",
"rna",
"isolation",
"molecular",
"biology",
"techniques",
"bacteria",
"extraction",
"techniques",
"bacterial",
"pathogens",
"mapk",
"signaling",
"cascades",
"research",
"and",
"analysis",
"methods",
"pseudomonas",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"immune",
"response",
"immune",
"system",
"biomolecular",
"isolation",
"signal",
"transduction",
"host",
"cells",
"cell",
"biology",
"virology",
"physiology",
"secretion",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"organisms",
"signaling",
"cascades"
]
| 2016 | A Novel Mechanism of Host-Pathogen Interaction through sRNA in Bacterial Outer Membrane Vesicles |
Opacification of the ocular lens , termed cataract , is a common cause of blindness . To become transparent , lens fiber cells undergo degradation of their organelles , including their nuclei , presenting a fundamental question: does signaling/transcription sufficiently explain differentiation of cells progressing toward compromised transcriptional potential ? We report that a conserved RNA-binding protein Celf1 post-transcriptionally controls key genes to regulate lens fiber cell differentiation . Celf1-targeted knockout mice and celf1-knockdown zebrafish and Xenopus morphants have severe eye defects/cataract . Celf1 spatiotemporally down-regulates the cyclin-dependent kinase ( Cdk ) inhibitor p27Kip1 by interacting with its 5’ UTR and mediating translation inhibition . Celf1 deficiency causes ectopic up-regulation of p21Cip1 . Further , Celf1 directly binds to the mRNA of the nuclease Dnase2b to maintain its high levels . Together these events are necessary for Cdk1-mediated lamin A/C phosphorylation to initiate nuclear envelope breakdown and DNA degradation in fiber cells . Moreover , Celf1 controls alternative splicing of the membrane-organization factor beta-spectrin and regulates F-actin-crosslinking factor Actn2 mRNA levels , thereby controlling fiber cell morphology . Thus , we illustrate new Celf1-regulated molecular mechanisms in lens development , suggesting that post-transcriptional regulatory RNA-binding proteins have evolved conserved functions to control vertebrate oculogenesis .
Interaction of RNA-binding proteins ( RBPs ) with target mRNA is necessary for every aspect of its control , including its processing , export , localization , stability , and translation into protein [1] . These events are collectively defined as post-transcriptional control of gene expression and are essential to determine the proteome of a cell . While the role of transcription factors in vertebrate organogenesis is established–for example , a detailed transcriptional regulatory network governing lens development is now derived [2]–that of RBPs , functioning in post-transcriptional gene expression control , is not well defined [3] . This represents a significant knowledge gap , especially considering that vertebrate genomes encode similar numbers of transcription factors and RBPs [4] . Evolution of the ocular lens has enabled high-resolution vision in animals , and its development involves precise spatio-temporal control of gene expression that drives the formation of a transparent tissue containing anteriorly localized epithelial cells and posteriorly localized terminally differentiated fiber cells , which contribute to the bulk of the tissue [2 , 5] . To achieve transparency , lens fiber cells elongate , produce high amounts of refractive proteins called crystallins , and remove their organelles . Because they lose their nuclei , and with it their transcription potential , a fundamental question in ocular lens development is whether signaling and transcription are sufficient to explain the regulation of fiber cell differentiation . One hypothesis is that post-transcriptional regulatory mechanisms may have evolved to control these differentiation events as they provide additional layers of gene expression control [6 , 7 , 3] . However , the significance of post-transcriptional regulation , especially mediated by RBPs , is not characterized in lens development . Here we used a bioinformatics tool iSyTE ( integrated Systems Tool for Eye gene discovery ) to identify a conserved RBP Celf1 ( CUGBP , Elav-like family member 1; also known as Cugbp1 ) as a new post-transcriptional regulator of lens development , and by applying a variety of cellular , molecular and animal model approaches elucidate the detailed molecular mechanism of Celf1-based post-transcriptional control in the lens . Celf1 is shown to directly bind to specific mRNAs through its three RNA recognition motifs ( RRMs ) and control their distinct post-transcriptional fates namely , alternative splicing , stability and translation [8 , 9] . Celf1 is well conserved at the amino acid level differing in only two amino acids between mouse and human , and importantly the three RRMs are highly conserved across vertebrates , suggesting its functional conservation in vertebrates [8–10] . Celf1 deficiency in mouse causes spermatogenesis defects [11] and its mis-regulation is associated with myotonic dystrophy in human [12] . Further , in cardiomyocytes , Celf1 is known to control alternative splicing [13] . We now provide new advances into how Celf1-mediated distinct post-transcriptional mechanisms spatiotemporally control the proteome of differentiating fiber cells in lens development . Here , we show that Celf1 is essential for eye development in diverse vertebrates such as fish ( zebrafish ) , amphibians ( Xenopus ) and mammals ( mouse ) . Characterization of lens-specific Celf1 deletion mice by various phenotypic analyses reveals embryonic-onset defects involving abnormal cell morphology and defective degradation of nuclei that culminate into cataract . Further , molecular and cellular approaches uncover the distinct mechanisms underlying these defects . First , genome-level microarray expression profiling in combination with an effective filtering criteria identify key lens genes among the mis-regulated targets in Celf1 lens-specific conditional knockout mice . Using Celf1-RNA-immunoprecipitation ( RIP ) and cross-linking immunoprecipitation ( CLIP ) assays on lens tissue , we identify the fiber differentiation-associated nuclease Dnase2b and the ubiquitously important cell cycle regulator p27Kip1 ( Cdkn1b ) mRNAs among the endogenous direct targets of Celf1 in wild-type lens . Interestingly , Dnase2b is significantly down-regulated on the transcript level , while p27Kip1 is abnormally elevated on the protein level in Celf1 deficient lenses . Further , another cyclin-dependent kinase inhibitor , p21Cip1 ( Cdkn1a ) , is also up-regulated on both the mRNA and protein levels in Celf1 deficient lenses . Using reporter analysis in lens cells , we show that Celf1 inhibits p27Kip1 translation via its 5’UTR , which harbors a GU-rich Celf1-binding motif , whereas Celf1 over-expression causes an elevation of Dnase2b 3’UTR fused-reporter transcripts . These findings are physiologically relevant because in differentiating lens fiber cells cyclin-dependent kinase inhibitors need to be down-regulated so that Cdk1 ( Cyclin-dependent kinase 1 ) can be activated for phosphorylation of lamin A/C , resulting in nuclear envelope breakdown [14] , which in combination with Dnase2b [15] is necessary for nuclear degradation and lens transparency . In agreement , we find lamin A/C phosphorylation defects in Celf1 deficient mouse lenses . Thus , our data shows that Celf1 is necessary for high levels of the DNA-degrading enzyme Dnase2b , as well as its access to fiber cell nuclear DNA ( by controlling p27Kip1 , p21Cip1 , and lamin A/C phosphorylation ) to facilitate degradation of nuclei . Furthermore , we uncover yet other new Celf1 targets that provide an explanation for lens fiber cell morphology defects in Celf1 deficient mice . We find the cell membrane organization/stability protein Sptb ( Spnb1 , Beta-spectrin ) splice isoforms to be mis-regulated in Celf1 deficient mice lenses suggesting that Celf1 controls alternative splicing in the lens . Additionally , we find that Celf1 directly binds to the mRNA of the F-actin-binding protein Actn2 , which is down-regulated in Celf1 deficient lens , thus explaining the defects in their fiber cell morphology . These findings provide a new RBP-associated molecular mechanism for the pathology of the eye disease cataract , while demonstrating that core regulators of cell division , as well as those involved in determining cell morphology , can be recruited by RBPs to coordinate cellular differentiation .
We used an eye gene discovery tool iSyTE , which has previously led to the characterization of several cataract-linked genes [16–18] , to identify Celf1 as a potential new regulator of lens development . The iSyTE approach is based on meta-analyzed microarray gene expression data on wild type mouse lens development [17 , 19] . It uses a strategy termed “Whole embryonic body tissue ( WB ) in silico subtraction” which is a comparative analysis of lens data with mouse WB , which allows identification of candidate genes with high lens-enriched expression . Analysis of the mouse genome using iSyTE tracks identified Celf1 with a high lens-enriched expression score–among top 2% ( S1A and S1B Fig ) . The spatio-temporal expression pattern of Celf1 in embryonic lens development is conserved in fish ( Zebrafish , Danio rerio ) , amphibians ( Frog , Xenopus laevis ) and mammals ( Mouse , Mus musculus ) ( Fig 1A–1C , S1C–S1F’ Fig ) . In zebrafish , celf1 mRNA is expressed early in lens development at 1 day post fertilization ( dpf ) ( Fig 1A ) and elevates by 4dpf in the posterior lens where fiber cells differentiate ( S1C–S1E’ Fig ) , which is also validated by transgenic enhancer analyses ( Tg-1 . 2celf1:nucGFP ) ( S2A–S2I” Fig ) . In X . laevis , celf1 mRNA ( S2F and S2F’ Fig ) and protein ( S2J and S2J’ Fig ) expression in the lens is detected at early tail-bud stage ( St . 23 ) and increases at late tail-bud stage ( St . 27–30 ) ( Fig 1B ) . In mouse , Celf1 mRNA and protein expression is high in lens fiber cells and low in the lens epithelium ( Fig 1C–1F ) . While Celf1 protein stays high in fiber cells , its expression progressively increases in the epithelium ( S2M–S2P”‘ Fig ) . Celf1 expression in the mouse lens is also confirmed by knock-in reporter analysis ( S2Q and S2Q’ Fig ) . Together , these data indicate that Celf1 expression is conserved in vertebrate eye development . To investigate Celf1 function in mammalian lens development , we generated a new Celf1 conditional knockout mouse line ( Celf1cKO/cKO ) using a lens-expressed Cre-driver line ( Pax6GFPCre ) [20] . To minimize inefficient conditional gene deletion , we generated compound conditional knockout mice ( Celf1cKO/lacZKI ) that carried one conditional knockout allele and one germline knockout allele ( S3A–S3F Fig ) . In Celf1cKO/lacZKI mice , low level of Celf1 protein was observed at E14 . 5 ( S3E Fig ) . We additionally characterized Celf1 germline knockout/knock-in mice ( Celf1lacZKI/lacZKI ) that were previously generated [11] . In zebrafish and X . laevis , morpholinos were used to generate celf1 knockdown ( celf1KD ) morphants ( S4A–S4E Fig ) . Celf1 deficiency in all systems resulted in eye defects , suggesting an evolutionarily conserved requirement for Celf1 in eye development ( Fig 1G–1I’ ) . In zebrafish , celf1 morphants exhibit lens defects and cataracts at 4dpf ( Fig 1G and 1G’; S5A Fig ) while in X . laevis , majority of celf1 morphants ( 55 . 7% , n = 136 ) exhibit microphthalmia ( small eye ) ( Fig 1H and 1H’ ) and a minority ( 27% , n = 66 ) exhibit a skewed eye defect ( defined as distorted eyes with abnormal appearance of retina and lens tissue ) ( S5B–S5D Fig ) . In mouse , all three Celf1 homozygous deletion mouse genotypes–Celf1cKO/cKO , Celf1cKO/lacZKI and Celf1lacZKI/lacZKI–exhibit severe lens defects including cataracts at early postnatal ( P ) stages ( Fig 1I–1K’; S6A and S6B’ Fig ) . In Celf1cKO/lacZKI mice , lens fiber cells exhibit abnormal spaces at E16 . 5 ( Fig 1L and 1L’ ) , suggesting that perturbation of the fiber differentiation program causes cellular morphological defects that underlie the cataract phenotype . In sum , while the cataracts were obvious in zebrafish celf1 morphants and mouse Celf1 knockouts , and the frog celf1 morphants had small lenses , obvious cataracts were not detected in frogs . In addition to ocular defects , frog celf1 morphants exhibit somite segmentation defects ( S7 Fig ) , similar to earlier reports [21] . Collectively , these findings demonstrate that Celf1 is necessary for lens development in vertebrates . To understand the molecular mechanism underlying Celf1 function in mouse lens development , we first performed genome-level expression profiling in newborn Celf1cKO/lacZKI lenses , which identified several differentially expressed gene candidates ( DEGs ) ( Fig 2A ) . Further , analysis of the Celf1cKO/lacZKI lens DEGs by comparing them to normal mouse lens gene expression data in iSyTE shows that majority of the genes down-regulated in Celf1cKO/lacZKI lenses exhibit highly enriched expression in lens development , while up-regulated genes had no such pattern ( Fig 2B ) –indicating that Celf1 is necessary for expression of genes associated with differentiating fiber cells . To identify high-priority targets among Celf1cKO/lacZKI lens DEGs , we next applied an effective filtering criteria that in the past has successfully pointed to key genes that explained the lens phenotype in other gene knockout mice [18 , 22–24] . These analyses identify known–as well as new–candidate genes involved in different aspects of fiber cell differentiation , in turn offering avenues for detailed investigation ( see below ) for explaining the cataract pathology observed in Celf1 deficient lenses . Interestingly , among the Celf1cKO/lacZKI lens high-priority DEGs , Dnase2b –a lysosomal enzyme that is highly enriched in normal lens development [15 , 25] , is down-regulated in Celf1cKO/lacZKI lenses ( Fig 2A ) . This is of significance because Dnase2b is necessary for fiber cell nuclear degradation during terminal differentiation , and mice deficient for this gene exhibit cataracts [15 , 26] . We therefore examined the consequence of Dnase2b down-regulation in Celf1lacZKI/lacZKI early postnatal lens and detected the abnormal retention of nuclei in centrally located fiber cells ( Fig 3A–3B’ and S8 Fig ) . This defect is observed in all three types of Celf1 knockout mouse lenses , even at later stages , indicating that the nuclear degradation pathway is not simply delayed but is fundamentally defective ( S9A–S9F” Fig ) . Strikingly , zebrafish celf1 morphant lenses also exhibit defective fiber cell nuclear degradation , suggesting a functional conservation of Celf1 in developing fish and mammalian lenses ( Fig 3C–3D’; S9G–S9H’ Fig ) . We next examined the microarray data for altered expression of other factors involved in nuclear degradation , but did not uncover significant changes in such factors in the Celf1cKO/lacZKI lens data . While this analysis does not rule out translational level changes in these factors , it reinforced Dnase2b as a key candidate for further investigation in Celf1cKO/lacZKI lens . Dnase2b mRNA down-regulation in Celf1cKO/lacZKI lenses is confirmed by RT-qPCR ( Fig 3E ) . Further insight into Celf1-mediated Dnase2b control is gained from RNA-immunoprecipitation ( RIP ) and cross-linking immunoprecipitation ( CLIP ) assays that demonstrate an enrichment of Dnase2b mRNA in the Celf1 antibody pulldown on normal mouse lens ( stage P15 ) ( Fig 3F and 3G ) . These findings identify Dnase2b as a direct target of Celf1 and lead to the hypothesis that this RBP-RNA interaction is necessary for elevated Dnase2b transcript levels . Indeed , Celf1 overexpression in NIH3T3 cells that carry a luciferase reporter-Dnase2b 3’ UTR fusion construct result in elevated levels of luciferase reporter transcripts ( Fig 3H; S10A and S10B Fig ) . Together , these data indicate that Celf1 functions to control Dnase2b mRNA levels through interactions with its 3’UTR . In normal fiber differentiation , in addition to optimal levels of Dnase2b , phosphorylation of nuclear envelope proteins lamins A and C is also necessary for the breakdown of the nuclear membrane so that Dnase2b can gain access to fiber nuclear DNA [14] . In Celf1cKO/lacZKI lenses , fiber cell nuclei exhibit reduced phosphorylation of lamin A/C ( Fig 4A–4D’; S11A–S11F Fig ) . Interestingly , there is variation in the levels of phospho-lamin in the fiber cell nuclei of Celf1cKO/lacZKI lens , which may reflect the presence of residual Celf1 protein in a subset of fiber cells in the Celf1cKO/lacZKI lens . Together , these findings indicate that defective phosphorylation of nuclear lamins and reduced Dnase2b levels together contribute to the fiber cell nuclear degradation defects in Celf1cKO/lacZKI lenses . We next investigated the molecular basis of the lamin A/C phosphorylation defect in Celf1cKO/lacZKI mice . It is established that expression of the cyclin-dependent kinase ( Cdk ) inhibitor protein p27Kip1 ( Cdkn1b ) gets elevated in epithelial cells located in the lens transition zone , where it functions in coordinating their cell cycle exit and commitment to fiber differentiation [27] . In later stages of fiber differentiation , a sharp reduction of p27Kip1 protein is necessary for activation of the cyclin-dependent kinase Cdk1 , which in turn phosphorylates nuclear lamin A/C to initiate nuclear envelope disassembly [14 , 28] . However , the mechanism underlying the sharp reduction of p27Kip1 is not understood [29] . We find that p27Kip1 protein levels are abnormally high in Celf1cKO/lacZKI lens fiber cells without an accompanying significant increase in its mRNA ( Fig 4E–4I; S12A Fig ) . This lack of transcript-level changes of p27Kip1 in the Celf1cKO/lacZKI lens suggests that this factor may be regulated at the post-transcriptional level in the lens . In the Celf1cKO/lacZKI lens , at E16 . 5 , a few nuclei of the lens epithelium exhibited elevated p27Kip1 protein expression compared to control , but this subtle defect was not observed by stage P0 ( S12B Fig ) . However , the Celf1cKO/lacZKI lens fiber cells continued to express high p27Kip1 protein levels at P0 ( S12C and S12D Fig ) . Together , these data suggest that Celf1 normally functions to post-transcriptionally inhibit p27Kip1 protein expression in late stages of fiber differentiation . In support of this , p27Kip1 transcripts are enriched in Celf1-RIP and CLIP pulldown assays ( Fig 4J and 4K ) , suggesting that Celf1 directly associates with p27Kip1 mRNA in the developing mouse lens . Several studies have shown that Celf1 binds to GU-rich motifs in target RNAs . These are based on SELEX ( systematic evolution of ligands by exponential enrichment ) assays [30] and by identifying conserved motifs within various Celf1-controlled transcripts [31–33] . Further , the structural bases for the strong preference of Celf1 for GU-rich element binding has also been demonstrated by biophysical approaches [34] . Accordingly , the p27Kip1 mRNA 5’UTR has a GU-rich element that represents a potential Celf1 binding region ( Fig 4L ) . Therefore , to test the hypothesis that Celf1 directly represses p27Kip1 translation , we generated stable shRNA-mediated Celf1-knockdown ( Celf1-KD ) ( S13 Fig ) in a well characterized mouse lens-derived cell line 21EM15 [35] and transfected it with p27Kip1 5’UTR fused to a luciferase ORF reporter construct driven by the SV40 promoter . Celf1-KD cells had significantly elevated luciferase reporter levels compared to control ( Fig 4M ) , indicating a release of p27Kip1 translational inhibition upon Celf1 reduction . Together , these data suggest that Celf1 represses the translation of p27Kip1 in differentiating fiber cells through direct interaction with its 5’UTR . Interestingly , Celf1-mediated translational repression of p27Kip1 , potentially through interference of an internal ribosome entry site ( IRES ) in its 5’UTR , is observed in cultured breast cancer cells [36] , but has never before been described in developing tissue . Further , similar to p27Kip1 , the Cdk-inhibitor p21Cip1 ( Cdkn1a ) functions to regulate the cell cycle by inducing growth arrest [37] . While in normal lens development , p21Cip1 is repressed [27] , we find both p21Cip1 mRNA ( among the high-priority DEGs ) and protein to be abnormally elevated in Celf1cKO/lacZKI lens ( Fig 2A , Fig 4N–4O’ ) . This is of significance given that elevated levels of p21Cip1 has been linked to cataract [38] . We also performed Celf1 CLIP-RTqPCR on mouse lenses for p21Cip1 , but did not detect p21Cip1 transcripts to be enriched in the Celf1 pulldowns . This may be because the lens does not normally express p21Cip1 transcripts [27] . However , the direct interaction of Celf1 with p21Cip1 has been investigated in HeLa cells that have abundant expression of p21Cip1 mRNA [39] , and in this dataset CLIP identifies p21Cip1 as a direct binding target of Celf1 . Together , our findings indicate that Celf1 is necessary to negatively control the expression of the cell cycle regulators p27Kip1 and p21Cip1 and the phosphorylation status of lamin A/C to orchestrate fiber cell nuclear envelope breakdown in lens development . The severe fiber cell defects observed in Celf1cKO/lacZKI mice ( Fig 1L and 1L’ ) suggest its potential function in controlling fiber cell morphology/organization . In agreement with this , among the Celf1cKO/lacZKI lens DEGs , transcripts for Actn2 ( α-actinin 2; F-actin crosslinking protein ) and Sptb ( also known as Spnb1; β-spectrin; protein required for cell membrane organization/stability ) are among the high-priority candidates that are significantly downregulated ( Fig 2A ) . The reduced Actn2 transcript levels were confirmed by RT-qPCR ( Fig 5A ) . Actn2 is a spectrin family protein that contains a conserved actin-binding domain to facilitate crosslinking of actin [40] . Interestingly , Actn2 knockdown in zebrafish results in lens defects and microphthalmia [41] . Further , iSyTE analysis identifies Actn2 as a lens-enriched gene in development ( Fig 2A ) . These findings offer a hypothesis that the abnormalities in fiber cell morphology in Celf1cKO/lacZKI lenses may be reflective of Actn2 downregulation-mediated cytoskeletal defects . Therefore , we next tested potential interactions between Celf1 protein and Actn2 mRNA by performing RNA immunoprecipitation ( RIP ) on normal mouse lenses ( stage P15 ) . RIP analysis identified Actn2 as an enriched transcript in Celf1-pull down lysates but not in the IgG control ( Fig 5B ) , suggesting that Celf1 directly interacts with Actn2 mRNA . Next , we investigated Sptb ( β-spectrin ) expression in more detail in the Celf1cKO/lacZKI lens . Spectrins ( classified as α- and β-spectrins ) are membrane skeletal proteins that line the cell membrane mediating its stability by positioning transmembrane proteins and thereby controlling cell shape [42] . Further , spectrins crosslink with actin filaments , which is important for fiber cell packing [43 , 44] . We performed CLIP-RTqPCR on mouse postnatal day 13 lens and identified Sptb transcripts in the Celf1 pulldown , suggesting that Celf1 directly regulates Sptb transcripts ( Fig 5C ) . We identified two differentially expressed Sptb mRNA splice isoforms in normal lenses , the high-expressed Sptb isoform 1 ( ENSMUST00000021458 , contains 35 exons and codes for a splice isoform with a longer C-terminus region ) and the low-expressed Sptb isoform 2 ( ENSMUST00000166101 , contains 31 exons and codes for a splice isoform with a shorter C-terminus region ) ( S14 Fig ) . Using RT-qPCR , we find the abundance of these isoforms to be affected in the Celf1cKO/lacZKI lenses , with Sptb isoform 1 being reduced and Sptb isoform 2 being elevated ( Fig 5D ) . Spbt isoform 1 contains the pleckstrin domain in its extended C-terminal region , which is involved in membrane interactions , and based on its expression , is the predominant isoform in the lens . Thus , the abnormal levels of the two Sptb isoforms , along with reduced Actn2 , may contribute toward the fiber cell morphology defects in Celf1cKO/lacZKI lenses . To determine the impact of these mis-regulated cyto- and membrane-skeletal factors , we investigated F-actin pattern in Celf1cKO/lacZKI and control lenses in mouse and fish . In mouse , phalloidin staining of lens sections shows that while F-actin levels are not altered significantly , it appears abnormal , likely secondary due to cytoskeletal defects or the gross cellular disorganization in Celf1cKO/lacZKI lenses ( Fig 5E and 5E’ ) . Similarly , in zebrafish , compared to control , F-actin appears abnormal in celf1 knockdown lenses ( Fig 5F and 5F’ ) . These findings demonstrate a critical function for Celf1 in maintaining fiber cell cytoskeletal structure in lens development . Further , in mouse , scanning electron microscopy ( SEM ) was performed to analyze cortical fiber cells ( located in the lens outer cortex ) and nuclear fiber cells ( located near the core of lens ) at stage P15 . In control lenses , both cortical and nuclear fiber cells are arranged in a discrete parallel arrangement that interlink with the neighboring fiber cells through membrane protrusions ( Fig 5G and 5H ) . In contrast , both cortical and nuclear fiber cell organization is severely abnormal in Celf1cKO/lacZKI lens , exhibiting irregular arrangement of membrane protrusions and inter-digitations between neighboring cells ( Fig 5G’ and 5H’ ) . Collectively these data suggest that Celf1 deficiency causes severe alterations in mRNA levels or specific isoform abundance of key genes Actn2 and Sptb that results in abnormal fiber cell morphology .
While signaling , transcriptional , epigenetic and non-coding RNA-mediated gene expression control is well characterized , the importance of RBP-driven post-transcriptional control in vertebrate organogenesis is not well defined , barring a few exceptions [13 , 45] . Further , for the past 50 years , the ocular lens has been studied as an extreme example of cell type-specific gene expression , largely focusing on the transcriptional control of crystallin genes in cells that eventually degrade their nuclei and other organelles to achieve transparency . Here , we demonstrate that a conserved RBP is necessary for the dynamic spatio-temporal control over the lens fiber cell proteome , and its deficiency in different vertebrates results in cataract . Thus , the long-standing dogma that transcription is the predominant factor in regulating lens fiber differentiation is refuted by these findings that highlight distinct post-transcriptional roles of Celf1 in the lens . Significantly , these data reveal a new RBP-based mechanistic layer of control–involving mRNA translation , stability or alternative splicing–over p27Kip1 , Actn2 , Dnase2b and Sptb expression during lens development . These data establish Celf1 as an important regulator in vertebrate lens development , while shedding new light on the regulation of known–as well as several new–target genes . For example , ( 1 ) while it was known that germline deletion of Dnase2b in mouse caused nuclear degradation defects and cataract [15] , the mechanism of how Dnase2b was expressed at high levels was not understood; ( 2 ) Similarly , it was shown that down-regulation of p27Kip1 is necessary for nuclear degradation in differentiating fiber cells [28] , but how this p27Kip1 control was precisely accomplished in these cells was not understood; ( 3 ) Additionally , elevated expression of p21Cip1 was linked to cataract [38] , but which factor negatively controlled p21Cip1 in the lens was not understood; ( 4 ) Alternative splicing of key genes was long suspected to be important for lens development [3] , but no factor controlling this post-transcriptional control mechanism was characterized in the lens; and ( 5 ) It is long known that the characteristic cellular morphology is critical for lens transparency but the regulatory mechanism to control cytoskeletal factors , including F-actin , in these elongated cells was not well understood [44] . Data in the present manuscript establish that the RBP Celf1 mediates post-transcriptional regulation to control known ( p27Kip1 , p21Cip1 , Dnase2b ) , as well as novel ( Actn2 , Sptb ) , factors thereby orchestrating fiber differentiation during lens development . Although other post-transcriptional RBPs such as Tdrd7 and Caprin2 are known to function in lens development [16 , 46] , their gene knockout mouse models do not exhibit nuclear degradation defects that are observed in the Celf1cKO/lacZKI lens . This indicates that different RBPs have distinct function in lens development . From these new data , and in light of previous findings [14 , 15 , 28 , 38] , a model is defined for the molecular mechanism of Celf1 function in lens development ( Fig 6 ) . In one pathway , Celf1 positively regulates a lysosomal nuclease Dnase2b in fiber cells while also indirectly facilitating its access to nuclear DNA by negatively regulating p27Kip1 and p21Cip1 . Down-regulation of these Cdk-inhibitors results in the activation of Cdk1 , which phosphorylates nuclear lamins A/C to initiate fiber cell nuclear envelope breakdown , thus allowing Dnase2b access to degrade nuclear DNA . This model shows how components of the mitotic machinery–normally involved in nuclear disassembly during cell division–are rewired by RBPs to regulate cell differentiation . In separate pathways , Celf1 is necessary for appropriate levels of the F-actin crosslinking protein Actn2 ( α-actinin 2 ) , which is previously associated with lens defects [41] , and also for the high abundance of the specific alternative splice isoform of Sptb , β-spectrin , which is necessary for cell membrane integrity [43] . Deficiency of Celf1 reduces these factors , culminating into fiber cell morphology defects . Together , the fiber cell nuclear degradation and morphology defects cause cataract in the Celf1 deficient lens . The findings described here provide new control mechanisms for universally important factors such as p27Kip1 , p21Cip1 , alpha-actinin , Beta-spectrin , which are involved in differentiation and morphology of a multitude of different cell types . Indeed , the Celf1-mediated p27Kip1 post-transcriptional control mechanism described here in the lens may serve to inform on the complexity of p27Kip1 regulation in other developing tissues , and the impact of its mis-regulation on cell growth and cancer , as well as on the pathobiology of other Celf1-associated defects such as myotonic dystrophy . The translational control of p27Kip1 has been investigated in different cell types . The 5’UTR of p27Kip1 has an IRES that supports cap-independent translation and other elements that allow translational control in specific phases in the cell cycle [47–49] . Importantly , the present data show for the first time that Celf1 negatively controls p27Kip1 translation in a developing tissue . Interestingly , the ELAV family protein HuR has been shown to inhibit p27Kip1 translation via the IRES in its 5’UTR in HeLa cells [50] . We have identified HuR in mouse embryonic lens [16] and it will be interesting to investigate in future studies whether this protein functions with Celf1 to cooperatively regulate p27Kip1 translation in the lens . Further , our new findings on Celf1-based translational regulation of p27Kip1 , along with the findings that mutations in the 5’UTR of L-ferritin mRNA mis-regulate its translation into protein and result in hyperferritinaemia and cataract [51] , together serve to highlight the importance of translational control of key factors linked to cataractogenesis . Further , while Celf1 is implicated in control of p21Cip1 , depending on the specific cell line investigated , it has been reported to be either a positive regulator or a negative regulator of p21Cip1 [52 , 53] . Here , we show for the first time that in lens fiber cells , Celf1 negatively regulates p21Cip1 . The findings in the present study are generally significant because RBP function in RNA processing is increasingly recognized as a critical factor for fully comprehending human disease phenotypes [54] . Together with the finding that deficiency of the post-transcriptional regulatory protein TDRD7 causes juvenile cataracts in human , mouse and chicken [16] , these data highlight the importance of RBP-mediated post-transcriptional regulatory networks for precise spatiotemporal control of cellular proteomes in vertebrate organogenesis .
The University of Delaware Institutional Animal Care and Use Committee ( IACUC ) reviewed and approved the animal protocols ( number 1226 ) . Animal experiments were performed according to the Association for Research in Vision and Ophthalmology ( ARVO ) statement for the use of animals in ophthalmic and vision research . A new conditional knockout mouse model ( Celf1cKO/cKO ) was generated in this study wherein Celf1 exon five is flanked by loxP sites ( Celf1flox/flox ) ( Clinique de la Souris , Strasbourg ) . In addition , previously generated Celf1 germline targeted knockout animals ( referred to as Celf1lacZKI/lacZKI ) [11] were also used in this study . Further , to address potentially incomplete Celf1 deletion in Celf1cKO/cKO mice , Celf1 compound conditional knockout mice ( referred to as Celf1cKO/lacZKI ) were generated as follows . Germline Celf1 heterozygous mice ( Celf1lacZKI/+ ) were initially crossed with Pax6GFPCre transgenic mouse line [20] that express Cre recombinase in the embryonic day ( E ) 9 . 5 lens placode , to first generate Celf1 heterozygous and Pax6GFPCre heterozygous ( Pax6GFPCre+/-: Celf1lacZKI/+ ) animals . Pax6GFPCre+/-: Celf1lacZKI/+ animals were then crossed with Celf1flox/flox mice to generate Celf1 compound knockout mice ( Pax6GFPCre+/-: Celf1flox/lacZKI; referred to as Celf1cKO/lacZKI ) that remove Celf1 exon five in one of the Celf1 alleles in the developing lens starting at E9 . 5 , while the other Celf1 allele is germline deleted . Celf1 conditional knockout animals ( Pax6GFPCre+/-: Celf1flox/flox; referred to as Celf1cKO/cKO ) were generated by crossing Celf1flox/flox mice with Pax6GFPCre+/-: Celf1flox/+ mice . To generate Celf1 germline knockout , Celf1lacZKI/+ males were crossed with either Celf1lacZKI/lacZKI or Celf1lacZKI/+ females . Embryos were staged by designating the day that the vaginal plug was observed as embryonic day ( E ) 0 . 5 . Post- natal mice were staged by designating the day of birth as post-natal day 0 ( P0 ) . Control mice used in this study were either Celf1flox/flox and/or Pax6GFPCre+/-: Celf1+/+ genotype , which did not exhibit any lens defects . Genotyping was performed on tail-DNA prepared from post-natal or embryonic tissue using the Direct PCR Lysis reagent ( Viagen Biotech , Los Angeles , CA ) . Animals were genotyped by using the following primers . Celf1 germline deletion allele ( Celf1lacZKI ) is amplified by primers: KODEL2S-Forward-5’- GAATTATGGCCCACACCAGT-3’ and KODEL2R-Reverse-5’-GAGGGTTTTGGCTCCTATCC-3’ . Celf1 floxed allele ( Celf1flox ) are amplified by primers: LF-Forward-5’- CCATAACATGAAGGTCCTCCCTGGGGT-3’ and LR-Reverse-5’- GGCTGAACTGCAGGATCACAGCACC-3’ . Cre genomic region is amplified by primers: Forward-5’-TTCAATTTACTGACCGTACACC-3’ and 5’-CCGACGATGAAGCATGTTTAG -3’ . Wild-type AB and TL Zebrafish ( Danio rerio ) strains were used for this study . Fish were maintained at 28 . 5°C on a 14 hour light/10 hour dark cycle in accordance with University of Texas at Austin , IACUC provisions . To generate celf1- knockdown ( KD ) zebrafish ( celf1KD ) , celf1 pre-mRNA was targeted by celf1 antisense ( celf1-MO ) and control celf1 mismatch ( celf1-MM ) morpholinos ( MOs ) purchased from Open Biosystems and Gene Tools ( Philomath , OR ) , respectively . The predicted outcome of the morpholino-mediated knockdown on the celf1 protein is a frame shift which is expected to result in 12 incorrect amino acids being translated before a premature stop codon . Both MOs were injected at a concentration of 2 . 2 ng/embryo at the 1–4 cell stage into wild type embryos . MOs sequences are the following: celf1-MO 5'-AACATTTTCTCACCCCTGGAAGAAT-3' ( Celf1- specific morpholino ( MO ) , test ) and celf1-MM 5'-AAGATTTTGTCACCGCTGCAACAAT-3' ( Celf1 mismatch morpholino ( MM ) , control ) , wherein underline depicts nucleotide mismatches in the MM compared to the control . To confirm the splice-altering efficacy of the morpholino , RT-PCR was performed on both groups of injected zebrafish embryos ( celf1-MO and celf1-MM ) using the celf1-specific primers , Forward 5'- ATGAATGGGTCTCTGGACCAC-3' and Reverse 5'-CATTGTTTTTCTCACTGTCTGCAGG-3' . To confirm splice-defect induced celf1 knockdown , DNA from the appropriate size bands isolated by agarose gel electrophoresis was purified using QIAquick Gel Extraction Kit ( Qiagen , Hilden , Germany ) and validated by the Sanger sequencing . A ~1 . 2kb ( 1152 bp ) celf1 genomic ( potential enhancer ) sequence that is located in the upstream region of celf1 start codon was PCR amplified using the following primers; Forward: 5ʹ-GTACAGGTACCGCTTTCTCTTCCTGC-3ʹ and Reverse: 5ʹ-GTAGACACTAGTTTCTTCAGGCCTTC-3 and the amplicon was cloned into the Pgem-T Easy Vector ( Promega , Madison , WI ) . This genomic ( potential enhancer ) region encompasses the celf1 5’ UTR and extends from +9808 to +10959 downstream of the transcription initiation site ( +1 ) ( Ensemble genome sequence ( ID: ENSDARG00000005315 ) ) . This region begins at position 85 bp downstream from the start of exon 3 and includes 39bp of exon 3 as well as the first 1113 bp of intron 3 . The start codon ( ATG ) of the zebrafish celf1 is located in exon 4 . A GFP expression vector was then constructed using the celf1 1 . 2kb genomic sequence , nuclear EGFP and SV40 polyA sequence in a Tol2 transposon as previously described [55] and according to the manufacturer’s instruction from the MultiSite Gateway Three Fragment Vector Construction Kit ( Invitrogen , Carlsbad , CA ) . For transgenesis , 25pg ( picograms ) of DNA expression construct and 25pg of transposase mRNA were injected into one-cell stage embryos using a microinjector ( Harvard Apparatus , Medical Systems Research Products , Holliston , MA ) . Injected embryos were examined under a fluorescence microscope ( Leica Microscope MZ 16F , Buffalo Grove , IL ) at various developmental stages to assess for expression of the EGFP reporter gene . EGFP injected embryos ( F0; founder fish ) were grown up 3–4 months . F0 fish harboring the transgene were mated with wild type fish to generate transgenic stable lines ( F1 ) . Xenopus laevis embryo manipulations , embryo staining and histological methods were performed as previously described according to approved protocols [21] . Briefly , the celf1-specific morpholinos 5’-TTGTGCCATTCATTATCTTAGAAAT- 3’ and 5’-ATTGTGCCATTCATTTTCTTGGAAA-3’ ( NCBI Accession number BG513033 ) , were used for generating celf1-knockdown . The specific nucleotides complementary to the ATG initiation codon are underlined . The morpholino 5’-CCTCTTACCTAGTTACAATTTATA- 3’ was used as standard control . To assay the nuclear degradation defects in celf1 knockdown embryos , 4-5dpf embryos were fixed and sectioned as previously described [56] . Sections were rehydrated with PBTD ( 0 . 1% Tween-20 , 1% DMSO in 1X PBS ) and nuclei staining solution ( SytoxGreen , Molecular probe , Eugene , Oregon ) was added at 1:1000 dilution and incubated overnight at 4°C . Slides were washed three times with PBTD and mounted with vectashield mounting media ( Vector Laboratory Inc , Burlingame , CA ) and imaged with confocal microscopy . In situ hybridization ( ISH ) for detecting RNA was performed as previously described for mouse [17] using Celf1 open reading frame ( ORF ) -specific probe ( amplified by the primers: Celf1-F-5’- GCTATTTAGGTGACACTATAGACCCTGAGCAGCCTCCACCC-3’ , Celf1-R-5’- TTGTAATACGACTCACTATAGGGGCCACTGCTGCCCAGACCAC-3’ , where the underlined nucleotides in the forward primer denote the SP6 promoter sequence while the underlined nucleotides in the reverse primer denote the T7 promoter sequence ) . Mouse E12 . 5 embryonic head tissue fixed overnight in 4% paraformaldehyde ( PFA ) at 4°C were used for obtaining coronal sections ( 16 μm thickness , cryosectioned ) that were used for ISH analysis . For zebrafish , a celf1 full-length ORF-specific probe was used according to an established in situ protocol [57] . Zebrafish eyes from 1–4 dpf embryos were fixed and cryosectioned for ISH analysis . Mouse head tissue from developmental stages E11 . 5 , E14 . 5 , E16 . 5 and P0 was fixed in 4% PFA for 30 minutes on ice , and equilibrated in 30% sucrose overnight at 4°C before mounting in OCT ( Tissue-Tech , Doral , FL ) . Sections ( 16um thickness ) at similar depths were used in all the experiments to compare the expression of proteins between Celf1cKO/lacZKI and control lens . Frozen sections ( 16 μm thickness ) were blocked in either 5% chicken serum ( Abcam , Cambridge , UK ) or 5% donkey serum ( Jackson ImmunoResearch , West Grove , PA ) or 10% BSA ( Sigma-Aldrich , St . Louis , MO ) along with 0 . 1% Tween for one hour at room temperature . The following primary antibodies were purchased from Abcam , Cambridge , UK and Santa Cruz Biotechnology , Dallas , TX and used in the given dilutions in the respective blocking buffers: Celf1 ( ab-9549 , 1:500 dilution ) , p27Kip1 ( SC- 528 , 1:100 ) , Lamin A/C ( ab-58528 , 1:100 dilution ) , p21Cip1 ( SC-397 , 1:100 ) . In addition , a previously generated polyclonal antibody raised against X . laevis celf1 [21] , was used at 1:500 dilution . After one hour blocking , the sections were incubated with the primary antibody overnight at 4°C . Slides were washed and incubated with the appropriate secondary antibody conjugated to Alexa Fluor 594 ( 1:200 ) ( Life Technologies , Carlsbad , CA ) and the nuclear stain DRAQ5 ( 1:2000 ) ( Biostatus Limited , Loughborough , UK ) . Slides were washed , mounted using mounting media as described [58] . For F-actin staining , mouse lens tissue at P0 from both control and Celf1cKO/lacZKI animals were blocked with 2% BSA ( Sigma-Aldrich , St . Louis , MO ) for one hour at room temperature and stained with Alexa Fluor 568 labeled phalloidin at 1:200 dilution ( A12380 , Invitrogen , Carlsbad , CA ) , 0 . 25% Triton X-100 , and DAPI at 1:2000 dilution ( D21490 , Invitrogen , Carlsbad , CA ) for overnight at 4°C . Lenses were washed three times with 1X PBS containing 0 . 1% Triton X-100 and mounted using mounting media as described . All sections were imaged using the Zeiss LSM 780 confocal microscope configured with Argon/Krypton laser ( 488 nm and 561 nm excitation lines ) and Helium Neon laser ( 633 nm excitation line ) ( Carl Zeiss Inc , Oberkochen , Germany ) . Optimal adjustment of brightness/contract was performed in Adobe Photoshop ( Adobe , San Jose , CA ) and applied consistently for all images . Fiji imageJ software ( NIH , Bethesda , MD ) was used to quantify the differences in the fluorescence signal intensity of Lamin A/C and p27Kip1 between control and Celf1cKO/lacZKI lens . To measure the fluorescence intensity , images were split into single channel and the fluorescence intensity of the region of interest ( individual nuclei ) was measured in the red channel or the blue channel ( Draq5 staining ) for normalization and quantification of the intensity ratios in three biological replicates and a student t-test was performed to estimate statistical significance . Lenses from control and Celf1cKO/lacZKI animals were dissected and homogenized in ice-cold lysis buffer ( 50mM Tris-HCl at pH 8 , 150mM NaCl , 1% nonidet P40 , 0 . 1%SDS , 0 . 5% sodium deoxycholate , along with protease inhibitors ( Thermo Fisher Scientific , Waltham , MA ) . For cell line lysate preparation , lysis buffer ( 1 mL ) was directly added to the cell culture plate and incubated at 4°C for 30 min . Cell debris was removed by centrifuging lysates at 14 , 000 RPM for 30 min at 4°C . Protein concentration was determined using Pierce BCA protein kit ( Thermo Fisher Scientific , Waltham , MA ) according to the manufacturer’s instructions . Total protein ( 25–50 μg ) was resolved on TGX stain free polyacrylamide gels ( Bio-Rad , Hercules , CA Hercules , CA ) and transferred onto PVDF membrane ( Thermo Fisher Scientific , Waltham , MA ) . Blots were blocked with 5% non-fat dry milk for 1 hour at room temperature and incubated with primary antibody ( p27Kip1 BD Bioscience , San Jose , CA ) 610241 and Celf1 ab-9547 at 1:500 and 1:1000 dilutions , respectively ) over night at 4°C . Blots were incubated with secondary antibodies conjugated to horseradish peroxidase ( Cell Signaling Technology , Danvers , MA ) for one hour at room temperature , and the signals were detected with SuperSignalTM West Femto Maximum Sensitivity Substrate ( Thermo Fisher Scientific , Waltham , MA ) . Control and Celf1 deficient mouse eyes were dissected in 1X PBS and imaged by light microscopy ( Zeiss Stemi SV dissecting microscope ) . Mouse eyes were further dissected to isolate lenses for grid imaging . For the grid images , lenses from control and Celf1 deficient mice at stage P30 were placed on a 300-mesh electron microscopy grid ( Electron Microscopy Sciences , Hatsfield , PA , Catalog No . 6300H-Cu ) and imaged to evaluate the refractive properties of the lens as previously described [18] . Hematoxylin and Eosin ( H&E ) staining was performed on sections from mouse embryonic head tissue or postnatal eye tissue as described [58] . Mouse eyes from control and Celf1 deficient mice at stage P15 were processed as previously described [18] . Samples were imaged with a field emission scanning electron microscope , Hitachi S-4700 ( Tokyo , Japan ) . Four lenses for each of three biological replicates were collected from control and Celf1 deficient mice at stage P0 and total RNA was extracted using RNeasy mini kit ( Qiagen , Hilden , Germany ) . cDNA synthesis and RT-qPCR was performed as described [18] on ABI7300 Real-Time PCR system ( Applied Biosystems , Foster City , CA ) using Power Syber Green PCR master mix ( Invitrogen Life technologies , Carlsbad , CA ) . Transcript levels were normalized to the housekeeping gene Gapdh . For each sample , differential expression was determined using ΔΔCT method . The following primers were used for qRT-PCR: Celf1-F-5’-ACAGATGAAGCCTGCTGACA-3’ and Celf1-R-5’-CTCTGCTCAAGCCATCAGGT-3’; Dnase2b-F-5’-TGCTCTGGGGAGGACCTTAC-3’ and Dnase2b-R-5’-CCCCTGCGTTCTGTTCCATA-3’; p27Kip1-F-5’-GCCAGACGTAAACAGCTCCGAATT-3’ and p27Kip1-R-5’-AGGCAGATGGTTTAAGAGTGCCT-3’; p21Cip1-F-5’-CGGTGTCAGAGTCTAGGGGA-3’ and p21Cip1-R-5’-CATGAGCGCATCGCAATCAC-3’; Actn2-F-5’-GAATCAGATAGAGCCCGGCG-3’ and Actn2-R-5’-ATGTTCTCGATCTGGGTGCC-3’; Sptb-001 ENSMUST00000021458—Exon 36 -F-5’-TGGCTACAGAGCATGAGCAC-3’ and Exon 36 -R-5’-TCCTTTTCCTTCTTGCCAAC-3’; Sptb-002 ENSMUST 00000166101- Exon 32-F-5’- AGAGGAGGAAGGCGAGACAG-3’ and Exon 32-R-5’-GGAACTAGACAAGCGGGACA-3’; Gapdh-F-5’-GATCGTGGAAGGGCTAATGA-3’ and Gapdh-R-5’-GACCACCTGGTCCTCTGTGT-3’; B2M-F-5’- TGGTGCTTGTCTCACTGACC-3’and B2M-R-5’- CCGTTCTTCAGCATTTGGAT-3’ . Primers for Sptb isoforms RT-PCR: Sptb-001 ENSMUST00000021458 Exon 32-36-F-5’-AGAGGAGGAAGGCGAGACAG-3’ and Sptb-001 ENSMUST00000021458 Exon 32-36-R-5’-TCCTTTTCCTTCTTGCCAAC-3’; Sptb-002 ENSMUST00000166101 Exon 32-F-5’-AGAGGAGGAAGGCGAGACAG-3’ and Sptb-002 ENSMUST00000166101 Exon 32-R-5’-CTCTGGCAGCAGCGACTC-3’ . For fish , total RNA from 1dpf embryos was isolated using Trizol reagent ( Invitrogen Life technologies , Carlsbad , CA ) , and cDNA was synthesized using iScript cDNA synthesis kit ( Bio-Rad , Hercules , CA ) . Total RNA from wild-type and Celf1cKO/lacZKI mouse lenses at stage P0 was isolated ( two lenses for each biological replicate ) as described above and used for microarray gene expression profiling analysis on the Illumina MouseWG-6 v2 . 0 BeadChip platform ( Illumina , San Diego , CA ) . Raw microarray data files were imported to ‘R’ statistical environment ( http://www . r-project . org/ ) and background corrected using lumi package at Bioconductor ( www . bioconductor . org ) , followed by normalization by Rank Invariant method as previously described [59] . Differentially expressed genes ( DEGs ) were identified at a significant p-value < 0 . 05 and fold change cut-off of ±2 . 0 or ±2 . 5 . High-priority candidates were identified using iSyTE-based analysis and previously established filtering criteria [22 , 59] . The microarray data generated in this study are deposited in the Gene Expression Ominbus database ( www . ncbi . nih . gov/geo ) and the accession number is GSE101393 . Celf1 transcript expression in developing lens was investigated in previously generated wild-type mouse lens microarrays ( E10 . 5 , E11 . 5 , E12 . 5 ) ( GSE32334 ) in the iSyTE database [17] . Scatter plot analysis was performed , as previously described [59] , to investigate if DEGs in Celf1lacZKI/lacZKI lens microarrays showed preferential expression in the lens . Briefly , iSyTE assigns lens-enrichment scores to genes based on their comparison with mouse whole embryonic tissue ( WB ) [17] . For iSyTE lens- enrichment ( fold-change ) , P0 lens microarrays ( GSE165333 ) were compared to WB ( GSE32334 ) . In a quadrant plot , Celf1lacZKI/lacZKI lens DEGs ( ±2 . 0 fold-change , p-value < 0 . 05 ) were plotted against their iSyTE lens enrichment fold-change . Celf1/RNA-immunoprecipitation was performed according to manufacturer instructions ( EMD Millipore , Billerica , MA , 17–700 ) . Briefly , wild-type P15 mouse lens lysates were used ( n = 15 P15 stage lenses per replicate ) . Pre-conjugation of Celf1 antibody ( EMD Millipore , Billerica , MA , 03–104 ) and IgG antibody with magnetic beads was performed for 45 min . at room temperature and unbound antibody was removed by washing . Lens protein lysate was added to the beads-antibody complex and incubated overnight at 4°C . Bound RNA isolated by phenol-chloroform extraction was used in RT-PCR analysis . Freshly dissected wild-type stage P13 mouse lenses were UV irradiated three times at 4000 μJ/cm2 and 254 nm on ice for cross-linking , and stored at −80°C . Celf1/RNA complexes were immunoprecipitated from lens protein extracts as described [60] , except that the RNase treatment was omitted . The co- immunoprecipitated RNA was analyzed by RT-qPCR . B2M ( Beta-2-Microglobulin ) is used as a negative control in CLIP-RTqPCR and RIP-RTqPCR experiments because B2M mRNA levels are not affected by Celf1 inactivation and there is no evidence of an interaction between B2M and Celf1 . The mouse lens epithelial cell line 21EM15 was directly obtained from Dr . John Reddan ( Oakland University , MI ) . Cells were cultured in 100mm cell culture treated plates ( Eppendorf ) under standard conditions ( 10 mL of DMEM with 4 . 5 g/L glucose , L- glutamine , and sodium pyruvate included ( Corning Cellgro , Manassas , VA , 10-013-CV ) ) , 10% Fetal Bovine Serum ( Fisher Scientific , Pittsburg , PA , 03-600-511 ) , and 1% penicillin- streptomycin ( GE Healthcare Life Sciences , Logan , UT , SV30010 ) . Cells were incubated at 37°C in a humid chamber with 5% CO2 as described [16] . Five different lentiviral particles containing shRNA ( small hairpin RNA ) sequences targeting mouse Celf1 mRNA ( Sigma-Aldrich , St-Louis , MO , clone ID: NM_198683 . 1-1279s21c1; NM_198683 . 1-869s21c1; NM_198683 . 1- 1739s21c1; NM_198683 . 1-1320s21c1; NM_198683 . 1-868s21c1 ) were used to transduce mouse lens cell line 21EM15 as previously described [16] . Non-targeting shRNA control transduction particles were used as a control ( Sigma-Aldrich , Catalog No . SHC0-016V- 1EA ) . Standard manufacturer protocol was used to infect 21EM15 cells with lentiviral particles . Briefly , anti-Celf1 shRNA containing viral particles ( 106TU ) were used to infect 1 . 6 x 104 21EM15 cells in the presence of 8 μg/ml polybrene ( Sigma , St-Louis , MO ) . Media was changed 24 hours after transduction to prevent cell death from virus toxicity . Clones were selected at a final concentration of 6μg/ml puromycin and clones were selected using Pyrex cloning cylinders ( Sigma-Aldrich ) . The extent of Celf1 knockdown was determined by Western blot analysis . To test the Celf1 mediated translational repression of p27Kip1 , a plasmid containing p27Kip1 5’UTR sequence upstream of firefly luciferase reporter , pGL3- p27Kip1 LUC ( Addgene original plasmid # 23047 ) and an internal control vector , pRL-TK Renilla luciferase ( Promega ) were transiently co-transfected into Celf1 knockdown and control 21EM15 lens cell lines . After 48 hours of transfection cells were collected and firefly luciferase and Renilla luciferase activity was measured using Promega Dual luciferase reporter assay system ( Promega , Madison , WI , E1910 ) according to the manufacturer’s instructions . Signals were measured with the PromegaTM GloMaxTM 20/20 Luminometry System ( Promega , Madision , WI ) . To study Celf1 mediated regulation of Dnase2b mRNA , reporter constructs of ( a ) Dnase2b 3’UTR and ( b ) Celf1 ORF ( Celf1 over-expression ) were generated . To generate the Dnase2b 3’UTR plasmid , wild-type Dnase2b 3’UTR was cloned downstream of the firefly luciferase gene in the pmirGlo vector ( Promega , Madision , WI ) using the Gibson Assembly Master Mix kit ( New England Biolabs , Ipswich , MA , NEB#E2611S/L ) with the following primers: Forward-5’-tagttgtttaaacgagctCACACCCTCTGTCCTTGAA-3’ and Reverse-5’-atgcctgcaggtcgactCCTATATTTATTCACTTCCTTTACTGTC-3’ . The nucleotides corresponding to the target vector for the Gibson assembly are in lowercase . To generate the Celf1 over-expression plasmid , the Gateway cloning system ( Thermo Fisher Scientific , Waltham , MA ) was used . Briefly , the full-length coding sequence of Celf1 flanked by attb sites was generated by PCR according to the manufacturer’s instructions using the following primers: Forward- 5’-GGGGACAAGTTTGTACAAAAAAGCAGGCTTCACCAT-3’ and Reverse-5’- GGGGACCACTTTGTACAAGAAAGCTGGGTCCTATCAGTAGGGCTTACTATCATTCTTGGATGGC TGCGTTTAAGTTGGATT-3’ . The PCR product was used in BP recombination reaction with attp containing pDONR221 vector ( Thermo Fisher Scientific , Waltham , MA ) to create the entry vector . Next , using LR recombination reaction , the Celf1 gene from BP clone was moved into a destination vector , pDEST47 ( Thermo Fisher Scientific , Waltham , MA ) and confirmed by Sanger Sequencing . Celf1 over-expression plasmid was transfected into NIH3T3 cells for up to 72 hours and cells were assayed for Celf1 elevated levels by Western blotting . For over- expression assays , both Celf1 over-expression plasmid and the dual luciferase-Dnase2b 3’UTR plasmid were transiently transfected into NIH3T3 cells for 48 hours . Cells were collected , total RNA was isolated and cDNA were synthesized , and RT-qPCRs were performed as described above using the following primers to amplify the luciferase product: Firefly Luc Forward-5’-GCCCCAGCTAACGACATCTA-3’ , Firefly Luc Reverse-5’- TCTTTTGCAGCCCTTTCTTG-3’ . All experiments were performed in three biological replicates unless stated otherwise . Statistical significance for RT-qPCR data was determined using nested ANOVA as previously described [18] . Statistical significance for the fluorescence intensity measurement and luciferase assays was determined by two-tailed student t-test . | Besides associated with aging , the eye disease cataract can occur early in life because of defects in lens development . Lens fiber cells degrade their nuclei to achieve lens transparency , which poses a fundamental question: how is differentiation regulated in a cell progressing toward compromised transcriptional potential ? We demonstrate that this is achieved by distinct post-transcriptional gene expression control mechanisms mediated by a conserved RNA-binding protein Celf1 . Celf1 regulates key factors required for normal nuclear degradation and cell morphology of fiber cells by controlling abundance of target mRNAs and/or their translation into protein . These data identify new Celf1-regulated molecular mechanisms necessary for lens transparency and suggest that conserved post-transcriptional regulatory RNA-binding proteins have evolved to control eye development in vertebrates . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"fish",
"engineering",
"and",
"technology",
"lens",
"disorders",
"ocular",
"anatomy",
"messenger",
"rna",
"vertebrates",
"animals",
"cell",
"differentiation",
"animal",
"models",
"osteichthyes",
"developmental",
"biology",
"model",
"organisms",
"experimental",
"organism",
"systems",
"optical",
"equipment",
"eyes",
"research",
"and",
"analysis",
"methods",
"optical",
"lenses",
"mouse",
"models",
"head",
"biochemistry",
"zebrafish",
"rna",
"eukaryota",
"anatomy",
"nucleic",
"acids",
"equipment",
"cataracts",
"ophthalmology",
"lens",
"(anatomy)",
"ocular",
"system",
"biology",
"and",
"life",
"sciences",
"organisms"
]
| 2018 | The RNA-binding protein Celf1 post-transcriptionally regulates p27Kip1 and Dnase2b to control fiber cell nuclear degradation in lens development |
Chikungunya virus ( CHIKV ) has emerged as one of the most important arboviruses of public health significance in the past decade . The virus is mainly maintained through human-mosquito-human cycle . Other routes of transmission and the mechanism of maintenance of the virus in nature are not clearly known . Vertical transmission may be a mechanism of sustaining the virus during inter-epidemic periods . Laboratory experiments were conducted to determine whether Aedes aegypti , a principal vector , is capable of vertically transmitting CHIKV or not . Female Ae . aegypti were orally infected with a novel ECSA genotype of CHIKV in the 2nd gonotrophic cycle . On day 10 post infection , a non-infectious blood meal was provided to obtain another cycle of eggs . Larvae and adults developed from the eggs obtained following both infectious and non-infectious blood meal were tested for the presence of CHIKV specific RNA through real time RT-PCR . The results revealed that the larvae and adults developed from eggs derived from the infectious blood meal ( 2nd gonotrophic cycle ) were negative for CHIKV RNA . However , the larvae and adults developed after subsequent non-infectious blood meal ( 3rd gonotrophic cycle ) were positive with minimum filial infection rates of 28 . 2 ( 1∶35 . 5 ) and 20 . 2 ( 1∶49 . 5 ) respectively . This study is the first to confirm experimental vertical transmission of emerging novel ECSA genotype of CHIKV in Ae . aegypti from India , indicating the possibilities of occurrence of this phenomenon in nature . This evidence may have important consequence for survival of CHIKV during adverse climatic conditions and inter-epidemic periods .
Chikungunya virus ( CHIKV ) ( genus Alphavirus , family Togaviridae ) is a mosquito-borne pathogen , native to Africa that is transmitted between non human primates , mainly by forest dwelling Aedes species . The virus is also widespread as an urban infection throughout the old world tropics and subtropics , transmitted by two species of mosquito- Aedes aegypti and Ae . albopictus , both closely associated with the human peridomestic environment [1] . In Asia , the Ae . aegypti mosquitoes are primarily responsible for the maintenance of urban cycle , while in Africa , CHIKV transmission involves a sylvatic cycle , primarily with Ae . furcifer and Ae . africanus mosquitoes [2] . Autochthonous cases have also occurred in Europe , most notably in 2007 in an epidemic in northeast Italy that affected nearly 300 people [3] . In this case the vector was Ae . albopictus , an invasive species that is rapidly expanding its distribution in Europe and is already present in at least 27 countries . In humans , chikungunya fever is a self-limiting illness . In acute phase , it may involve some or all of the following: sudden onset of fever , headache , fatigue , nausea , vomiting , rash , myalgia and severe polyarthralgia . Symptoms may last up to 10 days , but crippling arthralgia can persist for months , even years in some patients [4] . CHIKV genome consists of 5′ capped positive sense single-strand RNA of ∼11 . 8 kb that harbors a poly ( A ) tail in its 3′ end . The genome is composed of two open reading frames ( ORFs ) embedded between non-translated regions ( 5′ NTR and 3′ NTR ) . The ORF located at the 5′ end of the genome encodes a polyprotein precursor of nonstructural proteins ( nsP1 , nsP2 , nsP3 , nsP4 ) with replicative and proteolytic activities . The second ORF encodes the polyprotein precursor of the structural proteins ( C , E1 , E2 ) [5] . Vertical transmission is the passage of virus between generations via the egg stage . Virus that infects the ovaries must persist through the larval instars , survive histolysis in the pupal instar and continue through to the adult stage [6] . Vertical transmission is considered to be a primary means by which some arboviruses are maintained during adverse environmental conditions . During this period , the arthropod hosts are either inactive or unable to survive , thus acting as a mechanism for virus persistence in environments where amplifying hosts are temporarily absent or immune . Because the Aedes eggs are desiccation resistant , these can survive for longer durations , leading to the possibility of persistence of CHIKV in the eggs [7] . The mechanisms responsible for prevalence of CHIKV during unfavourable periods , especially during winter seasons are unknown . So , vertical transmission is considered as an unresolved issue that has important bearing on the persistence of virus in periods when horizontal transmission is low or non-existent . Low rates of vertical transmission of the three main groups of mosquito-borne arboviruses- flaviviruses , alphaviruses , and bunyaviruses have been demonstrated in the field [8]–[10] . The existence of vertical transmission has also been demonstrated experimentally in these three groups . [11]–[14] . Among the alphaviruses , Ross River virus , Sindbis virus , western equine encephalomyelitis virus , and CHIKV have been isolated from adult Aedes species reared from larvae collected from natural habitats , confirming existence of natural vertical transmission [15]–[20] . Lindsay and coworkers isolated Ross River virus from wild-caught male Aedes mosquitoes , further providing evidence of natural vertical transmission [21] . However , to the best of our knowledge , there is no evidence of experimental vertical transmission among alphaviruses other than some conflicting reports in CHIKV [13] , [22] , [23] . In reviewing the literature on laboratory infections we noted that in nearly all studies , the infective blood meal was given to nulliparous mosquitoes [11] , [12] and detection of virus was limited to the offspring of the first gonotrophic cycle , whereas , when studies continued through subsequent cycles , rates of vertical transmission were found much higher [11] , [24] , [25] . We speculated that the difference could be attributable to two factors: ( i ) the first batch of eggs is laid several days before virus has begun rapid replication after passing via the gut wall into the hemolymph , and/or ( ii ) the enormous increase in the volume of the ovaries during oogenesis might increase its permeability to virus . We explored these possibilities by investigating the occurrence of vertical transmission in experimentally infected Ae . aegypti , the principal vector of CHIKV .
CHIKV belonging to novel Indian Ocean lineage ( IOL ) of ECSA genotype , obtained from an epidemic in India in 2006 ( DRDE06 ) ( GenBank Accession No . EF210157 ) . Initially it was isolated in BHK-21 cells and subsequently passaged in C6/36 cells . The virus is maintained at Virology Division , DRDE , Gwalior . In the present study , it was used at passage level 10 in C6/36 cells . Titre was found to be 108 PFU/ml through plaque assay [26] in Vero cells ( African green monkey kidney cells ) . The virus was aliquoted and stored in −80°C until use . Ae . aegypti used in this study , were collected from Gwalior district , Madhya Pradesh , India in 2005 and maintained as laboratory colony in Vector Management Division , Defence Research and Development Establishment ( DRDE ) , Gwalior at 27±2°C with 70% relative humidity and 14∶10 light∶dark photo period . Adult mosquitoes were provided with 10% sucrose solution soaked in cotton pads and larvae were provided with 2–3 yeast tablets per day in a pan containing tap water . The female Ae . aegypti mosquitoes were provided with three serial blood meals during this experiment to investigate vertical transmission . Schematic representation of the experimental design is shown in Fig . 1 . At day 14 post infection , 10 Ae . aegypti parental females ( after laying eggs from 3rd gonotrophic cycle ) were randomly selected and their midgut , legs & wings were tested for the presence of CHIKV to analyze infection and dissemination status . The eggs obtained after infectious blood meal and second non-infectious blood meal were reared under standard laboratory conditions i . e . 27±2°C at 70% relative humidity . The 4th instar larvae and 4–5 days old adults were screened for the presence of virus . The larvae and adults were processed in pools ( ≤20/pool ) . Each pool of adults and larvae were homogenized in 2 ml tubes with 800 µl of Eagles Minimum Essential Medium ( EMEM ) ( Sigma , St . Louis , USA ) and stainless steel beads in Tissuelyser LT ( Qiagen , Hilden , Germany ) . The homogenate was clarified by centrifugation at 4500× g for 10 minutes . 140 µl of supernatant was used to extract RNA using QIAamp viral RNA mini kit ( Qiagen , Hilden , Germany ) according to the manufacturer's protocol . The RNA was finally eluted in 50 µL elution buffer and stored at −80°C until use . CHIKV specific SYBR Green I based one step real time quantitative RT-PCR targeting to E1 gene was performed to screen the presence of CHIKV RNA in test samples [27] . Briefly , the quantitative RT-PCR was carried out using SS III Platinum one step qRT-PCR kit ( Invitrogen , Carlsbad , USA ) in Mx3005P system ( Stratagene , La Jolla , USA ) . Samples were assayed in a 25 µL reaction volume containing 12 . 5 µL of 2× master mix , 0 . 125 µL ( 0 . 25 µmol ) each of forward and reverse primer , 0 . 25 µL of enzyme mix comprising of Taq DNA polymerase and MMLV Reverse transcriptase , 9 . 5 µL of nuclease free water and 2 . 5 µL of RNA . The thermal profile comprised of 30 min of reverse transcription at 50°C , 10 min of polymerase activation at 95°C , followed by 40 cycles of PCR at 95°C for 30 s , 55°C for 60 s , and 72°C for 30 s . Following amplification , a melting curve analysis was performed with the melting curve analysis software of the Mx3005P according to the instructions of manufacturer . Positive and negative template control was also included along side in all experiments . CHIKV RNA mean titre in larvae and adults were analyzed . Comparison between two groups was performed by 2 tailed unpaired t-test using GraphPad Prism software ( Version 6 . 04 ) .
CHIKV RNA was detected in body , legs & wings of all the 10 randomly selected female Ae . aegypti that were provided with an infectious blood meal . This indicates 100% infection and dissemination of virus at day 14 post infection . The mean titre of CHIKV RNA in midgut , legs & wings of female Ae . aegypti was found to be 106 . 4±101 . 4 and 106 . 0±101 . 0 per reaction respectively . The total number of larvae and adults obtained after infectious blood meal were 230 and 485 respectively . All these larvae and adults were found to be negative for CHIKV . The total number of larvae and adults obtained after second non infectious blood meal were 284 and 693 respectively . These were divided and processed in pools ( ≤20/pool ) . Out of a total 30 pools of larvae , 8 pools were found positive . Out of 33 pools of adults , 14 were found positive . Minimum infection Rate ( MIR ) was calculated by the following formula: No . of positive pools/No . of individuals tested X 1000 [25] . The minimum infection rates achieved for larvae and adults were 28 . 2 and 20 . 2 respectively . MIR is also expressed as ratio i . e . Proportion of positive mosquitoes by uninfected mosquitoes . Thus the ratio was found to be 1∶35 . 5 for larvae and 1∶49 . 5 for adults [25] ( Table 1 ) . The results indicated above are cumulative of three experiments . Because the feeding efficiency was low and fewer mosquitoes underwent oviposition with successive gonotrophic cycles , fewer larvae and adults were obtained . The CHIKV RNA titre as determined by real time RT-PCR in larvae was 102 . 8 to 103 . 2 ( 103 . 0±100 . 1 ) . The CHIKV RNA titre in adults was 102 . 8 to 104 . 5 ( 103 . 8±100 . 6 ) ( Fig . 2 ) . Significant difference was observed between the two groups ( p = 0 . 0012 ) with t = 3 . 788 and df = 20 . The result indicated the amplification of virus following transition from larvae to adult stage .
The identification of vertical transmission of CHIKV in both natural and experimental settings , confirms the existence of this transmission pattern . The present study indicated that vertical transmission is a more common phenomena in mosquitoes during subsequent gonotrophic cycles following an arboviral infection . In view of the desiccation resistance nature of Aedes eggs , vertical transmission is likely to facilitate the persistent survival of virus during unfavourable inter-epidemic periods . This survival of virus has immense epidemiological implication further enhancing the risk of potential future outbreaks . | Although vertical transmission of arboviruses has been recognized for nearly a century , rates of transmission in laboratory experiments are low and their significance in terms of survival of virus during periods of low transmission appears debatable . Recently , major urban outbreaks of chikungunya have been recorded in many parts of Asia , Africa , and Europe . The occurrence of random sporadic cases of the disease in years following a major outbreak prompted us to investigate whether these might be attributable to survival of the virus by vertical transmission . Our experiments were designed to test two hypotheses: ( 1 ) The development of an egg-batch derived from an infectious blood meal is too rapid for the infection to reach ovaries; ( 2 ) The enormous distension of the membrane enveloping ovaries and ovarioles following oviposition , might facilitate virus penetration . We conclude that after the infected blood meal , oogenesis and oviposition were complete before virus had disseminated to infect the ovaries . Because similar experiments with infection in first gonotrophic cycle did not lead to infected progenies , it is presumed that expanded parous ovaries might support efficient infection . Therefore , it may be concluded that vertical transmission is a more common phenomena in mosquitoes during subsequent gonotrophic cycles following arboviral infection . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"virology",
"biology",
"and",
"life",
"sciences",
"microbiology",
"viral",
"diseases"
]
| 2014 | Evidence of Experimental Vertical Transmission of Emerging Novel ECSA Genotype of Chikungunya Virus in Aedes aegypti |
Ticks are considered the second vector of human and animal diseases after mosquitoes . Therefore , identification of ticks and associated pathogens is an important step in the management of these vectors . In recent years , Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry ( MALDI-TOF MS ) has been reported as a promising method for the identification of arthropods including ticks . The objective of this study was to improve the conditions for the preparation of tick samples for their identification by MALDI-TOF MS from field-collected ethanol-stored Malian samples and to evaluate the capacity of this technology to distinguish infected and uninfected ticks . A total of 1 , 333 ticks were collected from mammals in three distinct sites from Mali . Morphological identification allowed classification of ticks into 6 species including Amblyomma variegatum , Hyalomma truncatum , Hyalomma marginatum rufipes , Rhipicephalus ( Boophilus ) microplus , Rhipicephalus evertsi evertsi and Rhipicephalus sanguineus sl . Among those , 471 ticks were randomly selected for molecular and proteomic analyses . Tick legs submitted to MALDI-TOF MS revealed a concordant morpho/molecular identification of 99 . 6% . The inclusion in our MALDI-TOF MS arthropod database of MS reference spectra from ethanol-preserved tick leg specimens was required to obtain reliable identification . When tested by molecular tools , 76 . 6% , 37 . 6% , 20 . 8% and 1 . 1% of the specimens tested were positive for Rickettsia spp . , Coxiella burnetii , Anaplasmataceae and Borrelia spp . , respectively . These results support the fact that MALDI-TOF is a reliable tool for the identification of ticks conserved in alcohol and enhances knowledge about the diversity of tick species and pathogens transmitted by ticks circulating in Mali .
Ticks are bloodsucking arthropods that parasitize most of the vertebrates in the world and occasionally bite humans [1] . About 900 tick species have been identified and classified worldwide [2] . In Africa , the number of tick species indexed is 223 , including 180 hard and 43 soft ticks [2] . Currently , ticks are considered the second most important vector of human disease after mosquitoes and can transmit bacterial [1] , viral [3] and protozoan pathogens [4] . A significant number of these pathogens are of exceptional importance , as they are responsible for high morbidity and mortality in humans and animals [1] . Identification of tick species is an important step in epidemiological studies , in order to establish tick species distribution maps and to characterize tick fauna and seasonal trends [5 , 6] . In Mali , a West African country , livestock farming is an essential economical factor . At present , there are few studies on tick species that infest cattle or tick-borne diseases transmitted in Mali . To date , 23 tick species belonging to six genera have been categorized in Mali [7–9] . Among them , Amblyomma ( Am . ) variegatum , Rhipicephalus ( Rh . ) spp . and Hyalomma ( Hy . ) spp . are the main ticks monitored by Malian veterinarians for their effects on livestock healthcare and productivity [10] . Other public health problems , such as tuberculosis , AIDS or malaria , take precedence over tick-borne diseases ( TBDs ) , which are little explored by medical doctors . Several bacteria were detected in ticks from Mali . Spotted fever group rickettsiae were detected , including Rickettsia africae in Am . variegatum , R . aeschlimannii in Hy . marginatum rufipes , and R . massiliae in Rhipicephalus spp . , all three being human pathogens [11] . An Ehrlichia sp . of unknown pathogenicity , Ehrlichia Erm58 , was detected in Rh . mushamae [11] . More recently , Borrelia theileri , the agent of bovine and equine borreliosis , and B . crocidurae , agents of relapsing fever in humans , have been detected in Rh . geigyi and Ornithodoros sonrai , respectively [12–14] . To study and control ticks and TBD transmission , accurate identification of tick species and determination of their infectious status are essential [1] . Currently , tick identification is principally conducted by observing morphological characteristics . However , it is limited by entomological expertise , dichotomous keys availability , tick integrity or engorged status [9] . Molecular tools have been used as an alternative to overcome the limitations of morphological identification [15] . Sequencing of several genes has been used , including ribosomal sub-units ( e . g . , 12S , 16S or 18S ) , the cytochrome c oxidase unit I ( COI ) , or the internal transcribed spacer [16] . These techniques are generally time-consuming , laborious and can be expensive , preventing their use in large scale studies [17–20] . Moreover , the absence of a consensus gene target sequence for tick identification and/or the comprehensiveness of genomic databases are additional factors hampering their use [16] . Recently an alternative tool based on the analysis of protein profiles resulting from matrix-assisted laser desorption/ionization time-of-flight mass spectrometry ( MALDI-TOF MS ) analysis has been explored to identify arthropods [21] . MALDI-TOF MS has been used to identify tick species [22–24] and to determine tick infectious status [25–27] . However , tick collection usually takes place far from analytical laboratories and therefore requires proper storage of samples . Ticks are generally stored either alive , at -20°C , or in alcohol . Although alcohol storage is cheaper and easier , especially in African countries , previous studies reported that the use of fresh ( i . e . , recently dead ) or frozen specimens led to more reproducible and better MS spectra compared to the alcohol preservation mode for ticks [24] [28] , and also for other arthropod families[29 , 30] . In a recent study , it was demonstrated that long-term tick storage in alcohol altered MS profiles , which did not provide conclusive identification following in-house MS reference spectra database-querying containing MS spectra from counterpart fresh tick species . Nevertheless , the upgrading of the in-house MS reference spectra database of specimens stored in alcohol allowed correct identification of ticks at the species level , also underlining the reproducibility and specificity of MS profiles for tick specimens stored in alcohol [31] . The goal of the present work was to determine tick population diversity and associated pathogens from alcohol stored specimens collected on cattle in Mali by using MALDI-TOF MS and molecular approaches with specimens collected in the field . First , optimized sample preparation conditions for ticks stored in ethanol for MALDI-TOF MS analysis were established . Second , based on morphological and molecular identification of ticks , an MS reference spectra database was created and tested blindly using new tick specimens . In addition , tick-associated bacterial pathogens were screened by molecular biology on half-tick body parts and leg MS spectra from ticks mono-infected or not by bacterial pathogenic agents , and they were compared to assess the efficiency of this proteomic tool for classification of ticks according to their infectious status .
Tick collection protocols were developed as of a large study under the GIRAFE programme , UMI 3189 and MSHP-MRTC HFV project . The protocols were cleared by the FMPOS IRB in 2015 and 2016 . Verbal informed consent was obtained from managers of the livestock selected for tick sampling directly on mammals . The collection of ticks on domestic animals did not involve national parks or other protected areas or endangered or protected species . Ticks were collected from three localities in Mali , including Bamako , Kollé and Bougoula Hameau , in September 2015 and August 2016 ( Fig 1 ) . Bamako , the capital city of Mali , is an urban area surrounded by hills . The climate is Sahelian-type with two distinct seasons , the dry season ( i . e . , from November to May ) and the rainy season ( i . e . , from June to October ) . The total amount of precipitation was less than 900 milliliters in 2009 . Kollé is a rural village located about 60 km southwest of the capital . Agriculture , livestock farming and small businesses are the main economic activities of the village . The village , located on a flat land with submersible and dry areas , presents a Sahelian-type climate with two distinct seasons , a rainy ( i . e . , from June to November with maximum rainfall in August-September of 350 to 400 milliliters ) and a dry season ( i . e . , from December to May with a cool period in December- February and a warm period in March-May ) . The third site was Bougoula Hameau , a suburban village , located at 4 km of Sikasso town and it was situated at 374 km southeast of Bamako by road . The climate is of Sudanese type , under the influence of the humid forest with a rainy ( i . e . , from May to October ) and a dry season ( i . e . , from November to April ) . The annual rainfall can vary from 1 , 200 to 1 , 800 milliliters , depending on the year . These climatic conditions are appropriate for agricultural and livestock farming . Ticks were collected from domestic animals and cattle . Examination of all body parts was conducted from the tail to the head of the animal to detect ticks on the skin . All ticks ( engorged and non-engorged ) were collected manually with forceps . The ticks of the same animal were counted , pooled in the same tube and stored at room temperature in 70% v/v ethanol ( ticks collected in September 2015 ) or frozen at -20°C ( ticks collected in August 2016 ) until morphological , molecular and MALDI-TOF MS analyses . Ticks were transferred from MRTC ( Bamako , Mali ) to the URMITE laboratories ( Marseille , France ) for analysis . Ticks were identified morphologically to the species level firstly by a PhD student and then checked by expert tick entomologists using previously established taxonomic identification keys [9] . Tick identification and gender determination were performed under microscope at a magnification of ×56 ( Zeiss Axio Zoom . V16 , Zeiss , Marly le Roi , France ) . The tick genera , species , gender , host and animal number , collection site and date were codified to include this information on the tube . Each tick was dissected with a new sterile surgical blade to remove the legs , which were used for MALDI-TOF MS analyses . The rest of the tick was longitudinally cut in two equal parts . The half part with legs cut off was immediately used for molecular biology , and the second half was stored frozen as a backup sample for any additional analysis . Each half-tick without legs was transferred to a 1 . 5 mL tube containing 180 μL of G2 lysis buffer and 20 μL proteinase K ( Qiagen , Hilden , Germany ) , and incubated at 56°C overnight . DNA extraction from the half-tick was performed with an EZ1 DNA Tissue Kit ( Qiagen ) according to manufacturer recommendations . The DNA from each sample was eluted with 100 μL of Tris-EDTA ( TE ) buffer ( Qiagen ) and was either immediately used or stored at -20°C until use . Standard PCR , using an automated DNA thermal cycler amplifying a 405-base pair fragment of the mitochondrial 12S RNA gene ( Table 1 ) , was used for tick identification to the species level , as described previously [31] . The 16S RNA gene was used to confirm all Rhipicepalus ( Boophilus ) microplus identification . DNA from Am . variegatum specimens reared at the laboratory was used as positive control . PCR products of the positive samples were purified and sequenced as described previously [31] . The sequences were assembled and analyzed using the ChromasPro software ( version 1 . 34 ) ( Technelysium Pty . Ltd . , Tewantin , Australia ) , and were then blasted against GenBank ( http://blast . ncbi . nlm . nih . gov ) . Quantitative PCR was performed according to the manufacturer's protocol using a PCR detection system; a CFX Connect™ Real-Time ( Bio-Rad ) with the Eurogentec Takyon qPCR kit ( Takyon , Eurogentec , Belgium ) . The qPCR reaction contained 10 μl of Takyon Master Mix ( Takyon , Eurogentec , Belgium ) , 3 . 5 μl sterile distilled water , 0 . 5 μl of each of the primers and probe and 5 μl of the DNA extract . A total of 471 samples were screened using primers and probes , targeting specific sequences of the following bacterial pathogens: Rickettsia spp . , Anaplasmataceae spp . , Borrelia spp . , Bartonella spp . and Coxiella burnetii ( Table 1 ) . For Borrelia spp we used 2 genes , the 16S Borrelia gene first and all the ticks that were positive for this gene were retested by ITS4 for confirmation . Only samples positive for both genes ( 16S borrelia and ITS4 ) were considered positive . Positive samples for Rickettsia spp . were then submitted to a qPCR system specific for detecting R . africae [32] . Negative samples for R . africae but positive for Rickettsia spp . were submitted to gltA gene sequencing to determine Rickettsia species [33] . All ticks positive either for Anaplasmataceae spp . were submitted to amplification using standard PCR and sequencing to identify the bacteria species [34 , 35] . Ticks that were positive for Borrelia spp for both the 16S Borrelia gene and ITS4 were submitted to amplification using standard PCR and sequencing [33] . PCR tests were considered positive when the cycle threshold ( Ct ) was lower than 36 [36] . The DNA from Rickettsia montanensis , Bartonella elizabethae , Anaplasma phagocytophilum , Coxiella burnetii and Borrelia crocidurae was used as positive controls and mix as negative controls in PCR , respectively . All these bacteria come from the strains of culture of our laboratory and Borrelia crocidurae was cultured in Barbour-Stoenner-Kelly ( BSK-H ) liquid medium supplemented with rabbit serum . Only samples considered as negative ( i . e . , Ct ≥ 36 for all bacteria tested ) , were submitted to 12 S tick gene amplification to control the correctness of DNA extraction . The homogenized tick legs were centrifuged at 2000 g for 30 seconds and 1 μL of the supernatant from each sample was carefully dropped onto the MALDI-TOF target plate as previously described [28] . Each spot was then recovered with 1 μL of CHCA matrix solution composed of saturated α-cyano-4-hydroxycynnamic acid ( Sigma , Lyon . France ) , 50% acetonitrile ( v/v ) , 2 . 5% trifluoroacetic acid ( v/v ) ( Aldrich , Dorset , UK ) and HPLC-grade water [16] . The target plate , after drying for several minutes at room temperature , was introduced into the Microflex LT MALDI-TOF Mass Spectrometer device ( Bruker Daltonics , Germany ) for analysis . The loading of the MS target plate , the matrix quality , and the performance of the MALDI-TOF were performed as previously described [28] . Protein mass profiles were obtained using a Microflex LT MALDI-TOF mass spectrometer ( Bruker Daltonics , Germany ) using parameters previously described [31] . The spectrum profiles obtained were visualized with Flex analysis v . 3 . 3 software and exported to ClinProTools software v . 2 . 2 and MALDI-Biotyper v . 3 . 0 . ( Bruker Daltonics , Germany ) for analysis [25] . The reproducibility of spectra was evaluated by analyzing ten Am . variegatum specimens from Kollé per sample preparation protocol as previously described [29] . The selected protocol was analyzed using an unsupervised statistical test classifying specimens according to MS spectra ( i . e . , Principal Component Analysis , PCA test , ClinProTools v2 . 2 software ) . The Composite Correlation Index ( CCI ) tool ( MALDI-Biotyper v3 . 0 . software , Bruker Daltonics ) , was used to assess spectra variations within each sample group according to protocol tested , as previously described [37] . CCI was computed using the standard settings of mass range 3000–12000 Da , resolution 4 , 8 intervals and autocorrelation off . Higher the log score values ( LSVs ) and correlation values ( expressed by mean ± standard deviation [SD] ) reflect higher reproducibility of MS spectra and were used to determine the best protocol for sample preparation . Based on the correlation of morphological and molecular results of tick identification , two to five specimens per species were used to assess MS spectra reproducibility from specimens of the same tick species , and the MS spectra specificity from specimens of distinct tick species . These analyses were performed with the average spectral profiles ( MSP , Main Spectrum Profile ) obtained from the four spots of each individual tick specimen using Flex analysis v . 3 . 3 and ClinProTools 2 . 2 softwares ( Bruker Daltonics ) . Tick species exhibiting reproducible and specific MS spectra were then included in-house MS spectra reference database . To upgrade the database , MSP reference spectra were created using spectra from at least 2 specimens per species of both genders by the automated function of the MALDI-Biotyper software v3 . 0 . ( Bruker Daltonics ) . MS spectra were created with an unbiased algorithm using information on the peak position , intensity and frequency [38] . The spectra files are available on request and transferable to any Bruker MALDI-TOF device . A blind test was performed with new tick specimens collected in Mali stored in 70% alcohol or frozen . A total of 451 MS spectra from tick legs , including 340 stored in alcohol and 111 frozen specimens were tested successively against the in-house MS reference spectra database ( Database 1 ) and its upgraded version , which includes the 20 MS spectra from specimens of the 6 tick species collected in Mali and alcohol-preserved ( Database 2 ) . Among the 451 ticks tested 51 Am . variegatum and 23 Rh ( B ) microplus were fully engorged . Database 1 was composed of specimens of fresh or frozen arthropods ( Table 2 ) [16 , 24 , 30 , 31 , 39] . Database 2 includes database 1 plus MS spectra of tick legs from 6 species stored in ethanol from the present study ( Tables 3 and 4 ) . The reliability of species identification was estimated using the LSVs obtained from the MALDI-Biotyper software v . 3 . 3 , which ranged from 0 to 3 . These LSVs correspond to the degree of similarity between the MS reference spectra database and those submitted by blind tests . An LSV was obtained for each spectrum of the samples tested . Moreover , to decipher incoherent results obtained between morphological and MS identification , molecular identification of ticks was performed for the respective specimens . These comparative analyses to determine the infectious status of ticks were made by ClinProTools v . 2 . 2 software ( Bruker Daltonics , Germany ) . Only tick leg MS spectra from species with at least five mono-infected or pathogen-free specimens were included in this analysis . The spectra of 30 specimens of A . variegatum infected by R . africae were compared to those of 12 uninfected specimens from the same species . Moreover , MS spectra of 36 uninfected specimens of Hy . truncatum were also compared with the spectra of 23 specimens of Hy . truncatum infected by C . burnetii .
A total of 1 , 333 ticks were collected from the three sites including 406 engorged ( Fig 1 ) . A total of 1 , 217 were found on 44 bovine specimens and 116 on 9 dogs . Nineteen engorged females of the Hyalomma genus ( 1 . 55% of ticks collected ) were not morphologically identified to the species level . Morphologically , six distinct tick species belonging to three genera were identified among ticks collected in September 2015 ( Table 3 ) . Am . variegatum ( n = 877 , 71 . 79% ) was the overall predominant tick species collected from different sites , followed by two species of the Hyalomma genus , Hy . truncatum ( n = 260 , 21 . 27% ) and Hy . m . rufipes ( n = 28 , 2 . 29% ) . The three other tick species , Rh . ( Bo . ) microplus , Rh . e . evertsi and Rh . sanguineus sensus lato , represented less than 3 . 10% ( n = 38 ) . The five Rh . sanguineus sl [40] specimens were all collected on a dog . All 111 ticks collected in August 2016 were identified as Rh . sanguineus sl . Three hundred sixteen of the 1 , 222 ticks collected from three sites in 2015 and 111 ticks in 2016 had specimens of six species randomly selected for molecular and proteomic analyses ( Table 3 ) . A total of 20 specimens , including 2 to 5 specimens per species and all specimens of Rh ( Bo . ) microplus , were randomly selected for molecular analysis . A GenBank query revealed that 12S gene sequences were available for the 6 tick species . BLAST analysis indicated high identity ( i . e . , a range from 99% to 100% ) of 12S rRNA gene sequences among specimens classified per species according to morphological identification ( Table 4 ) . BLAST analysis revealed that these 6 tick species and all specimen of Rh ( B ) microplus had high sequence identity with their respective homolog species available in GenBank ( i . e . , range 96 . 5% to 100%; Table 4 ) . Among the ticks tested , 41 . 8% ( 197/471 ) were negative for the six bacteria tested , 37 . 4% ( 176/471 ) were positive for one bacterium and 20 . 8% ( 98/471 ) were found co-infected by two or three of the screened bacteria . Among the 274 specimens found positive for at least one bacteria tested , 76 . 6% ( 210/274 ) were infected by Rickettsia spp . , among which R . africae was found in 87 . 6% ( 184/210 ) ( Table 4 ) . The amplification of the ompA fragment in the remaining ticks positive for Rickettsia spp . and negative for R . africae ( n = 26 ) was used for identification of these Rickettsia spp . R . aeschlimannii and R . mongolitimonae were detected in 24 and 2 tick specimens , respectively ( Table 5 ) . Screening of all ticks for Coxiella burnetii revealed that 37 . 6% ( 103/274 ) of the specimens were positive ( Table 4 ) . Fifty-seven ticks , 20 . 8% ( 57/274 ) were positive in qPCR targeting the 23S rRNA of Anaplasmataceae . Among them , 23S rRNA amplification and sequencing was successful for 50 samples . The BLAST found broad agreement that 43 ticks were positive for E . ruminantium ( GenBank accession number NR 077002 . 1 ) , 2 ticks were positive for Ehrlichia sp . urmitei TCI148 ( GenBank ACCN KT 364334 . 1 ) and 1 tick for Ehrlichia sp . rustica TCI141 ( GenBank ACCN KT 364330 . 1 ) . A . marginale was detected in 3 ticks and A . sp . ivoriensis TCI50 ( GenBank ACCN KT 364336 . 1 ) in 1 specimen ( Table 5 ) . Borrelia spp . was detected in 1 . 1% ( 3/274 ) of ticks by qPCR . However , all standard PCR for determination of Borrelia species failed . No Bartonella spp . was detected in the ticks tested . A comparison of our current reference sample preparation method ( i . e . , “de-alcoholization” ) with the “dry” and “direct” methods was performed [31] . The best method was selected on the following criteria: reproducibility and intensity of MS spectra , low handling and simplicity of the protocol . To exclude inter-individual variability , protocols were successively compared by pairs , and then the four right legs were used for one protocol and the four left legs from the same tick for the other . Then , ten specimens of both genders tested per protocol , five males and five females , were included . For all these experiments , morphologically identified ticks from Kollé ( Am . variegatum ) were used . The first comparison concerned the “de-alcoholization” and “dry” protocols ( Fig 2A ) . The visual comparison of MS profiles between these two groups using the gel view tool and the superimposition of average MS profiles in each condition using ClinProTools software ( Bruker ) did not reveal differences in peak position between the two protocols ( Fig 3A and 3B ) . This reproducibility of the profiles was analyzed using an unsupervised statistical test classifying specimens according to MS spectra ( i . e . , Principal Component Analysis , PCA test , ClinProTools software ) . The mixing of both groups on the graphical representation confirmed the absence of differences between both groups ( Fig 3C ) . Thus , the “dry” protocol was preferred compared to the “de-alcoholization” protocol , the latter considered to be more time-consuming and fastidious . The second comparison concerned the “dry” and “direct” protocols , using ten Am . variegatum specimens from both genders ( Fig 2B ) . The comparison of MS profiles between these two groups , either by gel view , superimposition or PCA ( Fig 4A , 4B and 4C ) , could not determine the more relevant method . The Composite Correlation Index ( CCI ) tool revealed a higher CCI ( LSV mean±SD: 0 . 783±0 . 101; Fig 4D ) for the “dry” protocol compared to “direct” ( LSV mean±SD: 0 . 755±0 . 175; Fig 4D ) . These results were in agreement with the gel view showing a higher visual homogeneity of the MS spectra from the “dry” group . Finally , the “dry” protocol appeared consistently to be the more reproducible and low-handling procedure for the preparation of ethanol-stored ticks for MS analysis , and was chosen for the next experiments of the present study . Twenty ticks , including several specimens per species coming from distinct localities , were identified by sequencing 12S tick gene . Their non-infected status was also controlled for the microorganisms tested in the present work by q PCR . These specimens were selected for evaluating intra-species reproducibility and inter-species specificity of MS spectra . Comparison of the MS spectra with Flex analysis software indicated reproducibility of the MS profiles between tick specimens from the same species ( Fig 5A ) . Moreover , the visual comparison of MS profiles indicated a clear distinction of spectra according to species . To reinforce the specificity of MS profiles according to tick species , MS profiles from these 20 specimens were used to generate a dendrogram and PCA ( Fig 5B and 5C ) . Clustering analysis revealed a gathering on distinct branches of ticks according to species . However , at the genus level , all specimens from the Rhipicephalus genus were not clustered in the same part of the dendrogram . The profile of the spectra of specimens preserved in alcohol was different from those of fresh specimens of the same species; this difference was also observed between manual sample homogenization and automated sample crushing using the TissueLyser apparatus . To assess the efficacy of the in-house MS reference spectra database , named database 1 ( DB 1 ) to correctly identify tick specimens preserved in alcohol , half of the MS spectra from ticks included in the present study were randomly selected . Then , MS spectra from 178 specimens including 60 Am . variegatum , 64 Hy . truncatum , 16 Hy . m . rufipes , 26 Rh . ( Bo . ) microplus , 7 Rh . e . evertsi and 5 Rh . sanguineus sl were queried against the DB 1 spectra database . The blind test against DB 1 revealed correct identification for some specimens of Hy . truncatum ( n = 5 ) and Hy . m . rufipes ( n = 5 ) , with LSVs > 1 . 8 ( Table 6 ) . For the remaining ticks ( n = 168 ) , all LSVs were ˂ 1 . 8 [24] . Tick MS spectra from 20 specimens , including 6 species identified morphologically and molecularly in this work , were added to DB 1 , which was then renamed DB 2 ( Table 4 ) . Thereafter , the leg spectra of the 451 morphologically-identified ticks , including 340 stored in alcohol and 111 frozen , were queried against DB 2 . Among the 451 ticks tested 51 Am . variegatum and 23 Rh ( Bo . ) microplus were fully engorged . The results of this second interrogation ( blind test 2 , BT2 ) showed 96 . 7% ( 325/340 ) concordance between morphological identification and MALDI-TOF MS identification . The percentage of concordant identification with morphology was 100% for Rh ( Bo ) microplus , Rh . e . evertsi and Rh . sanguineus sl stored in alcohol , with LSVs ranging from 1 . 89 to 2 . 71 ( Table 6 ) . A total of 15 specimens presented divergent identification between morphological and MALDI-TOF MS identification . To eliminate any doubt , these 15 specimens were submitted to molecular identification . Sequencing of the 12S gene confirmed the identification obtained by MALDI-TOF MS for 13 specimens ( Table 6 ) . The remaining 2 specimens identified as Hy . m . rufipes by MALDI-TOF MS were finally classified as Hy . truncatum by molecular biology , confirming morphological identification . All fully engorged ticks were correctly identified by MALDI-TOF MS . The percentage of correct MALDI-TOF MS identification for all species was 99 . 6% ( 449/451 ) ( Table 6 ) . The comparison of MS profiles between 30 Am . variegatum uninfected and 12 infected by R . africae using the gel view tool and Principal Component Analysis by ClinProTools software ( Bruker ) , revealed no differences between the two groups ( S1 Fig ) . The same observation was made by comparing of MS profiles of 36 Hy . truncatum uninfected and 23 Hy . truncatum infected by C . burnetii ( S1C and S1D Fig ) .
MALDI-TOF MS has revolutionized clinical microbiology by its use in the routine identification of bacteria [41 , 42] and archaea [43] . Even if the MALDI-TOF MS device acquisition could be expensive , its use for entomological analyzes induces low additional costs because reagents used for this high-throughput technique are economical and data analyses are simple and rapid compared to morphological and molecular methods [44] . This fast , economical and accurate proteomic tool has since been applied to the identification of arthropods: culicoides biting midges [45] , mosquitoes[39 , 46 , 47] , phlebotomine sand flies [48 , 49] , fleas [30] and tsetse flies [50 , 51] . MALDI-TOF MS has also been proposed for identifying tick species which are laboratory-reared , collected in the field or on mammalian hosts , by analyzing whole specimens [22] or legs only [23 , 24] . More recently , preliminary studies have investigated the capacity of MALDI-TOF MS to differentiate ticks infected or not by Borrelia spp . or spotted fever group rickettsiae [25–27] , and to detect the Plasmodium in anopheles [44] . However , tick collection is usually far from the analytical laboratories , requiring proper storage of samples . Although the alcohol storage mode is cheaper and easier , especially in African countries , previous studies reported that the use of fresh ( i . e . , recently dead ) or frozen specimens led to more reproducible and better MS spectra , compared to the alcohol storage mode for ticks [24 , 28] , and also for other arthropod families [29 , 30] . Recently , the application of MALDI-TOF MS for identification of ticks collected in the field in East Africa and preserved in alcohol has allowed reliable identification [23] . More recently , the discriminatory power of MALDI MS-TOF for the correct identification of ixodid tick specimens collected in the field in Ethiopia , which were preserved in 70% ethanol for about two years , was reported [31] . In this study , the morphological identification of ticks revealed the presence of six species , including Am . variegatum , Hy . truncatum , Hy . m . rufipes , Rh . ( Bo ) microplus and Rh . e . evertsi that were collected from cattle and Rh . sanguineus sl from dogs . Rh . e . evertsi was found only in Bamako , while all other species of ticks were found on cattle in the three locations . In support of these morphological identification results , several studies have reported the presence of these tick species in Mali , except for Rh . ( Bo ) microplus [7–9] . Rh . ( Bo ) microplus , which is a southeast Asian tick , was introduced in the southeast of Africa ( South Africa , Zambia , Tanzania and Malawi ) by cattle from Madagascar [9] . It was reported in West Africa ( Ivory Coast ) for the first time in 2007 [52] . The presence of Rh . ( Bo ) microplus has only been found in three other countries of West Africa ( Mali , Benin and Burkina Faso ) [53] . Biguezoton et al ( 2016 ) and Boka et al ( 2017 ) found that Rh ( B ) microplus represent 70% and 63 . 2% of ticks in Burkina Faso and Benin and Ivory Coast respectively [54 , 55] . Our study confirms the presence of this species in 3 localities in Mali , which could indicate its rapid spread and its probable installation in Mali . As expected , Am . variegatum was the most prevalent species in the three sites of the present study [10] . To confirm the morphological identification of tick specimens that were used for creating the MALDI-TOF MS database , sequencing of the 12S rRNA gene was performed . The 12S rRNA gene was chosen to validate identification because this gene is known as a reliable tool for molecular identification of ixodid ticks [16] . The coverage percentages and identity between the sequences of specimens of the same species were from 99 to 100% for all species of ticks . Percentages of identity and coverage of sequences Am . variegatum , Hy . m . rufipes , Rh . ( Bo ) microplus , Rh . e . evertsi and Rh . sanguineus sl were 99–100% with sequences of the same species available in GenBank . Interestingly , lower sequence identities ( 96–97% ) of Hy . truncatum compared to the corresponding reference sequence in GenBank were observed . It could be hypothesized that the sequence differences could correspond to genetic variation within ticks of the same species adapted to different geographic regions of a country or countries , as previously described [56] . The difference between the sequences of 12S rRNA genes of Hy . truncatum collected in Mali and that available on GenBank tick collected in Zimbabwe [57] could explain these genetic variations . In the future , the sequencing of a second gene target , such as 16S or COI , could be performed to further study these variations [58] . In this study , DNA from Rickettsia spp . was detected in 76 . 6% of infected ticks collected from cattle , among which R . africae was found in 87 . 6% ( 184/210 ) . R . africae was detected in 92 . 2% of Am . variegatum , a cattle tick found throughout sub-Saharan Africa . Such high prevalence of R . africae in Am . variegatum has already been reported [59–61] . R . africae was also detected in Rhipicephalus spp . and Hyalomma spp . , respectively 7 . 9% and 9 . 2% . Other recent studies have detected R . africae in other tick genera , including Rhipicephalus and Hyalomma [59 , 62 , 63] . R . africae is the etiological agent of African tick-bite fever in humans ( ATBF ) [64] . R . aeschlimannii have been observed in Hyalomma spp . , with 9% and 52 . 6% respectively in Hy . truncatum and Hy . m . rufipes . These data are comparable with those of previous studies that reported 45% to 55% of Hy . m . rufipes and 6% to 7% of Hy . truncatum were DNA carriers of R . aeschlimannii in Senegal [63] , and 44% and 11% in Ivory Coast [59] . The sequences of R . aeschlimannii identified in our work were identical to those of R . aeschlimannii , previously detected in Hy . truncatum collected in Senegal ( GenBank accession number HM050276 . 1 ) . R . aeschlimannii is an agent of spotted fever in humans [64] . R . aeschlimannii is found in sub-Saharan Africa , North Africa , Europe and Asia [11 , 65] . Our results confirm a large prevalence of this pathogen in Mali . For the first time , the presence of R . mongolitimonae was identified in Hy . truncatum from Mali . It had been previously detected in Hy . truncatum from the countries bordering Mali , including Niger [11] and Senegal [63] . R . mongolitimonae 12S sequence of the present study were 99% identical with the same sequence fragment of a strain previously isolated from a patient from Algeria ( GenBank DQ097081 . 1 ) . Until now , two Borrelia species have been identified in Mali , B . crocidurae in the soft tick ( O . sonrai ) and B . theileri in the hard tick ( Rh . geigyi ) [12 , 13] . Our results show the presence of Borrelia spp . in 2 specimen of Am . variegatum and 1 of Hy . truncatum by qPCR using 16S Borrelia and ITS4 genes . Similarly , Ehounoud et al . previously reported the presence of Borrelia spp . in the same tick species in Ivory Coast [59] . Unfortunately , no PCR products using standard amplification were obtained for any of these ticks . This failing could be explained by the higher sensitivity of qPCR compared to standard PCR [66] . In the present work , C . burnetii , the agent of Q fever , was detected for the first time in ticks in Mali , with a prevalence of 33 . 4% in the six tick species identified . These results differ from those of Ehounoud et al . in Ivory Coast , who found only one tick infected with C . burnetii [59] . Q fever is a ubiquitous zoonotic disease caused by C . burnetii . It is poorly documented in Africa . A recent study conducted in febrile African patients found one male adult patient ( 0 . 3% ) infected with C . burnetii in Algeria and six patients ( 0 . 5% ) in Senegal [67] . However , in another study conducted in Senegal , C . burnetii was detected in humans as well as in ticks [68] . The Anaplasmataceae bacteria family was previously considered to be pathogens of veterinary importance [59] . However , in recent decades , many agents of this family have been described in humans [69] . Here , we reveal the presence of A . marginale in 11 . 5% of Rh ( Bo . ) microplus . This is the first demonstration of the presence of A . marginale , the agent of bovine anaplasmosis [70] in Mali . A . marginale is an intracellular bacterium responsible for bovine anaplasmosis which manifests with anemia and jaundice [64] . Also , E . ruminantium was found in Am . variegatum , Hy . truncatum , Rh ( Bo . ) microplus , and Rh . e . evertsi . The prevalence of E . ruminantium was 13 . 9% in ticks . Potential new species of Ehrlichia and Anaplasma ( E . sp urmitei TCI148 , E . sp rustica TCI141 and A . sp ivoriensis TCI50 ) have been detected in Rh ( Bo . ) microplus and Hy . truncatum . These bacteria had already been detected in ticks from Ivory Coast [59] . However , co-infections have been found in the ticks in this study . The percentage of co-infected ticks was 23 . 1% ( 109/471 ) , and we describe for the first time multiple co-infections in ticks in Mali . Recently , multiple co-infections in ticks have been reported in Ivory Coast; these co-infections systematically involved R . africae [59] . The percentage of ticks co-infected was higher in our study than that obtained in Ivory Coast [59] . To avoid bias , we choose to query the MS spectra of 178 specimens of ticks , including 6 species against DB 1 which includes several families of arthropods , including mosquitoes . We constantly improve it with new specimens collected in the field and find it more relevant to carry out a total interrogation without the knowledge without any filter on a specific family . The results of the blind test revealed correct identification in 10 specimens only with high log score values , even though this database contained the same tick species that were also preserved in alcohol . This misidentification could be attributed to several factors: ( i ) the method used for sample crushing ( initially manually , and here an automatic apparatus was used as previously described [28] , ( ii ) the difference in storage time ( 6 months here vs 3 years in the previous study ) , ( iii ) the geographical distance ( Mali vs . Ethiopia ) , which could have consequences on MS spectra profiles , as observed also at the genetic level . This last phenomenon had already been reported in other studies of sand flies [71] , mosquito immature stages [46] and ticks [31] . Conversely , when database 1 was upgraded with 20 spectra of the six tick species of our study , the blind test of all ticks revealed 95 . 60% ( 325/340 ) correct identification for tick species stored in alcohol . However , the remaining fifteen ticks ( 4 . 40% ) with inconsistent identification between morphological and MALDI-TOF MS tools were subjected to molecular biology to determine the real identification of these specimens . The molecular biology results confirmed those of MALDI-TOF MS for 13 of these specimens . Two Hy . truncatum specimens were misidentified by MALDI-TOF MS . The reasons for the misidentification of the two specimens remain unknown . Additionally , all ticks frozenly stored were correctly identified by the blind test . The results of this work show that MALDI-TOF MS is superior to morphological identification , as the correct identification percentage is 99 . 6% for all tested . It is also interesting because there are fewer entomologists able to identify ticks and the morphological identification keys are not always available . Another advantage of MALDI-TOF is that it can identify ticks that are completely engorged or damaged , for which morphological characteristics can be deformed or even disappear making morphological identification difficult or impossible . Conversely , the proteomic strategy proposed here , does not require specific skill or expertise , reagents are very cheap so the running cost is very low compared to a molecular biology . The current limiting factors of MALDI-TOF MS analysis are the small diversity of tick species included in the MS spectra reference database and the relative elevate cost to acquire the machine . Nevertheless , it high-throughput and large application for microorganisms identification either in research or medical diagnosis , do of this emerging tool a highly competitive method also for medical entomology studies . It is likely that MALDI-TOF MS will realize similar revolution in medical entomology as it was occurred in microbiology . Our results confirm those of previous studies , according to which MALDI-TOF MS could be used for identification of ticks preserved in alcohol , but it requires the creation of a database with specimens stored in the same condition [31] . In our work , MALDI-TOF MS analysis was not able to differentiate ticks which were infected or not by the bacteria that were screened . However , preliminary studies from our laboratory seemed promising , as MALDI TOF analysis allowed differentiation of ticks infected or not by Borrelia spp . or spotted fever group rickettsiae [26 , 27] . The failing of bacteria-pathogen detection by MALDI-TOF MS could be attributed to several factors . The storage mode , fresh versus alcohol , might play a role . Moreover , the infectious status of these ticks was controlled against some bacteria pathogens , however , it was possible that they were infected by others pathogens not researched in the present study , which could impaired the determination of specific MS profiles for each associated pathogens . These factors could alter MS spectra profiles between uninfected and infected ticks . More studies are needed to explore the capacities of MALDI TOF to detect tick infectious status . To conclude , the present work has confirmed that MALDI-TOF MS may represent a rapid and inexpensive alternative tool for accurate identification of ticks collected in the field and stored in alcohol . The recent demonstration of the use of MALDI-TOF MS for identification of ticks and associated pathogens requires further investigation . | Ticks are among the most important vectors and reservoirs of several animal and human pathogens such as viruses , bacteria and protozoa . However , very few studies have been done on ticks in Mali . At present , little information is available about tick species infesting livestock or human tick-borne diseases transmitted in Mali . The identification of tick species and the determination of pathogens associated are essential to evaluate epidemiology and risks of human and animal diseases: the One Health approach . Current identification methods are time consuming , expensive and laborious . Previous studies have shown that MALDI-TOF mass spectrometry analyses may allow accurate tick species identification . A recent study suggested that it was possible to identify ticks preserved in alcohol by MALDI-TOF MS . The aim of the present study was to improve tick leg sample preparation conditions for their identification by MALDI-TOF MS from Malian ethanol-preserved specimens collected in the field . This study provided 99 . 4% concordance between morphological and MALDI-TOF identification . The detection of microorganisms was also performed by molecular biology revealing the presence of the presence of Rickettsia spp . , Coxiella burnetii , Borrelia spp . and Anaplasmataceae . These results support the use of MALDI-TOF MS in entomology , tick diseases epidemiology and improve the knowledge of tick species-diversity and tick-borne pathogens circulating in Mali . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"legs",
"ixodes",
"chemical",
"compounds",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"limbs",
"(anatomy)",
"animals",
"organic",
"compounds",
"rickettsia",
"ticks",
"alcohols",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"coxiella",
"burnetii",
"infectious",
"diseases",
"spectrum",
"analysis",
"techniques",
"musculoskeletal",
"system",
"medical",
"microbiology",
"microbial",
"pathogens",
"mass",
"spectrometry",
"chemistry",
"borrelia",
"disease",
"vectors",
"matrix-assisted",
"laser",
"desorption",
"ionization",
"time-of-flight",
"mass",
"spectrometry",
"arthropoda",
"arachnida",
"analytical",
"chemistry",
"anatomy",
"organic",
"chemistry",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"species",
"interactions",
"organisms"
]
| 2017 | Molecular and MALDI-TOF identification of ticks and tick-associated bacteria in Mali |
Tsetse flies are vectors of the protozoan parasite African trypanosomes , which cause sleeping sickness disease in humans and nagana in livestock . Although there are no effective vaccines and efficacious drugs against this parasite , vector reduction methods have been successful in curbing the disease , especially for nagana . Potential vector control methods that do not involve use of chemicals is a genetic modification approach where flies engineered to be parasite resistant are allowed to replace their susceptible natural counterparts , and Sterile Insect technique ( SIT ) where males sterilized by chemical means are released to suppress female fecundity . The success of genetic modification approaches requires identification of strong drive systems to spread the desirable traits and the efficacy of SIT can be enhanced by identification of natural mating incompatibility . One such drive mechanism results from the cytoplasmic incompatibility ( CI ) phenomenon induced by the symbiont Wolbachia . CI can also be used to induce natural mating incompatibility between release males and natural populations . Although Wolbachia infections have been reported in tsetse , it has been a challenge to understand their functional biology as attempts to cure tsetse of Wolbachia infections by antibiotic treatment damages the obligate mutualistic symbiont ( Wigglesworthia ) , without which the flies are sterile . Here , we developed aposymbiotic ( symbiont-free ) and fertile tsetse lines by dietary provisioning of tetracycline supplemented blood meals with yeast extract , which rescues Wigglesworthia-induced sterility . Our results reveal that Wolbachia infections confer strong CI during embryogenesis in Wolbachia-free ( GmmApo ) females when mated with Wolbachia-infected ( GmmWt ) males . These results are the first demonstration of the biological significance of Wolbachia infections in tsetse . Furthermore , when incorporated into a mathematical model , our results confirm that Wolbachia can be used successfully as a gene driver . This lays the foundation for new disease control methods including a population replacement approach with parasite resistant flies . Alternatively , the availability of males that are reproductively incompatible with natural populations can enhance the efficacy of the ongoing sterile insect technique ( SIT ) applications by eliminating the need for chemical irradiation .
Tsetse flies are the sole vector of Human African Trypanosomiasis ( HAT ) , also known as sleeping sickness , caused by the protozoan Trypanosoma brucei spp . in sub-Saharan Africa . Recent figures released by the World Health Organization ( WHO ) indicate that the devastating HAT epidemics , which started in the early 1990s , are coming under control and may no longer represent a major public health crisis [1]–[3] . While this news is welcoming , about 60 million people continue to live in tsetse infested areas at risk for HAT in 37 countries , and those at high risk are in remote areas where disease control is difficult to implement [2] . Diseases caused by trypanosomes in animals continue to be rampant in Africa and result in severe economic and nutritional losses . The ability to curb infections in animals stands to increase both economic and nutritional status of the continent . Unfortunately , the disease toolbox remains very limited . To date , no vaccines have been developed for HAT , therapeutic treatments are expensive and have serious side effects , and diagnostic tools are inadequate [1] . Reduction of tsetse populations , however has proven as an effective method for disease control [1] . Although effective , implementation of vector control methods in remote regions of Africa where the disease is rampant is difficult , expensive and relies on extensive community participation and thus has not been widely exercised for human disease control [4] . During an endemic period however , vector control can be particularly advantageous in the absence of continued active case surveillance [5] . Mathematical models indicate that parasite infection prevalence in the tsetse host is an influential parameter for HAT epidemiology and disease dynamics [5] . Thus , reducing vector populations or reducing the parasite transmission ability of flies can be most effective in preventing disease emergence . Advances in tsetse biology offer novel strategies , one being a population replacement approach to modify tsetse’s parasite transmission ability ( vector competence ) by expressing trypanocidal molecules in the gut bacterial symbiont fauna , termed paratransgenic transformation strategy [6]–[10] . For the paratransgenic approach to be successful , gene drive mechanisms need to be discovered to spread parasite resistant phenotypes into natural populations . An alternative vector control approach currently being entertained on the continent involves a population eradication method , through sterile male releases ( SIT ) [11] . Genetic methods that induce reproductive male sterility are superior to the currently available SIT strategy that relies on chemical irradiation to induce male sterility . Tsetse flies are infected with multiple bacterial symbionts . Two of the symbionts are enteric: the obligate Wigglesworthia glossinidia reside within bacteriocytes in the midgut bacteriome organ as well as in milk gland accessory tissue [12] , while commensal Sodalis glossinidius reside both inter- and extra-cellularly in various tissues [13] . A large portion of Wigglesworthia’s proteome encodes vitamin products that may be necessary to supplement the strictly vertebrate blood meal diet of tsetse [14] . Without the bacteriome population of Wigglesworthia , tsetse flies have reduced egg development and are infecund [15]–[18] . The third symbiont , Wolbachia resides mainly in the reproductive tissues [13] , [19] , [20] . Tsetse females have an unusual viviparous reproductive biology . Females develop a single oocyte per gonotrophic cycle . The oocyte is ovulated , fertilized and undergoes embryonic development in-utero . The resulting larva hatches and is carried in the intrauterine environment through three larval instars before being deposited . During its intrauterine life , the larva receives all of its nutrients in the form of milk secreted by the female accessory glands , milk glands . While Wolbachia is transovarially transmitted , the enteric symbionts are maternally transmitted into tsetse’s intrauterine larva through mother’s milk secretions [14] . By providing ampicillin in the blood meal diet , it has been possible to clear the extracellular Wigglesworthia in the milk without damaging the intracellular Wigglesworthia in the bacteriome [21] . Thus , such females remain fecund but give rise to sterile progeny that lack Wigglesworthia ( both bacteriome and milk gland populations ) but retain Wolbachia and Sodalis . As a result of the obligate role of Wigglesworthia , it has not been possible to use tetracycline treatment to cure Wolbachia infections , and the biological significance of Wolbachia infections in tsetse has thus remained elusive . Wolbachia infections associated with various insects have been shown to cause a number of reproductive modifications in their hosts , the most common being CI [22]–[24] . CI occurs when a Wolbachia infected male mates with an uninfected female , causing developmental arrest of the embryo . In contrast , Wolbachia infected females can mate with either an uninfected male or a male infected with the same Wolbachia strain and produce viable Wolbachia infected offspring . This reproductive advantage of infected females results in the spread of Wolbachia infections along with other traits that infected insects may exhibit [25] , [26] . Empirical studies and previously developed models have shown that the reproductive advantage provided by Wolbachia may be able to drive desired phenotypes along with other maternally inherited genes , organelles and/or symbionts into natural populations [27]–[30] . The Wolbachia type found in the tsetse species Glossina morsitans morsitans belongs to the Wolbachia A super group [20] . In a number of insect systems , Wolbachia strains belonging to the A super group have been associated with the CI phenotype in the different hosts they infect [31] . Here we investigated the possible role of Wolbachia symbionts that can be used to drive desirable tsetse phenotypes into natural populations , or to induce natural reproductive male sterility for field applications . We developed a dietary supplementation method that can restore fecundity of tsetse in the absence of their natural symbiotic fauna , including obligate Wigglesworthia and Wolbachia . We report on the fitness parameters of the engineered symbiont-free lines and on the level of CI expression after wild type and aposymbiotic flies are crossed . A mathematical model was also developed to ascertain whether Wolbachia infections in tsetse could be used to drive a disease refractory phenotype into a natural population .
In many insect systems , tetracycline supplemented diet is used to generate Wolbachia free lines to demonstrate the functional role of Wolbachia through mating experiments . Inseminated tsetse females maintained on tetracycline-supplemented blood meals however do not generate any viable progeny . This is because tetracycline treatment damages the obligate intracellular Wigglesworthia present in the midgut bacteriome structure ( Figure S1 ) [21] . These results are similar to prior reports where damage to Wigglesworthia had been found to reduce host fecundity [17] , [21] , [32] . The fecundity of fertile females maintained on various diets was evaluated ( Figure 1A ) . Specifically , the diet combinations were as follows: ( a ) blood only , ( b ) blood and ampicillin , ( c ) blood and tetracycline , ( d ) blood and yeast , ( e ) blood , ampicillin and yeast , and ( f ) blood , tetracycline and yeast . We monitored the number of larva deposited in each group over a 40-day period when females undergo two gonotrophic cycles ( defined as time required for the development of a single progeny in-utero ) . Under optimum conditions the first gonotrophic cycle takes about 20–22 days for development from egg to parturition . In subsequent gonotrophic cycles females produce a larva every 9 to 11 days . As we had previously shown , ampicillin treatment does not reduce fecundity since it does not damage Wigglesworthia resident within bacteriocytes in the midgut , unlike tetracycline , which clears all bacteria including Wigglesworthia and Wolbachia and induces sterility . Accordingly , ampicillin-receiving flies remained fecund while tetracycline receiving flies were rendered sterile . Yeast extract ( 10% w/v ) provisioning of the blood meal rescued fecundity of the females receiving tetracycline to similar levels as that of wild type and ampicillin receiving flies ( 65% , 55% and 64% over the first gonotrophic cycle and 53% , 58% and 49% over the second gonotrophic cycles , respectively ) . However , yeast provisioning at 10% w/v had a cost on fecundity when compared to flies maintained on normal blood meals , ( 92% versus 55% over the first gonotrophic cycle and 92% and 58% over the second gonotrophic cycle , respectively ) . Nevertheless , yeast supplementation was able to rescue the tetracycline-induced sterility to levels comparable to those observed for GmmWt receiving yeast or ampicillin supplemented blood meals , respectively ( Figure 1A ) . Thus yeast supplemented dietary regiment allowed us to develop two lines to analyze the functional role of Wolbachia symbionts in tsetse biology; one lacking all symbionts ( GmmApo ) and another lacking Wigglesworthia but still retaining Wolbachia and Sodalis ( GmmWig− ) . The GmmApo progeny resulting from the first and second depositions of tetracycline treated mothers were tested for the presence of Sodalis , Wigglesworthia and Wolbachia by a bacterium-specific PCR-assay . The PCR-assay demonstrated the absence of all three symbionts as early as the first deposition in both the male and female GmmApo adults ( Figure 1B ) . The absence of Wolbachia from the reproductive tissues of GmmApo females was also verified by Fluorescent In Situ Hybridization ( FISH ) analysis ( Figure 1E ) . In contrast , Wolbachia was present in egg chambers during both early and late developmental stages in GmmWt females ( Figure 1C & D ) . For analysis of longevity , survivorship curves were compared using the Kaplan-Meier and log rank tests . Longevity of F1 GmmApo females was compared to that of GmmWt adults maintained on the same yeast-supplemented blood meal over 40 days ( two-gonotrophic cycles ) . No difference ( X2 = 0 . 71 , df = 1 , P = 0 . 4 ) was observed in survivorship comparisons between the two groups ( Figure 1F ) . The second line ( GmmWig− ) generated from ampicillin treated females still retain their Wolbachia and Sodalis symbionts , while lacking both Wigglesworthia populations as evidenced by FISH and PCR amplification analysis [21] . When maintained on yeast-supplemented blood , this line ( similar to GmmApo ) also did not display any longevity differences from the GmmWt adults sustained on the same diet . Tetracycline treatment has been shown to have a negative impact on the fertility of Drosophila simulans males [33] . To determine if the fertility of GmmApo males is negatively affected , we mated GmmWt females with either GmmWt or GmmApo males and maintained all flies on yeast-supplemented blood meals . Larval deposition and eclosion rates from both crosses were compared using arcsin ( sqrrt ( x ) ) transformed data to ensure normality . No significant difference was observed between the crosses for two gonotrophic cycles ( P>0 . 05 ) ( Table 1 ) . The mean larval deposition rate for GmmWt females crossed with GmmWt males was 0 . 68 and 0 . 65 for the first and second gonotrophic cycles respectively , while the mean larval deposition rate for GmmWt females crossed with GmmApo males was 0 . 87 and 0 . 89 for the first and second gonotrophic cycles , respectively ( Table 1 ) . Similarly , no difference in eclosion rates was observed between the two groups ( P>0 . 05 ) ( Table 2 ) . Of the pupae obtained in the first gonotrophic cycle from the GmmWt cross , 82% underwent eclosion compared to 83% for the cross between GmmWt females and GmmApo males . For the second gonotrophic cycle , we observed 89% average eclosion for pupae from GmmWt crosses and 93% for pupae from GmmWt females crossed with GmmApo males ( Table 2 ) . Taken together , these results demonstrate the preservation of reproductive fitness in GmmApo males and rule out possible paternal effects of Wolbachia in tsetse . To determine the expression of Wolbachia-induced CI , cage population crosses were setup between GmmWt and GmmApo individuals . Cages were the experimental units and the data were arcsin ( sqrrt ( x ) ) transformed to ensure normality . To estimate the possible cost of reproductive fitness due to loss of Wigglesworthia , we made use of GmmWig− flies . Since GmmWig− flies still retained Wolbachia infections but lacked Wigglesworthia ( as described earlier and in Figure 1A ) , this line served as the control for the CI cross in order to measure potential fecundity effects due to loss of Wigglesworthia in the GmmApo line and possible yeast-supplementation effects . Although CI typically manifests itself as embryonic lethality , given the viviparous nature of reproduction in tsetse , we measured larval deposition rates , which are reflective of both successful embryogenesis and larvagenesis ( Table 1 ) . Differences in larval deposition rates ( number of larva deposited per female ) over the two gonotrophic cycles for all crosses were significant by ANOVA on arcsin ( sqrrt ( x ) ) transformed data ( ANOVA; first deposition , F4 , 9 = 20 . 6 , P<0 . 0001 , second deposition , F4 , 10 = 21 . 9 , P≤0 . 0001 ) . No differences in larval deposition were observed between the crosses GmmWt × GmmWt , GmmWig− × GmmWig− and GmmApo × GmmApo ( Table 1 ) . However differences were observed in comparisons of the GmmApo × GmmWt cross with all other crosses for the first and second gonotrophic cycles ( Table 1 ) . Given that the GmmWig− females that lack Wigglesworthia are equally fecund as GmmWt , the strong incompatibility we observed in GmmApo females when crossed with GmmWt males is likely due to Wolbachia mediated reproductive affects , and not due to nutritional effects resulting from loss of the obligate symbiont Wigglesworthia . We found that GmmWt females were compatible with all male infection types , while GmmApo females were only compatible with GmmApo males . Crosses of GmmApo females and GmmWt males demonstrated a pattern of unidirectional CI ( Table 1 ) . Spermathecae dissections of females in incompatible crosses that did not deposit a larva revealed the presence of sperm , suggesting females were inseminated and that lack of deposition was the result of CI . We also found that larval deposition rates and pupal eclosion rates showed similar patterns to large cage experiments when measured in single-pair crosses ( Table S2 ) . Differences were observed in larval deposition rates ( number of larva deposited per female ) over the two gonotrophic cycles for all single-pair crosses ( Kruskal-Wallis; first deposition , χ2 = 9 . 3 , df = 3 , P = 0 . 03 , second deposition , χ2 = 9 . 5 , df = 3 , P = 0 . 02 ) . No differences in larval deposition were observed in pair-wise comparisons of the crosses GmmWt × GmmWt , GmmWt × GmmApo and GmmApo × GmmApo ( Table S2 ) . However differences were observed in comparisons of the incompatible GmmApo × GmmWt cross with all other crosses for the first and second gonotrophic cycles ( Table S2 ) . These results support strong CI expression driven by the Wolbachia infection status in female flies . Other than reproductive modifications , Wolbachia infections have been shown to affect the fitness of their insect hosts [34] , [35] . In this study , differences in eclosion rates ( Table 2 ) were observed in the first gonotrophic cycle of crosses of GmmApo , GmmWt , and GmmWig− individuals on arcsin ( sqrrt ( x ) ) data ( ANOVA , first gonotrophic cycle , F4 , 11 = 7 . 5 , P = 0 . 0036 , second gonotrophic cycle , F3 , 8 = 2 . 5 , P = 0 . 13 ) ( Table 2 ) . No differences in eclosion rates were observed in single pair crosses for both gonotrophic cycles ( Kruskal-Wallis; first gonotrophic cycle , χ2 = 0 . 74 , df = 3 , P = 0 . 86 , second gonotrophic cycle , χ2 = 0 . 31 , df = 2 , P = 0 . 85 ) ( Table S2 ) . To determine if observed differences in eclosion rates were due to Wolbachia infection we compared the GmmWig− × GmmWig− and the GmmApo × GmmApo cross , since both strains lack Wigglesworthia infection , but one ( GmmWig− ) harbors Wolbachia infection . There were no significant differences however between these crosses ( P>0 . 05 ) ( Table 2 ) , suggesting no extensive effect of Wolbachia infection on host eclosion rates . The CI phenotype was further examined by analyzing the reproductive tract physiology of tsetse females between incompatible and compatible crosses during embryonic development . Under normal conditions a single oocyte undergoes and completes oogenesis during larvagenesis . In compatible crosses ( ♀ GmmWt × ♂ GmmWt ) we observed that the reproductive tract contains a developing larva in the uterus and a developing or completed oocyte in one of the two ovaries ( Figure 2A ) . In an incompatible cross ( ♀ GmmApo × ♂ GmmWt ) a developing oocyte is observed in one of the ovaries in the absence of a developing larva in the uterus , suggesting a disruption of embryogenesis or early larval development ( Figure 2C ) . The observation of an incomplete oocyte in the absence of a developing larva in the uterus suggests the failure and abortion of either an embryo or very young larva . These observations differ from older GmmWt virgin females . Typically , GmmWt virgin females undergo oogenesis but do not undergo ovulation , which results in the development and eventual accumulation of two oocytes in each of the ovaries . Larvae are never observed in the uterus as developed oocytes are never ovulated , or fertilized in adult virgin females ( Figure 2B ) . From the experimental data , we estimated the impact of CI on tsetse population biology using a Bayesian Markov chain Monte Carlo method . The transmission failure of Wolbachia from mothers to developing oocytes was moderate: 10 . 7% [0 . 07% , 22 . 7%] of progeny produced by GmmWt mothers were Wolbachia uninfected ( Table 3 ) . In addition , the incompatibility between GmmWt males and GmmApo females was strong: 79 . 8% [63 . 0% , 90 . 3%] of matings between GmmWt males and GmmApo females did not result in viable larvae as measured by pupal deposition . There was a significant fecundity ( number of larval progeny deposited ) benefit for Wigglesworhia infection: GmmWt females had 28 . 4% [8 . 5% , 54 . 2%] higher fecundity than GmmWig− females . Furthermore , Wolbachia infection alone was estimated to give a fecundity benefit of 19 . 3% [−9 . 2% , 57 . 9%] . This is an estimate of the fecundity difference between hypothetical females carrying Wigglesworthia and Sodalis but not Wolbachia and the experimental GmmWt females . Most importantly , our model demonstrates that , given a large enough initial release , Wolbachia infected individuals will successfully invade a tsetse population ( Table 4 ) . The fixation prevalence of Wolbachia is estimated to be 96 . 9% [85 . 6% , 99 . 8%] . There may exist a release threshold , which an initial release must be above in order for Wolbachia to invade: the median was no release threshold ( i . e . 0% ) , but the upper end of the 95% credible interval was a release of the size of 39 . 6% of the native population . The median threshold value is zero because , despite imperfect maternal transmission , the fecundity benefit of Wolbachia is strong enough to allow Wolbachia to invade a naïve tsetse population from any size initial release , no matter how small . In addition , the time to reach fixation from a release of the size of 10% of the native population can be relatively short: the median value was 529 days , however the upper end of the 95% credible interval was undefined because in more than 2 . 5% of samples , 10% initial release was below the release threshold . Sensitivity analysis showed that the model results are sensitive to both assumed and estimated parameters ( supplementary material Text S1 ) . In particular , time to fixation had the largest sensitivity to the time to first deposition and large elasticities to Wolbachia- and Wigglesworthia-related parameters , suggesting that improving the estimates of these parameters would most effectively improve the fidelity of the estimate of time to fixation .
Here , we report for the first time on the functional role of Wolbachia infections in tsetse , which support the expression of CI . Microscopic analyses of the CI expressing females show that loss of fecundity results from early embryogenic failure . Essential for our studies we have discovered that we can maintain Wolbachia cured tsetse lines fertile by dietary provisioning of tetracycline supplemented blood meals with yeast extract , despite the fact that such flies lack the obligate mutualist Wigglesworthia , which is essential for tsetse’s fecundity . When incorporated into a mathematical model , our results suggest that Wolbachia can be used successfully as a gene driver and , the time to reach fixation is relatively short given a large enough initial release: on the order of 1 to 2 years . These results provide a first insight into the role of Wolbachia infections in a viviparous insect and indicate that Wolbachia mediated CI can potentially be used to drive desirable tsetse phenotypes into natural populations . Our data presented here as well as previous results from other studies indicate that in the absence of Wigglesworthia , tsetse females are rendered sterile . Our prior studies where we maintained inseminated flies on ampicillin supplemented blood diets resulted in progeny deposition . This is because ampicillin treatment did not affect the intracellular Wigglesworthia resident in the bacteriome organ in the midgut , which provides essential nutrients to maintain tsetse host fecundity [21] . Antibiotic ampicillin treatment however eliminated the extracellular Wigglesworthia population present in the milk gland essential for symbiont transmission , and thus the resulting progeny from such females lacked Wigglesworthia ( GmmWig− ) . Such progeny were reproductively sterile although they retained the symbiont Wolbachia . The tetracycline diet eliminated both intracellular and extracellular forms of Wigglesworthia and thus we did not obtain any viable progeny from inseminated females that were maintained on the tetracycline only diet . Prior studies showed that tetracycline blood meals supplemented with vitamin B1 could partially rescue fertility [15] , but in our experiments vitamin supplementation could give rise to at most one progeny deposition , which either did not hatch or did not survive as an adult ( data not shown ) . In sharp contrast , supplementation of the blood meal diet with 10% ( w/v ) yeast-extract reverted sterility in tetracycline treated flies to levels comparable to GmmWt and GmmWig− females receiving the same diet ( Figure 1A ) . Although we have compared the fecundity of all three lines for two gonotrophic cycles here , yeast supplemented flies continue to deposit four to five progeny ( data not shown ) . Given the complex nature of the yeast extract ( peptides , amino acids , vitamins and other yeast cell components ) , it is difficult to know the exact nature of the essential nutrients it provides , but we believe that it could be working via supplementation of lipids and/or essential vitamins that are lacking in the strict blood diet of tsetse . However , we did observe some negative effect attributable to the yeast diet when the fecundity of GmmWt flies receiving yeast supplemented blood meals is compared to those receiving normal blood diets . As such , we are further investigating the use of different yeast supplementations and/or concentrations in an effort to improve the diet efficiency . Nevertheless the availability of Wolbachia-cured flies ( GmmApo ) allowed us to begin to understand the functional role of this symbiosis . In addition to Wolbachia symbiont specific PCR amplification , we confirmed the absence of Wolbachia from the reproductive tissues of GmmApo females by FISH analysis . We show the presence of Wolbachia in GmmWt females , isolates to a pole late in development ( Figure 1C ) . There are a number of studies in other model systems that have investigated the link between Wolbachia localization during spermatogenesis and density effects on CI [36] , [37] . However , other studies have found no correlation between Wolbachia density and CI during spermatogenesis [38] , [39] . There have also been a number of studies investigating Wolbachia localization during oogenesis [40]–[42] . Different Wolbachia strains in Drosophila embryos display posterior , anterior , or cortical localization congruent with the classification based on the wsp gene sequence [39] . A positive correlation between levels of Wolbachia at the posterior pole and CI has been suggested , but this has yet to be examined in detail [42] . Not withstanding , assessing the role of Wolbachia during oogenesis is important , given that factors promoting CI rescue are deposited in the egg cytoplasm during oocyte development [43] and bacterial deposition in the oocyte is an essential even for efficient maternal transmission . Before we could perform crossing experiments to assess for CI , we evaluated the effect of Wolbachia clearance on male reproductive capacity . This evaluation is important given that tetracycline has been shown to negatively affect reproductive fitness in Drosophila simulans [33] . Additionally , the importance of this finding is highlighted by a study of the mosquito A . albopictus system in which the natural Wolbachia strains ( wAlbA and wAlbB ) were cleared and transinfected with the Wolbachia strain wRi from D . simulans [44] . Their results showed that the wRi transinfected males have a reduced mating capacity compared with the wild type super infected males [44] . In contrast , in our system , no decrease in mating capacity was observed in GmmApo males compared with GmmWt males under the laboratory conditions . Our observation agrees with the evolutionary model proposed by Charlat et al . , [45] , where Wolbachia is exclusively maternally transmitted therefore males may be considered an evolutionary dead end in terms of Wolbachia infection [46] . Consequently , no direct selection by Wolbachia can be theoretically expected on paternal reproductive fitness . Loss of fecundity in the cross ( ♀ GmmApo x ♂ GmmWt ) could conceivably arise from loss of Wigglesworthia-mediated nutritional benefits in GmmApo females rather than to Wolbachia mediated CI . To test this possibility , we compared the larval deposition rates in crosses between ♀GmmApo × ♂ GmmApo and ♀GmmWig− × ♂ GmmWig− flies ( Table 1 ) . Our results show no statistically significant differences between these crosses indicating that loss of fecundity in the CI cross is not due to loss of Wigglesworthia . Our empirical results were used to parameterize a population genetic model of the spread of Wolbachia . Our model demonstrated that GmmWt would successfully invade an uninfected natural population with a large enough release given CI rates . Indeed , uninfected natural populations and natural populations with low infection prevalence have recently been identified for multiple tsetse species [47] . This modeling result is consistent with the natural spread of Wolbachia in Drosophila populations [48]–[50] . In addition , the rise to the predicted fixation prevalence of between 86% and 100% is rapid . Apparently , the Wolbachia-mediated CI has the potential to rapidly and effectively drive a desirable phenotype into natural populations . We have previously been able to culture and genetically transform the commensal symbiont of tsetse , Sodalis glossinidius [51] . It has also been possible to reintroduce the transformed Sodalis into tsetse , called a paratransgenic approach [52] , [53] . Given that Sodalis resides in close proximity to pathogenic trypanosomes in tsetse’s midgut , products expressed in recSodalis can have an immediate effect on trypanosome biology . The potential paratransgenic strategy in tsetse could harness the Wolbachia mediated CI to drive a recombinant Sodalis strain that would encode parasite resistance genes into natural populations [6] , [10] . Our studies on the maternal transmission dynamics of tsetse’s symbionts in the laboratory indicated perfect transmission of both Wolbachia and Sodalis into tsetse’s sequential progeny [54] . This high transmission fidelity of the two symbionts , coupled with strong nearly 100% CI caused by Wolbachia would serve paratransgenic applications favorably . An alternative control strategy to paratransgenic population replacement strategy would be use CI as part of an incompatible insect technique ( IIT ) , which is analogous to a SIT approach [29] , [55]–[58] . In a Wolbachia-based SIT approach female sterility is artificially sustained by repeated releases of cytoplasmically incompatible males . Similar to SIT , the increasing ratio of incompatible matings over time can lead to population suppression . The benefit of an IIT strategy is that it would not require the use of irradiation or chemosterilants to sterilize males prior to release , which often reduces the fitness of released males , but would rely on the naturally induced sterility of an incompatible Wolbachia infection [59] . A Wolbachia-based paratransgenic and IIT control strategy for tsetse would rely upon the introduction of a novel infection type into a population with an existing infection that could result in bi-directional CI or the introduction of a novel infection into an uninfected host population . Typically , in other insect systems novel Wolbachia infections are established by embryonic microinjections [60] , [61] . This would be difficult in tsetse given their viviparous reproductive biology , in that adult females carry and nourish their offspring for their entire larval developmental cycle making injections of embryos difficult . Future studies however can focus on the introduction of novel infection types via microinjection in aposymbiotic and naturally infected adult flies [62] . Maternal intrathoracic injections of Wolbachia infection establishment has also been successful in Aedes aegypti [63] . There has been a growing interest in understanding the variety of Wolbachia induced phenotypes in arthropods given the impact that Wolbachia infections could potentially have on genetic variation and host speciation impacting evolution of the species . Our data add to this growing field , as this is the first demonstration of the biological significance of Wolbachia infections in tsetse . Interestingly , CI in tsetse appears to be strong in that by the second gonotrophic cycle 0% of the females in an incompatible cross give rise to progeny . This is an exception given that in many insect systems incomplete CI is observed [27] , [64] . Future studies with natural populations would now be important to confirm some of the parameters we report here including maternal transmission rates , infection prevalence and the maternal linkage efficacy between Wolbachia and other maternally transmitted symbionts such as Sodalis , which is being entertained for paratransgenic applications . Additionally , the aposymbiotic lines generated in this study are currently being used to address the interactive role of trypanosome transmission in tsetse . The importance of which is highlighted by recent studies that have shown that Wolbachia infections may impact host immune biology , limiting pathogen proliferation in insect hosts [65]–[70] .
The Glossina morsitans morsitans colony maintained in the insectary at Yale University was originally established from puparia collected in Zimbabwe . Newly emerged flies are separated based on sex and mated at three to four days post eclosion . Flies are maintained at 24±1°C with 50 – 55% relative humidity and fed defibrinated bovine blood ( HemoStat Laboratories , CA ) every forty eight hours using an artificial membrane system [71] . Selective elimination of natural tsetse endosymbionts was obtained as described below . Wild type ( GmmWt ) fertile females were maintained on blood meals supplemented with 10% ( w/v ) yeast extract ( Becton Dickinson ) and 20 ug/ml of tetracycline . The yeast extract was briefly boiled in water before being added the blood meal each time . Flies were fed every 48 h using an artificial membrane feeding system ( as above ) for the duration of their life span . The resulting progeny are aposymbiotic ( GmmApo ) in that they lack their natural endosymbionts , Wigglesworthia and Wolbachia . These GmmApo lines were maintained on blood meals supplemented with 10% ( w/v ) yeast extract without tetracycline . GmmWt fertile females were maintained on blood meals supplemented with 50 ug/ml of ampicillin . The resulting progeny do not have Wigglesworthia ( GmmWig− ) , and were maintained on blood meals supplemented with 10% ( w/v ) yeast extract without ampicillin . Newly eclosed aged matched females and males were divided into six groups and copulation observed . Three of these groups were provided with either normal blood meals ( control ) or blood meals supplemented with ampicillin at 50 ug/ml or tetracycline at 20 ug/ml . Whereas the remaining three groups received blood meals supplemented with 10% ( w/v ) yeast extract with either ampicillin ( 50 ug/ml ) or tetracycline ( 20 ug/ml ) . The cages were monitored daily for pupal deposition and fly mortality over two gonotrophic cycles ( 40 days ) . Fecundity was quantified by determining the number of fecund females relative to total number of females alive at the end of the gonotrophic cycle to give an average percent of females depositing pupae . Each group was setup with 100 females per cage . Total DNA was extracted from adults eight days post eclosion using the Qiagen Blood and Tissue extraction kit under manufacturers conditions ( Qiagen Kit # , 69506 . CA ) . The presence of the symbionts Sodalis , Wigglesworthia and Wolbachia was determined by a species-specific PCR amplification assay using the primer sets and conditions described ( Table S1 ) . For input DNA quality control , the tsetse gene β-tubulin ( GmmTub ) specific primer set was used . All PCR reactions were performed in an MJ-Research thermocycler and the amplification products were analyzed by electrophoresis on a 1% agarose gel and visualized using image analysis software . Dissected reproductive tracts from GmmWt and GmmApo females were fixed in 4% paraformaldehyde ( PFA ) , embedded in paraffin , cut into 5 mm thick sections and mounted on poly L-lysine coated microscopy slides . After dewaxing in methylcyclohexane and rehydration the sections were processed using the FISH protocol previously described in Anselme et al . 2006 [72] . Slides were covered with a drop of 70% acetic acid and incubated at 45°C until drop had dried , followed by dehydration and a 10 min deproteinization step in 0 . 01N HCl/pepsine at 37°C . Slides were then dehydrated again , prehybridized for 30 min at 45°C and hybridized for 3 h at 45°C with 5′ end rhodamine labeled 16S RNA probes ( 5′-AAT CCG GCC GAR CCG ACC C -3′ ) and ( 5′-CTT CTG TGA GTA CCG TCA TTA TC -3′ ) . Microscopic analyses were conducted using a Zeiss Axioskop2 microscope equipped with an Infinity1 USB 2 . 0 camera and software ( Lumenera Corporation ) . Fluorescent images were taken using a fluorescent filter set with fluorescein , rhodamine and DAPI specific channels . GmmApo and GmmWt flies that emerged within a 24-hour period ( teneral ) were collected , mated with GmmApo males at a ratio of 5∶2 and copulation was observed . After six days males were removed from experimental cages . Six independent cages were set-up for both GmmApo and GmmWt groups , comprising of a total of 169 GmmApo and 170 GmmWt females , respectively . Both the males and females used represented offspring acquired from different gonotrophic cycles ( 1st and 2nd ) . All flies were maintained on yeast extract supplemented blood meals and fly mortality was monitored daily over a 40-day period . To determine the expression of CI , reciprocal crosses were set up between GmmApo , GmmWt and GmmWig− flies , in triplicate . Cages with a minimum of 15 females and 7 males each were set-up in the following combinations: 1 ) ♀ GmmWt × ♂ GmmWt , 2 ) ♀ GmmWt × ♂ GmmApo , 3 ) ♀ GmmApo × ♂ GmmApo , 4 ) ♀ GmmApo × ♂ GmmWt and 5 ) ♀ GmmWig− × ♂ GmmWig− . All flies received yeast supplemented blood meal diets . Flies were observed over two-gonotrophic cycles with daily recording of mortality , larval deposition dates , pupal eclosion dates and sex of emergent progeny . Larval deposition rates for each gonotrophic cycle were determined by dividing the number of larvae deposited per day by the number of remaining females in the cage on the day of larviposition and summing the values for each gonotrophic cycle . At the conclusion of the experiment , all females were checked for insemination by examination of dissected spermatheca for the presence of sperm microscopically . Additionally , single line crosses consisting of a single female and male per cage were set up ( Table S2 ) . For the ♀ GmmWt × ♂ GmmWt a total of 31 crosses were set up . Also set up were 40 crosses for ♀ GmmWt × ♂ GmmApo , 20 for ♀ GmmApo × ♂ GmmApo and 33 for ♀ GmmApo × ♂ GmmWt . Both the males and females used in these crosses represented offspring acquired from different gonotrophic cycles to rule out batch affects . Spermathecae of females was also dissected to confirm insemination . Here we will briefly describe the mathematical modeling used in this study; full details are available in the supplementary material ( Text S1 ) . The data from mating crosses were modeled as samples from the standard binomial random variable , with probability of larval deposition per mated female per gonotrophic cycle , and using a different probability for each cross . Following the empirical findings regarding Wolbachia -mediated CI in Drosophila [48] , the probabilities were then defined in terms of four mechanistic parameters: the probability of reproduction success ( larval deposit ) from a cross between an GmmApo female and an GmmApo male ( ) , the proportion of Wolbachia-free eggs of Wolbachia-carrying mothers ( ) , the relative benefit to reproduction success of Wolbachia infection to females ( ) , the relative benefit to reproduction success of Wigglesworthia infection to females ( ) , and the proportion of fertilizations of Wolbachia-free eggs by Wolbachia-affected sperm that are not viable ( ) . The larval-deposition probabilities in terms of these parameters arewhere the subscripts refer to the types of the female and male , respectively , with for wild type ( GmmWt ) , for tetracycline treated ( GmmApo ) , and for ampicillin treated ( GmmWig− ) . In addition to these mechanistic parameters , we also estimated population-genetic quantities fundamental to the invasion of Wolbachia into a novel tsetse population . Again following existing models for Wolbachia-induced CI in Drosophila [38] , a mathematical model was developed for the temporal evolution of tsetse abundance with and without Wolbachia infection . We incorporated the Wolbachia-mediated CI trade-off of the fitness cost to male hosts in reducing their mating success with uninfected females versus the fitness benefit to female hosts in allowing them to successfully mate with both infected and uninfected males ( in addition to direct effects of Wolbachia on fecundity and mortality ) . For some values of the mechanistic parameters , these models exhibit a threshold for Wolbachia invasion into the host population: if , in a novel population , the proportion that is initially Wolbachia infected is above the threshold , Wolbachia will continue to stable fixation in the population at a high level . If the proportion infected is below the threshold , Wolbachia will be driven out of the population over time . This threshold level was calculated , along with the prevalence of Wolbachia at fixation , and the time to fixation . For the population-genetic model , several parameters could not be estimated from the data on mating crosses . Thus , we also performed a sensitivity analysis on these parameters , along with the parameters estimated from the mating-cross data . To estimate both the mechanistic parameters for CI and the population-genetics quantities derived from these parameters , a Bayesian Markov chain Monte Carlo ( MCMC ) method was used with uninformative prior distributions for the parameters [49] . | Infections with the parasitic bacterium Wolbachia are widespread in insects and cause a number of reproductive modifications , including cytoplasmic incompatibility ( CI ) . There is growing interest in Wolbachia , as CI may be able to drive desired phenotypes such as disease resistance traits , into natural populations . Although Wolbachia infections had been reported in the medically and agriculturally important tsetse , their functional role was unknown . This is because attempts to cure tsetse of Wolbachia by antibiotic treatment damages the obligate mutualist Wigglesworthia , without which the flies are sterile . Here we have succeeded in the development of Wolbachia free and still fertile tsetse lines . Mating experiments for the first time provides evidence of strong CI in tsetse . We have incorporated our empirical data in a mathematical model and show that Wolbachia infections can be harnessed in tsetse to drive desirable phenotypes into natural populations in few generations . This finding provides additional support for the application of genetic approaches , which aim to spread parasite resistance traits in natural populations as a novel disease control method . Alternatively , releasing Wolbachia infected males can enhance Sterile Insect applications , as this will reduce the fecundity of natural females either uninfected or carrying a different strain of Wolbachia . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"mathematics",
"applied",
"mathematics",
"biology",
"microbiology"
]
| 2011 | Wolbachia Symbiont Infections Induce Strong Cytoplasmic Incompatibility in the Tsetse Fly Glossina morsitans |
Mathematical models of scientific data can be formally compared using Bayesian model evidence . Previous applications in the biological sciences have mainly focussed on model selection in which one first selects the model with the highest evidence and then makes inferences based on the parameters of that model . This “best model” approach is very useful but can become brittle if there are a large number of models to compare , and if different subjects use different models . To overcome this shortcoming we propose the combination of two further approaches: ( i ) family level inference and ( ii ) Bayesian model averaging within families . Family level inference removes uncertainty about aspects of model structure other than the characteristic of interest . For example: What are the inputs to the system ? Is processing serial or parallel ? Is it linear or nonlinear ? Is it mediated by a single , crucial connection ? We apply Bayesian model averaging within families to provide inferences about parameters that are independent of further assumptions about model structure . We illustrate the methods using Dynamic Causal Models of brain imaging data .
Mathematical models of scientific data can be formally compared using Bayesian model evidence [1]–[3] , an approach that is now widely used in statistics [4] , signal processing [5] , machine learning [6] , natural language processing [7] , and neuroimaging [8]–[10] . An emerging area of application is the evaluation of dynamical system models represented using differential equations , both in neuroimaging [11] and systems biology [12]–[14] . Much previous practice in these areas has focussed on model selection in which one first selects the model with the highest evidence and then makes inferences based on the parameters of that model [15]–[18] . This ‘best model’ approach is very useful but , as we shall see , can become brittle if there are a large number of models to compare , or if in the analysis of data from a group of subjects , different subjects use different models ( as is the case for a random effects analysis [19] ) . This brittleness , refers to the fact that which is the best model can depend critically on which set of models are being compared . In random effects analysis , augmenting the comparison set with a single extra model can , for example , reverse the ranking of the best and second best models . To address this issue we propose the combination of two further approaches ( i ) family level inference and ( ii ) Bayesian model averaging within families . We envisage that these methods will be useful for the comparison of large numbers of models ( eg . tens , hundreds or thousands ) . In the context of neuroimaging , for example , inferences about changes in brain connectivity can be made using Dynamic Causal Models [20] , [21] . These are differential equation models which relate neuronal activity in different brain areas using a dynamical systems approach . One can then ask a number of generic questions . For example: Is processing serial or parallel ? Is it linear or nonlinear ? Is it mediated by changes in forward or backward connections ? A schematic of a DCM used in this paper is shown in Figure 1 . The particular questions we will address in this paper are ( i ) which regions receive driving input ? and ( ii ) which connections are modulated by other experimental factors ? This paper proposes that the above questions are best answered by ‘Family level inference’ . That is inference at the level of model families , rather than at the level of the individual models themselves . As a simple example , in previous work [19] we have considered comparison of a number of DCMs , half of which embodied linear hemodynamics and half nonlinear hemodynamics . The model space was thus partitioned into two families; linear and nonlinear . One can compute the relative evidence of the two model families to answer the question: does my imaging data provide evidence in favour of linear versus nonlinear hemodynamics ? This effectively removes uncertainty about aspects of model structure other than the characteristic of interest . We have provided a simple illustration of this approach in previous work [19] . We now provide a formal introduction to family level inference and describe the key issues . These include , importantly , the issue of how to deal with families that do not contain the same number of models . Additionally , this paper shows how Bayesian model averaging can be used to provide a summary measure of likely parameter values for each model family . We provide an example of family-level inference using data from neuroimaging , a DCM study of auditory word processing , but envisage that the methods can be applied throughout the biological sciences . Before proceeding we note that the use of Bayesian model averaging is a standard approach in the field of Bayesian statistics [4] , but has yet to be applied extensively in computational biology . The use of model families is also accomodated naturally within the framework of hierarchical Bayesian models [1] and is proposed to address the well known issue of model dilution [4] .
Dynamic Causal Modelling is a framework for fitting differential equation models of neuronal activity to brain imaging data using Bayesian inference . The DCM approach can be applied to functional Magnetic Resonance Imaging ( fMRI ) , Electroencephalographic ( EEG ) , Magnetoencephalographic ( MEG ) , and Local Field Potential ( LFP ) data [22] . The empirical work in this paper uses DCM for fMRI . DCMs for fMRI comprise a bilinear model for the neurodynamics and an extended Balloon model [23] for the hemodynamics . The neurodynamics are described by the following multivariate differential equation ( 1 ) where indexes continuous time and the dot notation denotes a time derivative . The th entry in corresponds to neuronal activity in the th region , and is the th experimental input . A DCM is characterised by a set of ‘exogenous connections’ , , that specify which regions are connected and whether these connections are unidirectional or bidirectional . We also define a set of input connections , , that specify which inputs are connected to which regions , and a set of modulatory connections , , that specify which intrinsic connections can be changed by which inputs . The overall specification of input , intrinsic and modulatory connectivity comprise our assumptions about model structure . This in turn represents a scientific hypothesis about the structure of the large-scale neuronal network mediating the underlying cognitive function . A schematic of a DCM is shown in Figure 1 . In DCM , neuronal activity gives rise to fMRI activity by a dynamic process described by an extended Balloon model [24] for each region . This specifies how changes in neuronal activity give rise to changes in blood oxygenation that are measured with fMRI . It involves a set of hemodynamic state variables , state equations and hemodynamic parameters , . In brief , for the th region , neuronal activity causes an increase in vasodilatory signal that is subject to autoregulatory feedback . Inflow responds in proportion to this signal with concomitant changes in blood volume and deoxyhemoglobin content . ( 2 ) Outflow is related to volume through Grubb's exponent [20] . The oxygen extraction is a function of flow where is resting oxygen extraction fraction . The Blood Oxygenation Level Dependent ( BOLD ) signal is then taken to be a static nonlinear function of volume and deoxyhemoglobin that comprises a volume-weighted sum of extra- and intra-vascular signals [20] ( 3 ) where is resting blood volume fraction . The hemodynamic parameters comprise and are specific to each brain region . Together these equations describe a nonlinear hemodynamic process that converts neuronal activity in the th region to the fMRI signal ( which is additionally corrupted by additive Gaussian noise ) . Full details are given in [20] , [23] . In DCM , model parameters are estimated using Bayesian methods . Usually , the parameters are of greatest interest as these describe how connections between brain regions are dependent on experimental manipulations . For a given DCM indexed by , a prior distribution , is specified using biophysical and dynamic constraints [20] . The likelihood , can be computed by numerically integrating the neurodynamic ( equation 1 ) and hemodynamic processes ( equation 2 ) . The posterior density is then estimated using a nonlinear variational approach described in [23] , [25] . Other Bayesian estimation algorithms can , of course , be used to approximate the posterior density . Reassuringly , posterior confidence regions found using the nonlinear variational approach have been found to be very similar to those obtained using a computationally more expensive sample-based algorithm [26] . This section reviews methods for computing the evidence for a model , , fitted to a single data set . Bayesian estimation provides estimates of two quantities . The first is the posterior distribution over model parameters which can be used to make inferences about model parameters . The second is the probability of the data given the model , otherwise known as the model evidence . In general , the model evidence is not straightforward to compute , since this computation involves integrating out the dependence on model parameters ( 4 ) A common technique for approximating the above integral is the Variational Bayes ( VB ) approach [27] . This is an analytic method that can be formulated by analogy with statistical physics as a gradient ascent on the ‘negative variational Free Energy’ ( or Free Energy for short ) , , of the system . This quantity is related to the model evidence by the relation [27] , [28] ( 5 ) where the last term in Eq . ( 5 ) is the Kullback-Leibler ( KL ) divergence between an ‘approximate’ posterior density , , and the true posterior , . This quantity is always positive , or zero when the densities are identical , and therefore is bounded below by . Because the evidence is fixed ( but unknown ) , maximising implicitly minimises the KL divergence . The Free Energy then becomes an increasingly tighter lower bound on the desired log-model evidence . Under the assumption that this bound is tight , model comparison can then proceed using as a surrogate for the log-model evidence . The Free Energy is but one approximation to the model evidence , albeit one that is widely used in neuroimaging [29] , [30] . A simpler approximation , the Bayesian Information Criterion ( BIC ) [11] , uses a fixed complexity penalty for each parameter . This is to be compared with the free energy approach in which the complexity penalty is given by the KL-divergence between the prior and approximate posterior [11] . This allows parameters to be differentially penalised . If , for example , a parameter is unchanged from its prior , there will be no penalty . This adaptability makes the Free Energy a better approximation to the model evidence , as has been shown empirically [6] , [31] . There are also a number of sample-based approximations to the model evidence . For models with small numbers of parameters the Posterior Harmomic Mean provides a good approximation . This has been used in neuroscience applications , for example , to infer based on spike data whether neurons are responsive to particular features , and if so what form the dependence takes [32] . For models with a larger number of parameters the evidence can be well approximated using Annealed Importance Sampling ( AIS ) [33] . In a comparison of sample-based methods using synthetic data from biochemical networks , AIS provided the best balance between accuracy and computation time [13] . In other comparisons , based on simulation of graphical model structures [6] the Free Energy method approached the performance of AIS and clearly outperformed BIC . In this paper model evidence is approximated using the Free Energy . Neuroimaging data sets usually comprise data from multiple subjects as the perhaps subtle cognitive effects one is interested in are often only manifest at the group level . In this and following sections we therefore consider group model inference where we fit models to data from subjects . Every model is fitted to every subjects data . In Fixed Effects ( FFX ) Analysis it is assumed that every subject uses the same model , whereas Random Effects ( RFX ) Analysis allows for the possibility that different subjects use different models . This section focusses on FFX . Given that our overall data set , , which comprises data for each subject , , is independent over subjects , we can write the overall model evidence as ( 6 ) Bayesian inference at the model level can then be implemented using Bayes rule ( 7 ) Under uniform model priors , , the comparison of a pair of models , and , can be implemented using the Bayes Factor which is defined as the ratio of model evidences ( 8 ) Given only two models and uniform priors , the posterior model probability is greater than 0 . 95 if the BF is greater than twenty . Bayes Factors have also been stratified into different ranges deemed to correspond to different strengths of evidence . ‘Strong’ evidence , for example , corresponds to a BF of over twenty [34] . Under non-uniform priors , pairs of models can be compared using Odds Ratios . The prior and posterior Odds Ratios are defined as ( 9 ) resepectively , and are related by the Bayes Factor ( 10 ) When comparing two models across a group of subjects , one can multiply the individual Bayes factors ( or exponentiate the sum of log evidence differences ) ; this is referred to as the Group Bayes Factor ( GBF ) [16] . As is made clear in [19] the GBF approach implicitly assumes that every subject uses the same model . It is therefore a Fixed Effects analysis . If one believes that the optimal model structure is identical across subjects , then an FFX approach is entirely valid . This assumption is warranted when studying a basic physiological mechanism that is unlikely to vary across subjects , such as the role of forward and backward connections in visual processing [35] . An alternative procedure for group level model inference allows for the possibility that different subjects use different models . This may be the case in neuroimaging when investigating pathophysiological mechanisms in a spectrum disease or when dealing with cognitive tasks that can be performed with different strategies . RFX inference is based on the characteristics of the population from which the subjects are drawn . Given a candidate set of models , we denote as the frequency with which model is used in the population . We also refer to as the model probability . We define a prior distribution over which in this paper , and in previous work [19] , is taken to be a Dirichlet density ( but see later ) ( 11 ) where is a normalisation term and the parameters , , are strictly positively valued and can be interpreted as the number of times model has been observed or selected . For the density is convex in -space , whereas for it is concave . Given that we have drawn subjects from the population of interest we then define the indicator variable as equal to unity if model has been assigned to subject . The probability of the ‘assignation vector’ , , is then given by the multinomial density ( 12 ) The model evidence , , together with the above densities for model probabilities and model assignations constitutes a generative model for the data , ( see figure 1 in [19] ) . This model , can then be inverted to make inferences about the model probabilites from experimental data . Such an inversion has been described in previous work , which developed an approximate inference procedure based on a variational approximation [19] ( this was in addition to the variational approximation used to compute the Free Energy for each model ) . The robustness and accuracy of this method was verified via simulations using data from synthetic populations with known frequencies of competing models [19] . This algorithm produces an approximation to the posterior density on which subsequent RFX inferences are based . As we shall see in the following section , unbiased family level inferences require uniform priors over families . This requires that the prior model counts , , take on very small values ( see equation 24 ) . These values become smaller as the number of models in a family increases . It turns out that although the variational algorithm is robust for , it is not accurate for . This is a generic problem with the VB approach and is explained further in the the supporting material ( see file Text S1 ) . For this reason , in this paper we choose to take a Gibbs sampling instead of a VB approach . Additionally , the use of Gibbs sampling allows us to relax the assumption made in VB that the posterior densities over and factorise [19] . Gibbs sampling is the Monte-Carlo method of choice when it is possible to iteratively sample from the conditional posteriors [1] . Fortunately , this is the case with the RFX models as we can iterate between sampling from and . Such iterated sampling eventually produces samples from the marginal posteriors and by allowing for a sufficient burn-in period after which the Markov-chain will have converged [1] . The procedure is described in the following section . We have so far described procedures for Bayesian inference over models . These models comprise the comparison set , . This section points out a number of generic features of Bayesian model comparison . First , for any data set there exists an infinite number of possible models that could explain it . The purpose of model comparison is not to discover a ‘true’ model , but to determine that model , given a set of plausible alternatives , which is most ‘useful’ , ie . represents an optimal balance between accuracy and complexity . In other words Bayesian model inference has nothing to say about ‘true’ models . All that it provides is an inference about which is more likely , given the data , among a set of candidate models . Second , we emphasise that posterior model probabilities depend on the comparison set . For FFX inference this can be clearly seen in equation 7 where the denominator is given by a sum over . Similarly , for RFX inference , the dependence of posterior model probabilities on the comparison set can be seen in equation 14 . Other factors being constant , posterior model probabilities are therefore likely to be smaller for larger . Our third point relates to the ranking of models . For FFX analysis the relative ranking of a pair of models is not dependent on . That is , if then for any two comparison sets and that contain models and . This follows trivially from equation 7 as the comparison set acts only as a normalisation term . However , for group random effects inference the ranking of models can be critically dependent on the comparison set . That is , if then it could be that where is the posterior expected probability of model given comparison set . The same holds for other quantities derived from the posterior over , such as the exceedance probability ( see [19] and later ) . This means that the decision as to which is the best model depends on . This property arises because different subjects can use different models and we illustrate it with the following example . Consider that comprises just two models and . Further assume that we have subjects and model is preferred by 7 of these subjects and by the remaining 10 . We assume , for simplicity , that the degrees of preference ( ie differences in evidence ) are the same for each subject . The quantity then simply reflects the proportion of subjects that prefer model [19] . So , and for comparison set model 2 is the highest ranked model . Although the differences in posterior expected values are small the corresponding differences in exceedance probabilities will be much greater . Now consider a new comparison set that contains an addditional model . This model is very similar to model such that , of the ten subjects who previously preferred it , six still do but four now prefer model . Again , assuming identical degrees of preference , we now have , and . So , for comparison set model is now the best model . So which is the best model: model one or two ? We suggest that this seeming paradox shows , not that group random effects inference is unreliable , but that it is not always appropriate to ask which is the best model . As is usual in Bayesian inference it is wise to consider the full posterior density rather than just the single maximum posterior value . We can ask what is common to models two and three . Perhaps they share some structural assumption such as the existence of certain connections or other characteristic such as nonlinearity . If one were to group the models based on this characteristic then the inference about the characteristic would be robust . This notion of grouping models together is formalised using family-level inference which is described in the following section . One can then ask: of the models that have this characteristic what are the typical parameter values ? This can be addressed using Bayesian Model Averaging within families . To implement family level inference one must specify which models belong to which families . This amounts to specifying a partition , , which splits S into disjoint subsets . The subset contains all models belonging family and there are models in the th subset . Different questions can be asked by specifying different partitions . For example , to test model space for the ‘effect of linearity’ one would specify a partition into linear and nonlinear subsets . One could then test the same model space for the ‘effect of seriality’ using a different partition comprising serial and parallel subsets . The subsets must be non-overlapping and their union must be equal to S . For example , when testing for effects of “seriality” , some models may be neither serial or parallel; these models would then define a third subset . The usefulness of the approach is that many models ( perhaps all models ) are used to answer ( perhaps ) all questions . This is similar to factorial experimental designs in psychology [36] where data from all cells are used to assess the strength of main effects and interactions . We now relate the two-levels of inference: family and model . So far , we have dealt with inference on model-space , using partitions into families . We now consider inference on parameters . Usually , the key inference is on models , while the maximum a posteriori ( MAP ) estimates of parameters are reported to provide a quantitative interpretation of the best model ( or family ) . Alternatively , people sometimes use subject-specific MAP estimates as summary statistics for classical inference at the group level . These applications require only a point ( MAP ) estimate . However for completeness , we now describe how to access the full posterior density on parameters , from which MAP estimates can be harvested . The basic idea here is to use Bayesian model averaging within a family; in other words , summarise family-specific coupling parameters in a way that avoids brittle assumptions about any particular model . For example , the marginal posterior for subject and family is ( 27 ) where is our variational approximation to the subject specific posterior and is the posterior probability that subject uses model . We could take this to be under the FFX assumption that all subjects use the same model , or under the RFX assumption that each subject uses their own model ( see equation 14 ) . Finally , to provide a single posterior density over subjects one can define the parameters for an average subject ( 28 ) and compute the posterior density from the above relation and the individual subject posteriors from equation 27 . Equation 27 arises from a straightforward application of probability theory in which a marginal probability is computed by marginalising over quantities one is uninterested in ( see also equation 4 for marginalising over parameters ) . Use of equation 27 in this context is known as Bayesian Model Averaging ( BMA ) [4] , [37] . In neuroimaging BMA has previously been used for source reconstruction of MEG and EEG data [9] . We stress that no additional assumptions are required to implement equation 27 . One can make small or large . If we make , the entire model-space , the posteriors on the parameters become conventional Bayesian model averages where . Conversely , if we make , a single model , we get conventional parameter inference of the sort used when selecting the best model; i . e . , . This is formally identical to using under the assumption that the posterior model density is a point mass at . More generally , we want to average within families of similar models that have been identified by inference on families . One can see from equation 27 that models with low probability contribute little to the estimate of the marginal density . This property can be made use of to speed up the implementation of BMA by excluding low probability models from the summation . This can be implemented by including only models for which ( 29 ) where is the minimal posterior odds ratio . Models satisfying this criterion are said to be in Occam's window [38] . The number of models in the window , , is a useful indicator as smaller values correspond to peakier posteriors . In this paper we use . We emphasise that the use of Occam's window is for computational expedience only . Although it is fairly simple to compute the MAP estimates of the Bayesian parameter ( MAP ) averages analytically , the full posteriors per se have a complicated form . This is because they are mixtures of Gaussians ( and delta functions for models where some parameters are precluded a priori ) . This means the posteriors can be multimodal and are most simply evaluated by sampling . The sampling approach can be implemented as follows . This generates samples from the posterior density . For each sample , , and subject we first select a model as follows . For RFX we draw from ( 30 ) where the th element of the vector is the posterior model probability for subject , ( we will use the expected values from equation 14 ) . For FFX the model probabilities are the same for all subjects and we draw from ( 31 ) where is the vector of posterior model probabilities with th element equal to . For each subject one then draws a single parameter vector , from the subject and model specific posterior ( 32 ) These samples can then be averaged to produce a single sample ( 33 ) One then generates another sample by repeating steps 30/31 , 32 and 33 . The samples then provide a sample-based representation of the posterior density from which the usual posterior means and exceedance probabilities can be derived . Model averaging can also be restricted to be within-subject ( using equations 30/31 and 32 only ) . Summary statistics from the resulting within-subject densities can then be entered into standard random effects inference ( eg using t-tests ) [19] . For any given parameter , some models assume that the parameter is zero . Other models allow it to be non-zero and its value is estimated . The posterior densities from equation 27 will therefore include a delta function at zero , the height of which corresponds to the posterior probability mass of models which assume that the parameter is zero . For the applications in this paper , the posterior densities from equation 27 will therefore correspond to a mixture of delta functions and Gaussians because for DCMs have a Gaussian form . This is reminiscent of the model selection priors used in [39] but in our case we have posterior densities .
Our first family level inference concerns the pattern of input connectivity . To this end we assign each of the models to one of input pattern families . These are family A ( models 1 to 64 ) , F ( 65 to 128 ) , P ( 129 to 192 ) , AF ( 193 to 256 ) , PA ( 257 to 320 ) , PF ( 321 to 384 ) and PAF ( 285 to 448 ) . Family PA , for example , has auditory inputs to both region P and A . The first two numerical columns of Table 1 show the posterior family probabilities from an FFX analysis computed using equation 21 . These are overwhelmingly in support of models in which region P alone receives auditory input ( alternative probability ) . The last two columns in Table 1 show the corresponding posterior expectations and exceedance probabilities from an RFX analysis computed using equation 25 . The conclusions from RFX analysis are less clear cut . But we can say , with high confidence ( total exceedance probability , ) that either region A alone or region P alone receives auditory input . Out of these two possibilities it is much more likely that region P alone receives auditory input ( exceedance probability ) rather than region A ( exceedance probability ) . Figure 2 shows the posterior distributions , from an RFX analysis , for each of the model families . Having established that auditory input most likely enters region we now turn to a family level inference regarding modulatory structure . For this inference we restrict our set of candidate models , , to the 64 models receiving input to region . We then assign each of these models to one of modulatory families . These were specified by first defining a hierarchy with region P at the bottom , A in the middle and F at the top; in accordance with recent studies that tend to place F above A in the language hierarchy [40] . For each structure we then counted the number of forward , , and backward , , connections and defined the following families: predominantly forward ( F , ) , predominantly backward ( B , ) , balanced ( BAL , ) , or None . The first two numerical columns of Table 2 show the posterior family probabilities from an FFX analysis . We can say , with high confidence ( total posterior probability , ) that . The last two columns in Table 2 show the posterior expectations and exceedance probabilities from an RFX analysis . These were computed from the posterior densities shown in Figure 3 . The conclusions we draw , in this case , are identical to those from the FFX analysis . That is , we can say , with high confidence ( total exceedance probability , ) that . Family level posteriors are related to model level posteriors via summation over family members according to equation 21 for FFX and equation 22 for RFX . Figure 4 shows the how the posterior probabilities over input families break down into posterior probabilities for individual models . Figure 5 shows the same for the modulatory families . The maximum posterior model for the input family inference is model number 185 having posterior probability . Given that all families have the same number of members , the model priors are uniform , so the maximum posterior model is also the one with highest aggregate model evidence . This model has input to region P and modulatory connections as shown in Figure 6 ( a ) . The model evidence for the DCMs fitted in this paper was computed using the free energy approximation . This is to be contrasted with previous work in which ( the most conservative of ) AIC and BIC was used [17] . One notable difference arising from this distinction is that the top-ranked models in [17] contained significantly fewer connections than those in this paper ( one sample t-test , ) . The top 10 models in [17] contained an average 2 . 4 modulatory connections whereas those in this paper contained an average of 4 . 5 . This difference reflects the fact that the AIC/BIC approximation to the log evidence penalizes models for each additional connection ( parameter ) without considering interdependencies or covariances amongst parameters , whereas the free energy approximation takes such dependencies into account . We now follow up the family-level inferences about input connections with Bayesian model averaging . As previously discussed , this is especially useful when the posterior model density is not sharply peaked , as is the case here ( see Figure 4 . All of the averaging results in this paper are obtained with an Occam's window defined using a minimal posterior odds ratio of . For FFX inference the input was inferred to enter region P only . We therefore restrict the averaging to those 64 models in family P . This produces 16 models in Occam's window ( itself indicating that the posterior is not sharply peaked ) . The worst one is with . The posterior odds of the best relative to the worst is only ( the largest it could be is ) , meaning these models are not significantly better than one another . Four of the models in Occam's window are shown in Figure 6 . Figure 7 shows the posterior densities of average modulatory connections ( averaging over models and subjects ) . The height of the delta functions in these histograms correspond to the total posterior probability mass of models which assume that the connection is zero . For RFX inference the input was inferred to most likely enter region P alone ( posterior exceedance probability , ) . In the RFX model averaging the Occam's windowing procedure was specific to each subject , thus each subject can have a different number of models in Occam's window . For the input model P family there were an average of models in Occam's window and Figure 8 shows the posterior densities of the average modulatory connections ( averaging over models and subjects ) . Both the RFX and FFX model averages within family P show that only connections from P to A , and from P to F , are facilitated by speech intelligibility .
This paper has investigated the formal comparison of models using Bayesian model evidence . Previous application of the method in the biological sciences has focussed on model selection in which one first selects the model with the highest evidence and then makes inferences based on the parameters of that model . We have shown that this ‘best model’ approach , though useful when the number of models is small , can become brittle if there are a large number of models , and if different subjects use different models . To overcome this shortcoming we have proposed the combination of two further approaches ( i ) family level inference and ( ii ) Bayesian model averaging within families . Family level inference removes uncertainty about aspects of model structure other than the characteristic one is interested in . Bayesian model averaging can then be used to provide a summary measure of likely parameter values for each family . We have applied these approaches to neuroimaging data , specifically a DCM study of auditory word processing using fMRI . Our results indicate that spoken words most likely stimulate a region in posterior STS and that if the word is intelligible connections are strengthened both to anterior STS and an inferior frontal region . These conclusions were drawn based on family level inference and Bayesian model averaging . The model evidence for the DCMs fitted in this paper was computed using the free energy approximation whereas previous work used ( the most conservative of ) AIC and BIC [17] . This resulted in the highly ranked models containing significantly more connections than in the previous study . This is due to a bias in the AIC/BIC criterion which leads to overly simple models being selected . Previous work in graphical models favours the free energy approach over BIC [6] and work on biochemical models finds AIS to be the best of the more computationally expensive sampling methods . The relative merits of the different model selection criteria , as applied to brain imaging models and data , will be addressed in a future publication . The family level inference procedures described in this paper can be applied whatever method is used for estimating the model evidence . Interestingly , the use of BMA produced an average network structure with speech input to region P , and modulatory connections from P to A and from P to F . This is exactly the winning model from earlier work [17] ( based on AIC/BIC approximation of model evidence ) . It is not , however , the best model as indicated by the free energy . The model with the highest free energy ( see figure 6 ( a ) ) does not , however , have significantly higher evidence than the second best model , or indeed , any model in Occam's window . This indicates that in the particular example we have studied the use of Bayes factors or posterior odds ratios would be inconclusive , whereas clear conclusions can be drawn from family level inference . This paper has also introduced a Gibbs sampling method for RFX model level inference when the number of models is large . This sampling method should be preferred to the previously suggested VB method [19] when the number of models exceeds the number of subjects ( ie . ) . We do emphasise , however , that for RFX model level inferences involving a small number of models ( as in previous work [19] ) the VB approach is perfectly valid , and is indeed the preferred approach because it is faster . The issue of family versus model level inference is orthogonal to the issue of random versus fixed effects analysis . The same critera re . FFX versus RFX apply at the family level as at the model level . For the data in this paper one might use RFX analysis as auditory word processing is part of the high level language system and one expect might expect differences in the neuronal instantiation ( eg . lateralisation ) . If the issue remains unclear one could adopt a more pragmatic approach by first implementing a FFX analysis , and if there appear to be outlying subjects , then one could follow this up with an RFX analysis . Family level inferences under FFX assumptions are simple to implement . Families with ( the same and ) different numbers of models are accommodated by setting model priors using equation 20 , model posteriors are computed using equation 7 , and family level posteriors using equation 21 . This is a simple non-iterative procedure . Family level inferences under RFX assumptions are more subtle and have been the main focus of this paper . Families with ( equal and ) unequal numbers of models are accommodated using the model priors in equation 24 , model posteriors are computed using an iterative Gibbs sampling procedure , and family level posteriors are computed using equation 22 . We envisage that family level inference under RFX assumptions will be particularly useful in neuroimaging studies of high level cognition or for clinical groups where there is a high degree of intersubject variability . Where subjects can be clearly divided into two or more groups on behavioural or other grounds ( e . g . patients and controls ) , then it would be correct to group the models accordingly , and proceed with a between group analysis on selected parameters of the averaged models . Finally , we comment on the broader issue of comparison of discrete models ( the ‘Discrete’ approach adopted in this work ) versus a hierarchical approach embodying Automatic Relevance Determination ( ARD ) in which irrelevant connections are ‘switched off’ during model fitting [41] ( for the case of DCMs the ARD approach is currently hypothetical as no such algorithm has yet been implemented ) . The ARD approach provides an estimate of the marginal density directly without recourse to Bayesian model averaging . The Discrete approach allows for quantitative family-level inferences about issues such as whether processing is serial or parallel , linear or nonlinear . Additionally , Bayesian Model Averaging can be used with the Discrete approach to provide estimates of the marginal density . Overall , the ARD approach is probably the preffered method if one is solely interested in the marginal density over parameters , because it will likely be faster . If one is additionally interested in quantitative family-level inference then the Discrete approach would be the method of choice . We expect that the comparison of model families will prove useful for a range of model comparison applications in biology , from connectivity models of brain imaging data , to behavioural models of learning and decision making , and dynamical models in molecular biology . | Bayesian model comparison provides a formal method for evaluating different computational models in the biological sciences . Emerging application domains include dynamical models of neuronal and biochemical networks based on differential equations . Much previous work in this area has focussed on selecting the single best model . This approach is useful but can become brittle if there are a large number of models to compare and if different subjects use different models . This paper shows that these problems can be overcome with the use of Family Level Inference and Bayesian Model Averaging within model families . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"neuroscience/cognitive",
"neuroscience",
"neuroscience/theoretical",
"neuroscience",
"computational",
"biology/transcriptional",
"regulation",
"computational",
"biology/computational",
"neuroscience",
"computational",
"biology/metabolic",
"networks",
"mathematics/statistics",
"computational",
"biology/systems",
"biology"
]
| 2010 | Comparing Families of Dynamic Causal Models |
Rare copy number variants ( CNVs ) are frequently associated with common neurological disorders such as mental retardation ( MR; learning disability ) , autism , and schizophrenia . CNV screening in clinical practice is limited because pathological CNVs cannot be distinguished routinely from benign CNVs , and because genes underlying patients' phenotypes remain largely unknown . Here , we present a novel , statistically robust approach that forges links between 148 MR–associated CNVs and phenotypes from ∼5 , 000 mouse gene knockout experiments . These CNVs were found to be significantly enriched in two classes of genes , those whose mouse orthologues , when disrupted , result in either abnormal axon or dopaminergic neuron morphologies . Additional enrichments highlighted correspondences between relevant mouse phenotypes and secondary presentations such as brain abnormality , cleft palate , and seizures . The strength of these phenotype enrichments ( >100% increases ) greatly exceeded molecular annotations ( <30% increases ) and allowed the identification of 78 genes that may contribute to MR and associated phenotypes . This study is the first to demonstrate how the power of mouse knockout data can be systematically exploited to better understand genetically heterogeneous neurological disorders .
Mental retardation ( MR ) is defined as an overall intelligence quotient lower than 70 , and is associated with functional deficits in adaptive behaviour , such as daily-living skills , social skills and communication . This disorder affects 1%–3% of the population and results from extraordinarily heterogeneous environmental and genetic causes [1] . Genetic changes underlying MR are still poorly resolved , especially for the autosomes that provide the largest contribution to disease aetiology [2] . Microscopically visible chromosomal rearrangements detected by routine chromosome analysis are the cause for MR in ∼5%–10% of patients [3] . Such rearrangements represent gains or losses of more than 5–10 Mb of DNA and affect many genes thereby almost inevitably leading to developmental abnormalities during embryogenesis . The most common effect of these variants is cognitive impairment , but they can also be frequently associated with other abnormalities such as heart defects , seizures and dysmorphic features [4] . Many recent genomic microarray studies have indicated that smaller , submicroscopic rearrangements , such as copy number variations ( CNVs ) , frequently underlie MR ( Table S1 ) . However , CNVs , defined as DNA deletions or duplications greater than 1 Kb [5] , are also widespread in the general population which considerably hinders the clinical interpretation of patients' CNVs [6] . Until now , most clinical CNV studies have focused on the identification of rare de novo CNVs [7]–[9] , as the rate of de novo large ( >50 kb ) CNVs in the general population is comparatively low [10] , [11] . Nevertheless , discriminating between benign and pathogenic CNVs solely on the basis of size and lack of inheritance is crude and provides no insights into how CNVs exert their phenotypic effects . Fortunately , the genomics era has amassed a wealth of data that have long promised to associate the disruption of a particular molecular function or cellular pathway with clinical observations; in short , to forge links between genotype and disease phenotype . These genomic data include behavioural , physiological and anatomical examinations following the disruption of more than 5000 individual mouse genes [12]–[14] . These mouse phenotypic measurements more closely resemble observations from human clinical examination than any other systematic genome-wide data source . They might be especially relevant to human gene deletion variants , which represent a large majority among the rare disease-associated CNVs considered here ( Table 1 and Table S2 ) . Available genomic data also include functional annotations such as from the Gene Ontology resource [15] , tissue expression levels [16] and carefully curated pathway data such as the Kyoto Encyclopedia of Genes and Genomes ( KEGG ) [17] . Our approach was to test the null hypothesis that genes present in MR–associated CNVs randomly sample all human genes . In particular , are they a random sample of genes ( i ) that , when disrupted in mice , result in particular phenotypes , or ( ii ) that are predominantly expressed in the human brain , or ( iii ) that participate in specific human disease pathways ? To ensure that we correctly account for the application of multiple tests , we have controlled the false discovery rate ( FDR ) [18] such that there is only a small 5% likelihood that any annotation term has been identified as over-represented in our tests simply by chance . Only if any particular set of genes present within MR–associated CNVs form a significantly ( FDR<5% ) non-random sample can we be truly justified in predicting single genes , among the dozens commonly overlapped by such CNVs , as contributing to MR disease aetiology . In this study , we show both significant and substantial enrichments in phenotypic annotations whose power in predicting pathoetiology greatly exceeds that of molecular annotations .
We first tested whether MR–associated CNVR genes were enriched in 33 major categories of mouse phenotypes ( see Materials and Methods ) . Although for All MR–associated CNVRs none of these terms was significant , the set of Loss MR–associated CNVRs showed a strong and significant enrichment in genes whose knockouts in mice produced a nervous system phenotype ( +13 . 6% , or 1 . 14-fold , enrichment , p = 3×10−3 , FDR<5%; Figure 1 ) . An enrichment of genes associated with nervous system phenotypes was not observed within the Gain CNVRs ( +0 . 2% ) . Given the significant enrichment within the Loss set , we then tested this set against each of 147 finer-scale mouse nervous system phenotypes . Two of these terms were significantly enriched ( FDR<5% ) : abnormal axon morphology ( obs = 19 , exp = 7 . 1 , +170% enrichment , p = 3×10−5 ) , and abnormal dopaminergic neuron morphology ( obs = 9 , exp = 2 . 5 , +260% enrichment , p = 3×10−4 ) ( Figure 1 ) . Both of these mouse neural phenotypes are relevant to human MR phenotypes owing to these mouse phenotype's abnormalities in neuronal and cerebral cortex morphologies ( see Discussion ) . Within Gain CNVRs , we observe a non-significant enrichment of genes associated with abnormal axon morphology ( obs = 6 , exp = 2 . 7 , +120% enrichment , p = 5×10−2 ) but a non-significant depletion of genes associated with abnormal dopaminergic neuron morphology ( obs = 0 , exp = 0 . 95 , −100% deficit , p = 0 . 38 ) . The neurological phenotypes of MR patients suggested that MR–associated CNVs might contain an unusually high density of genes that , when mutated , are involved in human neurological disease . Considering those genes classified by KEGG to be involved in 6 neurodegenerative pathways , we indeed found MR–associated CNVRs to be significantly enriched in genes involved in the Parkinson's disease pathway ( obs = 8 , exp = 2 . 7 , +196% enrichment , p = 3×10−3 , FDR<5%; Figure 2 ) . While enrichments of this pathway's genes were observed both for Loss CNVRs ( obs = 7 , exp = 2 . 1 , +230% enrichment , p = 3×10−3 , FDR<5% ) and for Gain CNVRs ( obs = 2 , exp = 0 . 8 , +151% enrichment , p = 0 . 19 ) , significance was reached only for Loss CNVRs . As Parkinson's disease is a condition characterized by the degeneration and dysfunction of dopaminergic neurons [19] , these enrichments corroborate our finding that orthologues of genes whose disruption in mouse gives rise to abnormal dopaminergic neuron morphology are enriched in MR–associated CNVRs ( see above ) . The allelic changes underlying MR phenotypes might also be expected to preferentially involve ‘brain-specific’ genes , those that are highly expressed in the human brain relative to other human tissues . Indeed , All MR–associated CNVRs were significantly enriched in brain-specific genes ( +24% enrichment , p = 1×10−2; Figure 3 ) , specifically for Loss ( +31% enrichment , p = 8×10−3 ) but not for Gain CNVs ( +4% enrichment , p = 0 . 45 ) . The significant enrichments observed when testing mouse phenotypes are thus corroborated by enrichments in human gene expression . These findings would have little or no predictive potential if apparently ‘benign’ CNVs ( those present in the general human population ) also exhibit such biases . However , in contrast to the above results , benign CNVs show no significant enrichments of ( i ) genes that are highly-expressed in the brain ( −11% deficit , p = 0 . 2; Figure 3 ) , ( ii ) genes present in neurodegenerative disease pathways ( −32% deficit , p = 0 . 1; Figure 2 ) , or ( iii ) genes with nervous system phenotypes when disrupted in mice ( −11% deficit , p = 0 . 01; Figure 1 ) . Instead , benign CNV genes show significant tendencies to encode proteins with roles in immunity and host defense [20] , [21] . Each of these three features thus may be exploited to distinguish MR–associated CNVR genes from benign CNVR genes . MR–associated and benign CNVs show no significant tendency to overlap ( p = 0 . 1 ) . Nevertheless , by excluding all genes in MR–associated CNVs whose gain/loss-matched copy number change is also seen in benign CNVs we enhanced the discrimination of genes whose copy number change is predicted to contribute to MR aetiology . This was specifically the case for mouse fine-scale nervous system phenotypes and human neurodegenerative disease pathways ( Figure 1 and Figure 2 ) . Moreover , after excluding benign CNV-overlapped genes , not only Parkinson's disease pathway genes , but genes from 5 other neurodegenerative disease pathways ( namely , Alzheimer's disease , Amyotrophic Lateral Sclerosis , Huntington's disease , Dentatorubropallidoluysian atrophy and Prion Diseases ) when considered together , became significantly enriched ( +60% enrichment; p = 0 . 02 ) in this analysis . These results would be explained if MR-causative alleles segregate more with sequence that is copy number variable in MR individuals than with CNVs observed in the general population . We considered whether our method could identify significant associations between mouse and human patient phenotypes other than MR . We investigated 7 clinical features that were present in our patient population in addition to the MR phenotype , namely brain- , cleft palate- , eye- , facial- , heart- or urogenital- abnormalities and seizures ( see Materials and Methods ) . We tested whether CNVs from individuals with these specific clinical features were significantly enriched in genes associated with phenotypically-relevant mouse phenotypes . In order to limit the large number of statistical tests that could be performed we matched mouse phenotype categories ( each containing between 129 and 220 terms ) to each of the 7 clinical features based on clinical experience ( see Materials and Methods ) before performing the association tests . We found that 4 of the 7 additional clinical features were significantly associated ( FDR<5% ) with between 1 and 6 mouse phenotypic terms ( Figure 4 ) . For example , the CNVRs of the 8 MR patients presenting with cleft palate were significantly enriched with genes whose mouse orthologues , when disrupted , also exhibited cleft palate ( Figure 4 ) . Importantly , no significant associations were observed between CNVs from humans without a particular clinical feature apart from MR and any mouse phenotype category matched to patients with that clinical feature , with the notable exception of ‘abnormal axon morphology’ that thus appears to be a term of broad relevance to the primary MR presentation ( Figure 4 ) . These findings demonstrate the relevance of mouse gene knockout observations to both the MR phenotype and associated phenotypes in patients . The distinctions between MR–associated and benign CNVR genes , described above , allowed the identification of genes whose copy number change may contribute to MR and associated phenotypes . To identify such candidate genes , we could not exploit Gene Ontology annotations ( Figure S1 ) or brain expression enrichments ( Figure 3 ) as these enrichments provide insufficient discriminatory power ( <30% increase over expected ) . Of the 4 , 009 genes present in the 148 MR–associated CNVs , 55 are annotated with either a mouse knockout phenotype ( n = 29 ) and/or a neurodegenerative disease pathway ( n = 29 ) that was significantly over-represented in MR–associated Loss CNVRs ( Table 2 ) . 50 of the MR–associated CNVs ( 33% ) contain at least 1 of these 55 candidate genes . We calculate that our list represents a ∼120% increase of likely phenotype-contributing genes over the random expectation ( see Materials and Methods ) . Similarly , 34 genes were identified as potential candidates for additional clinical features such as cleft palate , facial or brain abnormalities , or seizures , 23 of which were not associated with MR itself ( Table 2 ) . We note that whilst some of these candidate genes might have been prioritized from among the 4 , 009 CNVRs genes using a priori subjective expectations , our method is the first to generate a candidate gene set on the basis of objective and statistically sound criteria .
If de novo MR–associated CNVs do not contribute to disease etiology their gene contents would not be expected to exhibit biases in gene function or expression . Instead , we demonstrate the first evidence for significant tendencies of MR–associated CNV genes to be brain-expressed , to belong to neurodegenerative pathways , and to present particular phenotypes when disrupted in mice , all of which validate the assumption that large de novo CNVs commonly underlie MR phenotypes . These results could not have been obtained without collating data from a number of sources . For example , essentially all ( 147 of 148 ) CNVs were required to obtain a significant enrichment of genes whose mouse orthologues' knockout produced a nervous system phenotype ( Figure S2 ) . It was only by harnessing the statistical power of a research community's large data set that this meta-analysis achieved significance of statistical associations ( see Materials and Methods ) . The significant signals seen in Loss CNVs , but not in Gain CNVs , imply that MR phenotypes commonly result from gene dosage sensitivity ( haploinsufficency ) . However , we cannot discount that they may occur from the uncovering , by DNA loss , of rare recessive alleles . While we did not observe an enrichment within the Gain CNVRs of genes associated with abnormal dopaminergic neuron morphology or of genes that showed brain-specific expression , we did observe non-significant enrichments of genes associated with abnormal axon morphology and of Parkinson's disease pathway genes . Given that the Gain CNVRs overlap 38% of the number of genes overlapped by the Loss CNVRs ( Table 1 ) , it is plausible that these enrichments might reach significance as more Gain MR–associated CNVs are reported and analysed . Our results are in contrast with previously-reported sporadic and familial cases of MR whose associated genes are enriched in both X-chromosome location and enzymatic function [22] . Nevertheless , this is explained by Wright's physiological theory of dominance: haplosufficient genes , such as those lying on the X chromosome , have an expected tendency to encode enzymes , whereas haploinsufficient genes , such as those expected to underlie our autosomal MR disorders , have an expected tendency to encode transcription regulatory genes [23] . Indeed , we do observe a significant enrichment of genes associated with transcriptional regulation within MR–associated CNVRs ( Figure S1 ) . In contrast to X-linked MR genes , of which approximately one quarter encode postsynaptic proteins [24] , we observe a small and non-significant depletion ( p = 0 . 39 ) of postsynaptic protein genes among our MR–associated CNVs . None of the human CNVs recorded in this study represent homozygous losses . Thus it may initially appear problematic to compare human phenotypes directly with those from mice harbouring homozygous gene disruptions . Nevertheless , without sequence information confirming the genetic integrity of the surviving haplotype we cannot be certain that these human hemizygous loss CNVs do not contain independent disruptions of each allelic copy . To gain some insight into this issue we considered 21 of the 55 candidate genes that contribute to a significantly enriched mouse knock-out phenotype identified in our study ( Table 2 ) , and whose phenotype has been recorded in the MGI resource when in the hemizygous state . Of these 21 , four ( namely , En1 , Mn1 , Plp1 and Pmp22 ) also exhibit the phenotype of interest when hemizygously disrupted [25]–[28] . Of the remaining 17 genes , all exhibit abnormal phenotypes , and thus are haploinsufficient , with the exceptions of Mapt and Slc6a3 [29] , [30] . Importantly , these mouse hemizygous phenotypes are often closely-related to the homozygous phenotypes , while some hemizygous phenotypes appear particularly relevant to the associated human phenotype . For example , Scn1a ( which contributes to the tremors phenotypic enrichment we find to be associated with patients presenting with seizures ) exhibits a seizures phenotype when in the hemizygous state in mice [31] . Does our analysis allow us to link particular mouse gene knockout phenotypes to human CNV phenotypes ? Obviously , a direct comparison between mouse neural phenotypes and human MR phenotypes is hindered because the invasive procedures of brain biopsies in patients are unacceptable . Results from a limited number of post-mortem studies of MR patients suggest that abnormalities of dendritic spines are a general neuropathological feature of MR [32] . The mouse gene knockout phenotypes do provide a plausible explanation for the brain phenotypes observed in some patients as a consequence of the structural variation identified in their genomes . An example of this is the myelin-associated glycoprotein ( MAG ) gene that is deleted in one patient ( case 123 , Table S2 ) and duplicated in another ( case 124 ) , whilst the knockout of its orthologous gene in mice leads to both abnormal axon morphology and tremors phenotypes [33] . Underexpression of MAG in transfected Schwann cells is known to lead to hypomyelinisation [34] . Therefore , the delayed brain myelinisation observed in the patient with the MAG deletion could be caused by under-expression of MAG during brain development . By contrast , over-expression of MAG is known to lead to accelerated myelinisation [35] . Whether the macrocephaly in the patient with the MAG duplication is related to over-expression of MAG during brain development remains unknown . Our enrichment analysis revealed 8 genes associated with cleft palate in humans , present in 6 different patients ( cases 10 , 13 , 27 , 48 , 96 , and 141 ) . Seven of these genes were located in Loss CNVs on human chromosomes 1p31 . 1p31 . 3 ( containing LHX8 ) , 1q41q42 . 13 ( DISP1 ) , 2q24 . 3q31 . 1 ( DLX1 , DLX2 and GAD1 ) , 4q31 . 21q31 . 23 ( EDNRA ) and 22q12 . 1 ( MN1 ) , and one with a Gain CNV on human chromosome 16p13 . 2–p13 . 3 9 ( CREBBP ) . Except for DISP1 , all these genes have been associated with cleft palate in mouse models [26] , [36]–[39] , whereas only LHX8 and GAD1 have been associated with cleft palate disorders in humans [40] , [41] . This strongly suggests that our approach revealed 6 novel orofacial cleft ( OFC ) candidate genes in humans . Strikingly , the hemizygous loss of five of these OFC candidate genes may also contribute to MR . Absence of both Dlx1 and Dlx2 in mice results in abnormal differentiation within the forebrain [36] , [42] . Both genes also regulate Arx , a homeobox transcription factor required for the migration of interneurons , whose human equivalent ARX , when mutated , is associated with X-linked MR and epilepsy [43] . In addition , mutations and deletions of CREBBP causes the Rubinstein-Taybi syndrome which is characterized by MR [44] . Ednra is involved in cranial neural crest cell migration from the posterior midbrain and hindbrain to the arches [45] . Lhx8 is required for the development of many cholinergic neurons in the mouse forebrain [46] , whereas GAD1 , which encodes the GABA-producing enzyme , may play a role in the development and plasticity of the central nervous system [39] . In conclusion , it appears that our approach identified a large number of interesting and plausible novel candidate genes for both MR and associated clinical phenotypes . Mouse phenotype data have not previously been exploited in a systematic genome-wide analysis , and our results clearly show its utility in addressing a particularly difficult and contemporary challenge in the field of neurological genomic disorders . The functional biases we see for MR–associated CNV genes can now be exploited to prioritise genes for further investigation in MR individuals without large de novo CNVs ( Table 2 ) . We suggest that all human genes whose orthologues present specific phenotypes when disrupted in mice ( Figure 1 ) deserve particular scrutiny for fine-scale insertion , deletion or point mutations contributing to MR . Mouse orthologue knockout data are available currently for only ∼25% of all human genes . More specifically , of the 4 , 009 genes overlapped by the MR–associated CNVs considered here , 830 ( ∼21% ) have available phenotypic annotations . Thus , we would expect that many more candidate genes possessing these annotations will be discovered within MR–associated CNVs as further knockouts are generated . Furthermore , we consider all genes that are involved in the specific molecular pathways we have identified , such as Parkinson's disease and other neurodegenerative disorder pathways , to represent candidates for MR and/or associated phenotypes when hemizygous . We propose that the contribution of these candidate genes ( Table 2 ) to many MR phenotypes can now be investigated thoroughly in mouse model systems: specifically , the 55 genes whose hemizygous deletions may be associated with MR are now amenable to study using hemizygous knockout mouse models . Our study has exploited CNVs identified using several different platforms . As the identification technologies have improved , CNVs called using earlier technologies have been shown to over-estimate the true extent of a CNV's boundaries [47] . Thus , we expect enhanced resolution of pathogenic CNVs to also increase the power by which genic enrichments can be identified . However , it should also be noted that CNVs have been shown to affect the expression of neighbouring genes and it is possible that pathogenic CNVs may exert their genetic effect through outlying genes [48] . Finally , there is no reason why this approach can not be applied successfully to other complex neurological diseases , including schizophrenia and autism , which show a high frequency of rare de novo CNVs [8] , [9] , [49]–[51] . Many studies that are currently under-powered to demonstrate significance after correcting for multiple testing may yet prove informative of the genetic etiology of complex genomic disorders . For this , it will be crucial to collect large disease-associated CNV sets from well-phenotyped cohorts , as our analysis has shown that only then is there sufficient power to detect significant associations ( Figure S2 ) .
For this study we collected 148 rare structural variants associated with MR from the literature , the Decipher database ( https://decipher . sanger . ac . uk/ ) , as well as from our own in-house diagnostic microarray group [52] ( Table S1 ) . The majority of these CNVs ( n = 135 , 91% ) were proved to have occurred de novo in the patient and all were independently validated . Thirteen rare autosomal CNVs for which parental samples were unavailable were included , as were seven rare maternally inherited CNVs on the X chromosome in male patients that are considered to be as clinically relevant as de novo CNVs on the autosomes . Importantly , at the point of discovery none of these CNVs were known to greatly ( >50% ) overlap with a collection of >15 , 000 CNVs identified in healthy individuals as collected in the Database of Genomic Variants version 3 ( http://projects . tcag . ca/variation/ ) . All CNVs were mapped to NCBI35 coordinates . The median number of Entrez genes within a CNV was 35 . Overlapping CNVs were merged to obtain a non-redundant set of 112 CNV regions ( CNVRs ) totalling 440 Mb of unique sequence ( 14 . 3% of the total NCBI35 human genome assembly; Table 1 ) . CNVR sets were also formed separately from Gain and from Loss CNVs ( Table 1 ) . For 121 of the 148 CNVs , information regarding distinct anatomical or physiological abnormalities presented by the patient in addition to MR was available ( Table S2 ) . These clinical features were used to form 7 non-exclusive groupings for additional tests . We obtained 25 , 196 CNVs identified in 270 individuals from Redon et al . [11] . To these , we added 1 , 276 inherited CNVs identified in 494 individuals with a 32 k BAC tiling path array . This last set is described in Nguyen et al . [53] and , together with the Koolen et al . [52] MR–associated CNV data , are available from the Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/ ) with accession number GSE7391 . Combined , these apparently benign CNVs represent 430 Mb of unique sequence ( 14 . 0% of the total NCBI35 human genome assembly; Table 1 ) . In the absence of information suggesting that any of the individuals present with MR , we conservatively assume that genes overlapped by these apparently benign CNVs do not contribute to the MR phenotypes . Assignment of protein-coding genes depended upon the particular analysis performed: for protein-coding gene counts and the Gene Ontology analysis , we assigned genes to CNVs according to Ensembl [54] ( Ensembl mart version 37 ) , whereas for KEGG pathway and MGI analyses we assigned genes to CNVs according to Entrez genes [55] . Information on human NCBI genes whose mouse orthologues' disruption had been assayed were obtained from the Mouse Genome Informatics ( MGI ) resource ( http://www . informatics . jax . org , version 3 . 54 ) [12]–[14] . We employed the MGI's human/mouse orthology and marker assignment to map MGI mouse marker phenotypes to Human Entrez genes [55] . We mapped , using unambiguous gene orthology relationships , 5 , 075 different MGI phenotypic annotation terms to 4 , 999 human genes . We considered all phenotypic annotations from all experimental methodologies described within the MGI resource . While the vast majority of these annotations are derived from the disruption of mouse genes , some phenotypes were derived from experiments in which mutant alleles are introduced into the mouse ( e . g . [56] ) . Nonetheless , we regard the phenotypic information from these experiments as remaining informative of the biological functions or pathways to which the gene contributes . It is noted , however , that the phenotypes of all genes underlying the phenotypic enrichments we report in this work ( Figure 1 and Figure 2; Table 2 ) were obtained through gene disruption experiments . The MGI phenotypic annotations are categorised non-exclusively into 33 over-arching terms ( Table S3 ) . When examining finer phenotypic terms beneath an over-arching term ( s ) we considered only those finer terms that possessed at least 1% of the genes annotated with the over-arching term ( s ) . This allowed a reduction in the number of tests performed thereby limiting spurious and uninformative results . The phenotypes associated with the Entrez genes overlapped by a given set of genomic regions were compared to the frequency of that phenotype across the whole genome . All p-values were obtained by application of the hypergeometric test and were subject to a false discovery rate ( FDR ) of <5% [18] ( see below ) . Given the large number of phenotypic terms and the unrealistic assumption of terms' independence when applying an FDR , application of this significance threshold is likely to be conservative . Many of the MR patients used in this study show additional clinical features . We tested for associations between commonly occurring non-MR clinical features in patients and a subset of MGI phenotypes . We scored patients for the presence of 7 common features derived from the London Dysmorphology Database [57] . These were: ( i ) seizures/abnormal EEG , ( ii ) facial dysmorphism , ( iii ) cleft palate , ( iv ) heart , general abnormalities , ( v ) eye abnormalities , ( vi ) brain , general abnormalities , and ( vii ) urogenital system abnormalities . Patients were excluded if specific phenotypic data were unavailable ( all 19 cases from the Decipher database ) . As these secondary clinical feature-grouped CNVs were fewer in number than the entire set of MR–associated CNVs , and therefore relatively diminished in statistical power , the most relevant MGI phenotypic categories were selected ( from a total of 33; Table S3 ) in order to reduce the number of tests . Two pairs of paralogous genes , DLX1 & DLX2 and SELE & SELP , contributed to the significant phenotypic enrichments reported within the secondary clinical feature grouped CNVs ( Table 2 ) . However , significant phenotypic enrichments that these pairs of paralogues contributed to all remained significant after removing one of the paralogous pairs ( p<0 . 05; single test ) . Nevertheless , we note that an increased penetrance of a resulting phenotype might be expected if these pairs of paralogues provided a degree of redundancy to one another , and therefore the concurrent copy number variation of both paralogues may prove even more significant than variation involving only one [42] . Annotations of genes involved in neurodegenerative pathways were obtained from KEGG [17] . KEGG genes were collated if they belonged to KEGG Pathways section 5 . 3 , namely Alzheimer's disease ( KEGG pathway 05010 ) , Parkinson's disease ( KEGG pathway 05020 ) , Amyotrophic Lateral Sclerosis ( KEGG pathway 05030 ) , Huntington's disease ( KEGG pathway 05040 ) , Dentatorubropallidoluysian atrophy ( KEGG pathway 05050 ) and Prion Diseases ( KEGG pathway 05060 ) . KEGG genes were mapped to NCBI Entrez genes using associations provided by KEGG . For human gene expression data , we used GNF's gene atlas data for the MAS5-condensed human U133A and GNF1H chips , considering all 74 non-cancer tissues [16] . Expression levels were mapped to LocusLink identifiers and to 11 , 594 Ensembl Ensmart 37 ( NCBI35 ) genes using the annotation tables supplied by GNF . To identify genes that are highly expressed in the brain we selected those genes whose expression in the whole brain exceeded by 4-fold their median expression in all other non-brain tissues after excluding cancerous tissues . This resulted in 435 genes ( 3 . 75% ) being classified as exhibiting strong expression in the brain relative to other tissues . However , the significant enrichments reported in the Results were also found when brain-specificity was redefined at 2- , 3- , 7- , 10- , 11- , 12- , 13- , and 14-fold expression in the brain above the median across all other tissues . A set of postsynaptic protein genes was obtained from Collins et al . [58] and matched to human orthologues using Ensembl Compara [59] . Over- or under-representation of these genes within human CNVs was assessed using the hypergeometric distribution and all human Ensembl genes as the background set . The significance of enrichments or deficits of genes associated with particular MGI knockout phenotypes , genes involved in KEGG neurodegenerative pathways , genes associated with particular GO terms and brain-specific genes were evaluated using hypergeometric tests . Where multiple tests were performed , a False Discovery Rate ( FDR ) multiple testing correction was applied to ensure a less than 5% likelihood of any significant term being a false-positive [18] . Explicitly , an FDR correction was applied when testing for enrichments of genes: ( i ) associated with MGI phenotypic terms , ( ii ) belonging to individual KEGG neurodegenerative pathways or ( iii ) annotated with Gene Ontology terms ( Figure S1 ) . All other tests performed were single tests . Calculation of the fold-enrichment within MR–associated CNVs for the final set of 55 MR–associated candidate genes was performed by random sampling . 1000 gene sets , matched in gene number to that within the Loss MR–associated CNVRs , were obtained by random sampling and the median expected number of genes , 23 ( std . dev . = 4 . 6 ) , annotated with one or more significantly-enriched terms ( Figure 1 and Figure 2 ) was recorded . Given the 50 candidate genes within the Loss CNVRs , we thus estimate a ∼2 . 2-fold enrichment over the number expected by chance . | Mental retardation ( MR; also known as learning disability ) affects 1%–3% of people and is often associated with the presence of genomic copy number variations ( CNVs ) such as deletions and duplications . Most of these CNVs are rare and they often involve tens , sometimes hundreds , of genes . Pinpointing exactly which particular gene or genes are responsible for MR in an individual patient is therefore challenging and limits diagnostic applications . In this study , the functions of genes present within a large collection of MR–associated CNVs were investigated by comparing them to data from large-scale mouse knock-out experiments . We found that MR–associated CNVs contain greater than expected numbers of genes that give specific nervous system phenotypes when disrupted in the mouse . Not only does this study confirm that CNVs frequently cause MR , but it narrows down the list of genes whose changes lead to this disorder from thousands to several dozen . This reduced list of genes brings wide-spread genetic testing for MR one step closer . It also provides a better understanding of the biology behind MR that could , eventually , yield medical treatments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"neurological",
"disorders/cognitive",
"neurology",
"and",
"dementia",
"genetics",
"and",
"genomics/functional",
"genomics",
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"neurological",
"disorders/neuropsychiatric",
"disorders",
"neurological",
"disorders/developmental",
"and",
"pediatric",
"neurology",
"genetics",
"and",
"genomics/bioinformatics",
"genetics",
"and",
"genomics/medical",
"genetics"
]
| 2009 | Forging Links between Human Mental Retardation–Associated CNVs and Mouse Gene Knockout Models |
The olfactory organ of vertebrates receives chemical cues present in the air or water and , at the same time , they are exposed to invading pathogens . Nasal-associated lymphoid tissue ( NALT ) , which serves as a mucosal inductive site for humoral immune responses against antigen stimulation in mammals , is present also in teleosts . IgT in teleosts is responsible for similar functions to those carried out by IgA in mammals . Moreover , teleost NALT is known to contain B-cells and teleost nasal mucus contains immunoglobulins ( Igs ) . Yet , whether nasal B cells and Igs respond to infection remains unknown . We hypothesized that water-borne parasites can invade the nasal cavity of fish and elicit local specific immune responses . To address this hypothesis , we developed a model of bath infection with the Ichthyophthirius multifiliis ( Ich ) parasite in rainbow trout , Oncorhynchus mykiss , an ancient bony fish , and investigated the nasal adaptive immune response against this parasite . Critically , we found that Ich parasites in water could reach the nasal cavity and successfully invade the nasal mucosa . Moreover , strong parasite-specific IgT responses were detected in the nasal mucus , and the accumulation of IgT+ B-cells was noted in the nasal epidermis after Ich infection . Strikingly , local IgT+ B-cell proliferation and parasite-specific IgT generation were found in the trout olfactory organ , providing new evidence that nasal-specific immune responses were induced locally by a parasitic challenge . Overall , our findings suggest that nasal mucosal adaptive immune responses are similar to those reported in other fish mucosal sites and that an antibody system with a dedicated mucosal Ig performs evolutionary conserved functions across vertebrate mucosal surfaces .
Olfaction is a vital sense for all animals [1] . To receive an olfactory signal , terrestrial vertebrates inhale gases containing volatile chemical substances , while aquatic vertebrates like teleost fish actively draw water containing dissolved chemicals into the olfactory organs [2] . Simultaneously , during this process , the olfactory organs are constantly stimulated by toxins and pathogens in the air or water [3] . Therefore , there is an evident need to defend the large , delicate surface of olfactory organs from pathogenic invasion . In mammals , nasopharynx-associated lymphoid tissue ( NALT ) , is a paired mucosal lymphoid organ containing well-organized lymphoid structures ( organized MALT , O-MALT ) and scattered or disseminated lymphoid cells ( diffuse MALT , D-MALT ) and is traditionally considered the first line of defense against external threats [4] . Similar to the Peyer’s patches in the guts of mammals , O-NALT has distinct B-cell zones [5 , 6] , and humoral immune responses occur in response to infection or antigenic stimulation [7] . Importantly , the higher percentage of IgA+ B-cells in D-NALT compared with that in O-NALT indicates that D-NALT may play an important role in nasal antibody-mediated immunity [8] . Interestingly , NALT in early vertebrates like teleost fish has structures and components similar to those of mammalian NALT [1] . Teleost NALT has thus far been described as D-NALT but lacks O-NALT . Teleost NALT includes B-cells , T cells , myeloid cells and expresses innate and adaptive related molecules [9] . Thus , from an evolutionary viewpoint , NALT in teleost fish is equipped to rapidly respond to antigens present in the water environment [3] . Teleost fish represent the most ancient bony vertebrates with a nasal-associated immune system [10] and containing immunoglobulins ( Igs ) [9] . So far , only three Ig classes ( IgM , IgD , and IgT/Z ) have been identified in teleosts [11] . Teleost IgM has been considered the principal Ig in plasma , and strong parasite-specific IgM responses have been induced in systemic immunity [12–14] . Although secreted IgD ( sIgD ) has been found in the coating of a small percentage of the microbiota at the gill mucosa surface , its function remains unknown [15] . In contrast , teleost IgT ( also called IgZ in some species ) has been identified at the genome level and found to play a specialized role in response to pathogen infection in mucosal tissues [15–17] . Moreover , IgT+ B-cells represent the predominant mucosal B-cell subset , and the accumulation of IgT+ B-cells has been detected after infection in trout gut- , skin- , and gill-associated lymphoid tissues ( GALT , SALT , and GIALT ) [15–17] . Interestingly , in mammals , parasite-specific IgA has been mainly induced after pathogenic infection , and it has mediated nasal-adaptive immunity [18–20] . However , in teleosts , the role of the three Ig classes and B-cells in the olfactory organ is still unknown . Thus , given the abundance of IgT+ B-cells as well as the high concentration of IgT in the olfactory organ [9 , 21] , we hypothesized that IgT is the major Ig involved in the pathogen-specific immune responses in the NALT of teleost fish . To test the aforementioned hypothesis , here , we studied the nasal B-cell and parasite-specific Ig responses to the ciliated parasite Ichthyophthirius multifiliis ( Ich ) in rainbow trout , a model species in the field of evolutionary and comparative immunology [22 , 23] . Our findings show that the olfactory system of rainbow trout is an ancient mucosal surface that elicits strong innate and adaptive immune responses to Ich infection . In addition , we demonstrate that IgT is the main Ig isotype playing a critical role in nasal adaptive immune responses . Furthermore , we show for the first time the local production of IgT at the nasal mucosa and proliferation of IgT+ B-cells after a parasitic challenge in the olfactory organ of teleost fish . These results demonstrate that NALT is both an inductive and effector immune site in teleost fish .
Here , nasal mucosa IgT was detected by Western blot consistent with the reported molecular mass using anti-trout IgT antibody [15–17] . To understand the protein characterization of nasal IgT , we collected the nasal mucosa ( 2 . 7 mg/ml ) of rainbow trout and loaded 0 . 5 ml of processed mucus into a gel filtration column . From these results , we found that a portion of IgT in the nasal mucosa was present in polymeric form , as it eluted at a fraction similar to that of trout nasal IgM , a tetrameric Ig ( Fig 1A ) , and simultaneously , some IgT was consistently eluted in monomeric form . Next , nasal mucosa polymeric IgT ( pIgT ) migrated to the same position as a monomer by SDS-PAGE under non-reducing conditions , indicating that nasal pIgT is associated by non-covalent interactions ( Fig 1B , right panel ) . However , unlike IgM and IgT , IgD in nasal mucosa eluted at 8 . 5 and 9 . 5 as a monomer at the molecular weight range previously studied for serum IgD [15] . Using Western blot , under the same immunoblot conditions , we found that nasal IgD and IgM migrated as a monomer and polymer , respectively ( Fig 1B , left and middle panel ) . These findings are similar to those previously reported in the gut [16] , skin [17] , and gill [15] . Finally , using Western blot , we compared and analyzed the concentrations of three Igs and the ratios of IgT/IgM and IgD/IgM in nasal mucus and serum ( Fig 1C–1F ) , respectively . Our results showed that the protein concentration of IgT was ~ 164- and ~ 602-fold lower than that of IgM in nasal mucus and serum , respectively ( Fig 1C and 1D ) . Although IgM was found to be the highest Ig in nasal mucosa , the IgT/IgM ratio in nasal mucus was ~ 4-fold higher than that in serum ( Fig 1E ) , whereas there was no obvious difference between them in terms of the IgD/IgM ratio ( Fig 1F ) . In mammals , pIgR can mediate the transepithelial transport of secretory IgA ( sIgA ) into the nasal mucosa [24 , 25] . In trout , we previously found that the secretory component of trout pIgR ( tSC ) is associated with secretory IgT ( sIgT ) in the gut [16] , skin [17] , and gills [15] and pIgR expression is very high in control rainbow trout olfactory organ [9] . Here , using pIgR polyclonal antibody [16] , tSC was detected in the nasal mucosa but not in the serum ( Fig 2A ) . By coimmunoprecipitation assay and immunoglobulins , we showed that antibodies against trout IgT was able to coimmunoprecipitate tSC in nasal mucus ( Fig 2B ) . Moreover , using immunofluorescence microscopy , most of the pIgR-containing cells were located in the OE of trout , some of which were stained with IgT ( Fig 2C; isotype-matched control antibodies ( S1A Fig ) . To evaluate whether the trout olfactory organ expresses immune-related genes after parasite challenge , we firstly selected the Ich parasite bath infection model and ( S2A Fig ) . Ich is a parasite that directly invades the mucosal tissues of fish , such as the skin , gill , and fin , and it might elicit a strong immune response [15 , 17] , however , it has never been reported to infect the olfactory organ of fish . At 7 days post-infection , the phenotype of the small white dots appeared on the trout’s skin and fin surface ( S2B Fig ) , and by examining paraffin sections of olfactory organs stained with H & E , the Ich parasite was found within the nasal cavity and mucosa , interestingly , most of which were present in lateral regions compared with tips of nasal lamina propria ( S2C Fig ) . In addition , by reverse transcription quantitative real-time PCR ( RT-qPCR ) , we detected the expression of Ich-18SrRNA in the olfactory organ , gills , skin , head kidney , and spleen of trout after 7 days infection and controls ( S2D Fig ) . Ich-18SrDNA expression levels were comparable in the nose and the gills , one of the target organs of Ich , highlighting the importance of Ich nasal infections in trout . A time series study of Ich-18SrRNA expression showed that parasites levels in the nose peaked at day 7 post-infection with a second wave occurring at day 21 . Interestingly , Ich levels dropped dramatically on day 28 but increased again 75 days post-infection ( S2E Fig ) . Using RT-qPCR , we measured the expression of 26 immune-related genes in the olfactory organ of trout at days 1 , 7 , 14 , 21 , 28 , and 75 post-infection . Overall , greatest changes in expression of pro-inflammatory and complement-related genes , occurred at 7 days post-infection ( Fig 3A , primers used are shown in S1 Table ) when parasite levels were highest . Expression of IgT and IgM heavy chain genes , in turn , increased later during infection , starting at day 7 and peaking at day 28 , whereas no obvious change in IgD heavy chain expression was observed ( Fig 3A and 3B ) . IgT expression was the most up-regulated ( ~ 258-fold ) compared to IgM ( ~ 116-fold ) on day 28 and remained up-regulated on day 75 ( ~ 112-fold ) compared to IgM which was only moderately higher than controls ( ~ 8-fold ) ( Fig 3B ) . Moreover , using histological examination , the lamina propria ( LP ) at the tip of the nasal lamella ( ~ 100 μm away from the apex ) showed a significant enlargement ( Fig 3C and 3D ) and numbers of goblet cells on the nasal lamella increased significantly in the fish at day 7 after infection with Ich ( Fig 3C and 3E ) . Interestingly , by days 28 and 75 , the tissue reaction was smaller , the LP showed some enlargement and abundant goblet cells appeared in nasal lamella ( Fig 3C–3E ) . Combined , these results demonstrate that apart from infecting gills , and skin , Ich is able to chronically infect the trout olfactory organ and induce strong long-lasting IgT responses . The high expression of IgT in the olfactory organ of trout after an Ich parasite challenge led us to hypothesize a critical role of IgT in nasal immunity . Using immunofluorescent micrographs , Ich trophonts could be easily detected in the olfactory organ of trout after 28 days of infection ( Fig 4 ) using an anti-Ich antibody ( isotype-matched control antibodies , S3 Fig ) . Interestingly , most parasites detected in the olfactory organ of trout were intensely coated with IgT , while only some parasites were slightly coated with IgM and nearly no parasites were coated with IgD ( Fig 4 ) . In addition , we found few IgT+ and IgM+ B-cells in the nasal epithelium of control fish ( Fig 5A; isotype-matched control antibodies , S1B Fig left ) . Interestingly , a moderate increase of IgT+ B-cells was observed in the nasal epithelium of trout after 28 days of infection ( Fig 5B; isotype-matched control antibodies , S1B Fig middle ) . It is worth mentioning that a notable accumulation of IgT+ B-cells was detected in the nasal epithelium of survivor fish ( 75 days post-infection ) compared with control trout ( Fig 5C; isotype-matched control antibodies , S1B Fig right ) . In contrast , the abundance of IgM+ B-cells did not change in the infected and survivor fish when compared to the controls ( Fig 5A–5C ) . Next , we analyzed the percentages of IgT+ and IgM+ B-cells in the olfactory organs of control , infected , and survivor fish . We observed that , similar to the result obtained by immunofluorescence microscopy , the percentages of IgT+ B-cells in the infected group ( ~ 3 . 66 ± 0 . 2% ) and survivor group ( ~ 4 . 43 ± 0 . 28% ) increased significantly compared to those of the control group ( ~ 1 . 72 ± 0 . 08% ) ( Fig 6A ) . In contrast , the percentages of nasal IgM+ B-cells did not change in the three groups ( Fig 6A ) . Unlike the results in the olfactory organ , the percentage of IgM+ B-cells of the head kidney in the infected group ( ~ 11 . 34 ± 0 . 39% ) showed a significant increase compared to that of the control group ( ~ 6 . 53 ± 0 . 27% ) , while the percentage of IgM+ B-cells in the survivor group ( ~ 8 . 65 ± 0 . 67% ) showed no significant change ( Fig 6B ) . In contrast , the percentages of IgT+ B-cells remained unchanged in both the infected groups and the survivor groups ( Fig 6B ) . In agreement with the increased IgT+ B-cells observed in the olfactory organ of infected and survivor fish , the IgT concentration in the nasal mucosa of these fish increased by ~ 2- and ~ 6-fold when compared with control fish , respectively . However , IgM and IgD protein concentrations did not change in any fish groups ( Fig 6C ) . In serum , a ~ 3-fold and ~5-fold increase in IgT concentration was observed in the infected and survivor group , respectively , whereas that of IgM in both the infected and survivor group increased by ~ 5-fold with respect to control fish ( Fig 6D ) . As expected , the IgD protein concentration did not change significantly in infected or survivor fish ( Fig 6D ) . The results of large increases of IgT+ B-cells and IgT protein levels in the olfactory organs of infected and survivor fish , together with the observation that parasites in the olfactory organ of infected fish appear intensely coated with IgT , suggested that parasite-specific IgT might be secreted in the nasal mucosa response to Ich infection . To verify this hypothesis , using a pull-down assay , we measured the capacity of nasal Igs to bind the Ich parasite ( Fig 7 ) . We found a significant increase in parasite-specific IgT binding in up to 1/10 ( ~ 3 . 2-fold ) and 1/100 ( ~ 2 . 8-fold ) of the diluted nasal mucus of infected ( Fig 7B ) and survivor fish ( Fig 7C ) , respectively , when compared to that of control fish . However , in serum ( Fig 7D–7F ) , parasite-specific IgT binding was detected only in 1/10 dilution of survivor fish ( Fig 7F ) . In contrast , parasite-specific IgM binding in up to 1/1000 and 1/4000 of the diluted serum of infected ( Fig 7E ) and survivor fish ( Fig 7F ) increased by ~ 2 . 9-fold and ~ 4 . 3-fold , respectively . Finally , in the nasal mucosa and serum of both the infected and survivor fish , Ich-specific IgD showed no change when compared to that of control fish ( Fig 7A–7F ) . To further evaluate whether an increase of IgT+ B-cells in the olfactory organ of survivor fish that survived 75 days after first infection was derived from the process of local IgT+ B-cell proliferation or from influx of B cells from systemic lymphoid organs , we performed in vivo proliferation studies of IgT+ B-cells and IgM+ B-cells stained with EdU , which can incorporate into DNA during cell division [26] . Immunofluorescence microscopy analysis showed a significant increase in the percentage of proliferating cells in the olfactory organ of survivor fish ( ~ 0 . 048 ± 0 . 0006% ) when compared with that of control animals ( ~ 0 . 019 ± 0 . 0003% ) ( Fig 8A and 8B ) . Interestingly , we detected a significant increase in the proliferation of EdU+ IgT+ B-cells in survivor fish ( ~ 5 . 21 ± 0 . 23% ) when compared with that of the control fish ( ~ 0 . 58 ± 0 . 05% ) ( Fig 8A and 8C ) . However , no difference was found in the percentage of EdU+ IgM+ B-cells of control fish and survivor fish ( Fig 8A–8C ) . Using flow cytometry , similar results were obtained ( S4A Fig ) , with the percentage of EdU+ IgT+ B-cells increased significantly in the olfactory organ of survivor fish ( ~ 5 . 67 ± 0 . 10% in all IgT+ B-cells ) when compared with that of control fish ( ~ 3 . 46 ± 0 . 26% in all IgT+ B-cells ) , while no difference in the percentage of EdU+ IgM+ B-cells was detected between control and survivor fish ( S4A Fig ) . In the head kidney , the percentage of EdU+ IgM+ B-cells of the olfactory organ was detected in survivor fish , and it presented a large increase when compared with that of control fish . In contrast , these two groups showed no difference in proliferating IgT+ B-cells ( S4B Fig ) . The local proliferation of IgT+ B-cells in the olfactory organ and the detection of parasite-specific IgT in nasal mucus ( Fig 7 ) suggest that specific IgT in the trout olfactory organ is locally generated rather than produced and transported from systemic lymphoid organs . To further address this hypothesis , we measured parasite-specific Igs titers from medium of cultured olfactory organ , head kidney , and spleen explants from control and survivor fish ( Fig 9 ) . We detected parasite-specific IgT binding in 1/40 diluted medium ( ~ 3 . 6-fold ) of cultured olfactory organ explants of survivor fish , whereas low parasite-specific IgM titers were detected only at the 1/10 dilution in the same medium ( Fig 9A and 9D ) . In contrast , dominant parasite-specific IgM binding ( up to 1/40 dilutions ) was observed in the medium of head kidney and spleen explants , and low parasite-specific IgT responses were detected in the same medium ( Fig 9B–9F ) . Interestingly , negligible parasite-specific IgD titers were detected in the medium of cultured olfactory organ , head kidney , and spleen explants from control and survivor fish ( Fig 9A–9F ) . Together , our results showed that specific IgT responses were highest in the olfactory organ , although specific IgT was also detected in head kidney and spleen .
Protozoans are the most common parasites of freshwater and marine fish [27–29] . Ich is one of the most problematic parasites in freshwater ecosystems infecting many different fish species [30 , 31] . Ich has been traditionally associated with skin and gill lesions in rainbow trout [15 , 17] , however , the teleost olfactory organ is constantly exposed to the aquatic environment and therefore may represent a route of entry for any pathogen . Here we report for the first time that Ich can infect the olfactory organ of rainbow trout when the fish are exposed to the parasite by bath , the natural route of exposure . Importantly , we found that parasite loads in the olfactory organ were the highest along with the gills , suggesting that the olfactory route of infection may be one of the main targets of this parasite . Given that the impacts of Ich invasion via the nose have until now been overlooked , further investigations are required to determine the impacts of Ich nasal infections in the fish host health . The olfactory organ of teleosts , similar to that of mammals , is coated by mucus containing Igs . In this study , we characterized in detail all three Ig classes in the nasal mucus of rainbow trout , including sIgD , secreted IgM ( sIgM ) , and secreted IgT ( sIgT ) . Trout nasal IgT existed for the most part as a polymer , similar to the characterized IgA in the nasal mucosa from humans [32 , 33] . On the contrary , nasal IgD was in monomeric form , as previously reported in the gill [15] and serum [34] . Interestingly , in agreement with the descriptions for gut [16] , skin [17] , and gill [15] sIgT , all subunits of polymeric nasal IgT in rainbow trout were associated by noncovalent interactions . In addition , we detected the concentrations of all three Igs in nasal mucosa and serum and found that although the concentration of IgT was lower than that of IgM in both nasal mucosa and serum , the ratio of IgT/IgM in nasal mucosa was higher than that in serum , in agreement with a previous report by Tacchi et al [9] . Combined , these findings underscore that mucus secretions in teleosts consist of mixtures of all three Ig isotypes and that Ig protein concentrations of each isotype differ among the four teleost MALT [15–17] as they do in mammals [35–37] . In mammals , NALT has been considered a mucosal inductive site for IgA [38–40] . Yet , it is not clear whether in fish , which lack organized lymphoid structures ( adenoids and tonsils ) in the teleost olfactory organ [9] , NALT acts as an inductive and/or effector mucosal lymphoid tissue . In our Ich infection model , we found large increases in the concentration of IgT but not IgM or IgD at the protein level in the nasal mucosa of infected and surviving fish exposed to Ich , which correlated with the large accumulation of IgT+ but not IgM+ B-cells appearing in the olfactory epidermis of the same fish . These results might indicate that the large concentration of IgT was secreted by the accumulation of IgT+ B-cells in the olfactory epidermis . In addition , we showed a striking abundance of IgT coating on the Ich parasite surface in the olfactory organ of rainbow trout . However , much lower or negligible levels of IgM or IgD coating were detected on the same parasites . These results suggested that a strong IgT but not IgM response to Ich takes place in the local olfactory environment . Interestingly , similar results were also discovered in our previous studies in the gut , skin , and gills [15–17] . In mammals , a dramatic increase of IgA secretion and significant accumulations of IgA-antibody forming cells ( IgA-AFC ) were induced in the nasal mucosa following intranasal infection with a small volume of influenza virus [18] and Naegleria fowleri parasite [41–43] , respectively . Based on our findings , it is clear that teleost NALT is a mucosal inductive site . Whether NALT-induced IgT+ B-cell and plasma cell responses seed effector sites such as the gut lamina propria remains to be characterized in this or other models . Finally , our results strengthen the notion that despite anatomical differences and the absence of organized NALT structures in teleost , IgT and IgA carry out vital roles in nasal adaptive immune responses . Immunoglobulins are of particular relevance in the context of Ich infections since previous studies have demonstrated that antibody ( IgM ) mediated responses against Ich i-antigen trigger the exit of the parasite from the fish host skin conferring host disease resistance [44 , 45] . Similarly , in our model , Ich was being expulsed at day 28 and minority stay in nasal cavity , but interestingly , expelled Ich was mainly coated by IgT . In addition , we recorded the greatest upregulation in the expression of the IgT heavy chain gene in the trout olfactory organ 28 days after Ich exposure , the same time point when Ich levels dropped dramatically , suggesting that IgT might play a crucial role in the nasal immune response to Ich infection and may contribute to parasite clearance or exit . At this point , IgM expression levels had also increased in the olfactory organ and some detectable titers of parasite-specific IgM were found in trout nasal mucosa . Nasal IgM titers might be the result of Ich-instigated microlesions in the olfactory system and consequent leakage of parasite-specific IgM or plasma from the blood . Thus , specific IgT responses appear to be the most critical antibody response against Ich in the nasal environment and further studies should address how IgT contributes to parasite clearance from trout mucosal surfaces . Interestingly , even though IgT expression was still high at 75 days infection , we detected an increase in Ich loads and the downregulation in the expression of the IgM heavy chain gene at this time point , suggesting that the reduction in IgM expression might affect the ability of IgT to combat Ich in the olfactory organ . Therefore , further studies will investigate the potential cooperation between both immunoglobulins during parasite infection . IgD , a primordial antibody , exists almost in all the teleosts [46] . Previous studies have shown that transcripts of IgD are primarily distributed in both systemic and mucosal lymphoid tissues , including head kidney , spleen , blood , liver , gill , skin and gut [46–49] . IgD gene expression is differentially modulated following challenge with different aquatic pathogens in several fish species [46–54] . Although these findings may suggest that IgD plays a role in both systemic immunity and mucosal immunity , its function remains poorly understood due to lack of clear responses recorded at the protein level . Interestingly , here the gene expression of IgD did not change in the olfactory organ following bath infection with Ich parasite , suggesting that Ich does not elicit IgD nasal immune responses . Moreover , negligible parasite-specific IgD responses were induced in both the nasal mucosa and serum after Ich challenges , similar to previous studies in gills . However , because of the detectable concentrations of IgD in both the nasal mucosa and serum , we cannot exclude the possibility that relevant IgD may be induced in the nasal mucosa or systemic compartment when using different pathogens or stimulation routes . Thus , future studies are needed to investigate the role of nasal and systemic IgD in the parasite-specific immune responses of teleost fish against different pathogens . The accumulations of IgT+ B-cells observed in the olfactory epidermis correlated with high parasite-specific IgT titers in the same fish led us to hypothesize the local proliferation and production of the parasite-specific IgT+ B-cell response . IgT+ B-cell proliferation responses were detected in the olfactory organ but not in the systemic immune organs ( head kidney and spleen ) of the same fish , which strongly suggests that the accumulation of IgT+ B-cells in the olfactory organ is due to local proliferation rather than migration from other organs . We also show that olfactory tissue explants produce specific anti-Ich IgT antibodies , demonstrating the presence of specific plasma cells in the local nasal mucosa . Interestingly , these results parallel our previous finding in the trout gill , and the proliferation rates we detected here was similar to the ones we have described in gills [15] but higher than those in olfactory organ in response to IHNV [21] , which might be due to the different duration after infection/immunization with different pathogen , respectively . It is worth noting that similar results were found in the NALT of mammals . For instance , previous studies have shown that intranasal immunization with N . fowleri could induce the secretion of IgA and IgG in nasal mucosa but pathogen-specific IgA mainly mediates local nasal immunity in mammals [55 , 56] . By in vitro culture of NALT cells following virus infection , parts of the virus-specific antibody-forming cells ( AFCs ) were observed to originate from B-cell precursors in NALT [18] . Moreover , in the nasal mucosa from 53 humans with chronic inflammation , most IgA seemed to be produced locally by IgA-producing plasma cells [57] . Hence , our results indicate that the local proliferation of mucosal B-cells and production of secretory Ig responses in the nasal mucosa happens not only in tetrapod species but also in early vertebrates such as teleost fish . The fish olfactory mucosa is a complex neuroepithelium in which lymphoid and myeloid cells are found in a scattered manner [9] among basal cells , sustentacular cells , olfactory sensory neurons , goblet cells and epithelial cells ( Fig 10 ) . In agreement with histological changes , strong immune responses including the upregulation of cytokine expression and complement genes were detected in the olfactory organ , especially at the early stages of the infection , preceding the onset of Ig responses . Interestingly , Ich stimulation resulted in changes in cytokine gene expression such as IL-10 , IL-6 and IL-11 . This cytokine profile largely resembles that previously found in mammalian models in the nose [41] . Both local nasal immune cells and recruited immune cells in the olfactory epithelium may be the sources of these cytokines [41] . Additionally , we cannot rule out the contribution of other cell types including epithelial cells , goblet cells and different neuronal populations . Future studies will address what the roles of nasal cytokine responses may be and how they aid in the production of specific antibodies against parasites . Thus , despite the lack of organized lymphoid structures in the olfactory organs of teleosts [3 , 9] , teleost fish might mount strong cytokine responses or enhance the specific humoral immunity upon pathogen invasion or immunization with antigenic analogues . In conclusion , our results provide the first evidence that parasite infection , antigen presentation , local B-cell activation and proliferation , as well as parasite-specific IgT production occur in the olfactory organ of teleost fish ( Fig 10 ) . Thus , although parasites such as Ich can infect the olfactory organ of fish , local IgT+ B-cells and parasite-specific IgT appear to be a major mechanism by which the host acquires resistance to this parasite . Our findings not only expand our view of nasal immune systems from an evolutionary perspective but also suggest that nasal vaccination may be an effective way to prevent aquatic parasitic diseases .
All experimental protocols involving fish were performed in accordance with the recommendations in the Guide for the care and use of Laboratory Animals of the Ministry of Science and Technology of China and were approved by the Scientific Committee of Huazhong Agricultural University ( permit number HZAUFI-2016-007 ) . All efforts were made to minimize the suffering of the animals . Rainbow trout ( 20–30 g ) were obtained from fish farm in Shiyan ( hubei , China ) , and maintained them in aquarium tanks using a water recirculation system involving thermostatic temperature control and extensive biofiltration . Fish were acclimatized for at least 2 wk at 15 °C and fed daily with commercial trout pellets at a rate of 0 . 5–1% body weight day-l , and feeding was terminated 48 h prior to sacrifice . The method used for Ich parasite isolation and infection were described previously by Xu et al [17] with slight modification . Briefly , heavily infected rainbow trout were anaesthetized with overdose of MS-222 ( 250 mg/L ) and placed in a beaker with water to allow trophonts and tomonts exit the fish . The trophonts and tomonts were left in the water at 15 °C for 24 h to let tomocyst formation and subsequent theront release . For parasite infection ( as shown in S2A Fig ) , two types of challenges with Ich were performed . The first group , fish were exposed to a single dose of ~ 7500 theronts per L for 3 hours and then migrated into the aquarium containing new aquatic water . Tissue samples and fluids ( serum and nasal mucus ) were taken after 28 days ( infected fish ) . The second group , fish were monthly exposed during 75 days period ( survival fish ) with the same dose . Fish samples were taken two weeks after the last challenge . Experiments were performed at least three independent times . Control fish ( mock infected ) were maintained in a similar tank but without parasites . During the whole experiment periods , the fish were raised in a flow through aquaria at 15 °C and fed daily with commercial trout pellets at a rate of 0 . 5–1% body weight day-l . For sampling , trout were anaesthetized with MS-222 and serum was collected and stored as described [16] . To obtain fish nasal mucus , we modified the method described previously [9 , 15] . Briefly , trout olfactory tissue was excised rinsed with PBS three times to remove the remaining blood . Thereafter , olfactory tissue was incubated for 12 h at 4 °C , with slightly shaking in protease inhibitor buffer ( 1 × PBS , containing 1 × protease inhibitor cocktail ( Roche ) , 1 mM phenylmethylsulfonyl fluoride ( Sigma ) ; pH 7 . 2 ) at a ratio of 100 mg of olfactory tissue per ml of buffer . The suspension ( nasal mucus ) was transferred to an Eppendorf tube , and then the supernatant was vigorously vortexed and centrifuged at 400 g for 10 min at 4 °C to remove trout cells . Furthermore , the olfactory organ was taken and fixed into 4% neutral buffered formalin for hematoxylin and eosin ( H & E ) staining and immunostaining . The leucocytes from head kidney were obtained using a modified methodology as described previously [15 , 58] . To obtain trout nasopharynx-associated lymphoid tissue ( NALT ) leukocytes , we modified the existing protocol as explained by Tacchi et al [9] . Briefly , rainbow trout were anaesthetized with MS-222 ( 50 mg/L ) and blood was collected from the caudal vein . The olfactory organ was taken and washed with cold PBS to avoid blood contamination . Leucocytes from trout olfactory organ were isolated by mechanical agitation of both olfactory rosettes in DMEM medium ( supplemented with 5% FBS , 100 U ml -1 penicillin and 100 μg ml -1 streptomycin ) at 4 °C for 30 min with continuous shaking . Leukocytes were collected , and the aforementioned procedure was repeated four times . Thereafter , the olfactory organ pieces were treated with PBS ( containing 0 . 37 mg ml -1 EDTA and 0 . 14 mg ml -1 dithiothreitol DTT ) for 30 min followed by enzymatic digestion with collagenase ( Invitrogen , 0 . 15 mg ml -1 in PBS ) for 1 h at 20 °C with continuous shaking . All cell fractions obtained from the olfactory organ after mechanical and enzymatic treatments were washed three times in fresh modified DMEM and layered over a 51/34% discontinuous Percoll gradient . After 30 min of centrifugation at 400 g , leucocytes lying at the interface of the gradient were collected and washed with modified DMEM medium . Nasal mucus ( 40 μl ) and serum ( 0 . 5 μl ) samples were resolved on 4–15% SDS-PAGE Ready Gel ( Bio-Rad ) under non-reducing or reducing conditions as described previously [15–17] . For western blot analysis , the gels were transferred onto PVDF membranes ( Bio-Rad ) . Thereafter , the membranes were blocked with 8% skim milk and incubated with anti-trout IgT ( rabbit pAb ) , anti-trout IgM ( mouse monoclonal antibody ( mAb ) ) or biotinylated anti-trout IgD ( mouse mAb ) antibodies followed by incubation with peroxidase-conjugated anti-rabbit , anti-mouse IgG ( Invitrogen ) or streptavidin ( Invitrogen ) . Immunoreactivity was detected with an enhanced chemiluminescent reagent ( Advansta ) and scanned by GE Amersham Imager 600 Imaging System ( GE Healthcare ) . The captured gel images were analysed by using ImageQuant TL software ( GE Healthcare ) . Thereafter , the concentration of IgM , IgD and IgT were determined by plotting the obtained signal strength values on a standard curve generated for each blot using known amounts of purified trout IgM , IgD or IgT . To analysis the monomeric or polymeric state of Igs in trout nasal mucus , gel filtration analyses were performed using as described previously for gut [16] and gill mucus [15] . In short , fractions containing the IgM , IgT or IgD were separated by gel filtration using a Superdex-200 FPLC column ( GE Healthcare ) . The column was previously equilibrated with cold PBS ( pH 7 . 2 ) , and protein fractions were eluted at 0 . 5 ml min -1 with PBS using a fast protein LC instrument with purifier systems ( GE Healthcare ) . Identification of IgM , IgD and IgT in the eluted fractions was performed by western blot analysis using anti-IgM , anti-IgD and anti-IgT antibodies , respectively . A standard curve was generated by plotting the elution volume of the standard proteins in a Gel Filtration Standard ( Bio-Rad ) against their known molecular weight , which was then used to determine the molecular weight of the eluted IgT , IgM and IgD by their elution volume . For flow cytometry studies of B cells in the head kidney and NALT , leukocyte suspensions were double-stained with monoclonal mouse anti-trout IgT and anti-trout IgM ( 1 μg ml −1 each ) at 4°C for 45 min . After washing three times , PE-goat anti-mouse IgG1 and APC-goat anti-mouse IgG2b ( 1 μg ml −1 each , BD Biosciences ) were added and incubated at 4 °C for 45 min to detect IgM+ and IgT+ B cells , respectively . After washing three times , analysis of stained leucocytes was performed with a CytoFLEX flow cytometer ( Beckman coulter ) and analysed by FlowJo software ( Tree Star ) . The olfactory organ of rainbow trout was dissected and fixed in 4% neutral buffered formalin ( 1:10 ) overnight at 4 °C and then transferred to 70% ethanol . Samples were then embedded in paraffin and 5 μm thick sections stained with haematoxylin / eosin ( H & E ) . Images were acquired in a microscope ( Olympus ) using the Axiovision software . For the detection of Ich parasite at the same time of IgT+ and IgM+ B cells , sections were double-stained with rabbit anti-trout IgT ( pAb; 0 . 49 μg ml-1 ) and mouse anti-trout IgM ( IgG1 isotype; 1 μg ml-1 ) overnight at 4°C . After washing three times , secondary antibodies Alexa Fluor 488-conjugated AffiniPure Goat anti-rabbit IgG or Cy3-conjugated AffiniPure Goat anti-mouse IgG ( Jackson ImmunoResearch Laboratories Inc . ) at 2 . 5 μg ml-1 each were added and incubated at temperature for 40 min to detect IgT+ and IgM+ B cells , respectively . After washing three times , mouse anti-Ich polyclonal antibody ( 1 μg ml-1 ) were added and incubated at 4 °C for 6 h , after washing three times , secondary antibody Alexa Fluor 647-goat anti-mouse ( Jackson ImmunoResearch Laboratories Inc . ) with 5 μg ml-1 were added and incubated at temperature for 40 min to detect Ich parasite . For detection of trout nose pIgR , we used the same methodology described to stain gill pIgR by using our rabbit anti-pIgR [16] . As controls , the rabbit IgG pre-bleed and the mouse-IgG1 isotype antibodies were labelled with the same antibody labelling kits and used at the same concentrations . Before mounting , all samples were stained with DAPI ( 4′ , 6-diamidino-2-phenylindole; 1 μg ml-1: Invitrogen ) for the sections . Images were acquired and analysed using Olympus BX53 fluorescence microscope ( Olympus ) and the iVision-Mac scientific imaging processing software ( Olympus ) . For proliferation of B cells studies , we modified the methodology as previously reported by us [15] . Briefly , control and survivor fish ( ~ 30g ) were anaesthetized with MS-222 and intravenous injected with 200 μg EdU ( Invitrogen ) . After 24 h , leucocytes from head kidney or olfactory tissue were obtained as described above , and cells were incubated with 10 μM of EdU ( Invitrogen ) for 2 hours . Thereafter , leucocytes were incubated with mAb mouse anti-trout IgM and anti-trout IgT ( 1 μg ml-1 each ) at 4 °C for 45 min . After washing three times , Alexa Fluor 488-goat anti-mouse IgG ( Invitrogen ) was used as secondary antibody to detect IgM+ or IgT + B cells . After incubation at 4 °C for 45 min , cells were washed three times with DMEM medium and fixed with 4% neutral buffered formalin at room temperature for 15 min . EdU+ cell detection was performed according to the manufacturer’s instructions ( Click-iT EdU Alexa Fluor 647 Flow Cytometry Assay Kit , Invitrogen ) . Cells were thereafter analysed in a CytoFLEX flow cytometer ( Beckman coulter ) and FlowJo software ( Tree Star ) . For immunofluorescence analysis , as described above , we used the paraffin sections of olfactory organ from control and survival fish previously injected with EdU and incubated at 4 °C for 45 min with rabbit anti-trout IgT ( pAb; 1 μg ml-1 ) and mouse anti-trout IgM ( IgG1 isotype; 1 μg ml-1 ) . After washing with PBS , paraffin sections were incubated for 2 h at room temperature with Alexa Fluor 488-conjugated AffiniPure Goat anti-rabbit IgG or Cy3-conjugated AffiniPure Goat anti-mouse IgG ( Jackson ImmunoResearch Laboratories Inc . ) at 2 . 5 μg ml-1 each . Stained cells were fixed with 4% neutral buffered formalin and EdU+ cell detection was performed according to the manufacturer’s instructions ( Click-iT EdU Alexa Fluor 647 Imaging Kit , Invitrogen ) . Cell nuclei were stained with DAPI ( 1 μg ml-1 ) before mounting with fluorescent microscopy mounting solution . Images were acquired and analysed using an Olympus BX53 fluorescence microscope ( Olympus ) and the iVision-Mac scientific imaging processing software ( Olympus ) . To assess whether the parasite-specific IgT responses were locally generated in the olfactory organ , we analysed parasite-specific immunoglobulin titers from medium derived of cultured olfactory organ , head kidney and spleen explants obtained from control and survivor fish as previously described by us [15] . In short , control and survivor fish were anaesthetized with an overdose of MS-222 , and blood was removed through the caudal vein to minimize the blood content in the collected organs . Thereafter , olfactory organ , head kidney and spleen were collected . Approximately 20 mg of each tissue was submerged in 70% ethanol for 1 min to eliminate possible bacteria on their surface and then washed twice with PBS . Thereafter , tissues were placed in a 24-well plate and cultured with 200 ml DMEM medium ( Invitrogen ) , supplemented with 10% FBS , 100 U ml -1 penicillin , 100 μg ml -1 streptomycin , 200 μg ml -1 amphotericin B and 250 μg ml -1 gentamycin sulfate , with 5% CO2 at 17 °C . After 7 days culture , supernatants were harvested , centrifuged and stored at 4 °C prior to use the same day . The capacity of IgT , IgM and IgD from serum , nasal mucus or tissue ( olfactory organ , head kidney and spleen ) explant supernatants to bind to Ich was measured by using a pull-down assay as described previously [15 , 17] . Briefly , approximately 100 tomonts were pre-incubated with a solution of 0 . 5% BSA in PBS ( pH 7 . 2 ) at 4 °C for 2 h . Thereafter , tomonts were incubated with diluted nasal mucus or serum or tissue ( olfactory organ , head kidney and spleen ) explant supernatants from infected , survivor or control fish at 4 °C for 4 h with continuous shaking in a 300 ml volume . After incubation , the tomonts were washed three times with PBS and bound proteins were eluted with 2 × Laemmli Sample Buffer ( Bio-Rad ) and boiled for 5 min at 95 °C . The eluted material was resolved on 4–15% SDS-PAGE Ready Gel under non-reducing conditions , and the presence of IgT , IgM or IgD was detected by western blotting using anti-trout IgT , IgM or IgD antibodies as described above . Total RNA was extracted by homogenization in 1 ml TRIZol ( Invitrogen ) using steel beads and shaking ( 60 HZ for 1 min ) following the manufacturer’s instructions . The quantification of the extracted RNA was carried out using a spectrophotometry ( NanoPhotometer NP 80 Touch ) and the integrity of the RNA was determined by agarose gel electrophoresis . To normalize gene expression levels for each sample , equivalent amounts of the total RNA ( 1000 ng ) were used for cDNA synthesis with the SuperScript first-strand synthesis system for qPCR ( Abm ) in a 20 μl reaction volume . The synthesized cDNA was diluted 4 times and then used as a template for qPCR analysis . The qPCRs were performed on a 7500 Real-time PCR system ( Applied Biosystems ) using the EvaGreen 2 × qPCR Master mix ( Abm ) . All samples were performed following conditions: 95 °C for 30 s , followed by 40 cycles at 95 °C for 1 s and at 58 °C for 10 s . A dissociation protocol was carried out after thermos cycling to confirm a band of the correct size was amplified . Ct values determined for each sample were normalized against the values for housekeeping gene ( EF1α ) . To gain some insights on the kinetics of the immune responses that takes place after Ich infection , twenty-six immune relevant genes , such as cytokine , complement and Igs genes were detected in the olfactory organ . The relative expression level of the genes was determined using the Pfaffl method [59] . The primers used for qRT-PCR are listed in S1 Table . We followed the same strategy to detect the association of pIgR to IgT in gut , skin and gill mucus as we previously described [15–17] . To detect whether polymeric trout IgT present in the nasal mucus were associated to a secretory component-like molecule derived from trout secretory component–like molecule ( tSC ) , we performed co-immunoprecipitating analysis using anti-trout IgT ( pAb ) antibodies with the goal to potentially co-immunoprecipitate the tSC . To this end , 10 μg of anti-IgT were incubated with 300 μl of trout nasal mucus . As control for these studies , the same amount of rabbit IgG ( purified from the pre-bleed serum of the rabbit ) were used as negative controls for anti-IgT . After overnight incubation at 4 °C , Dynabeads Protein G ( 10001D; 50μl; Invitrogen ) prepared previously was added into each reaction mixture and incubated for 1 h at 4 °C following the manufacturer’s instructions . Thereafter , the beads were washed five times with PBS , and subsequently bound proteins were eluted in 2 × Laemmli Sample Buffer ( Bio-Rad ) . The eluted material was resolved by SDS-PAGE on 4–15% Tris-HCl Gradient ReadyGels ( Bio-Rad ) under reducing ( for tSC detection ) or non-reducing ( for IgT detection ) conditions . Western blot was performed with anti-pIgR and anti-IgT antibodies as described above . An unpaired Student’s t-test and one-way analysis of variance with Bonferroni correction ( Prism version 6 . 01; GraphPad ) were used for analysis of differences between groups . Data are expressed as mean ± s . e . m . P values less than 0 . 05 were considered statistically significant . | The olfactory organ is a vitally important chemosensory organ in vertebrates but it is also continuously stimulated by pathogenic microorganisms in the external environment . In mammals and birds , nasopharynx-associated lymphoid tissue ( NALT ) is considered one of the first lines of immune defense against inhaled antigens and in bony fish , protecting against water-borne infections . However , although B-cells and immunoglobulins ( Igs ) have been found in teleost NALT , the defensive mechanisms of parasite-specific immune responses after pathogen challenge in the olfactory organ of teleost fish remain poorly understood . Considering that the NALT of all vertebrates has been subjected to similar evolutionary forces , we hypothesize that mucosal Igs play a critical role in the defense of olfactory systems against parasites . To confirm this hypothesis , we show the local proliferation of IgT+ B-cells and production of pathogen-specific IgT within the nasal mucosa upon parasite infection , indicating that parasite-specific IgT is the main Ig isotype specialized for nasal-adaptive immune responses . From an evolutionary perspective , our findings contribute to expanding our view of nasal immune systems and determining the fate of the host–pathogen interaction . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"fish",
"immune",
"cells",
"body",
"fluids",
"immune",
"physiology",
"immunology",
"vertebrates",
"parasitic",
"diseases",
"animals",
"olfactory",
"organs",
"osteichthyes",
"respiratory",
"system",
"antibodies",
"immune",
"system",
"proteins",
"white",
"blood",
"cells",
"mucus",
"animal",
"cells",
"proteins",
"immune",
"response",
"trout",
"antibody-producing",
"cells",
"biochemistry",
"eukaryota",
"anatomy",
"cell",
"biology",
"b",
"cells",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"nasal",
"mucosa",
"organisms"
]
| 2018 | Mucosal immunoglobulins protect the olfactory organ of teleost fish against parasitic infection |
Single-stranded polyanions ≥40 bases in length facilitate the formation of hamster scrapie prions in vitro , and polyanions co-localize with PrPSc aggregates in vivo [1] , [2] . To test the hypothesis that intact polyanionic molecules might serve as a structural backbone essential for maintaining the infectious conformation ( s ) of PrPSc , we produced synthetic prions using a photocleavable , 100-base oligonucleotide ( PC-oligo ) . In serial Protein Misfolding Cyclic Amplification ( sPMCA ) reactions using purified PrPC substrate , PC-oligo was incorporated into physical complexes with PrPSc molecules that were resistant to benzonase digestion . Exposure of these nuclease-resistant prion complexes to long wave ultraviolet light ( 315 nm ) induced degradation of PC-oligo into 5 base fragments . Light-induced photolysis of incorporated PC-oligo did not alter the infectivity of in vitro-generated prions , as determined by bioassay in hamsters and brain homogenate sPMCA assays . Neuropathological analysis also revealed no significant differences in the neurotropism of prions containing intact versus degraded PC-oligo . These results show that polyanions >5 bases in length are not required for maintaining the infectious properties of in vitro-generated scrapie prions , and indicate that such properties are maintained either by short polyanion remnants , other co-purified cofactors , or by PrPSc molecules alone .
Infectious prion diseases such as Creutzfeldt Jakob disease ( CJD ) and other related human disorders , chronic wasting disease ( CWD ) , bovine spongiform encephalopathy ( BSE ) , and scrapie are associated with the conversion of a host-encoded glycoprotein ( PrPC ) into a misfolded conformer , PrPSc [3] . Several biochemical studies utilizing the protein misfolding cyclic amplification ( PMCA ) technique have shown that PrPSc is an essential component of the infectious agent [4] , [5] , [6] . Interestingly , a variety of prion strains with distinctive infectious phenotypes , such as selective neurotropism and characteristic disease incubation times , have been isolated and propagated in vivo and in vitro [7] , [8] . In some cases , infection with specific prion strains produces PrPSc molecules with distinctive biochemical characteristics , suggesting that multiple self-propagating PrPSc conformations may provide the structural basis for the existence of multiple prion strains [9] , [10] . Several biochemical and cell culture studies have implicated polyanions such as single stranded nucleic acid and glycosaminoglycan ( GAG ) molecules as potent cofactors in the process of infectious prion propagation [1] , [5] , [6] , [11] , [12] , [13] , [14] , [15] , [16] , [17] . Moreover , it has been shown that polyanions are selectively incorporated into physical complexes with purified PrP molecules in infectious hamster prions , and that a minimum length corresponding to a 40 base single stranded oligonucleotide is required for a polyanion to facilitate hamster prion formation in vitro [1] . Neuropathological analysis of scrapie-infected animals has shown that nucleic acids and GAG-containing proteoglycans co-localize with PrPSc aggregates in situ [1] , [2] . Collectively , these observations raise the possibility that moderately long polyanions might be needed to maintain hamster prion infectivity or strain properties by acting as a structural support for PrPSc molecules . Nuclease-resistant RNA molecules have been detected in purified prion preparations [18] , and analytical methods suggest that the average size of polynucleotides in such preparations is ≤25 nucleotides , assuming a ratio of 1 polynucleotide molecule per infectious unit [19] . More recently , Jeong et al . reported reduction of prion infectivity and PrPSc by treatment of hamster 263K scrapie brain homogenates with LiAlH4 ( lithium aluminum hydride ) , a highly reactive reducing agent capable of degrading RNA molecules [20] . However , because LiAlH4 is a non-specific reducing agent , which can react with a variety of biological molecules other than RNA , including proteins , it is not possible to ascribe definitively the detrimental effect of LiAlH4 on prion infectivity to the degradation of RNA . Previous photo-irradiation studies of naturally occurring prions [21] , [22] suggested that specific nucleic acid sequences are not necessary for prion infectivity , and this conclusion was confirmed by the de novo formation of purified prions using synthetic , homopolymeric poly ( A ) RNA molecules [5] . However , these prior studies did not exclude a structural , non-coding role for polyanions such as RNA in maintaining the infectious conformation of PrPSc since ultraviolet ( UV ) light mutates pyrimidine bases but does not significantly degrade the polynucleotide backbone of nucleic acids or other types of polyanions such as GAGs . Here we report the results of studies using purified synthetic prions containing a photocleavable ( dT ) 100 oligonucleotide ( PC-oligo ) , which can be selectively degraded in situ into 5-mers by exposure to long wave UV light . This approach directly addresses the question of whether polyanions might serve as a structural backbone for infectious prions in a chemically defined model system .
Hamster prion strain Sc237 was kindly provided by Dr . Stanley Prusiner ( UCSF , San Francisco , CA ) . Three-week-old female Golden Syrian hamsters used in inoculation experiments and seven-week-old Golden Syrian hamsters used to make brain homogenates were purchased from Charles River ( Wilmington , MA ) . Monoclonal antibody ( mAb ) 6D11 was purchased from Covance ( Princeton , NJ ) . PC-oligo was chemically synthesized by placing a photocleavable group in between every five base of ( dT ) 100 , using ( 3- ( 4 , 4′-Dimethoxytrityl ) -1- ( 2-nitrophenyl ) -propan-1-yl-[ ( 2-cyanoethyl ) - ( N , N-diisopropyl ) ]-phosphoramidite ) . PC-oligowas synthesized by and purchased from Gene Link ( Hawthorne , NY ) . Stock solutions of PC-oligo were made by dissolving lyophilized powder into 50% DMSO/ 50% TE pH 8 . 0 ( 10 mM Tris 8 . 0 , 1 mM EDTA ) to a final concentration of 5 mg/ml . The dT-oligo was synthesized by and purchased from Operon ( Huntsville , AL ) . Wild type mouse recombinant PrP was expressed in E . Coli and purified as previously described [6] . Benzonase nuclease ( 50-230-8706 ) was purchased from EMD Chemicals ( Gibbstown , NH ) . sPMCA reactions were performed as previously described [4] , [5] . Reaction tubes were subjected to PMCA for 24 h using a Misonix s3000 programmable sonicator equipped with a microplate horn ( Misonix , Farmingdale , NY ) containing 350 ml of water . The temperature was maintained inside the horn by a circulating water bath , pumping warmed water through aluminum coils surrounding the horn . Sample tubes were held in a plastic rack which prevents lid opening and hold tubes ∼3 mm from the surface of the horn . After 24 h , the tubes were removed from the sonicator , and 10 µl of the reaction mixture was added to a new tube containing fresh substrate . A portion of each sample was treated with 25 µg/ml proteinase K for 30 min at 37°C shaking at 750 rpm in an Eppendorf Thermomixer ( Fisher Scientific , Pittsburg , PA ) . An equal volume of 2x sodium dodecyl sulfate ( SDS ) loading buffer was added to each tube and then boiled at 95°C for 10 min . SDS-polyacrylamide gel electrophoresis ( PAGE ) was then performed using 1 . 5 mm 12% polyacrylamide gels ( 29∶1 acrylamide∶bisacrylamide ) The gel was transferred to a methanol-charged polyvinylidene difluoride membrane ( Millipore , Billerica , MA ) using a Transblot SD semidry transfer cell ( Bio-Rad , Hercules , CA ) . The transfer was set to achieve 3 mA/cm2 for 30 min . After transfer , the membrane was incubated in 20% ( v/v ) instant nonfat dry milk ( Nestle , Vevey , Switzerland ) dissolved in TBST ( 10 mM Tris pH 7 . 1 , 150 mM NaCl , 0 . 1% Tween 20 ) . The membrane was then incubated overnight at 4°C with mAb 6D11 in TBST ( final concentration 80 ng/ml ) . The membrane was washed 3 times with TBST for 10 min and then incubated for 1 h with horseradish peroxidase-labeled anti-mouse immunoglobulin G secondary antibody conjugate ( GE Healthcare ) diluted 1∶5 , 000 in TBST . Again the membrane was washed 4 times for 10 min in TBST . Blots were developed with West Pico ( Pierce , Rockford , IL ) chemiluminescence substrate . Images were captured using a Fuji LAS-3000 chemiluminescence documentation system ( Fujifilm , Tokyo , Japan ) . Relative molecular masses were determined by comparison to prestained standards from Fermentas ( Hanover , MD ) . Immunopurified PrPC was prepared as described previously [5] . To produce substrate mixtures , 37 . 5 µl of immunopurified PrPC was mixed with 10 µl of 10x reaction buffer ( 200 mM MOPS pH 7 . 5 , 10% Triton X-100 , 500 mM imidazole , 50 mM EDTA , 1 . 5 M NaCl ) , 7 . 5 µl of dT-oligo or PC-oligo ( 0 . 3 mg/ml for a total of 2 . 25 µg ) and 35 µl of molecular grade water ( Mediatech , Herndon , VA ) . On the first day , 10 µl of Sc237 brain homogenate was added to a 0 . 5 ml thin-walled PCR tube ( Axygen , Union City , CA ) and subjected to sPMCA as described above . The sonicator delivered bursts of 30 s every 30 min at output 6 . 5 . sPMCA was continued for 15 days , effectively diluting the original inocula 1015 fold . The negative control line was treated as stated above with the omission of nucleic acid from the substrate cocktail . dT-oligo and PC-oligo were placed in their own ProxiPlate 96-well plate ( PerkinElmer , Waltham , MA ) . The plates were placed into a UV Stratalinker 2400 ( Stratagene , La Jolla , CA ) equipped with bulbs that emit light at 315 nm . The 96-well plates were placed on top of tip boxes and racks so samples were ∼2 . 5 cm from the light source . Each light treated sample received 4 pulses of light at an energy level of 500 , 000 µJ . Non-light treated samples were also placed in their own ProxiPlate 96-well plate but were then placed inside a drawer to protect from light . Treated and non-treated oligos were either added to sPMCA reactions as described above , or directly analyzed as described below . For benzonase protection assays 6 µg of either dT-oligo or PC-oligo were incubated for 10 min at 37°C and shaking at 750 rpm with or without 300 µg recPrP . After incubation samples were sonicated for one 30 s pulse followed by another 10 min incubation at 37°C and 750 rpm . After second incubation 7 µl of benzonase was added ( 175 units ) and further Incubated for 1 h at 37°C and 750 rpm . The oligonucleotides were then extracted as described below . Concentrations of PrPSc containing either dT-oligo or PC-oligo were compared by Western blot and diluted to achieve equal concentrations . Of the normalized material , 500 µl of each sample was placed in its own ProxiPlate 96-well plate ( PerkinElmer , Waltham , MA ) . Samples were then treated as described above with one modification . In between each light pulse , samples were removed from the 96-well plate , placed in a 1 . 5 ml tube and subjected to sonication in a cold sonicator . Briefly , 350 ml of chilled water was placed in the microplate horn , and tubes were positioned in a rack as described above and subjected to a 20 s pulse at output 6 . 5 . Samples were returned to new wells in the 96-well plate and subjected to another round of light or dark treatment . To analyze the effect of light on the nucleic acids 150 µl of each sample was brought to 250 µl by adding 1x reaction buffer . An equal volume of Phenol∶Chloroform∶Isoamyl alcohol 25∶24∶1 ( Sigma , St . Louis , MO ) was added to each tube and vortexed . Samples were spun at 14 , 000× g for 2 min . The top aqueous phase was carefully removed and the process repeated twice more . After three extractions with the Phenol∶Chloroform∶Isoamyl alcohol , an equal volume of chloroform was added to each tube , vortexed and spun as above . The top aqueous phase was removed and to it 25 µl of 3 M sodium acetate pH 7 . 0 was added followed by 750 µl of 95% EtOH . Tubes were vortexed and incubated at −20°C for 15 min . After incubation tubes were spun at 14 , 000× g for 10 min to pellet the DNA . The pellet was washed with 80% EtOH , spun and dried . The pellet was resuspended in equal volumes of TE ( 10 mM Tris pH 7 . 0 , 1 mM EDTA ) and 2x TBE-Urea sample buffers ( Invitrogen , Carlsbad , CA ) . Samples were boiled at 95°C for 5 min and then loaded into pre-cast 15% TBE-Urea acrylamide gels ( Invitrogen , Carlsbad , CA ) . Gels were run in a X-Cell SureLock apparatus ( Invitrogen , Carlsbad , CA ) at 200 V for 50 min . Molecular weight ladder was made by mixing oligo dT 100mer , oligo dT 30mer and oligo dT 10mer ( 10 mg/ml ) in a ratio of 1∶3∶9 with 2x TBE-Urea sample buffer ( oligos purchased from Operon , Huntsville , AL ) . Gels were then stained for 1 h with Sybr Gold ( Invitrogen , Carlsbad , CA ) diluted 1∶10 , 000 in TBE ( 89 mM Tris-base , 89 mM Boric Acid , 2 mM EDTA pH 8 . 3 ) . After staining , gels were visualized on a transilluminator and photographed with a Canon Power Shot A650 IS ( Canon , Tokyo , Japan ) equipped with proper filters . A 10% hamster brain homogenate ( w/v ) was made in conversion buffer ( PBS with 150 mM NaCl , 1% Triton X-100 , 4 mM EDTA ) plus a Complete PI tablet with EDTA ( Poche , Indianapolis , IN ) using a polytron tissue grinder . The homogenate was then spun at 200× g for 30 s . The supernatant was then aliquotted into 0 . 5 ml thin-walled PCR tubes ( 90 µl reactions ) and frozen at −70°C until used . PrPSc molecules treated with or without light were serially diluted into conversion buffer ( PBS with 150 mM NaCl , 1% Triton X-100 , 5 mM EDTA ) to create 10 titers ranging from 10−1–10−10 . Each titer was used to start a reaction by adding 10 µl to 90 µl of brain homogenate prepared as described above . Samples were subjected to 3 rounds of sPMCA as described above . The sonicator for this experiment was set to deliver 30 s pulses at output 8 . 5 . Each sample was Western blotted as described above . For each of the lines injected , 20 µl of day 15 material was spiked into each of 5 reaction tubes containing 200 µl of fresh substrate cocktail and subjected to 1 cycle of PMCA as described above . After 24 h tubes were removed and combined . This material was then treated with or without light as described above . After treatments , 100 µl of each sample was diluted into 900 µl of sterile PBS with 5 mg/ml bovine serum albumin ( BSA ) to prepare the inoculum . The DNA alone inoculum was prepared by mixing 2 . 25 µg of each nucleic acid into 100 µl of 1x reaction buffer . This material was then diluted 1∶10 into 5 mg/ml BSA , as described above . Injections were performed in laminar-flow biosafety cabinets using disposable syringes , gloves and surfaces . Intracerebral inoculations were performed using 28-gauge hypodermic needles inserted into the parietal lobe . Each animal received 50 µl of inoculum . Hamsters were examined daily by veterinary staff blinded to experimental groups . Standard diagnostic criteria were used to identify animals showing signs of scrapie . Animals were euthanized by CO2 inhalation until death was imminent . Their brains were quickly removed using sterile dissection instruments on disposable surfaces . Brains were immersed in 10% buffered formalin for 1 week . Sagittal and parasagittal sections were made and brains placed into tissue-processing cassettes . The cassettes were immersed into 90% formic acid for 1 hr and then placed into 10% formalin . Tissues were next embedded in paraffin and cut into 4 µm sections . Sections were stained with hematoxylin and eosin . A single neuropathologist , who was blinded to the experimental groups examined all slides and scored them according to the degree of vacuolation . Scores were made according to previously published standards [5] . All animals were handled in strict accordance with good animal practice , as defined by the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The Dartmouth College Institutional Animal Care and Use Committee approved the animal work ( assurance number A3259-01 ) . Inoculations were performed under isoflurane anesthesia , and all efforts were made to minimize suffering .
We chemically synthesized a 100-base poly-dT molecule containing with a UV-photocleavable group ( PC-oligo ) every five bases ( Figure 1 ) . To confirm that PC-oligo can support PrPSc propagation , sPMCA reactions containing purified Syrian hamster PrPC supplemented with either PC-oligo or dT-oligo were seeded with Sc237 brain homogenate . The PrPC substrate used in these studies has been previously characterized extensively by a variety of analytical methods , and was found to contain stoichiometric quantities of co-purified endogenous lipids; but no other proteins , nucleic acids , or specific metal ions [5] . The results of our preliminary experiment show that both PC-oligo and dT-oligo support PrPSc propagation through three rounds of sPMCA reactions whereas control reactions without added polyanions failed to propagate ( Figure S1 in Supporting Information S1 ) . After proteinase K ( PK ) digestion , PrPSc molecules formed with either oligonucleotide showed a similar shift in electrophoretic mobility to 27–30 kD ( Figure S1 in Supporting Information S1 ) . To determine the optimal irradiation conditions for the photolysis of PC-oligo , pure solutions of each oligonucleotide in buffer were treated with varying amounts of UV light ( 315 nm ) and analyzed by acrylamide gel electrophoresis . Prior to light treatment , PC-oligo can be seen a 5-base ladder , representing the collection of full length and incomplete products formed during the sequential conjugation chemical synthesis protocol ( Figure S2 in Supporting Information S1 , lane 3 ) . Upon exposure to two Joules of light , PC-oligo was completely degraded to 5-base oligonucleotide units ( Figure S2 in Supporting Information S1 , compare lanes 3 and 7 ) while dT-oligo remained intact ( Figure S2 in Supporting Information S1 , compare lanes 1 and 2 ) . We next tested the effect of light-induced degradation on the ability of PC-oligo to facilitate prion propagation in sPMCA reactions . As anticipated , dT-oligo supported propagation of Sc237 prions in sPMCA reactions using immunopurified hamster PrPC substrate even when the control oligonucleotide was pretreated with UV light ( Figure 2 , left gels ) . PC-oligo pretreated with UV light did not support propagation ( Figure 2 , bottom right gel ) , whereas PC-oligo mock-treated in the dark supported Sc237 propagation in sPMCA for three rounds ( Figure 2 , top right gel ) . These results confirm our previously published results that an oligonucleotide must be at least 40 bases long to support propagation , and that our light treatment protocol completely degrades PC-oligo to a length smaller than that required for efficient propagation in sPMCA . We previously found that single stranded nucleic acids form a nuclease-resistant complex with both PrPC and PrPSc molecules during sPMCA reactions[1] . To study the penetration of UV light into nucleoprotein complexes , we combined a stoichiometric excess of full-length recombinant mouse PrP ( recPrP ) with our test oligonucleotides . Following PMCA sonication , both PC-oligo and dT-oligo became completely protected from nuclease digestion , whereas both oligonucleotides were nuclease-sensitive in the absence of PrP ( Figures 3A and 3B ) . UV irradiation induced the degradation of PC-oligo , but not dT-oligo , within these nuclease-resistant complexes containing excess PrP ( compare Figure 3A , lanes 7–8 and Figure 3B , lanes 7–8 ) . Thus , UV light is able to penetrate PrP nucleoprotein complexes , providing us with a unique opportunity to control the degradation of PC-oligo in situ within purified prions . Having established conditions for UV light penetration of PrP nucleoprotein complexes , we sought to determine the effect of light-induced degradation of PC-oligo in situ ( i . e . already incorporated into complexes PrPSc molecules ) on the conformational stability , sPMCA seeding activity , and biological infectivity of synthetic PrPSc molecules . PrPSc molecules containing either PC-oligo or control dT-oligo were generated by performing sPMCA for 15 rounds in order to dilute the initial prion seeds several orders of magnitude beyond the calculated endpoint [4] , [5] , and then either treated with UV light or mock-treated in the dark . A PK digestion assay revealed no significant differences in PrPSc protease-resistance between light-treated versus mock-treated samples ( Figure S3 in Supporting Information S1 ) . To measure sPMCA seeding activity , samples were serially titrated , and the resulting dilutions were used to seed three rounds of sPMCA using fresh hamster brain homogenate as substrate , and samples from the final round of sPMCA were analyzed for the presence of PrPSc by Western blot . The results of this quantitative end-point titration assay show that PrPSc molecules containing either dT-oligo or PC-oligo were equally susceptible to UV irradiation; both sets of samples displayed ∼1 log decrease in seed titer upon light treatment , presumably due to non-specific effects ( Figure 4 ) . A small amount of each sample was concurrently analyzed by acrylamide gel electrophoresis to confirm that UV treatment was successful in degrading the PC-oligo below detection limits ( Figure S4 in Supporting Information S1 ) . Taken together , the results indicate that light-induced degradation of incorporated PC-oligo into 5-mers had no specific effect on the ability of PrPSc to seed sPMCA reactions of normal brain homogenate . Light- and dark-treated PrPSc containing either dT-oligo or PC-oligo were injected intracerebrally into Golden Syrian hamsters for bioassay . In addition , negative control samples consisting of either the original Sc237 seed propagated for 15 rounds in PrPC substrate without nucleic acid ( designated PrPC ) or a cocktail containing the dT-oligo and PC-oligo ( designated dT/PC-oligo alone ) were inoculated in parallel . The scrapie incubation time for all 4 experimental groups was ∼150 days , with all of the inoculated animals developing similar clinical signs of ataxia , trembling , and circling ( Table 1 ) . Animals injected with the PrPC and dT/PC-oligo control samples did not develop signs of clinical scrapie during the time frame of the experiment ( Table 1 and figures 5A and 5B ) . Neuropathological analysis showed significant vacuolization throughout the brains of animals in all 4 groups , including the brainstem and hippocampus ( Figure 5A and 5B ) , and a blinded comparison of vacuolization scores from different brain regions showed no difference between the experimental groups ( Figure 6 ) . Biochemical analysis showed no differences in the protease sensitivity , glycoform ratio , or electrophoretic mobility of the PrPSc molecules produced in the brains of animals from all 4 experimental groups ( Figure S5 in Supporting Information S1 ) . There was also no apparent difference in the conformational stability of PrPSc molecules in the brains of animals injected with light-treated versus control inocula , as determined by a urea denaturation assay ( Figure S6 in Supporting Information S1 ) . Therefore , these results indicate that degradation of an incorporated polyanion does not significantly alter the biological infectivity or the strain-dependent properties of purified Sc237 prions .
The major finding of this work is that selective degradation of an incorporated photocleavable polyanion cofactor does not alter the catalytic activity , infectivity , or strain properties of in vitro generated prions . These results contrast with those of Jeong et al . , who showed that treatment of scrapie-infected brain homogenates with LiAlH4 caused degradation of endogenous RNA and increased prion incubation time from ∼100 to ∼340 days [20] . There are several possible reasons for the discrepancy between the two studies: ( 1 ) Although both LiAlH4 and UV light treatment are capable of reacting non-specifically with non-polyanionic molecules , under the conditions used for each study , it is possible that LiAlH4 treatment is more non-specifically damaging than UV light . For instance , it could reduce PrPSc molecules and/or directly damage protein structure . ( 2 ) The work of Jeong et al . used scrapie brain homogenate as the starting infectious material , whereas the experiments reported here used purified prions formed from native PrPC substrate in vitro using sPMCA . Such purified prions are composed only of PrP , stoichiometric amounts of an endogenous lipid molecule containing 20 carbon fatty acids , and a synthetic polyanion [5] . ( 3 ) Whereas LiAlH4 is capable of hydrolyzing RNA molecules completely , UV light degrades the PC-oligo used in our study to ( dT ) 5 oligonucleotides . Although such small nucleic acids do not support prion propagation in vitro , our results cannot formally exclude the possibility that short oligonucleotides , once incorporated , might be able to act as a reinforcing backbone for PrPSc . ( 4 ) Although we did not detect any intact PC-oligo after light treatment using a highly sensitive intercalating agent , it is possible that a small , undetectable amount of PC-oligo was protected from UV-irradiation by virtue of being complexed with PrPSc molecules . This possibility is difficult to address experimentally because it is technically difficult to recover and detect such small amounts of DNA that cannot be amplified . It is worth noting that UV light degrades PC-oligo quantitatively within nuclease-resistant complexes with recombinant PrP . Our results are compatible with the scenario in which prion infectivity might be exclusively encoded by PrPSc structure in the absence of non-proteinaceous cofactors , as proposed by the “protein only” hypothesis [23] , [24] . However , this hypothesis cannot be confirmed in our experimental system because both endogenous lipid and remnant ( dT ) 5 oligonucleotide molecules remain present in purified prions generated with PC-oligo following light treatment . Nonetheless , we are able to conclude that it is unnecessary for prions to contain polyanions >5 bases in length to maintain infectivity , whereas polyanions ≥40 bases are required for the process of prion formation in vitro . This conclusion places a significant geometric constraint on the mechanism by which polyanionic cofactors could be theoretically used to maintain PrPSc architecture . Stoichiometric quantities of endogenous lipids containing 20-carbon fatty acids co-purify with the PrPC substrate used in these studies [5] , and therefore these lipid molecules could also become incorporated into the infectious PrPSc product during PMCA reactions . Future studies will be required to test whether endogenous lipid molecules or other non-nucleic acid cofactors , such as those recently described in mouse brain homogenates [25] , [26] , might be required to maintain prion infectivity and strain phenotypes . | Prions are unorthodox infectious agents whose composition remains undetermined . Previous experiments have shown that long , negatively charged polymers such as nucleic acid and carbohydrate molecules promote the formation of purified prions in test tube chemical reactions . Various classes of negatively charged polymers have also been found to co-exist within prion complexes in the brains of infected animals . These observations suggest that negatively charged polymers might act as a structural support necessary for prion infectivity . We tested this possibility by chemically synthesizing a negatively charged polymer that can be degraded by exposure to ultraviolet light . Prions containing this light-sensitive polymer remained infectious after light exposure , indicating that negatively charged polymers are not necessary to maintain the structural shapes of infectious prions . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"infectious",
"diseases",
"veterinary",
"diseases",
"zoonotic",
"diseases",
"veterinary",
"science",
"prion",
"diseases"
]
| 2011 | In Situ Photodegradation of Incorporated Polyanion Does Not Alter Prion Infectivity |
The DNA damage response is a signaling pathway found throughout biology . In many bacteria the DNA damage checkpoint is enforced by inducing expression of a small , membrane bound inhibitor that delays cell division providing time to repair damaged chromosomes . How cells promote checkpoint recovery after sensing successful repair is unknown . By using a high-throughput , forward genetic screen , we identified two unrelated proteases , YlbL and CtpA , that promote DNA damage checkpoint recovery in Bacillus subtilis . Deletion of both proteases leads to accumulation of the checkpoint protein YneA . We show that DNA damage sensitivity and increased cell elongation in protease mutants depends on yneA . Further , expression of YneA in protease mutants was sufficient to inhibit cell proliferation . Finally , we show that both proteases interact with YneA and that one of the two proteases , CtpA , directly cleaves YneA in vitro . With these results , we report the mechanism for DNA damage checkpoint recovery in bacteria that use membrane bound cell division inhibitors .
The DNA damage response ( DDR , SOS response in bacteria ) is an important pathway for maintaining genome integrity in all domains of life . Misregulation of the DDR in humans can result in various disease conditions [1 , 2] , and in bacteria the SOS response has been found to be important for survival under many stressors [3–5] . The DNA damage response in all organisms results in three principle outcomes: a transcriptional response , which can vary depending on the type of DNA damage incurred , DNA repair , and activation of a DNA damage checkpoint [6–9] . In eukaryotes , the G1/S and G2/M checkpoints are established by checkpoint kinases , which transduce the signal of DNA damage through inactivation of the phosphatase Cdc25 [7] . Checkpoint kinase dependent inhibition of Cdc25 leads to accumulation of phosphorylated cyclin dependent kinases , which prevents cell cycle progression [7] . In bacteria , the SOS-dependent DNA damage checkpoint relies on expression of a cell division inhibitor , though the type of inhibitor varies between bacterial species . In Escherichia coli , the SOS-dependent DNA damage checkpoint is the best understood bacterial checkpoint [6] . Upon activation of the SOS response , the cytoplasmic cell division inhibitor SulA is expressed [10] . SulA accumulation leads to a block in septum formation by preventing the assembly of the cytokinetic ring by FtsZ , a homolog of eukaryotic tubulin [11 , 12] . SulA binds directly to FtsZ [13] and inhibits FtsZ polymerization [14 , 15] . Recovery from the SulA-induced checkpoint occurs through proteolysis of SulA . Lon is the primary protease responsible for clearing SulA [16–18] , although ClpYQ ( HslUV ) were found to contribute to SulA degradation in the absence of Lon [19–21] . Thus , the mechanisms of DNA damage checkpoint activation by the cytoplasmic protein SulA and subsequent recovery are well understood in E . coli . The SulA-dependent checkpoint , however , is restricted to E . coli and a subset of closely related bacteria . It is becoming increasingly clear that most other bacteria use a DNA damage checkpoint with an entirely different mechanism of enforcement and recovery . An evolutionarily broad group of bacterial organisms have been shown to use a notably different DNA damage checkpoint mechanism [22–25] . In these Gram-positive and Gram-negative organisms , a small protein with a transmembrane domain is expressed that inhibits cell division without targeting FtsZ . One example is in the Gram-negative bacterium Caulobacter crescentus , where the SidA and DidA proteins bind to the essential membrane bound divisome components , FtsW/N that contribute to peptidoglycan remodeling [22 , 26] . Another example is the Gram-positive bacterium Bacillus subtilis in which the SOS-dependent cell division inhibitor is YneA [23] . YneA contains an N-terminal transmembrane domain with the majority of the protein found in the extracellular space [27] . Upon SOS activation , LexA-dependent repression of yneA is relieved and yneA is expressed [23] . Increased expression of yneA results in cell elongation , though FtsZ ring formation still occurs [27] , suggesting YneA inhibits cell division through a mechanism distinct from that of SulA . Further investigation found that overexpressed YneA is released into the medium , and that full length YneA is likely the active form of the protein [27] . The mechanism ( s ) responsible for YneA inactivation is unknown . Therefore , although the use of a small , membrane bound cell division inhibitor is wide-spread among bacteria , in all cases studied the mechanism of checkpoint recovery remains unknown [22–26] . We report a set of forward genetic screens to three different classes of DNA damaging agents using transposon mutagenesis followed by deep sequencing ( Tn-seq ) . Our screen identified two proteases , YlbL and CtpA , that are important for growth in the presence of DNA damage . Mechanistic investigation demonstrates that YlbL and CtpA have overlapping functions , and in the absence of these two proteases , DNA damage-dependent cell elongation is increased and checkpoint recovery is slowed . A proteomic analysis identified accumulation of YneA in the double protease mutant . We also found that DNA damage sensitivity of protease mutants depends solely on yneA . Further , we show that both proteases interact with full length YneA in a bacterial two-hybrid assay , and that CtpA is able to digest YneA in a purified system . With these results , we present a model of DNA damage checkpoint recovery for bacteria that use the more wide-spread mechanism employing a small , membrane bound cell division inhibitor .
In order to better understand the DNA damage response in bacteria , we performed three forward genetic screens using B . subtilis . We generated a transposon insertion library consisting of more than 120 , 000 distinct insertions ( S1 Table ) . The coverage of each transposon mutant in the library was plotted against the genome coordinates , which showed that the distribution of insertions was approximately uniform across the chromosome in the population of mutants ( Fig 1A ) . Two small exceptions were detected where coverage decreased . Decreased coverage corresponds to regions where many essential genes are clustered ( Fig 1A , arrow heads ) . With the goal of identifying mutants important for the DNA damage response , we grew parallel cultures of either control or DNA damage treatment over three growth periods , modelling our experimental design after a previous report [28; Fig 1B] . Mitomycin C ( MMC ) , methyl methane sulfonate ( MMS ) , and phleomycin ( Phleo ) were chosen for screening because these agents represent three different classes of antibiotics that damage DNA directly . MMC causes inter- and intra-strand crosslinks and larger adducts [29 , 30] , MMS causes smaller adducts consisting of DNA methylation [31] , and Phleo results in single and double stranded breaks [32 , 33] . As a result , we reasoned that the combined data would provide a collection of genes that are generally important for the DNA damage response . After sequencing , we performed quality control analysis . First , given that sequencing data are count data , the distribution of the coverage should be log-normal [34] . Indeed , the distribution of each replicate for the initial library and starter culture samples is approximately log-normal ( S1A Fig ) . We also found that the distributions for the remaining time points of the pooled replicates followed an approximate log-normal distribution ( S1B Fig ) . The sequencing data and viable cell count data ( S1 Table ) were used to calculate the fitness of each insertion mutant in each condition [Fig 1C; 35 , 36] . The relative fitness of each insertion was calculated by taking the ratio of treatment to control ( Fig 1C ) , thereby isolating fitness effects of the treatments . The relative fitness of each gene was determined by averaging the relative fitness calculated for each insertion within a gene ( Fig 1C ) . To verify that a t-test would be appropriate for determining relative fitness deviating significantly from one , we plotted the distribution of insertion relative fitness . All the distributions were normal with a mean close to one ( S1C Fig ) . We determined the relative fitness for every gene with sufficient data ( see supplemental methods ) , and report the relative fitness values and the adjusted p-values [37] in S2 Table . Initial inspection found that several genes known to be involved in DNA repair ( recA , ruvAB , recN , and recOR [38] ) had decreased relative fitness in growth period one of all experiments ( S2 Table ) . A closer analysis of recN , addA , and polA , three genes that are found toward the top of the lists in all treatments , showed that relative fitness is less than one in most cases , though in the Phleo experiment , it appears that the cultures were adapting to the treatment by growth period three ( Fig 1D ) . For comparison , we also plotted the relative fitness of thrC , a gene involved in threonine biosynthesis , and found the relative fitness to be about one in all conditions examined ( Fig 1D ) . Importantly , insertion in uvrA , a component of the nucleotide excision repair machinery [38 , 39] which helps repair MMC adducts but not MMS or Phleo related damage [40 , 41] , decreased relative fitness in growth periods 2 and 3 with MMC , but did not significantly decrease relative fitness in MMS or Phleo ( Fig 1D and S2 Table ) . Taken together , these results validate the approach by demonstrating that we were able to identify genes known to be involved in DNA repair . We also wondered whether our results contained false positives . To test this , we decided to experimentally validate the genes with the forty lowest relative fitness values from growth period two in the MMC experiment . We found that eight of the forty gene deletions were not sensitive to MMC in a spot titer assay ( Table 1 ) . Several genes that were false positives are located in the genome near genes with validated phenotypes , suggesting that polar effects explain some of the false positives ( Table 1; see supplemental results for detailed analysis ) . To identify genes required generally as part of the DNA damage response , we examined the 200 genes from growth period two with the lowest relative fitness and an adjusted p-value less than 0 . 01 from all three experiments ( S3 Table ) . We found that 21 genes overlapped for all three experiments ( Fig 1E ) , some of which are known to be involved in DNA repair ( recN , addB , polA , radA ) , while several genes have no known function ( e . g . , ylbL and ctpA ) ( S3 Table ) . Among the genes important for growth in the presence of DNA damage , we focused on two putative proteases YlbL and CtpA . YlbL is predicted to have three domains: a transmembrane domain , a Lon protease domain , and a PDZ domain ( Fig 2A ) . CtpA is predicted to have four domains: a transmembrane domain , a S41 peptidase domain , a PDZ domain , and a C-terminal peptidoglycan ( PG ) binding domain ( Fig 2A ) . In all three Tn-seq experiments , the relative fitness of insertions in either ylbL or ctpA was significantly less than one in the second and third growth periods ( Fig 2B ) , suggesting that absence of either protease results in sensitivity to DNA damage . In contrast , a control gene amyE , which is involved in starch utilization , had a relative fitness of approximately one in all conditions examined ( Fig 2B ) . To verify the Tn-seq results , we constructed clean deletions of ylbL and ctpA and found both mutants to be sensitive to DNA damage in a spot titer assay ( Fig 2C ) . Each phenotype was also complemented by ectopic expression of each protease in its respective mutant background ( Fig 2C ) . To identify putative catalytic residues , we aligned the protease domain of YlbL to LonA and LonB from B . subtilis and Lon from E . coli . The sequence alignment revealed that YlbL contains a putative catalytic dyad consisting of a serine ( S234 ) and a lysine ( K279 ) ( S2A Fig ) . Similarly , we aligned CtpA to its homologs CtpB from B . subtilis and Prc from E . coli , which identified a putative catalytic triad consisting of a serine ( S297 ) , a lysine ( K322 ) , and a glutamine ( Q326 ) ( S2B Fig ) . To test whether these putative catalytic residues were required for function , we attempted to complement the DNA damage sensitivity phenotype via ectopic expression of serine and lysine mutants . Both the serine and lysine mutants of YlbL and CtpA failed to complement the deletion phenotypes ( Fig 2C ) . The variants and the wild-type proteases were ectopically expressed to the same level in vivo ( Fig 2D ) , suggesting that the lack of complementation is not due to instability caused by the amino acid changes . With these results , we conclude that protease activity is required for YlbL and CtpA to function in response to DNA damage . The similarity in phenotypes led us to hypothesize that YlbL and CtpA have overlapping functions . To test this , we performed a cross-complementation experiment using spot titer assays for MMC sensitivity . Over-expression of YlbL , but not YlbL-S234A , complemented a ctpA deletion ( Fig 3A & 3B ) . Similarly , over-expression of CtpA , but not CtpA-S297A , complemented a ylbL deletion ( Fig 3A & 3B ) . In addition , deletion of both proteases rendered B . subtilis hypersensitive to MMC , even more so than loss of uvrA , which codes for the protein responsible for recognizing MMC adducts as part of nucleotide excision repair [Fig 3C; 38 , 39] . To further test the hypothesis that YlbL and CtpA have overlapping functions , we over-expressed each of the proteases separately in the double protease mutant background and observed a complete rescue of MMC sensitivity upon expression of the wild type ( WT ) , but not the serine variants ( Fig 3D & 3E ) . The experiments performed thus far cannot distinguish between sensitivity to MMC resulting from cell death , growth inhibition or both . To determine whether sensitivity arises from cell death , we performed a survival assay using an acute treatment of MMC . We detected a slight decrease in percent survival as MMC concentration increased in the ΔylbL and the double mutant strain ( S4A Fig ) . We compared the decrease in percent survival in single and double protease mutants to a ΔuvrA strain , which has been shown previously to be acutely sensitive to MMC [40] . The strain lacking uvrA was very sensitive to an acute treatment of MMC ( S4A Fig ) , whereas , the double protease deletion strain was significantly less sensitive to acute exposure compared with ΔuvrA ( compare Figs 3C & S4A ) . Taken together , we conclude that MMC sensitivity of the protease mutants observed in spot titer assays is primarily caused by growth inhibition . We hypothesized that sensitivity to DNA damage resulting from growth inhibition could also be explained by inhibiting cell proliferation , or by inhibiting cell division rather than cell growth . To distinguish between these two possibilities , we measured cell length , because inhibition of proliferation should be observed as an increase in cell length , consistent with a failure in checkpoint recovery . Thus , we designed a MMC recovery assay , reasoning that following treatment with MMC , cells lacking YlbL , CtpA , or both , would remain elongated showing slower checkpoint recovery relative to the WT strain . We grew cultures either in a vehicle control or in the presence of MMC . After a two-hour treatment , the MMC containing media was removed and cells were washed . Cells were then transferred to fresh media without MMC and allowed to continue growth to assay for checkpoint recovery . Although cells appeared to be elongated in the ΔylbL and double mutant strains , there was heterogeneity in the population ( Fig 4A ) . As a result , we measured the cell length of at least 900 cells for each genotype and each condition and plotted the cell length distributions as histograms ( Fig 4B ) . There was no difference in the vehicle control cell length distributions ( Fig 4B ) . The MMC treatment of all strains resulted in a rightward shift in the distribution for all strains ( Fig 4B , compare upper panels ) . When comparing the protease deletions to the WT , the difference in distribution could be visualized by considering the percentage of cells greater than 6 . 75 μm in length , which is about three cell lengths of 2 . 25 μm each . We found that deletion of ylbL resulted in an increase in the percentage of cells longer than 6 . 75 μm in MMC treated cultures and after both two hours and four hours of recovery ( Fig 4C ) . Deletion of ctpA , however , resulted in a very slight , though significant ( p-value = 0 . 0142 for one-tailed Z-test ) , increased percentage of cells longer than 6 . 75 μm after 4 hours of recovery ( Fig 4C ) . The double mutant resulted in a percentage of cells slightly greater than ΔylbL alone after both two hours ( p-value = 0 . 0001 for one-tailed Z-test ) and four hours ( p-value = 0 . 0088 for one-tailed Z-test ) of recovery ( Fig 4C ) . Taken together , we conclude that YlbL is the primary protease under these conditions , with CtpA also contributing . We also conclude that cells lacking YlbL or both YlbL and CtpA take longer to divide following exposure to MMC , which is consistent with DNA damage sensitivity resulting from inhibition of cell proliferation . Further , the observation of inhibition of cell proliferation suggests that YlbL and CtpA proteases could be important for DNA damage checkpoint recovery ( see below ) . A potential model to regulate YlbL and CtpA in response to DNA damage is to increase protein levels following exposure to DNA damage . Increased protease levels in response to DNA damage could promote the DNA damage checkpoint recovery when needed . To test this model , we monitored YlbL and CtpA protein levels via Western blotting over the course of the MMC recovery assay . YlbL and CtpA protein levels did not change relative to the loading control DnaN throughout the course of the experiment ( S4B & S4C Fig ) . As a positive control , we performed the same experiment and monitored RecA protein levels and found that , indeed , RecA protein levels increased ( S4B & S4C Fig ) , as expected because recA is induced as part of the SOS response [42 , 43] . We conclude that YlbL and CtpA protein levels are not regulated by DNA damage . The data presented thus far led us to hypothesize that in the absence of YlbL and CtpA , a protein accumulates which results in inhibition of cell division ( Fig 5A ) . To identify the accumulating protein , we performed an analysis of the entire proteome of WT and double protease mutant cell extracts . We chose to analyze the proteomes of cells after two hours of recovery , because the cell length distributions differed most between WT and the double protease mutant ( Fig 4B ) . We found that the normalized spectral count data had similar distributions for both WT and the double mutant , which were approximately log normal ( S5A Fig ) . We verified that the distribution of the test statistic ( the difference in double mutant average and WT average ) was normally distributed ( S5B Fig ) , thus allowing a t-test to be used . We also performed a principle component analysis and found that WT replicates and double mutant replicates each clustered together ( S5C Fig ) . In total , 2329 proteins were detected ( S4 Table ) , and 183 proteins were found to be differentially represented ( p-value < 0 . 05 ) in the double mutant relative to WT ( S5 Table ) . Of the proteins differentially represented in the double mutant , 104 had a fold change greater than one ( Fig 5B , red points ) . There are three major mechanisms that have been reported in B . subtilis to inhibit cell division: 1 ) Noc dependent nucleoid occlusion [44] , 2 ) FtsL depletion [45 , 46] , and 3 ) expression of YneA [23] . One possibility was that Noc protein levels were higher in the double mutant , but we observed no difference in Noc levels ( S5D Fig ) . Another possibility was that FtsL or the protease RasP , which degrades FtsL , was affected in the protease mutant background [46] . We found no difference in relative protein abundance of FtsL or RasP ( S5D Fig ) , ruling out the FtsL/RasP pathway . Among the top 10 proteins that were more abundant in the double mutant was YneA , the SOS-dependent cell division inhibitor ( S5 Table ) . We asked if the enrichment of YneA was simply because it is SOS induced . We analyzed the relative abundance of several other proteins that are known to be SOS induced , including RecA , UvrA , UvrB , DinB , and YneB [42] , which is in an operon with YneA [23] . We found that none of these other proteins were enriched in the double mutant ( S5E Fig ) . These results suggest that YneA accumulation is not a result of increased SOS activation , and regulation of YneA accumulation is likely to be post translational , because the protein levels of another member of the operon , YneB were unchanged . Taken together , our proteomics data suggest that YlbL and CtpA promote DNA damage checkpoint recovery through regulating YneA protein abundance . We directly tested for YneA accumulation in protease mutants throughout the MMC recovery assay using Western blotting . YneA accumulated in all protease deletion strains after 2 hours and 4 hours of recovery , though YneA accumulation in ΔctpA was slight ( Fig 5C ) . In the double mutant , YneA accumulated in the MMC treatment condition in addition to both recovery time points ( Fig 5C ) . In the double mutant we observed multiple YneA species , which we hypothesize to be the result of unnaturally high YneA protein levels resulting in non-specific cleavage by other proteases . With these results , we suggest that YneA is a substrate of YlbL and CtpA , both of which degrade YneA allowing for checkpoint recovery . Although accumulation of YneA fit our data well , we considered that the other proteins enriched greater than five-fold in the double mutant may have contributed to the DNA damage sensitivity phenotype . To test this , we constructed deletions of each gene in WT and the double mutant and tested for MMC sensitivity . We found that no single deletion of each of the 10 genes resulted in sensitivity to MMC ( S6A Fig ) . In the double mutant , only deletion of yneA was able to rescue the sensitivity to MMC ( S6A Fig ) . We verified that deletion of yneA could rescue MMC sensitivity in all protease mutant backgrounds ( Fig 6A ) . We examined cell length in the DNA damage recovery assay . As expected , deletion of yneA resulted in less severe cell elongation relative to WT ( compare WT in Fig 4B and ΔyneA∷loxP in Fig 6B ) . In addition , deletion of ylbL , ctpA , or both no longer changed the cell length distribution in the absence of yneA at the two-hour recovery time point ( Figs 6B , 6C and S6B ) . In the MMC treatment , we did observe a slight increase ( p-value = 0 . 0004 for one-tailed Z-test ) in the percentage of cells greater than 6 . 75 μm in the double protease deletion strain compared to WT ( Fig 6C ) . Given that MMC sensitivity and most cell elongation in protease mutants depends on yneA , we hypothesized that expression of YneA alone would be sufficient to inhibit growth to a greater extent in the protease mutants . Indeed , strains lacking YlbL , CtpA or both were more sensitive to over-expression of yneA from an IPTG inducible promoter than WT ( Fig 6D ) . Further , we show that YneA accumulated in the protease mutant strains following yneA ectopic expression ( Fig 6E ) . We conclude that YneA accumulation results in severe growth inhibition in cells lacking YlbL and CtpA . To test the hypothesis that YneA is a direct substrate of the proteases we purified YneA ( a . a . 28–103 ) , CtpA ( a . a 38–466 ) , and YlbL ( a . a 36–341 ) lacking their N-terminal transmembrane domains to allow for isolation . We were unable to detect protease activity from YlbL using YneA , lysozyme , or casein as substrates ( S7 Fig; see discussion ) . When purified CtpA was incubated with YneA , we observed digestion of YneA over time , but no digestion was observed using CtpA-S297A ( Fig 7A ) . To test if CtpA activity against YneA was specific we completed the same reaction using lysozyme as a substrate and detected no activity ( Fig 7B ) . We conclude that YneA is a direct and specific substrate of CtpA . Although we could not detect YlbL protease activity in vitro we asked if YlbL could interact with YneA . To test this , we used a bacterial two-hybrid assay [47 , 48] . We used this assay because it is effective at detecting interactions between membrane proteins [22 , 26 , 49] . We tested YlbL-S234A and CtpA-S297A to prevent digestion of YneA , and assayed for interaction with full length YneA or YneA without its transmembrane domain ( YneAΔN ) as a control . We found that YlbL and CtpA both interacted with full length YneA ( Fig 7C ) , but no interaction was detected with YneAΔN ( Fig 7C ) , likely due to YneA failing to localize to the membrane . Given that YlbL did not have activity in vitro and we detected an interaction with YneA in the bacterial two-hybrid assay , we suggest that YneA is a direct substrate of YlbL and that YlbL requires full length YneA for interaction .
How do YlbL and CtpA recognize YneA as a substrate ? An intriguing facet of this checkpoint recovery mechanism is the use of unrelated proteases . YlbL and CtpA both have transmembrane domains and PDZ domains , but the peptidase domains are very different . CtpA has a S41 peptidase domain and is homologous to Tail-specific protease or Tsp ( also Prc ) , which recognizes the C-terminus of its substrate through its PDZ domain [50 , 51] . We suggest that CtpA also recognizes YneA through the PDZ domain and that this mechanism explains how CtpA recognizes its other cognate substrates . In fact , the study by Mo and Burkholder identified a residue at the C-terminus of YneA ( D97 ) which when mutated to alanine stabilizes YneA [27] . It is tempting to speculate that D97 in YneA is important for CtpA to recognize YneA . The mechanism by which YlbL recognizes YneA is less clear . YlbL has a unique domain organization not found in other studied proteases . YlbL does not have the AAA+ ATPase domain common in other Lon proteases , which is logical given that YlbL likely resides extracellularly in order to degrade YneA . Instead of an ATPase domain , YlbL has a PDZ domain , which could act as a substrate recognition domain or as an inhibitory domain similar to the PDZ domain of DegS [52–54] . The bacterial two hybrid assay suggests that YlbL does recognize YneA directly ( Fig 7C ) , though we cannot rule out the possibility that there is an adaptor protein that recognizes YneA when these proteins are tethered to the membrane . We also did not identify a potential adaptor in the Tn-seq data further suggesting a direct interaction does occur between YlbL and YneA . A final possibility is that YlbL recognizes YneA through the transmembrane domain , which then activates the Lon peptidase domain to degrade or cleave YneA . This would also explain the reason we were unable to detect activity using YlbL lacking its transmembrane domain . In any case , further experiments are necessary to elucidate the mechanism by which YlbL recognizes YneA . All organisms control cellular processes through regulated signaling . To regulate a cellular process a signaling pathway must have mechanisms of activation and inactivation . Many bacteria use a small membrane protein as an SOS-induced DNA damage checkpoint protein [22–25] . The mechanism of checkpoint recovery , however , for organisms using membrane protein checkpoints has remained unclear . Our comprehensive study identified a dual protease mechanism of DNA damage checkpoint recovery ( Fig 7D ) . Proteases YlbL and CtpA are constitutively present in the plasma membrane of cells even in the absence of DNA damage . After encountering DNA damage , YneA expression is induced . We hypothesize that YlbL and CtpA activities become saturated by increased YneA expression , which results in a delay of cell division . Following DNA repair , expression of YneA decreases and YlbL and CtpA clear any remaining YneA allowing cell division to resume . DNA damage checkpoints are of fundamental importance to biology , and we have discovered the pathway responsible for checkpoint inactivation and cell cycle re-entry in B . subtilis . These findings represent an important advance in identifying how checkpoint recovery occurs in bacteria . The membrane bound cell division inhibitors identified to date [22–25] , are not homologs of YneA , and in fact the only unifying feature is that they are small membrane bound proteins [22–25] . This poses a great challenge because the components of checkpoint enforcement and recovery need to be experimentally identified . Our study serves as a model to identify the checkpoint recovery proteases through forward genetics , which in turn could be used to identify the enforcement protein through proteomics . The critical feature of our approach was the use of several growth periods in Tn-seq , which allowed us to identify both proteases . Thus , we propose a strategy using the combined approaches of forward genetics and targeted proteomics to identify the DNA damage checkpoint pathways in genetically tractable bacterial pathogens . Recovery from a DNA damage checkpoint is a critical process for all organisms . One theme found throughout biology is the use of multiple proteins with overlapping functions . In eukaryotes , the phosphorylation events that establish the checkpoint are removed by multiple phosphatases [55 , 56] . In E . coli , there are two cytoplasmic proteases , Lon and ClpYQ , that have been found to degrade the cell division inhibitor SulA [16 , 18–20] . Our study further extends the use of multiple factors in regulating checkpoint recovery to B . subtilis , by describing a mechanism using two proteases . In eukaryotes , multiple proteins with overlapping functions often exist due to spatial or temporal restrictions , which appears to at least partially explain the use of multiple factors in checkpoint recovery [55 , 56] . In E . coli , ClpYQ was found to be important at higher temperatures in the absence of Lon [19] , again suggesting that each protease functions under specific conditions . In the case of YlbL and CtpA , however , there appears to be a shared responsibility in rich media . Deletion of each protease results in DNA damage sensitivity and the double mutant has a more severe sensitivity . In contrast , during growth in minimal media , YlbL appears to be the primary protease , as the cell elongation phenotype is more pronounced in cells lacking ylbL . Still it is unclear how or when each protease functions . Why isn’t one protease sufficient to degrade YneA ? Do the proteases occupy distinct loci in the cell , requiring that each protease degrades a specific YneA pool ? Another possibility is that protease levels are constrained by another evolutionary pressure , such as substrates unique to each protease . Thus , the cell cannot maintain the individual proteases at levels required to titrate YneA as part of the DDR , because the levels of another substrate would be too low . Another explanation is that using multiple factors is an evolutionary strategy that increases the fitness of an organism . It is clear that checkpoint recovery is crucial , because the fitness of cells lacking ylbL or ctpA is significantly decreased in the presence of DNA damage ( S2 Table ) . Although the dual protease mechanism described here resolves an important step in the DDR , our data also reveal the complexity of the system . After we exposed cells to MMC the cells elongated . We noticed however , that not all elongation depended on yneA ( see Fig 6B ) , suggesting another mechanism for cell cycle control . In B . subtilis , there have been reports of yneA-independent control of cell division following replication stress [45 , 57 , 58] . The essential cell division component FtsL has been reported to be unstable and depletion leads to inhibition of cell division [58] . Further , ftsL transcript levels were reported to decrease following replication stress independent of the SOS response [45] , thus linking depletion of the unstable FtsL protein to cell division control following replication stress . A study using a replication block consisting of the Tet-repressor bound to a Tet-operator array , observed cell division inhibition independent of yneA , noc , and FtsL [57] . Interestingly , recent studies of Caulobacter crescentus uncovered two cell division inhibitors that are expressed in response to DNA damage , with one inhibitor SOS-dependent and the other SOS-independent [22 , 26] . In B . megaterium , a recent study found that the transcript of yneA is unstable following exposure to DNA damage [59] , suggesting yet another layer of regulation . No factor was identified to regulate yneA transcripts in the previous study , though it is possible that one of the genes of unknown function identified in our screens could regulate yneA mRNA . Together , these studies highlight the complexity of regulating the DNA damage checkpoint in bacteria .
Bacterial strains , plasmids , and oligonucleotides used in this study are listed in S6 Table and the construction of strains and plasmids is detailed in the supplemental methods . All Bacillus subtilis strains are isogenic derivatives of PY79 [60] . Bacillus subtilis strains were grown in LB ( 10 g/L NaCl , 10 g/L tryptone , and 5 g/L yeast extract ) or S750 minimal media with 2% glucose ( 1x S750 salts ( diluted from 10x S750 salts: 104 . 7g/L MOPS , 13 . 2 g/L , ammonium sulfate , 6 . 8 g/L monobasic potassium phosphate , pH 7 . 0 adjusted with potassium hydroxide ) , 1x metals ( diluted from 100x metals: 0 . 2 M MgCl2 , 70 mM CaCl2 , 5 mM MnCl2 , 0 . 1 mM ZnCl2 , 100 μg/mL thiamine-HCl , 2 mM HCl , 0 . 5 mM FeCl3 ) , 0 . 1% potassium glutamate , 2% glucose , 40 μg/mL phenylalanine , 40 μg/mL tryptophan ) at 30°C with shaking ( 200 rpm ) . Mitomycin C ( MMC ) , methyl methane sulfonate ( MMS ) , and phleomycin were used at the concentrations indicated in the figures . The following antibiotics were used for selection in B . subtilis as indicated in the method details: spectinomycin ( 100 μg/mL ) , chloramphenicol ( 5 μg/mL ) , and erythromycin ( 0 . 5 μg/mL ) . Selection of Escherichia coli ( MC1061 or TOP10 cells for cloning or BL21 for protein expression ) transformants was performed using the following antibiotics: spectinomycin ( 100 μg/mL ) or kanamycin ( 50 μg/mL ) . A transposon insertion library was constructed similar to [61] with modifications described in the supplemental methods . Tn-seq experiments were designed with multiple growth periods similar to a prior description [28] , with a detailed description in the supplemental methods . Sequencing library construction and data analysis were performed as described previously [35 , 61] with modifications described in the supplemental methods . Sequencing data were deposited into the GEO database with accession number GSE109366 . B . subtilis strains were struck out on LB agar and incubated at 30°C overnight . The next day , a single colony was used to inoculate a 2 mL LB culture in a 14 mL round bottom culture tube , which was incubated at 37°C on a rolling rack until OD600 was 0 . 5–1 . Cultures were normalized to OD600 = 0 . 5 and serial diluted . The serial dilutions were spotted ( 4 μL ) on the agar media indicated in the figures and the plates were incubated at 30°C overnight ( 16–20 hours ) . All spot titer assays were performed at least twice . Survival assays using an acute treatment of mitomycin C were performed as previously described [62] . Cultures were grown to an OD600 of about 1 , and triplicate samples of 0 . 6 mL of an OD600 = 1 equivalent were taken and cells were pelleted via centrifugation: 10 , 000 g for 5 minutes at room temperature ( all subsequent centrifugation steps were identical ) . Cells were washed with 0 . 6 mL 0 . 85% NaCl ( saline ) and pelleted via centrifugation . Cell pellets were re-suspended in 0 . 6 mL saline , and 100 μL aliquots were distributed for each MMC concentration . MMC was added to each tube to yield the final concentration stated in the figure in a total volume of 200 μL , and cells were incubated at 37°C for 30 minutes . Cells were pelleted via centrifugation to remove MMC , re-suspended in saline , and a serial dilution yielding a scorable number of cells ( about 30–300 ) was plated on LB agar to determine the surviving fraction of cells . Each experiment was performed three times in triplicate for each strain . Purified proteins ( see supplemental methods for purification protocols ) were submitted to Covance for antibody production using rabbits . Two rabbits were used in the 77 day protocol , and the serum with the least background was used for Western blots . For YlbL , CtpA , RecA , and DnaN Western blots , a cell pellet equivalent of 1 mL OD600 = 1 was re-suspended in 100 μL 1x SMM buffer ( 0 . 5 M sucrose , 0 . 02 M maleic acid , 0 . 02 M MgCl2 , adjusted to pH 6 . 5 ) containing 1 mg/mL lysozyme and 2x Roche protease inhibitors at room temperature for 1 or 2 hours . Samples were then lysed by addition of 6x SDS loading dye ( 0 . 35 M Tris , pH 6 . 8 , 30% glycerol , 10% SDS , 0 . 6 M DTT , and 0 . 012% bromophenol blue ) to 1x . Samples ( 12 μL ) were separated via 10% SDS-PAGE , and transferred to nitrocellulose using a Trans-Blot Turbo ( BioRad ) according to the manufacturer’s directions . Membranes were blocked in 5% milk in TBST ( 25 mM Tris , pH 7 . 5 , 150 mM NaCl , and 0 . 1% Tween 20 ) at room temperature for 1 hour or at 4°C overnight . Blocking buffer was removed , and primary antibodies were added in 2% milk in TBST ( αYlbL , 1:5000 or 1:8000; αCtpA , 1:5000; αRecA , 1:4000; αDnaN , 1:4000 ) . Primary antibody incubation was performed at room temperature for 1 hour or overnight at 4°C . Primary antibodies were removed and membranes were washed three times with TBST for 5 minutes at room temperature . Secondary antibodies ( Licor; 1:15000 ) were added in 2% milk in TBST and incubated at room temperature for 1 hour . Membranes were washed three times as above and imaged using the Li-COR Odyssey imaging system . All Western blot experiments were performed at least twice with independent samples . Molecular weight markers were used in the YlbL and CtpA blots . YlbL migrates to a location between the 30 and 40 kDa markers , consistent with its predicted molecular weight of 37 . 6 kDa . CtpA migrated to a location between the 40 and 80 kDa markers consistent with its molecular weight of 51 . 1 kDa . For YneA Western blots , cell pellets , 10 mL OD600 = 1 for MMC recovery assay and 25 mL OD600 = 1 for over-expression , were re-suspended in 400 or 500 μL , respectively , of sonication buffer ( 50 mM Tris , pH 8 . 0 , 10 mM EDTA , 20% glycerol , 2x Roche protease inhibitors , and 5 mM PMSF ) , and lysed via sonication . SDS loading dye was added to 2x and samples were incubated at 100°C for 7 minutes . Samples ( 10 μL ) were separated using 16 . 5% Tris-Tricine-SDS-PAGE ( BioRad ) and transferred to a nitrocellulose membrane using a Trans-blot Turbo ( BioRad ) according to the manufacturer’s directions . All subsequent steps were performed as above with a 1:3000 primary antibody dilution . An LB agar plate grown at 30°C overnight was washed with pre-warmed S750 minimal media and used to inoculate a culture of S750 minimal media at an OD600 = 0 . 1 . The cultures were incubated at 30°C until an OD600 of about 0 . 2 ( 2–2 . 5 hours ) . MMC was added to 100 ng/mL and cultures were incubated at 30°C for 2 hours . Cells were pelleted via centrifugation ( 4 , 696 g for 7 minutes ) and the media was removed . Cell pellets were washed in an equal volume of 1x PBS , pH 7 . 4 , and pelleted again via centrifugation as above . Cell pellets were re-suspended in an equal volume of pre-warmed S750 minimal media and incubated at 30°C for four hours . Samples for microscopy and Western blot analysis were taken after the two hour MMC treatment and at two and four hours following recovery , as indicated in the figures . The vehicle control samples were treated for 2 hours with an equivalent volume of the vehicle in which MMC was suspended ( 25% ( v/v ) DMSO ) . A 500 μL sample from the MMC recovery assay above was taken and FM4-64 was added to 2 μg/mL and incubated at room temperature for 5 minutes . Samples were then transferred to 1% agarose pads made of 1x Spizizen’s salts . Images were captured using an Olympus BX61 microscope . Cells were scored for cell length using the measuring tool in ImageJ software . For each image scored , all cells that were in focus were measured . The number of cells scored for each strain/condition is stated in the figures ( n = cells measured ) . The histograms were generated using ggplot2 in R . All scoring was done using unadjusted images . Representative images shown in the figures were modified in ImageJ by subtracting the background ( rolling ball radius method ) and adjusting the brightness and contrast . Any adjustments made were applied to the entire image . Samples ( 5 mL OD600 = 1 ) were harvested from cultures grown as described in the MMC recovery assay section at 2 hours recovery via centrifugation: 4 , 696 g at room temperature for 10 minutes . Samples were washed twice with 500 μL 1x PBS , pH 7 . 4 and pelleted via centrifugation: 10 , 000 g at room temperature for 5 minutes . Samples were frozen in liquid nitrogen and stored at -80°C . Samples were submitted for mass-spectrometry analysis to MS Bioworks . Further sample processing and data analysis was performed by MS Bioworks as described in the supplemental methods . The raw data files are available upon request and the processed data tables are provided as supplemental tables ( S4 and S5 Tables ) . YneA digestion reactions were prepared as a 20 μL reaction in 20 mM Tris pH 7 . 5 , 20 mM NaCl , and 20% glycerol containing 150 μM YneA , and 2 μM CtpA . Reactions were incubated at 30°C for the time indicated in the figure . Reactions were stopped by addition of 6x SDS-dye to 1x and incubating at 100°C for 5 minutes . Reaction products were separated via 16 . 5% Tris-Tricine SDS-PAGE . Proteins were detected by staining with coomassie blue . Lysozyme digestion assays were performed as for YneA using 2 mg/mL lysozyme and reactions were incubated at 30°C for 3 hours . Bacterial two hybrid assays were performed as previously described [47 , 48 , 63] . Briefly , T18 and T25 fusion plasmids were co-transformed into BTH101 cells and co-transformants were selected on LB agar + 100 μg/mL ampicillin + 50 μg/mL kanamycin at 37°C overnight . Cultures of LB + 100 μg/mL ampicillin + 50 μg/mL kanamycin were inoculated using several colonies and incubated at 37°C for 90 minutes . Cultures were diluted 250-fold and 4 μL were spotted on LB agar + 100 μg/mL ampicillin + 50 μg/mL kanamycin + 0 . 5 mM IPTG + 40 μg/mL X-gal and incubated at 30°C for 48 hours , then at room temperature for 24 hours . The brightness and contrast of the images were adjusted using Adobe Photoshop with changes applied to the entire image . All bacterial two-hybrid assays were performed at least twice . | Prokaryotes and eukaryotes coordinate cell division to genome integrity using DNA damage checkpoints . Many bacteria express a small , membrane binding protein to slow cell division when obstacles to DNA replication are encountered . Cell division inhibitors of this class have been identified in several bacterial species , yet the mechanism used to alleviate inhibition has remained unknown . Using forward genetics , we identified two unstudied genes , coding for the proteases YlbL and CtpA , that when deleted result in sensitivity to drugs that directly damage DNA . We show that sensitivity to DNA damage in protease mutants is a result of accumulation of the cell division inhibitor . Further , we show that YlbL and CtpA are responsible for degrading the cell division inhibitor allowing for cell division to resume . Importantly , these two proteases are not homologs , demonstrating a striking example of a bacterium using non-homologous enzymes to degrade a single substrate . Our investigation uncovers the previously unknown mechanism used to remove a cell division inhibitor , while also illuminating a potential strategy that bacteria can use to regulate signaling pathways . The use of multiple , unrelated proteins to perform a single function may represent a strategy employed throughout biological systems . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"blood",
"serum",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"cell",
"cycle",
"and",
"cell",
"division",
"pathogens",
"cell",
"processes",
"bacillus",
"enzymology",
"microbiology",
"gene",
"regulation",
"membrane",
"proteins",
"dna",
"damage",
"prokaryotic",
"models",
"experimental",
"organism",
"systems",
"dna",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"bacterial",
"pathogens",
"homologous",
"recombination",
"research",
"and",
"analysis",
"methods",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"cell",
"membranes",
"biochemistry",
"blood",
"cell",
"biology",
"nucleic",
"acids",
"anatomy",
"physiology",
"bacillus",
"subtilis",
"genetics",
"biology",
"and",
"life",
"sciences",
"proteases",
"dna",
"recombination",
"immune",
"serum",
"organisms"
]
| 2018 | Discovery of a dual protease mechanism that promotes DNA damage checkpoint recovery |
Social determinants can affect the transmission of leprosy and its progression to disease . Not much is known about the effectiveness of welfare and primary health care policies on the reduction of leprosy occurrence . The aim of this study is to evaluate the impact of the Brazilian cash transfer ( Bolsa Família Program-BFP ) and primary health care ( Family Health Program-FHP ) programs on new case detection rate of leprosy . We conducted the study with a mixed ecological design , a combination of an ecological multiple-group and time-trend design in the period 2004–2011 with the Brazilian municipalities as unit of analysis . The main independent variables were the BFP and FHP coverage at the municipal level and the outcome was new case detection rate of leprosy . Leprosy new cases , BFP and FHP coverage , population and other relevant socio-demographic covariates were obtained from national databases . We used fixed-effects negative binomial models for panel data adjusted for relevant socio-demographic covariates . A total of 1 , 358 municipalities were included in the analysis . In the studied period , while the municipal coverage of BFP and FHP increased , the new case detection rate of leprosy decreased . Leprosy new case detection rate was significantly reduced in municipalities with consolidated BFP coverage ( Risk Ratio 0 . 79; 95% CI = 0 . 74–0 . 83 ) and significantly increased in municipalities with FHP coverage in the medium ( 72–95% ) ( Risk Ratio 1 . 05; 95% CI = 1 . 02–1 . 09 ) and higher coverage tertiles ( >95% ) ( Risk Ratio 1 . 12; 95% CI = 1 . 08–1 . 17 ) . At the same time the Family Health Program had been effective in increasing the new case detection rate of leprosy in Brazil , the Bolsa Família Program was associated with a reduction of the new case detection rate of leprosy that we propose reflects a reduction in leprosy incidence .
According to WHO “leprosy is a chronic infectious disease caused by Mycobacterium leprae” . It can lead to physical disability , social stigma and suffering . Although significant improvements have been achieved in disease control , leprosy remains a public health problem in many countries with high incidence and transmission , mainly in tropical Africa , the Indian subcontinent , Pacific and Indian Ocean Islands and South America [1] , [2] . The new case detection rate ( NCDR ) of leprosy remains high in several parts of the world , including Brazil , although the known prevalence in the world has been reduced [3] . In area and population Brazil is the largest country in South America , and the fifth largest in the world . It has the highest leprosy occurrence in the American continent . The country contributed with 16% of new cases detected worldwide in 2011 [2] . Leprosy cases are concentrated in the poorest regions of the country , especially the North , Middle West and Northeast [3] , with the last region having the highest proportion of families receiving and benefiting from social programmes such as Bolsa Família Program ( BFP ) [4] , [5] . In 2012 the overall known prevalence of leprosy in Brazil was 1 . 5 per 10 , 000 ( equivalent to 29 , 311 individuals in treatment ) and the new case detection rate ( NCDR ) was 17 . 2 per 100 , 000 ( 33 , 303 new cases ) [6] . The known leprosy prevalence is calculated from the number of patients in treatment in a population reflecting the total patients in the moment of the analysis . It is related to the quality of treatment and the time that patients remain with active record in the health system . Paucibacilary patients remains in treatment for 6 months and multibacilary for 12 unless there are complications [7] . Hidden prevalence includes undiagnosed cases ( which are mainly responsible for transmission of the leprosy ) . The NCDR of leprosy which reflects the incidence is calculated from the number of new cases detected in a given population [7] , [8] . Because the average time in treatment in less than one year , the known prevalence should be lower than NCDR . Leprosy is a disease of poverty . Key risk factors reported to be associated with leprosy are crowding , low educational level , lack of hygiene , social inequality , food shortage and malnutrition [9] , [10] , [11] , [12] . It is not clear which influences the risk of infection and which influences the risk of evolution from infection to disease . Historically , the decline in leprosy is likely to have resulted from socioeconomic development: leprosy started to decline in Spain [13] and disappeared from Japan [14] and Norway [15] before implementation of the WHO multi-drug strategy . The disappearance in Hawaii , was attributed to economic development influencing family crowding , schooling , and nutritional status and others factors [16] , [17] . Chabot et al . ( 1995 ) [18] , argued that economic crisis had a negative impact on health care and on poverty related diseases in Africa , including leprosy . Furthermore , economic , political , demographic and social changes in Brazil during the last 40 years had a clear impact on social determinants of Brazilians' health [19] . During this period , there was an expansion of programs and activities in education , health , employment , housing , social security and social development [20] . This probably contributed to the reduction of infectious diseases but it is not clear how this affected leprosy in country . Conditional cash transfer programs are strategies that have increasingly garnered attention as a means to reduce poverty and inequalities in low and middle-income countries . These programs provide an income for poor families if they comply with specific conditions in education and health [21] . Cash transfers can significantly increase household consumption , reduce food insecurity , increase school enrollment and retention and improve health and nutritional outcomes under certain conditions [22] . Literature on cash transfer programs and their impact on leprosy is currently non-existent . However , recently evidence of this effect has been shown for HIV prevention programmes and other sexually transmitted diseases in underdeveloped countries [23] , [24] . Other studies discuss the positive effect of socio economic interventions , like cash transfer programs , in strengthening tuberculosis control by improving household's living conditions and therefore decreasing the exposure to biological risk factors ( such as malnutrition ) leading to better access and variety to food and health-seeking behavior thus reducing people's vulnerability to infection and disease [25] , [26] . The “Bolsa Família” Program ( BFP ) , introduced in Brazil in 2003 , was aimed at families in poverty and extreme poverty . It has three main objectives: to transfer income ( promoting an immediate relief of poverty ) , to improve access to education and health care and to offer complementary social programs ( enabling families to end their condition of vulnerability ) [27] . BFP is the largest cash transfer program in the world with 13 . 7 million families benefiting in 2012 . At the time the program aimed to transfer cash to those defined as “extremely poor families” with monthly per capita income $35 or less and “poor families” ( monthly per capita income between $35 and $70 and with children 17 years old or younger or pregnant or lactating women ) after enrollment in register of social programs ( CadÙnico , in Portuguese ) . Benefits range from $18 to $175 per month [28] . Enrolled families have to meet education and health conditions of BFP ( education and health conditionalities ) : up to date vaccination , nutritional surveillance of children under 7 years , attendance to ante natal care by pregnant women and post natal care after delivery [29] . It is well established that BFP reduces extreme poverty and contributed to mitigating the social and economic inequalities in Brazil [30] , [31] . The observed effect is explained by increased income , improves the food consumption and supplies related to health among the poor and extremely poor individuals [28] . The Family Health Program ( FHP ) , was introduced in 1994 , and contributed to the expansion of the Unified National Health System ( SUS ) . SUS principles include decentralization , universality and equity . According to the programme guide: “The FHP is a nationwide program , aimed at broadening access to public health services , especially in deprived areas , by offering free community based primary care” [32] . By 2013 , the program was implemented in 96% of Brazil's municipalities , covering 56 . 4% of the national population [33] . The FHP is widely decentralized and is managed , following national regulations , at the municipality level . It consists of multiprofessional teams with physicians , nurses , community health agents , oral health agents and dentists . Each FHP team is responsible for a well defined population , within an area , with systematic visits , to deliver health care , promotion and prevention . Actions include prenatal , neonatal and under-5 care , immunization and , more relevant for this analysis , prevention , and management of infectious diseases [32] . FHP contributes to leprosy control by supporting early detection and treatment of cases , contact tracing , control of disabilities and other preventive measures [34] . Increased access to primary care ( PHC ) achieved in Brazil mainly by FHP implementation has been shown to increase new case detection rate of leprosy [35] . There is clear evidence of the effectiveness of BFP and FHP in reducing malnutrition , childhood mortality , and other outcomes related to maternal and child health [36] , [37] , [38] , [39] , [40] . The objective of this study is to evaluate the impact of the Bolsa Família Program and Family Health Program on new case detection rate of leprosy in Brazil during the period 2004–2011 .
A study with a mixed ecological design , a combination of an ecological multiple-group and time-trend study design was carried out , with the municipality as unit of analysis , over the period from 2004 to 2011 . Of the 5 , 570 Brazilian municipalities 1 , 358 were selected because they belong to high risk clusters for leprosy detection previously described [3] , [41] . The annual new case detection rate of leprosy ( NCDR ) , was calculated as the number of reported new cases of leprosy ( defined by the code A30 in the International Classification of Diseases - 10th revision ) , per 100 , 000 people [33] . There are two possible indicators of BFP coverage from the number of families in the program: a ) Coverage of target population ( poor and extremely poor ) was obtained from Ministry of Social Development database . It is defined as “number of families included in the program by municipality divided by the number of eligible families ( according to BFP criteria ) in the same municipality” [42] and b ) Coverage of total population was defined as: “number of individuals enrolled in the BFP ( obtained by multiplying the number of beneficiary families by the average family size ) divided by the total population of the same municipality” [40] . The indicators obtained were combined and four categories were created according to the tertiles of the distribution of BFP coverage in the total population: low ( BFP coverage of the total population of the municipality from 0 . 0 to 27 . 75% ) , intermediate ( 27 . 76–48 . 10% ) , high ( > = 48 . 11% ) and consolidated ( BFP coverage of the total population of the municipality>48 . 11% in the presence of BFP coverage of the target population ≥100% for at least the last 4 years ) . The yearly coverage of the FHP was calculated as the number of individuals with records in any of the FHP facilities of the municipality in that year divided by the population of the municipality [43] . FHP coverage was categorized according to tertiles of the distribution ( 1st tertile: 0–72 . 02% , 2 st tertile: 72 . 03–95 . 06% and 3 st tertile: over 95 . 06% ) . A group of covariates was selected as potential leprosy determinants based on the literature [9]: percent of the population younger than 15 years , illiteracy rate , unemployment rate , urbanization rate , average number of residents per household , percentage of poor people in the city ( proportion of individuals with per capita household income equal to or less than U$ 35 , 00 monthly ) and Gini Index that is a measure of income distribution . Gini Index is defined as “measures the extent to which the distribution of income or consumption expenditure among individuals or households within an economy deviates from a perfectly equal distribution . A Gini index of 0 represents perfect equality , while an index of 100 implies perfect inequality” [44] . We dichotomized the covariates according to the median value of their distribution . The data used were collected from different information systems: Leprosy NCDR: the Notifiable Diseases Information System ( SINAN ) of the Ministry of Health [33] . BFP coverage: the Ministry of Social Development database [42] . FHP coverage: the Primary Care Information System ( SIAB ) [33] . Population and Socioeconomic variables: The Brazilian Institute of Geography and Statistics [45] . Some variables were extracted from the 2000 and 2010 national demographic census databases; in these cases , values for 2004–09 were estimated by linear interpolation and by linear extrapolation for 2011 . A descriptive analysis to describe trends in mean BFP and FHP coverage and in the variables . We measured the impact of BFP and FHP on the NCDR of leprosy using multivariable negative binomial regression models for panel data with fixed-effects specification , crude and adjusted for relevant covariates . As the outcome in this study was a rate ( the new case detection rate of leprosy ) negative binomial regression models were used as it is suitable for count data with overdispersion [46] . In these models , the rate is decomposed in a count using the logarithm of the population as an offset variable . Longitudinal panel data models as used here include a disturbance ( or error ) term and a term for unmeasured time-invariant characteristics of each unit of analysis , such as quality of the municipality management and other sociocultural or historical characteristics of the municipalities . Fixed-effect ( FE ) was used to control for the correlation between the time-invariant term with the coverage of the intervention under study , providing unbiased estimates of impact [47] . The Hausman specification test was used in order to confirm the appropriateness of the FE specification [48] . A total of 1 , 358 municipalities were selected to be included in the study . Seven municipalities without cases during the eight years of the study were not included in the model fitting because they had no cases and the fixed-effects model algorithms could not handle this [46] , [48] . The analyses were performed using Stata version 10 [49] . The Ethics Committee in Research of Institute of Collective Health - Federal University of Bahia ( protocol n ° 181 . 078 ) , approved this study .
The selected 1 , 358 municipalities originate over 50% of the new leprosy cases detected each year in Brazil and the annual NCDR of leprosy decreased from 74 . 8 to 45 . 6 per 100 , 000 people over the study period from 2004 to 2011 . This is a considerably higher reduction than in the total of the Brazilian municipalities ( Table 1 ) . Table 2 shows that in the selected municipalities , during the study period , there was a marked expansion of the average municipal BFP coverage both in all population ( from 24 . 6 to 44 . 7% ) and in the target population ( from 57 . 1 to 96 . 4% ) . There was also an increase in the mean municipality FHP coverage reaching 79 . 7% in 2011 . Marked improvements in the socioeconomic conditions was observed in the selected municipalities during the study period . The mean urbanization rate reached 61 . 3% in 2011 . There were reductions in percentage of poor people in the municipality ( from 43 . 8 to 29 . 8% ) , Gini Index ( from 0 . 56 to 0 . 53 ) , illiteracy rate ( from 23 . 1 to 19 . 6% ) , unemployment rate ( from 9 . 0 to 6 . 9% ) , average number of residents per household ( from 3 . 9 to 3 . 5 ) and mean percentage population aged less than 15 years ( from 34 . 7 to 28 . 3% ) . Table 3 shows the crude and adjusted association between new case detection rate of leprosy with BFP and FHP coverage levels . Increase in BFP coverage exhibited a significant dose–response reduction in new case detection rate of leprosy , and the effect is maintained after the controlling for demographic and socio-economic variables . When compared with municipalities with low coverage , municipalities with intermediate , high and consolidated BFP coverage have significant reductions in the new case detection rate of leprosy in crude and adjusted models . For instance , reduction in municipalities with BFP consolidated coverage was 27% over the period ( RR = 0 . 73; 95% CI = 0 . 69–0 . 77 ) on the crude model and 21% in the model adjusted for selected covariates ( RR = 0 . 79 95% CI = 0 . 74–0 . 83 ) . The analysis shows a significant increase in NCDR of leprosy as FHP coverage increases . In the adjusted model , compared with the low tertile of FHP coverage , in the medium tertile of FHP coverage ( 72 . 03–95 . 08% ) there was an increase of 5% over the period ( RR = 1 . 05 95% CI = 1 . 02–1 . 09 ) and for the higher tertile and increase of 12% over the period ( RR = 1 . 12 95% CI = 1 . 08–1 . 17 ) . All selected covariates except urbanization rate were significantly associated with the new case detection rate of leprosy .
This is the first evidence of the join impact of a conditional cash transfer and of a primary health care programmes on the incidence/detection of leprosy . BFP was associated with significant reduction in the NCDR of leprosy , and FHP was associated with significant increase in the NCDR of leprosy . Both effects were statistically significant and showed a dose-response effect . We postulate that the first effect - reduction in new case detection rate with the BFP - reflects a reduction in incidence of leprosy , consistent with the cash transfer component of BFP leading to improving living conditions . Poverty itself is a determinant of leprosy [9] , [10] , [11]; cash transfer reduces not only poverty but also specific aspects of poverty associated with leprosy , like inequality [9] , undernutrition and food shortage [9] , [10] , [11] . There is consistent evidence that conditional cash transfer programs increase food expenditure [50] , [51] , [52] , [53] . In Brazil , BFP increased access to food and improved food quality and diversity [53] , [54] . The second finding was an increase in new case detection rate of leprosy associated with the FHP coverage . We postulate that this reflects not a genuine increase in incidence , but an increased detection of cases that would otherwise remain undiagnosed - the hidden prevalence . FHP increases contact of individuals to health services and therefore is likely to facilitate self-reporting and diagnosis of leprosy cases in primary health care units . Other studies in Brazil showed increased coverage of primary health care contributing to an increase in new case detection rate of leprosy [35] , [55] , [56] . In Brazil leprosy has been a nationally notifiable disease for many decades . Brazil has a single surveillance information system . Each reported case is included in the database of the secretary of health of the municipalities and transmitted to the Ministry of Health . The NCDR depends of the capacity of health facilities identify the signs and symptoms of leprosy for diagnosis . Treatment was decentralized offering health care in a larger number of municipalities [3] , [35] . The National Leprosy Control Programme recommends treatment with multidrug therapy ( MDT ) according to World Health Organization recommendation and distributes it free of charge . The amount of MDT blister packs needed is estimated based on reported data , which guarantees an approximate relation between cases reported and cases treated [57] . Although better detection leads to a short-term increase in the NCDR , we fully expect that better detection will eventually lead to a long term reduction in incidence , as a result of lower number of infectious cases due to reduced hidden prevalence and earlier diagnosis and treatment of clinical cases , identification of contacts and better outcome of treatment [55] , [56] . Social interventions can have an impact on the leprosy transmission or clinical disease progression . The mean incubation period of leprosy is 2–5 years , but can be as long as 20 years [11] . Therefore would be necessary to analyze a longer period to infer whether the BFP and FHP had an impact on the transmission of leprosy . As our inference level is ecologic - we want to determine the effectiveness of social and health policy at an aggregate level – we do not commit ecological fallacy . The ecologic design also allows measurement of the effect of externalities of the BFP , which can represent an important part of its global effect [40]: the relief from poverty of a relevant proportion of the population in a small municipality can make the local economy grow , and families that are not recipients of the program are going to benefit from this spill-over effect . Furthermore , leprosy affects mainly the poor and extremely poor individuals and many of them are eligible for BFP and live in deprived areas where FHP is priority implemented . We used municipality as the unit of analysis because the National Social Assistance System ( SUAS ) and National Health System ( SUS ) are decentralized in Brazil and BFP and FPH were implemented at the municipality level [19] , [29] , [32] . Moreover , the application of a sophisticated statistical methodology allowed us to analyze a time series for each municipality in the data set . Negative binomial regression of panel data , widely used in econometric literature , has recently been introduced in health studies [36] , [37] , [39] , [40] . Panel data essentially defined a time series analysis for each municipality and contrasted the trends between them , making this a more rigorous approach than a simple purely cross-sectional data [48] . We used a coverage indicator combining BFP coverage of the total population of the municipality and BFP coverage of the target population ( poor and extremely poor ) . We did this to estimate the “spill over” effect of the BFP on inhabitants of the municipality that were not enrolled in the programme [40] . Additionally , because leprosy is a highly focused disease in some regions of Brazil , only municipalities located in areas with high disease burden were included in the analysis . Therefore , the results can not be generalized to municipalities in areas of low prevalence of leprosy . Leprosy clusters were formed by different groups of neighboring municipalities . Some municipalities in these clusters had lower case detection rates than the average case detection rate in Brazil . It is possible that fewer cases were detected because of limitations of the healthcare system , such as low population coverage and the inability of healthcare professionals to diagnosis leprosy . Municipalities with a low detection rate that are located in high-risk areas have to intensify case finding and treatment [3] . Another possible limitation was that the annual values of sociodemographic variables were obtained from linear interpolation and extrapolation from decennial census data . Since we did not expect substantive changes in these trends is unlikely that these estimates introduced any significant source of error . However , the categorization of variables can limit the possible bias introduced by the techniques of crude interpolation by smoothing sharp fluctuations artificially introduced by the method . Making socioeconomic covariate data at the municipal level available for inclusion in multivariate analyses strengthens the case for the effectiveness of health programs . This is particularly important for the case of Brazil and several other countries in Latin America , where the expansion of health services in the last decade has occurred simultaneously with other forms of social progress , such as improvements in sanitation infrastructure , educational attainment , and economic development [58] . We did not think it necessary to include in the model a variable representing time as in our view any secular trend was controlled by the use of rate ratios , contrasting different groups of coverage changes according to the same time trends . Moreover relevant confounding factors , which could have been represented by an artificial time variable , have been included in the models , and the individual-specific term of the fixed effects model control for time-invariant unobserved confounding variables [48] . Sensitivity analysis showed that the introduction of a time variable created an over specification problem in the models . One of the many strengths the study is that expansion of BFP and the FHP at different rates in the Brazilian municipalities in recent decades created the opportunity to investigate their effects on new case detection rate of leprosy . Despite the limitations , the results of this study are consistent and illustrate the contribution FHP in improving diagnosis and therefore of the control of leprosy . It also point for a positive effect of the BFP cash transfer in reducing leprosy , confirming the contribution of the social determinants to leprosy control . The conditional cash transfer programs has steadily increased around the world , including in leprosy endemic countries located in Africa and Asia , such as Nigeria and Indian [2] , [21] , [22] . Conditional cash transfer programs are one way to boost demand and reduce barriers to access for health services particularly in primary health care units to poor and extremely poor individuals . Thus , it is necessary an effective primary health care in these populations able to comply with basic health needs and have attending conditions required by the conditional cash transfer programs in these countries . Given the expansion of cash transfer programs and their relevance to public health it is necessary to accumulate evidence of mechanisms and pathways through which cash transfers affect epidemiologically related factors leprosy and other poverty related disease . Social interventions , such as conditional cash transfer programs for the poorest groups , improvements in health care , and progress in social and environmental determinants are essential for the control of poverty related infectious diseases and in particular leprosy [59] . It is expected that these results contribute with arguments to the discussion on the relationship between distributive social policies , primary health care and health conditions of the population in developing countries worldwide . | Leprosy is considered a poverty related disease . Not much is known about the effectiveness of welfare and primary health care policies on reduction of leprosy occurrence . We conducted a study to evaluate the impact of the Brazilian conditional cash transfer ( Bolsa Família Program ) and the Primary Health Care ( Family Health Program ) on the new case detection rate of leprosy in the period 2004–2011 in the Brazilian municipalities . All variables were obtained from national databases and a total of 1 , 358 municipalities were included in the analysis . The new case detection rate of leprosy was significantly reduced in municipalities with intermediate , high , and consolidated BFP coverage . There was a significant increase in new case detection rate of leprosy as Family Health Program coverage increased . We interpret this to mean that at the same time the primary health care had been effective increasing the new case detection rate of leprosy in Brazil , there is an impact of conditional cash transfer in the reduction of the new case detection rate of leprosy due to reduction in leprosy incidence . We expect that these results contribute with arguments to the discussion on the relationship between distributive social policies and health conditions of the population in developing countries worldwide . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"social",
"epidemiology",
"bacterial",
"diseases",
"infectious",
"diseases",
"public",
"and",
"occupational",
"health",
"plant",
"science",
"medicine",
"and",
"health",
"sciences",
"primary",
"care",
"infectious",
"disease",
"epidemiology",
"epidemiology",
"plant",
"pathology",
"biology",
"and",
"life",
"sciences",
"population",
"biology",
"tropical",
"diseases",
"neglected",
"tropical",
"diseases",
"leprosy",
"health",
"care"
]
| 2014 | Effect of the Brazilian Conditional Cash Transfer and Primary Health Care Programs on the New Case Detection Rate of Leprosy |
Many recurrent chromosome translocations in cancer result in the generation of fusion genes that are directly implicated in the tumorigenic process . Precise modeling of the effects of cancer fusion genes in mice has been inaccurate , as constructs of fusion genes often completely or partially lack the correct regulatory sequences . The reciprocal t ( 2;13 ) ( q36 . 1;q14 . 1 ) in human alveolar rhabdomyosarcoma ( A-RMS ) creates a pathognomonic PAX3-FOXO1 fusion gene . In vivo mimicking of this translocation in mice is complicated by the fact that Pax3 and Foxo1 are in opposite orientation on their respective chromosomes , precluding formation of a functional Pax3-Foxo1 fusion via a simple translocation . To circumvent this problem , we irreversibly inverted the orientation of a 4 . 9 Mb syntenic fragment on chromosome 3 , encompassing Foxo1 , by using Cre-mediated recombination of two pairs of unrelated oppositely oriented LoxP sites situated at the borders of the syntenic region . We tested if spatial proximity of the Pax3 and Foxo1 loci in myoblasts of mice homozygous for the inversion facilitated Pax3-Foxo1 fusion gene formation upon induction of targeted CRISPR-Cas9 nuclease-induced DNA double strand breaks in Pax3 and Foxo1 . Fluorescent in situ hybridization indicated that fore limb myoblasts show a higher frequency of Pax3/Foxo1 co-localization than hind limb myoblasts . Indeed , more fusion genes were generated in fore limb myoblasts via a reciprocal t ( 1;3 ) , which expressed correctly spliced Pax3-Foxo1 mRNA encoding Pax3-Foxo1 fusion protein . We conclude that locus proximity facilitates chromosome translocation upon induction of DNA double strand breaks . Given that the Pax3-Foxo1 fusion gene will contain all the regulatory sequences necessary for precise regulation of its expression , we propose that CRISPR-Cas9 provides a novel means to faithfully model human diseases caused by chromosome translocation in mice .
Rhabdomyosarcoma ( RMS ) is the third most common soft-tissue sarcoma in children with an annual incidence of five new cases per million . It accounts for 5–8% of all pediatric cancer . RMS belongs to the family of small round blue cell tumors of childhood and exhibits histological features of skeletal muscle . Two major histological subtypes of RMS can be distinguished , embryonal ( E-RMS ) and alveolar ( A-RMS ) . E-RMS has its highest incidence in infants and young children whereas A-RMS is more frequent in older children and adolescents . A-RMS has a more aggressive clinical behavior with early dissemination , a poor response to chemotherapy , frequent relapses , and a 5-year failure-free survival of 65% after treatment [1] . A-RMS is found predominantly in the extremities ( 42% ) , parameningeal ( 17% ) , head and neck ( 11% ) and other locations ( 21% ) [1] including the trunk , perirectal and perianal areas [2 , 3] . Cytogenetically A-RMS is distinguished from E-RMS by one of two recurrent chromosome translocations: t ( 2;13 ) or t ( 1;13 ) , which result in fusion of PAX3 or PAX7 to FOXO1 , respectively [4] . In spite of multiple attempts to identify the cell of origin in which the t ( 2;13 ) occurs the question remains unanswered . It was shown previously that transcription occurs at a few hundred discrete nuclear sites called transcription factories [5] . Some genes frequently involved in a recurrent chromosome translocation ( MYC and IGH in B lymphoid progenitors , TMPRSS2 and ERG or ETV1 in prostate cancer , RET and H4 in in radiation-associated papillary thyroid cancer ) co-localize to the same transcription factory [6–9] . Initial chromosome conformation capture experiments in activated mouse B cells suggested that physical proximity of the IGH and MYC loci is a minor contributor to the frequency of chromosome translocation [10] . However , combined high resolution Hi-C mapping and genome-wide translocation sequencing in transformed mouse pre-B cells found good coincidence between chromosomal translocation and spatial proximity [11] . A possible driver of double strand DNA breaks might be the co-localization of replication stress-induced early replication fragile sites ( ERFSs ) with highly expressed gene clusters [12] . Though it was demonstrated that ectopic expression of PAX3-FOXO1/Pax3-Foxo1 can transform mouse mesenchymal stem cells in vitro [13] as well as Myf6+ myofibers in vivo [14] in view of the above these cell types seem unlikely hosts for the chromosome translocation given that they do not express Pax3 . In fact , the suggestion that Myf6+ myofibers might be the host of the PAX3-FOXO translocation was recently rectified [15] . In contrast , Pax3 is expressed in activated myoblasts upon muscle injury or in growing muscles during normal development [16] . Moreover , PAX3-FKHR , in cooperation with loss of p16INK4A expression , transforms both fetal and postnatal primary human skeletal muscle cell precursors [17] . Together these observations suggest that translocation might occur in a population of activated myoblasts that express PAX3 ( PAX3+ ) . It has been shown that Pax3 expression differs among different muscles in the mouse [18 , 19] . There are many more Pax3+ cells in fore limb than in hind limb muscles [19] . Muscle satellite cells from the masseter and soleus did not express Pax3 while only 7% of those from the extensor digitorum longus ( EDL ) did . In contrast 49% of satellite cells from the biceps were Pax3+ . In addition , most ventral trunk muscles were Pax3-positive and 64% of satellite cells from the diaphragm expressed Pax3 . Importantly , primary myoblast cultures of Pax3+ satellite cells remain Pax3+ , while Pax3- satellite cells from hind limb remain negative [19] . Studies addressing the relation between spatial chromosome proximity and translocation have been performed in cells of the B-lymphoid lineage or of hormone-responsive lineages mostly using transformed cell lines [6 , 7 , 9] . Recently CRISPR-Cas9 nuclease ( Clustered Regularly Interspaced Short Palindromic Repeats ( CRISPR ) /CRISPR-associated systems ) [10] was used to engineer human tumor-associated translocations [20] . To answer the question if locus proximity of Pax3 and Foxo1 in low-passage primary mouse myoblasts contributes to the frequency of Pax3-Foxo1 fusion gene formation we used the CRISPR-Cas9 system to induce double strand DNA breaks ( DSBs ) , which spurred by non-homologous end joining repair ( NHEJ ) produce chromosome translocations between these two loci . We used synthetic single-guide RNAs ( sgRNA ) to program Cas9 to induce DNA double-strand breaks ( DSBs ) in Pax3 and Foxo1 [21–23] . Unlike the human PAX3/7 and FOXO1 genes , mouse Pax3/7 and Foxo1 are in opposite orientation on their respective chromosomes ( 1 , 4 , and 3 ) . Compared with human chromosome 13 , Foxo1 is part of an inverted 4 . 9Mb syntenic region on mouse chromosome 3 . Although a recurrent complex inversion/translocation event involving the oppositely oriented ETV6 and c-ABL genes in humans gives rise to the ETV6-ABL fusion gene in some myeloid and lymphoid malignancies , the frequency of this event is extremely low [24] . Therefore , to successfully generate a CRISPR-Cas9-mediated Pax3-Foxo1 fusion gene we used chromosomal engineering via Cre recombinase-mediated genetic alterations to create a mouse in which the Foxo1 containing 4 . 9 Mb syntenic region is inverted ( Foxo1-inv+/+ mice ) . Previously , Cre recombinase-mediated inversions of large fragments of chromosomes have been used to create balanced chromosomes [25–29] . We show that myoblasts isolated from fore and hind limb keep their Pax3-expressing identity and co-localization of Pax3 and Foxo1 loci strongly correlates with the level of Pax3 expression and generation of a CRISPR-Cas9 induced t ( 1;3 ) , which is more frequent in fore limb myoblasts . Our Foxo1inv+/+ mice will be a valuable tool for studying mechanisms underlying the initial stages of the A-RMS implicated chromosome translocations resulting in development of better animal models for this pediatric cancer and other human diseases caused by chromosome translocations .
Since close physical proximity of translocation partners might facilitate chromosome translocation , we determined if Pax3 and Foxo1—the translocation partners in A-RMS—are co-localized in actively proliferating low-passage primary mouse myoblasts . DNA-FISH analyses of the Pax3 and Foxo1 loci in interphase nuclei of primary limb myoblasts of newborn pups , after one week in culture showed 13% co-localization , which was significantly higher than in similarly cultured MEFs ( 2% , the background of this method; Fig . 1A ) . We hypothesized that co-localization of Pax3 and Foxo1 loci in myoblasts reflects the percentage of Pax3+ cells in the original newborn muscles . To test this hypothesis we isolated myoblasts from hind and fore limbs of newborn pups and compared the frequency of co-localization of Pax3 and Foxo1 loci in proliferating myoblasts from these two sources . It was shown previously with a Pax3 knock-in reporter gene that many more satellite cells in fore limb muscle express Pax3 than in hind limb muscle [19] . In accordance , the percentage of co-localized Pax3 and Foxo1 loci was notably higher in fore limb than in hind limb myoblasts in 8 independent experiments ( Fig . 1A–C ) . In addition , Q-RT-PCR of RNA from these myoblasts confirmed that expression of Pax3 was six-fold higher in fore limb myoblasts ( Fig . 1D ) . These results are in agreement with the published observation that satellite cells maintain their Pax3+ identity upon activation in vitro . Expression of other genes such as Foxo1 and Pax7 was similar in the two types of myoblasts ( Fig . 1D ) . The results for Pax3 expression were reproducible given that a number of independent experiments produced similar data ( S1 Fig . ) . Because diaphragm was shown to contain the highest number of Pax3+ myoblasts [19] , we compared by FISH the co-localization of the Pax3 and Foxo1 loci in myoblasts isolated from fore limb , hind limb , and diaphragm of the same adult mouse . Indeed , diaphragm myoblasts showed a higher co-localization of the two loci ( 20% ) than fore limb ( 11% ) or hind limb ( 9% ) myoblasts . The mouse Foxo1 gene is located on chromosome 3 in a 4 . 9 Mb DNA fragment ( ch3:52 , 059 , 615–56 , 995 , 963 ) that is syntenic with human chromosome 13 ( ch13:41 , 254 , 213–34 , 463 , 185 ) but positioned in the opposite orientation ( Fig . 2A ) . This places Foxo1 in the mouse in a reverse transcriptional direction with respect to that of the Pax3 or Pax7 genes . To engineer a mouse capable of acquiring productive Pax3/7-Foxo1 fusion genes via a simple balanced t ( 1;3 ) or t ( 4;3 ) , we performed two consecutive rounds of ES cell targeting in which we introduced two pairs of non-compatible LoxP sites at either border of this syntenic region with the goal to create a Cre-recombinase mediated permanent inversion of the 4 . 9Mb DNA fragment ( S2 Fig . ) . Without inversion there would only be two ways to produce a productive fusion: 1 ) Via a translocation in which the resulting chromosomes would carry a double centromere and no centromere , respectively , an option likely to be non-viable in primary myoblasts and 2 ) Via a complex inversion/translocation event as described for the human ETV6-ABL fusion gene [24] , a rare event , which likely would reduce the frequency of fusion gene formation below detectable levels . The centromeric border of the mouse/human syntenic region is located 15 kb upstream of the Foxo1 start codon ( Fig . 2B ) . To precisely target this border in ES cells we used recombineering in E . coli [30] to modify the RP24–391O12 BAC ( bacterial artificial chromosome ) clone , so that it carries non-compatible mutant 511-ILoxP and wtLoxP sites [31] flanking the hph ( hygromycin B resistance ) and tk ( HSV1-thymidine kinase ) selectable marker genes ( Figs . 2B , S2 ) . The precise targeting of the border of the syntenic region minimizes the chance of disturbing any potentially important regulatory sequences that might affect Foxo1 expression ( Fig . 2B , top ) . We targeted ES cells with linearized RP24–391O12-LoxP-hygro-TK BAC DNA and counter selected hygromycin B resistant clones carrying random integrations by screening for the presence of vector sequences remaining on either side of the insert . Colonies containing such vector segments were discarded [32] . The remaining clones were subjected to the ‘loss-of-native-allele’ assay using real-time quantitative PCR [33] . For copy number control of stably integrated target DNA we used the telomeric border of the syntenic region as a reference . In total 273 clones were analyzed , two of which contained a single copy of the wild type locus ( Fig . 2C ) . These clones were submitted to FISH analysis and karyotyping which confirmed the presence of only two native signals on chromosome 3 when hybridized with a wild type BAC RP24–391O12 probe ( Fig . 2D ) . For consecutive targeting of the telomeric border of the syntenic region we selected clone XIIB3 , which had a 100% normal diploid karyotype . For targeting of the telomeric border of the syntenic region we followed the same strategy and engineered a BAC clone carrying the 511-ILoxp-Neo-TK-wtLoxP cassette inserted at the precise syntenic border ( Fig . 2B , bottom ) . The recombinant RP23–422I13-LoxP-Neo-TK BAC was linearized in such a way that only very short vector fragments remained at either side of the insert . After targeting in ES cells , analysis with the ‘loss-of-native-allele’ assay of 48 clones proved sufficient to obtain the desired recombinant . Two clones carrying a single copy of the wild type telomeric locus ( Fig . 2E ) were analyzed by FISH using the RP24–391O12 and RP23–422I13 BAC probes . One of them ( 13D3 ) showed two native signals on chromosome 3 with either BAC probe ( Fig . 2F ) . This clone had a 90% normal diploid karyotype and we next determined if it carried cis- or trans-targeted borders of the 4 . 9 Mb syntenic region . To discriminate between these two possibilities we transiently transfected a Cre recombinase plasmid into the double-targeted 13D3 ES cells and DNA isolated from the pool of electroporated cells was analyzed by PCR using only forward or reverse primers from both targeted borders . Both PCRs produced bands indicating that the pool contained cells carrying the 4 . 9Mb inversion . The same pool of cells was counter selected with FIAU and 23 resistant clones were analyzed by PCR ( Fig . 3A ) . Sixteen clones harbored the 4 . 9Mb inversion and two of these were selected , A6 and C5 , which had a 100% and 93% normal karyotype , respectively . Inversion of the 4 . 9Mb region in these clones was subsequently confirmed by FISH analysis ( Fig . 3B ) using the RP24–391O12 and RP23–422I13 BAC probes . The chromosome containing the inversion showed split hybridization signals while the wild type chromosome produced contiguous signals with these probes . These ES cell clones were used to generate chimeric mice that transmitted the inversion of the Foxo1 syntenic region to heterozygous Foxo1-inv+/- offspring . Foxo1-inv+/- mice were fertile and produced Foxo1-inv+/+ offspring at the expected Mendelian frequency . Foxo1-inv+/+ animals did not exhibit any obvious phenotypic abnormalities and showed normal fecundity and life span . Moreover , western blot analysis confirmed that Foxo1-inv+/+ primary myoblasts and wild type myoblasts expressed equal amounts of Foxo1 protein ( Fig . 3C ) and co-localization of the Pax3 and Foxo1 loci was equal in Foxo1-inv+/+ and wild type myoblasts ( 8% , S5 Fig . ) . Finally , DNA-FISH analysis of Foxo1-inv+/+ fibroblasts with RP24–391O12 and RP23–422I13 BAC probes confirmed that both chromosomes 3 carried the 4 . 9Mb inversion ( Fig . 3D ) . Nuclear receptor-induced chromosomal proximity of TMPRSS2 and ERG or TMPRSS2 and ETV1 promotes the occurrence of nonrandom ligation sites upon translocation between these partner genes , thereby generating unique breakpoint “hot spots” [6] . It is possible that translocations in A-RMS are non-random and occur predominantly at sites , coming in close proximity during co-regulated expression . We hypothesized that directing DSBs to sites in mouse Pax3 and Foxo1 homologous to those in PAX3 and FOXO1 in an ARMS cell line carrying a t ( 2;13 ) might increase the chance of generating a t ( 1;3 ) in proliferating Foxo1-inv+/+ myoblasts after Cas9 induced DSBs . We chose to mimic the breakpoints of the widely used ARMS cell line RH30 ( S3 and S4 Figs . ) . Alignment of human and mouse Pax3 and Foxo1 sequences mapped the RH30-like breakpoints at positions 78105273 on mouse chromosome 1 and 52300558 on mouse chromosome 3 ( Fig . 4A ) . We chose unique protospacer sequences followed by a 5’-GGT PAM as close as possible to the RH30-like breakpoints in both Pax3 and Foxo1 ( Fig . 4B ) . Cas9 introduces DSB three nucleotides downstream of the two PAM sequences , which would result in DSBs between nucleotides 78105248 and 78105247 on chromosome 1 in intron 7 of Pax3 and between nucleotides 52300541 and 52300542 ( coordinates in the non-inverted sequence ) on chromosome 3 in intron 1 of Foxo1 ( Fig . 4B ) . For gene delivery to the myoblasts we cloned the human codon optimized Cas9 ( hCas9 ) into the pCL20C [34] lentiviral vector downstream of the MSCV promoter and upstream of an IRES-YFP fluorescent marker ( Fig . 4C ) . In order to express two different sgRNAs form a single vector we constructed a second pCL20C dual sgRNA vector in which the Pax3-specific sgRNA was driven by the human U6 promoter and the Foxo1-specific sgRNA by the mouse U6 promoter ( Fig . 4C , D ) . We first determined that with our current batch of serum maximum co-localization of Pax3 and Foxo1 occurred at 7–8 days of culture after myoblast isolation . This time point synchronized with Cas9 and sgRNA expression should therefore maximize the probability of introducing DSB in closely positioned Pax3 and Foxo1 loci . Hence 24 hours after isolation we transduced primary fore and hind limb myoblasts of Foxo1-inv+/+ pups , Foxo1-inv+/+ MEFs and fore limb myoblasts from wild type mice with Cas9 lentivirus ( Fig . 4C ) . After FACS sorting for YFP , cells were expanded and transduced with lentivirus expressing the RH30-like sgRNAs at day 7 after isolation and with an SV40 large T antigen expressing lentivirus at day 8 . The latter was done to prevent senescence of the myoblasts during puromycin selection and allows subsequent expansion of the culture . To detect the Pax3-Foxo1 fusion DNA fragments from Cas9/sgRNAs expressing myoblasts and MEFs we used the Pax3-RH30F ( forward ) and Foxo1-RH30R ( reverse ) primers for PCR analysis , which are positioned downstream and upstream of the putative Cas9-induced Pax3 and Foxo1 DSBs ( Fig . 5A ) . PCR amplification of DNA from 104 cells produced bands of 250 bp or shorter in Cas9/sgRNAs expressing hind limb and fore limb Foxo1-inv+/+ myoblast ( Fig 5B , lanes 2 and 4 ) . However , no product was detected upon PCR amplification of DNA from 104 hind and fore limb Foxo1-inv+/+ myoblasts not treated with Cas9/sgRNAs ( Fig . 5B , lanes 1 and 3 ) or from 104 Cas9/sgRNAs expressing Foxo1-inv+/+ MEFs or wild type fore limb myoblasts ( Fig . 5B , lanes 5 and 6 ) . As a control we verified that the difference in translocation frequency between myoblasts and MEFs was not caused by differences in CRISPR-Cas9’s accessibility to chromatin , given that Pax3 is not expressed in MEFs . The CRISPR-Cas9 breakpoint in Pax3 falls within a MaeIII restriction endonuclease site and that in Foxo1 within a DdeI site . Therefore we PCR amplified the Pax3 and Foxo1 fragments spanning the breakpoints and digested them with MaeIII or DdeI . This showed that 96% ( Pax3 ) and 97% ( Foxo1 ) of the PCR products of CRISPR-Cas9 treated myoblasts were resistant to MaeIII or DdeI digestion , whereas in CRISPR-Cas9 treated MEFs these numbers were 72% for both enzymes ( Fig . 5C , D ) . Thus , there was no great difference in chromatin accessibility . Moreover , the Pax3 and Foxo1 chromatin in MEFs was equally accessible to CRISPR-Cas9 , despite the fact that Foxo1 is and Pax3 is not expressed in these cells . Cloning of the CRISPR-Cas9 induced fusion DNAs , followed by sequencing of 45 individual clones of each of the PCR products , produced 39 and 34 translocation breakpoint sequences from fore and hind limb myoblasts , respectively . This identified 6 different breakpoint sequences from fore limb and 3 different breakpoint sequences from the hind limb myoblasts . This represents the minimal number of translocation events per 104 cells ( Fig . 5E , top and bottom ) . Taking into account the percentage of locus co-localization ( Fig . 5F ) these numbers translate to a minimal translocation frequency of 1 in 150 in fore limb and 1 in 200 in hind limb myoblasts , respectively . The only sequence in common between the fusion fragments from these two types of myoblasts was the cleanly re-ligated fusion , without missing or added base pairs . The other 7 ( 5 from fore limb myoblast and 2 from hind limb myoblast ) were all unique and carried NHEJ-mediated deletions varying from 6 to 71 bp . Superimposed on the deletion , two of the clones also contained randomly added base pairs . Notably , three additional breakpoint sequences obtained from an independent experiment ( S5 Fig . ) were different from the 7 shown in Fig . 5E and underline the mutation-prone repair of the NHEJ DNA-repair machinery during the translocation event . Together these results show excellent correlation between the frequency of translocation , co-localization , and expression of the Pax3 and Foxo1 loci in primary myoblasts . It was highest in fore limb myoblasts , lower in hind limb myoblasts and undetectable in MEFs . Although wild type myoblasts show the same frequency of locus co-localization as Foxo-inv+/+ myoblasts ( S6 Fig . ) , the opposite orientation of Foxo1 prevented the formation of a productive Pax3-Foxo1 fusion gene . Next we performed RT-PCR on equal amounts of total RNA from fore limb and hind limb myoblasts to detect the Pax3-Foxo1 fusion mRNA . In support of the higher frequency of chromosome translocation in fore limb myoblasts , we were able to RT-PCR amplify the Pax3-Foxo1 cDNA from these myoblasts ( Fig . 5G ) but not from the hind limb myoblasts using an equal amount of input RNA ( not shown ) . Sequence analysis of the cDNA confirmed the correctly spliced Pax3 exon 7-Foxo1 exon 2 fusion ( Fig . 5G ) . To further characterize the t ( 1;3 ) we repeated the experiment in Foxo1-inv+/+/Ink4a-ARF-/- myoblasts . Due to loss of a functional p53 pathway Ink4a-ARF-/- myoblasts do not senesce during further experimental manipulation . Based on the Pax3 and Foxo1 co-localization data at the time of induction of the t ( 1;3 ) ( 11% in fore limb myoblasts and 7% in hind limb myoblasts ) we assumed that the frequency of translocation events in these myoblasts should not be lower than in the myoblasts used in Fig . 5 , i . e . at least 6 independent translocation events per 104 fore limb myoblasts . This frequency is too low for further molecular and functional analyses . To enrich the cell pool for the t ( 1;3 ) carrying cells , we evenly distributed 104 cells between the wells of three 96-well plates ( on average 30 cells per well ) . PCR analyses of the DNA of 95 wells from the first plate identified 3 potentially t ( 1;3 ) -enriched cell pools ( S7 Fig . ) . Pool 1E10 was lost during the freeze-thawing cycle but FISH analyses detected the reciprocal t ( 1;3 ) in 64% of pool 1G3 metaphase cells ( Fig . 6A–C ) and in 4% of pool 1D10 metaphase cells . Both the derivative chromosomes 1 and 3 were detected in all t ( 1;3 ) positive cells , confirming that the translocation was reciprocal . To determine if the t ( 1;3 ) resulted in expression of Pax3-Foxo1 protein we immunoprecipitated three cell lysates each of the t ( 1;3 ) -negative ( 1H3 ) and t ( 1;3 ) -positive ( 1G3 ) pools with either an anti-Pax3 or an anti-Foxo1 antibody . The Pax3 IPs were then immunoblotted with the anti-Pax3 antibody , showing the Pax3 and Pax3-Foxo1 bands ( Fig . 6D , Pax3/Pax3 panel ) , or with anti-Foxo1 antibody showing only the Pax3-Foxo1 band ( Fig . 6D , Pax3/Foxo1 panel ) . Similarly , immunoblotting the Foxo1 IPs with anti-Pax3 antibody again showed the Pax3-Foxo1 fusion protein ( Fig . 6D , Foxo1/Pax3 panel ) while immunoblotting with the Foxo1 antibody showed both Foxo1 and the fusion protein ( Fig . 6D , Foxo1/Foxo1 panel ) . This confirmed that the engineered t ( 1;3 ) expressed the fusion protein , which allowed us to assess if it affected the expression of Pax3-Foxo1’s transcriptional targets . We performed RNA-seq analysis comparing the mapped sequence reads of presumed PAX3-FOXO1 target genes [2] in the 1G3 pool ( 64% Pax3-Foxo1 positive ) with those in the 1H3 pool ( Pax3-Foxo1 negative ) ( S1 Table ) . This showed that roughly half the targets of PAX3-FOXO1 were correctly up or down regulated in the 1G3 pool . The same comparison with a PAX3-FOXO1 expression signature obtained with the ectopic PAX3-FOXO1 expressing ERMS cell line RD [35] , also showed coincident regulation of half the targets ( S2 Table ) , suggesting that the t ( 1;3 ) generated fusion protein is active .
For the precise modeling of human recurrent chromosome translocations and their impact on disease development in mice , reenactment of the actual translocation would be the closest possible recapitulation of the sequence of events in humans . Until now such reenactment was a daunting task as the translocation would require introduction of LoxP [36 , 37] or Frt recombination sites into both translocation partners via homologous recombination in ES cells , followed by expression of Cre or Flp recombinase to create DSBs that would mediate the translocation . As shown by others [20] and here , the availability of the CRISPR-Cas9 system has paved the way to implementing this approach without such major technical or time investment . Given the high homology between mouse and human genes and their regulatory sequences , this approach is likely to include all sequences that are important for the precise regulation of the mouse fusion gene as it occurs in humans . The first and only published model for ARMS [38] in which expression of a conditional Pax3-Foxo1 knock-in fusion oncogene is induced by a Myf6 driven Cre had a low incidence and long latency of tumor development , requiring the presence of two Pax3-Foxo1 alleles on a Trp53-null or Ink4a/Arf-null background . One reason for this might be that the level of expression of the fusion oncogene in this KI model is inadequate for shorter latency tumor development . An argument against this possibility is that a high level of PAX3-FOXO1 expression induces cell death [39] , most likely due to transcriptional activation of the Pax3-Foxo1 pro-apoptotic target gene Noxa1 [40] . Unlike other studies [41 , 42] , the KI Pax3-Foxo1 gene contained some Foxo1 genomic sequences that allowed expression of the fusion gene in adult mice , but despite their presence the construct might lack sequences that mediate human-like regulation of fusion gene expression , which in turn might be crucial for efficient tumor development . In agreement with published data [19] we established that co-localization of Pax3 and Foxo1 in our culture system was higher in forelimb than in hind limb myoblasts , which coincided with higher Pax3 expression in forelimbs . Due to experimental variability the percentage of co-localization of the two loci varied in 8 independent experiments , but co-localization in the fore limbs was always higher than in the hind limbs . Therefore our myoblast model represents a graded system to determine if these features contributed to the frequency of chromosome translocation in low passage primary myoblasts upon introduction of targeted DSBs . To perform these experiments and to eventually develop a precise mouse model of ARMS , the transcriptional orientation of Foxo1 on chromosome 3 needed to be inverted . We followed the Cre-dependent one-way inversion of a DNA fragment in mice as was previously demonstrated by Schnütgen and colleagues [43] . To avoid disturbing the transcriptional regulation of the inverted Foxo1 , we decided to invert the mouse/human 4 . 9 Mb syntenic region encompassing Foxo1 , rather than the gene itself . Although the centromeric border of this region is only 15 kb upstream of Foxo1 , we reasoned that all important Foxo1 regulatory sequences should be contained within this region otherwise it would not be syntenic with human FOXO1 on chromosome 13q14 . 1 . Although we did not analyze the detailed expression of Foxo1 in Foxo1-inv+/+ mice during pre- and postnatal life , the animals did not show any obvious phenotypic abnormalities . In addition , they had a normal lifespan , normal fecundity , and the level of Foxo1 protein expression and co-localization of the Pax3 and Foxo1 loci in myoblasts were identical to those of wild type mice . Together these observations made the Foxo1-inv+/+ myoblasts suitable for our translocation experiments . To determine if the level of co-localization of Pax3 and Foxo1 in primary myoblasts affected the frequency of chromosome translocation between these loci upon induction of targeted DSB , we transduced the cells with Cas9 and dual sgRNA expressing lentiviruses . Combining the three genes into a single lentiviral vector failed to produce viral particles . We targeted the CRISPR-Cas9 DSBs to sequences in Pax3 and Foxo1 that mediated the t ( 2;13 ) in the A-RMS cell line Rh30 . Both breakpoints are present in sequences conserved between the mouse and human genes , suggesting that they occurred in non-redundant sequences that might bind factors with a role in expression regulation of both genes . Currently we do not know if this affects the frequency of translocation , which is a possibility that can be tested in future by choosing sgRNAs targeting non-conserved sequences within the target Pax3 and Foxo1 introns . We found excellent positive correlation between the frequency of the t ( 1;3 ) and the percentage of locus co-localization using FISH analysis . This also correlated with the level of Pax3 expression , which is much higher in fore limb than hind limb myoblasts and absent in MEFs , while Foxo1 expression is ubiquitous . Given that the frequency of CRISPR-Cas9 induced DSBs in Pax3 and Foxo1 is comparable in myoblasts and MEFs , it is the proximity of the loci in these primary cells that facilitates trans-chromosomal ligation producing the two expected derivative chromosomes during NHEJ DNA repair . The derivative chromosome 3 produced correctly spliced Pax3-Foxo1 mRNA , encoding active Pax3-Foxo1 protein that up/down-regulated expression of approximately half the presumed PAX3-FOXO1 targets in the 64% Pax3-Foxo1-positive cell pool ( S1 Table ) . The genes compiled in this table are differentially expressed in ARMS versus ERMS tumors or have been identified by forced expression of PAX3-FOXO1 in different cell lines , including NIH3T3 cells , MEFs , SAOS2 cells and C2C12 cells ( [2] and references therein ) . Because the cell background affects the range of PAX3-FOXO1 target gene expression [44] , none of the published scenarios reflect expression of Pax3-Foxo1 in primary p16/Arf-/- mouse myoblasts . Possibly this is the reason for the 45% match of reported PAX3-FOXO1 up or down regulated genes . Comparison with genes up or down regulated in the ERMS cell line RD transduced with PAX3-FOXO1 retrovirus [35] showed 52% coincident regulation ( S2 Table ) . Clearly , the t ( 1;3 ) generated Pax3-Foxo1 protein in mouse myoblasts is active and changes the expression of target genes in an ARMS-like manner . One would expect that the frequency of translocation in myoblast that show co-localization of the two translocation partners would be the same irrespective of the source of myoblasts . We found a frequency of 1/150 and 1/200 in fore and hind limb myoblasts , respectively , which we believe does not represent a difference given the uncertainty of how many translocation events actually took place ( we can only count those that give distinguishable fusion products ) . Our results in mouse myoblasts suggest that human myoblasts can be a cell of origin for the PAX3-FOXO1 translocation as they would provide a favorable environment for the translocation to occur , i . e . expression of both genes and spatial co-localization . It is curious that A-RMS is more frequent in the lower than in the higher extremities in humans , as reported by Neville and co-workers [45] . This apparent inconsistency with our mouse data might be explained by the possibility that humans may not have a difference in the distribution of PAX3 expression in the upper and lower extremities . In addition , the muscle mass and presumably the number of satellite cells in the lower extremities in humans is much higher than in the upper extremities , hence increasing the number of translocation-competent cells and frequency of translocation . By using CRISPR-Cas9 nuclease we showed that targeted chromosome translocations could be induced with high efficiency . Unlike other approaches that have relied on induction of chromosome translocation using γ−irradiation , DSB-inducing chemicals , or the lymphoid cell-specific gene rearrangement machinery , CRISPR-Cas9 can be employed in any cell type . Due to its specificity the system is suitable for use in vivo in cell culture or in mice . Application of this system will greatly facilitate the development of mouse models that precisely recapitulate chromosome translocation-induced human diseases .
A complete list of E . coli strains used for this work can be found in S1 Protocol . BAC clones RP24–391O12 ( centromeric border of the 4 . 9 Mb syntenic region ) and RP23–422I13 ( telomeric border of the 4 . 9 Mb syntenic region ) were purchased from the BACPAC Resource Center ( BPRC ) , Children’s Hospital Oakland Research Institute in Oakland , California , USA ( http://bacpac . chori . org ) . The complete list of PCR Primers and oligonucleotides can be found in S1 Protocol . A modified pNeoTKLoxP was recombineered into BAC RP23–422I13 ( telomeric border of the syntenic region ) . In pNeoTKLoxP we replaced the wild type ( wt ) LoxP site downstream of the TK gene with the 511-ILoxP sequence ( annealed oligonucleotides TK-511-ILoxP and TK-511-ILoxP-C ) . Then , via recombineering , we introduced the EM7 promoter upstream of the Neo gene . We therefore transformed electrocompetent DY380 Ecoli cells , containing the wtLoxPNeoTK-511-IloxP plasmid with the TK-EM7-Neo fragment ( ends of the annealed oligonucleotides TK-EM7 and EM7-NeoC had been filled-in with Klenow DNA polymerase ( Invitrogen ) following the manufacturer’s protocol ) . For recombineering we followed the protocol posted on the Frederick National Laboratory for Cancer Research web site: http://ncifrederick . cancer . gov/research/brb/protocol/Protocol1_DY380 . pdf . A short 5’-arm ( annealed phosphorylated oligonucleotides 5-tel-s and 5-tel-s-C ) was cloned downstream of 511-IloxP and a short 3’-arm was cloned upstream of wtLoxP ( annealed phosphorylated oligonucleotides 3-tel and 3-tel-C ) . A modified pBSLoxPTKhygro plasmid ( kind gift from Drs . M . Roussel and F . Zindy , SJCRH ) was recombineered into BAC RP24–391O12 ( centromeric border of the syntenic region ) . In this construct we inserted a 511-ILoxP sequence upstream of the TK-promoter-Neo sequence . Since the activity of TK promoter in prokaryotic cells was sufficient to ensure Hygromycin B resistance , we did not introduce the bacterial EM7 promoter in this construct . A short 5’-arm ( annealed phosphorylated oligonucleotides 5-cent-s and 5-cent-s-C ) was cloned upstream of the 511-IloxP site and a short 3’-arm was cloned downstream of wtLoxP ( annealed phosphorylated oligonucleotides 3-cent and 3-cent-C ) . The pCL20c-MSCV-IRES-YFP vector backbone was generated by replacing GFP of pCL20c-MSCV-GFP [46] with I-YFP from MSCV-I-YFP [38] . hCas9 [23] was then cloned downstream of MSCV into pCL20c-MSCV-IRES-YFP . The mU6 fragment was generated by PCR using pSicoR-GFP ( Addgene , Cambridge , MA , USA ) and cloned downstream of hU6 in pLKO . 1 ( Addgene , Cambridge , MA , USA ) . The cassette containing the human and mouse U6 promoters ( hU6 and mU6 ) followed by AgeI and EcoNI cloning sites was cloned upstream of the β-actin promoter of the modified pCL20c vector , containing the β-actin-puro cassette from pJ6 . OMEGA . puro [47] . The spacer sequence of hU6 driven sgRNA starts with GG followed by 18 specific nucleotides from the target sequence , and mU6 driven sgRNA starts with GT followed by 18 specific nucleotides from the second target sequence ( Fig . 4C ) . Synthetic ds-DNA fragments , coding Pax3_RH30 sgRNA and Foxo1_RH30 sgRNA were cloned into AgeI and EcoNI sites under control of hU6 and mU6 promoters of pCL20C-hU6-mU6-βact-puro , respectively ( Fig . 4C , D ) . pCL20C-MSCV-Luc2–2A-LgT was constructed by replacing IRES-YFP with a Luc2–2A-LgT cassette . Lentivirus was produced as described in [46] . F12 ( 129SvJ-derived ) embryonic stem ( ES ) cells were electroporated and selected for hygromycin B or G418 resistance using standard procedures . In short , 25–45 μg of linearized BAC DNA was electroporated into 2*107 ES cells followed by selection with 100 μg/ml Hygromycin B or 200 μg/ml G418 . RP24–391O12-LoxP-hygro-TK was linearized with PI-SceI ( NEB ) and RP23–422I13-LoxP-Neo-TK was linearized with NotI ( NEB ) . Drug resistant clones were picked after 7–9 days of selection . DNA from these clones was used for PCR analysis . Screening of homologously recombined ES cell clones was done by PCR and qPCR . The presence of vector arms remaining on either side of the insert was detected by PCR with primers pTARBAC1–3F and pTARBAC1–3R for 3’-located sequences and pTARBAC1–5F and RP24–5R for 5’-located sequences . The “loss-of-native allele” assay was performed as described in [33] with some minor modification . For quantitative ( q ) PCR we used SYBR®Green PCR Master Mix ( Applied Biosystems ) . qPCR was performed with the RP24-F and RP24-R primers to determine the copy number of the centromeric locus and the RP23-F and RP23-R primers to determine the copy number of the telomeric locus . Ratios between the copy numbers of the two loci were determined either by a standard dilution curve or by the Δct method . A double targeted ES cell clone was electroporated with a Cre-expressing plasmid ( pMC-CRE ) using the Amaxa™ Mouse ES Cell Nucleofector™ Kit ( Lonza , Germany ) according to the manufacturer’s protocol . After 5 days of selection with 0 . 2μM of fialuridine ( FIAU ) ( a kind gift of Bristol Myers ) ES cells were collected for DNA isolation as a pool or as single clones . Cre-mediated inversion was detected by standard PCR using the RP24-F/RP23-F2 , and RP24-R/RP23-R2 primer pairs . For FISH analyses of Pax3 and Foxo1 co-localization and detection of t ( 1;3 ) reciprocal translocation we used BAC probes RP23–260F1 and RP24–391O12 . For FISH analyses of targeted ES cells and Foxo1-inv+/+ fibroblasts we used BAC probes RP24–391O12 ( centromeric border of the syntenic region ) and RP23–422I13 ( telomeric border of the syntenic region ) . BAC probes were labeled with nick translation using either Green ( RP23–260F1and RP24–391O12 ) or Red ( RP23–422I13 ) dUTP ( Abbott Molecular ) . Probes were hybridized to metaphase and/or interphase cells either separately or as a 1:1 mixture in hybridization solution ( 50% formamide , 10% dextran sulfate , and 2X SSC ) . Slides were washed in 2X Saline-Sodium Citrate ( SSC ) buffer containing 50% formamide at 37°C for 5 minutes . Cells were counterstained with DAPI and analyzed using a Nikon E80i fluorescence microscope ( Nikon ) with a 100× oil immersion objective . Successfully targeted clones showed 2 native signals for the centromeric or telomeric targeted regions . Inverted chromosomes 3 appeared as two linked pairs of red and green signals on interphase cells , each pair representing one end of the inverted chromosome segment . Normal chromosomes 3 appeared as a single loosely paired red and green signal . One hundred interphase nuclei were scored for the presence of co-localization of Pax3/Pax7 and Foxo1 signals . Only nuclei with discernible red and green signals were scored . Fifty metaphase cells from CRISPR-Cas nuclease treated myoblasts were scored for the presence of cells containing the reciprocal translocation between Pax3 and Foxo1a . Experimental details are provided in S1 Protocol . Position of primers , used for LD-PCR , gel electrophoresis of LD-PCR products and sequences flanking the breakpoint in the Rh30 cell line are shown in S3 and S4 Figs . Cas9 induced translocation was detected by PCR of chromosomal DNA from 104 cells using Pax3-RH30F and Foxo1-RH30R primers . For each qRT-PCR reaction we used RNA isolated from either 2 . 5×103 ( data in Fig . 1 ) or 1 . 6×103 ( data in S1 Fig . ) cells . For each RT-PCR we used RNA isolated from 6 . 7×103 cells . For RT we used SuperScript III First-Strand Synthesis SuperMix ( Invitrogen ) with an equimolar mix of Pax3R primer and random hexonucleotides and performed the reaction following the manufacturer’s protocol . For the qPCR step we used TaqMan Gene Expression Master Mix ( Applied Biosystems ) . Ratios between gene expression in different cell lines were determined by a standard dilution curve . Myoblasts ( 5×106 ) were lysed in 0 . 5 ml CHAPS lysis buffer ( 40 mM HEPES [pH 7 . 4] , 1 mM EDTA 120 mM NaCl , 10 mM sodium pyrophosphate , 10 mM β-glycerophosphate , 0 . 3% CHAPS , 50 mM NaF , 1 . 5 mM NaVO , 1 mM PMSF , and 1 tablet of EDTA-free protease inhibitors [Roche] per 10 mL solution ) and freeze-thawed 3 times , followed by centrifugation at 20 , 000 ×g for 10 min at 4°C . After adding 2 μg anti-Pax3 antibody [48] or anti-Foxo1 ( C29H4 ) Rabbit mAb ( Cell Signaling ) Pax3 , Foxo1 and Pax3-Foxo1 were immunoprecipitated overnight at 4°C . Immunoprecipitated material was bound onto 10 μl protein G-coated Dynabeads ( Invitrogen ) for 90 minutes at 4°C , which were captured using a DYNA-Mag-2 magnet ( Invitrogen ) , washed 4 times with CHAPS buffer , and removed from the beads by heating to 70°C in 1 . 25xLDS loading buffer ( Invitrogen ) in CHAPS and separated on pre-cast 4%–12% bis-tris polyacrylamide gels . Western-blotting was performed using the same anti-Foxo1 and anti-Pax3 antibodies . To enrich for cells harboring the t ( 1;3 ) , 104 of the Cas9/sgRNAs expressing fore limb myoblasts were evenly distributed over three 96-well plates ( on average 30 cells per well ) . DNA from each cell pool was isolated and analyzed for the presence of t ( 1;3 ) translocation using PCR . Libraries were generated from ~ 500 ng total RNA of the 1H3 ( no Pax3-Foxo1 ) and 1G3 ( 64% Pax3-FOXO1 ) cell pools using the Illumina TruSeq Stranded mRNA Sample Preparation Kit . Libraries were sequenced on an Illumina HiSeq 2500 using paired-end 100 bp sequencing chemistry . Paired-end reads from RNA-seq were aligned to the following 4 database files using BWA ( 0 . 5 . 10 ) aligner: ( 1 ) the human GRCh37-lite reference sequence , ( 2 ) RefSeq , ( 3 ) a sequence file representing all possible combinations of non-sequential pairs in RefSeq exons , ( 4 ) AceView database flat file downloaded from UCSC representing transcripts constructed from human ESTs . The mapping results from ( 2 ) to ( 4 ) were aligned to human reference genome coordinates . In addition , they were aligned using STAR 2 . 3 . 0 to the human GRCh37-lite reference sequence without annotations . The final BAM file was constructed by selecting the best alignment among the five map events . We used HTSeq [49] to count the number of fragments that mapped to each gene ( Gencode v 15 ) , where each gene is considered as the union of all its exons . Then we normalized the count to FPKM ( fragments per kilobase of exons per million fragments mapped ) as the expression value of the gene . RNA-seq of both samples produced 55M reads each , with a 20X coverage of 43 . 561% of the exons in 1H3 and 43 . 992% of the exons in 1G3 . 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 Institutional Animal Care and Use Committee ( IACUC ) of St . Jude Children’s Research Hospital ( Animal protocol number 209–100171 ) approved the protocol . | Many cancers carry recurrent chromosome translocations , which often result in the formation of fusion genes that are directly involved in the tumorigenic process . Alveolar rhabdomyosarcoma , a muscle tumor in children , is typified by a translocation that fuses the PAX3 gene on chromosome 2 to the FOXO1 gene on chromosome 13 . For translocation to occur both genes need to break and the disparate ends need to fuse via a process called non-homologous end joining . We determined that physical proximity of Pax3 and Foxo1 in mouse muscle progenitor cells ( myoblasts ) facilitates fusion gene formation . Because Pax3 and Foxo1 in the mouse are in an opposite orientation , we used a chromosome engineering strategy to invert the orientation of Foxo1 so that upon translocation a productive Pax3-Foxo1 fusion gene is created . Co-localization of the Pax3 and Foxo1 loci is higher in fore limb than in hind limb myoblasts . Simultaneous induction of a targeted double strand DNA break in each gene by CRISPR-Cas9 nuclease generated more fusion genes in fore limb than in hind limb myoblasts . Thus , gene proximity facilitates fusion gene formation . We propose that CRISPR-Cas9 nuclease can be used for the precise modeling of chromosome translocations of human cancer in mice . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Modeling of the Human Alveolar Rhabdomyosarcoma Pax3-Foxo1 Chromosome Translocation in Mouse Myoblasts Using CRISPR-Cas9 Nuclease |
We recently demonstrated that the respiratory syncytial virus ( RSV ) NS1 protein , an antagonist of host type I interferon ( IFN-I ) production and signaling , has a suppressive effect on the maturation of human dendritic cells ( DC ) that was only partly dependent on released IFN-I . Here we investigated whether NS1 affects the ability of DC to activate CD8+ and CD4+ T cells . Human DC were infected with RSV deletion mutants lacking the NS1 and/or NS2 genes and assayed for the ability to activate autologous T cells in vitro , which were analyzed by multi-color flow cytometry . Deletion of the NS1 , but not NS2 , protein resulted in three major effects: ( i ) an increased activation and proliferation of CD8+ T cells that express CD103 , a tissue homing integrin that directs CD8+ T cells to mucosal epithelial cells of the respiratory tract and triggers cytolytic activity; ( ii ) an increased activation and proliferation of Th17 cells , which have recently been shown to have anti-viral effects and also indirectly attract neutrophils; and ( iii ) decreased activation of IL-4-producing CD4+ T cells - which are associated with enhanced RSV disease - and reduced proliferation of total CD4+ T cells . Except for total CD4+ T cell proliferation , none of the T cell effects appeared to be due to increased IFN-I signaling . In the infected DC , deletion of the NS1 and NS2 genes strongly up-regulated the expression of cytokines and other molecules involved in DC maturation . This was partly IFN-I-independent , and thus might account for the T cell effects . Taken together , these data demonstrate that the NS1 protein suppresses proliferation and activation of two of the protective cell populations ( CD103+ CD8+ T cells and Th17 cells ) , and promotes proliferation and activation of Th2 cells that can enhance RSV disease .
Following the identification of the NS1 protein of influenza A virus as a type I interferon ( IFN-I ) -antagonist more than a decade ago [1] , such proteins have been identified for many viruses . Their effects on the innate immune system include interference with IFN-I induction , IFN-I-activated signaling , and components of the IFN-I-induced antiviral state ( reviewed in [2] ) . The effects of viral IFN-I antagonists on the adaptive immune response are , however , largely unexplored . Meanwhile , IFN-I is known to have profound effects on key individual components of the adaptive immune system . These effects are diverse and may include both stimulatory and inhibitory roles . For example , several studies have shown that IFN-I promotes maturation of dendritic cells ( DC ) ( reviewed in [3] ) , although one study demonstrated an inhibitory effect on mouse Langerhans cells , which are DC located in the epidermis [4] . The effect of IFN-I on T cells is also quite complex and depends on various factors including the type of T cells ( such as CD8+ , CD4+ Th1 , or CD4+ Th2 ) , their stage of development , the concentration of IFN-I [5] , the experimental system used , and innate inflammatory signals produced during infection , which can be different across various pathogens [6] . For example , while some studies demonstrated that IFN-I stimulates activation and proliferation of T cells [7] , [8] , [9] , [10] , others demonstrated an inhibitory effect [11] , [12] , [13] , [14] . In the case of RSV , IFN-I has been demonstrated to inhibit proliferation of CD4+ T cells in vitro [15] . Thus , the net effect of viral IFN-I antagonist proteins on the adaptive immune response is difficult to predict since it may be the outcome of opposing ( i . e . stimulatory and inhibitory ) effects of IFN-I on various individual components of the adaptive immune system . RSV continues to be the most important viral agent of severe respiratory tract illness in infants and children worldwide and is an important cause of morbidity and mortality in the elderly and severely immunocompromised individuals ( reviewed in [16] , [17] ) . An unusual feature of RSV is its capability to re-infect symptomatically ( albeit with reduced disease ) throughout life without significant antigenic change [18] , [19] , suggesting an ability of the virus to partially suppress or evade the host's adaptive immune response . Several mechanisms contributing to immune evasion by RSV have been identified . One of them is the expression of the secreted form of RSV G protein , one of the two major neutralization antigens of RSV , which prevents efficient antibody mediated neutralization of the virus [20] . Furthermore , the RSV NS1 and NS2 accessory proteins are known to inhibit the production of IFN-I and type III interferon , and to inhibit signaling resulting from the engagement of the IFN-I receptor ( IFNAR ) [21] , [22] , [23] , [24] , [25] , [26] . The two proteins also were demonstrated to inhibit apoptosis [27] . Previously , we analyzed the effect of the NS1 and NS2 proteins on stimulation of human DC by RSV . We reported that the NS1 protein suppresses DC maturation , which was only partly due to suppression of the IFN-I response; furthermore , this effect was somewhat enhanced by the added deletion of the NS2 protein , whereas deletion of NS2 alone had no effect [28] . In the present study , we extend these findings by evaluating the role of the NS1 and NS2 proteins in RSV-specific human T cell responses by co-culturing primary human CD4+ or CD8+ T cells with autologous immature monocyte-derived DC that had been infected with recombinant RSVs lacking the NS1 and/or NS2 proteins . DC derived from primary human monocytes represent an appropriate model for lung-infiltrating DC , since monocytes give rise to mucosal DC [29] , which are highly relevant to the immune response to RSV that occurs in pulmonary mucosa [30] . Moreover , inflammation is a typical feature of acute infections with RSV and other respiratory viruses , and monocyte-derived DC are phenotypically and functionally similar to DC located at sites of inflammation in vivo [31] . Another important advantage of the human DC-T cell co-cultivation system is that it avoids limitations of the commonly used mouse model for RSV , which is based on a non-natural host that is only semi-permissive for RSV infection . Mice also have some incongruities in their immune system compared to humans; for example , IFN-I activates human and mouse T cells by different mechanisms [32] , [33] . In the present study , we identified three major effects of the NS1 protein on T cells that likely play a role in reducing the efficiency of the adaptive immune response to RSV and may contribute to disease severity . We also investigated the role of IFN-I in regulating these effects .
We investigated the effects of the RSV NS1 and NS2 proteins on the ability of human monocyte-derived DC to activate autologous T cells during co-culture in vitro ( Fig . 1 ) . DC were infected with 2 plaque forming units ( PFU ) per cell of wt RSV or RSV that lacks the NS1 , NS2 , or both genes and expresses enhanced green fluorescent protein ( GFP ) ( ΔNS1 , ΔNS2 , ΔNS1/2 [28] ) . The DC were washed and co-cultured with purified autologous CD8+ or CD4+ T cells and incubated for 7 days ( Fig . 1A ) . The T cells were then immunostained for surface markers and intracellular cytokines and analyzed by multi-color flow cytometry ( Fig . 1B ) . Flow cytometry data were analyzed for ( i ) proliferated CD8+ or CD4+ T cells measured by dilution of carboxyfluorescein diacetate succinimidyl ester ( CFSE ) intensity due to cell division , ( ii ) proliferated CD8+ or CD4+ T cells positive for each individual marker , and ( iii ) proliferated T cell populations positive for one or more markers and negative for the other markers , resulting in a total of 32 combinations ( Fig . 1B ) . Since essentially all adults have been exposed to RSV due to prior natural infections , the proliferating T cells would primarily represent activation of memory cells . The possible presence of trace numbers of RSV-specific naïve T cells would not affect the study . We note that viruses lacking the NS1 and/or NS2 proteins are similar to wt virus with regard to their efficiency of infection of human DCs and have only a modest reduction in viral gene expression [28] . Thus , changes in immune cells associated with the deletions ( below ) can be attributed to the absence of the NS1/2 genes rather than substantial changes in viral growth or gene expression . Co-cultivation of wt RSV-infected DC with CD8+ T cells resulted in significant proliferation of T cells , which was negligible with mock-infected DC ( Fig . 2 ) . As compared to wt RSV , deletion of the NS1 gene resulted in an increased proliferation of CD8+ T cells using cells from 6 donors and reduced proliferation with cells from 4 donors . Deletion of the NS2 gene resulted in an increased proliferation of cells from only 2 of 10 donors ( Fig . 2 ) . On average , deletion of the NS1 , NS2 , and NS1 and NS2 together resulted in an increase in mean CD8+ T cell proliferation of 20 , 13 and 17% , respectively , as compared to wt RSV , and none of these differences were significant . Thus , the effect of NS1 and/or NS2 deletion on total CD8+ T cell activation was modest and inconsistent . We then evaluated the effects of the NS1 and NS2 deletions on activation of specific subsets of CD8+ T cells by multicolor flow cytometry . The analysis involved staining for two T cell surface proteins: CD103 ( the αE subunit of the αEβ7 integrin ) , which is preferentially expressed by CD8+ T lymphocytes present in mucosal tissue , directs lymphocytes to mucosal tissues by binding to the epithelial cell marker E-cadherin [34] , [35] , [36] , and triggers cytolytic activity of CD8+ T cells in lung tissue [37] , and CD107a , which is a marker of de-granulation of CD8+ CTL [38] . We also analyzed intracellular staining of the cytokines IFNγ , IL-2 , and TNFα , which are considered markers of CD8+ CTL activation [39] . We found that deletion of the NS1 gene alone or in combination with the NS2 gene resulted in a dramatic increase in the number of proliferating CD8+ cells that express CD103 ( Fig . 3A , panel I ) . In addition to this increase in the number of CD103+ cells , deletion of NS1 and NS1/2 was also accompanied by an increase in the median fluorescence intensity ( MFI ) of CD103 expression of up to 32% as compared to wt RSV , an increase that was statistically significant in most , but not all cases ( Fig . S1A ) . In addition , deletion of NS1 or NS1/2 resulted in a mean increase of 30% and 58% , respectively , in total CD8+ T cells positive for CD107a , although in case of the NS1 deletion , the increase was not statistically significant ( Fig . 3A , panel II ) . The presence of this degranulation marker presumably is a consequence of cytotoxic activity of CD8+ T cells against the DC in the co-culture , and thus is a marker for activation . Incidentally , cytotoxic attack on DC is not necessarily an aberrant activity , and indeed has been suggested to occur in vivo as a means of down-regulating further T cell activation [40] . It also should be noted that CD107a transiently appears on cell surface and is rapidly internalized following degranulation [41] , and hence the level of CD107a observed on day 7 probably is a substantial underestimate of the total amount of degranulation occurring during the 7 day incubation . Deletion of NS1 or NS1/2 also resulted in substantial increases in CD8+ cells that were positive for CD103 plus IFNγ ( Fig . 3A , panel III and 3B ) and CD103 plus both IFNγ and IL-2 ( Fig . 3A , panel IV ) . In contrast , deletion of the NS2 gene alone did not alter the expression of these various markers . Incidentally , deletion of NS1 , but not NS2 , also was associated with a down-regulation of the number of total CD8+ T cells secreting TNFα in most of the donors ( Fig . 3A , panel V ) . As a possible explanation of this effect , deletion of NS1 is associated with increased secretion of IFN-I by RSV-infected DC [28] , and IFN-I has been shown to suppress the production of TNFα [42] . Data are conflicting as to whether TNFα plays a role in CD8+ CTL-mediated lysis of target cells [43] , [44] . Perhaps more importantly , TNFα down-regulates memory CD8+ T cell responses by limiting the duration of the CTL effector phase and magnitude of CD8 T cell memory response [45] . Taken together these data suggest that the RSV NS1 protein reduces the number of CD8+ T cells positive for the mucosal tissue homing molecule CD103 , reduces activation of these cells , as measured by expression of the de-granulation marker CD107a and IFNγ , and increases the number of CD8+ T cells positive for TNFα . We also evaluated the effects of deleting the NS1 and NS2 proteins on the activation of CD4+ T cells using the co-culture system . First , we compared the levels of proliferation of total CD4+ T cells following co-culture with autologous DC pre-infected with wt RSV or the NS1 and/or NS2 deletion mutants using the same experimental design as that used for CD8+ T cells ( Fig . 1 ) . Abundant proliferation of RSV-specific CD4+ T cells was detected by CFSE dilution assay when the cells were co-cultivated with DC pre-infected with wt RSV , whereas there was a near-absence of proliferation in response to mock-infected DC ( Fig . 4 ) . Deletion of the NS1 gene resulted in a significant reduction in the proliferation of CD4+ T cells ( Fig . 4A , B ) : the effect was observed in all samples and mean reduction associated with deletion of the NS1 and NS1/2 genes was 26% and 22% , respectively . In contrast , deletion of the NS2 gene alone was associated with no change or only a marginal reduction of proliferation ( Fig . 4A ) . Thus , the NS1 protein , but not NS2 , promotes CD4+ T cell proliferation . We also investigated the effects of the NS1 and NS2 proteins on the cytokine profiles of proliferating CD4+ T cells from the co-culture experiments described above by intracellular cytokine staining of T cells and multi-color flow cytometry . The analyzed cytokines included markers for the Th1 ( IFNγ ) , Th2 ( IL-4 ) , and Th17 ( IL-17 ) subsets , as well as general markers of activation ( IL-2 and TNFα ) . Unexpectedly , we found a sizable population of CD4+ T cells positive for both IFNγ and IL-4 , which made up 12 . 6% of total CD4+ T cells proliferating in response to stimulation with wt RSV-infected DC and also accounted for most of the IL-4+ population ( Fig . 5A , B ) . CD4+ T cells positive for both IFNγ and IL-4 were previously identified in mice and have been cloned and characterized in vitro [46] . A recent study demonstrated that infection of mice with lymphocytic choriomeningitis virus reprogrammed otherwise stably committed Th2 cells to adopt an IL-4+IFNγ+ “Th2+1” phenotype that was maintained in vivo for months [47] . However , to the best of our knowledge a similar human Th cell population has not yet been reported [48] . In the present study , control experiments excluded the possibility that the Th1/Th2 phenotype of CD4+ T cells was the result of fluorochrome spectral overlap: specifically , the IL-4+ IFNγ+ double positive cells were observed only with antibodies to both IFNγ and IL-4 and were not observed when the cells were stained with all antibodies while omitting those for IFNγ or IL-4 ( Materials and Methods and Fig . S2B ) . Deletion of the NS1 gene alone or with the NS2 gene , but not deletion of NS2 alone , resulted in a significant reduction in the number of cells secreting IL-4 , including total IL-4+ cells and those positive for both IL-4 and IFNγ ( Fig . 5A , B ) in most of the donors . Furthermore , analysis of IL-4 expression showed that deletion of NS1 , but not NS2 , also resulted in reduced IL-4 MFI for total IL-4+ as well as IL-4+IFNγ+ cells ( Fig . S1B ) . No change was observed in the total number of IFNγ or IL-2 secreting cells ( not shown ) . These results suggest that the NS1 protein acts as a factor promoting the Th2 response , acting to both shift the bias toward Th2 and to promote proliferation . The reduced percentages of not only IL-4+ but also IL-4+IFNγ+ cells are consistent with our previous studies demonstrating that the induction of a Th2 response during RSV infection in mice results in an increased production of not only the Th2 cytokines , but also of the Th1 cytokine IFNγ [49] , as well as the observation by others that the increased production of Th2 cytokines in atopic children is not associated with a reduced level of IFNγ [50] . Deletion of the NS1 gene or both the NS1 and NS2 genes , but not deletion of NS2 alone , also resulted in elevated levels of total IL-17+ cells , as compared to the levels associated with wt RSV , by 125% and 33% , respectively ( Fig . 5C , D ) . Interestingly , we detected a substantial population of the recently described human Th17/Th1 cells positive for both IL-17 and IFNγ [51] , and confirmed their double positive phenotype by staining with all antibodies while omitting those for IFNγ or IL-17 ( Fig . S2C ) . The percentage of this cell population was also increased due to deletion of NS1 or both the NS1 and NS2 genes , and the increases were even more dramatic than for total IL-17+ cells , 233% and 64% , respectively . These data indicate that the NS1 protein also antagonizes or alters the adaptive immune response by suppressing the number of Th17 cells , a subset that has recently been shown to be involved in clearance of viral infections in vivo ( Discussion ) . Unlike CD8+ T cells , CD4+ T cells demonstrated no change in the percentage of TNFα+ cells on deletion of NS1 or NS2 genes ( not shown ) . As noted ( Introduction ) , we previously showed that the NS1 protein suppresses DC maturation , due in part to antagonism of IFN-I production and signaling [28] . IFN-I also affects T cell activation , polarization , and proliferation ( Introduction , Discussion ) . Therefore , the suppression of CD103+CD8+ T cells and Th17 cells and enhancement of Th2 cells by NS1 might be mediated by either or both of the two following mechanisms . First , they might result from reduced production of IFN-I by DC , resulting in reduced IFN-I signaling in T cells . Second , they might result from a reduced level of DC maturation , thereby reducing the ability of the DC to activate T cells . As a first step in evaluating these possibilities , we investigated the role of IFN-I production and signaling in co-culture experiments in which IFN-I signaling was blocked with an antibody specific to the IFNAR beta subunit ( IFNAR2 ) . To determine the effectiveness of the IFNAR2-specific antibody in inhibiting IFN-I signaling , CD8+ ( Fig . 6A ) and CD4+ ( Fig . 6B ) T cells were incubated for 2 h with a range of antibody concentrations from 0 . 3 ( Fig . 6A ) or 0 . 03 ( Fig . 6B ) to 30 µg/ml and then challenged with 100 IU/ml of IFNα2a . The T cells were harvested at 3 , 6 , and 18 h post-IFN-challenge and the expression of the IFN-I-inducible Mx1 ( Fig . 6A and B ) and ISG-56 ( not shown ) genes were analyzed by quantitative RT-PCR ( QRT-PCR ) . This showed that the antibody effectively suppressed induction of either gene even at a concentration that was one-hundredth ( Fig . 6A ) or one-thousandth ( Fig . 6B ) of that used in subsequent blockade experiments . We then investigated the role of IFN-I in the effects on CD8+ T cells in co-cultures . In preliminary experiments , we treated the DC or T cells separately with IFNAR2-blocking antibody for various times , washed away free antibody , and performed co-culture . However , we found that we were unable to detect effects of the antibody blockade on T cell activation ( not shown ) . This perhaps was not surprising , since new surface IFNAR2 molecules likely appear during the long co-culture due to receptor recycling [52] and proliferation . Therefore , we modified the protocol so that the IFNAR2-blocking antibody would be present throughout the co-culture incubation . One unavoidable limitation of this experimental design is that it would not allow us to distinguish directly between effects on DC versus T cells . For these experiments , the IFNAR2-blocking antibody , or an isotype control antibody , was added at a 2-fold concentration ( 60 µg/ml ) to CD8+ T cells , and the cells were incubated for 1 h at 37°C , mixed with an equal volume of autologous DC that had been pre-infected for 4 h with wt RSV or ΔNS1/2 RSV or mock infected , and co-cultured with the antibody present . In one of the four donors , the blockade resulted in a substantial increase in the proliferation of CD8+ T cells compared to the isotype control antibody , and in the other donors , there was either no effect or a moderate reduction of proliferation ( Fig . 6C ) . Thus , a consistent effect was not observed , and usually there was little effect . We next analyzed the effect of IFNAR2 blockade on the proliferation of the CD103+CD8+ T cells , the population most significantly up-regulated by deletion of the NS1 protein , as previously shown in Fig . 3 . Somewhat surprisingly , this showed that the blockade increased the percentage of CD103+ cells , an effect that occurred with either virus ( Fig . 6D ) . This is the opposite of the effect that would have been expected if the increase in CD103+CD8+ T cells observed in Fig . 3 for the NS1 and NS1/2 deletion viruses had been due to increased IFN-I production and signaling . Next , we evaluated the effects of exogenously added IFN-I on CD8+ T cell activation and proliferation . Published studies have variously demonstrated stimulatory or anti-proliferative effects of IFN-I on CD8+ T cells ( Introduction ) . We added increasing amounts of an equal mixture of IFNα2a and IFNβ at the beginning of co-cultivation of wt RSV-infected DC with CD8+ T cells . This resulted in inconsistent effects between the two donors , with small , inconsistent increases or decreases in proliferation at various doses . We next analyzed the CD103+CD8+ T cell population and found that exogenous IFN-I reduced , rather than increased , its percentage , although in cells from one donor this effect was observed only at lower concentrations of IFN-I ( Fig . 6F ) . This is consistent with the antibody blockade results described above in Fig . 6D . An inhibitory , rather than stimulatory , effect of exogenous IFN-I on total CD103+ cells also was observed for most of the donors in the experiments shown in Fig . 3A panel I , in which a single dose of 133 IU/ml each of both IFNα2a and IFNβ was added ( similar to the concentration produced by DC infected with the ΔNS1 virus [28] ) , with the inhibition being statistically significant ( P<0 . 05 ) . The finding that IFN-I and IFN-I signaling suppress rather than stimulate the proliferation of CD103+CD8+ T cells indicates that the effects of NS1 deletion involve a mechanism other than increased IFN-I production and signaling . Next , we investigated the role of IFN-I in the effects on CD4+ T cells in the co-culture system . We added IFNAR2-blocking antibodies to CD4+ T cells 1 h prior to their co-cultivation with DC that had been pre-infected with wt RSV or ΔNS1/2 RSV , as described above for CD8+ T cells . With either virus , the blockade of IFN-I signaling during co-culture resulted in an increased proliferation of CD4+ T cells ( Fig . 6G ) . However , the subpopulations that were positive for IL-4 or IL-17 did not consistently change with the IFNAR2 blockade ( Fig . 6 H and I , respectively ) . We note that , if increased IFN-I production and signaling indeed had been responsible for the Th2 and Th17 effects observed in Fig . 5 , we would have expected to observe an increase in IL-4+ T cells in Fig . 6H and a decrease in IL-17+ T cells in Fig . 6I , which were not observed . We then examined the effect of exogenously added IFN-I on CD4+ T cell activation . Increasing amounts of a mixture of equal quantities of IFNα2a and IFNβ were added at the beginning of co-culture of wt RSV-infected DC and CD4+ T cells , as was done earlier with CD8+ T cells ( above ) . Addition of the IFN-I cocktail resulted in a dose-dependent reduction in the proliferation of CD4+ T cells ( Fig . 6J ) , which is consistent with our previously published study [15] . However , analysis of individual populations of CD4+ T cells demonstrated a lack of consistent effect on IL-4+ cells ( Fig . 6K ) , and a modest decrease in IL-17+ cells that was not dose-dependent ( Fig . 6L ) . We note that , if increased IFN-I production and signaling had been responsible for the Th2 and Th17 effects observed in Fig . 5 , we would have expected to observe a decrease in IL-4+ T cells in Fig . 6K and an increase in IL-17+ T cells in Fig . 6L , which were not observed . In summary , IFN-I production and signaling in the co-cultures had effects on proliferation of the total population of CD4+ T cells . However , the three major effects associated with the presence of the RSV NS1 protein , namely the reduced proliferation and activation of CD103+CD8+ T cells , the reduced activation and proliferation of Th17 cells , and the skew in the Th1/Th2 balance towards Th2 , do not appear to arise from direct effects of IFN-I signaling during co-culture . We also investigated the effects of IFN production and signaling on DC . We treated DC from four donors with the IFNAR2 blocking or isotype control antibody for 2 h and then infected them with wt RSV or ΔNS1/2 RSV followed by 48 h incubation in the presence of the blocking antibody . The cells were harvested , total RNA was purified , and mRNAs encoding 49 proteins relevant to DC maturation and function were analyzed by QRT-PCR using a microfluidic gene card format . The results are displayed as a heat map ( Fig . 7A ) and in some cases in linear plots ( Fig . 7B ) . We found that deletion of the NS1 and NS2 genes resulted in a dramatic transcriptional up-regulation in a number of mRNAs including ones encoding type I and type III IFNs and other cytokines , molecules involved in antigen presentation and signaling , co-stimulatory factors , and pattern recognition receptors . The IFNAR2-blocking antibody strongly reduced the expression of the IFN-I-stimulated genes CXCL9 , CXCL10 , and RIG-I ( Fig . 6A , B ) , suggesting that the blockade of IFN-I signaling was effective . The strong IFNAR2 blockade did not reduce the expression of IFNα mRNA ( Fig . 6A ) ; also , ELISA analysis of secreted IFNα in the medium demonstrated a 14-fold greater mean concentration for ΔNS1/2 RSV compared to wt RSV ( data not shown ) . This observed expression of IFNα in the face of the IFNAR2 blockade presumably arises by a mechanism that is not dependent on signaling from the IFNAR , such as through activation of IRF7 or IRF5 that are constitutively expressed in human monocyte-derived DC [53] . We found that the IFNAR2 blockade resulted in a partial transcriptional down-regulation of DC maturation markers including CD38 , CD40 and CD80 ( Fig . 6A , B ) , in a dose-dependant manner ( data not shown ) . This is consistent with our previous study , which showed that only part of the maturation of DC in response to wt RSV or the ΔNS mutants appeared to be IFN-I-dependent [28] . These results suggest that NS1/2 suppress multiple factors of DC maturation and that this suppression is only partially regulated by secreted IFN-I . For example , we note that the levels of IL-12β and IL-23α mRNA which enhance Th1 and Th17 responses , respectively , were increased 57-759-fold and 49-134-fold , respectively , in DC infected with the ΔNS1/2 virus compared to wt RSV ( Fig . 7A , B ) . The level of expression of each was increased rather than decreased by the IFNAR2 blockade: thus , the expression of IL-12β and IL-23α was suppressed by NS1 , but not due to suppressed IFN-I production and signaling . Analysis of cytokine expression by flow cytometry reliably determines the percentage of cells that are positive for expression , but since in vitro stimulation can induce high levels of expression , the MFI does not necessarily provide an accurate comparison of the level of expression . We therefore also analyzed selected cytokines by ELISA . We performed co-culture of wt and mutant RSV-infected DC with autologous CD8+ or CD4+ T cells , and after 7 days of incubation , analyzed the co-culture media supernatants by ELISA for IFNγ and TNFα ( CD8+ T cell co-cultures ) , or IFNγ , IL-2 , IL-4 , IL-12 and IL-17 ( CD4+ T cell co-cultures ) . We found that for both CD8+ and CD4+ T cells from most of the donors , the deletion of the NS1 gene was associated with increases in the concentration of secreted IFNγ ( mean increases of 70% and 31% , respectively ) , whereas deletion of both the NS1 and NS2 genes resulted in somewhat lesser increases ( Fig . 8 ) . The other cytokines analyzed by ELISA were either not detected or were detected at very low levels ( not shown ) , reflecting a limitation of this method . The addition of exogenous IFN-I at various concentrations ( data not shown ) or at a single dose ( 133 IU/ml of IFNα2a and IFNβ each , Fig . 8 ) during the co-cultures of wt RSV-infected DC with T cells did not significantly affect IFNγ production ( Fig . 8 and data not shown ) . Thus , the presence of the NS1 gene changed the cytokine microenvironment towards reduced secretion of IFNγ , and this effect was not altered by added IFN-I .
We demonstrated , using primary human immune cells , that the RSV NS1 protein induces quantitative and qualitative deficiencies in adaptive immunity . In particular , the NS1 protein ( i ) suppresses the CD103+ CD8+ T cell response , ( ii ) promotes a Th2 response , and ( iii ) suppresses the Th17 response ( Fig . 9 ) . Thus , NS1 suppresses two mechanisms that play a protective role during RSV infection ( CD103+ CD8+ T cells and Th17 cells ) and stimulates a mechanism contributing to enhanced disease ( Th2 cells ) . A role for CD8+ T cells in restricting and clearing RSV infection is well established [54] , [55] . Following infection with RSV or influenza virus , virus-specific CD8+ CTL accumulate in lungs at much greater concentration than in the peripheral blood [30] . Mucosal CD8+CTL , but not their peripheral blood counterparts with the same antigenic specificity , are highly positive for CD103 and have the effector memory phenotype [56] . The presence of CD103 on the surface of lung CD8+ CTL is important for respiratory epithelial cell-specific tropism and cytotoxicity due to expression of E-cadherin by epithelial cells , which specifically binds CD103 [34] , [57] . A study with human lung tumor tissues and tumor-specific CD8+ CTL demonstrated that the interaction of CD103 with E-cadherin promotes cytolytic activity by triggering lytic granule polarization and exocytosis [34] . Similarly , it was found that , in human bronchoalveolar lavages , a much greater fraction of both CD4+ and CD8+ T cells expresses CD103 , as compared to that in the peripheral blood , and that the CD103 molecule is required for effective lysis of E-cadherin-expressing target cells [35] , [57] . Moreover , CD103 plays role in CD8+ T cell retention in human lung tumors by a CCR5-dependent mechanism [37] . Thus , CD103 is a molecule expressed at high levels by mucosal and , in particular , pulmonary CTL , and is associated with their mucosal homing , cytotoxicity against respiratory tract epithelial cells , and their subsequent retention as effector memory T cells . Accumulating evidence thus indicates an important role for CD103 in pulmonary CTL-mediated lysis of virus-infected cells of the respiratory epithelium and thus protection against respiratory viruses . Therefore , the suppression of activation and proliferation of CD103+CD8+ CTL by the RSV NS1 protein demonstrated in this study identifies a new mechanism by which the virus suppresses an important protective component of the adaptive immune response . Th17 cells were initially found to have a pro-inflammatory effect and to be involved in autoimmune diseases [58] . Subsequent studies elucidated their role in innate and adaptive immune response against pathogens . One of the major functions of IL-17 is the recruitment of neutrophils [59] . Infection of bronchial epithelial cells with rhinovirus , a human respiratory pathogen , demonstrated that the IL-17-mediated recruitment of neutrophils is related to its ability to induce IL-8 , which is a neutrophil chemoattractant [60] . Neutrophils play an essential role in antibody-mediated neutralization of influenza virus in vivo [61] and play a critical role in the activation of natural killer cells [62] . Using Listeria monocytogenes , it was found that Th17 cells are much more strongly induced by the respiratory route of infection than by the intravenous route [63] . It has only recently been found that Th17 cells play a role in protection against viruses . IL-17 produced by Th17 cells facilitates clearance of vaccinia virus infection in mice , most likely by the attraction of neutrophils [64] . Th17 cells have also been shown to protect mice from a lethal dose of influenza virus by a mechanism independent of IFNγ , T cell helper function , or perforin-mediated cytotoxicity , and which possibly involves reduction in the severity of lung damage and/or effect on the rate of lung repair [65] . Th17/Th1 cells exhibit functional properties similar to those of Th17 cells [51] . Thus , the suppression of proliferation of IL-17+ CD4+T cells ( Th17 cells ) or IL-17+IFNγ+ CD4+ T cells ( Th17/Th1 cells ) by the RSV NS1 protein found in this study suggests yet another novel mechanism the virus uses to counteract the anti-viral effects of both the adaptive ( the Th17 and Th17/Th1 cells ) and the innate ( secreted IFNγ ) components of the immune system . Severe RSV disease likely is influenced by a variety of factors whose individual importance varies in the heterogeneous human population and likely also with age . One proposed scenario involves a skewing of the Th1/Th2 balance of the virus-specific response towards Th2 , which includes down-regulation of the CD8+ CTL response and a B cell switch from synthesis of virus-specific protective IgA and IgG to non-protective IgE ( reviewed in [16] , [66] ) . A Th2 bias also has been implicated for a formalin-inactivated RSV vaccine that was evaluated in infants and children in the 1960s and resulted in a greatly increased frequency and severity of disease upon subsequent natural RSV infection [16] , [66] . Moreover , a number of studies over the past decade have suggested that severe RSV disease may sometimes be linked to various genetic factors , including those controlling the expression of or response to Th2 cytokines [67] . In particular , polymorphisms leading to an increased activity of IL-4 and IL-13 appeared to be overrepresented in infants and children with severe RSV disease [68] , [69] , [70] , [71] . In the present study , NS1 expressed by wt RSV not only increased the percentage of CD4+ T cells positive for IL-4 ( Fig . 5A , B ) , but also enhanced proliferation of these cells by antagonizing the anti-proliferative effect of IFN-I ( Fig . 4 ) . NS1 also depressed the concentration of IFNγ in DC-T cell co-cultures ( Fig . 8 ) , thus favoring the Th2 response . These data suggest that the NS1 protein contributes to skewing of the Th1/Th2 balance towards Th2 during the priming of naive T cells or stimulation of memory T cells . The idea that NS1 can skew the T cell response towards Th2 was indicated in previous work . Specifically , when RSV-infected human DC were treated with small interfering RNA targeting the NS1 gene and co-cultivated with heterologous RSV-naïve cord blood CD4+ T cells , the percentage of IFNγ+ cells increased , and that of IL-4+ cells decreased [72] . However , what is surprising in the present study is that these effects do not appear to be dependent on IFN-I production and signaling , as described below . Since the RSV NS genes are IFN-I antagonists , the most obvious explanation for the effects of NS1 deletion observed in the present study would be that they are mediated by increased IFN-I production and signaling , which are known to affect DC maturation and T cell activation , polarization and proliferation . For example , as noted , NS1 suppresses DC maturation , due in part to inhibition of IFN-I production and signaling [28] . IFN-I has been reported to have stimulatory or inhibitory effects on CD8+ T cell proliferation , depending on the situation ( Introduction ) . Type I and type III IFN have been reported to suppress activation and proliferation of CD4+ T cells during co-culture with RSV-infected DC [15] . IFN-I has been shown to promote a Th1 response by increasing the frequency of human IFNγ-secreting CD4+ Th cells and antagonizing the suppressive effect of IL-4 on IFNγ production [73] , [74] . In the present study , one effect indeed did appear to be mediated directly by changes in IFN-I production and IFN-I signaling: specifically , proliferation of total CD4+ T cells was inhibited by IFN-I . However , the other observed effects , namely the suppression of CD103+CD8+ T cells and Th17 cells and enhancement of Th2 cells by NS1 , did not appear to be directly mediated by changes in IFN-I production and signaling ( Fig . 9 ) . This was suggested by several observations . For example , while both NS1 and NS2 , individually and in combination , antagonize IFN-I production and signaling in epithelial cells ( Introduction ) , the effects in the present study were seen only in response to deletion of NS1 and not the NS2 protein . The inclusion of IFNAR2-blocking antibody in the co-cultures provided evidence that the NS1-specific effects on CD103+CD8+ cells , Th17 cells , and Th2 cells were not due to changes in IFN-I production or signaling . This also was indicated in the experiments in which exogenous IFN-I was added to the co-cultures . An alternative possibility is that these NS1-mediated effects involve suppression of pathways that do not depend on IFN-I production and signaling . As noted , maturation of DCs in response to RSV is partly independent of IFN-I production and signaling [28] . Similarly , Lopez et al . previously demonstrated that the maturation of DC in response to negative strand RNA viruses involves intracellular IFN-I induction pathways , including activation of nuclear factor-kB , but not the released IFN-I or subsequent IFN-I signaling [75] . In the present study , infection of DC with the ΔNS1/2 mutant induced a strong up-regulation in the transcription of multiple genes involved in DC maturation and T cell activation ( Fig . 7A , B ) . Much of this up-regulation was suppressed by wt RSV . The role of IFN-I in this up-regulation was assessed with IFNAR2-blocking antibody . The blockade strongly reduced expression of CXCL9 , CXCL10 and RIG-I , which are known to be induced mainly by IFN-I [76] , [77] , [78] . In contrast , the blockade only partly reduced the up-regulation of several maturation-related markers including CD38 , CD40 , and CD80 among others . Thus , we suggest that the effects on the T cells observed in the present study result from NS1-mediated suppression of signaling pathways leading to DC maturation , which likely include signaling pathways leading to induction of IFN-I , but are independent of secreted IFN-I and IFN-I signaling . IL-12 and IL-23 also may be involved . IL-12 and IL-23 are involved in polarization of T cell response toward Th1 [79] and maintenance of the Th17 cell population [80] , respectively . The expression of each by RSV-infected DC was increased by deletion of NS1/2 . This did not appear to be a result of increased IFN-I production and signaling because the IFNAR2 blockade increased rather than decreased expression , indicative of suppression rather than stimulation by IFN-I ( Fig . 7A , B ) . Suppression of IL-12 expression by IFN-I also has previously been reported in humans and mice [81] , [82] . NS1 also reduced secretion of IFNγ by both CD8+ and CD4+ T cells . IFNγ promotes the differentiation of Th cells to Th1 and suppresses their development to Th2 phenotype , and IFNγ secreted by CD8+ T cells prevents Th2-driven pathology during RSV infections [83] , [84] . Thus , suppression of IFNγ secretion by CD4+ T cells , which was not reversed by exogenous IFN-I , may be involved in the Th2-promoting effect of the NS1 protein demonstrated in this study . In conclusion , the data presented in this study suggest that the RSV NS1 protein has quantitative and qualitative effects on the adaptive immune response , thus promoting infection and disease . The experimental system used in this study involved human primary cells from adult donors and thus represents a model of the immune system of RSV-immune children and adults for investigation of the effects of immunomodulating RSV proteins on the virus-specific memory T cells . However , the pattern of immune response to RSV depends on age and is biased towards Th2 in neonates [85] . Therefore , it is possible that , in RSV-naïve individuals , the three major effects of the NS1 protein , especially that of enhanced Th2 activity , might be even more pronounced than in adults .
Virus growth and purification were performed as described previously [28] , [86] . Briefly , Vero cells were maintained in Opti-MEM I mefdium ( Invitrogen , Carlsbad , CA ) supplemented with 5% heat-inactivated fetal bovine serum ( FBS ) ( HyClone , Logan , UT ) , 2 mM L-glutamine , 100 IU/ml penicillin and 100 µg/ml streptomycin sulfate ( Invitrogen ) . Recombinant wild type ( wt ) RSV strain A2 and RSV A2 deletion mutants lacking NS1 ( ΔNS1 ) , NS2 ( ΔNS2 ) or both NS1 and NS2 ( ΔNS1/2 ) genes , all expressing enhanced green fluorescent protein to monitor DC infection , described previously [28] , were propagated in Vero cells . Virus-infected Vero cell supernatants were harvested and viruses were purified by sucrose gradient centrifugation . In order to eliminate sucrose from purified viruses , virus bands were diluted with Advanced RPMI 1640 medium with 2 mM L-glutamine ( Invitrogen ) , and the viruses were pelleted at 8 , 000 x g for 2 h at 4°C , followed by re-suspension in the same medium . Aliquots of the viruses were stored at −80°C and titers were determined by plaque assay as previously described [87] . Elutriated monocytes obtained from healthy adult donors according to an approved clinical protocol were provided by the NIH blood bank and were differentiated into monocyte-derived DC cultures as described previously [28] . Briefly , CD14+ monocytes were purified by positive selection using anti-CD14 monoclonal antibody-conjugated magnetic microbeads and an Automacs separator ( Miltenyi Biotech , Auburn , CA ) , and were cultured with granulocyte-macrophage colony-stimulating factor ( GM-CSF ) and IL-4 . Cultures were incubated at 37°C for 7 days during which time the cells developed phenotypic features of immature DC ( CD1a+ CD14low CD38low CD11chigh ) [88] . Autologous elutriated lymphocytes from the same donors were also provided by the NIH blood bank and were purified by density centrifugation in Ficoll-Hypaque ( Cellgro-Mediatech , Manassas , VA ) , harvested and washed in Advanced RPMI 1640 medium ( Invitrogen ) supplemented with 10% FBS ( HyClone ) and 2 mM L-glutamine ( Invitrogen ) . Lymphocytes were cryopreserved in Advanced RPMI 1640 medium containing 10% FBS ( HyClone ) , 2 mM L-glutamine ( Invitrogen ) , 10% dimethyl sulfoxide ( DMSO ) ( Cellgro-Mediatech ) at a cell density of 4×107 cell/ml . Before use , the lymphocytes were thawed and cultured overnight in Advanced RPMI 1640 medium supplemented with 10% human serum ( Gemini Bio-products , West Sacramento , CA ) , 2 mM L-glutamine , 200 IU/ml penicillin , and 200 µg/ml streptomycin sulfate ( Invitrogen ) . The next day , CD8+ T cells were purified from lymphocytes by negative selection with an Automacs separator using a primary cocktail of antibodies conjugated to biotin and secondary anti-biotin antibody conjugated to magnetic microbeads to deplete non-CD8+ T cells , including CD4+ T cells , γ/δ T cells , B cells , NK cells , DC , monocytes , granulocytes , and erythroid cells . The cells were analyzed for purity by staining with anti-CD8 phycoerythrin ( PE ) -conjugated antibodies ( BD Biosciences , San Jose , CA ) and flow cytometry . CD4+ T cells were isolated from lymphocytes by positive selection with an Automacs using anti-CD4 monoclonal antibody-coated magnetic microbeads ( Fig . 1A ) . The purity of the cells was assessed by staining with anti-CD4 fluorescein isothiocyanate ( FITC ) -conjugated antibodies ( BD Biosciences ) and flow cytometry . The purity of CD8+ and CD4+ T cells was 85–95% and >98% , respectively . Purified CD8+ and CD4+ T cells were labeled with CFSE ( Invitrogen ) as per the manufacturer's instructions , in order to monitor cell proliferation as assessed by reduction in the CFSE fluorescence intensity due to cell divisions . On day 7 of culture , when monocytes had differentiated into DC , they were harvested and labeled either with CellVue Lavender ( Molecular Technologies , West Chester , PA ) or 7-hydroxy-9H- ( 1 , 3-dichloro-9 , 9-dimethylacridin-2-one ) ( DDAO ) ( Invitrogen ) fluorescent tracers , as per manufacturer's instructions , for co-culture with CD8+ and CD4+ T cells , respectively ( Fig . 1A ) . Following labeling , DC were inoculated with wt RSV , ΔNS1 , ΔNS2 or ΔNS1/2 RSV at MOI of 2 PFU/cell , mock treated or treated with Staphylococcal Enterotoxin B ( SEB ) ( Sigma-Aldrich , St . Louis , MO ) at 1 µg/ml . After 4 h of incubation , DC were washed to remove virus inoculum and were co-cultured with purified autologous CFSE-labeled CD8+ or CD4+ T cells at a DC: lymphocyte ratio of 1∶10 ( 2×105 DC: 2×106 lymphocytes ) in 2 ml of Advanced RPMI 1640 medium ( Invitrogen ) containing 10% human serum ( Gemini Bio-Products ) , 2 mM L-glutamine , 200 IU/ml penicillin , and 200 µg/ml streptomycin sulfate ( all Invitrogen ) in 12-well plates . Co-cultures were incubated at 37°C for 7 days . This is based on our previous finding that proliferation was minimal by day 4 and requires 7 days to be substantial [89] . On day 5 , 1 ml of fresh medium was added to the 2 ml of pre-existing medium of each co-culture . Since the deletion of NS1 protein significantly increases the production of IFN-I from infected DC , which may in turn be responsible for the T cell phenotypes observed in our experiments , a control was included where exogenous IFNα2a and IFNβ ( PBL Interferon Source , Piscataway , NJ ) were added at 133 IU/ml each to co-culture of DC pre-infected with wt RSV with CD4+ as well as CD8+ T cells to monitor such effects . The quantity of IFN-I added to this control was based on our previous studies [28] and represents the average amount produced by human monocyte-derived DC on infection with the viruses lacking NS1 . For IFN-I receptor blocking experiments , mouse monoclonal anti-human IFNAR2-neutralizing antibody ( PBL Interferon Source ) or its isotype control antibody ( R&D Systems , Minneapolis , MN ) was added to the purified CD8+ or CD4+ T cells at a final concentration of 60 µg/ml and incubated for 1 h at 37°C . We chose to use an IFNAR2-blocking antibody because the antiviral activity of IFN-I correlates well with signaling through IFNAR2 rather than IFNAR1 [90] . This mixture was then combined with autologous DC that had been pre-infected with wt RSV or ΔNS1/2 RSV at MOI of 2 PFU/cell or mock infected as described , resulting in final antibody concentration of 30 µg/ml . The co-cultures were incubated at 37°C for 7 days . In another format , the antibody was added to DC for 2 h , followed by infection and additional incubation for 48 h , wash and co-cultivation with T cells as above . To determine the effect of IFN-I , an equal mixture of IFNα2a and IFNβ ( PBL Interferon Source ) was added exogenously to a final concentration of 25 , 50 , 100 , 133 , 200 , 400 , 800 , or 1 , 600 IU/ml each to the co-cultures of DC pre-infected with wt RSV and CD8+ or CD4+ T cells . The co-cultures were incubated at 37°C for 7 days . After 7-days of co-culture , medium supernatants were collected for cytokine assay . The co-cultures were then prepared for intracellular cytokine staining by stimulation with phorbol-12-myristate-13-acetate ( PMA ) ( Sigma Aldrich , St . Louis , MO ) at 20 ng/ml and ionomycin ( EMD Chemicals , Gibbstown , NJ ) at 1 µM in the presence of brefeldin A ( Sigma Aldrich ) at 10 µg/ml , and were incubated at 37°C for 6 h; 5 min before the harvest , DNase I ( Calbiochem , Gibbstown , NJ ) was added at 150 µg/ml . Following harvest , cells were stained with Live/Dead Fixable Blue Dead Cell Stain ( Invitrogen ) to discriminate dead cells by flow cytometry . CD8+ T cells were also stained with PE-labeled anti-CD103 antibodies ( BD Biosciences ) , peridinin chlorophyll protein ( PerCP-Cy5 . 5 ) -labeled anti-CD107a antibodies ( Biolegend , San Diego , CA ) , and pacific orange-labeled anti-CD8 antibodies ( Invitrogen ) . After surface staining , cells were fixed and permeabilized using fixation/permeabilization kit ( BD Biosciences ) as per the manufacturer's instructions . Cells were then blocked and permeabilized by overnight incubation in phosphate buffered saline containing 0 . 1% saponin and 5% non-fat dry milk . CD8+ T cells were then stained for intracellular cytokines and cell markers with allophycocyanin-Alexa Fluor 750 ( APC-Cy7 ) -labeled anti-IFNγ ( Invitrogen ) , PE-Cy7-labeled anti- TNFα ( BD Biosciences ) , APC-labeled anti-IL-2 ( BD Biosciences ) , and PE Texas Red-labeled anti-CD3 antibodies ( Beckman Coulter , Brea , CA ) . CD4+ T cells were stained for intracellular cytokines with APC-Cy7-labeled anti-IFNγ ( Invitrogen ) , PE-labeled anti-IL-4 ( BD Biosciences ) , PerCP-Cy5 . 5-labeled anti-IL-17A ( eBioscience , San Diego , CA ) , PE-Cy7-labeled anti- TNFα ( BD Biosciences ) , and Alexa-Fluor 680-labeled anti-IL-2 antibodies ( Biolegend ) . The stained cells were suspended in staining buffer [phosphate buffered saline , 1% FBS ( HyClone ) ] for flow cytometry analysis . Ten-parameter and 9-parameter flow cytometric analysis was performed for CD8+ and CD4+ T cells , respectively , on a Becton Dickinson LSR II flow cytometer ( BD Biosciences ) . Data compensation was performed using mouse and rat CompBeads compensation beads ( BD Biosciences ) stained with individual antibodies used in the multicolor analysis . Data acquired from the T cells stained only with CFSE and that from DC stained only with CellVue Lavender dye or DDAO were used for compensations involving these fluorochromes . The data were analyzed with FlowJo software version 8 . 8 . 6 ( Tree Star , Ashland , OR ) . The data acquired from stained compensation beads and cells were used for automatic compensation using the FlowJo Compensation Wizard . In order to determine the accuracy of compensation between overlapping spectra and to define the positive cells , fluorescence-minus-one ( FMO ) staining controls were included . The FMO controls included staining of CD4+ T cells , derived from co-cultures with wt RSV-infected DC , with fluorochrome-conjugated antibodies for all cytokines omitting one antibody conjugate per control . The gating strategy followed to obtain proliferated CD8+ or CD4+ T cells for analyses is shown in Fig . 1B . Briefly , the compensated data were gated at UV Live/Dead stain versus CellVue Lavender dye or DDAO to eliminate dead T cells and all DC , and to obtain live T cells . Live T cells were then gated at UV Live/Dead dye versus CD3 to obtain live CD3+ T cells , which were then gated at forward scatter area versus forward scatter height to eliminate cell doublets and clusters and to identify single cells for analyses . Single cells thus identified were again gated at forward scatter area versus side scatter area to exclude any debris and determine cell size and granularity . These cells were then finally gated at CD3 versus CFSE to identify proliferated CD3+ T cells as assessed by reduction in fluorescence intensity of CFSE as a result of cell divisions . For CD8+ T cells , an additional gate was drawn after gate 2 at CD3 versus CD8 to identify CD3+ CD8+ T cells . The proliferated cells thus obtained were further gated for individual cytokines and activation markers including CD103 , CD107a , IFNγ , TNFα and IL-2 for CD8+ T cells , and IFNγ , IL-4 , IL-17 , TNFα and IL-2 for CD4+ T cells . The FMO controls demonstrated that cells stained with all but one cytokine antibody were indeed negative for that cytokine ( Fig . S2 ) . This suggests that the electronic compensations applied to the data were accurate , and that the cells found positive for a particular cytokine were indeed a result of specific staining with the corresponding antibody conjugate and were not false positives due to fluorescence contribution from the overlapping spectra . Proliferated T cells gated positive for individual cytokines were analyzed by Boolean gating to obtain frequencies for 32 possible cytokine combinations both positive and negative for individual cytokines using FlowJo software . The identified cell populations were further analyzed using Pestle and Spice software ( version 4 . 2 . 3; Mario Roederer and Joshua Nozzi , Vaccine Research Center and Bioinformatics and Scientific IT Program , NIAID , NIH , Bethesda , MD ) . Data obtained by Boolean gating were normalized to wt RSV and plotted for graphical representation using Prism 5 software ( GraphPad software Inc . , La Jolla , CA ) . DC were treated with the IFNAR2 blocking antibody or its isotype control for 2 h and then infected with wt RSV or ΔNS1/2 RSV , followed by 48 h-long incubation and analysis of maturation markers by QRT-PCR . Total RNA was extracted from DC using RNeasy kit ( Qiagen , Valencia , CA ) and used to synthesize cDNA using Superscript III reverse transcriptase kit ( Invitrogen ) as per manufacturer's protocol . The cDNA thus obtained was used to analyze 49 genes ( the full gene names and the NCBI gene reference numbers are provided in Table S1 ) in triplicate by QRT-PCR on a TaqMan micro fluidic gene card ( Applied Biosystems , Foster City , CA ) using TaqMan universal PCR master mix ( Applied Biosystems ) . The 18S ribosomal RNA was included on the gene card in triplicate and used as an endogenous control for analysis . PCR was run on 7900HT fast real-time PCR system ( Applied Biosystems ) . The threshold cycle ( Ct ) values obtained for DC infected with wt RSV without any antibody treatment were used as the calibrator for calculating the fold change in gene expression for the rest of the treatments . The entire experiment was performed four times using DC prepared from four individual donors . The expression fold change values were converted to log2 and analyzed with Genesis software version 5 ( Alexander Sturn , Institute for Genomics and Bioinformatics , Graz University of Technology , Petersgasse , Austria ) to perform the hierarchical clustering of data and generate a heat map . The fold changes for CXCL9 , CXCL10 , RIG-I , CD38 , CD40 , CD80 , IL-12β , and IL-23α were also plotted for graphical representation using Prism 5 software ( GraphPad software Inc . ) . 2×106 CD8+ or CD4+ T cells purified from elutriated lymphocytes as above were treated with IFNAR2 antibody at concentrations 0 . 03 , 0 . 3 , 3 , or 30 µg/ml , plated in 12-well plates and incubated for 2 h at 37°C . Next , IFNα2a was added to all cells at 100 IU/ml , followed by incubation at 37°C . Cells were harvested at 3 , 6 , and 18 h later , and total RNA was isolated using RNeasy kit ( Qiagen , CA ) . The mRNA expression of MX1 and ISG56 was investigated by QRT-PCR using TaqMan gene expression assay primers and probe set and one step RT-PCR master mix ( Applied Biosystems ) . The 18S ribosomal RNA was quantitated for all samples in triplicate , using TaqMan primers and probe , and was used as endogenous control . The reactions were run in triplicate in ABI 7900HT fast real-time PCR system ( Applied Biosystems ) and analyzed with SDS 2 . 3 and RQ Manager 1 . 2 softwares ( Applied Biosystems ) . Quantitation of MX1 expression in cells treated with IFNAR2 antibody and IFNα2a was analyzed relative to the calibrator , i . e . , cells treated with IFNα2a in the absence of IFNAR2 antibody . The data is presented as MX1 expression relative to the calibrator . Co-culture medium supernatants were analyzed for various cytokines by ELISA . Supernatants from CD8+ T cells were analyzed for IFNγ and TNFα while those from CD4+ T cells were assayed for IFNγ , IL-4 , IL-17 , IL-2 and IL-12 . All ELISAs were performed with Quantikine colorimetric sandwich ELISA kits ( R&D Systems , Minneapolis , MN ) . DC supernatants were analyzed for IFNα using IFNα multi-subtype ELISA kit ( PBL Interferon Source ) . All data involving the comparison of wt RSV with its NS1 and/or NS2 deletion mutants were analyzed for statistical significance using repeated measures ANOVA with Tukey's multiple comparison post test ( GraphPad Prism 5 software ) . In the experiments involving blocking of IFNAR2 or adding exogenous IFN-I , paired one-tailed T-test was used ( GraphPad Prism 5 software ) . For all statistical analyses data were considered significant at P<0 . 05 . | Respiratory syncytial virus ( RSV ) is a leading cause of pediatric lower respiratory tract disease . RSV has two IFN-I antagonist proteins , NS1 and NS2 . In this study , we infected primary human dendritic cells with recombinant RSV from which the NS1 and/or the NS2 genes were deleted , and evaluated effects on the proliferation of autologous T lymphocytes during co-culture in vitro . We found that NS1 , but not NS2 , has a suppressive effect on two cell populations , namely CD103+ CD8+ T cells and Th17 cells , which are known to protect against viral respiratory infections , and a stimulatory effect on Th2 cells , which are involved in enhanced disease caused by RSV . We also provide evidence that these effects are not due to suppressed IFN-I production or signaling in dendritic cells or T cells , and that they likely result from reduced maturation of dendritic cells caused by the NS1 protein . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"virology/virulence",
"factors",
"and",
"mechanisms",
"infectious",
"diseases",
"immunology/immune",
"response",
"immunology/innate",
"immunity",
"virology",
"infectious",
"diseases/viral",
"infections",
"virology/mechanisms",
"of",
"resistance",
"and",
"susceptibility,",
"including",
"host",
"genetics",
"immunology",
"virology/immune",
"evasion",
"virology/effects",
"of",
"virus",
"infection",
"on",
"host",
"gene",
"expression",
"immunology/immunity",
"to",
"infections",
"virology/host",
"antiviral",
"responses"
]
| 2011 | Respiratory Syncytial Virus Interferon Antagonist NS1 Protein Suppresses and Skews the Human T Lymphocyte Response |
Bordetella adenylate cyclase toxin-hemolysin ( CyaA ) penetrates the cytoplasmic membrane of phagocytes and employs two distinct conformers to exert its multiple activities . One conformer forms cation-selective pores that permeabilize phagocyte membrane for efflux of cytosolic potassium . The other conformer conducts extracellular calcium ions across cytoplasmic membrane of cells , relocates into lipid rafts , translocates the adenylate cyclase enzyme ( AC ) domain into cells and converts cytosolic ATP to cAMP . We show that the calcium-conducting activity of CyaA controls the path and kinetics of endocytic removal of toxin pores from phagocyte membrane . The enzymatically inactive but calcium-conducting CyaA-AC− toxoid was endocytosed via a clathrin-dependent pathway . In contrast , a doubly mutated ( E570K+E581P ) toxoid , unable to conduct Ca2+ into cells , was rapidly internalized by membrane macropinocytosis , unless rescued by Ca2+ influx promoted in trans by ionomycin or intact toxoid . Moreover , a fully pore-forming CyaA-ΔAC hemolysin failed to permeabilize phagocytes , unless endocytic removal of its pores from cell membrane was decelerated through Ca2+ influx promoted by molecules locked in a Ca2+-conducting conformation by the 3D1 antibody . Inhibition of endocytosis also enabled the native B . pertussis-produced CyaA to induce lysis of J774A . 1 macrophages at concentrations starting from 100 ng/ml . Hence , by mediating calcium influx into cells , the translocating conformer of CyaA controls the removal of bystander toxin pores from phagocyte membrane . This triggers a positive feedback loop of exacerbated cell permeabilization , where the efflux of cellular potassium yields further decreased toxin pore removal from cell membrane and this further enhances cell permeabilization and potassium efflux .
By instantaneously disrupting bactericidal functions of host phagocytes , the adenylate cyclase toxin-hemolysin ( CyaA , ACT , or AC-Hly ) plays a major role in virulence of pathogenic Bordetellae [1] . The toxin rapidly paralyzes phagocytes [1] , [2] by translocating across their cytoplasmic membrane an N-terminal adenylate cyclase enzyme ( AC ) domain ( ∼400 residues ) that binds cytosolic calmodulin and converts ATP to a key signaling molecule , cAMP [1] . In parallel , the multidomain ∼1300 residues-long RTX ( Repeat in ToXin ) cytolysin moiety of CyaA acts independently as a pore-forming leukotoxin and hemolysin [1] . This employs a hydrophobic pore-forming domain ( residues 500 to 700 ) , a domain with covalently palmitoylated lysine residues 860 and 983 , and a typical calcium-binding RTX repeat domain within the last 700 residues of CyaA that accounts for receptor binding [1] . CyaA can oligomerize into small cation-selective pores that mediate efflux of cytosolic potassium ions from cells [3]–[6] , eventually provoking colloid-osmotic cell lysis [7]–[9] . This activity synergizes with cytotoxic signaling of the translocated AC enzyme in bringing about the final cytotoxic action of CyaA [8] , [9] . The structure of CyaA in target membrane remains unknown . Accumulated indirect evidence strongly suggests that the cell-invasive AC and the pore-forming activities are associated with distinct subpopulations of CyaA conformers within target cell membrane . Indeed , translocation of the AC domain across cellular membrane and the pore-forming activity of CyaA can be dissociated by temperature , calcium concentrations , or altered acylation [5] , [10] , [11] . Moreover , the balance between the two activities can be largely shifted in either direction by specific residue substitutions . Several CyaA variants with strongly enhanced pore-forming activity and reduced or nil capacity to deliver the AC domain into cells could be generated . Recently , a CyaA defective in formation of toxin pores but intact in AC delivery into cells could also be constructed [3] , [6] , [12] . Summarizing this evidence , a model was proposed that predicts the co-existence of two distinct toxin conformers inside target cell membrane . These would operate in parallel and would employ the same segments of the hydrophobic domain in an alternative manner . One conformer would insert into cell membrane as toxin translocation precursor that accomplishes delivery of the AC domain into cells ( cell-invasive activity ) . The second conformer would account for formation of oligomeric CyaA pores [3] , [6] , [12] . The primary targets of CyaA appear to be host myeloid phagocytes , to which the toxin binds through their αMβ2 integrin , known as CD11b/CD18 , CR3 or Mac-1 [13] . Recently , we could show that AC translocation into cells occurs in two steps . First , the membrane–inserted translocation precursors of CyaA generate a calcium-conducting path across cell membrane and mediate influx of extracellular calcium ions into cells [14] . The incoming Ca2+ than activates calpain-mediated processing of the talin tether of CD11b/CD18 , which triggers relocation of the toxin-receptor complex into cholesterol-rich membrane lipid rafts . There the translocation of the AC domain into cytosol of cells across the cytoplasmic membrane of phagocytes is completed [15] . Unlike for most other enzymatic toxins , the capacity of CyaA to deliver its AC enzyme component into target cells does not depend on receptor-mediated endocytosis . Instead , the AC domain translocates into cytosol directly across the cytoplasmic membrane of phagocytes . Indeed , inhibitors of endocytic pathways interfere neither with translocation of the AC domain into cells and elevation of cytosolic cAMP , nor with AC domain-mediated delivery of inserted CD8+ T cell epitopes into the cytosol of antigen presenting cells [16]–[18] . Clathrin-dependent endocytic uptake of CyaA together with its receptor CD11b/CD18 has , however , previously been observed [19] , [20] and it was found to account for the capacity of CyaA-AC− toxoids to deliver cargo antigens into dendritic cells for endosomal processing and MHC II-restricted presentation to CD4+ T cells [18] . Therefore , we analyzed here the mechanism of endocytic uptake of CyaA from cell membrane and its relevance for toxin action . We show that by conducting calcium ions across the target membrane into cytosol of cells , the translocating CyaA molecules control the rate of toxin pore removal from cellular membrane and thereby support the permeabilization of phagocytes .
Recently we found that despite binding to CD11b/CD18 , the CyaA constructs lacking the hydrophobic domain failed to deliver a fused MalE antigen for presentation to CD4+ T lymphocytes by dendritic cells ( Figure S1 ) . To address the role of functional integrity of CyaA in its endocytic trafficking , we employed live microscopy imaging . To avoid the interference of massive phagocyte ruffling and death from ATP depletion and cAMP signaling provoked by enzymatically active CyaA at concentrations required for imaging [8] , [21] , we used fluorescently labeled and enzymatically inactive CyaA-AC− toxoids . To enable assessment of their capacity to deliver antigens for presentation on MHC class I and II molecules , the toxoids were further tagged by insertion of a CD8+ T-cell epitope ( SIINFEKL ) from ovalbumin at position 336 and of a MalE CD4+ T-cell epitope ( NGKLIAYPIAVEALS ) at position 108 of the AC domain , respectively . Trafficking of such tagged dCyaA was then compared to that of a doubly mutated CyaA-E570K+E581P-AC− toxoid ( dCyaA-KP ) that carried a combination of debilitating substitutions of glutamate residues at positions 570 and 581 within the pore-forming domain . That construct was previously shown to retain a full capacity to bind the toxin receptor CD11b/CD18 , without being able to conduct calcium ions into cells , to associate with lipid rafts and translocate the AC domain across cell membrane , or to form oligomeric CyaA pores within cellular membrane , respectively [3] , [15] . As shown in Figure 1 , insertion of the epitope tags and fluorescent labeling did not alter the expected properties of the dCyaA toxoid , which elevated [Ca2+]i in J774A . 1 cells ( Figure 1A ) and relocated into lipid rafts in J774A . 1 membrane ( Figure 1B ) . As also expected , the dCyaA-KP toxoid lacked all these activities . When examined by live cell imaging , the two fluorescently labeled toxoids exhibited intriguingly different kinetics and patterns of endocytic uptake . As documented by representative images in Figure 1C and quantified in Figure 1D , dCyaA was detected predominantly within the plasma membrane , or in fluorescent vesicles located beneath the membrane of J774A . 1 cells throughout the 20 minutes of incubation at 37°C ( Video S1 ) . In contrast , dCyaA-KP was taken-up much faster and accumulated massively within endocytic vesicles dispersed through the cytosol of cells already within 5 minutes from toxoid addition . The dCyaA-KP toxoid was then almost completely removed from cell membrane within 10 minutes ( Figure 1C , Video S2 ) , as quantified by counting of the intracellularly located fluorescent vesicles for a set of inspected cells ( Figure 1D , see Figure S2 for the method ) . Identical results were observed upon swapping of the fluorescent labels ( not shown ) , thus excluding the impact of the used dyes on toxoid trafficking . While the two toxoids entered into quite different endocytic pathways , the uptake of both was receptor-mediated and depended on toxoid binding to CD11b/CD18 . Blocking of the receptor by the M1/70 antibody abrogated , indeed , cell binding and endocytosis of both toxoids , as documented in Figure 1C and Figure 1E . To characterize the differing uptake pathways , we next examined the co-localization of the toxoids with transferrin , a marker of clathrin-dependent endocytosis [22] . As shown by representative images in Figure 2A and quantified by Pearson's correlation coefficients ( P . c . c . ) for compared channels in Figure 2B , the co-localization of dCyaA with transferrin ( Dyomics 547 ) increased in time and was near-complete after 60 minutes of co-incubation with cells . In contrast , a weak and progressively decreasing co-localization of dCyaA-KP with transferrin was observed over the same time interval . Hence , while dCyaA was transiting through the early and/or recycling compartment together with transferrin , the dCyaA-KP toxoid took a different pathway . To corroborate this observation , the uptake of the two toxoids was examined in RAW 264 . 7 macrophages expressing a dominant negative variant of the EPS-15 protein ( DN EPS-15-GFP , DIII ) , which selectively interferes with clathrin-dependent endocytosis [23] . As documented by a representative image in Figure 2C ( left ) , the accumulation of transferrin and dCyaA at intracellular sites was abrogated in DN EPS-15-GFP-transfected cells ( blue ) and no co-localization of cell-associated dCyaA with transferrin was observed upon removal of surface-associated transferrin with a low pH buffer wash . As further documented by the z-axis projections , few if any intracellular vesicles containing dCyaA were observed inside DN EPS-15-GFP-transfected cells and the membrane-associated dCyaA was located exclusively inside fluorescent patches on cell surface . In striking contrast , the endocytic uptake of dCyaA-KP was unaffected in DN EPS-15-GFP-transfected cells and intracellular vesicles containing dCyaA-KP were clearly observed in the z-axis projections ( Figure 2C , right panel ) . Neither of the two toxoids exhibited any co-localization with caveolin-1 ( Figure S3 ) . By difference to dCyaA , however , the dCyaA-KP exhibited a strong co-localization with soluble fluorescent BSA serving as a fluid phase uptake marker . As also shown in Figure 3C and quantified in Figure 3D , the kinetics of dCyaA-KP endocytic uptake was strongly decelerated upon pretreatment of cells with wortmannin ( 1 µM ) , a PI3 kinase inhibitor blocking macropinocytosis but not micropinocytosis [24] . In contrast , no inhibition of dCyaA-KP uptake was observed upon treatment with dynasore or chlorpromazine , the inhibitors of clathrin-dependent endocytosis [25] , [26] . It can , hence , be concluded that dCyaA was endocytosed through a decelerated clathrin-dependent pathway , while dCyaA-KP was internalized with the cytoplasmic membrane by a macropinocytic mechanism . Clathrin-dependent endocytosis was previously shown to enable dCyaA-mediated delivery of model antigens into CD11b+ dendritic cells for endosomal processing and MHC class II-restricted presentation to CD4+ T cells [18] . Therefore , we examined whether altered endocytic trafficking affected the antigen delivery capacity of the dCyaA-KP toxoid . As documented in Figure 4A and quantified in Figure 4B , following one hour of incubation with RAW 264 . 7 macrophages expressing Rab-7-EGFP , fluorescently labeled material derived from either of the toxoids accumulated within organelles positive for the lysosomal/late endosomal marker Rab7 . dCyaA and dCyaA-KP toxoid-derived fluorescence also co-localized to the same extent with the MHC II–EGFP molecules in a subset of intracellular vesicles of bone marrow derived dendritic cells from a MHC Class II–EGFP knock-in mice , within 120 minutes of incubation ( Figure 4C , 4D ) . Despite entering cells by different endocytic pathways , hence , the two toxoids or their fragments reached similar acidic endosomes . To test if faster macropinocytic uptake lead to alteration of dCyaA-KP processing , lyzates from toxoid-pulsed J774A . 1 cells were probed in Western blots with the 9D4 MAb that recognizes C-terminal RTX repeats of CyaA . As shown in Figure 5A , comparable amounts of the ∼200 kDa forms of both toxoids and of their fragments were detected in lyzates of cells pretreated with inhibitors of endocytosis and protease inhibitors , like 0 . 01% sodium azide plus 10 mM 2-deoxy glucose ( 2DG ) , or a protease inhibitor cocktail plus 1 mM chloroquine . As compared to dCyaA , however , a notably faster degradation dCyaA-KP occurred in uninhibited cells ( Figure 5A ) , as judged from the detected amounts of smear corresponding to partially digested molecules . dCyaA-KP toxoid , hence , appeared to be degraded faster and more completely , albeit the final protease-resistant fragments of both toxoids appeared to be of similar size . To examine if faster uptake and degradation affected the capacity of dCyaA-KP to deliver cargo epitopes for endosomal processing and MHC II-restricted presentation , the relative capacity of the two toxoids to deliver CD4+ T cell epitopes was assessed . As shown in Figure 5B by comparison to dCyaA , the dCyaA-KP toxoid exhibited an about ten-fold reduced capacity to deliver the MalE epitope for presentation by BMDCs to CRMC3 CD4+ T hybridoma cells [27] . To corroborate this observation , the MalE epitope was replaced by the OVA258–276 epitope recognized by MF2 . 2D9 CD4+ T hybridoma and new dCyaA and dCyaA-KP toxoids were produced . As shown in Figure 5C , however , a similarly reduced capacity to deliver the OVA258–276 epitopes for MHC II-restricted presentation was again observed with dCyaA-KP , as compared to dCyaA . This suggested that faster proteolytic destruction of the cargo epitope during trafficking through the macropinocytic-like pathway may have accounted for the reduced efficacy of dCyaA-KP in antigen delivery . To test whether the loss of capacity to mediate influx of Ca2+ ions into cells committed dCyaA-KP for the rapid macropinocytic uptake and subsequent degradation , the J774A . 1 cells were incubated with dCyaA-KP in the presence of ionomycin , a calcium ionophore that permeabilizes cells for extracellular Ca2+ . As documented in Figure 6A , Video S3 , and as quantified in Figure 6B , respectively , internalization of dCyaA-KP into fluorescent intracellular vesicles was strongly delayed upon treatment of J774A . 1 cells with 5 or 10 µM ionomycin in the presence of 1 . 9 mM Ca2+ . As also shown in Figure 6A ( Video S4 ) and quantified in Figure 6B and 6C , dCyaA-KP was redirected into a slower uptake pathway even more efficiently upon co-incubation at a 1∶1 ratio with intact dCyaA . This goes well with our previous observation that permeabilization of cells for Ca2+ in trans by intact dCyaA rescued in part the defect of a CyaA-KP construct and mobilized it into lipid rafts [15] . Upon co-incubation with dCyaA , a fraction of biotinylated dCyaA-KP protein was indeed found to float in sucrose gradients with detergent-resistant membrane ( Figure 6D ) . Furthermore , following co-incubation with dCyaA , the dCyaA-KP appeared to be endocytosed with the same kinetics and through the same pathway as the intact toxoid ( Videos S1 and S4 ) . The two proteins co-localized within the same endocytic vesicles at 30 and 60 minutes of incubation with cells ( Figure 6E , 6F ) . Hence , permeabilization of cells for Ca2+ by ionomycin or intact toxoid rescued dCyaA-KP from the rapid macropinocytic membrane uptake pathway and redirected it for decelerated and clathrin-dependent endocytosis . We have previously observed that the CyaA-ΔAC hemolysin construct lacking the AC domain of CyaA ( Δ1–373 ) was unable to promote Ca2+ influx into monocytes and was essentially unable to provoke lysis of J774A . 1 cells [14] . However , on planar lipid bilayers , or on sheep erythrocytes devoid of endocytic mechanisms , the CyaA-ΔAC hemolysin exhibited the same specific pore-forming and hemolytic activities as the full-length dCyaA ( CyaA-AC− ) [4] . We thus hypothesized that its inability to lyze J774A . 1 cells might be due to rapid removal of the CyaA-ΔAC pores from the cytoplasmic membrane of J774A . 1 cells . As shown in Figure 7A , the CyaA-ΔAC hemolysin elicited much slower efflux of cytosolic K+ from J774A . 1 cells then the full-length toxoid . In cells exposed to 3 µg/ml of enzymatically inactive CyaA-AC− a complete drop of cytosolic [K+]i concentration down to 10 mM was reproducibly observed within 20 minutes , while in cells exposed to equal amounts of CyaA-ΔAC the [K+]i only decreased to 80 mM ( Figure 7A ) . To determine if this was due to rapid removal of CyaA-ΔAC pores from cell membrane , we assessed the capacity of CyaA-ΔAC to elicit K+ efflux on cells with membrane trafficking inhibited upon preincubation in media containing 0 . 01% ( w/w ) sodium azide and 10 mM 2-deoxy-glucose ( 2DG ) . This treatment potentiated the cell-permeabilizing activity of full-length CyaA-AC− toxoid ( Figure 7A , 7B ) , but did not promote any significant K+ efflux from cells on its own despite causing an about five-fold drop of cellular ATP levels ( Figure 7A , 7B and 7C ) . As further documented in Figure 7E and 7F ( Videos S5 and S6 ) , the strong inhibition of the CyaA-ΔAC uptake was accompanied by a steep increase of the specific capacity of CyaA-ΔAC to promote K+ efflux ( Figure 7A and 7B ) from cells and lyze J774A . 1 monocytes ( Figure 7D ) . It can thus be concluded that rapid macropinocytic internalization with cell membrane was strongly restricting the cell permeabilizing and cytolytic capacities of the CyaA-ΔAC hemolysin pores . As further shown in Figure 7G , despite the prolonged persistence in the cytoplasmic membrane of cells upon inhibition of endocytosis , the CyaA-ΔAC hemolysin failed to associate with lipid rafts ( compare CyaA-ΔAC panel in Figure 7G and upper dCyaA panel of Figure 1B for comparable toxoid loading ) . This goes well with our previous observation that mobilization into lipid rafts with CD11b/CD18 depends on the capacity of CyaA to conduct calcium ions into cells to activate talin cleavage by calpain [15] . More importantly , this result shows that CyaA pores can form outside of lipid rafts within the bulk phase of cell membrane . We next tested if Ca2+ influx induced in trans would delay removal of CyaA-ΔAC from cytoplasmic membrane and extend thereby its capacity to permeabilize J774A . 1 cells . Mobilization of Ca2+ into cells with ionomycin , however , caused on its own a high unspecific leakage of K+ from J774A . 1 cells ( data not shown ) . Therefore , an alternative approach was used , exploiting the capacity of the 3D1 MAb to lock CyaA molecules in the conformation of membrane-inserted ‘translocation precursors’ that conduct calcium ions across the cytoplasmic membrane of cells [15] . CyaA-ΔAC was preincubated with 3D1 , or with an isotype control IgG1 MAb ( TU-01 , 20 µg/ml ) and the capacity of the CyaA-ΔAC to elicit K+ efflux from J774A . 1 cells was assessed . As documented in Figure 8A , while the 3D1 MAb alone did not cause any elevation of [Ca2+]i , the binding of 3D1 conferred on CyaA-ΔAC the capacity to promote rapid influx of Ca2+ into monocytes . As shown in Figure 8B , this allowed association of detectable amounts of CyaA-ΔAC molecules with lipid rafts . More importantly , the capacity of CyaA-ΔAC to permeabilize J774A . 1 cells for K+ efflux was strongly enhanced in the presence of 3D1 MAb and equaled the specific cell-permeabilizing activity of the intact and enzymatically active CyaA , as shown in Figure 8C and 8D . Furthermore , this enhancement of cell-permeabilizing activity was accompanied by a strong deceleration of endocytic uptake of CyaA-ΔAC ( Figure 8E , 8F ) . To test the hypothesis that toxoid-induced K+ efflux itself contributed further deceleration of endocytosis , the clathrin-mediated uptake of dCyaA was examined in media supplemented with 50 mM potassium ions . As documented in Figure 9A and quantified in Figure 9B , the endocytosis of dCyaA was clearly accelerated when efflux of K+ from cells was counteracted by addition of potassium ions into media . Collectively , hence , these results show that CyaA-mediated Ca2+ influx rescues hemolysin pores from rapid macropinocytic uptake from cell membrane and thus extends their cell-permeabilizing and cytolytic action . The pore-forming activity was previously shown to synergize with the ATP-depleting and cAMP-signaling activities of the cell-invasive AC domain of CyaA [8] , [9] . Therefore , we examined to which extent does endocytic uptake modulate the cytotoxicity of fully active ( AC+ ) toxin . We assessed first the impact of cAMP accumulation on the uptake of CyaA , using a fluorescently labeled CyaA-KP construct ( AC+ ) that exhibits only a residual capacity to deliver the AC enzyme into cells . Unlike active toxin , hence , CyaA-KP does not provoke cell death at the high concentrations ( 1 to 5 µg/ml ) employed in live cell imaging . As shown in Figure 10A , at such concentrations the dCyaA-KP produced easily detectable cAMP amounts in cells , while the extent of its endocytic uptake was the same as that of the enzymatically inactive ( AC− ) dCyaA-KP toxoid ( Figure 10B , 10C ) . Since both proteins were unable to mediate detectable Ca2+ influx into J774A . 1 cells ( ref . [15] and Figure 1B ) , it can be concluded that cAMP levels alone do not noticeably influence the rate of endocytic uptake of CyaA . It was next important to characterize the rate of endocytic uptake of CyaA at as low toxin concentrations as that presumably encountered by host phagocytes during Bordetella infections . Therefore , the kinetics of CyaA endocytosis by J774A . 1 cells was assessed by a flow cytometry assay measuring the amount of biotinylated CyaA that remains accessible to binding by streptavidin on cell surface . As shown in Figure 11 , pulsing of J774A . 1 cells for 5 minutes at 37°C with 200 ng/ml of CyaA-biotin yielded about 1 ng of toxin stably bound to 3×105 washed cells , when kept on ice . In contrast , the amounts of surface-exposed CyaA-biotin decreased progressively upon transfer of cells to 37°C , with ∼80% of CyaA being removed from cell surface within 15 minutes . As also shown in Figure 11 , the rate of CyaA-biotin uptake was reduced by about a factor of two in the presence of a 5-fold ( non-saturating ) excess of unlabeled CyaA-AC− toxoid ( 1 µg/ml ) that enhanced Ca2+ influx and K+ efflux across the membrane of J774A . 1 cells [14] . To address the relative contribution of cell surface retention of CyaA pores to the overall cytotoxic potency of low amounts of intact ( AC+ ) CyaA , the natively fatty-acylated Bp-CyaA purified from B . pertussis 18323/pHSP9 was used [28] . Bp-CyaA exhibits about fourfold higher specific pore-forming ( hemolytic ) activity than the recombinant rEc-CyaA produced in E . coli ( [11] and Figure S4 ) . Therefore the use of Bp-CyaA allowed maximizing the observable amplitude of changes in cell permeabilizing and cytolytic capacity that would result from alterations of the rate of toxin pore removal from cell membrane . As indeed shown in Figure 12 , as little as 100 ng/ml of Bp-CyaA promoted a detectable LDH release form J774A . 1 cells already in 2 hours of incubation . Inhibition of endocytic uptake of Bp-CyaA with 2DG and sodium azide treatment then increased the cytolytic potency of Bp-CyaA by a factor of 2 to 3 ( Figure 12A ) and strongly increased the capacity of Bp-CyaA to permeabilize cells for K+ efflux , which become well observable already at 300 ng/ml of the toxin ( Figure 12B ) .
We show here that by conducting Ca2+ ions across target cell membrane , CyaA decelerates its endocytic uptake and escapes from rapid macropinocytic removal from cell membrane and destruction in endosomes . By redirecting the toxin into a decelerated clathrin-dependent uptake pathway , the calcium-conducting activity of toxin translocation intermediates protracts toxin pore persistence within cytoplasmic membrane , thus extending phagocyte permeabilization and maximizing cytotoxic action of CyaA , as summarized in the model proposed in Figure 13 . This mechanism could be directly demonstrated here for the CyaA-AC− toxoid and CyaA-ΔAC hemolysin variants that lack the AC enzyme activity and could be used at sufficiently high concentrations for live cell imaging . Imposing on CyaA-ΔAC a conformation that enabled it to conduct Ca2+ into cells , indeed , rescued the hemolysin from the macropinocytic pathway , and by decelerating its removal form cell membrane , it particularly enhanced the otherwise very low cell-permeabilizing and cytolytic capacity of CyaA-ΔAC . Using intact toxin ( AC+ ) purified form B . pertussis cells at close to physiologically relevant concentrations ( 200 ng/ml ) , we found that inhibition of endocytic uptake of Bp-CyaA enhanced its capacity to lyze cells . It was , however , not possible to design an experiment that would directly prove the role of calcium-induced deceleration of CyaA uptake in its cytotoxic action . Such demonstration would only be possible if inhibition of Ca2+ influx would leave the other cytotoxic activities of CyaA unaffected . This would then allow to test whether cytolytic activity of CyaA is reduced by acceleration of the endocytic uptake of the intact toxin . However , Ca2+ influx into cells is a prerequisite for CyaA relocation into lipid rafts and subsequent AC domain translocation into cell cytosol [15] . Therefore , blocking of Ca2+ influx by whatever means inevitably ablates also the major component of the cytotoxic activity of CyaA , namely the capacity of its AC enzyme to reach cell cytosol and catalyze unregulated dissipation of cytosolic ATP into cAMP to impair cellular signaling . We have previously shown that upon initial binding of the CD11b/CD18 integrin , the CyaA toxin inserts into the membrane lipid bilayer as a translocation precursor , in which the segments of the AC domain cooperate with segments of the pore-forming domain in forming a novel calcium-conducting path across the phagocyte membrane that mediates influx of extracellular Ca2+ ions into cells [14] , [15] . Activation of calpain by incoming Ca2+ then yields cleavage of the talin tether of CD11b/CD18 and liberates the toxin-receptor complex for recruitment into membrane lipid rafts [15] . This process appears to be entirely independent of the cAMP-generating capacity of CyaA in CD11b+ J774A . 1 macrophage cells , as a quite comparable amplitude and even faster initial kinetics of calcium mobilization into cells is observed with the enzymatically inactive dCyaA variant ( CyaA-AC− ) than by intact CyaA [14] , [15] . The results presented in this study then show that it is rather the permeabilization of cells for influx of Ca2+ , than the toxin relocation into lipid rafts itself , which allows the toxoid to escape the rapid macropinocytic removal from cell surface with the cytoplasmic membrane . Moreover , lipid rafts are tiny microdomains in the membrane that would accommodate only limited amounts of the hemolysin , while a quantitative escape from rapid macropinocytosis of dCyaA-KP toxoid and its redirection for decelerated endocytosis was observed upon co-incubation with dCyaA ( Figure 6A ) . This indicates that by permeabilizing cells for calcium influx , the AC-translocating CyaA conformers provoke also deceleration of endocytosis of the bystander CyaA pores that can form outside the rafts , in the bulk phase of cell membrane . Such conclusion would also be in line with the observed opposing phenotypes of the CyaA-E509K+516K ( CyaA-KK ) and CyaA-E570Q+K860R ( CyaA-QR ) variants of CyaA . While the CyaA-KK exhibits a strongly enhanced specific pore-forming activity , its capacity to promote Ca2+ influx into cells is very low [14] , [29] . On the opposite , the CyaA-QR toxin exhibits a very low capacity to form pores , to permeabilize cells and elicit K+ efflux , while it mediates normal levels of calcium influx into cells [6] , [15] . This suggests that K+ efflux and Ca2+ influx are two parallel and independent processes that are associated with distinct conformational and/or oligomeric states of the two co-existing conformers of membrane-inserted CyaA . In this respect , it is puzzling that binding of the 3D1 antibody to CyaA-ΔAC exacerbated at the same time its capacity to conduct calcium ions into cells , as well as its cell permeabilizing capacity . This would , perhaps , be best explained by the bystander effect mentioned above . Due to association-dissociation equilibrium , the antibody would lock only a fraction of membrane-inserted CyaA-ΔAC molecules into a calcium-conducting conformation . The other fraction of hemolysin molecules , not bound with 3D1 , would hence be free to form the cell-permeabilizing oligomeric pores promoting K+ efflux , benefiting from the calcium-induced deceleration of the endocytic uptake of the hemolysin pores . It remains , nevertheless , to be conclusively shown that bridging of CyaA-ΔAC dimers by the bivalent 3D1 antibody does not also contribute to the enhanced cell permeabilization capacity and K+ efflux mediated by the CyaA-ΔAC hemolysin complexes with the antibody . 3D1 might , indeed , potentially stabilize the formed oligomeric pores against dissociation within the membrane . At present it can neither be formally excluded that the complex of membrane-inserted CyaA-ΔAC with 3D1 might be conducting at the same time the Ca2+ ions into cells and the K+ ions out of the cell . For intact CyaA the accumulated evidence strongly argues against such possibility ( see above ) . Due to absence of the translocated AC domain , however , the 3D1-locked conformers of CyaA-ΔAC may be capable of oligomerization into pores enabling K+ efflux , or the calcium-conducting path formed by these conformers may be accessible for efflux of cytosolic K+ ions in the absence of the AC domain . A particularly intriguing observation is that the Ca2+ influx and K+ efflux-promoting activities of the toxin synergize in manipulating the pathway and kinetics of CyaA endocytosis . Indeed , the decrease of intracellular potassium level was repeatedly shown to cause inhibition of clathrin-mediated endocytosis [30] , [31] . Hence , the more the cell becomes permeabilized for efflux of K+ , upon inhibition of macropinocytic uptake of toxin pores by incoming Ca2+ , the more the potassium leakage from cells through toxin pores further decelerates the clathrin-dependent removal of CyaA pores from cell membrane . Such cooperation of Ca2+ influx and K+ efflux-mediating activities of CyaA would thus generate a positive feedback loop , exacerbating potassium depletion due to steadily increasing extent of cell membrane permeabilization by persisting and/or newly forming CyaA pores . The above outlined positive feedback loop of potassium efflux was apparently operating under the used experimental conditions , since vesicles containing dCyaA were observed to accumulate as tightly attached to , or located just beneath the cell membrane , for over 20 minutes from toxoid addition ( Figure 1 and Video S1 ) . This shows that excision of clathrin-coated vesicles from cell membrane was inhibited and protraction of cell membrane permeabilization by the persisting pores then fed back into deceleration of endocytic uptake of dCyaA . The enzymatically active CyaA could not be used in this study for analysis of endocytic trafficking of CyaA by live cell microscopy imaging , as at the high concentrations of labeled proteins ( >1 µg/ml ) , required for this type of analyzes , the enzymatically active CyaA uncontrollably impairs intracellular trafficking by rapid depletion of ATP and induction of cell death [8] , [9] . Indeed , the cytotoxic action of enzymatically active CyaA on CD11b+ phagocytes was documented repeatedly at doses lower than 10 ng/ml , where less than 1 ng/ml of active CyaA toxin quantitatively inhibits oxidative burst in neutrophils [32] . As low CyaA doses as 5 to 10 ng/ml elicit monocyte ruffling and a near-instant inhibition of CR3-mediated opsonophagocytosis or arrest of the fluid-phase uptake , respectively . This appears to be due to cAMP signaling-mediated inhibition of the small GTPase RhoA and possibly of the PI3 kinase [21] . The results obtained here with the toxoids appear , nevertheless , to be relevant also to the understanding of endocytosis and of cytotoxic action of the enzymatically active CyaA . By using a newly developed FACS assay for cell surface accessibility of bound CyaA , we could approach here the kinetics of endocytic removal of intact CyaA from cell surface at close to physiologically low toxin concentrations ( 200 ng/ml ) . Under such conditions , the receptor-bound CyaA was found to be progressively removed from cell surface over 15 minutes of incubation with about one quarter of toxin molecules escaping the uptake into endosomes and remaining exposed on the surface of cell membrane . This endocytic uptake of intact CyaA from cell membrane was noticeably slowed down in the presence of a five-fold excess of enzymatically inactive CyaA-AC− molecules that promoted Ca2+ influx into cells and K+ efflux in trans . Moreover , inhibition of the endocytic uptake through inhibition of ATP re-synthesis strongly enhanced the capacity of native Bp-CyaA toxin to permeabilize and lyze cells already at as low toxin concentrations as 200 ng/ml . This points towards a more important contribution of the pore-forming activity to the overall cytotoxic action of CyaA than previously recognized [8] , [9] . Enzymatically inactive but pore-forming CyaA-AC− toxoids have been extensively used over the past 18 years for delivery of AC-inserted CD8+ and CD4+ T cell epitopes from viruses , mycobacteria or tumors into the MHC class I and II-restricted antigen presentation pathways of CD11b-expressing dendritic cells [33]–[36] . As dCyaA-based vaccines for cancer immunotherapy are currently in phase I of clinical studies ( www . genticel . com ) , the here reported insight into dCyaA endocytosis and trafficking paves the way towards deciphering of the efficacy of T cell vaccine delivery by CyaA toxoids . Moreover , an endotoxin-free CyaA-AC− ( dCyaA ) toxoid was recently observed to alter the expression levels of a dozen of genes involved in innate immune response signaling in bone marrow-derived macrophages [37] . This is likely due to the capacity of the toxoid to promote Ca2+ influx into cells and permeabilize cells for K+ efflux . A long-sought evidence for a role of the pore-forming capacity of CyaA in Bordetella pertussis infection has , indeed , been recently reported by Dunne and coworkers [29] . This study showed that by eliciting K+ efflux from dendritic cells , and perhaps some other CD11b-expressing phagocytes , the pore-forming activity of CyaA contributes to activation of the NALP3 inflammasome and thereby to induction of innate IL-1β response , which supports the clearance of Bordetella bacteria at later stages of infection . The results reported herein show that the capacity of CyaA to permeabilize cells for K+ efflux depends on the capacity of the toxin to promote Ca2+ influx into cells and escape the rapid macropinocytic removal from target cell membrane . This reveals a further layer of sophistication of CyaA action on host cells , thus underpinning the key role of this toxin with multiple ‘talents’ in the virulence of Bordetellae in mammals .
Mouse monoclonal antibody ( MAb ) anti-CD71 ( clone MEM-189 ) was from Exbio ( Vestec , Czech Republic ) , anti-NTAL MAb ( clone NAP-08 ) was a generous gift from Pavla Angelisova ( Institute of Molecular Genetics , Prague , Czech Republic ) and anti-CyaA MAbs clone 9D4 and 3D1 were kindly provided by Erik L . Hewlett ( University of Virginia School of Medicine , Charlottesville , USA ) . Fura-2/AM , PBFI/AM , Alexa Fluor 488 , LysoTracker-red DND-99 , DAPI ( 4′-6-diamidino-2-phenylindole ) and Mowiol were purchased from Molecular Probes ( Eugene , OR ) . Transferrin coupled to Dyomics 547 , BSA-Dyomics 547 and mouse monoclonal antibody against α-tubulin ( TU-01 , IgG1 isotype control ) were from Exbio ( Vestec , Czech Republic ) . Ionomycin , valinomycin , nigericin , wortmannin , chlorpromazine , dynasore , 2-deoxy-D-glucose , sodium azide , chloroquine and Pluronic F-127 were purchased from Sigma ( St . Louis , MO ) . Complete Mini protease inhibitors cocktail was purchased from Roche ( Basel , Switzerland ) . NHS-Sulfo-LC-Biotin was purchased from Pierce ( Rockford , IL , USA ) . The SIINFEKL peptide corresponding to the CD8+ T-cell epitope encompassing the chicken ovalbumin ( OVA ) residues 257–264 and to the CD4+ T-cell epitope NGKLIAYPIAVEALS peptide corresponding to the Escherichia coli MalE protein ( residues 100–114 [18] ) , respectively , were purchased from Neosystem ( Strasbourg , France ) . Construction of dCyaA was described earlier [18] . dCyaA-OVA258–276 , harboring the VQLTGLEQLESIINFEKLTEWTSS NVMEERKIKVYLPRIVH peptide from hen egg ovalbumin protein ( OVA , residues 249–284 ) and carrying the MHC Class II sequence IINFEKLTEWTSSNVMEER ( OVA258–276 ) was constructed according to the standard protocols by insertion of a corresponding double-stranded synthetic oligonucleotide between codons 107 and 108 of the cyaA gene ( Holubova et al . , Infect . Immun . doi:10 . 1128/IAI . 05711-11 , published online ahead of print on 3 January 2012 ) . The corresponding CyaA protein variants , harboring the E570K and E581P substitutions , dCyaA-KP and dCyaA-OVA258–276-KP were constructed by recombination with previously described plasmid constructs [18] , [38] . Toxoid constructs were genetically detoxified by insertion of the dipeptide sequence GlySer between residues 188 and 189 , thus disrupting the ATP binding site of the enzyme [39] . CyaA derivatives were produced using E . coli strain XL1-Blue ( Stratagene , La Jolla , CA ) in the presence of CyaC acyltransferase produced from the same plasmid and were purified by chromatography on DEAE-Sepharose and Phenyl-Sepharose as described earlier [40] . Enzymatically active CyaA ( rEc-CyaA or CyaA ) produced in E . coli XL1-Blue and Bp-CyaA produced in B . pertussis strain 18323/ pHSP9 [28] were purified from urea extracts by combination of chromatography on DEAE-Sepharose and calmodulin agarose . All experiments were repeated with proteins from at least two independent preparations . J774A . 1 macrophages ( 3×105 ) were incubated in D-MEM with 200 ng/ml of CyaA-biotin for 5 min at 37°C , prior to removal of unbound toxin by three washes in cold D-MEM medium . Cells were lyzed with 0 . 1% Triton X-100 for determination of cell-bound AC activity . Toxin-induced lysis of J774A . 1 cells was determined using the CytoTox 96 kit assay ( Promega , Madison , USA ) as the amount of lactate dehydrogenase released from 2×105 cells in 2 hours of incubation with CyaA at 37°C in D-MEM . 105 J774A . 1 cells were incubated with different concentrations of the CyaA derived constructs for 30 minutes in D-MEM medium without FCS ( fetal calf serum , Life Technologies , Gaithersburg , USA ) . The reaction was stopped by addition of 0 . 2% Tween-20 in 50 mM HCl and samples were boiled for 15 min to denature cellular proteins . The samples were neutralized by addition of 150 mM unbuffered imidazol and concentration of cAMP was determined by a competition imunoassay performed as previously described [8] . Toxoids were labeled with the amine-reactive Alexa Fluor 488 or Dyomics 647 dyes upon binding to Phenyl-Sepharose during the last purification step in 0 . 1 M NaHCO3 pH 8 . 3 at room temperature for 1 hour . Unreacted dye was washed-out from the resin with 50 mM Tris-HCl buffer ( pH 8 . 0 ) and labeled toxoids were eluted from Phenyl-Sepharose in TUE buffer ( Tris 50 mM , 8M Urea , 2 mM EDTA , pH 8 . 0 ) . The molar ratio of protein∶dye was approximately 1∶4 for all toxoid preparations . On-column biotinylation of rEc-CyaA toxin was performed after the DEAE-Sepharose purification step using NHS-Sulfo-LC-Biotin ( Pierce , Rockford , IL , USA ) to reach a biotin∶toxin molar ratio of 20∶1 . Biotin coupling was at room temperature and was stopped after 40 min by washing of the resin with 15 ml of 50 mM Tris-HCl , pH 8 . 0 , and then extensively with PBS ( phosphate-buffered saline ) . Purified biotinylated toxin was then eluted with 50 mM HEPES , 8 M urea , and 2 mM EDTA . J774A . 1 cells ( ATCC TIB 67 ) were maintained in RPMI 1640 medium containing 10% FCS and antibiotic/antimycotic solution ( 0 . 1 mg/ml streptomycin , 1000 U/ml penicillin and 0 . 25 µg/ml amphotericin ( Sigma , St . Louis , MO ) ) . For all experiments the RPMI 1640 medium used for cell cultivation was replaced by Dulbecco's modified Eagle's medium ( D-MEM ) containing 1 . 9 mM Ca2+ without FCS , to avoid uncontrollable chelation of calcium ions by the phosphate ions contained in RPMI 1640 medium , as calcium is required for CyaA activity . Bone marrow derived dendritic cells ( BMDCs ) from MHC Class II/EGFP knock-in mouse [41] were flushed from the femur bone marrow cavity with PBS/2 . 5% FCS , plated at approx . 106 cells/well in 2 ml of DMEM/25 mM HEPES/10% FCS without phenol red but containing 5 ng/ml GM-CSF ( Sigma , St . Louis , MO ) . Cells were cultured on 25-mm circular cover slips , with media replacement every second day . Non-adherent cells were removed by gentle washing on days 2 and 4 . BMDCs from conventional 6- to 8-week-old female C57BL/6 mice were obtained from femoral and tibial bones and cells were cultured in RPMI 1640 medium supplemented with 10% FCS , 20 ng/ml GM-CSF and antibiotic/antimycotic solution for 7 days , as previously described by [42] . RAW 264 . 7 ( ATCC TIB 71 ) cells ( 5×104 ) grown on coverslips were maintained in RPMI medium supplemented with 10% FCS and transfected with pEGFP constructs bearing cDNA encoding Eps-15 ( DIII ) or Rab-7 ( kind gift of J . Forstova , Charles University , Prague ) using a FuGENE-6 transfection reagent ( Roche ) . Detergent resistant membranes ( coalesced lipid rafts ) were separated by flotation in discontinuous sucrose density gradients . J774A . 1 cells ( 2×107 ) were washed with prewarmed DMEM and incubated with 500 ng of CyaA-derived proteins at 37°C for 10 min . Cells were washed with ice-cold PBS , scraped from the Petri dish and extracted at 4°C in 200 µl of TBS buffer ( 20 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl ) containing 1% Triton X-100 , 1 mM EDTA , 10 mM NaF and a Complete Mini protease inhibitor cocktail ( Roche ) for 60 min . The lyzate was clarified by centrifugation at 250× g for 5 min and the post-nuclear supernatant was mixed with equal volume of 90% sucrose in TBS . The suspension was placed at the bottom of a centrifuge tube and overlaid with 2 . 5 ml of 30% sucrose and 1 . 5 ml of 5% sucrose in TBS . Buoyant density centrifugation was performed at 150 , 000× g in a Beckman SW60Ti rotor for 16 h at 4°C . Fractions of 0 . 5 ml were removed from the top of the gradient . For immunodetection , the separated proteins were transferred onto Immobilon-P membrane , blocked with 5% non-fat milk powder in TBST buffer ( 20 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 0 . 05% Tween-20 ) and probed with the indicated mouse monoclonal antibody . CyaA toxoids were recognized in Western blots by the 9D4 antibody binding to the C-terminal RTX repeats of CyaA . The signal was developed using a secondary peroxidase-conjugated sheep anti-mouse IgG and chemiluminescent detection system ( SuperSignal West Femto Maximum Sensitivity Substrate chemiluminescence reagent kit , Pierce , Rockford , IL ) . J774A . 1 mouse monocytes were grown on glass coverslips to subconfluence in the absence of the pH indicator ( to avoid cellular autofluorescence ) . Cells were incubated with Alexa Fluor 488-labeled toxoid variants in serum free D-MEM medium at 37°C in the presence of 10 µg/ml transferrin-Dyomics 547 or in the presence of 50 µg/ml BSA-Dyomics 547 ( Exbio , Czech Republic ) . For in vivo imaging , J774A . 1 cells grown on glass bottom microwell dishes ( MatTek , USA ) were incubated in the presence of labeled toxoids at 37°C in HBSS medium alone , or in the presence of inhibitors . BMDCs from day 5 were incubated for 2 hours with 1 µg/ml of Dyomics-647-labeled toxoid variants at 37°C in DMEM without serum . For simultaneous visualization of early and recycling endocytic compartments , DCs were co-incubated with 25 µg/ml transferrin-Dyomics-547 for the last 30 minutes . Cells were fixed ( 4% PFA in PBS ) and mounted in Mowiol . Images were captured using a Leica confocal microscope TCS SP2 ( Wetzlar , Germany ) or a CellR Imaging Station ( Olympus , Hamburg , Germany ) based on Olympus IX 81 fluorescence microscope . For colocalization analysis the 3D stack of desired colour channels was acquired comprising most of the cell volume ( usually 10–15 planes in the z-axis ) . Analysis was performed using a macro in WCIF ImageJ software . Special care was taken of the pixel shift between individual colour channels ( calibration with fluorescent beads ) . The threshold levels for each image and channel was found using “Huang dark” method . Individual cells were selected ( as ROI ) and colocalization analysis was performed using an ImageJ plug-in ( http://www . uhnresearch . ca /facilities/wcif/imagej/colour_analysis . htm ) . Pearson's correlation coefficients for each image ( individual cell recorded in all z-axis planes ) were calculated for the given channels . Usually about 60 cells were analyzed and average Pearson's correlation coefficient ( P . c . c . ) of all z-axis planes ± standard deviation was calculated . A value of 1 represents perfect correlation , zero represents random localization . For this purpose a script based on WCIF ImageJ software was used ( see Figure S2 for detailed description ) . Calcium influx into cells was measured as previously described [14] . Briefly , the J774A . 1 cells were loaded with 3 µM Fura-2/AM ( Molecular Probes ) at 25°C for 30 min and time course of calcium entry into cells after addition of CyaA-derived proteins was determined as ratio of fluorescence excited at 340/380 nm . Fluorescence measurement of cytosolic K+ was performed as described before [6] . Briefly , J774A . 1 cells were loaded with 9 . 5 µM PBFI/AM ( Molecular Probes ) for 30 min at 25°C in the presence of 0 , 05% ( w/w ) Pluronic F-127 ( Sigma-Aldrich ) in the dark . Fluorescence intensity of PBFI ( excitation wavelength 340 , emission wavelengths 450 and 510 nm ) was recorded , ratio of these intensities are shown in the graphs . Calibration experiments were performed in solutions containing 50 mM HEPES ( pH 7 . 2 ) , with varying concentrations of potassium acetate ( 10 , 30 , 60 , or 140 mM ) and sodium acetate ( 5 , 85 , 115 , or 135 mM ) . Cellular plasma membrane was permeabilized for potassium ions and protons by valinomycin and nigericin ( 3 µM; Sigma-Aldrich ) for 30 min . The H-2b-restricted T cell hybridoma CRMC3 , recognizing the NGKLIAYPIAVEALS epitope of the MalE protein from E . coli ( MalE100–114 , abbreviated MalE ) , and the I-Ab-restricted T cell hybridoma MF2 . 2D9 , recognizing the IINFEKLTEWTSSNVMEER epitope of ovalbumin ( OVA258–276 ) , respectively , were used . BMDC ( 105 cells/well ) were pulsed with proteins or peptides for 4–5 hours , followed by medium disposal and cultivation with CRMC3 hybridoma ( 105 cells/well ) for 18 hours or the MF2 . 2D9 hybridoma ( 5×104 ) for 14 hours , respectively . Prior to CRMC3 hybridoma addition the BMDCs were washed with PBS . After cell co-incubation , cultures were frozen for at least 2 hours at −80°C . T cell stimulation was monitored by determination of IL-2 amounts released into culture supernatants using two alternative methods . CRMC3 culture supernatants ( 100 µl ) were added to cultures of the IL-2-dependent CTL-L cell line ( 104 cells/well , final volume 200 µl ) for 48 hours , followed by addition of [3H]thymidine ( 50 µCi/ml; Perkin Elmer , Courtabeuf , France ) . Cells were harvested 6 hours later using an automated cell harvester ( Molecular Devices , Lier , Norway ) and the incorporated thymidine was determined by scintillation counting . The concentration of IL-2 released into MF2 . 2D9 cell culture supernatants was determined using a sandwich ELISA with noncompeting pairs of anti–IL-2 mAbs ( JES6-1A12 and biotinylated JES6-5H4 , both from BD Pharmingen , San Diego , CA , USA . | The adenylate cyclase toxin ( CyaA ) of pathogenic Bordetellae eliminates the first line of host innate immune defense by inhibiting the oxidative burst and complement-mediated opsonophagocytic killing of bacteria . The toxin penetrates myeloid phagocytes , such as neutrophil , macrophage or dendritic cells , and subverts their signaling by catalyzing a rapid and massive conversion of intracellular ATP to the key signaling molecule cAMP . In parallel , the toxin forms cation-selective pores and permeabilizes the cytoplasmic membrane of phagocytes . This so-called ‘hemolysin’ activity synergizes with the enzymatic AC activity of CyaA in promoting apoptotic or necrotic cell death , depending on the toxin dose . Moreover , the pore-forming activity promotes activation of NALP3 inflammasome and release of interleukin IL-1β . We show here that the capacity of CyaA to permeabilize phagocytes depends on its ability to mediate influx of extracellular calcium ions into cells . This enables bystander CyaA pores to escape rapid macropinocytic removal from cell membrane and exacerbate the permeabilization of cells . These observations set a new paradigm for the mechanism of action of pore-forming RTX leukotoxins on phagocytes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"molecular",
"cell",
"biology",
"proteins",
"biology",
"microbiology",
"bacterial",
"pathogens",
"toxicology"
]
| 2012 | Calcium Influx Rescues Adenylate Cyclase-Hemolysin from Rapid Cell Membrane Removal and Enables Phagocyte Permeabilization by Toxin Pores |
The network of native non-covalent residue contacts determines the three-dimensional structure of a protein . However , not all contacts are of equal structural significance , and little knowledge exists about a minimal , yet sufficient , subset required to define the global features of a protein . Characterisation of this “structural essence” has remained elusive so far: no algorithmic strategy has been devised to-date that could outperform a random selection in terms of 3D reconstruction accuracy ( measured as the Ca RMSD ) . It is not only of theoretical interest ( i . e . , for design of advanced statistical potentials ) to identify the number and nature of essential native contacts—such a subset of spatial constraints is very useful in a number of novel experimental methods ( like EPR ) which rely heavily on constraint-based protein modelling . To derive accurate three-dimensional models from distance constraints , we implemented a reconstruction pipeline using distance geometry . We selected a test-set of 12 protein structures from the four major SCOP fold classes and performed our reconstruction analysis . As a reference set , series of random subsets ( ranging from 10% to 90% of native contacts ) are generated for each protein , and the reconstruction accuracy is computed for each subset . We have developed a rational strategy , termed “cone-peeling” that combines sequence features and network descriptors to select minimal subsets that outperform the reference sets . We present , for the first time , a rational strategy to derive a structural essence of residue contacts and provide an estimate of the size of this minimal subset . Our algorithm computes sparse subsets capable of determining the tertiary structure at approximately 4 . 8 Å Ca RMSD with as little as 8% of the native contacts ( Ca-Ca and Cb-Cb ) . At the same time , a randomly chosen subset of native contacts needs about twice as many contacts to reach the same level of accuracy . This “structural essence” opens new avenues in the fields of structure prediction , empirical potentials and docking .
The native structure of a protein is held intact by the complex and cooperative interplay of residue interactions . While a network of amino acid contacts is well defined given a native structure , it remains an open question if all the contacts are equivalent in terms of their contribution to the structural integrity . In part , this question has been addressed by studies where a partial contact network of a native structure is embedded into the three-dimensional space [1] , [2] . The resultant root mean square deviation ( RMSD ) of the embedding from the native structure quantifies the information content of the selected subset ( Figure 1 ) . Such a line of investigation also represents a logical extension to the current trends in structural biology . While most of the three-dimensional structures of proteins in the PDB are identified by X-ray crystallography and NMR , new experimental methods like EPR aim to broaden the horizon of structural proteomics and cover the protein universe [3]–[5] . In spirit , these techniques are similar to established NMR spectroscopic methods as they yield information about inter-residue proximity constraints . From a sufficient number of such experimentally derived constraints , the tertiary structure of the protein can be identified [6] . Identifying a minimal set of structure determining distance constraints a-priori from the sequence would not only minimize the experimental efforts , but would in fact imply a solution to the protein folding problem . As an intermediate step in this direction , analysing minimal subsets of structure determining contacts in known structures promises to provide preliminary insights into the question what the features of such an essential subset of contacts might be . For several years , the distance constraints and other stereo chemical and biophysical restraints have been employed in computational restraint-based protein modelling [7]–[13] . Specifically , a selected subset of native contacts is considered as distance constraints that efficiently define the fold and reconstruct the tertiary structure of the protein [14] . In order to obtain coordinates consistent with a given set of distance constraints , we implemented a contact map reconstruction method based on distance geometry [15] . Using the complete set of native contacts of a known protein structure as input , the reconstruction provides models that are within 2 . 0 Å Ca RMSD from the native structure ( Jose M Duarte et al . unpublished data ) . The other existing implementations of 3D reconstruction from contact maps are based on methods such as Discrete Molecular Dynamics ( DMD ) [16] , singular value decomposition [2] , [17] . The reconstruction accuracies of the alternate reconstruction methods vary and different contact definitions and datasets make a direct comparison difficult . Despite the quantitative differences in reconstruction accuracy , we reproduce qualitatively the same non-linear relationship of Ca RMSD with fraction of native contacts . In all such studies , below a certain fraction of the native contacts , the reconstruction accuracy deteriorates rapidly . With our reconstruction pipeline we could go as low as 20–30% of the native contacts ( Ca-Ca , Cb-Cb ) and still obtain an average reconstruction accuracy of ∼4 Å Ca RMSD . In summary , we confirmed that a contact map is highly redundant and a subset of native contacts is sufficient to determine the structure up to experimental accuracy . Together with an accurate 3D reconstruction method and the knowledge that a complete contact map is not required for recognizing the protein fold , the central question is to predetermine the nature and the number of ‘minimal distance constraints’ required to efficiently identify the tertiary structure of the protein ( Figure 1 ) . The current paper focuses on the methods used to derive a minimal set of contacts , the necessary and sufficient determinants to reconstruct any given protein fold , namely the ‘structural essence of a protein’ . An independent study by Chen et al showed that randomly picked subset of contacts could be used to successfully reconstruct the three-dimensional structure of the protein [16] . They further claimed that subsets selected with a rational strategy could only reconstruct as good as the random subsets and not better . Here , for the first time we demonstrate that a structural essence exists and provide a constructive algorithm for its calculation . We also characterize the structural essence from different folds . The results of this study facilitate the choice of contacts to obtain better models from experimental and computational restrain-based protein modelling .
To verify that a subset of native contacts is sufficient to reconstruct the native structure , we chose increasing fractions ( from 10%–90% ) randomly from native contacts and measured their reconstruction accuracy ( as Ca RMSD compared to the native structure ) . The reconstruction accuracies obtained are provided in Figure 2A . The dataset chosen for the study is given in Table 1 . We find that in all the proteins from the dataset , the reconstruction with a fraction of native contacts yields structures close to the native structure . Specifically , the 30%–50% random subsets show reconstruction accuracies comparable to those obtained from the complete contact map . The negligible increase in Ca RMSD between the 20% and 30% subsets provides a direct estimate of the size of the minimal subset ( Figure 2A ) . The existence of the structural essence is observed as a common feature across different folds . However , the size of the subset varied with different SCOP classes as shown in Figure 2A . For proteins from the all α , all β and the α/β SCOP classes , as low as 20% of the native contacts are sufficient to obtain a structure within Ca RMSD of ≤4 Å to the native structure . However in the α+β class more contacts ( 30% ) are required for acquiring the same reconstruction accuracy ( Figure 2A ) highlighting its higher topological complexity . Furthermore , it is worthwhile to note that 20% of contacts are sufficient for reconstruction across a range of protein sizes between 100 and 300 amino acids . Thus , there is a negligible effect of protein size on the reconstruction accuracy with our reconstruction method . The performance comparison of our contact definition and our reconstruction pipeline with other existing methods of contact map reconstruction namely FT-COMAR [1] , DMD [16] reveals differences in the reconstruction accuracies and the size of the minimal subsets . In order to systematically compare the differences , we have taken the dataset from Chen et al ( henceforth called Chen-set ) and repeated the reconstructions with our method and contact definition ( for details see methods ) . The results are shown in the Figure 2B . We find that the overall profile of the reconstruction accuracy of different random subsets from the Chen-set did not vary considerably between our method and DMD . However , there are differences observed in the size of the minimal subset . In contrast to the 70% contacts required by DMD to reconstruct to ∼3 . 5 Å of the native structure , our method required just 20%–30% contacts to achieve a similar accuracy . Further , the reconstructions with our contact definitions ( Ca 9 . 0 Å , Cb 8 . 0 Å ) for a 30% subset yielded RMSD of 3 . 25 Å; whereas using Cb 7 . 5 Å , we get a RMSD of 5 . 03 Å showing the improved performance of our contact definition . In case of FT-COMAR , 25% of native distance restraints were required for reconstructing up to ∼4 Å of the native structure [1] . However , a large Ca distance threshold ( >15 Å ) was used to define the contacts . In comparison , we used a four-fold sparser contact map ( 20% contacts ) and achieved better accuracy ( ∼3 . 4 Å ) . Thus , in the trade-off between the reconstruction accuracy and the size of the subset required for achieving a given accuracy , we observe that our reconstruction method along with our contact definition outperform FT-COMAR and DMD . An algorithm capable of picking the structurally important contacts should be able to generate sets with significantly better reconstruction accuracy than by random selection . On the same token , such an algorithm should also require fewer distance restraints as input . To measure the improvement we define a relative performance index ( PI ) as ( 1 ) where the size of the random subset equals the rational subset . An algorithm capable of picking a minimal subset that reconstructs better than a random subset scores a PI>1 . The sequence based information in combination with graph-based properties can be used as parameters in devising a rational strategy that identifies the structural essence . The sequence-range of a contact is defined as the separation in sequence between the amino acids i and j which are in contact ( for details see methods ) . While contacts from the lower sequence-range are determinants of the secondary structure , the long-range contacts determine the intricacies of the fold and the packing of the tertiary structure . Further , the number of long-range contacts and the long-range contact order influence the folding rate of proteins [18]–[21] . To evaluate the significance of contacts from different sequence-ranges , we selected predefined short and long sequence-range contacts ( see methods ) . The reconstruction accuracies of the chosen subsets compared to similar sized random subset are shown in Figure 3 . While the short-range subsets failed to produce a model anywhere near the native structure , the long sequence-range subsets reconstructed significantly better . However , in comparison to the random subsets , these results are not significant and the long-range contacts alone did not achieve a PI>1 . Although there is an effect of long-range contacts being more important , a set of long-range contacts alone is not sufficient to capture all the structural information . The concept of common neighbourhoods is used to analyze the significance of an edge and its local neighbourhood to the overall structure and stability of the network . For instance , common neighbourhoods are used in determining packing effects of atoms in crystals [22] , [23] . The common neighbourhood ( CNb ) of a contact is defined in methods and the concept illustrated in Figure 4 . For an edge Eij ( red ) between nodes i ( pink ) and j ( green ) , the edges formed by i and j with nodes k1 , k2 and k3 ( yellow ) constitute the neighbourhood of Eij . The triangle formed by Eij and its neighbours ( Eik and Ekj ( black ) ) forms the CNb triangle . A contact ( red ) embedded in its CNb is viewed as a representative of its neighbourhoods ( black ) . A contact map typically contains many contacts that have few common neighbours and few with many common neighbours . Thus , it is possible to rank contacts based on their CNb sizes . We hypothesize here that contacts that possess more common neighbours are structurally more significant compared to the small neighbourhood counterparts . By stripping the neighbourhoods from the contacts , the ability of contacts to represent their neighbourhood efficiently is tested . A simple rank-ordered selection of contacts was the initial strategy we employed in selecting structurally important contacts . Native contacts were ranked in the ascending order according to the sequence-range , CNb sizes and increasing fractions ( 10% to 90% ) were selected and reconstructed . The PIs of the rank-ordered subsets are given in Table 2 ( the reconstruction accuracies are shown in Figure S1 ) . Even with structurally important parameters like the sequence-range and the CNb sizes , a direct rank-ordered selection failed to distinguish the structurally essential from the non-essential contacts of the protein structure . This is evident when the rank-ordered subsets are visualized in a contact map . The rank-ordering samples only discrete regions of the contact map , while a random selection samples uniformly from different regions of the contact map ensuring better reconstruction . For instance , in the case of the sequence-range ordering , contacts are selected diagonally and clearly carry insufficient information about the protein's tertiary structure ( Figure S2 ) . This provides a possible explanation of why contact order ranked selection did not yield a better reconstructing subset for Chen and co-workers [16] . Thus , it is clear that even with a choice of parameters like sequence-range and CNbs that carry significant structural information , the method employed in selecting the best reconstructing subset can be considered as the biggest bottle-neck . Such a method should show better performance when the two parameters are combined in a most efficient way . The CNb sizes of contacts and the sequence-range are effectively combined with other network descriptors like degree in formulating the cone-peeling algorithm . The cone-peeling algorithm is based on the concept of common neighbourhood of edges . The CNb of an edge is defined in Eq . 3 and the concept explained in Figure 4 . For any given edge Eij , a CNb triangle can be defined with edges Eik1 and Ek1j . Here , we hypothesize that in every neighbourhood triangle if the edges Eik1 and Ek1j are redundant then every triangle can be reduced to just the edge Eij on some conditions . For instance , if Eik1 or Ek1j are low sequence-range edges , then Eij can successfully represent Eik1 and Ek1j and a single edge successfully represents the triangle . Thus , Eij is called the representative edge in its CNb triangle . This is meaningful when visualized in the context of the three-dimensional structure of proteins . Assuming Eij is present in a regular secondary structure such as an alpha-helix , the low sequence-range edges Eik1 or Ek1j would also be part of the same helix . Thus the presence of the representative edge Eij is sufficient and the edges Eik1 and Ek1j can be safely deleted . For the sake of illustration , the CNb triangles in contact maps can be visualized in 3D as occupying the base of a cone while the representative edges occupying the summit ( Figure 5A ) . The height of the cone is defined by the CNb size of the representative edge . In such a scenario , the algorithm peels the cone by deleting local contacts retaining only the summits . This is performed iteratively and every CNbs is replaced with its representative edge in the decreasing order of their degree and the common neighbour sizes of contacts . Thus , the strategy of retaining only higher neighbourhood long-range edges and deleting the low sequence-range neighbours has been implemented in the cone-peeling algorithm . A step-by-step implementation of the cone-peeling algorithm can be obtained from the pseudo code in the methods section . The long sequence-range and high CNb edges which emerge after cone-peeling is subjected to reconstruction and the accuracies compared with the random subsets in Figure 5B . Our ‘cone-peeled’ subsets from all the SCOP classes exhibit a PI of >1 . 5 ( Table 1 ) . Thus , our ‘cone-peeling’ of local contacts has filtered out the non-essential contacts , while retaining only the essential or structure-determining contacts . It is surprising to note that the minimal subsets of contacts selected from our approach are significantly sparse , on an average comprising about ∼5 . 8% of Ca and 11 . 1% of Cb contacts . The cone-peeled subset of contacts for CheY protein ( 1e6k ) is highlighted in Figure 5C . It can be seen that the structural essence as characterized by our algorithm has picked mostly the inter-secondary structural contacts and the contacts from loop regions that are crucial for packing in the protein core , while the ignoring intra-secondary structural contacts and the contacts on the surface . The overlay of five best reconstructed models of CheY ( 1e6k ) onto the native structure is shown in Figure 5D . It can be seen that the secondary structures and the inter-secondary structural regions are distinguished even with a sparse set of Ca and Cb contacts . With as little as 8% of native contacts ( Ca-Ca and Cb-Cb ) , our algorithm along with our reconstruction pipeline determines the structure of a protein at 4 . 8 Å ( Ca RMSD ) . At the same time , from a random selection of contacts , roughly twice the number of contacts is necessary to achieve such reconstruction accuracy . Thus , for the first time we report a method that successfully selects native contacts that determine the structure better than a random selection .
We have identified a structural essence from the non-covalent contacts of protein , which successfully determines structural features . The essence could be identified as a 20%–30% fraction of native contacts by random selection . We have proposed a rational strategy ( cone-peeling ) that outperforms a random contact selection and it successfully distilled the structural essence of a protein from the bulk of non-covalent contacts . The cone-peeling combines the sequence and network descriptors to select the essential contacts . The structural essence is only 8% of the native contacts that reconstructs to 4 . 8 Å ( Ca RMSD ) to the native structure . However , to attain a similar reconstruction accuracy with random selection about twice the number of contacts is required . Thus , our cone-peeling algorithm is the first rational strategy that characterizes the structural essence in protein structures . The concept of essential contacts in proteins can find further applications in the design of empirical contact potentials , in experimental and theoretical protein structure determination and also in constraint-based comparative sequence design .
A non-redundant dataset of proteins is selected from SCOP release 1 . 73 [27] . Only monomeric , monodomain proteins from the four main SCOP classes and from high populated folds are chosen such that all possible interactions that stabilize the native fold are taken into account . All proteins have resolutions better than 3 . 0 Å , R-factor lower than 0 . 3 as well as no missing or ambiguous conformational data ( Filippis , personal communication ) . A subset of 12 proteins , three per SCOP class , is selected from the dataset such that two fall in the size range of 100–120 amino acids and the third is thrice bigger . The PDB codes of the selected proteins are given in Table 1 . The protein structures are represented as graphs with amino acids as nodes and the interactions between the amino acids as edges . Specifically , the contacts between the Ca or Cb atoms are considered as edges . The distance thresholds of 9 . 0 Å and 8 . 0 Å are used respectively to define contacts between Ca and Cb atoms ( Ca for Gly ) . The contacts are visualized in a contact map with CMView [CMView: Interactive Contact Map Visualization and Analysis . http://www . molgen . mpg . de/~lappe/cmview/] . All covalent contacts are ignored . The Ca and the Cb contacts are passed as restraints to the distance geometry program ( distgeom ) of the Tinker molecular dynamics package [28] . The distgeom uses a variation of the EMBED algorithm [15] to find three-dimensional coordinates in agreement with a sparse set of distance restraints . It proceeds by calculating bounds for all pairs of atoms ( bounds smoothing ) , choosing particular distance values from within the bounds ( metrization ) and then embedding the resulting metric matrix . A final regularization step is performed by which the coordinates obtained are transformed so that their geometry , with respect to bond lengths and angles , is improved . For this purpose , we used the simulated annealing protocol offered by the distgeom program that minimizes an error function that measures the violations to the restraints . An ensemble of 50 models is generated for every protein in the dataset . Even after enforcing individual amino acids to the L-enantiomer in the refinement from simulated annealing , a solution to the given contact map can still be found such that the fold is ‘mirrored’ . These solutions are termed as ‘topological mirrors’ where the global fold possesses the wrong chirality in spite of individual amino acids being in the L-form . The conformations obtained from the distance geometry protocol cannot distinguish such topological mirrors and we overcome this problem by comparing the models with their native structure through Ca RMSD . The Ca RMSD values for the conformation ensemble are found to be distributed bimodally , by simply choosing the lowest fourth of models as ranked by RMSD we are sure to be selecting the correct models ( Figure S3 ) . The ensemble average is obtained from the fourth of models with the lowest Ca RMSD . We have considered the dataset of 5 proteins ( 1dd3 , 1nxb , 1igd , 1bxy , 1d0d ) from Chen and co-workers [12] . For every protein , contacts maps ( Ca 9 . 0 Å , Cb 8 . 0 Å ) were generated and reconstructed with Tinker . However , contact maps were also generated with Cb 7 . 5 Å , to compare the differences in reconstruction between Tinker and DMD . The covalent contacts are ignored and the ensemble average Ca RMSD is obtained as mentioned earlier . The properties of the contacts employed in selecting a minimal subset of a given contact map are discussed below . The algorithm employed in selecting subsets based on peeling of CNbs of contacts is shown in Figure 6 . | A protein structure can be visualized as a network of non-covalent contacts existing between amino acids . But not all such contacts are important structural determinants of a protein . We have attempted to identify a subset of amino acid contacts that are essential for reconstructing protein structures . Initially , we followed random sampling of contacts and tested their efficacy to successfully represent the three-dimensional structure . Further , we also developed an algorithm that selects a subset of amino acid contacts from proteins based on the sequence and network properties . The subsets picked by our algorithm represent protein three-dimensional structure better than random subsets , thereby offering direct evidence for the existence of a structural essence in protein structures . The identification of such structure-defining subsets finds application in experimental and computational protein structure determination . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"computational",
"biology/protein",
"structure",
"prediction",
"computational",
"biology/macromolecular",
"structure",
"analysis",
"computational",
"biology/protein",
"homology",
"detection",
"computational",
"biology"
]
| 2009 | Defining an Essence of Structure Determining Residue Contacts in Proteins |
Several brain diseases are characterized by abnormally strong neuronal synchrony . Coordinated Reset ( CR ) stimulation was computationally designed to specifically counteract abnormal neuronal synchronization processes by desynchronization . In the presence of spike-timing-dependent plasticity ( STDP ) this may lead to a decrease of synaptic excitatory weights and ultimately to an anti-kindling , i . e . unlearning of abnormal synaptic connectivity and abnormal neuronal synchrony . The long-lasting desynchronizing impact of CR stimulation has been verified in pre-clinical and clinical proof of concept studies . However , as yet it is unclear how to optimally choose the CR stimulation frequency , i . e . the repetition rate at which the CR stimuli are delivered . This work presents the first computational study on the dependence of the acute and long-term outcome on the CR stimulation frequency in neuronal networks with STDP . For this purpose , CR stimulation was applied with Rapidly Varying Sequences ( RVS ) as well as with Slowly Varying Sequences ( SVS ) in a wide range of stimulation frequencies and intensities . Our findings demonstrate that acute desynchronization , achieved during stimulation , does not necessarily lead to long-term desynchronization after cessation of stimulation . By comparing the long-term effects of the two different CR protocols , the RVS CR stimulation turned out to be more robust against variations of the stimulation frequency . However , SVS CR stimulation can obtain stronger anti-kindling effects . We revealed specific parameter ranges that are favorable for long-term desynchronization . For instance , RVS CR stimulation at weak intensities and with stimulation frequencies in the range of the neuronal firing rates turned out to be effective and robust , in particular , if no closed loop adaptation of stimulation parameters is ( technically ) available . From a clinical standpoint , this may be relevant in the context of both invasive as well as non-invasive CR stimulation .
Synchronization of oscillations is a generic mechanism in animate and inanimate systems [1–6] . In fact , oscillators of qualitatively different type may share fundamental synchronization mechanisms [1–6] . Synchronization processes may occur within as well as between different systems of the human body [3–6] , e . g . between heartbeat intervals and respiratory cycles [3] . Neuronal synchronization processes are relevant under normal as well as abnormal conditions [7] . A number of brain disorders are associated with abnormal neuronal synchrony , for example Parkinson’s disease [8–10] , tinnitus [11–15] and epilepsy [16–18] . Neuronal dynamics and , in particular synchronization processes crucially depend on the patterns and types of neuronal connections [19–21] . For instance , according to computational studies it makes a significant difference whether neurons interact through gap-junctions or synapses [20 , 21] . This is relevant for the emergence of different kinds of synchronization patterns [20–22] and epileptic seizures [23] . Connectivity and function are strongly connected and may undergo plastic changes throughout the life course [24] . The timing pattern of neuronal activity may strongly determine the strength of neuronal connections [25 , 26] . Spike-timing-dependent plasticity ( STDP ) is a pivotal mechanism by which neurons adapt the strength of their synapses to the relative timing of their action potentials [27–31] . Based on seminal experimental studies [28–30] a series of computational studies focused on how adaptive coupling and activity dependent synaptic strength influence the collective neuronal dynamics [21 , 23 , 32–42] . In the presence of STDP a plethora of qualitatively different stable dynamical regimes emerge [21 , 34 , 42] . Qualitatively different stable dynamical states may actually coexist . In fact , multistability is a typical feature of neuronal networks and oscillator networks equipped with STDP . Multistability was found in different neural network models comprising different STDP models , e . g . in phase oscillator networks with both symmetric and asymmetric phase difference-dependent plasticity , a time continuous approximation of STDP [32 , 34] as well as in phase oscillator networks with STDP [33] and in different types of neuronal networks with STDP [43–46] and other types of neural network models ( e . g . [47–56] and references therein ) . A number of computational studies were dedicated on desynchronizing synchronized ensembles or networks of oscillators or neurons [57–62] . The clinical need for stimulation techniques that cause desynchronization irrespective of the network’s initial state [63] , thereby being reasonably robust against variations of system parameters and , hence , not requiring time-consuming calibration , motivated the design of Coordinated Reset ( CR ) stimulation [64 , 65] . CR stimuli aim at disrupting in-phase synchronized neuronal populations by delivering phase resetting stimuli typically equidistantly in time , separated by time differences Ts/Ns , where Ts is the duration of a stimulation cycle , and Ns is the number of active stimulation sites [64 , 65] . The spatiotemporal sequence by which all stimulation sites are activated exactly once in a CR stimulation cycle is called the stimulation site sequence , or briefly sequence . Taking into account STDP [27–30] in oscillatory neural networks qualitatively changed the scope of the desynchronization approach: Computationally , it was shown that CR stimulation reduces the rate of coincident firing and , mediated by STDP , also decreases the average synaptic weight , ultimately preventing the network from generating abnormally increased synchrony [33] . This anti-kindling , i . e . , unlearning of abnormally strong synaptic connectivity and of excessive neuronal synchrony , causes long-lasting sustained effects that persist cessation of stimulation [33 , 43–45 , 66 , 67] . As shown computationally , anti-kindling can robustly be achieved in networks with plastic excitatory and inhibitory synapses , no matter whether CR stimulation is administered directly to the soma or through synapses [45 , 68] . In line with these computational findings , long-lasting CR-induced desynchronization and/or therapeutic effects were accomplished with invasive as well as non-invasive stimulation modalities . Long-lasting desynchronization was induced by electrical CR stimulation in rat hippocampal slices rendered epileptic by magnesium withdrawal [69] . Electrical CR deep brain stimulation ( DBS ) caused long-lasting therapeutic after-effects in parkinsonian non-human primates [70 , 71] . Bilateral therapeutic after-effects for at least 30 days were caused by unilateral CR stimulation delivered to the subthalamic nucleus ( STN ) of parkinsonian MPTP monkeys for only 2 h per day during 5 consecutive days [70] . In contrast , standard permanent high-frequency deep brain stimulation did not induce any sustained after-effects [70] , see also [72] . In patients with Parkinson’s disease electrical CR-DBS delivered to the STN caused a significant and cumulative reduction of abnormal beta band oscillations along with a significant improvement of motor function [73] . Non-invasive , acoustic CR stimulation was developed for the treatment of patients suffering from chronic subjective tinnitus [68 , 74] . In a proof of concept-study acoustic CR stimulation caused a statistically and clinically significant sustained reduction of tinnitus symptoms [74–76] together with a concomitant decrease of abnormal neuronal synchrony [74 , 77] , abnormal effective connectivity [78] as well as abnormal cross-frequency coupling [79] within a tinnitus-related network of brain areas . So far , the pre-clinical [74 , 80] and clinical [70 , 73] proof of concept studies for invasive and non-invasive CR stimulation were driven by computationally derived hypotheses and predictions . Theoretically predicted phenomena and mechanisms , such as long-lasting stimulation effects [33 , 43 , 45 , 66] , cumulative stimulation effects [67] , and improvement by weak stimulus intensity [81] were verified based on dedicated theory-driven study protocols for pre-clinical and clinical proof of concepts [70 , 73 , 74 , 80] . We here set out to investigate the impact of the CR stimulation frequency and intensity on the effects during stimulus delivery ( so-called acute effects ) , on transient effects emerging directly after cessation of stimulation ( so-called acute after-effects ) , and on effects outlasting cessation of stimulation ( so-called sustained after-effects ) . The ultimate goal of this study is to improve the calibration of CR stimulation , in particular , by providing computationally generated predictions that can be tested in subsequent pre-clinical and clinical studies . The computational study presented here is organized around three hypotheses: Hypothesis #1: Due to the inherently periodic structure of CR stimulation the relation between CR stimulation frequency and the spontaneous neuronal firing rates ( prior to stimulation ) matters . Periodic delivery of CR stimuli with fixed sequence basically constitutes a time-shifted entrainment of the different neuronal subpopulations [64 , 65] . A particular closed loop embodiment of CR stimulation , periodic stimulation with demand-controlled length of high-frequency pulse train , is basically a time-shifted entrainment of the different neuronal subpopulations with stimulus intensities adapted to the amount of undesired synchrony [64 , 65] . Accordingly , the duration of a stimulation cycle was selected to be reasonably close to the mean period of the synchronized neuronal oscillation [64 , 65] . In STDP-free networks of Kuramoto [82] and FitzHugh-Nagumo [83] model neurons the impact of CR stimulation intensity and frequency on the desynchronizing outcome of CR was studied in detail [81] . Hypothesis #2: Different embodiments of CR stimulation may differ with respect to effect size and robustness . In a series of computational studies [45 , 46 , 64 , 65 , 81 , 84 , 85] and in all pre-clinical [69 , 70] and clinical studies [73–78] performed so far , CR was applied either with fixed sequences or rapidly varying sequences ( RVS ) , where the sequence was randomly varied from cycle to cycle . In a recent computational study , it was shown that at intermediate stimulation intensities the CR-induced anti-kindling effect may significantly be improved by CR with slowly varying sequences ( SVS ) , i . e . by appropriate repetition of the sequence with occasional random switching to the next sequence [84] . However , this study was not performed for a larger range of CR stimulation frequencies . By definition , SVS CR stimulation features significantly more periodicity of the stimulus pattern . Accordingly , the dependence of resonance and/or anti-resonance effects on the CR stimulation frequency might be more pronounced for SVS CR as opposed to RVS CR . Hypothesis #3: Pronounced acute effects might provide a necessary , but not sufficient condition for pronounced sustained after-effects . In a pre-clinical study in Parkinsonian monkeys with CR-DBS delivered at an optimal and a less favorable intensity , it was shown that long and pronounced acute therapeutic after-effects coincide with long-lasting , sustained after-effects [74] . However , according to computational studies the relationship between acute after-effects and sustained long-lasting effects might be more involved , at least for particular parameter combinations [84] . Related to these hypotheses , to assess the robustness of CR stimulation against initial network conditions we performed our numerical simulations for different network initializations , respectively . In this study we did not systematically vary the stimulation duration . Rather , based on a pre-series of numerical simulations , we here used a fixed stimulation duration that is reasonably short , but nevertheless enabled to robustly achieve an anti-kindling for properly selected values of stimulation frequency and intensity . In fact , our goal was to find stimulation parameters enabling short , but notwithstanding effective CR stimulation . Keeping the stimulation duration at moderate levels may be beneficial for applying the CR approach to different invasive as well as non-invasive stimulation modalities . For instance , standard DBS , i . e . permanent electrical high-frequency pulse train stimulation delivered to dedicated target areas through implanted depth electrodes , used for the treatment of , e . g . Parkinson’s disease [86–88] may cause side effects . If side effects are caused by stimulation of non-target tissue , they may be reduced by adapting the spatial extent of the current spread to the target’s anatomical borders by appropriate electrode designs as introduced , e . g . by [89–91] , in particular , to spatially tailor stimuli by means of directional DBS [92–97] . However , some side effects may at least partly be caused by stimulating the target region itself [98 , 99] . Accordingly , no matter how precisely stimuli are delivered to DBS targets , the amount of stimulation should be decreased as much as possible . Another example refers to non-invasive applications of CR . In general , non-invasive CR stimulation requires the patients’ compliance to actually pursue treatment prescriptions . Obviously , patients might prefer shorter treatment sessions . To come up with favorable combinations of stimulation parameters , in our numerical analysis we used different data analysis methods , e . g . macroscopic measures assessing the average amount of the population’s synchrony and synaptic connectivity . These measures are appropriate to demonstrate relevant stimulation effects , such as stimulation-induced transitions from pronounced neuronal synchrony to desynchronized states . In summary , in this paper we first explain the computational model and analysis methods . We then apply RVS CR stimulation in a wide parameter range of stimulation frequencies and intensities . We repeat the same analysis for SVS CR stimulation and investigate the differential characteristics of RVS CR and SVS CR with respect to efficacy and robustness . Finally , we analyze the relationship between stimulation-induced acute effects and after-effects . Our results provide the foundation for the development of novel control techniques that will be the topic of a forthcoming study .
In this study we use the conductance-based Hodgkin-Huxley neuron model [100] for the description of an ensemble of spiking neurons . The set of equations and parameters are ( see also [101 , 102] ) : CdVidt=Ii−gNami3hi ( Vi−VNa ) −gKni4 ( Vi−VK ) −gl ( Vi−Vl ) +Si+Fi . dmidt=αm ( Vi ) ( 1−mi ) −βm ( Vi ) mi dhidt=αh ( Vi ) ( 1−hi ) −βh ( Vi ) hi ( 1 ) dnidt=αn ( Vi ) ( 1−ni ) −βn ( Vi ) ni The variable Vi , with i = 1 , … , N , describes the membrane potential of neuron i , and: αm ( V ) = ( 0 . 1V+4 ) /[1−exp ( −0 . 1V−4 ) ] , βm ( V ) =4exp[−V−6518] , αh ( V ) =0 . 07exp[−V−6520] , βh ( V ) =1[1+exp ( −0 . 1V−3 . 5 ) ] , αn ( V ) =0 . 01V+0 . 55[1−exp ( −0 . 1V−5 . 5 ) ] , βn ( V ) =0 . 125exp[−V−6580] . The total number of neurons is N = 200 , while gNa = 120 mS/cm2 , gK = 36 mS/cm2 , gl = 0 . 3 mS/cm2 are the maximum conductance per unit area for the sodium , potassium and leak currents respectively . The constants VNa = 50 mV , VK = −77 mV and Vl = −54 . 4 mV refer to the sodium , potassium and leak reversal potentials respectively . C is the constant membrane capacitance ( C = 1 μF/cm2 ) , Ii the constant depolarizing current injected into neuron i , determining the intrinsic firing rate of the uncoupled neurons . For the realization of different initial networks , we used random initial conditions drawn from uniform distributions , i . e . Ii ∈ [I0 − σI , I0 + σl] ( I0 = 11 . 0 μS/cm2 and σl = 0 . 45 μS/cm2 ) , hi , mi , ni ∈ [0 , 1] and Vi ∈ [−65 , 5] mV . The initial values of the neural synaptic weights cij are picked from a normal distribution N ( μc = 0 . 5 μA/cm2 , σc = 0 . 01 μA/cm2 ) ( see also [45 , 84] for more details ) . Hence , in this setup the neurons are not identical . The Si term refers to the internal synaptic input of the neurons within the network to neuron i , while Fi represents the current induced in neuron i by the external CR stimulation . The N = 200 spiking Hodgkin-Huxley neurons are placed on a ring and the Ns = 4 stimulations sites are equidistantly placed in space at the positions of neurons i = 25 , 75 , 125 , 175 . The neurons interact via excitatory and inhibitory chemical synapses by means of the postsynaptic potential ( PSP ) si which is triggered by a spike of neuron i [27 , 103] and modelled using an additional equation ( see also [104 , 105] ) : dsjdt=0 . 5 ( 1−sj ) 1+exp[− ( Vj+5 ) /12]−2sj . ( 2 ) Initially we draw si ∈ [0 , 1] ( randomly from a uniform distribution ) and then , the coupling term Si from Eq ( 1 ) ( see [45] ) contains a weighted ensemble average of all postsynaptic currents received by neuron i from the other neurons in the network and is described by the following term: Si=N−1∑j=1N ( Vr , j−Vi ) cij|Mij|sj , ( 3 ) where cij is the synaptic coupling strength from neuron j to neuron i and Vr , j is the reversal potential of the synaptic coupling ( 20 mV for excitatory and –40 mV for inhibitory coupling ) . In accordance with previous studies [45 , 84 , 85] the inhibitory reversal potential was set to −40 mV . The latter makes neurons’ more susceptible to input , e . g . stimuli . We performed the same sets of simulations for a subset of stimulation parameters with a different value of the inhibitory reversal potential , -80 mV as in [106] instead of -40 mV . In this way , we obtained very similar results . There are no neuronal self-connections within the network ( cii = 0 mS/cm2 ) . The term Mij , which describes the spatial profile of coupling between neurons i and j , is given by: Mij= ( 1−dij2/σ12 ) exp ( −dij2/ ( 2σ22 ) ) . ( 4 ) It has the form of a Mexican hat [107–109] and defines the strength and type of neuronal interaction: strong short-range excitatory ( Mij > 0 ) and weak long-range inhibitory interactions ( Mij < 0 ) . Here dij = d|i − j| is the distance between neurons i and j , while d=d0/ ( N−1 ) ( 5 ) determines the distance on the lattice between two neighboring neurons within the ensemble , d0 is the length of the neuronal chain ( d0 = 10 ) , σ1 = 3 . 5 , and σ2 = 2 . 0 . In order to limit boundary effects , we consider that the neurons are distributed in such a way that the distance dij is taken as: d ∙ min ( |i − j| , N − |i – j| ) when the i , j > N/2 . We follow the concepts described in [28 , 29] , regarding the synaptic coupling strengths change dependence on the precise timing of pre- and post-synaptic spikes . Hence , we consider all the synaptic weights cij to be dynamic variables that depend on the time difference ( Δtij ) between the onset of the spikes of the post-synaptic neuron i and pre- synaptic neuron j ( denoted by ti and tj respectively ) . Then the STDP rule for the change of synaptic weights is given by [29 , 45]: Δcij={β1e−Δtijγ1τ , Δtij≥0β2ΔtijτeΔtijγ2τ , Δtij<0 , ( 6 ) where β1 = 1 , β2 = 16 , γ1 = 0 . 12 , γ2 = 0 . 15 , τ = 14 ms and δ = 0 . 002 . According to the value of Δtij , the synaptic weight cij is updated in an event-like manner , i . e . we add or subtract an increment δ ∙ Δcij for excitatory or inhibitory connections respectively , with learning rate δ > 0 every time a neuron spikes . Furthermore , we restrict the values of cij on the interval [0 , 1] mS/cm2 for both excitatory and inhibitory synapses , ensuring in this way that their strengthening or weakening remains bounded . The maximal inhibitory synaptic weight cmax was set to be 1 in all our stimulations . However , a more detailed investigation about the effect and variation of this value was performed in [84] where when increasing cmax of the inhibitory neurons no significant impact was observed regarding ( de ) synchronization effects accompanied with a lower average network connectivity . The time window of the plasticity is adjusted with respect to the intrinsic firing rate of the neuron population in order to exhibit multistability , as also discussed in [45] . There , different time-windows ( via different choices of parameters ) were selected for the STDP for two different neuron models , i . e . one with bursting neurons ( FitzHugh-Rinzel ) and one for spiking neurons ( Hodgkin-Huxley ) . In our simulations , the STDP tends to simply stabilize the ongoing ensemble evolution and does not , by itself , ( de- ) synchronize the network . The parameters were , in general , chosen such that the ratio Δtijγ1 , 2τ is normalized , and the plasticity takes place within a time interval associated with the spiking period of the individual neurons . We analyzed two additional cases for small variation of the plasticity time-window ( τ = 12 and τ = 16 ) and obtained very similar general effects . The selected fixed value τ = 14 , used throughout the entire study , also allows us to compare our results with previously published studies . Coordinated Reset ( CR ) stimulation was applied to the neuronal ensemble of N spiking Hodgkin-Huxley neurons . This was done sequentially via Ns ( = 4 in this study ) equidistantly spaced stimulation sites [64]: one stimulation site was active during Ts/Ns , while the other stimulation sites were inactive during that period . After that another stimulation site was active during the next Ts/Ns period . All Ns stimulation sites were stimulated exactly once within one stimulation ON-cycle . Therefore , the duration of each ON-cycle is Ts ( in ms ) . The spatiotemporal activation of stimulation sites is represented by the indicator functions ρk ( t ) ( kϵ {1 , … , N} ) : ρk ( t ) ={1 , kthstimulationsiteisactiveatt0 , otherwise . ( 7 ) The stimulation signals induced single brief excitatory post-synaptic currents . The evoked time-dependent normalized conductances of the postsynaptic membranes are represented by α-functions given in [102]: Gstim ( t ) =t−tkτstime− ( t−tk ) /τstim , tk≤t≤tk+1 . ( 8 ) Here τstim = Ts/ ( 6Ns ) denotes the time-to-peak of Gstim , and tk is the onset of the kth activation of the stimulation site . Note that the period ( or frequency ) through the τstim parameter of the CR stimulation determines not only the random onset timing of each single signal but also its temporal duration . The spatial spread of the induced excitatory postsynaptic currents in the network is defined by a quadratic spatial decay profile ( see [102] for more details ) given as a function of the difference in index of neuron i and the index xk of the neuron at stimulation site k: D ( i , xk ) =11+d2 ( i−xk ) 2/σd2 , ( 9 ) with d the lattice distance between two neighboring neurons as defined in Eq ( 5 ) and σd = 0 . 8 the spatial decay rate of the stimulation current . Thus , the total stimulation current from Eq ( 1 ) is expressed by the following equations: Fi=[Vr−Vi ( t ) ]∙K∑k=1NsD ( i , xk ) ρk ( t ) Gstim ( t ) , ( 10 ) where Vr = 20 mV denotes the excitatory reversal potential , Vi the membrane potential of neuron i , K the stimulation intensity , and ρ , G , D are given by Eqs ( 7 ) , ( 8 ) and ( 9 ) respectively . For the RVS CR stimulation , sequences are randomly chosen from a set of Ns ! ( = 24 ) different possible sequences during each ON-cycle ( an example is shown in Fig 1A for CR stimulation period Ts = 10 ms for the first 90 ms of an activated CR period ) . Each newly drawn sequence is indicated by a different color and lasts exactly one ON cycle . On the other hand , for the SVS-l CR stimulation , one first randomly picks a sequence and repeats it l times before switching to another one , as shown by the example in Fig 1B ( again for Ts = 10 ms ) for l = 4 . The administered stimulation protocol consists of m:n = 3:2 CR ON-OFF cycles ( see [45 , 81 , 84] ) . Depending on the Ts value , more ( or less ) ON-cycles may be administered within a fixed time interval . In this panel , the total time spans up to two completed ON-and OFF cycles ( up to ~125 ms in this case ) and the color changes at each new sequence . The synaptic weights , being affected by the STDP and the different intrinsic periods of the neurons , change dynamically in time . One efficient way to measure the strength of the coupling within the neuronal population at time t is given by the following synaptic weight ( averaged over the neuron population ) : Cav ( t ) =N−2∑i , jsgn ( Mij ) cij ( t ) , ( 11 ) where Mij is defined in Eq ( 4 ) and sgn is the sign-function . Furthermore , one may additionally measure the degree of the neuronal synchronization within the ensemble , using the order parameter [82 , 110]: R ( t ) =|N−1∑jeiφj ( t ) | , ( 12 ) where φj ( t ) = 2π ( t − tj , m ) / ( tj , m+1 − tj , m ) for tj , m ≤ t < tj , m+1 is a linear approximation of the phase of neuron j between its mth and ( m + 1 ) th spikes at spiking times tj , m and tj , m+1 . R ( t ) is influenced by the synaptic weights , as the latter are time dependent due to the STDP . The order parameter R measures the extent of phase synchronization in the neuronal ensemble and takes values between 0 ( absence of in-phase synchronization ) and 1 ( perfect in-phase synchronization ) . In our numerical calculations , we estimate Cav [see Eq ( 11 ) ] and Rav . The latter quantity is averaged over the last 100 ∙ Ts . Whenever we plot the order parameter versus time , we determine the moving average <R> over a time window of 400 ∙ Ts , because of the presence of strong fluctuations . For the statistical description and analysis of the non-Gaussian distributed Cav and Rav data ( n = 11 samples ) , we use the median as well as the Inter-Quartile Range ( IQR ) [111] . The IQR measures the statistical dispersion , namely the width of the middle 50% of the distribution and is represented by the box in a boxplot . It is also used to determine outliers in the data: an outlier falls more than 1 . 5 times IQR below the 25% quartile or more than 1 . 5 times IQR above the 75% quartile . Selecting the appropriate sample size is a complex issue ( see e . g . http://www . itl . nist . gov/div898/handbook/index . htm ) , especially when the standard deviation is unknown . Following the steps described in ( http://www . itl . nist . gov/div898/handbook/prc/section2/prc222 . htm ) , we use the formula n= ( x1−αs/2+x1−βs ) 2 ( sdδs ) 2 to get a first rough estimation of the number of measurements ( n ) to be included in our sample , where αs refers to the risk of rejecting a true hypothesis , and βs is the risk of accepting a false null hypothesis when a particular value of the alternative hypothesis is true , sd the unknown standard deviation , δs the confidence interval , and x the values from student’s t-distribution . Using 11 samples , as minimum sample size , one is able to reach quite small p-values , much smaller than the significance level as = 0 . 05 . In this study , for each initial network of N = 200 non-identical-neurons and parameter conditions ( or simply “network” ) , we apply RVS and SVS CR signals ( different per network ) . For each network , the initial conditions for each neuron were randomly drawn from random distributions as given in the Hodgkin-Huxley Spiking Neuron Model subsection . We start the simulation with an equilibration phase , which lasts 2 s . Later on , we evolve the network under the influence of STDP ( which will be present until the end of the simulation ) . We then integrate the network for 60 s with STDP without any external stimulation yet , where a rewiring of the connections takes place , resulting in a strongly synchronized state . Next , we apply CR stimulation for 128 s ( resetting the starting time to t = 0 s ) . During this CR-on period three stimulation ON-cycles ( the stimulation is on ) alternated with two OFF-cycles ( the stimulation is off ) as in the example stimulation signals shown in Fig 2 . Each ON- and OFF-cycle lasts Ts . After 128 s the CR stimulation ceases permanently and we continue the evolution of the CR-off period for extra 128 s . In order to probe and chart the CR stimulation intensity and frequency parameter space , we restrict the CR stimulation intensity to values in the interval ( K ∈ [0 . 20 , … , 0 . 50] ) . This particular choice is based on our previous experience and numerical studies ( see e . g . [45 , 84] ) where it was found that weaker intensities were not able to sufficiently desynchronize the neuron ensemble while larger intensities did not significantly improve ( or sometimes even worsen ) the outcome of RVS and SVS CR stimulation signals . We then set an initial-central value for the CR stimulation period ( that defines the initial/starting frequency ) which in principle is selected close to the intrinsic firing rate of the strongly synchronized network . In this case , and before applying the CR stimulation , the intrinsic firing rate of the network is ~71 Hz which corresponds to Ts ≈ 14 ms . Hence , we begin with the CR stimulation period T0 = 16 ms which gives an initial stimulation frequency f0 = 1/T0 ( in a similar manner just like in [45 , 84] and adjusted to a value close to the intrinsic one ) . Then we define such a period interval [Tsmin , Tsmax] in ms ( Ts: integer ) that allows us to create an “approximately” equidistant grid on the frequency space: fstim∈ [25%f0 , … , 175%f0] . This initial T0 − value is also well studied for different types of signal patterns aiming to optimize the CR effect with the use of different type of CR stimulation sequences ( see e . g . [84] ) . Then , we define the ratio ( % ) of CR sequence frequency per ON-cycle ( fstim ) over the frequency of the reference stimulation frequency ( f0 = 62 . 5 Hz , T0 = 16 ms ) as r0 = ( fstim/f0 ) ∙ 100 and we end up in studying the intensity and frequency-ratio ( K , r0 ) – parameter space . In Table 1 , we show the conversion between the stimulation frequency-ratio and period . For comparison reasons , we also give the corresponding ratios rint ( % ) of CR stimulation frequency per ON-cycle ( fstim ) over the frequency of the intrinsic firing rate of the network frequency ( fint = 71 . 4 Hz , Tint = 14 ms ) without any external stimulation .
Before presenting the core of our findings , let us first start by discussing how the RVS CR stimulation duration affects the long-lasting anti-kindling of different initial randomly chosen networks . In Fig 2 , we show the evolution of the mean synaptic weight Cav as a function of time for different total CR-on time durations: t = 64 s ( Fig 2A ) , t = 128 s ( Fig 2B ) , and t = 256 s ( Fig 2C ) . 128 s is the standard CR-on period used throughout the paper . The CR stimulation intensity is K = 0 . 20 , and the period Ts = 10 ms . A general trend appears in this sequence of panels , i . e . the longer the CR stimulation lasts , less spread of the Cav regarding the long-lasting anti-kindling effect is observed after stimulation offset . This is shown in Fig 2D with boxplots . The last box ( corresponding to t = 256 s of total CR-on period ) has no outliers and shows a more “uniform” long-lasting effect ( as shown in Fig 2C ) for all 11 network initializations , not only during the CR-on period but also afterwards during the CR-off period . However , there is no statistically significant decrease of the median of the Cav from t = 64 s to t = 128 s ( right-sided Wilcoxon rank sum test [112] , p = 0 . 0209 , 5% significance level ) . Moreover , the median value of the Cav does not change significantly between t = 128 s ( Fig 2B ) and t = 256 s ( Fig 2C , both-sided Wilcoxon rank sum test , p = 0 . 8955 ) . Hence , the intermediate stimulation duration t = 128 s provides fairly good results . Furthermore , for considerably larger stimulation durations the anti-kindling is typically , but not always more pronounced . From a clinical standpoint , it is desirable to achieve reasonably pronounced stimulation effects without excessive stimulation durations . Accordingly , in this computational study we choose t = 128 s as total CR-on time , and t = 256 s as total CR-on/off time . For the different simulations , we use different random initial networks and CR signals . For the sake of generality , we do not pick any optimal combination of random initial network and RVS CR stimulation signal that would induce a favorable or biased behavior . This is to assess whether CR effects are robust with respect to different initial conditions . Fig 2B shows a typical example where 11 different random stimulation signals where applied to 11 different initial networks during the CR-on period , with CR stimulation intensity K = 0 . 20 and stimulation period Ts = 10 ms . The CR-on/off period lasts 128 ms respectively . During the CR-on period the mean synaptic weights Cav evolve in a similar manner for all networks , with little deviations between the different curves . They reach approximately the same small value at the end of the CR-on period . The latter corresponds to weak excitatory synaptic connectivity and , in most cases in this paper , to globally well-desynchronized states . However , the post-stimulation dynamics of Cav may be quite diverse . Some networks retain their weak average connectivity while others , like network 2 and 9 ( Fig 2B ) relapse back to states with strong synaptic connectivity . Next , we study what happens if we fix the CR stimulation signal for the 11 different initial networks ( Fig 2E ) . The results are similar to Fig 2B: The outcome at the end of the CR-on period is fairly uniform , while the post-stimulation dynamics of Cav is diverse . Replacing one random external stimulation signal by another one may improve the long-term outcome in some cases ( e . g . network 8 –green dotted line ) , but worsen the outcome in others ( e . g . network 3 –blue solid line ) . These plots indicate that both the random initialization of the network and the different stimulation signals during the CR-on period impact on the final outcome at the end of the CR-off period in a complex manner . Next , we investigate how stimulation intensity and stimulation frequency impact on the mean synaptic weight and synchronization at the end of the RVS CR-on period . Fig 3A shows the median of the mean synaptic weight Cav , and Fig 3B of the order parameter Rav ( averaged over the last 100 ∙ Ts ) as a function of stimulation intensity K and stimulation frequency fstim . The color bars show the median values which were calculated from 11 different random initial network configurations . Overall , at the end of the RVS CR-on period we observe a weak excitatory coupling . In other words , CR stimulation shifts the networks’ couplings towards more inhibition , the inhibitory couplings get stronger , and desynchronized states emerge for most of the ( K , r0 ) pairs , except for the two columns at fstim = 25%f0 ( Ts = 64 ms ) and fstim = 145%f0 ( Ts = 11 ms ) . For the former frequency , CR stimulation fails to weaken both the inter-neural connectivity and synchrony , whereas for the latter frequency CR down-regulates synaptic connectivity , but elevated levels of synchrony persist . Fig 3C and 3D show their Inter-Quartile-Range ( IQR ) respectively , which gives a measure of the data dispersion around these median values . All IQR values being close to zero indicate that the middle 50% of the distribution are very close to the median value . Fig 4 presents a global overview of the long-lasting impact of CR at the end of the CR-off period . Fig 4A shows the median of the mean synaptic weight Cav , and Fig 4B the median of the order parameter Rav . Fig 4C and 4D display the corresponding IQRs , showing that the dispersion around the median of the Cav results is very small in large parts of the parameter plane . In contrast , small IQRs are found only for small Rav , in regions with strong desynchronization . Fig 4A and 4B display two main bands in the ( K , r0 ) − parameter space associated with small dispersion: The first band is aligned along the horizontal axis , for weak stimulation intensities ( i . e . K = 0 . 20 and K = 0 . 25 ) and stimulation frequencies greater than 40% of the standard f0 corresponding to a stimulation period of T0 = 16 ms . The second band runs along the vertical stimulation intensity K axis , and for relatively high frequencies , i . e . for fstim = 160%f0 ( Ts = 10 ms ) and fstim = 175%f0 ( Ts = 9 ms ) which correspond to ~ 155% and ~ 140% of the firing rate of the synchronized neurons , respectively . For these ( bottom-horizontal and right-hand-side-vertical bands ) the dispersion around the median values is quite small for both Cav and Rav ( Fig 4C and 4D ) . In addition , the vertical stripe at the reference frequency value f0 ( "100%" , Ts = 16 ms ) , studied in [84] , but with a t = 64 s CR-on period , is also associated with robust long-lasting anti-kindling and desynchronization for all CR stimulation intensity values K . Another region with similar characteristics lies at the center of Fig 4A and 4B for intermediate stimulation intensity and frequency values . At a first glance , among those two bands in Fig 4A and 4B , where dark color dominates suggesting long-lasting anti-kindling after cessation of CR stimulation , the horizontal band seems especially intriguing . Along the lines of our model analysis , the horizontal band corresponds to pronounced desynchronizing outcome at favorably weak CR stimulation intensities within a range of stimulation frequencies . However , we have to keep in mind that the discrete grid is not very dense . Hence , in order to investigate whether this conclusion is justified , we calculated Cav and Rav for all the integer period Ts values for K = 0 . 20 , ranging from fstim = 175%f0 ( Ts = 10 ms ) to fstim = 40%f0 ( Ts = 40 ms ) . Fig 5 shows this fine-grained analysis . The boxplot for Cav is shown in Fig 5A , and for the Rav in Fig 5B . Note , in this figure the horizontal axis shows the CR stimulation period instead of the frequency . And it is sorted from larger to smaller values for an easier comparison between the two representations . The red and green dots indicate the reference stimulation period T0 = 16 ms and intrinsic firing rate period Tint = 14 ms ms respectively . For Ts ∈ [9 , … , 24 ms] we observe robust anti-kindling effects . In contrast , for Ts ∈ [25 , … , 28 ms] many networks tend to be in a synchronized state , while for Ts ∈ [29 , … , 38 ms] the anti-kindling is found to be robust again , before finally reaching the largest Ts value where the CR stimulation signals are not effective at all . In summary , at weak stimulation intensities favorable stimulation outcomes are achieved within wide ranges of the stimulation frequency . For further analyses of stimulation induced effects observed in particular ranges of the stimulation intensity/frequency parameter plane , we refer to the Supporting Information . For particular stimulation parameters , similar acute effects , as assessed with macroscopic quantities Rav and Cav , may lead to qualitatively different results . Neither prominent features of the connectivity matrix nor the dynamical states of the individually stimulated subpopulations at the end of the CR-on period enabled us to predict the long-term outcome ( see Supporting Information ) . Furthermore , this analysis revealed that CR may be effective without causing side-effects that are time-locked to the individual stimuli ( see Supporting Information ) . Next , we address the robustness of the long-lasting anti-kindling achieved by SVS CR stimulation in the ( K , r0 ) − parameter plane . We use SVS-100 CR stimulation , where the random switching occurs after 100 repetitions of the CR sequence ( for motivation see [84] ) . In Fig 6 we show the total outcome of Cav and Rav , obtained by delivering SVS-100 CR to the same 11 initial networks as in Figs 3 and 4 and varying the CR stimulation frequency and intensity . Let us compare these results with the results for RVS CR ( Fig 3A and 3B ) . Regarding the medians of the Cav , both RVS and SVS CRs ( Fig 3A vs Fig 6A ) overall the parameter dependence outcomes are similar , where the outcome plots of SVS CR ( Fig 6 ) contain more vertical stripes , associated with greater outcome variability . Let us consider some of the differences between RVS CR and SVS CR: For low intensity ( K = 0 . 20 ) and high frequencies fstim = 175%f0 , 160%f0 ( corresponding to Ts = 9 ms , 10 ms respectively ) SVS-100 does neither cause pronounced acute desynchronizing effects nor sustained long-lasting effects . For low CR frequency 25%f0 ( corresponding to Ts = 64 ms , leftmost column ) it requires even stronger intensities to induce an anti-kindling compared to RVS ( Fig 3A ) . Regarding the median of Rav ( Fig 3B vs Fig 6B ) for almost all ( K , r0 ) − parameters the networks are shifted to a desynchronized state at the end of the CR-on period , with only a few exceptions , in particular ( K , r0 ) = ( 0 . 20 , 175%f0 ) , ( 0 . 20 , 160%f0 ) and for 25%f0 . Moreover , for the frequency fstim = 145%f0 the SVS CR stimulation achieves more pronounced anti-kindling effects ( at the end of the CR-on period ) for all intensities K compared to the RVS CR stimulation . In Fig 6C and 6D we present the outcome for the medians of Cav and Rav at the end of the CR-off period for SVS-100 CR . The main differences compared to RVS CR ( Fig 4A and 4B ) are the ‘stripes’ at fstim = 115%f0 and , in particular , at fstim = 145%f0 where SVS-100 neither reduces Cav nor Rav for all K-values . Moreover , also for the lowest intensity value K = 0 . 20 and frequencies fstim = 175%f0 , 160%f0 , 25%f0 no anti-kindling is achieved . However , there is a substantial overlap of the ( K , fstim ) − parameter range where both RVS and SVS-100 CR lead to long-lasting anti-kindling , mainly for high frequencies fstim = 175%f0 , 160%f0 for K ≥ 0 . 25 as well as for 40%f0 ≲ fstim ≲ 100%f0 and a wide range of K-values . Interestingly , whenever SVS CR stimulation causes an anti-kindling , the long-term effects on the connectivity are particularly robust , irrespective of different network initializations and parameters . Let us now investigate a denser Ts period sample for the weakest intensity K = 0 . 20 , with the same format as in Fig 5 for RVS CR stimulation . Boxplots of Cav ( Fig 7A ) and Rav ( Fig 7B ) at the end of the CR-off period show that SVS CR stimulation at this weak intensity is overall less efficient in inducing long-lasting anti-kindling effects compared to RVS CR ( Fig 5 ) . In particular , there is no distinct range of Ts periods where SVS CR causes a pronounced anti-kindling . However , for a few values of Ts for the long-term outcome for SVS is stronger than for RVS , e . g . for Ts = 15 ms , 16 ms . Fig 8 enables us to display the stimulation’s global performance in a more concise manner . Namely , it shows the dependence of stimulation outcome on CR stimulation intensity and frequency for the time-averaged mean synaptic weights Cav ( Fig 8A and 8E ) and time-averaged order parameter Rav ( Fig 8B and 8F ) , both at the end of the CR-off period , with values belonging to the same intensity value K for RVS ( top row ) and SVS CR ( bottom row ) stimulation , respectively . Similar plots but now for values belonging to the same frequency ratio ( fstim/f0 ) ∙ 100 are shown in Fig 8C and 8G and Fig 8D and 8H respectively .
By systematically varying the CR stimulation frequency and intensity and comparing the stimulation outcome of the two different CR protocols , RVS and SVS CR stimulation , RVS CR proved to be more robust with respect to variations of the stimulation frequency . However , in accordance with a previous computational study , restricted to a fixed value of the stimulation frequency [84] , SVS CR stimulation can induce stronger anti-kindling effects . In our study , we obtained particular parameter ranges related to particularly favorable stimulation outcome . If no closed loop adaptation for the stimulation frequency is available , RVS CR stimulation at weak intensities and with stimulation frequencies in the range of the neuronal firing rates enables to effectively and robustly achieve an anti-kindling . To our knowledge , in our study in a plastic network the CR stimulation frequency and intensity were systematically varied for the first time to investigate the impact on acute and long-lasting stimulation outcome . Remarkably , pronounced acute desynchronization ( as measured by means of the standard order parameter from Eq ( 12 ) [82 , 110] ) does not necessarily lead to long-lasting desynchronization . On the one hand this finding might inspire future computational and pre-clinical studies aiming at specifically designing stimulation protocols for long-lasting ( as opposed to acute ) desynchronization . On the other hand , this finding is significant for the development of clinical calibration procedures for CR stimulation , see [113] . In a previous study in networks without STDP Lysyansky and coworkers [81] considered m:n ON-OFF CR stimulation with real rather than integer m and n and varied m and n systematically . For non-integer m incomplete CR stimulation cycles are delivered , intersected by incomplete pause cycles caused by non-integer n . This type of CR stimulation has not yet been used in pre-clinical or clinical studies and is somewhat remote to the initial CR concept that builds on the periodicity of both neuronal firing and stimulus patterns [64 , 65] . For the majority of the CR stimulation parameters used in this work , no drastic change was observed in the firing rates . The only exception was observed for very low stimulation frequency combined with comparably high intensities ( S2 Fig ) . Especially for the most relevant cases ( weak to intermediate intensities and frequencies around the reference stimulation frequency ) the firing rate of the neuron ensemble remains almost unchanged when compared with initial intrinsic firing rates before CR delivery ( about up to ±3% variation of the initial intrinsic firing rate ) . In this study , we focused on a network of spiking Hodgkin-Huxley neurons with STDP . Compared to STDP-free networks used before [81] , this is a step towards more complex and , in particular , plastic neural networks . Future studies should address yet more complex neural networks equipped with STDP to study parameter regions and stimulation protocols that are reasonably stable in different neural network models . In principle , we have to be careful about extrapolating findings obtained in one type of neural network model to network models of higher complexity . For instance , non-linear delayed feedback stimulation was introduced in globally coupled networks of limit cycle oscillators and phase oscillators [114] . It turned out to robustly cause desynchronization , nearly irrespective of the selected valued of the delay [115] . In contrast , linear delayed feedback [116] was shown to induce desynchronization only for a rather small subset of parameter pairs of delay and intensity , favoring delays close to half the intrinsic oscillation period and weak to moderate intensities [115 , 116] . However , in a more complex , microscopic neuronal network model consisting of a population of STN and a population of external globus pallidus ( GPe ) neurons [105] the parameter dependence for nonlinear delayed feedback was qualitatively different [117] . The parameter ranges of delay and intensity values associated with desynchronization were still greater for nonlinear delayed feedback as opposed to linear delayed feedback . However , in this microscopic STN-GPe network model nonlinear delayed feedback had to be properly calibrated and , in particular , the delay had to be adjusted to the intrinsic period of the neuronal oscillations , to enable desynchronization [117] . Note , the microscopic STN-GPe network model did not contain STDP [105] . Incorporating STDP to a neuronal network model substantially adds to the model’s complexity ( see e . g . [34] ) and might , hence , further impact on the dependence of the stimulation outcome on key parameters of both linear and nonlinear delayed feedback . Ultimately , we strive for using several neural networks with STDP as testbed for generating computationally based predictions and recommendations for favorable stimulus parameters and dosage protocols . However , different models may display similar or even identical spontaneous ( i . e . stimulation-free ) dynamics , but may have very different stimulus response properties ( see e . g . [62] ) . Accordingly , we cannot expect a stimulation technique to be generically effective , irrespective of the neural network model used . Nevertheless , stepwise adding further physiologically and anatomically relevant features to the neural network models employed may help to generate specific predictions and , ultimately , to further improve stimulation protocols and dosage regimes . In that sense , the finding that RVS CR stimulation at weak to moderate intensities and stimulation frequencies adapted to the neurons’ intrinsic firing rates causes a desynchronization in neural network models without STDP [81] and with STDP as shown in this study , is relevant and , in fact , in accordance with pre-clinical findings [71 , 74] . Furthermore , the fact that SVS CR stimulation might even be more effective , but requires more careful parameter adaptation may guide future development of calibration techniques as put forward in a forthcoming study [118] . In neural networks with STDP post-stimulation transients may be complex . For instance , for stimulation dosages just reaching the level required for an anti-kindling , a rebound of excessive synchrony may occur immediately following cessation of CR stimulation , while later on a full-blown , sustained desynchronization emerges [43 , 67 , 119] . This rebound selectively relates to synchrony , rather than synaptic connectivity . This phenomenon occurs when after CR delivery the neuronal population just reaches the basin of attraction of a favorable attractor . Upon entering the basin of attraction , the synaptic connectivity is still super-critical , so that synchrony emerges in the absence of stimulation . As the neuronal network relaxes towards the favorable attractor , the initially up-regulated synaptic connectivity fades away until , finally , the synaptic connectivity remains below a critical threshold , hence , preventing the population from getting synchronized [43 , 67 , 119] . However , it remains to be shown to which extent the rebound of synchrony phenomenon might be a generic after-effect occurring for just about sufficient CR dosage or simply an epiphenomenon specific to the computational model [120] used in those studies [43 , 67 , 119] , comprising networks of Morris-Lecar spike generators [120] transformed to burst mode by a slowly varying current [121 , 122] . We here demonstrated that over a wide range of stimulation parameters favorable acute effects do not automatically lead to favorable long-lasting , sustained after-effects . This is in agreement with a computational study in the same model , but performed in only a restricted parameter range [84] , as well as with an EEG experiment performed in tinnitus patients [123] . To characterize stimulation induced effects , we here used the average synaptic weight [Eq ( 11 ) ] and the average amount of neuronal synchrony [Eq ( 12 ) ] . These macroscopic quantities enabled us to effectively investigate the impact of variations of stimulation parameters on the stimulation outcome . However , in Supporting Information section , we have shown that pronounced differences of the average synaptic weight do not necessarily lead to pronounced differences of the average amount of synchrony ( S1 Fig ) . Another example in this context is the combination of weak average synaptic connectivity ( Fig 3A ) combined with increased levels of average neuronal synchrony ( Fig 3B ) at the end of the CR-on period . To further study the relationship between connectivity pattern and synchronization processes , macroscopic quantities may not be sufficient to grasp all relevant details of the connectivity matrix and the dynamical features of the resulting synchronization processes . In a number of previous studies , it was already shown that computational findings obtained in minimal models and , in particular , in models that are even simpler than the model studied in this manuscript , turned out to be of high clinical significance . The following computational predictions were obtained by studying minimal models , such as networks of phase oscillators with and without STDP: The computational predictions obtained in minimal models were actually used to design the pre-clinical and clinical studies referred to above . From a clinical standpoint , these computationally predicted and pre-clinically and/or clinically verified findings are significant and may ultimately enable to establish superior therapies that require stimulus delivery for only a few hours , on a regular or occasional basis . The Mexican hat connectivity was , e . g . used by [108 , 109] to study auditory responses , sensorineural hearing loss and tinnitus . These studies illustrate that the comparably simple Mexican hat connectivity model is able to capture relevant connectivity features . Accordingly , later on , the same connectivity profile was used in a several studies focusing on stimulus-induced desynchronization e . g . in the auditory cortex [45 , 84 , 85] In neural networks without STDP tested so far , CR stimulation works at higher intensities as well , see e . g . [81] . In that case , pronounced cluster states are induced , but coherent synchrony is reliably suppressed [81] . This is not the case in the neural network model with STDP studied here . For both RVS CR and SVS CR , for several parameters tested the long-term outcome deteriorates with increasing stimulation intensity ( Figs 4 , 6 and 8 ) . Accordingly , based on our results , in pre-clinical and clinical applications stimulation at higher intensities should be avoided . Another important aspect refers to the more pronounced periodicity of SVS CR pattern . In previous papers ( lacking a wider scan of the parameter space ) , SVS CR stimulation appeared to be superior to RVS CR stimulation [84 , 85] . In this paper , however , we show that SVS CR stimulation decisively depends on the appropriate choice of the stimulation frequency ( Figs 6 , 7 and 8 ) . This sensitivity may significantly reduce the performance in the presence of biologically realistic variations of the neuronal firing rates and might , hence , be the very reason , why the outcome of SVS CR stimulation is significantly better for smaller numbers of sequence repetitions [preliminary results presented by Wang et al . , Critical parameters determining efficacy of coordinated reset stimulation of subthalamic nucleus and related changes in primary motor cortical and subthalamic local field potentials in a parkinsonian monkey . Society for Neuroscience ( 2017 ) Poster 210 . 02 / I10] . There , based on first pre-clinical data , SVS CR stimulation at high numbers of sequence repetition appears to be inappropriate for an open loop application . However , its performance might be significantly improved by closed loop approaches as , e . g . computationally shown in [118] . For the development of CR stimulation , in a number of computational studies predominantly minimal models were used [33 , 43–45 , 61–68 , 119] , as opposed to biophysically realistic models [46 , 125] . These computational studies gave rise to qualitative non-trivial predictions , e . g . the emergence of long-lasting , sustained [33] as well as cumulative [67] effects and concerning the amplitude of the stimulation amplitude [81] . These predictions were verified in pre-clinical [71 , 74] and clinical studies [70 , 73 , 124] . In fact , the computational findings were used to design the study protocols and generate the underlying hypotheses . However , we have to keep in mind that the minimal-model based approach yields qualitative rather than quantitative predictions . Accordingly , we do not intend to provide “success rates” of the stimulation outcome since we do not intend to relate particular values of Cav and Rav with successful outcome for the following reasons . On the one hand , the mean synaptic weight Cav cannot be assessed in living humans and , hence , so far , no correlation between Cav and the extent of symptoms has been studied . On the other hand , the order parameter Rav cannot directly be assessed either , but is related to the amplitude of macroscopic/mesoscopic signals like the local field potential . However , for instance in Parkinson’s disease it is still a matter of debate whether there is a measurable quantity that reasonably represents the extent of symptoms , in a biomarker-like manner [126 , 127] . In fact , a number of studies provided results that are in contradiction of the biomarker notion [127–133] . Accordingly , so far , it is not possible to provide ranges of the amount of synchrony–reflected by Rav—that correspond to physiological as opposed to abnormal , Parkinsonian states . In recent studies , some alternative approaches have been proposed for the effective suppression of the global synchronization . In Kuramoto oscillator networks , the role of conformists , oscillators attracted to the mean field and tending to synchronize with it , and contrarians , repelled by the mean field and preferring a phase diametrically opposed to it , has been investigated , in order to suppress explosive synchronized activity . The latter refers to the transition from a non-synchronized state to a synchronized state in an abrupt/discontinuous manner ( see e . g . [134–137] ) . Different strategies have been proposed for exploiting the local ( contrarians ) versus total information , the role of the negative versus positive coupling in order to achieve this goal . In our work , when implementing CR stimulation , we use inhibitory and excitatory synapses where all neurons are connected to each other . However , in [45] ( Fig 12 ) , it was shown that CR stimulation can also desynchronize effectively such networks with topologies where a fraction of neurons is excitatory and the rest inhibitory when receiving the same type of stimulus . The impact of the CR stimulation induces an increase in the inhibitory coupling weights and a decrease in the excitatory ones via the STPD . In [33] , a multi-site CR stimulation , has proven to exhibit powerful long-term anti-kindling effects using a network of coupled phase oscillators with STDP . Furthermore in [66] , it was shown that with CR stimulation one can achieve robust long-term curative effects , irrespectively , of the ratio between excitatory and inhibitory impact . Hence , in principle , different types of stimulation-induced modifications of plastic couplings may counteract synchronization . Bistability and strong abnormal-explosive synchrony has also been studied recently in several physical and neuronal systems without synaptic plasticity ( see e . g . [137–140] and references therein ) . These theoretical findings , obtained in generic networks , shed light on the underlying mechanisms on the transition from abnormal to normal activity in networks due to e . g . the interplay of local structure the internal dynamics or the critical role of the coupling strength . Moreover , experimental efforts are made in implementing such ideas , showing that conditions for explosive synchronization in the human brain could suggest a potential mechanism for rapid recovery from the lightly-anesthetized state [141] . Synaptic plasticity is another source for bistability and multistability in oscillatory networks . In fact , bistability and multistability was found in different networks with qualitatively different synaptic plasticity mechanisms . For example , in [32] a slow varying coupling matrix was used in a generalized Kuramoto model of coupled phase oscillators . In that case , the synaptic weights were governed by the time-varying phase difference of pairs of oscillators in such a way that the coupling strength increases for synchronized oscillators and weakens for nonsynchronized pairs . In [33] , under spontaneous conditions two different states , a desynchronized and a synchronized state , were found in a system of coupled phase oscillators with asymmetric STDP . In addition , a similar bistability regime was found in a network of Morris–Lecar neurons with symmetric STDP in [66] . In network of phase oscillators with simplified , time-varying STDP multistability was shown to occur only for asymmetric STDP [34] . In that case , the coexistence of synchronized as well as desynchronized and cluster states depends on the distribution of the eigenfrequencies . In the network with Hodgkin-Huxley neurons with asymmetric STPD studied in this paper , a pronounced multistability was found , see also [45] . Though , not in the intended scope of our present study , the actual mechanism of action of CR stimulation deserves attention in future studies . In neural and oscillator networks without STDP , CR stimulation disrupts synchrony by causing phase resets of different subpopulations at different times [64 , 65 , 81] . The phase reset of a single subpopulation is time-locked to the stimulus affected that particular subpopulation [64 , 65 , 81] . However , in the present study we observed that CR stimulation may not just reorder the neurons’ phases . Rather , for particular stimulation parameters it may even cause a significant decrease of the neuronal firing rates , intriguingly associated with a particularly pronounced anti-kindling ( S2I Fig ) . Furthermore , in contradiction to the results obtained in networks without STDP [64 , 65 , 81] , CR stimulation may cause a full-blown anti-kindling without any phase resets of the subpopulations time locked to the corresponding stimuli ( S4 Fig ) . This is relevant for two reasons: ( i ) Since effective CR stimulation does not require phase resets time-locked to the individual stimuli , further computational studies should elucidate whether it makes sense to calibrate CR stimuli for pre-clinical and clinical applications by selecting stimulus parameters that favorably achieve phase resets . Corresponding results might be relevant for the design of calibration procedures and , in addition , challenge existing patents that are based on selecting parameters that optimally achieve phase resets of the stimuli delivered to the individual sub-populations ( e . g . [142] ) . ( ii ) By the same token , our results do not only challenge current hypotheses on the mechanism of CR stimulation , but also fundamental patents in the field of invasive ( e . g . [143] ) as well as non-invasive ( e . g . [144] ) CR stimulation . Accordingly , future computational studies should focus on the mechanism of action of CR stimulation in networks with STDP in order to actually understand and possibly improve anti-kindling protocols . Our goal is to accomplish an anti-kindling in a way as robust as possible , complying with clinically motivated constraints . For instance , striving for anti-kindling induced at minimal stimulation intensities led to the computational development of spaced CR stimulation [145] and two-stage CR stimulation with weak onset intensity [85] . The motivation behind these developments was to avoid side effects by substantially reducing stimulation intensities [85 , 145] . Another direction is to accomplish anti-kindling at moderate stimulation duration as computationally studied in this paper . This may be favorable from a clinical standpoint since it might help to reduce the occurrence of side effects as well as the requirement of the treatment on patients for their compliance , e . g . in terms of actually using non-invasive therapeutic devices . In this context , it might turn out to be beneficial that RVS CR stimulation causes sustained after-effects over a wide range of stimulation frequencies even at weak intensity ( Fig 4 ) . Accordingly , RVS CR stimulation might provide an appropriate stimulation protocol , in particular , if applied in an open-loop manner , without the ability to calibrate the stimulation parameters , especially the stimulation frequency by adapting it to the dominant peaks in the frequency spectrum of electrophysiological signals such as local field potentials or EEG signals . However , in a forthcoming computational study [118] we will use comparably simple closed-loop control modes to significantly improve the robustness of both RVS and SVS CR stimulation and , in particular , exploit the anti-kindling potential of SVS CR stimulation . | Abnormally strong neuronal synchronization is found in a number of brain disorders . To specifically counteract abnormal neuronal synchrony and , hence , related symptoms , Coordinated Reset ( CR ) stimulation was developed . CR stimulation employs basic plasticity and dynamic self-organization principles of the nervous system . Its fundamental goal is to induce long-lasting desynchronizing effects that persist cessation of stimulation . The latter are key to reducing side effects of invasive therapies such as deep brain stimulation . Furthermore , sustained stimulation effects pave the way for non-invasive neuromodulation treatments , where a few hours of stimulation delivered regularly or occasionally may provide substantial relief . Long-lasting CR-induced desynchronizing therapeutic effects have been verified in several pre-clinical and clinical studies . However , we here present the first computational study that systematically investigates the impact of key stimulation parameters on the stimulation outcome . Our results provide experimentally testable predictions that are relevant for pre-clinical and clinical studies . Furthermore , our results may contribute to stimulation techniques that enable to probe the functional role of brain rhythms in general . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"neural",
"networks",
"neurodegenerative",
"diseases",
"membrane",
"potential",
"signaling",
"networks",
"electrophysiology",
"neuroscience",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"synaptic",
"plasticity",
"functional",
"electrical",
"stimulation",
"network",
"analysis",
"neuronal",
"plasticity",
"computer",
"and",
"information",
"sciences",
"developmental",
"neuroscience",
"animal",
"cells",
"movement",
"disorders",
"cellular",
"neuroscience",
"cell",
"biology",
"neurology",
"neurons",
"parkinson",
"disease",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"neurophysiology"
]
| 2018 | How stimulation frequency and intensity impact on the long-lasting effects of coordinated reset stimulation |
The yeast prion [SWI+] , formed of heritable amyloid aggregates of the Swi1 protein , results in a partial loss of function of the SWI/SNF chromatin-remodeling complex , required for the regulation of a diverse set of genes . Our genetic analysis revealed that [SWI+] propagation is highly dependent upon the action of members of the Hsp70 molecular chaperone system , specifically the Hsp70 Ssa , two of its J-protein co-chaperones , Sis1 and Ydj1 , and the nucleotide exchange factors of the Hsp110 family ( Sse1/2 ) . Notably , while all yeast prions tested thus far require Sis1 , [SWI+] is the only one known to require the activity of Ydj1 , the most abundant J-protein in yeast . The C-terminal region of Ydj1 , which contains the client protein interaction domain , is required for [SWI+] propagation . However , Ydj1 is not unique in this regard , as another , closely related J-protein , Apj1 , can substitute for it when expressed at a level approaching that of Ydj1 . While dependent upon Ydj1 and Sis1 for propagation , [SWI+] is also highly sensitive to overexpression of both J-proteins . However , this increased prion-loss requires only the highly conserved 70 amino acid J-domain , which serves to stimulate the ATPase activity of Hsp70 and thus to stabilize its interaction with client protein . Overexpression of the J-domain from Sis1 , Ydj1 , or Apj1 is sufficient to destabilize [SWI+] . In addition , [SWI+] is lost upon overexpression of Sse nucleotide exchange factors , which act to destabilize Hsp70's interaction with client proteins . Given the plethora of genes affected by the activity of the SWI/SNF chromatin-remodeling complex , it is possible that this sensitivity of [SWI+] to the activity of Hsp70 chaperone machinery may serve a regulatory role , keeping this prion in an easily-lost , meta-stable state . Such sensitivity may provide a means to reach an optimal balance of phenotypic diversity within a cell population to better adapt to stressful environments .
Yeast prions are non-Mendelian genetic elements , most of which are amyloid aggregates formed by single proteins [1]–[8] . These aggregates , often referred to as seeds or propagons , whose number varies depending on the prion , serve as templates for the conversion of newly synthesized protein to the prion conformation [8]–[10] . The presence of an amyloid prion is often associated with phenotypes that arise from the partial loss of function of the prion-forming protein due to its sequestration in the aggregates . For example , [PSI+] and [URE3] , the prion forms of a translation termination factor and a transcriptional regulator , cause misreading of nonsense codons and misregulation of a set of genes involved in nitrogen catabolism , respectively [4] , [11] . In addition , a single prion-forming protein can take on different conformational states , thus resulting in prion “strains” having varying levels of severity of these heritable traits; so-called “weak” and “strong” [PSI+] are an example of such variants [12]–[15] . Of particular interest in regard to prion-associated phenotypes is the recently identified prion , [SWI+] , formed from Swi1 , an important component of the SWI/SNF chromatin remodeling complex [16] , [17] . [SWI+] cells exhibit a partial loss of SWI/SNF function , resulting in the impaired uptake of certain sugars , slow growth on synthetic media , and poor germination [16] . However , despite these mal-adaptive phenotypes , [SWI +] cells grow indistinguishably from wild-type cells on rich media and under at least one condition , the presence of microtubule-inhibiting fungicide benomyl , grow strikingly better than cells lacking the prion [18] . Propagation of yeast prions appears to be inexorably reliant on the function of molecular chaperones , proteins more generally known for their ability to facilitate protein folding and prevent misfolding [19]–[21] . In the case of amyloid prion propagation , they are required for the fragmentation of the aggregates needed to generate additional seeds [22]–[26] . Physical transmission of these aggregates/seeds to daughter cells is required for propagation of the prion in the cell population [8] , [27]–[29] . It has been known for some time that the action of the molecular chaperone Hsp104 , which functions as a dissaggregase by threading partially folded proteins through its central pore , is required for amyloid prion fragmentation [19] , [20] , [22] , [26] , [30] . More recent work has underscored the importance of the cytosolic Hsp70 Ssa and its co-chaperones [9] , [23]–[25] , [31]–[36] . Like all Hsp70 systems , Ssa functions in conjunction with two sets of co-chaperone proteins , J-proteins ( also known as Hsp40s ) and nucleotide exchange factors ( NEFs ) [21] . J-proteins and NEFs are critical because they affect the ATPase activity of Hsp70 and thus the cycle of interaction with client polypeptides such as prion proteins [21] . J-proteins act to stimulate the ATPase activity of Hsp70s , increasing the affinity for client polypeptides; NEFs stimulate the release of ADP from Hsp70 , and thereby facilitate polypeptide release [21] . J-proteins are particularly diverse in structure , with some binding client proteins and “delivering” them to their partner Hsp70 [21] , [37] . However , all have a highly conserved J-domain , which is responsible for ATPase stimulation [37] . Although Ssa has 12 J-protein partners [21] , [38] , only one , Sis1 , is required for the propagation of the three best-studied prions , [PSI+] , [RNQ+] , and [URE3] [24] , [25] , [33] . Current models assert that Sis1∶Ssa and Hsp104 act sequentially to fragment prion aggregates [23]–[25] . The global effects of [SWI+] raise the intriguing possibility that it may have had an impact during evolution , as it is likely that its presence or absence affects growth differentially under a variety of environmental conditions . Therefore , we initiated a genetic analysis of [SWI+] , concentrating on the effects of molecular chaperones on its maintenance . We found that , relative to other prions , [SWI+] propagation is highly sensitive to perturbations in the activity of the Hsp70 machinery . The idea that stress conditions may particularly affect the stability of [SWI+] in cell populations due to this sensitivity to chaperone activity is addressed .
As our first step in the characterization of [SWI+] and its dependence on molecular chaperones for its propagation , we determined its seed number . We employed an established method , a ‘propagon counting assay’ , which is based on monitoring prion-loss upon inactivation of Hsp104 activity [39] . The Hsp104 inhibitor GdnHCl was used [40] , as [SWI+] , like other yeast prions , requires Hsp104 activity for seed generation [16] . At various times after treatment of a culture of [SWI+] cells with GdnHCl , aliquots of cells were plated onto glucose-based medium . To determine the percentage of cells having lost the prion at each time point , cells from a minimum of 24 individual colonies were transformed with a plasmid containing the Asn- and Gln-rich segment of Swi1 , which contains the prion-forming domains , fused to YFP ( Swi1NQ-YFP ) . Resulting transformants were observed under the microscope and scored for Swi1 aggregation based on punctuate or diffuse fluorescence , indicative of presence or absence of the prion , respectively ( Figure 1A ) [16] . As expected , [SWI+] was lost from the population with time; after six generations , approximately 50% of the population was [swi−] ( Figure 1C ) . 13 generations after addition of GdnHCl , none of the colonies tested positive for the prion . No prion loss was observed for the control culture to which no GdnHCl was added . To confirm that the microscopic analysis provided an accurate indicator of the presence or absence of the prion , we also tested the growth of cells on medium containing raffinose as the carbon source . Growth of [SWI+] , but not [swi−] , cells is greatly impaired on such medium because raffinose transport into cells is reduced when activity of the SWI/SNF complex is impaired [16] . All 24 isolates taken from a culture prior to addition of GdnHCl grew poorly on raffinose-based medium; all 24 isolates taken from a culture 13 generations after GdnHCl addition grew well , confirming loss of the prion ( Figure 1B and Figure S1 ) . To estimate seed number , we fit our [SWI+] curing data to an established model in which prion seeds are assumed to be diluted two-fold upon cell division in the absence of Hsp104-mediated fragmentation [29] , [39] . This model allows comparison of relative seed number values among prions . A best-fit curve was generated when ∼45 seeds/cell was used as the initial estimate ( Figure 1C , bold line ) . Utilizing the same model , similar seed number estimates have been generated for other prions: 200–300 seeds/cell for strong [PSI+] strains [15] , [24] , [39] , ∼100 seeds for [RNQ+] [24] , 40–60 seeds/cell for the weak [PSI+] strain [PSI+]Sc37 [15] , [24] and 20–25 seeds/cell for [URE3] [24] , [41] . Thus , we conclude that [SWI+] has a low seed number , higher only than that observed for [URE3] . Sis1 is required for the propagation of at least three prions , [PSI+] , [URE3] , and [RNQ+] [24] , [25] , [33] . Therefore , we next tested the dependence of [SWI+] on this J-protein . Because Sis1 is an essential protein , we utilized a system having SIS1 under the control of the tetR promoter ( TETr ) . This system , which , upon addition of the drug doxycycline , allows repression of Sis1 synthesis to a minimal level required for cell growth ( Figure 2A ) , was previously used to analyze the role of Sis1 in the maintenance of other prions [23]–[25] . As a control , samples of cells cultured in the absence of drug were collected at time intervals and plated onto glucose-based media . As discussed above , the status of [SWI+] was assessed after subsequent transformation of resulting individual colonies ( n = 24 ) with a plasmid expressing Swi1NQ-YFP . sis1-Δ [TETr-Sis1] cells initially maintained [SWI+] at a high frequency , but showed gradual loss of the prion ( Figure 2B , ) as approximately 40% of the cells became [swi−] over 23 cell generations of cell culture in the absence of drug . Upon addition of drug to cells to repress Sis1 synthesis , [SWI+] propagation was severely affected . [SWI+] was lost from the population with sigmoidal kinetics , exhibiting ∼50% loss after only 9–10 generations and complete curing within ∼20 generations . Curing in these cultures was confirmed by testing for the restoration of robust growth on raffinose media ( Figure 2C , and Figure S1B ) . From these data we conclude that [SWI+] , like all other yeast prions thus far tested , is reliant upon Sis1 activity for continued propagation . A comparison of the control experiments in Figure 1C and Figure 2B indicates that [SWI+] is more stable in a wild-type background than in the sis1-Δ [TETr-Sis1] strains . To test whether this loss was due to lower than normal Sis1 expression , we asked whether the prion would be stabilized by supplemental expression from a second plasmid . However , we found that the presence of the second Sis1 plasmid exacerbated prion loss ( data not shown ) , suggesting that the instability of [SWI+] might be due to overexpression rather than underexpression . Thus , we tested two sis1-Δ strains , one expressing Sis1 from the native SIS1 promoter and one expressing Sis1 from the stronger GPD promoter , resulting in either normal or approximately two-fold higher Sis1 expression , respectively ( Figure 3A ) . After passage of the strains for one week on glucose-based media , the presence of [SWI+] was assessed by observing the growth of cells on medium containing raffinose as the carbon source . The cells overexpressing Sis1 from the GPD promoter grew more robustly , similar to the control [swi−] cells . Those having normal levels of Sis1 expression grew poorly , similar to the [SWI+] control ( Figure 3B ) , indicating the prion is indeed sensitive to overexpression of Sis1 . The maintenance or loss of [SWI+] in these strains was also confirmed using fluorescence analysis ( Figure 3C ) . The loss of [SWI+] upon Sis1 overexpression was somewhat surprising , as we previously reported that overexpression of Sis1 did not affect [RNQ+] , [URE3] , or several strains of [PSI+] [24] . However , these previous studies were not conducted in the 74D-694 genetic background that was used here . Therefore , we tested the effect of Sis1 overexpression on the maintenance of [RNQ+] and both weak and strong variants of [PSI+] in the same strain background , using the same expression constructs described above . [PRION+] cells were transformed with SIS1-bearing plasmids and serially passaged for two weeks on selective media before assaying for the continued presence of the prion . The results were the same as those obtained in the W303 background: no loss of [RNQ+] or [PSI+] was detected ( Figure S2 ) . Thus we conclude that the sensitivity of [SWI+] to Sis1 overexpression is a property of the prion , not a characteristic of the 74D-694 genetic background . Since Sis1 overexpression so potently affected [SWI+] , we decided to test two other J-proteins , Ydj1 and Apj1 , which had previously been found to affect yeast prions when overexpressed [12] , [24] , [42]–[45] . To this end , wild-type [SWI+] cells were transformed with either high-copy overexpression plasmids or empty vector . Transformants were repatched once before a second transformation with plasmid bearing Swi1NQ-YFP for prion scoring . The fate of at least 30 independent transformants was determined in each case . [SWI+] was maintained in >90% of control transformants receiving empty vector but , as expected , none of the transformants overexpressing Sis1 maintained the prion ( Figure 4 ) . Overexpression of both Ydj1 and Apj1 destabilized [SWI+] , with 6 out of 36 YDJ1 transformants and 4 out of 48 APJ1 transformants tested maintaining the prion ( Figure 4 ) . The expression of only a J-domain has been shown to be sufficient for carrying out some J-protein functions . For example , the severe growth defects of ydj1-Δ cells can be rescued by expression of only the J-domain of Ydj1 [38] . Given that overexpression of all three J-proteins tested thus far affected [SWI+] , we decided to determine whether this curing effect could also be accomplished by only a J-domain . We used a set of constructs that express J-domains attached to a C-terminal random peptide region and the hemagglutinin A tag ( HA ) [38] . These constructs are all expressed at high levels and are all sufficient to rescue the slow growth phenotype of ydj1-Δ cells ( [38] and unpublished observations , Sahi and Craig ) . As we suspected , [SWI+] was severely destabilized by all three J-domain constructs tested ( Figure 4 ) , demonstrating that overexpression of a J-domain is sufficient to destabilize the prion . Finally , to determine if the effects of J-protein and J-domain overexpression are specific to [SWI+] , and not the result of strain background , we also examined whether the vectors described above also cured [RNQ+] , or strong or weak strains of [PSI+] in the 74D-694 strain background . To do this , we transformed [PRION+] strains with vectors bearing J-proteins and passaged transformants for two weeks on media selective for the plasmid . As expected from results obtained using other strain backgrounds , none of the vectors used in these experiments had any discernable effect on [RNQ+] or either variant of [PSI+] ( Figure S3 ) , suggesting that J-protein and J-domain overexpression affects [SWI+] in a specific manner . Ydj1 is the most highly expressed J-protein in the yeast cytosol [46] . Since we found [SWI+] to be highly sensitive to J-protein levels , we decided to test the ability of cells lacking Ydj1 to propagate [SWI+] . To obtain ydj1-Δ strains , [SWI+] wild-type cells were transformed with a YDJ1 disruption cassette carrying a selectable marker . The presence or absence of [SWI+] was assayed by transformation with plasmid expressing Swi1NQ-YFP . Strikingly , all 19 isolated ydj1-Δ transformants exhibited the diffuse fluorescence indicative of [swi−] cells . 39 transformants that contained the selectable marker , but did not disrupt the YDJ1 gene , were used as controls . 38 retained the punctuate fluorescence pattern of [SWI+] cells . Thus prion loss strongly correlated with the absence of Ydj1 . As an additional test for requirement of Ydj1 in [SWI+] maintenance , we obtained cells which did not express Ydj1 by a different method , plasmid shuffling . First , we constructed a [SWI+] strain having a deletion of the YDJ1 gene on the chromosome , but carrying YDJ1 , driven by its native promoter in a URA3-based plasmid ( ydj1-Δ [YDJ1-Ydj1 , URA3] ) . Cells not expressing Ydj1 were obtained by selecting for resistance to 5-fluoro-orotic acid ( 5-FOA ) , a counter-selection against the URA3-based Ydj1 expression plasmid . Microscopic evaluation of cells subsequently obtained by transformation with Swi1NQ-YFP for [SWI+] scoring revealed loss of the prion in all 20 transformants evaluated , consistent with the requirement for Ydj1 in the maintenance of [SWI+] . We wanted to ensure that strain background did not account for the apparent variation in the requirement of different prions for Ydj1 . Our previous experiments that indicated that Ydj1 was not required for propagation of either [PSI+] or [RNQ+] were carried out in the W303 genetic background [24] . Therefore , we crossed two strains of the 74D-694 background: a [psi−] [rnq−] ydj1-Δ strain with a [PSI+] [RNQ+] wild-type strain . 22 wild-type and 22 ydj1-Δ haploids obtained from the cross were repatched for two weeks and then tested for the presence of the two prions . All 44 strains were [PSI+] [RNQ+] , indicated that , as we previously reported for the W303 strain background , [PSI+] and [RNQ+] are stably propagated in the absence of Ydj1 ( Figure S4 ) . Thus , we conclude that [SWI+] , unlike other prions , requires the expression of the abundant J-protein Ydj1 for continued propagation in yeast . The plasmid-shuffling system described above also opened up an avenue to test the possibility that overexpression of other J-proteins might be able to substitute for full-length Ydj1 in [SWI+] propagation . As a control to test this system , ydj1-Δ [YDJ1-Ydj1 , URA3] was transformed with a second plasmid , either one carrying a second YDJ1 gene , or one lacking an insert , thus serving as a vector control . As expected , after counter-selection against the URA3-based Ydj1-expressing plasmid , none of the 26 vector transformants tested were [SWI+] ( Figure 5 ) . However , 37 of 42 isolates having the Ydj1-expressing plasmid tested positive for [SWI+] . It is likely that the loss of the prion in a small portion of these transformants was due to the increased level of Ydj1 during the time that the cells carried two YDJ1 genes , because modest overexpression of J-domains can result in [SWI+] loss , as discussed above . Therefore , we included in our analysis a wild-type strain expressing Ydj1 from the endogenous gene , as a control for possible prion loss due to increased J-protein function during the construction of these strains , rather than a lack of Ydj1 function in the ydj1-Δ test strain . Wild-type control strains carrying either the vector or YDJ1 on the plasmid sustained some loss of [SWI+] , with 4 of 46 and 3 of 24 transformants being [swi−] , respectively . However , the stability was sufficient to allow use of this system to test the effectiveness of J-protein constructs to substitute for Ydj1 in [SWI+] propagation . Because the requirement of Ydj1 for the maintenance of [SWI+] is unique among yeast prions analyzed to date , we wanted to know whether the requirement was for Ydj1 specifically . As a first step , we tested an N-terminal containing fragment of Ydj1 , Ydj11–134 , which lacks the C-terminal domains involved in client protein binding , but contains the J-domain . Expression of Ydj11–134 could not support [SWI+] propagation , as none of the 23 ydj1-Δ transformants tested were [SWI+] ( Figure 5 ) , suggesting that Ydj1 J-domain function may not be sufficient . To test more directly whether the C-terminal client protein binding domain of Ydj1 is important for [SWI+] maintenance , we made use of a chimera between Sis1 and Ydj1 , S/Y , which contains the C-terminal client protein binding domain of Ydj1 and the N-terminal J-domain and glycine-rich region of Sis1 . We transformed our test and experimental plasmid-shuffling strains with centromeric plasmids expressing either the S/Y chimera or full-length Sis1 . Full-length Sis1 was not able to replace Ydj1 in the ydj1-Δ strain , as none of the 29 transformants analyzed were [SWI+] . It is possible that this failure to replace Ydj1 is due to loss caused by overexpression of Sis1 as discussed above . However , the prion was only mildly destabilized in our control strain , as 15 of 24 transformants maintained the prion . Interestingly , the chimeric S/Y protein was able to maintain the prion in most cases ( 18 out of 27 transformants ) ( Figure 5 ) . We conclude that [SWI+] propagation requires a function that can be accomplished by the C-terminal segment of Ydj1 , but not Sis1 . To begin to address the question of whether Ydj1 was functionally unique , we asked whether another J-protein that is more closely related to Ydj1 than Sis1 could substitute . Although Sis1 has a client protein binding domain bearing some structural similarity with Ydj1 , it does not possess the Zn2+-binding region characteristic of the so-called “class I” J-proteins [47] . Therefore we tested the ability of the class I J-protein Apj1 , which is normally expressed at much lower levels than Ydj1 [46] , to substitute for Ydj1 when overexpressed . As discussed above , Apj1 expression driven by the strong GPD promoter resulted in [SWI+] loss in the wild-type control strain , with only 4 of 48 transformants retaining the prion ( Figure 5 ) , indicating that its overexpression in the presence of wild-type Ydj1 destabilizes the prion . However , 17 of 25 ydj1-Δ transformants expressing Apj1 retained the prion , indicating that Apj1 can at least partially substitute for Ydj1 in [SWI+] propagation . To more directly test the idea that the C-terminal region of Apj1 can functionally substitute for that of Ydj1 , we constructed a Ydj1/Apj1 chimera , called Y/A , by substituting the Zn2+ and putative peptide-binding regions of Apj1 for that of Ydj1 . This chimera was competent to substitute for Ydj1 when expressed from a single-copy plasmid under a constitutive promoter , as 28 of 30 ydj1-Δ transformants remained [SWI+] . Thus , we conclude that there is functional overlap between Ydj1 and Apj1 and that a function of the C-terminal regions of Ydj1 or Apj1 is required for [SWI+] propagation . The above observations implicate several J-proteins in [SWI+] biology . Because Hsp70 ATPase stimulation is the only known function of a J-domain [21] , [37] , the sensitivity of [SWI+] to J-domain overexpression strongly indicates that this curing effect is likely mediated through alteration of Hsp70 activity . To test this more directly , we made use of a variant of Ydj1 , Ydj1H34Q , which lacks a histidine residue critical for J-domain-mediated stimulation of Hsp70 [21] , [37] , [48] . We subjected Ydj1H34Q to the tests described in the previous section . We found that it was unable to cure [SWI+] when overexpressed in a wild-type strain and unable to replace Ydj1 in [SWI+] maintenance ( Figure 5 ) , indicating that J-domain function , and therefore likely Ssa stimulation , is required in both [SWI+] curing and maintenance . We next wanted to test for Hsp70's involvement more directly . Since Ssa-type Hsp70s , the partner of both Sis1 and Ydj1 , are essential , we turned to a previously identified Ssa1 variant , Ssa1–21 known to affect the maintenance of other yeast prions [35] . Ssa1–21 bears a single amino acid substitution in the C-terminal domain ( L483→W ) . Expression of Ssa1–21 can destabilize [PSI+] , even in presence of the wild-type protein [35] . To test whether [SWI+] is similarly affected by Ssa1–21 expression , we transformed [SWI+] cells with a vector expressing Ssa1–21 under the constitutive TEF promoter . As controls , we also transformed cells with either vector expressing wild-type Ssa1 , or empty vector . Transformants were re-patched once before individual colonies ( n≥21 ) were transformed with Swi1NQ-YFP for [SWI+] scoring . [SWI+] was maintained in a high percentage of transformants regardless of whether cells expressed the wild-type Ssa1 ( 19 out of 21 ) , or empty vector ( 27 out of 30 ) . Ssa1–21 , on the other hand , greatly affected [SWI+] , dominantly curing the prion in all 22 transformants examined . Thus , we conclude that Ssa1 , along with its J-protein co-chaperones , is involved in [SWI+] propagation . The degree to which [SWI+] was cured by Ssa1–21 expression was surprising considering that [PSI+] is only mildly affected by Ssa1–21 [35] . However , because these observations were made in a different yeast genetic background , direct comparisons between [SWI+] and [PSI+] are not possible . To address whether strain background is a confounding factor in evaluating our data , we tested the effects of Ssa1–21 overexpression on [PSI+] in the 74D-694 background . To do this , we transformed strains bearing either weak or strong [PSI+] with the Ssa1–21 expression vector . After only one passage of the transformants on selective media , the time period used in the [SWI+] experiments , we observed no [PSI+] loss in any strain . Therefore , we continued passaging cells for two weeks to allow adequate time for prion curing . Indeed , while occasional [psi−] ( red ) colonies were observed in cultures of the weak [PSI+] strain ( not shown ) , [PSI+] was maintained in the overwhelming majority of cells in all cultures ( Figure S5 ) . These results indicate that strain background is not the causative factor for the high curing rate of [SWI+] , compared to [PSI+] , we observed when Ssa1–21 is expressed . Thus , we conclude that [SWI+] is markedly more sensitive than either weak or strong [PSI+] to this alteration in Hsp70 activity . The function of Hsp70 chaperone machinery requires the action of NEFs to stimulate ADP/ATP exchange , and subsequently , peptide release [21] . Consistent with this requirement , one particular NEF of the Hsp110 family , Sse1 , has been found to be important for the continued propagation of [URE3] and some weak strains of [PSI+] , but not [RNQ+] or strong [PSI+] [43] , [49] . To test whether [SWI+] also requires Sse1 , we created sse1-Δ strains by transformation of a wild-type [SWI+] strain with an SSE1 deletion cassette bearing the LEU2 selectable marker . The absence of Sse1 expression was verified by immunoblot analysis ( Figure 6A ) . All 31 resulting sse1-Δ transformants became [swi−] , as judged by both their ability to grow robustly on raffinose-based media and by the absence of punctate fluorescence ( Figure 6B–6D ) . In contrast , 11 transformants , which obtained the selectable marker but preserved wild-type Sse1 expression , maintained the prion , indicating that like [URE3] and weak [PSI+] strains , [SWI+] is lost upon SSE1 deletion . The Hsp110-type Sse protein family consists of two homologous isoforms Sse1 and Sse2 [50] . Because [SWI+] has exhibited a high sensitivity to ectopic chaperone expression , we also tested whether overexpression of either isoform would affect [SWI+] . To do this , [SWI+] cells were transformed with high-copy plasmids expressing either Sse1 or Sse2 from the constitutive GPD promoter or , as a control , empty vector . Overexpression of either isoform significantly destabilized [SWI+] relative to strains transformed with empty vector ( Figure 7 ) . Greater than 92% of the transformants overexpressing an NEF became [swi−] , while less than 4% of the vector control did . We conclude that stable [SWI+] propagation requires moderate expression of Sse proteins . Although Sse1 has NEF activity [51]–[53] , the fact that it has a domain structure similar to that of Hsp70s suggests that it might have additional functions as well . For example , Sse1 is known to interact with client proteins , though the relationship between client protein binding and NEF activity is less clear [54] , [55] . To assess whether loss of [SWI+] caused by Sse1 overexpression was due to increased NEF activity we took advantage of previously characterized SSE1 mutants . After transformation of plasmids carrying the mutant genes into a wild-type [SWI+] strain , transformants were tested for prion maintenance , using both the raffinose growth assay and visualization of Swi1 distribution . First , we tested whether expression of either the N-terminal ATP-binding domain or the C-terminal putative peptide-binding domain of Sse1 is sufficient to effect prion loss by expressing either Sse1Δ394–693 or Sse1Δ1–396 . In both cases , all of the 60 transformants tested positive for [SWI+] , indicating a requirement of both domains for prion curing ( Figure 8 ) . We then tested two point mutations encoding single amino acid alterations within the N-terminal domain: ( 1 ) G233→D in the ATP binding site , which impairs , both in vitro and in vivo , Sse1's NEF activity; ( 2 ) K69→Q in a site predicted to be required for ATP hydrolysis , which has no measureable effect on Sse1 NEF function [51] , [52] . 57 out of 60 transformants overfexpressing Sse1K69Q lost [SWI+] , while the prion was stable in those expressing Sse1G233D . These data are consistent with the hypothesis that Sse1 acts as a nucleotide exchange factor for Hsp70 in [SWI+] curing , rather than performing another uncharacterized function . Our results described above indicate that [SWI+] is highly sensitive to various perturbations of the activity of the Hsp70 chaperone machinery brought about by ectopic expression or mutation . As it is well understood that yeast naturally encounter stressful environmental conditions known to alter chaperone expression , we wanted to ask if [SWI+] is sensitive to such conditions . We subjected our wild-type [SWI+] strain to a variety of cell stresses , including heat and ethanol shock , as well as acute exposure to severe oxidative stress . No appreciable loss of the prion was found compared to untreated control cells for any of these conditions ( data not shown ) . However , because prion curing typically requires multiple cell divisions , we next tested whether extended growth at elevated temperatures , a condition known to cause prolonged alteration in chaperone activity altered [SWI+] stability . Indeed , cells grown for 8 days at 37°C reproducibly regained the ability to grow well on raffinose-based media , relative to cultures grown at 23°C ( Figure 9A ) . To confirm that the improved growth on raffinose was in fact due to loss of the prion , rather than some other alteration acquired during growth at 37°C , we also assayed these cultures for Swi1 aggregation using the YFP assay . The fraction of [SWI+] cells in a culture grown at 37°C was significantly less than that from a control culture grown at 23°C ( Figure 9B ) , confirming that unlike [PSI+] [35] , [SWI+] is destabilized by prolonged growth at elevated temperatures .
The J-protein Sis1 is required for the propagation of [SWI+] , as this prion is lost upon repression of Sis1 expression . This is not a surprising result . The three prions previously analyzed , [PSI+] , [RNQ+] and [URE3] , also require Sis1 [24] , [25] , [33] . However , amongst these four prions , [SWI+] is unique in that it also requires Ydj1 , as [SWI+] is lost when YDJ1 is deleted . Ydj1 , the most abundant J-protein in the cell , is involved in many physiological processes [21] , [46] . Because of these global roles , ydj1-Δ cells grow extremely poorly , thus raising the possibility that the requirement for Ydj1 is due to a general effect on cell growth , rather than a direct role in prion dynamics . However , our results point to a direct and important role of client protein binding in Ydj1 function in [SWI+] propagation . Ydj1 is structurally complex , containing an N-terminal J-domain , critical for functional interaction with Hsp70s , and a C-terminal region , capable of binding client proteins [21] . The N-terminal fragment Ydj11–134 was not able to provide the [SWI+] propagation function . However , as we previously reported [38] the J-domain itself is sufficient to restore robust growth . Thus , the ability of Ydj11–134 to suppress the slow growth phenotype of ydj1-Δ cells , but not the prion propagation defect , indicates that the requirement for Ydj1 is not an indirect effect of poor cell growth but rather a specific requirement for Ydj1 . The fact that the C-terminal region containing the client protein binding domain , when fused to the N-terminal J-domain containing region of Sis1 , was able to substitute for full-length Ydj1 in [SWI+] propagation , supports the idea that Ydj1 binding to a client protein , presumably Swi1 itself , is critical . However , the role that Ydj1 plays remains elusive . It is possible , that it , like Sis1 , functions in fragmentation to generate seeds . Alternatively , Ydj1 may play a role in conversion of the soluble form of Swi1 to the prion conformation . Although , because Ydj1 inhibits [URE3] fiber formation in vitro [56] , [57] , it seems more likely that Ydj1 may oppose polymerization , perhaps preventing the formation of dead-end aggregates . It is possible that Ydj1 plays a role , though clearly a non-essential one , in the maintenance of other prions . Indeed , Ydj1 has also been found to associate with Sup35 and Rnq1 , the proteins which form [PSI+] and [RNQ+] [32] , [58] , [59] . This idea is also supported by the observation that , at least for [PSI+] , deletion of the YDJ1 gene exacerbates the negative phenotypes of Ssa1–21 , indicating that Ydj1 may perform a beneficial function in [PSI+] maintenance under normal circumstances as well [60] . In addition , the observation that Apj1 , a J-protein predicted to have a structure quite similar to that of Ydj1 , could compensate for Ydj1 in [SWI+] maintenance when expressed at a sufficiently high level does support the idea that other J-proteins in the cytosol may be compensatory in the absence of Ydj1 , at least for prions other than [SWI+] . It is also interesting to note that Apj1 , named Anti-prion J-protein 1 , was originally identified in a screen for factors capable of curing a synthetic prion when overexpressed [42] . The data presented here support the idea that [SWI+] is more sensitive to the change in balance of the Hsp70 chaperone system than other prions such as [RNQ+] and [PSI+] . The dependence on Ydj1 for [SWI+] propagation described above is one example . The destabilization of [SWI+] , unlike [PSI+] and [RNQ+] , by overexpression of J-domains and NEFs , is another example . In addition , the observation that [SWI+] was cured in all cells expressing the Ssa1–21 variant underscores the idea that a variety of alterations in Hsp70 chaperone activity can cause destabilization of this prion . The explanation of the mechanism ( s ) behind the observed sensitivity is not obvious . For example , from the general understanding of the Hsp70 machinery , overexpression of J-proteins or NEFs would be expected to bias Hsp70 toward the ADP- or ATP-bound states , respectively . But how this precisely affects the overall rate of client protein cycling and the residence time spent associated with Hsp70 is not understood . Regardless of the precise mechanism , the overall picture that emerges is one in which [SWI +] propagation is delicately balanced , requiring a steady-state level of Hsp70 machinery activity , the disruption of which causes dramatic instability . It should be noted that while “strong” and “weak” variants of several prions have been identified , only one form of the recently identified [SWI+] prion is known . It will be of interest to analyze other stronger and weaker variants when they become available . One other prion , [URE3] , stands out as being sensitive to Hsp70 machinery activity . Although it does not require Ydj1 for propagation [24] , it is sensitive to overexpression of both J-domains and the NEF Sse1 [24] , [43] , [45] . Intriguingly , a recent analysis of the amino acid composition of known yeast prion- domains ( PrDs ) revealed several distinctive features placing the four prions discussed here into two groups: [PSI+] and [RNQ+] in one; [SWI+] and [URE3] in the other [61] . While all four are abundant in Q and N residues , a distinctive feature of yeast prion-forming proteins , the ratio of these residues ( Q∶N ) in the PrDs of Sup35 and Rnq1 is nearly 2∶1 , whereas the PrDs of Swi1 and Ure2 are N-rich , having Q∶N ratios approximating 2∶3 and 1∶3 , respectively . A recent study of Q/N-rich proteins in yeast revealed that those richest in N residues were more likely to form prions , supporting the idea that asparagines are more prionogenic than glutamines [18] . Congruent with this idea , the N-terminal 323 residue “N-domain” of Swi1 alone , which is N-rich , is sufficient for amyloid formation and prion induction , whereas the adjacent Q-rich region is not [62] . Additionally , PrDs of Swi1 and Ure2 are more abundant in bulky hydrophobic residues ( F , I , L , M , V , W ) known to promote amyloid formation [63]–[65] ( 19% and 15% , respectively ) than the corresponding domains of Sup35 and Rnq1 ( 4% and 9% ) . On the other hand , Sup35 and Rnq1 are highly abundant in glycine residues ( 17% vs . 3% and 6% for Swi1 and Ure2 ) . Taken together , the amino acid compositions of the PrDs which form prions that are particularly sensitive to chaperone function ( [SWI+] and [URE3] ) appear to be skewed in favor of amyloid formation relative to those that are less sensitive ( [PSI+] and [RNQ+] ) . This correlation also extends to a fifth prion [PSI+PS] . [PSI+PS] is the prion form of a chimeric protein in which the PrD of Sup35 from Saccharomyces cerevisiae ( Sup35-NSc ) is replaced by the corresponding domain from Pichia methanolytica ( Sup35-NPm ) [66] . Like [SWI+] , [PSI+PS] is also highly sensitive to ectopic chaperone expression , being destabilized by overexpression of Sis1 , Ydj1 , Apj1 or Sse1 , or by deletion of SSE1 [42] , [44] . Strikingly , like the PrD of Swi1 , Sup35-NPm is N-rich ( Q∶N≈1∶2 ) and has a lower content of glycine and a higher content of bulky hydrophobics than Sup35-NSc , consistent with the idea that PrDs which favor amyloid formation form prions with a higher degree of chaperone sensitivity . Thus , we think that the available data makes the idea that the amino acid composition of a PrD plays an important role in determining the sensitivity of a prion to chaperone activity an idea worthy of testing . Of course , it must be kept in mind that other factors , such as the character of the adjacent non-PrD , may play either a primary or secondary role . How might propensity to form amyloid relate to chaperone sensitivity ? One possibility is that prion forming proteins which are optimized for amyloid formation may form stable amyloid fibers which are more difficult to fragment than those of other prions . An alternate , but not mutually exclusive possibility is that these proteins may exhibit higher fiber extension rates in vivo . Rapidly formed fibers may laterally associate [67] , occluding chaperone interaction sites , before fibers can be fragmented by the Hsp70·Hsp104 chaperone machinery . In either case , the expected result would be prions with larger and less numerous prion seeds . Indeed , as discussed above , [URE3] and [SWI+] have a lower number of prion-forming seeds/cell than [RNQ+] or [PSI+] [15] , [24] , [41] . However , since all the parameters that determine seed number are unknown , other factors besides amyloid propensity may also be involved . For example , prions with a low number of seeds/cell may simply be less able to withstand even small changes in chaperone activity , which would lead to a further decrease in seed number , and failure to disseminate seeds efficiently to daughter cells . It is intriguing that the two prions found to be more sensitive to Hsp70 chaperone activity , [URE3] and [SWI+] , are formed by proteins that regulate S . cerevisiae's use of the essential nutrients , nitrogen and carbon , respectively . The presence of [URE3] results in indiscriminate utilization of nitrogen sources due to de-repression of genes typically involved in utilization of nitrogen under starvation conditions , while [SWI+] results in altered carbon source utilization [4] , [16] , [68] . It is easy to imagine that the presence and/or rapid loss of these prions could profoundly affect the ability of cells to survive and thrive under particular stressful environmental conditions [68] . For example , [SWI+] is clearly disadvantageous when grown in the presence of particular carbon sources such as raffinose , but has been reported to be advantageous in the presence of a mircotubule-inhibiting drug [18] . Thus , high chaperone sensitivity may be a beneficial counterbalancing factor to allow curing of the prion under certain unfavorable environmental conditions , and provide a means to reach an optimal balance of phenotypic diversity within a cell population . Our data indicating that [SWI+] is prone to being lost upon prolonged stress points to such a possibility . It is not known if [SWI+] is unique in this regard . However , [PSI+] has been reported to be quite stable in cells grown at elevated temperatures [35] . It is also interesting to note that two very recently identified prions , [MOT3] and [OCT+] , are formed by the proteins , Mot3 and Cyc8 , respectively , which play roles in regulation of gene expression [18] , [69] . Mot3 , like Ure2 , plays a rather specific role , regulating the expression of genes needed for robust growth when oxygen is limited . Cyc8 , on the other hand , is a global regulator , like Swi1 . It acts as a general co-repressor of RNA polymerase II , as well as playing a role in global chromatin structure [68] . It is possible that these prions might also exhibit high sensitivity towards chaperone activity , and thereby , along with [SWI+] and [URE3] , be candidates for prions that may play particularly important roles in adaptation of yeast to stressful environments .
Unless otherwise noted , the originally described [SWI+] 74D-694 strain ( [SWI+] [psi−] [rnq−] MATa ade1–14 ura3–5 leu2–3 , 112 trp1–289 his3–200 SNF5YFP::kanMX4 ) [16]was considered the wild-type strain used in all experiments to characterize the [SWI+] prion . All other strains are of the 74D-694 genetic background . To create [SWI+] cells for Sis1 repression , homozygous diploid cells ( [psi−] [RNQ+] ade1–14 ura3–52 leu2–3 , 112 trp1–289 his3–200 , a gift from Susan Liebman ) were transformed with a Δsis1::LEU2 PCR fragment . The recombinant SIS1/sis1-Δ::LEU2 heterozygous diploid ( Y1544 ) was transformed by a plasmid bearing SIS1 and a URA3 marker and the resulting transformants were sporulated and subjected to tetrad dissection . A resulting haploid strain ( [RNQ+] MATα sis1::LEU2 [SIS1-Sis1 , URA3] ) was mated to the wild-type [SWI+] strain . Two haploid strains ( [SWI+] [RNQ+] sis1::LEU2 [SIS1-Sis1 , URA3] ) were obtained from one complete tetrad and the presence of [SWI+] confirmed . Transformation of these strains with p414-TETr-SIS1 followed by passage onto 5-FOA , which counter-selects against the URA3 plasmid , resulted in the isolation of three [SWI+] , sis1::LEU2 , [TETr-Sis1] strains . To test the effect of YDJ1 deletion on [SWI+] , we transformed the wild-type [SWI+] strain with a PCR-generated ydj1::LEU2 integration cassette , and transformants selected on –Leu media . Transformants were identified as ydj1-Δ by slow growth at 30°C and 23°C and by the ability to rescue this phenotype by subsequent transformation with a plasmid expressing normal levels of Ydj1 protein . Original Leu+ transformants which grew normally at both temperatures were classified as YDJ1 and used as controls . To construct a strain suitable for YDJ1 gene plasmid shuffling experiments , a resulting [swi−] ydj1-Δ strain was transformed with a Ydj1-expressing plasmid [YDJ1-Ydj1 , URA3] and mated to a MATα [SWI+] strain which is otherwise isogenic to our wild-type strain . Isolated diploids were screened for [SWI+] and sporulated on media selective for the YDJ1 plasmid . One haploid strain ( [SWI+] ydj1::LEU2 [YDJ1-Ydj1 , URA3] ) was isolated from a single complete tetrad and used for plasmid-shuffling manipulations . A prion-cured version of this strain ( MATa ) was also mated to a [PSI+] , [RNQ+] MATα strain ( Y1682 , described below ) , and subjected to tetrad dissection and 5-FOA treatment to test the ability of these prions to propagate in a ydj1-Δ strain . Two haploid strains were used to test the effects of chaperone overexpression on other prions . One strain , Y2051 ( [PSI+]Sc37 [rnq−] ade1–14 ura3–52 leu2–3 , 112 trp1–289 his3–200 , a gift from Jonathan Weissman ) , bears the well-characterized weak [PSI+] variant [PSI+]Sc37 ( referred to in the text as “weak [PSI+]” ) [14] , [15] . The second strain , Y1682 ( [PSI+] [RNQ+] ade1–14 ura3–52 leu2–3 , 112 trp1–289 his3–200 ) carries both [RNQ+] ( derived from a strain Y1505 , a gift from Susan Liebman ) [12]and a strong [PSI+] variant generated by transient Sup35 overexpression . This strain was used in all other investigation involving [RNQ+] or [PSI+] ( referred to in text as “strong [PSI+]” ) . The [RNQ+] variant in this strain is a mitotically stable variant from the Liebman laboratory ( derived from strain L1842 ) [12] . This variant is resistant to curing by Ydj1 overexpression . To create sse1-Δ strains , a plasmid-based disruption construct sse1-Δ::LEU2 ( a gift from Kevin Morano ) was digested with Sac II and Pst I [52] . The resulting digestion mixture was used to transform the wild-type [SWI+] strain and transformants selected on –Leu media . SSE1 disruption was confirmed by immunoblotting . A complete list of plasmids used in this study is shown in Table 1 , and unless otherwise indicated , are based on the pRS plasmid series [70] . The URA3-marked plasmid carrying the Swi1NQ-YFP construct was described elsewhere [62] . A HIS3-marked version of this vector was constructed by digestion with the enzymes Spe I and Xho I , and ligation into the pRS vector p413TEF . The SSA1 gene was cloned by polymerase chain reaction ( PCR ) using genomic DNA from a wild-type yeast strain from the W303 genetic background . A double-stranded product was then digested with BamH I and Xho I and ligated into predigested p416TEF to created the plasmid p416TEF-SSA1 . The plasmid p416TEF-SSA1-21 was then constructed by site-directed mutagenesis PCR ( Quikchange ) to introduce a single Trp residue in place of Leu483 . The plasmid p414GPD-YDJ1H34Q was created by first subcloning the YDJ1 open reading frame into p414GPD and subsequently a single Gln was introduced in place of His34 by Quikchange . The plasmid p414ADH1-Y/A was constructed by “PCR sewing” to combine PCR amplified fragments encoding residues 1–113 of Ydj1 and 162–529 of Apj1 , flanked by restriction sites for the enzymes XbaI and SalI at the 5′ and 3′ ends , respectively . The amplified linear insert was subsequently double digested and ligated into the vector p414ADH1 . To test SWI/SNF-dependent phenotypes , cells were plated onto synthetic media supplemented with 1 µg/ml antimycin A from Streptomyces sp . ( Sigma ) and 2% raffinose as the carbon source ( defined as raffinose-based media throughout ) . The glucose-based rich media YEPD ( Teknova ) is used as a control . Cells were grown at 30°C for 2–3 days prior to imaging . Prion-mediated Swi1 aggregation was observed directly by transforming cells with a plasmid expressing the Asn- and Gln-rich regions of Swi1 ( residues 1–554 ) , which contains the prion-forming domains , fused to YFP ( Swi1NQ-YFP ) . [SWI+] cells exhibit a characteristic punctuate fluorescence in the cytoplasm against a dark cytosolic background , whereas [swi−] cells exhibit a diffuse cytosolic fluorescence in combination with occasional increased nuclear fluorescence due to Swi1 accumulation in the nucleus and/or a single extremely large non-prion aggregate in less than 5% of cells [16] , [62] . Transformation of [SWI+] cultures routinely results in >85% of cells presenting punctuate fluorescence [16] , [62] . To allow time for prion curing , cells were serially passaged by repatching onto solid media every 2 days for the duration specified . ydj1-Δ cultures were grown at 23°C; all other yeast cultures were grown at 30°C unless otherwise noted . The presence or absence of [PSI+] was confirmed by observation of colony color on rich media where [PSI+]-mediated aggregation of Sup35 , a translation termination factor , causes read-through of the premature nonsense codon in the ade1–14 mutant allele [71] , [72] . Strains which are otherwise wild-type for adenine production appear pink or white in the presence of [PSI+] or dark red in the absence of [PSI+] due to the accumulation of a red intermediate when adenine production is blocked [73] . [RNQ+] aggregates in cells were observed directly following transformation with a vector expressing Rnq1 fused to GFP ( Rnq1-GFP ) . [RNQ+] cells can be easily distinguished from [rnq−] cells when examined under a microscope by characteristic punctuate or diffuse fluorescence patterns , respectively [23] . Semi-denaturing detergent agarose gel electrophoresis ( SDD-AGE ) was also used to resolve [RNQ+]-dependent detergent resistant aggregates and was performed as described elsewhere [24] , [74] , [75] . To visualize aggregates , protein was transferred to a nitrocellulose membrane at 1A for 1 hr at 25°C in a tris-glycine/methanol buffer and probed with antibodies specific for Rnq1 . Time course experiments for [SWI+] curing were performed as previously reported for [PSI+] and [URE3] [24] . Cell cultures were maintained in exponential growth phase by continual subculturing in YEPD in the presence of either 5 µg/ml doxycycline ( Sigma ) or 4 mM GdnHCl when indicated . Sis1 depleted cells remained viable for the duration of the experiment with a typical growth rate of 2 . 0–2 . 5 hrs/generation . GdnHCl curing experiments used to estimate relative seed numbers utilize the ‘propagon counting assay’ model of Cox et al . 2003 , which may underestimate actual propagon number [76] , but enables direct and unbiased comparisons between estimates generated for different prions [29] . Cells grown at either 23°C or 37°C were subjected to heat shock ( 2 min . at 51°C ) , ethanol shock ( 2 min . with 12% EtOH ) , or exposure to severe oxidative stress ( 3 min . with 4mM H2O2 ) and then allowed to recover for in fresh media without additional additives at 30°C for 2 days before assaying for the presence of the prion . To test the effects of prolonged heat stress , cells were patched onto glucose-based media and grown at either 23°C or 37°C . Cells were grown for 4 days before repatching to fresh media and allowed to grow an additional 4 days . Cells were then repatched to fresh media and grown overnight to allow time to recover before being assayed for the presence of [SWI+] by raffinose growth assay or Swi1NQ-YFP transformation . Total protein extracts were prepared by harvested yeast cells in mid-log phase followed by lysis in NaOH . The resulting protein extracts were analyzed by SDS-PAGE and immunoblot analysis . Sse1 antibody used in this study was kindly provided by Jeff Brodsky . Antibodies specific to Sis1 and Rnq1 have been described elsewhere [23] . Densitometry measurements used to estimate relative protein expression levels were made using the program ImageJ [77] . | Yeast prions are heritable genetic elements , formed spontaneously by aggregation of a single protein . Prions can thus generate diverse phenotypes in a dominant , non-Mendelian fashion , without a corresponding change in chromosomal gene structure . Since the phenotypes caused by the presence of a prion are thought to affect the ability of cells to survive under different environmental conditions , those that have global effects on cell physiology are of particular interest . Here we report the results of a study of one such prion , [SWI+] , formed by a component of the SWI/SNF chromatin-remodeling complex , which is required for the regulation of a diverse set of genes . We found that , compared to previously well-studied prions , [SWI+] is highly sensitive to changes in the activities of molecular chaperones , particularly components of the Hsp70 machinery . Both under- and over-expression of components of this system initiated rapid loss of the prion from the cell population . Since expression of molecular chaperones , often known as heat shock proteins , are known to vary under diverse environmental conditions , such “chaperone sensitivity” may allow alteration of traits that under particular environmental conditions convey a selective advantage and may be a common characteristic of prions formed from proteins involved in global gene regulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"genetics",
"and",
"genomics/complex",
"traits",
"microbiology/microbial",
"growth",
"and",
"development",
"genetics",
"and",
"genomics/gene",
"function"
]
| 2011 | [SWI+], the Prion Formed by the Chromatin Remodeling Factor Swi1, Is Highly Sensitive to Alterations in Hsp70 Chaperone System Activity |
Despite their importance in maintaining the integrity of all cellular pathways , the role of mutations on protein-protein interaction ( PPI ) interfaces as cancer drivers has not been systematically studied . Here we analyzed the mutation patterns of the PPI interfaces from 10 , 028 proteins in a pan-cancer cohort of 5 , 989 tumors from 23 projects of The Cancer Genome Atlas ( TCGA ) to find interfaces enriched in somatic missense mutations . To that end we use e-Driver , an algorithm to analyze the mutation distribution of specific protein functional regions . We identified 103 PPI interfaces enriched in somatic cancer mutations . 32 of these interfaces are found in proteins coded by known cancer driver genes . The remaining 71 interfaces are found in proteins that have not been previously identified as cancer drivers even that , in most cases , there is an extensive literature suggesting they play an important role in cancer . Finally , we integrate these findings with clinical information to show how tumors apparently driven by the same gene have different behaviors , including patient outcomes , depending on which specific interfaces are mutated .
Cancer patients are extremely heterogeneous in their response to treatments and disease outcomes . The first step towards the understanding of this variability was the identification of the multitude of genes that cause cancer , the so-called cancer driver genes[1] . In that sense , the completion of The Cancer Genome Atlas ( TCGA ) and other large-scale cancer genomics projects was a watershed event , as it provided the critical mass of data needed to identify driver alterations in most types of cancers[2–15] . Moreover , cancer types that previously were thought to represent homogenous diseases were found to constitute different subtypes with different outcomes depending on the specific driver events in each patient[16] . Since the start of the TCGA project , the catalogue of cancer driver genes has increased and become more accurate[17] thanks not only to the data generated by the project itself , but also to the development of multiple , complimentary algorithms that search for cancer driver genes using different approaches . For example , some of these methods identify cancer drivers by searching for genes with higher than expected mutation rates[18 , 19] , whereas others identify genes that tend to accumulate damaging mutations[20] or contain regions with an unusually high proportion of mutations[21 , 22] . Nevertheless , the catalogue of cancer driver genes is far from complete and , because of extreme mutation diversity , it is hard to extend it by simply increasing the size of the datasets[19] . A complementary approach towards that goal is to use methods that integrate cancer mutation profiles with other types of biological knowledge to increase the statistical power of the analysis . For example , by integrating the information on the mutation profile of cancer patients with biological networks we can identify pathways and protein complexes that are recurrently mutated in cancer and are , therefore , likely drivers[23] . Note that these complexes can only be identified as drivers when adding the signals of all the components , because each individual protein is rarely mutated and , thus , missed by standard gene-centric approaches . In fact , a recent paper describes the crucial role played by the network topology in the final phenotypic effect of apparently deleterious mutations[24] . Similarly , we can include information on the structure of the protein coded by genes being analyzed to check enrichment in cancer mutations in specific structural regions[22 , 25–27] . The underlying idea for this approach is that genes ( and the proteins they encode ) are not monolithic entities , but instead consist of different regions usually responsible for different functions . In that context , it is possible that a given protein acts as a driver only when a specific region is mutated . This idea can be exploited to identify cancer driver genes by analyzing the distribution of mutations within a gene and looking for regions with unusually high mutation rates . Such fine grain approaches are not only capable of finding novel cancer drivers , but they also can help explain some of the variability between tumors or cancer cell lines apparently driven by the same gene[28] . We have previously developed an algorithm , e-Driver , which exploits this feature to identify cancer driver genes based on linear annotations of biological regions such as protein domains[22] . Despite encouraging results , the algorithm still had some limitations , as many structural features , like protein interaction interfaces , may be discontinuous at the sequence level and , hence , can not be analyzed without explicit use of protein structure information . Here we introduce an extended version of e-Driver that uses information on three-dimensional structures of the mutated proteins to identify specific structural features . Then , the algorithm analyzes whether these features are enriched in cancer somatic mutations and , therefore , are candidate driver genes . While technically the analysis can be applied to any structural feature or region , here we focus our attention on protein-protein interaction ( PPI ) interfaces . Many known cancer driver genes are located in critical regions of the PPI network ( interactome ) , usually in network hubs or bottlenecks[29] , warranting closer investigation of interaction interfaces . Moreover , while it is known that many cancer somatic alterations alter PPI interfaces , either destroying existing interactions or creating new ones[30–32] , this question has never been systematically analyzed across all known cancer somatic mutations with the specific goal of finding protein interfaces enriched in cancer mutations . Our analysis identified PPI interfaces enriched in somatic cancer mutations in a total of 103 genes ( interface driver genes ) . Thirty-two of these are well-known cancer driver genes , which are strongly enriched in somatic missense mutations and were previously identified using other algorithms and approaches . We also found that interface driver genes have an unusually high number of interactions in all known PPI interaction network models . This effect is especially pronounced for the 32 known cancer drivers , not only when compared to the rest of the genes in the interactome , but also when compared to non-interface cancer driver genes . The role of the remaining 71 genes as cancer drivers will obviously have to be verified experimentally , though we find some attributes as well as literature links that , albeit indirectly , support the prediction of them being cancer drivers . Interestingly , many of the new putative “interface driver genes” are involved in the immune response , particularly in HLA-like and complement systems . The role of the immune system in cancer treatment and evolution is gaining increasing attention[33 , 34] and our results provide new details regarding which interactions seem to be most affected by somatic mutations . Finally we show how , in many cases , depending on which interface or protein region is altered , tumors apparently driven by the same cancer gene might have radically different behaviors and patient outcomes .
We assembled a data set consisting of 5 , 989 tumors from 23 cancer types from The Cancer Genome Atlas[35] ( Table A in S1 Table ) . The number of samples per tumor type ranged from 56 for uterine carcinosarcoma , to 975 for breast adenocarcinoma ( Fig A in S1 Text ) . Consistent with previous reports[36] , the average number of missense mutations per sample is highly variable among cancer types ( Fig B in S1 Text ) , with melanoma having the highest ( 429 missense mutations per sample ) and thyroid carcinoma the lowest ( 11 missense mutations per sample ) . We then compiled a list of currently known , high-confidence PPI interfaces using 18 , 651 protein structures downloaded from PDB ( Online methods ) . In short , we defined a PPI interface as the set of residues from a given chain that are within 5 angstroms of any residue from a different chain in the same set of PDB coordinates ( Fig 1a ) . We identified 122 , 326 different PPI interfaces between 70 , 199 PDB chains ( Online methods ) . Finally , we used BLAST to map the residues from the PDB datasets to gene sequences in the ENSEMBL human genome . Overall we mapped the PDB coordinates to 11 , 154 protein isoforms in 10 , 028 different human genes . The mapping covers roughly 30% of the human proteome ( measured per amino acid ) , with 6% of the proteome being mapped to at least one PPI interface . Mutations from all cancer datasets ( n = 868 , 508 ) are distributed randomly across the proteome , with approximately 30% of mutations ( n = 285 , 942 ) being in regions mapped to structures and around 6% in PPI interfaces ( n = 67 , 174 ) . However , in the case of known cancer driver genes [1 , 17] , regions covered by structures have between 20% and 60% more missense mutations than expected by chance ( Figs C and D in S1 Text ) , as we could map , on average , 40% of all mutations in known driver genes to a structurally solved region . This enrichment , while variable and dependent on the cancer type , is even higher , between two and three-fold , in regions involved in the PPI interfaces . For example , PPI interfaces from cancer driver genes in breast adenocarcinoma , glioblastoma , lower grade glioma rectal adenocarcinoma or uterine carcinosarcoma ( Fig D in S1 Text ) have more than three times as many mutations as would be expected by chance . These results strongly suggest that , indeed , mutations in PPI interfaces play key roles in carcinogenesis . To analyze the potential role of mutations in other proteins in cancer development , we used e-Driver to analyze individual PPI interfaces for all human proteins in each of the 23 individual cancer projects , as well as in the Pan-cancer dataset consisting of the combination of all of them . Briefly , e-Driver compares the observed number of mutations in a specific protein region with the expected value based on the ratio between the length of the given region and the length of the protein . We had previously used e-Driver to analyze the distribution of cancer somatic mutations in PFAM domains and intrinsically disordered regions and showed , for example , that different domains in the same protein can drive different types of cancer[22] . Here , we adapted e-Driver to analyze features that are discontinuous along the protein sequence , such as PPI interfaces identified from 3D structures of protein complexes . The whole process is exemplified in Fig 1 for PIK3R1 and its interaction interface with PIK3CA . We identified a total of 103 interface driver genes in either one of the cancer projects or in the Pan-cancer analysis ( FDR < 0 . 01 , Figs 2 , 3 and Tables B-Z in S1 Table ) . There is significant overlap between the genes identified in this analysis and lists of known cancer genes . For example , 32 interface driver genes ( 31% ) are included in either a list of high-confidence driver genes derived from previous analyses of TCGA data[17] or are part of the Cancer Gene Census[1] ( p < 1e-10 , odds ratio 9 , when only taking into account genes with structurally solved regions ) . To further validate our findings we repeated the analysis using the PPI interfaces from Interactome3D[37] . While the pipeline used to define the interfaces in Interactome3D is different than the one that we used and the coverage is lower , the resulting list of driver interface genes is very similar ( Figs E-G in S1 Text and Table AE in S1 Table ) , supporting the robustness of our analysis . Note that , as expected , there are many known cancer driver genes ( n = 433 ) that are not picked up by our analysis . These genes might not have been identified either because their mechanism of action does not involve perturbing specific PPI interface , but also because we currently may not have a structure with a PPI interface to match to them and , thus , they were not included our analysis . For example , we can map only 40% of all the mutations in known driver genes to 3D structure models . Many of the remaining 60% of mutations might also be altering interactions , but we will not know that until we increase the structural coverage of the human proteome . Some of the driver interfaces identified here contain known cancer hotspots . For example , NFE2L2 , a gene involved in cancer progression and drug resistance , is usually activated by mutations that disrupt the interaction with its repressor KEAP1 . We mapped 36 mutations from NFE2L2 to the structure showing its interaction with its repressor KEAP1 ( PDB 2FLU , shown in Fig 2 ) . In agreement with previous observations[15] , all but two of the mutations ( 94% ) in NFE2L2 involve interface residues , likely disrupting the interaction between the two proteins and activating NFE2L2 . Our results also highlight similarities and differences across related driver genes . For example , receptor tyrosine kinases , particularly members of the ERBB and FGFR families , are mutated in many cancers and frequently act as drivers . We found two ERBB proteins , ERBB2 and EGFR , among the interface driver genes . These two proteins are both strongly enriched in mutations in their dimerization interfaces , while the ligand-binding region is rarely mutated ( Fig N in S1 Text ) . We also identified two proteins from the FGFR family: FGFR2 and FGFR3 . Again , these two proteins have similar mutation profiles , with both proteins having most of their missense mutations in the region that interacts with the ligand , while leaving the dimerization interface intact . This , however , contrasts with the mutation pattern of the ERBB receptors , where , as we have explained , the ligand-binding region is rarely mutated . Since some of the most successful therapeutic antibodies against EGFR target the dimerization interface identified by our method , it is possible that antibodies against FGF receptors need to target the ligand-binding region in order to be successful[38] . Next we analyzed the 71 interface driver genes that are currently not classified as cancer drivers to determine their potential role in cancer . We found several results supporting our hypothesis that these genes can be cancer drivers . For example , many genes in our new-driver predictions are close network neighbors of known cancer drivers ( Fig L and M in S1 Text ) . A subset of them can also be identified by other established methods ( such as OncodriveFM[20] or OncodriveCLUST[21] , Tables AC and AD in S1 Table ) . Furthermore , in several cases there is extensive literature and biological evidence supporting this hypothesis . This is the case , for example , for ARGHAP21 . This protein is a small Rho GTPase that is suspected to play a role in epithelial-mesenchymal transition[39] and interacts , probably through the interface identified by e-Driver , with the known oncogene ARHGAP26 . Another subset of these 71 potential new cancer driver genes has functions related to immunity . Given the growing body of evidence showing that the immune system plays a key role in cancer progression and patients outcomes[34 , 40] , we analyzed these interfaces in more detail to try to find novel insights about the interplay between tumors and immune cells . For example , a recent pan-cancer analysis identified a subnetwork of proteins around HLA class I proteins as being recurrently mutated in cancer[23] . Our analysis also identified several antigen-presenting molecules as potential cancer drivers , including one class I ( HLA-C ) , one class II ( HLA-DRB1 ) , and three HLA-like proteins ( CD1C , CD1E and MR1 ) . Note also that HLA-C has been recently identified as a likely driver in head and neck cancer[15] . Another interesting group of immune-related proteins identified in our analysis include several elements of the complement cascade ( C3 , C4B and C5 ) or complement regulators and inhibitors ( CFHR4 , CFI and CPAMD8 ) . The complement molecules C3 and C4 have been previously associated with cancer progression and activation of PI3K signaling[41] , whereas C5a is suspected to inhibit CD8 lymphocytes and natural killer ( NK ) cells , a subset of immune cells involved in the immune response towards tumors [42] . Our analysis with e-Driver not only supports the role of these proteins in cancer , but also suggests a specific mechanism for that role . Cancer driver genes are known to occupy critical positions in the interactome , as well as having more interactions and higher betweenness than the average gene[29] . Since our method identifies additional cancer driver genes , we hypothesized that they would have similar network positions as known cancer drivers . To test this hypothesis , we measured the degree and betweenness centrality of the interface driver genes in 16 different protein interaction and functional networks from 7 different sources ( Fig 3 and Figs H-K in S1 Text ) . In all but one of the networks interface driver genes correlated with higher degrees even after correcting by confounding variables such as number of publications citing the gene[43] , whether the gene had a PDB structure or not , or if the gene is a known driver ( Table AF in S1 Table ) . Interface driver genes also correlated with higher betweenness in all but 4 of the networks , after correcting by all the aforementioned variables . Remarkably , while interface driver genes have similar network properties to known cancer drivers ( Fig 3b , Table AI in S1 Table ) , genes that belong to both groups ( i . e . known drivers with interfaces enriched in somatic mutations ) are located in even more critical positions of the network ( Fig 3c , Table AJ in S1 Table ) . These results are consistent with the hypothesis that the main driver mechanism of the interface driver genes , particularly those with strong driver signatures that are picked by multiple methods , is the alteration of the PPI interfaces and the interactions they mediate . Even if the genes that we identified have more mutations than expected in some of their PPI interfaces , there are tumor samples with mutations in other regions of the same genes . With that in mind , we wondered if there are consistent differences between cancer samples belonging to each of these two groups . To explore this issue , we first used proteomics data[44] and compared the expression levels of different proteins in tumors with mutations in the predicted driver interfaces to that of tumors with mutations in other regions of the same gene . To limit the impact of intrinsic tissue-variability in the protein expression levels , we limited our analysis to tissue-specific driver interfaces . Though we could not analyze most of the interfaces due to lack of statistical power ( there were not enough samples with proteomics data in both groups ) , we did find some interface-specific protein changes . For example , glioblastoma samples with mutations in EGFR’s dimerization interface have higher levels of both EGFR and phosphorylated EGFR ( Y992 and Y1173 ) proteins than patients with other EGFR mutations ( Fig 4 ) , suggesting that EGFR signaling is stronger in these patients . Note that these results also agree with the hypothesis that the main molecular mechanism driving cancer in these genes is the disruption of certain interactions , as cancer cells have different signaling levels depending on whether the gene is mutated in the identified driver interfaces or in another region . Another example of interface-specific protein expression changes comes from TP53 and its interface with SV40 ( Fig 5 ) . Note that this interface is the same as the one that TP53 uses to dimerize and bind to DNA ( Fig 5b ) . Patients from eight different cancer types ( bladder , breast , colon , endometrial , glioma , stomach , lung and head and neck ) with mutations in this interface had significantly higher levels of TP53 protein than those with other or no TP53 mutations ( Fig H in S1 Text ) . Moreover , patients with breast cancer had significantly worse outcomes ( Fig 5d and Tables AG and AH in S1 Table ) , suggesting that these mutations are more aggressive than other mutations in TP53 and that maybe different therapeutic approaches are needed in these cases . The association was observed also after correcting by patient age , though because of insufficient statistics we were not able to test other potentially confounding variables such as ER status . Note that traditional gene-centric analyses or the previous version of e-Driver cannot find these differences among patient subpopulations .
In this manuscript we have explored the role of missense mutations in PPI interfaces as cancer driver events using our e-Driver algorithm and the mutation profiles of 5 , 989 tumor samples from 23 different cancer types . Though the interaction interfaces of many cancer driver genes have been studied before[45 , 46] , this is the first time that three-dimensional protein features , such as PPI interfaces , have been systematically used to identify driver genes across large cancer datasets . Previous large scale analyses are either limited to linear features[22 , 24 , 25 , 47] , or are not based on known functional regions in three-dimensional structures but , instead , identify de novo three-dimensional clusters of mutations[26] . Our analysis identified several driver PPI interfaces in known cancer driver genes , such as TP53 , HRAS , PIK3CA or EGFR , proving that our method can find relevant genes and that alteration of interaction interfaces is a common pathogenic mechanism of cancer somatic mutations . In fact , we found that cancer driver genes , as a group , are strongly enriched ( over two-fold in most cancer types , and over three-fold in some cases ) in mutations in their PPI interfaces . Moreover , there is a strong correlation between the fact that a cancer driver gene is recurrently mutated on its PPI interfaces and how critical it is to the stability of the interactome in terms of both number of interactions and network betweenness . We also identified a series of driver interfaces in genes that are currently not known as cancer drivers . Some of these genes interact with known cancer drivers or are implicated in key cancer functions , suggesting that they are , indeed relevant to carcinogenesis . Another group of potential cancer drivers identified here are proteins involved in the immune system . With the growing appreciation of the importance of malfunction of the immune system in allowing cancer progression[34 , 48 , 49] , the immunity genes identified here can be used to develop a series of specific hypotheses of how modifications of their interaction patterns many modify immunity response to specific cancers . Analysis of all the genes with cancer driver interfaces identified in this work is ongoing in our labs , but in the meantime we provide a complete list of such genes in the S1 Table and S1 Text , as well as in our on-line resource Cancer3D[50] , inviting other groups to analyze , confirm or refute our predictions . It is important to note that the analysis presented here was limited to high quality interfaces , predicted either from solved structures or from high quality homology models . However , about 70% of the human proteome currently has no high quality structural coverage . This fraction of the proteome includes both low complexity or disordered regions , and protein regions without reliable templates to model their 3D structures . Also , structures of many complexes are still unknown . In these cases , even if we know the structures of the subunits , we cannot define the PPI interfaces and these proteins were not included in our analysis . Finally , even though we did not explore this issue here , there are other mutations that can have an impact on PPI interfaces , such as in-frame indels or silent mutations[51] . Therefore , the results presented here represent only the tip of the iceberg of what can be achieved by including structural data in the analysis of cancer mutation profiles . We expect that our method will improve not only as more cancer genomes are added to existing repositories ( increasing the statistical power of the analysis ) , but also as the structural coverage of the human proteome increases . We expect such increase to come from both new experimentally determined structures in public databases and the use of better modeling tools[52 , 53] . Another important results of our analysis is that we found that tumors with mutations in the same driver gene can have surprisingly different behavior and outcomes depending on the specific PPI interface affected by the mutation . This adds to a growing body of evidence suggesting that the current gene-centric paradigm in biology , while successful in some cases , will probably not be enough to explain the complex genotype-phenotype relationships underlying a vast array of complex traits[54–59] . In the case of cancer , for example , it is known that the two most common mutations in PIK3CA , E545K and H1047L , contribute to carcinogenesis through different mechanisms[45] . The same is true for different types of mutations in KRAS[60] or , as we have shown here , for mutations in EGFR or TP53 . All of the above suggests that in order to predict the outcome of a patient or the best treatment option we will need to have more detailed knowledge about the consequences of a specific mutation than just the identity of the cancer driver gene where it is located . Such increase in detail and knowledge should include , in the case of missense mutations , not only information about the protein domain or PPI interface of the gene being altered , but also data about mutations in other regions of the network , as these can also influence the phenotype of a driver gene through synthetic interactions[61] . Finally , one must keep in mind that , while this work has focused on the analysis and interpretation of missense mutations , there are many other types of variations that can act as cancer drivers and have a significant impact in the outcomes of cancer patients . Examples include promoter mutations[62] , copy number variations[63] , silent mutations[51] or small insertions or deletions . It is likely that different types of variations of the same gene will have different consequences and , therefore , could need different therapeutic approaches . A clear example of this phenomenon are TP53-driven tumors . As we show here , it is likely that patients with missense mutations in this gene have different outcomes depending on the specific region of TP53 that is mutated , suggesting that they might need different treatments . Nevertheless , a patient whose tumor is driven by a TP53 copy-number loss might benefit from yet another therapeutic approach that would not help any of the above[64] . Therefore , in order to identify the optimal treatment of each patient we will need to integrate and properly analyze all molecular consequences of the different types of mutations present in its tumor .
All the supplementary information , the raw data and the algorithms used in this manuscript , as well as the results presented , can be downloaded from http://github . com/eduardporta/e-Driver . The link to the Dropbox folder containing all the raw data can be found in the README . md file . All the statistical calculations were done using R 3 . 1 . 0 . All figures have been generated using the R package “ggplot2” . We downloaded level 3 mutation data from the TCGA data portal ( https://tcga-data . nci . nih . gov ) for 5 , 989 tumor samples that belong to 23 different cancer types ( Table A in S1 Table ) . We then used the Variant Effect Predictor tool to derive the consequences of each mutation in the different protein isoforms where it mapped[65] . We used gene and protein annotations from ENSEMBL version 72 . We identified a total of 868 , 508 missense mutations in 19 , 196 proteins . Note that we only analyzed the longest isoform of each gene in order to minimize problems related to multiple testing . We identified 18 , 651 protein structures with multiple chains in PDB ( as of May 2014 ) . Then , we analyzed all such structures to find the residues implicated in PPI interfaces . To that end , we defined a protein-protein interface in a chain as all the residues with a heavy atom within 5 angstroms of another heavy atom from a different chain , an intermediate value between the 4 and 6 angstroms seen in other references[31 , 66] . If a chain was in contact with multiple other chains , we defined a different interface for each chain-chain pair . Note that any specific interface does not have to be linear in sequence and that the same residue can be involved in multiple interfaces from different structures . The complete dataset containing all the PPI structures and models from Interactome3D was downloaded on April 30th 2015 . As described previously , protein-protein interfaces are defined as those residues in a chain whose atomic distance falls within 5 angstroms from the partner chain . Since Interactome3D uses Uniprot protein sequences while we use ENSEMBL , we had to map the coordinates from one to the other . In order to do that we compared the two sets of sequences and kept only those interfaces in Uniprot proteins whose sequence matched exactly a protein from ENSEMBL . While this reduced significantly the number of potential interfaces from Interactome3D , from 26 , 383 interfaces to 11 , 169 , it ensured that the results obtained with each dataset would be comparable . The mapping between ENSEMBL and PDB is the same as the one used in Cancer3D . Briefly , we queried the full PDB ( March 2014 ) , including non-human proteins , with every protein from ENSEMBL using BLAST . Every time we identified a PDB-ENSEMBL pair with an e-value below 1e-6 we used the BLAST output to map the residues from the ENSEMBL sequence to the PDB structure[50] . We used e-Driver[22] to identify interfaces that are enriched in somatic missense mutations . The algorithm calculates the statistical significance of deviation from the null hypothesis that the mutations are distributed randomly across the protein using a right-sided binomial test: P ( MR , MT ) = ( MTMR ) ( PMutReg ) MR ( 1−PMutReg ) MT−MR Where “PMutReg” is the ratio between the number of residues involved in the interface and the number of aminoacids in the entire protein , “MR” is the number of mutations in the interface and “MT” is the total number of mutations in the protein . Since it is possible that only a fraction of the protein is covered by the structure , we adjusted the algorithm to limit all the parameters to the structure-mapped region of the protein ( for example “MT” refers to the total number of mutations in the region of the protein covered by the specific structure being analyzed , not the absolute total of mutations in the protein ) . The final step consists in correcting all the p values for multiple testing using the Benjamini-Hochberg algorithm . We considered as positives all of the interfaces with a q value below 0 . 01 . Given the large number of available human PPI networks and the variability in their quality , we decided to use 16 different networks from 7 different sources ( Figs H-K in S1 Text ) : HPRD[67] , Biogrid[68] , STRING[69] , HumNet[70] , PSICQUIC[71] , one PPI derived from unbiased experiments as well as curated literature[43] and another network derived from in silico predictions of PPI based on structures[72] ( which we will call “Kotlyar” from this moment ) . Three networks , STRING , HumNet and Kotlyar have scores that approximately correlate with the probability of the interaction being true . Therefore , we decided to divide these networks in different subsets , by selecting only those interactions above a certain threshold ( Figs H-J in S1 Text ) . We then calculated the different protein properties ( node degree and node and edge betweenness ) in each network using the R package “iGraph” . We identified several potentially confounding factors that could explain differences in network properties of the different proteins . For example , there is a clear bias introduced by the fact that we can only analyze proteins with structurally covered regions ( either by direct experimental structures or homology modeling ) . Another potential confounding variable is the number of publications of each protein , as it seems to correlate with the number of interactions[43] . In order to do that , we used the e-tools from Pubmed to retrieve the number of papers mentioning each gene symbol in the title or the abstract . We also took into account whether the gene is a known cancer driver or not , as cancer driver genes are also known to have high degree and betweenness centrality[73] . Finally , we fitted the aforementioned variables , as well as whether the gene was an interface driver or not , into a generalized additive model ( using the R package “gam” ) and calculated the correlation of each variable with the degree or betweenness centrality in every network . We measured the distance between known cancer driver genes and genes identified by e-Driver not yet known to play a role in carcinogenesis in the different networks . To that end we calculated the distance between both groups of genes using the random walk with restart ( RWR ) algorithm . The random walk on graphs is defined as an iterative walker’s transition from its current node to a randomly selected neighbor starting at a given source node . This algorithm has been extensively used for the predictions of disease-associated genes[74] as well as the analysis of cancer genomes[75–77] . It also allows the restart of the walk from the source nodes at each time with probability “r” . For a detailed explanation of the effects of different “r” values see Fig L and Fig M in S1 Text . The random walk is described by the equation: pt+1 = ( 1-r ) *W*pt+r*p0 Where W is a column-normalized adjacency matrix of the graph , pt is a vector in which the i-th element holds the probability of being at node i at time t and p0 is the initial probability vector . This vector has value 0 if the gene is not a known driver , and value 1/D if the gene is a known driver , where D is the number of known driver genes in the network . The algorithm iterates the equation until the L1 norm between pt and pt+1 is less than 10−6 . Then , we added the probabilities of all the candidate driver genes identified by e-Driver and compared it to 10 , 000 groups with the same number of random genes to calculate empirical right-sided p-values . We downloaded level 3 clinical and protein expression data , whenever it was available , from the TCGA data portal . Then , for each statistically significant interface , we classified each sample into one of three groups: samples with mutations in the interface , samples with mutations in other regions of the same protein , and samples with no mutations in that protein . Finally , in the case of proteomics data , we used a two-sided Wilcoxon test to identify proteins with statistically significant differences between the first group and the other two . As for the clinical data , we used the Cox proportional hazards model from the R package “survival” to estimate whether mutation of a specific interface was a predictive feature for survival ( p < 0 . 01 ) after correcting by age . | Until now , most efforts in cancer genomics have focused on identifying genes and pathways driving tumor development . Although this has been unquestionably a success , as evidenced by the fact that we now have an extensive catalogue of cancer driver genes and pathways , there is still a poor understanding of why patients with the same affected driver genes may have different disease outcomes or drug responses . This is precisely the aim of this work-to show how by considering proteins as multifunctional factories instead of monolithic black boxes , it is possible to identify novel cancer driver genes and propose molecular hypotheses to explain such heterogeneity . To that end we have mapped the mutation profiles of 5 , 989 cancer patients from TCGA to more than 10 , 000 protein structures , leading us to identify 103 protein interaction interfaces enriched in somatic mutations . Finally , we have integrated clinical annotations as well as proteomics data to show how tumors apparently driven by the same gene can display different behaviors , including patient outcomes , depending on which specific interfaces are mutated . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | A Pan-Cancer Catalogue of Cancer Driver Protein Interaction Interfaces |
Interest in larval source management ( LSM ) as an adjunct intervention to control and eliminate malaria transmission has recently increased mainly because long-lasting insecticidal nets ( LLINs ) and indoor residual spray ( IRS ) are ineffective against exophagic and exophilic mosquitoes . In Amazonian Peru , the identification of the most productive , positive water bodies would increase the impact of targeted mosquito control on aquatic life stages . The present study explores the use of unmanned aerial vehicles ( drones ) for identifying Nyssorhynchus darlingi ( formerly Anopheles darlingi ) breeding sites with high-resolution imagery ( ~0 . 02m/pixel ) and their multispectral profile in Amazonian Peru . Our results show that high-resolution multispectral imagery can discriminate a profile of water bodies where Ny . darlingi is most likely to breed ( overall accuracy 86 . 73%- 96 . 98% ) with a moderate differentiation of spectral bands . This work provides proof-of-concept of the use of high-resolution images to detect malaria vector breeding sites in Amazonian Peru and such innovative methodology could be crucial for LSM malaria integrated interventions .
The most widespread strategies to combat malaria rely on the distribution of long-lasting insecticide-treated nets ( LLINs ) [1] and the application of indoor residual spray ( IRS ) [2] that target endophagic and endophilic mosquito vectors . The decline in their efficiency is associated mainly with: a ) insecticide contact avoidance by early-exiting behavior of mosquitoes feeding indoors [3]; b ) increased outdoor feeding and transmission; c ) zoophilic behavior; and d ) insecticide resistance [4] . Regional and local mosquito populations in Latin America frequently display both exophagic and exophilic feeding preferences , reducing the usefulness of these two widely-accepted strategies [5] . The urgent need to redesign vector control tools for mosquito populations resistant to current interventions has led to the targeting of key environmental resources , increasing the relevance of larval source management ( LSM ) [5–7] . Gravid female Anophelinae have the potential to discriminate among water bodies and seek suitable breeding sites for oviposition , using visual and olfactory cues [8] . Therefore , knowledge of the characterization and identification of the most productive , positive water bodies would help to increase the impact of targeted larval mosquito control . The current measures associated with LSM are oriented toward the use of larvicides and biological control . LSM trials have been conducted in Africa in part because the habitats of African anophelines are well characterized; such trials have shown that larvicides can reduce malaria transmission from 70–90% [7] . In the neotropics , the efficacy of larval control using Bacillus sphaericus against Nyssorhynchus darlingi ( formerly Anopheles darlingi [9] ) was evaluated in gold-mining pools [10] and in fish ponds [11] in the Brazilian Amazon . However , few studies have been performed in natural breeding sites [12 , 13] . Two examples of studies highlighting successful larval control in natural breeding sites are one that employed B . sphaericus against Nyssorhynchus aquasalis in Venezuela in brackish mangroves [14] and another that implemented larvivorous nematodes in Colombia [15] . There are several impediments to identifying Ny . darlingi breeding sites in the Amazon basin . For example , potential breeding sites are periodically flooded , making field surveys difficult [16]; sometimes natural breeding sites are nearly impossible to detect visually by ground-truthing due to extensive , dense vegetation . Nyssorhynchus darlingi is the primary malaria vector across the Amazon basin , accounting for up to 85% of the Anophelinae fauna feeding on humans [17–19] . This species is behaviorally very plastic , mainly biting and resting outdoors ( exophily ) with fewer reports of endophily ( indoor resting; reviewed in [20] ) , and simultaneous endophagy and exophagy ( reviewed in [21 , 22] ) . In Amazonian Peru , there are regional records of both endo- and exophagy [17 , 23] , including behavioural shifts presumed to be in response to the implementation of LLINs [24] . In this region , mosquito abundance is linked to river levels [17 , 25] , which rise substantially during the rainy season , providing female mosquitoes with innumerable water bodies suitable for oviposition . However , in some specific situations , floods have been reported as one driver of Ny . darlingi population elimination [26] . Nyssorhynchus darlingi colonizes diverse water bodies , contributing to dispersal and diversification across its broad range , from natural areas [12 , 13 , 21 , 22 , 27 , 28] to artificial ( human-made ) such as fish ponds , agricultural settlements , highways , mining sites and urban areas [29–31] . Sun exposure has been denoted as one of the determinant variables affecting oviposition site suitability , together with presence of water plants and secondary vegetation , green algae and reduced water current [12 , 22 , 28 , 32] . Malaria transmission in the Peruvian Amazon is highly heterogeneous . Loreto Department ( northeastern Peru ) reports the vast majority ( >95% of national cases; e . g . 53 , 163 of 55 , 210 in 2017 ) of the malaria cases in the country , with an estimated proportion of 80% Plasmodium vivax and 20% P . falciparum [33 , 34] . However , there are areas punctuated by transmission pockets that account for most cases in the Department [34 , 35] . Transmission occurs mainly during the rainy season , January—June , linked to river levels and mosquito abundance [23 , 25 , 36] . Parker and collaborators [37] demonstrated that high human biting rates ( HBR ) , entomological inoculation rate ( EIR ) , and infectivity of Ny . darlingi are a signature of remote riverine malaria hot spots and hyperendemicity in certain areas of the Peruvian Amazon , revising previous assumptions that transmission is hypoendemic throughout the peri-Iquitos region [17 , 29 , 38] . Classical survey techniques of larval habitats , in general , achieve small spatial coverage , limiting research on Anophelinae breeding sites , i . e . , extended water bodies over large areas are not practical to survey from the ground due to the complex landscape and dynamic nature of such water bodies . Several studies have demonstrated the capability of satellite imagery to detect large Ny . darlingi breeding sites in several countries [39–41] . However , the spatial resolution of public ( ~30 meters/pixel ) or private ( ~1 meter/pixel ) satellite imagery is inadequate due to the high vegetation coverage and/or the quality of images related to climatic conditions in the Amazon Region , particularly during the extensive rainy season . Although there are applications for Unmanned Aerial Vehicles ( UAVs a . k . a . drones ) across many fields , such as monitoring crops [42] and forest [43] , few researchers have taken advantage of this technology to investigate anopheline breeding sites linked to transmission pockets . Two recent studies have used UAVs to map land use and Anopheles gambiae breeding sites [44 , 45] and to link malaria epidemiology with landscape ecology in Thailand [46] . Nevertheless , no parallel studies have been conducted in the Amazon Basin , which is operationally challenging with a considerable amount of potential Anophelinae larval habitat , especially during the rainy months . The current study explores the use of drones for mapping water bodies in four rural villages in the Peruvian Amazon . Our main objective was to provide proof-of-concept of the suitability of high-resolution imagery ( RGB band ) to map Ny . darlingi aquatic habitats . Multi-spectral imaging data ( including the normalized difference vegetation index- NDVI ) was used to achieve sufficient resolution to identify water bodies potentially colonized by Ny . darlingi . The public health-oriented deployment of this approach to identify and target water bodies for use in LSM campaigns is discussed . The data here allow us to postulate that , in combination with existing vector interventions such as LLINs and IRS , drones could be an attractive additional tool for malaria elimination in the Amazon and other places where mosquito behavior and larval breeding sites remain difficult to locate and identify .
Study protocols were approved by the Ethics Review Board of the Regional Health Directorate of Loreto ( 477–2016 ) , Universidad Peruana Cayetano Heredia in Lima ( 184-09-16 ) and WHO Ethics Review Committee ( 0002669 ) . These requirements were established by TDR/WHO despite the absence of human subject involvement in the present work . All the methods were carried out in accordance with the approved guidelines . The study was conducted in the Mazan district ( Maynas Province , Loreto Department , Peru ) that has been identified as a very high-risk district for malaria transmission [47] . To represent as broadly as possible the landscape of this area , four communities were selected in two ecologically different river microbasins in the Mazan district [48]; two communities in the blackwater Mazan River district: Visto Bueno ( 3 . 449° S , 73 . 317° W; population = 60 ) and Libertad ( 3 . 496° S , 73 . 234° W; population = 345 ) ; and two in the whitewater Napo River: Salvador ( 3 . 445° S , 73 . 154° W; pop = 431 ) and Urco Miraño ( 3 . 361° S , 73 . 064° W; pop = 240 ) . Map in Fig 1 was produced with QGIS 2 . 16 ( QGIS Development Team , 2016 . QGIS Geographic Information System . Open Source Geospatial Foundation Project ) and based on public geographical data from OpenStreetMaps ( www . openstreetmap . org ) . Detailed characteristics of these communities have been described elsewhere [47] . Mazan is a district in Loreto with sustained annual malaria transmission . The Regional Health Directorate of Loreto ( RHDL ) reported 1061 cases in 2016 caused mainly by P . vivax ( 68 . 5% ) and P . falciparum ( 31 . 5% ) , equivalent to an Annual Parasite Index ( API ) of 78 . 9 cases per 1000 inhabitants . The RHDL passive case report is based exclusively on light microscopy and some studies demonstrate a large sub-microscopic malaria reservoir [16 , 38] . In this area , a seasonal pattern of increase during the rainy season was observed in both malaria cases and vector abundance ( predominantly Ny . darlingi ) [25 , 37] . Drone surveys were carried out in the four communities between April 17 and 23 , 2017 . Mapping based on RGB and multispectral imagery was conducted simultaneously . In each community , water bodies were inspected at three time points—in September and November 2016 ( dry season ) and March 2017 ( rainy season ) —for the presence of Ny . darlingi immature stages; then , data from water bodies were available six months prior to the drone surveys . Drone surveys were carried out using a DJI Phantom 4 Pro ( DJI , Shenzhen , China ) quadcopter fitted with a DJI 4K camera ( 8 . 8 mm/24 mm; f/2 . 8; 1'' CMOS; 20 MP ) for conventional RGB imagery collection and a 3DR Solo ( 3D Robotics , California , US ) quadcopter fitted with a Parrot Sequoia sensor ( Parrot , France ) which is composed of single-band cameras ( Green , Red , Red Edge and Near Infrared—nir ) of 1 . 2 MP for multispectral imagery collection . The flight plan was programmed with Pix4D Capture app in an iPad Mini 4 ( Apple , California , US ) . The connection between the controller and DJI Phantom 4 Pro and 3DR Solo was set up using DJI GO 4 app and 3DR Solo app , respectively . For RGB mapping , in each community the DJI Phantom 4 Pro drone was flown to an altitude of approximately 100 m , which gave a ground sampling distance ( GSD ) or spatial resolution of 0 . 1 meter/pixel . Grids of 500m x 500m were drawn in Pix4D . Households and a buffer of at least 250m were covered using several grids in each community: 4 in Visto Bueno , 10 in Libertad , 9 in Salvador , and 8 in Urco Miraño . In each grid , 100 waypoints were automatically calculated to ensure an overlap of at least 70% between neighboring images , necessary to generate an orthomosaic [49] . The flight plan was preloaded onto the DJI Phantom 4 Pro drone and the flight path was followed automatically . A flying time of ~30 minutes without a change of battery was required to complete the survey in each grid . Multispectral mapping was conducted over 16 randomly sampled water bodies ( 51 . 6% of water bodies inspected for Ny . darlingi larvae during the study ) , located as follows: 5 in Visto Bueno , 2 in Libertad , 4 in Salvador , and 5 in Urco Miraño . In each water body , the 3DR Solo drone was flown to an altitude of approximately 50m , which assured a GSD of 0 . 02 meter/pixel . A grid of 200m x 200m was drawn in Pix4D and the Sequoia multispectral camera was set up to take an image each second during the 20-minutes flight time of the 3DR Solo drone . The image classification was conducted in Google Earth Engine ( GEE ) [52] . Briefly , GEE is a cloud-based platform for planetary-scale geospatial analysis that brings Google’s massive computational capabilities to bear on a variety of high-impact societal issues including deforestation , drought , disaster , disease , food security , water management , climate monitoring and environmental protection . It is unique in the field as an integrated platform designed to empower not only traditional remote sensing scientists , but also a much wider audience that lacks the technical capacity needed to utilize traditional supercomputers or large-scale commodity cloud computing resources [5] . All classification analyses were conducted in the online Integrated Development Environment ( IDE ) at https://code . earthengine . google . com ( repositories for data and code available in Supplementary information ) . All 8-band multispectral orthomosaics were uploaded to GEE assets and a supervised classification was performed using a Random Forest ( RF algorithm in GEE ) [53] . RF is a collection of decision trees , also called CART ( Classification and Regression trees ) that has been widely used for mapping land cover in general . This method aims to associate specific targets with specific values of a particular variable; the result is a decision tree in which each part identifies a combination of values associated with a particular prediction [6] . The RF algorithm in GEE was set to 500 trees for each classification and was conducted using all bands in the 8-band orthomosaics as input . Default GEE parameters were used for the RF classification as follows: cross-validation factor for pruning = 10; maximal depth level of initial tree = 10; minimal leaf population = 1; minimal split population = 1; minimal split cost = 1e-10; whether to impose stopping criteria while growing the tree = false; quantization resolution for numerical feature = 100; quantization margin = 0 . 1 . RF classification use pre-labeled data as input . A dataset of polygons was constructed for each community in the study area , of which 480 were on-ground polygons and 240 were on-water polygons . Each class was composed of 30 samples per community , in total 120 samples per class . The total number of polygons per approach are presented in S1 Table Classes ( or attributes ) of on-ground polygons were labeled by in situ and ground inspection , whereas the on-water polygons classes were labeled using the results of the larvae sampling at the study area . For the classification , a water body was considered consistently positive if Ny . darlingi larvae were registered in 50% or more of the total visits and negative if Ny . darlingi larvae were recorded in less than 50% of the visits . In other words , if the water body was positive at least in 2 out of 3 or 1 out of 2 visits , the water body was considered consistently positive for Ny . darlingi . Three approaches were used for the spatially explicit land cover classification: ( 1 ) a classifier with particular focus on identifying water bodies placing the orthomosaics into five groups: low vegetation , high vegetation , bare soil , urban and water bodies; ( 2 ) a classifier with a particular focus on differentiating water bodies with presence or absence of Ny . darlingi larvae , classifying the orthomosaics into six groups: low vegetation , high vegetation , bare soil , urban , water bodies positive for Ny . darlingi and water bodies negative for Ny . darlingi; and ( 3 ) a classifier with a particular focus on differentiating water bodies as positive or negative for Ny . darlingi classifying only the water bodies detected in approach 1 into two groups: water bodies positive and negative for Ny . darlingi . S1 Fig presents the workflow diagram of the three approaches . A k-fold cross validation was carried out to evaluate the performance of the RF classifier [54] , thus , polygons served as training and validation samples . Briefly , all samples were randomly divided into k subsets ( groups ) , for this study k was set to 5 ( S2 Fig ) . The classifier was trained using four ( k-1 ) groups and then tested with the remaining one . This procedure was repeated k times until all groups were used as a testing group . For each set of 4 training groups , the accuracy was calculated in the testing group . The mean accuracy of the k sets was considered as the overall accuracy ( OA ) . In order to assess the probability distribution of the overall accuracy , the k-fold cross validation was repeated 999 times , where on each iteration a new random sample of polygons was assigned to each k-subset . Two additional performance measures were conducted , producer’s accuracy ( PA ) , also called sensitivity , and consumer’s accuracy ( CA ) , alternatively called positive predictive value ( PPV ) . In addition , to account for the spatial autocorrelation and lack of independence of polygons randomly selected at both training and test sets [55] a non-random groups assignment was conducted using the communities as natural groups ( k = 4 ) . In order to measure the statistical separability between positive ( aquatic habitats consistently harboring Ny . darlingi >50% of the time ) —and negative ( aquatic habitats consistently harboring Ny . darlingi < 50% of the time ) —water body classes in approaches 2 and 3 , an interclass separability analysis was conducted using the Jeffries Matusita ( JM ) distance . Briefly , JM is a measure of the average difference between two-class ( positive and negative water body ) density functions by pair-wise comparison and ranges between 0 and 2 [56] . A JM distance of 0 imply no separation and 2 for full separation between land cover classes . In addition , a Monte-Carlo coefficient/p-value/sample-size ( CPS ) sensitivity analysis was conducted . A complete description of the Monte-Carlo CPS is provided in the Supplementary Methods . All the implementations above were accomplished using R v . 3 . 4 . 3 ( R Development Core Ream , R Foundation for Statistical Computing , Australia ) .
From all water bodies inspected , 18 ( 58% ) were considered negative and 13 ( 42% ) consistently positive for the presence of Ny . darlingi immature stages . Of these , 16 ( 51 . 6% ) were inside the mapped area of the 8-bands multispectral orthomosaics , and 8 were consistently positive for the presence of Ny . darlingi larvae . From the 16 water bodies sampled and multispectrally mapped , 4 ( 25% ) provided information for only 2 of 3 collections because they were dry during 1 of the 3 visits , all of them in Visto Bueno . Importantly , none of the water bodies were dry during the drone survey . The proportion of water bodies positive for Ny . darlingi by community and survey is presented in S3 Fig for all water bodies inspected and for the 16 water bodies selected for multispectral mapping . Several images were used to build the orthomosaic in each community . There were 386 RGB images in Visto Bueno , 1020 in Libertad , 805 in Salvador , and 958 in Urco Miraño; and there were 3804 Multispectral images in Visto Bueno , 7080 in Libertad , 6980 in Salvador , and 6940 in Urco Miraño ( note that Parrot Sequoia captures 4 individual spectral band images per shot ) . An orthomosaic for each community is presented in Fig 2 , and the 3D models in S4 Fig . The high spatial resolution of the resulting orthomosaics allowed for a clear identification of water bodies via simple visual inspection . However , is important to notice the limitations of the Structure-From-Motion algorithm ( SfM ) in Photoscan to match points in complex canopy environments where there is too much texture , poor illumination , and/or insufficient unique features , resulting in some gaps observed in S4 Fig [57] . Mean values and the standard errors for each band at each community are presented in Table 2; the RGB bands values are presented in 8-bit and the multispectral bands are in 16-bit . A heterogeneous spectral profile was observed between communities ( Fig 3 ) , presumably due to different environment and land cover composition . An example of a landscape using RGB , Multispectral and NDVI for each community is presented in Fig 4 . Three approaches were used for the spatially explicit land cover classification in Google Earth Engine ( GEE ) . The classified images for each community using the first approach are presented in Fig 5a . This approach showed high accuracy for differentiating among 4 land cover classes ( bare soil , low- and high- vegetation , and urban ) and water bodies . After 999 iterations , the overall accuracy of approach 1 was 86 . 73% ( SE = 0 . 031 ) . Classification approach 2 includes the differentiation of water bodies based on the presence of Ny . darlingi in the previous 6 months , in addition to the 4 land cover classes used in approach 1 , with an overall accuracy of 87 . 58% ( SE = 0 . 029 ) ( Fig 5b ) . In approach 3 , the 8-band composite image was masked using the water class obtained in approach 1 . This approach shows the highest overall accuracy , with an average of 96 . 98% ( SE = 0 . 025 ) ( Fig 5c ) . The three approaches consistently depict highly heterogeneous land cover composition among the communities in the study ( Fig 6 ) . As these communities are located in the same district , this may reflect a high diversity of locations at the microgeographical scale where Ny . darlingi can breed . Regarding the classification with non-random subsets , using communities as natural groups , this resulted in a diminished overall accuracy for approach 1 and 2 ( 63 . 92% and 65 . 70% , respectively ) . However , approach 3 still showed a high overall accuracy ( 92 . 26% ) . The overall accuracy of random and non-random assignment cross-validations is presented in Table 3; producer and consumer accuracies of each class are presented in S2 Table for random assignment and S3 Table for non-random assignment . In approach 2 , the resulting number of pixels classified as positive water bodies was 31'717 , 931 and 44'391 , 373 pixels for negative water bodies . A higher number of pixels was included in the analysis of approach 3 , 35'211 , 614 for positive and 46'894 , 706 for negative water bodies . The mean , standard deviation and comparison of each band are shown in Table 4 for both approaches . Overall , JM distances of each band between positive and negative water body classes are very low . The highest values of JM were shown in green_m and red_m bands in both approaches ( Table 5 ) . Consistently , Monte-Carlo CPS sensitivity analysis show that bands green_m , red_m , but also NDVI , show a noteworthy effect size for approach 2 . Green_m and red_m show increased values in positive water bodies whereas higher values of NDVI were observed in negative water bodies . Interestingly , all bands except edge_red and nir were statistically meaningful in approach 3 . The bands that showed increased values in positive water bodies are green_m and red_m . Conversely , blue , green , red , and NDVI bands showed higher values in negative water bodies ( S5 and S6 Figs ) .
Difficulty in identifying and detecting Ny . darlingi breeding sites arises from vast and often difficult to access places where this species can successfully breed . The portability of UAVs allows investigators to navigate moderately hostile and complex environments , such as the Amazon Basin . This study assessed the feasibility of using UAVs to generate maps with a higher resolution compared to those available through satellites , mainly when the imagery required is specific to a local scale within a community or limited area of interest at a microgeographical scale . Previous studies also propose the use of UAV for mapping environmental risk factors for zoonotic malaria in Malaysia and Philippines [44] , and vector habitats in Zanzibar [45] . The current study proved that in addition to RGB imagery , multispectral imagery collection is also feasible in rural areas , and the addition of this information boosted the distinction of environmental characteristics of water bodies that harbor Ny . darlingi larvae [58–60] . Capturing data multiple times in longitudinal entomological surveys potentially would provide the tools to study Anophelinae breeding site dynamics [45] . For instance , the adaptation to more permanent anthropogenic larval habitats has been hypothesized to be the cause of a resident population of Ny . darlingi in Porto Velho ( Brazil ) , leading to a perennial presence of this species and probably promoting and maintaining continual Plasmodium transmission [61] . The data reported here classified Ny . darlingi -positive and -negative water bodies . A high concordance of location and extent of water bodies was observed in the three approaches applied . An average accuracy between 87% and 97% with a relatively narrow distribution demonstrates a valid strategy to identify and prioritize water bodies for outdoor interventions such as LSM [62] , microbial larvicides [63] , or attractive toxic sugar baits ( ATSBs ) [64] . As the implementation of this classifier harnessed Google’s cloud-computing platform , a short length of time is required to complete the classification , overcoming computing resource limitations [65 , 66] . Modifications of ecosystems and natural resources frequently contribute to the emergence and spread of infectious disease agents . Specifically , land use changes including deforestation , irrigation , wetland modification and road construction , among others , have been identified as major drivers of infectious disease outbreaks and also can interfere in their transmission dynamics [67] . Malaria has been associated with these anthropogenic alterations in Asia [68] , Africa [69] and Latin America [70] and of special concern is the creation of new breeding sites that may be increasing the proliferation of mosquitoes [71] . For example , Ny . darlingi uses a range of natural and artificial sites for breeding and is able to exploit highly diverse habitats [22 , 26 , 72] including deforested areas with substantial surrounding vertical vegetation [29 , 41 , 73] . Recently , fish farming has been promoted as a way to increase economic opportunity in rural localities in Brazil and Peru , and throughout Latin America . Unfortunately , these fishponds also provide ideal breeding sites for Ny . darlingi ( holding 4-fold more Anophelinae larvae than natural water bodies ) , demonstrating a rapid adaptability to some new environmental niches , associated with concomitant increases in malaria case numbers , e . g . , in Mancio Lima , Acre state , Brazil and along the Iquitos/Nauta highway , Loreto , Peru [41 , 74] . The use of imagery acquired from drones may be helpful for the detection of landscape modifications in a rapidly changing environment that can affect mosquito population distribution . A recent study described distinct Ny . darlingi populations related to urban or rural settlements in Acre , Brazil with different grades of anthropogenic landscape modification [75] . Here , deforestation was the most plausible cause for loss of genetic diversity in the mosquito populations . Modifications in landscape affects physicochemical characteristics and/or ecological communities of Anophelinae breeding sites and this also may affect malaria transmission dynamics . For instance , Plasmodium transmission potential , including survival and extrinsic incubation period , has been demonstrated to be affected by larval food quantity in Anopheles stephensi [76] . Furthermore , Anopheles coluzzii has different permissiveness to Plasmodium depending on the nature of the diet associated with microbiota composition [77] . The findings in this study suggest strong differential microenvironmental composition of Ny . darlingi breeding sites compared with other less favorable water bodies that could be assessed with the combination of RGB and multispectral imagery . These differences were evaluated by the inspection of certain bands of the spectral profile between communities and the resulting land cover classification discussed above . As these patterns were observed in four communities in two microbasins of the Amazon region , these findings may be generalizable in similar contexts elsewhere and denote heterogeneous environmental characteristics at a microgeographical scale [78] . As discussed previously , Ny . darlingi dominates all these diverse microhabitats in the communities under study . Moreover , Parker et . al . [37] reported that An . darlingi comprised the majority of the mosquitoes collected in 21 sites along approximately 100 km of the Mazan river microbasin . Knowledge of rapidly changing patterns of human settlements and vector distribution is vital for predicting disease risks and effectively targeting disease control measures . Interestingly , Libertad and Urco Miraño , the sites with the highest and the lowest proportion of area of water bodies with Ny . darlingi larvae , were reported as communities with very high and low malaria transmission , respectively [47] . In addition , the high heterogeneity in malaria incidence [47] reported in the Mazan and Napo river microbasins may have arisen in part from the highly heterogeneous environmental composition of each community and the productivity of Anophelinae in these habitat types [79] . Considering that this study was not designed to demonstrate any association between malaria risk and microhabitat composition of Ny . darlingi , further research is needed to obtain a time-series of high-resolution imagery to detect fluctuations in the spectral profile of aquatic habitats , leading to the development of accurate risk maps and to the identification of potential effects on subsequent local malaria transmission . Drone-based mapping could have a wider range of applications . For instance , high resolution digital elevation models ( DEM ) are useful tools to analyze watersheds and small streams [80 , 81] , favorable to Ny . darlingi breeding sites that are shaped by intermittent heavy rain [82] [83 , 84] . Moreover , these DEMs support the identification of seasonally flooded areas , common in the Amazon basin , that possibly increase human-mosquito contact and therefore are associated with a higher risk of malaria [82 , 85–87] . Importantly , canopy coverage prevents DEM reconstruction in forested areas due to SfM photogrammetric issues , in consequence DEM must rely on other sensors such as Laser Imaging Detection and Ranging ( LIDAR ) , that are more expensive and logistically demanding . However , photogrammetric-based DEM could still be useful for localized characterization of terrain in the forest fringes where Ny . darlingi demonstrates a breeding site preference in rural Amazon , [29 , 41 , 73] . In 2011 the Amazon river ( and tributaries in Iquitos , Peru ) experienced an unusual flooding event , a peak of the river level over 10 m , most likely associated with climatic events ( El Nino Southern Oscillation-ENSO ) , altering the temporality and characteristics of water bodies and resulting in a replacement event of Ny . darlingi populations [88] . In Surinam , abnormal flooding of rivers with subsequent inundation of larval habitats was reported as one of the factors that destroyed a local Ny . darlingi population ( together with ITN distribution and other interventions ) [26] . Another key benefit of the UAVs for high-resolution mapping is the rapid assessment of house positions . This approach offers the opportunity to pinpoint the GPS coordinates of several human dwellings with a high accuracy in a single flight path , rather than the more laborious ground inspection of each dwelling . Also , this technology can help epidemiologists to understand spatial malaria transmission and human travel patterns [47] . The present study showed that in addition to traditional RGB mapping , multispectral bands add critical information to differentiate water bodies ( independently , whether or not they harbor Ny . darlingi larvae ) , and other types of land cover in the Amazon Region . A limited set of low-cost cameras and drones were tested , therefore an evaluation of a wider range of commercially available options is recommended . Despite initial capital cost , scaling up of drone flights in multiple settings and times would require small investments . It is important to note the limitation of the extent of covered area with drone flights due to energy consumption; if large areas are required to be covered in a short time period , multiple drones would be necessary , increasing the cost of this implementation . Importantly , due to the abundance , tangled distribution , and unclear boundaries of the water bodies in the Amazon Region , the classification approach showed in this study could be preferred over manually delineation demonstrated in other settings [45] . The computing time of a single classification in GEE is less than a minute , however training and test sets are only applicable to the Peruvian Amazon Region . The addition of training and test sets of contrasting locations should be included to test transferability to a variety of scenarios . We recognize some potential shortcomings in this study . The equipment used was of the highest quality and lowest price on the market at the time of the field study; this strategy would be more cost-effective as the number of surveys increase . To overcome this , several projects in other fields are proposing to utilize low-cost non-commercial UAVs that may help to spread the strategy [89 , 90] . Another caveat is the limited flight time of the drone . Thus , several flights over the locality are required to obtain a single map , and may represent some deviation in the time between scenes of the unique map , with a potential effect on the spectral signature , although this is likely relatively minor . Also , the flight of a UAV requires a certain degree of expertise , however , steps in flight path automation will overcome this difficulty [91 , 92] . Despite our use of the recommended overlap percentage in this study , some gaps in final imagery through forest canopy and some water bodies , as observed in S4 Fig , may have affected the final classification . Regrettably , land cover classification using Google Earth Engine depends on an internet connection . With this in mind , transport of imagery in any physical storage unit to a point with a stable internet connection is feasible; however , the number and cost of storage units should be taken into account . Because this methodology is at an early stage , there is a lack of methods for rapid data processing and developing strategies to speed up image processing methods , but these are currently expanding . Multispectral camera calibration and climatic conditions , such as heavy rain , may also jeopardize imagery collection . However , in this study all flights were conducted during the same time range ( over a few days ) and under low cloud coverage and wind conditions to reduce the effect on the spectral signature . Overall the most important methodological caveat in this study is the definition of negative water bodies . Although we sampled only 8 negative and 8 positive water bodies for the presence of Ny . darlingi , the water body type included streams , fishponds and palm swamps and we sampled in two distinctive river microbasins . As this is a proof-of-concept study , future work should consider more frequent surveillance of these and additional water bodies from more communities and additional flights over the survey localities at different times of the day and under various atmospheric conditions . In summary , the use of high-resolution imagery can provide a better understanding of environment-related disease changes and can play a meaningful part in the development of decision-support tools . Our findings back the use of a low-cost UAVs and a freely available planetary cloud-based platform to achieve a highly accurate classification of the differential spectral signature of water bodies that harbor Ny . darlingi larvae and those that do not , in the Amazon region . This strategy might be generalizable to similar contexts elsewhere , resulting in new ways to control and survey malaria in affected settings , in combination with existing approaches . | The most efficient malaria vector in the Latin American region is Nyssorhynchus darlingi ( formerly Anopheles darlingi ) . In Amazonian Peru , where malaria is endemic , Ny . darlingi feeds both indoors and outdoors ( endophagy , exophagy ) , depending on the local environment , and rests outdoors ( exophily ) . LLINs and IRS , the most common tools employed for vector control , target endophagic and endophilic mosquitoes . Thus , they are only partially effective against Ny . darlingi . Control of the aquatic stages of vector mosquitoes , larval source management ( LSM ) , targets the most productive breeding sites nearest to human habitation . In four riverine communities , we used drones with high-resolution imagery as a key initial step to analyze water bodies within the estimated flight range of Ny . darlingi , ~ 1 km . We found distinctive spectral profiles for water bodies that were positive versus negative for Ny . darlingi . The methodology and analysis reported here provide the basis for testing whether LSM can be combined successfully with LLINs and IRS to contribute to the elimination of transmission in malaria hotspots in the Amazon . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"surface",
"water",
"rivers",
"engineering",
"and",
"technology",
"tropical",
"diseases",
"geographical",
"locations",
"parasitic",
"diseases",
"animals",
"cameras",
"peru",
"developmental",
"biology",
"aquatic",
"environments",
"habitats",
"bodies",
"of",
"water",
"optical",
"equipment",
"insect",
"vectors",
"hydrology",
"infectious",
"diseases",
"south",
"america",
"marine",
"and",
"aquatic",
"sciences",
"life",
"cycles",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"eukaryota",
"freshwater",
"environments",
"equipment",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"malaria",
"species",
"interactions",
"larvae",
"organisms"
]
| 2019 | High-accuracy detection of malaria vector larval habitats using drone-based multispectral imagery |
Myriapods ( e . g . , centipedes and millipedes ) display a simple homonomous body plan relative to other arthropods . All members of the class are terrestrial , but they attained terrestriality independently of insects . Myriapoda is the only arthropod class not represented by a sequenced genome . We present an analysis of the genome of the centipede Strigamia maritima . It retains a compact genome that has undergone less gene loss and shuffling than previously sequenced arthropods , and many orthologues of genes conserved from the bilaterian ancestor that have been lost in insects . Our analysis locates many genes in conserved macro-synteny contexts , and many small-scale examples of gene clustering . We describe several examples where S . maritima shows different solutions from insects to similar problems . The insect olfactory receptor gene family is absent from S . maritima , and olfaction in air is likely effected by expansion of other receptor gene families . For some genes S . maritima has evolved paralogues to generate coding sequence diversity , where insects use alternate splicing . This is most striking for the Dscam gene , which in Drosophila generates more than 100 , 000 alternate splice forms , but in S . maritima is encoded by over 100 paralogues . We see an intriguing linkage between the absence of any known photosensory proteins in a blind organism and the additional absence of canonical circadian clock genes . The phylogenetic position of myriapods allows us to identify where in arthropod phylogeny several particular molecular mechanisms and traits emerged . For example , we conclude that juvenile hormone signalling evolved with the emergence of the exoskeleton in the arthropods and that RR-1 containing cuticle proteins evolved in the lineage leading to Mandibulata . We also identify when various gene expansions and losses occurred . The genome of S . maritima offers us a unique glimpse into the ancestral arthropod genome , while also displaying many adaptations to its specific life history .
Myriapods are today represented by two major lineages—the herbivorous millipedes ( Diplopoda ) and the carnivorous centipedes ( Chilopoda ) , together with two minor clades , the Symphyla , which look superficially like small white centipedes , and the minute Pauropoda [8] . All are characterised by a multi-segmented trunk of rather similar ( homonomous ) segments , with no differentiation into thorax or abdomen . All recent studies , molecular and morphological , support the monophyly of myriapods [3]–[5] , [8]–[10] suggesting that they share a single common ancestor . Myriapods , insects , and crustaceans have traditionally been identified as a clade of mandibulate arthropods , characterised by head appendages that include antennae and biting jaws [11] . Some molecular datasets have challenged this idea , suggesting instead that the myriapods are a sister group to the chelicerates [12] , [13] . The most comprehensive phylogenomic datasets thus far reject this , and strongly support the phylogeny that proposes that the chelicerates are the most basal of the four major extant arthropod clades , and the mandibulates represent a true monophyletic group [3] , [5] , [10] , [14]–[17] . Within the mandibulates , myriapods were believed until recently to share a common origin with insects as terrestrial arthropods . This view , based on a number of shared characters including uniramous limbs , air breathing through tracheae , the lack of a second pair of antennae , and excretion using Malpighian tubules , was widely supported by morphologically based phylogenies [9] , [18] . However , molecular phylogenies robustly reject the sister group relationship between insects and myriapods , placing the origin of myriapods basal to the diversification of crustaceans [5] , and identifying insects as a derived clade within the Crustacea [19]–[21] . As crustaceans are overwhelmingly a marine group today , and were so ancestrally , this implies that myriapods and insects represent independent invasions of the land ( with the chelicerates representing an additional , unrelated invasion ) . Their shared characteristics are striking convergences , not synapomorphies . We chose S . maritima as the species to sequence partly for pragmatic reasons: geophilomorph centipedes , such as S . maritima , have relatively small genome sizes , certainly compared to other centipedes [22] . More importantly , it is a species that has attracted interest for ecological and developmental studies [23]–[25] , especially the process of segment patterning [26]–[32] . S . maritima is a common centipede of north western Europe , found along the coastline from France to the middle of Norway . It is a specialist of shingle beaches and rocky shores , occurring around the high tide mark , and feeding on the abundant crustaceans and insect larvae associated with the strand line . It is by far the most abundant centipede in these habitats around the British Isles , sometimes occurring at densities of thousands per square metre in suitable locations [25] . Eggs can be harvested from these abundant populations in large numbers with relatively little effort during the summer breeding season [27] . They can be reared in the lab from egg lay to at least the first free-living stage , adolescens I [24] , [33] . Some aspects of S . maritima biology are not common to all centipedes . Notable among these is epimorphic development , wherein the embryos hatch from the egg with the final adult number of leg-bearing segments . Epimorphic development is found in two centipede orders: geophilomorphs ( including S . maritima ) and scolopendromorphs . In contrast , more basal clades display anamorphic development and add segments post-embryonically [34] . These anamorphic clades have relatively few leg-bearing segments , generally 15 , while geophilomorphs have many more , up to nearly 200 in some species [6] . These unique characteristics probably arose at least 300 million years ago , as the earliest fossils of the much larger scolopendromorph centipedes date to the Upper Carboniferous [35] . These share the same mode of development as the geophilomorphs , and are their likely sister group . Geophilomorphs are also adapted to a subsurface life style , the whole order having lost all trace of eyes [36] , [37] , though apparently not photosensitivity [38] . We have sequenced the genome of S . maritima as a representative of the phylogenetically important myriapods . In contrast to the intensively sampled holometabolous insects , our analysis of this myriapod genome finds conservative gene sets and conserved synteny , shedding light on general genomic features of the arthropods .
Genomic DNA from multiple individuals of a wild Scottish population of S . maritima was sequenced and assembled into a draft genome sequence spanning 176 . 2 Mb . This assembled sequence omits many repeat sequences including heterochromatin , which probably accounts for the difference between the assembly length and the total genome size estimate of 290 Mb . An analysis of repetitive elements within the assembly is presented in Text S1 . The assembly incorporates 14 , 992 automatically generated gene models , 1 , 095 of which have been additionally manually annotated . We re-sequenced four individuals comprising three females and one male . The frequency of identified polymorphism , with SNP density of 4 . 5 variants/kb , is comparable with the five variants per kb in the Drosophila genetic reference panel [39] . It is hard to say how typical this is for soil dwelling arthropods , as very little population data are available for such species . To understand general patterns of gene evolution in S . maritima we reconstructed the evolutionary histories of all of its genes , i . e . , the phylome . The resulting gene phylogenies , available through phylomeDB [40] , were analysed to establish orthology and paralogy relationships with other arthropod genomes [41] , transfer functional knowledge from annotated orthologues , and to detect and date gene duplication events [42] . Some 32% of S . maritima genes can be traced back to duplications specific to this myriapod lineage since its divergence from other arthropod groups included in the analysis . Functions enriched among these genes include those related to , among other processes , catabolism of peptidoglycans , sodium transport , glutamate receptor , and sensory perception of taste . Related to this latter function , two of the largest gene expansions specific to the S . maritima lineage detected in our analysis are the gustatory receptor ( GR ) and ionotropic receptor ( IR ) families encoding putative membrane-associated gustatory and/or olfactory receptors ( see Text S1 , and Chemosensory section below ) . No obviously differentiated sex chromosomes are apparent in the diploid S . maritima karyotype , which comprises one long pair of metacentric chromosomes , together with seven pairs of much shorter telocentric chromosomes ( P . Woznicki , unpublished data; J . Green et al . , unpublished ) . Read-depth data from the genome assembly show that a proportion of the genome is underrepresented compared to the bulk of the data . One obvious reason for underrepresentation would be sequences derived from sex chromosomes . To confirm this , the coverage of individual scaffolds from the assembly was examined in sequence obtained from single individuals . A distinct fraction of underrepresented scaffolds is present in DNA derived from a male , but absent in female sequence ( Figure 2 ) , implying an XY sex determination mechanism . Quantitative PCR from three scaffolds in the underrepresented fraction confirmed that they are present at approximately twice the copy number in females as in males , identifying them as X chromosome derived ( J . Green et al . , unpublished ) . Other scaffolds of this fraction contain male specific sequences , and therefore presumably derive from a Y chromosome ( J . Green et al . , unpublished ) [31] . Combined with the karyotype data , this finding suggests that S . maritima possesses a weakly differentiated pair of X and Y chromosomes . From the whole genome assembly , S . maritima scaffold scf7180001247661 was found to contain a complete copy of the mitochondrial coding regions , flanked by a TY1/Copia-like retrotransposon , which all together spanned approximately 20 kb . This is unusually large for a metazoan mitochondrial genome and , as mis-assembly was suspected , PCR was used to clone the DNA between the genes at either end of the scaffold . This enabled us to close the circle of the mitochondrial genome , correct frameshifts , and confirm an unusual gene arrangement , resulting in a final circular assembly of 14 , 983 bp ( Table S11 ) . The gene arrangement in the S . maritima mitochondrial genome is striking ( Figure S6 ) . It diverges dramatically from the basic arthropod genome arrangement and differs from all other known centipede mitochondrial gene arrangements [43] . Although small sections of the S . maritima gene order are conserved with respect to the arthropod ground pattern found in Limulus polyphemus and the lithobiomorph centipede Lithobius forficatus ( e . g . , trnaF-nad5-H-nad4-nad4L on the minus strand ) , other sections are completely rearranged to an extent unusual in arthropods , and metazoans ( ACR and MJT , unpublished ) . This confounds attempts to use S . maritima mitochondrial gene order in phylogenetic reconstructions . With the exception of some conserved local gene clusters , the location of genes on the chromosomes of Drosophila and other Diptera retains no obvious trace of the ancestral bilaterian gene linkage . Other holometabolous insects such as Bombyx mori and Tribolium castaneum do show significant conservation of large-scale gene linkage with other phyla , for example , in the chordate Branchiostoma floridae ( amphioxus ) and the cnidarian Nematostella vectensis [44] , [45] . The last common ancestor of these two lineages pre-dated the ancestor of all bilaterian animals , and yet the genomes of these species retain detectable conserved synteny: orthologous genes are found together on the same chromosomes , or chromosome fragments , far more often than would be expected by chance . We find the S . maritima genome also retains significant traces of the large-scale genome organisation that was present in the bilaterian ancestor . Although the assignment of scaffolds to chromosomes is not determined in S . maritima , there are sufficient gene linkage data within scaffolds to reveal clear retained synteny between amphioxus and S . maritima ( Figure 3 ) , at a higher level than any of the Insecta or Pancrustacea we have examined . Of the 62 scaffolds with at least 20 genes from ancestral bilaterian orthology groups , 37 show enrichment of shared orthologues with one or ( in the case of a single scaffold ) two chordate ancestral linkage groups ( ALGs ) at a significance threshold of p<0 . 0001 ( after Bonferroni correction for 1 , 116 pairwise ALG-scaffold comparisons ) . Of these scaffolds' genes that have predicted human orthologues , 57% are found in a conserved macro-synteny context . At a more relaxed significance threshold ( p<0 . 01 ) , 71% of these scaffolds have a significant association with at least one chordate ALG , and 17 of the 18 chordate ALGs hit at least one of these scaffolds . Stronger synteny is also detected for the genome of the nematode Caenorhabditis elegans with S . maritima than with insects or other Metazoa . The C . elegans genome is highly rearranged , and shows low synteny with higher insects , or with chordates [7] , [46] , [47] . As members of the Ecdysozoa , nematodes last shared a common ancestor with the arthropods more recently than with chordates . This shared ancestry allows traces of conserved genome organisation to be detected with slowly rearranging arthropod genomes , even when it is only weakly apparent with chordates . By implication , the last common ancestor of the arthropods retained significant synteny with the last common ancestor of bilaterians as well as the last common ancestors of other phyla , such as the Chordata . This conserved synteny is more complete with this S . maritima genome sequence , due to the relative scrambling of the genomes of those other arthropods that have been sequenced previously . The clustering of genes in a genome is often of functional significance ( e . g . , reflecting co-regulation ) , as well as providing important insights into the origins of particular gene families when clusters are composed of genes from the same class or family . Gene clusters can also be a useful proxy for the degree of genome rearrangement . The homeobox gene super-class is one type of gene for which clustering has been extensively explored . S . maritima has 113 homeobox-containing genes , which is slightly more than seen in other sequenced arthropods such as D . melanogaster , T . castaneum , and Apis mellifera . This is due to some lineage-specific duplications in S . maritima as well as the retention of some homeobox families that have been lost in other arthropods , including Vax , Dmbx , and Hmbox ( see Text S1 ) . The homeobox-containing genes of the Hox gene cluster are renowned for their role in patterning the anterior-posterior axis of animal embryos . S . maritima has an intact , well-ordered Hox cluster containing one orthologue of each of the ten expected arthropod Hox genes , except for Hox3 . There are two potential Hox3 genes elsewhere in the S . maritima genome [48] , but the true orthology of these genes remains slightly ambiguous; it remains possible that they are the first example of ecdysozoan Xlox ParaHox genes ( see Text S1 ) . The Hox cluster spans 457 kb ( labial to eve ) , a span similar to assembled Hox clusters in a range of other invertebrate groups ( crustacean , mollusc , echinoderm , cephalochordate ) . This suggests that the contrasting very large ( and frequently broken ) Hox clusters of Drosophilids and some other insects are a derived characteristic . However , the spectrum of alternatively spliced and polyadenylated transcripts encoded by the Hox genes of S . maritima is comparable with what is known from D . melanogaster ( details in Text S1 ) . Exceptionally among protostomes , the S . maritima Hox cluster retains tight linkage to one orthologue of evx/evenskipped , as it does in some chordates and cnidarians . Further instances of homeobox gene clustering and linkage , and reconstructions of ancestral states , are summarized in Figure 4 and Table 1 ( and see Text S1 ) . The Hox gene cluster is hypothesized to have evolved within the context of a Mega-homeobox cluster that existed before the origin of the bilaterians and consisted of an array of many ANTP-class genes [49]–[51] . By the time of the last common ancestor of bilaterians the Hox cluster existed within the context of a SuperHox cluster , containing the Hox genes themselves and at least eight further ANTP-class genes [52] . The conservative nature of the S . maritima genome has left several fragments from the Mega-homeobox and SuperHox clusters still intact ( Figure 4; Table 1 ) . Furthermore , homeobox linkages in S . maritima raise the possibility that further genes could have been members of the Mega-homeobox and SuperHox clusters , including the ANTP-class gene Vax , as well as the SINE-class gene sine oculis and the HNF-class gene Hmbox ( see Text S1 for further details ) . The chemosensory system of arthropods is best known in insects . During the evolutionary transition from water to terrestrial environments , insects evolved a new set of genes to detect airborne molecules ( odorants ) [53]–[55] . The independent colonization of land by insects and myriapods raises two interesting questions: ( 1 ) what are the genes involved in chemosensation in non-insect arthropods , and ( 2 ) what genes are responsible for the detection of airborne molecules in other terrestrial arthropods ? We searched the S . maritima genome for homologues of the insect chemosensory genes , included in six gene families , three ligand binding protein families: odorant binding proteins ( OBPs ) [56] , [57] , chemosensory proteins ( CSPs ) [58] , [59] , and CheA/B [60] , [61]; and three membrane receptor families: GRs [62] , [63] , odorant receptors ( ORs ) [64] , [65] , and IRs [66] , [67] . Of the ligand binding proteins , we found only two genes belonging to the CSP family , but no representatives of the OBP or CheA/B families . Among the membrane receptor families , we identified a number of genes of both the GR and IR families , but no OR genes . The GR family in S . maritima is represented by 77 genes , 17 of which seem to be pseudogenes , with similar numbers of genes and pseudogenes being fairly typical features of this gene family in other arthropods . A phylogenetic tree revealed that none of the S . maritima GR genes have 1∶1 orthology to other arthropod GRs . Instead , all S . maritima GRs cluster in a single clade , with six major subclades , representing separate expansions of the GR repertoire in the centipede lineage ( Figure 5A and see Text S1 ) . The IR family is known to be ancient [67] , but S . maritima has a relative expansion of this family . The search for IRs led to the annotation of 69 genes , 15 of which belong to the IGluR subfamily , which is not involved in chemosensation , but is highly conserved among arthropods and animals in general . Among the remaining 54 IRs , three are orthologues of conserved IR genes that have been shown to have an olfactory function in D . melanogaster . However , 51 of the S . maritima IRs do not have orthologues either in D . melanogaster or in Ixodes scapularis , clustering together in a single clade ( the expansion clade in Figure 5B ) . This finding suggests that most S . maritima IRs , as observed with GRs , have duplicated from a common ancestral gene exclusive to the centipede lineage . The absence of the insect OR family agrees with the prediction of Robertson and colleagues [54] that this lineage of the insect chemoreceptor superfamily evolved with terrestriality in insects , and it is also missing from the water flea Daphnia pulex [53] . The same appears to be true for the OBPs . We therefore infer that , as centipedes adapted to terrestriality independently from the hexapods , they utilized a novel combination of expanded GR and IR protein families for olfaction , in addition to their more ancestral roles in gustation . S . maritima , like all species of the order Geophilomorpha , is blind [37] . Nevertheless , it avoids open spaces and negative phototaxis has been demonstrated in other species of Geophilomorpha [38] , [68] . We searched the S . maritima genome for light receptor genes . Interestingly , we have found no opsin genes , no homologue of gustatory receptor 28b ( GR28b ) , which is involved in larval light avoidance behaviour in Drosophila [69] , and no cryptochromes . Thus , none of the known arthropod light receptors are present . Furthermore , there are no photolyases , which would repair UV light induced DNA damage . As a consequence , the critical avoidance of open spaces by S . maritima must either be mediated by other sensory instances than light perception , or S . maritima possesses yet unknown light receptor molecules . The absence of light receptors , particularly cryptochromes , also raises the issue of the entrainment and composition of a potential S . maritima circadian clock . Strikingly , we could not identify any components of the major regulatory feedback loop of the canonical arthropod circadian clock ( including period , cycle , b-mal/clock , timeless , cryptochromes 1 and 2 , jetlag [70] ) . The only circadian clock genes found ( timeout , vrille , pdp1 , clockwork orange ) are generally known to be involved in other physiological processes as well [71]–[73] . The extensive secondary gene loss of both light receptors and circadian clock genes raises questions about the actual existence of a circadian clock in S . maritima . One could hypothesize that a circadian clock may not be required in S . maritima's subsurface habitat , although other periodicities , such as tide cycles , might be important . If S . maritima does have a circadian clock then it must be operating via a mechanism distinct from the canonical arthropod system . Other blind or subterranean animals do maintain a circadian rhythm , despite complete loss of vision and connection with the surface ( e . g . , Spalax ) [74]–[76] . In other cases ( e . g . , blind cave crayfish [77] ) , despite the loss of vision , opsin proteins remain functional , and are hypothesized to have a role in circadian cycles . However , both these examples represent species that have become blind and subterranean relatively recently . To confirm that the loss of these genes is not general for all centipedes , we performed BLASTP analyses searching for the set of light sensing and circadian clock genes that are missing from S . maritima in RNAseq data from the house centipede Scutigera coleoptrata ( NCBI SRA accession SRR1158078 ) , a species with well-developed eyes . We find homologs to period , cycle , b-mal/clock , jetlag , cryptochrome1 , cryptochrome 2 , ( 6-4 ) -photolyase , and nina-e ( rhodopsin 1 ) , suggesting that both light sensing and circadian clock systems were present in ancestor of myriapods . Although we have no direct information about photoreceptors or circadian genes in other geophilomorph species , the fact that all geophilomorphs are blind suggests that the loss of the related genes is very ancient , and may date back to the origin of the clade . A defining characteristic of arthropods is an exoskeleton with chitin and cuticular proteins as the primary components . Although several families of cuticular proteins have been recognized , the CPR family ( Cuticular Proteins with the Rebers and Riddiford consensus ) is by far the largest in every arthropod for which a complete genome is available , with 32 to >150 members [78] . Proteins in the CPR family have a consensus region in arthropods of about 28 amino acids , first recognized by Rebers and Riddiford [79] , which was subsequently extended to ∼64 amino acid residues and shown to be necessary and sufficient for binding to chitin [80] . No clear instances of the Rebers and Riddiford ( RR ) consensus have been identified outside the arthropods . We identified 38 members of the CPR family in S . maritima . There are two main forms of the consensus , designated RR-1 and RR-2 , with the former primarily associated with flexible cuticle , the latter with rigid cuticle . Interestingly , while chelicerates studied to date have no members of the RR-1 subfamily ( as classified at CutProtFam-Pred , http://aias . biol . uoa . gr/CutProtFam-Pred/home . php ) , seven of the S . maritima CPR proteins clearly belong to this class . This would be consistent with the origin of the RR1-coding genes being in the mandibulate ancestor after this lineage had diverged from the chelicerate lineage . Further data are needed to verify that the identified proteins are indeed important constituents of the cuticle . Cell-to-cell communication in arthropods occurs via a variety of neurotransmitters and neuro-endocrine hormones , including biogenic amines , neuropeptides , protein hormones , juvenile hormone ( JH ) , and ecdysone . These signalling molecules and their receptors steer central processes such as growth , metamorphosis , feeding , reproduction , and behaviour . Most receptors for biogenic amines , neuropeptides , and protein hormones are G protein-coupled receptors ( GPCRs ) [81] . Intracellularly , the G proteins initiate second messenger cascades [82] . JH and ecdysone , however , are lipophilic and can diffuse through the cell membrane to bind with nuclear receptors [83] , [84] . In addition , ecdysone can also activate a specific GPCR , and initiate a second messenger cascade [85] . There is extensive cross-talk between these extracellular signal molecules . S . maritima possesses 19 biogenic amine receptors , a number similar to the 18–22 biogenic amine receptors that have been identified in other arthropods ( Table S19 ) . In S . maritima , there are four octopamine GPCRs , one octopamine/tyramine , one tyramine , four dopamine and three serotonin GPCRs , three GPCRs for acetylcholine , one GPCR for adenosine , and two orphan biogenic amine receptors . Although this distribution resembles very much that of Drosophila and other arthropods , there are some interesting differences with Drosophila , which expresses two additional β-adrenergic-like octopamine receptors compared to S . maritima , while S . maritima expresses two putative β-adrenergic-like octopamine receptors ( Sm-OctBetaRHK and Sm-D1/OctBeta ) , which are expressed in a number of insect and tick species , but not in Drosophila ( Table S20 ) [86] . The true functional identities of all the putative S . maritima biogenic amine GPCRs awaits their cloning , functional expression , and pharmacological characterization in cell lines . In addition , 36 neuropeptide and protein hormone precursor genes are present in this centipede . Each neuropeptide precursor contains one or more ( up to seven ) immature neuropeptide sequences ( Figure S20 ) . Interestingly , the centipede contains two CCHamide-1 , two eclosion hormone , and two FMRFamide genes , whereas these genes are only present as single copies in the genomes of most other arthropods [87] . In concert with the presence of 36 neuropeptide genes , we found 33 genes for neuropeptide receptors ( 31 GPCRs and two guanylcyclase receptors ) ( see Table S21 ) . As observed for the neuropeptide genes , a number of the neuropeptide receptor genes , which are only found as single copies in most other arthropods , have also been duplicated . S . maritima has two inotocin GPCR genes , two SIFamide , two corazonin , two eclosion hormone guanylcyclase receptor genes , two eclosion triggering hormone GPCR genes , three sulfakinin GPCR genes , and three LGR-4 ( Leu-rich-repeats-containing-GPCR-4 ) genes . The latter receptors are orphans ( GPCRs without an identified ligand ) and only present as single-copy genes in most other arthropods [88] . Several of these duplicated GPCR genes are located in close vicinity to each other in the genome ( Figure S21 , suggesting recent duplication events . Furthermore , duplications of both the eclosion hormone and its receptor genes and the duplication of the ecdysis triggering hormone receptor genes suggest that the process of ecdysis ( moulting ) has undergone some sort of modification , perhaps requiring more complex control in the lineage leading to centipedes . We summarize in Table S22 the neuropeptide/protein hormone signalling systems that are present or absent in selected arthropod genome sequences . Each arthropod species , including S . maritima , has its own characteristic pattern , or “barcode , ” of present/absent neuropeptide signalling systems . However , the relationship between the specific neuropeptide “barcode” and physiology remains to be elucidated . Insect JH is important for growth , moulting , and reproduction in arthropods [84] . This hormone is a terpenoid ( unsaturated hydrocarbon ) that is synthesized from acetyl-CoA by several enzymatic steps ( Figure S22 ) . In several insects the production of JH is stimulated by the neuropeptide allatotropin , while it is inhibited by either allotostatin-A , -B , or -C [89] , [90] . We found that S . maritima has orthologues of many of the biosynthetic enzymes needed for JH biosynthesis in insects ( Table S23 ) . Also , the JH binding proteins are encoded in the centipede genome as well as JH degradation enzymes ( Table S24 ) . This implies that the complete JH system is present in this centipede . Similarly , neuropeptides that could stimulate or inhibit the synthesis and release of JH , such as allatotropin and the allatostatins -A , -B , and -C , are also present in S . maritima ( Figure S22 , suggesting that the overall functioning of the JH system in centipedes might be very similar to that of insects ) ( Table S23 ) . To date , the existence of JH signalling systems has been demonstrated in insects , crustaceans , and recently in spider mites [89] , [91] , [92] . Its occurrence in S . maritima and spider mites ( Chelicerata ) indicates that JH signalling has deep evolutionary roots and we suggest that it might have evolved together with the emergence of the exoskeleton in arthropods . Certain signalling systems , including transforming growth factor ( TGF ) -beta , Wnt , and fibroblast growth factor ( FGF ) , are used throughout development across the animal kingdom . Various lineage-specific modifications of these systems have occurred , particularly within the arthropods . With regards to TGF-beta signalling we found single orthologues of all members of the Activin family , except Alp ( Activin-like protein ) ( see Figure S23; Text S1 ) . In the BMP-family , the S . maritima genome contains two divergent BMP sequences , as well as a clear orthologue of glass-bottom boat ( gbb ) and two decapentaplegic ( dpp ) orthologues . In addition , the S . maritima sequences confirm the ancestral presence of an anti-dorsalizing morphogenetic protein ( ADMP ) and a BMP9/10 orthologue in arthropods , which are both absent from Drosophila [93] . Most interestingly , the S . maritima genome includes the antagonistic BMP ligand BMP3 ( previously suggested to be present only in deuterostomes [94] ) , a potential gremlin/neuroblastoma suppressor of tumorigenicity , and two nearly identical bambi genes ( absent from Drosophila ) , and the BMP inhibitor noggin ( present in vertebrates but lost in most holometabolous insects ) . The multiple BMP-agonists and -antagonists indicate that considerable changes have occurred in the TGF-beta signalling system during arthropod evolution , particularly in the Holometabola . Reconstructions of Wnt gene evolutionary history suggest that the ancestral bilaterian possessed at least 13 distinct Wnt gene subfamilies [95] , [96] . This initial number has been secondarily reduced in many taxa . This trend of secondary gene loss is readily apparent within the arthropods , with holometabolous insects such as D . melanogaster retaining only seven Wnt subfamilies [97] , [98] . In contrast , the Wnt signalling complement in S . maritima comprises 11 of the 13 Wnt-ligand subfamilies ( Figure S24 ) . Phylogenetic investigation has identified these genes as wnt1 , wnt2 , wnt4 , wnt5 , wnt6 , wnt7 , wnt9 , wnt10 , wnt11 , wnt16 , and wntA . wnt3 and wnt8 are missing from the S . maritima genome . While the absence of wnt3 is common to protostomes , wnt8 or wnt8-like sequences occur in other protostome genomes , including insects , spiders , and another myriapod , Glomeris marginata [97] . The Wnt genes are known to display a degree of linkage and clustering in many arthropods . Some conservation of this is also found in S . maritima , with wnt1 , wnt6 , and wnt10 adjacent to each other on the same scaffold , possibly representing part of an ancient clustering ( Table S25 ) [99] . The primary receptors for Wnt ligands in the canonical Wnt signalling pathway are the trans-membrane receptors of the Frizzled family . Five of these have been identified: Frizzled1 , Frizzled4 , Frizzled5/8 , Frizzled7 , and Frizzled10 . As is the case for the wnt genes themselves , this is a larger number than is found in most arthropods . Other Fz-related genes are also present: smoothened , involved in Hedgehog signalling , and secreted frizzled related protein , which has inhibitory roles in Wnt signalling in other taxa . Putative non-canonical Wnt receptors are also encoded , including two subfamilies of receptor tyrosine kinase-like orphan receptor ( ror ) . In addition to ror2 , there is a lineage-specific duplication of ror1 , making a total of three ror genes , as opposed to only one in D . melanogaster . Another Wnt agonist , the R-spondin orthologue was also found . As part of the Wnt-binding complex we found one arrow-LRP5/6-like Wnt-coreceptor gene in the genome: lrp6 . Other LRP-molecules with potential Wnt-binding activity also exist: LRP1 , LRP2 , and LRP4 . Because of the absence of an intracellular signalling domain these could potentially function as Wnt-inhibitors . Together , the large number of ligand and receptor genes point towards both the conservation of an ancestral Wnt signalling system and to a certain degree of unusual complexity in of this system in S . maritima . Concerning the FGF pathway , we identified two closely related FGF receptors . These two S . maritima receptors are likely to stem from a duplication in the myriapod lineage that was independent from that which generated the two Drosophila FGFRs , Heartless and Breathless ( Figure S25 ) . The number of FGF ligands found in the genomes of insects such as D . melanogaster ( three fgf genes ) or T . castaneum ( four fgf genes ) is small when compared to 22 fgf genes found in the genomes of vertebrates . In the S . maritima genome , we identified three fgf-genes ( Figure S26 ) . One of them potentially represents an fgf 18/8/24 orthologue to which the fgf8-like genes of Tribolium and of Drosophila ( pyramus and thisbe ) are associated . The second S . maritima fgf groups with the fgf1 genes , while the third groups with the fgf 16/9/20 clade ( the first known arthropod member of this clade ) . Low support values for this grouping raise the possibility that it might actually be an orthologue of insect branchless genes . Other FGF-pathway genes present in S . maritima include stumps ( Downstream-of-FGF-signalling [DOF] ) and sprouty related . Kinases make up about 2% of all proteins in most eukaryotes , while they phosphorylate over 30% of all proteins and regulate virtually all biological functions . We identified 393 protein kinases in the S . maritima genome , representing 2 . 6% of the proteome . We classified these into conserved families and subfamilies , compared the kinome to those of 26 other arthropods and inferred the evolutionary history of all kinases across the arthropods ( Figure 6 ) . We predict that an early arthropod had at least 231 distinct kinases and see considerable loss of ancestral kinases in most extant species . S . maritima has the smallest number of losses among the arthropods , with only ten kinases lost relative to the arthropod ancestor . In contrast , the two chelicerates T . urticae and I . scapularis have lost 63 and 45 kinases , respectively , and D . melanogaster lost 30 , giving S . maritima the richest repertoire of conserved kinases of any arthropod examined . All but one of the losses in S . maritima have been lost in other arthropods , suggesting that these genes may be partially redundant or particularly prone to loss . The one unique loss is NinaC , which in Drosophila is required for vision , likely associated with other vision related gene loss described above . As in many other species , we also see some novelties and expansions of existing families: the SRPK kinase family , involved in splicing and RNA regulation , has expanded to 36 members , and the nuclear VRK family is expanded to 16 . A novel family of receptor guanylate cyclases ( nine genes ) and three clusters of unique protein-kinase-like ( PKL ) kinases , containing 28 genes in total , are also seen , though their functions are not known . DNA binding proteins with the capacity to regulate the expression of other genes are central players in the control of development and many other processes . Since one of the original interests in S . maritima was for its developmental characteristics , we carried out a survey of developmentally relevant transcription factors , with an emphasis on transcription factors suspected to be involved in processes of axial specification , segmentation , mesoderm formation , and brain development . We identified orthologues of ∼80 transcription factors of the Zinc finger and helix-loop-helix families in addition to the 113 homeobox-containing transcription factors already discussed ( see Text S1 ) . In no case did we fail to find at least one orthologue of the gene families expected from our knowledge of Drosophila , though individual duplications and losses among gene families were not uncommon . Among the set of pair-rule segmentation genes , for example , S . maritima has multiple homologues of paired , even-skipped , odd-skipped , odd-paired , and hairy-like genes , but only a single orthologue of sloppy-paired and runt-like genes , whereas Drosophila has multiple runt and sloppy-paired genes but only single orthologues of even-skipped and odd-paired . Where both lineages have multiple copies , ( paired , hairy , odd-skipped ) , sequence alone rarely defines one-to-one orthologous relationships , and the evolutionary history remains unclear [29] . Other notable duplications include caudal ( three genes ) and brachyury ( two genes ) . In a number of cases , transcription factors known to play a role in vertebrate development , but apparently missing from Drosophila and other insects , are retained in S . maritima . Examples include the homeobox genes Dmbx and Vax noted above , and the FoxJ1 , FoxJ2 , and FoxL1 subfamilies of forkhead/Fox factors . One of the developmental transcription factors provides an example where insects use isoforms to generate alternative proteins that are encoded by paralogous genes in S . maritima . Two centipede orthologues of the developmental transcription factor cap‘n’collar encode isoforms that differ at their N-terminal end . The longer protein , encoded by the gene cnc1 , contains sequence motifs that align to Drosophila cnc isoform C ( Figure S27 , which is broadly expressed throughout embryonic development ) [100] . S . maritima cnc1 is similarly expressed ubiquitously , whereas the other orthologue , cnc2 , shows a segment specific pattern of expression similar to that of the shorter Drosophila cnc isoform B ( VSH and MA , unpublished ) [100] . Arthropods can mount an innate immune response against pathogenic bacteria , fungi , viruses , and metazoan parasites . The nature of the responses to these invaders , such as phagocytosis , encapsulation , melanisation , or the synthesis of antimicrobial peptides , is often similar across arthropods [101] . Furthermore , key aspects of innate immunity are conserved between insects and mammals , which suggests an ancient origin of these defences . Previous studies have revealed extensive conservation of key pathways and gene families across the insects and crustaceans [102] . Beyond the Pancrustacea the extent of immunity gene conservation is unclear . Therefore , we searched the S . maritima genome for homologues of immunity genes characterised in other arthropods . We found conservation of most immunity gene families between insects and S . maritima ( Table S30 ) , suggesting that the immune gene complement known from Drosophila was largely present in the most recent common ancestor of the myriapods and pancrustaceans . The humoral immune response of insects recognises infection using proteins that bind to conserved molecular patterns on pathogens [103] . Sequence homologues for the major recognition protein families found in Drosophila , peptidoglycan recognition proteins ( PGRPs ) , and gram-negative bacteria-binding proteins ( GNBPs ) , were found with the expected protein domains . These proteins then activate signalling pathways [103] , and all four major insect immune signalling pathways ( Toll , IMD , JAK/STAT , and JNK ) are present in S . maritima , with 1∶1 sequence homologues of most pathway members . The cellular immune response of insects relies on receptors and opsonins including thioester-containing proteins ( TEPs ) , fibrinogen related proteins ( FREPs ) , and scavenger receptors [103] , [104] , and these are also present in S . maritima , often with protein domains in the same arrangement as Drosophila . We also find sequence homologues for effector gene classes including nitric oxide synthases ( NOS ) and prophenoloxidase ( PPO ) . However , we failed to identify any antimicrobial peptide homologues , possibly as these genes are often short and highly divergent between species . In insects , it is common to find that certain immune gene families have undergone expansions in certain lineages [105] . Again , this is mirrored in S . maritima , where we found lineage-specific expansions of the PGRP and Toll-like receptor genes ( TLRs ) ( Figure 7 ) . Overall , the presence of the main families of immunity genes suggests that there is also functional conservation of the immune response . The innate immune system is thought to rely on a small number of immune receptors that bind to conserved molecules associated with pathogens . This view was challenged by the discovery in Drosophila that the gene Dscam ( Down syndrome cell adhesion molecule ) , which has the potential to generate over 150 , 000 different protein isoforms by alternative splicing , functions as an immune receptor in addition to its roles in nervous system development [106] . Dscam family members are membrane receptors composed of several immunoglobulin ( Ig ) and fibronectin domains ( FNIII ) . In pancrustaceans one member of the Dscam family has a large number of internal exon duplications and a sophisticated mechanism of mutually exclusive alternative splicing , which enables a single Dscam locus to somatically generate thousands of isoforms , which differ in half of two Ig domains ( Ig2 and Ig3 ) and in another complete Ig domain ( Ig7 ) . This creates a high diversity of adhesion properties , useful for immune responses . We found that S . maritima has evolved a different strategy to generate a diversity of Dscam isoforms [107] . The genome contains 60 to 80 canonical Dscam paralogues and over 20 other Dscam related incomplete or non-canonical genes ( Figure 8 ) . In 40 Dscam genes , the exon coding for Ig7 is duplicated two to five times ( but not the exons coding for Ig2 and Ig3 , which are duplicated in pancrustaceans ) . Our analysis of transcripts suggests that many of those duplicated exons might be alternatively spliced in a mutually exclusive fashion , supporting the hypothesis that the mechanism of mutually exclusive alternative splicing of Dscam probably evolved in the common ancestor of both pancrustaceans and myriapods . According to our phylogenetic analysis , which included 12 paralogues , the S . maritima Dscams share a common origin and arose by duplication in the centipede lineage [107] . In the chelicerate I . scapularis , Dscam has also been duplicated extensively , both by whole-gene and by domain duplications [107] . These Dscam homologues however do not have a canonical domain composition and whether or not alternative splicing is also present in chelicerates remains unknown . The independent evolution of Dscam diversification in different arthropod groups ( one locus with dozens of exon duplications in pancrustaceans versus many gene duplications coupled with a few exon duplications in S . maritima ( Figure 8 ) suggests that the functional diversity in adhesion properties was important in the early evolution of arthropods . Whether all of these genes function in the immune system or nervous system development remains to be determined . The short-interfering RNA ( siRNA ) pathway is the primary defence of insects against RNA viruses , while the piRNA pathway silences transposable elements in the germ line and micro RNAs ( miRNAs ) function in gene regulation [108] . These RNAi pathways appear to be intact in S . maritima , as we found homologues of key genes , including Ago1 and Dicer-1 in the miRNA pathway , Ago2 and Dcr2 in the siRNA pathway , and Ago3 and piwi in the piRNA pathway ( Table S30 ) . We found two paralogues of Ago2 and three paralogues of piwi , suggesting that RNAi may be more complex than in D . melanogaster . In other arthropods , expansion of the piwi family has been linked to neo- or subfunctionalization of germ line and soma roles , and so it remains to be seen whether this is also the case for S . maritima . Selenoproteins are peculiar proteins including a selenocysteine ( Sec ) residue , a very reactive amino acid typically found in the catalytic site of redox proteins , which is inserted through the recoding of a UGA codon [109] . While vertebrates possess 24–38 selenoproteins [110] , insects have very few ( D . melanogaster has three ) or none at all . Several events of complete selenoproteome loss have been observed in insects [111] . These were ascribed to the fundamental differences in the insect antioxidant systems , which would favour selenoprotein loss or their conversion to standard proteins ( cysteine homologues ) . The analysis of a myriapod selenoproteome is then crucial for a phylogenetic mapping of such differences . The S . maritima genome was found to be surprisingly rich in selenoproteins: we have identified 20 predicted proteins ( Table S26 ) . Downstream of the coding sequence of each selenoprotein gene , we detected a selenocysteine insertion sequence ( SECIS ) element , the stem-loop structure necessary to target the Sec recoding machinery during selenoprotein translation . The full set of factors necessary for selenocysteine insertion and production was also found: tRNA-Sec , SecS , SBP2 , eEFsec , pstk , secp43 , SPS2 . The centipede selenoproteome is rather similar to that predicted for the ancestral vertebrate ( see [110] ) . This supports the idea that selenoprotein losses are specific to insects and can be ascribed to changes in that lineage , supporting the idea that a massive selenoproteome reduction occurred specifically in insects . A notable difference with vertebrates was found for the protein methionine sulfoxide reductase A ( MsrA ) . This enzyme catalyzes the reduction of methionine-L-oxide to methionine , repairing proteins that were inactivated by oxidation . A selenoenzyme from this family has been previously characterized in the green alga Chlamydomonas , and selenocysteine containing forms were also observed in some non-insect arthropods [112] . In contrast , only cysteine homologues are present in vertebrate and insect genomes . We found a Sec-containing MsrA in the centipede genome , as well as in arthropods D . pulex , I . scapularis , and also in the chordate B . floridae . This , along with phylogenetic reconstruction analysis , supports the idea that the selenoprotein MsrA was present in their last common ancestor , and was later converted to a cysteine homologue independently in insects and vertebrates . The two major antioxidant selenoprotein families in vertebrates , glutathione peroxidases ( GPx ) , and thioredoxin reductases ( TrxR ) , were also found with selenocysteine in the centipede genome . In contrast , all holometabolous insects possess only cysteine forms , and consistently , important differences were noted in these and other enzymes in the glutathione and thioredoxin system ( see [113] for an overview ) . Thus , on the basis of gene content , we expect the antioxidant systems of S . maritima to be more similar to vertebrates and other animals than to holometabolous insects like D . melanogaster . Invertebrate DNA methylation occurs predominantly on gene bodies ( exons and introns ) , via addition of a methyl group to a cytosine residue in a CpG context [114]–[116] . The exact function of gene body methylation is currently unknown , though it is correlated with active transcription in a wide range of species [116] , and has been implicated in alternative splicing [117] , [118] and regulation of chromatin organization [118] . Methylated cytosines are susceptible to deamination , to form a uracil , which is recognized as a thymine . Thus , over evolutionary time , highly methylated genes ( in germ-line cells ) will have comparatively low CpG content . The “observed CpG/expected CpG” ( CpG ( o/e ) ) ratio is an indicator of C-methylation: plots of CpG ( o/e ) for a gene set produce a bimodal distribution where a proportion of the genes have an evolutionary history of methylation [119] . In contrast , species without methylation systems , such as D . melanogaster , yield a unimodal distribution [119] . The S . maritima gene body CpG ( o/e ) plot has a trimodal distribution , with the majority of genes having a ratio close to 1 ( Figure 9; Text S1 ) . Underlying this major peak are two smaller peaks , one “low” and one “high” centred around ratios of 0 . 62 and 1 . 48 , respectively . This “high” peak , that contains genes with higher than expected CpG content , is unusual and is not seen in this analysis of other arthropods [91] , [119]–[121] . Applying the same analysis to 1 , 000 bp windows across the entire genome ( including both coding and non-coding regions ) reveals a similar peak of high CpG content ( Figure S29 ) . This implies that the peak of “high” CpG content seen in gene bodies is due to unusually high CpG content in some regions of the genome rather than a specific feature of those coding regions . The “low” peak , however , indicates that 9 . 5% of genes have been methylated in the germ-line over evolutionary time . The number of genes contained within the “low” peak in S . maritima is smaller than observed in insect species with methylation , which can be as high as 40% in exceptional species such as the pea aphid and the honeybee [119] , [120] , where the mechanism is likely involved in polyphenism and caste differences respectively . However , the number of genes methylated is less in non-social hymenopteran such as Nasonia vitripennis , in beetles , and in mites [91] , [121] , [122] . Consistent with the low-levels of germ-line methylation detected , the genome contains a single orthologue of the de novo DNA methylation enzyme Dnmt3 and four orthologues of the maintenance DNA methyltransferases Dnmt1 ( a–d ) . Two of the Dnmt1 orthologues have lost amino acids that are required for methyltransferase activity , but these genes are represented in the transcriptome data , and are thus unlikely to be pseudogenes . One Dnmt1 gene shows sex-specific splicing , with a shorter transcript producing a truncated protein seen in female-derived transcription libraries . We also find a single orthologue of Tet1 , a putative DNA demethylation enzyme [123] , [124] . Taken together these data indicate that S . maritima has an active DNA methylation system , and that over evolutionary time a small number of genes have been methylated in the germ-line , resulting in a lower than expected CpG dinucleotide content . We annotated over 900 homologues of known non-coding RNAs in the S . maritima genome , including over 600 predicted tRNAs ( plus an additional 300 tRNA pseudogenes ) , 71 copies of 5S rRNA and 12 5 . 8S rRNAs , 88 copies of RNA components of the major spliceosome , and three out of the four RNA components of the minor U12 spliceosome , and 54 microRNA genes . As is common for whole genome assemblies , we did not identify intact copies of the 18S or 28S rRNAs . Further details of our methodology are provided in Text S1 . The predicted tRNA gene set includes all anticodons necessary to code for the 21 amino acids , including four potential SeC tRNAs . We identify a massive expansion of the tRNA-Ala-GGC family , with 322 sequences classified as functional tRNAs by tRNAscan-SE and an additional 172 classified as pseudogenes . These appear scattered throughout the scaffolds of the genome assembly . It is highly likely that the majority of these genes are pseudogenes , and the expansion may represent co-option of the tRNA into a transposable element . Three S . maritima microRNA genes have been reported previously , and are available in the miRBase database ( version 18 ) [125] . Two of these , mir-282 and mir-965 , have homologues in crustaceans and insects . The third , mir-3930 , is specific to myriapods [15] . In addition , we found 52 homologues of known microRNAs ( Figure S34 ) . These include 28 homologues of the 34 ancient microRNA families found throughout the Bilateria [126] . Four of these families were previously reported to be lost at various stages during animal evolution and , consistent with this , we failed to identify them in the S . maritima genome . Surprisingly , we also could not identify the S . maritima homologue of mir-125 , a member of the ancient mir-100/let-7/mir-125 cluster , which is found in almost all bilaterians and has a well-established function in the regulation of development of many species [127]–[129] . Mir-100 and let-7 are well-conserved and localized within a 1 kb region on the same scaffold in S . maritima . Whilst we cannot rule out the possibility that the missing mir-125 is an artefact of the draft-quality genome assembly , the size of the scaffold strongly suggests that it is not present in the mir-100/let-7 cluster . We also identified 17 homologues of microRNAs common to ecdysozoans , and nine microRNAs known only from arthropods . Among the former , there are five homologues of mir-2 localized in close proximity to each other and downstream of mir-71 . This clustering is conserved across protostomes , and it has previously been shown that the mir-2 family underwent various expansions during evolution [130] . Finally , we discovered a homologue of mir-2788 , which was previously only known from insects , suggesting that this microRNA had an earlier origin . The sequencing of the centipede genome extends significantly the diversity of available arthropod genomes , and provides novel information pertinent to a range of evolutionary questions . Myriapods show a simple body organization that has remained relatively unchanged in comparison to their ancestors from the Silurian or even earlier [6] , leading to an expectation of general conservatism . The myriapods are descendants of an independent terrestrialisation event from the hexapods and chelicerates , opening the opportunity for studying convergent evolution in these taxa . Naturally , S . maritima itself has its own evolutionary history , including both lineage specific features of the geophilomorphs and adaptations to their subterranean environment , allowing us to identify specific genomic signatures of ecological adaptations . Finally , the phylogenetic position of the myriapods within the arthropods has been the subject of intense debate for several years , and the availability of genomic data for a myriapod should contribute to the future resolution of this debate . The morphological conservatism of centipedes is mirrored in many conservative aspects of the S . maritima genome . From the analyses of the various gene families outlined above it becomes clear that the S . maritima genome has undergone much less gene loss and rearrangement than the genomes of other sequenced arthropods , in particular those of the holometabolous insects such as D . melanogaster . This prototypical nature of the S . maritima genome is illustrated by the conservation of synteny relative to the arthropod and bilaterian ancestors , and the conservation of some ancient gene linkages and clustering , as seen for numerous homeobox genes . As such , the S . maritima genome can serve as a guide to the ancestral state of the arthropod genomes , or as a reference in the reconstruction of evolutionary events in the history of arthropod genomes . The independent terrestrialisation of the myriapods and insects is evidenced by the use of different evolutionary solutions to similar problems . Figure 10 summarizes some of the gene gains and losses observed . We see this most clearly in the independent expansions of gustatory receptor proteins in myriapods and insects and the differential expansions of ionotropic and odorant receptors to deal with terrestrial chemosensation in the two lineages . Similarly , though probably not for the same reasons , we see a divergent solution for the generation of Dscam diversity in the immune response through the use of paralogues instead of the insect strategy of alternative splicing . The chelicerates also attained terrestriality independently . However , our understanding of chelicerate genomes still lags behind our understanding of insect , and now myriapod , genomes . Thus , extending this comparison to chelicerates , intriguing as it may be , will have to await future analysis of their genomes . Lineage specific features of the S . maritima genome include the apparent loss of all known photoreceptors and a loss of the canonical circadian clock system based around period and its associated gene network . The characterization of whether S . maritima does have a circadian clock , and if it does how this is controlled , awaits further work , as does the pinpointing of when in their evolutionary history these systems were lost . The absence of the microRNA miR-125 is another surprising evolutionary loss . The extensive rearrangement of the mitochondrial genome is striking in comparison with the general conservatism seen in other known arthropod mitochondrial genomes , and especially in contrast with the conservative nature of S . maritima's nuclear genome .
The S . maritima raw sequence , and assembled genome sequence data are available at the NCBI under bioproject PRJNA20501 ( http://www . ncbi . nlm . nih . gov/bioproject/PRJNA20501 ) Assembly ID GCA_000239455 . 1 . The genome was sequenced using 454 sequencing technology , assembled using the celera assembler , annotated using a combination of the Maker 2 . 0 pipeline , and custom perl scripts followed by manual annotation of selected genes . Text S1 includes detailed methods for these steps , and additionally for the individuals sequenced , library construction and sequencing protocols used , repeat analysis , RNA sequencing , phylome db analysis , specific protocols for manual annotation of gene families , CpG analysis , and phylome and synteny re-construction . | Arthropods are the most abundant animals on earth . Among them , insects clearly dominate on land , whereas crustaceans hold the title for the most diverse invertebrates in the oceans . Much is known about the biology of these groups , not least because of genomic studies of the fruit fly Drosophila , the water flea Daphnia , and other species used in research . Here we report the first genome sequence from a species belonging to a lineage that has previously received very little attention—the myriapods . Myriapods were among the first arthropods to invade the land over 400 million years ago , and survive today as the herbivorous millipedes and venomous centipedes , one of which—Strigamia maritima—we have sequenced here . We find that the genome of this centipede retains more characteristics of the presumed arthropod ancestor than other sequenced insect genomes . The genome provides access to many aspects of myriapod biology that have not been studied before , suggesting , for example , that they have diversified receptors for smell that are quite different from those used by insects . In addition , it shows specific consequences of the largely subterranean life of this particular species , which seems to have lost the genes for all known light-sensing molecules , even though it still avoids light . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"sequencing",
"techniques",
"genome",
"sequencing",
"organismal",
"evolution",
"genome",
"complexity",
"genomics",
"genome",
"evolution",
"molecular",
"biology",
"animal",
"evolution",
"genetics",
"biology",
"and",
"life",
"sciences",
"comparative",
"genomics",
"molecular",
"biology",
"techniques",
"computational",
"biology",
"evolutionary",
"biology",
"evolutionary",
"immunology",
"evolutionary",
"genetics"
]
| 2014 | The First Myriapod Genome Sequence Reveals Conservative Arthropod Gene Content and Genome Organisation in the Centipede Strigamia maritima |
Killed , oral cholera vaccines have proven safe and effective , and several large-scale mass cholera vaccination efforts have demonstrated the feasibility of widespread deployment . This study uses a mathematical model of cholera transmission in Bangladesh to examine the effectiveness of potential vaccination strategies . We developed an age-structured mathematical model of cholera transmission and calibrated it to reproduce the dynamics of cholera in Matlab , Bangladesh . We used the model to predict the effectiveness of different cholera vaccination strategies over a period of 20 years . We explored vaccination programs that targeted one of three increasingly focused age groups ( the entire vaccine-eligible population of age one year and older , children of ages 1 to 14 years , or preschoolers of ages 1 to 4 years ) and that could occur either as campaigns recurring every five years or as continuous ongoing vaccination efforts . Our modeling results suggest that vaccinating 70% of the population would avert 90% of cholera cases in the first year but that campaign and continuous vaccination strategies differ in effectiveness over 20 years . Maintaining 70% coverage of the population would be sufficient to prevent sustained transmission of endemic cholera in Matlab , while vaccinating periodically every five years is less effective . Selectively vaccinating children 1–14 years old would prevent the most cholera cases per vaccine administered in both campaign and continuous strategies . We conclude that continuous mass vaccination would be more effective against endemic cholera than periodic campaigns . Vaccinating children averts more cases per dose than vaccinating all age groups , although vaccinating only children is unlikely to control endemic cholera in Bangladesh . Careful consideration must be made before generalizing these results to other regions .
Vibrio cholerae , the bacterium responsible for clinical cholera , has long been associated with the Bay of Bengal where it exists as an autochthonous member of the estuarine ecosystem [1] , [2] . This area of south Asia has been the origin for six of the seven cholera pandemics , and the burden of disease remains high . Today cholera is endemic in much of the Ganges River Delta with an estimated 350 , 000 treated cases per year in Bangladesh alone [3] . Improvements in water , sanitation , and hygiene are the long-term solutions for cholera , but oral cholera vaccines ( OCVs ) may constitute a shorter-term option to reduce morbidity and mortality from the disease . Oral cholera vaccines are safe and effective [4]–[6] . A recent large field trial in Kolkata , India , has shown that Shanchol , one of two World Health Organization prequalified OCVs , provides 65% protection over 5 years [5] . Further , successful demonstration campaigns conducted by the International Centre for Diarrhoeal Disease Research , Bangladesh ( icddr , b ) in urban and rural communities show promise for expanding vaccination coverage in Bangladesh [4] , [7] . The expanded use of OCV as a component of cholera control is supported by the recent decision by the World Health Organization to establish a global stockpile of 2 million doses of OCV , and the GAVI Alliance has committed to finance and leverage support for the global stockpile through 2018 [8] , [9] . Although OCV will likely be more widely used in the coming years , its effectiveness at a population level is not well understood . This information is crucial for planning vaccination programs on a large scale . It can be difficult to justify the widespread use of OCVs on economic grounds because: OCVs confer only moderate protection for a few years [5] , [10] , the incidence of cholera in most settings is relatively low , and the number of deaths attributed to cholera is relatively small because of the availability of inexpensive and effective treatment . However , large vaccine trials have shown that as OCV coverage increases , indirect protection from vaccination , also known as herd protection , increases [11] , [12] . When indirect protection is considered , the effectiveness of mass cholera vaccination can be high , and accounting for the effects of indirect protection appears to be necessary to make OCVs cost effective in the developing world [3] , [13]–[15] . Mathematical models can be used to predict the effectiveness of mass cholera vaccination , including indirect effects [16]–[20] , so modeling may be an essential component of any economic case for cholera vaccination . An earlier modeling study found that vaccinating 50 to 70% of the population of Bangladesh would virtually eliminate transmission [16] . Here , we expand upon that work , using a mathematical model to predict the effectiveness of targeting different age cohorts for vaccination at various coverage levels and schedules over a 20-year horizon .
Matlab is a rural community of approximately 220 , 000 people 30 kilometers southeast of Dhaka [21] . Data on cholera cases in Matlab were collected during long-term passive surveillance as described in detail elsewhere [22] . Briefly , between 1997 and 2001 , twice a month for three days a study physician attended to all patients presenting with acute watery diarrhea at the icddr , b clinic . Following orally obtained informed consent , these patients were tested for V . cholerae by rectal swab , and cultured on the same day in the Matlab laboratory [22] , [23] . Patient data , including age and sex , were obtained with identifying information removed . The Committee on Human Research of the Johns Hopkins University Bloomberg School of Public Health approved the research , and its guidelines were followed in the conduct of the clinical research . We developed an age-structured mathematical model of cholera transmission . Compartments in the model are unvaccinated susceptible ( S ) , vaccinated susceptible ( V ) , symptomatically infected ( I ) , asymptomatically infected ( A ) , or recovered and immune ( R ) from cholera ( Figure 1 ) . The concentration of V . cholerae in the environment ( water ) is tracked in an additional compartment ( W ) . Susceptible individuals may become infected by direct contact with infected individuals ( direct transmission ) or by exposure to V . cholerae in the environment ( indirect transmission ) . A complete description of the model is given in Text S1 . The model aggregates the population in compartments by disease status and age . Age cohorts represent children under 2 years old , pre-school aged children ( 2 to 4 years old ) , school aged children ( 5 to 14 years old ) , and adults ( 15 years old and older ) . Younger age groups are assumed to be more susceptible to infection [24] . Births are modeled by adding unvaccinated susceptibles to the youngest age cohort each year , and deaths are modeled by removing individuals from all age cohorts . Birth and age-specific mortality rates were based on data from the Matlab Health and Demographic Surveillance System [21] . Cohorts are aged by moving individuals into the next older age compartment at the appropriate rates . Frequency-dependent transmission rates are assumed for infections acquired through short cycle transmission ( person-to-person ) while a saturation ( Holling type II [25] ) function in terms of cholera concentration in water ( W ) is used to model the force of infection from long cycle transmission ( environmental exposure ) . A fraction p of the infections are symptomatic , a fraction r of which seek treatment . We refer to r as the reporting rate . The asymptomatically infected individuals ( proportion 1-p of all infections ) are less infectious and shed bacteria into the environment at a lower rate than cholera cases . Throughout this paper , “cholera cases” refers to the number of symptomatically infected individuals . The reporting rate of cholera cases is set to 10% in the main scenario and to 25% in alternative scenarios to test the sensitivity of results to this parameter [26] . Infected individuals recover after five days on average and are immune to infection until they transition back to the susceptible state after an average of 3 years [27] , [28] . An alternative scenario , assuming different duration of natural immunity protection across age groups , is also investigated . The model is calibrated to fit the dynamics of cholera cases recorded between 1997 and 2001 in Matlab . The proportion of recovered individuals at the beginning of 1997 is estimated based on time-series data of cholera incidence in Matlab [29] , the assumed reporting rate , and the duration of natural immunity . Two periods of increased environmental transmission occur annually with the first peak occurring in spring ( approximately April to May ) followed by a larger peak in autumn ( approximately September to November ) [30] , [31] . An iterative fitting procedure is implemented in which one cholera season ( 1997–1998 ) is simulated to estimate: i ) the initial distribution of the recovered individuals by age groups by running the model for 5 years and rescaling back to the estimated overall recovered proportion; ii ) the transmission rates for both short and long cycle transmission by fitting the number of monthly symptomatic infections based on data for reported cases; and iii ) the relative susceptibility of each age group by fitting the observed age distribution of cholera cases . Next , the estimated transmission rates are used to estimate the magnitude of elevated environmental risk during spring and autumn periods as well as the start time of the autumn period by fitting the monthly cases reported between 1998 and 2001 . The resulting best fit of the dynamics , minimizing the residual sum of squares for the number of reported cases per month , are presented in Figure 2 . Parameters used in the model are presented in Table S1 in Text S2 and the values of the estimated parameters are presented in Tables S2 & S3 in Text S2 . The magnitude of the elevated environmental risk during peak periods is sampled from the aggregated ranges in Table S3 in Text S2 , which represent the variation fitted over 5 consecutive annual cycles . Vaccinated susceptibles in the model are protected for an average of five years . Adults and children 5 years and older are 65% less likely to become infected upon exposure to cholera than unvaccinated individuals and children 1–4 years are 40% less likely to become infected [5] . OCVs may decrease the probability of developing symptoms upon infection [32] , but insufficient trial data is available to include this effect in the model . Therefore , we took a conservative approach and assumed that upon infection , vaccinated individuals have the same probability of becoming symptomatic as and are as infectious as non-vaccinated individuals . The model was implemented in Matlab R2012a ( The MathWorks , Inc . ) . We modeled vaccination programs that target one of three age groups: the entire vaccine-eligible population ( those one year old and older ) , all children ( ages one to fourteen years ) , and preschoolers ( ages one to four years ) . We did not model the vaccination of those under one year old , since no vaccine is currently licensed for that age group [33] . The age structure of the model does not precisely match the age cohorts targeted for vaccination . The age cohorts in the model were chosen to match the age groups in the available epidemiological data for calibration . The age cohorts for vaccination were chosen to match current vaccine licensing and logistical considerations . Vaccination of one-year-olds in all scenarios is modeled by targeting half of the population younger than 2 years old . If cholera vaccines are later licensed for use in infants ( i . e . , under one year old ) , one could vaccinate a larger fraction of the youngest age cohort in the model . We modeled three distinct schedules for vaccinating these target populations: one-time campaign , periodic campaigns , and continuous vaccination . For the one-time campaign , a proportion of the targeted population is vaccinated at the start of the first year only . For the periodic campaigns , every five years a proportion of all susceptible and recovered individuals are vaccinated . The period between campaigns was chosen to match the duration of vaccine protection . The continuous vaccination strategy is an approximation of an annual vaccination program . In this strategy , a proportion of the targeted population is vaccinated at the beginning of the first year , then starting in the second year the unvaccinated susceptible and recovered individuals are vaccinated at a fixed rate for the duration of the simulation . A detailed description of the implementations of all vaccination strategies in the model is in Text S1 . We define the overall effectiveness of mass vaccination to be the number of cholera cases prevented ( i . e . , the difference in the number of cases in a simulation without vaccination and the number of cases in a simulation with mass vaccination ) divided by the number of cases when there is no vaccination [34] . We measure the efficiency of a mass vaccination strategy by the number of vaccinations per case averted ( VPC ) , calculated as the number of people who are vaccinated divided by the number of cholera cases prevented .
Seasonal cholera transmission was simulated in a rural population in Bangladesh using a mathematical model calibrated to reproduce the two annual peaks ( Figures 2A and S1 ) and the age distribution of cases observed in surveillance from the community of Matlab ( Figure 2B ) . We found that infants and children younger than two years old , preschoolers ( ages 2 to 4 years ) and school children ( ages 5 to 14 years ) were 6 . 3 , 5 . 2 , and 1 . 8 times more susceptible than adults , respectively , to best fit the data assuming the same duration of immunity after infection across ages ( Table S2 and Figure S2 in Text S2 ) . We also estimated that asymptomatically infected individuals were 15% as infectious as the symptomatic cholera cases assuming that 20% of all infected individuals become symptomatic and 10% of cholera cases are reported . We compare the effectiveness over 20 years of one-time mass vaccination , recurring campaigns every five years , and continuous vaccination targeting 70% of all individuals one year old and older ( Figure 3 ) . All three vaccination strategies avert about 94% of the cholera cases in the first year ( Figures 3B and 3C ) . Vaccination of 50% of the population would reduce the incidence of cholera by 88% in the first year following vaccination ( Figure S3 in Text S2 ) . This is consistent with projections from a previous modeling study that found vaccination coverage of 50% would be sufficient to avert 93% of cholera cases in one season in Matlab [16] . With a one-time mass vaccination campaign , cholera incidence rebounds as protection from vaccine wanes and new susceptible individuals are born , and the overall effectiveness of the campaign is only about 20% after 20 years ( Figure 3C ) . Because protection conferred by vaccination lasts five years , one might choose to conduct campaigns once every five years for logistical reasons . However , susceptibility in the population accumulates between campaigns and the proportion of the population protected by vaccine drops to 20%–25% due to the waning of vaccine efficacy and the birth of new susceptible individuals ( Figure 3A ) . Vaccination campaigns every five years could result in 70% overall effectiveness over 20 years but cholera incidence oscillates and peaks in the years preceding each campaign ( Figure 3B ) . To avoid the fluctuations in vaccination coverage associated with 5-year campaigns , we modeled continuous vaccination in which people are vaccinated at a constant rate throughout the year every year after year 1 . When calibrated to use nearly the same amount of vaccine as the 5-year campaigns ( Figure 3D ) , 58% of the population is always protected by vaccine ( Figure 3A ) . When the population is continuously vaccinated , cholera incidence remains low over the 20 years with overall effectiveness above 95% ( Figure 3C ) , and onward cholera transmission is essentially interrupted after ten years . The continuous strategy achieves 25% higher overall effectiveness than the 5-year campaigns ( Figure 3C ) while using slightly less vaccine over 20 years ( Figure 3D ) . We compared the effectiveness of targeting different age groups with campaigns every five years . Our modeling results suggest that vaccinating everyone ( 100% of ) one year old and older at 5-year intervals would prevent 89% of cholera cases over 20 years ( Figure 4A , red boxes ) . The efficiency of the 5-year campaigns decreases with higher coverage , with the number of vaccinations per case averted ( VPC ) rising from 11 to 14 ( Figure 4B ) . Mass vaccination of all children 1 to 14 years old at 5-year intervals would prevent approximately 33% of cholera cases ( Figure 4A , blue boxes ) while vaccinating all preschoolers would prevent only 6% of cholera cases over 20 years ( Figure 4A , green boxes ) . Because the proportion of the population protected by vaccine drops between campaigns , this vaccination strategy is not able to suppress cholera activity over 20 years , even at 100% coverage ( Figure 4A ) . Targeting children ( 1 to 14 years old ) is most efficient; requiring about 11 VPC over a wide range of vaccination coverage levels ( Figure 4B ) . Targeting those 1 to 4 years old is less efficient , primarily because of the lower vaccine efficacy in this group . Continuous vaccination is associated with higher overall effectiveness than the 5-year campaigns . Coverage above 70% of the general population is sufficient to virtually interrupt onward transmission of cholera ( Figure 4C ) . VPC associated with the continuous vaccination declines with increasing coverage when children ( 1 to 14 years old ) are targeted . However , VPC increases , thus efficiency decreases , for coverage over 70% when the general population is vaccinated because effectively all cholera transmission is prevented above this level ( Figure 4D ) . If vaccine efficacy in young children were as high as that in adults , then children ages 1 to 4 years old would be the most efficient age group to target , and vaccinating 70% of them every 5 years would have a VPC of 7 , and maintaining 70% coverage with continuous vaccination would have a VPC of 6 . 5 ( Figure S4 in Text S2 ) . If the entire vaccine-eligible population were targeted , then this vaccine would be associated with only a modest increase in overall effectiveness compared to the vaccine with lower efficacy in young children ( Figure S4 in Text S2 ) . If the vaccine confers protection for only 3 years instead of 5 and has 65% efficacy among all age groups , one could achieve effectiveness similar to the 5-year campaigns described above by vaccinating every three years , but more vaccine would be required ( Figure S5 in Text S2 ) . Simulated cholera epidemics are sensitive to the assumed proportion of cases that seek treatment . An alternative scenario was calibrated assuming 25% of cholera cases seek treatment in Matlab , resulting in a lower underlying disease burden than the main analysis , which assumed a 10% reporting rate . This alternative scenario projects substantially smaller epidemics and consequently stronger impact of all vaccination programs ( Figures S6 and S7 in Text S2 ) . Approximately 30% coverage was enough for the 5-year campaigns and continuous vaccination to eliminate 90% of cholera cases . The same reduction is achieved by campaign vaccination of 80% or continuous vaccination of 60% of children ( 1 to 14 years old ) . However , the projected recovered fractions ( Figure S6B in Text S2 ) for all age groups are substantially lower compared to the extrapolations based on the Matlab data ( Figure S2 in Text S2 ) , which argues against the plausibility of this high reporting rate . We also modeled an alternative scenario in which children are protected against cholera for a shorter time than adults after infection . In this model , the fraction of susceptible individuals in each age group differed from those seen when all individuals become susceptible an average of three years after infection , and mass vaccination was somewhat more effective ( Figures S8–S9 in Text S2 ) .
We used a mathematical model to explore the potential effectiveness of mass cholera vaccination in rural Bangladesh and believe that the results apply more broadly to cholera endemic areas in Bangladesh . With the model , we were able to predict the overall effectiveness , which includes indirect effects , of different mass vaccination strategies . Our results indicate that maintaining 60% or higher vaccine coverage in the population would stop cholera transmission , which is consistent with an earlier modeling study [16] . However , a continuous vaccination schedule might be difficult to implement , as it requires a constant effort to keep a substantial proportion of the population protected by vaccine by identifying unvaccinated individuals and vaccinating them and revaccinating individuals as protection from vaccines wane . The continuous vaccination strategy as described is a mathematical idealization of vaccination efforts that occur throughout the year rather than vaccination campaigns that occur every few years . Vaccination campaigns that occur only once a year would maintain approximately the same level of vaccine protection in the population while being more logistically practical . We also model mass vaccination campaigns that occur once every five years , the average duration of protection from vaccination [5] . This strategy might be easier to implement , but as vaccine protection wanes and birth ( and possibly immigration ) introduces new susceptible individuals to the population between campaign years , large cholera outbreaks can occur . We found that vaccinating all vaccine-eligible children , ages 1–14 years , requires the fewest number of vaccinations per case averted compared to vaccinating preschool-aged children ( 1–4 years ) or the general population ( ages 1 year and older ) . Although preschool-aged children have the greatest burden of cholera , represented by both disease incidence and mortality [24] , [35] , selectively vaccinating this group is the least efficient strategy , primarily due to the lower modeled vaccine efficacy in this age group ( 40% ) [5] . Delivering OCV to children could build upon existing delivery mechanisms like the Expanded Programme on Immunization ( EPI ) or National Immunization Days [3] . However , our analysis suggests that endemic cholera is unlikely to be eliminated by vaccinating only children . A major consideration of immunization plans could be the most efficient use of the limited supply of doses , currently around two million globally but anticipated to expand over the next five years [8] , [9] . Vaccinating populations with the highest risk of disease is efficient and also supported by evidence from cost-effectiveness analyses , the priorities of decision makers , and health equity considerations [3] , [15] , [19] , [36] . Previous cost-effectiveness studies have found that untargeted mass cholera vaccination in Bangladesh may not be effective unless one accounts for indirect protection [3] , [13] . However , the magnitude of indirect protection is difficult to estimate without proper studies and/or mathematical modeling . The results from this study suggest significant indirect protection from OCV that may improve the economic case for expanding its use . Although we do not evaluate the cost-effectiveness of the modeled vaccination strategies , the number of vaccinations per case averted can be used to estimate cost-effectiveness . If it requires between 10–25 vaccinations per case averted ( Figures 4B , 4D ) and costs $5 . 33 to fully vaccinate an individual ( two doses at a public sector cost of $1 . 85 per dose [3] and a delivery cost of $1 . 63 per individual [4] ) , the vaccination programs considered here would cost between $53–$133 per cholera case averted , and assuming a 1 . 5% case fatality rate [3] , $3 , 500–$8 , 900 per death averted . There are several limitations to this study . The model was calibrated to match the demographic and epidemiologic characteristics of cholera in Matlab , Bangladesh; so extrapolating the results from this study to other settings requires careful consideration . We modeled transmission of cholera in an endemic setting where the incidence is much higher in children than adults . However , data from cholera outbreaks in non-endemic settings suggest a more even distribution of cholera incidence by age [24] , [28] , [37] . Therefore , the results described apply to regions that experience annual cholera outbreaks at a scale similar to Matlab's , but the model should be recalibrated for settings with substantially different epidemiology or demography . The actual cause of the relatively high cholera incidence among children in Bangladesh is not known , but it has been hypothesized to be due to higher levels of previous exposure in adults and differences in the immune system in children and adults [38]–[40] . The model assumptions required to create this differential susceptibility could affect the effectiveness of mass vaccination . We assumed that the differences in cholera incidence between age groups were due largely to differential intrinsic susceptibility . We had tested an alternative hypothesis that the duration of immunity conferred by infection differed between age groups and could explain the differences in incidence , but this model also required differential susceptibility to fit the data from Matlab . The model was not intended for the prediction of cholera activity for a particular year . We calibrated the model using five years of data from Matlab , assuming that the epidemiology of cholera will not change substantially , so the 20-year projections described here should be considered average outcomes over this time horizon . We assumed that current demographic trends , sanitation levels , and climate would remain constant over the next 20 years , but changes in population movement , development , rainfall , the frequency of severe flooding events , sea level , and ocean temperature could change the epidemiology of cholera [41]–[44] . There is growing momentum toward incorporating oral cholera vaccine into cholera control and outbreak response planning . Field and feasibility trials have been conducted in urban and rural Bangladesh and there appears to be interest to include targeted OCV use as part of comprehensive cholera control strategies [3] , [4] , [33] , [45] , [46] . As Bangladesh and other countries begin to consider the role of OCV in comprehensive cholera control plans , this work provides insight into how OCV may diminish cholera transmission dynamics . This analysis demonstrates that mass immunization with oral cholera vaccines may greatly reduce the burden of disease , and mathematical modeling can provide guidance on the targeting of populations and scheduling of campaigns to maximize impact . | Bangladesh has a high burden of cholera and may become the first country to use cholera vaccine on a large scale . Mass cholera vaccination may be hard to justify to international funding agencies because of the modest efficacy of existing vaccines and their limited duration of protection . However , mass cholera vaccination can induce high levels of indirect protection in a population , i . e . , protecting even unvaccinated individuals by lowering cholera incidence , and a case for cost-effective cholera vaccination could be made . Mathematical modeling is one way to predict the magnitude of indirect protection conferred by a proposed vaccination program . Here , we predict the effectiveness of various mass cholera vaccination strategies in Bangladesh using a mathematical model . We found that maintaining high levels of vaccination coverage in children could be very effective in reducing the burden of cholera , and secondary transmission of cholera would virtually stop when 70% of the population is vaccinated . Mathematical modeling may play a key role in planning widespread cholera vaccination efforts in Bangladesh and other countries . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"vaccines",
"bacterial",
"diseases",
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"cholera",
"cholera",
"vaccines",
"population",
"modeling",
"neglected",
"tropical",
"diseases",
"immunology",
"biology",
"and",
"life",
"sciences",
"infectious",
"disease",
"modeling",
"computational",
"biology",
"tropical",
"diseases",
"vaccination",
"and",
"immunization"
]
| 2014 | Comparative Effectiveness of Different Strategies of Oral Cholera Vaccination in Bangladesh: A Modeling Study |
C . elegans undergoes periods of behavioral quiescence during larval molts ( termed lethargus ) and as adults . Little is known about the circuit mechanisms that establish these quiescent states . Lethargus and adult locomotion quiescence is dramatically reduced in mutants lacking the neuropeptide receptor NPR-1 . Here , we show that the aroused locomotion of npr-1 mutants results from the exaggerated activity in multiple classes of sensory neurons , including nociceptive ( ASH ) , touch sensitive ( ALM and PLM ) , and stretch sensing ( DVA ) neurons . These sensory neurons accelerate locomotion via both neuropeptide and glutamate release . The relative contribution of these sensory neurons to arousal differs between larval molts and adults . Our results suggest that a broad network of sensory neurons dictates transitions between aroused and quiescent behavioral states .
Animals undergo periods of behavioral quiescence and arousal in response to changes in their environment and metabolic state . Arousal is defined as a state of heightened responsiveness to external stimuli coupled with increased motor activity whereas quiescence is associated with diminished responsiveness and motor activity [1] . Quiescence and arousal can persist for minutes to hours . Arousal is associated with fear , stress , hunger , and exposure to sexual partners [1] , while quiescence is associated with sleep and satiety [2] . Relatively little is known about the specific circuit mechanisms leading to arousal or quiescence . In particular , it is unclear if similar mechanisms mediate quiescence and arousal in response to different cues , or at different times during development . To address this question , we have analyzed arousal and quiescence of C . elegans locomotion . During each larval molt , C . elegans undergoes a prolonged period of profound behavioral quiescence , termed lethargus behavior , whereby locomotion and feeding behaviors are inactive for approximately 2 hours [3] . Lethargus has properties of a sleep-like state such as reduced sensory responsiveness and homeostatic rebound of quiescence following perturbation [4] . Several genes and molecular pathways involved in lethargus behavior have been identified [4–11] . Multiple sensory responses are diminished during lethargus , including those mediated by a nociceptive neuron ( ASH ) [12] , and by mechanosensory neurons [11 , 13] . Mutants lacking NPR-1 Neuropeptide Y ( NPY ) receptors have been utilized as a model for generalized arousal . NPR-1 inhibits the activity of a central sensory circuit that is defined by gap junctions to the RMG interneuron [14] . In npr-1 mutants , responses mediated by the RMG circuit ( e . g . pheromone and oxygen avoidance ) are exaggerated , and this heightened acuity is associated with exaggerated locomotion ( both during lethargus and in adults ) [11 , 14–16] . Mutations that increase ( e . g . npr-1 ) and decrease ( e . g . tax-4 CNG and osm-9 TRPV ) RMG circuit activity are associated with locomotion arousal and quiescence respectively [11 , 14 , 17 , 18] . We previously showed that locomotion quiescence during lethargus is dramatically reduced in npr-1 mutants and that this effect requires increased RMG sensory activity [11] . Subsequent studies showed that in microfluidic chambers npr-1 mutants have modest defects in lethargus quiescence when sensory cues are minimized but that dramatic quiescence defects are observed following brief stimulation with light or vibration [19 , 20] . Taken together , these papers suggest that npr-1 mutants exhibit aroused locomotion as a consequence of enhanced sensory activity . The arousing effects of the RMG circuit are mediated in part by secretion of a neuropeptide , pigment dispersing factor ( PDF-1 ) [11] . Activation of PDF receptors ( PDFR-1 ) in peripheral mechanosensory neurons enhances sensitivity to vibration , thereby accelerating locomotion . Thus , sensory evoked activity in the RMG circuit arouses locomotion during lethargus through changes in PDF-1 and PDFR-1 signaling . These results raise several interesting questions . Which specific sensory neurons are responsible for arousal ? Does the RMG circuit regulate arousal via multiple outputs ( i . e . in addition to PDF-1 ) ? Does the RMG circuit function similarly during lethargus and in adults ? Is diminished sensory acuity during lethargus required for behavioral quiescence ? Here we show that glutamatergic transmission promotes arousal , we identify glutamatergic neurons and glutamate receptors that mediate arousal , and we show that arousal occurs by distinct mechanisms in lethargus and adult animals .
Adult npr-1 mutants exhibit accelerated locomotion ( Fig 1A–1C ) , as shown in prior studies [21] . Faster adult locomotion suggests that locomotion circuit activity has been altered . Consistent with this idea , npr-1 mutant adults have enhanced sensitivity to the paralytic effects of a cholinesterase inhibitor ( aldicarb ) ( Fig 1D–1F and S2A Fig ) [22] , indicating increased excitatory transmission at neuromuscular junctions ( NMJs ) . To more directly assess changes in synaptic transmission , we recorded miniature excitatory post-synaptic currents ( mEPSCs ) in body muscles , which are evoked by acetylcholine ( ACh ) release at NMJs . The mEPSC rate observed in npr-1 adults was significantly higher than in wild type controls while mEPSC amplitudes were unaltered ( Fig 1G–1I ) . Faster mEPSC rates suggest that ACh release from motor neurons was increased whereas unaltered mEPSC amplitudes imply that muscle responsiveness to secreted ACh was unaffected . By contrast , neither ACh release evoked by depolarizing motor neurons with a stimulating electrode ( evoked EPSCs ) , nor transmission at GABAergic NMJs ( assessed by miniature inhibitory post-synaptic currents , mIPSCs ) was altered in npr-1 mutants ( S1 Fig ) . This constellation of electrophysiological defects suggests that tonic ACh release ( assessed by mEPSC rate ) was enhanced in npr-1 mutants , whereas other forms of neurotransmitter release ( evoked ACh release and tonic GABA release ) were unaffected . Enhanced tonic ACh release at NMJs could account for the accelerated locomotion rate observed in npr-1 adults . Prior studies showed that several behavioral phenotypes exhibited by npr-1 mutants are caused by enhanced sensitivity to environmental cues . In particular , sensory responses mediated by the RMG circuit are enhanced in npr-1 mutants [14 , 17 , 18] and this enhanced sensory acuity is required for accelerated locomotion rates during lethargus [11 , 20] . We did several experiments to determine if enhanced RMG circuit activity is also required for increased cholinergic transmission in npr-1 adults . A transgene restoring npr-1 expression in the RMG circuit ( using the flp-21 promoter ) rescued the accelerated locomotion ( Fig 1B ) , enhanced aldicarb sensitivity ( Fig 1D and S2A Fig ) , and faster mEPSC rate ( Fig 1G–1I ) defects of npr-1 adults . By contrast , an npr-1 transgene expressed in GABAergic neurons lacked rescuing activity ( Fig 1D–1H ) . These results indicate that NPR-1 acts in the RMG circuit to slow adult locomotion . Similarly , mutations inactivating ion channels required for sensory transduction ( TAX-4/CNG and OCR-2/TRPV ) in the RMG circuit suppressed the npr-1 adult locomotion ( Fig 1C ) , aldicarb sensitivity ( Fig 1E and 1F and S2B and S2C Fig ) , and mEPSC rate ( Fig 1J and 1K ) defects . Collectively , these results suggest that the accelerated adult locomotion exhibited by npr-1 mutants is caused by heightened activity in the RMG sensory circuit and , consequently , corresponds to an aroused state . We previously showed that the lethargus quiescence defects exhibited by npr-1 mutants are caused by increased secretion of Pigment dispersing factor ( PDF-1 ) by cells in the RMG circuit [11] . Because PDF-1 secretion is also increased in npr-1 adults [11] , we tested the idea that the hyperactive adult locomotion of npr-1 mutants is also caused by increased PDF signaling . Contrary to this idea , we found that pdf-1 and pdfr-1 ( PDF Receptor-1 ) mutations reduced but did not eliminate the aldicarb hypersensitivity ( Fig 2A and 2B and S2D and S2E Fig ) , the accelerated locomotion ( Fig 2C ) , and increased mEPSC rate ( Fig 2D and 2E ) defects of npr-1 adults . Collectively , these results suggest that additional excitatory outputs from the RMG circuit ( i . e . beyond PDF-1 ) must contribute to the aroused locomotion of npr-1 adults . Many C . elegans sensory neurons are glutamatergic , including two neurons in the RMG circuit ( ASH and ASK ) and the body touch neurons [23] . To determine if glutamate release by sensory neurons is required for accelerated locomotion in npr-1 mutants , we analyzed mutations that inactivate the vesicular glutamate transporter ( eat-4 VGLUT ) , which is primarily expressed in sensory neurons [23] . eat-4 VGLUT mutations blocked the increased motile fraction and locomotion speed of npr-1 mutants both during the L4-Adult ( L4/A ) molt ( Fig 3A–3C ) and in adults ( Fig 3D and 3E ) . eat-4 mutations also blocked the hypersensitivity to aldicarb ( Fig 3F and S2F Fig ) and increased mEPSC rate ( Fig 3G and 3H ) defects of npr-1 adults . Transgenes restoring EAT-4 expression in touch neurons and ASH neurons partially reinstated both lethargus ( Fig 3B and 3C ) and adult locomotion ( Fig 3D and 3E ) defects in eat-4; npr-1 double mutants , whereas transgenes expressed in ASK lacked rescuing activity ( Fig 3B and 3C ) . eat-4 transgenes had no effect on lethargus quiescence in wild type animals ( S3 Fig ) . These results suggest that glutamate released by ASH and touch neurons arouses locomotion in L4/A and adult npr-1 mutants . The preceding results suggest that ASH synaptic output arouses locomotion in npr-1 mutants . We did several additional experiments to test this idea . If altered ASH output were required for aroused locomotion , we would expect that npr-1 mutants lacking ASH neurons would have increased locomotion quiescence . To test this idea , we induced ASH cell death with a transgene that expresses the pro-apoptotic caspase CED-3 . Killing ASH significantly decreased the L4/A motile fraction and locomotion rate in npr-1 mutants ( Fig 4A–4C ) . By contrast , ASH ablation had little effect on the locomotion rate of npr-1 adults ( Fig 4D ) . To determine if ASH activity is increased in npr-1 mutants during lethargus , we examined sensory-evoked calcium responses in ASH , using the genetically encoded calcium indicator Cameleon . ASH mediates avoidance responses to copper and hyper-osmotic stimuli . Consistent with a recent study [12] , the magnitude of copper ( Fig 4E and 4F ) and glycerol-evoked ( S4A and S4B Fig ) calcium transients in ASH was significantly decreased during lethargus in wild-type animals . Decreased ASH responsiveness to copper and glycerol during L4/A lethargus was blocked in npr-1 mutants , whereas ASH responsiveness in adults was unaltered in npr-1 mutants ( Fig 4E and 4F and S4A and S4B Fig ) . Transgenes expressing NPR-1 in the RMG circuit ( using the flp-21 promoter ) or in ASH ( using the sra-6 promoter ) reinstated the L4/A decrease in copper and glycerol-evoked ASH calcium transients in npr-1 mutants ( Fig 4G and 4H and S4C and S4D Fig ) . These results suggest that NPR-1 acts in ASH to inhibit sensory responses and that increased ASH activity is required for accelerated locomotion of npr-1 mutants during lethargus but not in adults . To determine if increased ASH activity is sufficient to arouse locomotion , we analyzed locomotion after artificially depolarizing ASH neurons . For this experiment , we utilized transgenic animals that express rat TRPV1 capsaicin receptors in ASH neurons [24] . In these animals , capsaicin treatment evokes ASH-mediated avoidance behaviors [24] . A 5-hour capsaicin treatment had little effect on L4/A motile fraction and locomotion velocity [11] , whereas capsaicin treatment significantly accelerated adult locomotion and increased aldicarb sensitivity ( Figs 4I–5J and S2G Fig ) . These effects were not observed in animals lacking TRPV1 expression in ASH neurons ( Fig 4I and 4J ) . Thus , forced ASH depolarization was sufficient to arouse adult but not lethargus locomotion . Collectively , these results suggest that diminished and heightened ASH activity is associated with locomotion quiescence and arousal respectively; however , the magnitude of ASH’s arousing effects differ between lethargus and adult animals . Which glutamate receptors arouse locomotion in npr-1 mutants ? Glutamate-activated cation channels , AMPA ( GLR-1 and -2 ) and NMDA ( NMR-1 and -2 ) receptors , mediate excitatory transmission at ASH-interneuron [25–27] . The npr-1 L4/A quiescence defect was abolished in glr-2; npr-1 double mutants ( Fig 5A–5C ) , while glr-1 mutations had no effect ( Fig 5D and 5E ) . By contrast , glr-1 , glr-2 , and nmr-1 mutations had little effect on npr-1 adult locomotion ( Fig 5F and S5 Fig ) . Similarly , glr-2 mutations did not block the increased mEPSC rate in npr-1 adults ( Fig 5G ) . These results suggest that GLR-2 AMPA receptors are specifically required for the aroused locomotion during the L4/A lethargus in npr-1 mutants . Which synaptic targets of ASH and touch neurons mediate locomotion arousal ? To address this question , we identified the neurons in which GLR-2 function is required . Aroused L4/A locomotion requires GLR-2 but not GLR-1 receptors; consequently , we reasoned that the relevant neurons are likely to express GLR-2 but not GLR-1 . GLR-1 and GLR-2 are co-expressed in many neurons; however , a few GLR-2-expressing neurons lack GLR-1 , including DVA ( a stretch-activated neuron ) and AIA ( an interneuron in the head ganglia ) [25–27] . The L4/A quiescence defect was partially restored in glr-2; npr-1 double mutants by transgenes expressing GLR-2 in DVA and AIA neurons , whereas transgenes expressed in the ventral cord interneurons ( using the glr-1 promoter ) failed to rescue ( Fig 5B and 5C ) . Transgenic expression of GLR-2 in DVA or AIA had no effect on lethargus quiescence in wild type worms ( S3 Sig ) . These results suggest that GLR-2 AMPA receptors expressed in AIA and DVA neurons arouse L4/A locomotion in npr-1 mutants . DVA receives direct synaptic input from the touch neuron PLM while AIA receives direct input from ASH [28] . Thus , increased transmission at ASH-AIA and PLM-DVA synapses could account for GLR-2’s effects on locomotion rate . Because we only observed partial rescue by glr-2 transgenes expressed in AIA and DVA , it is likely the GLR-2 function is required in additional ( as yet unidentified ) neurons . How do AIA and DVA arouse locomotion ? AIA neurons provide synaptic input to ASK and ASI , both of which express PDF-1 [11 , 29] . Thus , heightened AIA activity could arouse locomotion by enhancing PDF-1 secretion . To assess the level of PDF-1 secretion , we analyzed PDF-1::YFP fluorescence in the endolysosomal compartment of coelomocytes , which are specialized scavenger cells that internalize proteins secreted into the body cavity [30 , 31] . Inactivating GLR-2 did not alter PDF-1::YFP fluorescence in coelomocytes in both adult and L4/A animals ( Fig 6 ) . These results suggest that the arousing effects of GLR-2 are not mediated by changes in PDF secretion . DVA neurons receive direct synaptic input from the PLM touch neurons [32] , and secrete NLP-12 ( a neuropeptide that accelerates locomotion ) [33] . Thus , increased DVA activity could contribute to locomotion arousal in npr-1 mutants . Three results support this idea . First , PLM neurons exhibit enhanced touch-evoked calcium responses in adult npr-1 mutants ( S6 Fig ) . Thus , PLM neurons have increased sensory acuity in npr-1 mutants , similar to the effect we previously showed for ALM neurons [11] . Second , inducing DVA cell death ( with a CED-3 transgene ) significantly reduced npr-1 locomotion rate during L4/A lethargus ( Fig 5H–5J ) , but not in adults ( Fig 5K ) . Third , DVA secretion of NLP-12 is significantly increased in npr-1 mutants [33] , indicating increased DVA activity . These results suggest that PLM neurons provide enhanced excitatory input to DVA in npr-1 mutants , which promotes aroused L4/A locomotion .
Multiple classes of sensory neurons arouse locomotion during lethargus and in adults , including: mechanosensory neurons ( ALM and PLM ) , a nociceptive neuron ( ASH ) , a pheromone sensing neuron ( ASK ) , and a stretch sensing neuron ( DVA ) . Lethargus quiescence is accompanied by diminished sensory-evoked responses in ALM , PLM , and ASH ( this study and [11–13] ) . PDF-1 secretion from ASK neurons is significantly reduced during lethargus , implying that ASK neurons also have diminished activity during lethargus [11] . npr-1 mutations prevent the dampened ALM ( mechanosensory ) and ASH ( nociceptive ) responses during lethargus and this was accompanied by decreased locomotion quiescence ( this study and [11] ) . The arousing effects of npr-1 mutations are blocked ( or diminished ) by mutations that decrease sensory responsiveness ( e . g . tax-4 CNG and osm-9 TRPV mutations ) [11] , or by ablating sensory neurons ( e . g . ASH and DVA ) . Forced activation of ASH neurons arouses adult locomotion . Collectively , these results imply that a broad network of sensory neurons arouses locomotion , which allows C . elegans to adapt its behavior across a broad range of developmental and physiological circumstances . NPR-1 promotes behavioral quiescence by diminishing the sensitivity of many sensory modalities . NPR-1 directly inhibits ASH responses and indirectly inhibits other sensory neurons ( ALM , PLM , and DVA ) via decreased glutamate and neuropeptide release . Thus , gating of sensory perception by NPR-1 provides a circuit mechanism for producing aroused and quiescent locomotion in C . elegans . Our results do not exclude the possibility that additional mechanisms ( beyond sensory gating by NPR-1 ) contribute to arousal and quiescence . Both quiescence ( during lethargus ) and arousal ( following molts ) persist in microfluidic chambers where many sensory cues are minimized [19] . In particular , oxygen tension is likely to be very low in these chambers , which would greatly diminish NPR-1’s effects on behavior [15 , 16] . Thus , the quiescence and arousal exhibited in microfluidic chambers implies that additional mechanisms beyond NPR-1 must contribute to expressing these behavioral states . It will be interesting to determine if these NPR-1 independent mechanisms also act by gating sensory activity . Sensory neurons release glutamate and/or neuropeptides in response to external cues , which then engage downstream motor circuits in behavioral outputs . Our prior study shows that sensory-evoked PDF-1 secretion promotes locomotion arousal by enhancing touch neuron responsiveness . Neuropeptides also mediate arousal in flies ( PDF ) [34] , fish and mammals ( orexin/hypocretin ) [35 , 36] . Here we show that sensory evoked glutamate release also plays a role in arousal . Mutations inactivating the EAT-4/VGLUT decreased locomotion arousal in lethargus and in adults . EAT-4 is almost exclusively expressed in sensory neurons [23] and transgenes restoring EAT-4 in touch neurons and ASH neurons re-instates locomotion arousal in npr-1 mutants . These results suggest that sensory neurons utilize both glutamate and neuropeptides as excitatory outputs to arouse locomotion . Our results suggest that exaggerated glutamate release at ASH-AIA and PLM-DVA synapses arouses locomotion during lethargus in npr-1 mutants . ASH and PLM neurons have enhanced sensory evoked activity in npr-1 mutants , which is expected to produce enhanced glutamate release at ASH-AIA and PLM-DVA synapses . GLR-2 receptors are expressed in AIA and DVA . glr-2 mutations block the aroused L4/A locomotion of npr-1 mutants and arousal is re-instated by transgenes expressing GLR-2 in AIA and DVA . Finally , calcium responses in AIA [14] , and neuropeptide secretion from DVA [33] are both enhanced in npr-1 mutants , indicating that these neurons have increased activity . We observed only partial rescue of aroused locomotion by transgenes restoring EAT-4 expression in ASH and touch neurons or by those expressing GLR-2 in AIA or DVA; consequently , it is likely that glutamate released by other sensory neurons also contributes to the aroused L4/A locomotion in npr-1 mutants . Much less is known about the role of glutamate in arousal in other systems . Glutamate release has widespread effects throughout the brain in mammals , which complicates the analysis of its effects on arousal . Microinjection of glutamate or AMPA into lateral hypothalamic area increased locomotor activity and duration of waking episodes in rodents [37 , 38] , while microdialysis of CNQX , an AMPA receptor antagonist , into the thalamus promotes sleep in cats [39] . Glutamate also induces fictive locomotion in lamprey [40] . In these cases , however , the circuit mechanisms underlying glutamate’s arousing effects are not known . Mutants lacking NPR-1 exhibit accelerated locomotion in adults and during lethargus [11 , 18] . Several results suggest that locomotion arousal in adult and lethargus is established by a shared central sensory circuit . First , in both adult and lethargus , enhanced activity in the RMG sensory circuit accelerates locomotion , whereas decreased sensory transduction in the RMG circuit ( i . e . by inactivating TAX-4 or OSM-9 ) abolishes npr-1’s hyperactive locomotion defect [11 , 14] , suggesting that the RMG circuit activity stimulates arousal in both awake and quiescent states . Second , EAT-4 acts in ASH and touch neurons to mediate hyperactive locomotion of npr-1 adult and lethargus stage animals , suggesting that glutamate release from these sensory neurons is required for locomotion arousal in npr-1 mutants . On the other hand , several results suggest that the mechanisms that arouse locomotion differ between adult and lethargus animals . Inactivating GLR-2 AMPA receptors blocks the hyperactive locomotion of npr-1 mutants during lethargus but not in adults . Aroused locomotion in npr-1 adults persists in glr-1 , glr-2 , and nmr-1 mutants , indicating that other glutamate receptors are responsible for arousing adult locomotion . Similarly , artificial activation of ASH accelerates adult but not lethargus locomotion . Collectively , our results suggest that multiple sensory circuits govern locomotion arousal throughout development but that the relative contribution of each circuit to arousal differs depending on the developmental stage .
Strain maintenance and genetic manipulation were performed as described [41] . Animals were cultivated at 20°C on agar nematode growth media ( NGM ) seeded with OP50 ( for imaging and behavior ) or HB101 E . coli ( for electrophysiology ) . Wild type reference strain was N2 Bristol . Strains used in this study are as follows: KP6048 npr-1 ( ky13 ) X DA609 npr-1 ( ad609 ) X KP6064 npr-1 ( ok1447 ) X PR678 tax-4 ( p678 ) III CX4544 ocr-2 ( ak47 ) IV LSC27 pdf-1 ( tm1996 ) III KP6340 pdfr-1 ( ok3425 ) III MT6308 eat-4 ( ky5 ) III KP0004 glr-1 ( n2461 ) III VM487 nmr-1 ( ak4 ) II KP6057 ocr-2 ( ak47 ) IV;npr-1 ( ok1447 ) X KP6058 ocr-2 ( ak47 ) IV;npr-1 ( ky13 ) X KP6060 tax-4 ( p678 ) III;npr-1 ( ky13 ) X KP6061 tax-4 ( p678 ) III;npr-1 ( ok1447 ) X KP6100 pdf-1 ( tm1996 ) III;npr-1 ( ky13 ) X KP6410 pdfr-1 ( ok3425 ) III;npr-1 ( ky13 ) X KP6349 eat-4 ( ky5 ) III; npr-1 ( ky13 ) X CX4978 kyIs200[sra-6p::VR1 , elt-2p::NLS-gfp] ( Gift from Cori Bargmann ) KP6414 nmr-1 ( ak4 ) II; npr-1 ( ky13 ) X KP6415 glr-1 ( n2461 ) III;npr-1 ( ky13 ) X VM1123 dpy-19 ( n1347 ) glr-2 ( ak10 ) III KP6740 dpy-19 ( n1347 ) glr-2 ( ak10 ) III; npr-1 ( ky13 ) X KP7362 npr-1 ( ky13 ) X; nuIs439[nlp-12p::GFP]; nuIs519[nlp-12p::ced-3::GFP , vha-6::mCherry] KP6693 nuIs472 [pdf-1p::pdf-1::venus , vha-6p::mCherry] KP6743 npr-1 ( ky13 ) X; nuIs472 KP7194 dpy-19 ( n1347 ) glr-2 ( ak10 ) III; nuIs472 KP7195 dpy-19 ( n1347 ) glr-2 ( ak10 ) III; npr-1 ( ky13 ) X; nuIs472 AQ906 bzIs17[mec-4p::YC2 . 12] KP6681 npr-1 ( ky13 ) X; bzIS17 CX9396 npr-1 ( ad609 ) X;kyEx1966[flp-21p::npr-1 SL2 GFP , ofm-1p::dsRed] ( Gift from Cori Bargmann ) KP6051 npr-1 ( ad609 ) X;nuEx1519[unc-25p::npr-1::gfp , myo-2p::NLS-mCherry] KP6053 npr-1 ( ad609 ) X;nuEx1520[unc-30p::npr-1::gfp , myo-2p::NLS-mCherry] KP7149 , KP7150 eat-4 ( ky5 ) III; npr-1 ( ky13 ) X; nuEx1613-1614[sra-6p::eat-4 , myo-2p::NLS-mCherry] KP7176 , KP7177 eat-4 ( ky5 ) III; npr-1 ( ky13 ) X; nuEx1615-1616[sra-9p::eat-4 , vha-6p::mCherry] KP7198 , KP7199 eat-4 ( ky5 ) III; npr-1 ( ky13 ) X; nuEx1640-1641[mec-4p::eat-4 , vha-6p::mCherry] KP7442 npr-1 ( ky13 ) X; nuEx1684[sra-6p::ced-3::GFP , sra-6p::mCherry , vha-6p::mCherry] KP7633 nuEx1613[sra-6p::eat-4 , myo-2p::NLS-mCherry] KP7634 nuEx1640[mec-4p::eat-4 , vha-6p::mCherry] AQ3304 ljEx239[sra-6::YC . 360] KP7353 npr-1 ( ky13 ) X; ljEx239 KP7443 npr-1 ( ky13 ) X; ljEx239; nuEX1607[flp-21p::npr-1 , myo-2p::NLS-mCherry] KP7495 npr-1 ( ky13 ) X; ljEx239; nuEX1683[sra-6p::npr-1 , vha-6p::mCherry] KP7191 dpy-19 ( n1347 ) glr-2 ( ak10 ) III; npr-1 ( ky13 ) X; nuEx1637[nlp-12p::glr-2 ( gDNA ) , myo-2p::NLS-mCherry] KP7192 dpy-19 ( n1347 ) glr-2 ( ak10 ) III; npr-1 ( ky13 ) X; nuEx1638[gcy-28 ( d ) p::glr-2 ( gDNA ) , vha-6p::mCherry] KP7354 , KP7355 , KP7356 dpy-19 ( n1347 ) glr-2 ( ak10 ) III; npr-1 ( ky13 ) X; nuEx1642-1644[glr-1p::glr-2 ( gDNA ) , vha-6p::mCherry] KP7635 nuEx1637[nlp-12p::glr-2 ( gDNA ) , myo-2p::NLS-mCherry] KP7636 nuEx1638[gcy-28 ( d ) p::glr-2 ( gDNA ) , vha-6p::mCherry] Transgenic strains were generated by microinjection of various plasmids with coinjection markers ( myo-2p::NLS-mCherry ( KP#1480 ) and vha-6p::mcherry ( KP#1874 ) ) . Injection concentration was 40–50 ng/μl for all the expression constructs and 10 ng/μl for coinjection markers . The empty vector pBluescript was used to bring the final DNA concentration to 100 ng/μl . The flp-21 promoter ( which is expressed in the RMG , ASH , ADL , ASK , URX , and ASI neurons [14] ) was used to express transgenes in the RMG circuit . Lethargus locomotion was analyzed as previously described [11] . Well-fed late L4 animals were transferred to full lawn OP50 bacterial plates . After 1 hour , locomotion of animals in lethargus ( determined by absence of pharyngeal pumping ) was recorded on a Zeiss Discovery Stereomicroscope using Axiovision software . Locomotion was recorded at 2 Hz for 60 seconds . Centroid velocity of each animal was analyzed at each frame using object-tracking software in Axiovision . Motile fraction of each animal was calculated by dividing the number of frames with positive velocity value with total number of frames . Speed of each animal was calculated by averaging the velocity value at each frame . Quantitative analysis was done using a custom written MATLAB program ( Mathworks ) . Statistical significance was determined using one-way ANOVA with Tukey test for multiple comparisons and two-tailed Student’s t test for pairwise comparison . Locomotion of adult animals was analyzed with the same setup as lethargus locomotion analysis described above , except that well-fed adult animals were monitored 1–1 . 5hr after the transfer to full lawn OP50 bacterial plates . For the capsaicin treatment ( Fig 4I ) , 1 day-old animals were transferred to NGM plates containing 50 μM capsaicin ( with food ) , treated with capsaicin for 5 hours , and recorded for their locomotion . Statistical significance was determined using one-way ANOVA with Tukey test for multiple comparisons and two-tailed Student’s t test for pairwise comparison . Neurons were ablated in npr-1 ( ky13 ) mutant worms by transgenes co-expressing CED-3 and a fluorescent protein ( GFP or mCherry ) under the sra-6 ( ASH ablation ) or nlp-12 ( DVA ablation ) promoters . ASH or DVA ablations were confirmed after locomotion analysis by fluorescence microscopy . Sensitivity to aldicarb was determined by analyzing the time course of paralysis following treatment with 1 mM aldicarb ( Sigma-Aldrich ) as previously described [42] . Briefly , movement of animals was assessed by prodding animals with a platinum wire every 10 minute following exposure to aldicarb . 20–30 animals were tested for each trial . For the capsaicin treatment ( Fig 4J ) , adult animals were transferred to NGM plates containing 50 μM capsaicin ( with food ) , treated with capsaicin for 2–3 hours , and assayed for their paralysis on 1 mM aldicarb plates containing 50 μM capsaicin . Electrophysiology was performed on dissected adult worms as previously described [43] . Worms were superfused in an extracellular solution containing 127 mM NaCl , 5 mM KCl , 26 mM NaHCO3 , 1 . 25 mM NaH2PO4 , 20 mM glucose , 1 mM CaCl2 , and 4 mM MgCl2 , bubbled with 5% CO2 , 95% O2 at 20°C . Whole cell recordings were carried out at –60 mV using an internal solution containing 105 mM CsCH3SO3 , 10 mM CsCl , 15 mM CsF , 4mM MgCl2 , 5mM EGTA , 0 . 25mM CaCl2 , 10mM HEPES , and 4 mM Na2ATP , adjusted to pH 7 . 2 using CsOH . Under these conditions , we only observed endogenous acetylcholine EPSCs . To record GABAergic postsynaptic currents , the holding potential was 0 mV , at which we only observe mIPSCs . All recording conditions were as described [44] . To record evoked EPSCs , a 0 . 4 ms , 30 μA square pulse was applied to a motor neuron cell body with a stimulating electrode placed near the ventral nerve cord ( one muscle distance from the recording pipette ) . Statistical significance was determined using one-way ANOVA with Tukey test for multiple comparisons and two-tailed Student’s t test for pairwise comparison . Quantitative imaging of coelomocyte fluorescence was performed as previously described [11] using a Zeiss Axioskop equipped with an Olympus PlanAPO 100x ( NA = 1 . 4 ) objective and a CoolSNAP HQ CCD camera ( Photometrics ) . Worms were immobilized with 30 mg/ml BDM ( Sigma ) . The anterior coelomocytes were imaged in L4/A lethargus ( determined by absence of pharyngeal pumping ) , and 1 day-old adult animals . Image stacks were captured and maximum intensity projections were obtained using Metamorph 7 . 1 software ( Universal Imaging ) . YFP fluorescence was normalized to the absolute mean fluorescence of 0 . 5 mm FluoSphere beads ( Molecular Probes ) . Statistical significance was determined using Kolmogorov-Smirnov test . Using Dermabond topical skin adhesive , individual worms were glued to 2% agarose pads in extracellular saline ( 145 mM NaCl , 5 mM KCl , 1 mM CaCl2 , 5 mM MgCl2 , 20 mM D-glucose , and 10 mM HEPES buffer [pH7 . 2] ) . To image copper and glycerol responses , single animals were placed in a perfusion chamber ( RC-26GLP , Warner Instruments ) under a constant flow rate ( 0 . 4 ml min-1 ) of buffer using a perfusion pencil ( AutoMate ) . Outflow was regulated using a peristaltic pump ( Econo Pump , Bio-Rad ) . 10mM CuCl2 ( copper ( II ) chloride dihydrate , Sigma ) or 500mM glycerol ( Fisher ) were delivered using the perfusion , pencil and switch between control and stimulus solutions was done using manually controlled valves . Solutions contained either 10mM CuCl2 in M13 buffer or 500mM glycerol in 40mM NaCl , 1 mM MgSO4 , 1 mM CaCl2 and 5 mM KPO4 . The stimulus was delivered for 10 seconds starting on the 10th second from the beginning of the movie . Optical recordings were performed on a Zeiss Axioskop 2 upright compound microscope equipped with a Dual View beam splitter and a Uniblitz Shutter . Images were recorded at 10 Hz using an iXon EM camera ( Andor Technology ) and captured using IQ1 . 9 software ( Andor Technology ) . For ratiometric imaging , ROIY tracked the neuron in the yellow channel , and in the cyan channel , ROIC moved at a fixed offset from ROIY . F was computed as FY/FC following a correction for bleed through . No correction for bleaching was required . Ratio changes were detected and parametrized using scripts for MATLAB ( The Mathworks ) . Briefly , the scripts average the F value for 5 preceding and including the marked start stimulus frame ( F0 ) and the 5 frames centered on the marked peak frame ( F1 ) . ΔF was equal to ( F1—F0 ) / F0 x 100 . Touch-evoked calcium responses in PLM neurons were analyzed as previously described [11] . Statistical significance was determined using one-way ANOVA with Tukey test for multiple comparisons . | Animals switch between periods of behavioral arousal and quiescence in response to environmental , developmental , and circadian cues . Little is known about the circuit mechanisms that produce these behavioral states . During larval molts , C . elegans exhibits a sleep-like state ( termed lethargus ) that is characterized by the absence of feeding and profound locomotion quiescence . We previously showed that mutants lacking the neuropeptide receptor NPR-1 exhibit increased arousal during larval molts , which is in part mediated by increased secretion of an arousal peptide ( PDF-1 ) . Here , we compare the circuits regulating arousal in larval molts and adults . We show that a broad network of sensory neurons arouses locomotion but that the impact of each neuron differs between lethargus and adults . We propose that this broad sensory network allows C . elegans to adapt its behavior across a broad range of developmental and physiological circumstances . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Sensory Neurons Arouse C. elegans Locomotion via Both Glutamate and Neuropeptide Release |
Human African Trypanosomiasis is a vector-borne disease of sub-Saharan Africa that causes significant morbidity and mortality . Current therapies have many drawbacks , and there is an urgent need for new , better medicines . Ideally such new treatments should be fast-acting cidal agents that cure the disease in as few doses as possible . Screening assays used for hit-discovery campaigns often do not distinguish cytocidal from cytostatic compounds and further detailed follow-up experiments are required . Such studies usually do not have the throughput required to test the large numbers of hits produced in a primary high-throughput screen . Here , we present a 384-well assay that is compatible with high-throughput screening and provides an initial indication of the cidal nature of a compound . The assay produces growth curves at ten compound concentrations by assessing trypanosome counts at 4 , 24 and 48 hours after compound addition . A reduction in trypanosome counts over time is used as a marker for cidal activity . The lowest concentration at which cell killing is seen is a quantitative measure for the cidal activity of the compound . We show that the assay can identify compounds that have trypanostatic activity rather than cidal activity , and importantly , that results from primary high-throughput assays can overestimate the potency of compounds significantly . This is due to biphasic growth inhibition , which remains hidden at low starting cell densities and is revealed in our static-cidal assay . The assay presented here provides an important tool to follow-up hits from high-throughput screening campaigns and avoid progression of compounds that have poor prospects due to lack of cidal activity or overestimated potency .
Human African Trypanosomiasis ( HAT ) or sleeping sickness is an endemic disease of sub-Saharan Africa . It is caused by two subspecies of Trypanosoma brucei , T . b . rhodesiense and T . b . gambiense , the latter of which is responsible for 95% of all cases of HAT [1] . Progress has been made in reducing the incidence of the disease and in 2009 , for the first time in 50 years , the number of new reported cases has dropped below 10 , 000 [2] . Currently , the early-stage drug treatments are pentamidine and suramin for T . b . gambiense and T . b . rhodesiense respectively , whereas late stage disease ( when the parasites have spread to the central nervous system ) is treated with nifurtimox eflornithine combination therapy ( NECT ) for T . b . gambiense and the arsenic based compound melarsoprol for T . b . rhodesiense . In spite of the apparent success of the current treatments they are associated with a series of problems , including complexity of administration , adverse side-effects and the emergence of resistance [3] . Thus , it is clear that it is essential to continue the development of new drugs . There are two main avenues to identify new chemical matter with anti-trypanosomal activity; target-based screening and phenotypic screening . Both methods have their own merits , and can complement each other [4] , [5] . Whichever method is used , it is desirable to identify compounds with fast trypanocidal ( killing of the parasites ) , rather than trypanostatic ( inhibition of parasite growth ) activity , as it is more likely that such compounds can be developed into fast acting drugs with improved treatment regimens compared to the current options . This is reflected in the current DNDi Target Product Profile for HAT which specifies that new candidate drugs should ideally be cidal [6] . Experiments determining the cidal versus static nature of compounds are routinely carried out in the antibacterial field [7] , and have also been described for malaria [8] and Trypanosoma cruzi [9] . High-throughput phenotypic screens for T . brucei are often carried out using an endpoint growth assay with either a redox indicator [10] , [11] , [12] , [13] , [14] or ATP-dependent luminescence [15] , [16] as a measure for trypanosome cell density . Typically the trypanosomes are grown in the presence of test compounds for several days with a starting cell-density well below the detection limit of the assay . This makes distinguishing compounds with a cidal effect from compounds with a static effect impossible . Follow-up experiments to demonstrate the cidal nature of a compound are relatively labour intensive and often fall outside the remit of hit discovery programmes as they are difficult to automate and not suitable for screening large amounts of compounds . In our own experience we have progressed apparently potent hits from a primary resazurin-based T . brucei screen all the way to in vivo efficacy studies where the series failed . Further studies showed that the minimum cidal concentration for these compounds was much higher than expected from our routine T . brucei assay . This led to the development of the high-throughput 384-well trypanosome growth assay presented here which provides an initial , quantitative indication of the cidal nature of anti-trypanosome compounds . We further show that the starting-density used in trypanosome growth assays can have a significant effect on the shape of potency curves , and that at low starting-density important information regarding the mode of action of a compound may be hidden .
Melarsoprol was a gift from Rhone-Poulenc ( France ) . Resazurin , pentamidine , nifurtimox , eflornithine ( D , L-α-difluoromethylornithine; DFMO ) and suramin were obtained from Sigma . White clear bottom plates for T . brucei growth assays were obtained from Greiner and echo plates were from LabCyte . Bloodstream-form T . b . brucei ( strain 427 , ‘single marker’ cloned line ) was grown at 37°C in presence of 5% CO2 in HMI9-T medium [10] . For routine culturing cells were diluted to 2×104 ml−1 for 48-h passages or diluted to 2×103 ml−1 for 72-h passages . Cells were maintained in vented T75 flasks . Cultures used for screening were split to 2×105 ml−1 the day before plating in 384-well plates . T . brucei cells were counted and dispensed into a 384-well plate at the following densities: 0 , 5×103 , 1×104 , 2×104 , 5×104 , 1×105 , 2×105 , 5×105 , 1×106 and 2×106 cells ml−1 . Each cell density was plated into 16 wells . Resazurin was then added at 0 . 05 mM final concentration and the plates were incubated for 4 h at 37°C . Fluorescence intensity was then measured using a Perkin Elmer Victor 3 plate-reader ( excitation 528 nm , emission 590 nm ) . The limit of detection was calculated as the number of cells giving a signal greater than the mean signal of the blank wells plus 3 times the standard deviation of the blank wells . For the preparation of potency curves: 30 µl of compound at 10 mM in DMSO was manually dispensed into 384-well compound-holding plates . A ten-point three-fold dilution curve in DMSO was then created on a Perkin Elmer Janus liquid handling robot . For the preparation of assay plates , serial dilution curves were transferred to LabCyte Echo certified plates , and 250 nl of each concentration was dispensed into white clear bottom 384-well assay plates using a LabCyte Echo acoustic dispenser . For static-cidal assays three replicate assay plates were created , one for each time point . Columns 11 , 12 , 23 and 24 of each plate contained DMSO only . The final concentration of DMSO in all assay wells was 0 . 5% ( v/v ) . Standard growth assays were carried out as described previously [17] . In short , bloodstream-form T . brucei cells in fresh medium were plated into columns 1–22 of 384-well assay plates using a Wellmate dispenser ( Thermo Fisher ) ( 5×103 ml−1 , 50 µl/well ) and incubated for 68 h at 37C , 5% CO2 . Next resazurin was added at 0 . 05 mM final concentration and the plates were incubated for a further 4 h . Fluorescence was then measured using a Perkin Elmer Victor 3 plate-reader ( excitation 528 nm , emission 590 nm ) . For the static-cidal assay 3 replicate assay plates were prepared and trypanosomes were seeded into each plate at 4×105 ml−1 ( 50 µl/well ) , unless otherwise indicated . Immediately thereafter resazurin was added to one of the plates , and all plates were incubated at 37°C , 5%CO2 . After 4 h the time = 0 plate was read as above . Twenty hours later the second plate was processed in the same way , and at 44 h the last plate was processed . Z-factors were calculated for each plate using the following formula: . A minimum z-factor of 0 . 6 was set as quality cut-off . Growth rates were determined for both assays by harvesting cells every 24 hours from 384-well plates and counting them in a CASY Counter ( Roche ) . This was carried out on three occasions and growth curves were plotted . Dose-response curve fitting was carried out either in IDBS ActivityBase or using the Excel Add-in XL-fit ( IDBS ) . For monophasic fits the following 4 parametric equation was used:where A = % inhibition at bottom , B = % inhibition at top , C = EC50 , D = slope , x = inhibitor concentration and y = % inhibition . For biphasic fits equation: was used , with A = % inhibition at mid-plateau , B = slope , C = log ( EC50A ) , D = log ( EC50B ) ) . Inhibition at the bottom of the curve is fixed to 0% and at the top to 100% . Potency values are given as pEC50 ( the negative logarithm of the EC50 in molar units ) . All data are the result of at least 3 independent measurements , reported as arithmetic mean of pEC50 values . Growth curves for the static-cidal assay were produced in Microsoft Excel by plotting the raw resorufin fluorescence measurement for each drug concentration against time . The average of three independent measurements was calculated and plotted together with the standard deviation . The Minimum Cidal Concentration ( MCC ) was defined as the lowest concentration of drug that results in a decrease of resorufin signal over time . For compound concentrations that allowed significant growth during the 0–24-h window ( resorufin RFU>10 at 24 h ) followed by a decrease in signal during the 24–48-h window , we set the additional requirement that the signal at 48 h had to be equal to , or less than the starting signal ( t = 0 h ) to account for the noise in the assay at the 48-h time point and to assure that cidal action was indeed occurring .
The routine T . brucei screening assay employed in our unit involves plating the parasites at a density of 250 cells per well ( in 50 µl ) into 384-well plates with pre-dispensed compounds . After 68 h of incubation the live-cell indicator resazurin is added and incubated for 4 h followed by measuring resorufin fluorescence in a plate-reader . As this starting-density is below the detection limit of the assay ( 2 , 500 cells/well , dotted line in Figure 1 ) , it is impossible to distinguish cidal compounds from static compounds , or even compounds that effect a moderate level of growth inhibition . To obtain a measurement of the cidal nature of compounds we altered our standard assay to allow the monitoring of cell growth from the moment of addition of the trypanosomes to the compounds . We increased the starting-density 80-fold ( 2×104 cells per well ) , well above the detection limit of the assay , which allowed us to use the resazurin method immediately after mixing parasites with compounds to obtain a time zero measurement of cell density . We then repeated the resazurin measurement at 24 h and 48 h to obtain a growth curve ( later time points could not be measured as the cells cease growing due to nutrient depletion or acidification of the media ) . After the final time point , growth curves were created for each concentration of compound . A panel of known inhibitors of T . brucei growth were tested in this assay ( Figure 2 ) . A decline in signal over time indicates a reduction in the number of trypanosomes and thus cidal activity . A quantitative measurement of the lethality of a compound is given by the MCC , the lowest concentration at which a decline in signal is seen during the 48-h drug exposure ( see methods ) . Cidal activity was observed for melarsoprol ( MCC 0 . 1 µM ) , suramin ( MCC 0 . 5 µM ) , nifurtimox ( MCC 33 µM ) and pentamidine ( MCC 0 . 03 µM ) . DFMO only had a growth reducing effect . A series of compounds ( unpublished ) from our HAT drug discovery programme were tested in this assay and the growth curves for a representative compound is also shown ( DDU1; MCC 50 µM ) . The structure for DDU1 is given in supplementary figure 1 . We next compared the potencies obtained in our standard growth assay with the MCCs from the static-cidal assay . As shown in Table 1 , the fold change between the two measures was markedly larger for compound DDU1 compared to the cidal control compounds , but similar to the cytostatic drug DFMO . To investigate this further we fitted potency curves using the data obtained in the static-cidal assay at the 48 h time point , and compared these to potency curves obtained in the standard assay . The data for the standard assay were obtained at 48 h and 72 h , to assess any effects of exposure time , and no significant differences were observed ( data not shown ) . Figure 3A shows the dose-response curves for compound DDU1 . The data obtained in the standard assay fitted best to a monophasic dose-response curve ( left panel ) while the data from the static-cidal assay fitted much better to a biphasic curve ( right panel ) . Melarsoprol , which showed a small difference between pEC50 ( std ) and pMCC exhibited monophasic curves under both conditions ( Figure 3B ) . The main difference between the two assays is the starting cell-density; we thus investigated whether the appearance of the biphasic dose-response curve was cell-density dependent . Figure 4 shows potency curves obtained for DDU1 with three different starting-densities ( 4×105 , 4×104 and 5×103 ml−1 ) . As expected , the data fitted a monophasic model at 5×103 ml−1 . Biphasic behaviour was observed for both 4×104 and 4×105 ml−1 starting-densities , but interestingly the mid-plateau was positioned much higher at the lower of the two densities ( ∼80% compared to ∼50% ) . These data confirm that the starting-density can have a major effect on the shape of dose-response curves . The explanation for this behaviour lies in the effect of the assay detection limit on the end-point measurements . In the standard assay , which starts at a cell-density well below the assay detection limit , the cell-density in a well exposed to a compound that inhibits growth by 50% , i . e . a twofold increase in the doubling time , will show 95% inhibition compared to untreated control ( Figure 5 ) . Potency curves will thus show near complete inhibition for a concentration of compound that only moderately inhibits cell growth . This problem is resolved by increasing the starting-density to above the detection limit as done in our static-cidal assay , and as a result any level of inhibition can now be detected . Reducing the starting density will result in increasing overestimation of compound activity , which explains why the mid-plateau seen in Figure 4 moves up with decreasing starting-density . Following from this , EC50A of the biphasic curves should match the EC50 obtained from the monophasic curves using the low starting-density . Figure 6 shows this correlation for the set of compounds with biphasic behaviour in the static-cidal assay , and indeed good correlation ( R2 = 0 . 9 ) is seen between pEC50A ( static-cidal ) and pEC50 ( std ) . Taken together these data show that the starting-density used in the assay has a major impact on the resulting potency curves , and , when too low , that important biologically relevant characteristics of compounds like the biphasic behaviour shown here may be hidden .
Here we describe a straightforward 384-well T . brucei growth assay that estimates the extent to which a compound is cidal . The quantitative MCC can be used to compare compounds and should be an essential component of the decision-making process for progression of a compound . Important advantages of this assay over existing procedures are the high-throughput that can be achieved and the straight-forward implementation of the assay as it is similar to the standard 384-well high-throughput assay [14] and does not require any additional resources . We routinely screen upwards of twenty 384-well plates in a single batch , which allows the static-cidal testing of nearly 200 compounds . We do not see this assay as a platform for primary screening; instead it is more suitable as a secondary assay to follow up hits from high-throughput screening campaigns run with the standard assay . As such screens can yield large numbers of hits , the throughput achieved in our static-cidal assay is essential to allow an assessment of the cidal nature of these hits . The MCC , together with other relevant criteria , can then be used to progress the most promising compounds into the hit-to-lead phase . On the down-side , the assay does not provide unequivocal proof of cidal activity and , to obtain this , more detailed wash-out experiments are required . Such studies can be carried out at a later stage in development for the most promising compounds . We investigated the cidal activity of commonly used HAT drugs using our assay . As expected , we saw cidal activity for melarsoprol , suramin , pentamidine and nifurtimox . The known trypanostatic effect of DFMO was confirmed in our assay as we only detected a growth-retarding effect for this compound [18] , [19] . A series of factors need to be taken into consideration when interpreting the results provided by the assay . The mode of action of the compound , concentration range tested , duration of exposure , compound stability , and cell density may all affect the growth-inhibition curves and the resulting MCCs . Compounds that act as polypharmacological agents may be static at lower concentrations when only one of the targets is inhibited , whereas at higher concentration they may be cidal due to the inhibition of other targets . Similarly , specific inhibition of a target may result in static behaviour , whereas cidal behaviour at higher concentrations may be the result of non-specific toxicity . The high cell density used in our static-cidal assay may affect the final intra-cellular concentration of the drug , in particular for compounds like pentamidine that are actively concentrated by the parasite [20] , [21] . So , while extensive studies are required to properly characterise the cidal nature of a compound , our assay provides a powerful tool , using a set of standard conditions , to rank large sets of compounds and to aid in choosing the right candidate compounds for progression in a drug development programme . We employed our assay to test a series of compounds , represented here by DDU1 , which while appearing potent in the standard assay , and in spite of good ADME properties , did not show any in vivo efficacy ( data not shown ) . The results revealed that there was a much larger fold-difference between the EC50 and MCC for this series ( >200 fold ) compared to the cidal control compounds ( ≤12 fold ) ( Table 1 ) . This striking disparity made us investigate the differences between the standard assay and our static-cidal assay and led to the observation that obtaining potency curves using starting-densities below the detection limit may hide important characteristics of compounds relevant to their mode of action . In our case , the biphasic nature of one of our compound series remained hidden using our routine assay . This resulted in a severe overestimation of compound potency and progression of these compounds into in vivo efficacy studies , wasting a significant amount of resource . The main explanation for this behaviour is that moderate levels of growth inhibition may result in cell-densities below or close to the detection limit at the end of the assay , suggesting complete inhibition of cell growth . The implication is that standard resazurin-based T . brucei assays with low starting density do not actually report trypanosome viability; instead they show growth inhibition to the extent that the detection limit of the assay is not reached . By increasing the starting density substantially above the detection limit moderate levels of inhibition can be detected and in our case reveal the biphasic nature of the compound series in question . This biphasic behaviour is likely indicative of the compounds exerting their effect through at least two modes of action . Differential susceptibility in the cell population could be another explanation; however the fact that we use a clonal cell line in this study , and the observation that the position of the biphasic mid-plateau is cell-density dependent rule this out . An important consideration is the effect of starting density on the rate of cell growth across the time course . As can be seen on Figure 5 , cell growth remains close to exponential for a 3-day time course when the low starting-density of 5×103 ml−1 is used . However , with the high starting-density ( 4×105 ml−1 ) cells start reaching stationary phase within 48 h . This should be avoided as it affects the accuracy of the percent inhibition calculations . This highlights the challenge when choosing drug-exposure time and starting density for screening fast growing organisms such as T . brucei , as these parameters can have a significant effect on apparent drug action ( fast versus slow acting , monophasic versus biphasic behaviour , static versus cidal action ) . There is no one set of parameters that can be universally applied and suitable conditions need to be chosen depending on the questions being addressed . The effect of starting cell-density described here is likely not unique to trypanosomes and may apply to growth assays used in , amongst others , the anti-bacterial and anti-cancer fields . | Trypanosoma brucei is a protozoan parasite causing African sleeping sickness . Current treatments for this disease have significant limitations , underlining the need for better and safer drugs . To identify new chemical starting points for drug development , large compound collections are screened against the parasite . Such screens typically do not distinguish between compounds that slow the growth of the parasite and compounds that actually kill the parasite ( cidal compounds ) . Here , we present the development of an assay to identify such compounds . The main advantage of our assay is that it marries a relatively high-throughput to increased understanding of mode of action . Many active compounds ( hits ) are usually identified in T . brucei primary screening campaigns , making it difficult to select which compounds should undergo further development . Our assay allows testing of all of the hits for cidal activity so that only the most promising compounds are progressed . We show that the starting cell density used in the T . brucei growth assay can have a significant effect on the shape of dose response curves , and that important information regarding the mode of action of a compound can remain hidden at low starting densities as used commonly in T . brucei screening assays . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"biotechnology",
"medicine",
"infectious",
"diseases",
"african",
"trypanosomiasis",
"parasitology",
"neglected",
"tropical",
"diseases",
"biology",
"microbiology",
"drug",
"discovery"
]
| 2012 | A Static-Cidal Assay for Trypanosoma brucei to Aid Hit Prioritisation for Progression into Drug Discovery Programmes |
Dosage compensation has been thought to be a ubiquitous property of sex chromosomes that are represented differently in males and females . The expression of most X-borne genes is equalized between XX females and XY males in therian mammals ( marsupials and “placentals” ) by inactivating one X chromosome in female somatic cells . However , compensation seems not to be strictly required to equalize the expression of most Z-borne genes between ZZ male and ZW female birds . Whether dosage compensation operates in the third mammal lineage , the egg-laying monotremes , is of considerable interest , since the platypus has a complex sex chromosome system in which five X and five Y chromosomes share considerable genetic homology with the chicken ZW sex chromosome pair , but not with therian XY chromosomes . The assignment of genes to four platypus X chromosomes allowed us to examine X dosage compensation in this unique species . Quantitative PCR showed a range of compensation , but SNP analysis of several X-borne genes showed that both alleles are transcribed in a heterozygous female . Transcription of 14 BACs representing 19 X-borne genes was examined by RNA-FISH in female and male fibroblasts . An autosomal control gene was expressed from both alleles in nearly all nuclei , and four pseudoautosomal BACs were usually expressed from both alleles in male as well as female nuclei , showing that their Y loci are active . However , nine X-specific BACs were usually transcribed from only one allele . This suggests that while some genes on the platypus X are not dosage compensated , other genes do show some form of compensation via stochastic transcriptional inhibition , perhaps representing an ancestral system that evolved to be more tightly controlled in placental mammals such as human and mouse .
Monotremes are unique mammals that exhibit a mix of reptilian and mammalian features , as they lay eggs , yet have fur and produce milk for their young . Represented only by the fabled platypus and four species of echidna , they are distantly related to humans and other eutherian ( ‘placental’ ) mammals , having diverged from therian mammals ( eutherians and marsupials ) 166 million years ago ( MYA ) [1] . Monotreme genomes also show a curious mixture of reptilian and mammalian characteristics . They have a smaller genome than therian mammals [2] , and their karyotype comprises a few large chromosomes , and many small ones , somewhat reminiscent of chicken macro and microchromosomes . Most curious of all is the sex chromosome system of monotremes . Although monotremes , like other mammals , subscribe to an XY system of male heterogamety , they have multiple X and Y chromosomes [3] which form a multivalent translocation chain during meiosis [4] . Platypus ( Ornithorhynchus anatinus ) have ten sex chromosomes; males have five X chromosomes ( X1X2X3X4X5 ) and five Y chromosomes ( Y1Y2Y3Y4Y5 ) , and females five pairs of X chromosomes [5] . During male meiosis , X and Y chromosomes pair within terminal pseudoautosomal regions [6] , forming a chain of alternating X and Y chromosomes ( numbered by their order in the chain X1–Y1–X2–Y2–X3–Y3–X4–Y4–X5–Y5 ) which segregate into five X-bearing ( female-determining ) and five Y-bearing ( male-determining ) sperm [7] . The sex chromosomes of therian mammals are remarkably conserved . The X chromosomes of all placental mammals have virtually identical gene contents , and the marsupial X chromosome shares two thirds of the human X , defining it as the ancient X conserved region [8] . The largest platypus X was also thought to share this ancient region [9] . However , comparisons of the gene contents of platypus , human and marsupial sex chromosomes reveal that the ancient region of the therian X is entirely homologous to platypus chromosome 6 [6] . Instead , platypus X chromosomes share considerable homology with the chicken Z chromosome , including DMRT1 , a dosage-sensitive gene that is a candidate for bird sex determination [6] , [10] . The monotreme sex chromosome complex is proposed to have evolved by repeated autosome translocation onto an original bird-like ZW pair [5] , [11] . The possession of a chain of nine sex chromosomes by the echidna , seven of which are shared with platypus [12] , means that the chain is at least 30 M years old . How a ZW system of female heterogamety was transformed into an XY system of male heterogamety has been vigorously debated [13] . Mammalian Y chromosomes are much smaller and more variable than their X chromosome partners , but share homology within pseudoautosomal regions , and also between coding genes on the X and Y . This supports the theory that heteromorphic sex chromosomes evolved from a pair of homologous autosomes in a mammal ancestor after one member of the pair acquired a sex determining locus , which lead to suppression of recombination and ultimately resulted in differentiation between members of the pair ( reviewed in [14] , [15] ) . A similar scenario is proposed for the evolution of the bird Z and W from an ancient autosomal pair [16] . Comparative gene mapping between the mammal X and bird Z [17] , [18] shows that they arose from different autosomal pairs . Although they are non-homologous , the XY of therians and ZW of birds do possess similar general properties . The bird Z , like the mammal X , is highly conserved between species [18] , whereas the W is degraded to different extents in different bird groups . Also , the bird Z and the mammal X are large chromosomes carrying many genes , and are well conserved between species , whereas the heterogametic chromosome ( W and Y ) is small , heterochromatic and varies greatly in size and gene content . The X and Z chromosomes both appear to have sex-biased gene content . For example , the human X chromosome is enriched with genes involved in brain function , sex and reproduction [19]–[21] , and in male ( but not female ) specific genes [22] , and the chicken Z is enriched with genes involved in male ( but not female ) reproduction [23] . Despite these similarities between the mammal XY and the bird ZW sex chromosome systems , the extent to which genes on the X and Z are dosage compensated is remarkably different . X chromosome inactivation overcomes differences in gene dosage between XX females and XY males in therian mammals . In somatic cells of female humans and mice , genes on one X become genetically inactive [24] and transcriptionally silenced [25] early in embryogenesis , a state that is somatically heritable . In marsupials , too , genes on one X chromosome are inactivated [26] . X inactivation mechanisms in eutherians and marsupials differ in a number of important aspects . In somatic cells of eutherians , inactivation is random between maternally and paternally derived X chromosomes , whereas in marsupials only the paternal X is silenced . X inactivation in eutherians is more stable and complete than in marsupials [26] , although it was recently discovered that between 5% [27] and 15% [28] of genes on the human X escape inactivation , mostly on the region added recently to the X in the eutherian lineage [28] . At the molecular level , eutherian X inactivation results from a complex process controlled by a master locus ( the X inactivation centre XIC ) , which includes the non-coding XIST gene [29] , [30] . An array of epigenetic mechanisms , including binding with variant histones [31] , histone modifications [32] , [33] and differential DNA methylation [34] , [35] , contribute to the transcriptional silencing of the X-borne genes . An accumulation of LINE1 elements may provide “booster stations” for the propagation of silencing signal along the chromosome [36] . The molecular mechanism of X inactivation seems to be much simpler in marsupials . The region homologous to the XIC in eutherians is disrupted in marsupials and monotremes and no evidence of XIST has been found in the regions that juxtapose flanking markers [37]–[39] . XIST may have evolved in eutherians from relics of an ancient protein-coding gene [40] . Molecular mechanisms shared between marsupial and eutherian inactivation so far have been limited to late replication [41] and histone underacetylation of the inactive X [42]; DNA methylation does not seem to be involved in marsupial X inactivation [43] . It was suggested that marsupial X inactivation might represent an ancestral form of paternally imprinted X inactivation [26] , [44] , and this hypothesis is supported by imprinted inactivation in mouse extra-embryonic tissues [45] , which , like marsupial X inactivation , is less stable and incomplete , and does not involve DNA methylation [46] . However , unlike marsupials , this imprinted X inactivation in mice requires Xist [47] , [48] . The XIC , along with an accumulation of LINE1 elements on the X , may control random inactivation in eutherians and its absence correlates to the absence of XIST and LINE1 accumulation on the marsupial X [49] . The dosage difference for Z-borne genes between ZZ male and ZW female birds is equally as extreme as for the mammal X . Yet birds do not appear to achieve dosage compensation by silencing one Z chromosome in males , since both alleles can be demonstrated to be active by RNA-FISH and SNP analysis [50] , [51] . Quantitative PCR showed that nine of ten Z-borne genes have a male-female ratio close to 1∶1 [52] , but in microarrays , 40 zebrafinch and 964 chicken Z-borne genes showed a range of male to female ratios from 2∶1 ( ∼10% of genes ) to 1∶1 ( ∼10% of genes ) , with a mode in the middle [53] . In chicken embryos , the mean male to female ratio is 1 . 4–1 . 6 for Z-linked genes , consistent with an absence of complete dosage compensation [54] . This incomplete dosage compensation suggests that differences in gene dosage may be critical for only a few genes on the bird Z compared to the mammal X . The molecular mechanisms behind bird dosage compensation are yet to be elucidated . Differences in male to female ratios between Z linked genes suggest that at least some are regulated at the transcriptional level . A region on the short arm of the Z chromosome containing over 200 copies of a 2 . 2 kb repetitive sequence called MHM ( male hypermethylated ) , is hypermethylated on the Z chromosomes in male embryos , but hypomethylated on the Z in females [55] . MHM is transcribed only in females and accumulates as non-coding RNA near the DMRT1 locus in the nucleus . A higher proportion of genes subject to dosage compensation are clustered in this MHM region [56] . This suggests that dosage compensation in birds is via upregulation of gene expression in females , controlled by MHM [57] . The platypus presents a fascinating system in which to study dosage compensation . The need for such a system would appear to be acute , since the five X chromosomes of the complex account for 15% of the haploid genome , and are mostly unpaired by the five Y chromosomes , which together account for only 6% , and are at least half heterochromatic . Thus 12% of the genome is subject to 1:2 dosage differences . The homology of the platypus sex chromosomes with the bird Z , and lack of homology with the mammal X , raises questions of whether dosage compensation is incomplete and bird-like , or related to the mammal X inactivation system–or is completely different from both . There are almost no studies of dosage compensation in monotremes , and none using any molecular techniques . Early studies of replication timing of platypus X1 found no asynchronous replication of the unpaired region of this chromosome [58] . This suggests that if the platypus does compensate for gene dosage , it is unlikely to do so by X inactivation . Determining whether the platypus X chromosomes are dosage compensated has previously been difficult in the absence of knowledge of the genes on platypus X chromosomes . The assignment of genes to four of the five X chromosomes as part of the platypus genome project now presents an opportunity to investigate dosage compensation in this species . We used three different approaches to determine activity of genes located on four of the five platypus X chromosomes , and present evidence of significant transcriptional silencing of platypus X-borne genes .
We determined male to female gene expression ratios for two autosomal genes and 19 genes on platypus X1 , X2 , X3 and X5 , 10 of which are X-specific and nine pseudoautosomal ( shared with the Y chromosomes adjacent in the meiotic translocation chain ) . Genes chosen were from BAC ( Bacterial Artificial Chromosome ) clones mapped to platypus X chromosomes as part of the genome project [6] , as this localization indicated directly whether genes were X-specific or pseudoautosomal . BAC-end sequences from mapped BACs were aligned to the genome to reveal the genomic sequence contained within each BAC . Genes within BACs were identified using the platypus genome Ensembl database ( http://www . ensembl . org/Ornithorhynchus_anatinus/index . html ) ( Oana5 . 0 ) . The presence of these genes within the BACs was confirmed by PCR and sequencing , and expression of these genes in fibroblasts was determined ( Table 1 ) . We used RNA isolated from independently derived primary fibroblast cell lines representing 16 different individuals ( eight males and eight females ) . Expression of these genes was normalized to the expression levels of the housekeeping gene ACTB , an autosomal gene located on platypus chromosome 2 . Male to female ratios were calculated for the normalized data for each gene . The ratio was near 1 for both autosomal control genes ( G6PD and HPRT1 ) on platypus chromosome 6 . We also measured expression levels for nine pseudoautosomal genes with copies on X and Y . The expression ratios of seven genes were high ( 0 . 86–1 . 49 ) , indicating that the Y-borne , as well as the X-borne , alleles are active . However , two pseudoautosomal genes ( CDX1 and GMDS ) had ratios of about 0 . 5 , suggesting that the Y locus is not active . For five of the ten X-specific genes , ratios were high ( 0 . 81–0 . 99 ) , as would be expected if genes were largely or fully compensated . However , for three X-specific genes , the ratio was near 0 . 5 , which would be expected if the genes were not compensated between XY males and XX females . Two genes had intermediate ratios ( ∼0 . 7 ) , suggesting partial dosage compensation ( Table 2 ) . Statistical tests of the null hypothesis that there is no difference in expression levels between males and females , were compromised by the high variability between individuals , which resulted in p-values supporting the null hypothesis ( p = 0 . 05 ) for all X-specific genes . This variation could not be attributed to particular cell lines consistently showing higher or lower expression for the different genes tested ( see Figure S1 ) . The trend towards a higher level of expression in females than in males for X-specific genes suggests that different genes may be incompletely compensated to different extents . We used a bioinformatics approach to identify SNPs in genes on four of the five platypus X chromosomes ( details in Materials and Methods ) . We searched the Ensembl database for exonic sequence from predicted genes on platypus chromosomes X1 , X2 , X3 and X5 and compared these to platypus whole genome traces . Within these alignments we searched for single nucleotide mismatches appearing more than once at the same site . Possible SNPs were found in the platypus genome sequence within 57 genes on platypus chromosomes X1 ( 29 ) , X2 ( 6 ) , X3 ( 6 ) and X5 ( 16 ) . We validated a subset of these SNPs by sequencing PCR products derived from genomic DNA isolated from the same female animal ( “Glennie” ) used for the genome sequencing project and tested expression of these genes in fibroblast RNA isolated from this same individual . Of ten genes tested , seven were found to be expressed in fibroblasts ( ss76901227–ss76901236 ) ( Table 3 ) . BAC clones for these seven potentially X-specific SNP-containing genes were isolated , by using sequence up to 100 kb either side of the gene to search the platypus trace archive for BAC-end sequences . We confirmed that BACs contained the gene ( s ) of interest by PCR and direct sequencing . BACs were mapped by DNA-FISH to male metaphase chromosomes to confirm their location on an X and determine whether they have Y homologues ( data not shown ) . Three genes with validated SNPs on X1 were found to be pseudoautosomal , and based on genome assembly co-ordinates , all other unvalidated X1 SNPs are predicted to likewise fall within the pseudoautosomal region . Similarly , the SNP on X2 was shown to have a homologue on Y2 by FISH . However , the three X5 genes containing SNPs are X-specific . Sequencing of X-specific SNPs revealed that all genes were biallelically expressed ( Figure 1 ) , as were the pseudoautosomal SNPs ( data not shown ) . Allele specific real-time PCR was used to determine if alleles were expressed to the same extent for the pseudoautosomal gene GMDS and the X specific genes . No significant difference from a 1∶1 ratio was observed , implying the absence of imprinting ( Table 4 and Figure S2 ) . Biallelic expression with equivalent expression from alternate alleles for the three X-specific genes eliminates the possibility that genes on platypus X5 are subjected to complete paternal inactivation ( as is observed in marsupials ) , and directed our approaches to examining the probability of transcription from the two loci by RNA-FISH . RNA-FISH detects the sites of primary transcription in interphase cells by hybridization with large intronic sequences that are spliced from cytoplasmic mRNA . Thus large genomic probes were required for the genes of interest . BAC clones mapped to platypus X chromosomes as part of the genome project and found to contain genes expressed in fibroblast , were used for RNA-FISH experiments ( Table 1 ) . These included the four clones discussed above ( one from X2 and three from X5 ) . We also included BAC OaBb_24M14 ( GenBank Accession No . AC152941 ) containing DMRT2 , which had been fully sequenced previously and whose expression had been confirmed in fibroblast cell lines [10] . A BAC containing the HPRT1 gene located on chromosome 6 , OaBb_405M2 ( GenBank Accession No . AC148426 ) , was used as an autosomal control . HPRT1 was detected in the platypus fibroblast EST library sequenced as part of the genome project ( GenBank Accession No . EG341684 ) . The 14 BACs together contained 19 genes; two pseudoautosomal BACs contained four and two genes respectively and one X-specific BAC contained two genes ( Table S1 ) . Transcription of the 14 BACs described above was initially examined by RNA-FISH in female and male fibroblasts ( Figure 2 ) . As a control , RNA-FISH was followed by DNA-FISH to ensure that RNA signals were located near one ( X-specific genes in males ) or both of the alleles ( X-specific genes in females , autosomal and pseudoautosomal genes ) . Only those cells with two DNA-FISH signals per nucleus ( or one signal for X-specific genes in males ) were included in analysis . Data from the male RNA-FISH experiments was used to determine the efficiency of detection for each gene which was then used to extrapolate the expected percent of nuclei with biallelic expression in females , which is expected if there is no X inactivation ( Table 5 - refer to Table S2 for complete RNA-FISH dataset ) . HPRT1 , an autosomal control gene located on chromosome 6 , was expressed from both alleles in 96–97% of nuclei ( Figure 3A ) . Genes within four pseudoautosomal BACs on X1 , X2 ( including GMDS ) and X3 were also expressed from both alleles in most female nuclei ( 77–84% ) , as well as in most male nuclei ( 62–92% ) , showing that the Y , as well as the X , alleles are active ( Figure 3B ) . Two pseudoautosomal BACs used for RNA-FISH contain more than one gene , so it remains possible that not all genes within these BACs have an active Y copy . We obtained quite different results from the BAC containing CRIM1 , a X1-Y1 pseudoautosomal gene which was expressed from only one allele in most male ( 81% ) and female cells ( 71% ) ( Figure 3C ) . Except for this locus , we conclude that for the pseudoautosomal loci we tested , both X alleles are active in females , and both X and Y alleles are active in males . We then tested transcription from nine X-specific BACs on platypus X1 , X3 and X5 . Transcription from both alleles was observed on average in only 45% of nuclei ( Figure 3D ) . Different genes showed a range of transcription of both alleles , from 20% ( SEMA6A ) to 53% ( Ox_plat_124086 ) . These X-specific genes were therefore expressed very differently from the autosomal and pseudoautosomal genes , and significantly different to that expected for biallelic expression , indicating some level of transcriptional inactivation for these genes . Two colour RNA FISH was performed with genes FBXO10 and SHB , located within 500 kb of each other . Co-location of the two RNA signals showed the same X in all of the 51% of cells expressing from only one allele . ( Figure 4 ) . A few cells ( 12% ) displayed biallelic expression from SHB with monoallelic expression of FBXO10 , and in 37% of nuclei , both genes were expressed from both alleles . As a control , this experiment was performed on male nuclei showing that RNA-FISH signals co-located in all nuclei in which genes were expressed . This experiment was carried out only for two genes lying close together , as results from genes situated further apart ( and hence with a gap between signals expressed from the same chromosome ) would make results from cells expressing only one of each gene , difficult to interpret . RNA-FISH results were validated for a subset of genes ( HPRT1 , CRIM1 , GMDS , SEMA6A and DMRT2 ) on four other independently derived primary fibroblast cell lines from different individuals ( one male and three females ) . Results for each cell line are shown in Table S3 . As observed ( Figure 2 ) , the autosomal gene HPRT was expressed from both alleles in most nuclei ( 88% male and 83–90% female ) , as was the pseudoautosomal gene GMDS ( 86% , 85–90% ) . The pseudoautosomal gene CRIM1 , as before , was expressed from both X chromosomes in only 24–56% of female nuclei and X and Y in only 24% of male nuclei . As observed ( Figure 2 ) , both X-specific genes ( DMRT2 and SEMA6A ) were expressed from the single X in 99% of male nuclei , and both X chromosomes in half of female nuclei ( 45–60% and 38–43% respectively ) . Although there was some variation between individuals , overall results were similar between all six cell lines tested in this study . Statistical analysis revealed that only the two X-specific genes had a significant difference between the males and females for the number of nuclei expressing only one allele ( p = 0 . 0006 and 0 . 0008 respectively ) .
Together , our findings have parallels in observations of some genes on the marsupial X and the mouse X in extra-embyronic tissues , whose paternal alleles are partially inactive , or “escaper” genes on the recently added region of the human and mouse X , which are partially expressed from the inactive X . The observations of partial inactivation in all three major mammalian lineages suggests that partial inactivation observed here in platypus represents a basic form of mammalian X inactivation , which has come under tighter control during therian evolution , ultimately resulting in the highly stable and complex form of inactivation typical of most eutherian X-borne genes . Partial inactivation has been documented for two marsupial genes ( out of a total of five ) in some tissues . PGK1 isozyme variants showed strong expression from the maternal allele and weaker expression from the paternal allele in cells from heterozygous female kangaroos , even in single clones [59] , and G6PD from hybrid marsupials showed a heteropolymer band , diagnostic of expression from both alleles in a single cell [60] . Differences between species , tissues and even between genes make it difficult to generalize about the nature of marsupial X inactivation , and these experiments could not distinguish whether partial expression from the paternal X is due to low expression from paternal X chromosomes in every cell , or to a mixture of two X-active and one X-active cells . RNA-FISH was used to show that the tammar wallaby X-borne gene SLC16A2 was expressed from only one allele in most fibroblast cells [61] . The partial silencing displayed for platypus X-specific genes also has some parallels to genes on the human X that escape inactivation . X inactivation in humans was initially thought to involve all genes on the X chromosome , but in recent years it was found that 5% to as many as 15% of human genes escape inactivation in lymphoblastoid [27] and fibroblast cell lines [28] respectively . Remarkably , transcription of some of these genes in fibroblasts varies between individuals , as seems to be the case for platypus . Partial expression of genes on the inactive X has also been observed in other eutherians , including the mouse , cow and mole [62] , [63] . Typically , these escaper genes are fully expressed from the active X and partially expressed from the inactive X [28] , [64] . We propose that partial inactivation was the mechanism for compensating differences in gene dosage in an ancestral mammal . To date , it has been difficult to differentiate between the alternative hypotheses that partial inactivation is due to a lowered rate of transcription in all cells , or from a lowered probability of expression per cell in the population . Ohlsson et al [65] argued that genes transcribed at a low level show a low probability of transcription in the cell population , rather than a uniformly low transcription level . They propose that genomic imprinting and X chromosome inactivation evolved by regulating , not the activity of each locus , but the probability that it is expressed , and making this parent specific [65] . This radical hypothesis is supported by our RNA-FISH data , which show that platypus genes differ in the frequency of nuclei in which one or both alleles are transcribed , giving an overall partial dosage compensation that differs from gene to gene . The data from the bird Z is equivocal; the variability between genes is thought to reflect differences in the rate of transcription , but could equally well reflect differences in the probability that a locus is transcribed . RNA-FISH of five chicken genes shows that most are transcribed from both alleles in most cells [50]; however , the low efficiency of signal detection ( about a quarter of nuclei had no signals ) , and the different tissues used makes this hard to interpret . Efficient RNA-FISH on the chicken Z genes for which we have data in platypus would test the hypothesis that partial inactivation of the Z in male birds operates by altering the probability of transcription , rather than uniformly downregulating transcription . Our finding that two genes located 500 kb apart are expressed from the same chromosome implies that the stochastic expression of X-specific genes is coordinated in cis . Furthermore , a recent study has shown that this type of probabilistic expression is widespread on human autosomes , with their data suggesting that as many as 1000 human genes are subject to stochastic monoallelic expression [66] . Around 80% of these genes also showed some level of biallelic expression . Unlike the hypothesis put forward by Ohlsson et al [64] , this type of expression is not limited to those with low levels of expressions . Is partial expression in therian mammals explained by stochastic expression ? Data on partial expression of genes on the paternal X in marsupials are equivocal; the partial expression of the maternal PGK1 allele in clones , and the fainter paternal isozyme heteropolymer band for G6PD are explained equally well by both hypotheses . The few data that would distinguish these hypotheses for escapers on the inactive human X do not conclusively eliminate either hypothesis . Assays of the partially expressed human X-borne gene CHM ( REP1 ) in single cells showed that CHM was expressed from the inactive X in most ( 70% ) but not all cells from one cell line , and in only seven out of ten hybrid cell lines carrying an inactive X [67] . More recently , a study on dosage compensation in human lymphoblastoid cell lines found that genes escaping X inactivation were not subject to the higher levels of variation found for fibroblast cell lines , suggesting that the expression of the escaper genes is not stochastic but subject to tight regulation [27] . RNA-FISH performed on both fibroblasts and lymphoblastoid cells for these escaper genes would conclusively rule out stochastic expression . It is important to note the difference in the number of genes in human which escape inactivation between fibroblast cell lines , where 15% of genes are said to escape inactivation [28] and lymphoblastoid cell lines where only 5% of genes escape [27] . Similarly in marsupials , differences have been found in the inactivation status of genes between tissues [26] . Our study has only used fibroblast cell lines due to the difficultly in obtaining tissue samples in large enough sample sizes , as the platypus is listed as a “vulnerable” species . A comparison of results for other tissues may show different results . Several human X-borne genes that escape from inactivation have a widely expressed Y homologue , and some others have homology to a Y-borne pseudogene that represents a recently inactivated partner on the Y . The Y homologue of an X/Y pair often has a lower level of expression than its partner on the X ( reviewed in [68] ) , similar to the lower level of expression exhibited by alleles on the inactive X in females . However , the presence of a Y homologue does not necessarily negate the need for dosage compensation , as some Y alleles have evidently taken on functions different from those of their X homologue . Nearly all escaper genes are part of the region added to the eutherian X chromosome and only recently recruited to the inactivation system , suggesting that their partial escape from X inactivation correlates with progressive assimilation of genes into the X inactivation systems once the Y paralogue has degenerated . In eutherian mammals , small terminal regions of the X and Y are homologous , and pair and recombine at male meiosis . These pseudoautosomal regions ( PARs ) are relics of the X added region that have not yet degraded [15] . Genes within the PAR have no need of dosage compensation . There are two PARs on the human X . PAR1 on the short arm represents a relic of ancient XY homology , and contains genes that are expressed from the Y , and not inactivated on the X [69] . The smaller PAR2 was added very recently to the long arm of the Y from the long arm of the X , but two genes in the region ( SYBL1 and SPRY3 ) are subject to inactivation , not only on the inactive X , but also on the Y [70] . We observed that seven of the nine platypus genes from the pseudoautosomal regions displayed as much or more expression from males than females , as assessed by quantitative RT-PCR , suggesting that they are expressed from Y as well as the X alleles . RNA-FISH of these genes showed that both alleles were expressed in most cells in females ( two X alleles ) and males ( X and Y alleles ) . Two of these BACs contained multiple genes , so detection of predominantly two signals per cell does not necessarily mean that all genes are active on both chromosomes; however , expression analysis of transcripts from each of these BACs confirms that most of these genes ( 3/4 in BAC 286H10 and 2/2 BAC 271I19 ) have active Y homologues . Two pseudoautosomal genes CDX1 and GMDS had male∶female expression ratios near 0 . 5 but an almost equal probability of expression , suggesting that either both alleles are downregulated in males , or alternatively , the Y allele sequence has sufficiently diverged from that of the X homologue , leaving it unable to be amplified by our primers . A fifth platypus pseudoautosomal gene showed a completely different expression pattern . CRIM1 ( cysteine rich transmembrane BMP regulator 1 ) , located on platypus X1-Y1 , had equivalent expression in males and females , but was usually expressed from only one allele in both males ( 81% of nuclei ) and females ( 69% ) . There are two possible explanations . Firstly , the Y homologue may have evolved a new male-specific function like many genes on the human Y [15] , and be testis specific , so silencing of one X in females evolved to equalize expression of the X homologue . Alternatively , inactivation of both X and Y could be equivalent to the silencing of PAR2 genes on the long arm of the human X . SYBL1 and SPRY3 undergo silencing on both the X and Y , the product of their evolutionary history as a block transposed from the X ( where it was subject to inactivation ) to the Y , where it was dosage compensated to match the X [70] . Thus for most pseudoautosomal genes there is no need for dosage compensation on the X because the Y allele is active , and no dosage compensation is observed . The chromosome-wide X inactivation in mouse and human has given rise to the expectation that dosage compensation for genes on sex chromosomes is critical for life . However , this does not seem to be the case in birds . Dosage compensation for the 964 genes on the bird Z chromosome extends over a range from complete compensation ( ∼10% genes ) to no compensation ( ∼10% genes ) with most falling between these extremes [53] . This suggests either that the necessity for strict dosage compensation has been over-emphasized , or that genes on the bird Z chromosome are much more tolerant of dosage differences than genes on the therian X [71] . By no means are all genes dosage sensitive [71] . For instance , many protein products , such as enzymes , are controlled at different levels in the cell , so transcriptional control is not essential . For some genes , a dosage difference may even be essential for function; for instance , a 2∶1 dosage of DMRT1 has been suggested to define male versus female development in birds [72] . One gene that does not display equal expression between males and females and may even be hypertranscribed in females of both platypus and zebrafinch is SEMA6A , a gene on platypus X5 and the avian Z . From our data , platypus SEMA6A appears not be subject to dosage compensation by real-time RT-PCR , yet RNA-FISH results show that it predominantly has only one allele active per cell . In zebrafinch liver , SEMA6A is expressed more than two-fold more in females with just one copy than males with two copies [53] . Although these results were obtained from different cell types in the different species , it is intriguing that in both cases there is some evidence of hypertranscription in females . It is therefore likely that only a minority of genes on the mammalian X really need to be dosage compensated . The difference in the level of control of sex chromosome activity may therefore be a side-effect of the mechanism used for dosage compensation . Eutherian mammals subscribe to a whole-X mechanism in which inactivation spreads along the X . The bird Z , however , seems to have a piecemeal dosage compensation system in which different genes appear to show different levels of compensation , and compensated genes are clustered [56] . The alternative is that the genes on the bird Z and therian X evolved under different selective pressures . We know that the gene content of these chromosomes is different , having originated from two different pairs of autosomes , and we also know that the gene content of sex chromosomes is biased toward sex-specific expression . The human X is enriched for genes involved in brain function , and sex and ( particularly male ) reproduction [19]–[22] . The chicken Z chromosome gene content is male-biased yet noticeably deficient in female-biased genes [23] . Commenting on the finding that dosage compensation in birds is much less tightly controlled than in therian mammals , Graves and Disteche [71] suggested that expression differences in Z-borne genes between males and females may have been selected for to control sex-specific characters . Since platypus sex chromosomes show considerable homology to the bird Z , the functions of platypus X-borne genes are likely to be equivalent to those on the chicken Z . Perhaps , then , partial and variable silencing in the platypus dosage compensates some essential genes , leaves some genes uncompensated where dosage differences are essential for sex-specific function , and partially compensates most genes in proportion to their dosage-sensitivity , as is evidently the case for birds . We found that genes on the multiple platypus X chromosomes show partial and variable dosage compensation . This is very similar to the partial and variable dosage relationships of genes on the chicken Z chromosome , with which the platypus X chromosomes share considerable homology . However , unlike birds , platypus dosage compensation involves transcription from only one of the two alleles in a proportion of cells and is coordinated at least on a regional level . Transcriptional inhibition is a property shared by X chromosome inactivation in therian mammals . Thus , platypus dosage compensation has features shared with dosage compensation of the bird Z and the mammal X .
BAC-end sequences from CHORI-236 BAC clones ( http://bacpac . chori . org ) , mapped to platypus X chromosomes as part of the genome project , were aligned against the genome sequence . Genes within the genomic region contained between the BAC-end sequences were identified by using the Ensembl database ( http://www . ensembl . org/Ornithorhynchus_anatinus/index . html ) . An additional four BACs were chosen because they span genes with SNPs that were potentially X-specific . These BACs were identified by searching the platypus sequence trace archives containing BAC-end sequence data ( http://www . ncbi . nlm . nih . gov/Traces ) with genomic sequence from 100 kb up and downstream of the gene of interest . PCR was performed on the BACs to confirm that the genes predicted to be contained within the BAC were present . The PCR cycling conditions for all primers were as follows: an initial denaturing step of 94°C for 2 min , 30 cycles of 94°C for 30 sec , annealing for 30 sec at the appropriate temperature ( Table S4 ) , 72°C for 1 min and a final extension at 72°C for 10 min . To determine whether genes within BACs were expressed in fibroblasts , total RNA was extracted from female and male fibroblast cell lines using Gene Elute Mammalian Total RNA Miniprep extraction kit ( Sigma ) . RNA was treated with DNA-free ( Ambion ) to remove any contaminating DNA and Superscript III ( Invitrogen ) was used to generate cDNA using random hexamers as primers for first strand synthesis . To ensure there was no genomic DNA contamination in the cDNA sample , a RT-negative control was made by excluding the Superscript III enzyme from the first strand synthesis reaction and was used as a negative control in all RT-PCR experiments . Where possible , primers were designed to span introns . Primers , annealing temperatures and product sizes are listed in Table S4 . PCR was carried out using the same cycling conditions described above . Each set of primers was tested on female and male RT-positive and RT-negative samples as well as genomic DNA . PCR products were gel purified using a QIAquick Gel Extraction kit ( Qiagen ) and directly sequenced by AGRF ( Brisbane ) . For the four BACs not previously mapped , 1 µg of DNA from these BACs was labeled by nick translation with digoxigenin –11-dUTP ( Roche Diagnostics ) , Spectrum-Orange or Spectrum-Green ( Vysis ) . Unincorporated nucleotides were removed from Spectrum-Orange and Spectrum-Green labeled probes using ProbeQuant G50 micro columns ( GE Healthcare ) . Probes were precipitated with 1 µg platypus C0t1 DNA and hybridized to male and/or female platypus metaphase chromosomes and fluorescent signals for digoxigenin labeled probes were detected using the protocol described by Alsop et al [73] . A Zeiss Axioplan2 epifluorescence microscope was used to visualize fluorescent signals . Images for DAPI-stained metaphase chromosomes and fluorescent signals were captured on a SPOT RT Monochrome CCD ( charge-coupled device ) camera ( Diagnostic Instruments Inc . , Sterling Heights ) and merged using IP Lab imaging software ( Scanalytics Inc . , Fairfax , VA , USA ) . Total RNA was extracted from eight different male and eight different female fibroblast ( toe web ) cell lines ( at passage 6 to 8 ) to represent a total of 16 individuals . First-strand cDNA was synthesized by oligo ( dT ) priming using Superscript III ( Invitrogen ) . Primers for each gene were designed using the Plexor program ( Promega ) ( Table S4 ) . PCR reactions were carried out using Quantitect SYBR Green PCR kit ( Qiagen ) according to the manufacturer's instructions . Amplifications were performed and detected in a Rotorgene 3000 cycler ( Corbett Research ) . To determine the detection range , linearity and real-time PCR amplification efficiency for each primer pair , standard curves were calculated over a 10-fold serial dilution of fibroblast cDNA . A series of two-fold serial dilutions were also carried out to confirm the ability of the PCR conditions to detect this level of difference in expression . All dilutions and samples were run in triplicate . Cycling conditions consisted of an initial hold cycle of 95°C for 15 min , 40 cycles of 94°C for 15 sec , annealing at the appropriate temperature listed in Table S4 for 15 sec and extension at 72°C for 20 sec for data acquisition . Melting curves were constructed from 45°C–95°C to confirm the purity of the PCR products and direct sequencing of products was performed to confirm their identity . Relative expression of each gene was determined by normalization to ACTB expression using the formula where the ratio of ACTB to target = ( 1+ERef ) CtRef/ ( 1+ETarget ) CtTarget [74] . Statistical significance was assessed , for the null hypothesis that there was no difference between male and female expression levels , using an unrelated samples 2-tailed t test with unequal variance . Exonic sequence from predicted genes on platypus chromosomes X1 , X2 , X3 and X5 were extracted from the Ensembl 46 database , using the Biomart tool ( http://www . ensembl . org/biomart/martview ) . These sequences were compared to the platypus whole genome shotgun sequence traces ( “Ornithorhynchus anatinus WGS” ) deposited on the trace archive at NCBI ( http://www . ncbi . nlm . nih . gov/Traces ) , using MegaBLAST [75] . Potential single-nucleotide polymorphisms ( SNPs ) were discovered by manually searching within the BLAST output for single nucleotide mismatches occurring in approximately 50% of target traces . The chromatogram files containing a potential SNP were extracted from the trace archive and assembled using Sequencher™ 4 . 7 ( Gene Codes Corporation , Michigan ) . This assembled sequence ( including surrounding intronic sequence ) was tested for uniqueness within the platypus genome using BLAT [76] on the UCSC test browser ( http://genome-test . cse . ucsc . edu ) . To validate identified SNPs and test expression in fibroblasts , DNA was extracted from the “Glennie” fibroblast cell line using the Dneasy Blood and Tissue kit ( Qiagen ) and RNA was extracted as described above . First strand synthesis was performed on RNA using the Supercript III First-Strand Synthesis System for RT-PCR kit ( Invitrogen ) according to manufacturer's instructions . PCR and RT-PCR was carried out using the primers listed in Table S4 . To quantify the expression level of SNPs for three X-specific SNPs and one pseudoautosomal gene , allele-specific real-time PCR was carried out . Allele specific primers were designed with the 3′end base of either the forward or reverse primer corresponding to the specific allele ( refer to Table S5 for primer sequences and corresponding annealing temperatures ) . The different alleles were amplified in separate tubes . Real-time PCR was performed using Quantitect SYBR Green PCR kit ( Qiagen ) with amplifications performed and detected in a Rotorgene 3000 cycler ( Corbett Research ) . Cycling conditions are the same for those described in the quantitative PCR section with all samples run in triplicate . Genomic DNA for “Glennie” was included as a control since the allele frequency ratio should be 1∶1 , permitting allele-specific amplification bias to be detected and corrected . Known homozygous cDNA samples and pooled homozygous samples with varying ratios of each allele ( 0 . 2 , 0 . 4 , 0 . 6 , 0 . 8 ) were included to ensure the technique was sensitive enough to detect small differences . Allele relative expression levels were calculated using the formula: frequency of allele A = 1/ ( 2EΔCt+1 ) [77] , where ΔCt = ( AcDNA−BcDNA ) − ( AgDNA−BgDNA ) and converted to a ratio of allele A to allele B . PCR products were sequenced to confirm the identity of products . Male and female fibroblast cells ( from toe web ) were cultured on gelatin-coated coverslips in AminoMax C100 medium ( Invitrogen ) at 30°C in an atmosphere of 5% CO2 . Cells on coverslips were washed with PBS , permeabilized for 7 minutes on ice using CSK buffer plus Triton X ( 100 mM NaCl , 300mM sucrose , 3 mM MgCl2 , 10 mM PIPES pH 6 . 8 , 2 mm Vanadyl Ribonucleoside Complex ( VRC ) , 0 . 5% Triton X ) and fixed in 3% paraformaldehyde for 10 minutes . Coverslips were dehydrated via a series of ethanol washes ( 70% , 80% , 95% , 100% ) , air-dried and denatured . Probes were labeled as described in the DNA-FISH on metaphase chromosomes section . Hybridization buffer ( 4×SSC , 40% dextran sulphate , 2 mg/ml BSA , 10 mM VRC ) was added to each probe . Probes were denatured at 75°C for 7 min and allowed to preanneal for 20 min . 10 µl of probe was added to each coverslip and hybridized overnight in a humid chamber at 37°C . Coverslips were washed in 0 . 4×SSC with 0 . 3% Tween 20 at 60°C for 2 minutes followed by a wash in 2×SSC with 0 . 1% Tween 20 for 1 min at room temperature . Coverslips were fixed in 3% paraformaldehyde for 10 minutes , treated with 0 . 1 mg/ml RNase for 1 hour at 37°C and subjected to DNA-FISH following the same hybridization protocol described for DNA-FISH on metaphase chromosomes . Nuclei were viewed under a fluorescence microscope in several different focal planes , with 100 nuclei examined for each probe for both males and females . Efficiency ( p ) of RNA-FISH hybridisation was determined from the results obtained in male fibroblasts and extrapolated to determine the expected frequency of nuclei with two signals , one signal and no signal per cell using the formula p2+2pq+q2 = 1 , where p2 is the number of nuclei with two signals , 2pq ( q = 1−p ) represents nuclei with one signal and q2 is the number with no signal . P-values were determined by a χ2 test with two degrees of freedom . Inconsistencies between RNA-FISH results in previous experiments examining transcription have been attributed to the inability to detect weak signals , which could be overcome by , not only using a combination of RNA and DNA-FISH , but also by amplifying the RNA-FISH signal [78] . In order to ensure that the differences between autosomal , pseudoautosomal and X-specific genes were not due to the inability of the technique to detect both transcripts , an experiment where BACs containing SEMA6A and CRIM1 were labeled with either Spectrum Green or Spectrum Orange ( Vysis ) or with biotin-16-dUTP ( Roche Diagnostics ) was performed . Biotin-labeled probes were detected with avidin-FITC ( Vector Laboratories Inc . ) , with FITC signals amplified by additional layers of biotinylated anti-avidin ( Vector ) and avidin-FITC . No differences between direct labeling and biotin labeling followed by amplification were detected . | Dosage compensation equalizes the expression of genes found on sex chromosomes so that they are equally expressed in females and males . In placental and marsupial mammals , this is accomplished by silencing one of the two X chromosomes in female cells . In birds , dosage compensation seems not to be strictly required to balance the expression of most genes on the Z chromosome between ZZ males and ZW females . Whether dosage compensation exists in the third group of mammals , the egg-laying monotremes , is of considerable interest , particularly since the platypus has five different X and five different Y chromosomes . As part of the platypus genome project , genes have now been assigned to four of the five X chromosomes . We have shown that there is some evidence for dosage compensation , but it is variable between genes . Most interesting are our results showing that there is a difference in the probability of expression for X-specific genes , with about 50% of female cells having two active copies of an X gene while the remainder have only one . This means that , although the platypus has the variable compensation characteristic of birds , it also has some level of inactivation , which is characteristic of dosage compensation in other mammals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/genomics",
"genetics",
"and",
"genomics/epigenetics"
]
| 2008 | The Status of Dosage Compensation in the Multiple X Chromosomes of the Platypus |
Decoding models , such as those underlying multivariate classification algorithms , have been increasingly used to infer cognitive or clinical brain states from measures of brain activity obtained by functional magnetic resonance imaging ( fMRI ) . The practicality of current classifiers , however , is restricted by two major challenges . First , due to the high data dimensionality and low sample size , algorithms struggle to separate informative from uninformative features , resulting in poor generalization performance . Second , popular discriminative methods such as support vector machines ( SVMs ) rarely afford mechanistic interpretability . In this paper , we address these issues by proposing a novel generative-embedding approach that incorporates neurobiologically interpretable generative models into discriminative classifiers . Our approach extends previous work on trial-by-trial classification for electrophysiological recordings to subject-by-subject classification for fMRI and offers two key advantages over conventional methods: it may provide more accurate predictions by exploiting discriminative information encoded in ‘hidden’ physiological quantities such as synaptic connection strengths; and it affords mechanistic interpretability of clinical classifications . Here , we introduce generative embedding for fMRI using a combination of dynamic causal models ( DCMs ) and SVMs . We propose a general procedure of DCM-based generative embedding for subject-wise classification , provide a concrete implementation , and suggest good-practice guidelines for unbiased application of generative embedding in the context of fMRI . We illustrate the utility of our approach by a clinical example in which we classify moderately aphasic patients and healthy controls using a DCM of thalamo-temporal regions during speech processing . Generative embedding achieves a near-perfect balanced classification accuracy of 98% and significantly outperforms conventional activation-based and correlation-based methods . This example demonstrates how disease states can be detected with very high accuracy and , at the same time , be interpreted mechanistically in terms of abnormalities in connectivity . We envisage that future applications of generative embedding may provide crucial advances in dissecting spectrum disorders into physiologically more well-defined subgroups .
Despite their increasing popularity , two challenges critically limit the practical applicability of current classification methods for functional neuroimaging data . First , classifying subjects directly in voxel space is often a prohibitively difficult task . This is because functional neuroimaging datasets ( i ) typically exhibit a low signal-to-noise ratio , ( ii ) are obtained in an extremely high-dimensional measurement space ( a conventional fMRI scan contains more than 100 , 000 voxels ) , and ( iii ) are characterized by a striking mismatch between the large number of voxels and the small number of available subjects . As a result , even the most carefully designed algorithms have great difficulties in reliably finding jointly informative voxels while ignoring uninformative sources of noise . Popular strategies include: preselecting voxels based on an anatomical mask [18] , or a separate functional localizer [20] , [21]; spatial subsampling [22]; finding informative voxels using univariate models [3] , [11] , [12] or locally multivariate searchlight methods [23] , [24]; and unsupervised dimensionality reduction [4] , [25] . Other recently proposed strategies attempt to account for the inherent spatial structure of the feature space [23] , [26] , [27] or use voxel-wise models to infer a particular stimulus identity [28]–[30] . Finally , those submissions that performed best in the Pittsburgh Brain Activity Interpretation Competition ( PBAIC 2007 ) highlighted the utility of kernel ridge regression [31] and relevance vector regression [31] , [32] . The common assumption underlying all of these approaches is that interesting variations of the data with regard to the class variable are confined to a manifold that populates a latent space of much lower dimensionality than the measurement space . The second challenge for classification methods concerns the interpretation of their results . Most classification studies to date draw conclusions from overall prediction accuracies [33] , [11] , the spatial deployment of informative voxels [19] , [34] , [18] , [35]–[39] , the temporal evolution of discriminative information [40] , [37] , [41] , [42] , [26] , or patterns of undirected regional correlations [43] . These approaches may support discriminative decisions , but they are blind to the neuronal mechanisms ( such as effective connectivity or synaptic plasticity ) that underlie discriminability of brain or disease states . In other words: while some conventional classification studies have achieved impressive diagnostic accuracy [14] , their results have not improved our mechanistic understanding of disease processes . Generative embedding for model-based classification may provide a solution to the challenges outlined above . It is based on the idea that both the performance and interpretability of conventional approaches could be improved by taking into account available prior knowledge about the process generating the observed data ( see [44] for an overview ) . ( The term generative embedding is sometimes used to denote a particular model-induced feature space , or so-called generative score space , in which case the associated line of research is said to be concerned with generative embeddings . Here , we will use the term in singular form to denote the process of using a generative model to project the data into a generative score space , rather than using the term to denote the space itself . ) Generative embedding rests on two components: a generative model for principled selection of mechanistically interpretable features and a discriminative method for classification ( see Figure 1 ) . Generative models have proven powerful in explaining how observed data are caused by the underlying ( neuronal ) system . Unlike their discriminative counterparts , generative models capture the joint probability of the observed data and the class labels , governed by a set of parameters of a postulated generative process . One example in neuroimaging is dynamic causal modelling ( DCM ) [45] . DCM enables statistical inference on physiological quantities that are not directly observable with current methods , such as directed interregional coupling strengths and their modulation , e . g . , by synaptic gating [46] . ( We use the term DCM to refer both to a specific dynamic causal model and to dynamic causal modelling as a method . ) From a pathophysiological perspective , disturbances of synaptic plasticity and neuromodulation are at the heart of psychiatric spectrum diseases such as schizophrenia [47] or depression [48] . It is therefore likely that classification of disease states could benefit from exploiting estimates of these quantities . While DCM is a natural ( and presently the only ) candidate for obtaining model-based estimates of synaptic plasticity ( cf . [46] , [49] ) , the most widely used approach to classification relies on discriminative methods , such as support vector machines ( SVMs ) [50] , [51] . Together , DCM and SVM methods thus represent natural building blocks for classification of disease states . Generative embedding represents a special case of using generative kernels for classification , such as the P-kernel [52] or the Fisher kernel [53] . Generative kernels have been fruitfully exploited in a range of applications [54]–[66] and define an active area of research [67]–[70] . In the special case of generative embedding , a generative kernel is used to construct a generative score space . This is a model-based feature space in which the original observations have been replaced by statistical representations that potentially yield better class separability when fed into a discriminative classifier . Thus , an unsupervised embedding step is followed by a supervised classification step . In previous work , we suggested a concrete implementation of this approach for the trial-by-trial classification of electrophysiological recordings [61] . In this paper , we propose a DCM-based generative-embedding approach for subject-by-subject classification of fMRI data , demonstrate its performance using a clinical data set , and highlight potential methodological pitfalls ( and how to avoid them ) . DCM [45] views the brain as a nonlinear dynamical system of interconnected neuronal populations whose directed connection strengths are modulated by external perturbations ( i . e . , experimental conditions ) or endogenous activity . Here , we will use DCM to replace high-dimensional fMRI time series by a low-dimensional vector of parameter estimates . The discriminative part of our approach will be based on an SVM with a linear kernel . This algorithm learns to discriminate between two groups of subjects by estimating a separating hyperplane in their feature space . Since this paper brings together techniques from different statistical domains that tend to be used by different communities , we have tried to adopt a tutorial-like style and introduce basic concepts of either approach in the Methods section . Generative embedding for fMRI may offer three substantial advantages over conventional classification methods . First , because the approach aims to fuse the strengths of generative models with those of discriminative methods , it may outperform conventional voxel-based schemes , especially in those cases where crucial discriminative information is encoded in ‘hidden’ quantities such as directed ( synaptic ) connection strengths . Second , the construction of the feature space is governed and constrained by a biologically motivated systems model . As a result , feature weights can be interpreted mechanistically in the context of this model . Incidentally , the curse of dimensionality faced by many conventional feature-extraction methods may turn into a blessing when using generative embedding: the higher the temporal and spatial resolution of the fMRI data , the more precise the estimation of the parameters of the generative model , leading to better discriminability . Third , our approach can be used to compare alternative generative model architectures in situations where evidence-based approaches , such as Bayesian model selection , are not applicable . We will deal with these three points in more detail in the Discussion . The remainder of this paper is structured as follows . First , we summarize the general ideas of generative embedding and the specific generative and discriminative components used here , i . e . , DCM and SVM . We then inspect different procedures of how generative embedding could be implemented practically while distinguishing between approaches with and without bias . Third , we illustrate the utility of our approach , using empirical data obtained during speech processing in healthy volunteers and patients with moderate aphasia . These data have been explored in a previous study , in which DCM and Bayesian model selection ( BMS ) were applied to investigate the effective connectivity among cortical areas activated by intelligible speech [71] . In a subsequent study , we extended this analysis to patients with aphasia ( Schofield et al . , in preparation ) . In the present paper , we ask whether subject-specific directed connection strengths among cortical regions involved in speech processing contain sufficiently rich discriminative information to enable accurate predictions of the diagnostic category ( healthy or aphasic ) of a previously unseen individual . In brief , we found that ( i ) generative embedding yielded a near-perfect classification accuracy , ( ii ) significantly outperformed conventional ‘gold standard’ activation-based and correlation-based classification schemes , and ( iii ) afforded a novel mechanistic interpretation of the differences between aphasic patients and healthy controls during processing of speech and speech-like sounds .
The study was approved by the local research ethics committee at UCL , and all participants gave informed consent . Most methods for classification attempt to find a linear function that separates examples as accurately as possible in a space of features ( e . g . , voxel-wise measurements ) . Such discriminative classification methods differ from generative methods in two ways . First , rather than trying to estimate the joint density of observations and class labels , which is not needed for classification , or trying to estimate class-conditional probability densities , which can be difficult , discriminative classifiers directly model the class an example belongs to . Second , many discriminative methods do not operate on examples themselves but are based on the similarity between any two examples , expressed as the inner product between their feature vectors . This provides an elegant way of transforming a linear classifier into a more powerful nonlinear one . ( Note that the term discriminative methods is used here to collectively describe the class of learning algorithms that find a discriminant function for mapping an example onto a class label , typically without invoking probability theory . This is in contrast to discriminative models , which model the conditional probability , and generative models , which first model the full joint probability and then derive . ) The most popular classification algorithm of the above kind is the -norm soft-margin support vector machine ( SVM ) [50] , [51] , [72] , [73] . The only way in which examples enter an SVM is in terms of an inner product . This product can be replaced by the evaluation of a kernel function , which implicitly computes the inner product between the examples in a new feature space , . The -norm SVM is a natural choice when the goal is maximal prediction accuracy . However , it usually leads to a dense solution ( as opposed to a sparse solution ) in which almost all features are used for classification . This is suboptimal when one wishes to understand which model parameters contribute most to distinguishing groups , which will be the focus in the Section ‘Interpretation of the feature space . ’ In this case , an SVM that enforces feature sparsity may be more useful . One simple way of inducing sparsity is to penalize the number of non-zero coefficients by using an -regularizer . Unlike other regularizers , the -norm ( also known as the counting norm ) reduces the feature-selection bias inherent in unbounded regularizers such as the - or -norm . The computational cost of optimizing an -SVM objective function is prohibitive , because the number of subsets of items which are of size is exponential in . We therefore replace the -norm by a capped -regularizer which has very similar properties [74] . One way of solving the resulting optimization problem is to use a bilinear programming approach [75] . Here , we use a more efficient difference-of-convex-functions algorithm ( Ong & Thi , under review ) . In summary , we will use two types of SVM . For the purpose of classification ( Section ‘Classification’ ) , we aim to maximize the potential for highly accurate predictions by using an -norm SVM . For the purpose of feature selection and interpretation ( Section ‘Interpretation of the feature space’ ) , we will focus on feature sparsity by using an approximation to an -norm SVM , which will highlight those DCM parameters jointly deemed most informative in distinguishing between groups . Most current applications of classification algorithms in neuroimaging begin by embedding the measured recordings of each subject in a -dimensional Euclidean space . In fMRI , for example , a subject can be represented by a vector of features , each of which corresponds to the signal measured in a particular voxel at a particular point in time . This approach makes it possible to use any learning algorithm that expects vectorial input , such as an SVM; but it ignores the spatio-temporal structure of the data as well as the process that generated them . This limitation has motivated the search for kernel methods that provide a more natural way of measuring the similarity between the functional datasets of two subjects , for example by incorporating prior knowledge about how the data were generated , which has led to the idea of generative kernels , as described below . Generative kernels are functions that define a similarity metric for observed examples using a generative model . In the case of a dynamic causal model ( DCM ) , for example , the observed time series are modelled by a system of parameterized differential equations with Gaussian observation noise . Generative embedding defines a generative kernel by transferring the models into a vectorial feature space in which an appropriate similarity metric is defined ( see Figure 1 ) . This feature space , which we will refer to as a generative score space , embodies a model-guided dimensionality reduction of the observed data . The kernel defined in this space could be a simple inner product of feature vectors , or it could be based on any other higher-order function , as long as it is positive definite [76] . In conclusion , model-based classification via generative embedding is a hybrid generative-discriminative approach: it merges the explanatory abilities of generative models with the classification power of discriminative methods . The specific implementation for fMRI data proposed in this paper consists of four conceptual steps which are summarized in Figure 1 and described in the following subsections . First , a mapping is designed that projects an example from data space onto a multivariate probability distribution in a parametric family . In our case , we use the fMRI data from each subject to estimate the posterior density of the parameters of a DCM ( Sections ‘DCM for fMRI’ and ‘Model inversion’ ) . Second , a probability kernel is constructed that represents a similarity measure between two inverted DCMs . Here , we use a simple linear kernel on the maximum a posteriori ( MAP ) estimates of the model parameters ( Sections ‘Strategies for unbiased model specification and inversion’ and ‘Kernel construction’ ) . Third , this kernel is used for training and testing a discriminative classifier ( Section ‘Classification’ ) . Here , we employ a linear SVM to distinguish between patients and healthy controls . Fourth , the constructed feature space can be investigated to find out which model parameters jointly contributed most to distinguishing the two groups ( Section ‘Interpretation of the feature space’ ) . We will conclude with an example in which we distinguish between patients with moderate aphasia and healthy controls ( Sections ‘Experimental design , data acquisition , and preprocessing , ’ ‘Implementation of generative embedding , ’ and ‘Comparative analyses’ ) . DCM regards the brain as a nonlinear dynamic system of interconnected nodes , and an experiment as a designed perturbation of the system's dynamics [45] . Its goal is to provide a mechanistic model for explaining experimental measures of brain activity . While the mathematical formulation of DCMs varies across measurement types , common mechanisms modelled by all DCMs include synaptic connection strengths and experimentally induced modulation thereof [46] , [77]–[80] . Generally , DCMs strive for neurobiological interpretability of their parameters; this is one core feature distinguishing them from alternative approaches , such as multivariate autoregressive models [81] which characterize inter-regional connectivity in a phenomenological fashion . DCMs consist of two hierarchical layers [82] . The first layer is a neuronal model of the dynamics of interacting neuronal populations in the context of experimental perturbations . Critically , its parameters are neurobiologically interpretable , representing , for example , synaptic weights and their context-specific modulation; electrophysiological DCMs describe even more fine-grained processes such as spike-frequency adaptation or conduction delays . Experimental manipulations enter the model in two different ways: they can elicit responses through direct influences on specific regions ( e . g . , sensory inputs ) , or they can modulate the strength of coupling among regions ( e . g . , task demands or learning ) . The second layer of a DCM is a biophysically motivated forward model that describes how a given neuronal state translates into a measurement . Depending on the measurement modality , this can be a set of nonlinear differential equations ( as for fMRI [83] ) or a simple linear equation ( as for EEG [84] ) . While the forward model plays a critical role in model inversion , it is the parameters of the neuronal model that are typically of primary scientific interest . In this paper , we will use the classical bilinear DCM for fMRI [45] as implemented in the software package SPM8/DCM10 , ( 1 ) ( 2 ) where represents the neuronal state vector at time , is a matrix of endogenous connection strengths , represents the additive change of these connection strengths induced by modulatory input , and denotes the strengths of direct ( driving ) inputs . These neuronal parameters are rate constants with units . The haemodynamic forward model is given by the function , a nonlinear operator that links a neuronal state to a predicted blood oxygen level dependent ( BOLD ) signal via changes in vasodilation , blood flow , blood volume , and deoxyhaemoglobin content ( see [83] for details ) . This forward model has haemodynamic parameters and Gaussian measurement error . The haemodynamic parameters primarily serve to account for variations in neurovascular coupling across regions and subjects and are typically not of primary scientific interest . In addition , the haemodynamic parameters exhibit strong inter-dependencies and thus high posterior covariances and low precision [83] , which makes it difficult to establish the distinct contribution afforded by each parameter . For these reasons , the model-induced feature spaces in this paper will be based exclusively on the neuronal parameters . In summary , DCM provides a mechanistic model for explaining measured time series of brain activity as the outcome of hidden dynamics in an interconnected network of neuronal populations and its experimentally induced perturbations . Inverting such a model ( see next section ) means to infer the posterior distribution of the parameters of both the neuronal and the forward model from observed responses of a specific subject . Its mechanistic interpretability and applicability to single-subject data makes DCM an attractive candidate for generative embedding of fMRI data . Bayesian inversion of a given dynamic causal model defines a map that projects a given example ( i . e . , data from a single subject ) onto a multivariate probability distribution in a parametric family . The model architecture specifies the neuronal populations ( regions ) of interest , experimentally controlled inputs , synaptic connections , and a prior distribution over the parameters . Given the model and subject-specific data , model inversion proceeds in an unsupervised and subject-by-subject fashion , i . e . , in ignorance of the subject label that will later be used in the context of classification . ( The literature on DCM has adopted the convention of denoting the hidden states by and the data by . Here , in order to keep the notation consistent with the literature on classification , we use for the data and for the labels . A distinct symbol for the hidden states is not required here . ) DCM uses a fully Bayesian approach to parameter estimation , with empirical priors for the haemodynamic parameters and conservative shrinkage priors for the coupling parameters [85] , [45] . Combining the prior density over the parameters with the likelihood function yields the posterior density . This inversion can be carried out efficiently by maximizing a variational free-energy bound to the log model evidence , , under Gaussian assumptions about the posterior ( the Laplace assumption; see [86] for details ) . Given parameters , model inversion thus yields a subject-specific probability density that can be fully described in terms of a vector of posterior means and a covariance matrix . Model specification and selection is an important theme in DCM [87] . In this paper we are not concerned with the question of which of several alternative DCMs may be optimal for explaining the data or for classifying subjects; these issues can be addressed using Bayesian evidence methods [88] , [89] or by applying cross-validation to the classifications suggested by each of the models , respectively ( see [61] for an example ) . However , an important issue is that model specification cannot be treated in isolation from its subsequent use for classification . Specifically , some procedures for selecting time series can lead to biased estimation of classification accuracy . In the next section , we therefore provide a detailed assessment of different strategies for time series selection in DCM-based generative embedding and highlight those procedures which safeguard against obtaining optimistic estimates of classification performance . For conventional fMRI classification procedures , good-practice guidelines have been suggested for avoiding an optimistic bias in assessing classification performance [8] , [10] . Generally , to obtain an unbiased estimate of generalization accuracy , a classifier must be applied to test data that have not been used during training . In generative embedding , this principle implies that the specification of the generative model cannot be treated in isolation from its use for classification . In this section , we structure different strategies in terms of a decision tree and evaluate the degree of bias they invoke ( see Figure 2 ) . The first distinction is based on whether the regions of interest ( ROIs ) underlying the DCM are defined anatomically or functionally . When ROIs are defined exclusively on the basis of anatomical masks ( Figure 2a ) , the selection of voxels is independent of the functional data . Using time series from these regions , the model is inverted separately for each subject . Thus , given subjects , a single initial model-specification step is followed by subject-wise model inversions . The resulting parameter estimates can be safely submitted to a cross-validation procedure to obtain an unbiased estimate of classification performance . Whenever functional contrasts have played a role in defining ROIs , subsequent classification may no longer be unbiased . This is because a functional contrast introduces statistics of the data into voxel selection , which usually generates a bias . In this case , we ask whether contrasts are defined in an across-subjects or a between-groups fashion . In the case of an across-subjects contrast ( which does not take into account group membership ) , one might be tempted to follow the same logic as in the case of anatomical ROI definitions: a single across-subjects contrast , computed for all subjects , guides the selection of voxels , and the resulting DCM is inverted separately for each subject ( Figure 2b ) . Unfortunately , this procedure is problematic . When using the resulting parameter estimates in a leave-one-out cross-validation scheme , in every repetition the features would be based on a model with regions determined by a group contrast that was based on the data from all subjects , including the left-out test subject . This means that training the classifier would no longer be independent of the test data , which violates the independence assumption underlying cross-validation , a situation referred to as peeking [10] . In consequence , the resulting generalization estimate may exhibit an optimistic bias . To avoid this bias , model specification must be integrated into cross-validation ( Figure 2c ) . Specifically , in each fold , we leave out one subject as a test subject and compute an across-subjects group contrast from the remaining subjects . The resulting choice of voxels is then used for specifying time series in each subject and the resulting model is inverted separately for each subject , including the left-out test subject . This procedure is repeated times , each time leaving out a different subject . In total , the model will be inverted times . In this way , within each cross-validation fold , the selection of voxels is exclusively based on the training data , and no peeking is involved . This is the strategy adopted for the dataset analysed in this paper , as detailed in the Section ‘Implementation of generative embedding’ . When functional contrasts are not defined across all subjects but between groups , the effect of peeking may become particularly severe . Using a between-groups contrast to define regions of interest on the basis of all available data , and using these regions to invert the model for each subject ( Figure 2d ) would introduce information about group membership into the process of voxel selection . Thus , feature selection for both training and test data would be influenced by both the data and the label of the left-out test subject . One way of decreasing the resulting bias is to integrate model specification into cross-validation ( Figure 2e ) . In this procedure , the between-groups contrast is computed separately for each training set ( i . e . , based on subjects ) , and the resulting regions are used to invert the model for the test subject . This means that the class label of the test subject is no longer involved in selecting features for the test subject . However , the test label continues to influence the features of the training set , since these are based on contrasts defined for a group that included the test subject . This bias can only be removed by adopting the same laborious procedure as with across-subjects contrasts: by using a between-groups contrast involving subjects , inverting the resulting model separately for each subject , and repeating this procedure times ( Figure 2f ) . This procedure guarantees that neither the training procedure nor the features selected for the test subject were influenced by the data or the label of the test subject . In summary , the above analysis shows that there are three practical strategies for the implementation of generative embedding that yield an unbiased cross-validated accuracy estimate . If regions are defined anatomically , the model is inverted separately for each subject , and the resulting parameter estimates can be safely used in cross-validation ( Figure 2a ) . Otherwise , if regions are defined by a functional contrast , both the definition of ROIs and model inversion for all subjects need to be carried out separately for each cross-validation fold ( Figure 2c , f ) . Given a set of inverted subject-specific generative models , the kernel defines the similarity metric under which these models are assessed within a discriminative classifier . In generative embedding , the choice of an appropriate kernel depends on the definition of the generative score space . A straightforward way to create a Euclidean vector space from an inverted DCM is to consider the posterior means or maximum a posteriori ( MAP ) estimates of model parameters of interest ( e . g . , parameters encoding synaptic connection strengths ) . More formally , we can define a mapping that extracts a subset of MAP estimates from the posterior distribution . This simple -dimensional vector space expresses discriminative information encoded in the connection strengths between regions , as opposed to activity levels within these regions . Alternatively , one could also incorporate elements of the posterior covariance matrix into the vector space . This would be beneficial if class differences were revealed by the precision with which connection strengths can be estimated from the data . Once a generative score space has been created , any conventional kernel can be used to compare two inverted models . The simplest one is the linear kernel , representing the inner product between two vectors and . Nonlinear kernels , such as quadratic , polynomial or radial basis function kernels , transform the generative score space , which makes it possible to consider quadratic ( or higher-order ) class boundaries and therefore account for possible interactions between features . Nonlinear kernels , however , have several disadvantages for generative embedding . As the complexity of the kernel increases , so does the risk of overfitting . Furthermore , feature weights are easiest to interpret in relation to the underlying model when they do not undergo further transformation; then , the contribution of a particular feature ( i . e . , model parameter ) to the success of the classifier can be understood as the degree to which the neuronal mechanism represented by that parameter aids classification . A simple linear kernel will therefore be our preferred choice . In summary , in this paper , we define a mapping from a subject-specific posterior distribution of model parameters to a feature vector . We then use a linear kernel for this model-based feature space . Together , these two steps define a probability kernel that represents a similarity metric between two inverted models and allows for mechanistic interpretations of how group membership of different subjects is encoded by spatiotemporal fMRI data . While a kernel describes how two subjects can be compared using a generative model of their fMRI data , it does not specify how such a comparison could be used for making predictions . This gap is filled by discriminative classification methods . As described in the Section ‘Combining generative models and discriminative methods’ , a natural choice is the -norm soft-margin support vector machine ( SVM ) , which currently represents the most widely used kernel method for classification [72] . An estimate of classification performance with minimal variance can be obtained by leave-one-out cross-validation . In each fold , the classifier is trained on subjects and tested on the left-out one . Using the training set only , the SVM can be fine-tuned by carrying out a simple line search over the regularization hyperparameter ( Eqn . 1 ) , a procedure known as nested cross-validation [90] , [91] . There are many ways of assessing the generalization performance of a classifier . Here , we are primarily interested in the balanced accuracy , that is , the mean accuracy obtained on either class , ( 3 ) where , , , and represent the number of true positives , false positives , true negatives , and false negatives , respectively [92] . The balanced accuracy represents the arithmetic mean between sensitivity and specificity . If the classifier performs equally well on either class , it reduces to the ordinary accuracy ( i . e . , the ratio of correct predictions to all predictions ) . If , however , the classifier has taken advantage of an imbalanced dataset , then the ordinary accuracy will be inflated , whereas the balanced accuracy will drop to chance ( 50% ) , as desired . The balanced accuracy thus removes the bias from estimates of generalizability that may arise in the presence of imbalanced datasets . A probability interval can be computed by considering the convolution of two Beta-distributed random variables that correspond to the true accuracies on positive and negative examples , respectively . A p-value can then be obtained by computing the posterior probability of the accuracy being below chance [92] . Most classification algorithms can not only be used for making predictions and obtaining an estimate of their generalization error; they can also be used to quantify how much each feature has contributed to classification performance . Such feature weights can sometimes be of greater interest than the classification accuracy itself . In the case of a generative score space , as defined above , each feature is associated with a neurobiologically interpretable model parameter . Provided there are no complex transformations of feature weights ( see above ) , they can be interpreted in the context of the underlying model . As described in the Section ‘Combining generative models and discriminative methods’ , the -norm soft-margin SVM is a natural choice when the goal is maximal prediction accuracy . However , its solution usually implies that almost all features are used for classification . This is suboptimal when one wishes to understand which model parameters , and thus mechanisms , contribute most to distinguishing groups . Therefore , for the purposes of interpreting the model-induced feature space , we use an -regularizer . This approach allows us to characterize the feature space by counting how often a particular feature has been selected in leave-one-out cross-validation . In order to illustrate the utility of generative embedding for fMRI , we used data from two groups of participants ( patients with moderate aphasia vs . healthy controls ) engaged in a simple speech-processing task . The conventional SPM and DCM analyses of these data are published elsewhere; we refer to [71] and Schofield et al . ( in preparation ) for detailed descriptions of all experimental procedures . The two groups of subjects consisted of 26 right-handed healthy participants with normal hearing , English as their first language , and no history of neurological disease ( 12 female; mean age 54 . 1 years; range 26–72 years ) ; and 11 patients diagnosed with moderate aphasia due to stroke ( 1 female; mean age 66 . 1; range 45–90 years ) . The patients' aphasia profile was characterized using the Comprehensive Aphasia Test [93] . As a group , they had scores in the aphasic range for: spoken and written word comprehension ( single word and sentence level ) , single word repetition and object naming . It is important to emphasize that the lesions did not affect any of the temporal regions which we included in our model described below ( see Schofield et al . , in preparation , for detailed information on lesion localization ) . Subjects were presented with two types of auditory stimulus: ( i ) normal speech; and ( ii ) time-reversed speech , which is unintelligible but retains both speaker identity and the spectral complexity of normal speech . Subjects were given an incidental task , to make a gender judgment on each auditory stimulus , which they indicated with a button press . Functional T2*-weighted echo-planar images ( EPI ) with BOLD contrast were acquired using a Siemens Sonata 1 . 5 T scanner ( in-plane resolution 3 mm×3 mm; slice thickness 2 mm; inter-slice gap 1 mm; TR 3 . 15 s ) . In total , 122 volumes were recorded in each of 4 consecutive sessions . In addition , a T1-weighted anatomical image was acquired . Following realignment and unwarping of the functional images , the mean functional image of each subject was coregistered to its high-resolution structural image . This image was spatially normalized to standard Montreal Neurological Institute ( MNI152 ) space , and the resulting deformation field was applied to the functional data . These data were then spatially smoothed using an isotropic Gaussian kernel ( FWHM 8 mm ) . In previous work , these data have been analysed using a conventional general linear model ( GLM ) and DCM; the results are described in Schofield et al . ( in preparation ) . Here , we re-examined the dataset using the procedure shown in Figure 2c , as described in detail in the next subsection . We compared the performance of generative embedding to a range of alternative approaches . To begin with , we examined several conventional activation-based classification schemes . The first method was based on a feature space composed of all voxels within the predefined anatomical masks used for guiding the specification of the DCMs . As above , we used a linear SVM , and all training sets were balanced by oversampling . We will refer to this approach as anatomical feature selection . The second method , in contrast to the first one , was not only based on the same classifier as in generative embedding but also used exactly the same voxels . Specifically , voxels were selected on the basis of the same ‘all auditory events’ contrast as above , which is a common approach to defining a voxel-based feature space in subject-by-subject classification [11] , [12] , [10] . In every cross-validation fold , only those voxels entered the classifier that survived a t-test ( , uncorrected ) in the current set of subjects . Training sets were balanced by oversampling . We will refer to this method as contrast feature selection . The third activation-based method employed a locally multivariate ‘searchlight’ strategy for feature selection . Specifically , in each cross-validation fold , a searchlight sphere ( radius 4 mm ) was passed across all voxels contained in the anatomical masks described above [23] . Using the training set only , a nested leave-one-out cross-validation scheme was used to estimate the generalization performance of each sphere using a linear SVM with a fixed regularization hyperparameter ( ) . Next , all spheres with an accuracy greater than 75% were used to form the feature space for the current outer cross-validation fold , which corresponds to selecting all voxels whose local neighbourhoods allowed for a significant discrimination between patients and healthy controls at . Both outer and inner training sets were balanced by oversampling . We will refer to this method as searchlight feature selection . To illustrate the location of the most informative voxels , we carried out an additional searchlight analysis , based on the entire dataset as opposed to a subset of size , and used the results to generate a discriminative map ( see Figure S1 in the Supplementary Material ) . The fourth conventional method was based on a principal component analysis ( PCA ) to reduce the dimensionality of the feature space constructed from all voxels in the anatomical masks described above . Unlike generative embedding , PCA-based dimensionality reduction finds a linear manifold in the data without a mechanistic view of how those data might have been generated . We sorted all principal components in decreasing order of explained variance . By retaining the 22 top components , the resulting dimensionality matched the dimensionality of the feature space used in generative embedding . In addition to the above activation-based methods , we compared generative embedding to several approaches based on undirected regional correlations . We began by averaging the activity within each region of interest to obtain region-specific representative time series . We then computed pairwise correlation coefficients to obtain a 15-dimensional feature space of functional connectivity . Next , instead of computing spatial averages , we summarized the activity within each region in terms of the first eigenvariate . Thus , in this approach , the exact same data was used to estimate functional connectivity as was used by DCM to infer effective connectivity . Finally , as suggested in [43] , we created yet another feature space by transforming the correlation coefficients on eigenvariates into z-scores using the Fisher transformation [96] . In addition to conventional activation- and correlation-based approaches , we also investigated the dependence of generative embedding on the structure of the underlying model . Specifically , we repeated our original analysis on the basis of three alternative models . For the first model , we constructed a feedforward system by depriving the original model of all feedback and interhemispheric connections ( Figure 5a ) ; while this model could still , in principle , explain neuronal dynamics throughout the system of interest , it was neurobiologically less plausible . For the second and third model , we kept all connections from the original model but modelled either only the left hemisphere ( Figure 5b ) or only the right hemisphere ( Figure 5c ) . In summary , we compared the primary approach proposed in this paper to 4 conventional activation-based methods , 3 conventional correlation-based methods , and 3 generative-embedding analyses using reduced and biologically less plausible models .
The classification performance of generative embedding was evaluated using the procedure described in Figure 2c . This procedure was compared to several conventional activation-based and correlation-based approaches . As an additional control , generative embedding was carried out on the basis of three biologically ill-informed models . In all cases , a leave-one-subject-out cross-validation scheme was used to obtain the posterior distribution of the balanced accuracy [92] as well as smooth estimates of the underlying receiver-operating characteristic ( ROC ) and precision-recall ( PC ) curves [97] . Results are presented in Table 2 and Figure 6 . The strongest classification performance was obtained when using generative embedding with the full model shown in Figure 3 . The approach correctly associated 36 out of 37 subjects with their true disease state , corresponding to a balanced accuracy of 98% . Regarding conventional activation-based methods , classification based on anatomical feature selection did not perform significantly above chance ( balanced accuracy 62% , p≈0 . 089 ) . Contrast feature selection ( 75% , p≈0 . 003 ) , searchlight feature selection ( 73% , p≈0 . 006 ) , and PCA-based dimensionality reduction ( 80% , p<0 . 001 ) did perform significantly above chance; however , all methods were outperformed significantly by generative embedding ( p≈0 . 003 , p≈0 . 001 , and p≈0 . 045 , paired-sample Wald test ) . Regarding conventional correlation-based methods , all three approaches performed above chance , whether based on correlations amongst the means ( 70% , p≈0 . 011 ) , correlations amongst eigenvariates ( 83% , p<0 . 001 ) , or z-transformed correlations amongst eigenvariates ( 74% , p≈0 . 002 ) . Critically , however , all were significantly outperformed by generative embedding ( p<0 . 001 , p≈0 . 045 , p≈0 . 006 ) . Regarding generative embedding itself , when replacing the original model shown in Figure 3 by a biologically less plausible feedforward model ( Figure 5a ) or by a model that captured the left hemisphere only ( Figure 5b ) , we observed a significant decrease in performance , from 98% down to 77% ( p≈0 . 002 ) and 81% ( p≈0 . 008 ) , respectively , although both accuracies remained significantly above chance ( p≈0 . 001 and p<0 . 001 ) . By contrast , when modelling the right hemisphere only ( Figure 5c ) , performance dropped to a level indistinguishable from chance ( 59 . 3% , p≈0 . 134 ) . In order to provide a better intuition as to how the generative model shown in Figure 3 created a score space in which examples were much better separated than in the original voxel-based feature space , we produced two scatter plots of the data ( see Figure 7 ) . The first plot is based on the peak voxels of the three most discriminative clusters among all regions of interest , evaluated by a searchlight classification analysis . The second plot , by analogy , is based on the three most discriminative model parameters , as measured by two-sample t-tests in the ( normalized ) generative score space . This illustration shows how the voxel-based projection ( left ) leads to classes that still overlap considerably , whereas the model-based projection ( right ) provides an almost perfectly linear separation of patients and controls . The low dimensionality of the model-based feature space makes it possible to visualize each example in a radial coordinate system , where each axis corresponds to a particular model parameter ( see Figure 8 ) . When using parameters that represent directed connection strengths , this form of visualization is reminiscent of the notion of ‘connectional fingerprints’ for characterizing individual cortical regions [98] . In our case , there is no immediately obvious visual difference in fingerprints between aphasic patients and healthy controls . On the contrary , the plot gives an impression of the large variability across subjects and suggests that differences might be subtle and possibly jointly encoded in multiple parameters . One way of characterizing the discriminative information encoded in individual model parameters more directly is to estimate class-conditional univariate feature densities ( see Figure 9 ) . Here , densities were estimated in a nonparametric way using a Gaussian kernel with an automatically selected bandwidth , making no assumptions about the distributions other than smoothness [99] . While most densities are heavily overlapping , a two-sample t-test revealed significant group differences in four model parameters ( denoted by stars in Figure 9 ) : the self-connection of L . HG ( parameter 4 ) ; the influence that L . HG exerts over L . PT ( parameter 5 ) ; the influence R . MGB on R . PT ( parameter 13 ) ; and the influence of R . HG on L . HG ( parameter 14 ) . All of these were significant at the 0 . 001 level while no other parameter survived p = 0 . 05 . An extended plot of all bivariate feature distributions , illustrating how well any two features jointly discriminated between patients and healthy controls , can be found in the Supplementary Material ( Figure S2 ) . In order to better understand which DCM parameters jointly enabled the distinction between patients and controls , we examined the frequency with which features were selected in leave-one-out cross-validation when using an SVM with a sparsity-inducing regularizer [75] , [74] ( see Figure 10 ) . We found that the classifier favoured a highly consistent and sparse set of 9 ( out of 22 ) model parameters; the corresponding synaptic connections are highlighted in red in Figure 3 . Notably , this 9-dimensional feature space , when used with the original -norm SVM , yielded the same balanced classification accuracy ( 98% ) as the full 22-dimensional feature space , despite discarding more than two thirds of its dimensions . The above representation disclosed interesting potential mechanisms . For example , discriminative parameters were restricted to cortico-cortical and thalamo-cortical connection strengths , whereas parameters representing auditory inputs to thalamic nuclei did not contribute to the distinction between patients and healthy controls . This finding implies that , as one would expect , low-level processing of auditory stimuli , from brain stem to thalamus , is unimpaired in aphasia and that processing deficiencies are restricted to thalamo-cortical and cortico-cortical networks . In particular , the discriminative connections included the top-down connections from planum temporale to Heschl's gyrus bilaterally; the importance of these connections had also been highlighted by the previous univariate analyses of group-wise DCM parameters in the study by Schofield et al . ( in preparation ) . Furthermore , all of the connections from the right to the left hemisphere were informative for group membership , but none of the connections in the reverse direction . This pattern is interesting given the known specialization of the left hemisphere in language and speech processing and previous findings that language-relevant information is transferred from the right hemisphere to the left , but not vice versa [100] . It implies that aphasia leads to specific changes in connectivity , even in non-lesioned parts of the language network , with a particular effect on inter-hemispheric transfer of speech information . This specificity is seen even more clearly when considering only those three parameters which were selected 100% of the time ( i . e . , in all cross-validation folds ) and are thus particularly meaningful for classification ( bold red arrows in Figure 3 ) . The associated connections mediate information transfer from the right to the left hemisphere and converge on the left planum temporale which is a critical structure for processing of language and speech [101] , [102] . In summary , all selected features represented connectivity parameters ( as opposed to stimulus input ) , their selection was both sparse and highly consistent across resampling repetitions , and their combination was sufficient to afford the same classification accuracy as the full feature set .
Generative embedding for subject-by-subject classification provides three potential advantages over conventional voxel-based methods . The first advantage is that it combines the explanatory strengths of generative models with the classification power of discriminative methods . Thus , in contrast to purely discriminative or purely generative methods , generative embedding is a hybrid approach . It fuses a feature space that captures both the data and their generative process with a classifier that finds the maximum-margin boundary for class separation . Intuitively , this exploits the idea that differences in the generative process between two examples ( observations ) might provide optimal discriminative information required to enable accurate predictions . In the case of DCM for fMRI , this rationale should pay off whenever the directed connection strengths between brain regions contain more information about a disease state than regional activations or undirected correlations . Indeed , this is what we found in our analyses ( cf . Figure 6 ) . Using a DCM-informed data representation might prove particularly relevant in psychiatric disorders , such as schizophrenia or depression , where aberrant effective connectivity and synaptic plasticity are central to the disease process [48] , [47] . The second advantage of generative embedding for fMRI is that it enables an intuitive and mechanistic interpretation of features and their weights , an important property not afforded by most conventional classification methods [103] , [104] . By using parameter estimates from a mechanistically interpretable model for constructing a feature space , the subsequent classification no longer yields ‘black box’ results but allows one to assess the relative importance of different mechanisms for distinguishing groups ( e . g . , whether or not synaptic plasticity alters the strengths of certain connections in a particular context ) . Put differently , generative embedding embodies a shift in perspective: rather than representing sequential data in terms of high-dimensional and potentially highly complex trajectories , we are viewing the data in terms of the coefficients of a well-behaved model of system dynamics . Again , this may be of particular importance for clinical applications , as discussed in more detail below . It is also interesting to note that models like DCM , when used in the context of generative embedding , turn the curse of dimensionality faced by conventional classification methods into a blessing: the higher the spatial and temporal resolution of the underlying fMRI data , the more precise the resulting DCM parameter estimates; this in turn should lead to more accurate predictions . The third advantage provided by generative embedding is related to model comparison . For any given dataset , there is an infinite number of possible dynamic causal models , differing in the number and location of nodes , in connectivity structure , and in their parameterization ( e . g . , priors ) . Competing models can be compared using Bayesian model selection ( BMS ) [89] , [83] , [86] , [88] , where the best model is the one with the highest model evidence , that is , the highest probability of the data given the model [105] . BMS is a generic approach to distinguish between different models that is grounded in Bayesian probability theory and , when group-specific mechanisms can be mapped onto distinct models , represents a powerful technique for model-based classification in itself . However , there are two scenarios in which BMS is problematic and where classification based on generative embedding may represent a useful alternative [61] . First , BMS requires the data to be identical for all competing models . Thus , in the case of current implementations of DCM for fMRI , BMS enables dynamic model selection ( concerning the parameterization and mathematical form of the model equations ) but not structural model selection ( concerning which regions or nodes should be included in the model ) . Second , BMS is limited when different groups cannot be mapped onto different model structures , for example when the differences in neuronal mechanisms operate at a finer conceptual scale than can be represented within the chosen modelling framework . In this case , discriminability of subjects may be afforded by differences in ( combinations of ) parameter estimates under the same model structure ( see [106] for a recent example ) . In both these scenarios , the approach proposed in this paper may provide a solution , in that the unsupervised creation of a generative score space can be viewed as a method for biologically informed feature extraction , and the performance of the classifier reflects how much class information is encoded in the model parameters . This view enables a form of model comparison in which the best model is the one that enables the highest classification accuracy . This procedure can be applied even when ( i ) the underlying data ( e . g . , the chosen regional fMRI time series ) are different , or when ( ii ) the difference between two models lies exclusively in the pattern of parameter estimates . In this paper , we have illustrated both ideas: structural model selection to decide between a full model and two reduced models that disregard one hemisphere; and dynamic model selection to distinguish between different groups of subjects under the same model structure . In summary , BMS evaluates the goodness of a model with regard to its generalizability for explaining the data , whereas generative embedding evaluates a model in relation to an external criterion , i . e . , how well it allows for inference on group membership of any given subject . This difference is important as it highlights that the concept of a ‘good’ model can be based on fundamentally different aspects , and one could imagine scenarios where BMS and generative embedding arrive at opposing results . If , for example , discriminability of groups relies on a small subspace of the data , then one model ( which provides a good accuracy-complexity trade-off for most of the data except that subspace ) may have higher evidence , but another model that describes this subspace particularly well but is generally worse for the rest of the data may result in better classification performance ( cf . our discussion in [106] ) . We will examine the relation and complementary nature of BMS and generative-embedding approaches in future work . As discussed in this paper , there are three valid strategies for the implementation of generative embedding in fMRI that allow for an unbiased estimate of classification accuracy ( Figure 2 ) . If regions ( and thus time series ) are defined anatomically , the model is inverted separately for each subject , and the resulting parameter estimates can be safely used in cross-validation . If regions are defined by a functional contrast , both time series selection and model inversion for all subjects need to be carried out separately for each cross-validation fold . These procedures clearly have higher computational demands than conventional classification techniques , but the subject-wise nature of model inversion means that generative embedding for fMRI can exploit methods for distributed computing and can thus be implemented even for larger numbers of subjects . In order to demonstrate the utility of generative embedding for fMRI , we acquired and analysed a dataset consisting of 11 aphasic patients and 26 healthy controls . During the experiment , participants were listening to a series of speech and speech-like stimuli . In an initial analysis ( Schofield et al . , in preparation ) , we designed a dynamic causal model to explain observed activations in 6 auditory regions of interest . Here , we extended this analysis by examining whether patients and healthy controls could be distinguished on the basis of differences in subject-specific generative models . Specifically , we trained and tested a linear support vector machine on subject-wise estimates of connection strengths . This approach delivered two sets of results . First , we found strong evidence in favour of the hypothesis that aphasic patients and healthy controls may be distinguished on the basis of differences in the parameters of a generative model alone . Generative embedding did not only yield a near-perfect balanced classification accuracy ( 98% ) . It also significantly outperformed conventional activation-based methods , whether they were based on anatomical ( 62% ) , contrast ( 75% ) , searchlight feature selection ( 73% ) , or on a PCA-based dimensionality reduction ( 80% ) . Similarly , our approach outperformed conventional correlation-based methods , whether they were based on regional means ( 70% ) or regional eigenvariates ( 83% and 74% ) . Furthermore , it is interesting to observe that group separability was reduced considerably when using a less plausible feedforward model ( 77% ) . Finally , performance decreased significantly when modelling only the left hemisphere ( 81% ) , and it dropped to chance when considering the right hemisphere by itself ( 60% ) , which is precisely what one would expect under the view that the left hemisphere is predominantly , but not exclusively , implicated in language processing . Taken together , our findings provide strong support for the central idea of this paper: that critical differences between groups of subjects may be expressed in a highly nonlinear manifold which remains inaccessible by methods relying on activations or undirected correlations , but which can be unlocked by the nonlinear transformation embodied by an appropriate generative model . Second , since features correspond to model parameters , our approach allowed us to characterize a subset of features ( Figure 10 ) that can be interpreted in the context of the underlying model ( Figure 3 ) . This subset showed four remarkable properties . ( i ) Discriminative parameters were restricted to cortico-cortical and thalamo-cortical connection strengths . On the contrary , parameters representing auditory inputs to thalamic nuclei did not contribute to the distinction between patients and healthy controls . ( ii ) We observed a high degree of stability across resampling folds . That is , the same 9 ( out of 22 ) features were selected on almost every repetition . ( iii ) The set of discriminative parameters was found to be sparse , not just within repetitions ( which is enforced by the underlying regularizer ) but also across repetitions ( which is not enforced by the regularizer; see Figure S3 in the Supplementary Material ) . At the same time , the set was considerably larger than what would be expected from univariate feature-wise t-tests ( Figure 9 ) . ( iv ) The sparse set of discriminative parameters proved sufficient to yield the same balanced classification accuracy ( 98% ) as the full set . These results are consistent with the notion that a distinct mechanism , and thus few parameters , are sufficient to explain differences in processing of speech and speech-like sounds between aphasic patients and healthy controls . In particular , all of the connections from the right to the left hemisphere were informative with regard to group membership , but none of the connections in the reverse direction . This asymmetry resonates with previous findings that language-relevant information is transferred from the right hemisphere to the left , but not vice versa [100] , and suggests that in aphasia connectivity changes in non-lesioned parts of the language network have particularly pronounced effects on inter-hemispheric transfer of speech information from the ( non-dominant ) right hemisphere to the ( dominant ) left hemisphere . It is worthwhile briefly commenting on how the present findings relate to those of the original DCM study by Schofield et al . ( in preparation ) . Two crucial differences are that the previous study ( i ) applied Bayesian model averaging to a set of 512 models and ( ii ) statistically examined each of the resulting average connection strengths in a univariate fashion . They found group differences for most connections , highlighting in particular the top-down connections from planum temporale to primary auditory cortex bilaterally . In our multivariate analysis , these two connections were also amongst the most informative ones for distinguishing patients from controls ( Figure 3 ) . Schofield et al . also found group differences for interhemispheric connection strengths between left and right Heschl's gyrus , but their univariate approach did not demonstrate any asymmetries . In contrast , our multivariate approach yielded a sparser set of discriminative connections , highlighting the asymmetries of interhemispheric connections described above ( Figure 3 ) . The example described in this paper was chosen to illustrate the implementation and use of generative embedding for fMRI . It is important to emphasize that this example does not represent the sort of clinical application that we envisage in the long term . Clearly , there are few diagnostic problems when dealing with aphasia and usually a clinical examination by the physician is sufficient . However , this example is useful for demonstrating the utility of generative embedding since the diagnostic status of each subject is known without doubt and the networks involved in speech processing are well characterized . In the future , we hope that our approach will be useful for addressing clinical problems of high practical relevance , for instance for dissecting psychiatric spectrum disorders , such as schizophrenia , into physiologically defined subgroups [47] , or for predicting the response of individual patients to specific drugs . While an increasing number of studies have tried to describe neurobiological markers for psychiatric disorders [22] , [107] , [108] , [3] , [109] , [110] , [14] , [15] , we argue that these studies should be complemented by model-based approaches for inferring biologically plausible mechanisms . Such approaches will be useful in two domains of application: they can be used to decide between competing hypotheses ( as in traditional applications of DCM and BMS ) ; and they can harvest the potentially rich discriminative information encoded in aspects of synaptic plasticity or neuromodulation to build classifiers that distinguish between different subtypes of a psychiatric disorder on a physiological basis ( using techniques such as generative embedding ) . In the case of the illustrative dataset analysed in this paper , generative embedding yielded stronger classification performance than conventional methods , whether they were based on activations or regional correlations . One might think that this superior ability to accurately classify individual subjects determines the clinical value of the approach . Instead , we wish to argue that its clinical value will ultimately depend on whether patients that share the same symptoms can be differentially treated according to the underlying pathophysiology of the disorder . Generative embedding , using biologically plausible and mechanistically interpretable models , may prove critical in establishing diagnostic classification schemes that distinguish between pathophysiologically distinct subtypes of spectrum diseases and allow for predicting individualized behavioural and pharmacological therapy . | Neurological and psychiatric spectrum disorders are typically defined in terms of particular symptom sets , despite increasing evidence that the same symptom may be caused by very different pathologies . Pathophysiological classification and effective treatment of such disorders will increasingly require a mechanistic understanding of inter-individual differences and clinical tools for making accurate diagnostic inference in individual patients . Previous classification studies have shown that functional magnetic resonance imaging ( fMRI ) can be used to differentiate between healthy controls and neurological or psychiatric patients . However , these studies are typically based on descriptive patterns and indirect measures of neural activity , and they rarely afford mechanistic insights into the underlying condition . In this paper , we address this challenge by proposing a classification approach that rests on a model of brain function and exploits the rich discriminative information encoded in directed interregional connection strengths . Based on an fMRI dataset acquired from moderately aphasic patients and healthy controls , we illustrate that our approach enables more accurate classification and deeper mechanistic insights about disease processes than conventional classification methods . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"computer",
"science",
"computer",
"modeling",
"mathematics",
"neurological",
"disorders",
"statistics",
"neurology",
"statistical",
"methods",
"neuroimaging"
]
| 2011 | Generative Embedding for Model-Based Classification of fMRI Data |
Protein aggregation , arising from the failure of the cell to regulate the synthesis or degradation of aggregation-prone proteins , underlies many neurodegenerative disorders . However , the balance between the synthesis , clearance , and assembly of misfolded proteins into neurotoxic aggregates remains poorly understood . Here we study the effects of modulating this balance for the amyloid-beta ( Aβ ) peptide by using a small engineered binding protein ( ZAβ3 ) that binds with nanomolar affinity to Aβ , completely sequestering the aggregation-prone regions of the peptide and preventing its aggregation . Co-expression of ZAβ3 in the brains of Drosophila melanogaster expressing either Aβ42 or the aggressive familial associated E22G variant of Aβ42 abolishes their neurotoxic effects . Biochemical analysis indicates that monomer Aβ binding results in degradation of the peptide in vivo . Complementary biophysical studies emphasize the dynamic nature of Aβ aggregation and reveal that ZAβ3 not only inhibits the initial association of Aβ monomers into oligomers or fibrils , but also dissociates pre-formed oligomeric aggregates and , although very slowly , amyloid fibrils . Toxic effects of peptide aggregation in vivo can therefore be eliminated by sequestration of hydrophobic regions in monomeric peptides , even when these are extremely aggregation prone . Our studies also underline how a combination of in vivo and in vitro experiments provide mechanistic insight with regard to the relationship between protein aggregation and clearance and show that engineered binding proteins may provide powerful tools with which to address the physiological and pathological consequences of protein aggregation .
Of the neurodegenerative disorders that have been linked to protein misfolding and aggregation [1] , Alzheimer's disease ( AD ) is the most common [2] , [3] . Transgenic animal models have shown that aggregation of the Alzheimer β-peptide ( Aβ ) causes memory impairment [4] , [5] and cognitive deficits [6] similar to those seen in patients suffering from AD . Aβ aggregation precedes neuritic changes [7] , and there is a quantitative correlation between the propensities of mutant forms of Aβ to aggregate and their neurotoxicity [8] . In vitro aggregation of Aβ proceeds from the initial association of monomers into oligomeric , but still soluble , assemblies that ultimately form highly structured and insoluble amyloid fibrils [1] , [9] , [10] , [11] . Evidence suggests that the primary neurotoxic species are the soluble oligomeric aggregates [4] , [5] , [12] , [13] and that a fundamental building block may be dimeric Aβ species [14] . However , despite this progress , the details of Aβ aggregation in vivo , the structure of toxic aggregates , the mechanism of toxicity , and in particular , the relationship between aggregate formation and peptide clearance are not known . We set out to investigate a novel approach to study the dynamics of Aβ aggregation in vitro and neurotoxicity or degradation in vivo by using a conformation-specific Aβ binding protein , the ZAβ3 Affibody [15] , [16] . Affibody molecules are engineered binding proteins , which are selected by phage display from libraries based on the three-helix Z domain [17] , [18] . The ZAβ3 Affibody was selected [15] to bind specifically to Aβ monomers with nanomolar affinity ( dissociation constant Kd≈17 nM ) [16] . It forms a disulfide-linked dimer to which Aβ binds and folds by induced fit [19] into a hairpin conformation such that its two aggregation-prone hydrophobic faces become buried within a tunnel-like cavity in the ZAβ3 dimer [16] , [19] . The specificity and well-characterized structural features of ZAβ3 binding to Aβ make it an ideal candidate for studying the effects of Aβ monomer binding in vivo . We find that the presence of the Affibody molecule , achieved by co-expression , can eliminate Aβ neurotoxicity in a fruit fly ( Drosophila melanogaster ) model of AD [20] , [21] , and we used biochemical and biophysical experiments to identify the molecular mechanism by which this process occurs .
We first generated Drosophila strains transgenic for ZAβ3 . As ZAβ3 is most effective in binding Aβ when it is in its dimeric form , we also generated Drosophila in which two copies of ZAβ3 are connected head-to-tail— ( ZAβ3 ) 2—to enable the disulfide-linked dimer to form more readily . Drosophila transgenic for the wild-type Z domain were used as controls . These three Affibody fly lines were then each crossed with Drosophila transgenic for Aβ42 , Aβ42e22g [22] , or Aβ40 , and the co-expression of both transgenes together in the brain or in the eye was initiated by crossing with appropriate driver flies [20] , [21] . Expression of Aβ42e22g in the brain of Drosophila causes rapid neurodegeneration resulting in a drastic reduction in lifespan from 38 ( ±1 . 8 ) to 9 ( ±0 . 5 ) days , consistent with the findings of previous studies [8] . Co-expression of ZAβ3 with Aβ42e22g , however , increases the lifespan to 20 ( ±0 . 2 ) days . Strikingly , if the-head-to-tail dimer ( ZAβ3 ) 2 is co-expressed with Aβ42e22g , the toxic effects of the peptide are yet further reduced and the lifespan increases to 31 ( ±0 . 8 ) days , which is almost as long as in wild-type controls ( Figure 1A , Table S1 ) and indicates that the neurotoxicity of Aβ has been almost entirely abolished . Co-expression of the Z domain , which has no affinity for Aβ , does not affect Aβ42e22g toxicity , demonstrating that the rescue of Aβ toxicity in vivo is specific to ZAβ3 . Co-expression of ZAβ3 with wild-type Aβ42 also significantly prolongs the lifespan of these flies ( from 28 , ±0 . 4 , to 32 , ±0 . 7 , days ) . Again , the ( ZAβ3 ) 2 head-to-tail-dimer is even more effective , completely eliminating the toxicity associated with Aβ42 ( lifespan 40 , ±1 . 2 , days ) , whereas the Z domain control has no effect ( Figure 1B ) . Expression of the less aggregation-prone Aβ40 has no effect on lifespan , and none of the Affibody molecules or the control significantly affected the lifespan of flies expressing Aβ40 or wild-type Drosophila ( Figure 1C and 1D ) . The ability of ( ZAβ3 ) 2 to abolish the toxic effects of Aβ42e22g was confirmed physiologically by its ability to abolish the abnormal eye morphology associated with Aβ42e22g expression in the photoreceptors in the fly ( Figure 2 ) . To determine the mechanism by which ZAβ3 mediates suppression of Aβ toxicity , we assessed the levels of Aβ42 in the brains of flies co-expressing Aβ42e22g and either ZAβ3 , ( ZAβ3 ) 2 , or the Z domain by Western blotting . Fly brains were homogenized in 1% SDS , subjected to electrophoretic separation , and probed using an antibody against the N-terminus of Aβ , which detailed structural studies reveal remains exposed in the Aβ:ZAβ3 complex [16] . SDS soluble Aβ can clearly be detected in flies expressing Aβ42e22g , but it is absent in flies co-expressing ZAβ3 or ( ZAβ3 ) 2 ( Figure 3A ) . The specificity of this effect is confirmed by the continued presence of the Aβ42e22g in flies that co-express the non-binding Z domain . The ZAβ3:Aβ complex is stable in 1% SDS ( B . Macao , unpublished ) , and Aβ remaining in complexes or in SDS insoluble aggregates in the fly brain might therefore not be detectable by Western blot . In order to address this possibility , fly brains expressing Aβ42e22g with or without ( ZAβ3 ) 2 , ZAβ3 or the Z domain were homogenized in 5 M GdmCl , conditions known to dissociate both Aβ aggregates and Aβ:ZAβ3 complexes . The total level of Aβ42e22g in these extracts was then measured by a sensitive ELISA assay ( Figure 3B ) . Flies expressing both ( ZAβ3 ) 2 and Aβ42e22g show a 97% ( ±3% ) reduction in the concentration of Aβ42e22g compared to flies co-expressing Aβ42e22g and the inert Z domain ( the most appropriate control for the non-specific effects of expressing a second transgene on the levels of Aβ ) . Decreased Aβ42e22g levels in the presence of different Affibody constructs correlate well with corresponding reduction in neurotoxicity measured by the survival assay ( Figure 1 ) . The prevention of Aβ42e22g aggregation by ZAβ3 and ( ZAβ3 ) 2 is demonstrated by immunohistochemical detection of Aβ42e22g in whole mount brain preparations analyzed by confocal microscopy . Flies expressing Aβ42e22g under the control of the OK107-Gal4 driver , which drives expression in a subset of adult neurons , contain abundant deposits in the brain recognized by the anti-Aβ 6E10 antibody , whereas flies co-expressing Aβ42e22g and ( ZAβ3 ) 2 have almost no visible 6E10 immunoreactive deposits ( Figure 3C ) . In good agreement with the results of the ELISA analysis , co-expression of ZAβ3 results in a significant reduction in the burden of aggregates but does not result in their complete removal , whereas co-expression of the Z domain gives levels of Aβ deposits similar to those present in flies expressing Aβ42e22g . In order to determine whether the presence of Aβ42e22g had altered the levels of ZAβ3 or ( ZAβ3 ) 2 present in the fly brain , brain homogenates were analyzed using either anti-cMyc antibodies to detect ZAβ3 or anti-Affibody antibodies to detect ( ZAβ3 ) 2; both dimeric Affibody molecules can be observed as 12 kDa dimers under non-reducing conditions . The levels of these Affibody species are not detectably altered in flies co-expressing Aβ42e22g ( Figure 3D ) despite the marked reduction of the levels of soluble Aβ42e22g ( Figure 3A ) . While this experiment suggests that Aβ clearance could be occurring without the corresponding clearance of its binding partner ZAβ3 , the quantities seen by Western blot represent the equilibrium levels of these two proteins , and so would not detect any turnover in ZAβ3 that may also be occurring . We have established that the reductions in the levels of Aβ42e22g peptide in the fly brain are not due to altered gene regulation in flies co-expressing Z , ZAβ3 , or ( ZAβ3 ) 2 , because the levels of Aβ42e22g transcription are not significantly reduced in any case ( Figure 3E ) . In summary , ZAβ3 causes a reduction in Aβ42e22g levels by actively promoting its clearance from the brain . The clearance does not involve any specific antibody-mediated process , since Drosophila lacks an adaptive immune system [23] . In order to determine at which stages of the Aβ aggregation process the ZAβ3 Affibody can intervene , we analyzed the effects of ZAβ3 on the dynamic interconversion of monomeric , oligomeric , and fibrillar Aβ species in vitro . Sequestration of the hydrophobic regions of Aβ40 and Aβ42 ( Figure 4A and Figure S1 ) allows ZAβ3 to inhibit amyloid fibril formation completely , even that of the extremely aggregation-prone Aβ42e22g variant , as judged by thioflavin T ( ThT ) fluorescence assays indicative of amyloid fibril formation ( Figure 4B–D , Figure S2 , and Figure S3 ) . The addition of ZAβ3 to Aβ40 or Aβ42 aggregation reactions has the same effect on the aggregation kinetics as reducing the Aβ concentration by the equivalent amount ( Figure 4C and Figure S3A ) , demonstrating that inhibition of fibril formation occurs by sequestration of monomeric Aβ . When a molar excess of ZAβ3 is added at different times during the aggregation process , it effectively inhibits all further aggregation ( Figure 4B and Figure S3B ) , indicating that not only does ZAβ3 effectively block aggregation even after its initiation , but also that monomeric Aβ is accessible for binding throughout the process of fibril formation . We noted , however , during the course of the experiments that the ThT fluorescence signal tends to fall after the addition of ZAβ3 at advanced stages of the fibril formation reaction , suggesting that ZAβ3 may also act to reverse the aggregation process ( Figure 4D and Figure S3C ) . To determine the kinetics of fibril dissolution by ZAβ3 in vitro , we set up experiments in which Aβ40 monomers dissociating from pre-formed fibrils are captured by ZAβ3 . We used 15N-labelled Aβ40 for these experiments so that monomeric Aβ40 in complex with ZAβ3 could be identified by solution nuclear magnetic resonance ( NMR ) spectroscopy at low micromolar concentrations . The large fibrillar aggregates of 15N-Aβ40 ( Figure 4E ) did not generate an observable NMR spectrum even after 24 h of data collection , as expected , due to slow molecular tumbling and no highly mobile residues . The addition of ZAβ3 , however , generated resonances from ZAβ3-bound monomeric 15N-Aβ40 , indicating a gradual dissolution of the fibrils ( Figure 4F and Figure S4 ) . Only a small fraction of the Aβ40 , however , dissociates from the fibrils over the first three weeks; thereafter the dissolution process becomes very slow , even for fibrils fragmented by sonication ( Figure 4G ) . Still , under these conditions the observed level of dissolution does not represent the equilibrium state , as the pre-formed Aβ40:ZAβ3 complex is stable in the presence of Aβ40 fibrils ( Figure S5 ) . Hence , even though binding of the ZAβ3 Affibody to monomeric Aβ40 can act to dissolve fibrils , the dissociation kinetics are too slow , at least in vitro , for dissolution to be achievable in practice under ambient conditions . In order to determine the critical issue of whether or not ZAβ3 can prevent the formation of smaller Aβ aggregates ( oligomers ) , we examined their formation in vitro by size exclusion chromatography ( SEC ) in the presence and absence of ZAβ3 ( Figure 5A to 5D and Figure S6 ) . Oligomeric species [24] appear within hours in solutions of Aβ42 , prepared by dilution from alkaline conditions [25] , where the monomeric species is initially dominant . The partitioning between monomeric and oligomeric Aβ then reaches an interim steady state after ∼10 h before the onset of the formation of amyloid fibrils ( Figure 5A ) . By contrast , in the presence of the ZAβ3 , oligomer formation is completely inhibited ( Figure 5B ) , a result that can be attributed to the sequestration of Aβ42 within the complex formed with the Affibody . Isolated Aβ42 oligomers contain elements of well-defined β-sheet structure as measured by circular dichroism ( CD ) , but the β-sheet content is lower than in mature fibrils ( Figure 5E ) . Their stability is also lower as isolated oligomers dissociate into monomers and convert into amyloid fibrils ( Figure 5C ) . Addition of the ZAβ3 Affibody results in dissolution of the oligomers after a few days ( Figure 5D , 5F , and 5G and Figure S7 ) . This is because binding of monomeric Aβ acts to shift the dynamic monomer-oligomer equilibrium such that the oligomer population is reduced , and NMR ( Figure 5H ) and SEC analyses ( Figure S6 ) consequently also reveal monomeric Aβ42 in complex with ZAβ3 . The presence of the ZAβ3 Affibody in vivo results in the effective inhibition of Aβ toxicity and the promotion of Aβ degradation . These effects can be attributed to the ability of the ZAβ3 Affibody to act in three distinct ways on the Aβ aggregation process . First , monomeric Aβ will be sequestered by ZAβ3 , the result of which is that toxic Aβ aggregates will not be able to form in the brain . Second , if Aβ aggregation were to occur , it can be slowed , halted , and even reversed by the action of ZAβ3 on the dynamic Aβ monomer-aggregate equilibria . Furthermore , the presence of ZAβ3 not only prevents or reverses Aβ aggregate formation , it also promotes clearance from the brain . We envisage that this could occur either by intracellular lysosomal or proteasomal degradation , or alternatively by the secretion and uptake by phagocytic cells of the ZAβ3:Aβ complex . The results furthermore demonstrate how engineered binding proteins , such as Affibody molecules , that target specific protein conformations can be used to gain important insights into the dynamics of the Aβ aggregation process and its toxic consequences both in vivo and in vitro .
Drosophila melanogaster transgenic for Aβ40 , Aβ42 , and Aβ42e22g have been described previously [20] . Drosophila transgenic for the Z domain , ZAβ3 , and the ( ZAβ3 ) 2 head-to-tail dimer were created by standard p element mediated germ line transformation using pUAST ( Brand and Perrimon ) as the expression vector . Affibody cDNA was inserted into the multiple cloning site of pUAST using EcoR1 and Xho1 , except for ( ZAβ3 ) 2 , which was cloned between EcoR1 and Xba1 sites . Each transgene was preceded by the same secretion signal peptide ( MASKVSILLLLTVHLLAAQTFAQ ) , derived from the Drosophila necrotic gene , in order to target its expression to the secretory pathway . Transgenes were injected into w1118 embryos . Drosophila transgenic for Aβ40 , Aβ42 , and Aβ42e22g were each crossed with Drosophila transgenic for Z , ZAβ3 , and ( ZAβ3 ) 2 to create stable double transgenic stocks . Expression of the transgenes was achieved using the UAS-Gal4 system . UAS-Tg flies were crossed with flies expressing Gal4 under the control of either a neuronal promoter ( elavc155 or OK107 ) or eye specific promoter ( gmr ) . All fly crosses were maintained on standard cornmeal/agar fly food in humidified incubators . Crosses to generate flies expressing Affibody molecules or Aβ were performed at 29°C . Survival assays were performed as described previously [20] . Briefly , 100 flies of each genotype were collected , divided into tubes of 10 flies , and kept at 29°C . The number of live flies was counted every 2–3 days and recorded . Survival curves were calculated using the Kaplan-Meier method , and differences between genotypes were assessed using the log-rank test . Transgenes were expressed in the eye by crossing with gmr-Gal4 flies . Crosses were performed at 29°C . Flies were collected on the day of eclosion and sputter coated using 20 nM of Au/Pd in a Polaron E5000 . SEM images were collected using a Philips XL30 Microscope . Fifty flies were snap frozen in liquid nitrogen and decapitated for each genotype . Fly heads were homogenized in PBS/1% SDS containing protease inhibitors ( Complete , Roche Applied Science , UK ) . Homogenates were then centrifuged at 12 , 100 g for 1 min to remove insoluble material , and the supernatants were collected for analysis . Protein concentration in each supernatant was determined using the DC Protein Assay ( Biorad ) . Equal quantities of protein for each genotype were loaded on to 4%–12% Bis/Tris SDS PAGE gels ( Invitrogen ) for detection of Affibody molecules and 4%–12% Tris/glycine SDS PAGE gels ( Invitrogen ) for detection of Aβ . Electrophoresis was performed under non-reducing conditions , and protein was transferred to nitrocellulose membranes for Western blotting . ZAβ3 was detected using a mouse monoclonal anti-c-Myc antibody ( clone 9E10 , Abcam ) , and ( ZAβ3 ) 2 was detected using a goat anti-Affibody antibody ( Abcam ) . Aβ was detected using a mouse monoclonal anti-Aβ antibody directed against the N terminus of Aβ ( 6E10 , Signet ) . All blots were developed using Supersignal West Femto Maximum Sensitivity ECL substrate ( Pierce ) . Heads from flies expressing Aβ42e22g with or without Affibody domains were subjected to mechanical homogenization in 5 M GdmCl , 50 mM Hepes , and 5 mM EDTA followed by 4 min of sonication in a water bath . Homogenates were centrifuged for 7 min at 12 , 100 g to pellet any GdmCl insoluble material . Supernatants were diluted in 50 mM Hepes and 5 mM EDTA with protease inhibitors to a final concentration of 1 M GdmCl . A sandwich ELISA was performed on the supernatants using biotinylated 6E10 ( Signet ) and a C terminal Aβx-42-specific antibody 21F12 ( kind gift of D . Schenk , Elan ) . Protein levels were measured using a Sector Imager ( Meso Scale Discovery ) and normalized to a percentage of the level obtained for flies expressing Aβ42e22g alone . Flies of all genotypes were crossed with OK107-Gal4 flies ( Bloomington Stock No . 854 ) to drive expression in a subset of neurons that includes , but is not limited to , the mushroom bodies . For each genotype fly brains were dissected in PBS with 0 . 05% Triton X-100 and fixed in 4% paraformaldehyde for 1 h at room temperature . The brains were then washed three times in PBS/0 . 05% Triton X-100 and blocked in 5% w/v bovine serum albumin in PBS for 1 h at room temperature . Fly brains were incubated overnight in mouse anti-Aβ ( 6E10 , Signet ) diluted 1∶1000 in blocking buffer . After three further washes in PBS/0 . 05% Triton X-100 , brains were then incubated in goat anti-mouse IgG Alexa 546 ( Invitrogen ) and counterstained with TOTO-3 ( Invitrogen ) to detect nuclei before mounting in Vectashield ( Vectorlabs ) anti-fade mounting medium . Confocal serial scanning images were acquired at 2 or 4 µm intervals ( for high magnification and low magnification images , respectively ) using a Nikon Eclipse C1si on Nikon E90i upright stand ( Nikon ) . The image stacks were projected using ImageJ ( version 1 . 42k ) , and the resulting composite images were processed using Photoshop CS4 software ( Adobe Systems ) . Concentrations of mRNA were determined using quantitative real time PCR ( RT-PCR ) . Twenty-five flies per genotype were collected and snap frozen in liquid N2 . RNA was extracted from each group of 25 fly heads using TriZol followed by DNAse treatment to remove residual genomic DNA and reverse transcription to produce cDNA . Each sample was subjected to two separate quantitative PCR reactions to detect Aβ mRNA and the control gene Actin5c . Real time amplification of cDNA was monitored using SYBR Green fluorescence in a Bio-Rad iQ Cycler . ZAβ3 was produced in Escherichia coli and purified as described elsewhere [16] . Aβ peptides were obtained from a commercial source ( rpeptide , Bogart , GA , USA ) , synthesized in-house , or produced ( with an N-terminal methionine ) by recombinant co-expression of Aβ and ZAβ3 in E . coli [26] . Experiments were carried out in 20 mM sodium phosphate , 50 mM NaCl , except for the NMR experiments where NaCl was not included , and pH 7 . 2 . 10 µM ThT was added prior to fluorescence measurements . Fibril formation assays were carried out as described previously [16] . TEM images were obtained using a LEO 912 AB Omega microscope . CD spectra were recorded on a JASCO J-810 spectropolarimeter . Fibrils were prepared from Aβ40 at a concentration of 100 µM with the same set-up and conditions as for the fibril formation assays , but in the absence of ThT . After 3 days of incubation at 37°C , fibrils were isolated by centrifugation at 16 , 000 g . To remove any residual soluble peptide , fibrils were washed by resuspension in buffer F [20 mM sodium phosphate , pH 7 . 2 , 0 . 1% sodium azide , complete protease inhibitor ( Roche; at the concentration recommended by the manufacturer ) ] , followed by centrifugation . Fibrils were resuspended in buffer F supplemented with 10% D2O to a final concentration of 300 µM Aβ40 and investigated by 15N HSQC NMR with 24 h of data collection on a Varian Inova 900 MHz NMR spectrometer ( equipped with a cryogenic probe ) or on a Varian Inova 800 MHz spectrometer . The intensity of resonances originating from bound Aβ40 detected in the presence of 325 µM of unlabeled ZAβ3 was followed over time by recording a series of 24 h 15N HSQC NMR spectra . Five µM of 15N-ZAβ3 served as an internal concentration reference , assuming identical NMR-sensitivities of the intense resonances of the three C-terminal residues of bound Aβ40 and free ZAβ3 . Sonication was achieved by placing the NMR tube with the fibril sample into a Misonix water bath sonicator for 2 min before acquisition of NMR data . Oligomer formation was induced by adjusting the pH of alkaline ( pH∼10 . 5 ) solutions of Aβ42 ( concentration ≤100 µM ) in 20 mM sodium phosphate and 50 mM sodium chloride to pH 7 . 2 ( with 1 M HCl ) [25] . The samples were incubated at 21°C and oligomer formation was monitored with SEC and ThT fluorescence . Fifty µl ( for analytical runs ) or 1 ml ( for preparative oligomer isolation ) aliquots were injected onto an ÄKTA Explorer system ( GE Healthcare , Uppsala , Sweden ) equipped with a Superdex 75 10/300 column , and the elution was monitored by UV absorbance at 220 nm . Preparative oligomer isolation was carried out 4–20 h after induction of oligomer formation and yielded oligomer solutions at 10–20 µM total Aβ42 concentration . The elution volumes of the ZAβ3:Aβ42 complex and free ZAβ3 were determined in separate runs of the isolated complex or free Affibody , respectively , and conformed to previous SEC studies [19] . The amounts of Aβ42 in the monomeric , oligomeric , or ZAβ3-bound fraction were determined from the elution peak areas obtained by integration using the Unicorn software provided with the chromatography system . The data were normalized by setting to unity the sum of the oligomer and monomer peak areas in the first SEC profiles ( at t = 0 . 2 h for oligomer formation in Figure S6A , and at t = 0 . 5 h for oligomer dissolution in Figure S6C ) . The fraction of high molecular weight aggregates that did not enter the column bed was calculated as the difference between unity and the sum of the monomer and oligomer fractions . The fraction of ZAβ3-bound Aβ42 shown in Figure 5D was obtained by comparison of the integrated ZAβ3:Aβ42/free ZAβ3 peak area with those obtained in calibration runs of free ZAβ3 ( set to 0 ) and ZAβ3:Aβ42 complex ( set to 1 ) using the same protein concentrations as in the dissolution experiment . The fraction of Aβ42 bound to ZAβ3 was determined by 15N HSQC NMR employing an internal concentration standard . | Alzheimer's disease is thought to be a result of neuronal damage caused by toxic aggregated forms of the Aβ peptide in the brain . There is no cure and existing treatments are ineffective in reversing or preventing disease progression . Here we describe a novel strategy that makes use of an engineered “Affibody” protein to study the disease and potentially combat its underlying causes . The Affibody occludes the aggregation-prone regions of Aβ peptides , preventing their aggregation into toxic forms , and it also acts to dissolve pre-formed Aβ aggregates . It is functional in vivo , as its co-expression with Aβ peptides in transgenic fruit flies prevents the neuronal damage and premature death that result from expression of Aβ peptides alone . Moreover , we show that the origin of this protection is the enhanced clearance of Aβ peptides from the brain . These findings open up new opportunities for using engineered binding proteins to probe the origins of Alzheimer's disease and potentially to develop a new class of therapeutic agents . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"chemical",
"biology/protein",
"chemistry",
"and",
"proteomics",
"biochemistry/protein",
"folding",
"biochemistry/protein",
"chemistry",
"biotechnology/protein",
"chemistry",
"and",
"proteomics"
]
| 2010 | Sequestration of the Aβ Peptide Prevents Toxicity and Promotes Degradation In Vivo |
Whole-chromosome imbalances affect over half of early human embryos and are the leading cause of pregnancy loss . While these errors frequently arise in oocyte meiosis , many such whole-chromosome abnormalities affecting cleavage-stage embryos are the result of chromosome missegregation occurring during the initial mitotic cell divisions . The first wave of zygotic genome activation at the 4–8 cell stage results in the arrest of a large proportion of embryos , the vast majority of which contain whole-chromosome abnormalities . Thus , the full spectrum of meiotic and mitotic errors can only be detected by sampling after the initial cell divisions , but prior to this selective filter . Here , we apply 24-chromosome preimplantation genetic screening ( PGS ) to 28 , 052 single-cell day-3 blastomere biopsies and 18 , 387 multi-cell day-5 trophectoderm biopsies from 6 , 366 in vitro fertilization ( IVF ) cycles . We precisely characterize the rates and patterns of whole-chromosome abnormalities at each developmental stage and distinguish errors of meiotic and mitotic origin without embryo disaggregation , based on informative chromosomal signatures . We show that mitotic errors frequently involve multiple chromosome losses that are not biased toward maternal or paternal homologs . This outcome is characteristic of spindle abnormalities and chaotic cell division detected in previous studies . In contrast to meiotic errors , our data also show that mitotic errors are not significantly associated with maternal age . PGS patients referred due to previous IVF failure had elevated rates of mitotic error , while patients referred due to recurrent pregnancy loss had elevated rates of meiotic error , controlling for maternal age . These results support the conclusion that mitotic error is the predominant mechanism contributing to pregnancy losses occurring prior to blastocyst formation . This high-resolution view of the full spectrum of whole-chromosome abnormalities affecting early embryos provides insight into the cytogenetic mechanisms underlying their formation and the consequences for human fertility .
Human reproduction is inefficient , with pregnancy loss estimated to occur in approximately 70% of all conceptions [1] . The majority of pregnancy losses take place before 12 weeks of gestation [2] and are mostly explained by whole-chromosome abnormalities [3 , 4] as well as structural aberrations [5] . Our study focuses on numerical abnormalities affecting whole chromosomes , the detection of which has been extensively validated [6] . These errors can be broadly classified as polyploidy ( non-diploid multiples of the haploid chromosome set ) and aneuploidy ( other configurations of extra or missing chromosomes ) . Given the strong implications for fertility , a clear understanding of the rates and molecular mechanisms contributing to various classes of whole-chromosome abnormalities is an important goal in reproductive medicine and human biology in general . It has long been established that incidence of aneuploidy affecting maternal chromosome copies increases with maternal age [7] . This pattern is driven mostly by errors occurring during maternal meiosis , which arrests at the diplotene stage until it resumes at ovulation many years later [3] . These meiotic errors were at first thought to arise primarily via whole-chromosome non-disjunction—the failure of homologous chromosomes or sister chromatids to separate [3] . Later work , however , demonstrated a greater role of unbalanced predivision—the premature separation and subsequent missegregation of sister chromatids [8]—in contributing to maternal age-related meiotic error and implicated breakdown of cohesin proteins as a possible mechanism [9–11] . Both non-disjunction and unbalanced chromatid predivision result in a chromosome gain in one daughter cell with a corresponding chromosome loss in the other daughter cell , but can be distinguished when both oocytes or embryos and their corresponding polar bodies are analyzed . A recent study used this approach to confirm the preponderance of unbalanced chromatid predivision and also identify a non-canonical segregation pattern whereby sister chromatids separate at the meiosis I ( MI ) , followed by non-random segregation at meiosis II ( MII ) favoring separation of homologous chromosomes [12] . In addition to meiotic errors , mitotic errors are extremely common during the initial post-zygotic cell divisions and produce mosaic embryos containing multiple distinct karyotypes [13] . It has been estimated that a large proportion of embryos tested during in vitro fertilization ( IVF ) are mosaics [6 , 14–18] , though the incidence of mosaicism varies widely depending on embryonic stage investigated and method of analysis [13] . This high rate of mitotic error is presumably due to relaxed cell cycle control during the initial embryonic cell divisions , but which is reestablished prior to blastocyst formation [19] . While a subset of mosaic embryos may survive due to self-correction [20] , a substantial proportion of mosaic embryos are inviable and arrest before the blastocyst stage [21–23] . Despite these apparent consequences for embryonic survival , rates of mitotic error are yet to be reported for individual chromosomes based on a large survey of early embryos . The greatest source of information about chromosome abnormalities is preimplantation genetic screening ( PGS ) conducted during IVF , whereby cells are biopsied from day-3 or day-5 embryos and the copy number of one or more chromosomes is determined . Embryos that test euploid are then recommended for transfer to improve the rate of implantation and live birth in IVF . While early versions of PGS employed fluorescence in situ hybridization ( FISH ) and were thus limited to testing only a few chromosomes at a time , microarray-based approaches are capable of assaying ploidy status of all chromosomes simultaneously . One powerful approach , termed array comparative genomic hybridization ( aCGH ) , measures copy number aberrations by contrasting relative signal intensities of test and reference samples on a DNA microarray [24] . By combining this approach with single nucleotide polymorphism ( SNP ) profiling , more recent microarray-based technologies can differentiate between maternal and paternal homologs , thus shedding additional light on the parental origin and timing of chromosome missegregation . A recent study applied 24-chromosome SNP-microarray PGS to 15 , 169 trophectoderm ( TE ) biopsies from day-5 embryos , documenting ploidy at the blastocyst stage with extreme precision [25] . Our data demonstrate , however , that day-5 embryos contain a biased subset of whole-chromosome abnormalities that have already been filtered by self-correction and selection at the onset of zygotic genome activation at the 4–8 cell stage . In fact , embryonic arrest before day 5 can be responsible for the loss of more than half of IVF embryos , the vast majority of which are non-euploid [26] . Indeed , screening of blastocyst biopsies is preferable in the context of IVF , in part because survival to day 5 is an indicator of developmental competence [27] . Thus , the full spectrum of whole-chromosome abnormalities is only observable at earlier developmental time points , motivating a comparably large study of cleavage-stage embryos with 24-chromosome SNP-based PGS . Here , we present 24-chromosome PGS results from 28 , 052 individual day-3 blastomeres and 18 , 387 multi-cell day-5 TE biopsies collected from a total of 6 , 366 IVF cycles , characterizing frequencies of both common and rare ploidy states and contrasting those detected at each sampling point . These data suggest that embryos purged in early development often experienced catastrophic mitotic errors , while meiotic errors—which tend to result in minor aneuploidy or polyploidy—are comparatively viable through blastocyst formation . Consistent with this interpretation , we show that patients referred for PGS due to repeat IVF failure had higher rates of mitotic error than patients with other clinical indications , suggesting that some of these patients suffer systematically higher rates of preclinical pregnancy loss due to mitotic aberrations . By inferring ploidy status of all chromosome pairs and distinguishing parental origin of the affected homologs , we achieve a high-resolution view of whole-chromosome abnormalities that provides key insights into the characteristics of chromosome segregation and the impacts on early development .
PGS was conducted for a total of 6 , 366 anonymous IVF cases referred to Natera between February 2009 and March 2014 by a total of 181 IVF centers . The mean maternal age was 36 . 3 ( including egg donors ) and the mean paternal age was 40 . 3 ( Fig 1 ) . Of these cases , a total of 890 cases utilized egg donors , among whom the mean age was 26 . 3 ( Fig 1 ) . The data are composed of PGS results from 28 , 052 individual day-3 blastomeres and 18 , 387 5–10 cell day-5 TE biopsies . Patients submitted means of 9 . 6 blastomere biopsies or 5 . 1 TE biopsies per case , with the number of submitted samples of both types declining with increasing maternal age ( day-3 blastomeres: β = −0 . 0232 , SE = 0 . 00203 , P < 1 × 10−10; day-5 TE biopsies: β = −0 . 0257 , SE = 0 . 00251 , P < 1 × 10−10; Fig 2 ) . A total of 3 , 767 cases ( 59 . 2% ) reported the reason for their referral , with advanced maternal age , recurrent pregnancy loss , and gender selection constituting the most common indications ( Table 1 ) . For 40 cases , no reliable ploidy calls could be made for any embryo , limiting our analysis to the remaining 6 , 326 cycles . Of 46 , 449 total samples , DNA was not detected for 2 , 653 samples ( 5 . 7% ) . These samples were therefore excluded from all subsequent analyses . An additional 1 , 071 samples had low-confidence calls ( < 80% confidence ) for greater than four chromosomes and were also excluded for quality-control purposes . All other low-confidence calls were masked and considered as missing data . At least one error affecting a whole chromosome was detected for total of 15 , 842 ( 62 . 1% ) of the remaining 25 , 497 blastomeres as compared to 7 , 623 ( 44 . 3% ) of 17 , 219 TE biopsies , a highly significant difference [χ2 ( 1 , N = 42 , 716 ) = 1323 . 8 , P < 1 × 10−10] . For an additional 323 blastomeres and 146 TE biopsies , only segmental deletions or duplications were detected . We note that in the face of mosaicism , which is common in cleavage-stage embryos , ploidy of individual blastomeres will not necessarily reflect the ploidy status of the entire embryo , but rather provides a snapshot of a single cell at this early developmental stage . An additional caveat of our analysis is that mosaicism within TE biopsies may not be detectable if a small proportion of cells are affected . Ploidy inference was conducted using the Parental Support algorithm [6] , which does not explicitly test for mosaicism , but uses a Bayesian methodology to calculate likelihoods of ploidy hypotheses ( assuming the absence of mosaicism ) given the data . This limitation must therefore be considered when comparing results between sample types . Many studies have demonstrated that the incidence of aneuploidy increases with maternal age , starting in the mid-thirties , driven primarily by errors of maternal meiotic origin [3 , 28 , 29] . Our data replicate these results , with a significant increase in per-case proportion of samples with whole-chromosome abnormalities with increasing maternal age ( Fig 3 ) . This relationship is best fit by a third-order polynomial model in the case of the blastomere biopsies ( S1 Table; McFadden’s pseudo-R2 = 0 . 270 ) , and a second-order polynomial in the case of the TE biopsies ( S1 Table; McFadden’s pseudo-R2 = 0 . 189 ) . The model provided poor fit for TE biopsy patients in the upper tail of the age distribution , who had strikingly lower rates of whole-chromosome abnormalities than expected . While the sample size is limited in this upper tail , the consistently lower rates of abnormalities in day-5 embryos from mothers > 45 years old suggest a potential selection bias . The small subset of patients in this age group who are capable of producing embryos surviving to day 5 may have systematically lower rates of meiotic or mitotic error , a hypothesis that merits future investigation . Given the well-established association between fertility and embryonic euploidy , it should be noted that specific rates of meiotic and mitotic error reported in this study are likely particular to the IVF population . Previous studies also demonstrated that ovarian stimulation and IVF culture conditions can both influence rates of chromosome abnormalities [30] . Nevertheless , high rates of meiotic and mitotic error have been observed even for unstimulated cycles and for patients without obstetrical or gynecological pathologies [31 , 32] , suggesting that basic insights provided by this study can be generalized to better understand natural human fertility . In support of this conclusion , we detected no significant difference in rates of whole-chromosome abnormalities among day-3 blastomeres [F ( 1 , 2647 ) = 1 . 693 , P = 0 . 193] or day-5 TE biopsies [F ( 1 , 3165 ) = 0 . 615 , P = 0 . 433] in embryos from fertile egg donors and non-donor IVF patients after accounting for maternal age effects ( S2B Fig ) . The pattern of association between whole-chromosome abnormalities and maternal age was strongly chromosome-specific ( S1 Fig ) , as has been reported based on smaller samples from different developmental time points [33] . The maternal age effect on aneuploidy is considered the primary reason for the corresponding age-associated decline in female fertility , both in the contexts of natural conception and IVF . Our data are consistent with this interpretation , as the relationship between maternal age and the rate of whole-chromosome abnormalities closely mirrored the relationship between maternal age and various measures of IVF success in public data obtained from the 2011 CDC National Summary Report [34] ( S2A Fig ) . The question of whether risk for whole-chromosome abnormalities is affected by paternal age is contentious and is complicated by the fact that maternal and paternal ages are often highly correlated [35] . This was also the case in our study , with strong correlation between parental ages ( Fig 1; r = 0 . 334 , P < 1 × 10−10 ) , especially when egg donors were excluded ( Fig 1; r = 0 . 536 , P < 1 × 10−10 ) driving a strong relationship between whole-chromosome abnormalities and paternal age in blastomeres ( β = 0 . 0268 , SE = 0 . 00288 , P < 1 × 10−10 ) and TE biopsies ( β = 0 . 0449 , SE = 0 . 00342 , P < 1 × 10−10 ) . When limiting analysis to egg donor cases , among which maternal and paternal ages were not correlated ( r = −0 . 0672 , P = 0 . 204 ) , a marginal association was still detected between the rate of whole-chromosome abnormalities and paternal age in day-3 blastomeres ( β = 0 . 0123 , SE = 0 . 00612 , P = 0 . 0456 ) , but not in day-5 TE biopsies ( β = −0 . 0119 , SE = 0 . 0181 , P = 0 . 514 ) . Intrigued by the potential effect of paternal age , we employed complementary statistical approaches to control for maternal age and test for a residual paternal age effect on risk for whole-chromosome abnormalities . Spearman partial correlation detected a significant association between paternal age and proportion of affected samples upon holding maternal age constant for both variables in day-3 blastomeres ( rxy . z = 0 . 0613 , P = 0 . 00191 ) , but not in day-5 TE biopsies ( rxy . z = 0 . 0167 , P = 0 . 362 ) . As expected , the same approach applied to maternal age upon holding paternal age constant detected a much stronger association in both sample types ( day-3 blastomeres: rxz . y = 0 . 430 , P < 1 × 10−10; day-5 TE biopsies: rxz . y = 0 . 302 , P < 1 × 10−10 ) . Using an alternative approach , we stratified the sample into two groups: fathers younger than the median paternal age ( 39 . 6 for blastomere cases and 38 . 9 for TE biopsy cases ) and fathers equal to or older than the median paternal age . We then matched cases sampled from each paternal age group on maternal age ( within 0 . 1 standard deviations of the maternal age ) , dropping unmatched cases and randomly breaking ties . Controlling for maternal age in this manner , we found that increased paternal age was again marginally associated with an increased rate of whole-chromosome errors in day-3 blastomere biopsies [χ2 ( 1 , N = 12 , 263 ) = 4 . 980 , P = 0 . 0256] but not in day-5 TE biopsies [χ2 ( 1 , N = 8055 ) = 0 . 028 , P = 0 . 857] . While significant , the blastomere biopsy effect was small , with 61 . 7% affected blastomeres for fathers less than 39 . 6 years old versus 63 . 9% for fathers greater than or equal to 39 . 6 years old . In contrast , a logistic GLM including paternal age as a predictor variable did not provide significantly better fit than a reduced model that only accounted for maternal age for either day-3 blastomere biopsies [F ( 1 , 2549 ) = 2 . 169 , P = 0 . 141] or day-5 TE biopsies [F ( 1 , 2970 ) = 0 . 994 , P = 0 . 319] . We next sought to stratify different forms of whole-chromosome abnormalities to better understand the mechanisms underlying their formation . Certain chromosomal signatures are strongly indicative of either meiotic or mitotic error , while other signatures can arise via either process . One signature of chromosome gains that can help identify meiotic error is the presence of three unmatched haplotypes in a given region of the embryo’s genome ( i . e . two non-identical , but homologous chromosomes inherited from a single parent; Fig 4 ) . This signature , which we termed ‘both parental homologs’ ( BPH ) error , is unique to meiosis , and previous literature suggests that it primarily arises due to unbalanced chromatid predivision [8 , 10] . Mitotic errors , as well as meiotic errors in the absence of recombination , produce chromosome gains in which the extra chromosome is identical to another chromosome over its entire length ( Fig 4 ) . We refer to these chromosome gains as ‘single parental homolog’ ( SPH ) errors . This logic was first introduced by Johnson et al . [6] ( also see [36] ) and was recently employed to distinguish meiotic-origin aneuploidies for a genome wide association study of aneuploidy risk [37] . We tabulated the total counts of different forms of whole-chromosome abnormalities , contrasting those observed at day 3 and day 5 ( Table 2; Fig 5 ) . We found that errors affecting few chromosomes ( single trisomies , single monosomies ) were biased in their impact on maternal homologs , supporting a maternal meiotic origin of formation . Consistent with this interpretation , more than half of maternal trisomies carried the BPH signature in both day-3 blastomeres and day-5 TE biopsies ( Table 2 ) . Meanwhile , complex aneuploidies involving multiple chromosomes were approximately balanced in their effect on maternal and paternal homologs ( Fig 5 ) , suggesting that the mechanism underlying these errors is primarily post-zygotic and does not discriminate based on parental origin . Errors involving multiple chromosomes were strongly biased toward chromosome losses over chromosome gains , and were largely depleted by day 5 of development ( Table 2; Fig 5 ) . Triploidy ( and near-triploidy ) primarily affected maternal homologs ( Table 2; Fig 5 ) . We note that this excess of maternal ( digynic ) triploidy compared to paternal ( diandric ) triploidy is at odds with some earlier literature [38] . This discrepancy is explained , however , by the current widespread use of intracytoplasmic sperm injection ( ICSI ) in place of conventional IVF [39] . While most cases of triploidy in conventional IVF are due to dispermic fertilization and produce diandric embryos , these errors are essentially eliminated by ICSI , while the rate of digynic tripronuclear ( 3PN ) zygote formation is estimated at 2 . 5–6 . 2% [40] . These 3PN zygotes are formed when a diploid oocyte—arising via meiotic error—is fertilized by a haploid sperm [41] . The relative occurrences of digynic and diandric triploidies in our data ( Table 2 ) are therefore consistent with the literature , as 80–90% of IVF cases analyzed by Natera utilized ICSI . Haploid ( and near-haploid ) embryos possessing only the maternally inherited genome greatly exceeded those possessing only the paternally inherited genome ( Table 2; Fig 5 ) . This is also consistent with previous data from ICSI and is thought to arise primarily due to gynogenesis , whereby the oocyte is stimulated by a sperm factor , but the sperm chromatin fails to decondense [42] . Complex aneuploidies were non-random in their composition , with co-occurrence of certain forms of aneuploidy being more common than others ( Fig 6 ) . Maternal monosomy and maternal trisomy , the most prevalent forms of aneuploidy , frequently co-occurred within individual embryo biopsies at both day 3 and day 5 . Aneuploidies involving multiple forms of chromosome loss ( maternal monosomy , paternal monosomy , and nullisomy ) were extremely prevalent at day 3 of development , but rare among day-5 biopsies , again suggesting strong selection against this common class of whole-chromosome abnormality ( Fig 6 ) . The diverse forms and distinct characteristics of whole-chromosome abnormalities revealed by these analyses motivated us to separately investigate errors of meiotic and mitotic origin to shed light on the cytogenetic mechanisms that produce these patterns and the consequences for reproduction and early development . Focusing first on meiotic errors , we analyzed rates of maternal and paternal BPH errors , their associations with parents’ ages , and their tendencies toward particular chromosomes . Maternal BPH errors increased dramatically with maternal age , consistent with the interpretation that maternal meiotic errors drive the maternal age association with aneuploidy ( Fig 7 ) . This association was observed for both day-3 blastomere biopsies ( β = 0 . 110 , SE = 0 . 00404 , P < 1 × 10−10 ) and day-5 TE biopsies ( β = 0 . 120 , SE = 0 . 00599 , P < 1 × 10−10 ) . Decreased maternal BPH error was also responsible for the surprisingly low rate of whole-chromosome abnormalities in TE biopsies from patients > 45 years old , described in the previous section ( Figs 7 and 3 ) . Maternal BPH errors did not affect all chromosomes equally , with elevated rates of BPH error of chromosome 16 ( d3: 6 . 40% , d5: 4 . 12% ) , 22 ( d3: 6 . 00% , d5: 4 . 02% ) , 21 ( d3: 5 . 28% , d5: 3 . 04% ) and 15 ( d3: 5 . 24% , d5: 3 . 26% ) ( Fig 8A ) . A high rate of aneuploidy of these chromosomes is consistent with previous studies applying PGS to diverse developmental stages , from oocyte polar bodies [9 , 43 , 44] ( though see [45] for why this might not reflect the status of the embryo ) to products of conception from clinical miscarriages [46–51] . Together , these results support the suggestion these chromosomes are both inherently more susceptible to meiotic error and that meiotic errors are relatively viable into late development . Chromosome-specific rates of maternal BPH trisomy and maternal monosomy were also highly correlated with one another ( r = 0 . 840 , P = 5 . 471 × 10−7 ) , reflecting the fact that many maternal BPH trisomies and maternal monosomies likely share a common origin of unbalanced chromatid predivision . Further supporting this conclusion , chromosomes 16 , 22 , 15 , and 21 , which had the highest rates of maternal BPH trisomy and monosomy , also displayed the strongest increases with maternal age . This increase greatly exceeded the maternal age effects on other chromosomes ( S3 Fig ) , corroborating recent findings by Franasiak et al . [25] . A generalized linear model confirmed the presence of a length-by-age interaction effect on probability of maternal BPH trisomy of particular chromosomes ( β = 2 . 494 × 10−10 , SE = 6 . 423 × 10−11 , P = 1 . 15 × 10−4; S2 Table ) . While chromosome-specific rates of BPH error were lower at day 5 of development , this was almost exclusively due to selection against triploidies and haploidies ( as well as near-triploidies and near-haploidies ) rather than BPH aneuploidies affecting small numbers of chromosomes . Chromosome-specific rates of maternal BPH error at day 3 and day 5 were highly correlated ( Fig 8B; r = 0 . 978 , P < 1 × 10−10 ) , consistent with weak selection against minor meiotic-origin aneuploidies between these stages . Chromosome-specific rates of maternal BPH error were also negatively correlated with chromosome length at both day 3 ( Fig 9; r = −0 . 623 , P = 0 . 00148 ) and day 5 ( Fig 9; r = −0 . 556 , P = 0 . 00586 ) . This correlation was still significant after removing chromosomes 15 , 16 , 21 , and 22 from the analysis ( day-3 blastomeres: r = −0 . 734 , P = 0 . 000346; day-5 TE biopsies: r = −0 . 525 , P = 0 . 0209 ) . Upon excluding co-occurring cases of putative mitotic error , we observed that maternal BPH errors very rarely affected intermediate numbers of chromosomes , instead tending toward few chromosomes ( aneuploidy ) or the entire complement ( polyploidy; Fig 8C ) . In contrast to frequent meiotic errors in the egg , meiotic errors in sperm are rare , with paternal BPH error detected in only 1% of biopsies in our study . No significant association was detected between paternal age and these rare paternal BPH errors in blastomere samples ( Fig 10; Logistic GLM , β = −0 . 00360 , SE = 0 . 00933 , P = 0 . 700 ) , while a significant , but weak , negative association was observed in TE biopsies ( Fig 10; Logistic GLM , β = 0 . 0342 , SE = 0 . 0146 , P = 0 . 0194 ) . These results suggest that residual correlation between maternal and paternal age , rather than increased susceptibility to paternal meiotic error , may have been responsible for the marginal positive association between aneuploidy and paternal age detected in our previous analysis . While the BPH signature can be used to identify chromosome gains of likely meiotic origin , mitotic errors must be identified by separate chromosomal signatures . Gain or loss of at least one paternal chromosome copy is a good indicator of mitotic error , as previous studies have determined that fewer than 5% of sperm are aneuploid [52] and paternal BPH error affected only 1% of samples in our study . This logic was recently used to classify mitotic errors in the same dataset , correlating their occurrence with a maternal effect genetic variant in the region containing the gene PLK4 [37] . In contrast to extensive data supporting a maternal age effect on maternal meiotic-origin aneuploidy , previous studies have reached conflicting conclusions about whether mitotic error is influenced by maternal or paternal age . We observed that the incidence of mitotic error was not associated with maternal age for either day-3 blastomere biopsies ( Fig 11; β = −0 . 00186 , SE = 0 . 00322 , P = 0 . 564 ) or day-5 TE biopsies ( Fig 11; β = 0 . 0119 , SE = 0 . 00616 , P = 0 . 0526 ) . This finding thus contradicts several previous studies [53–55] , but is consistent with several others [56 , 57] . While not statistically significant , the day-5 biopsies show a trend in the positive direction , and a more comprehensive approach to classifying mitotic error may reveal a statistically significant association . Nevertheless , the sample size of our study is so large that any effect , should it exist , would be extremely weak in comparison to the strong maternal age effect on meiotic error . Similarly , no significant association was detected between incidence of mitotic error and paternal age at either day 3 ( Fig 12; β = 0 . 00315 , SE = 0 . 00253 , P = 0 . 213 ) or day 5 ( Fig 12; β = 0 . 000540 , SE = 0 . 00454 , P = 0 . 905 ) . This finding again suggests that the weak paternal age effect on the rate of whole-chromosome abnormalities detected in our previous analysis , may indeed be driven by confounding effects of maternal age on maternal meiotic error . In contrast to meiotic errors , larger chromosomes were more susceptible to mitotic error , with increased mitotic error rates observed for these chromosomes at day 3 ( Figs 13A and 14; r = 0 . 734 , P = 1 . 011 × 10−4 ) , but not at day 5 ( Figs 13A and 14; r = 0 . 351 , P = 0 . 109 ) . Despite a strong depletion of mitotic-origin aneuploidies between days 3 and 5 , chromosome-specific rates of mitotic error were still positively correlated between these developmental stages ( Fig 13B; r = 0 . 762 , P = 2 . 326 × 10−5 ) . Unlike meiotic errors , which tended to affect few chromosomes or the entire complement , mitotic errors frequently affected intermediate numbers of chromosomes in blastomere biopsies , but not in TE biopsies ( Fig 13C ) . Current evidence suggests that post-zygotic mitotic errors primarily arise via mechanisms termed anaphase lag , mitotic non-disjunction , and endoreplication [56 , 58–61] , but the relative frequencies of these mechanisms are the subject of debate [13] . Endoreplication refers to genome duplication without cell division , resulting in binucleate blastomeres and balanced polyploidy that is undetectable with our SNP microarray approach ( and would also be indistinguishable from normal mitotic cells immediately following S-phase ) [61] . Anaphase lag occurs due to failure of chromatids to connect to the mitotic spindle or due to slow migration of chromatids toward the spindle poles , resulting in chromosome loss or incorporation into micronuclei [62] . Mitotic non-disjunction is similar to meiotic non-disjunction and refers to the failure of sister chromatids to separate during mitotic anaphase , resulting in a trisomy in one daughter cell with a corresponding monosomy in the other daughter cell . The counts of mitotic chromosome gains and losses can thus be compared to assess relative frequencies of mitotic error mechanisms . Previous studies have argued that a predominance of chromosome loss compared to chromosome gain can be attributed to a high rate of anaphase lag [60 , 63] . On a per-chromosome basis , we observed that chromosome losses indeed exceeded chromosome gains 54 , 626 to 12 , 907 for samples affected by putative mitotic errors . Our data , however , suggest that these chromosome losses are not primarily due to isolated cases of anaphase lag , but tend to be driven by more catastrophic errors in mitosis . These aneuploidies may be driven by centrosome or mitotic spindle abnormalities [64] and are consistent with chaotic cell division reported in previous studies [59] . The discovery that variation encompassing the gene PLK4 influences mitotic-origin aneuploidy provides one clue that dysregulation of centrosome duplication may be an important factor underlying spindle abnormalities and aneuploidy in cleavage-stage embryos [37] . Centrosome overduplication , as is induced by overexpression of PLK4 , can result in multipolar cell division [65] or centrosome clustering and a high rate of anaphase lag [66] , with both mechanisms having the potential to cause multiple chromosome loss . We next explicitly contrasted the two developmental stages to gain additional insight into selection occurring during preimplantation development . While a plurality of errors affected only one chromosome , greater than 80% of errors in day-3 blastomeres affected two or more chromosomes ( Fig 15A ) . Compared to individual day-3 blastomere samples , fewer complex errors affecting multiple chromosomes were detected in day-5 TE biopsies ( Fig 15A; Table 2 ) . We compared the two sample types by calculating the percent difference in rates of non-euploidy between the blastomere and TE samples , stratifying by the total number of affected chromosomes ( Fig 15B ) . This metric reflects the proportion of embryos that were either lost or self-corrected between the two sampling stages . Due to the design of our study , we could not distinguish between embryonic arrest and self-correction , as blastomere and TE biopsy data were fully independent . Nevertheless , we observed that errors affecting increasing numbers of chromosomes were increasingly depleted in TE biopsies relative to blastomeres , plateauing at approximately 11 chromosomes affected ( Fig 15B ) . This difference became less extreme when greater than 18 chromosomes were affected ( Fig 15B ) , suggesting a slight relative viability of polyploidies compared to complex aneuploidies . Together , these results provide strong evidence of early selection against complex aneuploidy of primarily mitotic-origin . Our analysis revealed distinct characteristics of whole-chromosome abnormalities generated by meiotic versus mitotic mechanisms . Variation in meiotic and mitotic error rates in embryos from different parents is likely due to a wide array of environmental and genetic factors , not least of which are age [3] and PLK4 genotype [37] . Several researchers have noted a striking elevation of aneuploidy rates in embryos from particular patients , unrelated to maternal age [59] . Together , these findings led us to hypothesize that different forms of infertility—and thus , different referral reasons—would be specifically associated with increased rates of either meiotic or mitotic error . We therefore tested observed rates of meiotic and mitotic error against reasons for PGS referral , regressing out the effect of maternal age where appropriate ( Fig 16 ) . Chromosomes of parental carriers of reciprocal translocations form quadrivalent structures during meiosis [67 , 68] . A fraction of their gametes will thus be unbalanced and the resulting embryos are generally inviable due to these meiotic errors . Known carriers of translocations had expectedly higher rates of whole-chromosome abnormalities in both day-3 blastomere biopsies ( β = 0 . 438 , SE = 0 . 123 , P = 0 . 00038 ) and day-5 TE biopsies ( β = 0 . 502 , SE = 0 . 119 , P = 2 . 57 × 10−5 ) when compared to other clinical indications . Consistent with the known etiology of aneuploidy susceptibility in these patients , translocation status was associated with increased rates of meiotic ( day-3 blastomeres: β = 0 . 743 , SE = 0 . 127 , P = 6 . 54 × 10−9; day-5 TE biopsies: β = 0 . 366 , SE = 0 . 150 , P = 0 . 0146 ) , but not mitotic error ( day-3 blastomeres: β = −0 . 0285 , SE = 0 . 129 , P = 0 . 826; day-5 TE biopsies: β = 0 . 135 , SE = 0 . 172 , P = 0 . 435 ) . Previous studies have also demonstrated an association between IVF failure and patient-specific rates of embryonic aneuploidy [69] . Our data replicate this result , as previous IVF failure was associated with increased error rates in both day-3 blastomere biopsies ( β = 0 . 181 , SE = 0 . 0752 , P = 0 . 0160 ) and day-5 TE biopsies ( β = 0 . 138 , SE = 0 . 0585 , P = 0 . 0187 ) . Given our previous results suggesting selection against putative mitotic-origin aneuploidies between days 3 and 5 , we hypothesized that this association would be driven by errors of mitotic origin . Consistent with this hypothesis , previous IVF failure was associated with increased rates of mitotic error in both day-3 ( β = 0 . 191 , SE = 0 . 0756 , P = 0 . 0114 ) and day-5 embryos ( β = 0 . 213 , SE = 0 . 0830 , P = 0 . 0104 ) , but not associated with meiotic error ( day-3 blastomeres: β = 0 . 0656 , SE = 0 . 0773 , P = 0 . 396; day-5 TE biopsies: β = 0 . 00934 , SE = 0 . 00701 , P = 0 . 183 ) . Given the recent finding that PLK4 influences rates of mitotic-origin aneuploidy , we also tested maternal PLK4 genotype for association with previous IVF failure , but found no significant signal of this transitive association ( β = 0 . 0719 , SE = 0 . 0826 , P = 0 . 384 ) . The wide confidence interval around the coefficient estimate ( 95% CI [-0 . 091 , 0 . 233] ) , however , reflects the small sample size and does not rule out the existence of a moderate effect , such as might be expected given the PLK4 effect size and the relatively modest association between mitotic error and IVF failure . While mitotic error likely contributes to preclinical pregnancy loss , nearly all whole-chromosome abnormalities observed in products of conception from clinical miscarriages are attributable to meiotic error [3] . We thus hypothesized that recurrent pregnancy loss would be associated with increased rates of meiotic error , even after controlling for the well-documented maternal age effect . Recurrent pregnancy loss was indeed associated with increased error rates in day-5 TE biopsies ( β = 0 . 131 , SE = 0 . 0494 , P = 0 . 00831 ) , but not in day-3 blastomeres ( β = −0 . 00945 , SE = 0 . 0630 , P = 0 . 880 ) , possibly due to limitations in sample size and the predominance of mitotic errors at this developmental stage . Consistent with our hypothesis , the day-5 TE association was driven by an underlying association with meiotic ( β = 0 . 127 , SE = 0 . 0584 , P = 0 . 0299 ) , but not mitotic error ( β = 0 . 120 , SE = 0 . 0728 , P = 0 . 100 ) .
Our study is the largest genetic survey of IVF embryos to date . By leveraging parent and embryo genotypes measured via SNP-microarray , we inferred ploidy of all 24 chromosomes of embryo biopsies and assigned detected errors to individual parental homologs . This allowed us to classify whole-chromosome abnormalities of putative meiotic and mitotic origin to separately investigate parental age associations and chromosome-specific profiles . Our results demonstrate that this classification approach is important , given the diversity of cytogenetic mechanisms , karyotypic profiles , and consequences for preimplantation development . Our data support the current understanding that whole-chromosome abnormalities are primarily due to errors in maternal meiosis , as well as frequent mitotic errors arising during post-zygotic cell division . Variation in rates of mitotic error may be explained in part by maternal factors ( e . g . [37] ) , as the initial post-zygotic cell divisions are controlled by maternal gene products [70] . In comparison to maternal meiotic and mitotic errors , we found that paternal meiotic errors were rare , though they are of demonstrated importance in some cases [71] . This is consistent with the finding that relatively common male infertility ( affecting approximately 7% of men ) is primarily due to factors other than whole-chromosome abnormalities [72] . Errors did not affect all chromosomes equally . Maternal BPH ( meiotic ) trisomies and maternal monosomies were correlated in their bias toward smaller chromosomes , consistent with common mechanisms of formation . Both non-disjunction and unbalanced chromatid predivision produce corresponding chromosome gains and losses . While these mechanisms are indistinguishable in our data , unbalanced chromatid predivision has been shown to be much more common [9–11] , and a study of metaphase II oocytes and corresponding polar bodies found that this mechanism tended to affect smaller chromosomes [73] . A similar study found that these predivision errors were most strongly affected by maternal age [74] . Thus , inherent susceptibility of smaller chromosomes to premature separation of sister chromatids is likely responsible for the chromosome-specific age association observed in our study . Recent results from a large survey of TE biopsies [75] also support this interpretation , as this study detected a disproportionate maternal age effect on aneuploidies of chromosomes 13 , 15 , 16 , 18 , 19 , 21 , and 22 . Our study is the first , however , to characterize chromosome-specific rates of putative mitotic errors based on a large sample , demonstrating that these errors have a modest , but statistically significant bias toward larger chromosomes . Our study replicated the well-documented association between maternal age and incidence of maternal meiotic error . Using complementary statistical approaches , we also detected a significant association between paternal age and day-3 blastomere aneuploidy while controlling for the correlated effect of maternal age . By stratifying errors of putative maternal meiotic , paternal meiotic , and mitotic origin , however , we demonstrated that the paternal age association is likely driven by maternal meiotic error . Given the small effect size and the lack of plausible biological mechanism for such an association , we conclude that the paternal age effect is likely a statistical artifact due to residual correlation between maternal and paternal ages . This may help explain conflicting results in previous literature [35] . We also detected a weak negative association between paternal age and paternal BPH error in day-5 TE biopsies , but note that a recent study of trisomy 21 also found a negative association [76] . The central finding of our study was a high incidence of complex mitotic-origin aneuploidies in day-3 blastomere biopsies , which we conclude are purged by selection preceding blastocyst formation . Complex mitotic-origin aneuploidies were strongly biased toward chromosome losses over chromosome gains , and tended to involve random combinations of maternal monosomy , paternal monosomy , and nullisomy . This error profile strongly suggests a post-zygotic mechanism that does not discriminate among maternal and paternal homologs . Blastomeres containing such aneuploidies may have been sampled from mosaic embryos undergoing chaotic cell division due to centrosome abnormalities or other mitotic aberrations [5] . Complex mitotic-origin aneuploidies are under-appreciated in the literature , potentially due to technical limitations of alternative PGS technologies . FISH , for example , could systematically underestimate the extent of these errors because ploidy statuses of only a few chromosomes are assayed at once . Chromosome losses are also difficult to distinguish from FISH hybridization failure , potentially causing many complex mitotic-origin aneuploidies involving multiple chromosome losses to be falsely attributed to technical artifacts . Similarly , aCGH cannot distinguish between maternal or paternal identity of homologs and lacks resolution to distinguish meiotic and mitotic errors from single biopsies . One important limitation of our study is the fact that mitotic-origin aneuploidies present in diploid-aneuploid mosaic embryos may not be detectable if impacting a small proportion of cells in a biopsy . As intra-sample mosaicism is only present in multi-cell day-5 TE biopsies , this could confound our interpretation that selection purges these aneuploidies prior to blastocyst formation . While this limitation may be overcome by novel methods to detect mosaicism in multi-cell DNA extractions , we stand by our interpretation based on supporting results from previous disaggregation studies . Early applications of FISH-based PGS consistently demonstrated that extensive mosaicism was present at a higher rate in arrested embryos than in embryos surviving to the blastocyst stage [22 , 26 , 77–79] . We believe that these FISH results are trustworthy despite the technical limitations of the technology [80] , as these limitations should not systematically affect arrested versus non-arrested embryos . Because these studies were limited to assaying few chromosomes , however , they lacked resolution to characterize the full extent of mitotic errors . Consistent with our finding that mitotic errors often affect large numbers of chromosomes , recent results from a survey of 385 cleavage-stage embryos found that survival to blastocyst formation correlates with the number of chromosomal abnormalities [81] . A previous microarray-based study by Natera [82] showed perfect concordance in ploidy between disaggregated TE fractions , again supporting the conclusion that mosaicism is rare on a per-cell ( but not per-embryo [18 , 83] ) basis by the blastocyst stage . In line with these PGS results , a high rate of mitotic spindle and cell division abnormalities have been documented in early embryos by independent methods , but embryos with aberrant spindles and abnormal cell division rarely survive to blastocyst formation . One previous study used confocal laser scanning microscopy to show that mitotic spindle abnormalities and multiple chromosome loss affect a large proportion of cleavage-stage embryos [64] . The proportion of normal mitotic spindles detected in this study increased from 50% at the cleavage stage to 87% at the blastocyst stage [64] , again consistent with our observation that mitotic-origin aneuploidies are common in day-3 blastomeres but rare in day-5 TE biopsies . A recent study used time-lapse imaging to show that 0 of 18 two-pronuclear ( 2PN ) zygotes that underwent aberrant multipolar cell division survived to the blastocyst stage [65 , 84] . Strong , efficient selection against complex mitotic error means that these errors will rarely , if ever , contribute to clinical miscarriage . This does not imply that these errors are unimportant . Indeed , leveraging patient referral data , we detected a significant association between previous IVF failure and incidence of putative mitotic error . This finding in turn suggests that post-zygotic mechanisms of aneuploidy formation are an important factor limiting human fertility that may also help explain fertility differences among individuals . The latter suggestion is supported by the recent identification of a maternal genetic variant influencing mitotic-origin aneuploidy risk [37] . Fewer than 30% of conceptions progress to live birth , even for young , otherwise fertile couples [1] . Our findings highlight one reason for this low rate of human fertility , providing evidence that complex mitotic-origin aneuploidies abound in cleavage-stage embryos , but are purged early in preimplantation development .
Embryos were vitrified at the IVF clinics , shipped to Natera on dry ice , and analyzed within two weeks of arrival . Genetic material was obtained from oocyte donors ( buccal swabs ) , fathers ( peripheral venipuncture ) , and embryos ( either single-cell day-3 blastomere biopsy or multi-cell day-5 trophectoderm biopsy ) . Single tissue culture and egg donor buccal cells were isolated using a sterile tip attached to a pipette and stereomicroscope ( Leica; Wetzlar , Germany ) . For fresh day-3 embryo biopsy , individual blastomeres were separated via micromanipulator after zona pellucida drilling with acid Tyrode’s solution . Single cells for analysis were washed four times with buffer ( PBS buffer , pH 7 . 2 ( Life Technologies , 1mg/mL BSA ) . Multiple displacement amplification ( MDA ) with proteinase K buffer ( PKB ) was used for this procedure . Cells were placed in 5μl PKB ( Arcturus PicoPure Lysis Buffer , 100 mM DTT , 187 . 5 mM KCl , 3 . 75 mM MgCl2 , 3 . 75 mM Tris-HCl ) incubated at 56°C for 1 hour , followed by heat inactivation at 95°C for 10 min , and held at 25°C for 15 min . MDA reactions were incubated at 30°C for 2 . 5 hours and then 65°C for 10 min . Genomic DNA from buccal tissue was isolated using the QuickExtract DNA Extract Solution ( Epicentre; Madison , WI ) . Template controls were included for the amplification method . Amplified single cells and bulk parental tissue were genotyped using the Infinium II ( Illumina; San Diego , CA ) genome-wide SNP arrays ( HumanCytoSNP12 chip ) . The standard Infinium II protocol was used for parent samples ( bulk tissue ) , and Genome Studio was used for allele calling . For single cells , genotyping was accomplished using an Infinium II genotyping protocol . Detection and classification of various forms of whole chromosome abnormality was achieved using the Parental Support algorithm previously described by Johnson et al . [6] . This approach uses high-quality genotype data from the father and the mother ( or oocyte donor ) to infer the presence or absence of homologs in embryo genotype data . This procedure is useful because embryo biopsies incur a high allelic dropout rate due to limited starting material and whole-genome amplification . Johnson et al . [6] performed validation of the Parental Support method on single cells biopsied from both aneuploid and euploid cell lines , comparing the approach to the ‘gold standard’ of metaphase karyotyping . The authors found that the sensitivity ( 97 . 9%; 323/330 ) and specificity ( 96 . 1%; 125/129 ) of the SNP-microarray based Parental Support approach were comparable to metaphase karyotyping . Furthermore , confidence scores obtained from this approach were strongly correlated with false-detection rates . Consistent with these results , Treff et al . [80] demonstrated that SNP-microarray-based approaches are more consistent in detecting aneuploidy than widely-used FISH technology . Ploidy detection in day-5 blastocysts is complicated by the possibility of mosaicism within a single TE biopsy . Nevertheless , several studies have suggested that while the overall rate of mosaicism in blastocysts is higher than in individual blastomeres , due to the greater number of cells , the proportion of cells affected is much lower . We consider this limitation more extensively in the Discussion section . All statistical analyses were conducted using the R statistical computing environment [85] . Separate analyses were performed on day-3 blastomere biopsies and day-5 TE biopsies , which had different proportions of specific forms of whole-chromosome abnormalities . We used Poisson regression to test for association between the numbers of samples submitted per case and maternal age . In order to model overdispersion , we did not fix the dispersion parameter ( i . e . quasi-Poisson ) . We used logistic regression to test for associations between maternal and paternal ages and specific forms of meiotic and mitotic error . For each IVF case , we counted the number of embryos in which a particular form of whole-chromosome abnormality was detected , while considering all other embryos as controls . For the model of overall rate of whole-chromosome abnormalities , we added polynomial terms for maternal age in increasing order until the addition of a higher order term did not provide significantly better fit , as indicated by an F-test . In order to model overdispersion , we did not fix the dispersion parameter in these generalized linear models ( GLMs ) ( i . e . , quasi-binomial ) . We added egg donor status as a predictor to the best-fit age models , comparing the respective models with F-tests . To test various clinical indications against rates of whole-chromosome abnormalities , we utilized complementary procedures to ensure that our findings were robust . We first fit a model of maternal age versus error rates , including polynomial terms for maternal age in increasing order until the addition of a higher order term did not provide significantly better fit . We then added to this model each clinical indication , testing whether addition of that clinical indication provided significantly better fit using an F-test . We also performed backward model selection with the Akaike information criterion ( AIC ) , starting with a full model that included all clinical indications ( except for advanced maternal age , as maternal age was separately included in the model ) . Fig 16 depicts the results from the full regression model , as all significant predictors ( P < 0 . 05 ) were also significantly associated when tested using the complementary statistical procedures described above . Regression coefficients ( β ) were exponentiated to calculate odds ratios . We used Pearson correlations to assess the relationships between chromosome-specific rates of error affecting different developmental stages and chromosomes with different lengths as well as the relationship between chromosome-specific rates of different forms of whole-chromosome abnormality . To test for an interaction between chromosome-specific rates of maternal meiotic error and chromosome length , we fit a logistic GLM with the response variable encoded as counts of BPH and non-BPH blastomeres for each chromosome for cases stratified into maternal age groups ( rounding to the nearest year ) . Predictor variables included maternal age , chromosome length , and an interaction of age and chromosome length . In order to model overdispersion , we did not fix the dispersion parameter ( i . e . , quasi-binomial ) . To calculate relative difference in rates of whole-chromosome abnormalities for the two different sample types , we stratified errors by the total number of affected chromosomes , then used the formula: ( p − q ) /p , where p is the proportion of affected blastomeres and q is the proportion of affected TE biopsies with a given number of whole-chromosome abnormalities out of the total sample . The Stanford University Research Compliance Office deemed this work to not meet the Federal definition of human subjects research , and it was thus exempted from IRB review . This determination was based on the facts that 1 ) the work involved no intervention or interaction with study subjects , 2 ) researchers did not obtain or receive individually identifiable private information , and 3 ) the data or specimens were collected for purposes other than the current research , the identifiers for the data or specimens were replaced with a code , and the research team was prohibited from obtaining the key to the code . Natera , Inc . also received an IRB exemption for this retrospective examination of the de-identified prenatal genetic screening data in a review conducted by Ethical & Independent Review Services . Analysis code and auxiliary files are available via GitHub: https://github . com/rmccoy7541/aneuploidy_analysis . De-identified primary data are shared with [37] , and ploidy calls are available in supplemental materials of that publication . Questions regarding the detection of aneuploidy and the underlying genotype data should be addressed to Zachary Demko ( zdemko@natera . com ) . | By day 3 of development , more than half of human embryos contain at least one cell that deviates from the typical 46-chromosome complement . These whole-chromosome abnormalities include polyploidies , which affect the entire chromosome set , as well as aneuploidies , which involve gains and losses of particular chromosomes . The rate of aneuploidy increases with maternal age , primarily due to chromosome segregation errors arising during egg formation ( maternal meiosis ) . While some forms of aneuploidy , such as Trisomy 21 , are compatible with live birth , most aneuploid embryos do not survive to term . Our study applied genetic techniques to screen early embryos from in vitro fertilization cycles , demonstrating that while diverse whole-chromosome abnormalities can be observed at early developmental stages , these errors are strongly filtered during preimplantation development . Specifically , errors occurring during the initial post-fertilization cell divisions often result in the simultaneous loss of multiple chromosomes , a pattern consistent with abnormal cell division . Our data provide evidence of selection against this class of aneuploidy before day 5 of development , thus reducing fertility . Patients referred for genetic screening due to previous IVF failure had higher rates of mitotic error , highlighting its clinical relevance and indicating that patient-specific genetic and environmental factors influence error rates . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Evidence of Selection against Complex Mitotic-Origin Aneuploidy during Preimplantation Development |
Schistosomiasis japonica is a zoonotic parasitic disease . After nearly 70 years of control efforts in China , Schistosomiasis transmission has been reduced to a much lower level . The absence or near absence of infections in humans or livestock , based on traditional fecal and serological tests , has made the targets and priorities of future control efforts difficult to determine . However , detection of schistosome cercariae in waters using sentinel mice could be an alternative way of identifying remaining foci of infection , or even serve as a tool for evaluation of control efficacy . This method has been employed in China over last forty years . We therefore performed a meta-analysis of the relevant research to investigate if infections in sentinel mice mirror the ongoing trend of schistosomiasis transmission in China . We conducted a meta-analysis of studies reporting infection rates of S . japonicum in sentinel mice in China before Sep 1 , 2018 in accordance with the PRISMA guidelines . We retrieved all relative studies based on five databases ( CNKI , WanFang , VIP , PubMed and Web of Science ) and the reference lists of resulting articles . For each individual study , the infection rate in sentinel mice is presented together with its 95% confidence interval ( CI ) . Point estimates of the overall infection rates and their 95% CIs were calculated . Subgroup analyses were performed according to study periods , seasons or regions . We identified 90 articles , including 290 studies covering eight endemic provinces . The overall rate in sentinel mice was 12 . 31% ( 95% CI: 10 . 14–14 . 65% ) from 1980 to 2018 . The value of 3 . 66% ( 95% CI: 2 . 62–4 . 85% ) estimated in 2004 to 2018 was significantly lower than in 1980 to 2003 ( 22 . 96% , 95% CI: 19 . 25–26 . 89% ) . The estimate was significantly higher in the middle and lower reaches than in the upper reaches of the Yangtze River . The highest estimates were obtained in Hunan ( 30 . 11% , 95% CI: 25 . 64–34 . 77% ) followed by Anhui ( 26 . 34% , 95% CI: 12 . 88–42 . 44% ) and then Jiangxi ( 13 . 73% , 95% CI: 6 . 71–22 . 56% ) . Unlike the other provinces in the middle and lower reaches , no significant reduction was seen in Hubei after 2003 . Even in Hubei two studies carried out after 2014 reported infections in sentinel mice , although no infected snails were reported across the province . Infections were most found in April ( 17 . 40% , 95% CI: 1 . 13–45 . 49% ) , July ( 24 . 98% , 95% CI: 15 . 64–35 . 62% ) and October ( 17 . 08% , 95% CI 5 . 94–32 . 05% ) . High degrees of heterogeneity were observed . This meta-analysis provides a comprehensive analysis of schistosome infection in sentinel mice across China . The estimates largely mirror the ongoing trends of transmission in terms of periods and regions . Infections were most likely to occur in April , July and October . In areas where no infected snails were reported infections in sentinel mice were still observed . Due to the presence of snails and infected wildlife , detection of schistosomes in waters using such a highly sensitive method as the deployment of sentinel mice , remains of importance in schistosomiasis monitoring . We would suggest the current criteria for transmission interruption or elimination of schistosomiasis in China be adjusted by integrating the results of sentinel mice based surveys .
Schistosomiasis , caused by blood flukes of the genus Schistosoma ( phylum Platyhelminthes; class Trematoda ) , is a zoonotic parasitic disease and is one of the 18 neglected tropical disease listed by the World Health Organization [1 , 2] . Currently , an estimated 240 million people are infected with the parasites , and about 800 million are at risk of infection in 78 tropical and subtropical countries [3 , 4] . The majority of human infections and morbidity are caused by three schistosome species: Schistosoma mansoni , Schistosoma haematobium , and Schistosoma japonicum [5] . Among three species , Schistosoma japonicum is considered to cause the most serious disease due to its female worms’ highest egg output and a possible longest lifespan of adult worms [6] . In China , Schistosomiasis is mainly caused by the infection of S . japonicum . Schistosomiasis japonicum is a water-borne parasitic disease with amphibious Oncomelania hupensis snails serving as its intermediate host . In water infected snails release cercariae , which infect humans or animals when they have water activities nearby . After infection , schistosomula migrate to the liver where they develop and mate with an opposite-sex parasite . The paired worms then migrate into mesenteric veins where they reside and lay eggs . A fraction of the eggs are discharged out of the body with host’s stool , and then in water eggs hatch into free-swimming miracidia , which penetrate snails and then develop into cercariae [7] . Schistosomiasis japonicum was once a major public health problem in China with up to 11 . 6 million human cases in 1950s . After nearly 70 years of control efforts , the number of infected people has been gradually reduced to nearly 37 . 6 thousand in 2017 [8 , 9] . In 2014 , the central government of China proposed a two-stage roadmap with aims to achieve transmission interruption by 2020 and then to achieve disease elimination by 2030 [10 , 11] . However , due to the complex lifecycle of S . japonicum and its easy colonization of new snail populations [12] , the widely distributed snail habitats [9] , and the existence of infected wildlife [13] , schistosomiasis elimination in China remains a great challenge , and recently schistosomiasis has even been believed to be more serious than previous thought [14] . One important issue we have encountered in China is that much lower infection prevalences of S . japonicum , based on the traditional fecal and serological tests , in both humans and livestock have been frequently reported and documented [15] , which may have led to no or unclear targets of further control efforts . However , detection of existence of schistosome cercariae in waters could be an alternative way in determining potential foci of transmission , or even serve as an evaluation of current schistosomiasis situation . There are several approaches for detecting water infectivity , including use of sentinel mice or rabbits , sticking cercariae with specific membrane , detection of parasite DNA with PCR , and so on [16–18] . Among these , the sentinel mice method has been most commonly employed because of its high sensitivity and simple operability [19] , and even recommended as one control measure [20] . The procedure is to put a group of 5 to 10 mice into a wire cage , which is tied with foam plastics at two ends and can float on water surface . This ensures mice to expose to water on limbs , tail and lower abdomen . The period of water exposure in practice generally lasts for a few hours per day and for two to three days . The exposed mice are then raised in the laboratory for 28 to 35 days and dissected for recovering worms or eggs [21] . This method has been employed in China over last forty years . Several research even reported an infection rate of up to 100% of S . japonicum in mice [22–30] . As infections in sentinel mice could also be an index of endemic situation in areas , we therefore performed a meta-analysis of research performed over the last 40 years to estimate the overall prevalence of S . japonicum in sentinel mice to see if it mirrors the ongoing trend of the parasite transmission in China . Subgroup analyses according to study periods , seasons and regions were also performed . To the authors’ knowledge , this is the first time to assess the potential and direct threat of S . japonicum infection in natural environments to humans and livestock . The purpose was to increase our knowledge on how to facilitate schistosomiasis control , particularly in China with low infection prevalence of the parasite in both humans and cattle [15] .
A comprehensive literature search was carried out for publications published before September 1 , 2018 . Three Chinese and two English electronic bibliographic databases , namely China National Knowledge Infrastructure ( CNKI ) , Wanfang , Chinese Scientific Journal Databases ( VIP ) , PubMed and Web of Science ( SCI ) , were searched to include all published studies that reported the infection rate of S . japonicum in sentinel mice within mainland china . We used search terms ‘Sentinel mice ( or mouse ) ’ , ‘Schistosomiasis’ , and ‘China’ in the English databases and ‘shaoshu’ , ‘xuexichong and/or xuexichongbing’ in the Chinese databases . No restrictions were imposed . To find additional studies , we also manually checked the relevant eligible literatures through cross-references of the identified articles in the reference lists . We did not contact authors of original studies for additional information . No attempt was made to identify unpublished studies . Full text articles were downloaded or obtained through library resources . All papers were imported to the literature management software Endnote X7 to eliminate duplicated records . Two authors ( CQ and HZ ) independently conducted an initial screening of identified titles and abstracts and then the full-text articles were downloaded for a second screening . Studies were considered eligible only if they: ( i ) were carried out within mainland China; ( II ) were neither experimental studies nor review articles; ( iii ) clearly reported the time performed , as least specific to year; ( iv ) provided geographical location , at least specific to provinces; ( v ) provided numbers of dissected and infected mice , or could calculate by formula; ( vi ) were available in full texts . Studies were excluded if they did not fulfill any of these criteria . We deemed data regarding S . japonicum infection rates in sentinel mice from the same place at the same time as a single study , and so sometimes an article can contain several studies . If the same study data were published in both English and Chinese sources , the articles with less detailed information would be excluded from this study . The detailed characteristics of each study were extracted using a pre-designed data-collection excel form . Information was recorded as follows: surname of first author , year of publication , year of study , location of study , numbers of the infected and dissected mice . The pooled infection rate and its 95% confidence intervals ( CI ) of S . japonicum in sentinel mice were calculated with the Freeman-Tukey double arcsine transformation [3 , 31] . Besides addressing the problem of variance instability , this approach also solves the problem of confidence limits falling outside the 0 to 1 range , as transformed infection rates are weighted slightly towards 50% and thus studies with infection rates of zero or one can be included in the analysis . Forest plots were used to visualize the results of each study and the heterogeneity among studies . Between studies heterogeneity was assessed using the Cochran’s Q ( reported as p values ) , which is quantified by I2 values . The I2 index indicate the variation between studies attributed to heterogeneity rather than chance , with values of 25 , 50 and 75% corresponding to low , moderate , and high degrees of heterogeneity , respectively [32] . When there was evidence of heterogeneity ( I2 > 50% ) , infection rates were combined by using a random-effects model; otherwise , rates were combined by using a fixed-effects model [33] . We also conducted subgroup analyses to investigate potential sources of heterogeneity . Such analyses were performed on the following variables: 1 ) by study periods , i . e . 1980 to 2003 and 2004 to 2018 according to the schistosomiasis control strategy implemented [34]; 2 ) by river reaches , i . e the upper , middle , and lower reaches of the Yangtze River [35] . The upper reach includes Sichuan and Yunnan provinces , the middle includes Hunan , Hubei and Jiangxi provinces , and the lower includes Anhui , Jiangsu and Zhejiang provinces; 3 ) by provinces; 4 ) by seasons a study performed . The publication bias was visually examined by funnel plots and the statistical significance was assessed by the Egger’s regression asymmetry test [36] . A two-tailed p value < 0 . 05 was considered statistically significant . Extracted data were entered into Microsoft Office Excel 2016 and R3 . 5 . 1 was used in all statistical analyses . This study was carried out in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses ( PRISMA ) guidelines [37] , and the PRISMA checklist ( S1 Checklist ) was used as the basis for inclusion of relevant information .
We identified 495 potentially relevant publications through five databases and the reference lists , of which 205 articles were excluded when taking duplication into consideration . After the initial screening of titles and abstracts , a further 95 articles were excluded . Employing the selection criteria , we obtained quantitative data for our meta-analysis after reading through full texts . The search strategy finally resulted in 90 articles ( 3 in English and 87 in Chinese ) [22–30 , 38–118] , reporting 290 studies . Fig 1 shows our systematic workflow for identifying , screening , and including studies in this study . The years of the studies performed and published ranged from 1980 to 2018 and from 1985 to 2018 , respectively . A total of 91 studies were reported from Jiangsu province , 53 from Hunan , 40 from Jiangxi , 40 from Hubei , 30 from Anhui , 20 from Sichuan , 15 from Yunnan , and one from Zhejiang . A total of 63998 sentinel mice were dissected and 7521 were identified with S . japonicum infection . A total of 153 studies were carried out during 1980 to 2003 , and 137 during 2004 to 2018 . The infection rates of S . japonicum in sentinel mice among the included studies varied between 0 and 100% . The detailed characteristics of each study are provided in Supporting Information file S1 Table , and its infection rate with 95% CI are also plot in Supporting Information file S1 Fig . A substantial heterogeneity was observed among studies ( χ2 = 20412 . 5 , p < 0 . 0001; I2 = 98 . 6% , 95% CI: 98 . 5–98 . 6% ) . When calculated using a random-effects model , the overall infection rate was 12 . 31% ( 95% CI: 10 . 14–14 . 65% ) . The estimates of infection rates for different subgroups and heterogeneities are presented in Table 1 and Fig 2 . All pooled infection rates for each subgroup were calculated using a random-effects model because of the observed high heterogeneity among studies within subgroups . Based on study periods , the estimate had been significantly reduced since 2004 ( 1980–2003: 22 . 96% ( 95% CI: 19 . 25–26 . 89% ) , n = 153 vs 2004–2018: 3 . 66% ( 95% CI: 2 . 62–4 . 85% ) , n = 137 ) . In terms of the River reaches , the estimate was highest in the middle reach ( 15 . 85% , 95% CI: 12 . 54–19 . 45% , n = 133 ) , followed by that in the lower reach ( 12 . 80% , 95% CI: 9 . 55–16 . 42% , n = 122 ) , and the lowest in the upper reach ( 2 . 12% , 95% CI: 0 . 63–4 . 32% , n = 35 ) . At the level of provinces , the estimates ranged from 1 . 41% ( 95% CI: 0 . 05–4 . 09% , n = 20 ) in Sichuan to 30 . 11% ( 95% CI: 25 . 64–34 . 77% , n = 53 ) in Hunan . In terms of season , the estimate was highest in July ( 24 . 98% , 95% CI: 15 . 64–35 . 62% , n = 33 ) and lowest in September ( 5 . 36% , 95% CI: 2 . 25–9 . 57% , n = 27 ) . The forest plots for each subgroup are provided in Supporting Information file S2–S4 Figs . In addition , we also made a further stratification of meta-analyses within provinces according to study periods . As seen in Fig 3 , a rapid reduction in the pooled infection rate has been seen in all provinces since 2004 with the exception of Hubei province . Even in the latter , two studies reported infection rates of 1 . 75% in 2015 and 2 . 5% in 2016 . See Supporting Information file S1 Table . A potential publication bias was indicated by Egger linear regression test ( bias coefficients b = 0 . 16 , t = 3 . 97 , p < 0 . 0001 ) , which is showed in Fig 4 . Subgroup analyses also suggested a publication bias for studies during the period of 2004 to 2018 , within each of the three provinces ( i . e . Jiangsu , Jiangxi , and Anhui ) and in the lower reach of the Yangtze River ( see Table 1 ) .
Though this study provided information on schistosomiasis transmission in China over the last 40 years and also directly assessed the threat of infections to humans and livestock , which would be useful in aiding future schistosomiasis control , it is not devoid of limitations . First , the estimations of infection rates using a random-effect model may not absolutely invalidate the heterogeneity between studies . Secondly , we don’t conduct the sensitivity analysis , as nearly 290 studies were included . Finally , we observed some evidence of publication bias in this work . Publication bias may exist when there is a preference to publish studies with significant findings . However , there is no certainty for a paper with high or low infection rates of S . japonicum in sentinel mice to get easily published; moreover , not all subgroup analyses showed publication bias . We thus think that publication bias is unlikely to have distorted our results . This meta-analysis provides a comprehensive analysis of S . japonicum infection in sentinel mice across China . The estimates largely mirror the ongoing trends of S . japonicum infections in terms of periods and regions . Infections were most likely to occur in April , July and October . However , in areas where no infected snails were reported infections of S . japonicum in sentinel mice were still observed . Due to the wide distribution of snails and the existence of any infected wildlife , detection of schistosome in waters using such a highly sensitive method remains of importance in the monitoring and objective evaluation of the disease . We would suggest that the current criteria for transmission interruption or elimination of S . japonicum in China [137] be adjusted by integrating the results of sentinel mice method . The recently updated method will facilitate its wide application and make the index easily obtained . | With the continued activities of the prevention and control programme in China , the prevalence and intensity of Schistosoma japonicum infection have been reduced to low levels . This makes it impossible to detect any infections in humans or livestock using the traditional approach of fecal and serological testing , so as to evaluate properly the risk map of infection . However , detection of existence of schistosome cercariae in waters could be an alternative way of detecting a potential focus of transmission . We therefore performed a meta-analysis of studies performed over the last 40 years to estimate the overall infection rates of S . japonicum in sentinel mice . The estimate across China in 2004 to 2018 was 3 . 66% , significantly lower than in 1980 to 2003 . The highest estimates were observed in Hunan , followed by Anhui and Jiangxi . Two studies conducted in Hubei in 2015 and 2016 respectively , reported infected sentinel mice where no infected snails had been reported across the province since 2014 . Transmission was found be most likely in April , July and October . The estimates largely mirror the ongoing trends of S . japonicum infections in terms of periods and regions . Due to the presence of snails and other infected wildlife , detection of schistosome cercariae in waters with a highly sensitive tool remains of great importance in schistosomiasis monitoring and evaluation . To the authors’ knowledge , this is the first time that the potential threat of S . japonicum in nature , to humans and livestock has been assessed in this manner . We would suggest that the current criteria for transmission interruption or elimination of S . japonicum in China be adjusted by integrating the results of sentinel mice . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"schistosoma",
"invertebrates",
"livestock",
"medicine",
"and",
"health",
"sciences",
"helminths",
"china",
"statistics",
"metaanalysis",
"tropical",
"diseases",
"geographical",
"locations",
"parasitic",
"diseases",
"animals",
"gastropods",
"mathematics",
"neglected",
"tropical",
"diseases",
"snails",
"research",
"and",
"analysis",
"methods",
"schistosoma",
"japonicum",
"mathematical",
"and",
"statistical",
"techniques",
"molluscs",
"agriculture",
"people",
"and",
"places",
"helminth",
"infections",
"schistosomiasis",
"eukaryota",
"asia",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"statistical",
"methods",
"organisms"
]
| 2019 | A meta-analysis of infection rates of Schistosoma japonicum in sentinel mice associated with infectious waters in mainland China over last 40 years |
Control and coordination of eukaryotic gene expression rely on transcriptional and posttranscriptional regulatory networks . Evolutionary innovations and adaptations often require rapid changes of such networks . It has long been hypothesized that transposable elements ( TE ) might contribute to the rewiring of regulatory interactions . More recently it emerged that TEs might bring in ready-to-use transcription factor binding sites to create alterations to the promoters by which they were captured . A process where the gene regulatory architecture is of remarkable plasticity is sex determination . While the more downstream components of the sex determination cascades are evolutionary conserved , the master regulators can switch between groups of organisms even on the interspecies level or between populations . In the medaka fish ( Oryzias latipes ) a duplicated copy of dmrt1 , designated dmrt1bY or DMY , on the Y chromosome was shown to be the master regulator of male development , similar to Sry in mammals . We found that the dmrt1bY gene has acquired a new feedback downregulation of its expression . Additionally , the autosomal dmrt1a gene is also able to regulate transcription of its duplicated paralog by binding to a unique target Dmrt1 site nested within the dmrt1bY proximal promoter region . We could trace back this novel regulatory element to a highly conserved sequence within a new type of TE that inserted into the upstream region of dmrt1bY shortly after the duplication event . Our data provide functional evidence for a role of TEs in transcriptional network rewiring for sub- and/or neo-functionalization of duplicated genes . In the particular case of dmrt1bY , this contributed to create new hierarchies of sex-determining genes .
Control and coordination of eukaryotic gene expression rely on transcriptional and posttranscriptional regulatory networks . From an evolutionary point of view innovations and changes in given functional linkages of regulatory networks have to occur at the DNA level by alteration of the cis-regulatory sequence defining transcription factor binding sites . While such alterations may occur in any cis-regulatory module , they will have fundamentally different effects depending on where in the structure of the network they occur ( see [1] for review ) . After the discovery of the ubiquitousness of repeated sequences , a long standing hypothesis proposed that repeated sequences were likely to be active in the 5′ regions of genes controlling transcription [2] and that they could move and supply evolutionary variations [3] . From an evolutionary perspective , transposable elements ( TEs ) have recently been attributed an important role in shaping the gene regulation landscape [4] , [5] , [6] . In spite of and , to some extent , because of their selfish and parasitic nature , the movement and accumulation of TEs have exerted a strong influence on the evolutionary trajectory of their host genome [7] . Many ways have been illustrated through which TEs can directly influence the regulation of nearby gene expression , both at the transcriptional and post-transcriptional levels ( for review see [5] ) . Genome-scale bioinformatic analyses have shown that many promoters and polyadenylation signals of human and mouse genes are derived from primate and rodent–specific TEs respectively [8] , [9] . Hence , it is postulated that insertion of TEs harbouring “ready-to-use” cis-regulatory sequences probably contributed to the establishment of lineage-specific patterns of gene expression [10] . In addition to donating cis-elements and creating new regulatory networks , the movement and accumulation of TEs have recently been proposed to participate in the rewiring of pre-established regulatory networks ( see [5] for review ) . Such rewiring is especially important when rapid adaptation of existing regulatory networks or new networks become necessary . One system where fast changes obviously regularly occurred during evolution is the genetic control of sexual development [11] , [12] . It is well documented that different groups of organisms and sometimes even closely related species of different populations of the same species have fundamentally different modes of sex determination . Comparative evolutionary studies of the genetic cascades controlling sex determination in different species revealed that the master genes at the top of the regulatory hierarchy can change dramatically as new species evolve , while the downstream genes at the bottom of the hierarchy remain the same , exerting essentially identical functions from one species to the next ( see [12] , [13] for review ) . The most conserved downstream component characterized to date , a gene with homology to both the Drosophila doublesex and C . elegans mab-3 sex regulatory genes , is the Dmrt1 ( Doublesex and Mab-3 Related Transcription factor 1 ) gene of vertebrates [14] . All three genes encode proteins sharing a common DNA-binding domain and belong to the DM domain gene family , which has been shown to be involved in sex determination and differentiation in organisms as phylogenetically divergent as corals , worms , flies and all vertebrate groups ranging from fish to mammals . Of note , in humans , haploinsufficiency of the genomic region that includes DMRT1 and its paralogs DMRT2 and DMRT3 leads to XY male to female sex reversal [15] . In chicken and other avian species Dmrt1 is located on the Z chromosome , but absent from W , making it an excellent candidate for the male sex-determining gene of birds [16] , [17] . In the medaka fish ( Oryzias latipes ) , which has XY-XX sex determination , a duplicated copy of dmrt1 , designated dmrt1bY or DMY , on the Y-chromosome was shown to be the master regulator of male development [18] , [19] , similar to Sry in mammals . Interestingly , also in Xenopus laevis a W-specific duplicate of dmrt1 was shown to participate in primary gonad development [20] . Because medaka dmrt1bY acts , like Sry , as a dominant male determiner [21] , it is believed that it is functionally equivalent to the mammalian gene and might share many molecular features [22] , [23] . Although many of the early cellular and morphological events downstream of Sry have been characterized , as well as a number of genes involved in these processes ( for review see [24]–[26] ) , little is known about the mode of action and the biological targets of Sry [27] . Interestingly , dmrt1 , the ancestor of dmrt1bY , is one of these downstream effectors of Sry . Contrary to the situation with Sry , it is totally unknown how in medaka dmrt1bY expression is regulated . As a prerequisite to elucidate the function of dmrt1bY , information on its expression regulation at the transcriptional level is required . Here , we found a feed back down-regulation of the dmrt1bY promoter . Also dmrt1a , the autosomal ancestor of dmrt1bY , is able to down-regulate transcription of its paralog . Interestingly we found clear evidence that the major cis-regulatory element , pre-existing within a new medaka-specific TE at the time of its insertion , was co-opted in order to confer to Dmrt1bY its specific expression pattern after gene duplication . This is the first experimental evidence supporting a role of TEs for transcriptional network rewiring in sub- and/or neo-functionalization of duplicated genes . Additionally , in the particular case of dmrt1bY , this contributed to create new hierarchies of sex determining genes .
To obtain insights into the sequence evolution of the cis-regulatory region of the dmrt1bY gene on the Y-chromosome , we first compared its genomic region and that of its autosomal progenitor , the dmrt1a gene from linkage group 9 ( LG9 ) , with those of the available dmrt1 orthologs from other teleosts ( stickleback , Tetraodon , Fugu , zebrafish ) , chicken and human . This phylogenetic footprinting approach should point to the conservation of regulatory elements being putatively essential for vertebrate Dmrt1 gene expression ( Figure S1 ) . Furthermore , it could possibly indicate cis-regulatory subfunctionalization between the medaka dmrt1 paralogs as observed for other pairs of duplicated genes with spatio-temporal expression divergence [28] , [29] . However , our VISTA plots ( Figure S1 ) did not reveal sequence conservation in the promoter regions of neither dmrt1bY nor dmrt1a with other vertebrates except for stretches corresponding to the teleost MHCL gene , which are pseudogenes in both medaka dmrt1 5′ regions [30] . Conserved non-coding elements were also not found between the other vertebrate sequences suggesting that the regulatory regions of vertebrate Dmrt1 orthologs diverged strongly despite their conserved position in the sex determination cascade . High turn-over of cis-regulatory regions in the face of conserved expression is commonly found for vertebrate genes [31] . However , longer stretches of conservation between promoter regions of dmrt1bY and dmrt1a in medaka were evident ( Figure S1 ) . Thus , we compared in more detail the promoter regions of the medaka dmrt1 paralogs upstream of the transcriptional start site to the last exons of their upstream gene , KIAA00172 , which is a pseudogene on the Y chromosome but functional on the autosomal LG9 [30] ( Figure 1A ) . This region spans around 9 Kb on the Y chromosome but only around 6 Kb on autosomal LG9 . In the upstream sequence of dmrt1 paralogs , five regions contribute to length divergence between autosome and Y chromosome ( Table 1 , Figure 1A , Figure S1 and Figure S2 ) . Region I located 69 bp upstream of the transcriptional start site of dmrt1bY is over 2 Kb in length and absent from dmrt1a . Similarly , regions II–IV further upstream are only found on the Y chromosome but not on the autosome . Region V , in contrast , is missing from the Y chromosome but present on the autosome . This region contains two exons of the KIAA0172 gene and obviously has been lost during the pseudogenization of the Y chromosomal KIAA0172 copy after the duplication of the dmrt1 region [30] . Region III is directly adjacent to the MHCL pseudogenes [30] , [32] present in both dmrt1 promoters and a stretch of sequence similarity with other teleost MHCL orthologs is found within region III on the Y chromosome ( Figure S1 ) . Furthermore , region III is present only once in the medaka genome and does not constitute a repetitive element according to RepeatMasker analysis . Thus , region III seems to have been lost from the autosomal dmrt1a region after duplication , but has been kept on the Y chromosome . Next , we further characterized the additional regions I , II and IV for the presence of repetitive sequences . As each of these three regions is present in multiple copies in the medaka genome ( Table 1 ) , they constitute repetitive sequences that were rather added to the Y chromosomal region than lost from the autosome . Each repeat element is only present once in the analyzed section of the dmrt1bY region and is often found in poorly assembled regions of the medaka genome emphasizing their repetitive character . BLAST searches in other teleost genomes failed to find the repeat elements identified here in other species suggesting that they are specific to the medaka lineage . Particularly , region I is subdivided into three parts with a non-LTR retrotransposable Rex1 element of the LINE family ( class I transposable elements ) as middle segment ( “repeat 2” ) ( Table 1 ) . The surrounding parts ( “repeat 1a” , “repeat 1b” ) also represent repetitive sequences with multiple copies in the medaka genome . These two repeat regions are found side by side in other regions of the genome . Thus , they together build a larger repeat element ( “repeat 1” ) into which repeat 2 was inserted ( see below ) . Repeat 1 has a length of 1316 bp and is characterized by 27 bp terminal inverted repeats ( TIRs ) ( 5′-CAATGAGTTATATCACTAGAGGAGACA-3′ ) assigning it to DNA transposons ( class II transposable elements ) . However , it does not contain a transposase gene or any other open reading frame and thus constitutes a non-autonomous class II element . Only few diagnostic motifs are available to classify such elements [33] . Repeat 1 in the dmrt1bY promoter has a 8 bp target site duplication ( 5′- GTGTGGCT-3′ ) and other copies of this element in the medaka genome have target site duplications of the same length . Repeat 1 is found in multiple copies in the medaka genome ( Table 1 and Table S1 ) , which generally have target site duplications . This points to an active state of repeat 1 in the medaka genome . From the consensus sequence of the multiple repeat 1 elements in the medaka genome , a THAP protein domain composed of three putative exons was deduced ( Figure S2 and Figure S3 ) . In the repeat 1 element in the dmrt1bY promoter , the second putative exon of the THAP domain has been disrupted by the insertion of the repeat element ( Figure S2 and Figure S3 ) . The THAP domain is a DNA-binding zinc finger motif present in the P element transposases from Drosophila [34] . Furthermore , the terminal motif of repeat 1 is similar to the consensus sequence for the P element superfamily of DNA transposons ( 5′-YARNG-3′ ) [7] . Thus , we conclude that we have identified a new , medaka-specific non-autonomous P element element that we term Izanagi ( named after an ancient Japanese deity , “the male who invites”; for etymology see Text S1 ) . Vertebrate mobile DNA transposons of the P element family have been only found so far in zebrafish [35] , [36] . However , the THAP domain has been recurrently recruited from domesticated P elements during chordate evolution [37] . In the Izanagi family , the THAP domain is degenerated and , in the case of repeat 1 , has been additionally disrupted by the repeat 2 insertion . Region II ( “repeat 3” ) and region IV ( “repeat 4” ) also have multiple copies in the medaka genome ( Table 1 ) . They do not contain open reading frames and , like repeat 1 , they also lack similarity to known transposable elements . Furthermore , target site duplication or other diagnostic features could not be recognized preventing further classification as putative transposable elements . The sequence of the medaka dmrt1bY promoter region ( 9 . 107 Kb ) was next analyzed for the presence of putative transcription factor binding sites using the MatInspector program ( Figure 1A and Figure S4 and Table 1 ) . Most interestingly , the Izanagi element is characterized by an overrepresentation of putative binding sites for Sox5 ( Figure 1 and Figure S4 ) . In this region seven Sox5 binding sites are present while random prediction would expect 15 times less ( only 0 . 46 sites; MatInspector ) . This , together with the fact that Sox5 expression has been correlated with direct dmrt1 promoter down-regulation in zebrafish [38] suggests that region to be of primary interest for medaka Dmrt1bY transcriptional regulation , but remains to be investigated for the proposed functional role of Sox5 . Additionally , the 9 Kb dmrt1bY promoter region contains several other putative transcription factor binding sites such as Pax2 , HMG-box protein 1 , HMG-A , Sox9 , WT1 and SF1 binding sites that are reasonable candidates for gonadal-specific transcriptional regulation ( Figure 1A and Figure S4 and Table 1 ) . Several of them are conserved with the dmrt1a promoter ( Figure S4 ) and might be essentially required for dmrt1 expression . To evaluate the mechanisms regulating dmrt1bY transcription , a portion of the medaka gene from +117 bp to −8990 bp of the transcriptional start site was cloned upstream of the Gaussia luciferase gene ( pBSII-ISceI::9 Kb Dmrt1bY prom::GLuc ) and the activity of the promoter was measured in a variety of cell types using transient transfection analysis . Sequential deletions of the 9 Kb promoter were generated from pBSII-ISceI::9 Kb Dmrt1bY prom::GLuc . In all three cell types basal promoter activity was detectable when using the 3 Kb proximal region ( Figure S5 ) . In fibroblast cell lines ( Xiphophorus A2 and medaka HN2 ) , but not in Sertoli TM4 cells , a dramatic drop in promoter activity was observed when the region from bp −2985 to −6207 was added ( Figure S5 ) . Similarly , the same decrease of promoter activity was apparent in Sertoli TM4 cells when the bp −6207 to −8996 region was additionally inserted ( Figure S5 ) . This indicates the possible presence of Sertoli cell specific transcriptional repressing sequence ( s ) within the most distal part of the dmrt1bY promoter . The most proximal part of the promoter always accounted for the basal activity in all the cell lines tested . Interestingly , two adjacent binding sites for Steroidogenic factor 1 ( Sf1 ) are located at positions −5933 and −5524 ( Figure S4 ) . Being specifically expressed in Sertoli -and Leydig- cells , it is tempting to assume that the presence of these two distinct SF-1 binding sites nested in this −3 to −6 Kb fragment is accounting for this difference . A similar situation has been shown for the porcine Sry promoter for which SF-1 transactivation occurs at two SF1 binding sites [39] . Using the vertebrate Dmrt1 binding site matrix [40] , different dmrt1 promoters -including medaka dmrt1a and dmrt1bY- ( up to 9 KB upstream the ORF ) were scanned for such target site sequences . A unique and robust Dmrt1 binding site of high prediction probability was found only in the medaka dmrt1bY promoter ( CTGCAACAATGCATT; weight: 8 . 5 , pValue: 1 . 0 e-05 , lnPval:−11 . 492 ) ( Figure 1A and Figures S2 , S3 , S4 ) but not in the dmrt1a promoter ( lower threshold set to 0 ) . Interestingly , this predicted Dmrt1 binding site is nested within the above newly described Oryzias latipes Izanagi element in the proximal active part of dmrt1bY promoter ( Figure 1A and Figure S5 ) . The medaka putative Dmrt1 binding site is present at position −2132 within repeat 1b in close proximity to the Sox5 binding site-rich region ( Figure 1A and Figure S4 ) . We first asked about the origin of this Dmrt1 binding site . It might have either evolved de novo from sequence provided by repeat 1b or been an integral part of such repeats and then was inserted into the dmrt1bY promoter after duplication . We therefore blasted the region approximately 300 bp up-and downstream of the Dmrt1 binding site to the medaka genome and aligned the obtained repeat sequences ( Figure 1B ) . In total , we identified 28 elements that are highly similar to repeat 1b and that contain the same Dmrt1 binding site found in the dmrt1bY promoter ( Table S1 ) . Furthermore , the predicted Dmrt1 binding site is present in the derived Izanagi consensus sequence . Hence , this putative Dmrt1 binding site is a regular and conserved part of the Izanagi transposon family . Given that the Dmrt1 binding site donated to the dmrt1bY promoter by the Izanagi element has been important for the evolution of dmrt1bY function within the sex determining cascade , we asked about the timing of the Izanagi insertion in relation to the duplication of the medaka dmrt1 genes . The dmrt1 gene duplication occurred in a common ancestor of medaka ( O . latipes ) , O . curvinotus and O . luzonensis around 10 million years ago [41] . First , we estimated the sequence divergence between repeat 1 from the dmrt1bY promoter and the Izanagi element consensus and mapped it onto a linearized neighbour joining ( NJ ) tree of dmrt1 genes from the genus Oryzias , which was based on neutral sites only ( third codon positions ) . This analysis showed that the repeat 1 insertion occurred after the split from O . mekongensis but before the divergence of medaka , O . curvinotus and O . luzonensis ( Figure S6A ) . This is exactly the branch on which the dmrt1 duplication occurred . We also estimated the dmrt1 duplication by the same method . There has to be a note of caution with dating the age of the dmrt1 duplication due to the enhanced rate of molecular evolution of dmrt1bY after duplication [41] . Nevertheless , based on sequence divergence data the insertion of repeat 1 is certainly estimated to be younger than the dmrt1 duplication ( Figure S6A ) . Using a different nuclear marker to date the divergence of the Oryzias species , the tyrosinase a gene , a similar result was obtained ( Figure S6A ) . The analogous analysis for the secondary insertion of repeat 2 into repeat 1 , in contrast , revealed that this insertion is quite young and must have occurred in Oryzias latipes . We conclude that our sequence divergence estimates are consistent with an insertion of repeat 1 and thereby of the Dmrt1 binding site shortly after the dmrt1 duplication , supporting its importance for the evolution the Dmrt1bY sex determinator function . Thus far we could show feed back down-regulation of dmrt1bY and regulation by its paralog Dmrt1a in vitro . We next addressed whether this regulation indeed exists in vivo .
Sex determination involves a complex hierarchy of genes . Expression screen analyses have resulted in hundreds of candidate genes that show sex-specific expression pattern . However it has been difficult to place these genes into a network of gene regulation and function . Nevertheless , several genes encoding for transcription factors , with specific temporal and spatial expression patterns during early gonad induction , have been suggested to participate in this process . Among them , from C . elegans to mammals , genetic evidence has suggested that the dmrt1 gene is an important regulator of male development at a downstream position of the regulatory network . In medaka , a duplicated copy of dmrt1 has acquired the upstream position of the sex-determining cascade . The analysis presented here provides evidences that this evolutionary novelty , which is predicted to require a rewiring of the regulatory network is brought about by co-option of “ready-to -use” pre-existing cis-regulatory elements carried by transposing elements . We could show that the master sex determining gene of medaka , dmrt1bY , is able to bind to one of these elements in its own promoter . This binding leads to a significant repression of its own transcription . During early stages when the primordial gonad is formed , dmrt1bY is exclusively expressed and exerts its sex determining function [18] . The dmrt1a gene , with its proposed specification and maintenance function for the Sertoli cells , is expressed only when the testes are in the process of differentiation . Notably , the master sex determinator gene dmrt1bY , continues to be expressed . In adult testes , where both paralogs have been shown to be expressed , the predominant expression of dmrt1a compared to dmrt1bY ( 50 fold higher; [43] ) argues for a downregulation of dmrt1bY . Although additional post-transcriptional mechanisms accounting for dmrt1bY expression regulation , involving the 3′ UTR [44] , have been shown to be essential for spatial expression pattern in the embryo and restricted expression to the gonad in adult fish , the data presented here indicate that a feed back auto-regulation of dmrt1bY promoter activity and trans-regulation by its paralog Dmrt1a is a key mechanism of dmrt1bY transcriptional tuning ( Figure 8 ) . With respect to the evolutionary history of the two dmrt1 genes in medaka , it is of note that the newly generated paralog dmrt1bY , independently of any functional considerations , is kept back under tight transcriptional regulation of the ancestral dmrt1a gene . Consequently this avoids any kind of expression pattern redundancy in testes after their development is initiated and could then be a reasonable way of preserving both genes from any purification/degeneration processes after duplication , thus favouring a subsequent sub-neo-functionalization . So far no putative Dmrt1 binding site could be observed within the more than 10 Kb upstream medaka dmrt1a sequence inspected . Similarly such Dmrt1 target sites are absent from the zebrafish , fugu , stickleback , mouse or human 10 kB upstream dmrt1 promoter regions . This together with the apparent loss of Dmrt1 canonical cis-regulatory sequences ( such as Gata4 ) indicates a particular transcriptional context acquired by dmrt1bY during its evolution towards becoming a novel master sex determination gene . It was previously reported that multiple TEs inserted into the Y-specific region on medaka LG1 [30] . Interestingly , our study revealed that the cis-regulatory element containing the Dmrt1 binding site , pre-existing within the Izanagi element at the time of its insertion , was co-opted in order to confer dmrt1bY its specific expression pattern after gene duplication around 10 million years ago [19] , [41] . This fact has interesting evolutionary implications , since TEs are probably the most dynamic part of the genome . Dmrt1 possibly also regulates other genes in the proximity of Izanagi elements via the Dmrt1 binding site ( Table S1 ) . In the context of gene duplication and its correlated process of sub-/neo- functionalization ( see [45]–[47] for review ) , the contribution of TEs to the remodelling of the sex determination cascade ( see [12] , [48] for review ) is of prime interest . The case reported here for the medaka-specific Izanagi element bringing in a novel regulatory element into the dmrt1bY promoter is –at least to our knowledge- the first example showing that TEs not only change/rewire the expression of existing genes but surely lead to the creation of new regulatory hierarchies within recently duplicated genes . The present case is even more interesting since this new TE-derived TFBS confers transcriptional control from the ancestral gene against the duplicate and allows the dmrt1bY gene to take an upstream position in the sex determination cascade without excluding its dmrt1 ancestor from a role in sexual development . This supports a role of TEs for transcriptional network rewiring in sub- and/or neo- functionalization of duplicated genes in creating new hierarchies of sex determining genes .
Comparative analysis of vertebrate Dmrt1 genomic regions were performed with mVISTA at http://genome . lbl . gov/vista [49] using the Shuffle-LAGAN alignment program [50] . Medaka dmrt1a ( LG9 ) and dmrt1bY ( LG1 ) region sequences were obtained from [30] , all other regions from the Ensembl Genome Browser ( http://www . ensembl . org/; release 49 , March 2008 ) : stickleback groupXIII , scaf57; Fugu scaf4; Tetraodon chr12 , scaf14966; zebrafish chr5 , scaf463; chicken chrZ , supercontig194; human chr9 , supercontig NT_008413 . Screens for repetitive elements were performed with RepeatMasker ( http://www . repeatmasker . org/ ) . Additional copies of repeat elements and their genomic environment in the medaka genome ( version HdrR , Oct 2005 ) were identified with BLASTN with >85% sequence identity over >85% of query length ( Table 1 ) . Alignments of repeat elements were obtained with CLUSTALW as implemented in BioEdit [51] followed by manual improvement . 50% threshold frequency was used for inclusion in repeat consensus sequences . The putative THAP domain found in the Izanagi consensus was identified by comparison to the PFAM database ( http://pfam . sanger . ac . uk/ ) . Transcription factor binding sites were determined using MatInspector of the Genomatix portal ( http://www . genomatix . de/ ) . Binding sites for Dmrt1 in different genomes were identified using the matrix provided by [40] together with the Regulatory Sequence Analysis Tools portal; RSat ( http://rsat . ulb . ac . be/rsat/ ) . MEGA4 [52] was used to estimate sequence divergence between repeat 1 and the Izanagi element consensus ( 0 . 034 +− 0 . 005 ) as well as between repeat 2 and the Rex1 element consensus ( 0 . 010 +− 0 . 003 ) using the Kimura-2-parameter model . Linearized neighbor-joining trees of dmrt1 and tyrosinase a gene were obtained as described in ref . [41] , with the only exception that they were based on third codon positions only . Other models of sequence evolution gave similar results . Accession numbers are given in Figure S6 . For promoter analysis , a 9107 bp fragment upstream of the Dmrt1bY open reading frame ( ORF ) was isolated by restriction enzyme digestion ( XhoI/EcoRI ) from BAC clone Mn0113N21 [30] , was cloned into pBSII-ISceI plasmid ( pBSII-ISceI::9 Kb Dmrt1bY prom . plasmid ) . Subsequently , Gaussia luciferase gene from pGLuc-basic ( New England Biolabs ) plasmid was inserted between EcoRI and NotI sites of pBSII-ISceI::9 Kb Dmrt1bY prom ( pBSII-ISceI::9 Kb Dmrt1bY prom::GLuc plasmid ) . pBSII-ISceI::3 Kb Dmrt1bY prom::GLuc and pBSII-ISceI::6 Kb Dmrt1bY prom::GLuc plasmids were constructed the same way removing 5′ fragments of the 9107 bp Dmrt1bY promoter region using Eco47III and HindIII restriction enzyme digestion respectively and re-ligation . Mutation of the Dmrt1bY binding site was performed by PCR in the context of pBSII-ISceI::3 Kb Dmrt1bY prom::GLuc plasmid ( native form: AATGCATTGTTGCAG; mutated form: GCCGGCTTCCCACCA ) . All PCR-obtained fragments were sequenced . To generate plasmids for in vitro transcription , full-length cDNAs encoding medaka dmrt1a or dmrt1bY were subcloned into EcoRI/NotI digested pRN3 plasmid [53] . For establishment of transgenic lines , either GFP or mCherry open reading frames were inserted between EcoRI and NotI sites of pBSII-ISceI::9 Kb Dmrt1bY prom ( pBSII-ISceI::9 Kb Dmrt1bY prom::GFP or mCherry plasmids respectively ) . GFP fusion protein vector ( dmrt1bY::GFP and deltadmrt1bY::GFP ) were constructed as described in [43] . Mouse TM4 Sertoli cells , Xiphophorus embryonic epithelial A2 cells , and medaka spermatogonial ( Sg3 ) and fibroblast like ( HN2 ) cells were cultured as described [54] , [55] , [56] , [57] . Cells were grown to 80% confluency in 6-well plates and transfected with 5 µg expression vector using FuGene ( Roche ) or Lipofectamine ( Invitrogen ) reagents as described by the manufacturers . Gaussia luciferase activity was quantified using the Luciferase Reporter Assay System from Promega and normalized against co-tranfected firefly luciferase expressing plasmid ( ptkLUC+; [58] ) . When DNA amounts transfected are expressed as a ratio , the total amount of expression vector remained constant ( 5 µg ) by filling in the reaction with empty vector . Experiments for which error bars are shown result from at least three replicates and error bars represent the standard error of the mean . ( Dmrt1bY-Trgt ) 5′-AGCTTAATGCATTGTTGCAGAGCT-3′ , ( Competitor ) 5′-AGCTGACGGCCGCGAAGCAAGCT and respective complements were annealed by heating to 90°C for five minutes in 1X T4 PolyNucleotide Kinase ( PNK ) buffer ( 70 mM Tris-HCl ( pH 7 . 6 ) , 10 mM MgCl2 , 5 mM dithiothreitol ) ; slow-cooled to 50°C; held at that temperature for 5 minutes and then cooled to room temperature . For radioactive labelling 50 pmol of the duplex 5′ termini were used together with 50 pmol of gamma-[32P]-ATP and 20 units of T4 PNK in 1X adjusted T4 PNK buffer and incubated for 20 minutes at 37°C . Unincorporated nucleotides were removed through a Sephadex G-50 spin column . For producing Dmrt1a and Dmrt1bY proteins , pRN3::Dmrt1a or pRN3::Dmrt1bY plasmids were linearized using KpnI and then in vitro transcribed using mMessage mMachine kit ( Ambion ) . Finally , Dmrt1a or Dmrt1bY proteins were in vitro translated using Ambion's Retic Lysate Kit from the previously in vitro transcribed capped Dmrt1a or Dmrt1bY RNAs . DNA binding reaction contained 10 mMTris-HCl ( pH 7 . 9 ) , 100 mM KCl , 10% glycerol , 5 mM MgCl2 , 1 µg torula rRNA , 0 . 075% Triton X-100 , 1 mM DTT , 1 µg BSA , 0 . 5 ng radiolabeled duplex probe and 2 or 4 µL in vitro translation mix in a total volume of 20 µL . 1/10 volume heparin ( 50 mg/mL ) was added just before loading the binding reaction . For control reticulocyte lysate alone together with radiolabeled duplex probe was used and did not result in any shift ( data not shown ) . Binding reactions were performed on ice for ten minutes and complexes were resolved on a 5% native acrylamide ( 37 . 5∶1 ) gel in 0 . 5 X TBE and then directly subjected to autoradiography . Total RNA was extracted from 9KbDmrt1bYprom::Dmrt1bY::GFP::Dmrt1bY3′UTR transgenic fish ( Carbio genetic background ) using the TRIZOL reagent ( Invitrogen ) according to the supplier's recommendation . After DNase treatment , reverse transcription was done with 2 micrograms total RNA using RevertAid First Strand Synthesis kit ( Fermentas ) and random primers . Real-time quantitative PCR was carried out with SYBR Green reagents and amplifications were detected with an i-Cycler ( Biorad ) . All results are averages of at least two independent RT reactions and 2–5 PCR reactions from each RT reaction using each time three set of primer combination ( DMTYk: 5′-CCTTCTTCCCCAGCAGCCT-3′/eGFP3: 5′-AGTCGTGCTGCTTCATGTGGTC-3′; DMTYa2: 5′-CGACTCCATGAGCAGTGAAA-3′/eGFP3: 5′-AGTCGTGCTGCTTCATGTGGTC-3′; DMTYa2: 5′-CGACTCCATGAGCAGTGAAA-3′/eGFP5: 5′-GAACTTCAGGGTCAGCTTGC-3′ . Error bars represent the standard deviation of the mean . Relative expression levels ( according to the equation 2–DeltaCT ) were calculated after correction of expression of elongation factor 1 alpha ( elf1alpha ) and brain expression was set to 1 as a reference . For in vivo chromatin immunoprecipitation , the EpiQuik Tissue Chromatin Immunoprecipitation kit ( Epigentek ) was utilized according to the manufacturers instructions , using testis tissue samples either from dmrt1bY::GFP or deltadmrt1bY::GFP transgenic fish ( 20 testes for each ) and GFP antibody ( Upstate ) for immunoprecipitation . After immunoprecipitation [ ( Izanagi element Dmrt1bYspeF003 ) 5′-TCCGGTCTCTCCGGCGTGTGG-3′/ ( Izanagi element Dmrt1bYspeR00 ) 5′-TTGTAAGAGGACCTGCAACAATG-3′; ( Izanagi element F01 ) 5′-CTATCTTGGTGAGGTCGACGATGCC-3′/ ( Izanagi element R01 ) 5′-AATTTAAATTACATGTCAAAGAGGTC-3′; ( Dmrt1bYCtrF04 ) 5′-GTTCTGACTTTCAGCGTCTCACCTG-3′/ ( Dmrt1bYCtrR04 ) 5′-GGTTCTGGTCCAAATCTGTCAGAAG-3′] primer sets were used for enrichment quantification by real-time PCR . For the generation of stable transgenic lines the meganuclease protocol [59] was used . Briefly , approximately 15–20 pg of total vector DNA in a volume of 500 pl injection solution containing I-SceI meganuclease was injected into the cytoplasm of one cell stage medaka embryos ( Carbio strain ) . Adult F0 fish were mated to each other and the offspring was tested for the presence of the transgene by PCR from pooled hatchlings . Siblings from positive F1 fish were raised to adulthood and tested by PCR from dorsal fin clips as described [60] . Identically to the transgenic line expressing the Dmrt1bY protein fused to GFP [61] a second line lacking the Dmrt1bY DNA binding domain ( DM-domain between aminoacids 10 and 78 ) was established . These two lines were used for in vivo Chromatin Immunoprecipitation . Similarly , for in vivo Dmrt1bY promoter activity quantification another transgenic line expressing a 9Kbdmrt1bYprom driven dmrt1bY::GFP fusion protein was created . Dmrt1a prom::GFP transgenic medaka was generated following the BAC transgenic method [62] . The BAC clone including dmrt1a genomic region , ola1-171C06 ( NCBI accession numbers; DE071574 and DE071575 ) was obtained from NBRP . The followings were the primers to amplify EGFP fragments for homologous recombination into the BAC clone; Forward: 5′ -tctgacatgagcaaggagaagcagggcaggccggttccggagggcccggcTCAACCGGTCGCCACCATGG-3′ Reverse:5′-ttcagcggagacacgaagccgtggttccggcagcgggagcacttgggcatcGTCGACCAGTTGGTGATTTTG-3′ . | Evolutionary innovations and adaptations often require rapid changes in gene regulation . Transposable elements constitute the most dynamic part of eukaryotic genomes . Insertions of transposable elements can influence the expression of surrounding genes by donating new regulatory elements . A longstanding hypothesis postulates that the dispersal of transposable elements may rewire regulatory links between genes , thereby changing regulatory networks and shuffling regulatory cascades . A regulatory hierarchy of remarkable plasticity is the sex determination cascade . In the course of animal evolution , new master regulators frequently replace the sex determination gene on top of the hierarchy . In the medaka fish , a duplicate of the dmrt1 transcription factor gene , dmrt1bY , has become the sex master regulator . Its ancestor dmrt1a , in contrast , has a downstream position in the sex determination cascade . We show that after the duplication of the dmrt1 gene , the new hierarchy has been established by the insertion of a transposable element into the regulatory region of the dmrt1bY gene on the sex chromosome . This transposable element , harboring a Dmrt1 binding site , enables the self- and cross-regulation of dmrt1bY expression by Dmrt1 proteins . Our study therefore provides strong evidence for the important role of transposable elements in the rewiring of gene regulatory networks . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/gene",
"expression",
"evolutionary",
"biology/genomics",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics"
]
| 2010 | Transcriptional Rewiring of the Sex Determining dmrt1 Gene Duplicate by Transposable Elements |
Synthetic biology efforts have largely focused on small engineered gene networks , yet understanding how to integrate multiple synthetic modules and interface them with endogenous pathways remains a challenge . Here we present the design , system integration , and analysis of several large scale synthetic gene circuits for artificial tissue homeostasis . Diabetes therapy represents a possible application for engineered homeostasis , where genetically programmed stem cells maintain a steady population of β-cells despite continuous turnover . We develop a new iterative process that incorporates modular design principles with hierarchical performance optimization targeted for environments with uncertainty and incomplete information . We employ theoretical analysis and computational simulations of multicellular reaction/diffusion models to design and understand system behavior , and find that certain features often associated with robustness ( e . g . , multicellular synchronization and noise attenuation ) are actually detrimental for tissue homeostasis . We overcome these problems by engineering a new class of genetic modules for ‘synthetic cellular heterogeneity’ that function to generate beneficial population diversity . We design two such modules ( an asynchronous genetic oscillator and a signaling throttle mechanism ) , demonstrate their capacity for enhancing robust control , and provide guidance for experimental implementation with various computational techniques . We found that designing modules for synthetic heterogeneity can be complex , and in general requires a framework for non-linear and multifactorial analysis . Consequently , we adapt a ‘phenotypic sensitivity analysis’ method to determine how functional module behaviors combine to achieve optimal system performance . We ultimately combine this analysis with Bayesian network inference to extract critical , causal relationships between a module's biochemical rate-constants , its high level functional behavior in isolation , and its impact on overall system performance once integrated .
As potential solutions for this problem , we propose several increasingly robust variants of a synthetic gene network that are designed to maintain a steady level of -cells despite normal cell death and constant destruction of the -cells by the immune system . The synthetic gene networks continuously direct proliferation , quiescence , and stem cell differentiation into insulin producing -cells as needed ( Figure 1A ) . The resulting engineered circuits may be employed to regulate tissue homeostasis both in vitro where the cell culture is removed from natural cues , and in vivo when natural systems fail or tissue is ectopically transplanted ( for example , the Edmonton protocol involves implanting pancreatic islets including -cells to the liver [24] ) . The efforts described here are based on encouraging genetic engineering accomplishments that have demonstrated population control of bacteria and yeast [12] , [25] , mammalian cell proliferation [26] , and stem cell differentiation [27] , [28] . To mitigate some of the uncertainties involved in system construction , we restricted our designs to use only genetic parts and modules that have already been demonstrated experimentally . These include engineered cell-cell communication to determine population densities , a toggle switch , an oscillator , and a multi-input AND gate . To gain a detailed understanding of our proposed synthetic gene networks , we carried out theoretical analysis and computational simulations using Ordinary Differential Equations ( ODE's ) , Langevin , and Gillespie algorithms . The analysis revealed that while simple modular composition was useful for initial system design , various factors such as stochastic effects , feedback control , and module interdependence significantly impacted system function and hence had to be taken into account when evaluating system designs . Strikingly , we observed that system features typically associated with robustness , including cell-synchronization , noise attenuation , and rapid signal processing destabilized our systems . To overcome these problems , we propose and analyze mechanisms that generate population diversity , and through this symmetry breaking facilitate proportionate and homeostatic system response to population-wide cues . Endogenous mechanisms of cellular heterogeneity have been previously observed in many physiological processes , including differentiation [29] . In the synthetic biology context , however , these mechanisms may be either unavailable for integration into the synthetic genetic circuit or too poorly understood to fully utilize . As a result , we forward engineer modules to generate synthetic cellular heterogeneity . For example , we incorporate an asynchronous oscillator module into the design as an engineered generator of intrinsic variability . Ultimately , our analysis indicates that such modules greatly improve homeostatic robustness among an isogenic population of cells , and we identify several examples of natural analogs . We found that the design and optimization of modules for synthetic heterogeneity is both non-intuitive and multifactorial , and in general requires a framework for non-linear and multivariate analysis . For example , with the asynchronous oscillator , we could not a priori define a simple objective or ideal ‘phenotype’ since oscillator properties such as period , dynamic range , and asynchronicity affected overall system performance in complex and interdependent manners . Furthermore , even if ideal module phenotypes are known , understanding the physical parameters required to achieve such phenotypes also represents a challenge . To address these issues , we developed a new framework using a hierarchy of computational tools to understand the optimal phenotypic and physical characteristics of the synthetic heterogeneity modules with respect to overall system behavior . We developed a ‘phenotypic sensitivity analysis’ method to determine how functional module behaviors combine to achieve optimal system performance . Parametric sensitivity analysis then captures the dependency of a module's phenotypes on its underlying physical rate constants . Ultimately , we integrated both analyses using Bayesian network inference to extract critical , causal relationships between a module's biochemical rate constants , its high level functional behavior in isolation , and its impact on overall system performance once integrated . Importantly , we anticipate that our hierarchical optimization strategy prescribes directions for system design that readily apply to experimental systems facing high degrees of uncertainty in rate constants and cellular environment . We designed and modeled an artificial tissue homeostasis system where a population of self-renewing stem cells grow and differentiate in a regulated manner to sustain a steady population of adult cells which , in this case , are insulin-producing -cells ( Figure 1A ) . Here we present four iterations of system design , analysis , and redesign with increased sophistication for improved robustness in controlling tissue homeostasis ( Figure 1B ) . The initial model for artificial tissue homeostasis ( System 1 ) comprises four integrated modules , and is analyzed using ODE simulation and global stability analysis . We incorporate a toggle switch in System 2 to minimize undesired -cell population fluctuations observed in System 1 , and analyze the improved design using stochastic differential equations ( SDEs ) . Although System 2 represents an improvement , its homogeneous response to commitment cues results in poor performance , thereby motivating the incorporation of an oscillator module and a throttle module for Systems 3 and 4 , respectively . Using SDE simulations , we optimize these modules and their integration into the full system . Throughout the discussion , we focus on several aspects of system design , including module integration , optimization of rate constants for individual modules , and optimization of module phenotypic behaviors .
Simple mathematical analysis suggested that feedback regulation between the two populations of stem cells and adult cells was necessary for robust homeostatic control , and recent work has explored the essential role of feedback control in stem cell biology ( Text S1 , Sec . 2 . 1 , [30] ) . In all alternative system designs presented in this manuscript , we implemented feedback control through artificial cell-cell communication pathways . Our first design , System 1 , allows differentiation only with a high density of stem cells and a low density of -cells ( Figure 2A ) . The “Stem Cell Population Control” ( SPC ) module allows for differentiation only when the population density of self-renewing cells lies above some threshold . We also designed the SPC to suppress proliferation through the expression of a growth arrest factor ( GAF ) , currently under development in the Weiss lab . The “-Cell Population Control” ( BPC ) module produces high output and inhibits differentiation when the density of -cells reaches a threshold ( Figure 2A ) . We based the cell-cell communication systems in the SPC and BPC modules on previously described communication systems [7]–[9] . As a proof of concept , Supplementary Figure S1 A–B presents results for a signal-receiver circuit based on the LuxR protein that responds to 3-oxo-hexanoyl-homoserine lactone ( 3OC6HSL ) , that has been experimentally implemented in human embryonic kidney ( HEK293 ) cells . We model stem cell differentiation as a multistage process that can take several weeks to complete [31] . For example , directed in vitro differentiation of hES cells into insulin-producing cells involves stepwise administration of growth factors to first induce endodermal cell fate , followed by pancreatic specialization , expansion , and maturation [32] . This general process is modeled by four cell types: stem cells ( population size ) grow with a constant division rate . Upon maturation , they proceed through two intermediate populations of endodermic ( ) and pancreatic ( ) cells before becoming -cells ( ) , which die at a constant rate . We describe the sequential maturation of into , , and as first-order reactions with rates , , and . Feedback terms are modeled as Hill functions , where and represent the SPC and BPC module thresholds , respectively . ( 1 ) The differentiation process is generally long in vivo ( e . g . , 20 days [32] . For System 1 , such delay in the feedback could induce undesirable oscillations ( Figure 2B–C ) . As a result , System 1 failed to maintain homeostasis for a large range of parameter values ( Text S1 , Sec . 2 ) . The integration of several network modules presents a challenge on multiple levels , especially in the context of uncertain biological environments and complex module dynamics . In the following sections , we introduce a framework composed of computational modeling and analysis techniques that addresses these issues in optimizing Systems 2 , 3 and 4 . We first study overall system robustness to external parameters such as cell survival dynamics , and introduce time-scale analysis as a method for guiding module integration . We then optimize the population control module using a novel ‘clustered sensitivity analysis’ to comprehend global patterns of parametric sensitivity in the context of a detailed biochemical model . Finally , we analyze the synthetic heterogeneity modules with an approach that focuses on module phenotype rather than rate constants alone . Comparisons among the different system architectures ultimately provide guidance for experimental optimization .
In this work , we engineer mechanisms of robust control using synthetic generators of heterogeneity , and use a multi-faceted computational framework for design and optimization in the context of a relatively large-scale synthetic gene network . As a case study we chose tissue homeostasis control where individual cell decisions need to be coordinated to obtain desired multi-cellular behavior . To tackle this complex problem , we used top-down decomposition , achieving the overall task through the creation of interconnected modules , where each module has its own specific objective . Throughout this hierarchical optimization process we used different modeling approaches ( population-based , Langevin and Gillespie simulations , see Figure 1B ) , while ensuring that the population-based results are consistent between the models ( Supplementary Figure S17 ) . We designed System 1 by coupling four modules together , and simulated this system using a simplified ODE model . Computational analysis elucidated properties of global stability and demarcated regimes of steady vs . oscillatory homeostatic behavior in general tissue homeostasis systems . Analogous oscillatory homeostatic behavior from delayed feedback has been observed in natural mammalian systems , for example with hematopoiesis [40] and bacterial biofilms [41] . To mitigate the problem of population level oscillations , we created System 2 which includes a toggle switch module to implement faster feedback ( Supplementary Table S9 ) . Of note , various natural cell types regulate proliferation and differentiation by a switch similar in principle to that used in our system [42] . Analysis of System 2 using a stochastic Langevin model revealed how population-wide communication signals can be highly destabilizing to homeostasis , leading us to two new system designs . For Systems 3 and 4 , the addition of the oscillator or the throttle module , respectively , provides more robust performance compared to System 2 ( Figure 5 ) because these systems are less dependent on precise parameter values and are able to maintain sufficient population heterogeneity at lower levels of intrinsic molecular noise ( Supplementary Table S9 ) . Alternative mechanisms for generating population heterogeneity may exist . For example , the AND gate in System 2 could have been coupled with endogenously heterogeneous biological behavior such as Nanog expression ( discussed below ) [43] . Nonetheless , we chose to focus on the oscillator and throttle because they do not rely on potentially unpredictable endogenous mechanisms that would complicate computational modeling , and they represent two substantially distinct mechanisms for generating heterogeneity . The design and analysis methods developed in this work attempt to identify relationships between rate constants , module phenotypes , and overall system performance , while maintaining an appreciation for the high degree of uncertainty and incomplete system knowledge in the experimental setting . For example , relating overall system performance directly to phenomenological definitions of module behavior frees the analysis from constraints to a particular module architecture or set of rate constants . Nonetheless , when more detailed information is desired we can apply global optimization strategies to capture patterns of parametric sensitivity that remain consistent across a broad range of rate constant values . For example , our analysis of the cell-cell communication module used a detailed biochemical reaction model with a large number of unknown rate constants . This level of granularity allowed us to analyze hysteretic response , which is not possible in the more abstract models . Ultimately , we addressed uncertainty by employing a novel technique , clustered sensitivity analysis , that revealed distinct patterns of relative parametric sensitivity for hysteresis that persisted across a wide range of rate constants . Previous reports have shown that bistability and hysteretic responses exist for both natural and engineered bacterial QS systems [44] , [45] , and in this work such bistability drives undesired oscillations . Accordingly , we designed the population control module to avoid hysteretic response and identified specific properties affecting hysteresis in our system . The synthetic heterogeneity modules in our systems display complex and multivariate behaviors that depend on the cooperative influence of multiple rate constants . Since existing experimental and computational biological circuit optimization methods do not scale well with system complexity , we decomposed the analysis and optimization processes for Systems 3 and 4 by characterizing modules first in isolation and then by relating their phenotypes to the performance of the overall system . We correlated module phenotypic behaviors with overall system performance , and found several significant correlations that were non-intuitive . Similarly , we identified dependencies between particular rate constants and the ability to maintain homeostasis . While Systems 3 and 4 exhibited comparable overall performances , further analyses revealed several distinguishing strengths and weaknesses ( Supplementary Table S9 ) . For example , the oscillator in System 3 appears to insulate modules from each other , while the throttle mechanism in System 4 amplifies their coupling strength ( Figure 5 C–F ) . Our results suggest that the oscillator may mitigate problems associated with module integration , at least with respect to matching dynamics . However , the throttle mechanism is likely to be better suited for toggle switches with slow switching times ( similar to the one we report on experimentally in Text S1 , Sec . 1 ) . At a high level , our work describes strategies to exploit stochastic effects for enhancing stability of tissue homeostasis . This concept has been recently explored in a number of reports emphasizing the role probabilistic strategies play in natural mechanisms of cell-decision processing , including differentiation [29] , [46] , [47] . Furthermore , attempts have been made at engineering inherently stochastic processes for functions such as enhanced cellular reprogramming into induced pluripotent stem ( iPS ) cells [48] . Nonetheless , to our knowledge no efforts have yet been made that combine advances in synthetic biology with an appreciation of stochastic processes to engineer homeostatic tissue from isogenic cellular populations . The asynchronous oscillator stabilizes our system by generating population heterogeneity during conditions of environmental homogeneity and exogenous perturbation . Among natural systems , recent work has highlighted the role multistable feedback systems and stochastic switching play in appropriately priming cells for differentiation [49] . For example , evidence indicates the Nanog-Sox2-Oct4 network functions in part to generate population diversity by stochastically interrupting differentiation signals . Oscillators have been described as mediating cell-decisions in other biological systems , for example with p53 and NF-B oscillations in response to DNA damage or other stimulation . These oscillations are hypothesized to enable discrete single-cell decisions to achieve a proportionate population-wide response [50] . Intrinsic noise generated by the oscillator also affects spatiotemporal clustering in our system ( Supplementary Figures S18 B , E and S19 ) and natural analogues of this phenomenon exist . For example , non-genetic sources of cell-cell variability can cause recently divided cells to react more similarly to pharmacological treatment [51] . Similarly , lateral inhibition as proposed in the throttle mechanism of System 4 has also been observed in biological systems , for example in pattern formation [52] , segmentation [53] or in the Notch signaling pathway [54] . Consistent with these studies , our spatial simulations show strong bias towards closely spaced alternate cell types in System 4 ( Supplementary Figure S18 C ) . Our optimization process , as well as the different biological examples described above , aim at seemingly contradictory objectives: information has to be processed faithfully from the population control modules to a commitment signal while , at the same time , stochasticity has to be amplified to generate heterogeneity . To achieve the first objective , several of our modules exhibit digital-like behavior , allowing us to effectively match components such that downstream modules react appropriately and with relative certainty to changes in upstream module output , attenuating the effects of noise . At the same time , to generate population heterogeneity , we exploit stochasticity by amplifying its effects in nonlinear modules operating in a transient regime . As a consequence , our modules are optimized to exhibit nonlinear responses to their inputs and , depending on the objective of the module , are tuned to work far from the transition regions for robust processing of information , or near the transition region where the response is highly sensitive to stochastic effects and hence efficiently generates heterogeneity . We present here an integrated framework for forward-engineering large scale synthetic genetic circuits that combines several distinct computational approaches , and demonstrate its application to the design , analysis , and optimization of systems for controlling artificial tissue homeostasis . This framework represents a conceptual advancement for guiding experimental implementation by introducing hierarchical strategies that coordinate detailed biochemical models with modular phenotypes and optimization of module integration , all while considering parametric uncertainty and incomplete knowledge of the underlying biological context . With regard to methods development , future work may consider how to incorporate iterations of computational design with stepwise experimental implementation . Experiments could be designed to determine rate constants or high-level properties such as module phenotypes that most critically impact system performance , according to the computational modeling . Future work may also explore the limits of design automation . Network-level modeling could benefit from an integration with molecular modeling for directed optimization of molecular rate constants . Importantly , the modular design principles described in this work have been developed in part to facilitate redesign for improved performance or alternative applications . Artificial homeostasis systems have a range of potential applications in lower organisms , including co-culture systems for biosynthetic chemical production [55] , controlled microbial homeostasis for environmental applications [56] , and maintenance of microbial bio-sensors [57] . Medical applications may include a range of stem cell therapies currently being researched for treatment of degenerative diseases and traumatic injuries [58] , [59] . Forward-engineering efforts such as those presented here may elucidate roles of heterogeneity and homeostasis in diseases such as cancer , where tumor diversity potentially contributes to chemoresistance and metastasis [60] . Beyond guiding experimental implementation of the systems described herein , we believe the design principles and control motifs revealed by our analyses may offer more general insights into the role of population heterogeneity for robust behavior , with implications for both synthetic and systems biology .
Experimental implementations of the toggle switch and the cell-cell communication receiver were performed using immortalized human embryonic kidney cells ( HEK293FT; Invitrogen ) , further discussed in the Text S1 , Sec . 1 . Computational methods and models utilized a variety of software platforms . We examined Systems 1–2 using ODE stability analyses and simulations ( described in Text S1 , Sec . 2 ) , performed in Maple ( Maplesoft; Waterloo , ON , Canada ) and Matlab ( MathWorks; Natick , MA ) . Systems 2–4 were analyzed using stochastic simulations . Langevin chemical simulations [35] ( Text S1 , Sec . 3 ) were performed using custom C++ code based on the 2-stage stochastic Runge-Kutta integration method with optimized parameters as described in [61] . All equations and parameters are reported in the Text S1 , Sec . 3 and Table S1 , respectively . In addition to Langevin simulations , Gillespie simulations ( Figure 6 , Text S1 , Sec . 4 ) were implemented for Systems 2–3 using a standard rate-equation approach and the Gibson-modified Gillespie algorithm [62] . Transition rates were chosen to match the dynamics of the Langevin implementations ( Table S3 ) . For both the Langevin and Gillespie simulations , systems were described using a previously reported multicellular spatiotemporal simulation environment [15] , [63] . The simulation platform ( written in C++ ) tracks the temporal evolution of intracellular reactions within individual cells that grow and die on a 2D grid . Furthermore , the platform monitors the spatiotemporal evolution of the cells themselves and extracellular signaling molecules that diffuse among them ( Text S1 , Sec . 2 and 4 ) . We utilized a two-compartment ODE model of the UPC module for the GA optimizations ( Text S1 , Sec . 5 . 3 and Table S7 ) , and implemented the GA in C++ using a distributed computing cluster ( n = 40 processor nodes ) . RS-HDMR ( Text S1 , Sec . 5 . 1 ) was implemented as reported elsewhere [37] , [64] . A version of RS-HDMR [64] can be found online at http://www . aerodyne . com ( free for academic users ) . Partial least squares regression and support vector machine classification ( Text S1 , Sec . 6 . 2 ) were implemented using standard Matlab functions , and Bayesian network inference ( Text S1 , Sec . 5 . 2 . 4 ) was performed in Matlab using previously described software [65] . | Over the last decade several relatively small synthetic gene networks have been successfully implemented and characterized , including oscillators , toggle switches , and intercellular communication systems . However , the ability to engineer large-scale synthetic gene networks for controlling multicellular systems with predictable and robust behavior remains a challenge . Here we present a novel combination of computational methods to aid the iterative design and optimization of such synthetic biological systems . We apply these methods to the design and analysis of an artificial tissue homeostasis system that exhibits coordinated control of cellular proliferation , differentiation , and cell-death . Achieving artificial tissue homeostasis would be therapeutically relevant for diseases such as Type I diabetes , for instance by transplanting genetically engineered stem cells that stably maintain populations of insulin-producing beta-cells despite normal cell death and autoimmune attacks . To manage complexity in the design process , we employ principles of logic abstraction and modularity and investigate their limits in biological networks . In this work , we find factors often associated with robustness ( e . g . , multicellular synchronization and noise attenuation ) to be actually detrimental , and overcome these problems by engineering genetic modules that generate beneficial population heterogeneity . A combination of computational methods elucidates how these modules function to enhance robust control , and provides guidance for experimental implementation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"systems",
"biology",
"synthetic",
"biology",
"biology",
"computational",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
]
| 2012 | Modular Design of Artificial Tissue Homeostasis: Robust Control through Synthetic Cellular Heterogeneity |
Consecutive repetition of actions is common in behavioral sequences . Although integration of sensory feedback with internal motor programs is important for sequence generation , if and how feedback contributes to repetitive actions is poorly understood . Here we study how auditory feedback contributes to generating repetitive syllable sequences in songbirds . We propose that auditory signals provide positive feedback to ongoing motor commands , but this influence decays as feedback weakens from response adaptation during syllable repetitions . Computational models show that this mechanism explains repeat distributions observed in Bengalese finch song . We experimentally confirmed two predictions of this mechanism in Bengalese finches: removal of auditory feedback by deafening reduces syllable repetitions; and neural responses to auditory playback of repeated syllable sequences gradually adapt in sensory-motor nucleus HVC . Together , our results implicate a positive auditory-feedback loop with adaptation in generating repetitive vocalizations , and suggest sensory adaptation is important for feedback control of motor sequences .
Many complex behaviors—human speech , playing a piano , or birdsong—consist of a set of discrete actions that can be flexibly organized into variable sequences [1–3] . A feature of many variably sequenced behaviors is the occurrence of repetitive sub-sequences of the same action . Examples include trills in music , repeated syllables in birdsong , and syllable/sound repetitions in stuttered speech . A central issue in understanding how nervous systems generate complex sequences is the role of sensory feedback versus internal motor programs [4] ( Fig 1a ) . At one extreme ( the serial chaining framework ) , the sensory feedback from one action initiates the next action in the sequence; therefore sensory feedback is critical for sequencing the actions [5 , 6] . However , because of the delays in both motor and sensory processing in nervous systems , it has been argued that a sequence generation mechanism relying solely on sensory feedback would be too slow to account for the execution of fast sequences such as typing and speech [1] . At the other extreme , sequences are generated by internal motor programs controlling sequence production without the use of sensory feedback [7–9] . However , there is ample evidence that sensory feedback can affect action sequences [10–14] . Despite the ubiquity of sequencing in behavior , the neural mechanisms of how sensory feedback interacts with internal motor programs to influence discrete actions remain largely unexplored . Here , we study the role of sensory feedback in the production of repetitive vocal sequences using the Bengalese finch as a model system . The Bengalese finch produces songs composed of discrete acoustic events , termed syllables , organized into variable sequences ( Fig 1b ) . However , sequence production is not random [15] , as the transition probabilities between syllables are statistically reproducible across time [13 , 16] . A prominent feature of the songs of several songbird species , including the Bengalese finch , is syllable repetition [15 , 17–21] ( e . g . ‘b’ in Fig 1b ) . For a given repeated syllable , the number of consecutively produced repeats ( the repeat number ) varies . The first order Markov process , in which the probability of repeating a syllable is constant , is a simple model for generating syllable repetitions . Such a process produces a monotonically decreasing distribution of repeat numbers , with the most probable repeat number ( peak repeat number ) being one ( Fig 1c , black curve ) . Indeed , many repeated syllables in the songs of the Bengalese finch do have such distributions [20] . However , there are also repeated syllables that violate the predictions of the Markov process . These syllables are typically long repeated , and their distributions of repeat numbers are peaked , with the most probable repeat number being much greater than one [20–23] ( Fig 1c , red curve ) . In the songs of the Bengalese finch , the transition probabilities between syllables are altered shortly after deafening [24 , 25] or in real-time by delayed auditory feedback [13] , demonstrating that disturbing auditory feedback can disturb sequence generation . Songbirds are prominent models for studying the neural basis of complex sequence production . Experimental data from sensory-motor song nucleus HVC ( proper name ) of singing zebra finches have led to neural network models of the internal motor program for sequence generation that instantiate first-order Markov processes [26] . This suggests that additional mechanisms contribute to the generation of non-Markovian distributions of repeat numbers [20 , 21 , 26] . One possibility is that , because of sensory-motor delays , auditory feedback from the previous syllable interacts with the internal motor program to contribute to the transition dynamics for subsequent syllables [13 , 14 , 27 , 28] . For repeated syllables , we hypothesized that the interaction of auditory-feedback and ongoing motor activity forms a positive-feedback loop that contributes to sustaining syllable repetition beyond the predictions of a Markov process ( Fig 1a ) . However , such positive-feedback architectures are inherently unstable , prone to indefinite repetition ( i . e . perseveration ) . Across sensory modalities , a common feature of sensory responses to repeated presentations of identical physical stimuli is a gradual decrease of response magnitude ( i . e . response adaptation ) [29] . We therefore hypothesized that auditory inputs are subject to response adaptation , which gradually reduces the strength of the positive feedback loop over time . Thus , an auditory-motor feedback loop with response adaptation is predicted to contribute to the generation of non-Makovian repeated syllable sequences by both pushing repeat counts beyond the expectations of a Markov process and simultaneously preventing indefinite repetitions of the syllable . We tested these hypotheses using computational modeling combined with behavioral and electrophysiological experiments .
The critical features of our framework for repeat generation are: ( 1 ) the population of neurons generating a repeated syllable receives a source of excitatory input in addition to the recurrent excitation from the sequencing network , and ( 2 ) the strength of this input adapts over time during repeat generation . For concreteness , we instantiate this framework as a ‘branched-chain’ network with adapting auditory feedback , and place this network in nucleus HVC . In songbirds , HVC has been proposed to contain an internal motor program for the generation of song sequences [26 , 30–35] . HVC sends descending motor commands for song timing to nucleus RA ( the robust nucleus of the arcopallium ) , which in turn projects to brainstem areas controlling the vocal organs [36 , 37] ( Fig 2a ) . HVC also receives input through internal feedback loops from the brainstem [38] , via Uva ( nucleus uvaeformis ) and NIf ( the interfacial nucleus of the nidopallium ) [39] . Experiments in the zebra finch have shown sparse sequential firing of the RA projecting HVC neurons ( HVCRA ) during singing [30 , 31 , 35] . This has led to the hypothesis that the motor program for sequence production in HVC includes sequential “chaining” of activity , in which populations of HVCRA neurons responsible for generating a syllable drive the neuronal populations that generate subsequent syllables either directly within HVC or through the internal feedback loop [31 , 34 , 35 , 40 , 41] ( Fig 2b ) . Our model for generating syllable sequences starts with such a synaptic chain framework . The details of this model have been described previously [26] and are summarized in Materials and Methods . In synaptic chain models , each syllable is encoded in a chain network of HVCRA neurons ( Fig 2b ) . Spike propagation through the chain produces the encoded syllable by driving appropriate RA neurons . To generate variable syllable transitions , the syllable-chains are connected into branching patterns . At a branch point , syllable-chains compete with each other through a winner-take all mechanism mediated by the inhibitory HVC interneurons ( HVCI ) , allowing only one branch to continue the spike propagation . The selection is probabilistic due to intrinsic neuronal noise , which provides a source of stochasticity in the winner-take-all competition ( Fig 2b ) . In this model , syllable repetition is generated by connecting the syllable-chains to themselves at the branching points [26 , 34] . In branched chain networks , the transitions between the syllable-chains are largely Markovian , and for repeating syllables this implies that repeat number distributions should be a decreasing function of the repeat number—in particular , the most probable ( or “peak” ) repeat number will be one [26] ( Fig 1c ) . However , many repeated syllables in Bengalese finch song have repeat distributions that are highly non-Markovian , with peak repeat numbers much larger than one [20–23] . This implies additional processes beyond synaptic chains contribute to generating non-Markovian repeated sequences . Here we incorporate auditory feedback into the branching chain network model and show that , when this feedback is strong and adapting , non-Markovian repeat distributions emerge . In HVC , as in many sensory-motor systems , including the human speech system [42 , 43] , the same neuronal populations that are responsible for the generation of the behavior also respond to the sensory consequences of that behavior , i . e . the bird’s own song ( BOS ) [14 , 44–46] . HVC receives much of its auditory input from NIf [47–50] , which can provide real-time auditory feedback during singing ( Fig 2a ) [51] . However , because of the time it takes to propagate motor commands to the periphery ( 30–50 ms ) and process the subsequent auditory signals ( 15–20 ms ) ( Fig 2a ) , auditory feedback is necessarily delayed relative to the motor activity that generated it [1 , 13 , 14 , 28] . This sensory-motor delay for HVC ( 45–70 ms ) is on the order of the duration of a syllable , making it possible for auditory feedback to influence HVC motor programs and the transition dynamics between syllables [13 , 14 , 27] ( Fig 2a ) . We first tested the feasibility of this mechanism using biophysically detailed neural network models . To illustrate this model , we focus on generating sequences of the form ‘abnc’ , where syllable ‘a’ transitions to syllable ‘b’ , ‘b’ repeats a variable number of times ( n ) , and transitions to ‘c’ ( e . g . ‘abbbbbbbc’ ) . For concreteness , we model the adapting input as an auditory feedback signal to the network , though in principle this adapting input could reflect recurrent circuit-activity that is non-sensory . To incorporate auditory feedback into the previous model , each HVCRA neuron in chain-b is contacted by excitatory synapses carrying auditory inputs triggered by the production of syllable ‘b’ ( Fig 2c ) . We assume that the auditory synapses are made by axons from NIf , which is a major source of auditory input to HVC [47–50] and is selective to BOS [49] . When auditory feedback is present , the auditory synapses receive spikes from a Poisson process , assumed to be from the population of NIf neurons responding to syllable ‘b’ ( Materials and Methods ) ( Fig 2c ) . The auditory synapses are subject to short-term synaptic depression , resulting in gradual adaptation of responses to repeated inputs [52 , 53] . Specifically , due to the synaptic depression , the average strength of the auditory inputs to chain-b decreases exponentially during the repeats of syllable ‘b’ ( Materials and Methods ) . In Fig 3 , we show results from an example network in which the auditory input to chain-b is strong and the spiking dynamics produce repeats of syllable ‘b’ with large repeat numbers . A spike raster for a standard single run of the network is shown in Fig 3a . Once spiking was initiated in chain-a ( through external current injection ) , spikes propagated through chain-a , and activated chain-b . Chain-b repeated a variable number of times before the spike activity exited to chain-c and stopped once it reached the end of chain-c . As chain-b continued to repeat , the synapses carrying the feedback signal weakened over time due to adaptation ( Fig 3b ) . Analyzing multiple trials , we find that the probability of chain-b transitioning to itself ( repeat probability ) also decreases over time , though the repeat probability is only meaningful at the transition times—i . e . when the activity reaches the end of chain-b ( Fig 3c ) . Examining the feedback strength at these transition times across the same trials allowed us to understand how the instantaneous feedback strength affects the repeat probability ( Fig 3d ) . Not surprisingly , we found that the repeat probability increases with the strengths of the auditory synapses . Repeat probability pr as a function of the feedback strength could be well fit with the sigmoidal function ( Fig 3d , red curve ) p r ( A ) = 1 - c 1 + η A ν , ( 1 ) where A > 0 represents the strength of the auditory synapses , η , ν > 0 are parameters controlling the shape of the curve , and 0 < c < 1 is a parameter for the repeat probability when there is no auditory feedback ( i . e . A = 0 ) , which is determined by the connection strengths of the network at the branching point . Note that , when the auditory input A = 0 , the repeat probability is pr = 1 − c , and conversely , as A is large , pr approaches 1 . Initially , the strong auditory feedback biases the network toward repeating and so the repeat probability is close to 1 . If the strong excitatory input resulting from auditory feedback were constant , the network would perseverate on repeating syllable ‘b’ indefinitely ( a result of the positive feedback loop ) . However , because of the short-term synaptic depression , the auditory input to chain-b when syllable ‘b’ repeats decreases exponentially over time ( Fig 3b , red line; time-constant of τ = 148 ms for this particular network ) . Even so , the repeat probability stays close to 1 as long as the auditory input is strong enough . Further weakening of the feedback reduces the repeat probability more significantly , making repeat-ending transitions to chain-c more likely . For this network , this process produced a repeat number distribution peaked at 6 , as shown in Fig 3e . These results demonstrate that branched-chain networks receiving adapting excitatory inputs can generate repeat distributions that are non-Markovian . The repeat number distributions from our network model can be described using a simple statistical model with a small number of parameters . In our network model , the gradual reduction of excitatory drive from auditory feedback as a syllable is repeated reduces the probability that the syllable transitions to itself , and thus reduces the repeat probability . Eq ( 1 ) describes the dependence of the repeat probability pr on the auditory input strength , A . The synaptic depression model tells us how A changes with time . Sampling this at the transition times describes how A changes with the repeat number , n . At the end of the nth repeat of the syllable , A reduces to A ( n ) = a 0 e - n T / τ , ( 2 ) where a0 is the initial strength of the auditory feedback , τ is the time constant of the input decay , and T is the duration of the syllable . Combining this with the dependence of the repeat probability on A , shown in Eq ( 1 ) , we find that the repeat probability after the nth repetition of the syllable is given by p r ( n ) = 1 - c 1 + η a 0 ν e - n ν T / τ = 1 - c 1 + a b n , ( 3 ) where a = η a 0 ν and b = e−νT/τ . Therefore , there are effectively three parameters ( a , b and c ) for how pr depends on n . We call Eq ( 3 ) the sigmoidal adaptation model of repeat probability . The network sequence dynamics can be represented with a state transition model , in which a single state corresponds to the repeating chain . The state can transition to itself with a probability pr ( n ) given by Eq ( 3 ) , or exit the state with probability 1 − pr ( n ) . This single state transition model can accurately fit the repeat number distributions generated by the network simulations with varying parameters , as shown in Fig 4a ( all fit errors below their respective benchmark errors , which characterize the fitting errors expected from the finiteness of the data set—see Materials and Methods ) . This model contains the Markov model and a previously described ‘geometric adaptation’ model [20] as special cases ( Materials and Methods ) . Both of these models fail to fit the simulated data , even when a large number of states/parameters are used ( Fig 4b and 4c ) . On the other hand , we have shown that the sigmoidal model provides an accurate fit with a single state and a small number of parameters . Therefore , relative to other statistical models , the single-state transition model with sigmoidal adaptation parsimoniously and accurately replicates the syllable repetition statistics of our network model . Using the single state transition model with sigmoidal adaptation , we explored how peak repeat numbers depend on the initial feedback strength and the adaptation strength ( defined by the related parameter , α , in the synaptic depression model , Materials and Methods ) ( Fig 4d ) . Here we see that , for a given adaptation strength , there is a threshold feedback strength at which the peak repeat number is greater than 1 , and this threshold increases with increasing adaptation strength . This demarcates the transition between Markovian ( peak repeat number = 1 ) and non-Markovian ( peak repeat number > 1 ) repeat distributions ( black-to-red transition in ( Fig 4d ) ) . Further increases in the feedback strength result in larger peak repeat numbers . Conversely , for a given feedback strength , increasing the adaptation strength results in a reduction of the peak repeat number . Together , these results demonstrate that a large range of peak repeat numbers can be generated through various combinations of feedback and adaptation strengths , and suggest that there is a threshold feedback strength required to produce non-Markovian repeat distributions . To see whether the non-Markovian repeat distributions generated with our network model can accurately describe syllable repeat number distributions of actual Bengalese finch songs , we recorded and analyzed the songs of 32 Bengalese finches . We identified the song syllables and obtained the syllable sequences ( Materials and Methods ) . Our data set contains more than 82 , 000 instances of 281 unique syllables , of which 71 are repeating syllables . Since the simulations of the network model are slow , we used the single state transition model with sigmoidal adaptation to fit the repeat number distributions for these syllables . As demonstrated above , the statistical model ( Eq ( 3 ) ) captures the essential features of our network model , and succinctly represents the repeat number distributions produced by the network simulations . In Fig 5a , we show six examples of Bengalese finch repeat count histograms ( grey bars ) with different peak repeat counts ( peak repeat count increases across plots i-vi . ) , and the best-fit model distributions ( red lines ) . These examples show a range of distribution peaks and shapes , from small peak numbers with long rightward tails ( i ) , to large peak numbers with tight , symmetric tails . Interestingly , we found that three repeated syllables ( out of 71 ) had clear double-peaked distributions , with a prominent peak at repeat number 1 and another peak far away ( two of which are displayed in panels ii and vi ) . These double peaked distributions cannot be explained with a single state transition model . A simple explanation is that the single peak and the broad peak are generated by two separate states ( or neural substrate ) , as postulated in Jin & Kozhevinov ( the “many-to-one mapping” from multiple chains in HVC to the same syllable type ) [20] . Here we removed the single peak at repeat number 1 for these three syllables and only analyzed the longer repeat parts . The state transition with sigmoidal adaptation model does an excellent job of fitting the wide variety of peaks and shapes of the repeat distributions found in the Bengalese finches . The results comparing the fit errors from the sigmoidal adaption model to benchmark errors across all 71 repeating syllables are shown in Fig 5b ( Materials and Methods; see also [20] ) . The vast majority of fit errors from the feedback adaptation model are below their respective benchmark errors ( 86% of fit errors below the benchmark error ) , demonstrating that the model does an excellent job of fitting the diverse shapes of Bengalese finch song repeat number distributions . Therefore , the single state transition model with sigmoidal adaptation , and by extension the branched-chain model with adaptive auditory feedback , can successfully describe the syllable repeat number distributions in Bengalese finch songs . In our framework , auditory feedback from the previous syllable arrives in HVC at a time appropriate to provide driving excitatory input to HVC neurons that generate the upcoming syllable . For repeated syllables , this creates a positive feedback loop which is responsible for generating peak repeat numbers greater than 1 ( adaptation drives the process to extinction ) . Therefore , a key prediction is that without auditory-feedback driven excitatory input , the peak-repeat number should shift toward 1 . To test this prediction , we deafened six Bengalese finches by bilateral removal of the cochlea , and analyzed the songs before and soon after they were deafened ( 2–4 days ) ( Materials and Methods ) . We found that deafening greatly reduces the peak repeat-counts . For example , in Fig 6a , we display spectrograms and rectified amplitude waveforms of the song from one bird prior to deafening ( top ) and soon after deafening ( 2–3 days post-deafening ) . We see that deafening reduces the number of times that the syllable ( red-dashed box ) is repeated . The time course of repeat generation from this bird is examined in more detail in Fig 6b , where we plot the median repeat counts per song of the syllable from Fig 6a before deafening ( black ) and after deafening ( red ) . Here we see that , even in the first songs recorded post-deafening , there is a marked decrease in the produced number of repeats . This data further exemplifies that repeat counts per song is generally stable across bouts of singing within a day both before and after deafening . Across days , repeat counts continued to slowly decline with time since deafening , though the co-occurrence of acoustic degradation of syllables makes these later effects difficult to interpret [24 , 54] . Nonetheless , the rapidity of the effect of deafening underscores the acute function of auditory feedback in the generation of repeated syllables . Similar results were seen across the other repeated syllables . Fig 6c shows the repeat number distributions for two additional birds before ( black ) and after ( red ) deafening . In these cases , deafening resulted in repeat number distributions that monotonically decayed . The peak repeat numbers pre and post deafening for all 19 syllables in our data set are presented in Fig 6d . Across the 19 repeated syllables from 6 birds , deafening significantly reduced the number of consecutively produced repeated syllables ( Fig 6d , p < 0 . 01 , sign-rank test , N = 19 , medians demarcated in red , overlapping points are vertically shifted ) , although there was variability in the effect magnitude: the effect of deafening appeared larger for the repeat with larger initial repeat number ( compare upper and lower panels of Fig 6c ) . This suggests that the degree to which deafening reduces peak repeat number depends on the initial repeat number . We examined the change in peak repeat number resulting from deafening as a function of the peak repeat number before deafening ( Fig 6e , red dots correspond to data from individual syllables , overlapping points are horizontally offset for visual display ) . We found that the magnitude of decrease in peak repeat numbers after deafening grows progressively larger for syllables with greater peak repeat numbers before deafening ( R2 = 0 . 81 , p < 10−7 , N = 19 ) . This suggests that repeated syllables with larger repeat numbers are progressively more dependent upon auditory feedback for repeat production . Interestingly , after two days of hearing loss , one of the deafened Bengalese finches in our experiments had a repeat that was minimally affected by deafening , and several birds retained peak repeat number around 2 , not all the way to 1 as predicted for a Markov process ( Fig 6d ) . None-the-less , these deafening results are consistent with the hypothesis that the generation of repeated syllables is driven , in-part , by a positive-feedback loop caused by excitatory auditory input during singing . A key prediction of the adaptive feedback model for repeat generation is that auditory responses of HVC neurons should decline over the course of repeated presentations of the same syllable . To test this hypothesis , we examined the properties of HVC auditory responses to repeated syllables in sedated birds ( Materials and Methods ) . An example recording from an HVC multi-unit site in response to playback of the bird’s own song ( BOS ) stimulus is presented in Fig 7a , which displays the stimulus oscillogram ( top ) , and the average spike rate in response to the stimulus ( bottom ) . Multiple renditions of the repeated syllable are demarcated by red-dashed boxes , and we see that the evoked HVC auditory responses to repeated versions of the same syllable gradually declined . The example presented above suggests that auditory responses to repeated presentations of the same syllable adapt over time . However , in the context of BOS stimuli , the natural variations that occur in syllable acoustics , inter-syllable gap timing , and in the identity of the preceding sequence , make it difficult to directly compare responses to different syllables in a repeated sequence . Therefore , to examine how responses to repeated syllables are affected by the length and identity of the preceding sequence , for each bird we constructed a stimulus set of long , pseudo-randomly ordered sequences of syllables ( 10 , 000 syllables in the stimulus , one prototype per unique syllable , median of all inter-syllable gaps used for each inter-syllable gap , derived from the corpus of each bird’s songs , Materials and Methods ) . This stimulus allows a systematic investigation of how auditory responses to acoustically identical syllables depend on the length and syllabic composition of the preceding sequence [28] . Auditory responses at 18 multi-unit recordings sites in HVC from 6 birds were collected for this data set , which contained 40 unique syllables . Of these 40 syllables , 6 syllables in 4 birds ( with 11 recording sites ) were found to naturally repeat . We used these stimuli to systematically examine how auditory responses to a repeated syllable depend on the number of preceding repeated syllables . We found that HVC auditory responses gradually declined to repeated presentations of the same syllable . In Fig 7b , for each uniquely repeated syllable ( different syllables are colored from grey-to-red with increasing max repeat number ) , we plot the average normalized auditory response ( mean ±s . e . across sites ) to that syllable ( e . g . ‘b’ ) as a function of the repeat number ( e . g . repeat number 5 corresponds to the last ‘b’ in ‘bbbbb’ ) . Across HVC recordings sites and repeated syllables , the response to the last syllable declined as the number of preceding repeated syllables increased ( R2 = 0 . 523 , p < 10−10 , N = 24 , slope = -5% ) . Thus , auditory responses to repeated syllables gradually adapt as the number of preceding repeated syllables increases , providing confirmation of a key functional mechanism of the network model . To generate non-Markovian repeat distributions , we have proposed that the sequence generation circuitry is driven , in part , by auditory feedback that provides excitatory drive to sensory-motor neurons that control sequencing . Specifically , auditory feedback from the previous syllable arrives in HVC at a time appropriate to provide driving excitatory input to neurons that generate the upcoming syllable . This predicts that if HVC auditory responses are positively modulated by sound amplitude , feedback associated with louder syllables should provide stronger drive to the motor units , and thus generate longer strings of repeated syllables for a given rate of adaptation . This logic is supported by the sigmoidal adaptation model , which predicts a threshold auditory feedback strength at which the peak repeat number becomes greater than one ( i . e . non-Markovian , Fig 4b ) . Behaviorally , this predicts that non-Markovian sequences of repeated syllables should be composed of the loudest syllables in the bird’s repertoire . We tested this behavioral prediction by comparing the amplitudes of Bengalese finch vocalizations based on their repeat structure . Fig 8a plots the rectified amplitude waveforms ( mean ±s . d . ) of a few consecutively produced repetitions of a non-Markovian repeated syllable ( black ) , a Markovian repeated syllable ( red ) , and ‘introductory’ note ( grey ) from one bird . The non-Markovian repeated syllable is qualitatively louder than the other repeated vocalizations in the birds’ repertoire . To quantitatively test this prediction , we measured the peak amplitude of the 281 unique syllables in our data set , and normalized this to the minimum peak amplitude across syllables ( Materials and Methods ) . We categorized each syllable in our data set according to whether it was an introductory note ( Intro ) , a non-repeated syllable ( NR: repeats = 0 ) , a Markovian repeated syllable ( MR: peak repeat number = 1 ) , or a non-Markovian repeated syllable ( nMR: peak repeat number > 1 ) . In Fig 8b , we plot the mean ±s . e . of the normalized peak amplitudes of these syllable groups across the data set . As exemplified by the data in Fig 8a , we found that non-Markovian repeated syllables were significantly louder than the other vocalizations in a bird’s repertoire ( ***: p < 10−3 , **: p < 10−2 , sign-rank test , Bonferroni corrected for m = 3 comparisons ) . Therefore , syllables with non-Markovian repeat distributions are typically the loudest vocalizations produced by a bird . If amplitude is a contributing factor to repeat generation , then HVC auditory responses should be positively modulated by syllable amplitude . However , previous work in the avian primary auditory system has found a population of neurons that is insensitive to sound intensity [55] , and amplitude normalized auditory responses have been utilized in previous models of sequence encoding in HVC auditory responses [56] . Therefore , we first examined whether auditory responses were positively modulated by syllable amplitude . To make recordings from different sites/birds comparable , we normalized both the syllable amplitudes ( relative to mean ) and auditory responses ( relative to minimum ) . The scatter plot in Fig 8c plots the normalized syllable amplitudes vs . the normalized auditory responses ( averaged across sites within a bird ) , for the 40 syllables in in our data set [28] . We found a modest but significant positive correlation between auditory responses and syllable amplitude ( R2 = 0 . 30;p < 10−3 , N = 40 syllables ) . We next examined whether the increased amplitude of repeated syllables resulted in increased HVC auditory response to these syllables . We performed a paired comparison of normalized auditory responses to non-repeated syllables ( NR ) and non-Markovian repeated syllables ( nMR ) at the 11 sites where auditory responses to repeated syllables were collected ( Fig 8d ) . We found that repeated syllables had significantly larger auditory responses than non-repeated syllables ( p < 0 . 01 , sign-rank test , N = 11 sites ) . Thus , HVC auditory responses are sensitive to syllable amplitude , and repeated syllables elicit larger auditory responses than non-repeated syllables , likely due to being the loudest syllables that a bird sings . Therefore , the strong auditory feedback associated with these loud repeated syllables may be a key contributor to their non-Markovian repeat distributions .
We have provided converging evidence that adapting auditory feedback directly contributes to the generation of long repetitive vocal sequences with non-Markovian repeat number distributions in the Bengalese finch . A branching chain network model with adapting auditory feedback to the repeating syllable-chains produces repeat number distributions similar to those observed in the Bengalese finch songs . From the network model we derive the sigmoidal adaptation model for repeat probability , and show that it reproduces the repeat distributions of both the branching chain network and Bengalese finch data . Removal of auditory feedback by deafening reduced the peak repeat number , confirming one of the key features of the proposed mechanism . Furthermore , recordings in the Bengalese finch HVC show that auditory responses of HVC adapt to repeated presentations of the same syllable , providing evidence for another key feature of the proposed mechanism . Finally , we found that non-Markovian repeated syllables are louder than other syllables and elicit stronger auditory responses , suggesting that a threshold auditory feedback magnitude is required to generate long strings of repeated syllables , in agreement with modeling results . Together , these results implicate an adapting , positive auditory-feedback loop in the generation of long repeated syllable sequences , and suggest that animals may directly use normal sensory-feedback signals to guide behavioral sequence generation with sensory adaptation preventing behaviorally deleterious perseveration . In our framework , a positive feedback loop to a repeating syllable provides strong excitatory drive to that syllable and sustains high repeat probability . The strength of this feedback gradually reduces while the syllable repeats , preventing the network from perseverating on the repeated syllable . The combination of strong , positive feedback and gradual adaptation allows the production of non-Markovian repeat number distributions in the branching chain networks . It should be emphasized that this feedback mechanism is not necessary for repeat syllables with Markovian repeat number distributions [20] . Such Markovian repeats are short , and can be simply generated with self connections in the branching chain network model without auditory feedback [26] . We have conceptualized the adapting feedback as short-term synaptic depression of the NIf to HVC synapses resulting from auditory feedback . However , neither the exact source of the feedback nor the mechanism generating the adaptation is critical for our model . Indeed , the adaptation of auditory responses could arise from a variety of pre- and/or post-synaptic mechanisms anywhere in the auditory pathway , such as in the auditory forebrain [57] , the auditory responses of NIf [47–50] or other auditory inputs to HVC such CM ( caudal mesopallium ) [58] , or within HVC itself . The biophysical origin of the auditory adaptation in HVC observed in our experiments remains to be determined . Our experiments showing the adaptation of auditory feedback for the repeated syllables were performed on passively listening birds . Future experiments on singing birds are required to see whether such adaptation occurs in the singing state . Previous experiments that deafened Bengalese finches showed that removal of auditory feedback has immediate impact on the song syntax of the Bengalese finch [24 , 25 , 54] . The main effect reported was the increased randomness in the syllable sequences . However , the impacts on syllable repeats was not analyzed . Our own deafening experiments showed that long repeated syllables are particularly vulnerable to loss of hearing , and their repeat number distributions shift close to Markov distributions two days after deafening . The Markovian repeats , on the other hand , were not affected as much . These new observations supports the idea that non-Markovian repeats rely more on auditory feedback than Markovian ones , as suggested by our computational model . However , it should be noted that the deafening results are consistent with our model but do not prove it . There could be alternative explanations , including possible systematic changes in the stress level , the arousal states , the neural circuits in the auditory and motor areas during the recovery from deafening . Future experiments that directly manipulate auditory feedback online in intact brain will help to further test our model . After two days of hearing loss , one of the deafened Bengalese finches in our experiments maintained peaked repeat number distributions , and several birds retained peaked repeat numbers around 2 , not all the way to 1 as predicted for a Markov process . One possible explanation is the existence of multiple chains that produce syllables with similar acoustic features [20] . Such a “many-to-one mapping” could produce residual non-Markovian features in the repeat number distribution after deafening . Another possibility is that there are several internal feedback loops to HVC within the song system that could contribute to repeating syllables . For example , there are direct anatomical projections from RA back to HVC [59] as well as through the medial portion of MAN ( mMAN ) [60] . Furthermore , there are connections from vocal brainstem nuclei to HVC through Uva and NIf [38 , 61] . Although the signals transmitted through these internal feedback loops are poorly understood , they are likely to contribute to the temporal/sequential structure of song [62] . These internal feedback loops may also contribute to , or even be the main routes of connecting the syllable encoding chains in HVC , rather than the direct connections between the chains within HVC as assumed in our network model . Furthermore , such internal feedback loops could be one site of adapting excitatory drive that contributes to the generation of non-Markovian repeats . However , our deafening results suggest that auditory feedback is a primary source of excitatory drive for repeat generation . Our modeling results will not change if such internal feedback loops are used instead of the direct connections for sequence generation , or instead of auditory feedback as the route of adapting positive feedback . The feedback delay time plays an important role in our model , as the feedback signal must return to HVC in time to exert an influence on the selection of the next syllable . We have hypothesized a simple scenario where these feedback signals are auditory in nature . Each is tuned to a specific syllable in the bird’s repertoire and targets entire chains within HVC . In this case , there is a simple constraint on the delay time for the auditory feedback to exert its influence on the song sequence: the total delay time must be less than the duration of the syllable under examination . Different delay times conforming to this constraint would lead to slight changes in the repeat distribution due to small differences in the initial amount of adaptation experienced on the first repetition , but with no qualitative differences . This constraint could be pushed beyond its limit by very short syllables that terminate before the auditory feedback would return to HVC , precluding the ability of auditory feedback to influence the subsequent transition . If non-auditory internal feedback loops were to carry such a signal , the delay time—and thus the corresponding constraint—could be significantly shorter than predicted for the auditory case . Another possibility is that the delay makes the auditory feedback effective only after the syllable has repeated once or twice . The initial repeats could be sustained by the intrinsic self-connections of the chain network encoding the repeated syllable ( Fig 2c ) . The auditory feedback can then arrive to sustain a long repetition . If the self-connections are weak , the syllable tends to stop at one or two repetitions; but once it repeats more than once or twice , the arriving auditory feedback can take over and sustain a long repetition . This could be another mechanism for the double peaked repeat number distributions we have observed ( Fig 5a ) , in addition to the possibility of a “many-to-one” mapping from HVC to the syllable types . It will be interesting to distinguish these possibilities in future studies . We observed that non-Markovian repeated syllables are typically the loudest syllables in a bird’s repertoire . Furthermore , HVC responses to repeated syllables were significantly greater than responses to non-repeated syllables . Together , these results suggest that louder syllables provide stronger auditory feedback to HVC . This is consistent with our model , in which non-Markovian repeats are strongly influenced by auditory feedback to HVC , though by no means does our model predict such a result . The relationship between the syllable amplitude and repeat length can be further tested with experiments that manipulate syllable amplitudes online with realtime auditory feedback [22] . It should be noted , however , we are not suggesting that a syllable is loud because of a strong auditory input to HVC . The control of syllable amplitude could depend on multiple neural mechanisms . It remains to be investigated why the non-Markovian repeated syllables are louder than other syllables . Our framework can be extended to allow auditory feedback to influence transition probabilities beyond repeated syllables . In general , because the auditory-motor delay in HVC due to neural processing is on the order of a syllable duration ( Fig 2a ) , auditory feedback from the previous syllable arrives in HVC at a time to contribute to the motor activity for the current syllable [13 , 14 , 28] . For a diverging transition of syllable ‘a’ to either ‘b’ or to ‘c’ , as shown in Fig 2b , auditory feedback from syllable ‘a’ can be applied to chain-b and chain-c . Depending on the amount of feedback on each chain , the transition probability to ‘b’ or ‘c’ can be enhanced or reduced by the feedback . Our model for repeating syllables ( Fig 2c ) can be thought of as a special case of this general scenario , in which the repeating syllable-chain receives much stronger auditory input than the competing chain . The strong auditory feedback for repeated syllables may in part reflect synaptic weights that have been facilitated by Hebbian mechanisms operating on the repeated coincidence of auditory feedback with motor activity [28] . This framework is consistent with the observations that manipulating auditory feedback experimentally can change the transition probabilities [13] . Auditory feedback plays a secondary role in determining the song syntax in our proposed mechanism . The allowed syllable transitions are encoded by the branching patterns of the chain networks . Auditory feedback biases the transition probabilities , to varying degrees for different syllable transitions . The secondary role of auditory feedback on the syntax could be the reason for the individual variations seen in a previous deafening experiment [24] . Indeed , it was observed that one Bengalese finch maintained its song syntax 30 days after deafening [24] . The secondary role of feedback in our model is in contrast to the model of Hanuschkin et al , who relied entirely on auditory feedback for determining syllable transitions [27] . However , as in the Hanuschkin model , our model emphasizes the role of auditory feedback in shaping song syntax . We have theorized that auditory feedback provides direct inputs to HVCRA neurons in controlling syllable repetitions in the Bengalese finch . Whether auditory feedback can reach HVCRA neurons in the Bengalese finch is not yet known . Recent experiments that recorded projection neurons intracellularly in HVC of the zebra finch , whose song consists of fixed sequences of syllables , demonstrated that auditory feedback is gated off and does not provide inputs to the projection neurons during singing [63 , 64] . On the other hand , it was shown that the firing rates in HVC of the Bengalese finch changed during singing when the auditory feedback was manipulated [14] , suggesting that auditory feedback can influence HVC during singing in this species . It is possible that the differences in sequence complexity between these species may in part be due to different online sensitivities to auditory feedback [24] . Syllable repetitions are common in many other songbird species , including the canary [19] . It remains to be seen whether auditory feedback plays an important role in syllable repetitions in species other than the Bengalese finch . The differences of sensory-motor integration during singing in different species of songbirds need further investigations . Probabilistic state transition models have been used for describing variable birdsong syntax with high accuracy [20] . Multiple states for a single syllable are often required for the state transition model to capture the statistical properties of the syllable sequences , resulting in the partially observable Markov model with adaptation ( POMMA ) [20] . Such many-to-one mapping manifests as multiple peaks in the repeat number distributions in our data ( Fig 5a ) . However , some of the multiple states in POMMA could also be due to the inaccurate description of history-dependence of the transition probabilities . The geometric adaptation model for the repeat probability , used in the previous work [20] , often leads to multiple states to accurately capture the non-Markovian repeat number distributions , as shown in Fig 4b and 4c . In contrast , the sigmoidal adaptation model for the repeat probability , derived from our network model , enables accurate description of such distributions using a single state . Thus the sigmoidal adaptation model should reduce the complexity of POMMA for the Bengalese finch song syntax . For motor control with continuous trajectories , such as reaching movements or articulation of single speech phonemes , it has been proposed that internal models estimate sensory consequences of motor commands , compare these estimates to actual sensory feedback , and use the difference as error signals to correct ongoing motor commands [65–68] . Along these lines , recent recordings in the auditory areas Field-L and CLM ( caudolateral medopallium ) of the zebra finch showed that , during singing , a subset of neurons exhibit activity that is similar to , but precedes , the activity induced by playback of the birds own song [69] . These data have led to the hypothesis that the songbird auditory system encodes a prediction of the expected auditory feedback ( “forward model” ) used to cancel expected incoming auditory feedback signals [39 , 65 , 69 , 70] . According to such a forward model interpretation , as long as feedback matches expectation , auditory feedback does not reach HVC and therefore does not contribute to song generation during singing [64] . At the surface , this seems at odds with our framework in which auditory feedback has a direct role in song generation , in particular for repeats . One possible resolution is that due to the probabilistic syllable transitions , auditory feedback cannot be fully predicted and canceled by the forward model since the motor actions themselves are not entirely predictable . Such imperfect cancelation allows direct influence of auditory feedback on syllable sequences . Another possibility is that due to the increased loudness of non-Markovian repeated syllables , residual auditory input reaches HVC and contributes to song generation . Some similarities between non-Markovian syllable repetitions in birdsong and sound/syllable repetitions in stuttered speech have been observed in the past [71–73] . In persons who stutter , repeating syllables within words ( ‘to-to-to-today’ , for example ) is a prominent type of speech disfluency [74–76] . Auditory feedback plays an important , but poorly understood , role in stuttered speech . For example , altering auditory feedback , including deafening [74] , noise masking [77 , 78] , changing frequency [79] , and delaying auditory feedback reduces stuttering [80] . Conversely , delayed feedback can induce stuttering in people with normal speech [10 , 11] . Auditory processing may be abnormal both in zebra finches with abnormal syllable repetitions and in persons who stutter [71] . Our observation that deafening reduces syllable repetitions in Bengalese finch songs echoes the reduction of stuttering after deafening in persons who stutter [74] . In general agreement with our proposed role of auditory feedback in repeat generation , some theories suggest that persons who stutter have weak feed-forward control and overly rely on auditory feedback for speech production [67] . It will be interesting to see whether further quantitative analysis of the statistics of stuttered speech would reveal additional behavioral similarities , such as non-Markovian distributions and increased amplitude; to our knowledge no such examination exists . Such similarities could point to shared neural mechanisms with syllable repetition in birdsong , especially the possibility that auditory feedback plays a key role . However , our study also provides a cautionary note to the interpretation of repeated syllables in birdsong as ‘stutters’ . Our analysis shows that syllables with non-Markovian repeat distributions are loud and require strong auditory feedback . In contrast , syllables with Markovian repeat distributions are quieter and are less reliant on auditory feedback for their generation . We propose that it is the former type of syllable repetition that shares similarity to stuttering in humans .
All procedures involving animals were performed in accordance with established animal care protocols approved by the University of California , San Francisco Institutional Animal Care and Use Committee ( IACUC ) . The model neurons for the network simulations are a reproduction of those in previous works [26 , 35] . Below , we summarize the key aspects of these models . The reader is referred to these papers for exact details on the equations and constants . Since detailed information about the ion channels of HVC neurons is unavailable , we model both HVCRA and HVCI neurons as simple Hodgkin-Huxley type neurons , adding extra features to match available electrophysiological data . HVCI neurons exhibit prolonged tonic spiking during [35] . To simulate this we use a single-compartment model with the standard sodium-potassium mechanism for action potential generation along with an additional high-threshold potassium current that allows for rapid spike generation . A distinctive feature of HVCRA neurons is that their activity comes in the form of precise bursts during song production [30 , 35] . This bursting activity increases the robustness of signal propagation along chains of these neurons [32 , 35] . A study of the subthreshold dynamics of HVCRA neurons during singing suggests that this bursting is an intrinsic property of these cells [35] . We generate this intrinsic bursting behavior with a two-compartment model [26 , 32 , 35] . A dendritic compartment contains a calcium current as well as a calcium-gated potassium current . When driven above threshold , these currents produce a stereotyped calcium spike in the form of a sustained ( roughly 5 ms ) depolarization of the dendritic compartment . A somatic compartment contains the standard sodium-potassium currents for generating action potentials . These compartments are ohmically coupled so that a calcium spike in the dendrite drives a burst of spikes in the soma . All compartments also contain excitatory and inhibitory synaptic currents . Action potentials obey kick-and-decay dynamics . All synaptic conductances start at 0 . When an excitatory or inhibitory action potential is delivered to a compartment , the corresponding synaptic conductance is immediately augmented by an amount equal to the strength of the synapse . In between spikes , the synaptic conductances decay exponentially toward zero . The network topology underlying all of the more advanced models below is the branching synfire chain network for HVC [26] . HVCRA neurons are grouped into pools of 60 neurons . 20 pools are then sequentially ordered to form a chain . Except for the final pool , all neurons in a pool make an excitatory connection to every neuron in the next pool ( Fig 2b ) . The strengths of these synapses are randomly generated from a uniform random distribution between 0 and GEE , max = 0 . 09 mS/cm2 . Because of this setup , activating the neurons in the first group sets off a chain reaction where each group activates the subsequent group , leading to a signal of neural activity propagating down the chain with a precise timing . There is one chain for every syllable in the repertoire of the bird . Activity flowing down a given chain drives production of the corresponding syllable through the precise temporal activation of different connections from the HVCRA neurons to RA ( not explicitly modeled ) . To begin to impose a syntax on the song , the neurons in the final pool of one chain make connections to the initial pool of any chain whose syllable could follow its own . This branching pattern encodes the basic syllable transitions that are possible . When the activity in an active chain reaches a branching point , all subsequent chains are activated , however only one should stay active—the syllable chosen next . This selection is achieved through lateral inhibition between the chains intermediated by HVCI neurons . There is a group of 1 , 000 HVCI neurons . Each HVCRA neuron has a chance of making an excitatory connection to each HVCI neuron with a probability pEI = 0 . 05 . Each of these connections has a strength randomly drawn from a uniform distribution between 0 and GEI , max = 0 . 5 mS/cm2 . In turn , each HVCI neuron has a chance of making an inhibitory connection to each HVCRA neuron with a probability pIE = 0 . 1 . The strengths of these connections are randomly drawn from a uniform distribution between 0 and GIE , max = 0 . 7 mS/cm2 . This setup gives a rough approximation of global inhibition on the HVCRA neurons which is what leads to the lateral inhibition between the chains that they comprise . Noise is added to the network to make switching between chains a stochastic process . This noise is modeled as a Poisson process of spikes incident on each compartment of every neuron . The strength of each spike is randomly selected from a uniform distribution from 0 to Gnoise and every spike has an equal chance of being excitatory or inhibitory . Both compartments of HVCRA neurons receive noise at a frequency of 500 Hz; at the soma Gnoise = 0 . 045 mS/cm2 , while at the dendrite Gnoise = 0 . 035 mS/cm2 . The single compartment of the HVCI neurons receive noise at a frequency of 500 Hz with Gnoise = 0 . 45 mS/cm2 . In HVCRA neurons , this leads to subthreshold membrane fluctuations of ∼ 3 mV; in the HVCI neurons , the results is a baseline firing rate of ∼ 10 Hz . Each HVCRA neuron also receives an external drive that facilitates robust propagation of signals through the chains . This takes the form of a purely excitatory spike train modeled by a Poisson process with frequency 1 , 000 Hz . The strength of each spike is chosen from a uniform random distribution from 0 to 0 . 05 mS/cm2 . We incorporate auditory feedback into the branching synfire chain model in a manner similar to the external drive used in [26] . When a syllable is being produced and heard by the bird , some amount of auditory feedback can be delivered to any of the chains in the network in the form of external drives . The relative strength of this feedback drive between chains then biases transition probabilities so that auditory feedback plays an important role in determining song syntax . The first piece in our model for auditory feedback is determining when auditory feedback from a specific syllable is active . We assume that the first few pools in every chain encode for the silence between syllables . Furthermore , once a syllable is being produced , there is a delay before auditory feedback begins that represents how long it takes for the bird to hear the syllable and process the auditory information . In our simulations , the activity of the 4th pool of every chain is monitored ( by keeping track of the number of spikes in the previous 5 ms ) , with syllable production onset determined by when the population rate crosses a threshold of 43 Hz/neuron . After a delay of 40 ms , auditory feedback from that chain’s syllable begins . The auditory feedback takes the form of an external drive to all of the HVCRA neurons in a chain . Every chain can provide auditory feedback to every other chain , including itself . Thus , if there are N chains , then there are N2 auditory feedback pathways . Denote the strength of the auditory feedback from chain i to chain j as Gij . Every neuron in a chain will have N synapses , each one carrying the auditory feedback from one of the N chains in the network . The synapses carrying the auditory feedback from chain i to chain j have strengths drawn from a uniform random distribution between 0 and Gij . Setting Gij = 0 implies that there is no auditory feedback from chain i to chain j . When auditory feedback from a chain is active , the corresponding synapses are driven with Poisson processes at a frequency ffdbk . The model that each neuron receives only one synapse for each auditory feedback source is unrealistic . However , for computational simplicity , we model the feedback this way and consider each high-frequency synapse to be carrying spike trains from multiple sources . Since the kick-and-decay synapse model does not separate sources , this induces no real approximation . Auditory feedback parameters for Fig 2 were tuned to ffdbk = 1 , 340 Hz and Gbb = 1 . 9 mS/cm2 . To implement synaptic depression , we follow a simple phenomenological model used in Abbott et al . [53] . Whenever a synapse is used to transmit a spike , its strength g is decreased by a constant fraction α , so that g → ( 1 − α ) g . The parameter α is referred to as the depression strength . In between spikes , the synaptic strength recovers toward its base strength g0 with first order dynamics: τ R d g d t = - ( g - g 0 ) . ( 4 ) The parameter τR is called the synaptic depression recovery time constant . If such a depressing synapse carries a spike train with a constant frequency f , the large-scale effect is an exponential decay to a steady-state strength where recovery and depression are balanced . The time constant of this decay as well as the steady-state strength can be expressed as functions of the model parameters: τ ( τR , α , f ) and g∞ ( τR , α , f ) . See below for a derivation of the exact forms . In our simulations with synaptic depression on the synapses carrying auditory feedback ( in particular Fig 2 ) , we use τR = 3 . 25 s and α = 0 . 006 . It should be noted that , since these synapses actually represent the combined effect of multiple synapses ( see above ) , these model parameters should not be taken as biologically representative . However , by matching the large-scale dynamics ( τ and g∞ ) of the lower-frequency constituent synapses to that of the model synapse , one can find the more biologically relevant underlying depression parameters . Assume that each auditory feedback synapse represents the combined input of N constituent synapses , each carrying a spike train with a frequency f/N so that the model synapse carries a spike train with frequency f . Matching the large-scale dynamics is then expressed as ( primes representing biologically relevant parameters ) τ ( τ R ′ , α ′ , f / N ) = τ ( τ R , α , f ) , ( 5 ) g ∞ ( τ R ′ , α ′ , f / N ) = g ∞ ( τ R , α , f ) . ( 6 ) Since α , τR , and f are known from the model , we can solve for α′ and τ R ′ . With N = 50 this gives α′ ≈ 0 . 26 and τ R ′ ≈ 3 . 75 s—reasonable values for short-term depression in cortex [53] . Both the neural and synaptic depression models take the form of a large system of differential equations . A fourth-order Runge-Kutta scheme is used to numerically integrate these equations with custom code written in C++ . When action potentials are generated during a time-step , synaptic conductances and synaptic depression dynamics are immediately updated before the next time-step is taken . All analysis is done with custom code in the MATLAB ( The Mathworks , Natick , MA ) environment . To systematically examine how the repeat number distribution depends on the strength a0 of the auditory feedback and the adaptation strength α , we used the sigmoidal adaptation model , Eqs ( 2 ) and ( 3 ) , to generate repeat number distributions with combinations of these parameters . The decay time constant of the auditory feedback due to synaptic adaptation was set to τ = τ R 1 - τ R f log ( 1 - α ) , ( 7 ) where τR is the recovery time constant and f is the firing rate of NIf neurons during auditory feedback ( see below ) . Besides a0 and α , all other parameters are set using those from the network simulations shown in Fig 3 with T = 100 ms ( approximately the length observed in the simulations ) . To simulate a repeat bout , we sequentially generate random numbers xk from a uniform distribution between 0 and 1 and compare each number to pr ( k ) . The first time that xk > pr ( k ) signifies that a further repeat does not occur , so the bout contains k repeats . A distribution of repeats for a given ( a0 , α ) combination is produced by simulating the repeat bouts 10 , 000 times , and the results are shown in Fig 4b , where we plot the peak repeat numbers for the distributions . Because the peak repeat number can go to infinity as adaptation strength goes to 0 , for numerical stability we use a minimum adaptation strength of 0 . 001 . The sigmoidal adaptation contains two interesting special cases: ( 1 ) If we set the adaptation constant τ → ∞ , which is equivalent to no adaptation of the auditory synapses , we have b → 1 and the repeat probability becomes a constant , a hallmark of the Markov model for repeats . ( 2 ) If c = 1 , which means the repeat probability is zero when the repeat number is large , and the initial auditory input is small such that when ab ≪ 1 , we have pr ( n ) ≈ abn , i . e . the repeat probability decreases by a constant factor with the repeat number . This the geometric adaptation of repeat probability . It was used to describe the repeating syllables in a previous work on the song syntax of the Bengalese finch [20] . Any of these models can be extended to provide better fits to data by allowing multiple states . In these extended models , a repeated syllable is represented by multiple repeating states that all produce that syllable and are connected in series ( Fig 4b ) . The probability of the syllable repeating N times is given by P ( N ) = ( 1 - p r ( N ) ) Π n = 1 N - 1 p r ( n ) . ( 8 ) The observed repeat number probability Po ( N ) is computed by normalizing the histogram of the repeat numbers . The parameters a , b , c are determined by minimizing the sum of the errors E = ∑ N ( P ( N ) - P o ( N ) ) 2 , ( 9 ) while constraining the parameters ranges 0 < a < 108 , 0 < b < 1 , and 0 < c < 1 , using the nonlinear least square fitting function ‘lsqcurvefit’ in MATLAB . To avoid local minima in the search , 20 random sets of the initial values of the parameters were used for the minimization , and the best solution with the minimal square error was chosen . The difference between two probability distributions p1 ( n ) and p2 ( n ) is defined as d = max n | p 1 ( n ) - p 2 ( n ) | max n ( p 1 ( n ) , p 2 ( n ) ) , ( 10 ) i . e . the maximum absolute differences between the two distributions normalized by the maximum of the two distributions [20] . When fitting a functional form to a probability distribution , the difference between the empirical distribution and the fit is compared to a benchmark difference that represents the amount of error expected from the finiteness of data . For a given empirical distribution , the benchmark difference is computed by first randomly splitting the full data set into two groups of equal size and then computing the difference between the distributions resulting from each group . This process is repeated 1 , 000 times to produce a distribution of differences from simple resampling . The benchmark error is set at the 80th percentile of this bootstrapped distribution . The method of benchmark error was explained in detail previously [20] . Our model of synaptic depression characterizes the temporal dynamics of synaptic strength , g . Each synapse has a base strength , g0 . The depression model has two parameters: ( 1 ) depression strength , α: fraction of strength lost at each spike; ( 2 ) recovery time constant , τR: rate of exponential recovery toward g0 . Mathematically it can be described by two rules: 1 . at a spike: g → ( 1 − α ) g; 2 . between spikes: τRdg/dt = − ( g − g0 ) . We would like to characterize how this synapse will behave when transmitting a spike train that takes the form of a Poisson process . The analysis is simpler if we consider a regular spike train with frequency f as an approximation . Fortunately , this should still give the average behavior for the Poisson process case . We begin by deriving an iterative map that takes the strength right before one spike and gives the strength right before the next . Let the strength of the synapse right before a spike be g . Immediately after the spike , the strength will then be ( 1 − α ) g . Integrating the equation for recovery ( from an initial condition ( ti , gi ) to ( tf , gf ) ) yields: g f = g 0 + ( g i - g 0 ) e - ( t f - t i ) / τ R . ( 11 ) Since the spike train is assumed to be regular , we have tf − ti = 1/f . And since the recovery starts from gi = ( 1 − α ) g , the complete spike-to-spike iterative map is g → g 0 + ( ( 1 - α ) g - g 0 ) e - 1 / ( τ R f ) . ( 12 ) Using synaptic strength relative to g0 , i . e . g = Ag0 gives: A → 1 - e - 1 / ( τ R f ) + ( 1 - α ) e - 1 / ( τ R f ) A . ( 13 ) This iterative map has the form A → a+bA , with a = 1 − e−1/ ( τR f ) and b = ( 1 − α ) e−1/ ( τR f ) . If we start with A = 1 , then this map has a closed-form solution: A n = a 1 - b + 1 - a - b 1 - b b n - 1 . ( 14 ) This is a geometric decrease toward a steady-state value of a/ ( 1 − b ) with a ratio of r = b . In terms of our model parameters , this is g ∞ = g 0 1 - e - 1 / ( τ R f ) 1 - ( 1 - α ) e - 1 / ( τ R f ) , ( 15 ) r = ( 1 - α ) e - 1 / ( τ R f ) . ( 16 ) This discrete geometric decrease should be well-approximated by continuous exponential decay . The number of spikes needed to produce a fractional decrease of e−1 is given by rn = e−1 , so that n = −1/logr . Since the inter-spike interval is 1/f , the time constant of the continuous decay will thus be given by τ = n f = 1 f log [ ( 1 - α ) e - 1 / ( τ R f ) ] = τ R 1 - τ R f log ( 1 - α ) . ( 17 ) While this derivation is for a regular spike train , simulations ( not shown ) verify that it is also fits the large-scale dynamics of a Poisson spike train with the same mean frequency . 32 birds were used in this study . All 32 birds contributed to the behavioral analysis ( Fig 5 ) . Of these 32 birds , six birds were used in the deafening studies . A different subset of six birds were used in the electrophysiology experiments . During the experiments , birds were housed individually in sound-attenuating chambers ( Acoustic Systems , Austin , TX ) , and food and water were provided ad libitum . 14:10 light:dark photo-cycles were maintained during development and throughout all experiments . All behavioral analyses , as well as stimulus creation , were done using custom code written in MATLAB . Individual adult male Bengalese finchs were placed in a sound-attenuating chamber ( Acoustic Systems , Austin , Tx ) to collect audio recordings . An automated triggering procedure was used to record and digitize ( 44 , 100 Hz ) several hours of the bird’s singing . These recordings were then scanned to ensure that more than 50 bouts were obtained . Bouts were defined as continuous periods of singing separated by at least 2 seconds of silence . Bengalese finch songs typically consist of between 5–12 distinct acoustic events , termed syllables , organized into probabilistic sequences . Each bout of singing consists of several renditions of sequences , with each sequence containing between 1 and approximately 40 examples of a particular syllable . The syllables from 15–50 bouts were hand labeled for subsequent analysis . Birds were deafened by bilateral cochlear removal [81 , 82] . Complete removal of the cochlea , including the distal end of the auditory nerve , was visually confirmed using a dissecting microscope . After cochlear removal , some birds showed signs of vestibular disturbance that usually resolved in the first few days after surgery . Extra care was taken to ensure that such birds had easy access to seed and maintained full crops . Birds did not exhibit difficulty in perching , feeding , or interacting with other birds after returning to their home cages . The electrophysiological results presented in this study were collected as part of a larger study investigating how sequences and syllable features are encoded in HVC auditory responses . The data used in this study and the associated methods have been described previously [28] . Briefly , for neural recordings , birds were placed in a large sound-attenuating chamber ( Acoustic Systems , Austin , TX ) and stereotaxically fixed via a previously implanted pin . During electrophysiological recordings , birds were sedated by titrating various concentrations of isoflurane in O2 using a non-rebreathing anesthesia machine ( VetEquip , Pleasanton , CA ) . Throughout the experiment , the state of the bird was gauged by visually monitoring the eyes and respiration rate using an IR camera . Sites within HVC were at least 100 μm apart and were identified based on stereotaxic coordinates , baseline neural activity , and auditory response properties . Experiments were controlled and neural data were amplified with an AM Systems amplifier ( x1000 ) , filtered ( 300–10 , 000 Hz ) , and digitized at 32 , 000 Hz . Stimuli were band-pass filtered between 300-8 , 000 Hz and normalized such that BOS playback through a speaker placed 90 cm from the head had an average sound pressure level of 80 dB at the head ( A scale ) . Each stimulus was preceded and followed by 0 . 5-1 s of silence and a cosine modulated ramp was used to transition from silence to sounds . The power spectrum varied less than 5 dB across 300-8 , 000 Hz for white-noise stimuli . All stimuli were presented pseudo-randomly . To probe how repeated syllables are encoded in the population of HVC neurons , we used a stimulus set that consisted of 10 strings of 1000 pseudo-randomly ordered syllables was constructed . The details of this stimulus are described previously [28] . Briefly , for each bird , natural sequences ( i . e . sequences produced by a given bird ) and non-natural sequences ( i . e . sequences that were never produced by a bird ) of length 1 through 10 were concatenated with equal probability into 10 strings of 1000 syllables . For each syllable in the birds repertoire occurring in these stimuli , a single ‘prototype’ syllable was used based on the distributions of acoustic features of that syllable . The median of all inter-syllable gaps was used for each gap . BOS stimuli created with these elements ( synthesized BOS , prototype syllables and median gaps ) elicit HVC auditory responses of comparable magnitude to normal BOS stimuli . Additionally , responses to single syllables preceded by the same long sequences in the pseudo-random stimuli are not significantly different from responses in synthesized BOS . Thus , these stimuli isolate sequence variability from other sources of variability in song , and allow investigating how HVC auditory responses to individual syllables are modulated by the preceding sequence . Single units were identified events exceeding 6 standard deviations from the mean and/or were spike sorted using in house software based on a Bayesian inference algorithm . Multi-unit neural data were thresholded to detect spikes more than 3 standard deviations away from the mean . Both single and multi-unit spike times were binned into 5 ms compartments and then smoothed using a truncated Gaussian kernel with a standard deviation of 2 . 5 ms and total width of 5 ms . To characterize the responses to individual target syllables , we defined a response window , which started 15 ms after the onset of the syllable and extended 15 ms after the offset of that same syllable . All statistical tests were performed using either paired sign-rank tests or unpaired rank-sum tests . Throughout the paper , results were considered significant if the probability of Type I errors was α < 0 . 05 . Bonferroni corrections were used to adjust α-values when multiple comparisons were performed . | Repetitions are common in animal vocalizations . Songs of many songbirds contain syllables that repeat a variable number of times , with non-Markovian distributions of repeat counts . The neural mechanism underlying such syllable repetitions is unknown . In this work , we show that auditory feedback plays an important role in sustaining syllable repetitions in the Bengalese finch . Deafening reduces syllable repetitions and skews the repeat number distribution towards short repeats . These effects are explained with our computational model , which suggests that syllable repeats are initially sustained by auditory feedback to the neural networks that drive the syllable production . The feedback strength weakens as the syllable repeats , increasing the likelihood that the syllable repetition stops . Neural recordings confirm such adaptation of auditory feedback to the auditory-motor circuit in the Bengalese finch . Our results suggests that sensory feedback can directly impact repetitions in motor sequences , and may provide insights into neural mechanisms of speech disorders such as stuttering . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | An Adapting Auditory-motor Feedback Loop Can Contribute to Generating Vocal Repetition |
The notion of attractor networks is the leading hypothesis for how associative memories are stored and recalled . A defining anatomical feature of such networks is excitatory recurrent connections . These “attract” the firing pattern of the network to a stored pattern , even when the external input is incomplete ( pattern completion ) . The CA3 region of the hippocampus has been postulated to be such an attractor network; however , the experimental evidence has been ambiguous , leading to the suggestion that CA3 is not an attractor network . In order to resolve this controversy and to better understand how CA3 functions , we simulated CA3 and its input structures . In our simulation , we could reproduce critical experimental results and establish the criteria for identifying attractor properties . Notably , under conditions in which there is continuous input , the output should be “attracted” to a stored pattern . However , contrary to previous expectations , as a pattern is gradually “morphed” from one stored pattern to another , a sharp transition between output patterns is not expected . The observed firing patterns of CA3 meet these criteria and can be quantitatively accounted for by our model . Notably , as morphing proceeds , the activity pattern in the dentate gyrus changes; in contrast , the activity pattern in the downstream CA3 network is attracted to a stored pattern and thus undergoes little change . We furthermore show that other aspects of the observed firing patterns can be explained by learning that occurs during behavioral testing . The CA3 thus displays both the learning and recall signatures of an attractor network . These observations , taken together with existing anatomical and behavioral evidence , make the strong case that CA3 constructs associative memories based on attractor dynamics .
Theoretical work has shown how networks with excitatory recurrent connections can function as an associative memory [1]–[4] . Specifically , Hebbian plasticity at the synapses of recurrent connections leads to the association of the elements of a memory . Information stored in this way can be recalled given external input of a partial pattern , thus displaying “pattern completion” [5] , [6] . This “attraction” of network activity to a stored pattern provides a useful form of associative memory and has inspired much theoretical and experimental work . Among hippocampal sub-regions , CA3 is unique in having extensive excitatory recurrent connections [7] , [8] . This property , together with the finding that the synapses of these recurrent connections can undergo Hebbian plasticity [9] , [10] , has led to the hypothesis that CA3 has attractor dynamics and serves as the main site for associative memory storage in the hippocampus [11]–[16] . Despite the influence of the attractor concept , it has been difficult to obtain direct experimental support for attractor networks in the hippocampus . Experiments specifically designed to observe the electrophysiological signature of attractor dynamics in CA3 have been problematic ( for a review , see [17] ) . The experiments designed to identify attractors first established memories of two environments with different shapes ( square/round ) but unaltered distal cues; the environment was then gradually morphed from one to the other [18]–[20] . Given that attractor networks can display winner-take-all dynamics , the expectation was that , during such morphing , the network would first be attracted to one stored memory and would then make a sudden transition to the other . Because the firing patterns in CA3 changed gradually rather than suddenly , it has been argued that the results are inconsistent with the properties of attractor dynamics and that a different function of CA3 should be entertained [21] . Alternatively , it has been suggested that sudden transitions may not be an appropriate criteria for identifying attractor networks [22] . Here , we directly address this issue , which is central to understanding the hippocampal contribution to associative memory . We have developed a computational model of CA3 and its input structures and have used this model to simulate the morphing experiments described above [18]–[20] . This model was constrained not only by the properties of CA3 , but also by the properties of the input to CA3 from the dentate gyrus ( DG ) and entorhinal cortex ( EC ) . Furthermore , because CA3 cells fire selectively during only a small fraction of the theta cycle [23] , [24] , we modeled the dynamics of CA3 that could occur in a comparable short time segment . With this model , we have been able to account for the results obtained during morphing experiments and to analyze how the CA3 recurrent connections affect network function and dynamics . Our analysis clarifies the criteria that should be applied to identify attractor dynamics under the conditions of morphing experiments . These criteria are satisfied by the data . Our simulations show that CA3 only satisfies these criteria when recurrent excitation is present , leading us to conclude that intra-CA3 processes in fact support attractor dynamics . Notably , for small morphs , the pattern of activity in CA3 is attracted to a stored pattern , whereas the pattern in DG , a region that provides input to CA3 , is not . We have also analyzed an additional experimental observation , the hysteresis observed in CA3 recordings during morphing [19] . Our analysis suggests that this hysteresis arises from CA3 plasticity , thus suggesting a new method for observing how experience affects CA3 attractor dynamics .
The model that we developed is illustrated in Figure 1A . We modeled pyramidal cells of CA3 ( NCA3 = 10 , 000 ) and the granule cells of the dentate gyrus ( DG , NDG = 800 , 000 ) as one-compartment integrate-and-fire neurons [25] . The voltage of each neuron i is determined by the input feedforward excitatory input , IFF , the recurrent excitatory input , IREC , the recurrent inhibitory input , IGABA , and the after-hyperpolarization , IAHP , currents . Both DG and CA3 cells receive feedforward excitatory current from the EC ( NEC = 160 , 000 ) , IEC , and recurrent inhibition from their respective interneuron networks , IGABA . CA3 cells also receive feedforward excitatory current from DG cells , IDG , and recurrent excitatory current from the recurrent collaterals of CA3 , ICA3 . We use as parameters the average input resistance ( Rn = 33 MΩ ) [26] , the membrane time constant ( τn = 30 ms ) , and the firing threshold ( T = −50 mV ) . Voltage is reset to rest ( VREST = −65 mV ) after each spike . The after-hyperpolarization maximum current is set to AAHP = −2 nA with τAHP = 7 ms decay [27] . To emulate an absolute refractory period caused by sodium channel inactivation , cells that emitted a spike were not allowed to spike for the following period τSPIKE = 2 ms . Inhibition is global within each region ( CA3 and DG ) and occurs with a delay of 3 . 3±0 . 4 msec [28] relative to the first spike succeeding the previous inhibitory current discharge [29] . The membrane potential of neuron i evolves according to: ( 1 . 1 ) The feedforward excitatory input current of cell i , , is computed through the arithmetic sum of all excitatory post-synaptic currents from synapses reaching i multiplied by a feedforward gain factor . For DG cells: ; while for CA3 cells: . was estimated as 0 . 68 nA in order to allow network oscillation within gamma frequency ( ∼35 Hz ) . The entorhinal input current , IEC , of each CA3 and DG cell is computed as a linear combination of the inputs from the lateral and medial entorhinal cortices regulated by a mixing factor α ( Equation 1 . 2 ) . There is no data available to directly estimate α , so we quantitatively estimate it through a parametric search methodology . The inputs of the lateral and medial entorhinal cortices are computed independently and are normalized by the mean maximum value considering all positions ( Equations 1 . 3 and 1 . 4 ) . Normalization of the input allows an interpretation of α with respect to the overall size of the EPSC originated in each of the entorhinal cortices . Feedforward synaptic weights , WMEC and WLEC , are randomly assigned from a distribution that corresponds to the measured distribution of synapse size [30] , [31] . The number of EC cells that converge into the DG and CA3 can be estimated from the measured spine density of 2 . 3 spines/µm and dendrite length of 3000 µm [32] . Considering that each spine has one synapse , we estimate for the measured spine density and dendritic length [33] that each DG cell receives input from 1200 MEC and 1500 LEC cells , while CA3 cells receive inputs from 1400 MEC and 1500 LEC cells [34] , [35] . To emulate the morphing experiment ( see below ) , the total EC input of each cell i , IiEC , is defined for each of the Nr positions r ( x , y ) and Nc wall shapes c: ( 1 . 2 ) ( 1 . 3 ) ( 1 . 4 ) The DG input current to CA3 cells is set respective to the normalized activity of a randomly selected presynaptic DG cell multiplied by a relative gain factor β . Activity of DG cells is computed a priori as a mean rate ( see Data analysis section ) and is set constant for a specific position and morphing stage . The input of the DG to CA3 neuron i is defined as: ( 1 . 5 ) The recurrent CA3 input current of cell i , is determined by the non-linear threshold function of the arithmetic sum of the recurrent excitatory current , : ( 1 . 6 ) where = 20 nA is the asymptotic feedback excitatory current and is the recurrent excitation threshold , considering a non-linear mapping from input to excitatory current [36] . Recall threshold is defined as the threshold complementary . is computed as the sum of the product of the afferent activity , , and the specific synaptic weight , . is modeled as a step function with 3 msec duration and release lag , , of 1 msec [37]: ( 1 . 7 ) To emulate the experimental procedure and the ensuing hippocampal neuronal dynamics , the activity of the model was computed over the trajectory of real rats available online [38] . All trajectories were obtained from sessions in a square environment , and morphing was encoded in the activity of LEC neurons . We do not implement any direct source of noise in our simulations . Indirectly , noise arises naturally as fluctuation of EPSPs from the combination of a high-resolution spatial representation ( computed bins of 1 cm2 compared to place fields of >200 cm2 ) with the natural tracking of the position of the animal . Noise also arises from the IPSP delay , which is set probabilistically with a normal distribution . Neural activity was computed over sessions of 10 minutes ( T = 600 sec ) . Each gamma cycle ( t = 36 . 5 msec ) was computed independently with a randomly selected IPSP delay determined by a normal distribution with specific SD ( d = 3 . 3±0 . 4 msec ) and with the potential of all neurons initialized at rest ( Figure 1B ) . Different cells could assume multiple rate values ( see Data analysis section ) because the release of action potentials was probabilistic: the cells with the strongest input fire in every such simulated cycle and thus show a high probability ( rate , λ ) . Cells that fire just after the strongest cell will fire in most simulated cycles approximating the maximum rate , whereas cells with less excitation may fire late in the gamma cycle but often do not fire at all , thus displaying a low rate . Recurrent excitation is applied within the active window of the gamma cycle and can cause a cell with low feedforward excitation to produce an action potential , increasing its rate . Activity of EC cells is dependent on the current position of a virtual rat , r ( x , y ) , which navigates through the environment following empirically determined trajectories of real rats ( see below ) , and the current progression in the morphing procedure , c . The computation is analogous to previous work [34] . The activity of MEC cells is defined by a mathematical description of grid cells [39] and is made insensitive to morphing [19] , [40] ( Figure 1C , left ) , unless when noticed otherwise . To simulate the conditions that lead to global remapping in the CA3 [21] , [40] , grid cell realignment is implemented by setting a different angular and position phase but letting the same spatial frequency . Both MEC and LEC rate maps are tailored to fit the observed spatial information score [41] . Morphing is encoded in the activity of the LEC cells by allowing a sharp transition between two independent rate maps at a morphing degree specific for each cell defined randomly following a uniform distribution [34] ( Figure 1C , right ) . The LEC selectivity to morphing is grounded in the observed selectivity of LEC cells to objects [42] , [43] , the fact that it receives strong input from sensory driven areas [44]–[46] , and the finding that rate remapping in the CA3 is impaired by LEC lesions [20] . Importantly , although we assume a sharp transition for the response of LEC cells during morphing , the fact that the point of the transition is different for each cell and that the activity of many cells is summed makes our implementation equivalent to the case in which each cell had a smooth or non-uniform response to morphing . LEC and MEC activity was produced with a resolution of 1 cm×1 cm . Excitatory input for the EC and DG was computed using the virtual rat's position as a reference ( x , y ) . The model was updated with a step size of 1 msec from the beginning of the gamma cycle and the time of the first spike and 0 . 1 msec steps in the interval between the first spike and the release of GABA . Spikes were time stamped for further analysis . Due to the expected low activity levels of the DG [47] , [48] , only the cells with mean weight strengths within the upper 10% percentile were simulated . To emulate the rat's exposure to the square and round environments , the recurrent CA3 weight matrix is defined based on the history of the firing of cells on the square and round environments without recurrent excitation . As we are not interested in the dynamics of plasticity prior to morphing experiment , we used an interleaved procedure to define the recurrent CA3 weight matrix as follows: to enhance orthogonalization of CA3 activity following the CA3-DG interaction [16] , each CA3 cell is assigned to a cluster through a k-means algorithm using spatial correlation as a distance metric . The number of clusters is set to maximize the grouping of the data [49] . Cells belonging to the same cluster ( C ) are interconnected with a weight inversely proportional to the size of the cluster ( n ( C ) ) so that the sum of all synaptic weights to each cell is equal to 1: ( 1 . 8 ) The training is performed for the two extreme shapes of the environment , and the synaptic weight between two cells is defined as the maximum value over the two conditions: ( 1 . 9 ) Recurrent CA3 weights are updated at the end of every session: a temporary connection matrix is built using the above clustering method , and the weights interconnecting active cells are updated by a convex sum-ruled by a learning factor ( LRATE ) between the previous weights and the weights in the temporary connection matrix: ( 1 . 10 ) Data analysis includes construction of 16×16 bin rate maps , place fields analysis , population vector correlation , rate overlap , and spatial correlation and is performed following the same procedures and methods as reported for the experimental data [19] . In summary , the outcome of the simulations was a list of time-stamped spikes that could be related to the r ( x , y ) coordinate in which they were emitted ( r ( t ) ) . Space was discretized in 16×16 bins ( Nr = 256 , equivalent to 5×5 cm ) . For the specific case of the DG input to CA3 and the activity map of MEC and LEC cells , space was discretized in 80×80 bins ( Nr = 6400 , equivalent to 1×1 cm ) . For each bin , the firing rate was calculated by averaging the number of spikes at a certain position and dividing it by the average occupancy of that bin ( Figure 1D ) . Rate maps were smoothed by a Gaussian kernel ( g ) of h = 5 cm sd: ( 1 . 11 ) Cells with a mean firing rate above 0 . 1 Hz in at least one of the morphing steps were considered active . Place fields were determined by the existence of continuous bins ( n>8 and n<128 ) with a peak rate no less than 2 Hz , with all units above 20% of this peak value . The population vector ( PV ) correlation was calculated by correlating the response vector of all cells in a specific bin and correlating it to: ( a ) the same response vector under a different morphing condition and ( b ) a response vector of a different bin localized 50 cm away under the same morphing condition . Only active cells with a firing rate above 1 Hz in the two conditions were considered for the PV . The overlap between two rate maps was measured by dividing the mean rate displayed in the less active condition by the mean rate in the more active condition . The spatial correlation was defined as the pair-wise correlation of the rate maps considering each bin . In our simulations , to correct for sampling error , all comparisons in the morphing experiment between rate maps were performed using simulation data from different trajectories .
With this biologically constrained model , we computed the activity of CA3 and DG cells in different morph states by analyzing the spike probability as the simulated rat traversed the environment ( paths were taken from experimental data [38] , [53] ) . The rat's location was represented by the activity of grid cells of the medial entorhinal cortex ( MEC ) ( Figure 1C , left ) [38] , [53] , whereas sensory information about the walls of the environment was represented by the activity of the cells of the lateral entorhinal cortex ( LEC ) ( Figure 1C , right ) [42] , [43] . Both MEC and LEC maps were constrained by data ( see Methods ) . Rate maps were computed from the simulated neural activity and the trajectories ( Figure 1D ) . There were three open variables that we could not obtain from the literature: the relative strength of the input from LEC or MEC ( α ) ; the ratio of DG-to-EC input ( β ) ; and strength of the recurrent synapses ( 1-δ ) . We estimated these parameters computationally by searching the best fit to the experimental data using as reference the available metrics of both population and single-unit activity . This strategy allowed a direct comparison between the simulated data and experimental data using exactly the same methods . If the reader is not interested in the technical issue of parametrical optimization , he or she may wish to go directly to the next section , where we apply the model to the morphing data and analyze the evidence of attractor dynamics in the CA3 . Through the parametric optimization of the relative strength of the input from LEC or MEC ( α ) , the simulated DG population data reproduced the main features of the experimental data ( Figure 2 ) . In our simulations , an average of 3 . 5% of DG cells were active at each session , in accordance with experimental measurements [47] , [48] . DG cells exhibited place fields that independently rate remapped during morphing ( Figure 2A ) , as observed in previous modeling studies [31] , [34] and in the experimental data [19] . The distribution of the number of place fields per active cell was similar to that observed experimentally ( Figure 2B ) . Simulated place fields had a peak rate of 11 . 92 Hz±7 . 87 , comparable to 11 . 54 Hz±8 . 16 in Leutgeb et al . ( 2007 ) . The parameter α was optimized by searching in the range of valid α values ( 0–1 ) the value with which the simulated neural activity of the DG would better fit experimental data [19] . Both individual neurons , cumulative change in the average firing rate ( rate overlap ) and spatial correlation , and population activity metrics , PV correlation and PV autocorrelation , were used as metrics for the optimization process ( see Methods ) . When we analyzed the activity of individual neurons , we observed that , for small and large morphs , strengthening LEC influence ( high α ) led to a higher decorrelation between firing rate distributions of individual cells than when MEC influence was high ( low α ) ( Figure 2C ) . In the extreme case when considering only the LEC input ( α = 1 ) , the rate maps of individual DG cells in the two extreme shapes were uncorrelated . Strong LEC influence led to a higher change in the average firing rate of individual cells ( lower rate overlap ) compared to the condition of strong MEC influence ( Figure 2D ) . If we interpret these results in terms of rate remapping ( in which non-spatial information is encoded in the rate of spatially stable place fields ) [34] , [54] , we observe that there is a trade-off between the ability of the DG cells to encode the wall shape information in the peak rate of place fields and the maintenance of the position of place fields . Interestingly , the experimental observation indicates that there is a compromise with balanced MEC and LEC influence ( Figure S5 of [19] ) . When we analyzed the population activity of DG neurons , we observed that stronger LEC contribution yielded stronger PV decorrelation for every pair of box shapes if compared to conditions in which the MEC input was greater ( Figure 2E ) with the best model fit , as in the analysis of single cells , obtained with a balanced MEC and LEC input ( α = 0 . 5 ) . We next investigated whether the encoding of the wall shape information disturbed the ability of the DG population to produce orthogonal representations of unrelated positions by measuring the autocorrelation of PVs obtained in the same box shape but at positions located 50 cm away ( Figure 2F ) . High correlation would indicate a high overlap between representations of different positions , and a lower correlation would imply otherwise . We observed that strong LEC input resulted in PVs more strongly correlated at distant positions if compared to the condition with higher MEC influence . This observation indicates that also at the population level there is a trade-off between the ability of the DG to encode a specific position and a wall shape . Importantly , considering all population and individual cell metrics , an input with balanced MEC and LEC contribution provided the best fit to the experimental data [19] . Having established how to correctly simulate the DG and thus its input to CA3 , we analyzed CA3 responses during morphing . We first analyzed the CA3 population response without recurrent connections . In our simulations of such a network , CA3 cells exhibited several properties consistent with the data . The distribution of the number of place fields per active cell were similar to that observed experimentally ( Figure 3A , B ) . Peak place field firing was at 12 . 45 Hz±7 . 73 in simulation , which is comparable to 13 . 13 Hz±7 . 97 reported by Leutgeb et al . ( 2007 ) . However , although rate remapping was observed during morphing , it was not consistent with the experimental data ( Figure 3C–F ) , and this was true irrespective of the ratio ( β ) of DG-to-EC input . With increase of β , there was a general reduction of the correlation between rate maps in different environments ( Figure 3C ) and virtually no change in the average rate of the cells ( Figure 3D ) . With respect to the population response to morphing , high values of β resulted in an overall increase of PV decorrelation when compared to the condition with low β ( Figure 3E ) , thus not fitting the data [18]–[20] . Yet stronger DG input decreased the CA3 PV autocorrelation in distant positions ( Figure 3F ) . In conclusion , we were unable to fit the CA3 data using a model without recurrent collaterals . We next analyzed whether the morphing data could be accounted for if recurrent collaterals were included ( Figure 4 ) . Synaptic weights of the CA3 recurrent collaterals were set based on the population activity in environments 1 and 7 ( see Methods ) , emulating the experimental protocol in which the animals were familiarized to the two extreme shapes before the experiments . The addition of the excitatory feedback from the recurrent collaterals did not impair the formation of place fields and their ability to rate remap . Importantly , by increasing the strength of the recurrent synapses ( 1-δ ) , there was an increase of the correlation between rate maps of the same cell between different environments ( throughout all morphs ) , leading to an almost flat response , as observed experimentally ( Figure 4A ) . Such enhanced stability of the firing rate distribution of individual cells indicates that the place fields are present and unmoved throughout the morphing . We next examined whether single cells were still responsive to morphing by measuring the cumulative change in the average firing rate of individual cells as morphing progressed . We found that the addition of the recurrent collaterals affected the average change in rate differently for small and large morphs ( Figure 4B ) : for small morphs , the average change in rate was less than in the condition without recurrent collaterals; for large morphs , the average change in rate was higher than in the condition without recurrent collaterals . This indicates that not only are different wall shape conditions successfully encoded in the individual cells rate maps , but also that , for very similar inputs , the system attracts the average rate response to the stored pattern . Thus , the addition of the recurrent collaterals favors a code in which the information about the environment is encoded by the peak rate of place fields located at fixed positions . We found similar results when analyzing the population response to morphing: there was an overall increase in the PV correlation measured between sessions with different wall shapes approximating experimental observations ( Figure 4C ) . For the parameters that led to the best model fit ( 1-δ = 95% ) , there was a stronger increase in the PV correlation for the small morph than for the large morph ( Figure 4D ) . Moreover , we observed an additional reduction in the PV autocorrelation obtained in distant positions , approximating the observed value ( Figure 4E ) . This indicates that the activity of the recurrent collaterals enhances the ability of the network to discriminate between unrelated positions . Altogether , these analyses show that the addition of recurrent collaterals allows an accurate description of rate remapping as seen in both the single-cell and population responses . With all parameters set , we next directly compared the response to morphing in DG and CA3 . The simulated data not only provided a model fit of the individual region response to morphing , but also provided a reasonable description of the relation between the population response of the CA3 and the DG to morphing ( Figure 5 ) . The experimental finding that DG population activity was more strongly affected by the small morph than the CA3 population activity was only observed when recurrent collaterals were present ( Figure 5A ) . Notably , although the CA3 PV correlation was increased during both small and large morphs by the recurrent collaterals , this effect was stronger for small morphs than for large morphs , indicating the existence of a basin of attraction ( ΔPV correlation of 0 . 35 for small morph against 0 . 20 for large morph , Figure 4D ) . Further , we found additional evidence for a basin of attraction by analyzing how single CA3 cells changed their firing rate throughout morphing ( Figure 5B ) ; for the large morphs ( 1–7 ) in which little effect of the attractor dynamics is expected , we found a higher change in the average firing rate in CA3 when compared to DG in the presence of recurrent collaterals ( rate overlap in CA3 is ∼0 . 2 lower than in DG ) , setting the baseline of how the rate of DG and CA3 cells is affected by morphing . For the small morphs ( 1–2 ) , the condition in which the attractor dynamics would be effective , in the CA3 there was a lower change ( rate overlap in CA3 is ∼0 . 05 higher than in DG ) in the average firing rate in CA3 than in DG when recurrent collaterals were included . These results indicate that , even though CA3 cells are naturally more sensitive to changes in the environment when it is out of a basin of attraction ( as seen by the baseline results of the large morph ) , when we consider the conditions in which attractor dynamics are effective , there is a lower sensitivity to change in the CA3 cells . Notably , we also found that the addition of collaterals contributed to the spatial stability of place fields in the CA3; only in the presence of recurrent collaterals were individual rate maps of CA3 cells less affected by morphing than individual rate maps of DG cells ( Figure 5C ) . The analysis of the dynamics of CA3 rate coding also revealed the role of the feedforward excitation and competitive inhibition in pattern separation , as there is a considerable reduction in the PV autocorrelation between two distant and unrelated areas ( Figure 5D ) . Also , consistent with previous findings that a two-stage process increases spatial specificity [35] , we observe a reduction of the mean number of place fields in CA3 ( Figure 5E ) . These results allow the identification of the specific role of the neural circuits of DG and CA3 in memory: while the convergence of excitatory feedforward input and the internal inhibitory competition cause pattern separation , the recurrent excitation has a major role in pattern completion . Importantly , neither in the reported data nor in the simulation was there any evidence that morphing produced a sharp rather than a graded transition in any of the computed measurements ( Figure 5A ) . Thus , three important conclusions follow . First , the recurrent connections in CA3 do have an attractor function; during small morphs , they “attract” the dynamics toward a stored pattern ( e . g . , the square or the round shape ) . Second , this attractor dynamics modulates rate remapping , thereby leaving the spatial information intact . Third , despite this attractor function , the CA3 firing pattern undergoes a graded rather than abrupt change during morphing over intermediate states ( i . e . , 4 and 5 ) . To characterize the mechanisms by which the recurrent collaterals affect the population response to morphing , we analyzed the dynamics of single cells during small morphs in the presence and absence of recurrent excitation . The small morph is a condition in which there should be a moderate but still noticeable change in the input pattern to CA3 . In the presence of attractor dynamics , the input pattern will be within the stored pattern basin of attraction and thus pattern completion should be observed . We analyzed the firing pattern produced under these conditions and the subsequent influence of recurrent excitation . The small morph had three important effects ( Figure 6 ) . First , in cells whose total feedforward input , including the excitatory current from EC and DG , was strong ( 0 . 9 nA ) and led to a spike , the presence of recurrent excitation did not yield a significant increase in the probability of an action potential ( Figure 6A ) . Second , in CA3 cells that were part of the stored pattern but received DG/EC input after morphing that was subthreshold ( 0 . 6 nA ) , the recurrent input triggered a spike , thereby producing pattern completion ( Figure 6B ) . In the absence of recurrent excitation , such cells would not fire . This explains why there is a higher PV correlation between the population responses to a stored pattern and a small morph in the presence rather than in the absence of recurrent excitation ( Figure 5A ) . Thus , in this way , the internal dynamics provided by the CA3 recurrent synapses attracts a cell toward a stored pattern , thereby producing rapid pattern completion within a single gamma cycle [25] . Third , what the dynamics of the attractor cannot do is erase spikes that have already occurred . Consider that , after a small morph , a cell is strongly excited by DG/EC that is not part of the nearby stored pattern ( Figure 6C ) . Because this spike has occurred and cannot be erased , the total activity during the short firing period cannot be identical to the stored pattern . Likewise , activity induced by additive noise cannot be suppressed . Thus , although recurrent excitation can attract CA3 to a stored pattern , this attraction cannot be perfect . The understanding that attractor dynamics cannot eliminate spikes that are not part of the stored pattern has further implications . In the morphing experiments , the smallest morphs ( 1 , 2 ) displayed a PV less than 1 ( 0 . 9 ) . However , because two measurements in environment 1 ( albeit with intermediate sessions in all other environments , i . e . , 1-1′ is obtained from the sequence 1-2-3-4-5-6-7-1 with six intermediate sessions ) also showed a PV correlation of 0 . 9 [19] , it was suggested that an attractor mechanism made the response in environment 2 identical to that in environment 1 . Our analysis , however , suggests that such perfect attractor reconstruction cannot occur , and we suggest an alternative explanation: that the intermediate sessions between the two recordings in environment 1 altered the stored attractor , thereby reducing the correlation in the 1-1′ morphing to 0 . 9 . Thus , the 1-2 environments evoked different responses because the attractor system does not work perfectly as explained above , whereas the 1-1′ environments evoked different responses because of the learning produced in intermediate environments . We simulated the morphing procedure with varying learning rates and observed that the 1-1′ ( 1-2-3-4-5-6-7-1 ) and the 1-2 correlation were not equally affected by the exposure to different environments ( Figure 7A ) . Interestingly , because of the sequence in which the wall shapes were changed , the correlation of the 1-1 morphing changed more thoroughly to higher learning rates due to the fact that there were more intermediate trials ( n = 6 ) between the comparisons when compared to the correlation of the 1-2 environments ( n = 0 ) , which allowed that , for a specific learning rate , both comparisons are equivalent . Subsequent work supports our interpretation: when 1-1 comparisons are made without exposure to intermediate environments , the PV correlation was higher ( 0 . 93 in Figure 5 of [20] , 0 . 96 in Figure S5 of [55] ) . In the same studies , the PV correlation was lower [0 . 90 in 20 , 0 . 91 in 55] if there were intermediate exposures to other environments and was progressively reduced with the number of such exposures , as would be predicted if these exposures produced learning and a modification of the stored attractors . Further evidence of experience-dependent plasticity in the recurrent collaterals is that hysteresis was observed in CA3 , but not in the DG [19] . In our simulations with learning in the recurrent collaterals , we observed comparable levels of hysteresis in the CA3 rate maps ( Figure 7B ) . We next investigated how place cells respond under conditions in which environmental change does produce grid cell realignment . We investigated how grid cell realignment affects the population response to morphing in the CA3 . Grid cells were shown to realign when the animal is trained at the same location but in different boxes or at different locations but with the same box [40] . Under these conditions , place fields do not remain stable at the same positions , characterizing global remapping [40] , [54] . During morphing , grid cells seems to realign at an intermediate position , causing an abrupt change in the CA3 population neuronal activity [21] , [56] ( Figure 8 , left ) . To verify whether our model produces results in accordance to the literature , we realigned the grid cell population in the middle of morphing ( see Methods ) and computed the activity of CA3 cells ( Figure 8 , right ) . We observed that , following the realignment of the grid cells , there was an intense and abrupt change in the PV correlation . This effect is further supported by the observation that the change in the PV correlation during morphing is graded when grid cells are stable [19] . Our data thus corroborate the view that grid cell stability is required for rate remapping in the DG and CA3 .
We have addressed the question of whether the CA3 memory system can be considered an attractor network in the face of ostensibly conflicting experimental results . Using a simulation of the EC/DG/CA3 system , we show that firing patterns recorded in CA3 during the morphing of an environment are in accord with what is expected if CA3 is an attractor network . When the environment is subject to small morphs , DG granule cells , which do not have recurrent synapses , change their firing patterns substantially . In contrast , CA3 cells , which do share recurrent plastic connections , change much less , indicating an attraction to a stored pattern . Importantly , our simulated observations are in accord with experimental data [19] , [57] . Given that DG provides strong input to CA3 [50] , attraction of CA3 cells to a stored pattern must be due to recurrent activity within CA3 itself . Our simulations show that the recurrent collaterals of CA3 can produce these dynamics and do so within a short time interval consistent with the theta-phase specific firing of CA3 cells [24] . Importantly , our work clarifies the issue of whether sharp transitions during morphing are a requirement for demonstrating that a network follows an attractor dynamic . The argument that CA3 might not be an attractor network [21] was based on the observation that sharp transitions in PV correlation did not occur during morphing , thus not displaying a criterion of attractor networks . This criterion was suggested by work in which attractor networks were activated by brief external inputs and were then allowed to evolve to a stored pattern after the external input was removed [2] , [11] , [12] . The sharp transition occurs because without external input , attractor networks are all-or-none; with dynamics unconstrained by external input , the network uses internal dynamics to converge to the closest of the stored memories . For this reason , the final state of the network does not show intermediate states , and sharp transitions are expected . Such a feature is , however , not applicable to the hippocampus because external input from the EC and from DG to CA3 is never absent . Under these conditions , our simulations show that sharp transitions do not occur ( Figure 5C and 6A ) . Thus , under the conditions of the morphing experiment , sharp transitions are not an appropriate criterion for identifying an attractor network . There is a specific case in which sharp transition can be observed in the CA3 [21] , [56]: if the animal is familiarized with the two extreme shapes with altered distal cues , a different spatial coordinate system is assigned for each memory . As the EC globally remaps with different distal cues , a sharp transition in the CA3 will occur but will be caused by changes in cortical activity and cannot be attributed to attractor dynamics in CA3 [40] . In the experiments that we have analyzed here , distal cues were kept unaltered , and this prevents global remapping in MEC . For CA3 to function as an associative memory , the recurrent synapses must be able to undergo activity-dependent changes in their synaptic strength . Indeed , work in the slice preparation has clearly shown that these synapses can undergo long-term potentiation ( LTP ) [9] , [10] , but there has been no previous in vivo demonstration that these synapses can change in response to environmental stimuli . We argue that aspects of the data reported by Leutgeb et al . [19] strongly argue that the attractors formed in CA3 are continuously subject to learning . Indeed , this is demonstrated by the fact that exposing rats to intermediate environments is sufficient to produce a modest change in CA3 PV correlation and thus its synapses ( Figure 7 ) [20] , [55] . The key observation is hysteresis of the PV; if an altered environment is interposed between two test sessions in the same environment , the PV in the two identical environments will be slightly altered . Importantly , this hysteresis is not observed in DG [19] , strongly suggesting that it occurs because of the plasticity within the recurrent connections of CA3 . Indeed , we are able to reproduce these hysteresis effects in our model that simulates the effects of experience-dependent Hebbian plasticity in the CA3 excitatory recurrent connections . This analysis of morphing suggests future experiments investigating the role of attractors and their modification by learning . Given that attractor dynamics can now be more precisely identified , it would be of interest to test directly whether NMDAR action during learning of the square/round environments is necessary for attractor formation , as would be predicted based on in vitro studies analyzing pattern completion [58] . Indeed , following this prediction , NMDAR seems to be required during memory formation , as shown by the fact that pattern completion during subsequent recall is prevented [59] . NMDARs are not required during memory recall [60] . This is consistent with the observation that the latter effect depends on the fast dynamics of our model . Additionally , a second type of analysis could investigate discretization during learning [61]; it has previously not been possible to experimentally address the question of how finely the world is divided , but it is now approachable through the study of CA3 attractors in particular , by addressing both the temporal and the spatial ranges of this memory segmentation . In addition , we can speculate that GABA-dependent dendritic shunting of spike-time-dependent learning can assure that also the learning dynamics is restricted to single gamma cycles [62] . From the model presented here , we will be able to estimate the average size of the population of CA3 neurons that define a distinct memory , their interrelation , and the drift that they might be subject to . In addition , their embedding in a theta-gamma code raises the question of whether single memory segments defined in a single gamma cycle are , in turn , integrated in hierarchical structures following the theta rhythm . Further , the drift of CA3 memory that we have identified would suggest that , for a more permanent storage of memory segments , other structures will have to be engaged to solve the so-called plasticity-stability dilemma [63] . Finally , the ability of rather short exposures to altered environments to change the attractor properties of CA3 facilitates the study of learning in a defined network . This may allow the analysis of the spike patterns that lead to learning , the role of neuromodulators , and the role of repetition/replay in producing long-lasting synaptic modification . The demonstration that CA3 cells display the properties expected of an attractor network carries special significance because it provides the key remaining evidence , i . e . , analysis of in vivo data , that is necessary to establish the associative memory function of CA3 . As discussed , the existence of modifiable recurrent connections in CA3 suggested that CA3 is an attractor network . Consistent with this hypothesis , a mutation that disables synaptic plasticity in CA3 prevents behavior that is dependent on pattern completion [59] , [64] . Additionally , signatures of experience-dependent plasticity and pattern completion have been obtained in vitro with CA3 slices [58] . Thus , taken together , the anatomy , the behavioral experiments , in vitro electrophysiology , and our analysis of in vivo recordings make a strong case that CA3 is , in fact , an associative memory structure that follows attractor dynamics . The CA3 network analyzed here is thus among the very few cases in which the evidence regarding network , cellular , and anatomical properties has converged to explain an important aspect of memory and behavior . | A type of neural network called an “attractor network” is thought to underlie memory associations . Importantly , when such a network is presented with part of a memory , the network activity is attracted to the complete memory . However , it has been difficult to obtain clear experimental evidence for such attractor networks . Indeed , recent “morphing” experiments that were specifically designed to observe these attractor dynamics in the hippocampus did not obtain the expected results , leading to a controversy on the validity of the attractor hypothesis of memory . Here , we have built a computational model of the relevant hippocampal areas , including its core anatomical and physiological features , and through the use of large-scale computer simulations reveal in detail the physiological properties expected of the hippocampal attractor network during morphing experiments . We show that the experimental results obtained are actually those to be expected of an attractor network when the specifics of the experimental protocol are taken into account . Most importantly , the results directly demonstrate the attraction of CA3 activity to a stored pattern . Our results , together with previous behavioral and in vitro studies , provide strong evidence that CA3 is an attractor network for associative memory . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"circuit",
"models",
"neural",
"networks",
"computational",
"neuroscience",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"neuroscience",
"learning",
"and",
"memory"
]
| 2014 | A Signature of Attractor Dynamics in the CA3 Region of the Hippocampus |
Ankylosing spondylitis ( AS ) is a highly heritable immune-mediated arthritis common in Turkish and Iranian populations . Familial Mediterranean Fever ( FMF ) is an autosomal recessive autoinflammatory disease most common in people of Mediterranean origin . MEFV , an FMF-associated gene , is also a candidate gene for AS . We aimed to identify AS susceptibility loci and also examine the association between MEFV and AS in Turkish and Iranian cohorts . We performed genome-wide association studies in 1001 Turkish AS patients and 1011 Turkish controls , and 479 Iranian AS patients and 830 Iranian controls . Serum IL-1β , IL-17 and IL-23 cytokine levels were quantified in Turkish samples . An association of major effect was observed with a novel rare coding variant in MEFV in the Turkish cohort ( rs61752717 , M694V , OR = 5 . 3 , P = 7 . 63×10−12 ) , Iranian cohort ( OR = 2 . 9 , P = 0 . 042 ) , and combined dataset ( OR = 5 . 1 , P = 1 . 65×10−13 ) . 99 . 6% of Turkish AS cases , and 96% of those carrying MEFV rs61752717 variants , did not have FMF . In Turkish subjects , the association of rs61752717 was particularly strong in HLA-B27-negative cases ( OR = 7 . 8 , P = 8 . 93×10−15 ) , but also positive in HLA-B27-positive cases ( OR = 4 . 3 , P = 7 . 69×10−8 ) . Serum IL-1β , IL-17 and IL-23 levels were higher in AS cases than controls . Among AS cases , serum IL-1β and IL-23 levels were increased in MEFV 694V carriers compared with non-carriers . Our data suggest that FMF and AS have overlapping aetiopathogenic mechanisms . Functionally important MEFV mutations , such as M694V , lead to dysregulated inflammasome function and excessive IL-1β function . As IL-1 inhibition is effective in FMF , AS cases carrying FMF-associated MEFV variants may benefit from such therapy .
Ankylosing spondylitis ( AS ) is a common form of arthritis affecting primarily the spine and pelvic sacroiliac joints . Twin studies indicate that the disease is highly heritable , with over 90% of the risk of developing the condition being genetically determined [1 , 2] . To date 114 loci have been identified as being associated with the disease , contributing roughly 30% of the overall risk [3] . The most strongly associated variants at these loci are all common ( minor allele frequency ( MAF ) >5% ) or low-frequency ( MAF 1–5% ) variants , and no rare variant ( MAF<1% ) has yet been demonstrated to be AS-associated at genome-wide significance ( P<5×10−8 ) . AS is found in most ethnic groups with the notable exceptions of Africans and Australian Aboriginals , in whom the prevalence of HLA-B27 is very low , or as yet unidentified environmental factors protect against disease development . No GWAS has been performed to date in AS cases of Turkish or Iranian descent . These groups are of particular interest because of evidence that patients with the monogenic autoinflammatory disease , Familial Mediterranean Fever ( FMF ) , have an increased prevalence of sacroiliitis [4–9] , and that FMF-unaffected first-degree relatives of FMF patients have an increased frequency of AS [5] , suggesting an aetiopathogenic link with AS . There have also been four candidate gene case-control association studies of MEFV variants in AS , three of which have demonstrated nominal ( 10−5<P<0 . 05 ) association of the main FMF-associated MEFV-variant ( rs61752717 , M694V ) with AS [10–13] , with one marginally negative study ( P = 0 . 065 ) [14] . All of the participants in these studies were from Turkey . However , the sample sizes of these studies were relatively small ( number of patients ranging from 62 to 193 ) and , therefore , lacked power to achieve definitive significance . In order to identify new genetic variants associated with AS , and to investigate the potential association of MEFV variants in AS , we performed a genome-wide association study in case-control cohorts from Turkey and Iran .
As expected , SNPs in the major histocompatibility complex on chromosome 6p21 , and imputed HLA-alleles , were strongly associated with AS in both the Turkish and Iranian cohorts ( Table 1 ) . In the Turkish cohort the strongest SNP associations were with genotyped SNP rs17192932 ( OR = 22 . 84 , 95% CI = 17 . 55–29 . 72 , P = 5 . 64×10−120 ) , and in the Iranian cohort with rs117486637 ( OR = 31 . 08 , 95% CI = 22 . 27–43 . 38 , P = 8 . 62×10−91 ) . The strength of association of HLA-B27 with AS was similar in both Turkish and Iranian cohorts , but the proportion of HLA-B27 carriers was lower than in most studies of European descent AS cases ( 71% and 74% , respectively , compared with 80–95% in most European descent cohorts ) , consistent with previous reports . Stepwise conditional analyses confirmed risk associations with HLA-B40 subtypes in both cohorts , with HLA-B*5101 in the Turkish cohort , and with HLA-B*1503 in the Iranian cohort . HLA-B*5101 also achieved nominal significance in the Iranian cohort ( OR = 1 . 51 , 95%CI = 1 . 055–2 . 15 , P = 0 . 024 ) . Association at genome-wide significance ( P<5×10−8 ) was observed with two non-MHC loci in the Turkish cohort ( Table 2 ) , including the known AS-associated locus USP8 , and the MEFV locus , and with one known AS-associated locus in the Iranian cohort ( ERAP1; Table 3 ) . No additional loci were identified at genome-wide significance in the meta-analysis combining both cohorts . In the meta-analysis the strongest non-MHC association was with SNPs in the USP8 locus , which has a strong protective effect in AS ( exm1161045/rs148783236 , OR = 0 . 58 , P = 1 . 43×10−13 ) . The novel MEFV variant , rs61752717 , is also identified at genome-wide significance in the meta-analysis . Genome-wide significant association was also replicated with SNPs in chromosome 2p15 ( 2p15 ) and ERAP1 ( Table 4 ) . In each of these cases , the most strongly associated SNP is in strong linkage disequilibrium ( r2>0 . 8 ) with the previously reported AS-associated variant in European descent populations . Only nominal significance was seen with SNPs at IL23R , a known strongly AS-associated locus , in either dataset or in the meta-analysis analysis . There were 105 out of 113 non-MHC tag SNPs for known AS-associated loci in the combined cohort , while eight of them were missing in the dataset . Two of them ( 2p15 and ERAP1 ) were GWS in the combined cohort . One was at suggestive significance ( NOS2 ) . Twenty-one of them were at nominal significance . All these 24 tag SNPs were in the same direction of effect as they were in the cross-disease study . Suggestive associations ( 5×10−8 <P<1×10−5 ) were observed at nine novel loci in the Turkish cohort , and three novel and two known AS-associated loci ( 2p15 and FAS ) in the Iranian cohort ( Tables 2 and 3 , S1–S20 Figs ) . In the Turkish cohort , near genome-wide significant association was seen with SNPs in ADAM28 ( encoding ADAM metallopeptidase domain 28; rs1013210 , OR = 0 . 65 , 95% CI = 0 . 56–0 . 76 , P = 7 . 88×10−8 ) . No association was seen at this locus in the Iranian cohort despite the MAF being similar in Turkish and Iranian controls ( MAF = 0 . 28 and 0 . 25 , respectively ) . In the Turkish cohort , the strongest MEFV association observed was with the non-synonymous SNP rs61752717 ( OR = 5 . 34 , 95% CI = 3 . 31–8 . 62 , P = 7 . 63×10−12 ) , which encodes the M694V protein variant which is the most strongly associated FMF allele . Excluding four cases who have co-existent AS and FMF , this association remains genome-wide significant ( OR = 5 . 26 , 95% CI = 3 . 24–8 . 54 , P = 1 . 75×10−11 ) . This polymorphism was also nominally associated with AS in the Iranian cohort ( OR = 2 . 85 , 95% CI = 1 . 039–7 . 83 , P = 0 . 042 ) , its MAF being 3 . 3-times lower than in the Turkish controls ( MAF = 0 . 0033 in Iranian controls , 0 . 011 in Turkish controls; Table 5 ) . In the meta-analysis , association was seen with OR = 4 . 76 , 95% CI = 2 . 94–7 . 68 , P = 1 . 72×10−12 . Nominal association was seen with multiple other MEFV SNPs in both the Turkish and Iranian cohorts , but none of these are considered FMF-associated , and none were shared between the two cohorts ( S1 Table ) . Previous studies have suggested interaction between HLA-B27 and MEFV in AS ( OR = 8 . 69 in HLA-B27-positive , and 21 . 8 in HLA-B27-negative cases ) [11] . In the current study , rs61752717 shows stronger association in HLA-B27-negative patients ( P = 8 . 93×10−15 ) than in HLA-B27-positive patients against all controls ( P = 7 . 69×10−8 , Table 6 ) . In the Turkish cohort , in comparison of cases alone , rs61752717 risk allele carriage was higher in HLA-B27-negative than positive cases ( OR = 2 . 67 , 95% CI = 1 . 78–4 . 01 , P = 1 . 28×10−6 in whole cases cohort , OR = 2 . 39 , CI = 1 . 58–3 . 64 , P = 9 . 54×10−6 excluding four cases with co-existent FMF and AS , Fisher Exact Test , Table 6 ) , whereas no difference was observed in controls ( P = 0 . 25 ) . Compared with all-Turkish controls , the association of rs61752717 was much stronger in HLA-B27-negative cases ( OR = 7 . 8 , P = 8 . 93×10−15 in whole cohort , OR = 7 . 6 , P = 3 . 92×10−14 excluding four cases with co-existent FMF and AS ) than in HLA-B27-positive cases ( OR = 4 . 3 , P = 7 . 69×10−8 ) , although no interaction observed in a specific logistic regression analysis ( P = 0 . 97 ) . Similarly , rs61752717 risk allele carriage was higher in HLA-B27-negative than positive cases in the Turkish-Iranian combined cohort ( OR = 2 . 74 , 95% CI = 1 . 86–4 . 03 , P = 1 . 29×10−7 in entire cohort; OR = 2 . 47 , 95% CI = 1 . 66–3 . 67 , P = 5 . 78×10−6 excluding four cases with co-existent FMF and AS; Table 7 ) . However , no significant difference was observed in rs61752717 risk allele carriage between HLA-B27 positive and HLA-B27-negative patients ( OR = 1 . 44 , 95% CI = 0 . 23–6 . 40 , P = 0 . 70; Table 8 ) in the Iranian cohort , noting the extremely low power of this analysis given the rarity of rs61752717 in this cohort . The MEFV nsSNP rs61752717 is also associated with Behçet’s Disease ( BD ) [15–18] . Joint involvement is common in BD , and about 5% of the BD patients developed sacroiliitis [19] . The major genetic association with BD is with HLA-B51 [20] , which is also AS-associated [21] . rs61752717 was also more strongly associated with AS in HLA-B51-negative than in HLA-B51-positive restricted analyses ( OR = 5 . 88 , 95% CI = 3 . 31–10 . 42 , P = 1 . 36×10−9 vs OR = 4 . 82 , 95% CI = 1 . 72–11 . 66 , P = 2 . 11×10−3 , or OR = 5 . 88 , 95% CI 3 . 32–10 . 43 , P = 1 . 33×10−9 vs OR = 4 . 24 , 95% CI 1 . 58–11 . 42 , P = 4 . 22×10−3 if excluding four cases with co-existent FMF and AS; S2 Table ) but the difference was not statistically significant , nor was interaction observed in a specific logistic regression analysis ( P = 0 . 69 ) . No significant difference was observed in rs61752717 risk allele carriage between HLA-B51-positive and HLA-B51-negative patients ( OR = 1 . 44 , 95% CI = 0 . 23–6 . 40 , P = 0 . 70 ) in the Iranian cohort . Similarly , there was no significant interaction identified in the Iranian cohort ( P = 0 . 975; S3 Table ) nor in the combined cohort ( P = 0 . 994; S4 Table ) .
This study confirms that MEFV is associated with both HLA-B27-positive and–negative AS , even in the absence of clinical evidence of FMF . The association represents the first rare variant association with AS , and has the highest odds ratio for disease of any non-MHC reported locus to date , indicating a major effect on disease pathogenesis . The strength of association observed in the Turkish and Iranian populations was , as expected , closely related to the MAF of the polymorphisms involved , and therefore the absence of association of MEFV with AS in previous studies of other populations may simply due to ethnic differences in gene frequencies rather than differences in the underlying mechanisms driving disease . Further studies examining , for example , gene-expression in patients and healthy controls and in relation to disease activity , will be required to determine whether MEFV-driven autoinflammatory processes are a factor in AS in patients in populations where MEFV functional variants are much rarer . The strength of the association of the M694V MEFV polymorphism with AS is much stronger in HLA-B27-negative than–positive AS . Whilst a formal test of interaction between HLA-B27 and rs61752717 was negative , the stronger association in HLA-B27-negative cases is consistent with previous studies [11] , and suggests that this is the third example of epistasis found in AS genetics , following the previously demonstrated examples of ERAP1 variants and HLA-B27 and HLA-B40 . The strength of association in HLA-B27-negative subjects is considerable and is to our knowledge the strongest effect in terms of odds ratio of any non-MHC variant in a common immune-mediated disease . In the absence of a definite explanation as to how HLA-B27 induces AS , it is not possible to provide a functional explanation for to this finding . However , one possible hypothesis is that HLA-B27-positive disease is primarily driven by different immunological pathways than HLA-B27-negative MEFV-positive disease . The replication of the association of HLA-B51 with AS , previously observed in western European-descent AS [21] , in the Turkish and Iranian cohorts in the current study also provides further support that AS and Behcet’s disease have overlapping aetiopathogenesis . Behçet’s disease is also associated with MEFV polymorphisms , and Behçet’s disease cases carrying MEFV variants have more severe disease [15–17] , suggesting that autoinflammation may also contribute to its development . This study also provides further confirmation of the association of HLA-B40 alleles with AS [25 , 26] . The study sample size was too small to test whether gene-gene interaction with ERAP1 variants was observed , as has previously been demonstrated in AS cases and controls of western European-descent with both HLA-B27 and HLA-B40 [21] . We also demonstrated that serum IL-1β levels are higher in Turkish AS cases than healthy controls , and are higher in carriers of the MEFV M694V polymorphism . Pyrin , encoded by MEFV , has been shown to negatively or positively regulate caspase-1 and IL-1β activation , depending on the experimental system used [27–29] . In a loss of function model , MEFV variants operate by attenuating the apoptosis-associated speck-like protein containing a caspase recruitment domain ( ASC , encoded by Pycard ) -dependent and ASC independent inhibitory effects of pyrin on caspase activation and subsequent IL-β production [27 , 28] , whereas in a gain-of-function model they lead to caspase-1 activation through an ASC-dependent , NLRP3-independent , pyrin inflammasome [29] . In either model MEFV variants lead to excessive IL-1β production . Pyrin is also a member of the tripartite motif ( TRIM ) family proteins , which are critically involved in regulation of autophagy and innate immunity . Pyrin/TRIM20 , not only acts as a specific receptor for NLRP1 , NLRP3 and pro-caspase 1 , but also serves as a platform for assembly of key regulators ( ULK1 , Beclin1 , and ATG16L1 ) and effector factors ( mAtg8s ) of autophagy initiation machinery [30] . Through these abilities , pyrin leads to degradation of key inflammasome targets in a highly selective manner , resulting in suppression of caspase-1 activation and IL-1β production . Pyrin variants harboring FMF-associated B30 . 2 mutations , including the M694V variant , have been shown to be associated with a deficiency in the autophagic degradation of NLRP3 [30] . This finding adds more to the complexity and multiplicity of possible mechanisms underlying the excessive IL-1β production seen in patients with FMF and many other autoinflammatory diseases . Pyrin/TRIM20 , whose expression is significantly up-regulated by IFN-gamma [29] , is one of the required mediators of IFN-gamma-induced autophagy [30] , providing further support for pyrin’s regulating role in restraining excessive inflammation induced by innate immunity . In the light of above-mentioned findings , increased M694V prevalence in AS patients observed in our study adds to the mounting evidence that AS is an autoinflammatory disease [31] , and also provides support to the recent hypothesis that autophagy may be involved in AS pathogenesis [32] . In the latter context , it is worth mentioning a recent study that showed decreased expressions of autophagy-related genes ( MAP1LC3A , BECN1 , and ATG5 ) in the peripheral blood mononuclear cells of AS patients as compared to controls [33] . Notably , even further decreased levels of expression of MAP1LC3A and BECN1 were found in patients with more advanced spinal damage in the same study [33] . Further , there is substantial evidence of activation of the innate immune system in AS , with studies demonstrating that ILC3 , MAIT and γδ cells are major sources the pro-inflammatory cytokines IL-17 and IL-22 in the disease [34–36] , consistent with a role for autoinflammation in AS pathogenesis . Previous AS genetic studies have demonstrated genome-wide significant associations of the IL-1 receptor genes , IL1R2 and IL1R1 [37] , further supporting involvement of this cytokine pathway in the disease . Serum IL-1β levels have been shown previously not to be elevated in AS patients of western European descent [38 , 39] . Similarly , no difference in serum IL-1β levels were seen in a previous study of Turkish AS cases and controls [40] , although in this study no information is reported on the age and gender matching of the cases and controls . Put together , this evidence suggests that IL-1 related processes are important in AS in different ethnic populations , rather than being restricted to MEFV M694V carriers , but that in this group IL-1 plays a major immunopathogenic role . This may explain why minimal association was seen with IL23R variants in this population , perhaps suggesting that the IL-23 cytokine pathway is less important in this group of patients , where IL-1 effects may predominate . The low efficacy of IL-1 blockade with the IL-1 receptor antagonist anakinra in AS suggests that the disease , at least in the western European population studies , is driven by factors largely independent of IL-1 [22 , 41] . In contrast , IL-1 blockade is highly effective for FMF caused by the same MEFV polymorphism we demonstrate here to be associated with AS . Our genetic and cytokine level findings support the hypothesis that , at least in patients with the M694V MEFV variant , IL-1 blockade is a potential worthwhile therapeutic option . Anakinra has been shown to be effective in relieving signs and symptoms of spondyloarthritis associated with FMF [42–44] . Of particular interest , one of these cases had severe axial pain and elevated acute phase markers persisting despite treatment courses with adalimumab and etanercept , but both the clinical symptoms and acute phase reactant levels improved dramatically with IL-1 blocking therapy [44] . Bacterial pathogens are thought to also be drivers of inflammation in AS , as suggested by microbiome studies demonstrating differences in the gut microbiome in AS cases and controls [45] , and the association of the bacterial lipopolysaccharide innate immune receptor TLR4 variants with AS [3] . MEFV variants , and particularly the M694V polymorphism , are also associated with both ulcerative colitis and Crohn’s disease in Turkish patients , suggesting a particular role for MEFV in gut mucosal inflammation [46 , 47] . FMF patients have been shown to have intestinal microbiome dysbiosis , both whilst in remission and during attacks , consistent with models of FMF whereby gut bacterial interactions with a genetically primed host immune system drive disease . At genus level resolution , similarities between what was seen in FMF patients and AS cases are striking , with increases in carriage of Ruminococcus , Porphyromonadaceae , and reductions in Prevotella seen in both diseases [45 , 48] . Further studies in AS and FMF cases matched for ethnicity and other relevant covariates would appear indicated to determine if specific bacterial species or components are common drivers or have protective effects in these two diseases . In conclusion , this study demonstrates that the MEFV M694V ( rs61752717 ) SNP markedly increases the risk of AS even in patients not suffering from FMF , and is associated with increased serum IL-1β levels in those patients , suggesting that IL-1 inhibition may be effective in that subset of AS patients , and that MEFV-driven autoinflammation is a factor in the aetiopathogenesis of AS .
All case and control participants gave written , informed consent . The study was approved by the relevant research ethics authorities at each participating centre , and overall approval was given by Queensland University of Technology Health Research Ethics Committee ( approval number HREC/05/QPAH/221 ) . AS was defined according to the modified New York criteria . In total , the study involved 1001 Turkish AS patients , 1011 Turkish controls , 479 Iranian AS patients and 830 Iranian controls . All cases were specifically screened by recruiting clinicians for a history of FMF . DNA from all subjects were genotyped using the Illumina CoreExome chip following standard protocols at the Australian Translational Genomics Centre , Princess Alexandra Hospital , Brisbane . All Turkish samples were genotyped by CoreExome chips v1 . 0 . 479 Iranian AS cases and 197 Iranian controls were genotyped by CoreExome chips v1 . 0 , while the remaining 633 Iranian controls were genotyped by CoreExome chips v1 . 1 . Bead intensity data was processed and normalized for each sample and genotypes called using the Illumina Genome Studio software . A subset of 20 carriers of the MEFV M694V had the microarray genotypes tested by Sanger sequencing , with complete concordance of results . SNP genotypes from CoreExome v1 . 0 and v1 . 1 chips from the Iranian cohort were merged based on the manifest files and plink files . 349 SNPs were excluded due to uncertain strandedness or large difference ( > 10% ) in control allele frequencies between chip versions . We also removed the redundant SNPs with the same minor and major alleles in the same position but with different SNP IDs by keeping the one with lower sample missing rate . Quality control ( QC ) was completed separately on individual cohorts , and included assessment of missingness by individual ( threshold <5% ) , missingness by genotype ( threshold<5% ) , Hardy-Weinberg equilibrium in controls ( P < 1×10−6 ) , extreme heterozygosity ( threshold > 3 standard deviations from mean ) and identity by descent threshold at the inflation point of PI_HAT ( 0 . 2013 and 0 . 0981 for Turkish cohort and Iranian cohort , respectively ) . For each pair of related samples ( PI_HAT > threshold ) , the sample with the higher missing rate was removed , and where the pair involved a case and a control with similar call rates ( absolute difference in missing rate < 0 . 0005 ) , cases were preferentially selected for inclusion . Genotyped SNPs with MAF > 0 . 05 were then used to perform principal component analysis for ethnicity identification using SHELLFISH ( http://www . stats . ox . ac . uk/~davison/software/shellfish/shellfish . php ) . Principal components analysis ( PCA ) was performed with 0–10 eigenvectors . Ethnic and ancestry outliers ( more than 6 standard deviations from the mean on any principal component ) were excluded . Ski-plots of λ and λ1000 for different numbers of eigenvectors as covariates for the Iranian , Turkish and combined cohorts are shown in S21 and S22 Figs , and plots of the principal components analysis ( first vs second principal component ) in S23–S25 Figs . For the Iranian cohort , adding any principal components increased the λ nor λ1000 . The best λ was produced using no principal components ( 1 . 011 , λ1000 = 1 . 022 ) . In the Turkish cohort , use of the first five principal components produced λ at 1 . 022 ( λ1000 = 1 . 024 ) . Use of additional components did not reduce the λ nor λ1000 further . The quantile–quantile plots for the Iranian , Turkish and combined cohorts are shown in S26–S28 Figs , respectively . After QC , there were 1 , 828 Turkish samples ( 921 cases and 907 controls ) and 1 , 176 Iranian samples ( 422 cases and 754 controls ) . After SNP QC , there were total of 294 , 091 and 293 , 283 SNPs in the Turkish and Iranian cohorts , respectively . Genotypes were imputed in candidate regions for both cohorts using the 1000 Genomes Project ( Phase 1 integrated v3 ) . Genotype data were phased with SHAPEIT [49] , and genotypes were imputed with Impute2 [50] . SNPs with low imputation quality ( r2 < 0 . 6 ) were excluded . HLA alleles and Amino Acid Polymorphisms in HLA region were imputed by SNP2HLA with T1DGC reference ( collected by Type 1 Diabetes Genetics ) [51] . Logistic association analyses were performed using PLINK ( v1 . 90b3 . 36 , https://www . cog-genomics . org/plink2 ) [52] to perform with the first five principal components as covariates for population stratification control in the Turkish cohort . Covariates were not added in the Iranian cohort as there was little population stratification ( with any covariates the λ increases ) . Genome-wide significance was accepted at P<5×10−8 , and genome-wide suggestive at P<1×10−5 . Association analysis in imputed genotypes was assessed with PLINK best guess genotypes . Cluster plots for reported SNPs were checked manually in the cases and controls in the respective cohorts and with each different CoreExome chip version separately . The meta-analysis was performed by inverse-variance method implemented in the software package METAL [53] . Testing for interaction between HLA-B51 and HLA-B27 with the MEFV lead SNP rs61752717 in the corresponding cohort was performed by logistic regression fitting a dominant term for the respective HLA-B allele status ( positive or negative ) and an additive term for SNP rs61752717 , including a multiplicative interaction term , and the N corresponding principal components for population stratification correction ( N = 5 , 0 and 4 for Turkish , Iranian and combined cohort , respectively ) : Phenotype~rs61752717genotype+HLA-Bstatus+HLA-Bstatus*rs61752717genotype+PC1+⋯+PCN where Phenotype was coded as 2 ( patient ) and 1 ( healthy control ) , rs61752717 genotype was coded as 0 ( genotype CC ) , 1 ( genotype CT ) and 2 ( genotype TT ) to reflect additive effect , HLA-B allele status was coded as 0 or 1 to reflect dominant effect; HLA − Bstatus*rs61752717genotype was the interaction term for HLA-B and rs61752717 genotype; PCi codes was the ith principal component from PCA from the corresponding cohort . Concentrations of IL-1β and IL-17A in serum were measured using the Human IL-1β/IL-1F2 Quantikine HS ELISA kit ( R&D Systems ) and Human IL-17A High Sensitivity ELISA kit ( Life Technologies ) according to manufacturer’s instructions . Statistical significance of differences was evaluated using the Student’s t-test . SHELLFISH; http://www . stats . ox . ac . uk/~davison/software/shellfish/shellfish . php PLINK v1 . 90b3 . 36; https://www . cog-genomics . org/plink2 | Ankylosing spondylitis ( AS ) is a highly heritable immune-mediated arthritis . To identify new genetic associations with AS , we performed genome-wide association studies in Turkish and Iranian AS patients and controls . We identified a novel rare coding MEFV variant associated with AS . Rare polymorphisms of MEFV , which encodes the protein pyrin , are known to cause Familial Mediterranean Fever ( FMF ) , a monogenic , autosomal recessive , autoinflammatory disease which can be complicated by arthritis . 99 . 6% of Turkish AS cases , and 96% of those carrying the MEFV variant , did not have FMF , and the association with AS remains excluding cases with FMF . In Turkish subjects , the MEFV variant association was particularly strong in HLA-B27-negative cases , but also positive in HLA-B27-positive cases . This represents the first rare variant association with AS , and has the highest odds ratio for AS of any non-MHC reported hitherto , indicating a major effect on disease pathogenesis . We assessed serum cytokine levels in the cohort , and found that IL-1β , IL-17 and IL-23 levels were higher in AS cases . Furthermore , among AS cases , IL-1β and IL-23 levels were increased in MEFV variant carriers compared with non-carriers . This study has therapeutic implications; as IL-1 inhibition is effective in FMF , AS cases carrying FMF-associated MEFV variants may benefit from such therapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"statistics",
"immunology",
"variant",
"genotypes",
"alleles",
"genetic",
"mapping",
"multivariate",
"analysis",
"clinical",
"medicine",
"linkage",
"disequilibrium",
"mathematics",
"genome",
"analysis",
"molecular",
"genetics",
"research",
"and",
"analysis",
"methods",
"genomics",
"mathematical",
"and",
"statistical",
"techniques",
"principal",
"component",
"analysis",
"molecular",
"biology",
"genetic",
"loci",
"clinical",
"immunology",
"heredity",
"ankylosing",
"spondylitis",
"autoimmune",
"diseases",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"statistical",
"methods",
"human",
"genetics"
]
| 2019 | Genome-wide association study in Turkish and Iranian populations identify rare familial Mediterranean fever gene (MEFV) polymorphisms associated with ankylosing spondylitis |
A goal of many sensorimotor studies is to quantify the stimulus-behavioral response relation for specific organisms and specific sensory stimuli . This is especially important to do in the context of painful stimuli since most animals in these studies cannot easily communicate to us their perceived levels of such noxious stimuli . Thus progress on studies of nociception and pain-like responses in animal models depends crucially on our ability to quantitatively and objectively infer the sensed levels of these stimuli from animal behaviors . Here we develop a quantitative model to infer the perceived level of heat stimulus from the stereotyped escape response of individual nematodes Caenorhabditis elegans stimulated by an IR laser . The model provides a method for quantification of analgesic-like effects of chemical stimuli or genetic mutations in C . elegans . We test ibuprofen-treated worms and a TRPV ( transient receptor potential ) mutant , and we show that the perception of heat stimuli for the ibuprofen treated worms is lower than the wild-type . At the same time , our model shows that the mutant changes the worm’s behavior beyond affecting the thermal sensory system . Finally , we determine the stimulus level that best distinguishes the analgesic-like effects and the minimum number of worms that allow for a statistically significant identification of these effects .
A grand goal in understanding sensory systems is to predict the behavioral response of an organism to specific sensory stimuli; or , conversely , to infer the sensory stimuli from measurements of the behavioral response . Such a goal requires careful experimental design , precise control of the sensory inputs , and quantification of the behavioral outputs . While sensory stimuli can be carefully quantified and applied , the quantification of the behavioral output and its relationship to the stimulus is non-trivial because the behavior is often complicated and not well defined . In this work we address this grand goal in the context of studies of pain , or nociception ( i . e . , sensing of noxious stimuli , which damage or threaten to damage normal tissues ) , where behavioral quantification is especially hard . ( Note that , in this paper , we use nomenclature developed by the International Association for the Study of Pain , http://www . iasp-pain . org/Taxonomy . ) Pain studies on human subjects are difficult because of ethical constraints , difficulties in quantifying a psychophysical response , and subjectivity in self-reporting [1] . Partial conservation of molecular mechanisms of nociception among many different species [2–4] allows to solve some of these problems by using animal models . However , then the grand goal of quantifying the behavior and relating it to the stimulus becomes even harder: animal subjects cannot communicate their perceived noxious levels to us in an obvious fashion . Thus progress in using animal models depends crucially on our ability to quantitatively and objectively infer the perceived level of noxious stimuli from animal behavior . Historically , studies of nociception primarily used mammalian models [5 , 6] . For rodents , the tail flick test [7] , the hot-plate test [8 , 9] and the Hargreaves’ method [10] correlate pain perception with the reaction latency of different body parts to noxious stimuli . For larger animals such as canines [11] and primates [12 , 13] , similar nociceptive assays have also been developed . In recent years , new approaches started incorporating facial expressions in nociception quantification [14–16] . Although these mammalian models are extensively studied , several drawbacks hinder their use . First , ethical issues and risks arise for certain experiments . Second , compared to invertebrates , vertebrate subjects require more time and resources to maintain . Therefore , much effort has been devoted to investigations of the possibility of using invertebrate models in nociception research [3 , 17] . In experiments involving Drosophila larva , measures such as the response percentage of the total population [18 , 19] and the time to response [20] have been used to investigate changes in the ability of the animals to sense noxious stimuli . In experiments on Caenorhabditis elegans , behavioral features such as the turning rate [2] and the percentage of escape response [21] have been used to characterize nociception . All of these models share some common problems . First , the nociceptive assays focus on one particular coarse behavioral feature of the subject , such as avoidance behavior , orientation , or turning rate . Such features are selected in an ad hoc fashion , subject to a particular design of an experiment . This makes it difficult to compare results across different labs and experiments . Further , this does not solve the grand goal of quantifying the full stimulus-behavior relation , and thus the behavior may be providing additional information about the perceived noxious stimulus level that is not being captured by the coarse measures . Second , some assays report measurements as a percentage of a population , so that these measurements cannot be made for individuals . To overcome these problems , an ideal assay would infer a perceived noxious stimulus level of an individual animal on a continuous scale , using comprehensive , objective measurements of its behavioral profile . Solving the grand goal of quantifying the stimulus-behavior relation in the context of pain studies would allow one to use the assays to calibrate the perceived noxious stimulus level , and maybe even reductions in such levels due to analgesic-like effects of drugs , or mutations in the nociceptive pathways . At the same time , drugs or mutations can affect the motor response , rather than the nociception per se . Thus traditional pain assays mentioned above may convolve the perceived noxious stimulus reduction , if any , with behavioral changes . For example , a mutant defective for turning behavior will register a strong reduction in the turning rate , but it would be a mistake to interpret this as a reduction in nociception . Such concerns are very real , as is illustrated by a known fact that opioids can cause large behavioral changes [22] . To attribute a behavioral response difference to reduced nociception and not to motor changes , the response must be stereotyped and reflexive , which is often the case [6] . Further , only the response amplitude or frequency , but not the detailed temporal structure , should change in response to a drug or a mutation . Establishing the stability of the stereotypic response pattern requires solving the grand goal: analysis of the entire stimulus-triggered response behavior , rather than of its few selected features , as is done by most behavioral assays . In this work , we address these issues in the context of the nematode C . elegans , solving the grand goal in the context of its heat-evoked escape behavior , and hence developing the worm further as an animal model system for nociception research . The worm is a great model organism for such studies for a number of reasons . First , the behavioral dynamics of freely moving C . elegans is intrinsically low dimensional [23] . This makes quantification of its behavioral response relatively straightforward , providing an opportunity to use the entire motile behavior as a basis for assays . Second , the worms show a noxious response to a wide range of sensations including certain types of chemical [24 , 25] , mechanical [26 , 27] , and thermal [28 , 29] stimuli , and such a nociceptive response is different from and is transduced independently of the related taxis behaviors [30–34] . Third , at the molecular level , many details of heat nociception in the worm may be similar to vertebrate animals [3] . Fourth , there are powerful genetic and optical tools to reveal mechanisms of nociception in C . elegans . Finally , the low cost , small size , and absence of ethical constraints make the animal amenable to large scale pharmacological screens for new human analgesics [35] . We present combined experimental and modeling studies that show that the entire temporal behavioral profile during the heat-evoked escape response in C . elegans is highly stereotypical , with the frequency of the escape response and the amplitude of the escape velocity profile scaling with the stimulus level . By verifying the ability of the behavioral template to capture the response following a heat stimulus , the model we develop distinguishes changes in the sensory system from changes to the motor program . When a change is attributed to the sensory system , the model can infer the reduction in the perceived heat stimulus level following pharmacological or genetic treatments from the behavior of an individual worm . This quantification requires only about 60 worms to show statistically significant perceived stimulus reduction for a common human analgesic , and its statistical power quickly improves with an increasing number of subjects . Overall , this solution of the grand goal in the worm heat-evoked escape context suggests that , for C . elegans , it is possible to disambiguate perturbations to the sensory system from other perturbations affecting motility , and to quantify the reduction in the perceived heat stimulus from behavioral data . Combined with the previous evidence , this bodes well for further establishment of the worm as a model system for pain research . However , we stress that the differences between the worm and the human are so large that we do not want to overstate the importance of a C . elegans nociception model in the study of human health . Our analysis may be useful in the future for identifying molecular mechanisms of nociception ( which may be similar to those leading to pain in humans ) , or identification of drugs that affect them . Some of these drugs may even work in humans , but there is no reason to believe that pain ( and especially its emotional component ) occurs in C . elegans , or that drugs that affect nociception actually produce analgesic effect in worms . Thus the main contribution of our work is in addressing the grand goal , namely in quantitative characterization of regularities of complete heat-evoked escape behavior C . elegans on a single-subject level , and analysis of changes in these regularities under different pharmacological and genetic treatments , rather than potential applications of our findings to future discovery of new human analgesics . The paper is organized as follows . First we discuss the structure of the dataset and the model . Then we evaluate the performance of the model and discuss the stereotypical behavior we discovered . Further , we use the model to infer the heat stimulus level of worms in three different conditions: wild-type untreated , wild-type treated with ibuprofen , and a mutant with defects in thermal nociception . We argue that , while effects of the ibuprofen treatment can be attributed largely to reduced nociception , the mutant’s response shows changes to the behavior beyond nociceptive effects . Finally , we use the statistical model to discuss how the nociception experiments should be designed to achieve the highest statistical power to quantify the analgesic-like effects .
For presentation purposes , we bin the laser current of the heat stimulus into five distinct levels ( bins ) , defined to have an equal number of control worms in each bin ( 40 per bin ) . The maximum reverse escape velocity is indistinguishable among all three worm types for the largest stimulus level , indicating no gross defects to motility or noxious response . At the same time , the velocity at smaller stimuli levels shows substantial differences ( Z scores of up to 3 . 8 ) between the control and the treated worms , especially in the vicinity of ∼100 mA laser stimulation ( Fig 2A ) . Two-way ANOVA shows the difference between the ibuprofen and the control worms at p = 0 . 021 and between the mutant and the control at p = 1 . 5 ⋅ 10−5 across all five laser current groups . Mutant worms are especially different from the control over a wider stimulus range . However , as discussed above , it is unclear if such simple observed behavioral differences are indicative of the reduction of the perceived stimulus level or of other changes to the motor response . Furthermore , there might be additional changes in the detailed temporal structure of the heat-evoked escape dynamics , which would not be captured by simple statistics , such as the maximum reverse speed . To address this , instead of subjectively segmenting the complex escape behavior or choosing ad hoc metrics of the worm’s movement , we choose to infer the applied stimulus strength from a comprehensive model of the entire worm’s response velocity profile . Due to the considerable randomness and individual variability of responses , we choose to model them probabilistically ( see [39] for another example of Bayesian probabilistic modeling of nociception ) . Thus we are interested in estimating P ( I|v ) , the probability distribution of the applied laser current I conditional on the observed velocity of the escape response v ≡ {v ( t ) } . Using Bayes’ theorem , we write P ( I | v ) = P ( v | I ) P ( I ) P ( v ) = 1 Z P ( v | I ) P ( I ) , ( 1 ) where Z is the normalization factor , and P ( I ) is the prior distribution of stimuli . When we characterize the velocity profiles of the noxious response of C . elegans , we notice that the worm can react to the stimulus in two different ways . Some worms pause after the heat stimulus , even at large laser currents ( Fig 2B ) . These worms remain largely immobile for a few seconds , sometimes as long as the recording duration . Other worms actively reverse and follow the classic escape behavior ( Fig 1 ) . We choose to separate the active vs . the paused worms with a cutoff of 10 pixels/s , where 50 pixels is about one body length of the worm . To account for this heterogeneity in the behavior , we introduce the state variable s , which can take one of two values , a or p , for each individual worm . Then P ( v | I ) = P ( v | s = p , I ) P ( s = p | I ) + P ( v | s = a , I ) P ( s = a | I ) . ( 2 ) We model the probability of the paused state P ( p|I ) by a sigmoid function ( Fig 2C ) , P ( s = p | I , I 0 ) = 1 1 + ( I / I 0 ) 2 , ( 3 ) where I 0 is the pause current threshold . Then the probability of the active state is P ( s = a | I , I 0 ) = 1 - P ( s = p | I , I 0 ) = ( I / I 0 ) 2 1 + ( I / I 0 ) 2 , ( 4 ) We infer I 0 from data by maximizing ∏ i = 1 N type P ( s i | I i , I 0 ) , where Ntype is the number of trials with worms of the analyzed type , and Ii is the actual laser current for a particular trial . Note that each of the three data sets has its own pause current ( 25 . 9 ± 2 . 8 , 26 . 6 ± 1 . 9 , and 51 . 6 ± 6 . 3 mA for the control , ibuprofen , and mutant worms , respectively ) . Changes in this threshold , like that for the mutant , will result in different numbers of worms responding to the same stimulus , which can be consistent with the changes in the stimulus level , depending on whether the response profiles themselves stay stable . This is what we investigate next . Parenthetically , we note that the fraction of active worms is essentially the same as the percentage of the escape response , which has been used previously to quantify worm nociception [21] . Here we go further and additionally analyze the behavioral profiles of the responding worms . C . elegans locomotion consists of a series of stereotyped postures and behavioral states [40 , 41] . Further , in other animals , escape responses are stereotyped as well [6] . Therefore , it is natural to explore if the escape response of C . elegans is also stereotyped , separately for the paused and the active states . For paused worms , the escape velocity is small and independent of the laser current , and we model it as a multivariate normal variable , P ( v | p , I ) = 1 ( 2 π ) T 2 | Σ p | 1 2 exp - 1 2 ( v - u p ) T Σ p - 1 ( v - u p ) , ( 5 ) where up is the mean velocity profile of the paused worms measured from data , which we call the paused template velocity . Σp is the empirical covariance of the paused velocity , and T is the total number of effectively independent time points in the velocity profile time series , determined using the autocorrelation structure of the profile ( Materials and Methods ) . We expect that , in the active state , the worm escape is laser current dependent . Specifically , we seek to represent it by a current-dependent rescaling of a stereotypical escape velocity , v ∼ f ( I ) ua , where f is the scaling function , and ua is the active template velocity . Since various features of the worm escapes ( the maximum reverse speed , the maximum reverse acceleration , and the time to the omega turn ) , scale non-linearly and saturate with the laser current ( Fig 2A ) , the rescaling , f ( I ) , must be sigmoidal . Further , some worms have nonzero velocities even at zero laser current , so that f ( 0 ) may be nonzero . Finally , the overall scale of the template can be absorbed in the definition of ua . The simplest scaling function obeying these constraints has only two parameters f ( I ) ≡ f I 1 , I 2 ( I ) = I 1 + I 1 + I / I 2 , ( 6 ) where I 1 and I 2 are again constants , different for the three different worm types . With this , we write the probability of a velocity profile given the laser current I for the worm in an active state as a multivariate normal distribution P ( v | a , I ) = 1 ( 2 π ) T 2 | Σ a | 1 2 × exp - 1 2 ( v - f I 1 , I 2 ( I ) u a ) T Σ a - 1 ( v - f I 1 , I 2 ( I ) u a ) , ( 7 ) where Σa is the covariance of the average velocity profile . We find the constants I 1 and I 2 , ua and Σa by maximizing the likelihood of the observed data ( Materials and Methods ) . In summary , the probability of a velocity profile given the laser current in a certain trial is P ( v | I ) = 1 1 + ( I / I 0 ) 2 × 1 ( 2 π ) T 2 | Σ p | 1 2 exp - 1 2 ( v - u p ) T Σ p - 1 ( v - u p ) + ( I / I 0 ) 2 1 + ( I / I 0 ) 2 × 1 ( 2 π ) T 2 | Σ a | 1 2 exp - 1 2 ( v - f I 1 , I 2 ( I ) u a ) T Σ a - 1 ( v - f I 1 , I 2 ( I ) u a ) . ( 8 ) The overall model of the experiment , Eq ( 1 ) , also includes P ( I ) . To a large extent , this is controlled by the experimentalist , and details are described in Materials and Methods . The model we built assumes a stereotypical escape behavior . Is this assumption justified ? Velocities in the paused state are very small ( worms barely move ) . Thus whether the stereotypy assumption provides a good model of the data is determined largely by the stereotypy of the active worms . If the active stereotypical response template exists , then it should be possible to collapse the average velocity profiles onto a single curve by the following transformation v collapse = v a f I 1 , I 2 ( I ) . ( 9 ) Indeed , the means of different bins collapse relatively compactly , providing evidence for the existence of the stereotypy in active responses ( Fig 3 ) . We show the template velocities and the non-linear scaling function f inferred from the control , ibuprofen , and mutant in Fig 4 . Note that the three active template velocity profiles are very similar ( Fig 4A ) , but the mutant template velocity profile shows a response lag of 83 ms ( one time frame at 12 Hz ) compared to the control or ibuprofen data set . In other words , the pharmacological treatments and the mutations weakly affect the templated response , and the mutation slightly delays it . This bodes well for the assumption of the stereotypical response , definitely for ibuprofen and , to a somewhat lesser extent , for the mutant . While the existence of the stereotypical patterns and their similarity across treatments is encouraging , we still need to quantify how good the statistical models are . In the ideal case , the variance σ collapse 2 of the collapsed velocities , Eq ( 9 ) , calculated over individual trials , would be zero . However , there are a number of expected sources of variance in the velocity , such as the individual variability and the model inaccuracies . To establish how good the stereotypical model fits are , we need to disambiguate these contributions . For this , we again partition all velocity profiles into five current bins . We then write the total variance of all responses as σ total 2 = σ I 2 + σ ind 2 , ( 10 ) where σ ind 2 is the variance due to individual responses within each bin , and σ I 2 is the current-driven variance of the mean responses across the bins . Since the individuality of the worms is not accounted for in our model , σ I 2 represents the maximum potentially explainable variance in the data . The stereotypy-based model would be nearly perfect if σ I 2 were to drop to zero after the f−1 rescaling . To explore this , we plot the total variance of the active response σ total 2 ( Fig 5A ) , and the fraction of the potentially explainable variance , σ I 2 / σ total 2 ( Fig 5B ) . The latter varies from 20% to 40% of the total variance , depending on the time post-stimulus and on the treatment . In both panels , the mutant and the control dataset are nearly indistinguishable , while the ibuprofen worms show a smaller variance , and a smaller fraction of the explainable variance . This is consistent with a smaller stimulus-driven response for this analgesic-like treatment . At the same time , these figures suggest that the decrease in the maximum reverse speed in the mutant worm ( Fig 2A ) should not be attributed entirely to the reduced perceived heat stimuli . Indeed , the similarity of the variance and the explainable variance in the control and the mutant worms , which have very different mean maximum reverse velocities , suggests the existence of an additional ( explainable , non-templated ) component in the response behavior of the mutant , which is not present in the control . The explainable variance σ I 2 is further split into the variance explained by the model , σ m 2 , and the residual variance , σ res 2 , which the model fails to explain: σ I 2 = σ m 2 + σ res 2 . ( 11 ) In Fig 5C , we plot σ res 2 / σ I 2 , which is the fraction of the variance not captured by our model . Of the explainable variance , about 80% is captured by our model for the control and the ibuprofen worms in the window between 1 s and 3 . 3 s since the start of the trial , on average . This is a relatively large fraction for behavioral data , and provides an additional validation for our choice of a stereotypy-based model for representing escape behavior in these worms . Stability of the template itself between these two conditions , and the stability of the fraction of the explained variance suggest that much of the effect of ibuprofen can be attributed to the scaling of the templated response ( and also the fraction of active worms ) . In other words , ibuprofen decreases the sensed heat stimuli . In contrast , the unexplained variance for the mutant is about twice as large as that for the control , and approaches 100% at t > 2 . 7 s . This again illustrates that the templated response model is not very good for this treatment . Thus the mutations introduce changes in the behavior that are not consistent with a simple rescaling—mutations affect the fine motor behavior in addition to the sensory system per se . One of the goals of our study is to develop methods for quantitative assessment of the efficacy of pharmacological interventions to decreases sensed heat stimuli , at least in those cases where their action can be specifically interpreted as a change in thermal sensory transduction . We can use the developed statistical model for this . Specifically , taking the model derived from the control worms , we can infer the laser current from the behavior of all three different worm types . To the extent that the current inferred for the treated worms is smaller than that for the control worms at the same applied current , the heat stimuli level perceived by the treated worms is smaller . Fig 6 shows the overall structure of the inference done with the model . In the first row , we plot the conditional distribution of the inferred laser current given the actual applied current I for the three worm types , P ( Iinf|I , type ) . We again bin the trials using Ii into five bins Iμ , μ = 1 , … , 5 as before , and plot P ( I inf | I μ , type ) = ∑ i N type P control ( I inf | v i ) P type ( v i | I μ ) . ( 12 ) Here Ptype ( vi|Iμ ) is 1 if the stimulus on the i’th trial for this worm type was in the Iμ bin , and zero otherwise . Further , Pcontrol ( Iinf|vi ) is given by the full model , Eq ( 8 ) , with the parameters inferred for the control worm , and with the empirically observed velocities v in trial i for each worm type . We see that there is more probability concentrated at small Iinf for the ibuprofen and the mutant worms , suggesting a reduction in the perceived stimulus level . Similarly , in the second row in Fig 6 , we plot the expected value , I ¯ i , of the distribution of the current inferred using the control model , Pcontrol ( Iinf|vi ) , for each of the individual trials in each of the three worm types . To the extent that the values for the ibuprofen and the mutant worms are again somewhat lower than for the control worms , there is some reduction in the perceived current by this measure as well . However , these population averaged results wash out important differences in the structure of the stimulus-response relationship . To quantify these small effects more accurately , we now look at the perceived stimulus changes for individual worms in the datasets . Specifically , for each trial i in the control dataset , a trial j ( i ) with the closest value of the applied laser current is found in the ibuprofen / mutant dataset ( the mean magnitude of the current mismatch is <1 mA for both the ibuprofen and the mutant worms ) . We then use the control model to calculate the expected value of the inferred current for the jth trial in the ibuprofen / mutant datasets . This expectation is subtracted from the expectation value of the inferred current for the matched trial i for the control dataset . The difference of the expectation values , averaged over all control worms , is our measure of the reduction in the perceived stimulus level Δ I type = 1 N control ∑ i N control I i , control - I j ( i ) , type . ( 13 ) We evaluate ΔItype for different worm types and for control worms binned into the five usual current bins ( Fig 7 ) . To estimate the error of ΔItype , we bootstrap the whole analysis pipeline , see Materials and Methods . There is a statistically significant difference in stimulus perception between the ibuprofen and the control worms . The difference is most significant when the actual laser current is around 100–110 mA . This coincide with our observation ( Fig 2A ) that the most sensitive region of maximum reverse speed is around 100mA . Indeed , at smaller currents , the perceived stimulus level is small , many worms pause , and the behavior cannot be used to reliably estimate the stimulus level . At high current , the heat perception saturates , and all worms behave similarly , again reducing the ability to disambiguate the applied current level . This analysis of ibuprofen worms achieves one of our main goals . It proves our ability to reconstruct stimulus from the behavior , and shows that analgesic-like effects of pharmacological perturbations can be quantified from the behavior . At the same time , ΔImutant turns out to be insignificant ( Fig 7B ) , even though a large statistically significant difference exists between the mutant and the control behaviors ( Fig 2A ) . This failure to detect a significant perceived stimulus reduction is because the templated response model is not very good for the mutant worm; thus our analysis cannot reliably assign a mutant trajectory on a given trial to a specific stimulus level . In other words , the large error bars in Fig 7B serve as yet another check for self consistency: effects of the mutations cannot be attributed just to changes in stimulus perception . We expect our analysis to be useable for screening large numbers of chemicals for analgesic-like action . Since our approach targets one individual worm at a time , we need to estimate the number of worms needed to achieve statistical significance in such screening experiments . For this , we fix the number of control worms , arguing that these must be only analyzed once , and hence a relatively large number of them can be tested . We then focus on ibuprofen , whose action is analgesic-like in our experiments , and on the bin at 110 mA , where the worms experience the most significant perceived stimulus reduction . There are Nibuprofen , 110 = 94 worms in this bin . We randomly sample with replacements n < Nibuprofen , 110 worms from among these ibuprofen-treated worms and repeat our analysis pipeline , estimating the ΔIibuprofen ( n ) . Resampling 1000 times ( both the ibuprofen and the control datasets ) , we also estimate the variance of ΔIibuprofen ( n ) , and hence the Z score as a function of n ( Fig 7C ) . The plot here is an underestimate of the true Z score since resampling with replacements removes some stimulus values from the dataset , hence increasing the mismatch between control worms and the paired treated worms . Even with this , Z ≈ 2 is achieved at n ≈ 60 ibuprofen worms . In other words , in a typical screening experiment , one would need to test 200 or more worms to build the control model , and then at least ∼60 worms additionally for each treatment condition .
Typically a goal of sensory-response experiments is to develop a model that can predict the behavior in response to the stimuli . Here we wanted to do this in reverse . Our grand goal was to build a statistical model of the heat stimulus from careful measurements of the escape behavior of C . elegans , and to use this model to infer the changes in the perceived level of the stimulus felt by the organism due to perturbation in the sensory transduction pathway . Given this model , we could then measure changes in stimulus perception due to effects of chemicals and mutations , and use this as a basis to study the mechanism of sensory transduction in a genetically tractable organism amenable to high-throughput screens . As a representative data set , we choose to study the standard laboratory C . elegans strain N2 , N2 treated with ibuprofen , and a mutant with defects in TRPV function . Other chemicals are also studied , but only ibuprofen is selected due to two reason . First , we selected chemicals that did not affect normal motion without laser stimulus , to make it more likely that the stereotypical behavior was not affected . Second , we made sure that the worm nonetheless displayed visually different behavior after the laser stimulus compared to N2 strain . Only ibuprofen passed these tests . For the model to be successful , we had to meet a number of challenges . Since the worm could not communicate its perceived stimulus level to us directly we had to infer this level by reading the “body language” of the worm’s escape response . The difficulty with quantifying a behavioral response as a measure of perceived stimulus level is that drugs or mutations can affect locomotory behavior in addition to perturbing sensory transduction . So in an attempt to deconvolve these effects , we used the entire behavioral profile instead of making ad hoc measurements . We leveraged the fact that escape responses in C . elegans turn out to be highly stereotyped , so that the escape response can be modeled with a velocity profile template that scales non-linearly in response to an applied laser current . The success of the template in modeling the stereotyped wild-type escape response was confirmed by a functional collapse of the velocity profiles across different perceived stimulus levels . This discovery of invariance is important since it not only allowed us to effectively correlate escape behavior to the stimulus level , but it also allowed us to determine if the locomotory changes in our assay were due to changes specifically in the sensory transduction pathway or due to other general locomotory factors . By carefully accounting for the variation in our data and quantifying how much of this variation is captured by the model , we showed that the stereotypical behavior is unaffected by ibuprofen , save for changing the amplitude of the response . Thus this drug application likely reduced the perceived stimulus level in the worm . In contrast , a TRPV mutation changes locomotion in a way that is not as well captured by the template model . Thus we can be objectively critical about any inference made with this strain . The model was also useful in determining key experimental parameters for future measurements . After verification that the model works well with the native and ibuprofen treated stimulus-response data , we quantified the changes in heat perception due to ibuprofen treatment . Our modeling and experimental assessment of escape behavior identified the optimal stimulus range and required number of trials to determine statistically significant differences between the inferred current of N2 in the untreated and treated conditions . As a cautionary note , we point out that we avoid to call our heat stimulus as noxious , although the escape behavior of the worm in our experiment is similar to nociception . In the IASP definition , a nociceptive stimulus is an actually or potentially tissue-damaging event . The heat stimulus used in our experiment causes temperature increases of around 2°C in 0 . 1s , which does not have any evidence to damage the worm tissue . But previous research showed that the worm responses to small and rapid temperature increase in a nociception-like behavior [29] . Also a prolonged exposure to our heat stimulus is likely to cause damage to the worm tissue . Therefore although we are not calling our stimulus as noxious , we believe our model will be valid for nociceptive behavior . Also we point out that many noxious responses , especially in larger animals , are not stereotyped ( and hence less studied ) , and not all stereotyped behaviors are noxious responses . The stereotypy of escape in C . elegans has turned out to be helpful in disambiguating qualitatively different effects that ibuprofen and mutations have on nociception , and it is likely to be equally helpful in the future in characterizing effects of other mutations and perturbations . However , by itself the stereotypy should not be viewed as evidence for a nociceptive response , and neither should the absence of stereotypy be used as an evidence that a response is not noxious . In conclusion , we have solved what we defined as a grand problem in stimulus-response quantification and built a general model that connects stereotyped behavior to stimulus in the context of C . elegans heat-induced escape responce . With a language to describe this relationship , it is now possible to study quantitatively the effects of genetics and chemicals on this sensorimotor behavior . We believe that the utility of the model is quite general and could be applied to different model systems . However , we particularly hope that this work helps further establish C . elegans as a model for nociceptive research .
All worms were grown and maintained under standard conditions [42] , incubated with food at 20°C . Well fed worms were washed twice then gently spun down for 1 minute and the supernatant discarded by aspiration . We discovered empirically that ibuprofen affects the heat-induced escape response in our assay . For the drug application 100 μL of ibuprofen in M9 at 100 μM was added to the eppendorf tube . For the wild-type and mutant data set , M9 was used instead of the drug solution . Worms were then placed in an incubator for 30 minutes at 20°C . After that worms were poured onto a seeded agar plate and transferred to agar assay plates by a platinum wire pick . These assay plates were incubated at 20°C for 30 minutes , and then the experimental trials were done within the next 30 minutes . In total N = 201 worms for the control group , N = 441 worms for the ibuprofen group , and N = 100 worms for the mutant group ( ocr-2 ( ak47 ) osm-9 ( ky10 ) IV; ocr-1 ( ak46 ) ) group were tested . The mutant strain was obtained from the Caenorhabditis Genetics Center . The heat stimulation instrument has been described previously [29] . In summary , an infrared laser is directed to heat the head of a freely crawling worm ( ∼0 . 5mm FWHM ) on an agar plate . The laser pulse is generated with a randomly chosen laser current between 0 to 200 mA , with a duration of 0 . 1 s . The heating of the worm is nearly instantaneous , and it is directly proportional to the current , between 0 and 2°C for the current range used in our experiments . The temperature change at 60 mA current is 0 . 4°C ± 0 . 03°C , 100 mA current is 0 . 89°C ± 0 . 05°C and 150mA current is 1 . 4°C ± 0 . 2°C . Worms were stimulated only once and not reused . The movements of the worms are imaged using a standard stereomicroscope with video capture and laser control software written in LabVIEW . For each stimulus trial , a random worm is selected on the plate and its motion is sampled at 60 Hz for 15 s , and the laser is engaged 1 s after the start of the video recording . The recorded response videos were then processed with Matlab to calculate the time series of the worm centroid motion as described previously [29] . All the worms that were not stimulated near the head or were not moving forward in the beginning of the video were discarded . Numerical derivatives of the centroid times series were then taken and filtered with a custom 500 ms Gaussian filter , which was a one-sided Gaussian at the edges of the recorded time period , becoming a symmetric Gaussian away from the edges . This removed the noise due to numerical differentiation and also averaged out the spurious fluctuations in the forward velocity due to the imperfect sinusoidal shapes of the moving worm . We verified that different choices of the filter duration had little effect on the subsequent analysis pipeline . The direction of the velocity was determined by projecting the derivative of the centroid time series onto the head-to-tail vector for each worm , with the positive and negative velocity values denoting forward / backward motion , respectively . The filtered velocity profiles needed to be subsampled additionally . This was because the statistical model of the data , Eq ( 8 ) , involved covariance matrices of the active and paused velocity profiles , Σp and Σa ( note that velocity profiles are not temporally translationally invariant due to the presence of the stimulus , thus the full covariance matrix is needed , and not a simpler correlation function ) . To have a full rank covariance matrix , the number of trials must be larger than the number of time points . Alternatively , regularization is needed for covariance calculation . The autocorrelation function for all three worm types showed a natural correlation time scale of ≳ 0 . 2 s , whether the data was pre-filtered or not . Thus subsampling at a frequency > 5 Hz would not result in data loss . Therefore , instead of an arbitrary regularization , we chose to subsample the data at 12 Hz , leaving us with 37 data points to characterize the first 3 s of the worm velocity trace after the stimulus application , 1 ≤ t ≤ 4 s since the start of the trial . Eq ( 8 ) additionally needs knowledge of T , the number of effectively independent velocity measurements in the profile . This is obtained by dividing the duration of the profile by the velocity correlation time . An uncertainty of such procedure has a minimal effect on the model of the experiment since it simply changes log likelihoods of models by the same factor , not changing which model has the maximum likelihood . We then considered limiting the duration of the velocity profile used in model building: if velocities at certain time points do not contribute to the identification of I , they should be removed to decrease the number of unknowns in the model that must be determined from data ( values of the templates at different time points ) . The first candidate for removal was the period of about 10 frames ( 0 . 16 s ) after the laser stimulation since worms do not respond to the stimulus so quickly . However , removal of this time period had a negligible effect on the model performance , and we chose to leave it intact . In contrast , starting from 3 . 3 s ( 2 . 3 s after the stimulus ) the fraction of explainable variance drops to nearly zero ( Fig 5 ) since many worms already had turned by this time and resumed forward motion . Therefore , we eventually settled on the time in the 1 . 0…3 . 3 s range for building the model . Whenever needed , we estimated the variance of our predictions by bootstrapping the whole analysis pipeline [43] . For this , we created 1000 different datasets by resampling with replacement from the original control dataset and the mutant / ibuprofen datasets . Control statistical models ( the scaling function f and the velocity templates ) were estimated for each resampled control dataset . Standard deviations of these models were used as estimates of error bars in Fig 4 . For Fig 7 , we additionally needed to form the closest control / treatment worm pairs . These were formed between the resampled data sets for all worm types as well . Standard deviations of ΔItype evaluated by such resampling were then plotted in Fig 7 and used to estimate Z scores . Note that such resampling produces control / treatment paired worms that have slightly larger current differences than in the actual data; this leads to our error bars being overestimates . Model in Eq ( 1 ) requires knowing P ( I ) . In principle , this is controlled by the experimentalist , and thus should be known . In our experiments , P ( I ) was set to be uniform . However , as described above , some of the worms were discarded in preprocessing , and this resulted in non-uniformly distributed current samples . To account for this , we used the empirical Pemp ( I ) in our analysis instead of P ( I ) = const . In turn , Pemp ( I ) was inferred using a well-established algorithm for estimation of one-dimensional continuous probability distributions from data [44] . All of this analysis was implemented using Matlab , and the code is available for download from a public GitHub repository https://github . com/EmoryUniversityTheoreticalBiophysics/C . -elegans . The template for the paused state up is calculated by taking the average of all paused velocity profiles for each of the three worm datasets . The covariance Σp is then the covariance of the set of the paused velocity profiles . For active worms , we start with fixed putative parameter values I 1 and I 2 . We then calculate the active template ua and the covariance matrix ∑a by maximizing the likelihood in Eq ( 7 ) ∂ ∑ i N type , a log P ( v i | a , I i ) ∂ u a ∝ ∑ i N type , a v i f I 1 , I 2 ( I i ) - u a f I 1 I 2 2 ( I i ) = 0 , ( 14 ) ∂ ∑ i N type , a log P ( v i | a , I i ) ∂ Σ a ∝ ∑ i N type , a v i - u a f I 1 , I 2 ( I i ) 2 - ( Σ a ) - 1 = 0 , ( 15 ) where Ntype , a is the number of active worms of the analyzed type . This gives: u a ( I 1 , I 2 ) = ∑ i = 1 N type , a v i f I 1 , I 2 ( I i ) ∑ i = 1 N type , a f I 1 , I 2 2 ( I i ) , ( 16 ) Σ a = ∑ i N type , a v i - u a f I 1 , I 2 ( I i ) 2 . ( 17 ) Having thus estimated ua and Σa at fixed parameter values I 1 , I 2 , we maximize ∏i P ( vi|a , Ii ) over the parameters using standard optimization algorithms provided by MATLAB . We perform optimization from ten different initial conditions to increase the possibility that we find a global , rather than the local maximum . | A doctor assesses pain by asking her patient to “rate your pain on the scale of 1 to 10 . ” She may then prescribe some drugs and later ask the question again to see if they worked . New drugs are often developed using animal models , but we cannot ask an animal , especially a small invertebrate animal , to rate , similarly , the strength of its perceived noxious stimulus . In this paper , we successfully develop computational tools that read the “body language” of a roundworm C . elegans to estimate the strength of the heat stimulus that it experiences . Unlike previous attempts that have focused on ad hoc selected components of the overall behavior , our approach is based on quantifying the complete time series of the escape behavior , which we show to be captured by a behavioral “template” that scales in response to the stimulus strength . The existence of this template allows us to solve one of the hard questions in pain research: disambiguating analgesic-like effects of drugs or genetic perturbations from their other effects on animal behavior . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"velocity",
"invertebrates",
"somatosensory",
"system",
"medicine",
"and",
"health",
"sciences",
"classical",
"mechanics",
"engineering",
"and",
"technology",
"caenorhabditis",
"nociception",
"lasers",
"social",
"sciences",
"random",
"variables",
"neuroscience",
"animals",
"covariance",
"animal",
"models",
"caenorhabditis",
"elegans",
"model",
"organisms",
"mathematics",
"animal",
"behavior",
"statistics",
"(mathematics)",
"experimental",
"organism",
"systems",
"optical",
"equipment",
"zoology",
"research",
"and",
"analysis",
"methods",
"sensory",
"physiology",
"behavior",
"probability",
"theory",
"physics",
"statistical",
"models",
"psychology",
"pain",
"sensation",
"equipment",
"physiology",
"nematoda",
"biology",
"and",
"life",
"sciences",
"sensory",
"systems",
"sensory",
"perception",
"physical",
"sciences",
"organisms",
"motion"
]
| 2016 | Stereotypical Escape Behavior in Caenorhabditis elegans Allows Quantification of Effective Heat Stimulus Level |
The neurosteroid dehydroepiandrosterone ( DHEA ) , produced by neurons and glia , affects multiple processes in the brain , including neuronal survival and neurogenesis during development and in aging . We provide evidence that DHEA interacts with pro-survival TrkA and pro-death p75NTR membrane receptors of neurotrophin nerve growth factor ( NGF ) , acting as a neurotrophic factor: ( 1 ) the anti-apoptotic effects of DHEA were reversed by siRNA against TrkA or by a specific TrkA inhibitor; ( 2 ) [3H]-DHEA binding assays showed that it bound to membranes isolated from HEK293 cells transfected with the cDNAs of TrkA and p75NTR receptors ( KD: 7 . 4±1 . 75 nM and 5 . 6±0 . 55 nM , respectively ) ; ( 3 ) immobilized DHEA pulled down recombinant and naturally expressed TrkA and p75NTR receptors; ( 4 ) DHEA induced TrkA phosphorylation and NGF receptor-mediated signaling; Shc , Akt , and ERK1/2 kinases down-stream to TrkA receptors and TRAF6 , RIP2 , and RhoGDI interactors of p75NTR receptors; and ( 5 ) DHEA rescued from apoptosis TrkA receptor positive sensory neurons of dorsal root ganglia in NGF null embryos and compensated NGF in rescuing from apoptosis NGF receptor positive sympathetic neurons of embryonic superior cervical ganglia . Phylogenetic findings on the evolution of neurotrophins , their receptors , and CYP17 , the enzyme responsible for DHEA biosynthesis , combined with our data support the hypothesis that DHEA served as a phylogenetically ancient neurotrophic factor .
Dehydroepiandrosterone ( DHEA ) is a steroid , produced in adrenals , in neurons and in glia [1] . The physiological role of brain DHEA appears to be local , i . e . paracrine , while that produced from adrenals , which represents the almost exclusive source of circulating DHEA , is systemic . The precipitous decline of both brain and circulating DHEA with advancing age has been associated with aging-related neurodegenerative diseases [1] , [2] . It is experimentally supported that DHEA protects neurons against noxious conditions [3]–[6] . DHEA exerts its multiple pro-survival effects either directly modulating at micromolar concentrations γ-aminobutiric acid type A ( GABAA ) , N-methyl-D-aspartate ( NMDA ) , or sigma1 receptors , or following its conversion to estrogens and androgens . We have recently shown that nanomolar concentrations of DHEA protect sympathoadrenal PC12 cells from apoptosis [7] . PC12 cells do not express functional GABAA or NMDA receptors and cannot metabolize DHEA to estrogens and androgens [8] . The anti-apoptotic effect of DHEA in PC12 cells is mediated by high affinity ( KD at nanomolar levels ) specific membrane binding sites [9] . Activation of DHEA membrane binding sites results in an acute , transient , and sequential phosphorylation of the pro-survival MEK/ERK kinases , which , in turn , activate transcription factors CREB and NFκB , which afford the transcriptional control of anti-apoptotic Bcl-2 proteins . In parallel , activation of DHEA membrane binding sites induces the phosphorylation of PI3K/Akt kinases , leading to phosphorylation/deactivation of the pro-apoptotic Bad protein and protection of PC12 cells from apoptosis [10] . In fact , the anti-apoptotic pathways in sympathoadrenal cells initiated by DHEA at the membrane level strikingly resemble those sensitive to neurotrophin nerve growth factor ( NGF ) . NGF promotes survival and rescues from apoptosis neural crest–derived sympathetic neurons ( including their related sympathoadrenal cells ) and sensory neurons involved in noniception . NGF binds with high affinity ( KD: 0 . 01 nM ) to transmembrane tyrosine kinase TrkA receptor and with lower affinity ( KD: 1 . 0 nM ) to p75NTR receptor , a membrane protein belonging to the TNF receptor superfamily [11] . In the presence of TrkA receptors , p75NTR participates in the formation of high affinity binding sites and enhances NGF responsiveness , leading to cell survival signals . In the absence of TrkA , p75NTR generates cell death signals . Indeed , docking of TrkA by NGF initiates receptor dimerization and phosphorylation of cytoplasmic tyrosine residues 490 and 785 on the receptor . Phosphotyrosine-490 interacts with Shc and other adaptor proteins resulting in activation of PI3K/Akt and MEK/ERK signaling kinase pathways [11] . These signals lead to the activation of prosurvival transcription factors CREB and NFκB , the subsequent production of anti-apoptotic Bcl-2 proteins , and prevention of apoptotic cell death of sympathetic neurons and sympathoadrenal cells , including PC12 cells [12] . Intrigued by the similarities in the prosurvival membrane signaling of DHEA and NGF , we set out to examine in the present study whether the anti-apoptotic effects of DHEA are mediated by NGF receptors . To address this issue we employed a multifaceted approach , designing an array of specific experiments: we used RNA interference ( RNAi ) to define the involvement of TrkA and p75NTR receptors in the anti-apoptotic action of DHEA; we assessed membrane binding of DHEA in HEK293 cells transfected with the TrkA and p75NTR plasmid cDNAs , using binding assays , confocal laser microscopy , and flow cytometry; to investigate the potential direct physical interaction of DHEA with NGF receptors , we tested the ability of immobilized DHEA to pull-down recombinant or naturally expressed TrkA and p75NTR receptors; finally , we examined the ability of DHEA to rescue from apoptosis NGF receptor sensitive dorsal root ganglia sensory neurons of NGF null mice and NGF deprived rat superior cervical ganglia sympathetic neurons in culture [13] . We provide evidence that DHEA directly binds to NGF receptors to protect neuronal cells against apoptosis , acting as a neurotrophic factor .
To test the involvement of NGF receptors in the anti-apoptotic effect of DHEA in serum deprived PC12 cells we have used a combination of three different sequences of siRNAs for TrkA and two different shRNAs for p75NTR transcripts [14] . The effectiveness of si/shRNAs was shown by the remarkable decrease of TrkA and p75NTR protein levels in PC12 cells , observed by immunoblotting analysis , using GAPDH as reference standard ( Figure 1B ) . Scrambled siRNAs were ineffective in decreasing TrkA and p75NTR protein levels and did not significantly alter the effect of DHEA ( unpublished data ) . FACS analysis of apoptotic cells ( stained with Annexin V ) has shown that DHEA and membrane impermeable DHEA-BSA conjugate at 100 nM diminished the number of apoptotic cells in serum deprived PC12 cell cultures from 53 . 5%±17 . 6% increase of apoptosis in serum free condition ( control ) to 6%±1 . 4% and 13%±5 . 2% , respectively ( n = 8 , p<0 . 01 versus control ) ( Figure 1A ) . Decreased TrkA expression in serum deprived PC12 cells with siRNAs resulted in the almost complete reversal of the anti-apoptotic effects of NGF and DHEA or DHEA-BSA membrane-impermeable conjugate ( Figure 1A ) . Co-transfection of serum deprived PC12 cells with the si/shRNAs for TrkA and p75NTR receptors did not modify the effect of the TrkA deletion alone . Furthermore , transfection of serum deprived PC12 cells with shRNAs against p75NTR receptor alone did not significantly alter the anti-apoptotic effects of NGF and DHEA , suggesting that their anti-apoptotic effects are primarily afforded by TrkA receptors . Transfection of serum deprived PC12 cells with the siRNAs against the TrkA transcript fully annulled the ability of DHEA to maintain elevated levels of anti-apoptotic Bcl-2 protein ( Figure 1B ) . Again , transfection with the shRNA against p75NTR receptor alone did not significantly affect Bcl-2 induction by DHEA , further supporting the hypothesis that TrkA is the main mediator of the anti-apoptotic effect of DHEA in this system . It appears that the ratio of TrkA and p75NTR receptors determines the effect of DHEA or NGF on cell apoptosis and survival . Indeed , both NGF and DHEA induced apoptosis of nnr5 cells , a clone of PC12 cell line , known to express only pro-death p75NTR receptors ( Figure 1C ) , confirming the pro-apoptotic function of this receptor . Blockade of p75NTR expression by shRNA almost completely reversed the pro-apoptotic effect of both agents . The anti-apoptotic effect of NGF and DHEA was remarkably restored after transfection of nnr5 cells with the TrkA cDNA , the efficacy of reversal being proportionally dependent on the amount of transfected TrkA cDNA ( Figure 1C ) . DHEA was also controlling the response of NGF receptor-positive cells , by regulating TrkA and p75NTR receptor levels , mimicking NGF . Serum deprived PC12 cells were exposed to 100 nM of DHEA or 100 ng/ml of NGF for 12 , 24 , and 48 h; TrkA and p75NTR protein levels were measured in cell lysates with immunoblotting , using specific antibodies against TrkA and p75NTR proteins , and were normalized against GAPDH . Both NGF and DHEA significantly increased pro-survival TrkA receptor levels in the time frame studied , i . e . from 12 to 48 h ( n = 5 , p<0 . 01 ) ( Figure S1 ) . Furthermore , DHEA and NGF significantly decreased p75NTR receptor levels between 24 and 48 h of exposure ( n = 5 , p<0 . 01 ) . We have also tested the anti-apoptotic effects of DHEA in neural crest deriving superior cervical ganglia ( SCG ) , a classical NGF/TrkA sensitive mammalian neuronal tissue , containing primarily one class of neurons , principal sympathetic neurons . Indeed , NGF and TrkA receptors are absolutely required for SCG sympathetic neuron survival during late embryogenesis and early postnatal development [13] , [15] . TrkC receptors are barely detectable after E15 . 5 , and no significant TrkB receptors are present in the SCG at any developmental stage [16] . Dispersed rat sympathetic SCG neurons at P1 were isolated and cultured for at least 7 d in the presence of 100 ng/ml NGF before the experiments are performed , in order to obtain an enriched , quasi-homogenous ( 95% ) neuronal cell culture . Enriched SCGs were then incubated in the presence of 100 ng/ml NGF or in the same medium as above but lacking NGF and containing a polyclonal rabbit anti-NGF-neutralizing antiserum in the absence or the presence of 100 nM DHEA . Withdrawal of NGF strongly increased the number of apoptotic sympathetic neurons stained with Annexin V , while DHEA effectively compensated for NGF by decreasing the levels of apoptotic neurons . This effect was blocked by a specific TrkA inhibitor , thus suggesting the involvement of TrkA receptors as the main mediator of the anti-apoptotic action of DHEA ( Figure 2 ) . We have previously shown the presence of specific DHEA binding sites to membranes isolated from PC12 , primary human sympathoadrenal , and primary rat hippocampal cells , with KD at the nanomolar level [9] . The presence of DHEA-specific membrane binding sites on PC12 cells has been confirmed by flow cytometry and confocal laser microscopy of cells stained with the membrane impermeable DHEA-BSA-FITC conjugate . In contrast to estrogens , glucocorticoids and androgens displaced [3H]DHEA from its membrane binding sites , acting as pure antagonists by blocking the anti-apoptotic effect of DHEA in serum deprived PC12 cells [9] . In the present study , we repeated this series of experiments using membranes isolated from HEK293 cells transfected with the plasmid cDNAs of TrkA or p75NTR receptors . HEK293 cells ( not expressing TrkA or p75NTR ) were transfected with an empty vector ( control ) or a specific TrkA or p75NTR vector; transfection efficiency was assessed by Western blot ( Figure 3A and C , F inserts ) , confocal laser microscopy , and flow cytometry ( Figure 3B , D ) . Saturation binding experiments have shown that [3H]-DHEA bound to membranes isolated from HEK293 cells , transfected with the cDNAs of TrkA or p75NTR receptors . Membranes isolated from HEK293 cells transfected with the empty vector showed no specific binding . The KD values calculated after Scatchard analysis of saturation curves were , for incubation of membranes at 25°C for 30 min , 7 . 4±1 . 75 nM and 5 . 6±0 . 55 nM for TrkA or p75NTR , respectively ( n = 3 ) ( Figure 3A , C ) , and for overnight incubation of membranes at 4°C , 7 . 8±3 . 1 nM and 5 . 9±1 . 7 nM for TrkA or p75NTR , respectively ( n = 3 ) ( Figure S2 ) . DHEA was previously shown to bind with low affinity ( KD: 2 µM ) to androgen receptors ( AR ) [17] . We have thus tested the hypothesis that specific binding of DHEA to membranes of HEK293 cells transfected with the TrkA and p75NTR cDNAs might be due to the presence of AR receptors , induced by the transfection with NGF receptors . However , RT-PCR analysis showed no detectable levels of androgen receptors mRNA in RNA preparations isolated from naïve and TrkA or p75NTR transfected HEK23 cells ( Figure S3 ) . Transfection of PC12 cells , endogenously expressing NGF receptors , with shRNAs against both TrkA and p75NTR receptors resulted in a complete loss of [3H]-DHEA specific membrane binding ( Figure 3E , F ) . To rule out the possibility that the loss of specific binding might be due to the transfection process , we tested binding of [3H]-DHEA to membranes isolated from PC12 cells transfected with siRNA against GAPDH . Saturation binding and Scatchard analysis have shown that [3H]-DHEA bound to membranes from PC12-siRNA GAPDH cells with a KD = 1 . 068±0 . 43 nM ( Figure 3E ) . The selectivity of DHEA binding to HEK293TrkA and HEK293p75NTR cell membranes was examined by performing heterologous [3H]-DHEA displacement experiments using a number of non-labeled steroids or NGF . Binding of [3H]-DHEA to membranes isolated from both HEK293TrkA and HEK293p75NTR cells was effectively displaced by NGF ( IC50: 0 . 8±0 . 2 and 1 . 19±0 . 45 nM , respectively ) ( Figure S4 ) . NGF was also effective in displacing [3H]-DHEA binding on membranes isolated from PC12 cells ( IC50: 0 . 92±0 . 32 nM , unpublished data ) . Estradiol failed to displace [3H]-DHEA from its binding to membranes from HEK293TrkA and HEK293p75NTR cells at concentrations ranging from 0 . 1 to 1000 nM . In contrast , displacement of [3H]-DHEA binding to membranes from both HEK293TrkA and HEK293p75NTR cells was shown by sulfated ester of DHEA , DHEAS ( IC50: 6 . 1±1 . 1 and 8 . 1±1 . 2 nM , respectively , n = 3 ) , and testosterone ( Testo ) ( IC50: 5 . 3±2 . 1 and 7 . 4±3 . 2 nM , respectively ) . Glucocorticoid dexamethasone ( DEX ) effectively competed [3H]-DHEA binding to membranes from HEK293TrkA ( IC50: 9 . 5±4 . 6 nM ) but was ineffective in displacing DHEA binding to membranes from HEK293p75NTR cells . Homologous [125I]-NGF displacement experiments with unlabeled NGF confirmed the presence of specific NGF binding on membranes from both HEK293TrkA and HEK293p75NTR cells with IC50 0 . 3±0 . 09 and 1 . 7±0 . 38 nM , respectively . It is of note that in contrast to unlabeled NGF , DHEA was unable to displace binding of [125I]-NGF to membranes isolated from HEK293TrkA and HEK293p75NTR transfectants ( unpublished data ) . Incubation of PC12 cells with the membrane impermeable , fluorescent DHEA-BSA-fluorescein conjugate results in a specific spot-like membrane fluorescent staining [9] . In the present study , we have tested the ability of DHEA-BSA-FITC conjugate to stain HEK293TrkA and HEK293p75NTR transfectants . Fluorescence microscopy analysis revealed that DHEA-BSA-FITC clearly stained the membranes of HEK293TrkA and HEK293p75NTR cells ( Figure 3B , D ) . No such staining was found in non-transfected HEK293 cells ( unpublished data ) or in HEK293 cells transfected with the vectors empty of TrkA and p75NTR cDNAs ( Figure 3B , D ) . Furthermore , BSA-FITC conjugate was ineffective in staining both transfectants ( unpublished data ) . We have further confirmed the presence of membrane DHEA-BSA-FITC staining of HEK293TrkA and HEK293p75NTR cells with flow cytometry ( FACS ) analysis ( Figure 3B , D ) . Specific staining was noted in both transfectants . No such staining was seen in non-transfected HEK293 cells ( unpublished data ) or in HEK293 cells transfected with the empty vectors ( Figure 3B , D ) . In both fluorescence microscopy and FACS experiments membrane staining of TrkA or p75NTR proteins in HEK293TrkA and HEK293p75NTR cells was also shown using specific antibodies for each protein ( Figure 3B , D ) . Our binding assays with radiolabeled DHEA suggest that DHEA physically interacts with NGF receptors . To test this hypothesis we covalently linked DHEA-7-O- ( carboxymethyl ) oxime ( DHEA-7-CMO ) to polyethylene glycol amino resin ( NovaPEG amino resin ) and tested the ability of immobilized DHEA to pull down TrkA and p75NTR proteins . Precipitation experiments and Western blot analysis of precipitates with specific antibodies against TrkA and p75NTR proteins ( Figure 4A ) showed that immobilized DHEA effectively precipitated recombinant TrkA and p75NTR proteins , while pre-incubation of the recombinant proteins with DHEA or NGF in excess abolished the ability of DHEA-PEG to pull down both receptors . Similar results were obtained when cell extracts isolated from HEK293 cells transfected with TrkA and p75NTR cDNAs , PC12 cells , and whole rat brain were treated with immobilized DHEA ( Figure 4B , panels marked with A ) . No precipitation of TrkA and p75NTR proteins was shown with polymer-supported DHEA-7-CMO incubated with cell extracts from untransfected HEK293 cells or HEK293 cells transfected with the empty vectors . A control experiment was performed with NovaPeg amino resin ( no DHEA-7-CMO present ) , which was found ineffective in precipitating TrkA and p75NTR proteins ( Figure 4 ) . The presence of TrkA and p75NTR receptors in HEK293TrkA and HEK293p75NTR transfectants and in PC12 and fresh rat brain was confirmed with Western blot analysis using specific antibodies against TrkA and p75NTR proteins and GAPDH as reference standard ( Figure 4 , panels marked with B ) . Previous findings have shown that NGF controls the responsiveness of sensitive cells through induction of TrkA phosphorylation and regulation of the levels of each one's receptors [18] . We compared the ability of NGF and DHEA to induce phosphorylation of TrkA in HEK293 cells transfected with the cDNAs of TrkA receptors . HEK293TrkA transfectants were exposed for 10 and 20 min to 100 nM of DHEA or 100 ng/ml of NGF , and cell lysates were immunoprecipitated with anti-tyrosine antibodies and analyzed by Western blotting , using specific antibodies against TrkA receptors . Both NGF and DHEA strongly increased phosphorylation of TrkA as early as 10 min , an effect which was also maintained at 20 min ( Figure 5A ) . We also tested the effects of DHEA and NGF in PC12 cells , endogenously expressing TrkA receptors . Naive or siRNATrkA transfected PC12 cells were incubated for 10 min with DHEA or NGF , and cell lysates were analyzed with Western blotting , using specific antibodies against Tyr490-phosphorylated TrkA and total TrkA . Both NGF and DHEA strongly induced the phosphorylation of TrkA in naive PC12 cells , effects which were diminished in siRNATrkA transfected PC12 cells ( Figure 5A ) . The stimulatory effect of DHEA on TrkA phosphorylation might be due to an increase of NGF production . To test this hypothesis , we measured with ELISA the levels of NGF in culture media of HEK293 and PC12 cells exposed for 5 to 30 min to 100 nM of DHEA . NGF levels in culture media of control and DHEA-treated HEK293 and PC12 cells were undetectable , indicating that DHEA-induced TrkA phosphorylation was independent of NGF production . DHEAS mimicked the effect of DHEA and rapidly induced ( within 10 min ) the phosphorylation of TrkA receptors in HEK293 transfected with the TrkA cDNA expression vector ( Figure S5 ) . On the other hand , testosterone , while capable of displacing DHEA binding to TrkA receptors , was unable to increase phosphorylation of TrkA in the same system ( Figure S5 ) . We compared the ability of NGF and DHEA to induce phosphorylation of TrkA-sensitive Shc , ERK1/2 , and Akt kinases . Serum deprived naive or siRNATrkA transfected PC12 cells were incubated for 10 min with 100 nM DHEA or 100 ng/ml NGF and cell lysates were analyzed with Western blotting , using specific antibodies against the phosphorylated and total forms of kinases mentioned above . Both DHEA and NGF strongly increased phosphorylation of Shc , ERK1/2 , and Akt kinases in naive PC12 cells , effects which were almost absent in siRNATrkA transfected PC12 cells , suggesting that both DHEA and NGF induce Shc , ERK1/2 , and Akt phosphorylation via TrkA receptors ( Figure 5A ) . The effectiveness of DHEA to promote the interaction of p75NTR receptors with its effector proteins TRAF6 , RIP2 , and RhoGDI was also assessed . It is well established that NGF induces the association of p75NTR receptors with TNF receptor-associated factor 6 ( TRAF6 ) , thus facilitating nuclear translocation of transcription factor NFκB [19] . Furthermore , p75NTR receptors associate with receptor-interacting protein 2 ( RIP2 ) in a NGF-dependent manner [20] . RIP2 binds to the death domain of p75NTR via its caspase recruitment domain ( CARD ) , conferring nuclear translocation of NFκB . Finally , naive p75NTR interacts with RhoGDP dissociation inhibitor ( RhoGDI ) , activating small GTPase RhoA [21] . In that case , NGF binding abolishes the interaction of p75NTR receptors with RhoGDI , thus inactivating RhoA . We co-transfected HEK293 cells with the plasmid cDNAs of p75NTR and of each one of the effectors TRAF6 , RIP2 , or RhoGDI , tagged with the flag ( TRAF6 ) or myc ( RIP2 , RhoGDI ) epitopes . Transfectants were exposed to 100 nM DHEA or 100 ng/ml NGF , and lysates were immunoprecipitated with antibodies against flag or myc , followed by immunoblotting with p75NTR specific antibodies . Both DHEA and NGF efficiently induced the association of p75NTR with effectors TRAF6 and RIP2 , while facilitating the dissociation of RhoGDI from p75NTR receptors ( Figure 5B ) . NGF null mice have fewer sensory neurons in dorsal root ganglia ( DRG ) due to their apoptotic loss [13] . Heterozygous mice for the NGF deletion were interbred to obtain mice homozygous for the NGF gene disruption . The mothers were treated daily with an intraperitoneal injection of DHEA ( 2 mg ) or vehicle ( 4 . 5% ethanol in 0 . 9% saline ) . Embryos were collected at E14 day of pregnancy and sections were stained for Caspase 3 and Fluoro jade C , markers of apoptotic and degenerative neurons , respectively . ngf−/− embryos at E14 showed a dramatic increase in the number of Fluoro Jade C and Caspase 3 positive neurons in the DRG compared to the ngf+/− embryos ( Figure 6A , B ) . DHEA treatment significantly reduced Fluoro Jade C and Caspase 3 positive neurons in the DRG to levels of ngf+/− embryos . Furthermore , TrkA and TUNEL double staining of DRGs has shown that in ngf+/− embryos , numbers of TUNEL-positive apoptotic neurons were minimal , while TrkA positive staining was present in a large number of neuronal cell bodies of the DRG and their collaterals were extended within the marginal zone to the most dorsomedial region of the spinal cord . On the contrary , in DRG of ngf−/− embryos levels of TUNEL-positive apoptotic neurons were dramatically increased , while TrkA neuronal staining was considerably decreased and DRG collaterals of the dorsal funiculus were restricted in the dorsal root entry zone ( Figure 6C ) . DHEA treatment resulted in a significant increase of TrkA positive staining and the extension of TrkA staining within the marginal zone to the most dorsomedial region of the spinal cord similarly to the ngf+/− embryos ( Figure 6D ) , while staining of TUNEL-positive apoptotic neurons was decreased to levels shown in ngf+/− embryos .
DHEA exerts multiple actions in the central and peripheral nervous system; however , no specific receptor has been reported to date for this neurosteroid . Most of its actions in the nervous tissue were shown to be mediated via modulation , at micromolar concentrations , of membrane neurotransmitter receptors , such as NMDA , GABAA , and sigma1 receptors . DHEA may also influence brain function by direct binding , also at micromolar concentrations , to dendritic brain microtubule-associated protein MAP2C [22] . In the present study we provide evidence that DHEA binds to NGF receptors . This is the first report showing a direct binding of a steroid to neurotrophin receptors . Saturation experiments and Scatchard analysis of [3H]-DHEA binding to membranes isolated from HEK293 cells transfected with the cDNAs of TrkA and p75NTR receptors showed that DHEA binds to both membranes ( 7 . 4±1 . 75 nM and 5 . 6±0 . 55 nM for TrkA or p75NTR , respectively ) . Non-radioactive NGF effectively displaced [3H]-DHEA binding to both membrane preparations , with IC50: 0 . 8±0 . 2 and 1 . 19±0 . 45 nM , respectively . Furthermore , pull-down experiments using DHEA covalently immobilized on NovaPEG amino resin suggest that DHEA binds directly to TrkA and p75NTR proteins . Indeed , polymer-supported DHEA-7-CMO effectively pulled down recombinant TrkA and p75NTR proteins and precipitated both proteins from extracts prepared from cells expressing both receptors ( HEK293TrkA , HEK293p75NTR , and PC12 cells and freshly isolated rat brain ) . Interestingly , DHEA was unable to effectively displace binding of [125I]-NGF on membranes isolated from HEK293TrkA and HEK293p75NTR transfectants . It is possible that dissociation of binding of peptidic NGF from its receptors lasts longer due to the multiple sites of interaction within the binding cleft of this large peptidic molecule compared to smaller in volume steroid . Another explanation might be that NGF and DHEA bind to different domains of NGF receptors , the NGF domain being non-recognizable by DHEA . It is of note that antidepressant amitryptiline cannot chase NGF from TrkA receptors because it binds to a different domain on TrkA protein compared to NGF . Indeed , other small molecules , like antidepressant amitriptyline and gamboge's natural extract gambogic amide , bind in the extracellular and the cytoplasmic juxtamembrane domains of TrkA receptor , although with much lower affinity compared to DHEA ( Kd 3 µM and 75 nM , respectively ) [23] , [24] . The domains of TrkA and p75NTR proteins involved in DHEA binding were not defined in the present study . Mutagenesis assays combined with NMR spectroscopy are planned to map the domains of both receptors related to DHEA binding . Our findings suggest that binding of DHEA to NGF receptors is functional , mediating its anti-apoptotic effects . Indeed , blocking of TrkA expression by RNAi almost completely reversed the ability of DHEA to protect PC12 cells from serum deprivation-induced apoptosis and to maintain elevated levels of the anti-apoptotic Bcl-2 protein . Additionally , in dispersed primary sympathetic neurons in culture , DHEA effectively compensated NGF deprivation by decreasing the levels of apoptotic neurons , an effect which was reversed by a specific TrkA inhibitor , further supporting the involvement of TrkA receptors in the anti-apoptotic action of DHEA . Finally , DHEA effectively rescued from apoptosis TrkA-positive dorsal root ganglia sensory neurons of NGF null mouse embryos . It appears that the decision between survival and death among DHEA-responsive cells is determined by the ratio of TrkA and p75NTR receptors . In fact , DHEA and NGF induced apoptosis of nnr5 cells , a clone of PC12 cells expressing only pro-death p75NTR receptors . The pro-death effects of both agents were completely blocked by p75NTR shRNA and were remarkably restored after transfection of nnr5 cells with the TrkA cDNA . It is of note that during brain development the ratio of TrkA to p75NTR varies tempospatially [25] . Thus , the ability of DHEA to act in a positive or negative manner on neuronal cell survival may depend upon the levels of the two receptors during different stages of neuronal development . Binding of DHEA on both TrkA and p75NTR receptors was effectively competed by sulfated DHEA , DHEAS ( IC50: 6 . 1±1 . 1 and 8 . 1±1 . 2 nM , respectively ) , suggesting that DHEAS may also bind to NGF receptors . Testosterone displaced DHEA binding to TrkA and p75NTR ( IC50: 5 . 3±2 . 1 and 7 . 4±3 . 2 nM , respectively ) , while synthetic glucocorticoid dexamethasone displaced DHEA binding only to pro-survival TrkA receptors ( IC50: 9 . 5±4 . 6 nM ) . In a previous study we had shown that both steroids effectively displaced DHEA from its specific membrane binding sites of sympathoadrenal cells , acting as DHEA antagonists by blocking its anti-apoptotic effect and the induction of anti-apoptotic Bcl-2 proteins [9] . Our findings suggest that testosterone and glucocorticoids may act as neurotoxic factors by antagonizing endogenous DHEA and NGF for their binding to NGF receptors , explaining previously published data . Indeed , testosterone was shown to increase NMDA and GABAA-mediated neurotoxicity [26] , [27] . Our findings suggest that testosterone may act as a neurotoxic factor by also antagonizing the neuroprotective effects of endogenous DHEA . Furthermore , glucocorticoids show a bimodal effect on hippocampal neurons causing acutely an increase in performance of spatial memory tasks , while chronic exposure has been associated with decreased cognitive performance and neuronal atrophy [28] . Acute administration of glucocorticoids results in a glucocorticoid receptor-mediated phosphorylation and activation of hippocampal TrkB receptors , exerting trophic effects on dentate gyrus hippocampal neurons [29] , via an increase in the sensitivity of hippocampal cells to neurotrophin BDNF , the endogenous TrkB ligand known to promote memory and learning [30] . However , overexposure to glucocorticoids during prolonged periods of stress is detrimental to central nervous system neurons , especially in aged animals , affecting mainly the hippocampus . It is possible that part of neurotoxic effects of glucocorticoids may be due to their antagonistic effect on the neuroprotective effect of endogenous DHEA and NGF , via TrkA receptor antagonism . The decline of brain DHEA and NGF levels during aging and in Alzheimer's disease [28] might exacerbate this phenomenon , rendering neurons more vulnerable to glucocorticoid toxicity . Indeed , glucocorticoid neurotoxicity becomes more pronounced in aged subjects since cortisol levels in the cerebrospinal fluid increase in the course of normal aging , as well as in relatively early stages of Alzheimer's disease [28] . A number of neurodegenerative conditions are associated with lower production or action of both DHEA and NGF [31] , [32] . Animal studies suggest that NGF may reverse , or slow down the progression of Alzheimer's related cholinergic basal forebrain atrophy [32] . Furthermore , the neurotrophic effects of NGF in experimental animal models of neurodegenerative conditions , like MPTP ( Parkinson's disease ) , experimental allergic encephalomyelitis ( multiple sclerosis ) , or ischemic retina degeneration mice [33]–[35] support its potential as a promising neuroprotective agent . However , the use of NGF in the treatment of these conditions is limited , because of its poor brain blood barrier permeability . It is of interest that DHEA also exerts neuroprotective properties in some of these animal models [7] , [36] . These findings suggest that synthetic DHEA analogs , deprived of endocrine effects , may represent a new class of brain blood barrier permeable NGF receptor agonists with neuroprotective properties . We have recently reported the synthesis of 17-spiro-analogs of DHEA , with strong anti-apoptotic and neuroprotective properties , deprived of endocrine effects [37] , which are now being tested for their ability to bind and activate NGF receptors . We have previously defined the pro-survival signaling pathways that are initiated by DHEA at the membrane level [3] . These pathways include MEK1/2/ERK1/2 and PI3K/Akt pro-survival kinases . We now provide experimental evidence that DHEA activates these kinases via TrkA receptors . Down-regulation of TrkA receptors using siRNAs resulted in an almost complete reversal of the ability of DHEA to increase the phosphorylation of kinases Shc , Akt , and ERK1/2 . In addition to TrkA receptors , binding of DHEA to the low affinity NGF receptor was also functional , affording the activation of p75NTR receptors . Unlike TrkA receptors , p75NTR lacks any enzymatic activity . Signal transduction by p75NTR proceeds via ligand-dependent recruitment and release of cytoplasmic effectors to and from the receptor . Indeed , DHEA like NGF facilitated the recruitment of two major cytoplasmic interactors of p75NTR , TRAF6 and RIP2 proteins . Additionally , DHEA-mediated activation of p75NTR led to the dissociation of bound RhoGDI , a protein belonging to small GTPases and interacting with RhoA [21] . A schematic representation of our findings is shown in Figure 7 . Previous findings suggest that DHEA protects PC12 cells against apoptosis via pertussis toxin ( PTX ) sensitive , G protein-associated specific plasma membrane-binding sites [9] . Indeed , PTX was shown to partially reverse the anti-apoptotic effects of DHEA and its membrane impermeable DHEA-BSA conjugate , as well as their effects on prosurvival kinases PI3K/Akt , the activation of transcription factor NFkappaB , and the phosphorylation and inactivation of apoptotic protein Bad [10] . Interestingly , the prosurvival effects of NGF in sympathetic neurons and PC12 cells are also partially reversed by PTX [38] . Furthermore , the NGF-dependent activation of Akt is partially attenuated by PTX , indicating the participation of G ( i/o ) proteins . In the same study , NGF-induced phosphorylation of Bad and transcriptional activity of NFkappaB were also shown to be sensitive to PTX [38] . It appears that other NGF-driven pathways are sensitive to PTX too . For instance , in PC12 cells and primary cortical neurons the NGF-induced phosphorylation of tuberin ( a critical translation regulator holding a central role in NGF-promoted neuronal survival ) is partially blocked by PTX , suggesting the participation of G ( i/o ) proteins [39] . Finally , NGF-dependent activation of the p42/p44 mitogen-activated protein kinase ( p42/p44 MAPK ) pathway in PC12 cells was effectively blocked by PTX [40] . However in HEK293 cells transfected with TrkA receptors , PTX was unable to affect the induction of TrkA phosphorylation by NGF or DHEA ( Figure S5 ) . These findings considered together suggest that TrkA receptors may use down-stream G protein-coupled receptor pathways , after binding and activation by NGF or DHEA , to control neuronal cell survival . It is worth noticing that the interaction of DHEA with the NGF system was first suggested 15 years ago by Compagnone et al . , showing co-localized staining of CYP17 , the rate limiting enzyme of DHEA biosynthesis , and NGF receptors in mouse embryonic DRGs [41] . About one-fifth of CYP17-immunopositive DRG neurons in the mouse were found to be also TrkA-immunopositive . Among the TrkA-expressing cells , about one-third also express CYP17 , while p75NTR-expressing neurons represent only 13% of the cells in the DRG . Thus , about one-fifth of CYP17-immunopositive neurons may be able to respond to both DHEA and NGF stimulation , an observation compatible with our data , presented in Figure 6C . A recent report further supports the interaction of DHEA with NGF receptors . Indeed , DHEA was shown to act as a keratinocyte-deriving neurotrophic signal , mimicking NGF in promoting axonal outgrowth of NGF non-producing but TrkA positive sensory neurons , an effect blocked by TrkA inhibitor K252a [42] . CYP17 is expressed in invertebrate cephalochordata Amphioxus [43] . Amphioxus is also expressing TrkA receptor homologous AmphiTrk , which effectively transduces signals mediated by NGF [44] . Phylogenetic analysis of neurotrophins revealed that they emerged with the appearance of vertebrates ( 530–550 million years ago ) , when complexity of neural tissue increased [45] . Invertebrate cephalochordata like Amphioxus are positioned on the phylogenetic boundary with vertebrates ( 600 million years ago ) . It is thus tempting to hypothesize that DHEA contributed as one of the “prehistoric” neurotrophic factors in an ancestral , simpler structurally invertebrate nervous system [46]; then , when a strict tempospatial regulation of evolving nervous system of vertebrates was needed , peptidic neurotrophins emerged to afford rigorous and cell specific neurodevelopmental processes . In conclusion , our findings suggest that DHEA and NGF cross-talk via their binding to NGF receptors to afford brain shaping and maintenance during development . During aging , the decline of both factors may leave the brain unprotected against neurotoxic challenges . This may also be the case in neurodegenerative conditions associated with lower production or action of both factors . DHEA analogs may represent lead molecules for designing non-endocrine , neuroprotective , and neurogenic micromolecular NGF receptor agonists .
PC12 cells were transfected with specific si/shRNAs for blocking the expression of TrkA and/or p75NTR receptors . More specifically , three siRNAs and two shRNAs for TrkA and p75NTR , respectively , were obtained . The sequences for TrkA siRNAs ( Ambion ) were: GCCUAACCAUCGUGAAGAG ( siRNA ID 191894 ) , GCAUCCAUCAUAAUAGCAA ( siRNA ID 191895 ) , and CCUGACGGAGCUCUAUGUG ( siRNA ID 191893 ) . Sequences for p75NTR ( Qiagen ) were: GACCUAUCUGAGCUGAAA ( Cat . No . SI00251090 ) and GCGUGACUUUCAGGGAAA ( CatNo SI00251083 ) . Rat TrkA was expressed from the pHA vector backbone and rat p75NTR was expressed from the pCDNA3 vector backbone ( InVitrogen ) using a full length coding sequence flanked by an N-terminal hemagglutinin ( HA ) epitope tag . Plasmids to express RIP2 [19] and RhoGDI [36] were myc-tagged , while TRAF6 [19] was FLAG-tagged , as previously described . The origin of antibodies was as follows: Bcl-2 ( Cat . No . C-2 , sc-7382 , Santa Cruz Biotechnology Inc . ) , phospho TrkA ( Cat . No . 9141 , Cell Signaling ) , TrkA ( Cat . No . 2505 , Cell Signaling , was used for Western Blotting and Cat . No . 06-574 , Upstate , was used for immunostainings ) , p75NTR ( Cat . No . MAB365R , Millipore ) , c-myc ( Cat . No . 9E10 , sc-40 , Santa Cruz Biotechnology Inc . ) , phospho ERK1/2 ( Cat . No . 9106 , Cell Signaling ) , Erk1/2 ( Cat . No . 9102 , Cell Signaling ) , phospho-Shc ( Tyr239/240 ) Antibody ( Cat . No . 2434 , Cell Signaling ) , Shc ( Cat . No . 2432 , Cell Signaling ) , phospho-Akt ( Ser473 ) ( Cat . No . 9271 , Cell Signaling ) , Akt ( Cat . No . 9272 , Cell Signaling ) , anti-FLAG ( M2 ) mouse monoclonal ( Cat . No . F1804 , Sigma ) , pTyr ( Cat . No . sc-508 , Santa Cruz Biotechnology Inc . ) , active Caspase-3 ( Cat . No . ab13847 , Abcam ) , Tyrosine Hydroxylase ( Cat . No . ab6211 , Abcam ) , anti-rabbit-R-phycoerythrin conjugated ( Cat . No . P9537 , Sigma ) , anti-mouse-fluorescein conjugated ( Cat . No . AP124F , Millipore ) , anti-rabbit Alexa Fluor 488 ( Cat . No . A21206 ) , anti-rabbit Alexa Fluor 546 ( Cat . No . A10040 ) , and GAPDH ( Cat . No . 2118 , Cell Signaling ) . PC12 cells were obtained from LGC Promochem ( LGC Standards GmbH , Germany ) and nnr5 cells from Dr . C . F . Ibáñez ( Karolinska Institute ) . Both cell types were grown in RPMI 1640 containing 2 mM L-glutamine , 15 mM HEPES , 100 units/ml penicillin , 0 . 1 mg/ml streptomycin , and 10% horse serum , 5% fetal calf serum ( both charcoal-stripped for removing endogenous steroids ) at 5% CO2 and 37°C . HEK-293 cells were obtained from LGC Promochem . Cells were grown in DMEM medium containing 10% fetal bovine serum ( charcoal-stripped for removing endogenous steroids ) , 100 units/ml penicillin , and 0 . 1 mg/ml streptomycin , at 5% CO2 and 3°C . HEK-293 and PC12 cells were transfected with Lipofectamine 2000 ( InVitrogen ) according to manufacturer's instructions . Transfected cells were typically used on the 2nd day after transfection . PC12 cells were cultured in 12-well plates , and 24 h later they were transfected with the si/shRNAs for TrkA and/or p75NTR . Twenty-four hours later the medium was aspirated and replaced either with complete medium ( serum supplemented ) or serum free medium in the absence or the presence of DHEA or DHEA-BSA conjugate at 100 nM . Apoptosis was quantified 24 h later with annexin V-FITC and PI ( BD Pharmingen ) according to our protocol [8] . HEK293 cells were allowed to grow on gelatin-coated glass coverslips for 24 h in culture medium , and 24 h later they were transfected with the cDNAs for TrkA , and p75NTR receptors or the vector alone . Staining was performed 48 h after transfection . Culture medium was aspirated and transfectants were washed twice with PBS buffer . Primary antibodies against TrkA ( rabbit , Upstate , No . 06-574 , diluted 1∶100 ) or p75NTR ( mouse monoclonal ab , MAB365R , Millipore , dilution 1∶500 ) were added for 30 min at 37°C . Secondary antibodies , anti-rabbit-R-phycoerythrin conjugated ( Sigma , No . P9537 ) , and anti-mouse-fluorescein conjugated ( No . AP124F , Millipore ) were added at 1∶100 dilution and transfectants were incubated for 30 min at 37°C; then they were washed three times with PBS and counterstained with Hoechst nuclear stain ( Molecular Probes ) for 5 min . Transfectants were also incubated with the DHEA-BSA-FITC or the BSA-FITC conjugates ( 10−6M ) for 15 min at room temperature in the dark; then they were washed with serum free culture medium and incubated for another 15 min in serum free culture medium containing 4% BSA . Coverslips were mounted to slides with 90% glycerin and were observed with a confocal laser scanning microscope ( Leica TCS-NT , Leica Microsystems GmbH , Heidelberg , Germany ) , mounted with a digital camera . HEK293 cells were cultured in 12-well plates , and 24 h later they were transfected with the cDNAs for TrkA and/or p75NTR receptors , or the vector alone . Staining was performed 48 h later . Transfectants ( 5×105 cells ) were pelleted and incubated with 20 µl of the primary antibodies against TrkA or p75NTR receptors for 30 min over ice . Afterwards , transfectants were washed three times with PBS and 20 µl of the secondary antibodies , and anti-rabbit-R-phycoerythrin conjugated and anti-mouse-fluorescein conjugated were added , as described above . For DHEA-BSA-FITC binding on cells , 20 µl ( 100 nM ) were added on the pelleted cells for 10 min at RT , and then they were washed with serum free culture medium and incubated for another 15 min in serum free culture medium containing 4% BSA . Transfectants were washed twice with PBS , resuspended in 500 µl of PBS , and were analyzed in a Beckton-Dickinson FACSArray apparatus and the CELLQuest software ( Beckton-Dickinson , Franklin Lakes , NJ ) . NovaPEG amino resin ( loading value 0 . 78 mmol/g ) was purchased from Novabiochem . NMR spectra were recorded on a Varian 300 spectrometer operating at 300 MHz for 1H and 75 . 43 MHz for 13C or on a Varian 600 operating at 600 MHz for 1H . 1H NMR spectra are reported in units of δ relative to the internal standard of signals of the remaining protons of deuterated chloroform , at 7 . 24 ppm . 13C NMR shifts are expressed in units of δ relative to CDCl3 at 77 . 0 ppm . 13C NMR spectra were proton noise decoupled . IR spectra was recorded at Bruker Tensor 27 . Absorption maxima are reported in wavenumbers ( cm−1 ) . 3β-Acetoxy-17 , 17-ethylenedioxyandrost-5-ene ( 0 . 74 g , 1 . 98 mmol ) and N-hydroxy phthalimide ( 0 . 71 g , 2 . 2 mmol ) were dissolved in acetone ( 39 mL ) containing 1 mL of pyridine . The mixture was stirred vigorously at room temperature and sodium dichromate dihydrate ( 0 . 89 g , 3 mmol ) was added . Additional portions of solid sodium dichromate dihydrate ( 0 . 89 g , 3 mmol ) were added after 10 and 20 h stirring at room temperature . After reaction completion ( 48 h ) , the mixture was diluted with dichloromethane , filtered through a bed of celite , and the filtrate was washed with water , saturated sodium bicarbonate solution , and brine . The organic layer was dried over anhydrous sodium sulfate , the solvent evaporated in vacuo , and the residue purified by flash column chromatography using hexane/acetone/25% NH4OH ( 85∶15∶0 . 1 mL ) as eluent to afford 3β-acetoxy-17 , 17-ethylenedioxyandrost-5-ene-7-one ( 0 . 6 g , yield: 78% ) . 1H NMR ( CDCl3 , 300 MHz ) δ: 0 . 87 ( s , 3H ) , 1 . 21 ( s , 3H ) , 1 . 26–2 . 00 ( m , 14H ) , 2 . 05 ( s , 3H ) , 2 . 20–2 . 51 ( m , 3H ) , 3 . 84–3 . 92 ( m , 4H ) , 4 . 68–4 . 76 ( m , 1H ) , 5 . 70 ( d , J = 1 . 58 Hz , 1H ) . To a solution of 3β-acetoxy-17 , 17-ethylenedioxyandrost-5-en-7-one ( 0 . 1 g , 0 . 26 mmol ) in pyridine ( 1 . 9 mL ) was added O- ( carboxymethyl ) hydroxylamine hemihydrochloride ( 0 . 11 g , 0 . 52 mmol ) and the reaction mixture was stirred overnight under argon . After completion of the reaction , the solvent was evaporated and the residue was diluted with ethyl acetate . The organic layer was washed with water and brine , dried over anhydrous sodium sulfate , and the solvent was evaporated in vacuo to afford 3β-acetoxy-17 , 17-ethylenedioxyandrost-5-en-7-one7- ( O-carboxymethy1 ) oxime as a white foam ( 0 . 12 g , yield: 100% ) . 1H NMR ( CDCl3 , 300 MHz ) δ: 0 . 88 ( s , 3H ) , 1 . 13 ( s , 3H ) , 1 . 16–1 . 95 ( m , 12H ) , 2 . 04 ( s , 3H ) , 2 . 25–2 . 59 ( m , 5H ) , 3 . 84–3 . 95 ( m , 4H ) , 4 . 59 ( d , J = 2 . 29 Hz , 2H ) , 4 . 62–4 . 73 ( m , 1H ) , 6 . 51 ( d , J = 1 . 47 Hz , 1H ) . To a solution of 3β-acetoxy-17 , 17-ethylenedioxyandrost-5-en-7-one-7- ( O-carboxymethy1 ) oxime ( 0 . 12 g , 0 . 26 mmol ) in a mixture of acetone/water ( 5∶1 , 6 . 3 mL ) was added p-toluenesulfonic acid monohydrate ( 0 . 019 g , 0 . 10 mmol ) , and the reaction mixture was stirred until the starting material was consumed ( 48 h ) . The solvent was evaporated in vacuo and the residue was diluted with ethyl acetate . The organic layer was washed with water and brine , dried over anhydrous sodium sulfate , and the solvent was evaporated in vacuo to afford 3β-acetoxy-androst-5-en-7 , 17-dione 7- ( O-carboxymethy1 ) oxime as a white foam ( 0 . 11 g , yield: 100% ) . 1H NMR ( CDCl3 , 600 MHz ) δ: 0 . 90 ( s , 3H ) , 1 . 15 ( s , 3H ) , 1 . 20–1 . 95 ( m , 12H ) , 2 . 05 ( s , 3H ) , 2 . 09–2 . 68 ( m , 5H ) , 4 . 63 ( d , J = 4 . 18 Hz , 2H ) , 4 . 65–4 . 71 ( m , 1H ) , 6 . 56 ( d , J = 1 . 39 Hz , 1H ) . To a solution of 3β-acetoxy-androst-5-en-7 , 17-dione 7- ( O-carboxymethy1 ) oxime ( 0 . 11 g , 0 . 26 mmol ) in methanol ( 3 . 9 mL ) was added LiOH ( 1 . 5 mL , 1 . 5 mmol , 1N solution ) , and the reaction mixture was stirred until the starting material was consumed ( 4 h ) . The solvent was evaporated in vacuo and the residue was diluted with water . The solution was acidified with 10% hydrochloric acid and DHEA-7-CMO precipitated as a white solid , which was isolated by filtration ( 0 . 097 g , yield: 100% ) . 1H NMR ( CDCl3/CD3OD , 600 MHz ) δ: 0 . 90 ( s , 3H ) , 1 . 14 ( s , 3H ) , 1 . 20–2 . 75 ( m , 17H ) , 3 . 49–3 . 54 ( m , 1H ) , 4 . 54 ( s , 2H ) , 6 . 54 ( s , 1H ) . 3β-Hydroxy-17-oxoandrost-5-en-7-O- ( carboxymethyl ) oxime ( DHEA-7-CMO ) ( 192 mg , 0 . 511 mmol ) in DMF ( 5 mL ) was treated with HOBt ( 69 mg , 0 . 511 mmol ) and DIC ( 0 . 08 mL , 0 . 511 mmol ) , and the resulting mixture was stirred at room temperature for 30 min . This solution was added to NovaPEG amino resin ( 130 mg , 0 . 102 mmol , 0 . 78 mmol/gr ) ( pre-swollen with DMF for 1 h ) and the slurry was shaken at room temperature overnight . The mixture was filtered , the resin was sequentially washed with dichloromethane ( 3× ) , methanol ( 3× ) , and diethyl ether ( 3× ) , and was dried in vacuo overnight . Yield 175 mg ( 100% ) , loading value 0 . 61 mmol/gr . 13C NMR ( gel phase , CDCl3 ) δ: 220 . 66 , 170 . 15 , 157 . 10 , 154 . 15 , 113 . 11 , 72 . 57 , 66 . 59 , 49 . 92 , 47 . 86 , 42 . 15 , 38 . 46 , 37 . 08 , 36 . 53 , 35 . 49 , 31 . 20 , 30 . 71 , 24 . 96 , 20 . 15 , 18 . 05 , 13 . 95; IR: νmax/cm−1 2865 ( s ) , 1735 ( m ) , 1669 ( w ) , 1653 ( w ) , 1637 ( w ) , 1456 ( m ) , 1348 ( w ) , 1289 ( w ) , 1247 ( w ) , 1093 ( s ) , 946 ( w ) . HEK293 cells were transfected with the appropriate plasmids ( TrkA , p75NTR , RIP2 , TRAF-6 , and RhoGDI ) by using Lipofectamine 2000 ( Invitrogen ) . Cells were harvested 48 h after transfection and suspended in lysis buffer ( 50 mM Tris-HCl , 0 . 15 M NaCl , 1% Triton-X100 , pH 7 . 4 ) supplemented with protease inhibitors . Lysates were precleared for 1 h with Protein A-Sepharose beads ( Amersham ) and immunoprecipitated with the appropriate antibody ( pTyr , Flag , or c-myc ) overnight at 4°C . Protein A Sepharose beads were incubated with the lysates for 4 h at 4°C with gentle shaking . In the case of immobilized DHEA-7-CMO , HEK293 or PC12 cells lysates or purified receptors ( both from R&D Systems , Recombinant Mouse NGF R/TNFRSF16/Fc Chimera , Cat . No . : 1157-NR and Recombinant Rat Trk A/Fc Chimera , Cat . No . : 1056-TK ) were incubated overnight at 4°C with the NovaPEG amino resin alone or conjugated with DHEA . Beads were collected by centrifugation , washed four times with lysis buffer , and resuspended in SDS loading buffer . Proteins were separated by SDS/PAGE , followed by immunoblotting with specific antibodies . PC12 or HEK293 cells lysates were electrophoresed through a 12% SDS-polyacrylamide gel , and then proteins were transferred to nitrocellulose membranes , which were processed according to standard Western blotting procedures , as previously described [8] . To detect protein levels , membranes were incubated with the appropriate antibodies: Bcl-2 ( dilution 1∶500 ) , phospho TrkA ( dilution 1∶500 ) , total TrkA ( dilution1∶500 ) , p75NTR ( dilution 1∶500 ) , phospho Shc ( dilution 1∶1000 ) , total Shc ( dilution 1∶1000 ) , phospho Akt ( dilution 1∶500 ) , total Akt ( dilution 1∶500 ) , phospho ERK1/2 ( dilution 1∶500 ) , and total ERK1/2 ( dilution 1∶500 ) . Proteins were visualized using the ECL Western blotting kit ( ECL Amersham Biosciences , UK ) , and blots were exposed to Kodak X-Omat AR films . A PC-based Image Analysis program was used to quantify the intensity of each band ( Image Analysis , Inc . , Ontario , Canada ) . To normalize for protein content the blots were stripped and stained with GAPDH antibody ( dilution 1∶1000 ) ; the concentration of each target protein was normalized versus GAPDH . Where phosphorylation of TrkA or kinases was measured , membranes were first probed for the phosphorylated form of the protein , then stripped , and probed for the total protein . Superior cervical ganglia ( SCG ) were removed from newborn ( P0–P1 ) rat pups and dissociated in 0 . 25% trypsin ( Gibco , 15090 ) for 30 min at 37°C . After dissociation SCG neurons were re-suspended in culture medium ( Gibco , Neurobasal Cat . No . 21103 ) containing 1% fetal bovine serum ( FBS ) , 100 units/ml penicillin , 0 . 1 mg/ml streptomycin , 3 µg/ml araC antimitotic , and 100 ng/ml NGF ( Millipore , 01-125 ) . Cells were plated on collagen coated 24-well plates and cultured for 5 d prior to use . For NGF withdrawal experiments , cells were washed twice with Neurobasal containing 1% FBS and fresh culture medium lacking NGF and containing anti-NGF antibody at 1∶50 dilution ( Millipore , AB1526 ) . DHEA , TrkA-inhibitor ( Calbiochem , 648450 ) and anti-p75NTR ( mouse , MAB365R Millipore ) were used at 100 nM , 100 nM , and 1∶50 , respectively . ngf+/− mice [13] were obtained from the Jackson Laboratory and maintained on C57BL/6 background . All procedures described below were approved by the Animal Care Committee of the University of Crete , School of Medicine . Animals were housed in cages maintained under a constant 12 h light–dark cycle at 21–23°C , with free access to food and tap water . Genotyping was performed on tail DNA using the following primers: NGFKOU2 ( 5′CCG TGA TAT TGC TGA AGA GC3′ ) , NGFU6 ( 5′CAG AAC CGT ACA CAG ATA GC3′ ) , and NGFD1 ( 5′TGT GTC TAT CCG GAT GAA CC3′ ) . Genomic PCR reactions containing the 3 primers were incubated for 32 cycles at 95°C ( 30 s ) /59°C ( 30 s ) /72°C ( 1 min ) . Mice heterozygous for the NGF null mutation were interbred to obtain mice homozygous for the NGF gene disruption and the first day of gestation determined by the discovery of a copulation plug . The mothers were treated daily with a subcutaneous injection of DHEA ( 2 mg/day ) or vehicle ( 4 . 5% ethanol in 0 . 9% saline ) starting from the third day after gestation . Animals were collected at E14 . At the day of collection the mothers were deeply anesthetized with sodium pentobarbital ( Dolethal 0 . 7 ml/kg i . p . ) followed by transcardial perfusion with saline solution containing heparin for about 7 min , and with 4% PFA , 15% Picric Acid , 0 . 05% GA in phosphate buffer 0 . 1 M , for another 7 min . After the perfusion the embryos were collected and maintained in the same fixative overnight at 4°C . Embryos were then washed in 0 . 1 M phosphate buffer and cryoprotected by using 10% sucrose followed by 20% sucrose overnight at 4°C . Finally , embryos were frozen in OCT in iso-pentane over liquid nitrogen for 5 min and the frozen tissues were stored for later use at −80°C . The samples were sectioned ( 20 µm ) and mounted onto Superfrost plus slides ( Menzel-Glaser J1800AMNZ ) . Slides were left to air-dry overnight at room temperature ( RT ) and were then either used immediately or were fixed in cold acetone for 1 min and stored at −80°C for later use . Stored or fresh slides were fixed for 15 min in cold acetone at 4°C and left to dry for 10 min at room temperature . They were then washed in PB 0 . 1 M , then in TBS , and incubated for 45 min with 10% horse serum in TBS-T 0 . 1% . The normal serum was drained off and the primary antibodies ( anti-TrkA diluted 1∶400 and active Caspase-3 diluted 1∶50 ) , diluted in TBS-T 0 . 1% with 1% horse serum , were added . Sections were incubated for 4 h at RT and overnight at 4°C; they were then washed in TBS-T 0 . 1% and the anti-rabbit secondary antibodies ( Alexa Fluor 488 and Alexa Fluor 546 , 1∶1000 in TBS-T 0 . 1% ) were added for 6 h at RT . Sections were washed in TBS-T , TBS , and in PB 0 . 1 M and were coverslipped with Vectashield ( Vector , H-1400 ) and visualized in a confocal microscope . TUNEL ( Roche , Cat . No . 12156792910 ) and Fluoro-Jade C ( Millipore , Cat . No . AG325 ) staining of apoptotic and degenerating neurons , respectively , was performed according to the manufacturer's instructions . For the statistical evaluation of our data we have used analysis of variance , post hoc comparison of means , followed by the Fisher's least significance difference test . For data expressed as percent changes we have used the nonparametric Kruskal-Wallis test for several independent samples . | Dehydroepiandrosterone ( DHEA ) and its sulphate ester are the most abundant steroid hormones in humans , and DHEA was described as the first neurosteroid produced in the brain . DHEA is known to participate in multiple events in the brain , including neuronal survival and neurogenesis . However , to date no specific cellular receptor has been described for this important neurosteroid . In this study , we provide evidence that DHEA exerts its neurotrophic effects by directly interacting with the TrkA and p75NTR membrane receptors of nerve growth factor ( NGF ) , and efficiently activates their downstream signaling pathways . This activation prevents the apoptotic loss of NGF receptor positive sensory and sympathetic neurons . The interaction of DHEA with NGF receptors may also offer a mechanistic explanation for the multiple actions of DHEA in other peripheral biological systems expressing NGF receptors , such as the immune , reproductive , and cardiovascular systems . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"cell",
"death",
"molecular",
"neuroscience",
"neuroscience",
"membrane",
"receptor",
"signaling",
"signaling",
"pathways",
"signaling",
"in",
"cellular",
"processes",
"endocrinology",
"diabetes",
"and",
"endocrinology",
"biology",
"neuroendocrinology",
"molecular",
"biology",
"signal",
"transduction",
"cell",
"biology",
"molecular",
"cell",
"biology"
]
| 2011 | Neurosteroid Dehydroepiandrosterone Interacts with Nerve Growth Factor (NGF) Receptors, Preventing Neuronal Apoptosis |
Conserved noncoding elements ( CNCs ) are an abundant feature of vertebrate genomes . Some CNCs have been shown to act as cis-regulatory modules , but the function of most CNCs remains unclear . To study the evolution of CNCs , we have developed a statistical method called the “shared rates test” to identify CNCs that show significant variation in substitution rates across branches of a phylogenetic tree . We report an application of this method to alignments of 98 , 910 CNCs from the human , chimpanzee , dog , mouse , and rat genomes . We find that ∼68% of CNCs evolve according to a null model where , for each CNC , a single parameter models the level of constraint acting throughout the phylogeny linking these five species . The remaining ∼32% of CNCs show departures from the basic model including speed-ups and slow-downs on particular branches and occasionally multiple rate changes on different branches . We find that a subset of the significant CNCs have evolved significantly faster than the local neutral rate on a particular branch , providing strong evidence for adaptive evolution in these CNCs . The distribution of these signals on the phylogeny suggests that adaptive evolution of CNCs occurs in occasional short bursts of evolution . Our analyses suggest a large set of promising targets for future functional studies of adaptation .
Phenotypic evolution proceeds both by changes in protein coding sequences and by changes in gene expression that determine when , where , and how much genes are expressed [1–3] . Although recent genome-wide studies have begun the process of identifying genes that show signals of adaptive evolution in coding sequences [4] , much less is known about the adaptation of regulatory sequences . One avenue to studying adaptation of gene regulation is to identify regulatory elements that show rapid evolution at the DNA sequence level [2] . However , a challenge for this approach is that at present we have only limited knowledge of the DNA sequence elements that drive gene expression and regulation . One possible way forward is to study the evolution of conserved noncoding elements ( CNCs ) [5–7] . In recent years it has been shown that ∼3 . 5% of noncoding DNA sequence is substantially conserved across diverse mammals [8–10] , and that a smaller amount of noncoding sequence is also shared with more distant vertebrates , including chicken and even fish [9 , 11–13] . Some CNCs show extremely high levels of conservation; for example , Bejerano et al . [9] identified 481 segments longer than 200 bp that are absolutely conserved among the human , rat , and mouse genomes . Recent studies of CNCs , using varied definitions , have reported that most CNCs are segments of around 100–300 bp , and that they are widely distributed across the human genome [9 , 10 , 14–18] . CNCs are not preferentially located near genes [18] . In some cases , clusters of CNCs are found in gene deserts and a subset of these CNCs have been shown to play functional roles as enhancers [19–21] . It has been shown repeatedly that screening for CNCs is an effective method for identifying cis-regulatory modules of gene expression [18–25] . CNCs that are shared among humans and distant outgroups such as Fugu are heavily overrepresented near developmental regulator genes , and many serve as highly conserved regulators of these functionally conserved genes [13] . That said , there is still considerable uncertainty about the function of most CNCs , and it has been suggested that some CNCs may serve other kinds of functions , perhaps including roles in chromatin structure or structural connections between chromosomes [26] . In principle , another possibility might be that many CNCs could simply be regions of the genome with low mutation rates . However , two kinds of evidence argue convincingly that the low evolutionary rates of CNCs are indeed due to selective constraint . First , the allele frequency spectrum of human SNPs that lie within CNCs is skewed towards rare variants , consistent with the action of weak purifying selection [27 , 28] . Second , the rate of evolutionary change of CNCs is closer to the neutral rate in primates than in rodents [28 , 29] . The latter observation is probably due to reduced efficiency of weak purifying selection in primates , which have smaller effective population sizes . Hence , in this study , in view of the likely functional importance of CNCs , we set out to describe the patterns of evolutionary sequence change in these elements . We start with a simple null model in which the evolution of each CNC is characterized by a single substitution rate parameter r that accounts for varying levels of constraint and local mutation rate across CNCs . For each CNC we compare the null model to a hierarchy of alternative models that allow the CNC to have different evolutionary rates in different parts of the phylogeny . In the simplest alternative model , the CNC evolves at a single rate across the phylogeny except for one branch , which shows a change in rate ( Figure 1 ) . More complex alternative models allow multiple changes in rate . Increases in rate can be interpreted as evidence for positive adaptation or relaxation of functional constraint for the element in question . Decreases in rate are consistent with a tightening of selective constraint . Two recently published papers [5 , 7] have taken similar approaches to identify nongenic regions that show accelerated evolution specifically in the human lineage . Both studies concluded that human lineage-selection signals are enriched near neurological genes . In the study of Pollard et al . [5] , the most dramatically accelerated region was found to be part of a novel RNA gene that is expressed during cortical development . Here , we expand this kind of approach to look more broadly at evolutionary patterns of CNCs across the mammals .
To identify CNCs that have been targets of selection , we introduce a likelihood ratio test that we call the “Shared Rates Test” ( SRT ) . Under the null model , the divergence times of lineages are shared across CNCs , but each CNC may evolve faster or slower according to its local mutation rate and level of evolutionary constraint . For each CNC , we test whether any branches are surprisingly long or short compared to the others , indicating speed-ups or slow-downs of the substitution rate . For example , in Figure 1 , the first two trees evolve at different rates , but with the same tree “shape” ( i . e . , the ratios of branch lengths are the same ) . In contrast , the third tree has a longer-than-expected branch on the human lineage , suggesting the action of natural selection . In our model , each branch of the mammalian tree has a branch-length parameter vb , defined as the average number of substitutions per site on branch b for CNCs evolving under a constant level of constraint . ( Here , vb is defined as the average number of substitutions per site on branch b across all CNCs . ) In addition , under the null hypothesis , each CNC is associated with a single rate parameter r0 ( h ) ( where h indicates a particular CNC ) . Then the number of substitutions that occur in CNC h , on branch b has an expectation at each site of Nb , h , where Under the null model , there are seven branch length parameters for the tree that we consider , and one additional rate parameter for each CNC . As described in the Methods and Text S1 , we obtain a joint maximum likelihood estimate for all the parameters , assuming the Felsenstein 84 model of sequence evolution [31] . Our model is designed so that all CNCs have the same expected tree shape ( i . e . , the ratios of expected branch lengths are the same ) . However the total size of the tree is allowed to vary according to r0 ( h ) , in order to reflect variation in mutation rates and the level of selective constraint across CNCs . In addition , we place no constraints on the relative values of the vb , so that lineage-specific variation in mutation rates ( such as the higher substitution rate in rodents ) is reflected in longer estimates for those branch lengths ( Figures 1 and S1 ) . In summary , the null model allows mutation rates and levels of constraint to vary across CNCs , and it allows for the property that broad-scale mutation rates may vary across lineages . In addition to the basic null model , we consider a family of alternative models that allow additional rate parameters for particular CNCs . In the simplest alternative , a single branch on the tree evolves at a rate that is different from the background rate shared by the remaining lineages ( as for the third tree in Figure 1 ) . In the extreme alternative , each of the seven branches evolves with its own rate ri ( h ) , giving a total of seven rate parameters for the CNC in question . ( For simplicity of notation , we will henceforth drop the notation h on the rate parameters . ) In the extreme case , to test the hypotheses H0: r1 = r2 = ···= r7 ( = r0 ) versus HA: r1 ≠ r2 ≠ ···≠ r7 at a particular CNC , we compute the SRT as where L is the likelihood of the sequence data for the five mammalian species , maximized with respect to the rate parameters , and with the fixed estimate of branch lengths parameters ( ) and the sequence evolution model . Large values of the SRT indicate a substantially better fit of the alternative than the null model . Another example of alternative model is the case in which branches 2 and 3 have distinct rates r2 and r3 , while the other branches have a single “background” rate r0 , −2 , −3 . In this case , to test the hypotheses H0: r1 = r2 = ···= r7 ( = r0 ) versus HA: r2 ≠ r3 ≠ r1 = r4 = ···= r7 ( = r0 , −2 , −3 ) , we can compute the likelihood ratio statistic as In this paper , we perform two kinds of analyses . One analysis performs model selection using the SRT , while the other tests for individual branches with rate changes . When testing for a rate change on the ith branch only , it is convenient to transform the likelihood ratio statistic as follows . In this case , we will use special notation , denoted by SRTi: where sign ( x ) = 1 if x > 0 and otherwise sign ( x ) = −1 . Rewriting the SRT in this way provides the convenient property that SRTi > 0 implies that ri is larger than the background rate r0 , −i , and hence branch i shows a rate speed-up relative to the rest of the tree; conversely , SRTi < 0 implies a slow-down on branch i . As a convention , when we subscript SRT by a character or number , it will represent the signed likelihood ratio statistic testing for rate changes on the indicated branch . Otherwise , the notation SRT without subscripts will be used to indicate use of an unsigned test statistic , in the form of Equations 2 and 3 . Our SRT is a likelihood ratio test and , as such , standard theory suggests that under the null hypothesis the test statistic should asymptotically follow a chi-square distribution with degrees of freedom equal to the difference in the number of estimated parameters between the constrained ( null ) and less-constrained ( alternative ) models . Similarly , the signed root of this statistic for a one-dimensional parameter of interest is asymptotically standard normal . Therefore , when the null hypothesis is true and the number of sites in a CNC is large enough , the unsigned SRT might be expected to follow the chi-square distribution with the degrees of freedom equal to the difference in the number of rate parameters between the two models . For example there are six degrees of freedom in the global test ( Equation 2 ) and two degrees of freedom in the example in Equation 3 . Similarly , under the null , the signed test SRTi is constructed to have a standard normal distribution as the CNC size goes to infinity . Our simulation studies show that the asymptotic theory is reasonably accurate for both versions of the test statistic , except in the cases in which the lineages tested for selection are relatively short and are expected to accumulate few substitutions ( namely , the human and chimpanzee lineages; Figure S3 ) . Hence , to reduce computational burden , we calculate p-values using the asymptotic chi-square or normal approximations , except for tests on the human and chimpanzee branches for which , except where stated , we compute p-values based on the empirical null distribution in simulated data ( see Methods ) . An additional consideration is that we do not want the estimated null branch lengths ( vb ) to be heavily influenced by outlier CNCs with evidence for selection . To mitigate the impact of such CNCs , we first identify CNCs with clear overall departures from the null model ( SRT > 25 in the global six degrees of freedom test , corresponding to p < 0 . 00034 ) , and then reestimate the branch lengths after dropping those nonneutral CNCs , which represent 2 . 8% and 3 . 8% of the total mammalian and amniotic CNCs , respectively . In summary , then , our analysis performs the following steps: ( 1 ) Estimate maximum likelihood branch lengths and rates under the null; ( 2 ) identify outlier CNCs that have SRT > 25 comparing the seven- and one-parameter models; ( 3 ) drop outlier CNCs and recalculate the null branch lengths and rates; and ( 4 ) compute the shared rates test statistics for each CNC according to a range of alternative models . For reasons discussed below , in practice these analyses were performed in a sliding window of 50 consecutive CNCs , as defined by position in the human physical map . All analyses considered the mammalian and amniotic CNCs separately . It is well established that the extent of divergence among mammalian species varies substantially across large genomic regions [33–38] . For example , Gaffney and Keightley [38] showed that divergence between the mouse and rat genomes varied between and within chromosomes . While the causes and the scales of this type of variation are not completely understood , it has been shown that divergence correlates with various genomic features , including GC and CpG content , simple-repeat structures , and recombination rate , suggesting that these genomic features drive variation in mutation rates [35 , 37] . Variation in mutation rates or levels of CNC conservation across genomic regions should not be problematic for our method , provided that the substitution rate in any given region maintains a constant ratio to the average across the mammalian phylogeny . If a CNC is in a region with a higher , or lower , mutation rate than average , this effect should simply be absorbed into the rate parameter that we estimate for each CNC as part of our null model . However , if mutation rate variation is not stable across the phylogeny , this might produce false signals for our method . Therefore , we looked at whether the average tree shapes are significantly variable across chromosomes ( according to the human physical map ) as well as within chromosomes . We found that in fact there is nontrivial variation in tree shape , both at the chromosome level , and across genomic regions within chromosomes . For example , within Chromosome 2 there is a highly significant autocorrelation in the fraction of the tree occupied by the mouse lineage ( Figure 2 ) . This result implies that local variation in large-scale mutation rates is not conserved across evolutionary time; for example , genomic regions that evolve faster than average on some lineages may evolve slower than average elsewhere on the tree . If average tree shapes were constant across the genome , we could use CNCs from across the genome to estimate the tree shape for our null model . However , the observation that tree shape is not constant suggests that instead our model should allow for variation in tree shape across the genome . After some experimentation , we settled on using a sliding window of 50 consecutive CNCs to estimate the tree shape . That is , we test each CNC for significant departures from the tree shape in a 50-CNC window that , in the human physical map , is centered near the CNC in question ( see Methods ) . On average , this window size corresponds to 525 kb and 1 . 3 Mb ( median ) for mammalian CNCs and amniotic CNCs , respectively . Overall , we find that using the sliding window method produces only a modest impact on the rate of significant CNCs , but it should improve our inferences by taking into account the local variation in tree shapes ( Figures 2 and S4 ) . An obvious concern about using a sliding window based on the locations of CNCs in humans is that due to chromosomal rearrangements , CNCs that are close together in humans may not be close together in other mammals . Consequently , a sliding window based on the human map might not provide a suitable correction . Fortunately , our window size is relatively small compared to the typical size of syntenic blocks [8 , 39] and in Figure 3 , we show that the results of tests on the human lineage are highly concordant whether we use windows based on the human or mouse physical maps and , indeed , are only modestly different from the results using all CNCs together . Consequently , all subsequent results use 50-CNC windows based on the human map . Another plausible concern about our model stems from the prediction that selection against weakly deleterious mutations is more efficient in species with large populations than in small populations . This means that weakly constrained sites in CNCs are likely to evolve more quickly in primates than in rodents ( which have larger effective population sizes ) . This effect has been observed in a comparison between the evolutionary rates of CNCs and putatively neutral flanking sequences [29] . Hence—in contrast to our null model—one might expect the overall tree shape for a CNC to depend on its level of selective constraint . To investigate this issue , we classified CNCs into four different levels of conservation , according to their substitution rates on the dog lineage . We then separately compared the average human-to-chimpanzee divergence against the average mouse-to-rat divergence , within each of the four conservation levels ( Table S15 ) . We find that that as the level of constraint increases , the divergence in rodents indeed decreases faster than divergence in hominids , consistent with the results of Keightley et al . [29] . However , we find that the variation across CNCs is relatively small ( less than 11% change across different classes of CNCs ) and much less than when CNCs are compared to neutral sequences ( Table S3 ) . As shown below , we do not have the power to detect such small variations in tree shape at individual CNCs , so we conclude that it is not necessary to control for overall conservation level more carefully for the current study . For each CNC , we calculated SRTi for each of the seven branches of the mammalian tree to identify CNCs that have experienced a speed-up or slow-down on a particular branch . Figure 4A shows the histogram of p-values on the mouse lineage ( SRTm ) for the mammalian CNCs . The p-values are defined as P ( SRTi > srti ) where srti is the observed value . Hence , p-values near 0 indicate increased rates , and near 1 indicate decreased rates . The histogram is flat for intermediate p-values with peaks at both ends , suggesting that most CNCs fit the null distribution of SRTm , but with a substantial number of outliers . At the significance level of 0 . 001 , 1027 ( 1 . 2% ) and 503 ( 0 . 6% ) mammalian CNCs show speed-ups and slow-downs , respectively . Among amniotic CNCs , 228 ( 1 . 4% ) and 106 ( 0 . 6% ) show speed-ups and slow-downs , respectively on the mouse lineage . Figure 4B plots the expected and observed branch lengths on the mouse lineage for the CNCs that are significant at p < 0 . 001 in each tail . ( Similar plots for other lineages are shown in Figure S5 . ) The red points above the diagonal indicate CNCs with rate speed-ups . For the central 95% of the significantly fast-evolving CNCs , the observed branch lengths are between 0 . 04 to 0 . 13 substitutions per site , and are 2–4-fold higher than the expected branch lengths . The blue points below the diagonal are CNCs with reduced branch lengths . Nearly half of these CNCs accumulated no substitutions on the mouse lineage . The other long lineages show similar p-value histograms though with some variability in the proportion of significant CNCs . The dog lineage is the most enriched for signals , with 2 . 3% and 1 . 9% of mammalian CNCs showing speed-ups and slow-downs , respectively , at p < 0 . 001 ( in each tail ) . Even after a stringent Bonferroni correction , 186 and 46 CNCs , respectively , are still significant at p = 0 . 001 in the dog lineage . The overall results for amniotic CNCs are similar , but the fraction of significant results is slightly higher on each branch ( Table S8 ) . For most lineages , our significance threshold ( one-sided p-value < 0 . 001 on each end ) corresponds to a genome-wide false discovery rate ( FDR ) between 0 . 05 and 0 . 1 ( Table S9 ) . Since the distribution of SRTi on the human and chimpanzee lineages does not follow the standard asymptotic distribution , we simulated data under the null over a range of substitution rates that cover the observed range over all 50-CNC windows ( see Methods ) . We account for heterogeneity in the distribution of SRTi across bins of CNCs with different numbers of expected substitutions on the tested lineage by computing p-values based on the empirical null distribution of SRTi constructed in each bin ( unpublished data ) . At a significance level of 0 . 001 , 256 mammalian CNCs and 59 amniotic CNCs , respectively , show rate speed-ups on the human lineage ( Table S8 ) . Note that there is little power to detect rate reductions on these very short lineages . To better understand these SRTi results , we performed power simulations under a range of models . The simulation results , summarized in Figure S6 , show considerably greater power to detect speed-ups than slow-downs on all lineages , consistent with the results of Siepel et al . [40] . Thus , the fact that we detect more speed-ups than slow-downs does not necessarily imply that speed-ups are actually more common , and it is likely that many slow-down events are simply not detected by our analysis . Our human results allow a comparison to the human accelerated regions ( HARs ) identified by Pollard et al . [5] using a similar type of approach , based on regions that were highly conserved ( at least 96% identity ) across chimpanzee , mouse , and rat . Among the top 49 HARs , which include two coding regions , 34 overlap with CNCs in our dataset; however , generally , the HARs are considerably shorter and more conserved and lie within our CNCs . Perhaps not surprisingly , since the HARs are the top genome-wide hits in their data , the signals in our overlapping CNCs tend to be weaker . Among the 34 CNCs , just five CNCs are significant in our analysis at a genome-wide FDR less than 0 . 05 . Nonetheless , our CNCs that overlap HARs do show a strong enrichment of modest signals . Our human lineage p-values are <0 . 01 for 26 of the 34 CNCs overlapping HARs , and are <0 . 1 for 33 of the 34 ( Table S10 ) . Within our dataset , one of the most significant CNCs on the human lineage is a 144-bp amniotic CNC located on human Chromosome 21 starting at 33481809 ( q22 . 11 , NCBI Build 35 ) . It was not detected by Pollard et al . [5] because it fails their filtering threshold for similarity between chimpanzee , mouse , and rat . As illustrated in Figure 5 , the posterior expected number of substitutions ( see Methods for details ) on the human lineage is 5 . 2 , which is 26-fold higher than the value of 0 . 2 expected under the null model . The corresponding SRTh is 4 . 84 . The p-value for this CNC is so small that it is difficult to evaluate by simulation; however , the standard normal approximation suggests that p ≈ 6 × 10−7 ( our simulations indicate that this is conservative ) . In addition to the five nucleotide substitutions , there is also a 2-bp insertion on the human lineage that was not included in the statistical inference . Since the UCSC genome browser database was recently updated , we were able to inspect an alignment of 17 vertebrate species for this region . Manual inspection confirmed that all six of these substitutions occurred on the human lineage . The function of this CNC is unclear but the two nearest genes are C21orf54 , 17 kb upstream , and IFNAR2 , 42 kb downstream of the CNC . Not much is known about C21orf54 , but IFNAR2 codes for a type I membrane protein that forms one of the two chains of a receptor for interferons alpha and beta [41] . This CNC is strongly conserved among the other mammalian species and chicken but does not appear to be present in the fugu genome . In addition to the rapid evolution on the human lineage , there is weak evidence for slower evolution of this CNC on the mouse and dog lineages ( one-sided p-values = 0 . 011 and 0 . 023 , respectively; see Figure 5B ) . Thus far , we have focused on the simplest class of alternative models , in which a CNC changes substitution rate on a single branch only and has a constant background rate elsewhere on the tree . We now extend this approach in order to classify each CNC according to a family of more complicated models of evolutionary patterns . Our data are connected by a tree containing seven branches . The simplest model ( our “null” ) has a single rate parameter , and the most complicated alternative model has seven different rate parameters . In between , there are 876 ways of partitioning the seven branches into two or more different substitution rate groups . However , considering all of these partitions does not seem biologically meaningful or necessary , and here we focus on a reduced set of 126 alternative candidate models . The alternative models we consider can be divided into two distinct classes of models . In one class of models , each tree is assumed to have a “background” rate parameter . Then , each CNC may have between one and six “selected” lineages , and each selected lineage evolves at its own rate . In the other class of models , each tree may be split into subtrees that share a single rate , while the rest of the tree has a single background rate ( for full details , see Table S11 ) . We use a modified Akaike Information Criterion ( AIC ) procedure to classify each CNC into its best model . In brief , the method attempts to account for multiplicities of alternative models as well as the number of estimable parameters in each model ( see Methods ) . We have performed simulations to test the performance of this method , and we find that it provides suitable control over the rate of “false positives” ( i . e . , accepting models with more parameters than used to simulate the data ) . That said , our simulations show that it is often difficult to correctly classify complex models with multiple rate changes ( see Methods; Figure S7 ) . The results of our data analysis are summarized in Figure 6 and Table S12 . We estimate that ∼68% ( 54 , 643/81 , 957 ) of the mammalian CNCs evolve at a single rate . The remaining nonneutral CNCs show rate changes on at least one lineage . The number of CNCs assigned to each model category decreases with increasing model complexity . Among the 32% of CNCs with more than one rate , ∼75% ( 20 , 420/27 , 314 ) exhibit rate changes on a single lineage but not on the remaining lineages and ∼9% ( 2 , 419/27 , 314 ) exhibit rate changes on the primate or the rodent lineage that are inherited across all branches below . For the two-parameter models , the rate change events are easily classified as speed-ups or slow-downs . Counts for both types of event are shown in Figure 6B . For most lineages , there are slightly more speed-up events than slow-downs ( ∼55% versus ∼45% ) . However , there are 638 and 530 CNCs that show rate speed-ups on the human and chimpanzee lineages , respectively , far more than the four and eight CNCs , respectively , showing slow-downs . Presumably , these results are due in large part to the greater power to detect speed-ups , as well as differences in power across lineages ( Figure S6 ) . It is notable that the dog lineage shows a very large number of rate changes , which may not be fully explained by the long length of this lineage ( second longest among the seven ) . Since there is no strong tendency towards an excess of speed-ups over slow-downs on this lineage , it is unlikely that this can be explained by occasional CNCs with low-quality dog sequence . Perhaps a hint is that we have observed greater variation in the dog-lineage substitution rates at neutral sites than on other lineages . Perhaps there is greater fine-scale variation on the dog lineage that is not well captured by our 50-CNC window method ( see Methods; Figure S8 ) . As discussed above , we have identified many CNCs with significantly accelerated rates on one or more branches . However , it is unclear a priori whether these speed-ups reflect positive adaptation or relaxation of functional constraint . In order to address this issue , we estimated substitution rates in unconserved sequences near each CNC to estimate local neutral rates ( see Methods ) . We then determined how many of the CNCs showing rate speed-ups have an accelerated rate that actually exceeds the corresponding lineage-specific neutral rate . If the rate in a CNC actually exceeds the local neutral rate , this is strong evidence for adaptive evolution . However , a negative result here is difficult to interpret , since adaptive evolution in an otherwise slow-evolving sequence may not necessarily bring the total rate above the neutral background rate . Our results are summarized in Table 1 . We observe that most CNCs showing accelerations on the human and chimpanzee branches indeed have rate estimates exceeding the neutral rates; of these , more than half are actually significantly faster than the neutral rate at p < 0 . 05 . Meanwhile , the other branches of the mammalian tree all show smaller fractions of CNCs with rates that exceed the neutral rate , and very few of these are significantly faster than the neutral rate . One plausible explanation might be that if there is sufficiently rapid evolution on a long branch , this might cause an otherwise conserved element not to be classified as a “most conserved” region by the HMM [10] . However , some simple calculations suggest that this is likely to be a modest effect in practice . Moreover , we see the same effect for both the mammalian and amniotic CNCs ( Table 1 ) , even though the HMM data for the latter include the relatively long branch to chicken , and should therefore be much less susceptible to this effect . Instead , to explain these observations , we hypothesize that the rate speed-ups that we detect may often reflect rapid bursts of adaptation in which a CNC accumulates a series of sequence changes , thus modifying its function . A single burst of adaptation may produce enough sequence changes to exceed the neutral rate on a short branch , but not on a longer branch . In this model , we would have the most power to detect adaptive events on short branches . Our data argue strongly against a model in which a CNC adapts continuously over extended periods of evolutionary time , as such a model should also produce signals on the long branches . We have also performed analyses of the locations of CNCs showing branch-specific speed-ups , with respect to nearby genes . A recent report by Drake et al . [27] found that the frequency spectrum in CNCs is most skewed towards rare variants ( indicating weak purifying selection ) in introns and near genes , and is less skewed in CNCs that are far from genes . To test whether CNCs showing speed-ups on particular branches occur at higher rates near to or far from genes , we divided all our CNCs into four classes: intronic , within 10 kb of a gene , between 10 kb and 100 kb , and greater than 100 kb from any gene . We found that on the mouse and rat lineages , CNCs showing speed-ups ( p < 0 . 001 on the branch-specific test SRTi ) occur at higher rates in introns and within 10 kb of genes than among CNCs further from genes . However , this trend was not replicated on the other lineages of the tree ( Table S13 ) . We next looked at whether CNCs showing significant rate speed-ups are more likely to be in the proximity of particular kinds of genes [17] , using the PANTHER GO database [32] . A significant difficulty in this sort of analysis is that even for those CNCs that act as cis-regulators , it is unknown which of the nearby genes is being regulated . However , as a rather imperfect proxy for this we simply used , for each CNC , the nearest gene ( in either orientation ) . For each branch of the mammalian tree , we divided the CNCs into those with increased rate on that branch ( by AIC ) and used CNCs evolving under the null model as “neutral” controls . We looked at whether particular biological process categories were enriched among the nearest genes of the selected CNCs compared to the neutral CNCs . For mammalian CNCs , there is significant enrichment of the process categories “amino acid activation” and “other coenzyme and prosthetic group metabolism” on the dog and the lineage leading to the common ancestor of mouse and rat ( rodent lineage ) , respectively , at p < 0 . 05 after Bonferroni adjustment . We also tested whether any categories show repeated evidence for enrichment on different branches of the tree . For mammalian CNCs , the “sensory perception” category appears in the top ten enriched biological processes for three out of the seven lineages . However , in summary , we view these GO associations as rather tentative , since none of them is highly significant or highly repeatable across branches of the tree . Complete results from this analysis are presented in Tables S6 and S7 .
Our paper presents a new approach to studying the evolutionary patterns of CNCs . We find that a large fraction of CNCs ( ∼32% ) do not fit a simple model of evolution with a consistent substitution pattern across the mammalian tree . Among those CNCs that do not fit our null model , ∼75% show changes in evolutionary rate on a single branch of the mammalian tree , while the remainder have more complex substitution patterns . In many cases—particularly on the short branches of the phylogeny—CNCs with rate accelerations on a particular branch significantly exceed the neutral rate on that branch , suggesting that the changes are driven by adaptive evolution . The less extreme speed-ups may be due to either adaptation or a relaxation of selective constraint; however , we suggest that much of our signal on the longer branches may be due to short bursts of adaptation that do not generate enough changes to exceed the total neutral rate on a long branch . A very recent paper by Galtier and Duret [42] argues that many of the recently reported HARs [5] are likely the result of biased gene conversion ( BGC ) . One of the main characteristics of BGC is an excess of AT → GC transitions . In some of our CNCs showing accelerations on the human lineage , we also observe this transition bias , which seems to be larger with increased acceleration signals . However , for most fast-evolving CNCs , the numbers of AT → GC changes roughly match the distribution expected based on the overall distribution across random CNCs ( Figure S9 ) . In summary , these data suggest that some of the fast-evolving CNCs may in fact be due to BGC , however , that most fast-evolving CNCs do not show the signal expected for BGC . Overall , our results imply that either the levels of functional constraint or the functional roles of CNCs are reasonably changeable across the timespan of mammalian evolution . Although it lies beyond the scope of this paper , it will be of interest to use experimental approaches to probe the functional significance of the many CNCs that we have identified as having had bursts of rapid evolution [5 , 25] . Of course , in this type of study , there are inevitably features of the real data that are not fully accounted for in the models . We believe that our results should be reasonably robust to these issues , however , as follows . One natural concern is that our CNC alignments might occasionally align paralogs . This is a serious concern in principle; however , we have aimed to aggressively filter out CNCs with related paralogs to minimize this effect , in addition to making use of global alignments . Other model departures might inflate the variance of branch-specific substitution rates . These include the possibility of fine-scale , branch-specific changes in mutation rate , as well as variation in the branch lengths of the human and chimpanzee branches due to coalescent time variation [43] . On the whole these effects are likely to be fairly modest , since the observed rate changes are usually not significant unless they are quite dramatic ( significant rate changes are usually ∼2–4-fold on the mouse lineage , and larger on the shorter branches ) . For this reason , the analysis that uses a single global tree shape produces fairly similar overall results to the window-based analysis , despite evidence that the window-based analysis fits the data better ( Figure S4 ) . A related concern is that due to variation across lineages in effective population size , the evolutionary rates of CNCs with different levels of constraint might not scale linearly across the trees [29] . However , our data show that this is a modest effect relative to the size of change needed to produce a significant rate change in a CNC ( Table S15 ) . In this study , we aimed to classify CNCs according to their evolutionary patterns . To do so , we used a modified version of the AIC to find the model that best describes the pattern of evolution of each CNC . In order to reduce the space of alternative models , we restrict our alternatives in two classes of models . As we obtain genome sequences for increasingly more species , it will be worth revisiting these models , as we will be better able to distinguish among different modes of evolution [40] . In particular , two natural models for rate changes in a CNC are ( 1 ) that the CNC has a one-time change in evolutionary pattern ( for example , a burst of adaptation to acquire a new function ) , or ( 2 ) that the CNC changes function or evolutionary constraint in a way that is inherited across all branches below . With larger numbers of taxa , it should be possible to gain better insight into the relevance of these two possible modes of evolution . More broadly , as we obtain increasing information about the functions of CNCs , we will increasingly be able to interpret the biological relevance of the patterns of rate changes detected here .
From the UCSC genome browser [30] , we downloaded the genome-wide multiple alignment of eight vertebrate species: human , hg17 ( May 2004 ) ; chimpanzee , panTro1 ( November 2003 ) ; dog , canFam1 ( July 2004 ) ; mouse , mm5 ( May 2004 ) ; rat , rn3 ( June 2003 ) ; chicken , galGal2 ( February 2004 ) ; fugu , fr1 ( August 2002 ) ; and zebrafish , danRev1 ( November 2003 ) . We also downloaded the annotation of “most conserved” regions defined on the same multiple alignment by a phylogenetic HMM in June 2005 [10] . The most conserved regions are defined without regard to whether the sequence is coding or noncoding , and cover around 4 . 3% of the human genome . To define CNCs , we first extracted those conserved regions from the multiple alignment and then processed them by removing coding regions ( exons in the “known gene” annotation , UCSC genome browser ) , repetitive sequences ( marked by lower case letters in the alignment ) , and sites that are gaps or missing data in any of the five mammalian genome sequences . Conserved regions of less than 100 bp after the processing were discarded . The remaining 231 , 285 regions out of the initial 1 , 451 , 896 most conserved regions comprised our raw dataset of CNCs and spanned ∼48 Mb . We used BLAT [44] to exclude spuriously aligned CNCs . We restricted our data to unique CNCs in which the human version of a CNC does not find any similar sequence ( >50% sequence identity ) elsewhere on the human genome . This resulted in discarding 24 , 234 CNCs ( ∼5 . 4 Mb ) . Furthermore , we required that each nonhuman mammalian verion of a CNC find the human version as the best match when it is BLATed against the whole human genome . This resulted in discarding an additional 74 , 359 CNCs ( ∼10 . 7 Mb ) . Our statistical inferences are based on alignments of the five mammalian sequences . However , we used the aligned chicken and fugu sequences to classify CNCs into different conservation level groups , of which we analyzed the two largest , denoted as “mammalian” and “amniotic” CNCs . Roughly speaking , a CNC was classified into the mammalian group if it is conserved across the mammalian genomes but not chicken or fugu , and into the amniotic group if it is conserved among the mammals and chicken but not fugu . The classification depended on ( 1 ) the presence or absence of aligned chicken and fugu sequences , and ( 2 ) the mean identity between mammals and chicken and fugu . The details are given in Text S1 . To examine the scale of local variation in tree shape , we estimated tree shapes over a chosen set of window sizes of 10 , 30 , 50 , or 100 consecutive CNCs ( ordered according to the human genome position ) . Since the scale of local variation seems to vary across chromosomes , there is no clear boundary explaining the rate of decay of autocorrelations . Nonetheless , incorporating such variation into our model is important , since otherwise , regions that show a general pattern of evolution that departs from the shared pattern from all CNCs might produce clusters of spurious signals . After several trials , we decided to estimate tree shape using a sliding window of 50 CNCs with an overlap of 34 CNCs between successive windows . With this window size , we obtained enough data to stably estimate tree shapes , but were also able to capture much of the local variation . The one third of CNCs located in the center of each window use the estimated tree shape from that window . In order to reduce the effect of outliers ( defined as having SRT > 25 with degrees of freedom of six ) , we estimate branch lengths in each window , drop outliers , and then reestimate the divergence times after dropping those nonneutral CNCs . Through this procedure , we expect that our estimates are robust in the presence of outlier CNCs . However , when rate changes are spatially clustered , our locally estimated tree shapes may absorb some of the signal of variable rates , hence potentially reducing power . Our preliminary analysis of classifying CNCs using the modified AIC showed that the number of CNCs with signals on the chimpanzee lineage was 48% larger than on the human lineage . Closer examination indicated that often CNCs with low-quality chimpanzee sequence ( PanTro1 ) produced a large signal of rate changes on the chimpanzee lineage , since miscalled bases would appear as mutations . Therefore , we dropped any CNC that is classified by AIC into the group showing rate changes on the chimpanzee lineage but that has low-quality chimpanzee sequence . To identify those CNCs , the chimpanzee sequence in each CNC was BLATed to the chimpanzee genome ( PanTro1 ) . The best match position ( according to the BLAT score ) was found when it was available . Then , in the target region , we counted the number of sites that have low quality score ( ≤20 ) . If this count was larger than 15 , we considered the CNC to have low-quality chimpanzee data . A total of 378 mammalian and 89 amniotic CNCs that were significant on the chimpanzee lineage were dropped for this reason . The impact of occasional sequence errors is likely to be much smaller for the other species . The human genome sequence has very high accuracy ( the estimated error rate is one site per 100 kb , much lower than the human polymorphism rate [45] ) . Meanwhile , occasional sequence errors in the other species should have only a small effect due to the much longer branches leading to those taxa . We estimated branch lengths for an alignment using the Felsenstein 84 sequence evolution model and using the empirical base frequencies . To make computation feasible , the “peeling” algorithm [46] was used with the assumption that sites evolve independently and that given their common ancestor , branches evolve independently . Details of the evolution model and the “peeling” algorithm were described by Felsenstein and Churchill [31] . Note that there are many more general evolutionary models , but the Felsenstein 84 model , which is essentially the same as the HKY85 model [47] , seems to be sufficient for the purposes of our study [48] . Under the null hypothesis , our parameters are a set of seven branch lengths shared by all CNCs , and one additional local substitution rate for each CNC . Under an alternative , our parameters are a set of lineage-specific rates that explain a specific scenario for each CNC . Rather than maximizing the likelihood directly , we developed an expectation-maximization algorithm ( EM ) that efficiently maximizes many parameters jointly under the null model . The details are given in Text S1 , but essentially , in our EM algorithm , each branch length is updated sequentially by computing the posterior number of substitutions on each branch and updating the related parameters accordingly . We find that our EM algorithm is stable to choices of initial starting points . The estimates that we obtain for simple models match well with those computed by Phylip [49] and PAML [50] . There are many possible models of CNC evolution , ranging from the simplest case , where there is a single rate across the entire tree , to the most extreme case , where each lineage evolves with its own rate . Here , we address how to classify CNCs according to their evolutionary patterns . Each of the possible alternative models corresponds to a partition of the seven lineages into two or more blocks of substitution rates . There are 876 ways of partitioning the seven branches into two or more different substitution rate groups . However , to reduce the space of possible models , we restrict ourselves to a subset of 127 candidate models that seem biologically most natural . Our main class of alternative models consists of the models where there are k selected branches ( 1 ≤ k ≤ 6 ) , each with its own rate parameter , while the remaining branches share a single background rate parameter . Such models have k + 1 parameters , and there are such models for k = 1 , … , 5 and one additional model for k = 6 . This accounts for 121 candidate models . In addition , we also consider a further set of six models that seem biologically natural , that split the branches on an unrooted tree into two or three rate groups using an internal branch ( connecting two internal nodes ) . Thus , for example , we might hypothesize a single rate-changing event in the ancestor of mouse and rat that leads to a single altered rate on both the mouse and rat branches . To reduce the model space complexity , we assume that such rate change events occur at internal nodes on the tree . These six models are summarized in Table S11 . Since there are many possible models , correct classification of the CNCs is likely to be difficult . Here , we view the classification as a multiple testing problem rather than a model selection problem , where our first goal is to control the rate of over-estimating the number of model parameters . The scheme below , though ad hoc , provides a reasonable compromise in providing fairly good model choice while not having excessive rates of “false positives . ” For each CNC , we select the model that , among the 127 candidate models produces the highest value of the penalized likelihood , which is log ( L ) − ( k + 1 ) + log{ ( 7 − k − 1 ) ! } for our main class of alternative models , where L is the maximum likelihood and k is the number of selected lineages . The first penalization term ( k + 1 ) penalizes for the number of estimated parameters and is introduced for the same reasoning as in the standard AIC . The last term ( log{ ( 7 − k − 1 ) ! } ) aims to account for the multiplicity of different models within each level . This latter term was suggested previously as a prior weight for Bayesian classification in an analogous setting [51] . This term is motivated by thinking of each model as corresponding to a partition of the seven branches into one or more blocks of substitution rate groups; as a natural choice of partition distribution we use the Ewens sampling distribution [52] with concentration parameter λ of 1 ( see Text S1 ) . To evaluate the performance of the classification , we simulated data under the full range of null and alternative models . We used these simulations to compare among three possible choices of penalty functions: ( 1 ) using only the number of parameters ( AIC ) , ( 2 ) using only the Ewens prior ( Ewens ) , and ( 3 ) using both the number of parameters and Ewens prior ( AIC + Ewens ) , as detailed above . The penalty function computed for each model is summarized in Table S14 and the power simulation results are shown in Figure S7 . The AIC + Ewens penalization provides the best control against over-fitting and that is what we use for our data analysis . There are 6 , 037 mammalian and 1 , 497 amniotic CNCs that show branch-specific accelerations on at least one lineage at significance level of 0 . 001 ( based on the asymptotic distribution of SRTi ) . To see if these CNCs actually exceed the neutral rate on a particular branch showing speed-ups , we estimate the local “neutral” tree near each of those CNCs . Specifically , we take the surrounding 10 kb with each such CNC at the center , then exclude “most conserved” regions as well as exons to construct putatively neutral local regions . The genome-wide average of each branch length on trees estimated from those regions ( shown in Table S3 ) is very similar to that from CFTR nonexonic DNA region in Cooper et al . [15] . It is notable that the variation in neutral branch lengths of the dog lineage is considerably larger than that on other lineages ( Figure S8 ) . To test whether the accelerated rate exceeds the neutral rate on a particular branch i , we compute a likelihood ratio statistic testing the hypothesis H0: r0 , −i , ri = ri , neutral versus HA:r0 , −i , ri > ri , neutral only for those CNCs that have a substitution rate on the tested lineage ri that exceeds the neutral rate ri , neutral in the surrounding region . Based on the chi-square distribution with one degree of freedom , we compute p-values and reject the null hypothesis if p-value < 0 . 05 . Our results are summarized in Table 1 . We performed simulation studies with the Felsenstein 84 sequence evolution model using evolver13 , implemented in PAML [50] , to assess the distributions of the SRT and SRTi statistics under the null model and to evaluate p-values when the distribution is not well approximated by the asymptotic theory . We simulated two sets of 1 million CNCs under the null , one set for the mammalian and one for the amniotic CNCs , matching the distribution of base frequencies , the CNC size , the variation in the local substitution rates , and the overall tree shape . Specifically , each of the 1 million simulated CNCs was based on the characteristics of a randomly sampled CNC in our dataset ( sampling with replacement ) . We specified the branch lengths by rescaling the global tree estimated from all CNCs by the local substitution rate of the chosen CNC , and generated an alignment of five sequences of the corresponding size simulated on the specified tree . For each set of simulated CNCs , we obtained the empirical null distributions of statistics testing for various alternative scenarios ( the simplest and most extreme cases are shown in Figure S3 ) . As mentioned earlier , the asymptotic theory works reasonably well , except when testing for rate changes on short branches ( i . e . , the human and chimpanzee lineages ) . We also grouped sets of CNCs into bins with similar CNC sizes and local substitution rates and examined empirical distributions for each bin separately ( unpublished data ) . For each test statistic , the empirical distributions across bins were homogeneous , except for the cases in which the asymptotic theory does not work because of the small number of accumulated substitutions . For these cases , since the inhomogeneity across bins was mainly explained by differences in the expected number of substitutions on the tested lineage , we reconstructed bins of CNCs according to the number of expected substitutions and evaluated p-values within each bin separately . We also simulated a number of CNC sets under various alternative scenarios to examine the performance of the modified AIC method . Specifically , for each k selected lineages ( k = 0 , … , 6 ) , we simulated 100 , 000 CNCs in which each CNC has k branches that evolve with their own rates . These rates were higher or lower than the background rate with 50% probability each . To incorporate the variation in strength of signals in real data , the rate of each selected branch was simulated by multiplying or dividing the background rate by a scale factor that is drawn from 1 + Γ ( α , β ) distribution with a scale parameter β = 1 and a shape parameter α = 1 ( weak signals ) or α = 2 ( stronger signals ) . We downloaded a reference assembly ( seq_gene . md . gz ) that corresponds to the human genome build ( NCBI build 35 ) from ftp://ftp . ncbi . nih . gov/genomes/H_sapiens/ARCHIVE/BUILD . 35 . 1/mapview . We proceeded by extracting only reference genes ( release 3 [45] ) in autosomes only , and obtained 25 , 249 genes . We extracted the nearest gene for each CNC without considering gene orientation . Here the distance from a CNC to a gene is the minimum of distances from the middle of the CNC to either end of the gene . The PANTHER GO database was downloaded from http://www . pantherdb . org/panther/prowler . jsp in March 2006 . For each conservation group , we examined what kinds of biological process categories are enriched for being: ( 1 ) near CNCs in general and ( 2 ) near CNCs showing lineage-specific rate increases compared to near CNCs evolving under the null model . The nearest genes of CNCs were used for this analysis . For ( 1 ) , we compiled the list of all genes in the reference assembly and the list of genes near mammalian ( or amniotic ) CNCs . For each biological process category , we counted the number of genes in each list and compared them with a chi-square test . Within each list , genes are counted only once . For ( 2 ) , CNCs were first classified by the AIC ( in order to obtain disjoint categories of CNCs showing signals of speedups on each lineage ) . For each category , we counted genes near CNCs under the null and near CNCs in each of the seven selection groups that show rate speed-ups on a single lineage . In this case , however , individual genes were counted repeatedly each time they were the nearest neighbor of a relevant CNC . The reason for this is that multiple CNCs often have the same nearest neighbor . This effect is more pronounced in the null CNC group than in the selection groups . Consequently , if we count genes only once , then any biological functional category that is enriched near CNCs , in general , may be underrepresented in the null but overrepresented in each selection group . Since the numbers of selected CNCs are small for many gene categories , p-values were computed using Fisher's exact test . To account for multiple testing , the p-values were multiplied by the number of biological processes that were jointly tested . We have prepared a datafile that contains the list of all CNCs and summarizes our analysis results . It includes genomic properties and test statistics , as well as the best evolutionary pattern of each CNC . It will be downloadable from http://pritch . bsd . uchicago . edu/data . html .
The National Center for Biotechnology Information ( NCBI ) Entrez ( http://www . ncbi . nlm . nih . gov/gquery/gquery . fcgi ) gene accession numbers for the genes C21orf54 and IFNAR2 are GeneID:339629 and GeneID:3455 , respectively . | Conservation of DNA sequences across evolutionary history is a highly informative signal for identifying regions with important biological functions . In particular , conserved noncoding regions have been shown to be good candidates for containing regulatory elements that have roles in gene regulation . Recent studies have found that there are many thousands of conserved noncoding elements ( CNCs ) in vertebrate genomes and have suggested possible functions for some of these elements , but the function of most CNCs remains unknown . To study the evolution of CNCs , we developed a statistical method to identify CNCs that show changes in evolutionary rates on particular branches of the mammalian phylogenetic tree . Those rate changes may indicate changes in the function of a CNC . We applied our method to CNCs of five mammalian genomes , and found that , indeed , many CNCs have experienced rate changes during their evolution . We also found a subset of CNCs showing accelerations in evolutionary rate that actually exceed the neutral rates , suggesting that adaptive evolution has shaped the evolution of those elements . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"Supporting",
"Information"
]
| [
"evolutionary",
"biology",
"genetics",
"and",
"genomics",
"mammals",
"computational",
"biology"
]
| 2007 | Adaptive Evolution of Conserved Noncoding Elements in Mammals |
We have experimentally and computationally defined a set of genes that form a conserved metabolic module in the α-proteobacterium Caulobacter crescentus and used this module to illustrate a schema for the propagation of pathway-level annotation across bacterial genera . Applying comprehensive forward and reverse genetic methods and genome-wide transcriptional analysis , we ( 1 ) confirmed the presence of genes involved in catabolism of the abundant environmental sugar myo-inositol , ( 2 ) defined an operon encoding an ABC-family myo-inositol transmembrane transporter , and ( 3 ) identified a novel myo-inositol regulator protein and cis-acting regulatory motif that control expression of genes in this metabolic module . Despite being encoded from non-contiguous loci on the C . crescentus chromosome , these myo-inositol catabolic enzymes and transporter proteins form a tightly linked functional group in a computationally inferred network of protein associations . Primary sequence comparison was not sufficient to confidently extend annotation of all components of this novel metabolic module to related bacterial genera . Consequently , we implemented the Graemlin multiple-network alignment algorithm to generate cross-species predictions of genes involved in myo-inositol transport and catabolism in other α-proteobacteria . Although the chromosomal organization of genes in this functional module varied between species , the upstream regions of genes in this aligned network were enriched for the same palindromic cis-regulatory motif identified experimentally in C . crescentus . Transposon disruption of the operon encoding the computationally predicted ABC myo-inositol transporter of Sinorhizobium meliloti abolished growth on myo-inositol as the sole carbon source , confirming our cross-genera functional prediction . Thus , we have defined regulatory , transport , and catabolic genes and a cis-acting regulatory sequence that form a conserved module required for myo-inositol metabolism in select α-proteobacteria . Moreover , this study describes a forward validation of gene-network alignment , and illustrates a strategy for reliably transferring pathway-level annotation across bacterial species .
Inositol , or cyclohexanehexol , is one of the most abundant carbohydrates in freshwater and terrestrial ecosystems [1] . Phosphorylated and lipidated derivatives of inositol serve as important signaling molecules in eukaryotic cells and are critical components of cellular membranes . Among prokaryotes , several species of cyanobacteria , eubacteria and archaea are able to synthesize and derivitize inositol [2] . These molecules serve functional roles as antioxidants , osmolytes , cell membrane components , and as carbon storage substrates [3] , [4] . Inositol can also serve as the sole carbon and energy source for many bacterial species [5]–[9] and , in its phosphorylated forms , as a source of phosphorus [10] . Thus inositol is an important biomolecule that is involved in multiple aspects of eukaryotic and prokaryotic cellular physiology and is also a critical nutrient and energy source positioned at the intersection of environmental carbon and phosphorus cycles [1] . While cells can derivitize inositol into many different chemical species , the unmodified myo form of inositol ( cis-1 , 2 , 3 , 5-trans-4 , 6-cyclohexanehexol ) is among the most abundant species in the environment [1] . The myo-inositol degradation pathway has been characterized biochemically in Klebsiella aerogenes [5] , [11]–[13] and Bacillus subtilis [14] . In this pathway , seven proteins convert myo-inositol to CO2 , acetyl CoA and dihydroxy-acetone phosphate ( Figure 1 ) . Structural and regulatory genes required for myo-inositol catabolism have been identified and characterized in several gram-positive species , including B . subtilis [7] , [14] , Clostridium perfringens [8] , Corynebacterium glutamicum [9] , and Lactobacillus casei [15] , and in the gram-negative bacteria Rhizobium leguminosarum bv . viciae [6] , [16] , Sinorhizobium meliloti [3] and Sinorhizobium fredii [17] . Gram positives generally exhibit complete and contiguous catabolic operons that are adjacent to genes encoding myo-inositol transporters of the major facilitator superfamily; expression of these genes is controlled by transcriptional regulators of the DeoR or LacI families [7]–[9] , [18] . Among the gram negatives , genes involved in myo-inositol metabolism are more dispersed across the chromosome . This lack of chromosomal co-location and the difficulty in assigning function to transporter and regulatory proteins using sequence homology alone [19] , [20] has made comprehensive identification of the myo-inositol genetic modules more difficult in these species . In this study we have defined the structural and regulatory components of a genetic module controlling myo-inositol transport and catabolism in the gram-negative α-proteobacterium Caulobacter crescentus , and reliably extended this experimental functional annotation to other bacterial genera using a combination of computational network prediction and alignment methods .
C . crescentus strain CB15N ( NA1000 ) [21] and strains derived from it were grown in peptone/yeast extract ( PYE ) or M2 minimal broth [22] . Minimal broth was supplemented with either 0 . 2% ( w/v ) myo-inositol ( M2I ) or 0 . 2% ( w/v ) glucose ( M2G ) . Directed deletion strains were constructed by ligating approximately 500 base pair regions flanking the 5′ and 3′ regions of the gene to be deleted into the suicide plasmid pNPTS138 ( see Table 1 ) using the EcoRI and HindIII restriction sites . pNPTS138 carries the nptI gene to select for single integrants on kanamycin and the sacB gene for counterselection on sucrose . The pNPTS138-derived deletion plasmids were transformed into CB15N by electroporation . Initial selection was on 25 µg/ml kanamycin , which was followed by overnight growth in nonselective media and then plating on 3% sucrose to select for cells that had undergone a second crossover event to excise the gene . PCR was used to confirm chromosomal deletions . Cloned fragments to generate pNPTS138 deletion plasmids had the following chromosomal coordinates: ibpA upstream = 955 , 261–956 , 039; ibpA downstream = 956 , 772–957 , 359 . iatA upstream = 956 , 357–956 , 940; iatA downstream = 958 , 429–959 , 014 . iatP upstream = 957 , 949–958 , 539; iatP downstream = 959 , 410–960 , 005 . iolR upstream = 1 , 442 , 941–1 , 443 , 408; iolR downstream = 1 , 444 , 244–1 , 444 , 741 . All gene deletions were in-frame . The deletion of iolR ( CC1297 ) left the first and last 6 codons intact . The deletion of ibpA ( CC0859 ) left the first 45 and last 38 codons intact . The deletion of iatA ( CC0860 ) left the first 12 and last 9 codons intact . The deletion of iatP ( CC0861 ) left the first 30 and last 13 codons intact . A transporter complementation plasmid was generated by cloning the full transporter locus plus promoter region into the KpnI and NdeI sites of the replicating plasmid pMT630 [23]; cloned chromosomal coordinates = 955 , 261–960 , 056 . Sinorhizobium meliloti Rm2011 and strains derived from it were obtained from the lab of Anke Becker ( Bielefeld University , Germany ) [24] . S . meliloti was grown in either LB or GTS minimal medium [25] supplemented with 0 . 2% glucose ( GTS-G ) or 0 . 2% inositol ( GTS-I ) as the sole carbon source . All strains used in this study are listed in Table 1 . A library of ≈16 , 000 individual C . crescentus CB15N mutant strains carrying either the Mariner–based Himar-1 transposon [26] , the Mu-based HyperMu transposon ( Epicentre , Madison , WI ) , or the Tn5-derived EZ-Tn5 transposon ( Epicentre , Madison , WI ) was generated and stored in 96-well format ( Pritchard , Matteson and Viollier , unpublished ) . This transposon library was replica stamped from the 96-well plates onto M2 agar supplemented with either 0 . 2% ( w/v ) myo-inositol ( M2I ) , 0 . 1% cellobiose ( M2C ) or 0 . 2% glucose ( M2G ) . Strains that grew on M2C and M2G , but not M2I were considered to have inositol-conditional mutations . Mapping of the transposon insertion site in C . crescentus Himar-1 , Hyper Mu , and Ez-Tn5 mutant strains deficient for growth on M2I was determined by isolating chromosomal DNA , digesting with HinPI for 10 minutes at 37°C , ligating the digested genomic fragments into circles using T4 DNA ligase , and transforming 1 µl of this ligation reaction into electrocompetent E . coli EC100D pir-116 cells ( Epicentre , Madison , WI ) . These transposons all carry an R6K origin that replicates in a pir+ strain of E . coli . The circularized transposon plasmids were then isolated from E . coli and silica-column purified ( Novagen , Madison , WI ) . The location of the transposon insertion was determined via a single primer sequencing extension reaction from the purified , circularized transposon plasmids . The oligos used to map these transposons are as follows: Himar-1-GATATTGCTGAAGAGCTTGGCGGCGAA; Ez-Tn5- CTACCCTGTGGAACACCTACATCT; Hyper-Mu-AGAGATTTTGAGACAGGATCCG . Wild type C . crescentus CB15N cells were grown in either M2 minimal medium supplemented with 0 . 2% ( w/v ) glucose ( M2G ) or M2 supplemented with 0 . 2% myo-inositol ( M2I ) to OD660 of 0 . 3–0 . 4 . 5 ml of 4 replicate cultures ( for each carbon condition ) were spun down at 10 , 000× g for 30 seconds , the supernatant was removed , and the cell pellets were flash frozen in liquid nitrogen . RNA was isolated from these cells by incubating in 1 ml of Trizol ( Invitrogen , Carlsbad , CA ) at 65°C for 10 minutes , adding chloroform , vortexing , spinning , and extracting the aqueous layer . Nucleic acid in the aqueous layer was isopropanol precipitated overnight at −80°C followed by a 30 minute centrifugation at 16 , 000× g . The ethanol-washed and air-dried nucleic acid pellet was resuspended in 50 µl of nuclease-free water ( IDT , Coralville , IA ) . 1 µl of RNase-free DNase I ( Ambion , Austin , TX ) was added to the sample and incubated at room temperature for two hours to remove any residual DNA . The nucleic acid in this digested sample was then acid phenol-chloroform ( Ambion , Austin , TX ) extracted , ethanol precipitated at −80°C overnight , and centrifuged at 16 , 000× g to produce a DNA-free RNA pellet . RNA quality was assessed via agarose gel electrophoresis and RNA concentration determined by UV spectrophotometry using a Shimadzu UV-1650 ( Kyoto , Japan ) . Labeled indodicarbocyanine-dCTP ( Cy3 ) and indocarbocyanine-dCTP ( Cy5 ) cDNA was generated from 20 µg of total RNA by reverse transcription with Superscript II reverse transcriptase ( Invitrogen , Carlsbad , CA ) using 1 µg of random hexamer primers ( Invitrogen , Carlsbad , CA ) . 2 samples of cDNA from each RNA type ( M2I and M2G ) were Cy3 labeled and 2 were Cy5 labeled . Dye-swapped cDNA from the remaining two samples was generated in order to minimize dye bias in the microarray analysis . Paired Cy3 and Cy5 labeled cDNA from the M2G and M2I samples were hybridized onto spotted DNA oligo arrays using a protocol previously described [27] . After hybridization and washing , the arrays were scanned with a GenePix 4000B scanner ( Axon Instruments ) . Scanned spots were converted to ratios ( red/green ) with GenePix Pro 5 . 0 software . Expression ratio data ( glucose/inositol ) for the four biological replicates were normalized by median centering and analyzed using the Significance Analysis for Microarrays ( SAM ) package [28] . Genes that showed a 2-fold or greater mean expression change ( either up or down in myo-inositol relative to glucose ) and that were determined to be significant in SAM using a 5% false discovery cutoff are included in Table S1 . DNA microarray data have been deposited in the Gene Expression Omnibus ( GEO ) database ( http://ncbi . nlm . nih . gov/geo ) under accession number GSE12414 . To measure the promoter activities of inositol-regulated genes/operons , we first PCR-amplified promoter regions of three genes . The idhA promoter region extends from chromosomal coordinate 1 , 443 , 231 to 1 , 443 , 769; the iolC promoter region extends from coordinate 1 , 443 , 840 to 1 , 444 , 387; the ibpA promoter extends from coordinate 955 , 627 to 955 , 895 . These fragments were digested and cloned into the reporter plasmid pRKlac290 [29] using the EcoRI and HindIII sites . The resulting promoter-lacZ transcriptional fusion plasmids were introduced into C . crescentus CB15N or CB15NΔiolR by tri-parental conjugation using the E . coli helper strain FC3 , which carries the pRK600 plasmid [30] . β-galactosidase activity from the LacZ-promoter reporter strains was determined colorimetrically using cells in log phase ( 0 . 1–0 . 3 OD660 ) at 30°C; Z-buffer ( 60 mM Na2HPO4 , 60 mM NaH2PO4 , 10 mM KCl , 1 mM MgSO4 ) and an excess of o-nitrophenyl-β-D-galactopyranoside was added to chloroform-permeabilized cells and absorbance was measured at 420 nm on a Spectronic Genesys 20 Spectrophotometer ( ThermoFisher Scientific , Waltham , MA ) . Two palindromic consensus motifs in the iolC promoter - that we also identified upstream of several myo-inositol-regulated genes - were mutated by PCR using mismatched oligos . The site-directed mutagenesis PCR was followed by 1 hour of DpnI digestion , and 1 µL of the digested reaction was transformed into electrocompetent E . coli TOP10 cells ( Invitrogen , Carlsbad , CA ) . pCR-BluntII plasmids ( Invitrogen , Carlsbad , CA ) containing the mutant iolC promoters were amplified and purified , and the mutated iolC promoters were excised with EcoRI and HindIII , and sub-cloned into pRKLac290 to generate mutated PiolC-lacZ transcriptional fusions . The motif that is positioned between 104 and 119 bases upstream of the predicted start codon of iolC was mutated from TGGACCATATGTTCCA to TGTACCATATGTACAA . The motif positioned between 45 and 60 bases upstream of the predicted iolC start codon was mutated from TGGAATATGCGTTACA to TTGCATATGCGGTACA . Cell growth in different media types was measured in triplicate in bulk culture grown in 13 mm glass tubes in an Infors tube shaker ( ATR Biotech , Laurel , MD ) , at 30°C , 220 rpm . Density measurements for individual cultures were taken hourly up to 0 . 3 OD660 in a Genesys20 Spectrophotometer ( ThermoFisher Scientific , Waltham , MA ) and the growth rate was determined by fitting the data to an exponential growth equation:in Prism ( GraphPad Software , San Diego ) , where y0 is the initial cell density , k is growth rate , and t is time . For each of 305 sequenced prokaryotic genomes , we assembled a battery of different predictors of protein association including coexpression , coinheritance , colocation , and coevolution . We formulated the network integration problem as a binary classifier , where the goal is to distinguish functionally linked protein pairs ( L = 1 ) from non-interacting pairs ( L = 0 ) . In this formulation , a vector of interaction predictors is the input to a binary classifier function , which returns the integrated probability that two proteins are functionally linked . To calculate the mapping between raw interaction data and integrated probabilities , the classifier function is trained on a set of known interactions . Applying this classifier to predict interaction probabilities for all protein pairs in a genome yields a probabilistic protein interaction network . Specifically , we generated the training set of known interactions by using KEGG [31] classifications of individual proteins to produce an annotation of protein pairs . For each pair we recorded if the proteins had overlapping annotations ( L = 1 ) , if both were in entirely non-overlapping KEGG categories ( L = 0 ) , or if either protein lacked an annotation code or was marked as unknown ( L = ? ) . We also calculated four functional genomic and experimental predictors: 1 ) coexpression; the Pearson correlation between genes in publically-available DNA microarray expression data , 2 ) coinheritance; the Pearson correlation between protein phylogenetic profiles [32] , 3 ) coevolution; the Pearson correlation between protein distance matrices , taken elementwise [33] , and 4 ) collocation; the average chromosomal distance between ORFs . Each of these predictors is defined on a pair of proteins rather than an individual protein and can be arranged in a four dimensional vector:It can be shown empirically that the distribution of functionally linked protein pairs is shifted relative to the distribution of functionally unlinked pairs [34] . Intuitively , this means that each genomic evidence type is a predictor of protein functional interaction . We can combine these predictors to obtain the integrated probability of protein interaction via Bayes' rule [35] . In practice , the quotient formula for the Bayesian posterior probability is quite sensitive to fluctuations in the denominator . To deal with this , we used bootstrap aggregation [36] to smooth the posterior as follows:where M is the number of bootstrap replicates . Thus , for each pair of proteins , we have a value P ( L = 1|E1 , E2 , E3 , E4 ) which represents the integrated probability of protein interaction over several data types . Additional computational details underlying this protein network prediction strategy are discussed in Srinivasan , et al . [34] . A web interface for this functional networking database containing predicted networks for 305 bacterial species is available at http://networks . stanford . edu . Network alignment is a systems-biology analog of sequence alignment that compares protein association networks between different species in an effort to identify conserved functional modules . Such modules are sets of proteins that have both conserved primary sequences and conserved pairwise statistical associations between species . For automated network alignment , we used the experimentally- and computationally-defined myo-inositol network from C . crescentus as a query module . This module was used to conduct query-to-network alignment searches across computationally-predicted protein interaction networks of 5 related α-proteobacterial species [34]; these interaction networks had been previously defined using the statistical protein network prediction strategy outlined above . The bacterial species included in this alignment were Sinorhizobium meliloti , Mesorhizobium loti , Brucella melitensis , Agrobacterium tumefaciens , and Bradyrhizobium japonicum . Initial alignment identified the best match to the query in each protein interaction network . Specifically , we used the Graemlin algorithm [37] to perform automated cross-species alignment . Graemlin incorporates ideas from sequence alignment to perform query-to-network alignment accurately and efficiently . To search multiple networks for matches to a query module , Graemlin first aligns the query module to the evolutionarily closest network by identifying a high scoring pair of proteins within the query and network and aligning them . Then , Graemlin extends the alignment by aligning the pair of proteins that will increase the score of the alignment the most , continuing until it cannot further increase the score of the alignment . The score for aligning a pair of proteins is higher when the proteins are 1 ) sequence similar and 2 ) connected to many proteins in the current alignment . Once Graemlin aligns the query module to the evolutionarily closest ( i . e . highest scoring ) network , it aligns the resulting alignment to the next evolutionarily closest network . To perform this alignment it uses the same algorithm that it uses to perform the first alignment , with an adjusted scoring function [37] . Graemlin continues performing alignments in this fashion until it has aligned the query to every network . To date , Graemlin is the only algorithm capable of aligning a query module to more than three networks . Our benchmarks have shown that when aligning a query module to a single network , this method of alignment is more accurate and efficient than existing network alignment algorithms [37] . To improve the predictive power of the alignment , we manually refined the alignment to keep the best candidates in each species using the following criteria: 1 ) in each species , we considered only transporter operon candidates in which the three ABC transporter components were contiguous on the chromosome; this resulted in several candidate conserved operons in each species , 2 ) in each species , we assessed the similarity of each candidate operon to those in all other species in the alignment . We then calculated , for each protein in the candidate operon , the average BLAST significance score to its predicted counterpart in all other species; the candidate operon with the best average significance score ( i . e . lowest average p-value ) was selected for inclusion in the final cross-species module . Additional computational details underlying this protein network prediction strategy are discussed in Flannick et al . [37] . The network alignment tool Graemlin 2 . 0 is available under the GNU public license at http://graemlin . stanford . edu . We used MEME [38] to locate putative regulatory motifs in the upstream regions of genes in the C . crescentus myo-inositol module . In order to refine this motif , and also to investigate its conservation in other species , we used MEME to search 250 base pairs upstream of the predicted translation start sites of genes in the predicted inositol modules in each of the species present in our multi-species network alignment . The MEME search parameters were as follows: motif distribution , 0–1 per sequence; minimum motif width , 6; maximum motif width , 50 .
Using an arrayed library of ≈16 , 000 mutant C . crescentus strains carrying transposon insertions , we conducted a forward genetic screen for mutants that could not grow on myo-inositol as the sole carbon source . Three strains , FC354 , FC362 and FC536 , were discovered that were unable to grow on M2-myo-inositol medium ( M2I ) but exhibited normal growth on PYE , M2-cellobiose ( M2C ) and M2-glucose ( M2G ) . Strain FC536 has a transposon insertion in the myo-inositol 2-dehydrogenase ( idhA; CC1296 , NP_420109 ) gene . The IdhA homolog from B . subtilis has been characterized biochemically [39] , and is known to catalyze the first dehydrogenation reaction in the myo-inositol degradation pathway ( Figure 1 ) . Strain FC362 contains a transposon insertion in the iolD gene ( CC1299 , NP_420112 ) . IolD has also been characterized in B . subtilis where it was shown to catalyze hydrolysis and ring opening of the catabolic intermediate D-2 , 3-diketo-4-deoxy-epi-inositol to form 5-dehydro-2-deoxy-D-gluconate [14] . The transposon insertion in strain FC362 likely disrupts expression of not only iolD , but also genes downstream of iolD in the operon encoding other known myo-inositol catabolic enzymes ( Figure 2C and Figure 1 ) . The third strain identified in our screen , FC354 , contained a transposon that mapped to CC0860 , a gene encoding a ATPase protein in an operon predicted to encode an ATP-binding cassette ( ABC ) sugar transporter ( Figure 2B ) . This transporter operon is physically separated on the chromosome from the genes encoding the catabolic enzymes by ≈500 kilobases ( Figure 2 ) . ABC sugar transporters are inner-membrane transporters that employ three components - a periplasmic sugar binding protein , a transmembrane permease and a cytoplasmic ATPase - to move sugars from the periplasm to the cytoplasm [40] . To confirm that this transporter operon , CC0859–CC0861 , is required for growth on myo-inositol , we constructed strains with in-frame deletions of each of these genes individually: C . crescentus strains CB15NΔibpA ( CC0859 , inositol binding protein , NP_419676 ) , CB15NΔiatA ( CC0860 , inositol ABC transporter ATPase , NP_419677 ) , and CB15NΔiatP ( CC0861 , inositol ABC transporter permease , NP_419678 ) . Individual in-frame deletions of each of these genes abolished growth on defined medium containing myo-inositol as the sole carbon source , but not on defined minimal glucose medium ( Table 2 ) or PYE complex medium . Growth on myo-inositol in the individual in-frame transporter deletion strains was restored by complementation with a replicating vector carrying the entire ibpA-iatA-iatP locus under the control of its own promoter ( Table 2 ) . The inability of C . crescentus strains lacking any gene in the ibpA-iatA-iatP operon to grow in myo-inositol demonstrates that this operon encodes the only inner-membrane myo-inositol transporter in C . crescentus . Whole-genome transcriptional profiling using DNA microarrays was conducted to identify genes with differential regulation in myo-inositol relative to glucose as the sole carbon source . 50 genes were found to have transcript levels that were at least 2-fold higher in cells grown in myo-inositol than in glucose ( see Materials and Methods for data analysis parameters ) ( Figure 2A and Table S1 ) . Among these genes , as expected , are the catabolic genes idhA and the iolECBDA operon , as well as the gene , ibpA , encloding the periplasmic binding protein of the myo-inositol ABC transporter . The most highly induced gene in myo-inositol relative to glucose ( >4-fold ) is isocitrate lysase ( CC1764 , NP_420572 ) which catalyzes formation of glyoxylate and succinate from isocitrate . This result suggests that growth of C . crescentus on myo-inositol shifts energy metabolism toward the glyoxylate cycle relative to growth on glucose . The ATPase subunit of a HlyB-family ABC-transporter ( gene CC1314 , NP_420127 ) is also four-fold more abundant in myo-inositol than in glucose ( Table S1 ) . As discussed above , cells with mutations in the ibpA-iatA-iatP transporter operon fail to grow on myo-inositol as the sole carbon source after one week of incubation ( Table 2 ) providing evidence that this HlyB-family transporter is not a redundant myo-inositol transporter . However , this transporter may be involved in transporting derivatized versions of inositol ( e . g . inositol phosphates or lipidated inositols ) . The gene CC1297 ( NP_420110 ) is annotated as an RpiR-family transcriptional regulator and encodes a putative SIS ( Sugar ISomerase; Pfam 01380 ) domain at its N-terminus . Based on its predicted function as a sugar-binding transcription factor and its chromosomal location adjacent to the iol catabolic operon ( Figure 2C ) , we predicted that CC1297 would regulate transcription of the iol genes . To test this hypothesis , we constructed a strain with an in-frame deletion of this gene and measured expression from the idhA , iolC ( NP_420111 ) , and ibpA promoters in wild type and CC1297 deletion strains using promoter-lacZ fusions as transcriptional reporters . These assays revealed significant derepression of transcription from the idhA , ibpA and iolC promoters in a CC1297 deletion background when cells were grown in PYE complex medium ( Student's t-test; p<0 . 0001 ) ( Figure 3A ) . This result is consistent with the idea that CC1297 is a transcriptional regulator of the iol genes . As such , we have named this gene iolR . Notably , C . crescentus IolR is not homologous to the IolR proteins previously described in Bacillus subtilis , Corynebacterium glutamicum or Clostridium perfringens [8] , [9] , [18] and thus defines a new class of myo-inositol regulator proteins . In contrast with these unrelated myo-inositol regulator genes , which are induced by myo-inositol [7] , [9] , expression of C . crescentus iolR is not regulated by myo-inositol based on our microarray transcriptional profiling data . We then sought to identify possible regulatory motifs in the predicted promoter regions of genes in the myo-inositol metabolic module of C . crescentus . A MEME search [38] of the DNA sequence of these promoters suggested a consensus palindromic motif , GGAANATNCGTTCCA that is present upstream of ibpA , idhA , iolC and iolA ( NP_420115 ) ( Figures 2B and C & Figure 4 ) . The iolC promoter contains two copies of this motif with MEME e-values less than 10−8 ( Figure 4 ) and with good conservation of the palindrome . Motif 1 is 104 bp upstream of the predicted translation start site of iolC , while motif 2 is 45 bp upstream ( Figure 3C ) . We mutated each of these two motifs away from consensus ( Figure 3C ) , and measured expression from these mutant iolC promoters in complex medium ( PYE ) . Mutation of motif 1 results in significantly higher LacZ activity than the wild-type promoter ( Student's t-test; p<0 . 001 ) , demonstrating that motif 1 is involved in basal repression of iolC expression . Mutation of motif 2 does not affect measured promoter activity in PYE ( p>0 . 05 ) ( Figure 3B ) . These results demonstrate that , in the case of motif 1 , the palindromic sequence we have identified in the promoters of genes required for myo-inositol metabolism is a functionally relevant regulator of gene expression . Future analysis of the regulatory role of IolR , of motif 2 in the iolC promoter , and of the palindromic motifs in the idhA and ibpA promoters promises to provide insight into additional layers of regulation in this genetic module . Independent of our experimental work , we applied statistical methods that we previously developed to predict functional associations between genes in prokaryotic genomes ( see Materials and Methods and [34] ) . Figure 5 shows the computationally-predicted “myo-inositol module” of C . crescentus . This is a subset of our whole-genome C . crescentus integrated protein association network , containing proteins encoded in operons at just two distinct chromosomal loci . The first chromosomal locus contains genes ( C . crescentus gene numbers CC1296; CC1298–CC1302 ) that are predicted to be involved in catabolism of myo-inositol by sequence homology to known enzymes involved in myo-inositol catabolism [5] , [11]–[13] . The second locus is an operon containing genes ( CC0859–CC0861 ) that are predicted to encode the three components of a canonical ABC transmembrane sugar transporter [41]: a periplasmic sugar-binding protein , an ATPase subunit , and a transmembrane permease . However , the periplasmic sugar-binding protein of this transporter is only generally annotated as a member of the XylF superfamily in the Conserved Domain Database ( CDD score<e−15 ) [42] , and its true substrate was not known at the time we constructed our microbial protein association networks . The C . crescentus inositol module also contains the gene , CC1297 ( iolR ) , which is colocated with the predicted myo-inositol catabolic genes and is annotated as encoding a transcriptional regulator . We found that the transporter and catabolic proteins have strong intra-operon linkage ( >80% confidence ) , which is largely due to high colocation and coinheritance scores ( Figure 5 ) . The inter-operon association between the transporter , catabolic , and regulatory proteins , which are encoded from genes at two disparate chromosomal loci , primarily arises from moderate statistical correlations contained within the microarray coexpression component of our model . Using a 30% confidence cutoff , we deduce that the periplasmic sugar-binding protein CC0859 ( IbpA ) is functionally linked to several genes in the predicted myo-inositol catabolic operon ( Figure 5 ) . No other transmembrane transporters in the C . crescentus genome are predicted to associate with the myo-inositol catabolic genes in our network . This linkage between the myo-inositol catabolic proteins and the ABC sugar transporter is missed using a single association metric such as colocation , coinheritance or coexpression alone . An integrative statistical model , which incorporates multiple predictors of association , is required to identify this association . As discussed above , genetic and molecular experiments have confirmed the computationally-predicted association between the ABC sugar transporter and the myo-inositol catabolic genes . Specifically , we have shown that 1 ) proteins encoded by the transporter operon CC0859–CC0861 ( now annotated as IbpA , IatA , and IatP ) function to form the sole myo-inositol inner-membrane transporter in C . crescentus , 2 ) transposon disruption of the predicted catabolic locus encompassing CC1296; CC1298–CC1302 ( annotated as IdhA , IolC , IolD , IolE , IolB , IolA ) abolishes growth of C . crescentus on myo-inositol as the sole carbon source , 3 ) the transcriptional regulator gene CC1297 ( annotated as IolR ) functions to regulate expression of the myo-inositol transporter and catabolic genes . Using automated cross-species alignment in combination with manual post-refinement ( see Materials and Methods ) we identified genetic networks in other bacterial species that we predicted to be functionally homologous to the C . crescentus myo-inositol network ( Figure 6 ) . The cross-species alignment conducted in this study indicates significant conservation of the catabolic , regulatory , and transporter proteins across the six α-proteobacterial species aligned . In addition , there are conserved cross-protein functional linkages within each of these species ( Figure 6A ) . Linkage between the transporter and catabolic proteins is particularly strong in M . loti and S . meliloti as evidenced by the large number of association edges between transporter , catabolic , and regulatory genes in these species ( Figure 6A ) . The module is least conserved in B . japonicum , which is discussed further below . As discussed above , we discovered a palindromic motif ( GGAA-N6-TTCC ) with a moderate MEME e-value upstream of several genes in the C . crescentus inositol module ( Figure 4 ) . By reasoning that conservation at the gene system level may imply conservation at the level of gene regulation , we searched 250 bases upstream of the predicted translational start sites of genes in our cross-species network alignment for sequences related to the palindromic motif identified in C . crescentus ( Figure 4 ) . In these related sequences , we found 21 more examples of this same motif , which was particularly enriched in the predicted upstream homologs of iolC , idhA , and the myo-inositol ABC transporter operons in these species ( Figure 6B and 7 ) . Incorporating the upstream sequences from all species in the Graemlin alignment in a MEME motif search dramatically improved the significance score for this regulatory motif ( Figure 6B ) . Notably , B . japonicum is the only one of the six species in our multiple network alignment in which we could not identify this motif upstream of predicted inositol catabolic and transporter genes . Although it contains strong associations at the transporter nodes and for a number of the metabolic genes , it is missing several other myo-inositol catabolic genes and also does not encode a homolog of the regulatory protein IolR ( Figure 6 ) . The lack of conservation of several components of the myo-inositol network in B . japonicum decreases our confidence in the functional predictions presented for this species in Figure 6 relative to our predictions for S . meliloti , M . loti , B . melitensis , and A . tumefaciens . We propose that if B . japonicum can metabolize myo-inositol , it employs a different regulatory mechanism , and perhaps enzymatic strategy , than the other α-proteobacteria in our cross-species alignment . The cross-species network predicted that the operon Smb20712-4 ( NP_437959 , NP_437960 , NP_437961 ) in S . meliloti 1021 is a myo-inositol transporter ( Figure 6A ) . This ABC transporter operon in S . meliloti 1021 is annotated in GenBank as a putative rhizopine transporter , based on homology to the known MocB rhizopine transporter in S . meliloti strain L5-30 [43] . While S . meliloti 1021 cannot metabolize rhizopine [3] , rhizopine is derived from myo-inositol [43] suggesting that homology to MocB is a predictor of myo-inositol transport . However , a BLAST search of the S . meliloti 1021 genome using the sequence of C . crescentus IbpA inositol-binding protein as a query did not identify the periplasmic binding protein of the Smb20712-4 operon as the top hit , but rather another protein , Smb20072 , that is also annotated as a periplasmic rhizopine-binding protein . Indeed , a simple BLAST search revealed several different ABC transporter operons in S . meliloti with high probability scores to the experimentally-defined C . crescentus myo-inositol transporter ( see Table 3 for four candidate operons ) . Thus , simple pairwise comparisons with the known myo-inositol transporter of C . crescentus cannot easily distinguish the myo-inositol transport system in S . meliloti 1021 . Instead , several additional search criteria must be imposed before Smb20712-4 is assigned the highest confidence score as the principal myo-inositol transporter . Specifically , while other operons in S . meliloti showed higher overall homology with select subunits of the C . crescentus ABC transporter , the operon Smb20712-4 clearly showed the highest conservation across all six species in our network alignment ( Table 3 ) , and the promoter region of this operon also contained the conserved palindromic motif first identified in C . crescentus ( Figure 6B ) . To experimentally test our prediction , we tested the growth of strains of S . meliloti Rm2011 ( a direct derivative of S . meliloti 1021 ) carrying Tn5 transposon insertions in either iolA ( NP_384832 ) or in the predicted ABC transporter periplasmic binding protein gene , Smb20712 [24] , on GTS minimal medium [25] supplemented with either glucose or myo-inositol as the sole carbon source . Both Tn5 mutant strains grew normally in GTS-glucose and in Luria-Bertani ( LB ) medium . However , the iolA::Tn5 and Smb20712::Tn5 mutant strains did not grow on GTS with myo-inositol as a sole carbon source ( Table 4 ) . These results show that S . meliloti IolA is required for growth on myo-inositol and confirm our cross-species computational prediction that the protein Smb20712 is the functional homolog of the C . crescentus periplasmic inositol-binding protein IbpA . As such , we have annotated Smb20712 as ibpA , Smb20713 as iatA , and Smb20714 as iatP . Using forward and reverse genetic strategies , we have defined genes in C . crescentus involved in the metabolism of the abundant environmental sugar , myo-inositol . These experiments uncovered an ABC myo-inositol transporter , and identified a novel myo-inositol regulatory gene and conserved cis-acting promoter regulatory sequence that control gene expression . Together , these genes and regulatory sequences form a metabolic module that ensures C . crescentus can regulate gene expression in response to myo-inositol , transport the sugar across its inner membrane , and catabolize the sugar to form the central metabolite acetyl-CoA . Expanding upon these traditional genetic studies , we also presented a schema for generating reliable cross-species annotation of an entire functional genetic module . Specifically , using statistical and computational methods we leveraged our experimental work on C . crescentus myo-inositol metabolism to produce high-quality gene annotations for functionally-homologous myo-inositol transporters , catabolic enzymes , and transcriptional regulators in five related α-proteobacteria . The myo-inositol genes in all of these species were noncontiguous on their respective chromosomes ( Figure 7 ) , making it difficult to predict function using co-location . Our work has demonstrated the efficacy of combining a statistical protein network prediction algorithm [34] and the Graemlin network alignment algorithm [37] in the prediction and extrapolation of metabolic gene function . The method is significantly more robust than simple sequence comparison as a means to transfer annotation across species . Our network prediction and alignment protocol was validated on multiple levels: First , identification of the palindromic regulatory motif that was defined experimentally in C . crescentus ( Figure 4 ) in the upstream regions of homologous genes/operons in our cross-species network alignment ( Figure 6B & 7 ) provided excellent correlative validation of our alignment and refinement methodology . Second , we directly validated our cross-species functional prediction in S . meliloti , demonstrating that transposon disruption of the iolA gene and the Smb20712-Smb20714 transporter operon abolished growth on myo-inositol as the sole carbon source . As the number of microbial genome sequences continues to grow , it is imperative that we develop improved methods to define and assign functions to genes , and reliably propagate this functional information across species . This study demonstrates that combining directed genetic , genomic and molecular experiments , statistical functional prediction , and global network alignment provides a powerful means to define and propagate gene function at the pathway level . | More than 1 , 000 microbial genomes have been sequenced to date , containing millions of predicted genes . While the broad functional category of many of these individual genes can be reliably predicted using sequence homology , sequence information alone is often insufficient to assign a gene a specific cellular function . Closing this gap in our understanding of gene function will require tremendous experimental effort over a broad phylogenetic cross-section of model microbes , along with computational methods for high-confidence extrapolation of functional information from model organisms to other species . Here , we report the experimental identification of a novel genetic module in the model α-proteobacterium C . crescentus that controls transport and catabolism of the abundant environmental sugar myo-inositol . A combination of computational methods for probabilistic protein-network assignment and gene-network alignment were required to reliably extend the annotation of genes in this metabolic module to related bacterial genera . Our computational predictions of the operon encoding the ABC myo-inositol transporter and an essential enzyme for myo-inositol catabolism in S . meliloti were validated experimentally , demonstrating the feasibility of our method for high-confidence propagation of pathway-level annotation across species . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results/Discussion"
]
| [
"microbiology/environmental",
"microbiology",
"biochemistry/biocatalysis",
"physiology/sensory",
"systems",
"physiology/genomics",
"cell",
"biology/cell",
"growth",
"and",
"division",
"microbiology/microbial",
"evolution",
"and",
"genomics",
"molecular",
"biology",
"molecular",
"biology/bioinformatics",
"microbiology/microbial",
"physiology",
"and",
"metabolism",
"cell",
"biology/chemical",
"biology",
"of",
"the",
"cell",
"cell",
"biology/gene",
"expression"
]
| 2008 | Genetic and Computational Identification of a Conserved Bacterial Metabolic Module |
The nature of toxic effects exerted on neurons by misfolded proteins , occurring in a number of neurodegenerative diseases , is poorly understood . One approach to this problem is to measure effects when such proteins are expressed in heterologous neurons . We report on effects of an ALS-associated , misfolding-prone mutant human SOD1 , G85R , when expressed in the neurons of Caenorhabditis elegans . Stable mutant transgenic animals , but not wild-type human SOD1 transgenics , exhibited a strong locomotor defect associated with the presence , specifically in mutant animals , of both soluble oligomers and insoluble aggregates of G85R protein . A whole-genome RNAi screen identified chaperones and other components whose deficiency increased aggregation and further diminished locomotion . The nature of the locomotor defect was investigated . Mutant animals were resistant to paralysis by the cholinesterase inhibitor aldicarb , while exhibiting normal sensitivity to the cholinergic agonist levamisole and normal muscle morphology . When fluorescently labeled presynaptic components were examined in the dorsal nerve cord , decreased numbers of puncta corresponding to neuromuscular junctions were observed in mutant animals and brightness was also diminished . At the EM level , mutant animals exhibited a reduced number of synaptic vesicles . Neurotoxicity in this system thus appears to be mediated by misfolded SOD1 and is exerted on synaptic vesicle biogenesis and/or trafficking .
A number of neurodegenerative diseases have been associated with protein misfolding and aggregation , with a specific protein in each case observed to aggregate in a particular population of neurons . For example , in the case of amyotrophic lateral sclerosis ( Lou Gehrig's Disease ) , a dominantly inherited form of this condition , accounting for ∼2% of cases , is associated with mutant forms of the abundant cytosolic homodimeric enzyme superoxide dismutase ( SOD1 ) , which accumulate in insoluble aggregates in motor neurons [1]–[3] . Mutational studies of SOD1-linked ALS have uncovered single residue substitutions throughout the enzyme subunit [4] , and studies in vitro indicate that the substitutions generally destabilize the protein , disposing to misfolding and aggregation [5]–[7] . It remains unknown , however , exactly how apparent misfolding and aggregation of SOD1 exerts toxic effects on motor neurons . Is there a central common effect shared by the various mutant alleles that comprises a common pathway of motor neuron injury ? Mice transgenic for a variety of mutant SOD1 alleles also develop motor neuron disease resembling that of affected humans [8] , [9] , enabling a variety of pathological and biochemical studies . A survey of pathology reported for various alleles implicates a variety of potential physical sites of toxicity , including mitochondria , endoplasmic reticulum , and axonal traffic . For example , abnormal-appearing mitochondria have been observed in animals transgenic for G93A and G37R SOD1 [9]–[11] , and several mutant SOD1's have been coisolated with spinal cord mitochondria [12]–[15] . Concerning ER function , an unfolded protein response ( UPR ) was observed in spinal cord of G93A mice [16] , and a recent report suggests that mutant SOD1 induces this response by binding to the ER membrane component Derlin-1 , blocking retrograde traffic of ER proteins to the cytosol for proteasomal degradation ( ERAD ) at the level of ubiquitination [17] . Concerning axonal traffic , both anterograde and retrograde transport have been observed to be retarded in mice transgenic for mutant SOD1 [18]–[24] . Which , if any , of these effects is primary to mutant SOD1-induced motor neuron damage ? One approach to resolving this question is to produce mutant human SOD1 in neurons in an invertebrate system to inspect for effects on function , with the idea that this might reveal a minimal target of the toxic effect which could ultimately be further evaluated in the mammalian system . Such an approach has been taken , for example , with other neurodegenerative disease-associated proteins , expressing them in Drosophila or C . elegans , including polyglutamine repeat proteins [e . g . 25] , [26] and α-synuclein [e . g . 27] , [28] . Here we have taken such an approach with SOD1 using C . elegans , programming pan-neuronal expression of a mutant version of human SOD1 , G85R , that obligatorily misfolds . We observe a locomotor defect in transgenic animals expressing the mutant SOD1 , associated with aggregation of the mutant SOD1 protein and with synaptic dysfunction , involving deficient numbers and possibly deficient trafficking of pre-synaptic vesicles .
To examine the effects of an ALS-associated human mutant SOD1 on a collective of neurons in an optically accessible nervous system , we produced transgenic C . elegans expressing G85R mutant or wild-type human SOD1 ( referred to hereafter as SOD ) . G85R SOD has been identified in human cases of ALS [1] , and G85R SOD transgenic mice develop a similar disease [29] . In the latter setting , the protein has been shown to behave as a misfolded monomer [30] . That is , it fails to form the normal SOD homodimer , and it lacks the normal disulfide bond that is formed between Cys 57 and Cys 146 when the protein is properly folded . ( This disulfide bond is normally formed despite localization of SOD to the relatively reducing cytosol ) . To express the wild-type and G85R SOD proteins in as many of the 302 neurons of a C . elegans hermaphrodite as possible , a pan-neuronal promoter , the promoter of the C . elegans synptobrevin gene ( snb-1 ) , was used to drive the respective wild-type and G85R human cDNAs encoding SOD . To allow direct observation of the expressed SOD proteins , two additional constructs were employed that join a YFP reporter sequence via a flexible peptide linker to the C-terminus of the wild-type or mutant SOD [31] . Multiple stable transgenic C . elegans lines of both unfused and fused constructs were produced . We noticed immediately that G85R SOD transgenic animals exhibited minimal forward movement across the culture medium compared with normal movement of age and protein expression-matched wild-type SOD transgenic strains ( Figure 1A and Video S1; see Figure S1 for immunoblot analysis and activity blot analysis ) . The wild-type SOD transgenic animals exhibited a forward movement speed similar to that of the parental nontransgenic C . elegans N2 strain ( Figure 1A ) . In addition to severely reduced forward crawling of the mutant transgenic animals , their side-to-side thrashing movement in liquid was also severely affected ( Figure S2 and Videos S2 , S3 ) . The rates of forward movement of the animals transgenic for the fusion proteins , WTSOD-YFP and G85R-YFP , likewise showed a large difference ( Figure 1B ) , although the mutant now exhibited significant forward movement , and the wild-type fusion was somewhat slower than the nonfused wild-type transgenic animals ( compare Figure 1B with 1A ) . These same effects were observed on thrashing ( Figure S2 ) . Thus the addition of the YFP moiety has effects on the mutant and wild-type SOD proteins , exerted in opposite directions . Overall , however , even with the YFP fusion protein , there was still markedly slower movement of the G85R animals as compared with the wild-type ( Figure 1B ) . To assess whether the biological behavior of the G85R-YFP fusion protein has any relevance to the mammalian context where expression is associated with an identifiable clinical neuronal disorder , G85R-YFP was produced in transgenic mice from a human SOD genomic clone with YFP fused to the last coding exon . This fusion produced an ALS phenotype in mice at 3–9 months of age , with the age of onset of the motor deficit depending on copy number and expression level of the transgene ( JW , GF , KF , and ALH , unpublished ) . By contrast , expression-matched wild-type SOD-YFP transgenic mice produced in parallel did not develop disease even at ages of 2 years and beyond . Thus the SOD-YFP fusions tested here reflect the behavior of the nonfused counterparts in the context of production of ALS-like disease in the mammalian setting . An additional SOD mutant transgenic C . elegans strain , H46R/H48Q-YFP , containing an SOD double mutant allele that blocks copper binding by SOD and produces ALS in transgenic mice [32] , was also examined . Here it produced a movement defect less prominent than that seen in G85R-YFP ( Figure 1B ) . Fluorescence microscopy of transgenic C . elegans at both larval and adult stages revealed fluorescence in many neurons in the nerve ring ( head region ) , the ventral nerve cord , lateral body wall , and tail ganglia ( Figure S3A ) . Non-neuronal fluorescence of the spermatheca and two gonadal distal tip cells was also observed . The character of fluorescence of neuronal cell bodies of G85R-YFP transgenic animals differed from that of WTSOD-YFP transgenics as exemplified by cell bodies of motor neurons along the ventral nerve cord ( Figure 2A , B ) . [Neuronal processes , by contrast , did not exhibit altered fluorescence ( Figure S3B ) . ] Cell bodies from the wild-type animals exhibited a more diffuse cytosolic fluorescence pattern , whereas those from the G85R-YFP mutant exhibited a well-demarcated pattern , suggestive of aggregate formation . Consistent with this , a FRAP experiment ( fluorescent recovery after photobleaching ) showed very slow recovery of fluorescence in mutant cell bodies as compared with wild-type ( see Figure S4 ) . Aggregates were directly observed in the mutant by EM examination following chemical fixation ( Figure 2C , D ) , which revealed well-delineated electron-dense inclusions in the cytosol of ventral nerve cord cell bodies of G85R-YFP animals ( Figure 2C , Agg . ) . By contrast , there were no recognizable aggregates in WTSOD-YFP cell bodies ( Figure 2D ) . The location of G85R-YFP aggregates in a perinuclear position is reminiscent of aggresomes [33] or of perinuclear structures distinct from the centrosome known as JUNQ , involved in juxtranuclear quality control [34] , but aggregates formed in the unfused G85R animals exhibited a more diffuse , “fluffy” character ( Figure S5A , asterisks ) . When high pressure freezing/fixation was employed on the G85R animals , these regions were now observed as amorphous inclusions ( e . g . Figure 2E , Agg . ) , which at high magnification appeared to contain loosely stacked fibrillar material ( Figure 2F ) . To identify genetic modifiers of neuronal aggregation in the G85R SOD-YFP transgenic animals , we carried out an RNAi screen . Neurons of C . elegans are relatively resistant to RNA interference , so alleles previously identified to enhance interference activity , eri-1 ( mg366 ) and lin-15B ( n744 ) , were introduced [39] . Strikingly , the introduction of these alleles led to a significant reduction in fluorescence from the G85R-YFP protein as compared with the parental strain ( Figure S9A ) , associated with a decrease in the number of fluorescent inclusions observed in the ventral nerve cord ( Figure S9B ) and with somewhat improved movement . These results are consistent with a previous report that lin-15B could produce silencing of multicopy transgenes [43] . We observed that either allele alone could produce substantial silencing of the G85R-YFP transgene , although lin-15B exerted a greater effect . The combination of alleles proved to sufficiently reduce fluorescence and aggregate formation of G85R-YFP to a degree that could allow for a “dynamic” range of effects of RNAi screening , i . e . both reduction and enhancement of green fluorescence patterns reflecting aggregation would be detectable . A bacterial feeding library was employed for RNAi testing . Collectives of animals at all stages were transferred onto feeding plates and visually examined by multiple observers after 3–6 days for changes in the number and intensity of fluorescent aggregates in the nerve ring and ventral nerve cord as compared with the strain fed bacteria with an empty vector . In the general case of observing decreased fluorescent aggregates when animals were placed on a particular interfering bacterial strain , the development of the animals was generally also slowed and viability in many cases reduced , reflecting likely effects on general health or on general gene expression , and these interfering RNAs were not studied further . There was one exception to this , involving an interfering RNA for a ubiquitin specific protease , where fluorescent aggregation was reduced and viability somewhat improved , and this gene is under further study . In the general case of animals with increased fluorescent aggregates on a particular interfering bacterial strain , these animals generally exhibited larger numbers of fluorescent inclusions and diminished locomotion . There were 88 such hits ( Table S1 ) . 7 of these appeared likely to suppress eri-1; lin-15B action and were not studied further ( see last entries in Table S1 ) , while the remaining 81 hits were categorized into 10 groups ( Table 1 ) . The largest group of well-annotated hits comprised molecular chaperones and quality control components , amounting to about a quarter of the hits . These are discussed below . The other large group , of about the same size , comprised a collective of uncharacterized gene products . To independently confirm RNAi hits in the parental genetic background , loss-of-function alleles were obtained for 12 hits of interest and were crossed with the parental G85R-YFP strain . 11 out of 12 of these enhanced the inclusion phenotype of G85R-YFP . For example , heat shock factor 1 ( HSF1 ) , which transcriptionally regulates a number of stress components [44] , registered very strongly in the RNAi screen in increasing aggregate formation . Consistently , when the sy441 allele of hsf-1 was crossed into G85R-YFP , a strong increase in aggregate formation was observed ( Figure S10A ) , and locomotion of these animals was substantially decreased as compared with either parental strain ( Figure S10B ) . Multiple interference hits of strong magnitude were observed in a number of pathways ( Table 2 ) , indicating that these pathways are likely playing a role in preventing aggregation of the misfolded G85R-YFP protein or facilitating its turnover . For example , hits of a collective of chaperone components registered strong increases of aggregation , including an Hsp110 ( C30C11 . 4 ) , a DnaJ ( A2 ) ( dnj-19 ) , an Hsp70 ( stc-1 ) , and a neuron specific Hsp16 ( F08H9 . 4 ) . Three of these were validated as strongly increasing aggregation when mutant alleles were crossed with G85R-YFP . Two hits were identified in the ubiquitin-mediated turnover pathway at the level of E3 ligase SCF complexes , SEL-10 , an F-box protein , and RBX-1 , a RING finger protein , the latter confirmed with the ok782 knockout allele . Three hits were also observed in the sumoylation pathway by the RNAi screen , uba-2 , encoding the E1 enzyme that activates SUMO , ubc-9 , which encodes one subunit of the E2 SUMO-conjugating heterodimer , and an E3 SUMO-ligase component gei-17 ( homologous to PIAS1 ) . In validation of the interference effects , an allele of SUMO , smo-1 ( OK359 ) , increased aggregation when crossed into the parental G85R-YFP strain . The latter hits raised the question of whether G85R-YFP is itself sumoylated , and this is under study . In addition , three components involved with redox regulation were identified , PDI-2 , an orthologue of a human thioredoxin domain-containing protein ( C30H7 . 2 ) , and BLI-3 , a dual oxidase with both a peroxidase domain and a superoxide-generating oxidase domain . Other pathways were also identified through strong effects of interference . For example , a hit in the TGFβ component DBL-1 was confirmed by the allele nk3 . DBL-1 is expressed in neuronal cells , e . g . in the ventral cord , has recently been implicated in GABAergic synaptic transmission [45] , and has been shown to affect both body size and male tail development [46] . Notably , in mammalian contexts , TGF-β family members have exhibited effects on axon outgrowth and protection from excitotoxicity [e . g . 47] . A hit of the dopamine transporter DAT-1 was confirmed with allele tm903 . A link between neuronal activity and aggregation in the cases of GABAergic and dopaminergic transmission thus seems a consideration but requires further study . In the DNA replication/repair pathways , interference with top-1 , the gene for topoisomerase 1 , had a very strong effect , equivalent to the strong effect of hsf-1 interference . How such action could figure into aggregation behavior remains unclear . Similarly , interference of div-1 , a subunit of the DNA polymerase α/primase complex , increased aggregation . Also a strong effect was observed with interference of pha-4 , a FoxA transcription factor , implicated in calorie-restriction-mediated longevity , at least in part by regulating endogenous C . elegans SOD enzymes [48] .
We have described here that pan-neuronal expression of a human ALS-associated mutant SOD in C . elegans produces substantial locomotor defects associated with macroscopic aggregation in neuronal cell bodies . By contrast , a wild-type human SOD produced neither locomotor defects nor aggregation . Further testing indicated that the human SOD mutant , G85R , unable to fold properly and producing both soluble oligomers and aggregates , appears to produce neuronal dysfunction in C . elegans at the presynaptic level . This was indicated by three types of observation . First , there was a reduced number and brightness of puncta of fluorescently-labeled synaptobrevin , RAB-3 , and synapsin , as well as of the presynaptic active zone protein RIM1 in the dorsal nerve cord in transgenic G85R animals as compared with wild-type SOD transgenic animals . Second , mutant transgenic animals but not wild-type exhibited resistance to aldicarb paralysis , consistent with deficient acetylcholine release at cholinergic synapses in the mutant . Third , EM studies on a small number of animals revealed paucity of vesicles in the presynaptic region of nerve ring synapses of the mutant animals . In addition , EM showed reduced numbers of vesicles and mitochondria in ventral nerve cord processes ( axons ) of transgenic mutant animals . In contrast with the foregoing findings pointing to presynaptic dysfunction , there was a lack of postsynaptic effects . The cholinergic agonist levamisole produced the same efficient paralysis of both mutant G85R and wild-type SOD transgenic animals . In addition , EM inspection of muscle revealed normal morphology . The putative presynaptic defect could result , in the first instance , from defective biogenesis , axonal transport , or recycling of synaptic vesicles . The diminution of vesicles in dorsal and ventral nerve cord processes and their sparsity in the presynaptic regions as demonstrated from the EM and fluorescence studies suggest that biogenesis or transport is affected . Because accumulations of vesicles were not detected in cell bodies , it seems less likely that transport is affected . Yet the failure of recovery of the GFP-synaptobrevin fluorescence following photobleaching could be consistent with an ongoing transport or recycling defect . Further studies will be required to resolve the steps affected . Excitingly , effects of expression of human SOD1 on synaptic transmission in another invertebrate system have recently been reported . Watson et al [49] observed that expression of human SOD1 ( wild-type or mutant ) in motor neurons of Drosophila produced an age-progressive climbing defect that was associated in electrophysiological studies , stimulating muscle through the giant fiber circuit , with progressive loss of muscle response during high frequency stimulation . Presynaptic effects have also been observed in several contexts in C . elegans expressing a number of other neurodegeneration-associated proteins . Animals transgenic for both wild-type and FTDP-associated mutant forms of tau presented with locomotor defects , associated with aldicarb resistance ( and levamisole sensitivity ) , followed by appearance of macroscopic aggregates and then apparent neuronal loss [50] . Animals transgenic for α-synuclein exhibited locomotor defects when any of a number of synaptic proteins were knocked down , associated with aldicarb resistance and levamisole sensitivity [28] . In contrast with these studies , however , the studies of SOD1 presented here indicate a direct and specific effect of the mutant G85R SOD1 on presynaptic function . So far it is not evident how or whether the presence of G85R-SOD aggregates in cell bodies of specific neurons , comprising a variety of neuronal cell types in the transgenic mutant animals , directly relates to the apparent presynaptic defect . We note , however , that the degree of locomotor defect correlates roughly with the degree of aggregation observed . For example , for the G85R-YFP transgene , out of multiple stable integrant lines , those with the highest level of fluorescence and greatest level of aggregation exhibited the strongest locomotor defect . Thus , either the aggregates themselves or perhaps the apparent precursors , soluble oligomers ( Figure 3B ) , may be directly responsible for the presynaptic defect . The defect could come at the level of physical interactions of oligomers or aggregates either directly with synaptic vesicles themselves or with soluble or cytoskeletal components that are involved with vesicle biogenesis and traffic . A recent study producing pan-neuronal expression of dimeric versions of human wild-type and G85R SOD1 in C . elegans also observed locomotor defects associated with aggregation , but , in contrast to the direct correlation above , observed that a heterodimeric G85R-WTSOD-GFP molecule , while producing less aggregation than G85R-G85R-GFP , led to a greater paraquat-induced reduction of animal survival [51] . Thus in the context of a heterodimer , residual SOD enzymatic activity may contribute to toxicity . Notably , despite evident protein aggregation and synaptic dysfunction , neurons in the mutant animals studied here were not subject to cell death , even during later adult life . This resembles the observation in the recent report of Watson et al [49] where human SOD1 expressed in Drosophila motor neurons produced both focal SOD protein accumulation and measurable electrical dysfunction but no observable cell death . Similarly , in earlier studies of C . elegans PLM neurons transgenic for polyglutamine expansion , touch sensitivity function was abolished but cell death did not occur [52] . This lack of cell death may either relate to the state of disease progression or be a function of the invertebrate neuronal systems themselves . For example , concerning stage of disease , in the mammalian context , SOD1-affected motor neurons appear likely to be functionally affected for a period of time before being subject to cell death . In C . elegans or Drosophila , by contrast , the trajectory may not extend sufficiently in time to produce cell death , albeit that added insults such as oxidative toxicity [50] may be capable of producing cell death . Alternatively , these systems may differ from that of mammals . C . elegans has , for example , a limited number of glia , and they may not function as in the mammalian context to hasten death of affected neurons [53]–[55] . Alternatively , the absence of cell death could be a function of a different neuronal response to chronic exposure to misfolded protein as compared with mammalian neurons . The RNA interference screen conducted here , examining relative levels of G85R-YFP protein aggregation in relation to knockdown of various gene products , validated in many cases by crosses with corresponding mutant alleles ( Table 2 ) , provided evidence that a proteostatic network similar to that present in body wall muscles of C . elegans [56] , [57] and elsewhere [44] is operative in C . elegans neurons . Whether it is induced in response to mutant human SOD1 remains to be determined , and whether a higher level of expression of some or all members of the network could be protective remains to be tested . Two transcriptional regulators that lie at the top of such a network were identified , HSF-1 and PHA-4/FoxA , which have a broad range of targets in , for example , chaperone pathways [44] and redox regulation [48] , respectively . Components within the network were also identified , including chaperones , an E3 ligase , and redox components . Additional components had effects on aggregation but their mechanism of action remains unclear , including SUMO , the TGF-β homologue , DBL-1 , expressed mainly in neurons , and two components of DNA maintenance , topoisomerase I and a subunit of polα/primase complex , implicating DNA integrity in the response . Notably absent from both the interference screen and from EM studies was evidence for involvement of the autophagy system . Consistent with this , administration of rapamycin was without effect on the extent of aggregation ( data not shown ) . Do the phenotypic properties observed here bear any relation to mammalian disease induced by expression of mutant SOD1 ? Could the transgenic C . elegans inform usefully about mammalian disease ? As mentioned , the G85RSOD-YFP fusion protein expressed in mice indeed produces an ALS-like disease whereas WTSOD-YFP fusion produces no ill effect . Interestingly , the presynaptic effects observed here with the G85R transgenic C . elegans may have a parallel in mutant G85R SOD transgenic mice , as reported recently by Caroni and colleagues [58] , who examined motor neuron axons and NMJs in hind limb muscle of mutant transgenic animals of varying age [see also 59] . In vulnerable motor neuron axons ( FF and FR ) , they observed at early time ( 7 months of age; 2 months before end-stage ) localized synaptic vesicle accumulation associated with diminished overall density of vesicles , reflecting apparent stalling of vesicle traffic , followed at a later stage by severe loss of synaptic vesicles associated with loss of presynaptic active zone markers . The findings at later time generally agree with the observations here in dorsal cord of our G85R C . elegans , where both fluorescent synaptic vesicle protein markers and an active zone marker ( RIM1 ) appear to be reduced ( Figure 6 ) . The nature of the vesicle trafficking defect in either setting remains to be elucidated . Does it reflect failure of vesicles to be produced in the first instance , as suggested by the lack of organelles in ventral cord processes here ( Figure 5 ) , and/or is it a block of recycling , as might be suggested by the FRAP analysis of GFP-synaptobrevin ( Figure 6C ) ? Is it a direct effect of mutant SOD , forming physical association with synaptic vesicles ? Or is it a secondary effect , mediated e . g . via the motor/cytoskeletal trafficking system ? Questions concerning both the basis to vesicular defects and the overall pathway of toxicity of mutant SOD protein remain to be resolved .
The C . elegans snb-1 promoter , a PCR-amplified genomic DNA segment extending from minus 3021 bp to just upstream of the SNB start codon , was inserted in place of the unc-54 enhancer/promoter in the plasmid pPD30_38 ( Fire Lab Vector Kit , Addgene Inc . , Cambridge , MA ) , and the various human SOD cDNA-containing segments were then adjoined . SOD mutations were generated by PCR , and the derived coding sequences confirmed by sequencing them in entirety . Fusion constructs joining human SOD with YFP via a linker segment ( LQLQASAV ) were kindly provided by Dr . R . Morimoto , Northwestern University [31] . The N2 Bristol strain of C . elegans was used as the wild-type strain . Standard culturing and genetic methods were used [60] . Animals were maintained at 20°C unless otherwise indicated . Mutant strains obtained from the Caenorhabditis Genetics Center ( CGC ) , the National Bioresource Project in Japan , and the lab of Joshua Kaplan are listed in Suppl . Table 2 . Germline transformation was performed by injecting DNA solution containing 20 ng/µl of an SOD construct and 5 ng/µl of myo2::GFP into hermaphrodite gonads [61] . Multiple extrachromosomal lines were established based on the fluorescent markers . They were further treated with trimethylpsoralen/UV to generate integrated lines that stably expressed the transgenes . At least three independent stable lines were produced for each variant , and each line was backcrossed with the N2 strain four times . The transgenic SOD and SOD-YFP lines used in these studies are designated in the legend to Figure 1 . nuIs152 has the transgene Punc-129::GFP-snb-1 [39] . nuIs168 has the transgene Punc-129::YFP-Rab-3 [62] . nuIs163 and nuIs165 strains containing Punc-129::snn-1-YFP and Punc-129::unc-10-GFP were the kind gift of Joshua Kaplan [63] . For high resolution imaging , animals were immobilized with levamisole and examined by either differential interference contrast ( DIC ) or fluorescence with an Olympus IX81 microscope equipped with spinning disk confocal illumination . For fluorescence recovery after photobleaching ( FRAP ) , a laser confocal microscope was used ( Zeiss LSM510 META ) . For transmission electron microscopy , animals were prepared by conventional two-step chemical immersion fixation or high-pressure-freezing [64] , [65] . Serial thin sections were prepared and post-stained with heavy metals . At least four animals of each genotype were analyzed using a Tecnai 12 Biotwin at 80 kV . A video-based assay was used to assess the locomotion speed of C . elegans . Animals were transferred to a plate with a fresh bacterial lawn on which movement tracks could be traced . Immediately upon release , worms exhibit a maximum movement response for a short duration . A 30 second movie was shot for each worm , and the ratio of the movement distance to the body length , measured by the NIH ImageJ software , was used as a movement index ( see Video S1 ) . Mid L4 animals were transferred to freshly made NGM agar plates without bacterial food containing 1 mM aldicarb , and at different time points the animals were prodded on the nose to determine whether they had reached complete paralysis [42] . An identical experiment with 1 mM levamisole was also performed . To assess solubility of SOD protein , animals were disrupted by sonication on ice in 0 . 5 ml extraction buffer ( PBS , 1 mM EDTA , 1 mM EGTA , 1 mM TCEP , with half a tablet of Complete Mini protease inhibitor cocktail ( Roche ) ) and left on ice for 10 min to allow large debris , including cuticle , to sediment . The supernatant fraction was then centrifuged in a Beckman TLA-100 rotor at 53 , 000 rpm ( >120 , 000×g ) for 15 min at 4°C . Pellets were washed once by resuspension in the extraction buffer and sedimentation . Supernatant and SDS-solubilized pellet fractions were analyzed in SDS-PAGE under reducing conditions . Supernatant fractions ( 1 mg total protein ) were also subjected to gel filtration chromatography on a Superose 6 gel filtration column ( GE Healthcare ) and eluted with PBS supplemented to 0 . 1 mM TCEP at 0 . 5 ml/min . Individual fractions ( 0 . 5 ml ) were examined by SDS-PAGE and Western blotting . For Western analysis , antibodies to human SOD1 ( SOD-100 , Stressgen , Canada ) , antibodies raised in rabbits against purified YFP ( Cocalico ) , or antibodies to actin ( C4 , MP Biomedicals , Inc . , Aurora , Ohio ) were used . An RNAi feeding library of 16 , 757 bacterial clones was employed for screening ( GeneService , Cambridge , UK ) . Animals at mixed ages were screened in 96-well plates at 15°C as described [66] . “Hits” were identified by an increased number and intensity of fluorescent neuronal inclusions using a Leica fluorescence stereoscope with a 2 . 0× PLANAPO lens . All positives were subjected to secondary screening at both 15°C and 20°C in 6-well plates . The identities of all positive RNAi clones were confirmed by DNA sequencing of the plasmid insert . For selected hits , where loss-of-function alleles were available , the corresponding strains were crossed to the G85R-YFP parental strain ( line 8 ) . The genotypes of the product strains were verified by PCR or PCR/DNA sequencing and the phenotypes studied . | A new animal model of the human neurodegenerative disease amyotrophic lateral sclerosis ( ALS; Lou Gehrig's Disease ) is presented . Two percent of ALS cases result from heritable mutations affecting the abundant enzyme superoxide dismutase ( SOD1 ) . Such mutations have been indicated to impair the folding and stability of the enzyme , leading it to misfold and aggregate in motor neurons , associated with the paralyzing disease . Here , when a mutant form of human SOD1 was produced in neurons of C . elegans worms , it led to a severe locomotor defect—the worms were essentially paralyzed . The protein formed aggregates in the neurons , including an intermediate form of aggregate , soluble oligomers , that has been linked to toxicity to cells . By contrast , worms expressing the normal version of human SOD1 protein exhibited normal movement and no aggregation . The movement defect was further analyzed using chemical inhibitors and found to result from defective function of synapses , the connections made between neurons , and between neurons and muscle . Finally , in a screen using RNA interference , we observed that the worms' aggregation and locomotor condition was worsened when a class of molecules called molecular chaperones , which assist protein folding in the cell , were impaired in function . This is consistent with the idea that misfolded SOD1 is directly involved with causing the neuronal dysfunction . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"biology/neuronal",
"and",
"glial",
"cell",
"biology",
"biochemistry",
"cell",
"biology",
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"genetics",
"and",
"genomics/gene",
"function",
"genetics",
"and",
"genomics"
]
| 2009 | An ALS-Linked Mutant SOD1 Produces a Locomotor Defect Associated with Aggregation and Synaptic Dysfunction When Expressed in Neurons of Caenorhabditis elegans |
The rate of mutation is central to evolution . Mutations are required for adaptation , yet most mutations with phenotypic effects are deleterious . As a consequence , the mutation rate that maximizes adaptation will be some intermediate value . Here , we used digital organisms to investigate the ability of natural selection to adjust and optimize mutation rates . We assessed the optimal mutation rate by empirically determining what mutation rate produced the highest rate of adaptation . Then , we allowed mutation rates to evolve , and we evaluated the proximity to the optimum . Although we chose conditions favorable for mutation rate optimization , the evolved rates were invariably far below the optimum across a wide range of experimental parameter settings . We hypothesized that the reason that mutation rates evolved to be suboptimal was the ruggedness of fitness landscapes . To test this hypothesis , we created a simplified landscape without any fitness valleys and found that , in such conditions , populations evolved near-optimal mutation rates . In contrast , when fitness valleys were added to this simple landscape , the ability of evolving populations to find the optimal mutation rate was lost . We conclude that rugged fitness landscapes can prevent the evolution of mutation rates that are optimal for long-term adaptation . This finding has important implications for applied evolutionary research in both biological and computational realms .
Mutation is the ultimate source of genetic variation , and thus the rate at which spontaneous mutations appear is a fundamental evolutionary parameter . The mechanisms of DNA replication and repair are themselves genetically encoded and variable [1]–[5] , making mutation rates potential targets of evolutionary optimization . Two opposing forces contribute to the evolution of mutation rates . On the one hand , most mutations with phenotypic effects are deleterious , producing a genetic load that favors organisms with low mutation rates; on the other hand , beneficial mutations are necessary for adaptation . Given this trade-off between genetic load and adaptation , there should exist an intermediate mutation rate—hereafter referred to as the ‘optimal’ rate , or Uopt—that balances these forces and maximizes adaptation over the long-term [6]–[9] . It is important , however , to note that these two forces operate at different timescales . The costs of genetic load are continuously paid in the short-term , whereas the payoffs of adaptation come in the long-term [6]–[8] , [10]–[12] . Experiments have shown that genotypes with increased mutation rates can be favored by selection if they face novel or changing environments [1] , [13]–[21] . Similarly , recent work with RNA viruses has shown that certain high-fidelity genotypes have diminished fitness and virulence in mice [22] , [23] , which might reflect their restricted ability to create the genetic variability needed to escape from immune surveillance . However , another recent study with an RNA virus failed to observe a positive association between mutation rate and the rate of adaptation to a novel environment [24] . Despite their importance , these studies suffer from some unavoidable limitations . For example , it is unknown whether the observed mutation rates are the product of evolutionary optimization or , alternatively , if they are far from their optimal values . Also , it is often difficult to assess whether experimental observations reflect evolutionary equilibria or transient states . These limitations can be overcome using evolution with digital organisms owing to the speed and ease of data collection . Digital organisms are self-replicating computer programs that inhabit a virtual world where they reproduce , mutate , compete for resources , and evolve according to the same fundamental processes as biological organisms [25] . Here , we use digital organisms to study the ability of natural selection to adjust the mutation rate . We first validate the existence of an optimal mutation rate by extensively exploring a range of mutation rates and observing which rate maximizes adaptation over the long-term . Then we allow mutation rates to evolve under natural selection and assess whether the optimal rate is reached . Even in conditions highly favorable for mutation rate optimization , mutation rates systematically evolve that are far below the optimum , showing that natural selection fails to optimize mutation rates . We propose a novel hypothesis for these results based on the topology of the underlying fitness landscape , and we then proceed to experimentally test it .
We studied the evolution of mutation rates using the Avida digital evolution platform [25]–[34] . To test empirically whether there was an intermediate , optimal rate of mutation that maximized adaptation , we performed a series of evolution experiments . In each experiment , a genetically homogenous population was placed in a novel environment where it evolved for 150 , 000 updates ( ∼15 , 000 generations ) at a constant mutation rate ( see Methods ) . We explored 15 different mutation rates spanning six orders of magnitude ( 10−5 to 10 mutations per genome per generation ) . The final fitness values confirmed that there was an optimal mutation rate at an intermediate value , with Uopt≈4 . 641 ( Figure 1 ) . An analysis of the temporal dynamics of these experiments showed that this rate yielded the highest fitness from about generation 230 onward . Interestingly , for the very earliest time points ( before generation 50 ) , the lowest mutation rate ( 10−5 ) produced the highest fitness values , whereas for generations 50–230 a mutation rate of 2 . 2 gave the highest fitness values . To assess whether evolution would produce organisms with mutation rates near the long-term Uopt , we ran additional experiments in which mutation rates were allowed to change ( see Methods ) , starting from rates either below ( 10−3 ) or above ( 10 ) the optimum . Strikingly , mutation rates evolved to levels far below the long-term Uopt , regardless of the starting value ( Figure 1 ) . In light of our observation that the optimum rate can change over time , one might hypothesize that the typical mutation rate of an evolving population had actually followed a near-optimal trajectory throughout its evolution , but that the final mutation rate is not a good indicator of the ability to optimize the mutation rate . However , this explanation can be ruled out because the final average fitness of the populations whose mutation rates could change was significantly lower than the fitness levels of the populations that evolved at a constant Uopt . The log-transformed final fitness values for treatments with changing mutation rates were 4 . 61±0 . 70 and 1 . 23±0 . 15 ( mean±1 s . e . m . ) for the populations starting at high and low initial rates , respectively . Both of these values are significantly lower than the 14 . 45±0 . 64 obtained for populations evolved at Uopt ( Mann-Whitney tests , both P<0 . 001 ) . The fitness advantage for Uopt is also clear for nearly all intermediate time points ( Figure 2A ) . While populations starting below Uopt did experience a transient increase in their mutation rates ( Figure 2B ) , the mutation rates still stayed more than two orders of magnitude below Uopt . For populations starting above Uopt , the results were particularly striking because selection pushed the populations through the optimal rate on their way to an evidently very suboptimal rate ( Figures 1 and 2B ) . The finding that mutation rates evolved to be suboptimal was robust to diverse and substantial changes in the experimental conditions . First , we tested whether our results depended on the particular ancestral organism used . In the original experiments , the ancestor was a default , hand-coded organism . To assess whether this condition substantively influenced our results , we let a population founded by this organism adapt for 50 , 000 updates to an environment without any rewarded functions , using U = 4 . 641 . The most abundant genotype at the end of this preliminary run was then used as the ancestor in repetitions of our original experiments . Second , we modified the complexity of the environment by varying the number of rewarded functions . Third , we tested the effect of environmental fluctuations by introducing periodic changes in the set of rewarded functions . In some of these experiments the non-rewarded functions were neutral , and in others performing these functions reduced fitness . The rate at which environmental fluctuations occurred was also varied . Fourth , we experimented with different implementations of how mutation rates could themselves change over time . In the original experiments , each organism's mutation rate had a constant probability Π of changing every generation , and the magnitude of any resulting change was controlled by a dispersion parameter σ , with Π = 0 . 5 and σ = 0 . 1 . We conducted additional experiments in which we lowered Π , raised σ , or both by orders of magnitude . We also explored a configuration where increases in the mutation rate were more likely than decreases , as may happen in biological systems where it is more likely for mutations to harm than to improve an existing DNA repair pathway . Finally , we let the mutation rate apply reflexively to itself , such that high-fidelity genotypes rarely changed their mutation rates whereas low-fidelity genotypes did so frequently . In all of these additional experiments , mutation rates evolved to suboptimal levels ( data not shown ) . We conclude , therefore , that selection fails to optimize mutation rates for long-term adaptation in a broad range of experimental conditions . A possible explanation for why mutation rates evolved to be much lower than Uopt is that selection favored those genotypes that minimized the short-term fitness costs caused by deleterious mutations . This explanation is supported by the observations that , during the earliest generations of the evolution experiments , the lowest mutation rate yielded the highest fitness values . To test whether short-term selection would favor low mutation rates , we performed competition experiments between two kinds of organisms , designated A and B . These organisms were identical except for their mutation rate , which was set to Uopt for A and 0 for B; neither mutation rate was allowed to change during the competition . All competitions were conducted with the same environmental configurations as in the main experiments . In all of 50 runs , B drove A extinct in fewer than 40 generations . Competitions were also performed using U = 1 . 0 and U = 2 . 154 for B in order to address whether selection would also favor less extreme reductions in mutation rate . In both treatments , B drove A extinct in all 50 trials in fewer than 800 generations . These experiments confirm our hypothesis that natural selection was shortsighted and favored low mutation rates , even when such low rates precluded further adaptation . We conclude from the results presented thus far that the failure of the evolving populations to achieve or even maintain the mutation rates that maximize long-term adaptation reflect the conflict between the short-term cost of deleterious mutations and the long-term potential for adaptive evolution . We further hypothesize that the resolution of this tension may depend on the topology of the fitness landscape on which evolution occurs . In a rugged fitness landscape , where there are multiple peaks separated by maladaptive valleys [35] , [36] , populations at a local optimum must traverse regions of low fitness in the short-term in order to reach higher-fitness solutions in the long-term . This conflict leads us to hypothesize that the inability of natural selection to optimize mutation rates may depend on the ruggedness of the fitness landscape . The ideal test of this hypothesis requires comparing the evolution of mutation rates on fitness landscapes with and without fitness valleys . This test cannot be performed using the standard Avida setup , owing to the presence of extensive genetic interactions that make the fitness landscape complex and rugged [23] . We therefore modified Avida to allow simple , explicit , user-defined fitness functions that allowed us to manipulate the ruggedness of the fitness landscape ( Methods , Figure 3 ) . Adaptation occurs so fast when using these simple configurations that we also had to make the environment fluctuate between two ‘seasons’ in order to ensure a continual opportunity for beneficial mutations . These fluctuations mean that genotypes that are more fit in one season are less fit in the other ( Figure 3 ) . A quantitative investigation of mutation rates spanning orders of magnitude revealed , once again , that intermediate mutation rates were optimal over the long-term ( Figure 3 ) . We then allowed mutation rates to evolve starting at a genomic mutation rate either below ( 10−5 ) or above ( 1 ) the long-term optimum . Near-optimal values were efficiently selected in those landscapes without a fitness valley or with a narrow valley ( Figure 3 , rows 1 and 2 ) . However , as the width of the valley grew , mutation rates evolved to be orders of magnitude lower than Uopt ( Figure 3 , rows 3 and 4 ) . Fitness values were again used to judge the optimality of mutation rates . With no valleys or with narrow valleys , the average fitness in populations with variable mutation rates was slightly above that of populations with a constant rate of Uopt ( Figure 3 , rows 1 and 2 , Mann-Whitney test , P<0 . 001 in both cases ) , which indicates a small benefit of adjusting mutation rates during evolution [37] . In stark contrast , for wider valleys , the average fitness in populations with variable mutation rates was far below that of populations with a constant rate of Uopt ( Figure 3 , , rows 3 and 4 Mann-Whitney test , P<0 . 001 in both cases ) , confirming that the evolved mutation rates were suboptimal on the these rugged landscapes . These results show that there exists a conflict between short-term and long-term evolutionary strategies on rugged landscapes . In the short-term , low mutation rates are favored because they reduce the load of deleterious mutations , whereas in the long-term , high rates are favored because they increase the chance of producing beneficial mutants . Whether the short-term interests dominate , allowing genotypes with suboptimal mutation rates to spread , should be a function of the expected waiting time until the discovery of a beneficial mutant . To test this prediction , we competed genotypes with either optimal or suboptimal rates in the explicit fitness landscape with a valley size of three ( Figure 3 ) . In one set of experiments , we placed all organisms of both types on the low local fitness peak ( asterisk in Figure 3 ) and let them compete for 300 generations ( the duration of one season in the previous experiments ) . We then repeated the same experiments except that one of the individuals with the long-term optimal mutation rate started on the other side of the valley ( triangle in Figure 3 ) , such that the waiting time for the production of a beneficial mutant was eliminated . A comparison between these two sets of competition experiments shows that the probability that a genotype with a mutation rate that is below the long-term optimum can invade declines significantly when the waiting time to discover beneficial mutants is artificially eliminated ( Table 1 ) . This result illustrates why wider valleys , which create longer waiting times for beneficial mutants , cause the evolution of suboptimal mutation rates . The reader may also notice that the probability of invasion by the genotype with the suboptimal mutation rate was rather small in both sets of experiments ( Table 1 ) . This observation might seem , at first glance , to be at odds with the fact that mutation rates evolved over the long run to be extremely suboptimal ( Figure 3 , rows 3 and 4 ) . This difference makes sense , however , for two interrelated reasons . First , each environmental change that follows the fixation of a mutation on one adaptive peak requires another waiting period for a beneficial mutation , which provides another opportunity for invasion by a genotype with a suboptimal mutation rate that reduces the mutational load . Second , any reductions in the mutation rate become self-reinforcing , as the lower mutation rates make it less likely to generate a beneficial mutant on a distant peak , which increases the expected waiting time for the generation of the next beneficial mutants , thereby increasing the opportunity for a genotype with an even lower mutation rate to invade . Finally , we examined whether the frequency with which the mutation rate changes ( in essence , the mutation rate in the pathway that encodes the mutation rate ) , which we call Π , affects the evolutionarily stable mutation rate . Our intuition was that lower values of Π would make contests between lineages with different mutation rates less frequent , but that the long-term results of many such contests would remain the same . To test this prediction we again used the explicit landscape with a valley size of three . Even when Π varied over four orders of magnitude , it did not affect the final mutation rate that was reached ( Figure 4 ) . Hence , the inability of selection to optimize the mutation rate for long-term adaptation depends on the topology of the fitness landscape , but not on the frequency with which the mutation rate itself changes .
We have shown that mutation rates evolve to near-optimal levels on extremely smooth fitness landscapes . However , if fitness landscapes are rugged , and the maladapted valleys between nearby fitness peaks are wide , then the scarcity of immediately accessible beneficial mutations tips the scale such that short-term selection favors mutation rates that are far below the optimum that would produce the fastest long-term adaptation . Moreover , this process is self-reinforcing because the lower the mutation rate , the less likely it becomes to produce a genotype on the other side of the fitness valley , thereby effectively widening the valley . The digital organisms in the standard Avida configuration used in our first set of experiments exhibit extensive and variable genetic interactions , making the fitness landscape rugged [23] . In those experiments , populations invariably evolved to have mutation rates that were far below the rate that would maximize their long-term fitness gains . We hypothesized that the ruggedness of the landscape was responsible for this inability to optimize their mutation rate for long-term adaptation . In order to test this hypothesis rigorously , we had to change the fitness landscape in Avida from one that is an emergent feature of complex interactions among many instructions to a much simpler surface that could be tuned to be either smooth or rugged . We found that evolving populations were indeed able to achieve mutation rates that maximized their rate of adaptation on smooth landscapes , whereas they became stuck at much lower mutation rates when the valleys between fitness peaks became too large , thus confirming our hypothesis . A growing body of experiments with viruses , bacteria , yeast , and higher eukaryotes shows that epistatic interactions are widespread and vary in their sign and intensity , implying that natural fitness landscapes are also often rugged [35] , [36] , [38] . Thus , our finding that rugged fitness landscapes can impede the optimization of mutation rates for long-term evolutionary adaptation is relevant to the natural world . Our experiments were performed under conditions that were favorable for the optimization of mutation rates . First , the organisms reproduced asexually . Both theoretical [12] , [39] , [40] and experimental work [15] has shown that asexuality facilitates the evolution of elevated mutation rates , because sexual recombination breaks up the linkage between mutator alleles that increase mutation rates and the beneficial mutations that are generated by the mutators . Second , to ensure that beneficial mutations were always available , our experiments used either an environment with more rewarded functions than the organisms ever evolved during a run ( standard configuration ) or a changing environment ( explicit landscapes configuration ) . Third , population sizes were large and strong directional selection was imposed , so that drift was only a minor force in our experiments . Smaller populations might traverse maladaptive valleys more easily , owing to increased drift . However , small populations would be less likely to generate the multiple simultaneous mutations that would allow them to leap across these valleys in a single generation . In populations much larger than those we tested , the probability of an adaptive leap involving multiple simultaneous mutations would increase , but selection should be more powerful in preventing a multi-generation transition across a valley via drift . The effect of population size on the optimal mutation rate , and on the evolution of suboptimal mutation rates , thus remains an interesting area for future investigation . Nevertheless , while the optimal mutation rate and the precise width of the valley that is necessary to cause the evolution of a suboptimal rate may depend on population size , we would not expect that dependency to undermine the general conclusion of this paper , namely , that on sufficiently rugged fitness landscapes , mutation rates will evolve to be suboptimal for long-term adaptation . The inability of evolving populations to optimize their mutation rates for long-term adaptation , even with such favorable conditions , indicates that mutation rates will be suboptimal under a wide range of circumstances , at least when fitness landscapes are rugged and populations are far from a global fitness peak . While novel environments can promote increases in the mutation rate if many beneficial mutations become accessible [1] , [13]–[21] , [40] , our work suggests that this rise will be temporary and , moreover , that even the elevated mutation rates may be suboptimal ( Figure 2B ) . Also , given the difficulty of optimizing mutation rates that we have shown , it seems unlikely that stably high mutation rates , such as those for RNA viruses , are maintained primarily because of the rapid adaptive capacity they bestow , as has sometimes been argued [23] , [41] . Alternative explanations are needed . For example , the evolution of mutation rates is also influenced by the costs of replication fidelity [8] , [23] , and recent work has suggested that this cost might explain the high mutation rates observed in RNA viruses [24] , [42] . We expect that a cost of replication fidelity , all else being equal , will increase the evolved mutation rate . However , we would not expect the resulting increase to cause the optimization of mutation rates in general , although in a few fortuitous situations the cost of fidelity might increase the evolved mutation rate by just enough to push it near the optimal rate . Recent theoretical work by Gerrish et al . [43] has predicted that , contrary to our results , natural selection could favor a self-reinforcing increase in mutation rates in asexual populations . This process would continue even until a population suffered a mutational meltdown and went extinct , because a genotype with an increased mutation rate generates greater numbers of deleterious as well as beneficial mutations . Although not explicitly stated , the prediction of Gerrish et al . [43] of a run-away process toward higher mutation rates appears to assume a smooth fitness landscape . However , as we have shown here , the mutation rate typically evolves to a low value on a rugged fitness landscape , so that the runaway process explored by Gerrish et al . should not occur on such landscapes . Beyond their implications for understanding nature , our findings are also relevant for applied fields that use evolution to improve the performance of biological and computational systems , from molecular and microbial engineering to robotics and evolutionary computation [44] , [45] . Researchers using evolution in computational fields have long sought to use natural selection to adjust mutation rates automatically and “on the fly” , in such a way that would sustain and even optimize long-term adaptation [46]–[48] . These efforts were successful on simple “toy” problems [46] , but became frustrated when applied to more complex problems because self-adaptive mutation rates generally evolved to suboptimal levels [47] , [48] . Our results suggest an explanation: the toy problems had smooth fitness landscapes , whereas the complex problems had rugged landscapes with wide valleys that favored evolutionary conservatism . Our findings also imply that high , fixed mutation rates will often outperform self-adaptive rates on more complex problems , although what the fixed rate should be will depend on the particular problem at hand . In summary , natural selection is not universally effective at optimizing mutation rates for long-term adaptation; in fact , it is very poor in this respect for populations that evolve on complex fitness landscapes . Also , our results caution against making generalizations based on analyses of simple fitness landscapes , whether one is studying natural systems or using evolution for engineering . As we have shown , the mere inclusion of fitness valleys—which are presumably common to the vast majority of fitness landscapes—can yield radically different conclusions from those based on smooth fitness landscapes .
A general description of the Avida software can be found elsewhere [25] . Here , each experiment started with 3 , 600 identical digital organisms . Genome length was held constant at 100 instructions , with 26 possible instructions per site [27] . Reproduction was asexual . To replicate , an organism first had to copy its genome line by line by repeatedly executing the copy instruction; it then had to execute a divide instruction , which took the offspring and used it to replace a random organism from the population . During replication , each genomic instruction could mutate to another with probability μ , the genomic mutation rate being U = 100×μ . All instructions were equally likely to result from any given mutation . The mutation rate was held constant in some experiments , while in others the rate could change by evolving over time . In treatments where the mutation rate could change , μ had a constant and high probability Π of changing by a small amount during any replication cycle . The magnitude of any resulting change was obtained by drawing log2 ( μoffspring/μparent ) values from a Gaussian distribution ( 0 , σ2 ) . For the experiments in which mutation rates were more likely to increase than to decrease , we drew log2 ( μoffspring/μparent ) from a Gaussian ( bσ2 , σ2 ) , where b controls the upward bias , and tested values such that mutation rates were up to ∼1 . 6 times more likely to increase than decrease ( though seemingly small , this bias has a large cumulative effect over many generations ) . Organisms died when another organism's offspring replaced them or when they executed 2 , 000 instructions without producing an offspring of their own . All experiments using the standard configuration lasted 150 , 000 updates . Updates are an arbitrary unit of time in Avida; they represent the time during which each organism , on average , executes 30 instructions [25] . In this configuration , an update corresponded to roughly 0 . 1 generations , although the precise generation time varied depending on the complexity of the evolved organisms' phenotypes . Each organism's phenotype depended on the complex rules that governed how its genomic program was executed , and its fitness depended on the interaction between the resulting phenotype and its environment [25] . More specifically , each organism had a metabolic rate that affected how fast it executed instructions , which , in turn , affected its reproduction rate . The ancestral rate doubled with every rewarded logic function that an organism performed . The ancestral organisms could self-replicate but not perform any other function . The ability to perform logic functions evolved by mutation and selection during each run . An organism's fitness , therefore , represents its expected growth rate relative to others in the population and depended on both its replication efficiency and its ability to perform computations . All fitness values are expressed relative to the ancestor . In reporting fitness data , relative fitness values were first averaged over all organisms in a population , then log10 transformed , and finally averaged over all replicate populations ( independent trials ) in an experimental treatment . To perform logic functions , organisms used inputs consisting of three randomly generated 32-bit strings , which they manipulated to produce an output . The manipulation of these numbers occurred as organisms moved them on and off stacks or between registers by executing instructions such as push , pop , add ( combines the numbers in the two specified registers and places the result in a third ) , shift-r ( bit shift right ) , and so on . A function was rewarded only if the input to output conversion conformed to one of the 77 canonical one- , two- or three-input logic operations . For example , the two-input EQU ( ‘equals’ ) function requires inputting two strings and outputting a third string that had a 1 for each of the 32 bits where both inputs had the same value and a 0 where they differed . Avida runs are inherently stochastic with respect to mutation and death . Therefore , we performed 50 replicate runs for each treatment . Those replicates had identical initial conditions except for a random number seed . That seed affects the outcome of all subsequent stochastic events . The standard and the explicit Avida configurations differed in the instruction set , the fitness calculation and the mode of replication . We modified Avida to mimic a two-allele , 10-locus bit-string model used in a previous study [49] . Genome length was always 10 , while each “instruction” was either A or B; the ancestral genome was entirely A . Fitness depended only on the number of A or B instructions in an organism's genome , according to the seasonal scheme shown in Figure 3 . Every 300 generations the environment fluctuated between the two seasons , and the experiments ran for 15 , 000 generations . We found empirically that fluctuating the environment more or less frequently than every 300 generations produced smaller fitness differences between the optimal fixed mutation rate and suboptimal mutation rates ( data not shown ) . That high mutation rates are most fit at an intermediate rate of environmental change has been previously shown [49] . In the standard configuration , digital organisms had to copy their genomic instructions in order to replicate , and their fitness depended on their speed of replication as well as any rewards they obtained for performing computational functions . Under this alternative configuration , the organisms did not copy themselves , and only the number of A or B instructions mattered to their fitness . The rest of the setup , such as population size , was identical to the standard configuration . All experiments were performed with the Avida software , which can be downloaded for free at http://devolab . cse . msu . edu/software/avida . Default settings were used unless otherwise indicated . | Natural selection is shortsighted and therefore does not necessarily drive populations toward improved long-term performance . Some traits may evolve because they provide immediate gains , even though they are less successful in the long run than some alternatives . Here , we use digital organisms to analyze the ability of evolving populations to optimize their mutation rate , a fundamental evolutionary parameter . We show that when the mutation rate is constrained to be high , populations adapt considerably faster over the long term than when the mutation rate is allowed to evolve . By varying the fitness landscape , we show that natural selection tends to reduce the mutation rate on rugged landscapes ( but not on smooth ones ) so as to avoid the production of harmful mutations , even though this short-term benefit limits adaptation over the long term . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"computational",
"biology/evolutionary",
"modeling",
"evolutionary",
"biology"
]
| 2008 | Natural Selection Fails to Optimize Mutation Rates for Long-Term Adaptation on Rugged Fitness Landscapes |
The evolutionarily conserved nature of the few well-known anti-aging interventions that affect lifespan , such as caloric restriction , suggests that aging-related research in model organisms is directly relevant to human aging . Since human lifespan is a complex trait , a systems-level approach will contribute to a more comprehensive understanding of the underlying aging landscape . Here , we integrate evolutionary and functional information of normal aging across human and model organisms at three levels: gene-level , process-level , and network-level . We identify evolutionarily conserved modules of normal aging across diverse taxa , and notably show proteostasis to be conserved in normal aging . Additionally , we find that mechanisms related to protein quality control network are enriched for genes harboring genetic variants associated with 22 age-related human traits and associated to caloric restriction . These results demonstrate that a systems-level approach , combined with evolutionary conservation , allows the detection of candidate aging genes and pathways relevant to human normal aging .
Aging is a process that affects most living organisms and results in a progressive decline in life function and a gain in vulnerability to death [1] . In humans , aging is the main risk factor in a wide spectrum of diseases . The recent increase in human healthspan , also called ‘normal’ , ‘disease-free’ , or ‘healthy’ aging , is mostly due to improved medical care and sanitation [2 , 3] . While model organism results are critical to aging research , the identification of evolutionarily conserved molecular mechanisms remains a challenge . On the one hand , broad processes appear conserved , as illustrated by the consistent effect of caloric restriction in extending healthy lifespan in diverse species [4–8] . These common processes have led to the definition of general "hallmarks of aging" [9] , such as mitochondrial dysfunction , telomere attrition , or genomic instability . On the other hand , there appears to be very little conservation of orthologous genes in changing expression in aging [10–12] . Thus , although significant efforts have been made to uncover the identity of genes and pathways that affect lifespan , it is unclear to what extent the functional information of aging obtained from model organisms can contribute to understanding human aging . The main model species , fruit fly and nematode worm , are mostly post-mitotic [13–16] , whereas most human tissues are proliferative , which affects aging patterns [17 , 18] . Focusing on the process of aging in healthy individuals should improve the discovery of pathways important in natural aging . Yet this poses an additional challenge to the characterization of molecular mechanisms , as molecular changes influencing healthy aging are much more subtle than those in disease [19] . As gene-level studies are not sufficient to elucidate complex processes such as aging , systems-level analysis of large datasets is an important tool for identifying relevant molecular mechanisms . The integration of various data types contributes to identify pathways and marker genes associated with specific phenotypes [20 , 21] . Notably , co-expression network analyses can help to elucidate the underlying mechanisms of various complex traits [22 , 23] . To couple both evolutionary and functional age-related information , we integrated transcriptome profiles of four animal species from young and older adults: H . sapiens , M . musculus , D . melanogaster and C . elegans . As a source of gene expression , we used human data from the large-scale Genotype-Tissue expression ( GTEx ) project [24] , focusing on two post-mitotic organs , together with aging transcriptomes of model organisms . We identified the functional levels of conserved genetic modifiers important during normal aging , and related them to caloric restriction experiments and enrichments in age-related genome-wide association studies ( GWAS ) . We used gene families as evolutionary information across distant species in a two-step approach to observe age-related conserved mechanisms . Our results show how integrating both network information and evolutionary conservation is informative for a complex phenotype , such as aging . We strengthen the case for a conserved role of proteostasis in normal aging and in the reaction to dietary restriction .
We used an approach in three steps to integrate transcriptomes across distant species ( human and model organisms ) and to identify evolutionarily conserved mechanisms in normal aging ( Fig 1A ) . In the first step , we performed differential expression analysis between young and old samples in two tissues , skeletal muscle and hippocampus , from humans ( Homo sapiens ) and mice ( Mus musculus ) , and in whole body for the fly ( Drosophila melanogaster ) and the worm ( Caenorhabditis elegans ) . It was not possible to incorporate sex into the integration of aging effects between species , as data from both sexes were available only in human ( S1 Table ) . We also used transcriptome datasets related to caloric restriction in these species for validation . In the second step , we obtained 3232 orthologous sets of genes , ‘orthogroups’ , across those four species ( see Methods ) . Each orthogroup is defined as the set of the orthologous and paralogous genes that descended from a single ancestral gene in the last common ancestor to those four species ( H . sapiens , M . musculus , D . melanogaster , C . elegans ) and an outgroup species ( Amphimedon queenslandica ) . Each orthogroup can contain a different number of genes , and was treated as a single functional meta-gene common to four species . We corrected for the orthogroup sizes by applying Bonferroni correction on the gene p-values from differential expression analysis within the orthogroup . Then , we selected a representative gene per species within orthogroups . We took the minimum Bonferroni adjusted p-value of a species-specific age-related gene from differential expression analysis . This allowed us to build ‘age-related homologous quadruplets’ ( see Details in S1A Fig ) . The four p-values within each quadruplet were then summarized into a single p-value per quadruplet , by using Fisher’s combined test . We obtained 2511 gene quadruplets in skeletal muscle , 2800 in hippocampus , and 1971 in caloric restriction experiments ( S4 Table ) . We characterized their biological relevance by functional enrichment . In the third and final step , those quadruplets of age-related genes were used to build a co-expression network per species ( S1B Fig ) . These networks were then integrated together using order statistics into one cross-species age-related network . We performed community search algorithm on this network to obtain age-related and evolutionarily conserved modules . The modules were then tested for functional enrichment and for enrichment in GWAS hits . To study normal aging , we restricted ourselves to transcriptomic studies with at least one young adult and one old adult time-point , adult being defined as after sexual maturity ( Fig 1B ) . Transcriptomes had to come from control samples ( model organism datasets ) or from the GTEx dataset ( humans ) , which excludes notably HIV infection , viral hepatitis , and metastatic cancer [25] . While we cannot guarantee that these samples are from healthy individuals , they do represent “normal” aging , as opposed to studies of aging diseases . We defined young and old adults across species as follows: young: 3–4 months for M . musculus , 2–10 days for D . melanogaster , 3–6 days for C . elegans; old: 18–24 months for M . musculus , 20–50 days for D . melanogaster , 10–15 days for C . elegans . For the GTEx data , samples from all adults ( 20–70 years old ) were taken into account in a linear model to detect differentially expressed genes . In human and mouse , we focused on two tissues , skeletal muscle and hippocampus , because they are known to be profoundly affected by aging , and like most fly and nematode tissues are post-mitotic . During aging , skeletal muscle is affected by sarcopenia [26] . Changes in hippocampus function have a significant impact on the memory performances in elderly people [27] . Thus both tissues are susceptible to aging-related diseases . For human , we used transcriptomes of 361 samples from skeletal muscle tissue and 81 samples from hippocampus from GTEx V6p . For the other species we used diverse publicly available transcriptomic datasets ( information including sample numbers and data sources in S1 Table ) . The sample sizes for model organisms were variable , from 3 to 6 replicates per time-point . In order to compare samples between young and old age groups , we fitted linear regression models for each dataset . In addition , in the GTEx dataset we controlled for covariates and hidden confounding factors to identify genes whose expression is correlated or anti-correlated with chronological age , taking into account all samples ( see Methods ) . We observed uneven distributions of up- and down-regulated genes with aging across different species and datasets ( Fig 1C , S2 Table ) , suggesting variable responses to aging and different power of datasets . The human hippocampus shows substantially more age-related gene expression change than skeletal muscle ( 6083 vs . 5053 differentially expressed genes , FDR < 0 . 1 ) . However , mouse hippocampus shows less gene expression change than skeletal muscle ( 1639 vs . 2455 differentially expressed genes , FDR < 0 . 1 ) . These differences are due in part to the smaller sample size of the mouse skeletal muscle study . We limited our analysis to genes that were expressed in at least one age group , leading to detection of 15–40% of genes that exhibits age-related gene expression changes . Of note , these changes are often very small , typically less than 1 . 05 fold in human and less than 2-fold in animal models . It has been previously reported that there is a small overlap of differentially expressed genes among aging studies [28 , 29] . To make results easily comparable across species , the young and old adults of one species should correspond to young and old adults of another species [30] . Our clustering shows good consistency across age groups of samples between species , based on one-to-one orthologous genes with significant age variation ( FDR < 0 . 05 ) ( Fig 2A , S2 Fig ) . Yet there is a low overlap of one-to-one orthologous genes with significant expression change in aging ( S3 Table ) . This observation is in line with two studies showing that the overlap between individual genes associated with aging did not reach the level of significance [10 , 31] . To go beyond this observation , we correlated log-transformed fold change ( old/young ) between human and model organisms . We observed weak pairwise correlations ( S3 Fig ) when comparing single genes . This indicates that most transcriptional changes on the gene level are species-specific , and that there is little evolutionary conservation to be found at this level . To assess the age-related gene expression changes on a functional level per species , we performed gene set enrichment analysis ( GSEA ) using gene ontology ( GO ) annotations [32 , 33] . We then selected GO terms that we grouped into broader categories ( S4 Fig ) . We privileged sensitivity over specificity at this step ( FDR < 0 . 20 ) to highlight broad trends; thus results concerning individual GO terms should be treated with care . All species showed a general pattern of down-regulation of metabolic processes . The pattern of metabolic down-regulation was stronger in muscle for both human and mouse , while in hippocampus there was down-regulation of nervous system processes . This confirms that there is a tissue-specific signal in normal aging . Due to small samples size of the mouse skeletal muscle dataset , we were able to detect only down-regulated metabolic processes . In addition to metabolism , we observe strong immune systems response to aging in most samples . These results are consistent with known links between metabolism , immunity and aging [34] . We then aggregated processes on the functional level across four species using evolutionary information to observe common age-related mechanisms rather than tissue-specific mechanisms . We integrated differential expression analysis from each species , as described above . We obtained 2010 genes in skeletal muscle / whole body , 2075 genes in hippocampus / whole body , and 1962 genes in caloric restriction experiments ( Fisher combined tests , FDR < 0 . 10 ) . We examined their biological relevance using Gene Ontology enrichment analysis ( GEA ) based on human annotation . We did not take into account whether the processes that are shared across species are regulated in the same direction , but rather whether they are consistently perturbed during aging . We obtained 100 significant GO terms ( FDR < 0 . 05 ) related to biological processes , and aggregated them into broader GO categories . While our species-specific analysis mostly shows tissue-specific pathways , we found that terms with an evolutionarily conserved relation to normal aging are strongly enriched for processes involved in proteostasis , or protein homeostasis . The proteostasis-linked processes showed to be more conserved than expected by chance ( S6 Fig ) . The other conserved processes are related to transport , translation , transcription and post-transcriptional modifications , and protein ubiquitination ( Fig 2B , S5 Table ) . We also confirmed previously known evolutionarily conserved age-related pathways , such as cellular respiration and immune response . Integrating caloric restriction datasets across the four species showed enrichments in similar processes ( Fig 2B ) . While most of the shared processes have been previously linked to aging , we focused on proteostasis and related processes . To characterize in more detail the specificity of proteostasis-linked processes ( Fig 2C ) , we investigated their enrichment strength in the large human GTEx dataset . Since proteostasis perturbation is detected both through the GO domains of cellular localization and of biological process , we investigate these two domains . We show here the results from the skeletal muscle GTEx dataset , but similar results are observed in the hippocampus GTEx dataset ( S6 Fig , S6 Table ) . The most enriched cellular component terms in skeletal muscle were related to proteasome complex ( GO:0000502 , enrichment score: 1 . 99 ) and to mitochondrial matrix ( GO:0005759 , enrichment score: 1 . 38 ) . We also observed strong enrichment of ribosomal large ( GO:0000027 , enrichment score: 1 . 57 ) and small subunit ( GO:0000028 , enrichment score: 1 . 84 ) , of protein homotetramerization ( GO:0051289 , enrichment score: 1 . 18 ) , and of GO biological processes that are part of the protein quality control network . Overall , the translation and proteasome complexes appear to be the parts of the protein quality control network whose involvement in aging is both evolutionarily conserved across different species , and significant in human normal aging . Interestingly , we also detect mRNA splicing as a part of the conserved processes between species . The direction of the changes in conserved proteostasis processes in human is consistent with a relation between loss of proteostasis and aging ( Fig 3 ) . Although macroautophagy did not show a strong enrichment score in the Fig 2C ( GO:0016236 , enrichment score: 0 . 90 ) , there is down-regulation of the conserved genes associated with macroautophagy ( Fig 3A ) , translation ( Fig 3B ) , and the proteasome complex ( Fig 3C ) , which are important in the protein quality network . Similar results are observed in hippocampus , although not with a signal as strong as in skeletal muscle ( S7 Fig ) . The changes during normal aging in both tissues are rather subtle but significant ( Fig 3 , S7 Table ) . To characterize age-related processes at a systems-level and to prioritize conserved marker genes associated with normal aging , we constructed probabilistic networks . These were based on prioritization of co-expression links between conserved age-related genes across four species . The conserved age-related genes became nodes in the multi-species network . Thus the connections between those genes will be based on evolutionary conservation , and prioritized according to the their co-expression in each species . Our integrative network analysis initially identified 20 and 14 modules for skeletal muscle and hippocampus , respectively . We randomized our networks 100 times based on the same number of conserved genes per experiment and obtained significantly higher numbers of gene-gene connections than in the original network ( permutation test , p = 0 . 0198 ) ( S9 Fig ) . Thus aging networks appear to be lowly connected . We focused only on the modules larger than 10 genes; there were 12 such modules per tissue . These modules ranged in size from 16 ( M7 hippocampus ) to 191 genes ( M12 hippocampus ) ( Fig 4A and 4B , S8 Table ) . The networks were summarized to module level ( module as a node ) , and we observed strong inter-modular associations . This analysis provided several levels of information . First , it provided a small number of coherent gene modules that represent distinct transcriptional responses to aging , confirming the existence of a conserved modular system . Second , it detected conserved marker genes affected during aging , discussed below . To determine which of the conserved aging-associated modules are related to the main components of proteostasis network , we carried out functional enrichment analysis on these modules , based on human gene annotations . The enrichments were highly significant for all modules ( FDR < 0 . 01 ) , and confirmed the inter-modular associations ( S8 Table ) . Not all of the modules were related to proteostasis . Interestingly , M1 , M10 and M5 in the skeletal muscle network share strong associations with mitochondrion organization and distribution , regulation of cellular amino acid metabolic process and ubiquitin protein catabolic process , while M2 and M3 in hippocampus share associations with different types of protein transport . Other modules ( M1 , M6 , M7 , M8 , M11 , M12 in skeletal muscle; M2 , M3 , M4 , M5 , M12 in hippocampus ) support the impact of aging on genes related to the proteostasis-linked processes . This included processes related to protein polyubiquitination ( GO:0000209 ) , translational initiation ( GO:0006413 ) , protein transport ( GO:0015031 ) , regulation of macroautophagy ( GO:0016241 ) , and proteasome-mediated ubiquitin-dependent protein catabolic process ( GO:0043161 ) . In skeletal muscle tissue there were also a strong enrichment in splicing process ( M3 ) . Moreover , the connection between M2 , M10 and M6 in hippocampus , and between M1 , M5 and M12 in skeletal muscle indicates that there is a connection between mitochondrial and proteostasis-related processes . We also performed enrichment analysis based on genes coming from 22 GWAS studies ( See Methods , S9 Table ) . While results should be taken with caution because of the high associated FDR , these modules do appear enriched in metabolic or age-related diseases ( Fig 4 ) . To further characterize these modules , we studied how conserved modular genes associated with proteostasis and age-related GWAS diseases are changed in expression in humans , as a long-lived species . We looked deeper into the gene composition of two modules , M1 associated with SCF-dependent proteasomal ubiquitin-dependent protein catabolic process ( 79 genes ) and M4 associated with positive regulation of telomerase RNA localization to Calaj body ( 155 genes ) from the skeletal muscle and hippocampus networks , respectively . We defined network hubs , genes that exhibit a significantly high number of connections with other genes in the network , for each of these modules in muscle ( Fig 5A and 5B ) and hippocampus ( S10B Fig ) . We focused on the hubs with the highest scores in each module and examined their neighborhood . The top ranked genes in M1 of the skeletal muscle were CTSK , UBE2L3 and CPA3 ( Fig 5A ) . They are associated with protein quality network , related to protein degradation . Interestingly , the neighboring genes PSMB2 and PSMA1 are associated with the proteasome complex ( Fig 5A ) . The top ranked genes in M4 in skeletal muscle were related to the translational initiation process , with MAPRE3 , SPTBN2 and ATP6V0A1 as hub genes . Their network neighbors were tightly connected to the cytoskeleton and protein transportation ( Fig 5B ) . Other modules also show links to metabolism and to proteostasis . For example muscle module M12 and hippocampus module M3 are associated with the protein polyubiquitination process ( S10 Fig ) . The top-ranked hub genes in muscle M12 were DDX3X , KIF5B and USP7 ( S9A Fig ) . Those genes are related to DNA damage , translation and transport regulations in the cell . In the hippocampus module M3 ( S9B Fig ) , the three hub genes ( PPP3CB , DNM1L and ITFG1 ) are involved in hydrolase activity , apoptosis and programmed necrosis and modulating T-cell function . Although the hub genes with the highest scores were strongly related to metabolism and to tissue-specific functions in each of these two modules , their network neighborhood is associated with the protein quality control network . More specifically , the PSMB5 and PSMD3 genes are related to the proteasome complex and are connected to hub genes . We combined this hub gene analysis with GWAS association gene scores , and observed that PSMB5 , UBE2L3 , and PSMD3 ( Fig 5C , S9 Table ) are important in many age-related diseases or phenotypes , such as Alzheimer’s disease , HDL cholesterol , LDL cholesterol , triglycerides , and insulin resistance . Other genes related to translation and proteasome complex were also strongly associated to such diseases , such as PSMB5 with multiple sclerosis ( Pascal [35] gene score: p-value = 0 . 0348 ) and HDL cholesterol ( Pascal gene score: p-value = 0 . 0155 ) . Finally , we observed that the prioritized genes associated with age-related diseases from conserved functional modules change in opposite directions with normal aging and with caloric restriction ( Fig 5D ) . This differential expression is consistent with a causal role in these age related diseases , given the attenuating effect of caloric restriction on aging . We analyzed the association of the expression levels of candidate genes with lifespan in different tissues of mouse recombinant inbred lines used for population genetics analyses , such as the BXD [36 , 37] and LXS [38] strains . We observed an inverse correlation between lifespan and the transcript levels of not only PSMB5 but of many genes known to encode proteins of the proteasome complex ( Fig 5E ) in the hippocampus and spleen of the BXD strains and in the hippocampus and prefrontal cortex of LXS lines . Moreover , the GSEA showed negative correlation between the expression level of genes from the proteasome complex in those tissues and lifespan of the mice ( Fig 5F ) .
The challenge of detecting underlying mechanisms of normal aging that are evolutionarily conserved is thought to be a key impediment for understanding human aging biology [19] . In this work , we integrated evolutionary and functional information of normal ( non pathological ) aging gene expression to identify conserved age-related systems-level changes . We identified conserved functional modules by integration of co-expression networks , and we prioritized genes highlighted by GWAS of age-related diseases and traits . The observations on several functional levels allowed us to highlight the role of proteostasis , which includes all processes related to protein quality control network , as a strong core process associated with normal aging , at least in post-mitotic tissues . On the one hand , previous observations restricted to a small number of evolutionarily conserved genes with large effects in aging , or in age-related diseases , provide some evidence that aging mechanisms might be conserved among animals [28] . On the other hand , transcriptome level correlations of expression changes in aging between species are very low in our gene-level results , as in the literature [10–12] , which indicates low conservation . Yet the process of aging appears overall conserved , with notably common effects of interventions , such as caloric restriction , showing similar effects across species ranging from nematodes , flies to mammals [8 , 39] . The solution to this apparent paradox seems to be that pathways are evolutionarily conserved in aging [10] , even when single genes are not . Indeed , we have found strong similarities in age-related gene sets between human and other species . Of note , we restricted ourselves to two post-mitotic organs , and conservation patterns might be different between proliferative human tissues and fly or nematode . The fact that we compared whole body to specific organs remains a limitation of the available data , which we could not correct for . Moreover , we were only able to correct for sex effects on aging in human [40 , 41] . The mice samples are all male , whereas the fruit fly samples are all female ( detailed in S1 Table ) ; nematode worms are hermaphroditic . We might be missing an evolutionarily conserved signal if it is confounded by sex-bias in the available data . Other limitations include the large evolutionary distances and the small number of species sampled , which include only one long-lived species ( human ) . Because of this we might have missed aging signals which are not conserved over bilaterian evolution , or not conserved between short lived and long lived species . A final limitation is that the older model organisms were close to the limit of their life expectancy in captivity , whereas humans of 60–70 years old , while aging , are not at the limit of life expectancy of humans . Beyond individual pathways , the modular nature of aging has been previously reported at several levels , such as by protein-protein interaction network analysis during human and fruit fly brain aging [23] and blood over human aging [42] , human longevity network construction and identifying modules [43] , mouse age-related gene co-expression modules identification [44] , or aging and age-related diseases cluster detection in human aging [45] . Integrating co-expression networks across species , we identified 10 and 13 evolutionarily conserved functional modules for skeletal muscle and hippocampus , respectively . These conserved modules are not only enriched in processes known to be involved in normal aging , such as immune-related pathways , they significantly overlap with results from age-related GWASs . The latter is all the more interesting that finding causality for aging in GWAS is difficult , given its highly multifactorial nature [46] . Of note , these modules can be tissue-specific , for example related to energy and amino acids in muscle . Thus , aging is an evolutionarily conserved modular process , and this modularity is tissue-specific . An advantage of our approach is that it allows us to detect with good confidence processes whose changes in aging are quite subtle . This is important because normal aging is not a dramatic process , akin to embryonic development or cancer , but a gradual change in tissues and cell types which keep their defining characteristics . In other words , old muscle and young muscle are very similar at the molecular level , as shown by the log-fold change scale in Fig 3 . Yet we are able to detect processes associated to these changes with strong confidence , and these processes are mostly known in to be age-related . The largest changes , thus easiest to detect , include metabolism [47] , transcription [48] , translation [49] , and immune response . Changes in expression for proteostasis related genes are weaker , yet integrating at a systems-level between species provided us with a strong signal . More broadly , our results strengthen the case for further investigation into the molecular program that links proteostasis to normal aging . Indeed , one of the major hallmarks of aging is the loss of proteostasis . Loss of proteostasis is related to major human pathologies , such as Alzheimer’s and Parkinson’s disease , offering an opportunity to detect conserved candidate genes important in those age-related diseases [50] . The proteasome complex , one of the three major mechanisms of proteostasis [51] , is maintained throughout the life of the long-lived naked mole rat [52] . Perturbation of components of proteostasis has been shown to extend the lifespan of mice [53] . These observations on short and long lived rodents are consistent with a role of proteostasis in lifespan of these species [54] . Finally , caloric restriction , defined as a reduction of regular caloric intake by 20–40% , extends lifespan and delays the onset of age-related diseases in many species [4 , 6–8 , 55 , 56] , in part through effects on proteostasis networks . Moreover , “loss of proteostasis” is one of the nine proposed hallmarks of aging [9 , 57] . It is possible that this marker is stronger because of our focus on aging in post-mitotic organs . Aging involves a deregulation of the protein quality control network , and this is conserved between distant species . Changes in protein synthesis and protein degradation processes of proteostasis system may be fundamental to the response to normal aging because the accumulation of somatic and germline mutations can alter fine modulation of the protein homeostasis network and produce pathological alterations . Thus proteostasis provides a link between somatic genome-level changes and the phenotypic impact of aging . During normal aging , the alterations in proteostasis networks are rather subtle and discrete , by contrast to the strong down-regulation of metabolic processes . This suggests that perhaps there is a cascade of triggered pathways as aging proceeds . Moreover , we detect evolutionarily conserved links inside modules between mitochondrial deregulation ( hub genes ) and protein homeostasis ( neighboring genes ) in normal aging , consistent with a recent report [58] . The main evolutionarily conserved gene candidates from proteostasis , PSMB5 and PSMD3 , are related to the proteasome . These two genes were tightly connected to metabolic hub genes in skeletal muscle and to filament organization genes in the hippocampus . The proteasome complex is down-regulated during aging in our results , and in a transgenic mouse mutant proteasome dysfunction led to shorter lifespan [59] . Moreover , both genes showed significant association in GWAS studies with metabolic and disease traits . Although only the PSMB5 gene was an experimentally validated candidate gene in mice , the PSMD3 gene was related with coronary artery disease , HDL cholesterol and fasting proinsulin , and would also be worthwhile to explore further . The association with caloric restriction studies strengthens the relevance of the processes we report . We observed that gene-set signal was both evolutionarily conserved in caloric restriction , and shared between normal aging and caloric restriction experiments . Genes related to proteostasis change expression in opposite directions between human aging and caloric restriction . This indicates that these functions are maintained during caloric restriction in humans , and strengthens the case for a causal link between proteostasis and normal aging . Our observations are consistent with previous research in C . elegans , reporting improvement of proteostasis during caloric restriction treatments and extension of the lifespan [60 , 61] . Notably , PSMB5 and PSMD3 follow this trend in caloric restriction relative to aging , further suggesting that they are prime candidates to study transcriptional regulators underlying functional modules in normal aging . Integrating biological processes based on evolutionary conservation allows distinguishing relevant signals from noise , despite the weak patterns in aging transcriptomes . Moreover , the fact that a same process is involved in aging in very different species strengthens the case for causality . This provides a promising foundation to search for relevant biomarkers of healthy aging of specific tissues . In summary , the large-scale , comprehensive gene expression characterization in our study provided insights in underlying evolutionarily conserved mechanisms in normal aging . While metabolic and certain tissue-specific pathways play a crucial role in aging , processes affecting the protein quality control network show weaker but very consistent signal . Using both evolutionary and functional information , we detected evolutionarily conserved functional modules allowed us to identify core proteostasis-related genes . These genes were implicated as important hits in age-related GWAS . Together , the integrative systems-level approach facilitated the identification of conserved modularity of aging , and of candidate genes for future normal aging biomarkers .
To obtain a representative set of aging gene expression experiments , a set of raw RNA-seq and microarray datasets of four species ( H . sapiens , M . musculus , D . melanogaster , C . elegans ) were downloaded from the GEO database [62] and SRA database [63] ( S1 Table ) . For observing aging gene expression signatures in human and mouse , we selected hippocampus and skeletal muscle tissues . The aging gene expression experiments for fly and worm were available as whole-body experiments . All the normal or control samples came from two extreme age groups ( young and old adults ) that are counted from sexual maturity . This corresponds to 20–30 years old humans , 3–4 months old mice , 4–5 days old flies and 3–6 days old worms ( see Fig 1B ) in young adults . In old adult age group , this corresponds to 60–70 years old humans , 20–24 months old mice , 40–50 days old flies and 12–14 days old worms . The sample size per age group was 3–6 replicates . The GTEx V6p read counts were used as H . sapiens aging experiment ( V6p dbGaP accession phs000424 . v6 . p1 , release date: October , 2016 ) . The information about the sample ages was obtained through dbGAP annotation files of the GTEx project ( restricted access ) . Two RNA-seq datasets were matched for M . musculus and C . elegans; and the microarray platforms included were from Affymetrix: Mouse 430 A/2 . 0 , GeneChip Drosophila Genome array and C . elegans Genome array . From the downloaded GTEx V6p data , we extracted the gene read counts values for protein-coding genes by using Ensembl ( release 91 ) . For each tissue , the lowly expressed genes were excluded from data analysis according to the GTEx pipeline [24] . Prior to the age-related differential expression analysis , we used the PEER algorithm [64] in a two-step approach to account for known covariates as well as for hidden factors present in GTEx V6p data per tissue . From covariate files ( Brain_Hippocampus_Analysis . covariates . txt and Muscle_Skeletal_Analysis . covariates . txt ) , we used information about the three genotype principal components . From phenotype file ( phs000424 . v6 . pht002742 . v6 . p1 . c1 . GTEx_Subject_Phenotypes . GRU . txt ) , we used information about age , gender , ischemic time and BMI information . From attribute file ( phs000424 . v6 . pht002743 . v6 . p1 . c1 . GTEx_Sample_Attributes . GRU . txt ) , we extracted information about the sample associations with interested tissues , hippocampus and skeletal muscle . In the first step , the PEER algorithm discovers patterns of common variation; it created 15 and 35 assumed global hidden factors for hippocampus and skeletal muscle , respectively . In addition to global hidden factors , we provided age , BMI , sex and ischemic time as known covariates in PEER model . In the second step those hidden factors ( gene expression principal components ) that showed significant Pearson’s correlation coefficient with age ( p-value < 0 . 05 ) were excluded . The number of hidden factors that did not significantly correlate in hippocampus was 7/15 and in skeletal muscle were 22/35 that were selected for further linear model analysis . The sum of remaining hidden factors and known covariates were included in a linear regression model to obtain the genes differentially expressed during age in GTEx V6p data for each tissue ( Formula 1 ) . Yji=μ0+αjAgei+γjSexi+βjBMIi+θjIschemictimei+∑k=1nδjPCki+ϵi [1] where , Yji is the expression of a gene j in a sample i , where Age , Sex , BMI , Ischemic time of sample i , with their regression coefficients α , γ , β , θ . PCki ( 1 < k < n ) is the value of the k-hidden factors for the i-th sample with regression coefficient δ; n is a total number of factors that was not correlated with age , εi is the error term , and μ0 is the regression intercept . If α > 0 , gene j was treated as up-regulated , otherwise gene j was treated as down-regulated . The linear model ( Formula 1 ) was performed in limma voom , and the p-values were corrected for multiple testing by performing false discovery rate ( FDR ) correction using Benjamini-Hochberg method . Across this study , we have not fixed one threshold of p-value or of FDR ( false discovery rate ) . One can select entities , such as genes , based on any arbitrary FDR threshold depending on what is done with the selected set . When treated as a group one can make statements about properties of the selected entities , for example enrichment analysis ( e . g . [65] ) or use them for prediction ( e . g . Sup . Fig 10 of [66] ) . When the selected set of entities is not the end-point of the analysis , but is statistically tested for a property , any threshold can be applied and the enrichment test has to be controlled for type I error ( or FDR ) . An advantage of the FDR is that its interpretation relative to a list of results is intuitively clear . In most cases , we use an FDR of 10% , which is a good compromise between sensitivity and specificity . We deviate from this towards a higher FDR of 20% when we want to detect broad trends but not analyze specific results in detail , and towards a lower FDR of 5% at steps which provided the genes which were critical to the downstream analyses . For further discussion of statistical cut-off choices , see notably [67] . For microarray datasets ( both aging and caloric restriction experiments ) from skeletal muscle of M . musculus and whole-body of D . melanogaster , raw Affymetrix . CEL files were downloaded from the GEO database and preprocessed using RMA normalization algorithm [68] ( S1 Table ) . In case of multiple probes mapping to the genes on the array , the average of the probes was taken in further analysis . The annotation was used from Ensembl release 91 . In order to identify the features that exhibit the most variation in the dataset , principal component analysis ( PCA ) was performed on the expression matrices to detect outlier samples , gender and other batches . For RNA-seq datasets from two model organisms , M . musculus and C . elegans , the . sra files were downloaded from the SRA database [63] . Both datasets were sequenced on Illumina HiSeq 2000 with read length 50nt . The reads were mapped to species-specific reference genomes ( M . musculus: GRCm38 . p5 , C . elegans: WBCel235 ) using kallisto v0 . 43 . 1 ( for index building: kallisto index–i genome . idx genome . cdna . all . fa ( k-mer = 31 , default option ) ; for mapping: kallisto quant -i genome . idx–o output . file–single–l 200 –s 20 single . end . fastq . file ) [69] . Both M . musculus and C . elegans had single-end RNA-seq libraries in the experiments ( S1 Table ) . The transcript abundances were summarized at the gene-level [70] . For both species , we used GTF gene annotation files that were downloaded from Ensembl ftp site ( release 91 ) [71] . The transcript abundances were summarized at the gene-level to lengthscaledTPMs using tximport v1 . 6 . 0 [70] and used as an input to limma voom . The gene-level read counts were further analyzed in R v3 . 4 . 3 . The read counts were normalized by total number of all mappable reads ( library size ) for each gene . The limma voom results in a matrix of normalized gene expression values on log2 scale . The counts and normalized log2 limma voom expression values were used as a raw input for all the analysis . Outlier samples were checked by principal component analysis . For each species , genes that showed expression below 1 count per million ( cpm < 1 ) in the group of replicates were excluded from downstream analysis . To be able to obtain differentially expressed genes from different experiments that were normalized , we had to account for the possible batches present . Since we are not aware of all the batches in the studies , we used Surrogate Variable Analysis ( SVA ) to correct for batches [72] in microarray data analysis . The SVA method borrows the information across gene expression levels to estimate the large-scale effects of all factors absent from the model directly from the data . After species-specific expression matrices were corrected , they served as input into linear model analysis implemented in limma ( Affymetrix ) or limma voom ( RNA-seq ) [73] , for finding age-related differentially expressed genes between two extreme aging groups , young and old . Briefly , limma uses moderate t-statistics that includes moderated standard errors across genes , therefore effectively borrowing strength from other genes to obtain the inference about each gene . The statistical significance of putatively age-dependent genes was determined with a false discovery rate ( FDR ) of 10% . The GEO database was used to download caloric restriction datasets ( S1 Table ) . Only muscle tissue was available in H . sapiens , therefore we selected correspondingly muscle tissue in mouse , but whole body in fly and worm . The datasets were normalized using RMA normalization algorithm [68] ( S1 Table ) . In case of multiple probes mapping to the genes on the array , the average of the probes was taken in further analysis . The annotation was used from Ensembl release 91 . To call differentially expressed genes , we used limma between caloric restriction and control samples . The statistical significance of putatively age-dependent genes was determined with a false discovery rate ( FDR ) of 5% . For deriving one-to-one orthologs , human genes were mapped to the homologs in the respective species using biomaRt v2 . 34 . 2 . After detection of significant age-associated differentially expressed genes , we overlapped one-to-one orthologous genes between the species in order to observe the consistency of age groups between species . We took the limma voom corrected expression matrix for GTEx V6p and the expression matrices of model organisms , and selected only genes that were differentially expressed with an FDR of 5% . We then accounted for the laboratory batch effect by applying Combat on expression matrices [74] . To examine the relationship between aging in human and model organisms on single-gene level , we mapped one-to-one orthologous genes from human to model organisms and between the organisms downloaded from Ensembl [71] . We calculated Spearman correlations between sets of matched differentially expressed orthologous genes , between log2 fold-changes ( Supplementary S2 Fig ) . No cutoff for fold change was used . We downloaded hierarchical orthologous groups ( HOGs , in further text referring to orthologous groups ( OG ) ) across four species from the OMA ( orthologous matrix analysis ) database [75] at the Bilatera level ( Amphimedon queenslandica ( Cnidaria ) was used as a metazoan outgroup ) , which resulted in 3232 orthologous groups . Briefly , hierarchical orthologous groups are gene families that contain orthologs ( genes related by speciation ) and in-paralogs ( genes related by duplication ) at the taxonomic level which orthologous groups were defined . The sizes of orthologous groups in this study range from 4 to 246 genes . We filtered age-related genes per orthologous group per species in order to obtain representative species-specific genes per group . The genes within orthologous group were selected according to the P values from differentially expression analysis [76] . We applied Bonferroni correction on each orthologous group to the differential expression P values in order to correct for the size of the orthologous group . We then combined the corrected differential gene expression P values across species using Fisher’s combined probability test generating a new P value from χ2 distribution with 2k degrees of freedom ( Formula 2 ) . −2∑i=1kln ( Pi ) ∼χ2k2 , [2] where Pi is species-specific gene P value from differential expression analysis within a OG . We adjusted combined Fisher P values for multiple testing , and filtered orthologous groups with FDR of 10% for further analysis . This resulted in 2010 and 2075 common OGs for skeletal muscle and hippocampus , respectively . In caloric restriction experiments , we detected 1962 common OGs . We performed general GO enrichment analysis using Fisher’s test ( topGO R package ) on significant orthologous group genes and based on human gene set annotation to find functional enrichment of OGs in GO ‘biological process’ terms . To summarize the significantly enriched top 100 GO terms into main ones , we used the Wang GO semantic similarity method [77] that takes into account the hierarchy of gene ontology , and performed hierarchical clustering ( 11 clusters for skeletal muscle and 13 clusters for hippocampus , 10 clusters for caloric restriction ) on the semantic matrix for both aging and caloric restriction experiments ( S5 Table ) . The clusters were then named according to the common term of the cluster . We associated proteostasis-linked processes to GO terms associated with ‘translation’ , ‘protein folding’ , ‘proteasome assembly’ , ‘macroautophagy’ , ‘proteasome complex’ , ‘endoplasmic reticulum’ , ‘lysosome’ and others . To perform the randomizations , we selected random genes from the differential expression matrices with the same number as the number of orthologous groups selected for skeletal muscle and hippocampus . The p-values associated with the random genes per species were then combined with the Fisher’s combined test . The GO enrichment analysis was performed as for the observed data with focus on the ‘biological process’ and based on the human annotation . The procedure was repeated 100 times ( S6 Fig ) . We aimed to detect gene sets that are perturbed in aging in different species . We selected the genes from previously formed significant age-related OGs per species and constructed the species-specific co-expression networks by calculating Pearson correlation coefficient between age-related OGs genes . In the resulting species-specific co-expression network , nodes represent genes and edges connect genes that are above a set significant threshold from Pearson correlation calculation ( P value < 0 . 05 ) . Only positively correlated genes were taken into account , while the negatively correlated genes and genes correlating under the threshold were set to zero . Negatively correlated genes might be interesting to detect complex regulatory patterns , but are beyond the scope of this study . The cross-species network was obtained as follows [78] . Each co-expression link was assigned a rank within the species according to the Pearson correlation value . We then divided the species-specific ranks by the total number of OGs per tissue to normalize the ranks across the species ( Formula 3 , example for human , but same for other species ) . rn=rcxhNeog [3] , where rn is normalized gene pair rank , rcxh is the rank of co-expression link in human and Neog is the number of common evolutionary orthologous groups selected for tissue . The final gene-pair list was then obtained by integrating human , mouse , fly and worm ranked lists using robust aggregation , originally made for comparing two lists [79] . Briefly , using beta probability distribution on order statistics , we asked how probable is the co-expression link by taking into account the ranks of all four species . This method assigns a P value to each co-expression link in an aggregated list , indicating how much better it is ranked compared to the null model ( random ordering ) . This yielded networks with 2887 and 3353 significant gene-pairs ( edges ) ( P value < 0 . 001 ) for skeletal muscle and hippocampus , respectively . To confirm that the integrated age-related multi-species networks are significant , we selected randomly collected genes from each species . The numbers of selected genes was the same as in the OGs . We then formed the quadruplets and performed the same integration analysis as before . We repeated the procedure 100 times , and obtained 100 randomly integrated multi-species networks ( S7 Fig ) . In both cases , random and original analysis , the annotation was based on human . In order to identify aging-associated functional modules , we created networks containing 1142 nodes ( 2887 edges ) in skeletal muscle and 1098 nodes ( 3353 edges ) in hippocampus , from our prioritized gene pair list based on orthology and all edges between them . The negative logarithm ( base 10 ) of P values from aggregated list was assigned as edge weights in both integrated networks . We decomposed the skeletal muscle and hippocampus integrated networks into components and the further analysis was restricted to analysis of a giant component . The giant component contained 1050 genes ( nodes ) in skeletal muscle and 1067 genes ( nodes ) in hippocampus . As before , we used human annotation . The modules within the cross-species networks of each tissue were obtained by using a multilevel community algorithm that takes into account edge weights [80] from igraph [81] . Briefly , the multilevel algorithm [82] takes into account each node as its own and assigns it to the community with which it achieves the highest contribution to modularity . To obtain Fig 4 , we summarized groups of module nodes to single meta-nodes according to their multilevel-algorithm calculated module membership , and showed the inter-modular connectivity using a circular layout . We selected the modules with size greater than 10 , which returned 12 modules per tissue-specific cross-species network . We checked the functional enrichment of genes within selected modules in every network using Gene Ontology through topGO R package ( See Fig 4 ) . Moreover , we downloaded the pre-calculated file of gene-level summary statistics from 37 GWASs from the Pascal method [35] . We manually selected 22 out of 37 GWAS studies [83] ( S9 Table ) that are associated with metabolic , neurological , or age-related diseases . To perform enrichment of the module genes within GWAS age-related diseases categories , we selected top-ranking genes ( GWAS gene score < 0 . 1 ) within each disease and formed the categories for enrichment . We ran enrichment analysis on final network modules to find disease-related modules ( adjusted p-value < 0 . 2 ) . The human genome was used as a background gene set . Finally , we used Kleinberg’s hub centrality score to determine the hub genes within interested modules and observed the hub-gene neighborhood . The final genes were then selected to show their P value association within GWAS studies ( Fig 5C , S9 Table ) . Male and female mice from those strains were fed with normal ad libitum diet , and median and maximum lifespan were calculated to represent longevity across strains . Microarray data as well as lifespan data were downloaded from GeneNetwork . org . Microarray data from prefrontal cortex of LXS mice was generated by Dr . Michael Miles using animals with the average age of 72 days ( GN Accession: GN130 ) . Microarray data from hippocampus of LXS mice was generated by Dr . Robert Williams using animals with the average age of 73 days ( GN Accession: GN219 ) . Microarray data from spleen of BXD mice was generated by Dr . Robert W . Williams using animals with the average age of 78 days ( GN Accession: GN283 ) . Microarray data from hippocampus of BXD mice was generated by Dr . Gerd Kempermann and Dr . Robert W . Williams using animals with the average age of 70 days ( GN Accession: GN110 ) . For enrichment analysis , genes were ranked based on their Pearson correlation coefficients with the lifespan data of the BXD strains , and Gene Set Enrichment Analysis ( GSEA ) was performed to find the enriched gene sets correlated with the lifespan [84] . | Aging leads to changes in the activity of genes , but these changes are much more subtle than those between health and disease . This makes finding the genes involved in aging difficult . We have combined two ways to improve information on the involvement of genes in aging . First , important gene activity tends to be conserved in evolution , so we looked for genes whose changes in aging were conserved between distant animal species . Second , genes act in groups , so we looked for changes which affect groups of genes which act together . We were thus able to detect with good confidence subtle changes , which we could confirm with results of experiments modifying diet , and from large genetic scans in humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"genetic",
"networks",
"skeletal",
"muscles",
"brain",
"physiological",
"processes",
"developmental",
"biology",
"organism",
"development",
"network",
"analysis",
"genome",
"analysis",
"musculoskeletal",
"system",
"computer",
"and",
"information",
"sciences",
"genomics",
"gene",
"expression",
"muscles",
"evolutionary",
"genetics",
"gene",
"ontologies",
"aging",
"hippocampus",
"anatomy",
"physiology",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"human",
"genetics"
]
| 2019 | Cross-species functional modules link proteostasis to human normal aging |
Since 2001 , Haiti’s National Program for the Elimination of Lymphatic Filariasis ( NPELF ) has worked to reduce the transmission of lymphatic filariasis ( LF ) through annual mass drug administration ( MDA ) with diethylcarbamazine and albendazole . The NPELF reached full national coverage with MDA for LF in 2012 , and by 2014 , a total of 14 evaluation units ( 48 communes ) had met WHO eligibility criteria to conduct LF transmission assessment surveys ( TAS ) to determine whether prevalence had been reduced to below a threshold , such that transmission is assumed to be no longer sustainable . Haiti is also endemic for malaria and many communities suffer a high burden of soil transmitted helminths ( STH ) . Heeding the call from WHO for integration of neglected tropical diseases ( NTD ) activities , Haiti’s NPELF worked with the national malaria control program ( NMCP ) and with partners to develop an integrated TAS ( LF-STH-malaria ) to include assessments for malaria and STH . The aim of this study was to evaluate the feasibility of using TAS surveys for LF as a platform to collect information about STH and malaria . Between November 2014 and June 2015 , TAS were conducted in 14 evaluation units ( EUs ) including 1 TAS ( LF-only ) , 1 TAS-STH-malaria , and 12 TAS-malaria , with a total of 16 , 655 children tested for LF , 14 , 795 tested for malaria , and 298 tested for STH . In all , 12 of the 14 EUs passed the LF TAS , allowing the program to stop MDA for LF in 44 communes . The EU where children were also tested for STH will require annual school-based treatment with albendazole to maintain reduced STH levels . Finally , only 12 of 14 , 795 children tested positive for malaria by RDT in 38 communes . Haiti’s 2014–2015 Integrated TAS surveys provide evidence of the feasibility of using the LF TAS as a platform for integration of assessments for STH and or malaria .
Globally , lymphatic filariasis ( LF ) , soil transmitted helminths ( STH ) and malaria are frequently co-endemic , presenting opportunities for integration of programs targeting their control and elimination . The WHO Global Program to Eliminate LF ( GPELF ) was launched in 2000 with a commitment to the elimination of LF as a public health problem by 2020 through mass drug administration ( MDA ) [1] . By 2013 , more than 4 . 4 billion treatments of diethylcarbamazine or ivermectin plus albendazole ( DEC+ALB or IVM+ALB ) had been distributed in 56 countries , achieving an estimated 46% reduction in the population at risk of LF from 1 . 46 billion to 789 million people [1 , 2 , 3] . In 2012 , an estimated 1 . 5 billion people were infected with STH globally [4] . In 2014 , an estimated 269 million pre-school-aged ( PSAC ) and 576 . 6 million school-aged children ( SAC ) were living in areas endemic for STH , which WHO recommends be addressed with periodic administration of ALB or mebendazole ( MBZ ) preventive chemotherapy ( PC ) [4] . Globally , the WHO target is to treat at least 75% of children living in STH endemic countries with PC by 2020 [5] . Approximately 3 . 2 billion people live in areas where they are at risk for malaria transmission [6] . The WHO global technical strategy for malaria ( 2016–2030 ) aims to ensure universal access to malaria prevention , diagnosis and treatment , to accelerate efforts towards elimination and attainment of malaria-free status , and to transform malaria surveillance into a core intervention [7] . Occurring in the tropical and subtropical zones , LF and malaria are both transmitted by mosquito vectors , and in certain areas , by the same species . The island of Hispaniola is the only remaining Caribbean island that is endemic for both malaria and LF [8] , with Haiti bearing the greater burden of both diseases . In 2001 , Haiti’s National LF elimination program ( NPELF ) determined that nearly all communes were endemic for LF [9] and began administration of MDA ( DEC + ALB ) in select areas . Full national treatment coverage ( 140 communes ) was achieved by 2012 [10] . In 2014 , Haiti was one of the 25 countries in the Americas where PC was needed for STH , and one of seven in the region that achieved the ≥ 75% national coverage target [5] . Haiti’s malaria prevalence is low ( 0 . 4% in 2011 ) [11] , though 17 , 662 confirmed cases were reported in 2014 [6] . Documented asymptomatic parasitemia [12 , 13] highlights the need for surveillance strategies to identify remaining high transmission foci . To attain elimination of LF and malaria in Haiti , and control of STH , the three disease programs sought to identify remaining foci of disease transmission by integration of surveillance activities . In 2011 , the World Health Organization ( WHO ) recommended the Transmission Assessment Survey ( TAS ) , a standardized and statistically rigorous survey for measuring LF prevalence [14] . WHO recommends that LF elimination programs conduct the TAS in areas that have: 1 ) received 5 or more effective annual rounds of MDA; and 2 ) where spot-check and sentinel site surveys indicate microfilaria ( mf ) prevalence is less than 1% or antigenemia prevalence is less than 2% . The WHO has called for integration of neglected tropical diseases ( NTD ) activities , and in 2015 , released a protocol to integrate LF-TAS with an assessment for STH [15] . LF-TAS have successfully been integrated with STH assessments in various countries [16 , 17 , 18] . Results from these integrated surveys can be used to determine the frequency of school-based STH treatment needed after community-wide LF MDA stops . Haiti’s NPELF , National Malaria Control Program ( NMCP ) and partners saw an opportunity to synergize efforts for the LF and STH surveys in order to collect community-level information on malaria , as an integrated TAS for all three diseases ( ‘TAS-STH-malaria’ ) . The decision for the LF and malaria programs to work together was facilitated by the fact that the both programs are led by the same director , and are housed within the same office at the Ministry of Public Health and Population ( MSPP ) . To our knowledge , this was the first time the LF TAS was used as a platform for also assessing both STH and malaria . The aim of this study was to evaluate the feasibility of using TAS surveys for LF as a platform to collect information about STH and malaria .
Transmission assessment surveys were conducted according to study protocols approved by institutional review board ( IRB ) of the United States Centers for Disease Control and Prevention and the Ethics Committee of the Haitian Ministry of Public Health and Population ( MSPP ) . Following ministry policy , MSPP and IMA World Health staff recruited designated schools to participate in the survey and contacted the schools’ headmasters in advance to advise them of the purpose of the survey and to request that they notify parents . Children provided verbal assent at the time of the survey . At the time of the surveys , Haiti’s 10 departments were divided into 140 communes . Communes with the lowest baseline LF antigen prevalence based on the 2001 national survey were combined to form evaluation units ( EUs ) ; these EUs ranged from three or more communes to an entire department ( maximum of 10 communes , Sud Est ) ( Table 1 ) . Communes with greatest LF antigen prevalence at baseline were evaluated individually ( n = 8 EUs ) . Altogether , surveys were conducted in 14 EUs , composed of 47 communes in 6 departments , where TAS eligibility requirements had been met ( Fig 1 ) . Between November 2014 and June 2015 , TAS ( LF-only ) ( EU 1 ) , TAS-STH-malaria ( EU 2 ) and TAS-malaria ( EUs 3–14 ) were conducted . Due to the ambitious TAS schedule for 2014–2015 , MSPP approved piloting of the three disease protocol ( TAS-STH-Malaria ) in one EU for 2015 . The TAS is a school- or community-based survey which employs a sampling strategy ( cluster , systematic or census ) determined by the total number of children in the target age group ( six and seven years old ) , number of clusters ( schools or census enumeration areas ) , primary school enrollment rate , and vector and parasite species in predetermined EUs ( Table 2 ) . The TAS uses a critical cutoff for antigen prevalence in children , below which transmission is assumed to be no longer sustainable , even in the absence of MDA . When the number of LF positive cases among six and seven year olds is at or below the established threshold , the EU ‘passes’ the TAS and LF programs can decide to stop MDA . Surveys were designed using Survey Sample Builder ( SSB ) [19] with survey design for STH and malaria assessments the same as for TAS . The LF-only TAS teams were composed of a total of four people: ( i ) facilitator ( responsible for identifying and organizing eligible children ) ; ( ii ) enroller ( responsible for enrollment ) ; ( iii ) laboratory technician; and ( iv ) reader ( responsible for reading laboratory test and recording result ) . Some of the TAS-malaria teams had an additional laboratory technician , for a total of 5 team members . The TAS-STH-malaria teams had two additional individuals to collect and process the stool specimens and to perform Kato Katz , for a total of six or seven team members . Team members received intensive training prior to the surveys . The target population for all three survey parts was children aged six and seven years . This age group is selected based on the assumption that this age group of children would have been born just before or during the annual MDA campaigns for LF , and therefore , they should not have been exposed to bites of mosquitoes carrying infective larvae . Standard projection methods were used to estimate the population in the 14 EUs using data from Haiti’s most recent national census ( 2003 ) . Since comprehensive lists of school enrollment were not available from the Ministry of Education ( MOE ) , trained community-level workers visited schools within the EU and survey areas to generate lists of all schools with the number and ages of children enrolled . The numbers of children ( of the targeted ages ) enrolled in school were compared with the projected population of children in the target age group for each EU to estimate percent school enrollment rates and to inform survey design . Based on school enrollment rates , all 14 EUs were eligible to conduct school-based TAS [13] . The Haitian Ministry of Education indicated that 1st and 2nd grade students served as a reasonable proxy for six and seven year old children , therefore , selection criteria were set a priori to be conducted amongst 1st and 2nd grade students across the 14 EUs . At each school , eligible children were selected to receive an LF test according to the sample interval prescribed by SSB ( Table 2 ) . In the EUs where TAS-malaria was conducted ( 2–14 ) , all children selected for LF testing were also tested for malaria . In the 1 EU in which LF-malaria-STH TAS was conducted , an additional sampling interval was defined by SSB for selecting a subset of children to be tested for STH . All diagnostic tests for the integrated TAS were carried out in the field . The diagnostic tests used for LF , STH and malaria were the BinaxNOW Filariasis immunochromatographic test ( ICT ) ( Alere , Maine ) , Kato Katz ( Vestergaard-Frandsen , Denmark ) , and First Response Malaria Histidine-Rich Protein II ( HRP2 ) ( II3FRC30 ) ( Premier Medical Corporation , New Jersey ) rapid diagnostic test ( RDT ) , respectively . Dried blood spots ( DBS ) were collected on calibrated filter paper ( Cellabs , Australia ) for subsequent serological testing . From each enrolled child , technicians collected approximately 175 μL of blood from a single finger stick . Immediately following blood collection , 100 μL of blood was applied to the ICT and results were read at 10 minutes according to manufacturer instructions . Five μL of blood and 2 drops ( 60 μL ) buffer were applied to the RDT and read at 20 minutes , according to manufacturer instructions . Sixty μl of blood was applied to calibrated filter paper and dried individually . Serological assays are currently underway at CDC in Atlanta , GA , and results will be reported separately . Stool cups were distributed at the time of enrollment . Stool samples were immediately processed on site and two slides were prepared and examined from each sample . All enrollment information and diagnostic results were recorded directly into Blu® smart phones and uploaded to the cloud-based LINKS [20] server using software developed and supported by the NTD Support Center . At the conclusion of the survey at each school visit , teachers and administration were given the diagnostic results . Data were downloaded from the server in Microsoft Excel . They were cleaned , merged and analyzed using SAS 9 . 3 at CDC . Children in whom any of the three diseases were detected were provided treatment according to national guidelines , using medications provided by the survey team . Those found to have LF or STH were given DEC+ALB or ALB , respectively . Children found to have malaria were referred to the nearest health public facility to receive chloroquine and primaquine treatment free of charge , in compliance with the MSPP’s antimalarial first-line drug recommendation for treating uncomplicated malaria .
Team members spent a total of 316 working days ( working day = 1 person working for 1 day ) in the field to accomplish the TAS in 14 EUs . The LF-only TAS , the TAS-malaria and the TAS-STH-malaria required an average of 22 , 22 . 5 , and 27 working days per EU , respectively . In terms of productivity , LF-only TAS , TAS-malaria and TAS-STH-malaria teams completed activities at the same rate of 1 . 6 schools/day , accomplished by the additional personnel for the integrated TAS surveys . The cost of the TAS-malaria evaluation was an estimated 15% higher than the cost for the LF-only evaluation . The cost of the TAS-STH-malaria evaluation was 49% higher than the cost for the LF-only evaluation . The additional costs resulted from resources required for additional personnel and their transportation .
Integrated surveys have the potential to both optimize resource utilization ( human and financial ) and generate useful data across programs using a robust survey platform . The Integrated TAS-STH-malaria survey was found to be feasible and generated useful information for all three programs . Twelve of the 14 EUs passed the LF TAS , allowing the program to stop MDA for LF in 12 EUs , or 45 communes . In 36 of the communes that were able to stop MDA , baseline LF prevalence was low; however , the NPELF also succeeded in reducing LF prevalence sufficiently to pass TAS in 8 communes where LF prevalence was moderate or high at the time of mapping in 2000–2001 . This achievement underscores the quality of the MDA activities and the ability of the program to achieve adequate participation of the population . In the two EUs that failed TAS ( EU 12 , 13 ) , MSPP and partners will continue MDA for an additional two rounds , before re-evaluating in sentinel and spot check sites . Though EU 13 technically passed the TAS with 18 children positive out of 2 , 002 tested , the EU is surrounded by other areas of ongoing transmission , and thus MSPP and partners took the conservative decision to continue MDA there for an additional two rounds . According to the WHO TAS-STH manual , the 46 positive STH results from EU 2 would categorize the area as having a prevalence range of 10% to <20% [14] . This represents a decrease in STH prevalence from the 2002 national survey , but leads to the programmatic recommendation to conduct annual , school-based treatment for STH in that EU to maintain reduced STH levels . In all , malaria was detected by RDT in only 12 of 14 , 795 children in 38 communes . The inclusion of malaria RDTs in the LF TAS confirms that malaria prevalence in Haiti is low , and provides additional evidence in support of the decision to continue current programs and the development of new strategies in surveillance toward malaria elimination . Although few malaria RDT-positive individuals were identified amongst a limited population , antibody assays should provide a cumulative history of exposure and potentially define transmission foci for both diseases . The integrated TAS strategy has several advantages . Most importantly , the integrated TAS-STH-malaria enabled partners to make programmatic decisions for stopping LF MDA and for deworming frequency for STH . As the LF program nears elimination , new approaches to STH monitoring and control , including the integrated TAS which produces actionable results for STH , should be incorporated into work plans . Second , the integrated TAS establishes a potential framework for integrated surveillance and a more coordinated approach to community-based intervention strategies . By testing for multiple infections , we were able to generate actionable results for more than one program . The results also informed thinking about the testing needed to support malaria elimination . In a single , integrated and carefully planned survey , we are able to maximize returns on the investment of field activities while also reducing burden on both field teams and communities . These synergies are especially important for diseases in the elimination phase , since surveillance efforts will be resource intensive for increasingly rare conditions . Third , this activity fostered collaboration between ministries of education ( MOE ) and health ( MSPP ) , and across disease programs within MSPP . The integrated TAS marks the first time in which field activities for the three disease programs were combined in such a manner in Haiti . The NPELF and partners conveyed the common goals of each public health activity to encourage school administrators to allow field teams to conduct integrated TAS in schools . The support of school administrators will be particularly important as STH control moves towards school-based deworming following the end of LF MDA . The LF , malaria and STH programs and field teams worked well together in this mutually beneficial and informative activity , thereby developing and strengthening the new working relationship . Fourth , the integrated survey represented an overall cost savings , compared with performing similar assessments independently . The addition of the malaria assessment to the TAS ( TAS-malaria ) cost an estimated 15% more in Haiti than the LF-only TAS , due to the addition of a team member for processing and reading of malaria RDTs . As teams become more familiar with the workflow , it is anticipated an additional team member might not be necessary to perform the malaria RDT . The integrated TAS-STH-malaria costs an estimated 49% more than the LF-only TAS , due to the addition of two team members for processing of stool by Kato Katz . Obtaining the same information through separate surveys conducted by each independent program would have incurred substantially more expense . The integrated survey platform creates options to include additional tests to expand the utility of the survey . Lastly , the integrated TAS represented less of an intrusion for the communities than would three independent surveys . By performing multiple diagnostic tests in one coordinated visit , the amount of time that children were kept from classes was minimized . The overall activity was less time consuming , in that the school administration was approached only once on behalf of the three programs , and consent was obtained for all diagnostics concurrently . The one visit proved to take only slightly longer than the TAS-only activity , and provided immediate results to communities and treatment for infected children . This study identified a few challenges to performing the TAS in Haiti . First , the age of the children enrolled in first and second grades varies greatly , so the program was unable to rely on grades as a proxy for age . Testing children older than seven years ( i . e . born prior to the start of MDA ) , provides a more conservative estimate of LF transmission since it includes children were born before MDA began , however it fails to answer the question about the effect of MDA on recent transmission . For this reason , the decision was made to identify children by age in EUs 3–14 , rather than continuing to use grade as a proxy for age in order to better keep with the TAS protocol guidelines . Second , the estimate of student enrollment frequently exceeded actual child attendance on the day of the surveys . More schools needed to be visited in order to reach the predetermined sample size , which posed additional logistical burden on the field teams . The TAS survey guidelines rely on accurate estimations of the number of children of the target ages living in each EU as well as school enrollment rates , to ensure that survey design yields representative and appropriately distributed samples from across the targeted population . Prior to the survey , current school enrollment rates were not available from the MOE , so partners obtained school data directly from the schools in each EU . Third , field teams had to manage several logistical challenges including limited access to cold chain for storage of ICT cards before TAS and dried blood spots after TAS . Limited cellular service , accessibility and difficult terrain of some selected schools , and school holidays were a challenge for field teams impacting sample size . Finally , sensitization was not always sufficient , leading to refusals in some instances . There were also limitations associated with the malaria program . First , although testing more than 14 , 700 six and seven year old children confirmed the low prevalence of malaria in the surveyed areas , the rates of RDT positivity was low and consequently , the results of the malaria antibody testing planned for specimens collected during this study are likely to be more informative for programmatic decisions than the RDT results . Second , the integrated TAS survey design , including geographic distribution of EUs , was based on historical LF mapping data . Use of malaria distribution as the basis for the creating EUs would have likely influenced the choice of how to combine communes to form EUs .
The activities reported here from 14 TAS surveys provide evidence of the feasibility of using the LF TAS as a platform for integration of assessments for STH and or malaria . In 2014 , Haiti’s LF elimination program achieved eligibility for administration of TAS in 47 communes and after conducting TAS was able to stop MDA campaigns for approximately 1 , 981 , 920 people in 44 communes . Intensified activities in the next five years , including progressive implementation of TAS , stopping MDA , and phasing in post-MDA surveillance , will be essential to achieving the country’s elimination goal by 2020 [21] . With this in mind , the NPELF and partners plan to conduct TAS ( including TAS-STH-malaria and TAS-malaria ) in 58 additional communes in 2016 and 39 additional communes before the end of 2017 . Integrating assessment of STH infections will enable program managers to determine the effectiveness of proposed school-based STH programs , while Integrating malaria into the TAS platform will provide additional national data on recent malaria history , which is important for targeting malaria elimination efforts and ensuring progress towards a malaria-free Haiti . Despite experiencing many challenges , including the 2010 earthquake and cholera outbreak , integrated TAS results support the assertion that Haiti is on track to meeting the WHO’s 2020 LF global elimination targets . Although malaria elimination is admittedly a more ambitious goal , development of integrated surveillance strategies will help to achieve this goal . | Lymphatic filariasis and malaria are mosquito-borne parasitic infections that are endemic in Haiti . Soil-transmitted helminths are also present in Haiti , infecting large numbers of people every year . Since 2001 , Haiti’s National Program for the Elimination of Lymphatic Filariasis ( NPELF ) has worked to reduce the transmission of LF through annual mass drug administration with the aim of reducing LF prevalence in the population below a threshold , such that transmission is assumed to be no longer sustainable . By treating the entire population of Haiti with a combination of drugs , the elimination program has made tremendous progress towards eliminating the disease . By 2014 , Haiti’s NPELF had met the World Health Organization eligibility criteria to conduct LF transmission assessment surveys ( TAS ) and decided to use the LF TAS as a platform to collect information about STH and malaria . The WHO has called for the integration of program activities in the field , and the TAS is a platform that allows for such integration . In Haiti the integrated TAS reduced the burden of repeated surveys on communities by minimizing site visits and benefited all three disease programs by sharing the responsibilities of field data collection . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusions"
]
| [
"children",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"education",
"sociology",
"tropical",
"diseases",
"geographical",
"locations",
"social",
"sciences",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"north",
"america",
"age",
"groups",
"protozoans",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"caribbean",
"families",
"infectious",
"diseases",
"malarial",
"parasites",
"schools",
"haiti",
"people",
"and",
"places",
"helminth",
"infections",
"blood",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"malaria",
"soil-transmitted",
"helminthiases",
"organisms"
]
| 2017 | Partnering for impact: Integrated transmission assessment surveys for lymphatic filariasis, soil transmitted helminths and malaria in Haiti |
Despite dengue dynamics being driven by complex interactions between human hosts , mosquito vectors and viruses that are influenced by climate factors , an operational model that will enable health authorities to anticipate the outbreak risk in a dengue non-endemic area has not been developed . The objectives of this study were to evaluate the temporal relationship between meteorological variables , entomological surveillance indices and confirmed dengue cases; and to establish the threshold for entomological surveillance indices including three mosquito larval indices [Breteau ( BI ) , Container ( CI ) and House indices ( HI ) ] and one adult index ( AI ) as an early warning tool for dengue epidemic . Epidemiological , entomological and meteorological data were analyzed from 2005 to 2012 in Kaohsiung City , Taiwan . The successive waves of dengue outbreaks with different magnitudes were recorded in Kaohsiung City , and involved a dominant serotype during each epidemic . The annual indigenous dengue cases usually started from May to June and reached a peak in October to November . Vector data from 2005–2012 showed that the peak of the adult mosquito population was followed by a peak in the corresponding dengue activity with a lag period of 1–2 months . Therefore , we focused the analysis on the data from May to December and the high risk district , where the inspection of the immature and mature mosquitoes was carried out on a weekly basis and about 97 . 9% dengue cases occurred . The two-stage model was utilized here to estimate the risk and time-lag effect of annual dengue outbreaks in Taiwan . First , Poisson regression was used to select the optimal subset of variables and time-lags for predicting the number of dengue cases , and the final results of the multivariate analysis were selected based on the smallest AIC value . Next , each vector index models with selected variables were subjected to multiple logistic regression models to examine the accuracy of predicting the occurrence of dengue cases . The results suggested that Model-AI , BI , CI and HI predicted the occurrence of dengue cases with 83 . 8 , 87 . 8 , 88 . 3 and 88 . 4% accuracy , respectively . The predicting threshold based on individual Model-AI , BI , CI and HI was 0 . 97 , 1 . 16 , 1 . 79 and 0 . 997 , respectively . There was little evidence of quantifiable association among vector indices , meteorological factors and dengue transmission that could reliably be used for outbreak prediction . Our study here provided the proof-of-concept of how to search for the optimal model and determine the threshold for dengue epidemics . Since those factors used for prediction varied , depending on the ecology and herd immunity level under different geological areas , different thresholds may be developed for different countries using a similar structure of the two-stage model .
Dengue viruses ( DENV ) are the most widespread arthropod-borne viruses affecting humans . A recent study estimates that annually 390 million DENV infections occur worldwide with 500 , 000 severe cases and 25 , 000 deaths , mostly affecting children[1] . Infection with DENV can result in a range of outcomes from asymptomatic infection to clinical manifestations ranging from dengue fever ( DF ) to the life threatening complications of dengue hemorrhagic fever ( DHF ) and shock syndrome ( DSS ) . This mosquito-borne disease is caused by four serotypes of dengue virus ( DENV-1 to 4 ) , which belong to the family Flaviviridae , genus Flavivirus[2] . Infection by one serotype of DENV will provide lifelong immunity to that particular strain but not to the remaining three serotypes , which usually lead to the reduction of the susceptible population . However , immunity from prior infection might enhance the incidence of DHF through antibody-dependent enhancement mechanism even though the transmission of DENV is reduced[3 , 4] . The virus is transmitted to humans mainly by two mosquito vectors , Aedes aegypti or Aedes albopictus . In the absence of an effective vaccine or specific therapy , vector control remains the only way to prevent dengue viral transmission[5] . Increased travel with population movement , global trade , crowded urban living conditions , global warming , virus evolution and ineffective vector-control strategies are also increasing the risk of dengue transmission in the world[6 , 7] . Travelers infected with dengue virus during their trip returning home may place the local population at risk wherever mosquito vectors are present[8 , 9] . Therefore , the required conditions for the occurrence of a dengue outbreak in countries where dengue is not endemic include i ) the presence of dengue viruses through repeated introduction of imported cases , ii ) a sufficient density of competent vectors above the threshold , iii ) a sufficient number of susceptible population , and iv ) a favorable climatic and environmental condition for dengue transmission[10] . Furthermore , numerous studies suggested an effect of climate on DENV transmission through changes in vector population size and distribution . The relationships between entomological measures of risk and human infection are not well understood[11–13] . The mosquito vectors , principally A . aegypti , become infected when they feed on humans during the usual five-day period of viraemia . The virus passes from the mosquito intestinal tract to the salivary glands after an extrinsic incubation period , a process that takes approximately 10 days , which may vary depending on the ambient temperatures[14] . Mosquito bites after the extrinsic incubation period result in infection , which might be promoted by mosquito salivary proteins[15–17] . The abundance of dengue vector as well as dengue transmission generally exhibits seasonal variation depending on the local ecology and urban environment . Therefore , vector surveillance is recommended by the World Health Organization ( WHO ) and is a routine practice in many dengue-occurring countries to provide quantifiable measure of fluctuations in magnitude and geographical distribution of dengue vector populations[18 , 19] . The traditional standard protocol relies on surveys of larvae and pupae , which include three most commonly used indices: the House index ( HI ) , the Container index ( CI ) and Breteau index ( BI ) . A poor correlation with the abundance of adult mosquitoes has caused their sensitivity and reliability to be questioned[20 , 21] . The alternative , pupal indices developed by Focks et al[22] , has been suggested to better reflect the risk for transmission , but the utility for source reduction programs is still controversial[23 , 24] . The most accurate method of vector surveillance is the capture of adult mosquitoes by aspiration , which directly counts dengue vectors that are actively in search of a blood meal: adult female A . aegypti and occasionally A . albopictus mosquitoes . However , capturing adult mosquitoes is labor-intensive , and requires access to premises . Recently , fixed-position traps , designed to capture gravid mosquitoes using water-filled pots in which A . aegypti lay their eggs , are widely used as a simple sampling tool[25 , 26] . However , its correlation with the incidence of dengue is still controversial[27 , 28] . Kaohsiung City , a modern metropolis of 1 . 5 million people , has been afflicted by different serotypes of DENV and has become the focus of dengue virus activity in Taiwan during the recent decades[29] . During 2002–2011 , Kaohsiung City had annual outbreaks of variable scales , resulting in more than 6 , 000 confirmed cases[30] . Since 2005 , vector surveillance activities by the Department of Health , Kaohsiung City Government , were initiated by using specially trained personnel . Four different vector indices were chronically established . A previous study suggested that adult Aedes mosquito index from 2005–2009 showed temporal correlation with the peak of the DF activity with a lag period of 1–2 months[29] . However , the association between different vector indices and the occurrence of dengue cases has not been comprehensively evaluated . Therefore , the objectives of this study were to i ) evaluate the temporal relationship between meteorological variables , entomological surveillance indices and dengue confirmed cases , ii ) identify the suitable conditions for an epidemic occurrence , and iii ) establish the threshold for entomological surveillance indices as an early warning tool for dengue epidemic .
Although dengue virus epidemics have occurred annually in Taiwan for the past decade , the main focus of activity has been in Kaohsiung City ( Fig 1 ) . Kaohsiung City is a standard subtropical region with annual average rainfall from 1796 . 7 to 2821 . 4 mm and concentrated from May to September . In addition , the annual average temperature is from 24 . 9 to 25 . 7 degrees Celsius ( °C ) , with the lowest average 11 . 6°C in February and the highest average 31 . 5°C in June . After December 25 , 2010 , the area of Kaohsiung city expanded due to the combined administration area between Kaohsiung County and Kaohsiung City . Since our study period covered from January 2005 to December 2012 , the study area included the former Kaohsiung City , Fongshan , Daliao , and Linyuan districts as well as the adjacent Pingtung County and Tainan City in southern Taiwan , located between 120°10′32″ to 121°01′15″ east longitudes and 22°28′ to 23°28′ north latitudes . This study was approved by the Institutional Review Board ( Approval No . IRB-R-05-002 ) of Taichung Hospital , Ministry of Health and Welfare , Taiwan; and all analyzed data was anonymized . Since Taiwan is not a dengue-endemic country , the common season for the indigenous dengue cases to occur starts from May to December after repeated introduction of imported dengue cases as suggested by previous publication[8] . The dengue cases within this period comprised 98 . 7% of the annual total . Therefore , we focused our forecasting model from May to December . We used a two-week interval as a unit to divide the 8-year span into 151 intervals , counting the dengue case ( Y ) and averaged the meteorological data . The environmental factors included in the study were vector indices ( VI , including AI , BI , CI , and HI ) , mean temperature ( Temp , °C ) , mean rainfall ( RF , mm ) and relative humidity ( RH , % ) . We calculated the mean of the daily average over each week in the study period for all the weather factors , so that the corresponding 33rd and 67th percentiles can be determined . We then further transformed these factors into indicator variables of three levels ( low , medium , and high ) by using the percentiles as the cutoffs . Besides , both 2-week and 1-month lags of VI and meteorological factors were considered here . The 2-week lag took into consideration the blood feeding of mosquito on an undetected viremic subject and the 7–10 days interval to be able to re-infect a new subject , who requires 3–5 days to be symptomatic once infected [37] . On the other hand , the 1-month lag took into consideration the mosquito life cycle from the laying to hatching of eggs , which requires 2 weeks and another 2 weeks from feeding to infect a new subject . Therefore , the VI distinguishes into VI1 ( 2-week lag ) and VI2 ( 1-month lag ) , and each VI1 or VI2 was also calculated separately based on individual AI , BI , CI , and HI . To avoid colinearity among the VIs , we considered only one of the VIs at a time ( e . g . , AI 2-week lag ) joined by 6-weather variables ( RF1 ( 2-week lag ) , RF2 ( 1-month lag ) , Temp1 ( 2-week lag ) , Temp2 ( 1-month lag ) , RH1 ( 2-week lag ) , RH2 ( 1-month lag ) as potential predicting variables . We used the minimized Akaike’s information criteria ( AIC ) as the criterion for models selection , and each variable was either included or excluded; and therefore a total of 27 = 128 combinations were tried to select the optimal subset of predicting variables . Since there were 4 vector indices with 2 lag options , a total of 128×4×2 models were tried . Regression models were developed by using a two-stage approach , wherein we first performed an exploratory analysis to select the best models and used the selected model to create the indices for predicting the occurrence of dengue cases and then estimated the prediction accuracy . The stage 1 of the initial exploratory analysis used the Poisson regression to select the optimal subset of variables and time-lags for predicting the number of DF cases . The stage 2 used the optimized model selected from Stage 1 to establish the prediction threshold and defined the prediction accuracy from the ROC curve by logistic regression . In stage 1 , the univariate and multivariate lagged-time Poisson regression analysis was performed to assess the relationship between the environmental factors and dengue cases . A basic multivariate Poisson regression model was written as below . Y~Poisson ( λ ) , Ln ( λ ) =β0+β1⋅VI1+β2⋅VI2+β3⋅RF11+β4⋅RF12+β5⋅RF21+β6⋅RF22+β7⋅Temp11+β8⋅Temp12+β9⋅Temp21+β10⋅Temp22+β11⋅RH11+β12⋅RH12+β13⋅RH21+β14⋅RH22 where Y is the incidence of confirmed dengue cases and β0 is the intercept . VI1 and VI2 are indicator variables with value 1 if the 2-week lag and 1-month lag , respectively , are above the ( overall ) 33rd percentile; and 0 otherwise . RF11 , Temp11 , and RH11 are indicator variables with value 1 if the RF1 , Temp1 and RH1 , respectively , are between the ( overall ) 33rd and 67th percentiles; and 0 otherwise . Analogously , RF12 , Temp12 and RH12 are value 1 if the RF1 , Temp1 and RH1 , respectively , are above the 67th percentiles . In addition , RF21 , Temp21 and RH21 are indicator variables with value 1 if the RF2 , Temp2 and RH2 , respectively , are between the ( overall ) 33rd and 67th percentiles; and 0 otherwise . Similarly , RF22 , Temp22 and RH22 are value 1 if the RF2 , Temp2 and RH2 , respectively , are above the 67th percentiles . In stage 2 , we fitted Y on the 4 optimal subset of variables selected from stage 1 , and produced the ROC curves . We considered count of outbreak for bi-week greater than 1 , then definition of binary variable Y = 1 and = 0 if not . The final optimal models selected were variables including VI2 , RF1 , RF2 , Temp1 , and RH1 and the predicting coefficients for each VI were as below . For each ROC curves , we picked a point that looked farthest from the diagonal line , which has the largest area under curve ( AUC ) and with maximum total of sensitivity and specificity . The value decided by that point was then used as the predicting threshold . We defined the corresponding linear combinations of variables estimated from the logistic regression as a prediction index and the thresholds for each VI were calculated accordingly . As a result , there were 4 models for each index: AI , BI , CI , and HI . We conducted 2 cross-validations to examine the sensitivity of our method . The first is the leave-one-out approach , of which each observation ( a two-week period ) was removed and the rest were used to establish the classification criterion via multiple logistic regression and corresponding AUC , and then that criterion was used to classify the removed one . After iterating on all the periods , the performance was assessed by the average accuracy rate . The second approach was to leave one-year out instead of one period[38] . We repeated the above process on Model-AI , Model-BI , Model-CI and Model-HI separately . Two-tailed p<0 . 05 was regarded as statistically significant . The lagged-time Poisson regression analyses were performed by using SAS Version 9 . 3 for Windows ( SAS Institute Inc . , Cary , North Carolina , USA ) .
From January 2005 to December 2012 , Taiwan-CDC recorded 8 , 918 laboratory-confirmed cases of dengue virus infections in Taiwan , and 58 . 1% were from Kaohsiung City ( Fig 2A ) . The cases were detected by passive and active surveillance activities . The successive waves of dengue outbreaks with different magnitudes were recorded in Kaohsiung City , and they involved a dominant serotype of DENV during each epidemic , representing more than 80% of cases confirmed by virus detection in the specific year ( Fig 2B ) . The annual dengue incidence rate varied with the highest rate ( 1 , 176 cases ) observed in 2011caused by DENV-2 and DENV-3; and the lowest rate ( 102 cases ) in 2005 caused by DENV-3 without significantly secular trend . By dividing the annual dengue cases into four quarters of the year , the annual outbreak usually started one month before the third quarter and reached a peak in the fourth quarter . The dengue cases during the first quarter were residual cases from the outbreak in the previous year . Based on the geographical distribution , 98 , 1 . 9 and 0 . 2% of dengue cases were from the high , middle and low risk districts , respectively ( Fig 2B ) . From 2005 to 2012 , 159 total dengue cases were confirmed as imported , which distributed evenly throughout different months without a secular trend ( Fig 2C ) . The peaks of the confirmed cases detected by passive and active surveillance were also coincided as shown in S1 Fig . The average temperature in Kaohsiung City is around 25°C with the hottest season occurring in July to September and the coldest in January to March . The hottest season usually coincided with the strike of tropical hurricanes , which brought in significant amount of rainfall , causing a rise in the annual dengue epidemic ( Fig 2D ) . The relative humidity was relatively stable with an annual average of around 72 . 4% . The density of A . aegypti and A . albopictus in Kaohsiung City showed dynamic and periodic variation . We focused only on the high risk district for data analysis wherein about 97 . 9% dengue cases occurred and the inspection of immature and mature mosquitoes was carried out on a weekly basis . BI , CI or HI over 5 usually appeared in early summer and peaked during autumn in the high risk area ( Fig 3 ) . Vector data from 2005–2012 showed that the peak of the adult mosquito population was followed by a peak in the corresponding dengue activity with a lag period of 1–2 months ( Fig 3 ) . Since Kaohsiung City frequently had more dengue epidemics , which occurred annually , and had a more comprehensive vector surveillance data , the following statistical analyses was focused on the data from Kaohsiung City . Results of univariate analysis showed that the risk of an increased number of dengue cases was significantly associated with the increase in all vector indices ( including BI , CI , HI and AI ) on either 2-week- or 1-month-lag effect as shown in Table 1 . However , the meteorological variables showed different patterns of association of dengue epidemics . Either medium or high level of the temperature showed negative association with the increased risk of dengue epidemics at 2-week-lag effect . However , the medium temperature showed a positive association with the increased risk ( RR: 1 . 32; 95% CI: 1 . 23–1 . 41 ) , but the high level temperature showed a negative association ( RR: 0 . 77; 95% CI: 0 . 71–0 . 83 ) at 1-month lag with statistical significance ( p<0 . 05 ) . Similarly , both medium and high levels of RF showed negative associations with the increased risk of dengue epidemics at 2-week-lag effect . However , the medium RF showed positive association with increased risk ( RR: 1 . 12; 95% CI: 1 . 05–1 . 2 ) , but the high level temperature showed a negative association ( RR: 0 . 86; 95% CI: 0 . 80–0 . 92 ) at 1-month lag . The increase of RH showed consistently strong correlation with the increased risk of dengue cases either at 2-week- or 1-month-lag effect ( Table 1 ) . In order to find the best model for the prediction of dengue occurrence , a multivariable Poisson regression model was fitted to the data to search for independent factors by running different combinations of time lag effect . The final results of the multivariate analysis that best predicted the occurrence of dengue cases were selected based on the smallest AIC value as shown in Table 2 . The 1-month lag effect of all VI was selected in the multivariable Poisson model except AI , for which a 2-week-lag effect showed the best result . The 2-week-lag effect of all meteorological factors was also selected in the final multivariable Poisson model; however , RF and Temp showed a negative association with the occurrence of dengue cases . In contrast , RH and 1-month lag of RF showed a positive association . A slight difference with statistical significance was also noted at Model-CI , in which the 1-month lag of RF showed positive association when the RF level was medium ( RR:1 . 12; 95% CI: 1 . 04–1 . 21 ) and negative association when it was at a high level ( RR:0 . 89; 95% CI: 0 . 81–0 . 98 ) . Although the error between the observed and estimated counts was large ( range of R-square of four models: 0 . 16–0 . 3 ) , the prediction of peaks by the predictors selected from Poisson model quite coincided ( Fig 4 ) . Next , in order to establish the threshold for entomological surveillance indices as an early warning tool for dengue epidemics , a threshold , where good sensitivity and specificity both reach above 80% , was selected . A threshold with 100% sensitivity but poor specificity will lead to too many false alarms and exhaust public health resources . Therefore , we applied each selected VI models to multiple logistic regression models to examine the accuracy of predicting the occurrence of dengue cases based on the ROC analysis by selecting an operating point which provided an optimum tradeoff between false-positive and false-negative results . The results suggested that Model-AI , BI , CI and HI , based on the operating point selected , yielded a sensitivity of 82 , 87 , 86 and 85% , respectively; and a specificity of 76 , 80 , 80 and 80% , respectively ( Table 3 ) . The accuracy of Model-AI , BI , CI and HI in predicting the occurrence of dengue cases were 83 . 8 , 87 . 8 , 88 . 3 and 88 . 4% , respectively ( S2 Fig ) . The individual predicting thresholds for Model-AI , BI , CI and HI were 0 . 97 , 1 . 16 , 1 . 79 and 0 . 997 , respectively as shown below . Each of them when combined with meteorological factors had better performance compared to the prediction using AI , BI , CI and HI alone , where the value were only 69 . 2 , 78 . 7 , 80 . 2 and 78 . 7% accurate , respectively ( Table 3 ) . Model BI=3 . 04BI−0 . 357RF11−1 . 49RF12−0 . 096RF21+0 . 405RF22−1 . 5Temp1−0 . 888Temp2+0 . 634RH1+2 . 77RH2>1 . 16Model AI=2 . 48AI−0 . 385RF11−1 . 68RF12+0 . 719RF21+1 . 47RF22−1 . 78Temp1−0 . 619Temp2+1 . 02RH1+3 . 36RH2>1 . 79Model CI=3 . 39CI−0 . 462RF11−1 . 37RF12+0 . 013RF21+0 . 329RF22−1 . 84Temp1−1 . 54Temp2+0 . 58RH1+2 . 61RH2>0 . 97Model HI=3 . 44HI−0 . 596RF11−2 . 01RF12−0 . 337RF21+0 . 127RF22−1 . 78Temp1−0 . 866Temp2+0 . 738RH1+3 . 12RH2>0 . 997 where the variables definition are the same as Poisson regression variables . The estimates of AUCs , as obtained by leave-one-out cross-validation for Model-AI , Model-BI , Model-CI , and Model-HI , were 0 . 762 , 0 . 818 , 0 . 833 , and 0 . 829 , respectively; those by leave-one-year-out were 0 . 814 , 0 . 85 , 0 . 866 and 0 . 843 , respectively , and only slightly less ( 2~4% ) than the original AUC . The results suggest that our method is stable in predictive accuracy .
With the continuously high levels of worldwide dengue transmission , predicting dengue outbreaks in advance of their occurrence or establishing an early warning system through the combination of climate , environmental , host and vector-based data is of critical importance . The main purpose of an early warning system is the collection of information leading to timely decision making process , which triggers intervention strategies in order to reduce the burden and effect of the disease or outbreak on a specified population . Although mosquito vector is directly involved in virus transmission , the current entomological indicators do not reliably assess the risk of dengue case occurrence . Our study here provided the proof-of-concept results , utilizing a two-stage model to identify the best set of lag effects of meteorological and entomological variables , explaining dengue epidemics based on the data obtained from Taiwan , which is a dengue-non-endemic country . AI , BI , CI and HI of the vector indices when combined with the meteorological factors have better performances compared to the prediction using AI , BI , CI and HI alone , with 83 . 8 , 87 . 8 , 88 . 3 and 88 . 4% accuracy , respectively . The advantage of this two-stage model is not only to produce the unified set of predictors throughout two-stage modeling but also to keep as much information in the set as possible . Although the error between the observed and estimated counts could be large , the prediction of peaks by the co-variables selected from the Poisson models quite coincided ( S1 Fig ) . Further employing these co-variables in the second-stage logistic models for predicting the occurrence of outbreak came out with satisfactory results . Since same co-variables were employed in the two-stage model , the value above the threshold would not only predict the occurrence of dengue cases , but also the size of the outbreak based on the stage 1 model , either big or small . Therefore , each country should consider its own individual data and apply this two-stage modeling strategy to find the optimal predictive threshold for allocating public health resources and prevention strategies . Since only adult female Aedes mosquitoes are directly involved in dengue transmission , directly counting dengue vectors ( adult female A . aegypti and occasionally A . albopictus mosquitoes ) using fixed-position traps has been advocated to replace the traditional methods , because stegomyia indices are developed many decades earlier for yellow fever and the relationship with dengue transmission is usually ambiguous[39] . However , the stegomyia indices such as HI and BI remain central and are most widely used in the monitoring of dengue vector populations , but their critical threshold has never been determined for dengue virus transmission[40 , 41] . Traditionally , BI < 5 was proposed to prevent yellow fever transmission and three different risks of HI , with <0 . 1% as low , 0 . 1–5% as medium and >5% as high , were suggested by the Pan American Health Organization to prevent dengue transmission[42] . However , dengue transmission was observed with vector density below that and the appropriated entomologic level remains contentious[43] . A universal critical threshold applicable across many contexts has never been determined even though a simple threshold ( HI = 1% or BI = 5 ) has been used for many years and is only valid in some situations[44] . Since the population of mosquito vectors is influenced by the meteorological factors , a threshold combining VI and meteorological variables with different lag effects would provide a better prediction of dengue epidemic . In this study , four VI models were developed and integrated thresholds were estimated from the multivariate Poisson model with BI , CI , AI and HI of 1 . 16 , 1 . 79 , 0 . 97 and 0 . 997 , respectively . These integrated VI thresholds predicted better with accuracy higher than 80% , compared to using VI alone ( Table 3 ) . Furthermore , although choosing an arbitrary threshold of BI > 5 is more intuitive and interpretable , the prediction accuracy of dengue epidemic is only 77% in this study . The utilization of single global values of BI or other VI as thresholds for dengue transmission is unreliable and is not recommended based on the previous review[13 , 37 , 45] . Therefore , our study utilized a two-stage modeling , which is a simple and direct concept for estimating the thresholds in different locations or counties . An automatic smartphone application which uses the two-stage model to calculate the integrated VI thresholds from the collected data on a weekly basis would facilitate an early warning system for worldwide use . The meteorological factors ( temperature , rainfall and relative humidity ) were important variables which directly and indirectly affect the mosquito density and blood feeding behavior[46 , 47] . Overall , temperature affects the length of Aedes gonotrophic cycle , pupae development period and extrinsic incubation period of dengue virus , which are usually shorter at higher temperature[48–50] . Temperatures may also influence the vector body size and its biting behavior . Smaller mosquitoes feed more often than larger ones; and higher temperatures can augment immature development resulting in smaller mosquitoes[51] . Higher temperatures also speed blood meal digestion so that females need to feed more often[52] . Thus , all these factors directly and indirectly influence the contact rate between vectors , which leads to an increased risk of viral transmission from an infected mosquito to a susceptible host[3] . On the contrary , the effect of temperature on the mortality rate of larvae , pupae and adult mosquitoes can be U-shaped with a lower mortality rate seen when temperature ranged from 15 to 30°C[53 , 54] . This might explain the results in our study that showed a positive association of temperature at medium level at 1-month-lag effect with the risk of an increased number of dengue cases , but a negative association with the risk of an increased number of dengue cases at either 1-month lag only at a high level of temperature or 2-week lag , either in medium or high levels of temperature . Non-linear effect of the co-variables on the number of dengue cases could also be found at rainfall , which was also found in our and other’s studies . An increase in amount of rainfall leads to more breeding sites , which in turn lead to an increase in the number of mosquito density as suggested by previous studies[11] . However , too much rainfall might wash away the larvae or pupae inside the premise and decrease the mosquito density[55] . Adding quadratic term is one way to cope with the problem . However , due to the co-linearity between the linear and quadratic term , very few covariates would be significant . Another approach to cope with the nonlinearity is to trisect the covariate ( low , middle and high ) ; and we found it can come out with a more significant result for interpretation as shown in this current study . Other well-known factors may have contributed to the dynamic occurrence of dengue cases and epidemic . The shift of age structure from children to young adults during epidemics was previously reported[56] . The average age in confirmed dengue cases was 44 . 4 years old and slightly increased from 2005 to 2012 , which was consistent with the trend of gradual increase in age from the general population in Kaohsiung city . The slight increase of age in dengue cases was not significantly correlated with the annual dengue incidence rate ( P = 0 . 261 ) ( S3 Fig ) . Although the ratio of primary and secondary infections might change the epidemic dynamics and increase the disease severity , previous studies found the DHF/DF ratio increased through the epidemic and the disease severity was not correlated with the secondary infection in Taiwan[57] . Previous studies also suggested that certain strain or serotype of DENV with epidemic potential might increase viral growth in mosquito and enhance virus transmission[58 , 59] . Since not all the confirmed cases were determined by virus isolation or RT-PCR in this study , it was currently not feasible to incorporate the case ratio infected by different serotypes each year into the model . Additionally , herd immunity might affect the dengue epidemic as suggested in other studies but the results were not conclusive[45] . How the herd immunity , measured from the sero-prevalence data which is not available in this study , affects our model prediction requires further study . The results in this study should be interpreted within the context of strengths and limitations . First , entomologic data collected through routine systems could pose some limitations due to different vector control technicians for inspection , procedures that are not completely uniform and inspection cycles . We focused on high risk areas and inspected the premises for mosquito breeding sites on a weekly basis to minimize the bias . Second , the overall indices were calculated for communities defined by administrative boundaries , which do not constitute entomologically homogeneous units . The optimal geographical level for calculation would be under household and neighborhood level , which is usually difficult to obtain due to the protection of individual privacy . The consistent collection of vector indices under the same administrative boundaries available to be used for public domain would provide better predictions in the long term . Third , the surveillance and dengue case ascertainment did not allow us to detect asymptomatic infection , which likely varied through time and was underestimated in this study . Fourth , the present study was an ecological investigation; therefore , it is not possible to make inferences concerning the causative relationship between the mosquito larvae indices and dengue infection at the individual patient level . Fifth , the spatial heterogeneity was not considered in this study and will be the future focus for developing a better model[60 , 61] . Sixth , in this study we focused on the high risk district where 97 . 9% dengue cases occurred and inspection was carried out on a weekly . The potential bias is minimal since the timing of mosquito collection did not depend on the onset of dengue cases and the mosquito collection was not only done in the residential districts of the confirmed dengue cases . However , the threshold estimated in this study could only be applied to the high risk district . If the threshold is desired to be determined in the middle or low risk area , different lag effects of meteorological variables and monthly values of VIs would need to be determined separately . Seventh , when the case was confirmed , the environmental interventions carried out by the health services team would be implemented such as the chemical treatment of the location and the neighborhood of the confirmed case , the intensification of measures to control breeding areas and health education . These usually lead to the elimination of breeding grounds of immature and adult mosquitoes . Since our study was to establish a threshold for early case detection before any control measures is in place , the effect on our model prediction of the occurrence of dengue cases would be minimal . In conclusion , our study here provided the proof-of-concept of how to search for the optimal model and determine the threshold for dengue epidemics . Unlike other studies with a determined threshold , here the findings cannot be extrapolated to communities with different environment conditions or herd immunity levels . We currently are developing an automatic system allowing implementation of the data in a weekly basis and following the two-stage model to calculate the integrated VI threshold for worldwide use . This work provides an example of the practical utility of research projects in the operational public health field and reinforces the need for a multidisciplinary approach in the understanding and management of vector-borne diseases . | With the continuously high levels of worldwide dengue transmission , predicting dengue outbreaks in advance of their occurrence or identifying specific locations where outbreak risks are highest is of critical importance . However , only few studies have been conducted in dengue non-endemic countries to evaluate the association of vector index with the occurrence of dengue cases; and the establishment of an early warning signal would significantly enhance the public health intervention . Our study here provided the proof-of-concept results , utilizing a two-stage model to identify the best set of lag effects of meteorological and entomological variables , explaining dengue epidemics based on the data obtained from Taiwan , which is a dengue-non-endemic country . Each of the vector indices when combined with the meteorological factors has better performance compared to the prediction using AI , BI , CI and HI alone , with 83 . 8 , 87 . 8 , 88 . 3 and 88 . 4% accuracy , respectively . Because of the complex interplays between the size of human hosts and movement , environmental factors and dynamic changes of mosquito population and density , each country should consider its own individual data and situation and apply this two-stage model to find the optimal predictive models for allocating public health resources and prevention strategies . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | Re-assess Vector Indices Threshold as an Early Warning Tool for Predicting Dengue Epidemic in a Dengue Non-endemic Country |
Telomeres , the ends of linear eukaryotic chromosomes , have a specialized chromatin structure that provides a stable chromosomal terminus . In budding yeast Rap1 protein binds to telomeric TG repeat and negatively regulates telomere length . Here we show that binding of multiple Rap1 proteins stimulates DNA double-stranded break ( DSB ) induction at both telomeric and non-telomeric regions . Consistent with the role of DSB induction , Rap1 stimulates nearby recombination events in a dosage-dependent manner . Rap1 recruits Rif1 and Rif2 to telomeres , but neither Rif1 nor Rif2 is required for DSB induction . Rap1-mediated DSB induction involves replication fork progression but inactivation of checkpoint kinase Mec1 does not affect DSB induction . Rap1 tethering shortens artificially elongated telomeres in parallel with telomerase inhibition , and this telomere shortening does not require homologous recombination . These results suggest that Rap1 contributes to telomere homeostasis by promoting chromosome breakage .
Telomeres are specialized nucleoprotein complexes at the ends of linear eukaryotic chromosomes . The DNA component of telomeres typically comprises a double-stranded DNA ( dsDNA ) region of a tandem repeat and a 3’ protruding single-stranded DNA ( ssDNA ) region of the G-rich strand [1 , 2] . Both the dsDNA and ssDNA regions are covered with sequence-specific binding proteins . Telomeres protect chromosome ends from degradation or fusion [2 , 3] . Telomeres also promote DNA replication at the chromosome ends . Since conventional DNA polymerases cannot complete DNA synthesis at telomeres , linear chromosomes shorten progressively with every round of cell division . In most eukaryotes , continuous telomere shortening can be counteracted by telomerase [4] . The length of the duplex telomeric repeat is kept within a relatively narrow range in a cell-type specific manner [1] . In cells that express telomerase , telomere length homeostasis results from a balance between telomerase-dependent telomere addition and telomere shortening . The average telomere length varies between 5 and 15 kb in human , whereas much shorter telomeres ( ~300 bp ) are maintained in the budding yeast Saccharomyces cerevisiae . Telomere shortening results from incomplete DNA end replication and 5’-resection ( end-replication problem ) [2 , 5] . For example , in the absence of telomerase , telomeres of human and budding yeast cells lose 50–200 and 3–5 nucleotides per round of DNA replication , respectively [6–9] . Telomere shortening can also occur in a more accelerated manner called telomeric rapid deletion ( TRD ) , some of which involves intrachromosomal recombination events between telomeric repeats [10–13] . The structure of telomeres in budding yeast is typical of most eukaryotes , except that the repeat unit ( abbreviated TG1-3 ) is heterogeneous [2] . As in other eukaryotes , telomeres consist of double-stranded and single-stranded units in budding yeast . The G-rich strand extends to form a single-strand tail , which is covered with Cdc13 in complex with Stn1 and Ten1 [14–16] . The Cdc13-Stn1-Ten1 complex acts as a telomere cap to protect telomeres from degradation [2] . Telomerase comprises the catalytic subunit Est2 , two accessary subunits ( Est1 and Est3 ) and the RNA component Tlc1 in budding yeast [2 , 4] . Cdc13 interacts with Est1 and contributes to telomerase recruitment to chromosome ends [2 , 4] . Double-stranded telomeric DNA repeats are bound by the sequence-specific binding protein Rap1 . The budding yeast Rap1 protein is an essential protein involved in many diverse processes including transcription and telomere maintenance [17] . Rap1 consists of three conserved domains: a BRCT domain in the N-terminal region , a centrally located DNA-binding domain with two Myb-like folds , a transcription activation ( TA ) domain , and an independent C-terminal domain called RCT ( Rap1 C-terminal ) [17] . The central region , containing two Myb domains and the TA domain , plays an essential role in cell viability by regulating transcriptional activation of various genes [18] . The N-terminus has been shown to potentiate DNA bending of the Rap1-DNA complex [19] . The RCT domain regulates localization of Sir3 and Sir4 and promotes transcriptional silencing [20] . This domain also forms a complex with and recruits Rif1 and Rif2 to telomeres [21–24] . Targeting of the RCT , Rif1 or Rif2 to a telomere induces telomere shortening to an extent that is proportional to the number of targeted molecules [25 , 26] . This observation has led to the “protein counting model” of telomere length control , in which increasing numbers of telomere-bound Rap1 molecules negatively regulates telomerase-mediated telomere extension . The “protein counting model” has been defined by recent studies . As in other eukaryotes , the checkpoint protein kinases Mec1 and Tel1 control telomere length in budding yeast [2] . Although Mec1 is a central checkpoint kinase in the response to DNA replication block and DNA damage [27] , it has only minor functions in telomere homeostasis . In contrast , Tel1 plays a major role in telomere length maintenance in budding yeast [28] . Tel1 localizes to short telomeres as well as DNA double-strand breaks ( DSBs ) [29–32] . In turn , Tel1 recruits telomerase to short telomeres and promotes telomere addition [32–34] . Telomere extension increases Rap1 binding sites at telomeres . Rap1 collaborates with Rif1 and Rif2 and inhibits localization of Tel1 to DNA ends [35 , 36] . Thus , Rap1 cooperates with Rif1 and Rif2 to impede telomerase recruitment , therefore negatively regulating telomere length . The G-rich strand of telomeric TG repeats can fold into G-quadruplex ( G4 ) structures , which could cause replication fork blocks and induce DNA breaks [37] . Indeed , internal tracts of TG repeats are unstable and spontaneously converted to telomeres [38] . Rap1 binds internal tracts of TG repeat sequence as well as telomeres [21 , 22 , 35] . Since Rap1 binds to duplex DNA , binding of Rap1 to TG repeat sequence could inhibit G4 structure formation . However , there is evidence suggesting that Rap1 also binds to single-stranded DNA and stimulate G4 formation [39] . Alternatively , it is formally possible that DNA binding of Rap1 promotes DNA break induction independently of the G4 structure . In this study we provide evidence supporting that binding of multiple Rap1 proteins is involved in chromosome breakage during S phase . We devised systems to determine whether DNA binding of Rap1 stimulates DSB induction . We found that Rap1 mediates DSB induction at both non-telomeric and telomeric regions . While Rap1 operates together with Rif1 and Rif2 to inhibit telomerase recruitment , Rap1 acts independently of Rif1 or Rif2 function to induce chromosome breakage . Rap1 tethering prompts DSB generation in a copy number-dependent manner during DNA replication . While Rap1-mediated DSB induction shortens artificially extended telomeres , this telomere shortening does not involve homologous recombination events . Our results suggest that Rap1 contributes to telomere homeostasis by inducing DNA breaks .
We examined whether Rap1 is involved in DSB induction , thereby converting internal tracts of TG sequence to telomeres . To this end , we placed an 81 bp ( TG81 ) or a 250 bp telomeric TG ( TG250 ) repeat sequence between the KanMX marker and the K . lactis URA3Kl gene at the ADH4 locus ( Fig 1A ) . The TG81 or TG250 sequence , derived from endogenous telomeres , contains four or twelve Rap1 binding motifs , respectively [36 , 40] . There is no essential gene from the ADH4 locus to the chromosome end . Eukaryotes utilize two major pathways for DSB repair; homologous recombination ( HR ) and non-homologous end joining ( NHEJ ) [27] . In budding yeast , HR is the central DSB repair pathway . However , HR cannot efficiently repair centromere-proximal DSB ends generated between KanMX and URA3Kl because there is no homologous donor sequence available ( see below ) . Instead , telomerase-dependent telomere addition occurs at DNA ends with telomeric TG repeat sequence nearby [25 , 40] , a phenomenon which is referred to as “telomere healing” ( Fig 1B ) . Although URA3 cells cannot proliferate on medium containing 5-fluoroorotic acid ( 5-FOA ) , ura3 mutant cells can [41] . Therefore , if a DNA break is induced within or near the TG sequence , telomere formation can eliminate the distal portion including the URA3Kl marker gene , therefore generating 5-FOA resistant colonies . Cells were first maintained in medium selective for URA3 cells and then transferred to non-selective medium . Saturated cultures were diluted and spread on 5-FOA plates to monitor the rate of URA3Kl marker loss ( Fig 1C ) . The URA3Kl marker was stably maintained if there was no TG repeat ( TG0 ) sequence . Introduction of the TG250 repeat sequence , however , stimulated loss of the URA3Kl marker very efficiently ( 19 , 000-fold ) . Placement of the TG81 repeat increased URA3Kl marker loss ( 430-fold ) but much less efficiently compared with the TG250 sequence . All of the twenty Ura- cells examined possessed a telomere near the TG81 or TG250 sequence ( S1 Fig ) [38] . Consistent with telomerase-dependent telomere addition at TG sequences , inactivation of the RAD52-dependent HR pathway did not affect URA3Kl marker loss [40] ( Fig 1C ) . We examined whether chromosome breakage occurs near the TG250 repeat sequence by Southern blotting analysis . Cells carrying the TG0 or TG250 cassette were first cultured in medium selective for URA3 cells and then transferred to non-selective medium . Introduction of the TG250 repeat accumulated cells containing a DNA end nearby ( Fig 1D ) . Several lines of evidence have established the model in which persistent DNA fork stalling leads eventually to DSB induction [42] . Replication forks slow during their passage through telomeric TG tracts [43] . We confirmed that replication forks paused at the TG250 repeat sequence by two-dimensional gel electrophoresis analysis ( Fig 1E and S2 Fig ) . We investigated whether Rap1 is required for URA3Kl marker loss . Since the RAP1 gene is essential for cell proliferation , we examined the effect of partial Rap1 depletion using a copper-inducible rap1 degron , rap1- ( Δ ) [44] . Consistent with an essential role of Rap1 in cell proliferation , rap1- ( Δ ) mutants grew very poorly in the presence of 0 . 5 mM CuSO4 ( Fig 2A ) as the expression of Rap1 degron protein was decreased ( Fig 2B ) . In contrast , incubation with 0 . 05 mM CuSO4 did not significantly affect cell proliferation ( Fig 2A ) although the Rap1 expression level was decreased ( Fig 2B ) . We thus investigated the effect of rap1- ( Δ ) mutation on URA3Kl marker loss in the presence of 0 . 05 mM CuSO4 ( Fig 2C ) . Partial Rap1 depletion was found to decrease URA3Kl marker loss . These results are consistent with the hypothesis that Rap1 promotes DSB induction in a copy number-dependent manner . Rap1 comprises three conserved domains: a BRCT domain in the N-terminal region , a central region with DNA-binding Myb and TA domains , and a C-terminal RCT domain [17] ( Fig 1C ) . Although the central region is essential for cell viability , the BRCT or RCT domain is not [17 , 45 , 46] . To determine the role of the BRCT or the RCT domain , we examined the effect of rap1-ΔN or rap1-ΔC mutation on loss of the URA3Kl marker . However , neither the N-terminal nor the C-terminal deletion significantly affected the generation of Ura- cells ( Fig 1C ) , raising a possibility that the central region of Rap1 protein is involved in DSB induction . The above results support the model in which binding of multiple Rap1 proteins results in DSB induction . However , it remains possible that Rap1 binding mediates the formation of G-quadruplex ( G4 ) structures at TG repeats , which could cause replication fork blocks and induce DNA breaks [37] . The TG81 and the TG250 repeat sequence can potentially form two and eight G4 structures , respectively [47] . Moreover , Rap1 binding could stimulate telomerase activity at DNA ends . To exclude these possibilities , we set up a system that recruits Rap1 to non-TG sequences ( Fig 3A ) . We constructed a LacI-Rap1 fusion ( Fig 3B ) , which restores proliferation to rap1Δ mutants ( S3 Fig ) and functions as a negative regulator of telomere length ( see below ) . Expression of LacI-Rap1 fusion was driven from the GAL1 promoter in medium containing 2% galactose and 0 . 5% glucose ( galactose hereafter ) , thereby maintaining the expression level of LacI-Rap1 protein similar to that of endogenous Rap1 protein ( S4 Fig ) . To target LacI-Rap1 to non-TG sequences , we inserted cassettes with different copy numbers of the LacI-binding sequence ( lacO ) ( Fig 3A ) . No putative G4 forming sequence was found on the lacO sequences [47] . One LacI homodimer can bind to each lacO operator [48]; therefore , one lacO copy can be covered with two LacI-Rap1 molecules . However , the LacO4 cassette behaved like ~80 bp of telomeric TG sequence containing four Rap1 binding sites after LacI-Rap1 expression ( see below ) . The average telomere length is ~300 bp and the Rap1 binding motif appears every 20 bp at telomeres [2] . Thus , the LacO16 repeat in the presence of LacI-Rap1 could correspond to wild-type length telomere ( ~320 bp of TG repeat sequence ) in terms of Rap1 binding . We introduced a 3’-terminal truncation of URA3Kl adjacent to the lacO sequences at the ADH4 locus and a 5’-terminal truncation of URA3Kl on a different chromosome ( Fig 3A ) . In this system DNA breaks at the lacO sequences can be repaired by homologous recombination , generating the full-length URA3Kl marker gene . We thus determined the homologous recombination frequency between the truncated ura3Kl genes to estimate LacI-Rap1-mediated DSB induction . Cells carrying pGAL-LacI-RAP1 or the control vector were initially grown in sucrose and then incubated with galactose to express LacI-Rap1 . Saturated cultures were diluted and plated to score Ura+ cells ( Fig 3A ) . LacI-Rap1 expression stimulated interchromosomal recombination more efficiently when cells contained longer lacO sequences ( Fig 3B ) . For example , LacI-Rap1 expression led to 2 , 000-fold higher recombination events in cells containing the LacO16 cassette compared to cells containing no lacO ( the LacO0 cassette ) . LacI alone stimulated recombination near the LacO16 cassette but much less efficiently than LacI-Rap1 fusion ( Fig 3C ) . The observed effect is not specific to LacI-fusion , since TetR-Rap1 fusion also stimulated recombination near the tetO repeat ( S5 Fig ) . Since the N-terminus or C-terminus of Rap1 was dispensable for TG-mediated chromosome truncation ( see Fig 1C ) , we examined whether the central region of Rap1 stimulates interchromosomal recombination ( Fig 3C ) . We constructed a LacI-Rap1 fusion lacking both the N- and C-termini of Rap1 , named LacI-Rap1 ( 224–663 ) ( Fig 3B ) . The LacI-Rap1 ( 224–663 ) fusion stimulated recombination strongly compared with LacI alone but weakly compared with LacI-Rap1 ( Fig 3C ) . The expression level of LacI alone and LacI-Rap1 ( 224–663 ) was not significantly differently from that of LacI-Rap1 ( S4 Fig ) . Although the central domain of Rap1 triggered recombination less efficiently than the full-length Rap1 protein , deletion of the central region abolished recombination stimulation; LacI-Rap1 ( ΔM1-M2-TA ) behaved similarly to LacI alone ( S6 Fig ) . These results support the idea that the central region of Rap1 plays a key role in DSB induction . The central region of Rap1 consists of two Myb domains and a TA domain . Neither Myb nor TA domain was specifically involved in recombination stimulation , suggesting that the overall structure of the central region is critical for DSB induction ( S6 Fig ) . The C-terminal region of Rap1 mediates interaction with Rif1 or Rif2 for telomere homeostasis and Sir3 or Sir4 for transcriptional repression [2 , 17] . In agreement with the finding that the C-terminus is dispensable for Rap1-mediated DSB induction , neither rif1Δ rif2Δ nor sir3Δ sir4Δ double mutation affected recombination ( Fig 3D ) . It is possible that any transcriptional activation protein stimulates DSB induction similar to Rap1 . Gal4 is a potent transcription activator and the calculated molecular mass of Gal4 is similar to that of Rap1 . We examined the effect of LacI-Gal4 expression on DSB induction ( Fig 3C ) . LacI-Gal4 expression increased recombination compared with the control vector but behaved like LacI alone . The expression level of LacI-Gal4 was similar to that of LacI-Rap1 ( S4 Fig ) . Both Rap1 and LacI bind to the respective consensus sequence with a high affinity ( Kd = ~1x10-11M ) [49 , 50] . One explanation could be that DSB induction involves tight DNA binding . We addressed this possibility by using the LacI** variant that has weak affinity for the binding sequence but does not impair the accumulation at lacO repeat sequences [51] . We found that LacI**-Rap1 protein was defective in DSB induction ( Fig 3C ) although the expression of LacI-Rap1** was similar to that of LacI-Rap1 ( S4 Fig ) . Thus , anchoring of Rap1 appears to promote DSB induction rather specifically . We investigated whether LacI-Rap1 tethering induces DNA breaks at the LacO16 repeat by Southern blot analysis . We used a strain carrying the LacO16 cassette between KanMX and URA3Kl at the ADH4 locus ( Fig 4A ) . Cells transformed with pGAL-LacI-RAP1 or pGAL-LacI were grown in sucrose and then transferred to galactose medium for 4 hr . LacI-Rap1 expression induced DNA breakage near the LacO16 repeat whereas no apparent cleavage was detected with LacI expression ( Fig 4B ) . Neither Ku-dependent NHEJ nor Rad52-dependent HR significantly affected DNA breakage detection ( Fig 4B ) . This observation is consistent with the view that NHEJ is a minor pathway in budding yeast [27] and there is no homologous donor sequence available for DNA breaks generated between KanMX and URA3Kl . It was estimated that 3% of cells received a DNA break at the LacO16 locus 4 hr after LacI-Rap1 expression ( S7 Fig ) . The terminal deoxynucleotidyl transferase ( TdT ) has been used to end-label DNA ends including DSBs and telomeres [52] . To confirm that LacI-Rap1 tethering generates DNA breakage , we used an assay by combining TdT-mediated end-labeling and PCR ( Fig 4C and S8 Fig ) . In this assay PCR amplification detects the addition of G-tracts at 3’-DNA ends near the LacO16 repeat . PCR amplified DNA fragments that correspond to DSB induction at the LacO16 repeat in both directions after LacI-Rap1 expression whereas no discrete band was observed after LacI expression , indicating that LacI-Rap1 expression results in DSB induction near the LacO16 repeat . We next addressed in which cell cycle stage LacI-Rap1 expression induces DNA breaks by the TdT-based PCR assay . Cells were grown in sucrose and treated with α-factor or nocodazole to synchronize in G1 or G2/M phase , respectively ( Fig 5A ) . Synchronized cells were then incubated with galactose to induce LacI-Rap1 expression . No DSB induction was detected in G1 or G2/M-arrested cells . We next examined whether DNA breakage occurs in S phase ( Fig 5B ) . Cells arrested with α-factor in G1 were incubated with galactose to express LacI-Rap1 fusion . Cells were then released from α-factor arrest or remained arrested . DNA flow cytometry analysis confirmed that cells underwent S phase after α-factor release ( Fig 5B ) . DSB induction was detected in S phase after α-factor release but not in G1-arrested cells . Thus , LacI-Rap1 expression leads to DSB induction at the LacO16 repeat during S phase . To address whether DSB induction is coupled with DNA replication , we examined the effect of temperature-sensitive cdc17-1 mutation on break induction during S phase ( Fig 5C ) . CDC17 encodes a catalytic subunit of DNA polymerase α . Wild-type or cdc17-1 mutants carrying pGAL-LacI-RAP1 were arrested with α-factor in G1 and incubated with galactose at the restrictive temperature and then released from α-factor arrest . We then monitored DSB induction by Southern blotting analysis ( Fig 5C ) . While wild-type cells underwent DNA replication , cdc17-1 mutants arrested in early S phase ( Fig 5C ) . The cdc17-1 mutation suppressed DSB induction . These results show that replication fork progression in S phase is required for DSB induction . We analyzed replication fork progression at the LacO16 locus after LacI-Rap1 or LacI expression by two-dimensional gel electrophoresis ( Fig 6A and S9 Fig ) . Cells carrying pGAL-LacI-RAP1 , pGAL-LacI , or the control vector were initially grown in sucrose and then incubated with galactose for 4 hr to induce LacI-Rap1 or LacI expression . There was no replication fork pausing in cells carrying the control vector . Experiments using two different sets of restriction enzymes confirmed that replication forks pause at the LacO16 repeat . Replication fork pausing was seen after both LacI expression and LacI-Rap1 expression but corresponding signals were only two-fold more intense in cells expressing LacI-Rap1 than in those expressing LacI alone . As described above , however , LacI-Rap1 expression resulted in DNA breakage much more robustly than LacI expression ( Fig 4B and S7 Fig ) ; LacI-Rap1 was estimated to promote DSB induction ~100-fold more strongly than LacI alone ( Fig 3C ) . Thus , fork stalling per se did not fully explain the mechanism of DSB induction after LacI-Rap1 expression . The ATR/Mec1 checkpoint pathway has been proposed to facilitate replication progression and prevent chromosome breakage during replication stress [42 , 53] . We tested the possibility that Rap1 impairs Mec1 function that prevents chromosome breakage ( Fig 6B ) . If this were the case , LacI expression could generate DNA breaks as efficiently as LacI-Rap1 expression in mec1Δ mutants . However , DSB induction was still hardly detectable in mec1Δ mutants after LacI expression . Moreover , the introduction of mec1Δ mutation did not elevate DSB induction after LacI-Rap1 expression ( Fig 6B ) . Mrc1 and Tof1 stabilize DNA replication forks and contribute to the activation of the Mec1 pathway during replication stress [54–57] . Previous studies have suggested that Mrc1 and Tof1 have a role in telomere stability in addition to fork progression [58–60] . We examined the effect of mrc1Δ or tof1Δ mutation on DNA break induction at LacO16 after LacI-Rap1 expression ( Fig 6C ) . Neither mrc1Δ nor tof1Δ mutation increased the frequency of DSB induction . Thus , inactivation of Mec1 checkpoint function does not affect Rap1-induced DSB induction . We addressed whether Rap1 mediates DSB induction at telomeric regions . If the above model were applied to telomeres , Rap1 binding could generate DNA breakage at extended telomeres and truncate them . In parallel , however , Rap1 inhibits telomerase recruitment at extended telomeres and negatively regulates their length . To distinguish between these two different types of telomere shortening , we developed a system that generates an artificially elongated telomere . In this system the LacO16 sequence is integrated between a 33 bp TG sequence ( TG33 ) and the VII-L telomere ( Fig 7A ) . Non-telomeric sequences can be counted as telomeric sequences if Rap1 is anchored [25] . Therefore , the TG33-LacO16-telomere becomes an extended telomere mimic after LacI-Rap1 expression ( Fig 7B ) . Once converted to an extended telomere , telomeres become shorter gradually ( 3–5 nucleotides per generation ) because of the end-replication problem . While telomeric TG sequence is retained , Cdc13-telomere capping blocks DNA degradation [2] . However , once telomere shortening reaches the LacO16 sequence after 60–90 generations , the DNA end loses Cdc13-dependent protection and exonuclease activities start degrading the LacO16 repeat . It is estimated that DNA degradation occurs at the rate of 4 kb/hour [61] . Previous studies showed that 33 bp TG repeat sequences act as a telomere seed in vivo [25 , 62] . We have shown that DNA ends with 22 bp of TG repeat nearby are efficiently converted to telomeres [63] . If DNA degradation reaches to TG33 repeat sequences , telomere extension occurs using TG33 , generating TG-telomeres . However , DSB induction near or within LacO16 could skip gradual telomere shortening that results from the end-replication problem . Thus , DSB induction can be detected as swift conversion from the TG33-LacO16-telomere to TG-telomere . As mentioned above , there is no putative G4 forming sequence on the LacO repeat sequences . This system therefore enables us to examine the effect of Rap1 binding on DSB induction without increasing the length of TG repeat sequence that potentially generates G4 structures . We first confirmed that targeted LacI-Rap1 to the LacO4 repeat behaves in a manner identical to Rap1 with respect to telomere length control . We used a strain with the LacO4 repeat adjacent to VII-L telomere ( LacO4-telomere ) . Previous studies showed that the GAL4 binding site adjacent to telomeres behaves in a manner similar to a telomere sequence after expression of a Gal4-Rap1 fusion protein [25] . Cells containing the LacO4-telomere were transformed with pGAL-LacI-RAP1 or pGAL-LacI and cultured in sucrose to stationary phase . The cultures were diluted 1 , 000-fold in galactose and grown for 24 hr ( eight generations ) for successive serial dilutions . Cells were collected at each dilution point and the length of LacO4-telomere was monitored by Southern blot analysis ( Fig 7C ) . When LacI-Rap1 was expressed , the length of the telomere repeat tract was gradually decreased as telomerase activity is down regulated at longer telomeres [25] . In contrast , the expression of LacI alone did not have any effect on telomere length . LacO4-telomeres became about 80 bp shorter after the serial dilution culture ( Fig 7C ) . Thus , the lacO array sequence is counted as a part of the telomere after LacI-Rap1 expression , supporting the hypothesis that the TG33-LacO16-telomere with LacI-Rap1 coated on the lacO sequence behaves like an extended telomere . We then examined the effect of LacI-Rap1 or LacI expression on the length of TG33-LacO16-telomeres . Cells containing TG33-LacO16-telomere carrying pGAL-LacI-RAP1 or pGAL-LacI were cultured as above and telomere length was monitored by Southern blot analysis ( Fig 7D ) . As LacI expression had no impact on the length of LacO4-telomeres ( Fig 7C ) , it did not affect the length of TG33-LacO16-telomere . In contrast , if LacI-Rap1 was expressed , two different types of telomere shortening were observed . First , TG33-LacO16 telomeres ( PRE ) became gradually shorter consistent with the above finding that LacI-Rap1-covered lacO sequence is counted as a telomere sequence . This observation indicates that telomerase activity is inhibited because TG33-LacO16 telomeres behave like an extended telomere in the presence of LacI-Rap1 . Second , LacI-Rap1 expression generated shorter telomeres ( TG-telomeres ) corresponding to those extended from the TG33 repeat , consistent with the hypothesis that DSB induction occurs near or within the LacO16 repeat . As Rap1-mediated DSB induction occurs during S phase , continuous culturing increased the population size of cells containing TG-telomeres . 50% of the cells converted from the TG33-LacO16 telomere to the TG-telomere after three successive dilutions ( 24 generations ) ; it was estimated that 2% of cells received a DNA break within or adjacent to the LacO16 repeat per generation . To confirm that the rapid generation of TG-telomeres does not result from the inhibition of telomerase activity , we compared telomere shortening after LacI-Rap1 expression and telomerase depletion ( Fig 7E ) . We introduced a deletion mutation of EST1 , which encodes a regulatory protein for telomerase , in cells containing the TG33-LacO16 telomere . Growth of the est1Δ strains was maintained by wild-type EST1 on a URA3-marked plasmid . Wild-type cells and est1Δ mutant cells were transformed with pGAL-LacI-RAP1 and pGAL-LacI , respectively . To select for loss of the complementing EST1 plasmid , cells were grown on sucrose plates containing 5-FOA . The resulting single colonies were inoculated into galactose medium ( 1st dilution ) . After 24 hr culture , it was diluted 1000-fold for successive serial dilutions . Southern blotting analysis confirmed that the inhibition of telomerase activity caused slow telomere shortening ( ~30 bp during each dilution ) but did not promote TG-telomere formation ( Fig 7E ) . As discussed above , the generation of TG-telomeres was dependent on telomerase activity ( S10 Fig ) . To exclude the possibility that the emergence of TG-telomeres results from homologous recombination among telomeric sequences , we tested the effect of rad52Δ mutation on shortening of the TG33-LacO16-telomere after LacI-Rap1 expression ( Fig 7F ) . Cells were first grown in sucrose and cultured continuously in galactose to express LacI-Rap1 after serial dilutions . The rad52Δ mutation did not affect shortening of TG33-LacO16 telomeres after LacI-Rap1 expression . Thus , it is unlikely that Rap1 binding stimulates homologous recombination between telomeric TG repeat sequences . Collectively , our results support the model in which Rap1 binding is involved in telomere shortening by introducing DNA breaks .
Rap1 binds to double-stranded telomeric TG-repeat sequences and recruits Rif1 and Rif2 proteins via its C-terminal domain [21–24] . Previous studies have uncovered a negative-feedback mechanism that counts telomere-binding Rap1 protein to control telomere length [25 , 26] . In this feedback loop , Rap1 collaborates with Rif1 and Rif2 and inhibits the localization of the protein kinase Tel1 to adjacent DNA ends , thereby attenuating the recruitment of telomerase to long telomeres [30–32 , 34–36] . In this report we have provided evidence suggesting that an alternative Rap1-dependent mechanism operates to trim elongated telomeres . In this telomere shortening mechanism , Rap1 binding coupled with DNA replication promotes DSB induction independently of Rif1 or Rif2 . These observations support a model in which Rap1 negatively regulates telomere length through two distinctive mechanisms . DNA breaks adjacent to telomeric TG repeats can be readily converted to telomeres by telomerase [38 , 40] . It is therefore complicated to detect DNA breaks at telomeres . We have developed several experimental systems to monitor DSB induction and shown that binding of multiple Rap1 proteins induces DNA breaks at both telomeric and non-telomeric regions . Rap1-mediated DSB induction appears to operate in a dosage-dependent manner . First , longer TG repeats trigger chromosome truncation more efficiently than shorter TG repeats . Second , LacI-Rap1 expression promotes nearby recombination at longer lacO arrays more frequently than at shorter ones . As discussed above , the LacO16 repeat in the presence of LacI-Rap1 could correspond to wild-type length telomere in terms of Rap1 binding . To detect DSB induction at telomeres , we used TG33-LacO16-telomeres , which correspond to ~600 bp long telomeres in the presence of LacI-Rap1 . Two percent of TG33-LacO16-telomeres were estimated to receive DNA breaks near or within the LacO16 repeat every cell division in the presence of LacI-Rap1 . Although the frequency is low ( 2% per generation ) , approximately 50% of cells would receive DNA breaks after 32 generations . DNA ends adjacent to TG repeats , once generated , are protected from DNA degradation by Cdc13-mediated telomere capping [2 , 35] . Thus , Rap1-mediated DSB induction at telomeres seems to be a physiological phenomenon that keeps telomeres in normal-length ranges , although it remains possible that telomere-specific activity suppresses DSB induction at native telomeres . Rap1 collaborates with Rif1 and Rif2 to inhibit telomerase recruitment [64] . In contrast , Rif1 and Rif2 are dispensable for Rap1-mediated DSB induction . Telomerase inhibition steadily shortens telomeres 3–5 bp per cell division because of the end-replication problem [6 , 9] . In contrast , DSB induction can delete longer telomere sequence although it may not constantly occur . The extent of telomerase-dependent telomere extension varies at each telomere and is independent of telomere length [65] . Cells take full advantage of different telomere shortening modes to cope with the heterogeneous nature of telomeres . Replication forks in yeast and other organisms move more slowly through telomeric DNA than non-telomeric regions [43 , 66–68] . This replication problem is thought to result from the G-rich nature of telomeric DNA , which allows it to form G4 DNA . G4 DNA interferes with DNA replication and generates DNA breaks [37] . In addition to G4 structures , we have shown that binding of multiple Rap1 proteins generates a barrier that stimulates DSB induction during DNA replication . Persistent fork stalling could lead to fork collapse and subsequent DSB induction [42] . Consistent with this model , tight binding of LacI was critical for LacI-Rap1-mediated DSB induction . However , fork stalling by itself does not appear to result in DSB induction at Rap1 bound regions . We found that LacI-Rap1 expression induced DSB formation about 100-fold more efficiently than LacI expression alone whereas the rate of fork stalling after LacI-Rap1 expression was only two-fold higher compared with LacI expression alone . Moreover , larger complex formation with Rif1 , Rif2 , Sir3 and Sir4 did not increase the frequency of DSB induction . It seems less likely that DSB induction results from DNA bending because the N-terminus is dispensable [19] . Several lines of evidence have shown that paused forks by themselves are relatively stable [69] . Indeed , 40% of the replicating intermediates contained pausing forks at the LacO16 repeat after LacI expression but 0 . 02% of cells were expected to receive DNA breaks at the LacO16 repeat per generation . The ATR/Mec1 checkpoint pathway has been proposed to prevent chromosome breakage during replication stress [42 , 53] . However , inactivation of the Mec1 checkpoint pathway did not affect DSB induction after LacI or LacI-Rap1 expression . Thus , Rap1 appears to impair other mechanisms than Mec1 checkpoint function at DNA replication forks . We found that the overall structure of the central Rap1 region plays an important role in DSB induction . It has been shown that the central region of Rap1 inhibits NHEJ at telomeres [70] . This Rap1 region , once arrayed , might disrupt replication complexes as well as repair machinery along the DNA tract . We note that not all telomere-binding proteins have a negative impact on DNA replication . Taz1 in fission yeast and TRF1 in human have been shown to promote DNA replication at telomeres [67 , 68] . Interestingly , Taz1 and TRF1 possess a Myb domain as does the Rap1 central region . The central region of Rap1 also stimulates DSB induction for meiotic recombination [46]; however , meiotic DSB induction occurs through a different mechanism . Transcription factors including Rap1 generate nucleosome free regions where Spo11 binds to chromatin and catalyzes DSB [71] . Over-elongated telomeres can be shortened to normal length through a mechanism termed telomeric rapid deletion ( TRD ) [10] . TRD appears to result from an intra-chromosomal recombination event between telomeric repeats because TRD depends on the major recombination protein , Rad52 [10] . In addition , hpr1 mutation , which increases recombination between direct repeats , elevates the rate of TRD [10] . Similar telomere shortening has been observed in other systems including human cell lines and most likely involves homologous recombination-mediated removal of telomere loops [11–13] . While TRD largely depends on Rad52 function , some fraction of TRD was found to occur in a Rad52-independent manner [10] . Since DSBs near telomeric TG sequences are healed by telomere addition , DSB generation at over-elongated telomeres would lead to Rad52-independent TRD . G4 structure formation could promote DSB induction at telomeres [37] , thereby trimming telomeres independently of Rad52 function . However , our results suggest that Rap1-mediated DSB induction contribute to Rad52-independent TRD as well . Indeed , TG33-LacO16-telomeres became shorter independently of Rad52 function or the end-replication problem after LacI-Rap1 expression . In summary , we have provided evidence indicating that binding of multiple Rap1 proteins promotes DNA break induction during DNA replication . Since Rap1 binds extensively at telomeric DNA regions [72] , it seems likely that Rap1 binding promotes DSB induction at telomeres . Rap1 binds to telomeres and controls their function in other eukaryotes [73–75] . Given that telomeres consist of repetitive sequences and sequence-specific binding proteins , a similar system may function in other organisms as well .
The strain carrying the TG0-URA3 , TG81-URA3 , TG250-URA3 or LacO16-URA3 cassette was generated by the pNO-URA3 , pT81-URA3 , pT250-URA3 or pO16-URA3 plasmid after digestion with NotI and SalI , respectively . The strain carrying the LacO0-ura3-ΔC , LacO4-ura3-ΔC , LacO8-ura3-ΔC or LacO16-ura3-ΔC or TetO8-ura3-ΔC cassette was generated by the pUN-O0 , pUN-O4 , pUN-O8 , pUN-O16 or pUN-tetO8 plasmid after digestion with EcoRI and SalI , respectively . The strain containing the LacO4-telomere , TG33-LacO16-telomere , or TG-telomere was generated by the pO4-TG-HO , pTG33-O16-TG-HO or pTG81-HO plasmid after digestion with EcoRI and SalI , respectively . The ura3-ΔN-Hph cassette was introduced into the YER186 locus by a PCR-based method [76] . The copper-inducible RAP1 degron ( rap1- ( Δ ) ) construct has been described [44] . The N-terminal deletion of RAP1 ( rap1-ΔN ) has been described [46] . TG-telomere cells were generated from TG81-HO cells after HO expression [35] ( See S1 Fig ) . The mec1Δ , rad52Δ or sml1Δ mutation has been described [77] . The mec1-81 or tel1Δ mutation has been published [29] . The est1Δ mutation has been described [78] . The C-terminal rap1 truncation ( rap1-ΔC ) , rif1Δ and rif2Δ mutation have been described [35] . The rap1-ΔC allele encodes the same truncated Rap1 proteins as the rap1-17 mutation does [26 , 45] . Disruption of SIR3 was performed as described [79] . The copper inducible protein degradation system was described [80] . The SIR4 disruption plasmid was obtained from Masayasu Nomura . The strains used in this study are listed in S1 Table . The probe that detects the KanMX gene was obtained by NotI-digestion of the pFA6-kanMX4 plasmid [79] . The probe that detects DSB induction was a PCR fusion product of the KanMX coding sequence and the LTE1 locus , which were amplified by the primer pair KSX050/89 and KS3004/3005 , respectively . The probe that detects endogenous telomeres or telomere addition at the ADH4 locus has been described [35 , 78] . DNA probes were DIG-dUTP—or 32P-labeled by using the DIG prime ( Roche ) or the Random Primer DNA Labeling Kit ( Clontech ) , respectively . Genomic DNA was purified using a MasterPure yeast DNA purification kit ( Epicentre ) . Cells from a single colony were fully grown in uracil dropout medium . The culture was then diluted 1000-fold and grown in rich medium for 24 hr . Aliquots of the cultures were diluted and plated on plates containing 5-FOA or non-selectable rich medium . Rates per generation were calculated using the FALCOR program based on the Luria-Delbruck fluctuation analysis [81] . We confirmed that essentially all 5-FOA resistant cells derived from TG81-URA3 or TG250-URA3 are Ura- cells . The promoter of URA3Kl was replaced with the strong ADH1 promoter ( see S1 Text ) . More than 500 colonies of each TG81-URA3 , TG250-URA3 , TG250-URA3 rad52Δ or TG250-URA3 rap1- ( Δ ) cells on 5-FOA plates were tested by replica-plating to uracil drop-out plates . None of them grew on uracil-drop out plates . Cells were transformed with TRP1-marked plasmids and grown in tryptophan-dropout medium containing 2% sucrose overnight . The culture was then diluted and grown in tryptophan-dropout medium containing 2% galactose or 2% sucrose and 0 . 5% glucose . Aliquots of the cultures were diluted and plated on uracil-dropout or rich medium to estimate the URA3 recombination frequency after expression of LacI or LacI-fusion protein . Rates per generation were calculated using the FALCOR program [81] . Purified genomic DNA ( 100 ng ) was incubated with one unit of terminal deoxynucleotidyl-transferase ( TdT; New England BioLabs ) in a supplied reaction buffer supplemented with 0 . 2 mM dCTP at 37°C for 60 min , followed by PCR using a poly ( dG ) -oligonucleotide with either the TdT forward or reverse primer . The PCR condition was 33 cycles of denaturation at 94°C for 30 s , annealing at 62°C for 30 s , and elongation at 72°C for 60 s . Sequences of PCR primers are described in S2 Table . Immunoblotting or DNA flow cytometric analysis was performed as described [35 , 78] . Rap1 was detected with affinity-purified antibody ( gift from V . Zakian , Princeton University ) or antibody against the C-terminus ( yC-19 , Santa Cruz biotechnology ) . LacI fusions were detected with anti-LacI antibodies ( clone 9A5 , Millipore ) . Two-dimensional gel electrophoresis was performed in Tris-borate-EDTA as previously described [43] . Details of plasmid construction are described in S1 Text . | Telomere length is maintained primarily through equilibrium between telomerase-mediated lengthening and the loss of telomeric sequence through the end-replication problem . In budding yeast Rap1 protein binds to telomeric TG repeat and negatively regulates telomerase recruitment in a dosage-dependent manner . In this paper we provide evidence suggesting an alternative Rap1-dependent telomere shortening mechanism in which binding of multiple Rap1 proteins mediates DNA break induction during DNA replication . This process does not involve recombination events; therefore , it is distinct from loop-mediated telomere trimming . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Binding of Multiple Rap1 Proteins Stimulates Chromosome Breakage Induction during DNA Replication |
The Toxin Complex ( TC ) is a large multi-subunit toxin encoded by a range of bacterial pathogens . The best-characterized examples are from the insect pathogens Photorhabdus , Xenorhabdus and Yersinia . They consist of three large protein subunits , designated A , B and C that assemble in a 5∶1∶1 stoichiometry . Oral toxicity to a range of insects means that some have the potential to be developed as pest control technology . The three subunit proteins do not encode any recognisable export sequences and as such little progress has been made in understanding their secretion . We have developed heterologous TC production and secretion models in E . coli and used them to ascribe functions to different domains of the crucial B+C sub-complex . We have determined that the B and C subunits use a secretion mechanism that is either encoded by the proteins themselves or employ an as yet undefined system common to laboratory strains of E . coli . We demonstrate that both the N-terminal domains of the B and C subunits are required for secretion of the whole complex . We propose a model whereby the N-terminus of the C-subunit toxin exports the B+C sub-complex across the inner membrane while that of the B-subunit allows passage across the outer membrane . We also demonstrate that even in the absence of the B-subunit , that the C-subunit can also facilitate secretion of the larger A-subunit . The recognition of this novel export system is likely to be of importance to future protein secretion studies . Finally , the identification of homologues of B and C subunits in diverse bacterial pathogens , including Burkholderia and Pseudomonas , suggests that these toxins are likely to be important in a range of different hosts , including man .
Photorhabdus are Enterobacteriaceae which live in an obligate mutualistic association with entomopathogenic nematodes ( Heterorhabditis ) , which invade and kill insects in the soil [1] . Upon invasion of the insect's open blood system the nematode regurgitates Photorhabdus bacteria which release a plethora of toxins to kill the insect and protect the cadaver from invading scavengers and saprophytes [1] . An important class of secreted toxins are the Toxin Complexes ( TC ) [2] , [3] , [4] . These TC toxins constitute large multimeric protein complexes , some of which have been shown to exhibit oral toxicity to a range of insects [5] , [6] . This has made them potential candidates to augment the successful Bacillus thuringiensis Crystal-toxin crop protection technology . TCs were first characterized in the insect pathogens Photorhabdus and Xenorhabdus although it has now become clear that tc gene homologues are in fact widely distributed in a range of other pathogens [2] . These include the Gram-negative human disease agents such as Yersinia and Burkholderia and Gram-positive insect pathogens such as Paenibacillus and B . thuringiensis strain IBL200 ( accession NZ_ACNK01000119 ) . While the TCs of Yersinia entomophaga are active against insects [7] , homologues in other members of the genus , such as Yersinia pseudotuberculosis , appear to be adapted to act upon the mammalian gut [8] . They have also been implicated in mammalian gut colonisation in at least one strain of Y . enterocolitica [9] . The TC is a large multi-subunit toxin comprising three subunits , exemplified by the P . luminescens proteins TcdA , TcdB and TccC ( from here on also referred to as A , B and C-subunits [10] ) . The subunits themselves are large proteins with the examples of TcdA1 , TcdB1 and TccC5 being 2517 aa ( 283 kDa ) , 1477 aa ( 165 kDa ) and 939 aa ( 105 kDa ) respectively . In the Xenorhabdus nematophila TC , which consists of , XptA2 , XptB1 and XptC1 , the subunits apparently assemble in a 4∶1∶1 stoichiometry respectively . In this complex the A-subunits appear to form a tetramer of around 1120 kDa that is able to associate with a tightly bound 1∶1 sub-complex of the B and C-subunits [11] , [12] . Interestingly in Y . entomophaga , the TC structure is predicted to show five-fold symmetry and to also associate with a chitinase enzyme [7] . More recently a high resolution cryo-EM structural model has been proposed to describe the structure and conformational changes that the TcdA1 subunit pentomer can undergo . It is shown to likely perform an “injection-like” process which presumably facilitates the delivery of the toxic B+C sub-complex into the host cell [13] . There has been some progress ascribing biological function to certain TC subunits and domains . For example it has been shown that the Xenorhabdus A-subunits encode a host gut cell receptor binding function , targeted to the membranes of insect brush border cells . Two different A-subunits were shown to ascribe different species specificities . [12] , [14] . Furthermore , various sub-domains of Photorhabdus TC proteins have been investigated by transient expression in transfected mammalian cells [3] . More recently Lang et al demonstrate the mode of action of certain C-subunit C-terminal domains , which cause ADP-ribosylation of actin and RhoA [14] . The C-subunit family proteins all have a common conserved N-terminus and highly variable C-terminal domains . This bi-partite structure is now recognized as a common theme in “polymorphic toxin systems” . That is , many toxin families are seen to contain conserved N-terminal domains , which interact with various secretion systems , yet possess interchangeable and highly variable C-terminal “toxic” domains [15] , [16] , [17] . Certain regions of TC subunit proteins do exhibit homology to certain non-TC proteins . These include the first 361 amino acids ( aa ) of the TcdB1 B-subunit , which shows good homology to the N-terminus of the secreted Salmonella toxin , SpvB [18] . In addition aa154–290 of the TcdA1 A-subunit shows homology to SpvA protein . Like the A and B subunit genes , the spvA and spvB genes are also tightly linked , in this case on the Salmonella virulence plasmid [19] . Furthermore , the C-subunit proteins belong to a much larger family which includes the enigmatic Rhs proteins first discovered in E . coli [20] . Homologues of B and C-subunit genes are frequently seen tightly linked in other bacteria , often in the absence of an A-subunit homologue . Furthermore in strains of Burkholderia and Pseudomonas , homologues of B and C genes are present as genetic fusions , comprising a single long open reading frame . Indeed the increasing numbers of genome sequencing projects have revealed more distant tc B+C homologues in bacteria as diverse as Wolbachia , Mycobacteria , Plesiocystis pacifica and even fungi including Gibberella zeae and Podospora anserine . This supports a much wider role for these protein families beyond insect toxicity . Conversely , A-subunit gene homologues ( exemplified by TcdA1 and TcaAB like proteins ) are typically only seen encoded in genomes that also have B and C gene homologues . This suggests that the B and C sub-complex plays a central role in the biological activity of the TC while the A-subunits facilitate more host specific roles . Evidence suggests that the A-subunit is most likely a combined host-cell targeting and B+C subunit delivery system [13] [21] . It should be noted that when heterologously expressed at high levels , the A-subunit and B+C sub-complex have been shown to exhibit limited oral toxicity independently of one another , although together they form a far more potent complex [3] , [21] , [22] . The production of the TcdA1 protein in transgenic plants [23] also reconstituted partial activity , suggesting these proteins could provide a potential alternative or addition to the well-established cry-toxin pest control technology . Heterologous expression studies of TC homologues from species other than Photorhabdus and Xenorhabdus have also been published , including those of Serratia entomophila , Y . pseudotuberculosis and Y . pestis [8] , [24] , [25] . Furthermore , the heterologous production of TC toxins in Enterobacteria species which associate with termites has also been explored as a novel biological control strategy [26] . Recent work in our laboratory has demonstrated that the TC subunits of P . luminescens are post transcriptionally regulated , with the mRNA only being translated at the time of secretion [27] . Upon secretion the complex normally becomes associated with the outer surface of the cell . However some strains encode a small lipase in the tcd pathogenicity island ( pai ) [28] , named Pdl , which enhances the secretion and facilitates the release of the TC into the surrounding milieu [27] . It remains obscure how the bacterium is able to secrete and assemble such a large multimeric protein complex onto the cell surface . A lack of recognisable export sequences has confounded an understanding of the mechanism of secretion . Here we present an investigation into the functions of different domains of the B and C-subunit proteins . The large number of tc gene homologues in Photorhabdus makes the study of the export process in the original genus difficult . We therefore developed heterologous E . coli models to study the synthesis and export of the TC . We used both inducible expression systems and a cosmid model that utilises the native Photorhabdus expression signals [27] . The cosmid clone ( Fig . 1A ) encompasses the tcd1A , tcdB1 and tccC5 subunit genes from the tcd pai of P . luminescens strain W14 [28] . We have previously demonstrated that E . coli correctly synthesises and secretes an active Tcd complex from this cosmid with no loss of cell viability [27] . This confirms that all sequences required for synthesis and secretion are either associated with the tc genes themselves , or are common to the Escherichia genome . 2D-gel electrophoresis ( Fig . S5 ) and previous heterologous expression studies [3] , [12] indicate the B and C-subunits need to be produced together in the same cytoplasm in order to form a functional sub-complex . Here we demonstrate that the N-terminal domains of the B and C-subunit proteins , which are well conserved across family members , are both essential for secretion of the B+C sub-complex . We also confirm that while the C-terminal domain of the C-subunit is essential for toxicity , that it is irrelevant for secretion . Indeed , replacing the C-terminal domain with a FLAG-tag epitope reveals that it is possible to fuse alternative sequences to this gene and have it translated and exported . Interestingly an in-frame deletion of the B-subunit N-terminal domain strongly reduces production of both itself and a downstream C-subunit gene . These subunits can then no longer be detected in the supernatant . Despite this , the truncated-B+C sub-complex does still retain toxicity . We also show that the A-subunit is not required for the B+C sub-complex secretion . On the other hand we demonstrate that the A-subunit secretion ( but not synthesis ) is dependent upon the C-subunit . Even in the absence of the B-subunit , the C-subunit can facilitate secretion of the A-subunit . Using western blots we confirm that the C-subunit becomes localised to the periplasm and is unable to cross the outer membrane in the absence of a B-subunit . Furthermore , in the absence of a C-subunit , the B-subunit is unable to cross the inner membrane and remains associated with the cell spheroplast . Finally we demonstrate limited toxicity of the TccC5 C-subunit alone by injection into insects . We propose a model ascribing function to structure for the B and C-subunits based upon domain homologies and the experimental results presented here .
The tcdA1 , tcdB1 and tccC5 genes were amplified from P . luminescens strain W14 genomic DNA using rTth DNA polymerase ( Applied Biosystems ) . Polymerase chain reaction ( PCR ) conditions were 1 . 2–1 . 6 mM magnesium acetate , 2 mM each dNTP and 1 mM each primer . Thermocycling was performed as follows: 93°C for 30 s; 55°C for 30 s and 68°C for 8 min , for 30 cycles and final 68°C incubation for 10 min . PCR primers , used for cloning into the arabinose inducible expression vector pBAD30 , were designed to include unique restriction sites for subsequent cloning . The primer sequences ( 5′ to 3′ ) used for cloning the tcdA1 , tcdB1 and N-terminally truncated tcdB1 into pBAD30 were as follows: For tccC5 genes with a FLAG-tag , C5 , Csm and Cosp were first cloned into pFLAG-ctc to create FLAG-fusion genes . They were then subsequently PCR amplified from these templates for cloning into pBAD30 or pBAD30-tcdB1 ( downstream of the tcdB1 gene ) . The primer sequences ( 5′ to 3′ ) were as follows; C5-30XbaI-f: ATTCTAGAAAGGAAGTAAATATGGAAAACATTGACCC; SphI_Flag_R: ATGCATGCCGATCGAGAGATCGATCTTCACTTGTCG . For the N-terminal His-tag cloning of tccC5 ( pC5_N_his ) the gene was initially cloned into pET28a to create a His-fusion before subsequent PCR amplification using this template to create an amplicon to clone into pBAD30 . The primer sequences ( 5′ to 3′ ) were as follows: C5_NdeI_F: ATCATATGGAAAACATTGACCCAAAAC , C5_SphI_R: ATGCATGCTTAATTTGCACTGGATGA and pET28a_XbaI_F: CCTCTAGAAATAATTTTGTTTAACTTTAAGAAGGAGA . PCRs using primers C5_NdeI_F and C5_SphI_R were used for cloning into pET28a . PCRs using primers pET28a_XbaI _F and C5_SphI_R were used for cloning into pBAD30 . For creation of the C-terminal His-tag fusions of tccC5 ( pC5_C_his and pC5sm_his ) , the full length C5 and truncated C5sm genes were cloned into pET28a to create His-fusions before subsequent PCR amplification using these templates to create amplicons for cloning into pBAD30 . The primer sequences ( 5′ to 3′ ) were as follows: pET28a_SphI_R: ATGCATGCGCAGCCGGATCTCAGTGGT and ( 1 ) C5: C5_SacI_pET_R: ATGAGCTCATTTGCACTGGATGATTTGAAATTACCGG and ( 2 ) C5sm: C5sm2004_SacI_pET_R: ATGAGCTCTCCCTGAACATCAAATTGCGTCACCGGG . PCRs using primers C5-30XbaI-f and ( 1 ) or ( 2 ) above were used for cloning into pET28a; PCRs using primers C5-30XbaI-f and pET28a_SphI_R were used for cloning into pBAD30 . Following PCR , amplicons were purified ( using Millipore Montage PCR columns as per instructions ) , cut with the appropriate restriction enzyme ( s ) , and then re-purified prior to ligation and cloning . Cloning vector DNA from pBAD30 or pFLAG-ctc was prepared ( Qiagen miniprep kit as per instructions ) , with the relevant restriction enzyme ( s ) . Ligations were performed at a 3∶1 molar excess of insert to vector using the Promega T4 DNA ligase rapid ligation system . Aliquots of the ligation reaction were electroporated into commercial pre-prepared EC100 E . coli cells ( Epicentre ) and recovered on Luria–Broth ( LB ) agar containing 100 µg/ml ampicillin . Correct constructs were selected by restriction digest of DNA prepared from candidate clones and verified by sequencing before transformation into E . coli BL21 . Plasmid DNA from transformants were checked by restriction digest before storing the strains at −80°C in 15% glycerol-LB . Induction of expression strains was done as briefly described . Glycerol stocks were used to inoculate 5 ml of fresh LB media supplemented with 0 . 2% glucose ( w/v ) and the appropriate antibiotic for selection . Bacteria were grown overnight at 28°C with aeration , 1 ml of this culture was then harvested and resuspended in 100 ml of the same media and incubated at 28°C until an OD600 of 0 . 7–0 . 9 was achieved . Cells were then harvested at room temperature by 10 min centrifugation at 4000 rpm . The pellet was re-suspended in 100 ml of fresh LB , supplemented with the appropriate antibiotic and 0 . 2% ( w/v ) of the pBAD30 promoter ( Para ) inducer , L-arabinose . Cells or trichloroacetic acid ( TCA ) precipitated and concentrated supernatants from 3 hour induced cultures were collected for analysis using SDS-PAGE and western blot to confirm synthesis . Supernatants or cell extract samples were diluted in 1× phosphate-buffered saline ( PBS ) and applied to 1 cm3 disks of artificial wheat germ diet as previously described [22] . Treated food blocks were allowed to dry for 30 min and two neonate M . sexta larvae were placed on each food block before incubation at 25°C . Larvae were allowed to feed for 7 days before being weighed . Growth differences are then expressed as mean larval weight ( MLW ) in grams . Note there was no mortality incurred in these assays as discriminatory toxin dilutions were used . E . coli strains containing the required cosmids ( various tc gene insertional knock outs ) were grown for two days at 28°C with aeration . Cells were removed from culture supernatants by centrifugation and passed through a 0 . 22 µm filter ( Millipore ) . Cell extract samples from heterologous production strains were induced for 3 hours before harvesting by centrifugation , washing in 1×PBS and lysis by sonication . Cell debris and unbroken cells were removed by low speed centrifugation . Samples were mixed at the desired ratio at room temperature for 30 min , diluted as indicated , and then applied to artificial food disks as described for the Manduca sexta oral toxicity bioassay method . Heterologous production strains were induced for 3 hours . Cells were washed in 1×PBS , normalized to an OD600 optical density of 10 . 0 before lysis using sonication ( 10 s on and 10 s off , 45% power regime for 3 min ) . Unbroken cells and debris were removed by low speed centrifugation . Cell-free lysates were fractionated into soluble and insoluble sub-fractions by ultracentrifugation ( 28 000 r . p . m . for 2 h in a Beckman SW40Ti rotor ) . The insoluble fraction was resuspended in 1×PBS to restore its original concentration relative to the soluble fraction proteins in the lysate . Supernatant samples were prepared and concentrated using TCA precipitation . The spheroplast and periplasmic fractions were isolated using a Periplasting Kit as described by the manufacturer ( Epicentre Biotechnologies ) . Protein fractions were separated by 1 dimensional SDS-PAGE and western blotted onto Nitrocellulose using a semi-dry blotter ( Biorad ) . We probed these membranes using a standard protocol with the following antibodies: TcdB1C-terminus anti-peptide antibody ( raised against aa856-YSSSEEKPFSPPNDC-aa869 ) ; TccC1N-terminus anti-peptide antibody ( raised against aa347-FWRNQKVEPENRYVC-aa360 ) , a monoclonal anti-FLAG antibody ( SIGMA ) and monoclonal antibodies against the RNA polymerase andβ-lactamase ( Abcam ) . Immune-reactive bands were visualized using alkaline phosphatase-conjugated anti-rabbit secondary antibody ( SIGMA ) at 1∶5000 and developed using a NBT-BCIP reagent . NCBI accession numbers for the proteins described in this study are as follows: TcdA1 , AAL18486; TcdB1 , AAL18487; TccC5 , AAO17210 .
An E . coli cosmid clone model was previously characterised for the study of TC secretion at native expression levels [27] . From hereon we will use the terms A- , B- and C-subunits to refer specifically to the TcdA1 , TcdB1 and TccC5 proteins encoded on this cosmid . We used transposon insertions to determine the relevance of the A and C-subunit genes on secretion of the B-subunit ( Fig . 1A ) . The transposon insertion points were chosen so as to abolish the majority of each gene . As expected when any of the three subunit genes were knocked out ( KO ) by transposon insertion , then oral toxicity of the supernatant was lost ( Fig . S1 ) . Western blot analysis of these supernatants revealed that KO of the C-subunit prevented secretion of the B-subunit , with all the protein remaining in the whole cell samples ( Fig . 1B* ) . KO of the A-subunit gene did not prevent B-subunit secretion into the supernatant ( Fig . 1B-arrows ) . Note detection of TcdB1 synthesis in the AKO cosmid clone confirms there is no polar effect caused by the transposon insertion . We also constructed a series of inducible expression strains containing the B-subunit gene linked to different truncated versions of C-subunit genes . These truncations had FLAG tags fused at their C-termini to allow detection by western blot ( Fig . 2A ) . We used these constructs to assess the importance of the C-terminal domain of the C-subunit upon secretion of the B+C sub-complex . Figure 2B ( arrow ) illustrates that the N-terminus of the C-subunit ( aa1–600 ) is necessary and sufficient for the secretion of the full length B-subunit . Detection of the FLAG-epitopes in these fusions shows that the C-subunit is also being exported ( Fig . 2C* ) . This also demonstrates that it is possible to export non-TC protein sequences fused downstream of aa600 of the C-subunit without hindering the export process . While this data confirms that the C-terminus of the C-subunit is not relevant to secretion , complementation bioassays ( see below ) using these constructs confirm that it is necessary for toxicity as expected ( Fig . S2 ) . Previous studies have demonstrated that it is possible to mix A-subunits and the B+C sub-complex post-translation and recover the full toxic potential of the whole TC [22] . In the absence of an anti A-subunit antibody , we were able to use this observation to devise an oral toxicity “complementation bioassay” which we could use to determine the location and activity of different toxin subunits . This assay relies upon mixing sonicated cell extracts and culture supernatants from the cosmid model and bespoke heterologous production strains and then testing toxicity levels by feeding to Manduca sexta neonate larvae . Using these assays we were able to reveal the impact that the three different subunits have upon the synthesis and secretion of one another . Figure 3 demonstrates that when the C-subunit gene is inactivated ( CKO ) , that we can no longer detect bioactive A-subunit protein in the supernatants using the complementation bioassays . This is reflected as a lack of the ability of the CKO supernatant to complement the activity of B+C sub-complex ( Fig . 3* ) . Note this effect is specific to a defect in the secretion of the A-subunit not its synthesis . This is shown by the ability of sonicated cell extracts from this same CKO strain to complement the B+C sub-complex ( Fig . 3** ) . These experiments also show that the C-subunit can facilitate the secretion of the A-subunit in the cosmid model even when the B-subunit gene has been inactivated ( BKO ) ( Fig . 3*** ) . As expected these data also confirm ( from several combinations shown ) that while all three subunits are required for full toxicity , that some limited toxicity can be seen from the A-subunit alone when released from the cell by sonication . Figure 4 illustrates that when we mix supernatants from a cosmid clone in which the A-subunit is knocked out ( AKO ) with A-subunit protein released by sonication from cells of the A-subunit production strain ( pA ) , that we again reconstitute a fully toxic TC ( Fig . 4* ) . Note the oral toxicity of the A-subunit protein alone when released from the pA production strain by sonication is significantly lower ( Fig . 4** ) . This confirms that B+C sub-complex secretion from the cosmid clone is not dependent upon the presence of a functional A-subunit . We heterologously expressed truncated and C-terminal FLAG-tagged versions of the C-subunit in E . coli ( Fig . 5A ) . Western blots using an anti-C-subunit N-terminal anti-peptide antibody and an anti-FLAG antibody to detect the C-terminal tags confirmed that the C-subunit could not be secreted in the absence of the B-subunit ( Fig . 5B ) . This is in contrast to the results presented in figure 2 showing that when the B and C-subunits are co-expressed , that they are both secreted . We also expressed intact or N-terminally truncated copies of the B-subunit gene ( removing aa1–361 ) as bi-cistrons with C-terminal FLAG tagged copies of the full length C-subunit gene ( Fig . 6A ) . We used Western blots with anti-TcdB1 C-terminus and FLAG tag antibodies to determine the effect of the N-terminal truncation of the B-subunit on the secretion of the B+C sub-complex . The secretion of individually expressed copies of the B- and C-subunit genes was also examined in this way . Figure 6 shows that removal of the first 361 amino acids of the B-subunit results in a failure to secrete either the B or C-subunits into the supernatant ( Fig . 6B ) . It should be noted however that removal of these amino acids from the B-subunit also significantly reduced the relative synthesis levels both of itself and of the downstream C-subunit seen in the soluble cell fractions ( cytoplasm and periplasm ) ( Fig . 6* ) . The amount of B+C sub-complex seen in the membrane fraction was also reduced ( Fig . 6 arrows ) . It is not clear whether removal of the N-terminus of the B-subunit causes a complete abolition of secretion or whether we are failing to detect the lower levels of the B+C sub-complex that is secreted by these strains . Nevertheless , the importance of this region to the production level of the B+C sub-complex is clear . Complementation bioassays using supernatants from these strains confirmed that they showed a loss of toxicity ( Fig . 7* ) . Despite the drop in B and C-subunit production levels , this “truncated” sub-complex was still functional , able to show complementation with the A-subunit when released from the cell by sonication ( Fig . 7** ) . This indicates that the N-terminal 361 aa of the B-subunit are not essential for toxicity , although they do influence the synthesis and export levels of the B+C sub-complex . We PCR cloned the tcdB1 and tccC5 genes as C-terminal FLAG-tag epitope fusions into the arabinose inducible expression plasmid pBAD30 in E . coli ( Fig . 8A ) . Induced cells were then separated into spheroplast and periplasmic fractions . Western blots using the anti-FLAG antibody were then used to identify the location of these two subunits ( Fig . 8B ) . While we were able to detect the C-subunit in the periplasmic fraction , the B-subunit was restricted to the cell spheroplasts . This confirms that the N-terminus of the C-subunit is able to direct its secretion across the inner membrane and that the B-subunit N-terminus is then required to facilitate secretion , of the whole B+C-sub-complex across the outer membrane . This also demonstrates that the B-subunit requires the presence of the C-subunit in order to cross the inner membrane . Over-production of C-terminally truncated copies of the C-subunit ( Fig . 5A ) in the original P . luminescens W14 strain led to a significant reduction in the oral toxicity of the culture supernatant to M . sexta neonates ( Fig . S3 ) . This showed that the non-toxic overexpressed N-terminal portion ( aa1–600 ) of the C-subunit protein is able to interfere with either synthesis or secretion . Finally we tested the toxicity of full length and truncated heterologously expressed C-subunit proteins by injection into Galleria larvae . For these experiments we constructed N- and C-terminally His-tagged expression constructs and a tagged C-terminally truncated version of the gene ( Fig . S4A ) . When we injected sonicated cell samples of these production strains into Galleria it demonstrated that the full length C-subunit could show some level of toxicity independent of the A- and B-subunits ( Fig . S4B ) . As expected , the C-terminally truncated constructs showed no toxicity again confirming the C-terminal tail as the toxin encoding domain .
Our previous studies in P . luminescens W14 showed that abundant tc mRNA is constitutively transcribed for all the three tcd subunits , during laboratory culture in rich medium [27] . Nevertheless , Western blot analysis and oral toxicity bioassays confirmed that they are not translated until stationary phase at a time concurrent with secretion into the surrounding milieu [27] . This suggests that translation of the TC is tightly linked to export . Previous publications and our own experimental evidence ( Fig . S5 ) show that the TC toxin B and C -subunits become tightly bound to one another in the bacterial cytoplasm , at least during heterologous co-expression in E . coli [3] , [12] . In addition , when over-expressed together in E . coli BL21 , TcdB1 and TccC1 can show limited toxicity to M . sexta even in the absence of an A-subunit [3] . Conversely , when synthesised in separate cells , released by sonication and mixed together post-synthesis no toxicity is observed [3] . This suggests that these two proteins may either associate during the translation process or require chaperone activity to assemble together . We note that homology detection and structure predictions using HHpred [29] show that that C-terminus of the B-subunit ( aa1150–1450 ) and N-terminus of the C-subunit ( aa20–380 ) both contain OspA-like domains ( C5: P = 99 . 96 E-value = 1 . 9e-26; B1: P = 99 . 90 E-value = 7 . 7e-20 ) . This structural domain in the OspA protein of Borrelia burgdorferi is responsible for homo-dimerization [30] . We suggest that these domains are responsible for the association of the B and C -subunits in the cytoplasm . Interestingly , although both these regions show high structural similarity to the OspA domain , there is no identity between them . It is tempting to speculate that this may represent a mechanism to prevent homo-dimerization of the subunits , and drive the association of the B and C -subunits in a 1∶1 ratio . Examination of DNA databases reveals that genetic fusions of B and C –subunit gene homologues are encoded by a range of other pathogens , including the RhsT toxin of Psuedomonas and the YP_335336 protein of Burkholderia pseudomallei 1710b . We note that these large fusions only possess a single predicted OspA like domains ( e . g . aa1400–1800 in YP_335336 ) . The work presented here demonstrates that the N-termini of both the B and C -subunits are required for secretion of all three subunits of Tcd . Furthermore , neither the B nor C -subunits can be secreted when expressed independently indicating that they both contribute to the process . Localisation experiments confirmed that the N-terminus of the C-subunit is required for secretion across the inner membrane and that the N-terminus of the B-subunit is responsible for secretion across the outer membrane . Interestingly we observed that the C-subunit alone could enable the secretion of the A-subunit . However , we saw no evidence of the secretion of the A-subunit by B in the absence of C . Our truncated and tagged B+C synthesis constructs also confirmed that the C-terminal domain of the C-subunit is not required for secretion . This is consistent with previous publications that this domain encodes the actual toxin [14] , which is variable among homologues [2] . Interestingly , in our studies ( data not shown ) and in previously published work [12] , there is evidence that at least in some cases , the C-terminal domain of C-subunit proteins may be cleaved and is not covalently attached to the rest of the TC complex . The C-subunit proteins belong to a large and enigmatic family of proteins called the Rhs-family and previous studies have also implicated Rhs proteins in export processes in E . coli and Pseudomonas [31] , [32] . The large RhsT protein of Pseudomonas aeruginosa , which has recently been recognised as a secreted mammalian virulence factor [33] , can also be seen to contain domains homologous to both the B and C-subunit proteins . Although the authors did not suggest a secretion method for this protein , it also has a toxic C-terminal domain , which in that case can be cleaved off and enter the host cell . The normal route of TC delivery by Photorhabdus is direct release into the insect blood . It is interesting that we were able to show limited toxicity of the TccC5 C-subunit alone when overexpressed , released by sonication and injected directly into Galleria larvae . This effect was abolished by removal of the toxin encoding tail as expected . This nevertheless suggests that the C-subunit can gain entry into host cells independently of the other components of the TC . We speculate that the export mechanism of the C-subunit might actually be acting as a host cell import mechanism in this case . The similarity of the B-subunit N-terminus to that of the SpvB protein of Salmonella suggests that this domain represents a secretion device , as it does in SpvB [18] . In SpvB , the N-terminal 1–229 amino acids are sufficient to promote secretion into extracellular milieu of the whole protein [18] . We note however that unlike SpvB , the B-subunit proteins do not possess a general sec-pathway type II signal leader ( Fig . 9A ) . Our evidence suggests that the SpvB-like 361 N-terminal amino acids of the B-subunit is responsible for crossing the outer membrane only . Interestingly , when transiently expressed in mammalian cells , the N-terminus of TcdB1 was seen to accumulate in the nucleus [3] . We previously proposed that this region might represent a nuclear-targeting domain involved in toxicity to the host cell . However an alternative explanation is that this nuclear accumulation reflects a “one-way” membrane crossing activity of this N-terminal domain , which is normally used in secretion of Tcd across the Gram-negative outer membrane . Additional roles in crossing host cell membranes are not excluded by a role in secretion . In previous work we used anti-TcdB1N-terminus and anti-TcdB1C-terminus antibodies to examine the translation and export of the B-subunit in P . luminescens W14 and the E . coli cosmid model [27] . Consistent with the manifestation of oral toxicity , TcdB1 can be detected in culture supernatants after 1 day , but not earlier when using the anti-TcdB1C-terminus antibody . Conversely we observed only minor amounts of the TcdB1 in the supernatant when we used the anti-TcdBN-terminus antibody [27] . This suggests that the N-terminus may be cleaved off during the secretion process as the C and N terminal antibodies show similar detection efficacy against intracellular TcdB1 protein . In addition when using the anti-TcdB1C-terminus antibody , TcdB1 could be seen in the supernatants and the membrane fractions but not in the cytoplasm at three days [27] . It was possible to see only very small amounts of TcdB1 using the TcdB1N-terminus antibody in the membrane fraction only however . We hypothesized that this reflects a tight regulation between translation and export , and that the TcdB1 N-terminus is involved in this process . In support of this hypothesis , removal of the N-terminus of the B-subunit strongly reduced the production level of not only the B but also of the C-subunit . Taken together these observations suggest that under native expression level conditions , translation of the N-terminal domain of the B-subunit is rate limiting to synthesis and secretion and that this domain may also be removed during secretion . A previous report suggested that a B-subunit from Yersina ( TcaC ) could be secreted via a type III secretion system [34] . We argue this is not the normal route of export for Photorhabdus homologues , as our fully functional E . coli secretion model contains no type III system genes . Nevertheless a comparison of the amino acid sequences of TcaC from Yersinia pestis and TcdB1 from Photorhabdus show good identity along the length of the protein . As Photorhabdus encodes a type III secretion system it remains possible that B-subunits could also be secreted via this system . Our evidence suggests that the N-terminus of the C-subunit is responsible for export of the complex across the inner membrane . In this model , the C-subunit fulfils the analogous role of the type II secretion leader peptide of SpvB . We can draw further analogy between SpvB and the B+C sub-complex . The SpvB protein has a Type III secretion-independent N-terminal domain , fused by a proline stretch to a “toxic” ADP-ribosyltransferase C-terminal function . In the case of the B+C sub-complex , the N-terminal domains of the B and C-subunits provide the export system . The OspA-like domains of the B and C–subunits likely form the protein-protein interaction domains of the sub-complex while the variable C-termini of the C-subunit constitutes the active “toxin” domain . In the case of TccC5 this also constitutes an ADP-ribosyltransferase domain involved in forcing actin clustering [14] . Figure 9B shows a diagrammatic summary of the roles that the different subunit domains play in the export process . It was surprising that the C-subunit is able to export the A-subunit in the absence of the B-subunit . It is possible that this relies on the injection-like mechanism of TcdA1 as recently proposed by Gatsogiannis et al [13] . The A-subunit does not possess an OspA-like structural domain although previous studies have identified protein-protein interaction domains in the C-terminus [3] [13] . While there is no evidence to support this , we might speculate that the SpvA-like N-terminus of TcdA1 is able to substitute for the SpvB-like N-terminus of TcdB in this case . It was interesting that the over-production of truncated TccC5 proteins lacking the toxin C-terminal domain in P . luminescens W14 was able to supress oral toxicity of the supernatant . Different hypotheses may be proposed to explain this . Firstly , the over-expressed truncated subunits might be interfering with the normal export pathways , essentially blocking the export of the native orally toxic Tca and Tcd complexes . Alternatively the truncated subunits may be replacing the native full-length C-subunit proteins , producing predominantly non-toxic TC derivatives . If the latter were the case then it would suggest that the C-subunits are relatively promiscuous regarding which Toxin Complexes they interact with , in this case both Tca and Tcd . The initial work on Tca and Tcd in strain W14 suggested that they form distinct complexes , with Tca comprising TcaA , TcaB and TcaC and Tcd comprising TcdA1 and TcdB1 . Interestingly in these earlier studies the C-subunit proteins were not detected [5] , [35] , [36] . In conclusion we have investigated the role of various domains of TC subunits in secretion . Central to secretion are the N-terminal domains of the two subunits of the B+C sub-complex . These domains serve to export the TC in the absence of any other specialised Photorhabdus proteins . This suggests either that they encode a “self-contained” export system or that they interact with an as yet undefined export system also present in laboratory strains of E . coli . Furthermore , the presence of B and C-subunit gene homologues in a range of diverse bacterial species indicates that this toxin secretion mechanism is not restricted to the insect pathogens . | The Toxin Complex ( TC ) is a large multimeric protein complex first identified in the insect pathogens Photorhabdus and Xenorhabdus . TC isolates from these pathogens exhibit oral toxicity to a diverse range of insects . As such there is significant interest in developing them as candidates for crop protection strategies . Currently all insect resistant transgenic crops rely upon the production of Bacillus thuringiensis Cry toxins . However , to minimise the risk of insect resistance development it is imperative to develop additional toxin systems employing alternative modes of action . A barrier to the further development of TCs as agrochemical tools has been the complexity of their synthesis , secretion and assembly . Little is known about how the large TC subunits are secreted across the bacterial cell wall . We present here an investigation into the roles that the different domains of the B and C-subunit proteins play in secretion of the whole TC . The significance of this goes beyond these specific insect toxins as homologues of these two subunits are encoded in the genomes of a range of human pathogens , such as Burkholderia and Yersinia , in which they have been implicated in human virulence . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| []
| 2013 | The Role of TcdB and TccC Subunits in Secretion of the Photorhabdus Tcd Toxin Complex |
Post-mortem brains from Down syndrome ( DS ) and Alzheimer's disease ( AD ) patients show an upregulation of the Down syndrome critical region 1 protein ( DSCR1 ) , but its contribution to AD is not known . To gain insights into the role of DSCR1 in AD , we explored the functional interaction between DSCR1 and the amyloid precursor protein ( APP ) , which is known to cause AD when duplicated or upregulated in DS . We find that the Drosophila homolog of DSCR1 , Nebula , delays neurodegeneration and ameliorates axonal transport defects caused by APP overexpression . Live-imaging reveals that Nebula facilitates the transport of synaptic proteins and mitochondria affected by APP upregulation . Furthermore , we show that Nebula upregulation protects against axonal transport defects by restoring calcineurin and GSK-3β signaling altered by APP overexpression , thereby preserving cargo-motor interactions . As impaired transport of essential organelles caused by APP perturbation is thought to be an underlying cause of synaptic failure and neurodegeneration in AD , our findings imply that correcting calcineurin and GSK-3β signaling can prevent APP-induced pathologies . Our data further suggest that upregulation of Nebula/DSCR1 is neuroprotective in the presence of APP upregulation and provides evidence for calcineurin inhibition as a novel target for therapeutic intervention in preventing axonal transport impairments associated with AD .
Virtually all Down syndrome ( DS ) adults develop progressive neurodegeneration as seen in Alzheimer's disease ( AD ) , and overexpression of the amyloid precursor protein ( APP ) , a gene located on chromosome 21 , is thought to contribute to AD in DS [1]–[3] . Consistently , duplication of a normal copy of APP is sufficient to cause familial AD [4] , [5] , confirming that it is a key gene in AD neuropathologies seen in DS . This well-known connection between AD and DS provides a unique opportunity to identify the genetic and molecular pathways contributing to AD . In addition to APP , another gene likely to play a crucial role in both AD and DS is the Down syndrome critical region 1 gene ( DSCR1 , also known as RCAN1 ) . Intriguingly , post-mortem brains from AD patients show increased DSCR1 both at mRNA and protein levels [6]–[8] . Studies have also shown that oxidative stress and Aβ42 exposure can induce DSCR1 expression [8] , [9] . DSCR1 is located on human chromosome 21 and encodes a highly conserved calcineurin inhibitor family called calcipressin [10]–[15] . DSCR1 has been implicated paradoxically in both promoting cell survival in response to oxidative stress and in inducing apoptosis [8] , [9] , [16] , [17] . The role of DSCR1 in AD thus remains unclear and an important question is whether DSCR1 contributes to AD or plays a role in combating the toxic effects of APP overexpression . To elucidate the role of DSCR1 in modulating APP-induced phenotypes , we used Drosophila as a model system , which has been used successfully to investigate various human neurodegenerative diseases including AD , Parkinson's , and polyglutamine-repeat diseases [18]–[27] . Overexpression of APP in both fly and mouse models have previously been shown to cause age-dependent neurodegeneration and axonal transport defects [28]–[31] . Furthermore , impaired transport of essential organelles and synaptic vesicles caused by APP perturbation is thought to be an underlying cause of synaptic failure and neurodegeneration in AD [32]–[34] . However , mechanisms for how APP induces transport defects remain unclear . Here , we show that Nebula , the fly homolog of DSCR1 , delays neurodegeneration and reduces axonal transport defects caused by APP overexpression . We report that Nebula enhances anterograde and retrograde axonal trafficking as well as the delivery of synaptic proteins to the synaptic terminal . We find that APP upregulation elevates calcineurin activity and GSK-3β signaling , but Nebula co-upregulation corrects altered signaling to restore axonal transport . Together , our results indicate that Nebula/DSCR1 upregulation is neuroprotective in the presence of APP overexpression and further suggest that Nebula/DSCR1 upregulation may delay AD progression . In addition , our results for the first time link defective calcineurin signaling to altered axonal transport and imply that restoring calcineurin and GSK-3β signaling may be a feasible strategy for treating AD phenotypes caused by APP upregulation .
To examine the role of DSCR1 in modulating APP-induced neurodegeneration and axonal transport defects , we generated transgenic flies containing UAS-APP ( APP ) in the presence or absence of UAS-nebula ( nlat1 ) [15] . Targeted expression of human APP in the fly eyes using the Gmr-GAL4 driver caused age-dependent degeneration of the photoreceptor neurons , consistent with a previous report by Greeve et al [35] . As seen in Fig . 1A , staining with an antibody specific for the photoreceptor neurons ( 24B10 ) and antibody against the APP protein ( 6E10 ) revealed the presence of vacuoles in the retina ( arrow ) . Surprisingly , overexpression of nebula together with APP ( APP;nlat1 ) reduced neurodegeneration ( as determined by calculating the fold change in the percentage of area lost ) , suggesting that Nebula upregulation is neuroprotective ( Figs . 1A and 1B ) . By 45-days of age , flies expressing both nebula and APP started to show increased vacuole formation , but the extent of degeneration was significantly reduced compared to that of APP overexpression , further implying that Nebula delays the onset of neurodegeneration rather than completely preventing it . To confirm that Nebula indeed protects against neurodegeneration caused by APP upregulation , we expressed APP in nla1 , a previously characterized nla hypomorphic mutant [15] . Note that because nebula null alleles are lethal [15] , nebula hypomorphs were examined . Fig . 1 shows that decreasing Nebula level enhanced APP-induced neurodegeneration in the retina ( APP;nla1 ) , thus highlighting the importance of endogenous Nebula protein in conferring neuroprotection . We did not detect significant neurodegeneration in nla1 mutant and nla overexpression flies even by 45 days of age ( data not shown ) , indicating that APP is necessary for the observed phenotype . In addition , mitigation of photoreceptor degeneration by Nebula upregulation is not due to altered expression level of APP , since UAS-LacZ transgene was included to balance out the number of transgenes ( we found Gmr-GAL4 is particularly sensitive to number of transgenes ) . The level of APP protein in each fly line is also further confirmed by staining with the 6E10 antibody ( Fig . 1A ) and Western blot analyses ( Fig . S1 ) . Comparable level of APP was detected in all transgenic lines , suggesting that rescue by nebula overexpression is not due to altered APP level . We next determined if Nebula rescues functional defects in photoreceptor by measuring the ability of flies to see light . Flies are normally phototactic and will move toward light when placed in test tubes with light source on the opposite end [36] . We find that the severity of the vacuole phenotype was paralleled by impairments in phototactic behavior ( Fig . 1C ) . Flies overexpressing APP showed age-dependent decline in phototaxis that is delayed by APP and nebula expression ( Fig . 1C ) . Taken together , these results imply that nebula overexpression protects neurons structurally as well as functionally against the toxic effects of APP overexpression . We also noticed that APP overexpression caused formation of APP aggregates in the photoreceptor axons as detected by 6E10 antibody ( Fig . 1D; yellow arrow head ) . Previous studies have shown that APP phosphorylated on threonine 668 ( pT668-APP ) is preferentially transported in axons [37] , we thus further monitored the distribution of pT668-APP . We found that overexpression of APP led to pT668-APP accumulations in the photoreceptor axons , whereas APP and nebula co-overexpression significantly enhanced the delivery of pT668-APP to synaptic terminals in the medulla ( Fig . 1D ) . These results suggest that APP overexpression may lead to blocked transport that is alleviated by Nebula . Axonal transport abnormalities are thought to precede the onset of AD [30] , and APP overexpression has been shown to cause synaptic vesicle accumulations indicative of blocked axonal transport [28] , [29] . We thus further investigated the role of Nebula in modulating APP-induced vesicle aggregation in larval motor axons , which is an excellent system for monitoring vesicle transport because of the long axons and stereotypical innervation of the neuromuscular junction ( NMJ ) . As seen in Fig . 2A , APP overexpression in neurons using the Elav-GAL4 driver caused synaptic vesicle accumulation as detected by synaptotagmin staining in the motor axons , suggesting abnormal vesicle transport . Staining using the 4G8 antibody to detect APP revealed that APP aggregates frequently colocalized with synaptotagmin aggregates , implying that synaptotagmin and APP are either comparably inhibited by physical blockade within the nerve or that they are transported together as suggested by recent reports [38] , [39] . Co-upregulation of Nebula and APP significantly prevented APP-induced synaptotagmin and APP accumulations . Decreasing Nebula by crossing it into nla1 background increased the number of synaptotagmin and APP aggregates slightly , although not significantly ( Fig . 2B ) . As nla1 only reduces Nebula level by about 30% and that nla null alleles are lethal [15] , we used RNAi strategy to further decrease Nebula level ( Fig . S2 ) . Figs . 2A–2B show that greater reduction in Nebula level using the UAS-nla-RNAi transgene ( RNAi-nla ) further exacerbated the APP-induced aggregation phenotype . To ensure that the observed rescue in phenotype is not due to altered APP overexpression , we monitored the level of neuronal APP protein , as well as Nebula , in different fly lines . As seen in Fig . S3 , APP level was unaltered in flies containing different number of transgenes , and Nebula manipulations in APP overexpression background showed the expected changes . Similar results were obtained when performing western blot analyses using brains dissected from 3rd instar larvae ( Fig . S4 ) . Together , these results confirm that rescue of APP phenotype by Nebula is not due to altered APP expression . In addition , we examined the effect of altering Nebula levels alone on vesicle accumulation . Manipulations of Nebula levels alone did not cause synaptotagmin aggregate accumulation in nerves , suggesting the observed phenotype is APP-dependent ( Figs . S5A and S5B ) . To verify that the synaptotagmin aggregate accumulation phenotype is not due to a non-specific effect of expressing human APP , we also monitored the effect of Nebula on modulating endogenous fly Appl gene function . Fig . 2C shows that upregulation of APPL in neurons also caused synaptotagmin accumulation in axons . Nebula co-upregulation significantly reduced the number of synaptotagmin aggregates , whereas Nebula reduction using RNAi significantly exacerbated the phenotype ( Figs . 2C and 2D ) . Together , our results support earlier finding that mammalian APP and Drosophila APPL are functionally conserved [40] , and further indicate that APP and APPL-induced axonal transport defects are regulated by Nebula in a similar fashion . To determine to what degree aggregate accumulation corresponded to altered delivery of synaptic proteins to the synaptic terminal , we evaluated the levels of both synaptotagmin and APP in the NMJ . As demonstrated in Figs . 3A–3B , APP upregulation significantly reduced the level of average synaptotagmin intensity in the synapse while nebula co-overexpression enhanced the delivery of both synaptotagmin and APP to the synaptic terminal . This change is not due to altered overall synaptotagmin or APP levels ( Figs . S4 and S5C ) . Note that the 4G8 antibody does not detect endogenous fly APPL; therefore , we normalized the level of APP delivered to the synapse to flies overexpressing APP and nebula . We found Nebula reduction did not further reduce the amount of synaptotagmin reaching the terminal ( Fig . 3B ) , albeit it did increase the number of APP-induced aggregates in the axon ( Fig . 2B ) . This result indicates that either retrograde transport of synaptotagmin is altered , or the increase in aggregate number has not yet reached a critical threshold for further impairment . In addition , although no detectable synaptotagmin aggregate was seen in flies with Nebula reduction alone , a decrease in synaptotagmin staining was detected in the synapse ( Figs . S5B and S5D ) . This result suggests that Nebula itself may be required for reliable axonal transport . We also examined the effect of abnormal aggregate accumulations and reduced delivery of synaptic proteins on locomotor behavior . Overexpression of APP dramatically impaired larval movement ( Fig . 3C and Movie S1 ) . Nebula co-overexpression significantly rescued this locomotor defect , in further support of the hypothesis that Nebula upregulation exerts beneficial effects on synaptic functions by alleviating abnormal aggregate accumulations . Note that further reduction of Nebula in APP overexpression background did not significantly worsen the locomotor defect of APP overexpressing larvae , perhaps due to a threshold effect . Reducing Nebula alone was sufficient to induce a mild defect in locomotor activity ( Fig . S5E ) , suggesting delivery of synaptic proteins to the synaptic terminals is crucial for normal synaptic function . Similar to APP overexpression , upregulation of APPL decreased the delivery of synaptotagmin to the synapse . APPL and Nebula co-upregulation showed a higher level of synaptotagmin in the NMJ , confirming Nebula interacts genetically with APPL to rescue impaired in transport ( Fig . S6A and S6B ) . We also found that similar to RNAi-nla larvae , Appl null mutant ( Appld ) displayed a slight decrease in the level of synaptotagmin at the synapse independent of aggregate accumulation ( Figs . S6B and S6C ) . Reducing Nebula in neurons of Appld larvae with the RNAi-nla transgene driven by the pan-neuronal nSyb-GAL4 driver ( Appld; RANi-nla/nSyb-GAL4 ) did not further enhance the phenotype , suggesting that the two proteins act in the same pathway to modulate axonal transport . While monitoring synaptotagmin levels at the NMJ , we also noticed that APP overexpression triggered changes in synaptic morphology as previously reported [41] , [42] . Fig . 4 shows presynaptic terminals stained with HRP to outline the presynaptic terminals , which revealed an increase in the total number of boutons and satellite boutons brought upon by APP overexpression . Nebula co-upregulation also rescued APP-induced synapse proliferation phenotype , but not the number of satellite boutons ( Fig . 4B and 4D ) . Manipulating levels of Nebula alone without APP did not influence bouton number or morphology , suggesting that the satellite bouton phenotype is dependent on the presence APP in the synapse . Since reducing Nebula levels alone decreased the delivery of synaptotagmin to the synaptic terminal without altering synaptic morphology , axonal transport problems are not secondary consequences of altered synaptic morphology . A plausible mechanism by which Nebula suppresses the APP-induced over-proliferation phenotype is that Nebula co-upregulation restores the delivery of proteins required for normal synaptic growth such as Fasciclin II ( FasII ) , a cell adhesion molecule shown to influence synaptic morphology [43] , [44] . Previous reports suggest that changes in FasII levels differentially affect synaptic growth [42]–[44] , and that increasing FasII levels presynaptically can significantly suppress the increase in bouton number observed in APPL overexpression synapses [42] . We therefore quantified FasII levels in the NMJ ( Fig . S7 ) . We found that overexpression of APP reduced the level of FasII in the NMJ , whereas APP and nebula co-overexpression restored it ( Fig . S7 ) . While APP upregulation may play other roles in synapse formation , these results together with previous reports imply that depletion of FasII in the presynaptic terminal could partially contribute to the hyper-growth phenotype . Furthermore , our data reveal that Nebula upregulation is effective in protecting against multiple phenotypes caused by APP overexpression , including age dependent photoreceptor neurodegeneration , vesicle accumulations in axons , and changes in synaptic morphology . To directly evaluate the effect of Nebula on APP transport and to determine whether the observed axonal aggregates correspond to defective axonal transport , we performed live-imaging of human APP tagged with yellow fluorescent protein ( APP-YFP ) . APP-YFP vesicles in larval motor axons displayed movement in both the anterograde and retrograde directions over the 2-minute imaging period as represented by kymographs depicting distance traveled and time in the x- and y-directions , respectively ( Fig . 5A ) . Nebula co-overexpression had a mild , but significant , effect on APP-YFP movement . Nebula co-upregulation increased the percentage of anterograde moving vesicles and resulted in reduced number of stationary APP-YFP; knockdown of Nebula using RNAi increased the number of stationary APP-YFP ( Figs . 5A and 5B ) . Quantification of the average speed of APP-YFP movement revealed that overexpression of nebula also increased the speed of APP-YFP movement in both the anterograde and retrograde directions ( Fig . 5C ) . Together , these results suggest that Nebula upregulation enhances the transport of APP , consistent with the decreased aggregate accumulations of APP in axons and increased APP staining in the NMJ when Nebula is co-expressed ( Figs . 2A and 3A ) . To further confirm that Nebula facilitates synaptic vesicle movement in the presence of APP and to better assess the role of endogenous Nebula in regulating transport , we also monitored synaptotagmin movement in the motor axons of larvae expressing GFP-tagged synaptotagmin ( GFP-SYT ) . We find the movement of GFP-SYT to be highly dynamic with anterograde , retrograde , and bi-directional movement ( Fig . 6A ) . Overexpression of APP dramatically reduced the percentage of vesicles moving in both the anterograde and retrograde directions while nebula co-overexpression significantly facilitated synaptotagmin transport in both directions ( Figs . 6A and 6B ) , albeit retrograde transport was more effectively restored by Nebula . Reducing Nebula using RNAi further diminished APP-induced synaptotagmin transport in both directions , confirming interaction between Nebula and APP . Reduction in the overall movement was also accompanied by a decrease in anterograde and retrograde velocity ( Fig . 6C ) . Together , these results suggest that APP overexpression slows down the overall movement of vesicles , which may lead to accumulation of transported proteins . Nebula co-overexpression with APP partially restores the defect by increasing the movement and speed of transport in both the anterograde and retrograde directions . To understand the role of endogenous Nebula in axonal transport , we examined the effect of Nebula manipulations on GFP-SYT movement in the absence of APP overexpression . We find that Nebula upregulation alone did not significantly influence transport; decreasing Nebula through RNAi was sufficient to reduce the number of moving synaptotagmin vesicle in both directions , as well as the speed of anterograde transport ( Fig . 6 ) . This result is consistent with the decrease in synaptotagmin staining in the NMJ seen in static images , and further confirms that Nebula is required for efficient transport of synaptic proteins . To further determine if general axonal transport is affected by APP and Nebula upregulation , we also monitored mitochondrial transport . Proper distribution of mitochondria is vital for normal cell functions and defects in mitochondrial transport can adversely affect cell survival [45]–[47] . Time-lapse live imaging was performed in larvae with GFP targeted to mitochondria ( mito-GFP ) for the indicated genotypes ( Fig . 7 ) . APP upregulation severely impaired the movement of mitochondria in both the anterograde and retrograde directions both in terms of percent in motion and the speed of movement ( Figs . 7B and 7C ) . Nevertheless , the APP-induced mitochondrial transport defect was partially restored by Nebula co-upregulation ( Fig . 7 and Movie S2 ) , similar to what was observed for synaptic vesicle transport . Manipulations in the level of Nebula did not significantly alter the overall mitochondrial movement , except that nebula overexpression alone seemed to enhance both the proportion and the speed of mitochondria transported in the retrograde direction . This result is consistent with our observation that Nebula co-upregulation was more effective in restoring retrograde GFP-SYT transport . Together , our results suggest that Nebula influences general axonal transport that extends beyond synaptic proteins . Mitochondria are dynamic organelles whose distribution is tightly regulated to meet the energy demands within the polarized neuron [45] , [48] . We find that despite the decrease in mitochondrial movement in flies overexpressing APP , the distribution and density of mitochondria within the proximal axon where imaging was performed did not vary across genotypes ( Fig . S8A ) . These results imply that impaired synaptic vesicle transport is not likely caused by local depletion of mitochondria within the axon . Furthermore , mitochondria did not accumulate near the site of synaptotagmin aggregate formation in the axons ( Fig . S8B ) , suggesting that mitochondria are either able to move past the stalled synaptic vesicle accumulations or that mitochondria travel on other non-blocked microtubule tracks . Despite increasing evidence linking defective trafficking of presynaptic proteins , mitochondria , and signaling molecules to neuropathologies of AD , mechanisms for how APP overexpression affects axonal transport remain unclear . We first tested the possibility that APP upregulation impairs axonal transport by influencing overall microtubule integrity . To this end , we stained the axonal nerves and NMJs with antibodies against acetylated tubulin , β-tubulin , and Futsch ( Fig . 8 ) . Acetylated tubulin is a marker for stable microtubules [49]; Futsch is a microtubule binding protein homolog to human MAP1B and is involved in maintaining microtubule integrity at presynaptic terminals during NMJ growth [50] . Our data revealed that APP overexpression did not cause fragmentation of microtubules as revealed by both acetylated tubulin and β-tubulin staining in the axons ( Fig . 8A ) , and filamentous acetylated tubulin staining in the synaptic terminals across all genotypes ( Fig . 8B ) . Note that in Fig . 8B , we also highlighted the presynaptic boutons by HRP staining ( red ) , since acetylated tubulin in the muscles are also detected in the background . Western blot analyses of dissected larval brains further confirmed that the overall level of acetylated tubulin is not altered by APP overexpression ( Fig . 8C ) . Closer examination of Futsch staining also did not reveal differences in overall microtubule integrity ( Fig . 8D ) . Together , these results suggest that APP overexpression does not cause axonal transport problems by influencing microtubule stability , which is consistent with a recent report that showed normal microtubule stability and acetylated tubulin level in larvae overexpressing APP-YFP [51] . Nebula encodes an inhibitor of calcineurin that is highly conserved across species [15] , we therefore tested the hypothesis that calcineurin inhibition is an underlying mechanism for Nebula-mediated rescue of APP phenotypes . To this end , we genetically altered calcineurin activity in neurons using the UAS/GAL4 strategy . To elevate calcineurin activity , we expressed a constitutively active calcineurin ( CaNAct ) with its auto-inhibitory domain deleted ( Figs . S9A and S9B ) . To reduce calcineurin activity , RNAi strategy against the calcineurin B gene ( RNAi-CaNB ) , an obligatory subunit necessary for calcineurin activity , was used . We find that similar to Nebula upregulation , decreasing calcineurin using RNAi-CaNB in the presence of APP significantly reduced synaptotagmin aggregate accumulations and synaptic depletion , as well as restored larval locomotor behavior ( Figs . S9C–E ) . Overexpression of CaNAct together with APP further exacerbated the APP-induced phenotypes ( Figs . 9A and 9B ) , whereas co-overexpression of CaNAct and nebula diminished the ability of Nebula to protect against APP-induced transport defects . Similar to larvae with reduced levels of Nebula ( RNAi-nla ) , larvae expressing CaNAct did not show aggregate accumulations in axons but displayed a reduced level of synaptotagmin staining in the synapse ( Figs . S9D ) , indicating active calcineurin overexpression alone only has modest effect on axonal transport . As shown above , synaptotagmin aggregate accumulation in nerves and depletion in the synaptic terminals are reliable indicators of significant transport deficiencies; our results thus indicate that Nebula protects against APP-induced defects through inhibition of calcineurin . Furthermore , our data present for the first time that APP upregulation influences axonal transport through activation of calcineurin . This conclusion is further supported by direct measurement of calcineurin activity , in which we find that APP upregulation significantly elevated calcineurin activity but is further restored close to normal in flies overexpressing APP and nebula , or APP and RNAi-CaNB ( Fig . 9C ) . Overexpression of APP , CaNAct , and nebula together showed an intermediate phenotype in both calcineurin activity and aggregate accumulations , suggesting that the severity of aggregate accumulation correlated with the level of calcineurin when APP is upregulated . How does APP upregulation trigger calcineurin activation ? Because calcineurin phosphatase activity is dependent on intracellular calcium concentration [52] , we examined the possibility that APP overexpression elevates calcium levels . Using a genetically encoded fluorescent calcium sensor ( Case12 ) previously shown to detect calcium with high sensitivity [53] , [54] , we compared Case12 signal across different genotypes . Fig . S10 shows that larval brain expressing Case12 displayed a significant increase in signal following application of calcimycin , a calcium ionophore , confirming that the Case12 construct can indeed detect increases in calcium . Overexpression of APP alone or overexpression of APP and nebula also caused a significant elevation in Case12 signal in the larval brain and the ventral ganglion ( where the motor neuron cell bodies are located ) as compared to the control ( Figs . 9D and 9E ) . These data imply that an APP-mediated increase in calcium is triggering the increase in calcineurin activity . Furthermore , observations that co-overexpression of APP and nebula increased calcium while simultaneously restoring calcineurin activity indicate that Nebula is influencing axonal transport through calcineurin inhibition rather than acting at a step modulating calcium influx . Mechanisms by which calcineurin regulates axonal transport are not well understood , but one potential pathway is through regulation of GSK-3β activity . Aberrant activation of GSK-3β has been associated with AD and calcineurin has been shown to activate GSK-3β through dephosphorylation of Ser9 of GSK-3β in vitro [55]–[58] . It was suggested that GSK-3β may negatively influence axonal transport by altering microtubule stability through hyperphosphorylation of tau , by inhibiting kinesin motor binding to the cargo through phosphorylation of the kinesin light chain ( KLC ) , or by altering the kinesin motor activity [51] , [59]–[61] . These previous findings led us to investigate the possibility that Nebula restores APP-dependent transport problems through calcineurin-mediated regulation of GSK-3β in vivo . The activity of GSK-3β is regulated by phosphorylation and dephosphorylation: dephosphorylation of Ser9 by a number of phosphatases including calcineurin is required to activate GSK-3β [56] , [62] , and phosphorylation at Tyr216 site is necessary to enhance GSK-3β activity [63] , [64] . Interestingly , phosphorylation of GSK-3β at Ser9 can both inhibit GSK-3β activity and override the increase in activity even when phosphorylated at Tyr216 [65] . Because these phosphorylation sites are conserved between fly and human , we took advantage of phospho-specific antibodies to monitor GSK-3β activity . Western blot analyses using an antibody specific for phosphorylated Ser9 ( pSer9 ) of GSK-3β revealed that APP upregulation indeed reduced the level of pSer9-GSK-3β while APP and Nebula co-upregulation partially restored the level to normal ( Fig . 10A ) . This suggests APP upregulation leads to GSK-3β activation that is inhibited by Nebula upregulation . To verify that GSK-3β activation is due to calcineurin activation , we reduced calcineurin activity in APP overexpressing flies using RNAi-CaNB ( Fig . 10A ) . We find that APP and RNAi-CaNB co-overexpression in neurons , which was sufficient to restore calcineurin activity , completely prevented GSK-3β dephosphorylation at Ser9 site . This result indicates that APP-induced GSK-3β dephosphorylation at Ser9 is dependent on calcineurin activation in vivo . Note that we did not detect enhanced GSK-3β dephosphorylation when APP is expressed together with constitutively active calcineurin ( CaNAct ) , suggesting that calcineurin may in part directly influence transport through GSK-3β-independent pathways . Our data strongly implicate activation of calcineurin and subsequent GSK-3β induction to be a mechanism underlying APP-induced aggregate phenotype . Because activation of calcineurin alone did not result in synaptotagmin aggregate accumulation , we further hypothesized that APP upregulation also enhances GSK-3β activity through phosphorylation at Tyr216 . Western blot analyses show that the level of phosphorylated GSK-3β at Tyr214 ( conserved Tyr216 site in Drosophila ) is indeed elevated in flies overexpressing APP or APP and nebula ( Fig . 10B ) . Overexpression of CaNAct alone , however , failed to induce phosphorylation at Tyr214 , suggesting that phosphorylation of Tyr214 is not affected by calcineurin and dependent on the presence of APP . Together , our data demonstrate that in addition to activating GSK-3β by relieving inhibition through calcineurin , APP upregulation further enhances GSK-3β activity through phosphorylation at Y214 in fly . Active GSK-3β had been shown to phosphorylate KLC , leading to detachment of the cargo from the motor [59] , [66] . Since synaptotagmin transport was severely inhibited by APP overexpression , and that synaptotagmin transport can depend on kinesin 3 [67] , [68] and kinesin 1 ( both KLC and kinesin 1 heavy chain ) [69]–[73] , we tested the possibility that APP overexpression perturbs KLC and synaptotagmin interaction via immunoprecipitation . APP overexpression indeed reduced synaptotagmin ( cargo ) and KLC interaction while overexpression of APP and nebula preserved this interaction ( Fig . 10C ) . These results suggest that Nebula is likely to restore APP-induced axonal transport defects by correcting GSK-3β signaling and stabilizing cargo-motor interaction . Having demonstrated that APP activates calcineurin signaling to regulate GSK-3β phosphorylation , we next examined if reducing GSK-3β can restore axonal transport . In the presence of APP upregulation , decreasing Shaggy ( Sgg; fly homolog of GSK-3β ) in flies with APP overexpression ( sgg1;APP ) resulted in significant suppression of the APP aggregate phenotype ( Figs . 10D and 10E ) . This result is consistent with a recent report demonstrating mild enhancement of APP-YFP movement when GSK-3β is reduced [51] . Surprisingly , normal calcineurin activity was detected in these flies ( 1 . 00±0 . 16 fold of control for Sgg1;APP vs . 1 . 75±0 . 25 fold of control for APP ) . This result suggests the existence of feedback regulation of calcineurin activity and further implies that either a change in calcineurin activity or GSK-3β signaling could be responsible for the observed rescue . We therefore generated flies expressing APP and constitutively active calcineurin in sgg1 background ( sgg1;APP/CaNAct ) . Note that we used the hypomorphic allele sgg1 because sgg null animals are lethal [74] . Consistent with GSK-3β being downstream of calcineurin , reducing Sgg diminished the effect of CaNAct in enhancing APP phenotype ( Figs . 9B and 10E ) . We also expressed the constitutively active Sgg ( sggS9A ) together with APP , which surprisingly showed the same phenotype as APP overexpression . Calcineurin activity assay showed an unexpected decrease in calcineurin activity ( 0 . 74±0 . 06 fold of control ) in these flies , suggesting that constitutive GSK-3β activation in the absence of calcineurin activation is sufficient to disrupt axonal transport potentially through phosphorylation of KLC . Interestingly , we find that overexpression of the constitutively active Sgg in neurons alone was sufficient to induce aggregate accumulation similar to flies with APP overexpression ( Figs . 10D and 10E ) . Calcineurin activity assay revealed that these flies showed an increase in overall calcineurin activity ( 1 . 65±0 . 30 fold of control ) . This increase in calcineurin activity by active Sgg may be due to GSK dependent phosphorylation of Nebula , which has been shown to cause activation of calcineurin [75] . Since over-activation of calcineurin and GSK-3β pathway in the absence of APP upregulation fully replicated the aggregate accumulation phenotype , it suggests that abnormal activation of both the GSK-3β and calcineurin pathways are necessary for the severe axonal transport defect and aggregate accumulation phenotypes .
Although upregulation of APP had been shown to negatively influence axonal transport in mouse and fly models 28–31 , mechanisms by which APP upregulation induces transport defects are poorly understood . Several hypotheses have been proposed , including titration of motor/adaptor by APP , impairments in mitochondrial bioenergetics , altered microtubule tracks , or aberrant activation of signaling pathways [76] . The motor/adaptor titration theory suggests that excessive APP-cargos titrates the available motors away from other organelles , thus resulting in defective transport of pre-synaptic vesicles [29] . Our finding that Nebula co-upregulation enhanced the movement and delivery of both synaptotagmin and APP to the synaptic terminal argues against this hypothesis . In addition , earlier finding suggest that Nebula upregulation alone impaired mitochondrial function and elevated ROS level [77] , thus implying that Nebula is not likely to rescue APP-dependent phenotypes by selectively restoring mitochondrial bioenergetics . Furthermore , consistent with a recent report showing normal microtubule integrity in flies overexpressing either APP-YFP or activated GSK-3β [51] , our data revealed normal gross microtubule structure in flies with APP overexpression . Together , these results suggest that changes in gross microtubule structure and stability is not a likely cause of APP-induced transport defects . Instead , our results support the idea that Nebula facilitates axonal transport defects by correcting APP-mediated changes in phosphatase and kinase signaling pathways . First , we find that APP upregulation elevated intracellular calcium level and calcineurin activity , and that restoring calcineurin activity to normal suppressed the synaptotagmin aggregate accumulation in axons . The observed increase in calcium and calcineurin activity is consistent with reports of calcium dyshomeostasis and elevated calcineurin phosphatase activity found in AD brains [78]–[80] , as well as reports demonstrating elevated neuronal calcium level due to APP overexpression and increased calcineurin activation in Tg2576 transgenic mice carrying the APPswe mutant allele [81] , [82] . Second , APP upregulation resulted in calcineurin dependent dephosphorylation of GSK-3β at Ser9 site , a process thought to activate GSK-3β kinase [56] . APP upregulation also triggered calcineurin-independent phosphorylation at Tyr216 site , which has been shown to enhance GSK-3β activity [64] , [65] . The kinase ( s ) that phosphorylates APP at Tyr216 is currently not well understood , it will be important to study how APP leads to Tyr216 phosphorylation in the future . Based on our results , we envision that APP overexpression ultimately leads to excessive calcineurin and GSK-3β activity , whereas nebula overexpression inhibits calcineurin to prevent activation of GSK-3β ( Fig . S11 ) . Our findings that nebula co-overexpression prevented GSK-3β activation and enhanced the transport of APP-YFP vesicles are consistent with a recent report by Weaver et al . , in which they find decreasing GSK-3β in fly increased the speed of APP-YFP movement [51] . Furthermore , consistent with our result that APP upregulation triggers GSK-3β enhancement and severe axonal transport defect , Weaver et al . did not detect changes in GFP-synaptotagmin movement in the absence of APP upregulation . Active GSK-3β has been shown to influence the transport of mitochondria and synaptic proteins including APP , although the exact mechanism may differ between different cargos and motors [51] , [83] , [84] . One mechanism proposed for GSK-3β-mediated regulation of axonal transport is through phosphorylation of KLC1 , thereby disrupting axonal transport by decreasing the association of the anterograde molecular motor with its cargos [59] . Accordingly , we find that APP reduced KLC-synaptotagmin interaction while Nebula upregulation preserved it . Synaptotagmin transport in both the anterograde and retrograde directions were affected , consistent with previous reports showing that altering either the anterograde kinesin or retrograde dynein is sufficient affected transport in both directions [85] , [86] . Our results also support work suggesting that synaptotagmin can be transported by the kinesin 1 motor complex in addition to the kinesin 3/imac motor [67]–[73] . As kinesin 1 is known to mediate the movement of both APP and mitochondria [37] , [86]–[88] , and that phosphorylation of KLC had been shown to inhibit mitochondrial transport [89] , detachment of cargo-motor caused by GSK-3β mediated phosphorylation of KLC may lead to general axonal transport problems as reported here . However , GSK-3β activation may also perturb general axonal transport by influencing motor activity or binding of motors to the microtubule tract . Interestingly , increased levels of active GSK-3β and phosphorylated KLC and dynein intermediate chain ( DIC ) , a component of the dynein retrograde complex , have been observed in the frontal complex of AD patients [90] . Genetic variability for KLC1 is thought to be a risk factor for early-onset of Alzheimer's disease [91] . There is also increasing evidence implicating GSK-3β in regulating transport by modulating kinesin activity and exacerbating neurodegeneration in AD through tau hyperphosphorylation [21] , [51] , [55] . It will be interesting to investigate if Nebula also modulates these processes in the future . Although calcineurin had been shown to regulate many important cellular pathways , the link between altered calcineurin and axonal transport , especially in the context of AD , had not been established before . We show that calcineurin can regulate axonal transport through both GSK-3β independent and dependent pathways . This is supported by our observation that the severity of the aggregate phenotype was worse for flies expressing APP and active calcineurin than it was for flies expressing APP and active GSK-3β . These findings point to a role for calcineurin in influencing axonal transport directly , perhaps through dephosphorylation of motor or adaptor proteins . Our data also indicate that calcineurin in part modulates axonal transport through dephosphorylation of GSK-3β as discussed above; however , upregulation of APP is necessary for the induction of severe axonal transport problems , mainly by causing additional enhancement of GSK-3β signaling . GSK3 inhibition is widely discussed as a potential therapeutic intervention for AD , our results suggest that perhaps calcineurin is a more effective target for delaying degeneration by preserving axonal transport . DSCR1 and APP are both located on chromosome 21 and upregulated in DS [4] , [10] . Overexpression of DSCR1 alone had been contradictorily implicated in both conferring resistance to oxidative stress and in promoting apoptosis [8] , [9] , [16] , [17] . Upregulation of Nebula/DSCR1 had also been shown to negatively impact learning and memory in fly and mouse models through altered calcineurin pathways [15] , [92] . How could upregulation of DSCR1 be beneficial ? We propose that DSCR1 upregulation in the presence of APP upregulation compensates for the altered calcineurin and GSK-3β signaling , shifting the delicate balance of kinase/phosphatase signaling pathways close to normal , therefore preserving axonal transport and delaying neurodegeneration . We also propose that axonal transport defects and synapse dysfunction caused by APP upregulation in our Drosophila model system occur prior to accumulation of amyloid plaques and severe neurodegeneration , similar to that described for a mouse model [30] . DS is characterized by the presence of AD neuropathologies early in life , but most DS individuals do not exhibit signs of dementia until decades later , indicating that there is a delayed progression of cognitive decline [2] , [93] . The upregulation of DSCR1 may in fact activate compensatory cell signaling mechanisms that provide protection against APP-mediated oxidative stress , aberrant calcium , and altered calcineurin and GSK3-β activity .
Flies were cultured at 25°C on standard cornmeal , yeast , sugar , and agar medium under a 12 hour light and 12 hour dark cycle . The following fly lines were obtained from the Bloomington Drosophila Stock Center: Gmr-GAL4 , UAS-APP695-N-myc ( 6700 ) , sgg1/FM7a , UAS-sggS9A ( Sgg constitutively active ) , UAS-nla-RNAi ( 27260 ) , UAS-CaNB-RNAi ( 27307 ) , UAS-syt . eGFP ( 6925 ) , UAS-APP . YFP ( 32039 ) , and UAS-mitoGFP . Elav-GAL4 stock was kindly provided by Dr . Feany ( Harvard University ) , UAS-nlat1 , and nla1 flies were reported previously [15] . UAS-ΔCaNAct construct ( constitutively active calcineurin ) was generated by deleting the autoinhibitory domain of the CaNA gene Pp2B-14D and subcloned into the pINDY6 vector similar to that described [94] . UAS-Case12 was generated by inserting Case12 ( from Evrogen ) into pINDY6 vector [53] . Transgenic flies were generated by standard germline transformation method [95] . Adult Drosophila of 0 , 15 , 30 and 45 days of age were collected , decapitated and had their proboscis removed . Heads were incubated in Mirsky's fixative for 30 minutes , washed with PBS , and post-fixed in 4% paraformaldehyde for 20 minutes . Fly heads were then transferred to 25% sucrose overnight at 4°C and were subsequently embedded in Tissue-Tek O . C . T Compound for cryostat sectioning ( 10 µm ) . Photoreceptor axons were immunostained with 24B10 ( 1∶10; Developmental Studies Hybridoma ) , Phosphorylated APP ( 1∶400; Sigma ) , and 4G8 ( 1∶500; Signet ) . Flies were placed in 2 clear round bottom test tubes joined at the opening . After allowing 2 minutes for the flies to acclimate to the tubes , flies were lightly tapped and the percentage of flies that moved toward light in horizontal position within 30 seconds was counted . Wandering 3rd instar larvae were dissected in cold calcium-free dissection buffer and fixed with 4% paraformaldehyde in PBS for 25 minutes at room temperature ( RT ) . Samples were blocked in 5% normal goat serum in PBS+0 . 1% triton for 1 hour at RT and then incubated with primary antibodies overnight at 4°C . Antibodies included synaptotagmin ( 1∶1 , 000; gift from H . Bellen ) and mAb 4G8 ( 1∶1 , 000; Signet ) , β-tubulin ( 1∶1000; DSHB ) , acetylated tubulin ( 1∶500 , Abcam ) , Cy3-conjugated HRP ( 1∶200 , Jackson ImmunoResearch ) . Alexa-conjugated secondary antibodies were applied at 1∶500 and samples mounted in Pro -long Gold Antifade reagent ( Invitrogen ) . Images of motor axons and synaptic terminals from NMJ 6/7 in segment A2 or A3 were captured in a z-series using Zeiss LSM5 scanning confocal . The number of aggregates was determined manually by counting the number of punctate staining with intensity above background and size greater than 0 . 2 µm2 . For quantification of antibody staining intensities at the NMJ , dissected larvae were stained together using the same condition . Images were captured in a z-series and parameters were set to minimize saturation of pixel intensity . Intensity of Z-projected images was analyzed using ImageJ and fold changed calculated by comparing to the control . Wandering 3rd instar larvae expressing APP-YFP or GFP-SYT in combination with other transgenes were dissected in calcium free dissection buffer: 128 mM NaCl , 1 mM EGTA , 4 mM MgCl2 , 2 mM KCl , 5 mM HEPES , and 36 mM sucrose . Live imaging of GFP-SYT was done as described [96] . For imaging of mito-GFP , dissected larvae were bathed in HL-3 solution [97] . Time-lapse images were acquired at 5-s intervals using a Zeiss LSM5 confocal using minimum laser intensity to prevent photobleaching and damage to the tissues . Images were acquired for 5 minutes with a 63× lens and a zoom of 1 . 7 . All live imaging experiments were completed within 15 minutes starting from the time of dissection in order to ensure health of the samples . The Manual tracking Plugin in ImageJ was used to track individual vesicle and mitochondria movement . At least 10 frames ( >50 s ) were used to calculate the average speed of movement . Percentage of movement was determined by counting the percentage of moving vesicles over the imaging period . A vesicle is labeled as moving if it moved in three consecutive frames ( over a 15-s period ) over a distance of at least 0 . 1 µm . Direction of movement is determined by direction of net displacement of the vesicle at the start of imaging . Average speed was determined by tracking a vesicle for an uninterrupted run in either the anterograde or retrograde direction . The total distance of movement was divided by the total duration of movement in a specific direction . Student's t-test was used to determine statistical significance . Deficits in larval locomotor behavior were assessed as described previously [98] . Briefly , larvae were washed with PBS and placed in 60 mm petri dish filled with 1% agarose . Using a moistened paint brush , 3rd instar larvae were collected and allowed to habituate for 30 seconds . The number 0 . 5 cm2 boxes entered was counted for a 60-s period . Drosophila adults ( 1–2 days ) were collected on dry ice . Heads were removed and homogenized in cold RIPA buffer . The brains of 3rd instar larvae were dissected and collected on dry ice . Equal amount of protein per genotype ( 10–20 µg ) was run on SDS polyacrylamide gel and transferred to nitrocellulose membrane . Blocking for phosphorylated antibodies was performed using 5% BSA in PBS+0 . 1% tween ( PBS-TW ) . Blocking for non-phosphorylated antibodies was done using 5% milk in PBS-TW for one hour at RT . Membranes were incubated with the following primary antibodies overnight at 4°C: N-APP ( 1∶5 , 000; Sigma ) , β-tubulin ( 1∶500; Developmental Studies Hybridoma Bank ) , Nebula ( 1∶7 , 000 ) , Fasciclin II ( 1∶50 Developmental Studies Hybridoma ) , acetylated tubulin ( 1∶1 , 000 , Cell Signaling ) , phospho-GSK3β Ser9 ( 1∶1000 , Cell Signaling ) , phospho-GSK3β Tyr126 ( 1∶1000 , Cell Signaling ) , and GSK3 α/β ( 1∶2 , 000 , Cell Signaling ) . Secondary antibodies used were: anti-mouse Alexa 680 ( Invitrogen ) , anti-rabbit Dylight 800 ( Piercenet ) , anti-mouse coupled HRP or anti-rabbit coupled HRP . HRP signals were detected using ECL Reagents ( GE Healthcare ) . Alexa 680 and Dylight 800 signals were detected using Odyssey Imaging system ( LI-COR Biosciences ) . For reprobing , membranes were stripped using Reblot Plus strong antibody stripping solution ( Millipore ) and reprobed . NIH Image J software was used to measure signal intensity , and the fold change in specific protein level was normalized to a loading control and compared to the control flies . Fly heads were collected over dry ice , decapitated , and homogenized in lysis buffer ( 10 mM Tris pH 7 . 5 , 1 mM EDTA , 0 . 02% Sodium Azide ) . Calcineurin phosphatase activity was determined using the Ser/Threonine Phosphatase Assay Kit ( Promega ) following the manufacturer's protocol as done previously [15] . 5 µg of protein per genotype was used . Flies heads were collected on dry ice by passing through molecular sieves and homogenized in lysis buffer ( 10 mM HEPES , 0 . 1 M NaCl , 1% NP-40 , 2 mM EDTA , 50 mM NaF , 1 mM NA3VO4 ) plus Complete Mini protease inhibitor cocktail ( Roche ) . Lysates were pre-cleared by incubating fly extract with magnetic A/G beads ( Thermo Scientific ) for 1 hour at 4°C . Pre-cleared extract was then used for IP using GFP antibody conjugated to magnetic beads ( MBL International ) . Western blot analysis using an antibody against the kinesin light chain ( 1∶200; Novus Biologicals ) was used to confirm interaction . To determine the efficiency of GFP pull down , an antibody against GFP ( 1∶1000 , Abnova ) was also used . To eliminate signal contamination from IgG , we used HRP conjugated TrueBlot anti-rabbit IgG ( 1∶1000 , ebioscience ) that is specific for native IgG as secondary antibody . | Alzheimer's disease ( AD ) is a debilitating neurodegenerative disease characterized by gradual neuronal cell loss and memory decline . Importantly , Down syndrome ( DS ) individuals over 40 years of age almost always develop neuropathological features of AD , although most do not develop dementia until at least two decades later . These findings suggest that DS and AD may share common genetic causes and that a neuroprotective mechanism may delay neurodegeneration and cognitive decline . It has been shown that the amyloid precursor protein ( APP ) , which is associated with AD when duplicated and upregulated in DS , is a key gene contributing to AD pathologies and axonal transport abnormalities . Here , using fruit fly as a simple model organism , we examined the role of Down syndrome critical region 1 ( DSCR1 ) , another gene located on chromosome 21 and upregulated in both DS and AD , in modulating APP phenotypes . We find that upregulation of DSCR1 ( Nebula in flies ) is neuroprotective in the presence of APP upregulation . We report that nebula overexpression delays the onset of neurodegeneration and transport blockage in neuronal cells . Our results further suggest that signaling pathways downstream of DSCR1 may be potential therapeutic targets for AD . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | Nebula/DSCR1 Upregulation Delays Neurodegeneration and Protects against APP-Induced Axonal Transport Defects by Restoring Calcineurin and GSK-3β Signaling |
Root vacuolar sequestration is one of the best-conserved plant strategies to cope with heavy metal toxicity . Here we report that zinc ( Zn ) tolerance in Arabidopsis requires the action of a novel Major Facilitator Superfamily ( MFS ) transporter . We show that ZIF2 ( Zinc-Induced Facilitator 2 ) localises primarily at the tonoplast of root cortical cells and is a functional transporter able to mediate Zn efflux when heterologously expressed in yeast . By affecting plant tissue partitioning of the metal ion , loss of ZIF2 function exacerbates plant sensitivity to excess Zn , while its overexpression enhances Zn tolerance . The ZIF2 gene is Zn-induced and an intron retention event in its 5′UTR generates two splice variants ( ZIF2 . 1 and ZIF2 . 2 ) encoding the same protein . Importantly , high Zn favours production of the longer ZIF2 . 2 transcript , which compared to ZIF2 . 1 confers greater Zn tolerance to transgenic plants by promoting higher root Zn immobilization . We show that the retained intron in the ZIF2 5′UTR enhances translation in a Zn-responsive manner , markedly promoting ZIF2 protein expression under excess Zn . Moreover , Zn regulation of translation driven by the ZIF2 . 2 5′UTR depends largely on a predicted stable stem loop immediately upstream of the start codon that is lost in the ZIF2 . 1 5′UTR . Collectively , our findings indicate that alternative splicing controls the levels of a Zn-responsive mRNA variant of the ZIF2 transporter to enhance plant tolerance to the metal ion .
The transition metal zinc ( Zn ) is a micronutrient essential for optimal plant growth and development , serving as a catalytic co-factor or structural motif in numerous proteins required for basic cellular processes . However , its propensity to inactivate other crucial proteins also renders Zn a potentially toxic element that can adversely affect plant growth when present in excessive amounts [1] . To deal with these opposing effects and adjust to environmental fluctuations in Zn availability , plants developed a tightly controlled homeostatic network aimed at ensuring an adequate supply of the nutrient while preventing its toxic build-up at the cellular and whole-plant levels [2] . Zn is acquired from the soil as a divalent cation that once absorbed into the root epidermis moves mostly symplastically through the adjacent cell layers to the central stele . After active loading into the xylem vessels , Zn is translocated to the shoot via the transpiration stream and subsequently transferred to the phloem before allocation to aerial organs [3] . While plants adapt to zinc shortage supply prevalently by activating cellular Zn uptake particularly at the root-soil interface [4] , cellular Zn detoxification mechanisms are primarily intended at restricting the accumulation of free cytosolic Zn2+ , mainly through its extrusion to the apoplasm [5] , chelation with specific ligands [6] , [7] and/or vacuolar compartmentalisation [8] . At the whole-plant level , tolerance to excess Zn is achieved through the rearrangement of its tissue partitioning via enhanced Zn sequestration in leaves [9] , whereas within the root both immobilisation in the outer cell layers [10] and exclusion from the epidermis [11] contribute to limit Zn entry into the root symplasm . All steps in plant Zn homeostasis , from initial acquisition by the root to subsequent distribution among different tissues and subcellular trafficking , require the concerted action of multiple Zn membrane transport systems . Three groups of such carrier proteins have been primarily characterised in Arabidopsis thaliana . While members of the ZIP/IRT ( Zinc-regulated transporter , Iron-regulated transporter-like Protein ) family have been proposed to constitute the major cellular Zn uptake system [12]–[14] , HMAs ( Heavy–metal transporting ATPases ) and MTPs ( Metal Tolerance Proteins ) , which belong to the P1B-type ATPases and Cation Diffusion Facilitator ( CDF ) families , respectively , are essential for full Zn tolerance owing to their ability to drive either Zn whole-plant partitioning or organelle sequestration/exclusion [8]–[10] , [15]–[19] . The recent implication of two novel Arabidopsis transporters , PCR2 ( Plant Cadmium Resistance 2; [11] ) and the Major Facilitator Superfamily ( MFS ) carrier ZIF1 ( Zinc-Induced Facilitator 1; [20] ) , in Zn tolerance suggests that additional plant Zn transport components are still to be discovered . On the fringe of the vast majority of plants that like A . thaliana typically possess basal Zn tolerance , a specialized flora displays naturally selected tolerance to extremely high Zn concentrations , being able to accumulate considerable Zn amounts in their aboveground parts without developing toxicity symptoms . This hyperaccumulation strategy relies primarily on efficient xylem loading and vacuolar sequestration in leaves , thus inducing a systemic Zn-deficiency status that stimulates root Zn uptake [21] . Cross-species transcriptomic profiling particularly in the hyperaccumulator Arabidopsis halleri , a close A . thaliana relative , has established that key Zn transporters , such as HMA4 , MTP1 , HMA3 and ZIP/IRTs , are not different but rather constitutively overexpressed in hyperaccumulators [22]–[24] . Considering not only differential expression between the two Arabidopsis species but also individual regulation by exogenous Zn , Becher et al . [23] provided a shortlist of 50 promising candidate genes for a role in Zn homeostasis that notably includes At2g48020 encoding a yet uncharacterised MFS carrier , which given the Arabidopsis ZIF1 role in zinc tolerance [20] raises particular interest . MFS transporters are single-polypeptide secondary carriers using chemiosmotic gradients as an energy source and ubiquitous to all classes of organisms . With more than 120 predicted members , the MFS represents the second widest group of Arabidopsis transporters after the ATP-Binding Cassette ( ABC ) superfamily [25] , [26] . Interestingly , according to the current genome annotation ( TAIR V10; www . arabidopsis . org ) , 34 MFS genes undergo alternative splicing , a key generator of proteomic diversity and functional complexity in higher eukaryotes . Despite the fact that over 60% of the Arabidopsis multi-exon genes are currently estimated to produce more than one transcript via this mRNA processing mechanism [27] , its biological significance in plants is poorly understood , having been uncovered for only a dozen genes [28] , [29] . Here , we report that the MFS transporter encoded by the At2g48020 gene , which we named ZIF2 , sustains plant Zn tolerance in A . thaliana by mediating root vacuolar sequestration of the metal , thus preventing its translocation to the shoot . Importantly , we found that elevated Zn levels favour expression of a ZIF2 splice variant whose longer 5′ untranslated region ( UTR ) enhances the translation efficiency of the ZIF2 transporter and hence plant Zn tolerance .
To initiate the characterisation of the ZIF2 gene ( At2g48020 ) , we examined its organ- and tissue-specific expression pattern by means of reporter gene experiments . Staining of transgenic Arabidopsis lines stably expressing β-glucuronidase ( GUS ) under the control of the ZIF2 promoter revealed that the latter is active in most plant organs ( Figure 1A–H ) . In particular , ZIF2 was highly expressed at all stages of floral development ( Figure 1A ) , with flower buds displaying exclusive staining of the style , stamen filaments and sepals ( Figure 1B ) . Promoter activity was also detected in leaf mesophyll cells ( Figure 1C ) and strongly throughout the root system ( Figure 1D–H ) . Although homogenous GUS coloration was observed all along the primary root ( PR ) ( Figure 1D ) , the ZIF2 promoter was particularly active at the root tip ( Figure 1E ) . We detected very intense GUS staining in lateral root ( LR ) primordia ( Figure 1F , G ) , but ZIF2 was also abundantly expressed in elongating LRs ( Figure 1H ) . We further examined the precise root tissue distribution of ZIF2 promoter activity using transgenic plants expressing the green fluorescent protein ( GFP ) under the control of the native ZIF2 promoter ( Figure 1I–L ) . At the PR tip , ZIF2 promoter activity was confined to the endodermal and cortical cell layers of the elongation and transition zones , while absent from the apical meristem , the quiescent centre , columella cells and the lateral root cap ( Figure 1I ) . Similarly , in the mature portion of the PR , the GFP signal was detected in both the endodermis and the cortex ( Figure 1K ) . A slightly different layer-specific pattern was observed in LRs , with ZIF2 expression appearing largely restricted to the cortex , particularly at the tip ( Figure 1J , L ) . ZIF2 promoter activity in flowers and leaves was insufficient to allow detection of the GFP signal . Collectively , the above results suggest that ZIF2 exerts a prominent role in the root system , particularly in the PR . According to the current genome annotation ( TAIR V10 ) , the Arabidopsis ZIF2 gene contains 17 exons ( Figure 2A ) and generates two distinct mRNAs that differ solely in their 5′UTRs . In order to verify the accuracy of these predictions and examine the tissue-specific distribution of the ZIF2 transcripts , we performed RNA Ligase-Mediated-5′ Rapid Amplification of cDNA Ends ( RLM-5′RACE ) in different Arabidopsis vegetative and reproductive tissues , i . e . in seedlings , adult leaves and flowers . Given the high ZIF2 promoter activity detected throughout the root system ( see Figure 1 ) and the differential expression of the ZIF2 orthologue reported in the Zn hyperaccumulator A . halleri [23] , we also performed RLM-5′RACE in roots of seedlings exposed to Zn toxicity . As shown in Figure 2B , we were able to identify nine alternative ZIF2 5′UTRs and named the corresponding transcripts ZIF2 . 1 ( At2g48020 . 1 according to TAIR V10 ) to ZIF2 . 9 . A common start codon is predicted for all ZIF2 transcripts , and we verified the absence of an in-frame ATG or alternate start codon in the nine 5′UTRs as well as of any alterations in the coding sequence . Hence , all ZIF2 transcripts are predicted to encode the same 53-kDa full-size transporter , displaying the typical MFS transporter signature motif that includes two transmembrane domains , each consisting of six membrane-spanning segments , delimiting a central hydrophilic loop [25] . Distribution of the ZIF2 transcripts was found to exhibit tissue specificity ( Figure 2B ) . Interestingly , the ZIF2 . 1 and ZIF2 . 2 splice variants , which result from retention of a 139-nt intron in the 5′UTR , appeared to be predominantly expressed in roots challenged with high Zn concentrations , hinting at a role for this intron retention event in roots exposed to excess Zn . We then examined the tissue-specific distribution of the ZIF2 . 1 and ZIF2 . 2 splice variants by RT-PCR ( Figure 3A , B ) . Consistent with the promoter activity profile shown in Figure 1 , RT-qPCR analysis revealed that ZIF2 is expressed throughout plant development , particularly in roots and flowers of mature plants , and to a lesser extent in young seedlings , while lower transcript levels were detected in other aboveground tissues such as developing siliques , stems or leaves ( Figure 3B ) . By contrast , the ZIF2 . 1 and ZIF2 . 2 transcripts were detected at very low levels in all tissues analysed , except in roots of mature plants where the expression level of the two splice variants combined was found to represent more than half of total ZIF2 expression . ZIF2 . 1 and ZIF2 . 2 expression was also markedly induced during leaf senescence ( Figure 3A , B ) . We next asked whether ZIF2 expression would be responsive to exogenous Zn in A . thaliana roots . Both total ZIF2 expression and the equivalent amounts of the ZIF2 . 1 and ZIF2 . 2 splice variants detected in root tissues under control Zn conditions ( 30 µM ) appeared unchanged by Zn deficiency ( Figure 3C , D ) . However , concomitant with a significant induction of ZIF2 expression , a clear shift in the ZIF2 splicing pattern occurred after at least 48 h of challenge with high Zn ( Figure 3C , D; Figure S1 ) . Indeed , root ZIF2 . 2 expression was markedly up-regulated in a concentration-dependent manner following exposure to Zn excess , even as slight as 100 µM , while a considerable increase in ZIF2 . 1 steady-state levels was also observed but of lower magnitude and only under toxic Zn excess ( 250 µM ) . Thus , both total expression of the ZIF2 gene and alternative splicing of its precursor mRNA ( pre-mRNA ) are regulated by the plant's Zn status , with high levels of the metal ion promoting retention of the 5′UTR intron . To determine the subcellular localisation of the protein encoded by the ZIF2 transcripts , we generated C-terminal yellow fluorescent protein ( YFP ) or GFP fusions of each cDNA under the control of the 35S promoter . Confocal microscopy analysis of the YFP signal upon transient or stable expression in Arabidopsis protoplasts ( Figure 4A–C ) or root apices ( Figure 4D–F ) , respectively , suggested that ZIF2 localises to the vacuolar membrane . In co-localisation experiments in tobacco leaf epidermal cells using specific tonoplast and plasma-membrane RFP markers [30] , the ZIF2 . 2-GFP fusion protein co-localised exclusively with the tonoplast marker , whereas it did not match the distribution of the plasma membrane marker ( Figure 4G–R ) , confirming that the ZIF2 transporter is primarily targeted to the tonoplast of plant cells . The transport properties of the plant ZIF2 carrier were then explored by means of heterologous expression in Saccharomyces cerevisiae . Correct expression of the GFP-ZIF2 fusion protein in yeast was confirmed by immunoblotting ( Figure S2A ) and fluorescence microscopy ( Figure S2B ) . To determine whether the plant transporter is able to influence Zn delivery to yeast cells , the GFP-ZIF2 fusion was introduced into the Δzrt1zrt2 and Δzrc1cot1 mutant strains , which are sensitive to low and high Zn concentrations , respectively [31] , [32] . As seen in Figures 5 and S2C , yeast cells expressing GFP-ZIF2 displayed no significant growth difference with respect to cells transformed with the empty vector in normal nutrition medium ( 2 . 5 µM Zn2+ ) . Due to the lack of the Zn importers Zrt1 and Zrt2 , the Δzrt1zrt2 mutant exhibited as expected [32] a severe growth retardation under Zn-limiting conditions that was further exacerbated by GFP-ZIF2 expression . Conversely , GFP-ZIF2 was able to markedly alleviate the growth defects induced at high Zn concentrations by loss of the vacuolar transporters Zrc1 and Cot1 [31] . Similar although less pronounced trends were observed when ZIF2 was expressed in the wild-type background ( Figure 5 and Figure S2C ) . To correlate these effects with an enhanced Zn release capacity of yeast cells , we measured the total Zn concentration of the different strains grown under various Zn conditions ( Table 1 ) . While intracellular Zn accumulation was similar for all cell types under normal Zn supply , expression of the plant carrier significantly reduced the intracellular Zn concentration in all tested backgrounds when compared with cells carrying the empty vector under both low and high Zn supplies . These findings further confirm that the ZIF2 transporter is able to mediate Zn extrusion from yeast cells . By contrast , expression of the plant transporter was unable to rescue the growth defects induced at high Co concentrations by loss of the Zrc1 and Cot1 transporters [31] , nor was it able to influence yeast response to several other plant beneficial ions ( Figure S3A ) . ZIF2 , also known as ERD6 ( Early Response to Dehydration ) -like 7 , belongs to the Monosaccharide-like Transporter subfamily of MFS transporters and indeed one of its closest S . cerevisae homologues is the inositol transporter Itr1 [33] . Nevertheless , GFP-ZIF2 was unable to rescue the Δitr1 deletion mutant's deficient growth under limiting myo-inositol concentrations ( Figure S3B ) . Conversely , expression of GFP-ZIF2 conferred enhanced yeast resistance to acetate , malate and citrate ( Figure S3C ) , indicating that besides Zn , ZIF2 is also able to modulate weak acid sensitivity in yeast . To uncover the in vivo role of the ZIF2 transporter , we isolated an A . thaliana mutant allele ( SALK_037804 ) , harbouring a T-DNA insertion in the tenth intron of the ZIF2 gene , which we named zif2-1 ( see Figure 2A ) . RT-PCR analysis of ZIF2 expression in zif2-1 homozygous seedlings using primers annealing upstream of the insertion site revealed transcript levels comparable to wild-type plants , but no expression was detected when primers flanking or annealing downstream of the T-DNA segment were used ( Figure S4 ) . This indicated that the mutant allele produces a truncated version of the ZIF2 transcripts that lacks nearly the entire sequence corresponding to the second transmembrane domain and is thus unlikely to encode a functional membrane transporter [34] , strongly suggesting that zif2-1 is a true loss-of-function mutant . When grown in vitro under optimal conditions , zif2-1 mutant seedlings appeared morphologically indistinguishable from the corresponding wild type ( Col-0 ) , showing normal shoot growth , chlorophyll content and root system development ( Figure 6A , Table S1 ) . In the presence of excessive Zn amounts , Arabidopsis seedlings develop toxicity symptoms , typically including shoot growth retardation , leaf chlorosis and inhibition of PR elongation [1] . These three Zn toxicity hallmarks were visibly exacerbated in zif2-1 when compared to the wild type following exposure to a wide Zn toxicity range , though to a lesser extent than in the previously characterised [20] zif1-2 mutant ( Figure 6 ) . By contrast , no significant differences between zif2-1 and wild-type seedlings could be observed at the PR elongation level under conditions of Zn deficiency , even when assayed on media with low cation content [6] ( Figure S5A ) . In contrast to the zif1-2 mutant [7] , [20] , no zif2-1 phenotype was observed under conditions of Fe depletion ( Figure S5B ) or in the presence of excessive amounts of Cd or Ni ( Figure S6 ) . At least at the tested concentrations , sensitivity of the mutant to excess of other essential , beneficial or rhizotoxic cations ( Figure S6 ) or to myo-inositol ( Figure S7 ) was also unaltered , as was sensitivity to sucrose and glucose , whose protective effect on Zn toxicity was unchanged by the zif2-1 mutation ( Figure S8 ) . Taken together , these data suggest that ZIF2 acts specifically as a Zn detoxifier in Arabidopsis . To exclude the possibility that the observed phenotype results from disruption of another gene , we transformed the zif2-1 mutant with a genomic fragment spanning the entire ZIF2 gene and including the same promoter sequence used in the reporter gene experiments . These complementation lines exhibited complete restoration of wild-type sensitivity to excess Zn ( Figure S9 ) , thus confirming that the ZIF2 transporter participates in basal Zn tolerance in Arabidopsis . To gain insight into the physiological relevance of the ZIF2 5′UTR intron retention event promoted in roots by high Zn exposure , we selected three transgenic Arabidopsis lines independently expressing either the ZIF2 . 1 or ZIF2 . 2 cDNA under the control of the 35S promoter in the wild-type background that similarly displayed an approximately five-fold increase in expression of either splice variant when compared to wild-type plants ( Figure 7A , 7B ) . Importantly , our RT-PCR analyses also showed that ZIF2 . 1 levels in the ZIF2 . 2-overexpressing lines were similar to those detected in the wild type , indicating that the 5′UTR intron is not spliced from the transgenic ZIF2 . 2 cDNA . No obvious phenotypical alterations were observed between wild-type and ZIF2-overexpressing seedlings grown under control Zn conditions ( Figure 7C , Table S1 ) . However , ZIF2 . 1 or ZIF2 . 2 overexpression substantially attenuated the detrimental effects induced by excessive Zn amounts over a broad range of Zn supplies ( Figure 7C , D; Figure S10 ) . Together with the above results , these data demonstrate that ZIF2 deletion and overexpression confer exact opposite phenotypes in Arabidopsis upon challenge with Zn excess . Interestingly , the three ZIF2 . 2-overexpressing lines consistently exhibited significantly greater Zn tolerance levels than the three overexpressing ZIF2 . 1 ( Figure 7C , D; Figure S10 ) . Thus , in addition to being prevalently expressed under Zn stress ( see Figure 3C , D; Figure S1 ) , the ZIF2 . 2 splice variant contributes more significantly to plant Zn tolerance than ZIF2 . 1 . As a first step towards unravelling the mechanisms by which the ZIF2 transporter mediates Zn tolerance in Arabidopsis , we determined the Zn concentration of shoot and root tissues in seedlings grown under control Zn supply or moderate Zn stress ( Figure 8 ) . As expected , the Zn concentration of both tissues increased while the corresponding Zn shoot-to-root ratio decreased commensurately with Zn supply increment . Under any of the Zn supplies tested , the above-ground parts of the zif2-1 mutant accumulated on average 20% more Zn than wild-type shoots , whereas zif2-1 mutant roots concentrated significantly less Zn than those of the wild type . The opposite Zn partitioning trend was observed in ZIF2-overexpressing plants , which accumulate respectively less and more Zn in their shoot and root tissues than wild-type plants . This tendency , while also observed under control conditions , was substantially more pronounced at the two highest Zn concentrations ( Figure 8 ) . As the total amount of Zn quantified at the whole seedling level was globally comparable in all genotypes ( Figure S11 ) , the Zn shoot-to-root ratio was enhanced and reduced by ZIF2 loss of function and overexpression , respectively . These results indicate that ZIF2 function alters Zn partitioning between roots and shoots by driving root Zn immobilization and therefore shoot Zn exclusion , particularly under conditions of Zn excess . Importantly , the effects of ZIF2 overexpression on Zn distribution were significantly more pronounced in plants ectopically expressing the ZIF2 transcript with the longer 5′UTR , ZIF2 . 2 , than in the ZIF2 . 1 transgenics ( Figure 8 ) , in clear agreement with the physiological data obtained for both types of lines ( see Figures 7 and S10 ) . We next sought to understand the different functional impact of the two ZIF2 splice variants on Zn tolerance and whole-plant partitioning in Arabidopsis . Given that ZIF2 . 1- and ZIF2 . 2-overexpressing seedlings displayed very similar steady-state levels of the respective ZIF2 transcripts ( see Figure 7A , 7B ) , which encode the same transporter ( see Figure 2B ) , the differences in phenotype magnitude are unlikely to result from distinct mRNA levels and/or protein stability/activity , thus pointing to a translational regulatory mechanism . To assess the ZIF2 protein levels produced by each ZIF2 splice variant in planta , we took advantage of our ZIF2-YFP overexpression lines ( see Figure 4E , F ) , providing a setting in which the abundance of the ZIF2 transporter can be accurately linked to the amount of each splice variant through quantification of the corresponding YFP signal and transgene levels . Two transgenic Arabidopsis lines independently expressing either ZIF2 . 1-YFP or ZIF2 . 2-YFP under the control of the 35S promoter in the wild-type background were selected . Both the ZIF2 . 1-YFP and the ZIF2 . 2-YFP transgenic lines exhibited increased resistance to excess Zn when compared to the wild type ( Figure S12 ) , demonstrating the functionality of the ZIF2-YFP fusion proteins . Moreover , as with our other set of overexpression plants ( see Figures 7 and S10 ) , both ZIF2 . 2-YFP transgenic lines were significantly more tolerant to Zn toxicity than the two lines expressing ZIF2 . 1-YFP ( Figure S12 ) . Noticeably , striking differences in PR and PR tip ZIF2-YFP fluorescence levels were observed between the two types of transgenic plants , with a substantially higher ZIF2-YFP signal being detected in seedlings overexpressing the ZIF2 . 2-YFP fusion under control conditions ( Figure 9A , Figure S13 ) . Importantly , all four transgenic lines displayed comparable ZIF2-YFP transcript levels , with normalization of the YFP fluorescence intensity to the corresponding transgene transcript levels showing that the relative abundance of the ZIF2-YFP protein at the tonoplast of root tip cells is 2–3 fold higher in plants expressing the ZIF2 . 2 splice variant ( Figure 9B ) . Accordingly , higher ZIF2-YFP protein fusion levels were detected by western blot analysis in ZIF2 . 2-YFP than in ZIF2 . 1-YFP transgenic lines ( Figure 9C ) . This indicated that in planta the ZIF2 . 2 splice variant is more efficiently translated than ZIF2 . 1 . To investigate whether the ZIF2 . 1 and ZIF2 . 2 5′UTRs are sufficient to determine different translation efficiencies , the effect of each splice variant's 5′UTR on the expression of a reporter gene ( LUC ) was examined by means of an in vivo assay using isolated Arabidopsis protoplasts [35] , [36] . The ZIF2 . 15′UTR-LUC and ZIF2 . 25′UTR-LUC constructs were expressed under the control of the 35S promoter , and LUC activity was normalized to the GUS activity of the co-transfected Pro35S:GUS reporter construct . Strikingly , and despite the fact that nearly two thirds of the ZIF2 . 2-LUC transcript was spliced into the ZIF2 . 1-LUC transcript ( Figure S14A ) , under normal conditions ( 0 µM Zn ) about twice the LUC activity was detected upon transfection with ZIF2 . 25′UTR-LUC when compared with ZIF2 . 15′UTR-LUC ( Figure 9D ) . These results are consistent with the in planta data and indicate that the differences in translation efficiency of the two ZIF2 splice variants are solely attributable to their 5′UTRs . We next assessed the effect of Zn on translation driven by the two ZIF2 5′UTRs using the exact same experimental settings but in the presence of a range of Zn supplies ( Figure 9D ) . Importantly , both constructs were expressed at equivalent levels under all conditions and none of the applied Zn challenges promoted the intron retention event in the ZIF2 . 2-LUC transcript ( Figure S14A ) . Upon ZIF2 . 15′UTR-LUC transfection , LUC activity was induced by about two-fold in the presence of 5 µM Zn but this up-regulation was not further amplified at higher Zn concentrations . However , translation of the ZIF2 . 25′UTR-LUC transcript was strongly induced by Zn in a concentration-dependent manner , with a respective 2- , 2 . 5- and 4-fold induction in presence of 5 , 25 and 100 µM Zn when compared to the absence of the metal ion ( Figure 9D ) . By contrast , translation of the control LUC transcript was not significantly affected by the Zn challenge ( Figure S14B ) . Hence , our results show that the retained intron in the ZIF2 5′UTR enhances translation in a Zn-responsive fashion , markedly promoting protein expression under excess Zn levels . Interestingly , secondary structure prediction of the ZIF2 . 2 5′UTR indicated the formation of a stable imperfect stem loop immediately upstream of the ATG codon ( Figure 10A , B ) , which is predicted regardless of the length of ZIF2 sequence used , even with the full-length transcript . This structure is lost in the ZIF2 . 1 splice variant as the intron removes roughly half of the sequence participating in the structure formation . To determine whether the predicted 5′UTR secondary structure element plays a role in translational regulation of the two alternatively-spliced mRNAs , we performed two successive rounds of site-directed mutagenesis on the ZIF2 . 25′UTR-LUC construct to generate the ZIF2 . 25′UTRM-LUC ( where CTCA was mutated to AGTC ) and ZIF2 . 25′UTRR-LUC ( where TGAG was mutated to GACT ) constructs , in which the secondary structure was destabilized and restored , respectively ( Figure 10B ) . In the absence of Zn , a two-fold increase in LUC activity was detected upon transfection with any of the three ZIF2 . 25′UTR-LUC constructs when compared with the ZIF2 . 15′UTR-LUC transcript ( Figure 10C ) . However , destabilization of the secondary structure strikingly reduced the pronounced inductive effect that Zn exerts on ZIF2 . 25′UTR-LUC translation , while restoration of the structure fully rescued translational induction by the metal ion ( Figure 10C ) . The results from this last experiment indicate that the mechanism ( s ) underlying Zn regulation of translation driven by the ZIF2 5′UTR depend , at least to a large extent , on an RNA secondary structure that is present when the 5′UTR intron is retained .
Since the identification of ZIPs as the first plant Zn transporters [37] , several carriers from well-established metal transporter families have been linked to plant Zn homeostasis . The few plant MFS carriers characterised so far have been mainly implicated in sugar , nitrate , oligopeptide and phosphate transport [38]–[40] . Here we report that besides ZIF1 [20] , another MFS carrier contributes to Arabidopsis Zn tolerance , hinting at a broader role for this class of transporters in plant heavy-metal homeostasis . Conclusive evidence for a ZIF2 role in basal Zn tolerance stems from our functional analysis of Arabidopsis ZIF2 loss-of-function and overexpression lines , exhibiting respectively enhanced sensitivity and resistance to Zn toxicity . The ZIF2 tonoplastic localisation detected in planta is consistent with two previous proteomic studies of the Arabidopsis vacuole [41] , [42] and strongly suggests that ZIF2 is involved in vacuolar Zn compartmentalisation . Indeed , plant vacuoles play an essential role during metal detoxification and sequester up to 80% of cellular Zn [43] . In agreement with our physiological data , Zn concentration determination in seedlings exposed to high Zn shows that the shoot-to-root ratio of the zif2-1 mutant is unwedged to the shoot , while Zn is preferentially retained in ZIF2-overexpressing roots , demonstrating that ZIF2 function influences Zn root-to-shoot translocation under Zn excess . Given that ZIF2 is primarily expressed in the root endodermis and cortex , we propose that the encoded transporter protects from Zn toxicity by promoting vacuolar immobilization of the metal ion in these two cell layers , thus restricting symplastic movement towards the stele and subsequent xylem loading and translocation to the shoot . Richard et al . [44] showed that low root-driven Zn translocation rates to the shoot contribute to higher Zn tolerance , as reported for the Arabidopsis HMA3 , MTP3 , MTP1 and ZIF1 [10] , [19] , [20] , [45] . The partial expression pattern overlap among these vacuolar transporters and ZIF2 suggests either some degree of functional redundancy or a concerted action to regulate root Zn symplastic movement . Putative roles of the ZIF2 transporter in other plant organs where substantial ZIF2 expression was detected , such as in flowers , cannot be excluded as already shown for instance for the HMA2 and HMA4 transporters . Similarly to ZIF2 , HMA2 and HMA4 promoter activity is high in developing anthers , and a double hma2hma4 mutant exhibits a male-sterile phenotype [17] . The identification of a transporter's physiological substrate ( s ) can be instrumental in elucidating the molecular mechanisms governing its function , but remains an extremely challenging task . Importantly , our yeast heterologous expression and in planta phenotypical data point to a ZIF2 effect restricted to Zn , similarly to MTP1 and MTP3 that exhibit relatively high Zn selectivity [10] , [45] but in contrast to HMAs or ZIPs that often display broad metal selectivity contributing to plant metal homeostasis crosstalk [19] , [46] . More puzzling are our preliminary findings indicating that , despite belonging to the Monosaccharide-like subfamily , activity of the ZIF2 transporter does not seem to influence sugar sensitivity in planta nor in yeast . Instead , heterologous expression studies show that ZIF2 is able to mediate Zn efflux in yeast cells without requiring additional plant-specific factors , as already reported for HMA2 , HMA4 , MTP1 and PCR2 [5] , [11] , [16] . By contrast , the transport activity of ZIF1 is unable to complement a Zn hypersensitive yeast mutant , thus indicating that the mode of action of the two MFS carriers in mediating Zn tolerance is strikingly different . Haydon et al . [7] identified the substrate of ZIF1 as nicotianamine , a low molecular mass chelator with high affinity for a range of transition metals , and proposed that ZIF1 , by affecting nicotianamine vacuolar partitioning , specifically promotes Zn vacuolar sequestration via Zn2+/H+ antiporters . As no plasma-membrane Zn exporter has been identified so far in S . cerevisiae [47] , it seems reasonable to exclude ZIF2 activation of endogenous transporters catalysing Zn efflux . On the other hand , as MFS transporters are known to transport small solutes in response to chemiosmotic gradients and as ZIF2 affects weak acid delivery in yeast , it is tempting to speculate that ZIF2 indirectly mediates Zn vacuolar sequestration by transporting a small chelator or organic acid or by influencing the tonoplastic proton gradient . Indeed , plant cells are known to take advantage of proton gradients to actively sequester ions inside the vacuole [48] and the Arabidopsis H+-ATPase ( V-ATPase ) , which energizes transport across the tonoplast , is required for full Zn tolerance [49] . Our results also indicate that the Zn metal ion positively regulates ZIF2 expression at multiple levels . Firstly , root ZIF2 transcript levels are substantially induced following exposure to elevated Zn supplies , corroborating an essential role in Zn detoxification , as already inferred for ZIF1 [20] or MTP3 [10] . Secondly , splicing of the ZIF2 pre-mRNA is Zn-regulated , with high levels of the metal favouring retention of the 5′UTR intron and hence expression of the longer ZIF2 . 2 splice variant . The splicing factors regulating this intron retention event remain unknown . In animal systems , splice site selection is known to be primarily regulated by the highly conserved serine/arginine-rich ( SR ) and heterogeneous nuclear ribonucleoprotein particle ( hnRNP ) protein families , which bind specific exonic or intronic sequences in the pre-mRNA . While these splicing factors have been implicated in the modulation of alternative splicing also in plants , their direct endogenous targets and the mechanisms controlling differential splice site usage have not been elucidated [50] , [51] . Noticeably , our data indicate that the 5′UTR intron is not spliced from the ZIF2 . 2 transgene in our ZIF2 . 2-overexpressing lines , while splicing of this intron occurs at considerable rates from the ZIF2 . 25′UTR-LUC transcript transiently expressed in protoplasts . On the other hand , the inductive effect that Zn exerts on the intron retention event in wild-type roots is lost in the single-cell system . These apparent discrepancies could be due to the distinct tissue sources ( roots versus leaves ) and/or to differences in transcription rates , which are well known to affect alternative splicing [52] , between the different expression contexts . The molecular mechanisms underlying the regulation of alternative splicing in plants are poorly understood . Despite numerous reports that abiotic stress markedly affects plant mRNA splicing , virtually nothing is known about the effects of metal ions . Interestingly , in humans Zn has been shown to modulate alternative splicing of the hypoxia-inducible factor HIF1 [53] and the pro-apoptotic Bim [54] genes . Finally , and given that information on the functional relevance of alternative splicing in plant systems is also surprisingly scarce , a key finding of this study is that two ZIF2 splice forms differentially contribute to Zn detoxification . In fact , our two sets of overexpression data indicate not only that ZIF2 . 2 confers greater Zn tolerance than ZIF2 . 1 , but also that the different functional impact of the two mRNAs is not due to differences in their steady state levels . As the two alternative transcripts encode the same transporter , it is also highly improbable that differences in protein stability or activity account for the observed variations in resistance to Zn toxicity . Instead , we show that the stronger Zn tolerance phenotype conferred by ZIF2 . 2 results from its enhanced translation efficiency . Interestingly , Zn significantly promotes translation of both ZIF2 splice variants , but this effect is dramatically more pronounced for ZIF2 . 2 . Furthermore , our results indicate that the ZIF2 5′UTR alone is sufficient for this translational control . Indeed , 5′UTR sequence and structural features have been proposed to play an essential role in translational regulation in plants [55] , [56] , as described for other eukaryotes . It is interesting to note that expression of an Arabidopsis tonoplastic Zn2+/Mg2+ transporter , AtMHX , was shown to be repressed at the translational level through the inhibitory effect of an upstream open-reading-frame present in the 5′UTR of the corresponding gene [57] , [58] . Importantly , we were able to demonstrate that the mechanism ( s ) underlying Zn-responsive ZIF2 . 2 translation depend largely on a predicted stable stem loop structure that is lost when the ZIF2 5′UTR intron is spliced out . From the publicly available databases , the 5′UTR intron retention event in ZIF2 appears to be fairly specific to this Arabidopsis gene , as no significant sequence homology is found with other genes in A . thaliana nor in other plant species except with the ZIF2 orthologue in Arabidopsis lyrata . Similarly , the identified stable stem loop appears to be a unique feature of A . thaliana , as no common mRNA secondary structure could be predicted in the 5′UTR of plant ZIF2 orthologues . Selective recruitment of certain physiologically relevant mRNAs under stress conditions that trigger a global repression of the initiation step of translation has been widely reported in metazoans . Such a mechanism is beginning to emerge as a key process in plant adaptation to environmental stresses , interestingly for instance during sub-lethal cadmium poisoning [59] . In particular , secondary structures in the 5′UTRs have been implicated in cap-independent mRNA translation in maize under heat and other stress conditions [60] , [61] as well as in Arabidopsis under heat stress [55] . The findings presented here indicate that alternative splicing controls the levels of a Zn stress responsive mRNA variant of the ZIF2 transporter to enhance plant tolerance to the metal ion .
The Arabidopsis thaliana ( L . ) Heynh . , ecotype Colombia ( Col-0 ) , was used in all experiments . Seeds of the T-DNA insertion mutants zif2-1 ( SALK_037804 ) and zif1-2 ( SALK_011408; [20] ) were obtained from the Nottingham Arabidopsis Stock Centre ( NASC ) ( Nottingham , UK ) . The exact zif2-1 T-DNA insertion site was confirmed using gene-specific primers ( Table S2 ) and primers annealing at the T-DNA border , which also allowed PCR-based genotyping to identify homozygous lines . Plant transformation was achieved by the floral-dip method [62] using Agrobacterium tumefaciens strain EHA105 . Seeds were surface-sterilized and sown on Murashige and Skoog [63] medium solidified with 0 . 8% agar , stratified for 3 d , placed in a growth chamber and transferred to soil after 2–3 weeks . Plants were cultivated under long-day conditions ( 16-h light , 22°C/8-h dark , 18°C; 60% RH ) . RLM-5′RACE was performed using the FirstChoice RLM-RACE kit ( Ambion ) on total RNA extracted from 21-d old seedlings , adult leaves , flowers or roots exposed to high Zn according to the manufacturer's instructions and using various gene-specific primers ( Table S2 ) . Subsequently , PCR products were purified , cloned into the p-Gem-T-easy vector ( Promega ) and sequenced . For native ZIF2 promoter reporter gene experiments , a fragment including the 1233 bp immediately upstream of the start codon was PCR-amplified ( Table S2 ) from genomic DNA and inserted via the SacI/SacII restriction sites into the pKGWFS7 plasmid [64] . After agroinfiltration of this ProZIF2:GFP:GUS construct into wild-type plants , eight independent transformants all showing similar tissue-specific GUS expression patterns were recovered . Histochemical staining of GUS activity was performed as described by Sundaresan et al . [65] . RT-PCR analyses were conducted as previously described [66] using primers designed to detect ZIF2 , ZIF1 , ZIP1 , ROC10 ( CYCLOPHILIN ) and UBQ10 ( UBIQUITIN10 ) expression ( Table S2 ) . The results shown are representative of three independent experiments . Real-time RT-PCR was performed using specific primers ( Table S2 ) on a CFX384 Touch Real-Time PCR Detection System ( Bio-Rad ) using the Absolute SYBR Green ROX mix ( Thermoscientific ) according to the manufacturer's instructions . For each condition tested , two RNA extractions from different biological samples and two reverse transcription reactions for each biological repeat were performed . Data were processed using Q-Gene [67] that took the respective primer efficiency into consideration . To generate ZIF2 protein fusions with the YFP and GFP reporters , each ZIF2 transcript ( 5′UTR+coding sequence except the stop codon ) was PCR-amplified ( Table S2 ) using root cDNA as a template , and independently inserted under the control of the 35S promoter via the XhoI/PacI restriction sites into the YFP- or GFP-tagged versions of the pBA002 vector . Two transgenic lines displaying a strong fluorescence signal in the root were recovered upon transformation of wild-type plants with either Pro35S:ZIF2 . 1-YFP or Pro35S:ZIF2 . 2-YFP . These YFP constructs were also transfected into Arabidopsis protoplasts ( see below ) . Transient co-expression of the GFP constructs with the tonoplast marker γ-Tonoplast Intrinsic Protein ( TIP ) -mCherry or the plasma membrane marker Plasma membrane Intrinsic Protein 2A ( PIP2A ) -mCherry [30] and the pBIN-NA construct [68] in leaf abaxial epidermal cells of Nicotiana tabacum was performed via the agroinfiltration procedure described by Voinnet et al . [69] using A . tumefaciens strain GV3101 . The S . cerevisiae mutant strains Δzrt1zrt2 [32] and Δzrc1cot1 [31] , along with the corresponding wild-type strain DY1457 , were used in the Zn experiments . All other studies were performed in the parental strain BY4741 and the derived deletion mutant Δitr1 [33] . Cloning of the Arabidopsis ZIF2 full-length coding sequence into the pGREG576 vector ( [70]; Euroscarf collection ) and expression analysis of the corresponding GFP-ZIF2 fusion protein by fluorescence microscopy and western blotting were performed as described previously [71] . Growth curve and spot assays were conducted in MMB-U liquid or agarized medium [71] supplemented with the indicated compound ( Sigma-Aldrich ) at the desired concentration . Results presented are representative of three independent experiments . All assays were performed in a climate-controlled growth cabinet under long-day conditions . After 5 d of vertically-oriented growth on control medium ( 30 µM Zn ) , seedlings were transferred to fresh medium containing the indicated compound ( Sigma-Aldrich ) at the specified concentration . PR elongation , along with shoot biomass and chlorophyll content [72] were evaluated after an additional 1 and 3 weeks of growth , respectively . Results are representative of at least three independent experiments . For genomic complementation , a 4450-bp fragment encompassing the entire ZIF2 gene and including the 1233-bp promoter sequence described above was PCR-amplified ( Table S2 ) from genomic DNA and inserted into the promoterless version of pBA002 via the NcoI/XbaI restriction sites . Representative results for three lines recovered upon introduction of the construct into zif2-1 mutant plants are shown . ZIF2 overexpression constructs were generated as described for the YFP and GFP plasmids , except that the corresponding fragments were inserted into the pBA002 background via the XhoI/AscI restriction sites . After agroinfiltration of wild-type plants , three transgenic lines independently overexpressing each ZIF2 transcript were selected . Arabidopsis protoplasts were generated as described by Yoo et al . [35] and transfected by polyethylene glycol transformation [73] . To generate the Pro35S:ZIF25′UTR fusion constructs with the LUC reporter , the ZIF2 . 1 and ZIF2 . 2 5′UTRs were PCR-amplified ( Table S2 ) and independently cloned into the Pro35S:LUC vector [35] via the SalI/NcoI restriction sites . Site-directed mutagenesis of the Pro35S:ZIF2 . 25′UTR-LUC construct was performed using gene-specific primers ( Table S2 ) and the NZYMutagenesis kit ( NZYTech ) according to the manufacturer's instructions . Protoplast transient expression assays along with GUS and LUC activity measurements and RNA protoplast extraction were performed as described previously [74] . Results shown are representative of at least three independent experiments . Differential interference contrast and confocal images were taken with a DM LB2 microscope ( Leica ) and an LSM 510 laser scanning microscope equipped with a Meta detector ( Zeiss ) , respectively . Excitation/detection wavelengths used to detect fluorescence were 488/500–550 nm for GFP , 514/535–590 nm for YFP , 543/565–615 nm for mCherry , 543/>560 nm for propidium iodide and 458/>560 nm for autofluorescence . For YFP signal quantification , fluorescence detection parameters ( laser intensity , offset , gain and pinhole settings ) were set so the fluorescence signal emitted by ZIF2 . 2-YFPOX1 root tips was just below the saturation threshold . Twelve micrographs per genotype were captured in the median section of roots from 5-d old seedlings immerged in 50 mM Phosphate Buffer using identical confocal settings to allow comparison between genotypes . Post-imaging , average fluorescence intensity within the whole region ( ca . 20 µm2 ) spanning the extreme root apex to the first elongating cells or the zone immediately above the first elongating cells was recorded . Results are representative of three independent experiments . Total protein was extracted from 5-d old seedlings using 2X Laemmli Extraction Buffer [75] and equal amounts of extracts were resolved on a 8% SDS/polyacrylamide gel before proteins were transferred to a PVDF membrane ( Immobilon-P , Millipore ) , which was incubated with anti-GFP primary antibody ( Roche; 1∶500 dilution ) and then with anti-mouse peroxydase-conjugated secondary antibody ( Amersham Pharmacia; 1∶20000 dilution ) , before membrane-associated peroxydase activity was revealed by ECL . To measure plant zinc concentration , pooled shoot and root tissues from 3-week old seedlings grown on 30 , 125 or 250 µM Zn were processed as previously described [66] . The total Zn concentration of yeast cells was measured as described by Arrivault et al . [10] . The zinc concentration of the digests was quantified using the Atomic Emission Spectrometry – Inductively Coupled Plasma — Optical Emission System ( Perkin-Elmer Optical Emission , Optima 2100 DV ) at the Laboratório de Análises , Instituto Superior Técnico ( Lisbon , Portugal ) according to method 3120B described by Eaton et al . [76] . Zinc standards for analytical calibration were from Merck KGaA . Four independent samples were processed per genotype . Real-time RT-PCR results were analysed using the CFX Manager 3 . 0 software ( Bio-Rad ) and the Q-Gene application [67] . Microscopy images and scanned images of root assays and western blots were processed using LSM 510 software ( Zeiss ) and ImageJ ( http://rsbweb . nih . gov/ij/ ) , respectively . The entire 5′UTR ( plus 200 nt of coding sequence ) alternative structures and their minimum free energies ( MFEs ) were calculated using RNAFold [77] . The structures presented in Figure 10 are the centroid representations calculated with default parameters . | Alternative splicing , which generates multiple messenger RNAs ( mRNAs ) from the same gene , is a key posttranscriptional regulatory mechanism in higher eukaryotes whose functional relevance in plants remains poorly understood . The sequestration of metal ions inside the vacuole of root cells is an important strategy employed by plants to cope with heavy metal toxicity . Here , we describe a new vacuolar membrane transporter of the model plant Arabidopsis thaliana , ZIF2 , that confers tolerance to zinc ( Zn ) by promoting root immobilisation of the metal ion and thus its exclusion from the aerial parts of the plant . The ZIF2 gene is induced by exposure to excess Zn and undergoes alternative splicing , generating two mRNAs that differ solely in their non-coding regions and hence code for the same transporter . Interestingly , toxic Zn levels favour expression of the longer mRNA , which in turn confers higher plant tolerance to the metal . We show that the longer ZIF2 non-coding region markedly promotes translation of the downstream coding sequence into protein in a Zn-responsive fashion . Thus , our results indicate that by regulating translation efficiency of the ZIF2 mRNA , alternative splicing controls the amounts of the encoded membrane transporter and therefore plant Zn tolerance . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biotechnology",
"plant",
"science",
"cell",
"biology",
"gene",
"expression",
"plant",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"plant",
"genetics",
"molecular",
"cell",
"biology",
"plant",
"physiology",
"genetic",
"engineering",
"gene",
"function"
]
| 2014 | Intron Retention in the 5′UTR of the Novel ZIF2 Transporter Enhances Translation to Promote Zinc Tolerance in Arabidopsis |
Control of arbovirus transmission remains focused on vector control through application of insecticides directly to the environment . However , these insecticide applications are often reactive interventions that can be poorly-targeted , inadequate for localized control during outbreaks , and opposed due to environmental and toxicity concerns . In this study , we developed endectocide-treated feed as a systemic endectocide for birds to target blood feeding Culex tarsalis , the primary West Nile virus ( WNV ) bridge vector in the western United States , and conducted preliminary tests on the effects of deploying this feed in the field . In lab tests , ivermectin ( IVM ) was the most effective endectocide tested against Cx . tarsalis and WNV-infection did not influence mosquito mortality from IVM . Chickens and wild Eurasian collared doves exhibited no signs of toxicity when fed solely on bird feed treated with concentrations up to 200 mg IVM/kg of diet , and significantly more Cx . tarsalis that blood fed on these birds died ( greater than 80% mortality ) compared to controls ( less than 25% mortality ) . Mosquito mortality following blood feeding correlated with IVM serum concentrations at the time of blood feeding , which dropped rapidly after the withdrawal of treated feed . Preliminary field testing over one WNV season in Fort Collins , Colorado demonstrated that nearly all birds captured around treated bird feeders had detectable levels of IVM in their blood . However , entomological data showed that WNV transmission was non-significantly reduced around treated bird feeders . With further development , deployment of ivermectin-treated bird feed might be an effective , localized WNV transmission control tool .
West Nile virus ( WNV ) is an arthropod-borne flavivirus , and the leading cause of domestically acquired arboviral disease in the United States [1 , 2] , resulting in significant disease and death every year in humans , domesticated animals , and wildlife . From 1999–2017 , >48 , 000 cases of human WNV disease and >2000 deaths were reported to the CDC [3] , but the total number of individuals in the U . S . who have been made ill from WNV is estimated to be greater than 1 million , or approximately 1 of every 5 persons infected ( >5 million infected individuals ) [4] . Control of WNV transmission remains focused on vector control through larvicide and adulticide applications [5] . Larvicide applications are generally preferred to adulticide applications as they are more cost-effective and less environmentally-damaging due to more direct and efficient targeting of mosquitoes [6 , 7] . While previous studies have demonstrated the effectiveness of larvicide applications to catch basins , a common Culex larval habitat , in reducing the number of mosquitoes [8 , 9] , the efficacy may vary significantly with suboptimal catch basin design or environmental conditions [10 , 11] . Aerial spraying can be costly [12] , but is effective in reducing target mosquito populations [13–16] , and has been linked to reductions in human WNV cases in a treated area relative to an untreated area [15] and in entomological measures of WNV risk [16] . Similar ground ultra-low volume application of adulticides may reduce target mosquito populations under ideal conditions , but studies have provided inconclusive data on their effect on WNV infection rates in mosquitoes or subsequent virus transmission [17–20] . Additionally , off-target effects can occur despite optimal calibration of adulticide applications to host-seeking and active times for target vector species [21–23] . Insecticide applications also often face community opposition due to environmental and toxicity/allergenicity concerns [24–28] and are often restricted to urban and semi-urban communities that can afford to fund them [29 , 30] . WNV is maintained in an enzootic cycle between Culex mosquitoes and avian hosts . The highest WNV disease incidence occurs along the Great Plains region of the United States [31] , as the irrigated agriculture provides a supportive habitat for the main WNV bridge vector of the region , Culex tarsalis [32] . Therefore , blood meals by Cx . tarsalis from often-bitten avian species may be utilized to selectively target adult females through their blood feeding behavior . Given that the majority of Cx . tarsalis blood meals on the northern Colorado plains may come from select species during the WNV transmission season [33] , effective targeting of these preferred hosts with endectocide-treated bird feed could result in control of WNV transmission . Previous studies have assessed the use of systemic endectocides provided to wild animals to control tick vector populations . Pound et al . evaluated ivermectin ( IVM ) -treated corn that was fed to white-tailed deer ( Odocoileus virginianus ) in a treatment pasture to control tick populations [34] . Amblyomma americanum collections from treatment pastures showed a 83 . 4% reduction in adults , 92 . 4% in nymphs , and 100 . 0% in larvae compared to control pastures [34] . IVM-treated feed provided to O . virginianus , which is the definitive host for the reproductive stage of Ixodes scapularis , has also been explored as a method for controlling this vector of Lyme disease . Rand et al . provided an island community of white-tailed deer with IVM-treated corn for 5 consecutive spring and fall seasons [35] . A treatment effect was observed in island deer that reached target IVM sera concentrations resulting in reductions in adult tick density , engorgement , and oviposition rates as well as reduced rates of larval eclosion from any laid eggs compared to collections from untreated deer on a control island [35] . Dolan et al . also conducted a field study that targeted the rodent reservoirs of Lyme disease to reduce the infection prevalence of Borrelia burgdorferi and Anaplasma phagocytophilum with antibiotic-treated bait . Between treated and control areas , they found that B . burgdorferi prevalence was reduced by 87% and A . phagocytophilum by 74% in small mammals , and in questing nymphal ticks , B . burgdorferi prevalence was reduced by 94% and A . phagocytophilum by 92% [36] . A field study testing the passive application of topical acaricide during bait consumption showed reductions of 68% and 84% of nymphal and larval I . scapularis found on white-footed mice , accompanied by a 53% reduction in the B . burgdorferi infection rate of white-footed mice and a 77% decrease in the questing adult I . scapularis abundance between control and treated properties [37] . Rodent baits with feed-through and systemic insecticide activity have also been evaluated to control the phlebotomine sand fly vectors of zoonotic cutaneous leishmaniasis and visceral leishmaniasis . A wide variety of insecticides have been tested for efficacy against multiple phlebotomine sand fly species using larval and adult blood feeding bioassays in multiple rodents . Methoprene , pyriproxyfen , novularon , eprinomectin , ivermectin , and diflubenzuron have been tested for efficacy within the lab [38–41] , while fipronil has been additionally tested in field studies [40 , 42 , 43] . Systemic insecticides have also been used to target plague transmission , where field trials have assessed imidalcloprid-treated bait for controlling flea populations in California ground squirrels ( Spermphilus beechyi ) , black-tailed prairie dogs ( Cynomys ludovicianus ) , and other rodents [44–46] . To our knowledge , this strategy of endectocide-treated baits has not been evaluated in birds for arbovirus control . IVM use in birds is primarily off-label; however , IVM has been administered to treat multiple species of parasites that infest birds , including falcons , cockerels , and chickens [47–50] . Moreno et al . characterized the pharmacokinetics , metabolism , and tissue profiles of IVM in laying hens ( Gallus gallus ) with IVM delivered using intravenous ( IV ) and oral routes [51] . For both IV and oral routes , expected pharmacokinetic profiles and tissue distributions consistent for a highly lipophilic drug were observed [51] . Bennett et al . demonstrated transfer of IVM through crop milk when adult pigeon pairs were given 3 . 3 μg/mL IVM dosed in drinking water and housed with brooding squab , and IVM was subsequently detected in squabs following 3 days of daily adult pigeon IVM dosing [52] . In this present study , we evaluated endectocide-treated bird feed as a systemic endectocide to target Cx . tarsalis . 50% lethal concentrations for selamectin , eprinomectin , and ivermectin were determined in artificial blood meals . IVM-treated bird feed was evaluated for safety and consumption rates in chickens . Mosquitocidal effects in Cx . tarsalis fed on IVM-treated birds were also characterized . Lastly , we present the results of a pilot field trial conducted in Fort Collins , CO in 2017 that examined the safety of IVM-treated bird feed in the field and efficacy on entomological indices of WNV transmission .
Animal research was done under CSU IACUC study protocol 16-6552A . Animal euthanasia was applied using sodium pentobarbital as approved in the IACUC study protocol . Field research was done under Colorado Parks and Wildlife Scientific Collection License #17TRb2104 and Fort Collins Natural Areas Permit #914–2017 . Cx . tarsalis ( Bakersfield colony ) were reared in standard insectary conditions ( 28 ˚C , 16:8 light cycle ) . Approximately 150 larvae were reared in roughly 3 gallons of water and fed 2 . 5 grams of powdered Tetramin fish food daily until pupation . Adults were housed at approximately 300 per cage and fed ad libitum sugar and water until separated for bioassays . Mosquito bioassays were performed to determine the lethal concentrations resulting in 50% mortality ( LC50 ) by adding drug ( eprinomectin , selamectin , and IVM ) into defibrinated calf blood ( Colorado Serum Company ) at serial dilutions for artificial membrane feeding . Following blood feeding , Cx . tarsalis were knocked down with CO2 , and fully-engorged females were collected and held for 5 days in the same insectary conditions . For all bioassays , mosquito mortality was recorded every 24 hours and analyzed using Kaplan-Meier survival curves and compared using Mantel-Cox ( log-rank ) test . LC50 values were calculated using a nonlinear mixed model with probit analysis [53] . Artificial membrane blood feeds were also used to test the effects of IVM and WNV on Cx . tarsalis mortality . The WNV strain used was a 2012 Colorado isolate propagated in Vero cells . Negative controls were DMEM ( Dulbecco’s Modified Eagle Media ) and DMSO ( dimethyl sulfoxide ) at the same volumes as WNV and IVM , respectively . For the concurrent blood feed of WNV and IVM , IVM at 73 . 66 ng/mL ( LC75 ) and WNV at low titer ( 5x105 PFU/mL ) or high titer ( 107 PFU/mL ) were fed in a membrane blood meal to Cx . tarsalis and mortality was observed as described above . For the WNV-exposure followed by an IVM blood feed , mosquitoes were fed a first blood meal containing 107 PFU/mL of WNV or DMEM for a mock-exposure . Fully engorged females were sorted and held for 10 days , then fed a second blood meal containing 73 . 66 ng/mL IVM , after which fully blood fed females were sorted and mortality observed . 4–6 weeks old white leghorn chickens were divided into groups ( n = 4 ) that were housed separately , and which were provided clean water daily and control ( untreated ) diet consisting of a cracked corn mix ( Chick Start and Grow , Northern Colorado Feeders Supply ) mixed with any additives that were also added to IVM-treated diet for 3 or 7 consecutive days . IVM-treated diet consisted of two formulations: an Ivomec formulation where liquid Ivomec ( Merial ) was mixed directly into the cracked corn mix and a powder IVM formulation where powder IVM ( Sigma-Aldrich ) was mixed into all-purpose flour at 5% and then added to the cracked corn mixture to aid in even powder distribution . Chickens were fed ad libitum and feed consumed by each group was measured daily . Chickens were weighed daily and observed for clinical signs of toxicity , including diarrhea , mydriasis , ptosis , stupor and ataxia . The amount of chicken feed consumed was compared between groups using the students t-test and chicken growth rates were compared using linear regression . Blood was collected from these chickens through venipuncture at the end of their IVM diet regimen and for two days following IVM diet withdrawal . Serum was then isolated from the blood samples and stored at -80°C until further analysis . Eurasian collared doves ( Streptopelia decaocto ) were captured by mist net in Wellington , CO and brought back to CSU . They were housed in groups of three and provided ad libitum clean water and either control diet or powder IVM formulation diet of 200 mg IVM/kg of diet for 10 days . Three doves were fed each control and powder IVM formulation diet and then used for mosquito bioassays . Mosquito bioassays following blood feeding on birds were conducted on the last day of the IVM diet regimen for each group and for two days following IVM diet removal . For direct blood feeding on birds , the downy breast feathers were trimmed , and the exposed bird breast was placed on top of the mosquito cage . The birds were gently restrained for 30 minutes while the mosquitoes blood fed through the mosquito cage organdy . Given the difficulties of direct mosquito blood feeding on live chickens , supplemental serum-replacement membrane blood feeds were also performed , where frozen chicken serum was used in reconstituted blood meals using red blood cells from defibrinated calf blood [54 , 55] . All research with animals was reviewed and conducted under authorization by the Colorado State University Institutional Animal Care and Use Committee , protocol 16-6552A . Colorado State University Animal Care and Use is Public Health Service ( PHS ) and Office for Laboratory Animal Welfare ( OLAW ) assured ( #A3572-01 ) , United States Department of Agriculture ( USDA ) registered ( #84-R-0003 ) , and Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) accredited ( #000834 ) . All chemicals used in derivatization were HPLC grade and purchased from Sigma-Aldrich . IVM was extracted from serum following methanol precipitation [56] . 400 μL of methanol was added to 100 μL serum and vortexed for 1 . 5 min . Methanol precipitation was carried out at -80˚C overnight . Samples were centrifuged for 30 min at 16 , 000 x g . Supernatants were transferred and evaporated to dryness using a Speedvac concentrator ( Savant ) . The dry residue was dissolved in 20 μL acetonitrile . Samples were derivatized according to previously published literature [57] . A Waters 700 autosampler system was used to quantify IVM by high-performance liquid chromatography ( HPLC ) -fluorescence . A mobile phase of acetonitrile/water ( 3:1 , v/v ) was pumped through a C8 column ( Waters , XBridge BEH C8 XP , 130 Å , 2 . 5 μm , 2 . 1x100 mm ) at a rate of 0 . 45 mL/min . Excitation and emission spectra were 365 and 470 nm , respectively . 10 μL of derivatized sample was injected by the autosampler . Precision was quantified as coefficient of variation ( %CV ) . This was calculated interday and intraday , evaluating drug-free chicken serum samples ( n = 5 ) spiked with IVM at 25 , 50 , and 100 ng/mL . Instrument CV was 6 . 11% . Intraday CV ranged between 4 . 36 and 9 . 77% . Interday reproducibility was 15 . 39% . Retention time CV was 1 . 77% . The method was linear across the range of concentrations tested in the standard curve ( 3 . 125–100 ng/mL ) . Linear regression curves containing fortified IVM serum samples with concentrations of 3 . 125 , 6 . 25 , 12 . 5 , 25 , 50 , and 100 ng/mL had a R-square value of 0 . 9974 . Limits of detection and quantification were 1 . 56 ng/mL and 3 . 125 ng/mL , respectively . For the 2017 pilot trial , field sites were located in urban and suburban areas in the City of Fort Collins ( mainly in city open space areas and near water sources ) that were weekly mosquito trapping sites used by the city for WNV surveillance efforts and have been maintained since 2006 ( S1 Fig ) . Six field sites were chosen based on historical WNV surveillance data from all city trapping sites as those having the highest number of WNV-positive Cx . tarsalis pools since 2006 , while excluding trap sites in neighborhoods that are regularly treated with adulticides or used as sentinel sites for the Colorado WNV surveillance system by the state department of health . The 6 chosen sites were all in east Fort Collins and were randomly placed into the treatment group ( 3 sites; mosquito traps surrounded by IVM-treated bird feed stations ) or the control group ( 3 sites; mosquito traps surrounded by control un-treated bird feed stations ) . At each field site , an array of three bird feed stations was placed in an approximate triangular perimeter around the mosquito trap at a distance of 50 m ( S1 Fig ) . IVM-treated bird feed was used at a concentration of 200 mg/kg of diet , and the diet was a mixture of white proso millet , cracked corn , and flour ( 47 . 5:47 . 5:5 , v/v/v ) . IVM-treated bird feed was changed daily to account for any effects of IVM degradation due to exposure , which also allowed for daily monitoring of any obvious adverse effects of IVM in local fauna . Motion-activated trail cameras were used to document bird visits to feeders , with each field site having a motion-activated trail camera placed at one of the three feeders . Photos were screened using the Sibley Guide to Birds [58] . Due to an overabundance of pictures , only a random sampling of 6 days from the 2017 season were counted . Bird trapping and sampling of their blood was performed at two IVM sites . Birds were caught using mist nets placed approximately 10 m from an IVM-treated bird feeder . Blood was collected from netted birds using jugular venipuncture and placed into serum separator tubes . Bird sera were analyzed using HPLC-fluorescence and a subset of samples was analyzed using LC-MS . Because 200 μL of blood could not be drawn from the sparrows as needed for HPLC-fluorescence quantification , IVM analysis for sparrows was only documented as presence or absence . Control sera from house sparrows caught in spring 2014 were used as negative controls . Serum from one IVM-positive grackle was also used in a serum-replacement blood feed with colony Cx . tarsalis for a mosquito survival bioassay . Mosquitoes were processed as part of the Fort Collins WNV surveillance program according to established protocols [59] . Briefly , mosquitoes were collected weekly by Vector Disease Control International using miniature CDC light traps baited with CO2 . Mosquitoes were sorted to species and pooled into groups of typically no more than 50 . Mosquito pools were screened at CSU using qRT-PCR using the following primer sequences: forward 5’ 1160-TCAGCGATCTCTCCACCAAAG 3’ , reverse 5’ 1209-GGGTCAGCACGTTTGTCATTG 3’ , probe 5’ FAM-1186-TGCCCGACCATGGGAGAAGCTC 3’ [59] . Bird sampling was done under Colorado Parks and Wildlife Scientific Collection License #17TRb2104 and Fort Collins Natural Areas Permit #914–2017 . Chicken feed consumption was compared between groups using a t-test . Linear regression was done on chicken weights and the rate of weight gain was compared using Analysis of Covariance . For mosquito bioassays , survival was analyzed using Kaplan-Meier survival curves and compared using Mantel-Cox ( log-rank ) test . LC50 values were calculated using a nonlinear mixed model with probit analysis [53] . IVM sera concentrations from chickens were compared using ANOVA . IVM sera concentrations from individual chickens were correlated to cumulative mosquito morality from bioassays conducted on the respective chickens using Spearman correlation . The field trial utilized control and treatment sites located in the City of Fort Collins; however , it was an exploratory trial to test a new trial design and sites , and so was not powered for detecting differences in Cx . tarsalis abundance and WNV infection . Cx . tarsalis abundance from control and treatment sites were compared against each other using a generalized linear mixed model with negative binomial distribution that included site , week of trapping , and treatment . Cx . tarsalis abundance was also shown in comparison to historical data from 2006–2016 ( which lacked any bird feed stations surrounding the traps ) . WNV infection rate was calculated as maximum likelihood estimate ( MLE ) using the Excel PooledInfRate Add In [60] , but Fisher’s exact test was again used to compare the total number of WNV-positive and WNV-negative pools between control and treatment sites . Statistical analyses were done in GraphPad Prism ( Version 7 ) and R ( Version 3 . 3 . 1 ) .
Mosquitocidal concentrations of IVM , selamectin , and eprinomectin were determined with mosquito bioassays following blood feeds with serially diluted drug ( S2 Fig ) . IVM had the lowest LC50 concentration at 49 . 94 ng/mL ( Table 1 ) as compared to eprinomectin with a LC50 of 101 . 59 ng/mL and selamectin with a LC50 of 151 . 46 ng/mL . With the lowest effective concentrations , ivermectin was chosen for further characterization in birds . Potential interactions of IVM and WNV on Cx . tarsalis mortality were assessed in a simultaneous blood meal containing IVM ( LC75 ) and WNV . Feeding with IVM only resulted in significantly increased mortality compared to DMSO controls; however , the observed 41% and 83% mortality for IVM control groups ( Fig 1A and 1B ) reflect the variability of mosquito bioassays , especially for intermediate ranges of lethal concentrations . WNV ( both low and high titer ) exposure in the absence of IVM did not affect Cx . tarsalis mortality over 5 days immediately after the blood meal ( Fig 1A and Fig 1B ) , or following a second untreated blood meal 10 days later ( Fig 1C ) . On the other hand , Cx . tarsalis given a concurrent blood meal containing low-titer WNV and IVM exhibited significantly increased mortality at 51% compared to the control IVM group not fed WNV with 41% morality ( p = 0 . 0268 , χ2 = 4 . 904 ) ( Fig 1A ) . However , there was no significant difference ( p = 0 . 2529 , χ2 = 1 . 307 ) in mortality between Cx . tarsalis fed a concurrent blood meal containing high titer WNV and IVM compared to the control ( Fig 1B ) . Similarly , Cx . tarsalis given a first blood meal of either DMEM control or high titer WNV , and then a second blood meal containing IVM 10 days later , showed no significant differences in mortality ( p = 0 . 1637 , χ2 = 1 . 940 ) ( Fig 1C ) . Over 7 days of observation , there were no observable clinical signs of IVM neurotoxicity—diarrhea , mydriasis , ptosis , stupor , and ataxia–in groups that consumed either liquid Ivomec or powder formulations of IVM of 200 mg IVM/kg of diet . For the Ivomec formulation diet , the chickens consumed an average 59 . 3 g of feed per chicken daily . This was significantly less than the corresponding control group which averaged 121 . 6 g of feed per chicken per day ( p = 0 . 0045 , t = 3 . 490 ) . Consequently , there was also a significant difference ( p <0 . 0001 , F = 19 . 45 ) in the rate of weight gain between Ivomec and control groups ( S3A Fig ) . For the powder IVM formulation diet , the IVM group consumed 60 . 97 g of feed per chicken each day , which was not significantly different from daily control group consumption of 55 . 2 g of feed per chicken ( p = 0 . 2928 , t = 1 . 100 ) . This was also reflected in similar rates of weight gain between powder IVM and control groups ( p = 0 . 0680 , F = 4 . 022 ) ( S3B Fig ) . Cx . tarsalis mortality following blood feeding on IVM-treated chickens increased as IVM concentration within the diet increased ( S4 Fig ) . There were significant differences in mosquito mortality following blood feeding on chickens given 50 mg IVM/kg of diet ( p = 0 . 0132 , χ2 = 6 . 146 ) and 100 mg IVM/kg of diet ( <0 . 0001 , χ2 = 86 . 48 ) . However , the largest increase in mortality ( p<0 . 0001 , χ2 = 461 . 1 ) following blood feeding was at 200 mg IVM/kg of diet with 95 . 2% mortality in mosquitoes fed on IVM-treated chickens and 2 . 7% mortality in mosquitoes fed on control chickens . All subsequent experiments used IVM-treated feed at 200 mg IVM/kg of diet . For the Ivomec formulation at 200 mg IVM/kg of diet , there was a significantly increased mortality in mosquitoes blood fed on chickens consuming Ivomec-diet for either 3 or 7 days as compared to mosquitoes blood fed on control chickens ( Fig 2; left and right panels , respectively ) . On the last day of Ivomec feed administered , for both 3 or 7 days , there was a significant increase ( p<0 . 0001 , χ2 = 80 . 22 and χ2 = 76 . 41 , respectively ) in mortality between mosquitoes blood fed on chickens consuming an Ivomec diet with upwards of 80% mortality as compared to mosquitoes blood fed on control chickens with less than 40% mortality ( Fig 2A and 2B ) . This difference in mosquito mortality between treatment and controls decreased when the blood feed occurred 1 day following the withdrawal of the Ivomec diet in the treatment group ( Fig 2C and 2D ) . After 2 days following Ivomec diet withdrawal , there was no significant difference in mosquito mortality between those blood fed on Ivomec-consuming chickens as compared to mosquitoes blood fed on control chickens in the 3 day group , but there was a significant difference in the 7 day IVM group ( p = 0 . 0117 , χ2 = 6 . 354 ) which is likely due to the variability in mosquito bioassays ( Fig 2E and 2F ) . In addition , the time administered Ivomec-treated diets ( 3 vs . 7 days ) did not affect mosquito survival curves following direct blood feeding on chickens , regardless if the mosquitoes were blood fed on the last day of chicken time on the diets , or if the chickens were 1 or 2 days post withdrawal of the diets ( Fig 2 , left vs . right panels ) . There was also significantly increased mosquito mortality in mosquitoes blood fed on chickens consuming the powder formulation of IVM ( 200 mg IVM/kg of diet ) compared to mosquitoes fed on control chickens ( S5 Fig ) . Because bioassays from the Ivomec formulation and a preliminary powder formulation indicated no differences between mosquitocidal effects for groups given IVM for 3 or 7 days , these and subsequent experiments focused on the 7 day time point . A direct blood feed of mosquitoes on chickens given a powder IVM diet for 7 days resulted in 92 . 3% mosquito mortality as compared to 25 . 7% mosquito mortality from those blood fed on control chickens ( p<0 . 0001 , χ2 = 41 . 23 ) ( S5A Fig ) , while an indirect , serum-replacement blood feed using sera from chickens given a powder IVM diet for 7 days resulted in 79 . 0% mosquito mortality as compared to 16 . 7% mortality from those blood fed on control chicken serum ( p<0 . 0001 , χ2 = 42 . 83 ) ( S5B Fig ) . Furthermore , the mosquito survival curves between those blood fed directly on IVM-treated chickens as compared to sera from IVM-treated chickens were significantly different ( red lines in S5A Fig vs . S5B; p<0 . 0001; hazard ratio 2 . 007 ) . At 1 day post-powder IVM diet withdrawal , there was still a significant difference ( p = 0 . 001 , χ2 = 10 . 86 ) in mosquito mortality between those directly blood fed on IVM-diet vs control-diet chickens ( S5C Fig; 90 . 9% vs . 0% mortality ) . However , this mosquitocidal effect was not apparent in a serum-replacement blood feed derived from chicken blood taken 1 day after IVM diet withdrawal ( p = 0 . 7445 , χ2 = 0 . 1062 ) ( S5D Fig ) . As above , the mosquito survival curves between those blood fed directly vs . indirectly on treated chickens 1 day post-diet withdrawal were also significantly different ( red lines in S5C Fig vs . S5D; p<0 . 0001; hazard ratio 6 . 742 ) . At 2 days post-IVM diet withdrawal , blood/serum from treated chickens was no longer mosquitocidal in either direct blood feeding ( p = 0 . 8402 , χ2 = 0 . 04065 ) or serum-replacement ( p = 0 . 1792 , χ2 = 1 . 804 ) assays ( S5E and S5F Fig ) . Direct blood feeds of Cx . tarsalis were also conducted on six wild caught Eurasian Collared Doves fed either a powder IVM formulation diet of 200 mg IVM/kg or control diet in the laboratory ( Fig 3 ) . There was a significant difference in mosquito mortality ( p<0 . 0001 , χ2 = 60 . 34 ) with 88 . 5% mortality in Cx . tarsalis fed on IVM-treated doves as compared to 14 . 3% mortality from mosquitoes blood fed on control doves . Additionally , there were no clinical signs of IVM toxicity observed in this treated bird species . Neither the IVM formulation nor the time for which the chickens consumed IVM-treated diet resulted in significant differences in average IVM serum concentrations ( p = 0 . 2715 , F = 1 . 472 ) ( Fig 4A , blue vs . green bars ) . On the last day of IVM diet , the average IVM serum concentrations ( with SD ) were 88 . 575 ( ±43 . 613 ) ng/mL for 3-day Ivomec , 45 . 255 ( ±70 . 051 ) ng/mL for 3-day powder IVM , 21 . 910 ( ±20 . 914 ) ng/mL for 7-day Ivomec , 45 . 745 ( ±33 . 852 ) ng/mL for 7-day powder IVM . Chicken IVM serum concentrations decreased following withdrawal of the IVM diet and were nearly undetectable at 2 days post-withdrawal , which corresponded with mosquito bioassay results showing decreases in mosquitocidal activity following IVM-diet removal . Additionally , IVM serum concentrations were correlated to resulting mosquito mortality from blood feeding on these corresponding IVM-powder fed chickens ( Fig 4B ) . There was a higher correlation between IVM serum concentrations and mortality from serum-replacement feeds with a Spearman r of 0 . 8629 ( P = 0 . 0007 ) , while the correlation between IVM serum concentrations and mortality from direct blood feeds was 0 . 4153 ( p = 0 . 3062 ) . For a pilot trial testing IVM feed in a natural transmission cycle , feeder stations were placed in urban and suburban areas within the City of Fort Collins ( S1 Fig ) and randomized to treatment or control sites . Bird visits to IVM feeders at all sites were dominated by grackles with infrequent visits by house ( Passer domesticus ) and sagebrush sparrows ( Artemisiospiza nevadensis ) and black-capped chickadees ( Poecile atricapillus ) ( Table 2 ) . There were also two visits by blue jays ( Cyanocitta cristata ) , and a few other birds which could not be identified from the photographs . A more homogenous mix of grackles , house and brewers ( Spizella breweri ) sparrows , blue jays , black-capped chickadees , bushtits , and squirrels visited control feeders ( Table 2 ) . Birds were also caught by mist net and their sera assayed for IVM at the end of the field season . Ten grackles and 5 sparrows were caught over 4 mornings of sampling on August 30th and September 2nd , 3rd , and 7th . Most birds had been observed feeding from the IVM-treated feeder immediately preceding mist net capture . Nine grackles and 4 sparrows ( 87% of tested sera ) had detectable levels of IVM within their serum , and the negative control sparrow serum from 2014 had no detectable IVM ( Table 3 ) . Serum from grackle #5 ( Table 3 ) was plentiful and thus further used in a LC-MS assay to confirm the presence of IVM , and also tested in a serum-replacement bioassay . Interestingly , even though the IVM serum concentration in grackle #5 was measured as 5 . 7 ng/mL , there was strong mosquitocidal effect from this serum ( 100% mortality within 2 days; p<0 . 0001 , χ2 = 54 . 15 ) compared to control mosquitoes fed on control calf serum ( Fig 5 ) . Cx . tarsalis abundance over time in 2017 at the urban and suburban field sites was similar to historical data collected from the same traps for 10 years prior ( Fig 6A ) . A generalized linear mixed model with negative binomial distribution did not find a significant difference between Cx . tarsalis abundance at IVM sites compared to control sites ( p = 0 . 161 , z = 1 . 401 ) ( Fig 6B ) . The low number of WNV infections did not allow for robust statistical analysis , although MLE was calculated ( Fig 6C ) . A combined Fisher’s Exact Test of all 6 field sites showed a non-significant decrease in the proportion of WNV-positive pools to WNV-negative pools among control and treatment traps ( p = 0 . 2081 ) ( Fig 6D ) .
This study presents a novel characterization of IVM-treated bird feed as a systemic endectocide to control WNV transmission . Lab studies characterized the effects of IVM-treated bird feed in both domestic and wild birds , especially mosquitocidal effects in Cx . tarsalis blood fed on birds consuming this IVM-containing diet . In addition , a pilot field trial was performed over a WNV season to gather preliminary efficacy data on the effects of IVM-treated bird feed within a natural WNV transmission cycle between wild birds and mosquitoes . IVM was determined to be the most effective endectocide tested with the lowest lethal concentrations for Cx . tarsalis . In addition , there did not appear to be a synergistic effect of IVM and WNV on Cx . tarsalis mortality in either a simultaneous blood feed of IVM and high titer WNV or sequential blood feeds , the first containing WNV and the second containing IVM . There was a statistical difference between survival curves of Cx . tarsalis fed a concurrent blood meal of a low WNV titer IVM compared to Cx . tarsalis fed only IVM . However , this increased mortality was likely due to the variable survival response of mosquitoes to IVM particularly at intermediate lethal concentrations , rather than a biologically significant interaction between WNV and IVM as there was no mortality difference between mosquitoes fed a concurrent higher titer WNV+IVM blood meal compared to mosquitoes fed DMEM+IVM . There was also no difference between mosquitoes previously exposed to WNV and then fed IVM as compared to mosquitoes unexposed to WNV and then fed IVM . While there is a study suggesting that IVM can inhibit WNV replication by targeting NS3 helicase activity , this was an in vitro cell-culture study using mammalian cells , and the concentration of IVM needed to inhibit 50% of the RNA synthesis in the Vero cells infected with WNV was considerably higher than what was achieved in our chickens following IVM feed consumption [61] . No clinical signs of toxicity were observed in any of the birds consuming either formulation of IVM feed . This was not surprising as IVM is given therapeutically in bird species in a wide range of doses ( 0 . 2 mg/kg to 2 mg/kg ) , depending on route of administration . However , more detailed studies of IVM toxicity should be conducted in multiple bird species in future controlled experiments . Previous studies have identified neurotoxic effects in pigeons following long-term consumption of a diet containing avermectin [62 , 63] , of which IVM is a safer derivative [64] . Specifically , Chen et al . observed clinical signs of neurotoxicity , ranging from reduced activity and food intake following avermectin consumption for 60 days on a 20 mg/kg diet , to ataxia and spasms following avermectin consumption for 30 days on a 60 mg/kg diet [63] . On the other hand , a characterization of IVM pharmacokinetics , metabolism , and tissue distribution in laying hens treated intravenously ( 400 μg/kg ) or consuming IVM-treated water ( 400 μg/kg/day ) for 5 days did not report any ill effects in the birds [51] . Following the intravenous injection of the hens , the highest IVM plasma concentrations ( 739 . 6 ± 50 . 2 ng/mL ) were 30 minutes after administration and plasma concentrations remained below 10 ng/mL after 24 hours [51] . Mean IVM concentrations in our chickens fed exclusively on an IVM-containing diet for 3 and 7 days were approximately 45 ng/mL , and similarly we did not observe any neurotoxicity . It remains to be determined if these results vary among different bird species or longer times on the diet . However , in the field studies , it is unlikely that the IVM-treated bird feed was the sole or even primary source of food for the wild birds visiting the feeders given the abundance of alternative food sources during summer . While chickens on the powder IVM and control diets consumed equivalent quantities of food , there was a significant difference in feed consumption among chicken fed the Ivomec diet and their controls . This may be a result of the glycerol formal and propylene glycol carriers in Ivomec that could give an unpleasant taste , as propylene glycol has been identified as a unpleasant and unpalatable feed additive in cattle [65] . Consequently , the decreased Ivomec feed consumption relative to control feed consumption is likely responsible for the significantly reduced rate of weight gain in the Ivomec group as compared to controls . Chickens that consumed either a powder IVM or Ivomec diet reached mosquitocidal levels of IVM in their blood within 3 days , as demonstrated by both the IVM serum concentrations in the chickens as well as the significant difference in survival curves of mosquitoes blood fed on IVM-treated chickens compared to controls . There were no notable differences between either IVM diet formulations in mosquitocidal efficacy when considering either time to achieve a mosquitocidal effect and IVM persistence in chicken serum following IVM withdrawal . Furthermore , the time the chickens were placed on the two IVM diets ( 3 and 7 days ) did not significantly affect mosquito mortality , serum concentrations , or the elimination time of IVM from serum following feed withdrawal . This is corroborated by the similar IVM serum concentrations at all time points among the different IVM administration times and formulations . A mosquitocidal effect , but no observable bird toxicity , was demonstrated for wild-caught Eurasian collared doves following consumption of the 200 mg IVM/kg diet , indicating similar mosquitocidal efficacy of the approach in one other bird species and thus potential application to other wild bird species in field settings . The mosquito mortality in control groups had a greater variation for direct blood feeds ( 17 . 75% CV ) relative to control groups for serum-replacement blood feeds ( 3 . 57% CV ) , indicating that direct blood feeds results in more inherent variability in mosquito mortality . This increased variability could be a result of increased mosquito handling and rougher conditions during direct blood feeding on birds . It is also possible this higher variability is partly due to smaller sample sizes from the direct blood feeds due to the low success of our colony mosquitoes imbibing full blood meals from live chickens . Regardless , the higher variability among direct blood feed data led to a weaker correlation between IVM serum concentrations and mosquito mortality compared to that from serum-replacement blood feed data . However , despite this higher variability , cumulative mosquito mortality from these direct blood feeds was higher ( consistently above 75% ) compared to that from the serum-replacement feeds , and mostly independent of measured IVM concentration in the chickens’ sera . One likely possibility for this discrepancy is that the IVM concentration within serum extracted from venous blood may not always be an accurate representation of the IVM concentration in subdermal capillary blood on which mosquitoes blood feed . It has been previously proposed that because IVM is extremely lipophilic and sequestered in fatty tissues , there may exist a concentration gradient of higher IVM or IVM metabolite concentrations in adipose tissue and blood of the surrounding capillaries compared with venous blood [66] . This is also one explanation for the observation that the IVM serum concentrations in chickens correlated with higher cumulative mosquito mortality than would be predicted from the LCx values calculated using artificial membrane feeds . A useful future analysis would be to compare mosquito mortality results from direct skin blood feeding on chickens , membrane blood feeds using venous blood drawn from the chickens , and serum replacement blood feeds using unfrozen serum from the same chickens . The mosquitocidal effect from chickens on an IVM-containing diet did not extend past one day after IVM-feed withdrawal , and this corresponded with the IVM serum concentrations that were generally below detectable limits by two days post-IVM feed withdrawal . This could potentially be a concern for applying this strategy in the field as it would suggest that frequent bird visits would be necessary to maintain their mosquitocidal blood concentrations of IVM . However , our field data indicated that wild birds were visiting the bird feeders and did have detectable levels of IVM within their sera during multiple days throughout the trial . In addition , one grackle from our 2017 field trial had strongly mosquitocidal serum as assessed in a bioassay , even though the IVM concentration in that serum was surprisingly low . It is promising that a majority of the birds tested had detectable levels of IVM within their sera , indicating that there was an unexpectedly high coverage of IVM in captured birds . However , the placement of mist nets at roughly a 10 m distance from an IVM feeder may have biased the sampling towards birds that visited the feeder , so future studies should more intensively sample birds at wider radii from the feeders . Understanding IVM coverage and persistence within wild birds is an important component of determining the efficacy of this strategy and should be supplemented with detection of IVM in wild-caught blood fed Cx . tarsalis in future field seasons . This could also be coupled with mosquito survival bioassays using wild bird sera to assess mosquitocidal activity as we performed here . This use of IVM-treated feed as a systemic endectocide to control WNV transmission is based on targeting Cx . tarsalis by medicating its preferred host species . Previous studies in California implicate Cx . tarsalis as a regionally adaptive , opportunistic blood feeder with a preference for avian hosts , and the diversity of available blood meal sources is reflected in the composition of its blood meals [67–71] . Important avian hosts for Cx . tarsalis in small rural towns within Weld County , which is adjacent to our Fort Collins field site area , include American Robins , doves , and other Passeriformes [33] . American Robins are an important Cx . tarsalis blood meal source and WNV amplification host that does not frequent bird feeders and would not be targeted by this current strategy [33 , 72 , 73] . However , doves and passerines are preferred blood meal sources of Cx . tarsalis and contribute to the cumulative number of WNV-positive Cx . tarsalis at estimated rates of approximately 30% in June , 60% in July , and 85% in August [33] . This represents a large proportion of Cx . tarsalis blood meal sources and WNV-positive contributions from birds that consume grain and seed that could be targeted throughout the summer season . However , our trail camera data did not show a large proportion of visits from these species identified as regionally important . For example , grackles were predominantly visiting our IVM-treated feeders , while control feeders were visited mostly by grackles , blue jays , brewer’s sparrows , and squirrels . However , the single trail camera we employed per site may not have fully documented bird visits to other feeders at the field site . Camera placement was limited to tree-filled areas where a feeder could be placed with a camera locked to a tree across from the feeder , and this may have biased the camera data against bird species that feed in open space or brush rather than among trees . This limitation of the field camera data is illustrated by our detection of IVM in house sparrows caught by mist net , but we had no documentation of sparrow visits on the trail camera for this specific field site . An important future direction will also be to gather a more updated understanding of the Cx . tarsalis blood meal sources within urban and suburban area of the City of Fort Collins , which might allow for specific targeting of these bird species with attractive bird feed compositions and an optimized bird feeder design . In addition to a better characterization of avian blood meal sources for Cx . tarsalis , a more complete understanding of bird and Cx . tarsalis spatial dynamics is also important for determining the best placement for the IVM-treated feeders . Because our field sites were chosen based on historical mosquito and WNV surveillance , we did not account for crucial bird parameters that may have influenced mosquito sampling . For example , birds may have fed at the IVM-treated feeders and returned to their communal roosts where they would have been blood fed on by Cx . tarsalis [33 , 70 , 74] , representing a treatment effect in a different population of Cx . tarsalis than sampled at our traps . Accounting for these bird-mosquito spatial dynamics by placing IVM-treated feeders near communal roosts of granivorous birds and sampling mosquitoes within close range may show the greatest entomological treatment effect , especially as Kent et al . gives an example of a house sparrow roost serving as both a major blood meal and amplification source of WNV-positive Cx . tarsalis [33] . While communal bird roosts could present a critical target , this strategy should continue to be tested in areas of increased human use such as parks and backyards . This highlights that future studies should also consider the best placement of bird feeders in the context of both human land use , and bird and mosquito interactions . Our pilot field trial was ultimately inconclusive and did not find a significant difference in Cx . tarsalis abundance or WNV infection due to IVM treatment . This is likely due to three field sites for each trial arm being underpowered to observe a significant effect . However , these preliminary field data will serve as important effect size variables with which to properly power future field trials . In addition , this strategy of controlling vector pathogen transmission with an endectocide like IVM is based on shifting the mosquito population age structure in a treatment area from older , infectious mosquitoes to younger , non-infectious mosquitoes , and is less dependent on reducing total mosquito abundance . This has been modeled , as well as observed with empirical data , in trials testing IVM for malaria transmission control [75 , 76] . We would also expect to see a shift in the age structure of the population to fewer older , infectious Cx . tarsalis and more uninfected , younger mosquitoes . However , our preliminary results from ovary dissections and parity scoring according to Detinova [77] showed consistently high parous rates within the field-caught Cx . tarsalis . This suggested that autogeny , or the ability to develop a batch of eggs without imbibing a blood meal , could be present among the Cx . tarsalis in our study area and confounded our data , and we chose to not conduct further parity scoring during our pilot field trial . As determining age structure of the wild Cx . tarsalis population would be additional way to evaluate this control strategy , future studies should integrate other age-grading techniques such as near infrared spectroscopy ( NIRS ) [78 , 79] . Our characterization of IVM as a systemic endectocide in birds demonstrates its feasibility to be developed into a novel WNV transmission control tool . We have demonstrated that birds readily consume IVM-treated feed in the lab and field with our formulation and concentration , while not displaying any observable clinical signs of toxicity following consumption . Furthermore , Cx . tarsalis mosquitoes blood feed on these IVM-treated birds and often die as a result . Our pilot field trial testing IVM-treated feed in natural transmission cycles within wild birds and mosquitoes was ultimately inconclusive , but did provide critical effect size variables to inform future trial design . Important future directions will be to optimize treated bird feed formulations for the field and better characterize the pharmacokinetics and pharmacodynamics of this diet within multiple bird species , especially in relation to mosquitocidal activity and physiological/clinical signs of toxicity . In addition , a more-updated , regionally-specific understanding of the blood meal host preferences of Cx . tarsalis across urban , suburban and rural habitats would allow for better targeting of these preferred host species through the design of an attractive bird feed composition , discriminating bird feeders , and optimized bird feeder location for application to different geographic areas . Finally , our field study provides an important template for future field studies across multiple WNV seasons that will be adequately-powered for measuring effect sizes in entomological and other outcomes . | West Nile virus ( WNV ) is a mosquito-borne virus that causes significant disease and death every year in humans , domesticated animals , and wildlife . Control of WNV transmission is focused on controlling the mosquito vector through applications of insecticides directly to the environment . In this study , we evaluate a novel control strategy for WNV transmission by targeting the main mosquito bridge vector in the Great Plains region , Culex tarsalis , through its blood feeding behavior . Because Culex tarsalis favor taking blood meals from particular bird species , our strategy aims to target these bird species with endectocide-treated bird feed that will result in lethal blood meals for Cx . tarsalis . In this study , we developed a safe and effective formulation of ivermectin-treated diet that resulted in increased mortality for Cx . tarsalis blood fed on birds consuming this treated diet as compared to mosquitoes feeding on control birds . We also conducted a pilot field trial in Fort Collins , Colorado to test this strategy in a natural transmission cycle , which demonstrated promising results . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"death",
"rates",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"vertebrates",
"diet",
"animals",
"viruses",
"nutrition",
"rna",
"viruses",
"population",
"biology",
"insect",
"vectors",
"infectious",
"diseases",
"birds",
"medical",
"microbiology",
"microbial",
"pathogens",
"gamefowl",
"fowl",
"disease",
"vectors",
"insects",
"poultry",
"arthropoda",
"community",
"ecology",
"population",
"metrics",
"mosquitoes",
"eukaryota",
"blood",
"anatomy",
"flaviviruses",
"west",
"nile",
"virus",
"ecology",
"physiology",
"viral",
"pathogens",
"trophic",
"interactions",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"chickens",
"amniotes",
"organisms"
]
| 2019 | Evaluation of a novel West Nile virus transmission control strategy that targets Culex tarsalis with endectocide-containing blood meals |
Antimicrobial resistance ( AMR ) is currently one of the most important challenges to the treatment of bacterial infections . A critical issue to combat AMR is to restrict its spread . In several instances , bacterial plasmids are involved in the global spread of AMR . Plasmids belonging to the incompatibility group ( Inc ) HI are widespread in Enterobacteriaceae and most of them express multiple antibiotic resistance determinants . They play a relevant role in the recent spread of colistin resistance . We present in this report novel findings regarding IncHI plasmid conjugation . Conjugative transfer in liquid medium of an IncHI plasmid requires expression of a plasmid-encoded , large-molecular-mass protein that contains an Ig-like domain . The protein , termed RSP , is encoded by a gene ( ORF R0009 ) that maps in the Tra2 region of the IncHI1 R27 plasmid . The RSP protein is exported outside the cell by using the plasmid-encoded type IV secretion system that is also used for its transmission to new cells . Expression of the protein reduces cell motility and enables plasmid conjugation . Flagella are one of the cellular targets of the RSP protein . The RSP protein is required for a high rate of plasmid transfer in both flagellated and nonflagellated Salmonella cells . This effect suggests that RSP interacts with other cellular structures as well as with flagella . These unidentified interactions must facilitate mating pair formation and , hence , facilitate IncHI plasmid conjugation . Due to its location on the outer surfaces of the bacterial cell , targeting the RSP protein could be a means of controlling IncHI plasmid conjugation in natural environments or of combatting infections caused by AMR enterobacteria that harbor IncHI plasmids .
Infectious diseases , despite the availability of antibiotics , remain an important public health issue , representing the second leading cause of death worldwide [1] . Antimicrobial resistance ( AMR ) is in several instances underlying the evolution of fatal bacterial infections . The gradual increase in resistance rates of several important pathogens represents a serious threat to public health [2–4] . The dissemination of antibiotic resistance in gram-negative bacteria has been largely attributed to the acquisition of plasmid-located antibiotic resistance genes [5–7] by horizontal gene transfer ( HGT ) . Plasmids belonging to the incompatibility group ( Inc ) HI include mainly genetic elements encoding AMR determinants [8] and are widespread in Enterobacteriaceae . Based on the degree of DNA homology , IncHI plasmids have been classically divided into three subgroups , IncHI1 , IncHI2 and IncHI3 [9] , but two new Inc groups ( IncHI4 and IncHI5 ) have been recently described [10] . Regulation of conjugative transfer of IncHI plasmids shows a distinctive feature: transfer is repressed at temperatures encountered within a warm-blooded host ( 37°C ) and induced at temperatures found outside the host ( 22–30°C ) [11] . Within the genus Salmonella , IncHI plasmids account for a significant proportion of antibiotic resistance phenotypes in the most common invasive Salmonella serovars: S . enterica serovar Typhi and S . Paratyphi A [12] . A search for IncHI plasmids within S . Typhi strains has shown that more than 40% of all isolates harbor an IncHI plasmid [13] . A recent study has also shown that IncHI2 plasmids predominate in antibiotic-resistant Salmonella isolates [14] . IncHI-encoded AMR can also be present in other enterobacterial genera , such as Klebsiella pneumoniae [15] and Citrobacter freundii [16] . Over the last three years , a novel role of IncHI2 plasmids in AMR spread has been reported . The emergence of AMR gram-negative bacteria , especially those producing carbapenemases , reintroduced colistin as a last resort antibiotic for the treatment of severe infections [17] . In contrast to its limited use in humans , colistin is widely used in food-producing animals [18] . In the past , colistin resistance was associated with chromosomal mutations only [19] . Nevertheless , plasmid-mediated resistance conferred by a mobilized colistin resistance gene ( mcr-1 ) emerged recently . Since its discovery in 2016 in China [20] , mcr genes , including the mcr-1/2/3/4/5 variants , have been detected in bacterial organisms from human and animal microbiota , including clinical specimens and food samples in over thirty countries [21–25] . IncHI2 plasmids represented 20 . 5% of all plasmids encoding the mcr-1 gene worldwide , but up to 41% in Europe [26] . This finding highlights the role of IncHI plasmids in the global epidemiology of AMR . In addition to colistin resistance in the Enterobacteriaceae , IncHI2 plasmids have also been shown to encode fluoroquinolone resistance determinants in Salmonella [27] . Of special concern is the additionally present mcr-1 resistance determinant in Enterobacteriaceae carrying carbapenem resistance genes , such as blaNDM and blaKPC . The combination of these AMR determinants will seriously compromise the treatment of infections caused by pathogenic strains harboring these plasmids [28 , 29] . An AMR clone of the highly virulent E . coli ST95 lineage has been recently described [30] . E . coli ST95 isolates are causative agents of extraintestinal infections , such as neonatal meningitis and sepsis . They are usually sensitive to several antibiotics . The characterized clone harbors an IncHI2 plasmid that encodes , among others , resistance determinants to colistin and several other antibiotics , including the extended-spectrum beta-lactamase blaCTX-M-1 . The spread of such a clone could be a global threat to human health [30] . The plasmid R27 is the prototype of IncHI1 plasmids . It harbors the Tn10 transposon , which confers resistance to tetracycline ( Tc ) , and has been exhaustively studied for over 25 years . R27 replication and conjugation determinants are well characterized [31–33] , and its complete nucleotide sequence is available [34] . Several ORFs from R27 ( 66% ) do not show similarity to any known ORFs . IncHI plasmids share a common core of approximately 160 kb . The differences in size are due to the distinct presence of insertion elements . Many of them encode AMR determinants [35] . The immunoglobulin ( Ig ) -like domain can be identified in a large number of proteins with diverse biological functions , is widely distributed in nature , and is present in vertebrates , invertebrates , plants , fungi , parasites , bacteria , and viruses [36] . The structural feature of Ig-like domains is the presence of chains of approximately 70–100 amino acid residues present in anti-parallel β-strands and organized in two β-sheets that are packed against each other in a β-sandwich . Ig-like domains are widely distributed in bacteria . Bacterial proteins that contain an Ig-like domain ( Big ) are involved in several functions , such as conjugative transfer , adhesion and biofilm development [37] . We present in this report the identification and characterization of a novel protein containing an Ig-like domain that is encoded by IncHI plasmids and that , among other functions , plays a key role in plasmid conjugation . This newly identified protein may be of interest to develop new approaches to combat IncHI-mediated AMR .
In a previous report , we analyzed the secretome of the Salmonella strain SL1344 harboring the IncHI1 plasmid R27 . A large molecular mass protein ( 155 . 4 kDa ) could be detected in the cell-free supernatant fraction [38] . Protein identification by LC-MS/MS analysis showed that it corresponds to an R27-encoded gene product , the product of ORF R0009 . The R27 R0009 gene is one of the several R27 genes whose gene products previously had no function assigned . In this report , we have addressed the characterization of the R0009 gene product . From here on , we will term the R0009 gene product RSP ( R27-secreted protein ) and the R0009 gene rsp . To begin with the characterization of the protein , we first performed an in silico analysis by using the Phyre2 and PSIPRED secondary structure prediction algorithms ( S1 Fig ) . The structure predicted by PSIPRED shows that the RSP protein contains only 1% alpha helices and is formed by mainly β-sheets ( 61% ) . The use of the Conserved Domain algorithm ( see Material and Methods ) to identify conserved domains in the RSP protein led to the identification of a group 3 bacterial Ig-like domain ( Big 3 ) ( Fig 1 ) . The Phyre2 algorithm supports the PSIPRED prediction and shows that the C-terminal region of the RSP protein exhibits significant similarity with , among others , two large molecular mass bacterial proteins that exhibit adhesion properties ( i . e . , the Staphylococcus aureus SraP protein [39] and the Salmonella giant adhesion protein SiiE [40] ) ( S2 Fig and Fig 1 ) . A BLASTn search of the NCBI plasmid database using the R27 R0009 sequence showed that genes encoding RSP-like proteins are present in both IncHI1 ( 99% identity ) and IncHI2 plasmids ( 87%-74% identity ) ( S1 Table ) . All 34 different IncHI1 plasmids and 102 different IncHI2 plasmids included in the NCBI database contain an R0009 allele . To identify the RSP protein in the different cellular compartments , a Flag tag was added to the rsp gene ( see materials and methods section for details ) . Cultures of the strain SL1344 ( R27 RSP-Flag ) were grown in LB medium at 25°C to an O . D . 600 nm of 2 . 0 . Samples were then collected , and the different cellular fractions were obtained . RSP was detected by western blotting , using anti-Flag antibodies ( Fig 2 ) . The protein was detected in the periplasm , inner membrane and cytoplasmic fractions . The presence of this protein in different envelope compartments of the cell suggests that RSP can be translocated to the outer surface of the cell . Therefore , it could be identified when the secretome was analyzed . rsp maps in one of the R27 regions required for plasmid transfer , the Tra2 region [32 , 41] ( S3 Fig ) . In a previous study [32] , the rsp gene was disrupted by inserting a chloramphenicol resistance cassette , and its effect on the conjugation frequency of an R27 mutant derivative ( drR27 ) that exhibits a conjugation frequency that is higher than that of the wt plasmid was determined . rsp mutants exhibited a reduced conjugation frequency of the drR27 plasmid , but complementation experiments were not performed . We therefore decided to clarify the role of the RSP protein in wt R27 plasmid conjugation . Upon constructing an R27 derivative lacking the rsp gene ( the plasmid R27 Δrsp ) , the conjugation frequency of the strains SL1344 ( R27 ) and SL1344 ( R27 Δrsp ) growing at 25°C was compared . In three independent experiments , transfer of the R27 Δrsp plasmid could not be detected at a frequency higher than 3 x 10−7 ( Fig 3 ) . To correlate RSP loss with the observed effect on R27 plasmid conjugation , we cloned the rsp gene with its own promoter in the low copy number vector pLG338-30 , obtaining the plasmid pLG338-rsp . Transformation of this latter plasmid in the strain SL1344 ( R27 Δrsp ) resulted in R27 Δrsp transfer at a frequency only slightly lower than that of wt R27 ( Fig 3 ) . Previous reports suggested that the transcription of genes mapping in the R27 Tra2 region is thermoregulated [11 , 33] . We decided both to confirm RSP thermoregulation and to assess the effect of the growth medium on RSP expression . To that end , we constructed an rsp::lacZ transcriptional fusion and measured rsp transcription at 25°C and 37°C both in rich ( LB ) and minimal ( M9 ) media . Samples were both collected at the exponential and early stationary growth phases . In accordance with the observed effect of temperature on IncHI plasmid conjugation , low temperature influences rsp transcription , and this effect occurs in both culture media used ( Fig 4 ) . Nevertheless , we could also observe that when comparing cultures grown in both media , rsp transcription is significantly higher in cells grown in minimal medium than in cells grown in LB medium . In fact , rsp transcription at 37°C in M9 medium is only four times lower than rsp transcription at 25°C in cells growing in LB medium ( Fig 4 ) . We also used qRT-PCR to assess the effect of the growth temperature on rsp transcription in cells grown at 25 and 37°C in LB medium until the onset of stationary phase ( O . D . 600 nm of 2 . 0 ) . Transcription of the rsp gene is several orders of magnitude ( more than 40-fold ) lower at 37°C than at 25°C . We previously showed that acquisition of the R27 plasmid by Salmonella results in reduced motility [42] . Considering that the RSP protein is exported to the external surface of the cell and that it influences conjugation , we decided to determine whether this protein might play a role in the observed R27-dependent reduced motility of Salmonella cells . To assess this possibility , we performed a comparative motility assay with the Salmonella strain SL1344 and its derivatives incorporating the R27 , R27 Δrsp , and R27 Δrsp pLG338-rsp plasmids . The results obtained ( Fig 5 , S4 Fig ) show that R27-dependent motility loss requires the synthesis of the RSP protein . This phenotype can be complemented by providing the RSP protein in trans: the comparatively increased motility that is observed in the strain R27 Δrsp is reduced when the plasmid pLG338-rsp is incorporated ( Fig 5 ) . We next studied the mechanism by which the RSP protein is exported . To address this point , we first conjugated the R27 plasmid to the E . coli strain MG1655 and analyzed the secretome of the transconjugants . The RSP protein could be detected by SDS-PAGE as it has been detected in Salmonella ( S5 Fig ) . This observation suggests that ( i ) both the strains SL1344 and MG1655 harbor chromosomally encoded secretion determinants for the RSP protein , ( ii ) RSP is an autotransporter and is secreted via a type V secretion system , or ( iii ) the R27 plasmid encodes the determinants responsible for RSP export . Considering the limited secretory ability of E . coli MG1655 , it seems unlikely that this strain would express a secretion system that would account for RSP export . The use of the SSPred program ( http://www . bioinformatics . org/sspred/html/sspred . html ) suggested that the RSP protein could be exported through a type IV secretion system ( S6 Fig ) . To provide evidence supporting this hypothesis , we decided to knock out the ATPase encoded by the R27 trhC gene [43] and check the presence of the RSP protein in the secretome of the strain SL1344 ( R27 ΔtrhC ) . The RSP protein could not be detected in the secretome of the strain SL1344 ( R27 ΔtrhC ) but could be detected intracellularly ( Fig 6 ) . We next checked the complementation of RSP export in the strain SL1344 ( R27 ΔtrhC ) by providing in trans the gene encoding the ATPase cloned in the plasmid pBR322 ( plasmid pBR322-trhC ) . Complementation of RSP export could be observed ( Fig 6A–6C ) , thus suggesting that the R27-encoded type IV secretion system mediates export of the RSP protein . As shown above , the RSP protein can be detected in different cellular compartments of strain SL1344 cells but not in the outer membrane . To better understand the cellular location of RSP , we performed transmission electron microscopy studies by using gold-labeled antibodies raised against the Big 3 domain of the RSP protein . No gold particles were found to be associated with plasmid-free SL1344 cells or SL1344 ( R27 Δrsp ) cells ( Fig 7A and 7C ) . In contrast , gold particles could be found associated with mainly the flagellar filaments of SL1344 ( R27 ) cells ( Fig 7B ) . A detailed image of flagella fragments shows gold particles associated with specific structures attached to the flagella ( Fig 7D and 7F ) . Flagella containing the RSP protein are frequently present as broken fragments . Because of the observed interaction , we studied whether the R27 protein copurifies with flagellin , and this possibility was indeed occurred ( S7 Fig ) . Considering that the RSP protein is required for R27 plasmid conjugation and that it is associated with the flagella , we decided to analyze the role of the RSP protein in R27 conjugation in donor cells lacking flagella . To this end , we constructed an SL1344 ( R27 ) ΔflgE derivative . The strain SL1344 ( R27 ) ΔflgE does not synthesize the flagellar hook , it is nonmotile , and the RSP protein is expressed ( S7 Fig ) . We assessed whether the RSP protein differentially influenced conjugation in the wt and the flgE derivative of the strain SL1344 ( R27 ) . We used two experimental designs: conjugation in liquid medium or on nitrocellulose filters . When conjugation was performed in liquid medium , the strain SL1344 ( R27 ) ΔflgE showed a conjugation frequency similar to that of the wt strain ( Fig 8A ) . Expression of the RSP protein was also required for plasmid conjugation in the strain SL1344 ( R27 ) ΔflgE ( Fig 8A ) . When matings took place on nitrocellulose filters , the strain SL1344 ( R27 ) ΔflgE also showed a conjugation frequency similar to that of the wt strain ( Fig 8B ) . On the other hand , by using this latter experimental approach , transconjugants could be detected at a low frequency when the strain SL1344 ( R27 Δrsp ) was used as the donor ( Fig 8C ) . Notably , the conjugation frequency observed in the strain SL1344 ( R27 Δrsp ) ΔflgE was significantly higher than that in the strain SL1344 ( R27 Δrsp ) ( Fig 8C ) . Hence , loss of the RSP protein differentially influences conjugation in flagellated and nonflagellated Salmonella cells under specific mating conditions . Considering that the RSP protein is exported to the outer surfaces of Salmonella cells , we decided to assess whether opsonization of SL1344 ( R27 RSP-Flag ) cells with anti-Flag antibodies would result in increased phagocytosis . For this purpose , bone marrow-derived macrophages ( BMDMs ) were exposed to S . enterica serovar Typhimurium SL1344 expressing either RSP ( R27 wild type ) or RSP-Flag ( R27 RSP-Flag ) that had previously been incubated with anti-Flag antibodies , and the levels of internalized bacteria were measured by flow cytometry . Significantly higher numbers of infected macrophages were observed after exposure to opsonized RSP-Flag-expressing bacteria than of BMDM exposed to nonopsonized bacteria ( Fig 9 ) . Thus , binding of anti-Flag antibodies to the surface of RSP-Flag increases the capability of macrophages to phagocytose bacterial cells expressing this protein , further supporting the surface localization of RSP in SL1344 cells .
Details of IncHI plasmid conjugation , including the effect of temperature and the role of global regulators , such as the H-NS and Hha proteins , in the thermoregulation of IncHI conjugation , have been known for several years [11 , 31–33 , 44] . Nevertheless , the role of the RSP protein had been hitherto overlooked . Our study exemplifies the relevance of assigning function to the large percentage of sequenced genes that code for proteins of unknown function , as was the case for the rsp gene of the plasmid R27 . The fact that RSP protein homologs are encoded in all IncHI1 and IncHI2 plasmids hitherto sequenced and deposited in the NCBI database highlights the importance of RSP function in the biology of these plasmids and their hosts . By studying the RSP protein , we shed light on novel aspects of IncHI plasmid conjugation and focused on the design of a novel strategy to combat infections caused by bacteria harboring IncHI plasmids . As suggested from the fact that the rsp gene maps in the Tra2 region of the R27 plasmid ( S3 Fig ) , which includes several conjugative determinants , the RSP protein plays a relevant role in R27 conjugation . Previous studies that used a conjugation-derepressed mutant derivative of the R27 plasmid ( dR27 ) suggested that the rsp gene product reduces the conjugation frequency of the plasmid [32] . By using the wt R27 plasmid here , we show that the expression of the RSP protein is required for R27 conjugation in SL1344 cells growing in liquid medium . Inactivation of the rsp gene reduces the conjugation frequency to experimentally undetectable values , and complementation of the rsp mutation by the plasmid pLG338-rsp restores the conjugation frequency observed in wild-type cells . Semi quantitative RT-PCR analysis of transcription in the R27 Tra2 region where rsp maps suggested that transcription of genes mapping in Tra2 is thermoregulated [33] . Thermoregulation of IncHI plasmid transfer has long been known [45] . Classically , thermoregulation of IncHI plasmid conjugation was interpreted as a means of facilitating dissemination of antibiotic resistance determinants in natural water and soil environments [11 , 34 , 46] . Nevertheless , some studies could show that these plasmids facilitate the adaptation of enterobacteria such as Salmonella , to nonhost environments [42] . Incorporation of the IncHI plasmid R27 significantly impacts the Salmonella transcriptome when cells enter stationary phase and grow at low temperature ( 25°C ) . Optimal dissemination of IncHI plasmids at low temperatures can thus be interpreted as these plasmids favoring fitness of their bacterial hosts when they thrive in the environment or in hosts such as plants . R27 also modified the Salmonella transcriptome at 37°C [42] , thus suggesting that the R27 plasmid can also influence Salmonella physiology within the host [42] . Whereas these plasmids show very reduced transfer frequencies at 37°C , the facts that they also modify the bacterial transcriptome within the host and that they encode AMR determinants [8] provide evidence for their clinical role . We confirm in this report that rsp transcription is thermoregulated and provide further data about the regulation of rsp expression . The nature of the growth medium also influences rsp transcription . Growth in minimal medium at high temperature allows moderately high levels of rsp transcription . Hence , even within the host at 37°C , significant levels of RSP protein expression can occur . Some of the data presented in this report shed light on one of the roles of the RSP protein and hence on novel features of IncHI plasmid conjugation . The type IV secretion system encoded by the R27 plasmid likely mediates RSP export . This phenomenon has already been reported for several type IV secretion systems , which are known to transfer both proteins and relaxosomes [47–51] . The existence of a periplasmic RSP intermediate suggests that the protein is translocated by a type IV piston-like mechanism [52] . The flagellar filaments of SL1344 cells appear to be one of the targets of extracellular RSP protein . Interaction of the RSP protein with the flagellar filament may affect its structural stability . In fact , flagella interacting with the RSP protein are shorter and more breakable than those from cells not expressing the RSP protein . Although this latter observation may be a consequence of the manipulation of the bacterial cells for electron microscopy observation , it is apparent that the likely consequence of the interaction of the RSP protein with the flagella must be the alteration of flagellar function , which , in turn , leads to the observed reduced bacterial cell motility . IncHI plasmids reducing bacterial cell motility also appears to occur by another mechanism: R27 plasmid encoded regulators downregulate flagella synthesis [53] . The results presented both in this latter report and in the present paper support the view that the cell motility is reduced when IncHI plasmid conjugation is prompted . Nevertheless , we show here that motility reduction is only one of the events that promote efficient plasmid transfer: the RSP protein is also required by nonflagellated donor cells to efficiently transfer the R27 plasmid . Therefore the interaction of the RSP protein with the flagella is not the main reason why this protein plays a very important role in R27 conjugation . In spite of this , a relationship among the RSP protein , the flagella and the R27 conjugation frequency can be established when specific mating conditions take place . When the mating pairs are placed in nitrocellulose filters and flgE mutants are used as donors , the conjugation frequency of the plasmid R27 Δrsp is more than threefold higher than that observed when flgE cells harboring the wt R27 plasmid are used as donors . These results show a differential effect of the RSP protein in flagellated and nonflagellated cells . The absence of the flagella reduces the impact of RSP loss on the conjugation frequency . A likely hypothesis is that the interaction of the RSP protein with flagella may reduce motility , being a first step to favor conjugation but requiring additional RSP function ( s ) . RSP interacting with the flagella in cells grown in liquid medium is observed in most of the SL1344 ( R27 ) cells ( about 70% ) . Nevertheless , the R27 conjugation frequency between cells growing under these conditions is somewhat less than 10−3 . This clearly shows that expression of conjugation functions is just a requisite for conjugation to occur , but plasmid transfer requires several concatenated events to take place . Any interference in the process ( i . e . , disruption of the mating pairs ) may interrupt conjugation . Adhesion is a function displayed both by several bacterial proteins containing an Ig-like domain [37] and by two other proteins whose C-terminal domains show similarity to the RSP C-terminal domain: SraP and SiiE . The S . aureus SraP protein binds to sialylated receptors on platelets and lung epithelium [39 , 40 , 54] , and the S . enterica giant adhesion protein SiiE enables apical invasion into enterocytes [55] . It is therefore likely that the RSP protein may also facilitate cell-to-cell adherence , which is critical for IncHI plasmid conjugation . Hence , the RSP protein must play different roles in IncHI plasmid conjugation . By binding to flagella and subsequently reducing motility , RSP may facilitate random recipient/donor collisions , leading to cell-to-cell contact and generating mating pairs . Consolidation of the mating pairs must require the adhesion properties of RSP . Bacterial cell clumping has been shown to increase the conjugation frequency of some plasmids [56–58] . In some instances , the conjugative system encodes a clumping protein that promotes cell aggregation , which in turn results in an increased conjugation frequency [57] . Whereas large aggregates , such as those observed in these systems , are not apparent in the strain SL1344 ( R27 ) in the presence of recipient cells , RSP function in IncHI plasmid conjugation may resemble the function of clumping proteins in other systems . RSP-mediated adherence could also facilitate attachment to other surfaces . We show in this report that although RSP expression is thermoregulated , it can also be expressed at 37°C , with expression levels dependent on the nature of the culture medium . Taking into account the role of proteins such as SraP and SiiE in adherence to eukaryotic cells , RSP expression within the host , regardless of promoting conjugation , might also facilitate adherence of enterobacteria harboring an IncHI plasmid to specific receptors in host tissues . Infections by AMR Enterobacteriaceae in immunocompromised patients and others may result in fatality [59] . The recently reported relevant role of IncHI plasmids disseminating AMR , specifically colistin resistance [22–26 , 30 , 60] highlights the urgent need to control IncHI plasmid dissemination . Considering that ( i ) the RSP protein is expressed by both IncHI1 and IncHI2 plasmids; ( ii ) the RSP protein is exported to the outer surfaces of the bacterial cell; and ( iii ) Flag-tagged RSP is recognized by Flag-specific antibodies , leading to opsonization and increased phagocytosis by macrophages , it is apparent that targeting the RSP protein can become a strategy for both restricting the dissemination of IncHI plasmids and combatting infections caused by enterobacterial strains harboring any of them . Inhibiting bacterial conjugation has been suggested as an important strategy to reduce the persistence of antibiotic resistance in natural environments [7 , 61] . Considering that RSP loss results in IncHI plasmid conjugation inhibition , targeting of the RSP protein by nonpathogenic bacterial cells expressing RSP-directed nanobodies could represent a novel strategy focused on controlling the dissemination of IncHI plasmids in some natural environments , such as livestock farms or water treatment plants . Vaccination is one of the relevant approaches that should be fostered to combat AMR . In this context , multiantigen vaccines may favor competing bacteria in the different colonizing niches , thus reducing the incidence of AMR pathogens [62] . When proven to be antigenic , the RSP protein can be considered a candidate to be included in these vaccines .
The protocol requiring animal manipulation has been approved by the Institutional Animal Care and Use Committee ( IACUC ) from Parc Científic de Barcelona ( PCB ) ( project #9672 ) . The PCB Animal Facility is accredited and registered by the Generalitat of Catalonia government ( registration # B-9900044 ) as a breeding and user center for laboratory animal research and for the breeding and use of genetically modified organisms ( GMOs ) . IACUC-PCB considers that the abovementioned project complies with standard ethical regulations and meets the requirements of current applicable legislation ( RD 53/2013 Council Directive; 2010/63/UE; Order 214/1997/GC ) . The bacterial strains and plasmids used in this work are listed in S2 Table . The bacterial strains were routinely grown in Luria-Bertani ( LB ) medium ( 10 g l-1 NaCl , 10 g l-1 tryptone and 5 g l-1 yeast extract ) , or as indicated in the text , cells were also grown in M9 minimal medium [63] supplemented with glucose at a final concentration of 0 . 4% with vigorous shaking at 200 rpm ( Innova 3100 , New Brunswick Scientific ) . The antibiotics used were chloramphenicol ( Cm ) ( 25 μg ml-1 ) , tetracycline ( Tc ) ( 15 μg ml-1 ) , carbenicillin ( Cb ) ( 100 μg ml-1 ) and kanamycin ( Km ) ( 50 μg ml-1 ) ( Sigma-Aldrich ) . All enzymes used to perform standard molecular and genetic procedures were used according to the manufacturer’s recommendations . To introduce plasmids into E . coli and Salmonella , bacterial cells were grown until an O . D . 600 nm of 0 . 6 . Cells were then washed several times with 10% glycerol , and the respective plasmids or DNA were electroporated by using an Eppendorf gene pulser ( Electroporator 2510 ) . Deletions of the rsp ( ORF R0009 ) , trhC and flgE genes were performed in the strain SL1344 ( R27 ) by using the λ Red recombination method , as previously described [64] . The antibiotic resistance determinant of the plasmid pKD3 was amplified using the corresponding oligonucleotides RS0009_P1/RS0009_P2 and trhCP1/trhCP2 for the rsp and trhC genes , respectively ( S3 Table ) , and the resistance determinant of the plasmid pKD4 was amplified using the corresponding oligonucleotides SL1344flgEP1/SL1344flgEP2 for the flgE gene ( S3 Table ) . The mutants were confirmed by PCR using the oligonucleotides RS0009_Up_for/RS0009_down_rev for rsp , trhCP1up/trhCP2down for trhC and SL1344flgEP1up . 1/SL1344flgEP2down . 1 for the flgE gene ( S3 Table ) . A transcriptional lacZ fusion was made in the rsp gene from the R27 plasmid . The antibiotic resistance determinant from the plasmid R27 rsp was eliminated using an FLP/FRT-mediated site-specific recombination method , as previously described [65] , thus , generating the plasmid R27 Δrsp . A FRT-generated site was used to integrate the plasmid pKG136 [66] , thereby generating a transcriptional lacZ fusion . Recombinational transfer of the Flag sequence into the rsp gene was achieved by following the methodology described in [67] . The template vector coding for Flag and Kmr used was pSUB11 . The primers used for the construction of the Flag-tagged derivative were R27_p0103XP1 and R27_p0103XP2 ( S3 Table ) . The correct insertion of the Flag-tag was confirmed by PCR using oligonucleotides R27_p0103XP1UP and R27_p0103XP2DOWN ( S3 Table ) . To construct the plasmids pLG338-rsp and pBR322-trhC , ORF R0009 ( rsp ) and the trhC gene ( GenBank accession number NC_002305 . 1 , positions 11659–16099 and 28465–32972 , for R0009 and trhC genes , respectively ) were amplified using the oligonucleotides 09-EcoRI-pLG_For/09-BamHI-pLG_Rev and trhCBamHiFW/trhCBamHiRV ( see S3 Table for the sequences ) together with Phusion Hot Start II High-Fidelity DNA Polymerase ( Thermo Scientific ) following the manufacturer’s recommendations . rsp and trhC amplification with the above-referred oligonucleotides generated EcoRI/BamHI and BamHI/BamHI sites flanking the rsp and trhC genes , respectively . The corresponding EcoRI/BamHI and BamHI/BamHI fragments were cloned into the vectors pLG338-30 and pBR322 previously digested with the same enzymes , respectively . The resulting plasmids were Sanger sequenced and termed pLG338-rsp and pBR322-trhC , respectively . For polyclonal antibody production , the Big 3 domain encoded by the RSP protein was used . The Big 3 domain corresponds to 140 amino acids ( Y L Y I F D L T D L T N G S Y A A S F T V E N N S K N T S T Y N E P E S K L M L S D N P T L M V L K D G A A L A K R A P V Y F L N E I I V A A F Q G Q A G V A D I K A V T I D N K L V E L T P T N H K G I Y Y L P V G D D L E V N A D H E I T V I A E N L Y G K I V T F N T T F T Y Q P ) and is encoded in the central region of the RSP protein . Amplification of that region was achieved by performing PCR using the R27 plasmid as a DNA template and the primers RSPBig3_31FW and RSPBig3_31RV together with the Thermo Scientific Phusion Hot Start II High-fidelity DNA Polymerase following the manufacturer’s recommendations . The DNA was then purified using a Thermo Scientific GeneJet PCR Purification Kit and ligated into the pLATE31 vector according to the manufacturer´s instructions ( Thermo Scientific aLICator LIC cloning and expression system ) . The resulting plasmid , termed pLATE31-Big3 , was Sanger sequenced . BL21 DE3 cells were used for recombinant expression of the Big 3 domain . Cells transformed with pLATE31-Big3 plasmid were grown in LB medium supplemented with carbenicillin at a final concentration of 100 μg/ml at 37°C until O . D . 600 nm of 0 . 4 . Then , recombinant protein expression was induced by adding IPTG at a final concentration of 1 mM for 3 hours . Cells were then centrifuged at 7 , 500 xg for 30 minutes at 4°C . The pellet was subsequently resuspended in buffer A20 ( 20 mM HEPES pH 7 . 9 , 100 mM KCl , 5 mM MgCl2 , 10% glycerol , 20 mM imidazole ) plus protease inhibitor ( Complete Ultra Tablets , Mini , EDTA-free , EASYpack , Roche ) . Cells were then disrupted by sonication , and the insoluble fraction ( inclusion bodies ) was collected after centrifugation at 12 , 000 xg for 30 minutes at 4°C . Inclusion bodies containing the recombinant Big 3 protein were solubilized in buffer B ( 100 mM NaH2PO4 , 10 mM Tris pH 8 . 0 , 8 M urea ) for 30 minutes at room temperature . Upon centrifugation ( 12 , 000 xg at 4°C for 30 minutes ) , the supernatant was used for protein purification by by immobilized-metal affinity chromatography ( IMAC ) using HisPur Ni-NTA Superflow Agarose ( Thermo Scientific ) . Recombinant Big 3 protein was eluted from Ni-NTA resin by changing the pH first using Buffer D ( 100 mM NaH2PO4 , 10 mM Tris pH 6 . 3 , 8 M urea ) and then using buffer E ( 100 mM NaH2PO4 , 10 mM Tris pH 4 . 5 , 8 M urea ) . Both eluted fractions were collected and then concentrated using Amicon Ultra-15 Ultracel 3K ( Millipore ) according to the manufacturer´s instructions . The purified Big 3 protein was adjusted to 1 mg/ml and inoculated into rabbits according to standard protocols ( Unitat d'Experimentació Animal de Farmàcia–CCiTUB . Universitat de Barcelona , Barcelona , Spain ) . After immunization , preimmune serum and serum collected after the immunization period were tested by western blot against the RSP protein . The R27 plasmid was conjugated either in liquid as described previously [45] or on filters in the presence of a physical support ( 0 . 45 μm nitrocellulose filters , Millipore ) . For both protocols , cultures of donor and recipient strains were grown in Penassay broth ( 1 . 5 g l-1 meat extract , 1 . 5 g l-1 yeast extract , 5 g l-1 peptone , 1 g l-1 glucose , 3 . 5 g l-1 NaCl , 1 . 32 g l-1 KH2PO4 , 4 . 82 g l-1 K2HPO4 3H2O ) . Conjugations were performed using the recipient strains SL1344 ibplac ( Kmr ) [38] . Cultures of donor strains SL1344 that harbored the plasmids; R27 , R27 Δrsp , R27 Δrsp complemented with pLG338-rsp , R27 ΔflgE strain or R27 Δrsp ΔflgE strain and recipient strains SL1344 ibplac were grown overnight without shaking at 25°C in Penassay broth . Aliquots were washed to eliminate the antibiotics and resuspended in the same volume of initial culture . In the liquid protocol , 0 . 4 ml of the recipient strain culture and 0 . 1 ml of the donor strain culture were mixed and incubated at 25°C without agitation for 2 h . Mixtures were serially diluted and then plated in LB containing either Tc or Tc and Km . The mating frequency was calculated as the number of transconjugants per donor cell . In the filter protocol , 0 . 4 ml of the recipient strain culture and 0 . 1 ml of the donor strain culture were mixed . Then , 0 . 1 ml of the mixture was spotted in the center of a 0 . 45 μm filter laid on an LB plate . The plates were incubated at 25°C for 16 h . The filters were then washed with 1 ml of 10 mM MgSO4 , and the cells were collected , serially diluted and plated on LB plates containing either Tc or Tc and Km . The mating frequency was calculated as the number of transconjugants per donor cell . Student´s t-test was used to determine statistical significance , and the values were obtained by using the GraphPad Prism 5 software . A P value of less than 0 . 05 was considered significant . The oligonucleotides ( from 5’ to 3’ ) used in this work are listed in S3 Table . β-Galactosidase activity measurements were performed as previously described [63] . Values are given as Miller units . Student´s t-test was used to determine statistical significance , and the values were obtained by using the GraphPad Prism 5 software . A P value of less than 0 . 05 was considered significant . The motility assay was performed as described [42] . Briefly , motility was performed on tryptone broth ( TB ) plates ( 1% tryptone , 0 . 5% NaCl ) containing 0 . 35% agar . Overnight bacterial cultures grown in LB at 37°C were spotted ( 5 μl ) on the center of the plates and incubated for 24 h at 25°C . The experiments were repeated three times with three plates of each strain in each experiment . The colony diameter was measured and plotted , and standard errors were calculated . For flagellum isolation , cells were grown overnight at 25°C in LB medium supplemented with Tc ( for R27 selection ) . Cells were then centrifuged at 8 , 000 xg for 30 minutes at 4°C . Pellets were resuspended in 1/100 of the initial volume with 100 mM of Tris-HCl pH 8 . 0 and passed through a 21G syringe six times . Thereafter , the cells were centrifuged ( 8 , 000 xg , 20 minutes , 4°C ) . The resulting supernatants were centrifuged again at 12 , 000 xg for 30 minutes at 4°C . Again , the supernatants were ultracentrifuged at 40 , 000 xg for 1 hour at 4°C . The pellet , including the flagella , was resuspended in 100 mM of Tris , 2 mM EDTA pH 8 . 0 and analyzed by SDS-PAGE with a 10% gel [68] . Cell-free supernatants were prepared from cultures grown at 25°C until the beginning of stationary phase ( O . D . 600 nm of 2 . 0 ) . Ten milliliters of bacterial cells were centrifuged , and supernatants were filtered through a 0 . 22 μm filter ( Millipore ) . For each strain , 2 ml of cell-free supernatants containing secreted proteins were mixed with trichloroacetic acid at a final concentration of 10% . Incubation was performed on ice for 45 minutes , and the tubes were centrifuged for 30 minutes at 12 , 000 xg at room temperature . The pellets were washed once with cold acetone and again centrifuged for 30 minutes at 12 , 000 xg at room temperature . Proteins were solubilized with 1x Laemmli Sample Buffer ( Bio-Rad ) . Samples were boiled for 10 minutes and loaded into a SDS-PAGE with a 12 . 5% gel [68] . Protein samples were analyzed by SDS-PAGE with 10% or 12 . 5% gels [68] . Proteins were transferred from the gels to PVDF membranes using the Trans-Blot Turbo system ( Bio-rad ) . Western blot analysis was performed with a monoclonal antibody raised against the Flag-epitope ( Sigma ) diluted 1:10 , 000 in a solution of PBS , 0 . 2% Triton , 3% skimmed milk and incubated for 16 hours at 4°C . Membranes were washed for 20 minutes each with PBS , 0 . 2% Triton solution . The washing step was repeated three times . Thereafter , the membranes were incubated with horseradish peroxidase-conjugated goat anti-mouse IgG ( Promega ) diluted 1:2500 in a solution of PBS , 0 . 2% Triton for 1 hour at room temperature . Again , three washing steps of 45 minutes with PBS , 0 . 2% Triton solution were performed , and detection was performed by enhanced chemiluminescence using ImageQuant LAS54000 imaging system software ( GE Healthcare Lifesciences ) . Cell fractionation was performed as described [69] . We used 1 ml of bacterial cells from a culture entering stationary phase ( O . D . 600 nm of 2 . 0 ) for fractionation . Samples were resolved by SDS-PAGE with a 12 . 5% gel . Protein identification was performed as described [38] . Bacterial cells were grown until O . D . 600 nm of 2 . 0 . Then , 5 ml of cells were then mixed with a 0 . 2x volume of stop solution buffer ( 95% ethanol , 5% phenol ) , shaken and centrifuged ( 10 minutes , 6 , 000 x g ) . Bacterial pellets were subsequently frozen at -80°C until use . Total RNA was extracted from bacterial pellets using Tripure Isolation Reagent ( Roche ) according to the manufacturer’s instructions . Potential traces of DNA were removed by digestion with DNase I ( Turbo DNA-free , Ambion ) according to the manufacturer’s instructions . RNA concentration and RNA quality were measured using a Nano-Drop 1000 ( Thermo Fisher Scientific ) . The expression level of the rsp gene was determined by using real-time quantitative PCR . Briefly , 1 μg of previously isolated total RNA was reverse transcribed to generate cDNA using a High-capacity cDNA Reverse Transcription kit ( Applied Biosystems ) according to the manufacturer’s instructions . All samples within an experiment were reverse transcribed at the same time; the resulting cDNA was diluted 1:100 in nuclease-free water and stored in aliquots at –80°C until used . As a control , parallel samples in which reverse transcriptase was omitted from the reaction mixture were run . Real-time PCR was carried out using Maxima SYBR green/ROX qPCR master mix ( Thermo Scientific ) and an ABI Prism 7700 sequence detection system ( Applied Biosystems ) . Specific oligonucleotides complementary to the genes of interest were designed using primer3 software . Relative quantification of gene expression of mutants versus the wild-type strain was performed using the comparative threshold cycle ( CT ) method [70] . The relative amount of target cDNA was normalized using the gapA gene as an internal reference standard . For immunogold labeling of bacterial cells , 10 μl of the different bacterial suspensions were applied to carbon-coated grids for 10 min . Upon removal of the excess of liquid , grids were placed face down on drops of PBS and washed three times ( 1 minute each ) . Grids were blocked by floating on drops of PBS containing 1% bovine serum albumin ( BSA ) for 30 minutes and washed two times in PBS ( 1 minute each ) . Grids were then placed on drops of rabbit anti-RSP polyclonal antiserum ( 1:10 dilution ) for 45 min . After three washes in PBS ( 1 minute each ) , grids were incubated with goat anti-rabbit IgG conjugated to 12 nm gold particles for 45 min ( 1:20 dilution ) . Grids were then washed in PBS and distilled water drops , stained with 2% uranyl acetate and air dried . All incubations were carried out at room temperature . Controls were incubated with goat anti-rabbit IgG conjugated to 12 nm gold particles in the absence of specific antibodies . Samples were examined in a JEOL 1010 transmission electron microscope operating at 80 kV . The Phyre2 [71] and PSIPRED [72] web portals for protein prediction and analysis were used to predict the secondary structure of the RSP protein . The Conserved Domain algorithm was used to determine putative domains in the RSP sequence ( https://www . ncbi . nlm . nih . gov/cdd/ ) . BLASTn was performed using the nucleotide sequence of the rsp ( R0009 ) gene against the NCBI database , plus setting the search criteria organism to matching only plasmid sequences . To identify the IncHI1 and IncHI2 plasmids among the plasmid database , we used two reference plasmids previously classified by traditional incompatibility typing into the IncHI1 and IncHI2 groups: R27 and R478 , respectively . From all plasmids , we retrieved those that met two criteria: ( i ) they encode proteins that are homologs of more than half of all proteins encoded by any of the reference plasmids , and ( ii ) they encode replication initiation ( Rep ) proteins that are homologs to those of the reference plasmids , namely , RepHIA and RepHIB from the IncHI1 plasmids and RepHI2 from the IncHI2 plasmids . Following these parameters , we were able to discriminate those plasmids that share many proteins and encode the same replication machineries as the reference genomes ( thus satisfying both criteria ) from those that only share many proteins but not the Rep proteins or only Rep but very few other proteins . The results were filtered by a similarity cut-off > 85% , an alignment length between pairs > 85% and an e-value < 10−10 . C57BL/6 mice ( both males and females ) were obtained from Harlan and maintained under specific pathogen-free conditions at the animal facility of Parc Científic de Barcelona ( PCB ) , Spain . Murine bone marrow-derived macrophages ( BMDM ) were obtained from male and female C57BL/6 mice as previously described [73] . Briefly , bone marrow precursors were differentiated on Petri dishes for 8 days in Dulbecco’s modified Eagle medium ( DMEM ) , supplemented with 20% heat-inactivated fetal bovine serum ( FBS ) and 30% L-cell-conditioned medium as a source of macrophage-colony-stimulating factor . The Salmonella strains SL1344 ( R27 ) and SL1344 ( R27 RSP-Flag ) were transformed with a plasmid expressing red fluorescent protein ( RFP ) ( pBR-RFP . 1 ) [74] to render bacterial cells fluorescent . Before infection , the strains SL1344 ( R27 pBR-RFP . 1 ) and SL1344 ( R27 RSP-Flag pBR-RFP . 1 ) were grown at 25°C in the presence of 100 μg/ml of carbenicillin for 16 hours . Bacterial cells were opsonized with anti-Flag antibodies ( Sigma-Aldrich ) ( 50 ng/106-CFU ) in PBS for 2 hours at 4°C . Macrophages were plated in 6-well plates ( 1 . 5x106 cells/well ) containing DMEM-10% FBS for 24 hours before infection , as described [75] . Briefly , 15 minutes before infection , the cells were cooled at 4°C . For infection , either opsonized or nonopsonized bacteria were added to macrophage cultures at a multiplicity of infection of 15 . Macrophages were incubated with Salmonella cells for 30 min at 37°C and 5% CO2 . Negative control cells were incubated in the absence of bacteria . Additional control cells were incubated with bacterial cells at 4°C for 30 min . After infection , noninternalized bacteria were eliminated by three washes with ice cold PBS . Infected cells were fixed in 5% paraformaldehyde . Infection was analyzed by counting cells containing fluorescent bacteria ( RFP+ ) in a FacsAria I SORP sorter ( Becton Dickinson ) . Noninfected macrophages were used as a control for autofluorescence . Cells incubated with bacteria at 4°C were used to discriminate bacteria adhered to the macrophage cell surface from internalized bacteria . The infection index was calculated as follows: ( %RFP+ cells ) x ( mean fluorescence intensity in the RFP+ population ) . Statistical analysis was performed with GraphPad Prism 7 . 00 . Infection index values were compared using one-way ANOVA and Tukey´s post hoc test for multiple comparisons . | Dissemination of antimicrobial resistance ( AMR ) among different bacterial populations occurs due to mainly the presence of plasmids that encode AMR determinants . IncHI plasmids are one of the groups of bacterial plasmids that confer AMR to several enterobacteria . Recently , resistance to one of the last-resort antibiotics ( colistin ) for some multidrug-resistant infections has spread very rapidly . IncHI plasmids represent 20% of all plasmids transmitting colistin resistance worldwide and 40% in Europe . When analyzing the interactions of the IncHI1 plasmid R27 with Salmonella , we identified a large-molecular-mass protein that is encoded by this plasmid and is exported to the external medium . The R27 plasmid gene coding for that protein ( R0009 ) is widespread among IncHI plasmids . In this report , we characterize the protein , termed RSP . The presented data show that RSP plays a relevant role in IncHI plasmid conjugation and suggest that the protein is retained on the outer surface of the bacterial cells and facilitates cell-to-cell contact before plasmid DNA transfer . Considering that IncHI plasmids significantly contribute to AMR dissemination within enterobacteria , the findings reported in this paper suggest that the identified protein can be a target to control both IncHI-mediated AMR dissemination and infections caused by AMR enterobacteria that harbor these plasmids . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"plasmids",
"immunology",
"microbiology",
"plasmid",
"construction",
"bacterial",
"diseases",
"antibiotic",
"resistance",
"enterobacteriaceae",
"genetic",
"elements",
"forms",
"of",
"dna",
"dna",
"construction",
"dna",
"pharmacology",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"infectious",
"diseases",
"antimicrobial",
"resistance",
"white",
"blood",
"cells",
"animal",
"cells",
"bioinformatics",
"medical",
"microbiology",
"microbial",
"pathogens",
"biological",
"databases",
"molecular",
"biology",
"salmonella",
"pathogen",
"motility",
"biochemistry",
"macrophages",
"sequence",
"databases",
"cell",
"biology",
"nucleic",
"acids",
"virulence",
"factors",
"database",
"and",
"informatics",
"methods",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"genomics",
"mobile",
"genetic",
"elements",
"organisms",
"flagella"
]
| 2019 | Expression of a novel class of bacterial Ig-like proteins is required for IncHI plasmid conjugation |
It is widely accepted for humans and higher animals that vision is an active process in which the organism interprets the stimulus . To find out whether this also holds for lower animals , we designed an ambiguous motion stimulus , which serves as something like a multi-stable perception paradigm in Drosophila behavior . Confronted with a uniform panoramic texture in a closed-loop situation in stationary flight , the flies adjust their yaw torque to stabilize their virtual self-rotation . To make the visual input ambiguous , we added a second texture . Both textures got a rotatory bias to move into opposite directions at a constant relative angular velocity . The results indicate that the fly now had three possible frames of reference for self-rotation: either of the two motion components as well as the integrated motion vector of the two . In this ambiguous stimulus situation , the flies generated a continuous sequence of behaviors , each one adjusted to one or another of the three references .
Sensory stimuli serve an animal to organize its behavior . They may have a meaning or relevance . A stimulus may elicit a certain behavior , but often stimuli are ambiguous . A dark patch in the visual surround of an animal may be a potential hiding place or a predator . If the first evaluation triggers approach , this decision may have to be revised in the next moment as more information comes in . Sometimes , sensory ambiguities can be stable in time while their perception changes . This phenomenon is called multi-stable perception . It has been studied extensively over the last 50 years [1–4] and been observed not only in humans but also in nonhuman primates [5] and pigeons [6] . The capability to constantly reevaluate the current interpretation of a sensory input , with varying outcomes if it is ambiguous , may be a crucial procedure to gain more information from the stimulus so that it has evolved not only in vertebrates but also in insects . As motion vision has been extensively investigated in Drosophila , visual motion stimuli are used also in the present study . The flies are exposed to a stimulus consisting of two transparent wide-field motion components moving into opposite directions . The flies can respond adaptively in different ways . Indeed , like human subjects in multi-stable perception , they alternate stochastically with this experimental design between different , but equally adaptive , behaviors . We call this the transparent panorama motion paradigm ( TPMP ) . Coherent angular motion of the whole panorama signals self-rotation . In the TPMP , each of the two motion components signals self-rotation . Transparent motion of two components , each one covering the whole panorama , may never occur for a freely moving subject with panoramic vision like Drosophila . For smaller regions of the visual field , however , transparent motion regularly occurs , for instance , when a moving subject encounters a moving object . It has been shown that Drosophila can restrict its behavioral responses to parts of the visual field ( spatially selective visual attention ) and that it also possesses mechanisms to filter out unspecific motion stimuli impinging on wide-field motion detection [7] . A study on blowflies using a two-dimensional plaid stimulus [8] also suggests that blowflies might show component selectivity with transparent motion stimuli . The two-dimensional transparent “plaid motion” stimulus [9–11] is also an example for multi-stable perception in humans . There , a moving plaid pattern can be perceived either as the coherent , nontransparent motion of the plaid , or as the transparent movements of two individual gratings moving at an angle to each other . Little is known , however , how the fly visual system deals with completely overlapping transparent wide-field stimuli moving in the same dimension and what the behavioral response to such a stimulus might be . Traditional models of fly motion vision predict the optomotor turning response in horizontal direction to be a unimodal function of the orientation of the motion stimulus . This would include incoherently moving transparent wide-field motion stimuli [12 , 13] , as the models assume the integration of all motion signals detected by the appropriately oriented elementary motion detectors [14] . In the case of the TPMP , the signals from the two wide-field patterns would cancel each other . A study of human transparent motion vision shows that the perception of the transparency is only present under some stimulus conditions , namely when the local motion stimuli are unbalanced [15] . Here , we show component selectivity for these patterns in Drosophila . How extensively the fly activates the behaviors associated with the respective components depends strongly on certain properties of the visual stimuli , particularly pattern contrast and element density . These findings may relate transparent motion processing in the Drosophila visual system to interactions between the figure and wide-field motion vision systems [16] . The stochasticity of the alternations between the different behaviors indicates that these alternations are generated endogenously .
The fly's behavior was studied in the so-called flight simulator [17] . The fly was suspended at a torque meter [18] and positioned in the middle of a cylindrical light-guide arena . Its head was glued to the thorax and fixed in space . The horizontal component of the fly's angular momentum ( yaw torque ) was measured and transformed to generate the rotatory motion of the panorama the fly would have seen if it were free to rotate . In other words , the fly and the arena were in a negative feedback loop ( closed loop ) simulating the fly's horizontal angular motion . A fly confronted in this setup with a random dots pattern uniformly covering the whole inner wall of the arena will adjust its yaw torque to a level that keeps the panorama from rotating . This behavior in the flight simulator is called optomotor balance [19] . It is assumed to be a sensory-motor equivalent of flying straight in free flight . In the flight simulator , it can be challenged by adding a rotatory bias to the panorama . Now the fly must adjust its yaw torque to a new mean value to keep the panorama from rotating . So far , the motion stimulus was unambiguous . To provide ambiguity , we added a second transparent random dots pattern covering the entire arena . In both patterns , the individual pattern elements were opaque , so no change in local contrast occurred within the pattern elements when two of them overlapped . The two textures were identical , however , one was flipped vertically to prevent a complete overlap . If both were linked to the fly's yaw torque in a closed loop with the same coupling parameters , they moved coherently . To generate the ambiguity in the TPMP , we added a rotatory bias to each of the textures , a clockwise ( cw ) bias to one and a counterclockwise ( ccw ) bias to the other . This generated a relative angular motion of constant velocity between the two patterns , independent of the fly's yaw torque , while their velocities still depended upon the fly's yaw torque as well . The fly could stabilize each of the patterns but only one at a time . For this , it had to let the other pattern rotate without responding to that motion component . Now the fly had the options to generate optomotor balance with one or the other texture ( Fig 1A and 1B ) . This is what happened: The fly tied all the coherently moving motion elements of the entire visual field together and treated the two motion components as separate entities generating optomotor balance with one or the other pattern ( Fig 1A and 1B ) . We called the stabilization of one of the components single pattern stabilization ( SPS ) and distinguished the two behaviors according to the bias stabilized ( cw SPS [SPScw] and ccw SPS [SPSccw] ) . SPS was not the only type of orientation behavior the fly generated . It could also use the vector sum of the two bias components as a reference for straight flight . We called this motion average ( MA ) behavior ( Fig 1C ) . Further inspection of the yaw torque traces showed that the fly had various strategies at its disposition for stabilizing the motion components and the MA . Examples are shown in S1 Fig . The fly would , for instance , generate fast , large yaw torque modulations around the value that would stabilize one motion component ( S1D Fig ) . Or it would keep its yaw torque level at the zero baseline and suppress net rotation of one of the patterns by a sequence of saccades towards that side ( S1B and S1C Fig ) . Also , with MA behavior , different strategies could be observed ( Fig 1C; S1E Fig ) . This is not a special feature of the TPMP , as different strategies for optomotor balance could also be observed with unambiguous stimuli ( S2 Fig ) . As stated in the introduction , we took it that the three modes of pattern stabilization behavior ( SPScw , SPSccw , and MA ) reflected the flies’ responses to the three possible references for straight flight within the transparent motion stimulus . To calculate the duration and frequency of these modes , we classified them as follows: to be scored as ccw bias stabilization , yaw torque had to be between −2 × 10−10 and −6 × 10−10 Nm ( green colored domain in Figs 1 and 2 ) . To go as cw bias stabilization , yaw torque had to be in the blue domain ( 2 × 10−10 to 6 × 10−10 Nm ) , and it was taken as MA behavior in the range between −2 × 10−10 and 2 × 10−10 Nm ( yellow domain ) . As the coupling coefficient between pattern motion and yaw torque in the flight simulator was very low compared to freely rotating flies [20] , and very short yaw torque fluctuations therefore had little influence on pattern motion , the moving average over 2 s of the yaw torque ( dark red traces in Fig 1A–1C ) was used for scoring as it reflects the angular velocity of the patterns better than the nonaveraged yaw torque values . The low coupling coefficient was chosen to maximize the difference between the SPS values while keeping the relative angular velocity of the two biases low . The TPMP also worked with a higher coupling coefficient ( S3 Fig ) . The optomotor responses to a single random dots pattern at 37% contrast at different angular velocities showed that there was no difference in the strength of the motion stimulus in the range in which a nonstabilized pattern usually moved in the TPMP ( S4 Fig ) . The crucial property of the flies’ behavior in the TPMP , which allowed us to classify it as multi-stable , was its continuous alternation between the three stabilization modes . In Fig 1D , this property is visualized for a single 3-min experiment . For each of the three possible references for straight flight , an artificial position trace was calculated , assuming a constant forward velocity . The colored sections indicate where the respective behaviors were scored regarding the yaw torque domains in Fig 1 and Fig 2A–2C . As expected from the classification , this mostly applied to stretches of no or low curvature . The experiment in Fig 2A shows that SPScw and SPSccw were the predominant behaviors in the TPMP with random dots patterns and 37% pattern contrast . The flies preferably responded to the single motion components instead of the compound motion as is apparent from the bimodal distribution of the yaw torque . The peaks , however , are not at the exact SPS values , but shifted about 12 . 5% towards the other SPS value . As the yaw torque distribution of the stabilization of a single pattern without incoherent motion ( Fig 2B ) showed no peak shift , it cannot be a result of the rotatory bias . Hence , we hypothesized the peak shift in the TPMP to result from a residual response to one of the references not used , an issue we address further in a later part of this article . In S1 Fig , examples for the different strategies ( yaw torque patterns ) for SPS and MA behavior are described . A salient difference between these yaw torque patterns and the ones shown in Fig 1A–1C is the average amplitude of the yaw torque modulations . Because more of the strategies for SPS required stronger yaw torque fluctuations , we wanted to find out whether the individual mean yaw torque amplitude influenced the scoring of SPS and MA behavior . For each fly , the mean yaw torque amplitude was calculated and compared to the amount of SPS and MA behavior . No correlations were found ( S5 Fig ) . This suggests that the scoring of the different behaviors was independent of the yaw torque modulations . As Fig 2C shows , the component selectivity the flies expressed appeared not to depend on the feedback provided in the TPMP . With the same incoherent wide-field motion stimulus as in the closed-loop TPMP , the yaw torque distribution was much wider and possibly trimodal , compared to the distribution without any motion when the feedback was switched off . In the latter case , the distribution was unimodal and centered around zero . Obviously , the respective yaw torque ranges at which SPS and MA behavior were detected in the closed-loop TPMP held no meaning for the flies in the open-loop TPMP , which would have made the classification of these behaviors hard in open loop . The continuously alternating choice behavior in the TPMP did not require random dots textures . We also used 20 evenly spaced vertical stripes or regular dots ( Fig 2D ) . The results were similar , although we found a significant difference between the regular dots and the vertical stripes ( Fig 2E ) . With the latter two patterns , the elements completely overlapped every 0 . 45 s because of the bias condition , which resulted in a whole-field flicker . This did , however , not seem to have an influence on the overall behavioral choice of the flies . Because the basic horizontal pattern wavelength of the regular dots and the bar pattern was the same , some other pattern feature had to be the cause of the difference in SPS and MA behavior between the two . Interestingly , we found that the flies also showed a response to the individual components of the transparent motion stimulus with the regular dots patterns and without feedback ( S6 Fig ) . For the random dots patterns , as described above , this was not as surprising , as this is also the case in humans [15] . But locally balanced transparent motion stimuli , like the regular dots pattern used here , are not perceived as transparent by human subjects . Although the response to the individual motion components in the open-loop TPMP appeared to be slightly smaller with regular than with random dots patterns , the yaw torque distribution was still clearly multi-modal , which was not the case without the transparent motion stimulus . Because Drosophila shows no behavioral response to a whole-field flicker [21] , we considered this effect to be a response to the transparent motion stimulus . When the same relative rotatory bias between two patterns was injected into the feedback loop , but all of it was added to the closed-loop motion of one of the patterns , the flies stabilized the unbiased pattern significantly more often than the biased one ( Fig 2F ) , which required less yaw torque for SPS . After all , equal forces on both wings , which are in concordance with the internal zero-torque reference of the fly [20] , are more likely to generate straight flight and , presumably , are less demanding than strong turning behavior . As already mentioned above , we found the that time the flies spent with SPS in the TPMP was strongly dependent upon pattern contrast . The random dots patterns of Fig 2D were used . At the lowest contrast measured ( 8% ) , the flies spent about 60% of the time with SPS and less than 30% with MA behavior ( Fig 3 ) . At an intermediate contrast of 37% , which was also used in the experiments in Fig 1 , SPS was only slightly lower and MA behavior only slightly higher than at 8% contrast . At the highest contrast ( 91% ) , the flies predominantly generated MA behavior ( 82% ) and only rarely generated SPS ( 17% ) , with individual flies occasionally displaying constant , very stable MA behavior ( S7 Fig ) . Altogether , in the TPMP , ( Fig 3A and 3B ) SPS decreased with increasing contrast . For comparison , we tested stabilization with a single random dots pattern in the closed loop; the second one stationary ( Fig 3C ) . Flies showed reliable stabilization of a single pattern for contrast values between 8% and 91% , with stabilization increasing with contrast . We also tested a pattern contrast of 4% , but only found a very low pattern stabilization ( 16 . 89 ± 2 . 52 , n = 20 flies ) compared to 8% contrast ( t ( 38 ) = 6 . 029 , p < 0 . 0001 , t test ) , which is why the flies were not tested at contrast levels lower than 8% in the TPMP . We also measured the basic optomotor response ( Fig 3D ) . As with the single pattern in closed loop ( Fig 3C ) , it did not reflect the contrast dependence of SPS in the TPMP . Here , the optomotor response remained stable over all contrast values measured in the TPMP , except the 8% contrast condition , at which the optomotor response of the flies was significantly lower . In that contrast range , however , we observed no significant changes in the TPMP . The inverted contrast dependence of SPS was only found in the TPMP and could also be observed with the regular dots patterns ( S8 Fig ) . We hypothesized that it might be due to a rivalry between SPS and MA . Interestingly , the peak shifts in the yaw torque histogram at intermediate contrast ( 37% ) described above ( Fig 2A ) were not present at low contrast ( 8% ) with regular dots patterns ( S8A Fig ) and only very weak with random dots patterns ( Fig 3A ) . If , as hypothesized above , the peak shift was indeed a residual response to the nonstabilized pattern during SPS , this would make sense: at low contrast , the motion response to the patterns was weaker than at intermediate contrast ( Fig 3D , S8D Fig ) , therefore the response to the nonstabilized pattern in the TPMP at low contrast might also have been lower or more easily suppressed . To test this hypothesis further , we switched off the feedback for one of the two patterns ( Fig 3E ) . Now , the flies could only stabilize one pattern while suppressing their motion response to an open-loop motion component . They could also do MA behavior by integrating the closed- and the open-loop motion component , thereby also reaching the low net rotation that defines MA behavior . Additionally , they could follow the open-loop motion with an optomotor response , not actively controlling any part of the visual stimulus . Because one of the two patterns was not coupled to the flies’ yaw torque anymore , the relative angular velocity of the two patterns was no longer constant . By inducing this feedback asymmetry , we intended to find out whether this had any influence on the peak shift of SPS or on MA behavior . Fig 3F shows that , for SPS , the peak shifts were more pronounced than with both patterns in closed loop for low and intermediate contrast , whereas no peak shift was observed for the MA behavior at high contrast . As SPS was the prevalent behavior at low and intermediate contrast in the regular TPMP ( Fig 3A and 3B ) and MA behavior at high contrast , we conclude that this was also the case in the altered feedback situation . At all contrast values measured here , it must also be considered that at very high yaw torque values as they were measured here , resulting from the optomotor response to the open-loop motion , the closed-loop pattern rotated very fast , which may have led to a decreased motion stimulus of that pattern [18] and therefore to a decrease of its stabilization at low and intermediate contrast . At high contrast , this possible difference in the stimulus strength between the two motion components may have led to a decrease in MA behavior . In any case , by inducing feedback asymmetry in the TPMP , we could confirm that the SPS peak shift observed in the TPMP at intermediate contrast is a result of the motion of the nonstabilized pattern and not of the compound motion of the two motion components . When the two patterns differed in pattern contrast , the flies spent significantly more time stabilizing the high-contrast texture and less time with the low-contrast texture ( Fig 3F ) . A similar effect has been observed when flies were given the choice between differently contrasted vertical bars , where they predominantly chose the higher contrasted one [22] . In the TPMP with varying contrast values , we observed that higher contrast values led to increased MA behavior . With an increase in contrast , the single pattern elements increased in salience . Another way to vary the salience of single pattern elements is to vary pattern element density . To examine this closer , we switched from textures of randomly or evenly distributed dots to evenly spaced vertical bars ( Fig 2D ) and varied their number in the arena ( Fig 4A ) . With 20 bars in each panorama , the behavior of the flies resembled the one observed with the dot textures . The SPS at high pattern contrast ( 91% ) was very low , even slightly lower than with the random dots patterns . MA behavior was abundant . At 37% contrast , the two behaviors occurred about equally often ( Fig 4B ) . With a single bar per pattern , the flies were confronted with a profoundly different stimulus situation , two moving objects , neither of which was likely to reflect self-rotation . One might have expected to find only MA behavior . However , as is well known , flies tend to fixate isolated landmarks [19 , 23] . What we found is that the flies stabilized one bar in the frontal visual field on the side where its bias would move it progressively ( Fig 4C ) . With the data evaluation used in the present study , we found about the same amount of SPS and MA behavior as with the dot patterns ( Fig 3B , Fig 4A ) . Significantly , however , no inverted contrast dependence was observed for SPS with one or two bars per pattern ( Fig 4B ) . With two bars , the time the flies had to fixate one bar before a second bar entered the frontal visual field was correspondingly shorter . Hence , with an increasing number of bars , SPS strongly decreased ( Fig 4A ) . This reached a low plateau stretching from 4–8 bars per pattern , before SPS increased again with 10 bars and more . With 20 bars , no peaks in the position histogram indicated object fixation ( Fig 4D ) . With a single panorama , object fixation could be observed for up to eight evenly distributed bars , as the experiment of Fig 4E shows . There , Fourier transforms were performed on the cumulated position histograms of 20 flies , stabilizing a single pattern with a certain number of evenly spaced bars . As the fixation of individual bars resulted in peaks at the respective points in the position histograms , a Fourier transform would show peaks at the Fourier component corresponding with the number of bars in the pattern . Interestingly , the absence of peaks in the position histogram of a single pattern coincided with the re-increase of SPS in the TPMP ( 10 bars per pattern ) . Taking all these observations together suggests that with few isolated bars and , alternatively , with many bars or textures , SPS was mediated by different mechanisms . Turning to MA behavior , we found a different situation . As Fig 4A shows , MA behavior was most prominent with 4–8 bars per pattern . But short phases of MA behavior could also be observed with only a single bar per pattern , where they interrupted SPS right after the two bars had crossed and were now diverging ( Fig 4F and 4G ) . Altogether , MA behavior with few ( 1–3 ) bars occurred significantly more often when the bars were diverging than when they were converging ( Fig 4H ) , independently of where in the visual field they were previously fixated , although the fact that the bars were predominantly fixated on the side where their bias moved them progressively meant that MA usually occurred when both bars were diverging on the same side of the visual field . As the number of bar crossings increased with the number of bars per pattern , this explained the increase of MA behavior with an increasing number of bars as was found with 1–4 bars ( Fig 4A ) . With more than 8 bars per pattern , however , MA behavior decreased again , suggesting that now the flies no longer responded to the bars as separate objects but rather as the coherently moving elements of larger entities , therefore generating other behaviors but object fixation . One of the well-examined aspects of multi-stable perception in humans is the temporal dynamics of the perceptual alternations . Levelt [24] described that in humans the distribution of percept durations has a distinct right-skewed unimodal shape following a gamma distribution , as was later also found in other studies [25–28] . We wanted to know whether the switching behavior of Drosophila in the TPMP resembled multi-stable perception paradigms in other animals and humans and thus examined its temporal dynamics . One of the characteristic properties of the dynamics in higher animals and humans is its stability over time [25 , 29] . To measure this in flies , we extended the experiment to 6 min and evaluated the trend of SPS over this period . As we wanted to compare the dynamics to bi-stable perception in humans , we used a pattern contrast of 8% . Under that condition , MA behavior was the least frequent . Yaw torque distributions of individual flies performing in this experiment proved to be multi-modal ( Hartigans dip test; S1 Table ) , except for one fly that expressed MA behavior most of the time . All other flies alternated between the different interpretations of the stimulus in the TPMP , particularly between SPScw and SPSccw . No significant changes were observed for the overall time spent with SPS ( Fig 5A ) and for the number of SPS phases per minute ( Fig 5B ) . As an alternative to SPS duration , we also recorded the time between the onset of SPS with one pattern to the onset of SPS with the other pattern , which we termed inter-switch-phase ( ISP ) . The frequency of ISPs was also stable over time ( Fig 5C ) . Mean durations of ISPs were much larger and frequencies much lower than those of SPS phases . This was in part due to short interruptions in SPS phases by MA or “other” behavior , which were scored the same as the regular switches between behaviors . The mean SPS phases and ISPs of individual flies showed considerable variation ( Fig 5D and 5E ) , as is also typical for the alternation process in multi-stable perception [25 , 30–32] . There , the coefficient of variation for the individual mean duration of the percepts typically lies between 0 . 44 and 0 . 75 [32] . In the TPMP , the coefficient of variation for the individual mean duration of the ISPs was 0 . 59 , so it was within that range . As mentioned above , in human multi-stable perception , the distribution of percept durations follows a gamma distribution [25] . Its probability density is given by f ( t|k , λ ) =1λkΓ ( k ) t−ke−tλ The parameters λ and k determine the scale and shape , respectively , of the distribution . The gamma function Γ ( k ) represents the continuous extension of ( k − 1 ) ! . In our study , for a fit between the SPS phase or ISP durations and the gamma distribution above , we normalized the phase durations of all flies to the mean phase duration of each fly and pooled the data of all flies ( Fig 5F and 5G ) . With the SPS phases , we found a very good optimal fit ( R2 = 0 . 84 ) using the parameters λ = 0 . 66 and k = 0 . 42 ( for more details see Materials and methods ) . With the ISPs , the optimal fit with λ = 0 . 42 and k = 0 . 29 was not quite as good ( R2 = 0 . 55 ) , which could be attributed to the lower frequency of these events . The parameters of the two distributions were very similar , which might indicate that they show the same stochastic process . Unlike typically observed with humans and monkeys but as found in experiments on pigeons [6] , our values for k were smaller than 1 , reflecting the highest frequencies for the shortest durations . The pigeon study [6] found their data to also fit a single-parameter exponential function , which provides a more parsimonious model and might also be applicable with our data . Nevertheless , in the TPMP the temporal dynamics of both kinds of phase durations seemed to indicate that switches between behaviors were not tightly coupled to , for instance , a regularly occurring external stimulus but were influenced by a presumably endogenous stochastic process . Tang and Juusola [33] have shown that Drosophila alternates between cw and ccw motion responses , if these stimulus components are presented simultaneously , but each one only to one half of the visual field , a paradigm reminiscent of binocular rivalry . In the TPMP , both motion components were presented to both visual half-fields . Thus , binocular rivalry would not suppress one of the motion components unless the fly would use only progressive or regressive motion [34] for bias compensation . To test for binocular rivalry , we presented the visual motion only to one eye ( Fig 6A ) . We used the regular dots texture ( see above ) . In a first experiment , we presented only one of the patterns in closed loop , with the second one stationary . The flies were confronted with either a regressive or a progressive bias on the open side of the panorama . Using the classification for SPS above , we found a significantly more abundant stabilization of the progressive bias compared to the regressive one , while neither was different from the stabilization of a binocularly presented pattern ( Fig 6B ) . Also in the TPMP with the visual input restricted to only one eye , the flies could stabilize both the progressive and the regressive bias ( Fig 6C ) . Overall , SPS was as pronounced as with both eyes . However , the difference in the stabilization of the progressive and regressive bias was highly significant , whereas there was no difference between SPScw and SPSccw with both halves of the visual system in operation . To stabilize the regressive bias , the flies had to generate yaw torque towards the side where the visual motion input was blocked . With the progressive bias being visible at the same time , turning to the shielded side seemed to be even less attractive than with only the regressive bias . But although individual flies may have exclusively stabilized the pattern with the progressive bias ( S9 Fig ) , we also found flies that stabilized both patterns about equally often ( S10 Fig ) . Therefore , we concluded that the multi-stable behavior in the TPMP did not depend on binocularity , although the lack of binocularity influenced the choice behavior of the flies . Interestingly , MA behavior was significantly reduced with the monocular stimulation compared to the binocular one , which may lead to the assumption that the zone of binocular overlap could have a special relevance for MA behavior . It could also be at least partially dependent on the equivalence of the two opposing motion stimuli , which is not given in this case , because the stimulus strength of the progressively moving pattern was bigger than the one of the regressively moving pattern when both patterns were moving with the same absolute angular velocity . The fraction of time spent less on MA behavior was shifted to “other” behavior .
Perception is the process and result of the identification , organization , and interpretation of sensory stimuli to represent the environment [35] and provide a basis for directed behavior . Multi-stable perception can occur when a constant ambiguous stimulus allows for several interpretations . With the TPMP , we designed an ambiguous stimulus for Drosophila in which the ambiguity is persistent over time but that can still be actively controlled by the fly . The constant ambiguity evoked a multi-stable behavioral response , in which each of the behaviors was coupled to one reference for optomotor balance , i . e . , straight flight . Multi-stable phenomena in humans and nonhuman primates are studied as perceptual multi-stability [5 , 36–38] , which means that they rely on introspection and its report . In Drosophila , we measured behaviors that were functionally coupled to one of three references that defined the respective interpretations of the ambiguous stimulus . This raises the question of what the nature of the underlying ambiguity was . As the fly’s primary objective in the flight simulator without an ambiguous stimulus is to stabilize its flight , it can be assumed to also be so in the TPMP . The differing references for optomotor balance provided in the TPMP competed , leading to the multi-stability . Our results indicate that the TPMP produces two ambiguities . One is whether both motion components occur out in the world or one of them signals self-rotation . If only one component is indeed self-rotation , the second ambiguity is relevant: which components are which . The stimulus ambiguity in the TPMP can also be interpreted in the context of depth perception as either transparent or nontransparent , the latter facilitating MA behavior , because an asymmetry in the feedback alters SPS but not MA behavior ( Fig 3E ) . Also , the ratio between MA behavior and SPS ( Fig 2E , Fig 3B , Fig 4A ) is influenced by different parameters than the ratio between SPScw and SPSccw ( Fig 2F ) . When interpreted as transparent , either the cw- or the ccw-biased pattern can be perceived as the reference for optomotor balance , leading to SPS . The discovery of multi-stability in fly visual behavior using transparent panoramic motion stimuli reveals some interesting properties of Drosophila motion vision and behavior . The perception of transparent motion stimuli can also be multi-stable in humans [10 , 11] , although this has only been shown for two-dimensional transparent motion or plaid motion . A study investigating the processing of plaid motion in blowflies suggested that they express component selectivity in response to the individual moving gratings in a two-dimensional transparent motion stimulus [8] , although the physiological findings of a fly equivalent of pattern and component cells in the lobula plate did not fully explain the behavioral phenotype . The component selectivity of one-dimensional transparent wide-field motion stimuli observed in the TPMP in the form of SPS is not predicted by current models of fly motion vision , according to which motion signals of the same strength and opposite direction should cancel each other , if both are distributed evenly over the visual field [7 , 12 , 13] . Consequently , only the behavior associated with the integration of the two motion components , MA behavior should be found . In the present study , it can be shown that the visual system of Drosophila separates transparent motion stimuli , even if they completely overlap and cover most of the visual field . This does not only occur in closed-loop orientation behavior . Even without visual feedback , the two motion components are received separately and answered by syn-directional yaw torque . Studies of transparent motion vision in humans are usually conducted under gaze fixation , therefore providing much less visual feedback than the TPMP [15 , 39] . So , the open-loop TPMP provides a better comparison to these experiments than the normal TPMP . Here , an interesting difference between fly and human transparent motion vision is revealed: while in humans , the detection of transparency depends on locally unbalanced motion stimuli as they emerge with random dot patterns , this is not the case in Drosophila , because we also observe component selectivity with regular dot patterns—where the local motion is balanced ( S6 Fig , S8 Fig ) . How strongly the component selectivity is expressed in the orientation behavior of the flies in closed loop depends on at least two stimulus parameters , pattern contrast , and pattern element density . An increase in contrast and a decrease in pattern element density both increase the salience of the single pattern elements in a panorama pattern and thereby strengthen its figure features . Both also increase the relative abundance of MA behavior in the TPMP . With widely spaced vertical bars moving incoherently , MA appears to be a consequence of ipsilaterally diverging figures ( Fig 6F–6H ) . At intermediate contrast , with increasing density of the bars , MA behavior starts to decrease at a bar distance where fixation responses are no longer apparent . Parsimony would suggest that , as both high pattern contrast and high pattern element distance favor MA behavior in the TPMP and MA behavior at low pattern element density is elicited by figure responses , MA behavior is generally a consequence of figure detection and responses . SPS would then be a behavioral response to elementary wide-field motion in the absence of figure detection . This conjecture is supported by the fact that , regardless of the contrast level and the pattern element density , panorama pattern elements with stronger figure features , like vertical bars , which evoke edge detection [40] , result in higher MA behavior values than panorama pattern elements with weaker figure features , like vertical rows of dots ( Fig 2E , S8B Fig , Fig 4B ) . Moreover , the contrast effect only occurs when the figure features are not already very high due to a low pattern element density ( Fig 4B ) . It has been observed before that , when given the choice between stabilizing a background motion and fixating a bar , flies choose to fixate the bar [19] . Extrapolating this to the TPMP , the figure motion response resulting in MA behavior suppresses the elementary wide-field motion response , which would result in SPS . The weaker the figure features of the patterns become , the more often SPS is the preferred behavior . Therefore , we conclude that the rivalry between MA behavior and SPS is essentially one between the figure and the wide-field motion systems . It also means that figures can be detected by Drosophila on a much smaller spatial scale than previously assumed [20] . Many studies have been dedicated to the difference between figure motion and elementary wide-field motion vision in Drosophila , inquiring how the fly discriminates between the two and how it reacts if both are presented simultaneously [19 , 41–43] . Yet there has only been little research as to how the fly responds to transparent wide-field motion components when they do not differ in their pattern properties . Also , it remains unclear what defines a figure or wide-field motion stimulus as such , given that the motion stimulus of a panorama pattern always must consist of an array of objects , no matter the shape , size , or number of pattern elements . When shown a single , dark , vertical bar , a figure , on a light background in closed loop , the fly will fixate it in the frontal part of its visual field [19] . This behavior is supported by elementary motion vision , but is essentially independent of it [44 , 45] . With a panorama pattern containing numerous pattern elements , but no salient singularities , the fly will also show stabilization behavior of this pattern [19] , independent of the position of a particular pattern element . Lately , several studies have characterized how flies process figure motion [16 , 41 , 46] , pointing out that usually , moving figures possess both figure and elementary motion features , which are distinguished in the fly visual system . Interestingly , for a certain type of motion-blind flies it can be shown that their most basic motion response , their optomotor reflex to wide-field patterns , is completely abolished at low contrast [44] , but a very low , residual response seems to remain at intermediate [44] and high contrast [47] . This response is likely generated by the system processing figure motion [44] . This interpretation of the flies’ behavior does also make sense from an ecological point of view: If a fly interprets the stimulus in the TPMP as an array of objects , the most logical solution to maintain a straight trajectory is to integrate the motion components of the individual pattern elements . If it is interpreted as two panoramic flow-fields , which might both represent a stationary background , it makes more sense to choose a single one as a reference for straight flight . Previous studies examining a flies’ alternations between different responses to a constant stimulus situation [33 , 48–50] can ultimately all be explained by spatially selective visual attention , because the competing stimuli were presented in differing parts of the visual field . In human perceptual rivalries , attention is linked to multi-stability , but ultimately independent of it [2] . In the TPMP , the alternations between the two components of the compound visual stimulus in the TPMP are not a consequence of spatially selective visual attention or binocular rivalry , because we can show that they also occur without binocular stimulus presentation ( Fig 6 ) . The decrease in MA behavior we observe with monocular stimulus presentation might be a result of the strongly reduced visual input in the frontal part of the visual field of the noncovered eye , as MA behavior appears to be a kind of figure response and the response to figure motion components has been found to be stronger in the frontal part of the visual field [16] . But , as with SPS of the regressively biased pattern , it is not completely abolished , which represents another similarity of the multi-stable behavior of the fly in the TPMP to multi-stability in humans . There , if the stimulus strength and therefore the likelihood for one of the percepts is decreased , so is the time this percept is perceived , but while the ambiguity remains , however weak , the alternations remain [51] . The same can be said for the other parameters , like bias distribution , pattern contrast , and pattern element density , which influenced the choice behavior of the fly in the TPMP . When the likelihood for one of the interpretations of the transparent motion stimulus to be the “correct” one was increased , so was the time the flies spent with the corresponding behavior , and the time with the behaviors corresponding to the “less likely” interpretations was accordingly decreased , but the behaviors corresponding with the less likely interpretations were never fully abolished . In humans , visual multi-stable phenomena are considered an indication for vision being an active process . This means that the interpretation of a visual stimulus is shaped by the brain as well as by the stimulus , particularly in situations where a stimulus can be interpreted in more than one way [2 , 52 , 53] . As already mentioned in the introduction , also in the natural environment of a fly , visual stimuli may not always be unambiguous and its visual system needs a way to derive the optimal interpretation without getting stuck on a potentially wrong one . The temporal dynamics of the behavioral alternations of Drosophila in the TPMP also share most of the characteristics of human multi-stable perception . The stochastic component in the temporal activation pattern of the different behaviors in the TPMP suggests that the behaviors were activated endogenously . So far , this claim has been difficult to prove , but mechanisms providing such stochasticity are now being investigated [54 , 55] . To summarize , when Drosophila in stationary flight is exposed to transparent wide-field motion stimuli , its orientation behavior is multi-stable . This shows that Drosophila can process the individual components of a one-dimensional transparent motion stimulus separately and that this kind of stimulus is ambiguous to the fly . To what extent the fly expresses component selectivity depends on several properties of the stimulus , namely pattern contrast and element density . As component selectivity increases with a decrease of the figure features of the stimulus , we conclude it to be the result of wide-field motion vision in the absence of figure detection . The alternations between the different behaviors exhibit a stochasticity reminiscent of the temporal dynamics in human multi-stable perception .
Flies were cultured at 25°C on standard food medium [56] on a 12 h light/dark cycle with 60% relative humidity . Wild-type flies were of the Wildtype Berlin ( WTB ) strain . For tethering , 2- to 3-day-old female flies were cold-anesthetized and glued with dental composite ( ESPE Sinfony DO3 , 3M , Neuss , Germany ) to a copper rod ( Ø = 0 . 15 mm , length = 2 mm ) using a micromanipulator . The tip of the rod was positioned between the flies’ head and thorax to exclude independent motion of these two body parts . The glue was polymerized with blue LED light ( 10 s pulse , distance < 5 mm ) , and the flies were then kept isolated with access to water for 2 to 14 h prior to the experiment . Visual stimuli were presented in a cylindrical arena via fiber optics , and 32 x 180 lightguides connected the inner surface of the arena ( Ø = 90 mm , h = 90 mm ) with the rectangular frontplate . The arena covered 360° x +/−45° of the flies’ visual field . Computer-generated visual stimuli were displayed on a screen placed directly onto the frontplate . The visual stimuli were controlled by custom-made software written in VB . NET . The rod glued to the fly was positioned in the tip of a syringe , which was then attached to the torque meter and centered in the arena . In the flight simulator , the torque meter transduced yaw torque into an electrical voltage , which was read by a computer by using a data acquisition device ( USB-1208 FS; Measurement Computing , Germany ) . The angular motion of the visual panorama was calculated online from the yaw torque signal and then displayed in real time to the fly . Contrast values of the used patterns ranged between 4% and 91% . The RGB values of the white background input were always the same; however , the scattered radiation from the darker pattern elements , or lack thereof , resulted in illuminance values between 42 and 45 lux for the background . Luminance values of the darker pattern elements ranged between 2 lux and 40 lux , resulting in contrast values between 91% and 4% , measured as ( Ev ( max ) -Ev ( min ) ) / ( Ev ( max ) + Ev ( min ) ) . Random dots patterns ( Fig 2D , adapted from Letratone Sheet LT131 [Letraset] ) , regular dots patterns ( diameter , d = 7° , 20 columns , 6 horizontal rows ) or evenly spaced vertical stripes ( width , w = 6° ) were used as visual stimuli . For closed-loop experiments , the coupling coefficient was set to −5 ( i . e . , a yaw torque of 1 × 10−10 Nm ) , and resulted in an angular displacement of 5° against the direction of yaw torque . Unless otherwise stated , the rotatory bias used in the closed-loop experiments was set to ±20° per s . For the optomotor balance controls with just one pattern in closed loop the sign of the rotatory bias was randomly set to either positive or negative for each fly . In the incoherent motion paradigm , one bias value was set to 20° per s , the other to −20° per s . Unless otherwise stated , the experiment duration for the closed-loop experiments was set to 3 min . For the monocular condition , the stimulus input to one eye was eliminated by positioning a white virtual screen covering the arena from −180° to 20° or from −20° to 180° , respectively , over the panorama stimuli . The 20° on the noncovered side accounted for the 15° binocular overlap per side , plus a 5° error margin for positioning the fly in the arena to guarantee exclusive stimulus presentation to one eye . The optomotor response was tested by rotating one pattern around the fly with a second one stationary to provide the same contrast conditions as in the closed-loop experiments . The pattern was alternately rotated cw and ccw for 9 times for 15 s . The optomotor response was tested at three different angular velocities , 20° per s , 40° per s , and 60° per s . In the open-loop TPMP , the same settings were used as in the normal , closed-loop TPMP , but the feedback was switched off . Flies were tested under up to five stimulus conditions , provided they did not stop flying throughout the trial . If they were tested under more than one condition , the order of the stimulus conditions was randomized . If a fly stopped flying for more than three times throughout a trial , the trial was aborted . For the experiment with monocular visual stimuli , all flies were tested under every stimulus condition . Yaw torque recordings were stored on the measuring PC hard disc with a sampling rate of 20 Hz and evaluated after the experiment with custom-made software written in VB . NET . Controls with an asymmetric bias showed that the flies were prone to prefer the pattern with the lower bias ( Fig 2F ) . Because an out-of-alignment setting of the zero yaw torque value had the same effect as an asymmetric bias , flies that showed a strong bias towards one of the patterns ( > 75% of the time on one side ) under closed-loop conditions were excluded from the data evaluation . Following this criterion , dependent on pattern contrast , between 5% ( 91% contrast ) and 20% ( 8% contrast ) of the flies had to be excluded from evaluation . For the evaluation of the temporal dynamics experiment , all flies that stopped flying throughout the experiment were excluded from evaluation . The closed-loop experiments were evaluated by calculating the moving average over 2 s of the yaw torque values . The incidences of the resulting values within the range ( < ±2 × 10−10 Nm for MA behavior; < -2 × 10−10 Nm and > −6 × 10−10 Nm or > 2 × 10−10 Nm and < 6 × 10−10 Nm for SPS , respectively ) of a behavioral category were then reported as a percentage of the entire experiment time . For the evaluation of the optomotor response , the first 15-s period was discarded and the other 8 periods were averaged . As we found no difference in the optomotor response to the different angular velocities measured , they were pooled for further evaluation . The overall optomotor response was reported as the mean yaw torque of the last 5 s of the 15-s periods . Data were tested for normal distribution using a D’Agostino-Pearson omnibus normality test . When they were normally distributed , a Student t test was used to test two groups against each other . When no normal distribution could be assumed , a Mann-Whitney test was used to compare two groups . Comparison of more than two groups was achieved by a one-way ANOVA with Tukey’s multiple comparisons test when the data were normally distributed , and with a Kruskal-Wallis test with Dunn’s test for multiple comparisons when they were not normally distributed . Bonferroni corrections were used for multiple comparisons . The curve fitting was done with the method of least squares estimation . Statistical significance was demonstrated as ***p < 0 . 001 , **p < 0 . 01 , *p < 0 . 05 , and nonsignificant ( ns ) p > 0 . 05 . | Vision is considered an active process in humans and higher animals in which the stimulus is interpreted by the subject and can be perceived in different ways if it is ambiguous . We aimed to find out whether this also holds for lower animals , such as the fruit fly Drosophila melanogaster . To provide ambiguity , we exposed flies to transparent motion stimuli in a flight simulator and found their behavior to be multi-stable . These results show that the visual system of the fly can separate the individual components of a transparent motion stimulus , and that this kind of stimulus is ambiguous to the fly . The extent to which the fly shows component selectivity in its behavior depends on several properties of the stimulus , like pattern contrast and element density . The alternations between the different behaviors exhibit a stochasticity reminiscent of the temporal dynamics in human multi-stable perception . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"velocity",
"invertebrates",
"classical",
"mechanics",
"social",
"sciences",
"neuroscience",
"animals",
"animal",
"models",
"perception",
"drosophila",
"melanogaster",
"model",
"organisms",
"cognitive",
"psychology",
"experimental",
"organism",
"systems",
"vision",
"drosophila",
"research",
"and",
"analysis",
"methods",
"behavior",
"insects",
"torque",
"physics",
"arthropoda",
"psychology",
"eukaryota",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"sensory",
"perception",
"cognitive",
"science",
"organisms",
"motion"
]
| 2018 | Multi-stability with ambiguous visual stimuli in Drosophila orientation behavior |
Cytokine signaling is responsible for coordinating conserved epithelial regeneration and immune responses in the digestive tract . In the Drosophila midgut , Upd3 is a major cytokine , which is induced in enterocytes ( EC ) and enteroblasts ( EB ) upon oral infection , and initiates intestinal stem cell ( ISC ) dependent tissue repair . To date , the genetic network directing upd3 transcription remains largely uncharacterized . Here , we have identified the key infection-responsive enhancers of the upd3 gene and show that distinct enhancers respond to various stresses . Furthermore , through functional genetic screening , bioinformatic analyses and yeast one-hybrid screening , we determined that the transcription factors Scalloped ( Sd ) , Mothers against dpp ( Mad ) , and D-Fos are principal regulators of upd3 expression . Our study demonstrates that upd3 transcription in the gut is regulated by the activation of multiple pathways , including the Hippo , TGF-β/Dpp , and Src , as well as p38-dependent MAPK pathways . Thus , these essential pathways , which are known to control ISC proliferation cell-autonomously , are also activated in ECs to promote tissue turnover the regulation of upd3 transcription .
The digestive tract is uniquely challenged by its high degree of exposure to the external environment . The transit of nutrients through the gastrointestinal ( GI ) tract is accompanied by frequent introduction of biotic and abiotic stresses . In particular , digestive tissue is constantly exposed to a high density of microbes , including benign microbiota and invasive pathogens [1] . The gut epithelium performs a multifaceted role in maintaining the barrier between the host and its environment through immune responses and the maintenance of a continuous cellular monolayer [2] , while digesting and absorbing nutrients . Preservation of epithelial integrity in the GI tract requires continual tissue turnover by coordinated shedding of epithelial cells along with division and differentiation of intestinal stem cells ( ISCs ) [1 , 3] . Disorders in epithelial regeneration or intestinal immunity lead to intestinal maladies including inflammatory bowel disease ( IBD ) and colorectal cancer [4] . Cytokines , which are central to gut homeostasis , are produced by epithelial and immune cells to properly orchestrate immune and repair responses [2 , 3] . The control of cytokine signaling in the digestive tract is complex , and characterizing the regulators of cytokine expression is a critical step towards fully understanding the mechanisms underlying intestinal homeostasis . Drosophila melanogaster has emerged as a powerful model to study gut homeostasis , epithelial immunity and ISC regulation [1 , 5] , and acts as a model for intestinal infection and pathology [6] . Like the mammalian intestine , the midgut of Drosophila contains ISCs that divide and differentiate to replace the absorptive , polyploid enterocytes ( ECs ) and secretory enteroendocrine cells ( EEs ) [5] . During division , midgut ISCs self-renew and give rise to a pool of transient , differentiating precursor cells called enteroblasts ( EBs ) , which terminally differentiate into ECs . Similarly , EE cells are replaced via ISCs that divide and give rise to pre-EE progenitors [7] . Also like the mammalian intestine , the Drosophila midgut is regionalized . Specifically , it can be divided into five main regions: the cardia ( at the foregut-midgut junction ) , R1 and R2 composing the anterior midgut , R3 also known as the copper cell region , and R4 and R5 that constitute the posterior midgut [8 , 9] . In response to infection by microbial pathogens or , to a lesser extent , ingestion of dietary microbes , the midgut activates multiple layers of innate immunity . Among these are the induced synthesis of reactive oxygen species ( ROS ) by the NADPH oxidases Dual oxidase ( Duox ) and NADPH oxidase ( Nox ) , and the production of antimicrobial peptides under the regulation of the immune deficiency ( Imd ) and JAK-STAT pathways [10–13] . Imd pathway activation is triggered by the detection of bacteria via peptidoglycan recognition receptors ( PGRP-LE and PGRP-LC ) [14 , 15] while JAK-STAT pathway activation results from the expression and secretion from the gut epithelium of Drosophila IL-6 family cytokines: Unpaired 3 ( Upd3 ) and Unpaired 2 ( Upd2 ) [16] . In addition to immune activation , enteric infections also stimulate EC delamination and tissue turnover resulting in ISC-dependent tissue repair [12 , 17 , 18] . This regenerative process has been shown to depend strongly upon the activation of multiple pathways in progenitor cells , including the Hippo , Wingless , JAK-STAT and EGFR pathways [13 , 17 , 19 , 20] . Bacterial infection , as well as genetically induced apoptosis in ECs , triggers the transcription and secretion of Upd3 in ECs and EBs [13 , 17] , which subsequently initiates a homeostatic feedback loop and ultimately activates ISC-mediated regeneration . The Upd cytokines activate the JAK-STAT pathway in progenitor cells and visceral muscles , which in turn stimulates the release of epidermal growth factors ( EGFs ) by these cells [19 , 21 , 22] . Upd3-dependent secretion of the Epidermal Growth Factors ( EGFs ) Vein from visceral muscle and Spitz from EBs stimulates the EGFR pathway in ISCs to promote proliferation . Upd3-mediated JAK-STAT activity is also required to promote rapid EB differentiation , thus accelerating epithelium turnover upon infection [13 , 17] . Cytokines , such as Upd3 , therefore act as master regulators of intestinal homeostasis , as they are both required and sufficient to trigger immunity and tissue repair . Accordingly , the loss of Upd3 increases susceptibility to enteric infections , while ectopic induction of Upd3 induces dysplastic lesions in the gut [13 , 16] . However , a detailed knowledge of upstream enteric stress sensors as well as the downstream transcriptional regulatory network controlling Upd3 production in ECs remains elusive . In this study , we initiated analysis of the transcriptional regulation of upd3 , the primary cytokine responsible for inducing ISC proliferation and midgut renewal . We first identified two microbe-responsive enhancer sequences in the upd3 gene that direct its expression in ECs , and an additional enhancer that regulates upd3 induction in progenitor cells . A subsequent EC-specific RNAi knockdown screen of all the Drosophila transcription factors ( TFs ) was performed to determine which TFs govern the activity of the central infection-responsive enhancer region . From this screen , we identified 39 TFs required for enhancer induction , and 103 TFs that triggered aberrant induction when knocked down . This study was complemented by an in vitro , yeast one-hybrid screen as well as bioinformatic analyses of the enhancer sequence to identify TFs that may act as direct regulators of upd3 expression . Notably , we identified the Yorkie ( Yki ) /Scalloped ( Sd ) complex , the AP-1 complex ( D-Jun and D-Fos ) , Mad and Snail ( Sna ) as key regulators of upd3 transcription . We proceeded to explore the upstream regulatory pathways that control the activity of these major TFs . We determined that transcriptional induction of upd3 in ECs requires the Mitogen Activated Protein Kinases ( MAPKs ) p38b and D-ERK , downstream of Src oncogene ( Src ) Family Kinases ( SFKs ) and Raf , which converge on AP-1 activation . Surprisingly , the Stress Activated Protein Kinase ( SAPK ) cascade seems to be necessary for only a minimal portion of AP-1 function in ECs . In addition , a Misshapen ( Msn ) -Warts ( Wts ) -Yki/Sd pathway , independent of Hippo ( Hpo ) , is essential for full upd3 expression . Finally , we found that the Decapentaplegic ( Dpp ) pathway is also required for upd3 induction in ECs . Altogether , these results improve our understanding of the complex regulation of midgut tissue renewal by identifying the key TFs and pathways that control cytokine signaling in the intestinal epithelium in response to infection .
Upon oral infection by entomopathogenic bacteria like Erwinia carotovora ssp . carotovora 15 ( Ecc15 ) or Pseudomonas entomophila ( Pe ) , Upd3 acts as a signal to trigger antibacterial and reparative host responses [17 , 23] . We characterized this response through RT-qPCR measurements of midgut upd3 expression , taken over the course of a week following ingestion of Ecc15 or Pe . We found that upd3 transcription was strongly induced in response to ingestion of these pathogens and peaked between 8-24h post-infection before returning to basal levels within 96h ( S1A and S1B Fig ) . In addition , peak expression of upd3 , as well as the time that it takes to return to basal expression , increases with bacterial dose ( S1A and S1B Fig ) . These results demonstrate that upd3 is regulated by infection at the transcriptional level and varies with the amplitude of the given threat . As an initial step to characterize upd3 regulation in the digestive tract , we sought to identify the key enhancer regions that control its expression , especially its induction in response to pathogens . To this end , we generated twenty-one GFP transcriptional reporters covering the entire upd3 locus . Overlapping fragments of ~1–1 . 5Kb were cloned upstream of a GFP reporter , starting from 4 . 2Kb upstream of the upd3 start site and ending 7 . 3Kb downstream of the gene . Reporters were designated upd3-A-GFP through upd3-R-GFP ( Fig 1A , S1 Table ) . We first evaluated the transcriptional activity of these reporters both in unchallenged ( UC ) and orally infected flies . Seven lines gave no detectable signal in the digestive system under any condition ( enhancers A , D , F , J , N , O1 , O2 , see Fig 1A ) . The remaining enhancer regions were divided into five categories based on their expression profile: seven enhancer regions drove GFP expression constitutively , with little change in response to infection by Ecc15 . 1 ) For five of these lines , the signal was limited to specific regions of the gut , including the foregut and hindgut ( upd3-H-GFP ) , the foregut only ( upd3-K-GFP ) , the hindgut only ( upd3-P1-GFP ) , and the copper cell region ( upd3-G-GFP and upd3-P2-GFP ) ( S1C Fig ) . 2 ) The remaining two constitutive enhancer regions ( upd3-M-GFP and upd3-Q-GFP ) are active throughout the midgut in populations of small cells ( S1D Fig ) . Interestingly , cells expressing upd3-M-GFP accumulate upon infection ( Fig 1B ) . Overlap of the upd3-M-GFP signal and immunostaining of the progenitor marker esg-lacZ [24] revealed that the upd3-M-GFP reporter is specific to ISCs and EBs , and that the increase in total signal upon infection is thus secondary to progenitor cell proliferation ( Fig 1B ) . 3 ) Two additional enhancer regions ( upd3-E-GFP and upd3-E0F-GFP ) drove GFP expression in sporadic ECs of the R2 and R4 midgut segments upon infection ( S1E Fig ) . 4 ) One enhancer drove inducible upd3 expression only in the salivary glands ( upd3-L-GFP , S1F Fig ) . 5 ) Finally , we identified four infection-responsive enhancer regions , which show little or no GFP signal in UC conditions , but are activated upon infection: these include two overlapping regions of the upd3 promoter ( regions B-C ) , region I , and region R ( Fig 1A ) . Enhancer lines upd3-B-GFP , upd3-C-GFP , and upd3-I-GFP express GFP exclusively in ECs during infection , as shown by co-immuno-staining of GFP and the EC marker Myo-lacZ ( Fig 1C and 1D and S1G Fig ) . The upd3-I-GFP signal was stronger in the copper cell region and less consistent in the rest of the midgut . In contrast , upd3-R-GFP shows activity upon infection only in the ISC and EB cells , marked by esg-lacZ ( Fig 1E ) . Of note , the expression patterns identified in our study recapitulate the known upd3 signaling dynamics in the gut , including induction in ECs and progenitor cells upon stress [13 , 17 , 23] , as well as robust local expression in the middle midgut and in the cardia [8] , suggesting that we adequately captured the complexity of upd3 regulation . Altogether , these results indicate that upd3 expression is controlled by several classes of enhancers , including microbe-responsive and region and/or cell type-specific regulators . Upd3 acts as a major regulator of intestinal epithelial renewal and its expression is induced by a diversity of enteric stresses , not only limited to bacterial infections [16] . For instance , feeding bleomycin ( bleo ) , which induces gut epithelial cell loss , or dextran sulfate sodium ( DSS ) that disrupts basal membrane , induces upd3 transcription in the gut ( Fig 2A ) and promotes intestinal epithelial turnover ( Fig 2B ) [25] . Furthermore , basal levels of upd3 expression and subsequent tissue turnover have been shown to be regulated by the microbiota [16 , 26] . We confirmed that the guts of germ-free ( GF ) flies express a lower degree of upd3 than conventionally raised ( CR ) flies ( Fig 2C ) . These results suggest that the regulation of upd3 expression integrates signals from multiple stimuli , including various intestinal injuries and even benign gut microbes . We next examined whether these diverse stimuli all activate the microbe-responsive enhancers that we had previously identified . To this purpose , we fed upd3-C-GFP , upd3-I-GFP , and upd3-R-GFP flies damaging bacteria ( Ecc15 and Pe ) and harmful chemicals ( DSS and bleo ) at doses that trigger comparable epithelium renewal rates . Upd3-C-GFP induced GFP expression in response to every treatment except DSS ( Fig 2D ) . Enhancer region R responded to Ecc15 , Pe , bleo , and weakly to DSS by inducing GFP in progenitor cells ( Fig 2E ) . In upd3-I-GFP flies , a GFP signal was only detected upon infection with Ecc15 , and mostly in the copper cell region , while little signal was detected in response to Pe and no significant signal was observed in response to bleo or DSS treatment ( Fig 2F ) . Our findings imply that different stresses ( i . e . DSS vs other stressors ) may be interpreted through distinct cellular mechanisms and thus stimulate cytokine production via separate enhancers . They also suggest that all stressors that affect ECs ( Ecc15 , Pe , bleo ) stimulate upd3 expression mainly through enhancer region B-C . We next investigated whether the infection responsive enhancers C and R also react to the presence of microbiota . To this end , we generated CR and GF upd3-C-GFP and upd3-R-GFP flies and monitored their levels of GFP ( Fig 2G and S2A and S2B Fig ) . The basal GFP signals of CR flies is already very low with few GFP-positive cells detectable microscopically per midgut , rendering qualitative analysis challenging . We therefore estimated enhancer C and R activity by quantifying GFP levels by RT-qPCR . This revealed a significant reduction in enhancer C and R-driven GFP expression in GF midguts compared to CR ones ( Fig 2G ) . Our results demonstrate that both indigenous and pathogenic bacteria , as well as chemical stressors like bleo , all regulate upd3 expression through enhancers C and R , albeit to differing degrees . Altogether , these data suggest that enhancers C and R are microbe-responsive and act as stress sensing enhancers . We next aimed to identify the molecular mechanisms that control upd3 transcription in response to infection . As upd3 transcription is induced by infection in both ECs ( enhancers B-C and I in Fig 1C and 1D , S1G Fig and [12 , 16 , 17] ) and EBs ( enhancer R in Fig 1E and [23] ) , we began by determining which cell type contributes the most to global upd3 production in the midgut upon infection . RT-qPCR analysis of upd3 expression in guts in which upd3 was knocked-down by RNAi in ECs ( Myo-Gal4TS>UAS-upd3-IR ) or EBs ( Su ( H ) -Gal4TS>UAS-upd3-IR ) confirmed that ECs are the principal source of upd3 in the gut upon infection with Ecc15 or Pe ( S3A and S3B Fig ) . In agreement with this , knockdown of upd3 in ECs strongly reduced ISC proliferative activity ( S3C Fig ) . This suggests that the key enhancers controlling the levels of Upd3 in the gut are those functional in ECs ( regions B , C and I ) . As upd3-I-GFP responds only moderately to infection by Ecc15 , but not Pe ( Fig 2F ) , we decided to focus on enhancer regions B and C , which respond strongly to infectious bacteria and cellular stress . To further investigate the importance of the B-C enhancer region in activating upd3 expression in response to infection we created two new reporter lines , one that comprises the entire upd3 locus ( all enhancers included ) and encodes an NLS-GFP-tagged Upd3 protein ( full locus ) ( S4A Fig ) , and one in which the B-C sequence was deleted from the full locus ( full locus– ( B+C ) ) ( S4A Fig ) . While the complete upd3 sequence was able to direct an infection-induced GFP signal in the midgut , deletion of the B-C region eliminated all signal ( S4B Fig ) , demonstrating that enhancers B and C are central to upd3 regulation . In addition , quantification of upd3-C-GFP signal revealed that the kinetics of GFP induction upon infection is in accordance with total gut upd3 expression ( S1A and S3D Figs ) . Finally , the promoter of the upd3 reporter construct , upd3-lacZ , which covers regions B and C ( Fig 1A ) , drove a strong and consistent signal in the same cells that are marked by upd3-C-GFP ( S3E Fig ) . We conclude that the regulation of enhancer regions B and C ( and thus of upd3-lacZ ) is sufficient to induce upd3 with a faithful EC expression pattern during enteric infection . In order to identify the key regulators of upd3 acting through enhancers B and C , we initiated a comprehensive set of in vivo , ex vivo and in silico screens . First , a functional RNAi screen was performed by driving RNAi-mediated knockdown of 632 TFs ( 84% of all known and predicted TFs of D . melanogaster ) using all available UAS-RNAi transgenic lines of the TRiP collection ( Transgenic RNAi Project , Fig 3A ) [27] . The Gal4/Gal80TS system ( Myo-Gal4TS , upd3-lacZ ) allowed us to express RNAi specifically in the ECs of adult flies , thus minimizing developmental or systemic side effects . When available , two different UAS-RNAi lines were tested ( see S2 Table ) , bringing the total number of lines to 755 . Following one week of RNAi induction , five guts were dissected from both unchallenged ( UC ) and Ecc15 orally infected flies , and ß-galactosidase enzymatic activity levels were measured as a read-out of upd3 induction . F1 progeny ( Myo-Gal4TS , upd3-lacZ>UAS-RNAi ) with upd3-lacZ activity that was , compared to controls , increased or decreased by 40% upon infection and/or increased or decreased by 50% in UC conditions ( see methods section and S5A and S5B Fig ) were selected as positive hits . We further estimated the strength of the positive hit phenotypes by calculating their z-score compared to the entire population of crosses tested under the same conditions ( UC or Ecc15 infected ) ( S2 Table ) . Based on these criteria , we identified 149 lines with significantly altered upd3-lacZ expression in either challenged or unchallenged conditions . Positive hits were retested at least twice and 138 TFs were found to significantly alter upd3-lacZ expression when suppressed ( Fig 3A and 3B , S2 Table ) . Specifically , RNAi against 17 TFs in ECs resulted in reduced basal upd3-lacZ in UC flies , and knockdown of 66 TFs increased upd3-lacZ under the same UC conditions ( Fig 3A and S5C and S5D Fig ) . Furthermore , 24 TFs seemed required for upd3-lacZ expression upon infection while RNAi against 53 TFs increased Ecc15-induced upd3-lacZ activity ( Fig 3A and S5C and S5D Fig ) . These results indicate that the knockdown of many TFs results in upd3-lacZ induction rather than inhibition . This is in agreement with the fact that disrupting gut homeostasis by modulating key TFs such as GATAe , Ptx1 , Activating transcription factor 3 ( Atf3 ) , X box binding protein-1 ( Xbp1 ) , either in normal or stressed conditions , can indirectly result in higher expression levels of upd3 [8] . Based on EC-specific transcriptomic data obtained by Fluorescence-Activated Cell Sorting ( FACS ) of ECs coupled to RNA-seq , we established that 92% of TFs identified as positive hits by our screen are expressed in ECs ( RPKM ≥ 0 . 1 ) and 63% of the TFs required for upd3-lacZ expression are transcriptionally regulated ( fold RPKM induction ≥ 1 . 5 or ≤ -1 . 5 ) upon Pe infection ( S5E Fig ) [28 , 29] . This indicates that most of the TFs identified as upd3 regulators by our screen are expressed in ECs and regulated upon enteric infection , and serves as an indirect control of our screen quality . Surprisingly , TFs that alter upd3-lacZ expression in basal conditions or upon infection are poorly correlated with one another ( R2 = 0 . 24 , S5F Fig ) , suggesting that different mechanisms regulate upd3 expression in basal homeostasis and upon infection . Interestingly , positive hits in our screen were enriched for TFs involved in animal development and tissue growth rather than stress or immune responses , again suggesting that epithelial morphogenesis and dynamics are critical to upd3 regulation ( S5G Fig ) . Altogether , our functional genetic screen identified multiple TFs that have the capacity to modulate the expression of upd3-lacZ , particularly in response to infection . RNAi knockdown of TFs in ECs can influence upd3-lacZ expression in multiple ways: TFs could be acting via direct regulation of the upd3 promoter region , indirect regulation through secondary genes or even non-cell-autonomously through changes in gut physiology that subsequently alter upd3 expression . To complement our RNAi screen and identify the direct regulators of upd3 transcription , we thus undertook two parallel approaches . First , we performed a yeast one-hybrid screen to assess the direct interaction between the upd3 promoter and all Drosophila TFs ( Fig 3A’ ) . This additional screen identified 81 yeast one-hybrid-positive TFs ( S3 Table ) . Among these , 21 ( more than 25% ) showed altered upd3-lacZ expression when knocked down , suggesting a role in upd3 gene regulation ( Fig 3B ) . To further indicate the binding potential of TFs of interest , an in silico search for known TF-binding sites ( TFBS ) was performed in the same genomic region using the JASPAR and RedFly databases ( Fig 3B and 3C and S2 Table ) [30 , 31] . We identified seven TFs that are positive for all three approaches , thus specifying them as direct regulators of upd3: D-Fos or kayak , sd , Trithorax-like ( Trl ) , pangolin ( pan ) , giant , Ptx1 , and achintya ( achi ) ( Fig 3B and 3C ) . Knockdown of two of these TFs caused abnormal induction of upd3-lacZ ( Ptx1 and achi ) . The five others were found to be required for upd3-lacZ expression either basally ( giant ) or both during infection and in basal conditions ( D-Fos , sd , Trl and , pan ) ( Fig 3C ) . Of note , Sd and D-Fos have multiple binding sites in infection-responsive enhancers ( Fig 3D ) , and are critical for upd3 transcription in both UC and infected conditions ( Fig 3C ) . We therefore propose that these TFs act as direct , master regulators of upd3 expression in the gut . Next , we examined TFs that strongly alter upd3-lacZ expression upon knockdown , but lack evidence for binding potential to the upd3 promoter region . These important TFs required for upd3-lacZ induction include: Sna , a key regulator of epithelial to mesenchymal transition ( EMT ) ; Jra ( D-Jun ) , the partner of D-Fos in the AP-1 transcriptional complex; Yki , the transcriptional partner of Sd in the Hippo pathway; Mad , a transcription factor that mediates TGF-β/Dpp signaling; and one thus far uncharacterized TF ( CG33213 ) ( Fig 3C and S2 Table ) . Surprisingly , we found that the homeodomain TFs , Retinal Homeobox ( Rx ) and Ultrabithorax ( Ubx ) are also required for upd3-lacZ activity , primarily upon infection ( Fig 3C ) , suggesting these TFs could be involved in tissue repair . Among these TFs , Sna , Mad , Rx and Ubx were not found to bind to the upd3 promoter by the yeast one-hybrid assay , although there are some binding sites in the upd3 promoter region for these TFs according to the JASPAR database ( Fig 3C ) . This suggests that there is a possibility that they could act directly . Finally , global regulators of transcription such as the transcriptional corepressor CtBP , the H3K4 methyl-transferase Trithorax-Related ( Trr ) and MBD-like , a member of the NuRD complex also influenced the regulation of upd3-lacZ ( Fig 3C ) . On the opposite side of the spectrum , we also identified TFs that cause increased upd3-lacZ expression when knocked-down . For instance , Ptx1 , a master regulator of middle midgut identity , has TFBS sites in upd3 , interacts with upd3 in the one-hybrid screen and its knockdown strongly induces upd3-lacZ in both UC and infected guts ( Fig 3C ) [28] . This indicates that Ptx1 could act as a direct negative regulator of upd3 in the middle midgut . The TFs Anterior open ( Aop ) , Cyclic-AMP response element binding protein-17A ( CrebB-17A ) , Longitudinals lacking ( Lola ) , Atf3 , and Achi also show potential to bind to the upd3 promoter region in our one-hybrid screen and trigger upd3-lacZ induction when depleted in ECs . RNAi against GATAe , Xbp1 , deformed wings ( dwg ) , and hangover ( hang ) results in elevated levels of upd3-lacZ in both Ecc15 infected and UC conditions , but the absence of TFBS and association in our one-hybrid screen suggests that this is likely an indirect effect due to disruption of intestinal homeostasis . We also found a distinct set of epigenetic factors that strongly increase upd3-lacZ activity when knocked-down . Among these , there are known positive regulators of transcription such as MBD-R2 ( NSL complex ) ; the Tip60 acetylase; the histone acetyl-transferase Chameau ( Chm ) , Domino ( Dom ) of the SWI-SNF complex and Trl of the eponymous TRL complex . In summary , our combination of in vivo , in vitro , and in silico screens allowed us to identify putative direct positive and negative regulators of upd3 induction , as well as key transcriptional regulators of gut homeostasis . Among our positive hit TFs that are strongly required for upd3-lacZ induction , we took note of Sna , as well as the homeodomain TFs , Rx and Ubx , and the epigenetic regulator , Trl . Despite the fact that Trl was the only one with a yeast one-hybrid predicted TFBS , knockdown of any of these TFs blocked infection-induced upd3-lacZ activity by 40% or more ( Fig 3C and Fig 4A ) . RT-qPCR measurements of upd3 mRNA levels upon Ecc15 infection further confirmed the requirement of these TFs for proper upd3 transcriptional upregulation ( Fig 4B ) . Sna classically acts as a repressor of transcription [32 , 33] , suggesting that its positive effect on upd3 expression is indirect . We further confirmed that EC-specific RNAi against CtBP or Ebi , the co-repressors recruited by Sna to mediate transcriptional repression [34 , 35] , also suppressed upd3-lacZ activity during Ecc15 infection ( Fig 4A ) . It is notable that these phenotypes were found in ECs , despite the fact that Sna has been described as a marker and regulator of progenitors in the Drosophila midgut [28] . Surprisingly , we found that Sna itself is transcriptionally upregulated in response to both Ecc15 ( Fig 4C ) and Pe ( Fig 4D ) infections . In addition , most of its upregulation occurs in ECs ( Fig 4D ) . Altogether , our results suggest that , in response to infection , Sna is upregulated in ECs , and in turn promotes upd3 upregulation through an indirect mechanism . The Hippo pathway consists of a kinase cascade resulting in the phosphorylation of Wts , which in turn phosphorylates and inhibits the transcription factor Yki [36] . When released from phosphorylation-induced restraint , Yki is transported to the nucleus , where it dimerizes with other TFs to promote transcription of target genes [37] . Hippo regulation plays an important role in tissue regeneration and growth . In addition , Yki has been shown to control epithelium turnover , acting cell-autonomously in ISCs via a Hpo/Wts/Yki pathway and non-cell-autonomously in EBs via the Msn/Wts/Yki pathway [38] . As previously mentioned , Yki and its partner Sd were found in our TF RNAi screen to be required in ECs for upd3 transcription in both basal and Ecc15-infected conditions ( Figs 3C , 5A and 5B ) . In addition , Sd was found to interact with the upd3 promoter by yeast one-hybrid , suggesting that the Hippo pathway may be directly involved in basal and infection-induced upd3 expression . We also noted that Trr , a major constituent of the TRR histone H3 lysine 4 ( H3K4 ) methyltransferase complex , and Trl , which are both required for full Yki-Sd mediated transcription [39 , 40] , are also required during infection for upd3-lacZ induction ( Fig 5A ) . Conversely , overexpressing Yki , or knockdown of either wts or its activator , msn , in ECs was enough to induce the transcription of upd3-lacZ ( Fig 5C ) . However , RNAi mediated depletion of hpo , which encodes another Wts phosphorylating kinase , had no significant effect on upd3-lacZ ( S6 Fig ) . Finally , overexpressing msn in ECs inhibited usual upd3-lacZ activity in Ecc15 infected and unchallenged midguts ( Fig 5A and 5B ) . We confirmed the requirement of the Hippo pathway TF , Sd , for upd3 transcription in ECs during enteric infection by RT-qPCR ( Fig 5D ) . Our results suggest that the Hippo pathway , which has been shown to be important for upd3 regulation under basal conditions and in response to abiotic stress [41 , 42] , is additionally required in ECs for upd3 expression in response to oral infection by Ecc15 . The TGF-β/Dpp pathway has emerged as a major regulator of intestinal homeostasis in Drosophila , as it has been found to be involved in diverse processes including ISC proliferation , ISC quiescence , EC differentiation and EC protection [43–48] . Mad , a TF downstream of the Dpp pathway was found in our screen to be necessary for wild-type upd3-lacZ levels upon ingestion of Ecc15 as well as in basal conditions ( Fig 5A and 5B ) . Thus , we explored whether ECs require a fully functional Dpp pathway to regulate the transcription of upd3 . EC-specific RNAi against the Dpp type-1 receptors , thickveins ( tkv ) and saxophone ( sax ) , or the type-2 co-receptor punt ( put ) , all decreased infection-responsive upd3-lacZ activity ( Fig 5A ) . Furthermore , overexpression of Dpp triggered aberrant induction of upd3-lacZ ( Fig 5C ) . We additionally tested the Dpp pathway via manipulation of the glycogen-synthase-3-kinase Shaggy ( Sgg ) , which has been shown to negatively regulate Mad through phosphorylation of linker serines [49] . Overexpression of sgg in ECs blocked upd3-lacZ induction , while sgg knockdown increased upd3-lacZ basal activity ( Fig 5A and 5C ) . A role for the Dpp pathway in regulating upd3 was further supported by RT-qPCR of upd3 in flies expressing EC-specific RNAi against tkv or Medea ( Med ) , a TF that acts together with Mad [50] , as both led to decreased induction of upd3 upon Ecc15 infection ( Fig 5D ) . Altogether , our data demonstrate that the Dpp pathway is required for proper upd3 transcription in response to infection . D-Fos and D-Jun were among the TFs in our screen that most strongly impacted upd3-lacZ activity upon infection . When activated by upstream kinases these two TFs act together as the AP-1 transcription factor complex [51] . D-Fos also interacts ex vivo ( in our Y1H screen ) with the upd3 promoter , suggesting that AP-1 acts as a direct regulator of upd3 transcription . Accordingly , RNAi against D-Fos or D-Jun , or the expression of a dominant negative D-Jun ( UAS-JraDN ) significantly decreased upd3-lacZ activity ( Fig 6A and 6B ) . As an additional confirmation of these results , we found that RNAi mediated knockdown of D-Fos in ECs prevented infection-responsive upd3 expression as measured by RT-qPCR ( Fig 6C ) . We next aimed to identify the upstream pathway ( s ) that regulate ( s ) D-Fos and D-Jun in response to Ecc15 infection . Phosphorylation and subsequent activation of the AP-1 complex is carried out by both Stress Activated Protein Kinases ( SAPKs ) and Mitogen Activated Protein Kinases ( MAPKs ) [52] . SAPKs and MAPKs act in phosphorylation cascades that result in the activation of terminal kinases such as JNK , Basket ( Bsk ) , p38 and ERK ( S8F Fig ) . It has been previously shown that artificial activation of the Drosophila SAPK , Bsk , by overexpression of Hemipterous ( Hep ) induces upd3 transcription in the gut , possibly through the activation of apoptosis or by directly regulating AP-1 [17 , 19] . We first evaluated whether apoptosis is required for upd3 expression in response to microbes . To this end , we manipulated the expression of caspase and autophagy genes in ECs and measured the resulting upd3-lacZ activity . Our results confirmed that promotion of autophagy or apoptosis , by overexpression of Autophagy-related 1 ( Atg1 ) or Death regulator Nedd2-like caspase ( Dronc ) , respectively , induced upd3 ( S7A Fig ) . However , inhibiting either pathway by RNAi against Dronc , Death-associated APAF1-related killer ( Dark ) , Atg1 , Atg7 or Atg18 , or by overexpression of the caspase inhibitor P35 ( UAS-P35 ) , had no significant negative effect on upd3-lacZ levels during infection ( S7B Fig ) . Furthermore , detection of caspase activity in ECs by the UAS-Apoliner system ( S7C Fig ) [53] , in conjunction with immunostaining for upd3-lacZ-derived β-galactosidase , revealed that cytokine production during enteric infection is not restricted to ECs with increased caspase activity ( S7D Fig ) . Altogether these data suggest that apoptosis and autophagy are not the key inducers of upd3 expression upon infection . We next sought to evaluate the contribution of JNK to upd3 induction upon infection with Ecc15 . We first verified whether Ecc15 infection triggers JNK activation in ECs , via co-immunostaining of the phosphorylated form of JNK and an EC marker ( Myo-Gal4TS>UAS-GFP ) ( S8A Fig ) . In agreement with previous publications , ectopic activation of the JNK pathway in ECs , by overexpressing Bsk or a constitutively active form of Hep , strongly promoted upd3-lacZ transcription ( Fig 6D ) . However , EC-specific expression of a dominant negative form of Bsk ( UAS-BskDN ) , or knockdown of bsk expression , decreased upd3-lacZ activity following oral infection by only 20% ( S8C Fig ) . Additionally , RNAi knockdown of hep did not decrease upd3-lacZ induction significantly ( S8C Fig ) . This suggests that JNK only plays a minor role in upd3 regulation , and thus additional stress pathways may be responsible for stimulating AP-1 in response to oral bacterial infection . Another possible candidate for AP-1 regulation is the p38 family of stress responsive MAPKs . The p38 kinases can regulate the AP-1 complex ( S8F Fig ) , and have been shown to be involved in the response to oral infection in Drosophila [54] . Immunostaining for phosphorylated p38 kinases revealed a substantial increase in p38 phosphorylation in ECs upon infection ( S8B Fig ) . To investigate the role of the p38 pathway further , we knocked down the three p38 kinases of Drosophila ( p38a , p38b and p38c ) , independently . Only knockdown of p38b gave a mild , but significant ( p<0 . 05 ) decrease in upd3-lacZ induction upon infection ( Fig 6A ) . We similarly tested the involvement of the upstream p38 MAPKK , Licorne ( Lic ) , and found that knockdown of lic in ECs also blocks increased upd3-lacZ transcription in response to oral infection . These experiments suggest that the stress in ECs caused by enteric infection triggers activation of a Lic/p38b pathway that mediates part of the induction of upd3-lacZ . In addition to JNK and p38 kinases , the D-ERK kinase is also able to activate the AP-1 complex ( S8F Fig ) [51] . Thus , we decided to investigate whether the MAPK/D-ERK pathway could also act upstream of AP-1 to regulate upd3 upon infection . Immunostaining for the phosphorylated form of Rolled ( Rl ) , the Drosophila homologue of ERK , revealed that infection with Ecc15 triggers D-ERK activation in ECs within two hours ( Fig 6E ) . Furthermore , RNAi knockdown of rl in ECs resulted in a strong decrease in upd3-lacZ activity upon infection ( Fig 6A ) , suggesting that the MAPK/ERK pathway is necessary for infection-regulated upd3 induction . MAPKs are activated in a phosphorylation cascade downstream of MAPKKs and MAPKKKs ( S8F Fig ) . Two of the four Drosophila MAPKKs ( Lic and Hep ) were previously tested for a role in upd3 regulation , and thus we proceeded to test the remaining two: Downstream of raf1 ( Dsor1 ) and MAP kinase kinase 4 ( Mkk4 ) . As for ERK , Dsor1 was critical for full induction of upd3-lacZ upon infection ( Fig 6A ) . Accordingly , expressing a dominant negative form of the upstream MAPKKK , Raf , in ECs also decreased upd3-lacZ regulation by infection , while blocking other MAPKKKs , TGF-β activated kinase 1 ( TAK1 ) , Apoptotic signal-regulating kinase 1 ( ASK1 ) and MEKK1 , did not ( Fig 6A and S8D Fig ) . Furthermore , constitutively active Raf expression is sufficient to induce upd3-lacZ activity ( Fig 6D ) . These data together suggest the possibility of a Raf/Dsor1/ERK pathway that regulates upd3 expression via AP-1 in response to midgut infection or damage . Activation of Raf by phosphorylation is typically accomplished via Ras , downstream of growth factor receptors ( S8F Fig ) . However , although overexpression of constitutively active Ras is sufficient to induce upd3 ( Fig 6D ) , blocking Ras itself ( S8C Fig ) or signaling through the key Receptor Tyrosine Kinases ( RTKs ) EGFR and PDGF- and VEGF-receptor related ( Pvr ) ( UAS-RasDN , UAS-EGFRDN , UAS-Pvr-DN ) did not impair upd3-lacZ activity ( S8E Fig ) . Likewise , RNAi knockdown of the Pvr ligand , PDGF- and VEGF-related factor 2 ( Pvf2 ) , had no effect on upd3-lacZ regulation . Raf signaling can occur downstream of additional tyrosine kinases , including the Src family kinases ( SFKs , S8F Fig ) [55 , 56] . Immunostaining for the phosphorylated form of Src kinases revealed that infection with Ecc15 triggers Src activation in ECs ( Fig 6F ) . To determine if the Src complex is also required for upd3 regulation , we knocked down Src42A and Src64B by RNAi in ECs ( Fig 6A ) . Depletion of either Src42A or Src64B decreased upd3-lacZ induction upon infection . Conversely , the expression of a constitutively active form of Src42A in ECs triggered upd3-lacZ induction in absence of infection , suggesting that a Src/Raf/Dsor1/MAPK pathway is sufficient to activate upd3 transcription . We further confirmed our results by RT-qPCR of upd3 in response to infection while blocking expression of Dsor1 , p38b and Src42A in ECs by RNAi , as well as by activating the pathway by expression of a constitutively active form of Src42A ( Fig 6C ) . In summary , our results demonstrate that multiple kinase cascades ( Licorne-p38b and Src/Raf/Dsor1/ERK ) are activated in ECs following oral Ecc15 infection and converge on the regulation of upd3 . We next aimed to evaluate the physiological consequences of modulating in ECs the pathways that control upd3 transcription . The number of mitotically active ISCs ( phospho-Histone H3 positive cells ) following Ecc15 ingestion was significantly reduced by knockdown of AP-1 and Sd , as well as the MAPK , Rl , the Dpp receptor , Tkv , and the epigenetic regulator , Trl , using the temperature sensitive , EC specific driver line ( MyoTS ) ( Fig 7A ) . This suggested that pathways required for EC-derived Upd3 production are required for proper ISC activity upon infection . We therefore monitored the survival of flies expressing EC-specific RNAi against pathway components of the Hippo , Dpp and SFK/MAPK/AP-1 pathways as well as putatively indirect regulators ( Sna , Trl , Rx , Ubx ) of upd3 upon infection . These flies had significantly shorter lifespans following Ecc15 infection compared to wild-type controls , and LT50 values lower than controls by at least two days ( Fig 7B–7D , S4 Table ) . In addition , we also found that , under UC conditions , these knockdown flies have significantly shorter lives than wild-type ones , and correspondingly lower LT50 values ( S9A–S9C Fig , S4 Table ) , implying that the knockdown of these genes , or the subsequent reduction in basal Upd3 levels compromises midgut epithelial homeostasis . We further confirmed our results by altering the expression of our candidate genes in ECs using multiple independent transgenic UAS-RNAi lines for each gene and monitoring their survival in both infected and unchallenged conditions ( S4 Table ) . Altogether , our experiments demonstrated that the Hippo , Dpp and SFK/MAPK/AP-1 pathways are required in ECs for survival to oral infection and for normal aging .
We found that the upd3 gene is regulated by three classes of enhancers: region-specific , cell-specific and stress/microbe-responsive . This complexity likely reflects the multiple roles of the JAK-STAT pathway in the Drosophila midgut , where it acts to stimulate ISC proliferation , promote differentiation and serves as a regional determinant of cell identity , notably in the middle midgut [8 , 12 , 17 , 57] . We propose that the different functions of upd3 are therefore regulated independently by the diverse enhancer regions we identified . We further identified microbial responsive enhancers that are active either in ECs ( B-C and I ) or in progenitor cells ( enhancer R ) , supporting a distinct regulation of upd3 in different cell types . Interestingly , the progenitor-specific enhancer R is the only one to be induced by DSS feeding ( and only to a low degree ) , while the EC specific enhancers B-C and I do not promote transcription in these conditions . It has been speculated that DSS elicits stem cell proliferation through alteration of the basal lamina rather than by direct damage to ECs [25] , such as that caused by Ecc15 infection or by bleo treatment . This suggests that different cell-type specific enhancers allow for induction of upd3 expression in response to a broad variety of stresses . The regulation of host gene expression by bacteria in Drosophila relies mostly on dedicated pathways , Toll and Imd , that trigger effector induction in response to the detection of microbial patterns ( MAMPs ) , such as bacterial derived peptidoglycan [58] . The microbe-responsive enhancers of upd3 are activated by both pathogenic and benign microbes , such as Ecc15 and the gut microbiota , but are also stimulated by toxic chemicals such as bleo or DSS . This result suggests that cytokine production in the gut is primarily triggered in response to damage associated molecular patterns ( DAMPs ) rather than the detection of microbes alone . Considering that dietary microbes and the microbiota are constantly associated with the gut tissue , triggering perpetual , low-level Imd activation , responding to DAMPs could be a strategy to couple immune activation and tissue repair to the presence of pathogens rather than beneficial or commensal microbes . Accordingly , we found that upd3 activation is less pronounced by the microbiota than by pathogens . These pathways have been shown to be activated by various stresses and are central to upd3 regulation in ECs . A major source of stress in response to microbes , is the production of ROS , partly induced by NADPH oxidases Nox and Duox of the host immune response [59 , 60] . Notably , SAPKs and Src kinases are both sensitive to ROS and their activity is modulated by oxidative stress , indicating that a NADPH oxidase , ROS , Src , SAPK/MAPK axis could be involved in upd3 regulation . Future work should determine the link between infection-induced ROS , Src/SAPK activation and the control of gut homeostasis . We further focused on identifying the key TFs that regulate upd3 in the midgut . We found that altering the expression of 138 over the 708 Drosophila TFs significantly altered upd3 expression in the midgut . This number is surprisingly high , as it implies that a quarter of Drosophila TFs directly or indirectly regulate upd3 transcription . We interpret this high number as an indication that upd3 acts as a stress marker , and that any physiological alteration in the gut will result in a rupture of gut homeostasis and consequently in the induction of upd3 [8] . We therefore propose that upd3 acts as a global sensor of gut stress and in turn initiates a stereotypical immune and homeostatic program . This poses the question of how multiple stresses can converge on the activation of upd3 transcription . Our results suggest that in ECs , stresses are mostly integrated by one upd3 enhancer ( B-C ) that responds to both chemical and biotic stresses . Integration could occur either because all stresses result in one simple damage signal , for instance cell loss in the epithelium , or as a consequence of multiple types of gut damage . Interestingly , the TFs altering upd3 expression in basal and infected conditions are not the same , indicating that different cascades regulate upd3 expression under different conditions . Upon infection , our data show that the Dpp , Hippo , SAPK and MAPK pathways are all involved in the regulation of upd3 . We therefore propose a model in which the diverse transcriptional regulation of upd3 is required for its multiple roles in homeostatic regulation . The different transduction pathways we identified all respond to different cues . We find that the Dpp pathway is likely involved in the activation of enhancer B-C in the Drosophila midgut . The Dpp pathway is furthermore essential for EC differentiation , growth , survival to infection , and injury-induced Dpp negatively controls midgut homeostasis [43 , 45] . Upon enteric infection , the Dpp pathway displays complex behavior . In an early response , Dpp released from hemocytes has been shown to stimulate ISC proliferation , but in a second phase , the Dpp pathway promotes the reestablishment of a quiescent state in these same cells [61] . Our results suggest that upon infection with Ecc15 , the Dpp pathway also plays a role in ECs by promoting upd3 transcription , which could synergize with the early proliferative role of this pathway in ISCs . It remains unclear whether Dpp acts directly or indirectly on the upd3 promoter . We identified Mad and Med as required for upd3 expression , and TFBS for Mad are found in the promoter region of upd3; however , our yeast one-hybrid screen did not detect a direct interaction between these two components . We did find evidence of direct regulation of the upd3 gene by transcription factors downstream of the Hippo pathway and SAPK/SFK/MAPK cascades . The Hippo pathway regulates ISC proliferation in the midgut both cell-autonomously and non-cell-autonomously [42 , 62 , 63] . The upstream regulators of Hippo signaling remain uncharacterized in the midgut , but the MAPKKKK Msn has been shown to control Wts in progenitor cells [64] . Our data suggest that the Yki/Sd complex directly regulates upd3 in ECs upon infection , and that Msn , but not Hpo , is involved in that process . We furthermore identified D-Fos and D-Jun ( AP-1 complex ) as direct regulators of upd3 transcription , acting downstream of Src-Raf-Dsor1-ERK and Licorne-p38b kinase cascades . Stress responsive kinases , as well as SFKs , are key regulators of AP-1 [55] . It remains unclear whether the upstream stimuli inducing SAPK/SFK/MAPKs to regulate upd3 upon infection include oxidative stress , cytoskeletal modification or a combination of both , but all these stimuli occur upon infection and are possible candidates . We propose that the role of SFKs , MAPKs and SAPKs in the regulation of cytokine expression and cell proliferation is conserved across organisms . Indeed , AP-1 and these conserved pathways have been demonstrated to have an important role in the regulation of cytokine secretion and tumorigenesis [65 , 66] . Src kinases have been previously shown to be important for wound healing in multiple models , potentially downstream of ROS production [67 , 68] , suggesting that conserved pathways are used in both tissue repair and gut regeneration . Interestingly , the pathways we identified in our study are known to work cooperatively in other systems . For example , mammalian JNK kinases are capable of phosphorylating YAP ( Yki homologue ) , and can inhibit multiple constituents of the Hippo pathway during tumorigenesis [69] . In addition , mammalian Src has been shown to regulate YAP during inflammation [70] . Finally , it was recently found that binding sites for Yap/Taz/Tead ( Yki/Sd in Drosophila ) and AP-1 are associated genome-wide with enhancers of genes involved in oncogenic growth . Altogether , these results and our own suggest that the SAPK/SFK/MAPK pathways in coordination with Hippo and TGF-β pathways work together in a conserved regulatory network that controls tissue growth and repair . The maintenance of gut tissue homeostasis relies on the induction of ISC proliferation to compensate for the loss of cells in the epithelium in a homeostatic feedback loop . A simple model of homeostasis would hold that cell death directly triggers upd3 expression and subsequent ISC proliferation in a coupled manner . In agreement with this model , induction of apoptosis in ECs is sufficient to induce upd3 expression and trigger ISC proliferation [17] . However , a recent study using oral infection with a low dose of pathogenic bacteria in Drosophila demonstrated that cytokine-induced ISC proliferation can be elicited even by infections that do not induce epithelial cell death [71] . This indicates that the coupling of ISC proliferation with cell loss is not complete . In agreement with these results , we found that neither apoptosis nor autophagy alone appear to be necessary for Ecc15-induced upd3 expression . Rather , the results of Loudhaief et al . ( 2017 ) and our study suggest that cytokine signaling results from stress detection rather than cell death , and that regenerative processes can occur independently of apoptosis [71] . This is also in agreement with the fact that the gut microbiota , which induces basal levels of epithelial stress but does not induce massive cell death in the gut , also stimulates basal cytokine production [3 , 13 , 16 , 26] . Pathways such as Hippo regulate both cell death and apoptosis , as well as cytokine production in the gut . We therefore propose that coupling between cell death and cell renewal is a consequence of cross-talk between regulatory pathways , rather than renewal as a direct consequence of cell death . Another hypothesis is that cell loss without death is coupled to tissue repair . Accordingly , infection induces the loss of ECs from the epithelium prior to anoikis [19] . It is therefore possible that EC delamination , rather than death , is a key signal for regeneration as evidenced by the observation that loss of EC contact with the basal lamina of the midgut epithelium can trigger Upd3 production [72] . EMT is a process of tissue morphogenesis reminiscent of cell delamination , in which epithelial cells detach and are extruded from the epithelial sheet whereupon they migrate as loosely associated mesenchymal tissue . Curiously , our study shows that the transcription factor Sna , a main regulator of Drosophila EMT and a marker of progenitor cells in the midgut , is both transcriptionally induced in ECs upon infection and required for upd3 transcription [8 , 73] . Sna’s role as a negative regulator of transcription implies that this phenotype is likely a secondary effect . We thus propose that upd3 expression may be downstream of Sna-dependent , EMT-like shedding of ECs in response to enteric stresses . In such a scenario , cell loss would require an EMT like regulation in ECs and indirectly trigger upd3 transcription . Epithelial structure and tension modulated by delamination could also result in Src and Hippo pathway activation , and ultimately in upd3 induction . Future work will determine how ECs are extruded from the epithelial sheet and how cell loss modulates Upd3 production . The regulatory pathways that we find upstream of upd3 transcription in ECs appear to be the same pathways required in ISCs to control their proliferation . For instance , inactivation of the Hippo pathway or induction of the Dpp pathway in ISCs is sufficient to stimulate stem cell proliferation in the Drosophila midgut [42] [61] . Similarly , the MAPK pathway has been demonstrated to be critical in ISCs for division and differentiation downstream of EGFR [19] . However , the regulation of these pathways is not always identical between cell types: while the SAPK kinase cascade is strongly required cell-autonomously for ISC activity [13] , its effect on upd3 induction in ECs is only marginal . Along these lines , MAPKs act downstream of growth factor receptors in ISCs , while we found that Src kinases trigger their activation in ECs . We thus propose that a single regulatory network controls ISC proliferation both cell-autonomously and cell non-autonomously and that the two processes are linked by the secretion of cytokines and growth factors . Altogether , the results of our study illustrate key aspects of the regulation of cytokine expression by intestinal cells in the gut . We identify microbe-responsive enhancers in the promoter of upd3 that act as stress sensors , thanks to the cooperative regulation by multiple pathways . Dpp , Hippo , Src , SAPK and MAPK pathways all converge on the transcriptional regulation of upd3 , thus acting together as a genetic network dedicated to damage detection and response . Strikingly , this genetic network controls both proliferation in stem cells , as well as the expression of cytokines in ECs to subsequently induce ISC proliferation . This genetic regulatory network therefore links stem cell proliferation and cytokine production in one common molecular framework , and paves the way for future studies to decrypt the link between inflammation and cancer in the gut .
Drosophila stocks were maintained at room temperature ( ~23°C ) on standard fly medium ( sucrose , cornmeal , yeast , and agar ) . Control lines: as controls for Gal4 driver experiments , we used the F1 progeny of the driver line crossed to wild-type stocks such as Canton-S ( Cs ) ( BDSC: 64349 ) , and background matched stocks such as attp2 ( BDSC: 36303 ) and attp40 ( BDSC: 36304 ) . Gal4 Drivers: Myo1A-Gal4 , UAS-GFP , tub-Gal80TS; upd3-lacZ ( MyoTS , EC-specific ) , Su ( H ) GBE-Gal4;UAS-GFP , tub-Gal80TS ( Su ( H ) TS , EB-specific ) [17] . Conditional Gal4TS flies were obtained by crossing virgin females of the driver strain with males of the UAS-transgene line . For RNAi and overexpression experiments , F1 progenies ( driver > UAS-transgene ) were raised at 18°C until 3 days after emergence , to allow for full gut development . Flies were then switched to 29°C for a week to allow for maximum transgene expression and RNAi-mediated gene knockdown . UAS-transgene stocks: RNAi transgenic fly lines were obtained from Bloomington ( TRiP lines ) , VDRC ( Vienna ) or NIG ( Japan ) , as specified in S2 Table . UAS-Atf3 3xHA was obtained from FlyORF . UAS-Src42A , UAS-Src42AYF , UAS-Src64B , UAS-Src64BYF were generously provided by professor Tian Xu [74] . UAS-bskDN; IF/CyO ( BDSC: 6409 ) , UAS-Src42A-IR ( NIG-FLY: 7873R-2 ) , UAS-Src42A-IR ( NIG-FLY: 7873R-3 ) , UAS-Src64B-IR ( VDRC: 35252 ) , UAS-Src64B-IR ( NIG-FLY: 7524R-1 ) /CyO; MKRS/TM6B , UAS-Src42AYF5382B , sb/TM6B , UAS-Src64BYF161; sb/TM6B , yw;;Src64BYU1332 ( BDSC: 7342 ) , w-; IF/CyO; UAS-csk/TM6B , w-; UAS-cpb7/Cyo; MKRS/TM6B , w-; IF/CyO; UAS-cpa attB/TM6B [75] , w-; UAS-cpa-IRC10 [75] , w-; IF/CyO; UAS-cpb-IR/TM6B ( VDRC: 46668 ) , w-;;; zyxD41 [76] , were generously provided by Florence Janody . UAS-Apoliner and Tub-Apoliner were both generously provided by Jean-Paul Vincent [53] . Reporter lines: upd3 . 1-lacZ , esg-lacZ , Myo-lacZ , [17] . A complete list of the TRiP UAS-RNAi lines used in the TF screen can be found in S2 Table . A list of the additional transgenic lines used in this report can be found in S5 Table . Overlapping fragments of ~1 . 5Kb were cloned in front of GFP , starting from 4 . 2Kb upstream of the upd3 start site and ending 7 . 3Kb downstream of the gene . These sequence fragments were designated putative upd3 enhancer regions A-R and cloned into T vector followed by pH-stinger [77] to create 21 enhancer trap GFP vectors . Each vector was used to generate at least two enhancer trap GFP fly lines ( to account for insertion position effects ) , which were then screened for capacity to drive GFP expression in the adult midgut under both basal conditions as well as Ecc15 infection . In addition , two reporter transgenes expressing NLS-GFP , fused to the Upd3 protein and driven under the control of the full upd3 locus and endogenous promoter ( from 4 . 2Kb upstream of the upd3 start site , up to 7 . 3Kb downstream of the gene ) , as well as the same reporter with enhancer B and C sequence regions deleted , were created and inserted at the attP2 insertion site . Erwinia carotovora ssp . carotovora 15 ( Ecc15 ) and Pseudomonas entomophila ( Pe ) are two Gram-negative bacteria , pathogenic to the Drosophila midgut when ingested [1] . Bacteria were maintained on standard LB agar plates and Pe was plated from glycerol stocks for each experiment . Bacteria were cultured in LB broth at 29°C for 16 hours . Oral infection was performed as previously described [12]: flies were starved in empty vials for 2 hours at 29°C , then moved to fly vials in which the standard food was completely covered by a filter paper disc containing 150μl of either 2 . 5% sucrose solution ( control ) , or 5% sucrose solutions mixed in equal volume with OD600 = 200 bacterial pellet , or a solution of 500μg/ml of bleomycin or 6% DSS . Orally treated flies were incubated at 29°C until dissection . 3 to 5 day old flies were transferred on fresh fruit juice agar plates . After 1 day of habituation , flies were allowed to lay eggs for 4–6 hours . Eggs were first suspended in 1X PBS , rinsed in 70% EtOH for 1 minute and dechorionated using 10% bleach for ~10min . Eggs were then transferred under a sterile flow hood and further rinsed 3 times with sterile ddH2O . The eggs were finally transferred into sterile fly vials with sterilized fly food . Flies were tested for presence of bacteria after each experiment , by plating homogenates on MRS agar plates . After dissection , Drosophila midguts were fixed in 4% paraformaldehyde in 1X PBS for 45 to 90 minutes and successively washed 3 times with 0 . 1% TritonX in PBS . Guts to be immunostained were then incubated for an hour in blocking solution ( 1% bovine serum albumin , 1% normal donkey serum , and 0 . 1% Triton X-100 in PBS ) . Overnight primary antibody staining was performed at RT . Guts were washed 3 times with 0 . 1% TritonX in PBS and ≥2 hour secondary antibody staining was performed in PBS . Primary antibodies used: rabbit anti-pH3 ( 1:000 , EMD Millipore ) , rabbit anti-β-Galactosidase ( 1:1000 , MP Biomedicals ) , and mouse anti-Prospero ( 1:100 , DSHB ) . Secondary antibodies used: donkey anti-rabbit-555 ( 1:2000 , Thermo Fisher ) , donkey anti-mouse-488 ( 1:2000 , Thermo Fisher ) , and donkey anti-mouse-647 ( 1:1000 , Thermo Fisher ) . DNA was stained in 1:50 , 000 DAPI ( Sigma-Aldrich ) in PBS and 0 . 1% TritonX for 30min , and samples received a final three washes in PBS before mounting in antifade medium ( Citifluor AF1 ) . Imaging was performed on a Zeiss LSM 700 fluorescent/confocal inverted microscope . Myo-Gal4TS; Upd3-lacZ driver/reporter flies were crossed to RNAi or overexpression lines and their adult progeny were induced at 29°C for seven days , then treated with either sucrose ( control ) or Ecc15 for 16 hours . Five midguts were dissected for each sample and homogenized in 100μl Z-buffer ( 60mM Na2HPO4 , 60mM NaH2PO4 , 10mM KCl , 1mM MgSO4 , 50mM β-mercaptoethanol , adjusted pH to 8 with NaOH ) . Homogenates were then centrifuged and 40μl of supernatant was mixed with 250μl of 0 . 35mg/ml ONPG ( o-nitrophenol-β-D-galactoside ) in Z-buffer solution in the wells of a 96-well plate . Absorbance was then measured at 420nm in a plate-reader ( spectra max plus , Molecular Devices ) every minute for one hour at 37°C . Because the amount of ONPG added to the reaction is sufficient to saturate the β-Gal in the samples , the reaction rate ( absorbance vs time ) is proportional to the quantity of β-Gal in each sample , and thus the maximum reaction rate ( Vmax ) was used as a measure of the relative β-Gal quantity in each sample . For each experiment , the average of three controls was used as a reference and relative upd3-lacZ activity was calculated ( S2 Table ) . The three controls used were: progeny of Myo-Gal4TS; upd3-lacZ virgins crossed to either the wild type strain , Canton-S ( Cs ) , or the controls “attP2” and “attP40” . The attP2 and attP40 lines are background controls for the TRiP UAS-RNAi stocks , while Cs is a standard , laboratory wild-type fly line . We used the variation in upd3-lacZ activity between the three controls ( S5A and S5B Fig ) to determine a confidence interval and select positive hits in the screen results ( lower than 0 . 6 and higher than 1 . 4 upon infection , lower than 0 . 5 and higher than 1 . 6 in UC conditions ) . We further confirmed the significance of these results by the calculation of z-scores for each RNAi knockdown tested ( S2 Table ) . Myo-Gal4TS; upd3-lacZ driver/reporter flies were crossed to the UAS-RNAi lines and their progeny were raised at 18°C . At 3-days post eclosure , 20 adult females were shifted to 29°C , the temperature at which all survival experiments were done to allow constant expression of the RNAi constructs . Day seven post-induction was considered day 0 of the survival studies . The controls used were the F1 progeny of crosses between our driver and the wild-type stock Cs , as well as the background-matched lines “attP2” and “attP40” . To evaluate possible background or off-target effects , multiple RNAi lines were used for each gene and the survival of all parental lines alone was also monitored . Survival was recorded in unchallenged ( UC ) conditions , in which flies were kept on standard cornmeal medium , and upon constant exposure to Ecc15 ( flies were transferred to new tubes with fresh Ecc15 every 3 days ) . Deaths were monitored daily and plotted using the GraphPad Prism 7 . 0c software . Results of survival experiments are aggregates of 3 to 9 biological replicates and error bars represent standard errors . LT50s were determined using PROBIT analysis in R . Total RNA was extracted from 15 to 20 female fly midguts following standard protocol with Trizol ( Invitrogen ) . Reverse transcription ( RT ) was performed using the qScript cDNA synthesis kit ( Quanta ) and quantitative PCR with SsoAdvanced Universal SYBR Green Supermix ( Bio-Rad ) and a CFX96 TouchTM Real-Time PCR Detection System ( Bio-Rad ) . Measured mRNA quantities were normalized to control Rp49 ( RpL32 ) mRNA values . The upd3-lacZ sequence was cloned into 4 fragments fused to the HIS3 reporter to generate baits further tested in yeast one-hybrid . HIS3 encodes an imidazoleglycerol-phosphate dehydratase , that catalyzes histidine synthesis , and the inhibitor 3-amino-1 , 2 , 4-triazole ( 3AT ) competitively inhibits this activity . The higher the level of 3AT in the medium is , the higher HIS3 expression needs to be to insure yeast growth , thus testing the strength of transactivation of the bait-HIS3 in response to multiple TFs . Prior to the TF/bait interaction test , a self-activation test was performed to assess whether natural S . cerevisiae TFs are sufficient to induce basal HIS3 expression . This test was performed by measuring the growth of eight independently transformed yeasts for each bait on SC-His plates with varying concentrations of 3AT ( 0 , 10 , 20 , 40 , 60 , 80mM ) . For each bait , a transformant yeast that can grow on SC-His medium , but is unable to grow on medium supplemented with 3AT was selected . The yeast one-hybrid assay was performed as previously described [27 , 78 , 79] . Briefly , upd3-HIS3 baits were integrated in the genome of Saccharomyces cerevisiae and transformed with a collection of 670 plasmids containing Drosophila TF open reading frames fused to the Gal4 activation domain . Each colony was plated on synthetic complete medium lacking Histidine ( to select for the upd3-His construct ) and Tryptophan ( to select for the presence of the TF vector ) . Plates were incubated at 30°C for 3 , 7 , and 10d and imaged using a Bio-Rad gel doc system . Yeasts not transformed with any TF prey and yeasts transformed with the Gal4 activation domain alone served as negative control . Plate images were analyzed using the R package Gitter , that estimates colony surface and circularity . Sets of quadruplicate colonies that showed growth above background levels were deduced to have a direct interaction between the TF prey and the DNA bait , and the strength of the interaction was estimated and ranked ( from +/- to +++ ) by the ability of each yeast colony to grow on increasing concentrations of the HIS3 inhibitor 3AT as previously described [27 , 79] . | Tissue regeneration is a fundamental process that maintains the integrity of the intestinal epithelium when faced with chemical or microbial stresses . In both healthy and diseased conditions , pro-regenerative cytokines function as central coordinators of gut renewal , linking inflammation to stem cell activity . In Drosophila , the upstream events that stimulate the production of the primary cytokine Unpaired 3 ( Upd3 ) in response to indigenous or pathogenic microbes have yet to be elucidated . In this study , we demonstrate that upd3 expression is driven in different cell types by separate microbe-responsive enhancers . In enterocytes ( ECs ) , cytokine induction relies on the Yki/Sd , Mad/Med , and AP-1 transcription factors ( TFs ) . These TF complexes are activated downstream of the Hippo , TGF-β and Src-MAPK pathways , respectively . Inhibiting these pathways in ECs impairs upd3 transcription , which in turn blocks intestinal stem cell proliferation and reduces the survival rate of adult flies following enteric infections . Altogether , our study identifies the major microbe-responsive enhancers of the upd3 gene and sheds light on the complexity of the gene regulatory network required in ECs to regulate tissue homeostasis and stem cell activity in the digestive tract . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"invertebrates",
"cell",
"death",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"immune",
"physiology",
"cytokines",
"gene",
"regulation",
"immunology",
"cell",
"processes",
"animals",
"animal",
"models",
"developmental",
"biology",
"regulator",
"genes",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"molecular",
"development",
"gene",
"types",
"epigenetics",
"drosophila",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"transcriptional",
"control",
"genetic",
"interference",
"gene",
"expression",
"insects",
"immune",
"system",
"arthropoda",
"gastrointestinal",
"tract",
"biochemistry",
"rna",
"eukaryota",
"anatomy",
"nucleic",
"acids",
"cell",
"biology",
"physiology",
"apoptosis",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
]
| 2017 | Hippo, TGF-β, and Src-MAPK pathways regulate transcription of the upd3 cytokine in Drosophila enterocytes upon bacterial infection |
X chromosome inactivation in female mammals results in dosage compensation of X-linked gene products between the sexes . In humans there is evidence that a substantial proportion of genes escape from silencing . We have carried out a large-scale analysis of gene expression in lymphoblastoid cell lines from four human populations to determine the extent to which escape from X chromosome inactivation disrupts dosage compensation . We conclude that dosage compensation is virtually complete . Overall expression from the X chromosome is only slightly higher in females and can largely be accounted for by elevated female expression of approximately 5% of X-linked genes . We suggest that the potential contribution of escape from X chromosome inactivation to phenotypic differences between the sexes is more limited than previously believed .
Dosage compensation is a regulatory process that alters gene expression along entire X chromosomes resulting in equivalent levels of X-linked gene products in males and females . Dosage compensation has evolved independently several times and is achieved in various ways . In Drosophila melanogaster , for example , the male X chromosome is hypertranscribed , doubling the output of X-linked genes ( reviewed in [1] and [2] ) . This balances gene expression between the sexes and also satisfies the second requirement of a dosage compensation system , which is to balance X chromosome gene expression with that of the autosomes . The situation in mammals is more complex . Inactivation of one of the X chromosomes in female mammals [3] balances X-linked gene expression between males and females . However , if this were the only component of the mammalian system , the single active X chromosome of females and males would effectively make both sexes aneuploid with respect to autosome gene expression . Ohno hypothesized therefore that balance would be achieved by doubling the output from the male X ( and active female X ) [4] . The hypothesis was confirmed recently when 2-fold upregulation of the X chromosome was demonstrated for both human [5] and mouse [6] . The molecular mechanism for this X chromosome upregulation in mammals is unknown . By contrast X chromosome inactivation ( XCI ) has been known about for almost fifty years and is extensively characterized . XCI is a complex and tightly regulated process unique to mammals that results in heterochromatization and transcriptional silencing of one of the female X chromosomes ( for a review see reference [7] ) . In eutherian mammals , the maternal or paternal X chromosome is inactivated randomly early in embryogenesis , and once established the pattern is mitotically stable . XCI was first suggested in 1961 to explain mosaic phenotypes seen in female mice heterozygous for sex-linked mutations in coat colour genes [3] . The theory was supported by the observation that cloned fibroblasts from human females heterozygous for an electrophoretic protein variant from the X-linked gene G6PD expressed either the paternal or the maternal allele , but not both [8] . Similar observations were reported for HPRT [9] , PGK [10 , 11] and a number of other X-linked genes . This pattern of complete silencing of one allele in females is seen for the majority of X-linked genes tested . However , the finding that the steroid sulphatase gene ( STS ) was always expressed in female fibroblast clones with one STS-deficient allele , regardless of which X was inactivated , suggested that some genes are not subject to XCI [12] . This “escape” from XCI results in differential expression of STS loci on the active X ( Xa ) and inactive X ( Xi ) chromosomes: clones expressing STS from the Xi have approximately half the level of STS enzyme activity of clones expressing from the Xa [13] . The use of rodent-human hybrid cell lines retaining an inactive human X chromosome has contributed greatly to our knowledge of escape from X inactivation [14–16] . This approach has the advantage of being able to assay gene expression from the Xi without the interference of the active copy . The largest study of this type has estimated that minimally 16% of human genes escape from XCI [16] . In a complementary approach , the same authors compared expression from maternal and paternal alleles of 94 genes in a panel of cell lines with skewed XCI and found that 15% of these were consistently expressed from both X chromosomes [16] . These approaches can detect low levels of expression from Xi . However , measuring the effect on dosage compensation requires a method to compare gene expression between the sexes . Expression microarrays , designed using the annotated X chromosome sequence [17] , are suitable for such comparisons . Microarrays have previously been used to identify human genes escaping XCI by comparing gene expression in cell lines with supernumerary X chromosomes [18] , in male and female lymphocytes [19] , and in a range of male and female tissues [20] . A genome-wide survey of sex-differences in gene expression in lymphoblastoid cells also yielded several examples of X chromosome genes with elevated female expression levels [21] . In the light of this reported widespread escape from X inactivation , we sought to determine its effect on dosage compensation in the largest comparison to date of X chromosome gene expression in females and males . Here we report analysis of a microarray expression dataset obtained using lymphoblastoid cell lines from 210 individuals in four populations . We show that the proportion of X-linked genes with significantly higher expression in females is around 5% , and that dosage compensation in these cell lines is virtually complete .
Microarray data from the Gene Expression Variation project ( GENEVAR [22–24] ) were analysed for 210 unrelated individuals from the four HapMap populations [25] , designated CEU , CHB , YRI and JPT ( see Methods ) . First , we identified 371 genes on the X chromosome and 11 , 952 genes on autosomes that are expressed in the cell lines ( see Methods ) . We then compared gene expression from autosomes and the male X chromosome . The median expression value of the 11 , 952 autosomal genes was plotted against the median of the 371 X chromosome genes for 105 unrelated male individuals from four populations ( Figure 1A ) . There is a clear linear relationship between autosomal and X gene expression in all populations . The majority of data points fall along a diagonal close to that where the autosomal median is equal to the X median . We compared mean expression of the 371 X genes with randomly selected sets of 371 autosome genes ( n = 100 ) for 30 YRI males using Student's t-test and saw no significant difference ( data not shown ) . Average expression from the single X chromosome in males is , therefore , similar to expression from an autosome pair . Next , we compared the male X chromosome with each autosome pair separately . Results for the YRI population are shown in Figure 1B , and very similar results were obtained for the other populations ( data not shown ) . Median expression from the single X chromosome falls within the normal range seen for autosome pairs and is slightly above average . The latter accounts for the observation that most data points in Figure 1A lie above the diagonal . We conclude that expression from the single X in male cell lines is upregulated 2-fold relative to autosomes , thus achieving dosage parity between the X and autosomes . Upregulation occurs precisely and consistently in 105 individual male samples . This supports and extends the findings of Nguyen and Disteche [5] . The two X chromosomes in females are not equivalent as the majority of genes on one are subject to X inactivation; however , it is well established that a number of genes escape the silencing process . Therefore , we compared expression of X chromosome genes in females and males , reasoning that escape from XCI should produce a substantially higher level of expression in females . First we compared expression of the 11 , 952 expressed autosome genes and 371 expressed X genes in 30 males and 30 females from the YRI population . Figure 2A shows median male expression plotted against female expression for each gene . The autosome genes lie on a diagonal with the vast majority showing very similar expression in males and females . Most X genes lie on the same diagonal , indicating that X chromosome gene expression too is similar in males and females and is not proportional to the number of X chromosomes . These data suggest that for most X chromosome genes , dosage compensation is achieved between males and females . We then normalised expression of the 371 X chromosome genes to the median of the 11 , 952 autosome genes for each individual , and calculated the mean of the normalised X chromosome genes for 105 males and 105 females . In each population mean expression of X chromosome genes was higher in females than males ( Figure 2B ) . However , these differences are small , representing increased expression in females of just 2 . 6% , 3 . 4% , 1 . 5% , and 2 . 2% for CEU , CHB , JPT and YRI , respectively . Genes escaping X inactivation do not , therefore , have a great effect on the overall level of X-linked gene expression in the female cell lines . This indicates that dosage compensation is occurring effectively . Although the difference is small , there is a measurable and consistent increase in X chromosome gene expression in females compared to males . In order to identify the genes that contribute to this difference , we looked for X chromosome genes with significantly higher expression in females . We used a Student's t-test to assess differences in expression levels between females and males within each population separately for the 371 X chromosome genes and 11 , 952 autosome genes ( Table S1 ) . Figure 3 compares the proportion of X chromosome and autosome genes expressed more highly in females or males at three different levels of significance: p < 0 . 05 , p < 0 . 01 and p < 0 . 001 . The proportion of X chromosome genes expressed more highly in females is greater than that from autosomes in each population . The difference between X and autosomes is consistent across the three levels of significance and at p < 0 . 001 ranges from 3 . 5% to 4 . 9% of genes in different populations . In contrast , the proportion of X chromosome genes with higher expression in males remains similar to that of autosome genes at all levels of significance ( Figure 3 ) . The most likely explanation for the observations above is that a proportion of X chromosome genes escape from X inactivation to a measurable level in these female cell lines . We supposed that X-linked genes able to escape the silencing process would be expressed more highly in females in all populations . We also reasoned that other X-linked or autosomal genes reaching a significance threshold may not do so in all populations . We therefore assessed the population commonality of autosomal and X chromosome genes with higher female expression at different levels of significance ( Table 1 ) . At p < 0 . 05 and p < 0 . 01 , the proportion of genes with significantly higher female expression is almost identical for autosomes and the X chromosome . However , the distribution of genes among populations is strikingly different , with a far greater percentage of X chromosome genes achieving significance in all four populations . When the significance threshold is raised to p < 0 . 001 , the proportion of genes retained is now lower for autosomes than for the X chromosome , and no autosome gene is common to three or four populations ( Table 1 ) . By contrast , approximately 3% of X chromosome genes are significantly elevated in the females of all populations at p < 0 . 001 ( Table 1 ) . A single gene ( CD99 ) is expressed more highly in the males of all four populations ( p < 0 . 001 ) . This is the only notable difference between X and autosomes in respect of higher male expression ( Table 1 ) . Using the combination of significance values and population commonality as a filter ( Table 1 ) , we identified a group of 20 X chromosome genes that are remarkable in the consistency of their elevation across females ( Table 2 ) : ALG13 , CA5B , DDX3X , EIFIAX , EIF2S3 , FUNDC1 , HDHD1A , JARID1C , MSL3L1 , PCTK1 , PNPLA4 , PRKX , RPS4X , SMC1L1 , STS , UBE1 , USP9X , UTX , ZFX and ZRSR2 . Eleven of these were expressed more highly in females at p < 0 . 001 in all four populations ( Table 2 ) . This situation was not observed for any of 11 , 952 autosomal genes tested . Figure 4A illustrates the female to male ratio of expression for each of the genes in the four populations . Taking the mean of the four populations , there is a subset of six genes ( JARID1C , UTX , HDHD1A , PNPLA4 , DDX3X and EIF1AX ) for which expression in females is around 1 . 5-fold greater than in males . Most genes have a much smaller difference: EIF2S3 , USP9X , CA5B and PCTK1 , ZFX and SMC1L1 all have less than 1 . 2-fold higher expression in females compared to males . We interpret these ratios as expression from the active X that is equivalent to expression from the single X in males , combined with a lower level of expression from the inactive X that is more variable between genes . Some genes ( e . g . , DDX3X , STS ) have a much higher ratio in some populations than others , which may represent a biological difference in the extent of escape from X inactivation in different human populations . We observe higher female expression for 5 . 4% of the X chromosome genes expressed in the cell lines ( 20/371 ) . The possibility remains that other X-linked genes may escape from XCI in lymphoblastoid cells . However , we have determined that these 20 genes account for almost all of the difference in gene expression between males and females seen in Figure 2B ( data not shown ) . Therefore , any expression of additional genes from the inactive X chromosome must be very low and/or must occur in only a small fraction of female cell lines . In either case , the impact on dosage compensation at the population level would be minimal . Therefore , we conclude that 94 . 6% of X-linked genes are effectively dosage compensated in human lymphoblastoid cell lines . We hypothesized that escape from XCI may be partly stochastic , and that this might lead to greater variation in expression levels among females than males for some genes . Figure 4B shows box and whisker plots for four of the genes that can be considered to escape from XCI on the basis of their higher female expression . The other 16 genes follow a similar pattern ( data not shown ) . The size of the box ( interquartile range ) is a good indicator of the similarity of distribution between females and males . In the majority of cases there is little difference in the distribution or range of values between the sexes , although the female values have a higher median and therefore the entire plot is shifted upwards . There is no correlation between the size of the interquartile range and median expression level . We also calculated the variance for each gene and found no significant difference between males and females within each population ( unpublished data ) . We conclude that genes escaping XCI in these cell lines are not expressed more variably in females than males , which suggests that escape may be a tightly regulated rather than a stochastic event . While the medians and interquartile ranges are clearly different between males and females , there is considerable overlap between the distributions of the expression levels of the two datasets . Individual data points from males and females of the YRI population are shown for four genes escaping XCI as a scatter graph ( Figure 4C ) . JARID1C is unique in that all data points for females are higher than all data points for males . For the other 19 genes , the female and male datasets overlap to varying extents ( contrast EIF2S3 with RPS4X in Figure 4C ) . This can also be seen as overlap of whiskers in Figure 4B . This observation illustrates the extent of inter-individual variability of gene expression and highlights the importance of comparing large samples of males and females to identify differences in gene expression that are a consequence of escape from XCI . The genes from the pseudoautosomal regions ( PARs ) are a special case as they lie within regions of XY recombination and are essentially equivalent on the X and Y chromosomes . For genes in PAR1 , escape from XCI is generally believed to be a prerequisite for dosage compensation between the two female X chromosomes and the male X and Y . We found that twelve PAR1 genes show no significant difference between females and males across populations and are therefore dosage compensated ( Table S1 ) . The only exception is CD99 , which is expressed significantly more highly in males in all four populations ( p < 0 . 001 ) . We identified single nucleotide polymorphisms ( SNPs ) in PAR1 genes SLC25A6 , CXYorf3 , ZBED1 and CD99 and used a quantitative assay to measure relative expression from the X and Y alleles in heterozygous males . As shown in Table 3 , the relative contribution from the X and Y chromosomes is very similar for each of the four genes . We conclude that the majority of PAR1 genes escape from XCI and are dosage compensated , consistent with the expectation above . The majority of genes that have a functional Y chromosome homologue were found to be expressed in hybrid cells containing the Xi [16] . Therefore , we decided to test the possibility that genes with higher female expression might , like the genes in PAR1 , be compensated by functionally equivalent Y-linked copies . Eight of the 20 genes with higher female expression have functional Y-linked gametologues . We excluded PRKY and EIF1AY from the analysis . The PRKY probe has 94% sequence identity to PRKX and gives a strong signal in females . Expression of EIF1AY is approximately 13-fold greater than EIF1AX in males , suggesting that EIF1AY is not involved in a compensation mechanism . Gene expression for the remaining six X-Y gene pairs is shown in Figure 5 . USP9X expression is significantly higher in females at p < 0 . 001 in all populations . Interestingly , the sum of USP9X and USP9Y expression in males is not significantly different from the level of USP9X in females ( p = 0 . 822 [CEU] , p = 0 . 024 [CHB] , p = 0 . 245 [JPT] and p = 0 . 610 [YRI] ) . USP9Y expression therefore completely restores the USP9X dosage imbalance between males and females . Figure 5B–5D shows similar results for gene pairs RPS4X/RPS4Y1 , UTX/UTY and DDX3X/DDX3Y . In each case dosage from the X and Y copies in males is similar to that of the X copies in females . Expression of the Y copy is always much lower than expression of the X copy and appears to reflect the expression from the Xi . For these four genes , dosage compensation appears to be achieved by expression of the Y copy . In contrast , expression levels of JARID1C ( X ) and JARID1D ( Y ) ( Figure 5E ) are approximately equal and their combined expression in males is significantly greater than JARID1C expression in females . A similar picture is obtained with ZFX and ZFY in males ( Figure 5F ) . We have identified an X-Y gene pair ( TMSB4X/TMSB4Y ) whose X copy is dosage compensated according to our data . TMSB4Y is expressed at less than 1% of the level of TMSB4X and therefore does not affect dosage compensation between males and females . Genes that escape from XCI in hybrid cell lines are non-randomly distributed on the X chromosome [16 , 17] . In light of identifying a smaller proportion of genes with higher female expression , we assessed the relationship between gene expression and chromosomal location . We observed that the distribution of genes with elevated female expression is also non-random , with most lying on the short arm ( Figure 6 ) . The chromosome can be divided into strata that ceased to recombine with the Y chromosome at different times in evolutionary history [17 , 26] . The most ancient parts of the chromosome ( strata S1 and S2 ) , covering the long arm and proximal short arm , contain 287 of the 371 expressed genes , but only six that have higher female expression . By contrast , a larger fraction of genes have elevated female expression levels in regions that stopped recombining with the Y chromosome more recently , either in early eutherian mammals ( S3 ) or in primates ( S4 , S5 ) . Ten out of 66 S3 genes fit this picture , while all three expressed genes in S4 , together with the single example in S5 , are more highly expressed in females . These findings support the model that X-linked genes are recruited into the XCI system following the Y chromosome degeneration that occurs when regions cease to recombine [27] .
On the basis of these data , we suggest that dosage compensation in human lymphoblastoid cells is virtually complete . Gene expression from the single X chromosome in males is upregulated 2-fold compared to the autosomes . Expression from the female X chromosome pair is almost the same as from the male X , suggesting that few genes escape the silencing process to any great extent . Twenty genes in this study ( 5 . 4% ) have significantly higher female expression , and four of these could have dosage balance maintained through expression of a Y-linked homologue . Ohno predicted 40 years ago that the evolution of a dosage compensation mechanism in mammals must have involved a doubling of expression of each X-linked gene as the Y chromosome degenerated [4] . Two-fold upregulation of the X chromosome has now been demonstrated for both humans and mice [5 , 6] and in a range of tissues [5] . In D . melanogaster , a slight but significant overexpression of the X chromosome compared to the autosomes in all XX;AA samples has been reported [6] which may be due to inherent hypertranscription of the X chromosome . We have determined that human X chromosome expression is not significantly elevated above the autosome average , suggesting that the X chromosome is not hypertranscribed in the cell lines over and above the 2-fold upregulation . Our data show that upregulation occurs precisely and consistently in 105 individual male samples , also leading us to conclude that in lymphoblastoid cell lines gene expression is appropriately regulated . Upregulation of the X is not seen in mouse germ cells suggesting that it takes place in the developing embryo [5] . Two-fold upregulation may be a general feature of X chromosomes , affecting genes on Xa and those on Xi that escape inactivation . Alternatively , upregulation and silencing could be mutually exclusive choices for X chromosomes in embryogenesis , simultaneously achieving correct X gene expression and dosage compensation . Previously , genes have been classified as partially escaping XCI if their female to male expression ratio is below two . However , under the second model described , genes fully expressed from both Xi and an upregulated Xa in females would have a theoretical maximum expression that is 1 . 5-fold greater in females than males . We favour this model as we have identified six genes expressed approximately 1 . 5-fold more highly in females and the greatest ratio we observe for any gene is 1 . 56 , averaged across four populations . Is 2-fold upregulation a chromosome-wide phenomenon ? The PAR1 region is the only surviving remnant of a large autosomal addition to both sex chromosomes that still undergoes recombination in male meiosis . The X and Y chromosomes are equivalent in PAR1 and genes here are predicted to escape XCI . Accordingly , all PAR1 genes tested were expressed from Xi in hybrid cell lines [16] . The majority of PAR1 genes included in our study showed no significant difference in expression between females and males . This gives rise to two models for PAR1 gene expression . In the first , PAR1 genes have equal expression from Xa , Xi and Y copies and so no dosage compensation is necessary . Under this model , PAR1 is excluded from both the upregulation and the silencing components of dosage compensation . In the second model , the PAR1 is 2-fold upregulated on the Xa only and dosage compensation is achieved through equal expression from the Xi and Y copies . A corollary of this model is unequal expression of PAR1 alleles within a cell , but equal expression between males and females . We have tested three dosage compensated PAR1 genes in males and find that expression levels are similar from the X and Y alleles . Therefore , we favour the first model and propose that PAR1 is protected from both upregulation and silencing . CD99 is the only PAR1 gene found to be more highly expressed in males , yet has equivalent expression from the X and Y copies . CD99 is the gene that lies closest to the boundary between the PAR1 and X-linked material , and we suggest that spreading of the XCI signal across the pseudoautosomal boundary results in partial silencing of CD99 . Consistent with this hypothesis , the protein product of the CD99 gene was found to be present at lower levels in hybrids containing Xi than those containing Xa [28] . Outside the pseudoautosomal regions , twenty genes are expressed at significantly higher levels in female cell lines . We hypothesized that these genes would escape from XCI . Formally , the alternative explanation for the elevated female expression could be that these genes are hypertranscribed from the female active X chromosome . However , since all of these genes are included in previous reports of escape from XCI [12 , 15 , 16 , 29–33] , we prefer the former explanation for higher female expression . An intriguing observation is that the dosage of some of these genes could be effectively compensated by expression of a Y copy . An underlying assumption of this analysis is that the X and Y gametologues are functionally equivalent , despite their evolutionary divergence . The DEAD box RNA helicase proteins DDX3X and DDX3Y appear to be interchangeable , as both rescue a temperature-sensitive mutant hamster cell line incapable of growth at a non-permissive temperature [34] . The ribosomal proteins RPS4Y1 and RPS4X show functional equivalence in a similar rescue assay and can function interchangeably in ribosomes [35] . RPS4Y1 and RPS4X are among a very small number of genes from the ancient sex chromosomes that have functional Y copies [17] . Despite their considerable divergence time and very high synonymous substitution rate , their protein products share 93% identity and are the same length , consistent with their being functionally equivalent . Previous microarray studies have assessed escape from XCI by looking for increased expression in cell lines with supernumerary X chromosomes [18] or by comparing female and male expression [19 , 20] . More recently , a larger study assessed genome-wide sex differences in gene expression using lymphoblastoid cell lines from monozygotic twin pairs [21] . Each study reports some genes that are considered to be well established as escaping from XCI , but the four vary considerably in the number and identity of genes documented . These differences might be explained in part by variation in escape from X inactivation in different tissues , for which there is evidence [20] . However , a further possibility is that inter-individual variability in expression that is unrelated to XCI could increase the risk of false positives or negatives where the sample size is small . The scale of our study , which measured gene expression in 210 individuals for 81% of protein coding genes on the X chromosome , means that we can confidently detect significantly higher expression in females at the population level in spite of this factor . Notably the study by McRae et al . [21] , which assayed 38 lymphoblastoid cell lines , agrees most closely with our study in the identity and proportion of genes with higher female expression . We find that 5 . 4% of X-linked genes have increased female expression in the cell lines . Analysis of somatic-cell hybrids that retain Xi and of fibroblast cells with non-random XCI has put the proportion of genes escaping XCI at 15%–25% [16] . This difference could be explained by a lower proportion of genes escaping XCI in lymphoblastoid compared with fibroblast ( or hybrid ) cells . However , perhaps more important are the differences of approach between the two studies . Our study used a population based analysis of dosage compensation , whereas Carrel and Willard [16] detected expression in hybrid cell lines , or compared expression levels from Xa and Xi alleles in female cell lines , using methods capable of detecting very low levels of expression from Xi . Some genes , therefore , could be expressed from Xi but at a level that is insufficient to cause a dosage imbalance . Other genes could escape to a larger extent in a small number of females , as suggested by Carrel and Willard [16] . Neither of these , though , is substantial enough to generate a significant sex-difference at the population level . We conclude that dosage compensation in human lymphoblastoid cell lines is highly effective and tightly controlled . It will be interesting to extend these studies to other tissues , but it seems unlikely that this level of regulation would be restricted to this single cell type . Therefore , we propose that the contribution of escape from XCI to male-female phenotypic differences may be small and furthermore , we suggest that the number of genes contributing to phenotype in X chromosome aneuploidies is lower than previously thought .
Gene expression was assayed in lymphoblastoid cell lines of all 210 unrelated HapMap individuals [25] from four populations ( CEU: 60 ( 30 Male/30 Female ) Utah residents with ancestry from northern and western Europe; CHB: 45 ( 22M/23F ) Han Chinese in Beijing; JPT: 45 ( 23M/22F ) Japanese in Tokyo and YRI: 60 ( 30M/30F ) Yoruba in Ibadan , Nigeria ) . RNA preparation , labeling , hybridization to Sentrix Human-6 BeadChip ( Illumina ) , gene expression quantification and normalization of raw data were described previously [23] . Briefly , each RNA sample was labelled in duplicate and each labelled sample was hybridised to two separate arrays . Data were subjected to quantile normalization then were median normalized across all individuals . Final data points for each gene are the mean of the four normalized hybridisation values . Log2 transformed mRNA expression values were used throughout except where otherwise stated . Data can be downloaded from http://www . sanger . ac . uk/humgen/genevar/ and the Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/ ) entry GSE6536 . We established an appropriate cut-off point for evaluating gene expression by examination of four non-human probes ( lysA , pheA , thrB and trpF ) in 210 individuals: 836/840 data points had log2 expression values <6 . 4 . We also evaluated signals from Y-linked genes in female samples: 593/620 data points were found to have values <6 . 4 , excluding two probes that apparently cross-hybridised with X-linked genes . We conservatively chose to analyse genes with log2 median expression >6 . 4 in all four populations . We excluded redundant probes ( n = 21 ) for X chromosome genes and any X probes that matched autosomal exons . Data for 11 , 952 expressed autosome sequences and 371 expressed X chromosome genes were used in all downstream analyses . The complete set of X chromosome and autosomal genes represented in this study and median expression values are shown in Table S2 . Data from the four populations were considered separately throughout . We tested male and female datasets separately for skewness and kurtosis for 371 X chromosome probes and found no evidence for them , except for a very small number of genes in some populations where log2 expression was close to the expression cutoff of 6 . 4 . We therefore concluded that the gene expression data generally follow a normal distribution . We used median values to illustrate chromosome or population averages except where we have shown the standard deviation ( Figure 2B ) . To make comparisons between autosome genes and X genes , or between groups of individuals , we normalized all gene expression values to the median value of 11 , 952 autosomal probes for each individual . We compared the variances of male and female samples for X chromosome genes by placing the larger variance over the smaller to form an F statistic . We found that variances for females and males are not significantly different . We tested significance by calculating p-values associated with a Student's two sample homoscedastic t-test with a two-tailed distribution . The complete list of p-values is shown in Table S1 . SNaPshot was carried out on cDNA and genomic DNA from heterozygous males using the SNaPshot Multiplex Kit ( Applied Biosystems ) according to the manufacturer's instructions with the following modifications . Initial template generation was carried out using Platinum Taq polymerase ( Invitrogen ) in a standard reaction using touchdown polymerase chain reaction ( PCR ) : denaturation: 94°C 15 min; 20 cycles: 94°C 30 sec , 70°C , 30 sec reducing by 1°C per cycle , 72°C 45 sec; then 15 cycles: 94°C 30 sec , 50°C 30 sec , 72°C 45 sec; final extension 72°C for 7 min . PCR products were treated with 2 units of shrimp alkaline phosphatase ( USB ) and 1 . 5 units of Exonuclease I ( USB ) for 1 hour at 37°C to remove primers and nucleotides , then at 80°C for 15 mins . Primer extension products were analysed on an ABI 3730 DNA Analyzer with a POP-7 Polymer and a 36cm capillary array with the ABI standard run module . SNP data were analysed using ABI PRISM GeneMapper Software Version 3 . 0 . Primers used to generate template DNA for analysis and SNaPshot extension primers are shown in Table S3 . Peak heights from cDNA were normalized to genomic DNA values and expressed as an allelic ratio . | The males and females of many species are distinguished by their inheritance of different sets of sex chromosomes . This creates a significant imbalance in gene number between the sexes . Dosage compensation is the correction for this imbalance and is achieved by regulating gene activity across entire sex chromosomes . For example , human females have two X chromosomes and males have only one . Dosage compensation in humans involves X chromosome inactivation , which is the silencing of one X chromosome in female cells . Some genes are known to escape the silencing process and so are expressed at higher levels in females than males . We have investigated the extent to which such genes disrupt dosage compensation by comparing the activity of X chromosome genes in a large number of human male and female cell lines . We have shown that gene expression from the X chromosome pair in female cell lines is only slightly higher than from the single X in males . The small difference can be accounted for by increased female expression of approximately 5% of X chromosome genes . We conclude therefore that dosage compensation in these human cell lines is virtually complete , and we suggest that differences in X chromosome gene expression between males and females may be less extensive than previously thought . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"homo",
"(human)",
"genetics",
"and",
"genomics"
]
| 2008 | Large-Scale Population Study of Human Cell Lines Indicates that Dosage Compensation Is Virtually Complete |
Visceral leishmaniasis ( VL ) is a neglected tropical disease and is fatal if untreated . There is no vaccine available against leishmaniasis . The majority of patients with cutaneous leishmaniasis ( CL ) or VL develop a long-term protective immunity after cure from infection , which indicates that development of an effective vaccine against leishmaniasis is possible . Such protection may also be achieved by immunization with live attenuated parasites that do not cause disease . We have previously reported a protective response in mice , hamsters and dogs with Leishmania donovani centrin gene knock-out parasites ( LdCen-/- ) , a live attenuated parasite with a cell division specific centrin1 gene deletion . In this study we have explored the effects of salivary protein LJM19 as an adjuvant and intradermal ( ID ) route of immunization on the efficacy of LdCen-/- parasites as a vaccine against virulent L . donovani . To explore the potential of a combination of LdCen-/- parasites and salivary protein LJM19 as vaccine antigens , LdCen-/- ID immunization followed by ID challenge with virulent L . donovani were performed in hamsters in a 9-month follow up study . We determined parasite burden ( serial dilution ) , antibody production ( ELISA ) and cytokine expression ( qPCR ) in these animals . Compared to controls , animals immunized with LdCen-/- + LJM19 induced a strong antibody response , a reduction in spleen and liver parasite burden and a higher expression of pro-inflammatory cytokines after immunization and one month post-challenge . Additionally , a low parasite load in lymph nodes , spleen and liver , and a non-inflamed spleen was observed in immunized animals 9 months after the challenge infection . Our results demonstrate that an ID vaccination using LdCen-/-parasites in combination with sand fly salivary protein LJM19 has the capability to confer long lasting protection against visceral leishmaniasis that is comparable to intravenous or intracardial immunization .
Leishmaniasis is a disease with a wide spectrum of clinical manifestations caused by different species of protozoa belonging to the Leishmania genus that are transmitted by sand fly vectors [1] . The disease causes high morbidity and significant mortality throughout the world , where 350 million people in 98 countries are at risk of contracting the infection . Moreover , approximately 1 . 0 to 1 . 5 million cases of cutaneous leishmaniasis ( CL ) , and 200 , 000 to 500 , 000 cases of visceral leishmaniasis ( VL ) , are registered annually [2] . VL is fatal if not treated [2] . The treatment of leishmaniasis is still based on the use of the parenteral administration of pentavalent antimonial compounds . However , side effects associated with the treatment and increased parasite resistance have made control and elimination of VL a serious challenge [3 , 4] . Therefore , the development of new strategies to prevent leishmaniasis has become a high priority [5] . The development of a vaccine for VL has been the focus of several research groups . Among the various types of vaccines , genetically modified live-attenuated vaccines provide the immunized host with diverse and complex antigens and induce a potent protective immunity in murine models [5 , 6] . Importantly live attenuated parasites cause no pathology in experimental infections [7–14] , while inducing protection reflected by a significant reduction of parasite burden in animals challenged with virulent wild type strains [10 , 12 , 14–18] . We have previously reported on the LdCen-/- parasites as a live attenuated candidate vaccine in several animal models [12 , 14 , 18] . Infection with LdCen-/- was non-pathogenic i . e . , safe and highly immunogenic in mice , hamsters and dogs [12 , 13 , 18] . In addition , immunization with LdCen-/- induced protection against homologous challenge with wild type L . donovani and conferred cross-protection against infection with a heterologous challenge with L . braziliensis , L . mexicana and L . infantum [14 , 18] . However , previous studies with LdCen-/- parasites as immunogens were performed without any adjuvants . Since the adjuvants can activate a range of innate immune pathways it is difficult to predict on an empirical basis which adjuvant will work most effectively with live attenuated parasites . Since the adaptive response is the primary determinant of protective immunity generated by vaccination , immunomodulatory reagents that could supplement LdCen-/- induced immunity without causing rapid elimination of the vaccine antigen due to innate immune reactions could make LdCen-/- more effective as an anti-Leishmania vaccine . Saliva from sand flies contains potent pharmacologic components that facilitate blood meal acquisition and modulates the host inflammatory and immune responses [19 , 20] . Arthropod vector saliva also plays an important role in pathogen transmission from the sand fly to the vertebrate host [21] . Recent reports have shown the importance of some salivary proteins from sand fly vectors such as LJM19 , LJM11 or LJM17 as potential targets for vaccine development against Leishmania infection [11 , 19 , 22–30] . A specific immune response against salivary proteins has been reported in various animal models . For example , hamsters immunized with plasmid DNA coding for LJM19 , a Lu . longipalpis salivary protein , protected them from disease after challenge with wild type Leishmania infantum chagasi parasites plus saliva through the induction of a LJM19-specific immune response [26] . By comparison , salivary protein LJM11 provided partial protection that was not long lasting against virulent challenge [26] . Importantly , immunization with LJM19 induced higher ratios of IFN-γ/IL-10 and IFN- γ/TGF-β in the spleen , conditions consistent with a Th1 polarization [26 , 28] . These results suggested that salivary gland proteins such as LJM19 could be a potent supplement to the protective immunity with live attenuated Leishmania parasites . In our previous studies , we have tested different routes of immunization , including intravenous ( tail vein ) , intracardial and subcutaneous [12 , 13] . However , intradermal immunization can offer improved protective immunity and simplify the logistics of delivery as was previously demonstrated [26 , 28 , 29 , 31 , 32] . Therefore , it is of value to evaluate live attenuated parasite vaccines for their efficacy following intradermal immunization . Since our previous studies have shown that exposure to live attenuated parasites injected by intravenous or intracardial routes and without an adjuvant induced a strong protective immunity , we asked whether an intradermal immunization with LdCen-/- parasites in combination with LJM19 could further enhance vaccine induced protection . In the present study , we report for the first time the immunogenicity and protection outcome in hamsters intradermally primed with salivary protein LJM19 and boosted with genetically modified live attenuated L . donovani parasites ( LdCen-/- ) in combination with recombinant LJM19 . Immunized hamsters demonstrated a strong immune response comparable to that of intracardial immunization with LdCen-/- and resulted in long term protection against infection with virulent L . donovani parasites .
Two-month-old female Syrian golden hamsters ( Mesocricetus auratus ) were obtained from the Harlan Laboratories and kept in the Food and Drug Administration ( FDA ) animal facility . The experimental procedures used in this study were reviewed and approved by the Animal Care and Use Committees of the FDA and the National Institute of Allergy and Infectious Diseases ( NIAID ) . The animal protocol for this study has been approved by the Institutional Animal Care and Use Committee at the Center for Biologics Evaluation and Research , US FDA ( ASP 1995#26 ) . Further , the animal protocol is in full accordance with ‘The guide for the care and use of animals’ as described in the US Public Health Service policy on Humane Care and Use of Laboratory Animals 2015 ( http://grants . nih . gov/grants/olaw/references/phspolicylabanimals . pdf ) . Lu . longipalpis sand flies , Jacobina strain were reared at the Laboratory of Malaria and Vector Research , NIAID . Salivary glands were dissected from 5- to 7-day-old females and stored in PBS at -70°C . Before use , salivary glands were sonicated and centrifuged at 12 , 000×g for 2 min . The supernatant was collected and used immediately . LdCen-/- and L . donovani ( Ld1S ) promastigotes were grown at 26°C in medium 199 supplemented with 20% FCS . Three- to 4-month-old hamsters were immunized intradermally in the ear at two-week intervals between immunizations , using a 29-gauge needle ( BD Ultra-Fine ) in a volume of 20μl , using the following protocols . Group 1: prime with 2μg of LJM19 protein and boost with 107 stationary phase LdCen-/-promastigotes plus 2μg of LJM19 . Group 2: 107 stationary phase LdCen-/- promastigotes via intracardial injection . Group 3: 2μg of LJM19 protein ( two times ) . Group 4: BSA ( control ) ( S1 Fig ) . Each experimental group consisted of 6 hamsters . Five weeks after the last immunization , the animals were challenged ID with 105 stationary phase Ld1S promastigotes in combination with 0 . 5 pairs SGH . At 5 weeks post immunization , and 1 and 9 months post challenge , the parasite load was measured in the ear , lymph node ( from challenged ear ) , spleen and liver by the limiting dilution assay as previously described [33] . Whole blood from immunized hamsters was collected at the indicated time points before sacrifice in 1 . 1mL microcentrifuge tubes with serum gel-clotting activator ( Sartesdt , GE ) . The serum was separated by centrifugation and used in IgG assays . Total IgG , IgG1 and IgG2 responses to L . donovani soluble antigens were measured by ELISA as described [34] . The following clones were used in the study . IgG cocktail ( Catalog# 554010 ) ; IgG1 ( clone-G94-56 ) ; IgG2 ( clone-G192-3 ) ( BD Biosciences ) . The cut-off value of reactivity for SLA antigens was calculated as the mean plus 2 SD of the OD values observed in naive controls . Sera from 9 naïve hamsters were used for determining cut-off values . Splenocytes were collected from hamsters , macerated , lysed in Trizol for RNA extraction . Total RNA was extracted from the ear , lymph node ( superficial parotid lymph node ) , spleen and liver of infected hamsters using TRIzol reagent ( Invitrogen ) . First-strand cDNA synthesis was performed with ≈1–2μg of RNA using a Transcriptor High Fidelity cDNA Synthesis Kit ( Roche ) . Amplification conditions consisted of an initial pre-incubation at 95°C for 10 min , followed by amplification of the target DNA for 40 cycles of 95°C for 15 s and 60°C for 1 min with the LightCycler 480 ( Bio-Rad ) . The efficiency of each reaction was determined . The expression levels of genes of interest were normalized to β-Actin levels . The results are expressed in fold change of 2-ΔCt over control . Oligonucleotide primers used for real time PCR were: β Actin , reverse , ACA GAG AGA AGA TGA CGC AGA TAA TG , forward , GCC TGA ATG GCC ACG TAC A; IFN-γ , reverse , TGT TGC TCT GCC TCA CTC AGG , forward , AAG ACG AGG TCC CCT CCA TTC; TNF-α , reverse , TGA GCC ATC GTG CCA ATG , forward AGC CCG TCT GCT GGT ATC AC; IL-10 , reverse , GGT TGC CAA ACC TTA TCA GAA ATG , forward , TTC ACC TGT TCC ACA GCC TTG; IL-4 , reverse , ACA GAA AAA GGG ACA CCA TGC A , forward , GAA GCC CTG CAG ATG AGG TCT; IL-12/IL-23p40 , reverse , AAT GCG AGG CAG CAA ATT ACT C forward , CTG CTC TTG ACG TTG AAC TTC AAG , iNOS , reverse , ACC ACA CAG CCT CCG AGT CC , forward , CTG CCA GAT GTG GGT CTT CC . The primers and probes were synthesized at the Center for Biologics and Evaluation and Research , FDA Core facility . Statistical analysis was performed using GraphPad Prism 5 . 0 software ( GraphPad Software Inc . , USA ) . Non-parametric Kruskal-Wallis test followed by Dunns test were used to compare data from four groups ( G1 , G2 , G3 and G4 ) . Differences were considered significant when a p value ≤ 0 . 05 was obtained .
Since the efficacy of genetically modified Leishmania parasites in general , and LdCen-/- parasites in particular have not been tested through an intradermal route , we sought to determine the persistence of LdCen-/- parasites at the site of injection and its dissemination to other organs where relevant immune reactions might occur . Thus , we measured the parasite load in the ear , lymph node , spleen and liver 5 weeks post immunization ( 5wpi ) in hamsters primed with LJM19 and boosted with LJM19 plus LdCen-/- ( G1 ) . We observed an average of 300 and 30 viable LdCen-/- parasites in the ear and draining lymph node , respectively , measured by limiting dilution ( Fig 1A and 1B ) . However , we could not recover any LdCen-/- parasite from the spleen and liver from either G1 or from animals immunized with LdCen-/- intracardially . The analysis of sera from hamsters primed with LJM19 and boosted with LdCen-/- plus LJM19 indicated the occurrence of strong Leishmania-specific antibody responses . The SLA-specific IgGTotal and IgG1 production was significantly higher for immunized animals in G1 when compared to animals immunized IC with LdCen-/- ( IC , G2 ) , at 5wpi ( p<0 . 05 ) ( Fig 2A and 2B ) . However , increased levels of IgG2 were detected in G1 and G2 groups compared to G3 and control group G4 ( p<0 . 01 and p<0 . 001 , respectively; Fig 2C ) that show base line reactivity . Importantly , the IgG2/IgG1 ratio was significantly higher in G1 and G2 groups compared to G4 , the control group ( p<0 . 05 ) and to G3 ( p<0 . 05 ) that received LJM19 alone ( Fig 2D ) . Of note , the IgG2/IgG1 ratio was higher in G2 compared to G1 ( p<0 . 05 ) ( Fig 2D ) . In order to evaluate the protective immunity induced after ID immunization , we analyzed the mRNA expression of both Th1 and Th2 cytokines ( IFN-γ , iNOS , IL-12/IL-23p40 , IL-4 and IL-10 ) in the ear , the site of injection at 5wpi . The IFN-γ , iNOS and IL-12 expression levels were significantly higher in immunized hamsters in G1 compared with G2 , G3 and G4 groups ( p<0 . 05; Fig 3A , 3B and 3C , respectively ) . Expression of IL-4 and IL-10 cytokines was also higher in G1 animals after immunization , when compared to animals in groups G2 , G3 and G4 ( p<0 . 05; Fig 3D and 3E , respectively ) . However , the level of IFN-γ was higher than either IL-10 or IL-4 in G1 animals . Additionally , G1 animals showed up to a ~16-fold , ~12-fold and ~13-fold increase in IFN-γ , iNOS and IL-12 respectively , compared to BSA-immunized control group G4 ( Fig 3F ) . Similarly , mRNA levels of IL-4 and IL-10 were up-regulated ~9- and 8-folds , respectively , in G1 compared to G4 animals . The number of viable L . donovani parasites was determined by the limiting dilution assay in the draining lymph node and spleen of immunized hamsters a month post challenge ( mpc ) ( Fig 4 ) . The number of live parasites in the lymph node ( Fig 4A ) and in the spleen ( Fig 4B ) were significantly lower ( p<0 . 05 ) in the G1 and G2 groups either immunized with LJM19 then boosted with LdCen-/- plus LJM19 or immunized with LdCen-/- alone respectively , when compared with the groups of hamsters that received LJM19 alone or BSA alone . We measured antibody levels in the sera of immunized animals one mpc . No difference was observed in the levels of IgGTotal and IgG1 between the groups ( Fig 5A and 5B ) . The level of IgG2 was elevated in G1 immunized animals when compared to G3 and G4 hamsters ( p<0 . 05 and p<0 . 01 , respectively; Fig 5C ) . In addition , G1 group presented a significantly higher IgG2/IgG1 ratio in comparison to G3 and G4 control groups ( p<0 . 05 and p<0 . 01 , respectively ) ( Fig 5D ) , indicative of a Th1-type immune response post challenge . The mRNA expression level of Th1 and Th2 cytokines was estimated by qRT-PCR one mpc . In the lymph node ( LN ) , G1 and G2 groups presented a high expression of IFN-γ , when compared to G3 and G4 groups ( p<0 . 01 ) ( Fig 6A ) . A moderate increase in the expression levels of iNOS mRNA transcripts in LN was observed only in G1 , and G3 immunized hamsters compared to G4 group ( p<0 . 05 ) ( Fig 6B ) . Interestingly , cytokine IL-12 mRNA levels in LN were significantly higher in G1 , G2 and G3 groups compared to the G4 control group ( p<0 . 001 ) ( Fig 6C ) . Concomitantly , the mRNA levels of the Th2 cytokines IL-4 and IL-10 , primarily regulatory cytokine , were higher in LN of G4 control animals in comparison to G1 , G2 and G3 immunized groups ( p<0 . 05 ) ( Fig 6D and 6E ) . Additionally , IFN-γ , iNOS and IL-12 from G1 were significantly up-regulated by ~8 folds , ~14 folds and ~5 folds , respectively , compared to G4 group ( Fig 6F ) . In the spleen , IFN-γ was up-regulated one mpc in groups G1 ( p<0 . 001 ) , G2 ( p<0 . 01 ) and G3 ( p<0 . 01 ) , compared to control group ( G4 ) ( Fig 6G ) In addition , G1 presented a higher expression of iNOS when compared to G2 , G3 and G4 after challenge ( p<0 . 001 , p<0 . 005 and p<0 . 01 , respectively ) ( Fig 6H ) . Interestingly , IL-12 mRNA levels were not significantly different in spleens from animals of the 4 groups though G1 animals exhibited a trend for increased IL-12 expression ( Fig 6I ) . The IL-4 expression was decreased in G1 and G2 groups , but not significantly ( Fig 6J ) compared to G3 and G4 . However , there was a significant reduction in IL-10 expression ( p<0 . 05 ) in G1 when compared to G2 , G3 and G4 groups ( Fig 6K ) . The IFN-γ , iNOS and IL-12 mRNAs from G1 were significantly up-regulated by ~19 folds , ~32 folds and ~10 folds , respectively , compared to G4 group ( Fig 6L ) . In the liver , IFN-γ expression was upregulated one mpc in G1 and G2 groups in comparison with G3 and G4 animals ( p<0 . 01 ) ( Fig 6M ) . However , iNOS expression was significantly higher in G2 compared to G1 , G3 and G4 ( Fig 6N ) . Liver cells from group G1 immunized hamsters induced a significantly high expression of IL-12 ( Fig 6O ) . IL-4 expression was significantly higher in G4 animals compared to G1 , G2 and G3 hamsters ( p<0 . 01 ) , and lower in G2 group compared to G1 group , but was not significantly different compared to G3 group ( Fig 6P ) . On the other hand , hamsters from control group G4 showed a significant up-regulation ( p<0 . 001 ) in IL-10 expression compared to G1 , G2 and G3 ( Fig 6Q ) . The IFN-γ , iNOS and IL-12 mRNAs from G1 were moderately up-regulated by ~1 . 3 folds , ~0 . 39 folds and ~1 . 2 folds , respectively , compared to control group G4 ( Fig 6R ) . The animals in G1 and G2 groups showed robust protection 9 mpc as evident by a significant decrease in the lymph node parasite load ( Fig 7A ) , spleen ( Fig 7B ) and liver ( Fig 7C ) in comparison to G4 animals ( p<0 . 0001 ) . Immunization with salivary gland protein LJM19 alone ( G3 ) also provided significant protection ( p<0 . 01 ) , albeit weaker than that observed in animals in groups G1 and G2 , compared to G4 BSA immunized animals . Further , we wanted to test whether LdCen-/- immunization causes re-establishment of homeostatic conditions by comparing the spleen sizes . Both G1 and G2 presented a non-inflamed spleen ( median 3 . 4 and 3 . 5cm , respectively ) , as compared to a highly inflamed spleen in the control group G4 ( median 7 . 9cm ) ( Fig 7D ) . G3 animals showed an intermediate spleen size ( median 5 . 2 cm ) .
Previous work from our laboratories has shown that the LdCen-/- live attenuated vaccine is immunogenic in mice , hamsters and dogs [12 , 13 , 18] . Similarly , LJM19 protein from saliva of the vector Lu . longipalpis protected hamsters against challenge with L . infantum and L . braziliensis [26 , 28] . In the present work , we examined the value of combining the two immunization strategies for their potential to elicit protective immune responses in a hamster model of Leishmania donovani infection . ID needle inoculation of the ear has been extensively employed as the route of infection that most closely replicates the physiological ID and intra-epidermal deposition of parasites by the bite of an infected sand fly [33 , 35–37] . Additionally , the ID route presents the most practical route for vaccine delivery [26 , 28 , 29 , 31 , 32] . We hypothesized that a prime/boost strategy with LJM19 followed by LdCen-/- parasites plus LJM19 , all delivered intradermally , would induce long-lasting protective immunity against L . donovani particularly since LdCen-/- parasites can undergo limited replication in the immunized host and provide an array of antigens very similar to those produced by a virulent parasite . As such , in this prime/boost protocol , priming with LJM19 would generate a specific adaptive immune response to the sand fly salivary gland protein as was observed in previous studies [26 , 28] that could result in a potent supplement to the specific adaptive immune response to antigens of the LdCen-/- parasites . We had previously observed LdCen-/- parasites in the spleen up to 5 wpi after intracardial ( IC ) injection [12] . In the current study , we observed parasites in the immunized ear and the draining lymph node but not in the spleen , at 5wpi after ID injection , suggesting that either the parasites take a longer time to disseminate to the viscera and reach the spleen or alternatively they do not visceralize . Of interest , recent studies with dermotropic parasite strains ( L . donovani isolated from a cutaneous lesion and L . major ) that fail to persistently visceralize nevertheless produced protective immunity in a low-dose infection followed by challenge with L . donovani and L . infantum in mouse models [38 , 39] . This suggests that visceralization may not be a necessary pre-condition for protective immunity against VL to develop in the immunized mice . Consistent with this hypothesis , our results indicate that since the draining lymph nodes represent the immunological niche where relevant reactions between APCs that acquired the antigens and naïve T cells could occur , recovery of attenuated parasites from the lymph nodes 5 weeks post immunization suggested that parasite persistence , i . e . antigen availability , was adequate for protective immunity to be established . Immunization of hamsters with attenuated parasites associated with LJM19 protein elicited a biased Th1-type immune response at 5wpi at the site of injection . As expected , immunization by LJM19 alone provided protection against L . donovani parasites , however , it was considerably weaker compared to the one observed following a prime/boost ID immunization with LJM19 followed by LJM19 and LdCen-/- or an IC immunization with LdCen-/- . Of note , boosting with LdCen-/- along with LJM19 protein through the ID route resulted in a higher pro-inflammatory response compared to immunization with LdCen-/-alone through the IC route . Indeed , immunization with LdCen-/- through the intravenous ( IV ) route in mice and IC route in hamsters [12] and subcutaneous route in dogs [13] has been shown to promote a pro-inflammatory response , with the presence of IL-12p40 , IFN-γ , iNOS and TNF-α . Additionally , our finding that increased levels of IL-4 , IFN-α , iNOS and IL-12/IL-23p40 in immunized hamsters would suggest a mixed immune response ( Th1 biased ) triggered by LdCen-/- vaccination , as we observed in dogs immunized previously [18 , 38 , 40] . Our results suggest that the ID mode of immunization , at least in combination with LJM19 , is equally efficacious compared to the IC mode . Additionally , the high ratio of IgG2/IgG1 observed in groups G1 and G2 is considered an additional immune biomarker of protection [5 , 41–43] . During natural transmission , an infected sand fly deposits saliva and parasites into the skin of the host while feeding . To mimic the natural mode of infection with Leishmania , we injected L . donovani wild type parasites into the ear of hamsters along with sand fly salivary gland extract . After one month of infection , hamsters immunized with LdCen-/- either alone ( IC ) or with LJM19 protein ( ID ) demonstrated a reduced parasite burden in the lymph node and spleen . In our study , similar to the response observed post-immunization , challenged hamsters presented a significant increase of IgG2 production , and a high ratio of IgG2/IgG1 , as well as an enhanced production of IFN-γ and iNOS . Higher levels of IgG2 might also contribute to pathogen clearance in vaccinated animals [44 , 45] . Previously , it was reported that iNOS and concomitant high levels of NO were produced by macrophages in protected mice vaccinated with attenuated parasites after challenge with L . donovani [12 , 14] . In the present study hamsters immunized with either LdCen-/- alone or in association with recombinant LJM19 displayed increased IFN-γ and iNOS expression in lymph nodes and spleen one month after challenge with wild type parasites . Of importance , five weeks post-immunization we observed a higher production of IL-12 only in animals immunized with LdCen-/- in association with recombinant LJM19 . It can be speculated that LJM19 might be pre-conditioning the innate immune arm and thus allowing the antigen presenting cells such as DCs to produce IL-12 that is necessary for initiating a strong adaptive Th1 cell immunity . Certainly , the ability of LJM19 to produce a Th1 response in hamsters has been previously demonstrated [26] . As such , it may be argued that LJM19 might be enhancing the immunogenicity of LdCen-/- as a vaccine . An increased IFN-γ/IL-10 ratio has been observed when DNA vectors expressing KMP11 along with LJM19 were used as immunogens compared to either KMP11 or LJM19 alone at 5 months post challenge [46] . This increased IFN-γ/IL-10 ratio did not result in reduced splenic parasite burden between KMP11+LJM19 and either antigen or LJM19 alone groups at 5 months post challenge . Further in their study the authors also observed increased IFN-γ/IL-10 ratio after 5 months of infection in non-immunized animals which does not explain the role of increased IFN-γ levels in protection . The observed differences between da Silva [46] and our study in splenic parasite burden could be due to the ability of the live attenuated parasites to induce sustained immunological reactions ( Fig 6G and 6K ) because of a longer availability of a multitude of antigens compared to recombinant antigens that tend to have limited availability and diversity that is reflected in parasite control up to 9 months post challenge . Further , in the LdCen-/- immunized animals ( G1 and G2 ) down-regulation of IL-10 and a concomitant increase in IL-12 in lymph node , spleen and liver may explain the greater parasite killing observed at the challenge site . In murine and human VL , production of Th1 cytokines is desirable for resolution of infection [47–50] . In addition , IL-12 results in the generation of Th1 cells that produce both IFN-γ and IL-12 , thus favoring the development of a protective cellular immune response against Leishmania [51–53] . An important consideration is that the sustained Th1-type immune response and long-term protection generated here against L . donovani after ID injection of LdCen-/- parasites in combination with LJM19 is comparable to that observed following IC or intravenous immunization with LdCen-/- parasites alone [12 , 14] . A similar induction of Th1 immunity was also observed in dogs immunized subcutaneously with LdCen-/- parasites and challenged with L . infantum [18] . At 9 months post challenge , the parasite load in the lymph nodes and in the spleen was significantly reduced in all the immunized groups compared to the control group . The control of parasitemia in the spleen translated into lack of splenomegaly in immunized and challenged animals compared to control challenged animals . Importantly , the significant reduction of parasitemia after immunization with LJM19 alone in our current study argues that LJM19 contributes to the observed protection in G1 hamsters . This is corroborated by Gomes at al . [26] who observed a decrease in parasite load in the spleen and liver in hamsters immunized with LJM19 after 2 and 5 months post I . D . inoculation of L . infantum chagasi with sand fly salivary gland homogenate . Additionally , in an independent study , Tavares et . al [28] showed that hamsters immunized with LJM19 induced protection against infection with L . braziliensis . Taken together our data indicate the induction of a long-lasting protective immune response in the spleen , liver and lymph nodes in hamsters immunized intradermally with LJM19 and LdCen-/- after challenge with virulent parasites and reveal that a stronger immune response is elicited when Leishmania donovani live attenuated parasites are combined with a salivary gland protein . In summary , we have demonstrated the capability of a combined vaccine composed of live attenuated LdCen-/- parasite and a defined salivary gland protein from Lu . Longipalpis ( LJM19 ) delivered intradermally to confer strong long-lasting protection against L . donovani infection in a hamster model . | Leishmaniasis is a disease with a wide spectrum of clinical manifestations caused by different species of protozoa belonging to the Leishmania genus that are transmitted by sand fly vectors . Visceral infections of Leishmania cause significant mortality and morbidity and development of a vaccine to prevent leishmaniasis has become a high priority . We have previously reported that intravenous immunization with a live attenuated parasite vaccine comprised of Leishmania donovani parasites lacking the centrin gene conferred protection in mice , hamsters and dogs . In the current report , we describe the immunological response and associated protection to the ID immunization with attenuated parasites in combination with a sand fly salivary protein ( LJM19 ) . We observe that protection against experimental ID challenge with L . donovani resulting from ID immunization with live attenuated parasites in combination with LJM19 is comparable to intracardial immunization and offers improved protective immunity compared to immunization with salivary protein alone and non-immunized hamsters . This study supports the potential use of the genetically attenuated vaccine and a recombinant sand fly salivary protein for control of visceral leishmaniasis . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| []
| 2016 | Intradermal Immunization of Leishmania donovani Centrin Knock-Out Parasites in Combination with Salivary Protein LJM19 from Sand Fly Vector Induces a Durable Protective Immune Response in Hamsters |
Intervertebral disc metabolic transport is essential to the functional spine and provides the cells with the nutrients necessary to tissue maintenance . Disc degenerative changes alter the tissue mechanics , but interactions between mechanical loading and disc transport are still an open issue . A poromechanical finite element model of the human disc was coupled with oxygen and lactate transport models . Deformations and fluid flow were linked to transport predictions by including strain-dependent diffusion and advection . The two solute transport models were also coupled to account for cell metabolism . With this approach , the relevance of metabolic and mechano-transport couplings were assessed in the healthy disc under loading-recovery daily compression . Disc height , cell density and material degenerative changes were parametrically simulated to study their influence on the calculated solute concentrations . The effects of load frequency and amplitude were also studied in the healthy disc by considering short periods of cyclic compression . Results indicate that external loads influence the oxygen and lactate regional distributions within the disc when large volume changes modify diffusion distances and diffusivities , especially when healthy disc properties are simulated . Advection was negligible under both sustained and cyclic compression . Simulating degeneration , mechanical changes inhibited the mechanical effect on transport while disc height , fluid content , nucleus pressure and overall cell density reductions affected significantly transport predictions . For the healthy disc , nutrient concentration patterns depended mostly on the time of sustained compression and recovery . The relevant effect of cell density on the metabolic transport indicates the disturbance of cell number as a possible onset for disc degeneration via alteration of the metabolic balance . Results also suggest that healthy disc properties have a positive effect of loading on metabolic transport . Such relation , relevant to the maintenance of the tissue functional composition , would therefore link disc function with disc nutrition .
Degenerative changes of the intervertebral discs ( IVDs ) occur either in a pathological manner or as a consequence of aging , and seriously compromise the tissue capability to sustain the stressful loads transmitted throughout the spine [1] , [2] . The IVD is the largest avascular tissue in our body and is maintained by a relatively small number of cells , which further decreases with age [3] . Therefore , disc degenerative changes ( DDCs ) , are strongly suspected to be linked with a failure of nutrient transport from the peripheral blood vessels to the IVD cells [4] . Nutrient transport within the disc depends on the tissue composition and morphology , and is also coupled with the response to mechanical loads . While it is reported [5]–[7] that sustained mechanical stresses affect transport of solutes and that the nutrient pathway disturbance acts concomitantly to degenerative phenomena [4] , [8] , [9] , it has not been clearly investigated whether and how DDCs could affect solute transport . Anaerobic glycolysis has been recognized as the main source of energy for disc cells . Experimental work led to empirical equations modeling the interactions between the main two nutrients , i . e . oxygen and glucose and the relevant metabolic waste product , lactate [10] . For small neutral solutes such as oxygen , lactate and glucose , diffusion has been underlined as the main transport mechanism [11] , [12] . The effect of static compression on diffusion of Gadoteridol ( a small solute ) in the IVD has also been recently studied in vivo , suggesting a load-dependence process for the diffusion of the solute [5] . Due to the difficulty of measuring in vivo the solute distributions within the IVD [13] , finite element ( FE ) modeling of transport processes is often used to complement experiments and bring further insights in disc nutrition and degeneration issues . Such computational studies have shown the importance of biochemical couplings in the disc glycolytic metabolism [14] , and that of endplate obstruction to nutrients and waste products due to calcification and sclerosis [15] . Fluid velocity ( i . e . advective transport ) was suggested to have a negligible role in enhancing small solute transport [16] . Nevertheless , by coupling metabolic reactions together with a multiphasic mechanical model , different effects on solute concentrations ( oxygen and lactate ) were found induced by static and dynamic compressions [17] , suggesting a potential role of fluid advection enhancement . Using a similar theoretical framework , Magnier and coworkers [18] studied the disc solute transport and the effect of mechanical coupling under free swelling . The simulated transport process depended on porosity , cell density and endplate diffusion area , but was independent of disc stiffness . Nevertheless , under sustained loads and significant tissue deformation , solute transport is expected to depend on disc stiffness . Most of the abovementioned numerical studies did not consider any local strain-dependent diffusivity [16] , [19] , [20] and advective effect was not clearly studied [15] , [19] , [21] . Moreover , the few models that included both load-dependent advection and diffusion were coupled to bi-dimensional simplified geometries [17] , [18] , though considering 3D geometries may substantially change the transport predictions [19] . Since none of the abovementioned models fully coupled the diffusive , convective , and metabolic 3D transport equations with large mechanical strains , the metabolic solute transport interactions with multiple DDCs , i . e . loss of fluid proteglycan content , solid matrix stiffening , cell death , geometrical changes , etc . could not be properly explored . As such , a model including biphasic mechanics , advection-diffusion-reaction ( ADR ) , and 3D geometry is needed to better understand IVD nutrient transport in healthy and pathological situations . We aimed at contributing to the intricate mechanics-transport connections involved in the pathophysiology of the human IVD by using a coupled mechanics-ADR approach in a 3D FE model . First , we hypothesized that mechano-transport couplings are essential to predict the solute distributions in the healthy and degenerated discs . Second , we parametrically assessed which DDC would mostly affect the oxygen and lactate transport , and in which manner .
Poromechanical and transport properties corresponding to a “healthy” disc were used ( Table 1 ) . The compressive creep behavior of the healthy IVD model in terms of vertical height change and nucleus pulposus ( NP ) pore pressure were validated in a previous study [22] against published experimental results [23] . To explore the mechano-transport responses of the models , a simple diurnal cycle consisting in 16 hours of creep under compression at 0 . 5 MPa and 8 hours of rest at 0 . 1 MPa of compressive stress was applied and repeated during two days ( Fig . 1b ) . To establish the role of mechanics in transport predictions during a diurnal cycle , the healthy disc was tested with and without the abovementioned loading . When mechanical deformation was not considered , diffusivities were constant during the simulation and calculated from the initial porosities reported in Table 1 ( Eq . ( 8 ) in Methods ) . With mechanical loading , the roles of strain-dependent diffusivity , changes in diffusion distances and advective transport were also computed . The effect of metabolic coupling with lactate was assessed for the healthy disc by comparing with a simulation where pH was constant and equal to 7 . 1 , where oxygen reaction depends only on oxygen availability ( see Eq . ( 10 ) in Methods ) . Fig . 2 shows oxygen and lactate distributions in a sagittal section of the disc model with and without poromechanics-transport coupling , and for a transient analysis corresponding to the end of the second 16-hour creep period . For all cases , the anterior annulus fibrosus ( AF ) presented the lowest oxygen concentrations and the highest lactate levels . Local oxygen concentration is shown over time ( Fig . 3 and 4 ) for two nodes at the central NP and the anterior AF ( red dots in Fig . 1d ) . With initial conditions of zero oxygen and lactate within the disc , steady state concentrations were reached in approximately 16–18 hours without mechanical loading . When mechanical deformation was considered , a steady-state solution never occurred . Instead , a repetitive pattern was identified following loading and recovery phases . In all cases , with mechanical coupling , maximum oxygen and lactate concentration changes occurred at the end of the creep compression . Shortening the diffusion distance by 10% in the undeformed healthy disc height with constant diffusivities increased the oxygen levels in both the central NP ( up to 57% ) and the anterior AF ( up to 11% ) . Simultaneously , the lactate concentration decreased by a maximum of 27% in the NP and 22% in the AF , when compared to the undeformed case ( Fig . 3 ) . Including strain-dependent diffusivities resulted in a subsequent oxygen decrease in both the NP ( up to 17% ) and the AF ( up to 18% ) with a coupled counter-balanced lactate increase ( up to 16% in the NP and 10% in the AF ) , in comparison to the model with decreased diffusion distances and constant diffusivities . Merging together both strain-dependent diffusivity and diffusion distance changes caused an oscillating increase of oxygen for the central NP with a peak of 31% compared to the undeformed case , while a maximum of 9% oxygen decrease was calculated for the anterior AF . For lactate , with both strain-dependent diffusivities and distance changes , a similar decrease in both the anterior AF and the central NP up to 15% was found , compared to the undeformed case ( Fig . 3 ) . Advective effects were insignificant . Finally , when neglecting the lactate-dependence ( via pH changes ) of oxygen reaction , we found a 2 . 5% decrease in oxygen concentration in the anterior AF and 8% increase in the central NP , with respect to the lactate-coupled solution without poromechanical coupling ( Fig . 4 ) . A cyclic compression with an average load of 0 . 5 MPa was applied during 400 seconds . Two different frequencies ( 1 Hz and 0 . 1 Hz ) , and two different amplitudes ( A1 and A2 ) at 1 Hz were considered ( Fig . 1c ) . The initial conditions of the transport model during the dynamic loading simulations corresponded to the preconditioned solution of lactate and oxygen at the end of the two days of diurnal cycles . The different dynamic loading modes were compared to each other and with a creep compressive loading of 400 seconds ( A0 in Fig . 1c ) . We computed negligible effects ( <0 . 5% relative differences ) of amplitude A1 vs . A2 , frequency 1 Hz vs . 0 . 1 Hz , and amplitude A1 and A2 vs . creep loading amplitude A0 ( data not shown ) . The parameters shown in Table 2 were varied one by one in order to compare oxygen and lactate profiles to a base model ( with the healthy properties in Table 1 ) , including all mechanical and metabolic couplings . Such changes were related to DDCs reported in experiments to assess the sensitivity of the model under realistic parameter ranges . A reduction in the overall IVD cell density ( both AF and NP ) from the base model was simulated based on the experimental observation of substantial increase in cell apoptosis [24] and decrease in cell activity [25] with degeneration . Porosity decrease in both AF and NP from healthy to degenerated was also reported [26]–[29] . Values were thus reduced to simulate degeneration and study the sensitivity of the model within this range . Similarly , since solid phase stiffening during degeneration was reported in both AF [1] , [27] and NP [26] , the sensitivity of the model predictions was evaluated to global solid phase stiffness increase in both sub-tissues . Finally , a decrease in pH from healthy to degenerated IVDs and a decrease in NP swelling pressure [30] ( related to the proteoglycans loss [25] ) were studied as detailed in Table 1 and 2 . All the above parameter changes , following the corresponding experimental sources , referred to a grade of degeneration larger than 2 . 5 in commonly used grading systems [31] , [32] . The sensitivity of the metabolic transport outcomes was also evaluated with parameters from other IVD computational models [16] , [33]–[35] , even though such values might lack experimental support: ( i ) cartilage endplate permeability ( , see Methods ) , varied over several orders of magnitude in the literature [36] , [37] , [25] and thus was increased one order of magnitude ( as reported during ageing and degeneration [25] ) ; ( ii ) cartilage endplate stiffness and Poisson's ratio values taken from [33] were both lowered up to the values reported in [16] and ( iii ) bony endplate ( BEP ) stiffness , taken in the base model as 10 GPa [16] was decreased one order of magnitude as in [34] . Fig . 5 shows the sensitivity in terms of solute ( oxygen or lactate ) concentration under a given parameter change , normalized to the base model value , for both AF and NP . Fig . 5 refers to results at the end of the two simulated days . Within both the AF and the NP , oxygen and lactate concentrations were significantly more sensitive to porosity , cell density and swelling pressure changes than to variation of the other parameters , i . e . CEP permeability , pH and solid phase stiffening . In average , over both the AF and the NP , the oxygen concentration decreased by 44% and the lactate concentration increased by 36% when porosity was reduced . When cell density was reduced , the oxygen relatively increased by 42% and the lactate decreased by 21% . When the NP swelling pressure was reduced , a 44% relative increase in oxygen and a 23% relative decrease in lactate concentration were computed . As for the remaining parameters , less than 3% of relative changes in solute concentrations were found . Finally , CEP and BEP stiffness variations gave differences in oxygen and lactate concentrations lower than 0 . 2% ( data not reported in Fig . 5 ) . The sensitivity results in Fig . 5 were similar to those found at the end of the sustained creep ( data not shown ) . The healthy base model was compared with a model in which all the relevant DDC-related material properties reported in Table 1 were considered , i . e . disc height , porosity , cell density and swelling pressure reduction , and the solid phase stiffening ( Table 1 ) . However , the histological distinction between the two AF and NP subtissues was preserved , assuming a mild or moderate degenerated disc with a grade of 2 . 5–3 in the Thompson scale [38] , [32] . Fig . 6 shows the oxygen and lactate concentrations predicted at the end of the second 16-hours creep compression along the mid-transversal anteroposterior paths of both the healthy and degenerated disc models . Transport equations and mechanical deformations were alternatively coupled ( deformed ) and decoupled ( undeformed ) . While mechanical coupling tended to favor oxygen concentration and limit lactate accumulation within the healthy disc , it had only scarce effects within the degenerated IVD model .
A computational approach based on poromechanics and metabolism was developed and applied to investigate oxygen transport within the human IVD . First , the relevance of the mechanical and metabolic couplings was evaluated . Second , different degenerative changes were simulated with the proposed method to assess their effects on metabolic oxygen transport . In the healthy disc , predictions for oxygen distribution were most sensitive to mechanical coupling , this effect being predominant over that of the metabolic coupling . Advective transport by fluid movements was negligible in daily loading modes and after short period of cyclic compression . Mechanical effect acted via diffusion by local porosity changes , affecting diffusivities , and by global changes in geometry , affecting the diffusion distances . Unsurprisingly , mechanical effect was thus mainly observed when large volume changes were present , which was favored by healthy disc material properties . Disc height , cell density , NP swelling pressure and porosity DDCs affected more oxygen transport than AF and NP solid-phase stiffening . Effects of ( i ) daily disc deformation , and ( ii ) disc height change due to fluid loss with aging and degeneration on transport were studied . In the central disc , we found a general increase in oxygen concentration , induced by both sustained compression and permanent geometrical change . Such phenomena occurred in concomitance with a decrease of lactate levels , as a consequence of the oxygen-dependent lactate metabolic rate . These results are in agreement with other predictions [21] in which oxygen enhancement and lactate reduction were found following a volume loss , simulated by altering the IVD model dimensions . Our study further focused on the dual effect of a permanently reduced height together with daily deformations due to loading , as happens in a degenerated disc under sustained compression . While volume changes in the simulated healthy IVD during the load-recovery phases had a remarkable effect on solute levels , in the degenerated disc model , such effect was reduced because of decreased fluid content and swelling pressure , as well as increased influence of the solid phase ( Fig . 6 ) . This outcome could indicate how compression , here related to the fluctuation of solute concentrations , could be seen as a “healthy” condition for the disc [7] . Interestingly , the beneficial effect of compression through volumetric deformations could partly explain the fact that no differences in cell density were observed between male and female discs [3]: despite the increased diffusion distances in male discs , if the disc is sufficiently healthy , large dimensions would allow for increased deformability and improved disc maintenance through mechano-transport coupling . Moreover , the observed reduction in the normal load-dependent solutes pattern that occurred when degenerated material properties were simulated could be seen as an IVD degenerative catalyst . Distributions of oxygen and lactate predicted by our 3D model are difficult to validate as it would require invasive and complicated experimental procedures . Nevertheless , minimum oxygen pressure at disc mid-height was about 1 kPa in all models , which was close to the minimum value found in-vivo [13] in patients with scoliotic and back pain ( 0 . 71 kPa ) . Also , the maximum lactate concentrations , predicted nearby 3 . 5 nmol , were within the measured experimental range of 2–6 nmol [13] , and the location of minimum oxygen and maximum lactate contents were in agreement with other studies [17]–[19] . Fig . 7 shows the results of oxygen and lactate in the present study against the experimental range normalized with the boundary value found for each patient [13] . Normalization was performed because the solute concentrations reached within the disc strongly depend on the patient-specific vertebral blood supply . Although the comparison was made only with the lumbar discs , the experimental variability is still very high . However , a similar anteroposterior trend can be found between our study and the experimental results for both lactate and oxygen concentrations ( in particular for experimental curves in black in Fig . 7 ) . Such a comparison strongly suggests that the vertebral blood supply condition is a relevant factor that drives the absolute values predicted by any metabolic transport model and could also be relevant for the onset of possible degenerative changes [25] . Because of the large amount of calculations already performed in the present study , and because of scarce information in the literature about IVD boundary conditions in terms of oxygen concentration , such parameter was not explored . However , the sensitivity study of this work provides new information on the influence of parameter changes as measured in disc degeneration ( stiffening of solid phase as well as disc height , fluid content , swelling pressure and cell density decreases ) . The different responses of the AF and NP to mechanical coupling were related to the combination of subtissue-specific deformation modes and changes in diffusivity . In fact , the changes in diffusion distance caused an increase of oxygen and a decrease of lactate in both subtissues . Also , the strain-dependent diffusivity caused a decrease of oxygen and an increase of lactate when compared to a deforming disc with constant diffusivities , both in the AF and the NP . When strain-dependent diffusivity and shortening of diffusion distance were combined together , changes in solute concentrations were regionally opposite in the two substissues , because AF was significantly less sensitive than the NP to shortening of diffusion distance ( blue lines vs . red lines in Fig . 3 ) . In the anterior AF where the oxygen availability was already low , changes to diffusion-related parameters resulted in an oxygen deprivation . The location of such critical regions depends strongly on the disc geometry , distance from blood vessels , and regional water content loss , i . e . on patient specific characteristics . Finally , diffusion distance shortening due to mechanical loads , in the healthy IVD model gave 15% of disc height change after 7 hours under 0 . 5 MPa compression , which was comparable to in vivo deformations of 10–15% measured with comparable loads [39] . Fluid velocity enhancement for nutrients was negligible under higher frequency and amplitude loading modes . The highest fluid velocities were around 0 . 5 µm/s , at 1 Hz frequency and for the A2 load amplitude . Even if a particle was submitted to such velocity magnitude permanently during the simulated 400 seconds , the advective transport distance would have been 0 . 2 mm . Therefore , within the AF and the NP , considering both the low tissue permeability and the characteristic dimensions two orders of magnitude higher than the advective distance , fluid velocities are unlikely to transport small solutes . Moreover , AF and NP permeability changes ( although not simulated as a DDC following the experimental findings of [1] ) are known to have a negligible role in the displacement field calculation [40] and thus on the IVD volume changes . Therefore , the solute transport sensitivity is expected to be negligible via load-dependent diffusion . The negligible role of fluid velocity advection is congruent with other computational [16] and experimental evidences [11] , [41] , [12] . Yet , Huang and coworkers [17] predicted small solute enhancements due to dynamic loadings after 200 cycles and at a frequency of 0 . 1 Hz . However , advection and diffusion were not explicitly presented in the study of Huang et al . and dynamic vs . static loading differences could be also attributed in part to the loading-dependent diffusion . The endplates solute permeability change may also have played a role , while we did not simulate any alteration of solute input values at the endplates to avoid biasing the effect of the considered DDCs . Recent computational studies on IVD metabolic transport have included glucose [18]–[21] , which has a high influence on cell viability [20] , [42] . Since we did not consider local cell matrix synthesis and viability , glucose was ignored . In this first approach we demonstrated that cell density disturbance had a significant effect , since it is related with both reactive ( metabolic ) terms for oxygen and lactate . Cell density decrease has also been related to poor nutrient supply due to abnormalities of the CEP [3] . Thus , glucose concentration may play an important role if coupled to cell death . Likewise , high lactate concentrations resulting from a poor waste product removal could be detrimental to cell survival and should be taken into account via a pH and/or glucose dependent cell death , which would lead to the understanding of the degenerative processes under a cell biology-based chronology . On the other hand , since the total number of cells have also been shown to increase with degeneration [25] our simulated decrease in cell density could be optimistic with respect to a situation in which a non-functional cell number increase would occur and thus , a more severe condition in terms of oxygen consumption . In summary , the developed method to investigate the poromechanics and metabolic-coupled oxygen transport revealed that mechanical loads can significantly affect oxygen and lactate predictions when large and prolonged volume changes are involved . Such phenomenon was caused by the deformation-dependent nature of both tissue diffusivity and diffusion distances . The mechanism was regional- and solute-dependent within the disc . By applying the proposed approach together with IVD degenerative changes , we conclude that cell density drop , disc height reduction and decrease in NP swelling pressure due to loss of proteoglycans would significantly alter the interactions between mechanical loading and disc nutrition and be detrimental to the diffusion of nutrients . Thus , disc nutrition is most likely part of the synergy suggested between disc degeneration and physical activity [8] , [2] . In a healthy disc , it was found that mean compressive load variations in a daily cycle could be beneficial to disc maintenance . The approach could be used in regenerative and preventive medicine as a patient-specific numerical tool to explore the chronology of events in disc degeneration , in which the input parameters ( such as geometry , diffusivity , hydration ) could be derived from diagnostic images . The computational framework developed could also serve to study cell-loaded disc substitute materials .
A 3D FE poromechanical model coupled with an ADR transport model was created . The geometry of a L4–L5 IVD model was taken from an accurate spinal segment FE model [43] . All the relevant subtissues were represented: nucleus pulposus , annulus fibrosus , cartilage endplate ( CEP ) , bony endplate ( BEP ) , cortical bone and trabecular bone . For both the AF and NP , the solid porous skeleton was treated as a compressible Neo-Hookean material . In such a poro-hyperelastic formulation [44] , the solid grains were intrinsically incompressible and the porous material was considered to be saturated with an incompressible fluid phase , i . e . water . The total stress tensor caused by external loadings was the superimposition of the porous solid stress and the fluid pore pressure , p , that were respectively derived from a strain energy density function , , and Darcy's law: ( 1 ) ( 2 ) ( 3 ) In the above equations , G and K are respectively the shear and the bulk modulus , with the deformation gradient tensor , is the first strain invariant , is the pore fluid velocity , is the tissue porosity , and is the hydraulic permeability tensor of the tissue . With both phases being incompressible , in a large strain formulation the porosity varies with deformation with respect to an initial value as . Soft tissues permeability is highly strain-dependent and different exponential constitutive laws have been proposed [45] , [46] relating the isotropic hydraulic permeability with the porosity , the volumetric strain and with an initial value . The following laws were used: ( 4 ) ( 5 ) being and empirical coefficients , and the second-order unit tensor . In the AF , the porous solid strain energy density , , was the sum of Eq . 1 , and an additional term accounting for the anisotropic and nonlinear fibre-induced strengthening [47]: ( 6 ) The additional term depended on fibre stiffness parameters and and on the tension-only quantity , active only in two opposite fibre directions ( α = 1 , 2 ) and related to the fibre criss-cross distribution observed in AF . Different distributions of fibre orientation explained the regional differences in AF mechanical behaviour as detailed in our previous study [22] . For the NP , proteoglycan-induced NP swelling was described by considering the fluid pressure as a sum of the water chemical potential , , and a swelling-pressure related term , , that was assumed constant during deformation [48]: ( 7 ) For the ADR transport model , the tissue-averaged continuity equation can be expressed in the following fashion [49]: ( 8 ) where is the volume-averaged solute concentration , is the tissue diffusivity , and is the metabolic reactive term . Solute transport was solved by using a thermal-transport analogy [50] . To take into account advection and volumetric changes , a sequential approach was implemented so that the poromechanical results affected the transport solution . A Mackie-Meares diffusivity was used [51] , [52] , which relates the volume-averaged isotropic diffusivity of each solute with the updated porosity of the medium and the solute diffusivity in water : ( 9 ) Fluid velocities were computed from the poromechanical analysis and used as input for the advective term . All calculations were performed with ABAQUS 6 . 9 ( Simulia , Providence , RI , USA ) . However , since this commercial package did not allow modeling advection together with strain-dependent diffusion , the advective-diffusive flux was modified via user subroutine by adding a stabilization term to avoid oscillations in the results [53]: ( 10 ) This approach , dependent on the temporal and spatial discretizations , was verified against theoretical results using quadratic elements that fulfilled the condition , where was the Courant number , and and the time step and the mean element length , respectively . Finally , metabolic reactions from experiments on IVD cells [10] were used for: | Low back pain is a very common pathology in industrialized countries , often due to bad posture . It is also highly related to intervertebral disc aging . Aging of the disc is a normal process characterized by series of changes in its structure and function . The events that convert normal aging into degenerative disease is still not clear . Complications such as limited nutrition and possible reduction of disc cells with age make the issue intricate and multi-factorial . Using a numerical model that includes both nutritional and mechanical components , we found two different situations when looking at the effect of external loadings on two important cell solutes related with disc metabolism: oxygen and lactate . The effect of mechanical loading was greater when compressing a healthy disc than a degenerated one and promoted fluctuations of solutes concentrations . Also , changes in cell density seem fundamental in the process of disc degeneration and its causality with other degenerative changes should be further investigated . The importance of both mechanical and cellular patterns to maintain a healthy condition provides new insights to the field of disc regenerative medicine . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"biotechnology",
"bioengineering",
"biomedical",
"engineering",
"engineering"
]
| 2011 | The Effect of Sustained Compression on Oxygen Metabolic Transport in the Intervertebral Disc Decreases with Degenerative Changes |
The evolution of sterile worker castes in eusocial insects was a major problem in evolutionary theory until Hamilton developed a method called inclusive fitness . He used it to show that sterile castes could evolve via kin selection , in which a gene for altruistic sterility is favored when the altruism sufficiently benefits relatives carrying the gene . Inclusive fitness theory is well supported empirically and has been applied to many other areas , but a recent paper argued that the general method of inclusive fitness was wrong and advocated an alternative population genetic method . The claim of these authors was bolstered by a new model of the evolution of eusociality with novel conclusions that appeared to overturn some major results from inclusive fitness . Here we report an expanded examination of this kind of model for the evolution of eusociality and show that all three of its apparently novel conclusions are essentially false . Contrary to their claims , genetic relatedness is important and causal , workers are agents that can evolve to be in conflict with the queen , and eusociality is not so difficult to evolve . The misleading conclusions all resulted not from incorrect math but from overgeneralizing from narrow assumptions or parameter values . For example , all of their models implicitly assumed high relatedness , but modifying the model to allow lower relatedness shows that relatedness is essential and causal in the evolution of eusociality . Their modeling strategy , properly applied , actually confirms major insights of inclusive fitness studies of kin selection . This broad agreement of different models shows that social evolution theory , rather than being in turmoil , is supported by multiple theoretical approaches . It also suggests that extensive prior work using inclusive fitness , from microbial interactions to human evolution , should be considered robust unless shown otherwise .
The eusocial insects have occupied an important place in biology because of their extraordinary levels of cooperation [1–4] . In ants , termites , some bees , some wasps , and a few other taxa , certain individuals , called workers , give up their own reproduction in order to help others reproduce . Darwin was vexed over the question of how such reproductive altruism evolves or indeed how any traits of sterile workers evolve , but he believed that such sterility was due to some form of selection at the family level or at the group level [5] . Hamilton provided the first rigorous treatment of this idea , with a key insight being the importance of genetic relatedness [1] . A conditional gene causing a worker to give up reproduction could be favored if it provided sufficient help to a relative who would share that gene at above-random levels . He showed that this process , which became known as kin selection , could be analyzed by summing up an actor’s fitness effects , each multiplied by the actor’s relatedness to the individual receiving the fitness effect . When this sum , called the inclusive fitness effect , is positive , the trait should be favored by selection . For giving up one’s reproduction ( fitness cost c ) to benefit other individuals ( total fitness gain b ) related by r , the inclusive fitness condition is −c + rb > 0 . Kin selection and inclusive fitness became the dominant modes of thinking about the evolution of eusocial insects [4 , 6 , 7] , and their success in this area has led to them being applied to many other problems in social evolution [8–12] . Recently , this paradigm was criticized by Nowak et al . [13] , who argued that inclusive fitness was an inaccurate and unnecessary method and that kin selection was not a very useful way to think about social evolution . Both of these conclusions have in turn been extensively criticized as depending on multiple misconceptions [14–22] . We concur with many of these criticisms but do not revisit them here . Instead , we offer a different kind of critique of the Nowak et al . paper . To provide an example that bolstered their general arguments , Nowak et al . [13] also developed their own mathematical model of the evolution of eusociality , presenting it as an example of a modeling approach that is superior to inclusive fitness modeling . However , as has been recently pointed out [23] , this eusociality model has scarcely been addressed . We do not contest this modeling approach . Instead , we accept it as valid and use it to show that its implementation in Nowak et al . [13] led to errors of interpretation that greatly overstated any differences with standard inclusive fitness results . We do not address the exact quantitative match of the two approaches but instead focus on large apparent discrepancies of interest to empiricists . Because their model is claimed to be superior to inclusive fitness , we focus on three of their conclusions that seem at greatest variance with the conventional inclusive fitness and kin selection view of the evolution of eusociality . In each case , we will show that the kin selection view is essentially confirmed . Nowak et al . [13] also make other assertions about eusociality that are consistent with inclusive fitness theory , such as the importance of grouping and preadaptations . We ignore these in order to focus on the seemingly novel conclusions of the Nowak et al . model . The first two of these are fundamental qualitative differences from inclusive fitness , while the last is more a difference in degree . First , Nowak et al . [13] , following earlier work by Wilson [24 , 25] , claimed that relatedness was not an essential element in the evolution of eusociality . They wrote that “relatedness is better explained as a consequence rather than as the cause of sociality , ” that “grouping by family hastens the spread of eusocial alleles but it is not a causative agent , ” and that “relatedness does not drive the evolution of eusociality” [13] . In the same vein , they also contest empirical evidence that relatedness is important [13] . We take causality to mean that variation in relatedness leads to variation in the likelihood of evolving eusociality . As has previously been pointed out , the Nowak et al . model could not test this because it was based on groups of relatives , with no comparable model of unrelated individuals being presented [15 , 20] . Nowak et al . appear to have partially accepted this point: “One , we do not argue that relatedness is unimportant . Relatedness is an aspect of population structure , which affects evolution” [26] . However , this response leaves unanswered exactly how it affects evolution . At least one of the authors [27] continues to assert that relatedness only hastens the spread of alleles and that it is not causal . To test these claims , we extend their model to cases in which relatedness can vary . Second , whereas inclusive fitness theory has emphasized that cooperation occurs in the face of potential and actual conflicts among colony members with different interests [4 , 7 , 28 , 29] , Nowak et al . [13] assert that the colony as a whole is all that matters . They argue that “the workers are not independent agents , ” that “their properties are determined by the alleles that are present in the queen ( both in her own genome and in that of the sperm she has stored ) , ” that “the workers can be seen as ‘robots’ that are built by the queen , ” and that they “are part of the queen’s strategy for reproduction” [13] . Nor , contrary to earlier work by Wilson [24 , 25] , do they brook any conflicts between levels of selection: “there is only one level of selection , the hymenopteran colony , which is treated as an extension of the queen , whose genes are the units of selection” [13] . To test whether workers and queens are independent agents that are selected differently , we construct parallel models in which the genes determining whether their offspring stay and help are expressed in mothers or expressed in offspring . Finally , Nowak et al . [13] claim that eusociality is harder to evolve than has been appreciated . They write that “a key observation of our model is that it is difficult to evolve eusociality , because we need very favorable parameters” and that “despite the obvious and intuitive advantages of eusociality , it is very hard for a solitary species to achieve it” [13] . If there is any novelty in this conclusion , it must be that eusociality is harder to evolve than has been thought previously; that is , it is harder to evolve than predicted from inclusive fitness effects ( –c + rb > 0 ) . We explore how this conclusion changes with reasonable alterations in the fitness functions and the worker decision rules . If the three apparently novel conclusions of Nowak et al . are correct [13] , then inclusive fitness theory could be said to have made some serious errors , and we might have to throw out or rethink important elements of the last 50 years of social evolution theory . If instead our models reject those apparently novel conclusions in favor of results consistent with those obtained through inclusive fitness , it would show that different theoretical approaches yield broadly consistent results , as they ought to in a healthy science .
We modify the Nowak et al . [13] haploid model , which is simpler than their haplodiploid one but sufficient to demonstrate the important points . Our goal is not to exactly model eusociality in any particular organism but to examine the logic and truth of three general claims in Nowak et al . [13] , claims that pertain to both the haploid and haplodiploid models . The basic model includes solitary and eusocial genotypes expressed in offspring , where solitaries always leave to reproduce , while eusocials stay and help their mother with probability q and leave to reproduce with probability 1 – q . Mothers and offspring are genetically identical . Differential equations describe changes in the numbers of solitary individuals and eusocial colonies based on colony-size–specific queen birthrates ( bi ) and death rates ( di ) , as well as worker death rates ( α ) and density dependence ( η ) ( see Methods , Equation 1 ) . If larger colony size ( more workers ) sufficiently increases the queen’s birthrate and/or decreases her death rate , the eusocial type can be favored over solitary reproduction under some probabilities of staying q . Using these equations , we recovered results indistinguishable from those of Nowak et al . [13] ( e . g . , their Figure 4 ) . We then explored the effects of various assumptions by changing them one by one . First , the models of Nowak et al . [13] assumed eusocial offspring stay with their mother so that there was always genetic relatedness among participants . In the haploid model , this meant that helpers were genetically identical ( r = 1 ) to their mother and to the siblings they raised . To vary genetic relatedness in the haploid model , we allowed some offspring mixing between mothers before implementing their genetic helping rules . Each offspring has a probability r of being with her own mother before deciding whether to help her or leave to reproduce and a probability 1 – r of being with a random mother . This could result from offspring movement between nests , from mothers laying a fraction of their eggs in other nests , or from nest usurpation [30 , 31] . r is equivalent to relatedness to the new mother ( after movement ) because it represents identity to that mother above chance levels; a fraction r is identical to the head of their colony and her offspring ( r = 1 ) , while the remainder are randomly associated with colonies ( r = 0 ) . After this temporary mixing , offspring execute the original Nowak et al . strategies: offspring with the solitary genotype always leave to reproduce alone , and offspring with the eusocial genotype stay and help their colony with probability q . Differential equations implementing this model are given in the Methods ( Equation 2 ) . The filled circles in Fig . 1 show when selection on offspring favors eusociality under varying relatedness r , worker-assisted queen birthrate b , and probability of staying q ( other parameters continue to match the standard Nowak et al . Figure 4 parameter values ) . Lowering relatedness clearly makes it more difficult for eusociality to evolve; with lower r , a higher b is required to favor eusociality . In the extreme , when offspring are randomly associated with colonies so that relatedness is zero , even b = 500 ( a 1 , 000-fold increase in the queen’s birthrate due to helpers ) is insufficient to favor eusociality . As expected from inclusive fitness theory , relatedness is causal in the sense that some relatedness is necessary for eusociality and increasing relatedness increases the range of conditions allowing eusociality to evolve . Second , to address the issue of whether worker offspring are independent agents or simply robots carrying out the queen’s interests , we need to compare models of control by different agents . This means comparing models in which the decision to stay and help is made by genes in offspring bodies to models in which it is made by genes in the resident queens’ bodies . Though Nowak et al . [13] seem to argue for queen control , their models are for offspring control because they generally assume that genes expressed in worker bodies determine the decision to stay or leave . However , inclusive fitness theory predicts that when queen control is possible , it will generally be more favorable for evolving eusociality [7] unless relatedness is one , in which case no conflict is expected . To model queen control under varying relatedness in the haploid model , we allowed offspring to mix exactly as in the offspring control model above but then allowed the resident queen’s genotype to determine if her mixed offspring pool helps or not . If the mother has the solitary genotype , all of her mixed pool disperses to become reproductives; if the mother has the eusocial genotype , she causes a fraction q of her offspring pool to stay and help her , independent of offspring genotype . Differential equations governing this system are given in the Methods ( Equation 3 ) . As predicted by inclusive fitness theory , eusociality evolves much more easily under queen control ( Fig . 1 , all circles ) . The only exception , as expected under inclusive fitness theory , is when there is no mixing between nests so r = 1 and the two decision rules are selected identically . In fact , assuming that queens can control the trait , we see the expected opposite relationship with relatedness; the less related the queen is to the offspring in her colony , the more the queen is selected to cause them to be workers . The final claim that we examine is that eusociality is hard to evolve [13] . This depends on what is meant by “hard , ” but we can usefully ask whether eusociality is as difficult to evolve as is implied in the Nowak et al . [13] paper . Their claim seems based on particular and odd choices for fitness functions and worker decision rules . The fitness function that they generally explored was a threshold function in which workers add no fitness gains to the queen below a colony of size m and add a fixed gain ( increasing queen b or decreasing d ) in colonies at or above size m , regardless of how many workers are added . This means that workers in colonies below that threshold contribute nothing until enough further workers join and that workers above the threshold also add nothing extra unless other workers die , returning the colony to the threshold . If most workers are contributing nothing , then it is not surprising that eusociality would be hard to evolve . In the example most explored , the threshold colony size m was set at 3 ( their Figure 4 ) , such that two workers were needed to raise the queen’s birthrate from b0 = 0 . 5 to b = 4 and to lower her death rate from d0 = 0 . 1 to d = 0 . 01 ( they also let α = 0 . 1 and η = 0 . 01 ) [13] . This 8-fold increase in the queen’s birthrate allowed eusociality to evolve for some values of q , but lower values of b did not allow eusociality to evolve . Not surprisingly , requiring more workers before the queen increased fitness ( higher m thresholds ) made eusociality even more difficult to evolve . As noted above , the assumption that workers must stay with probability q , regardless of the state of the colony , means they may be maladaptively staying in colonies that are too large to gain further benefits . It should be easy for workers to avoid this problem . For example , they might instead implement the rule to stay when the colony is below some threshold size w and leave when it is at or above that size . We implemented differential equations to model this change of assumption ( see Methods , Equation 4 ) in the original Nowak et al . model with worker control and r = 1 ( i . e . , independently of the other changes explored above ) . Eusociality does evolve more readily . For example , for the same parameter values as in Figure 4 of Nowak et al . , eusociality can now be favored under a somewhat lower benefits threshold ( b = 3 ) , that is , when helped queens get a 6-fold advantage . In addition , the threshold fitness function assumed by Nowak et al . [13] prevents the earliest workers from contributing anything . However , it is easy to envision advantages that would come from having only a single worker [25 , 32] . To view this effect in isolation , we return to the Nowak et al . [13] decision rule ( stay with probability q ) and to their parameter values given above but allow a single worker to add half the contribution to the queen that two workers add ( for both birthrate and death rate ) ( m = 3 , b0 = 0 . 5 , d0 = 0 . 1 , d = 0 . 01 , α = 0 . 1 , η = 0 . 01 ) . This simple change ( implemented in Equation 1 ) makes it much easier to evolve eusociality , with b = 1 . 5 or only a 3-fold increase required ( Fig . 2 ) versus 8-fold with the threshold model . This analysis does not resolve what actual fitness functions and decision rules apply in nature , but we note that evolution tends to take the easiest paths available and eschew the difficult ones . This result appears very close to what is expected under inclusive fitness when r = 1: if two workers increase queen birthrate from 0 . 5 to 1 . 5 , each raises it by 0 . 5 , exactly the amount that the worker gives up by helping . However , the comparison is not accurate for two reasons . First , this comparison of birthrates neglects the workers’ effect on queen death rate in the model . Second , having gone back to the stay-with probability q decision rule , some workers waste their efforts by joining large colonies . In order to compare more closely with inclusive fitness , we altered both of these: the queen death rate is now unchanged by workers , and the stepwise birthrate function is implemented together with the stay-below-colony-size-w decision rule . For w = 3 , eusociality is not favored at b = 1 . 5 ( where inclusive fitness predicts it to be neutral [workers giving up 0 . 5 and adding 0 . 5 to the queen] ) but is favored to evolve at b = 1 . 6 . It is still possible to argue that eusociality is hard to evolve , depending upon one’s standard for what is hard , but it is considerably easier to evolve than implied by the initial Nowak et al . model and , not surprisingly hard relative to inclusive fitness predictions .
The controversy over the Nowak et al . paper has mostly been conducted at rather abstract levels; different researchers prefer different modeling strategies and may also interpret the evidence differently [13–20 , 26] . We take a different and more concrete approach by investigating their model for the evolution of eusociality more deeply . If their methods are superior and raise novel insights , we should welcome them and perhaps question our older theories . If instead their methods lead to no novel insights , it undermines the larger claims that the model is used to buttress , specifically that inclusive fitness has not been useful . We have therefore followed the recommendation of Nowak et al . [13] for modeling social evolution , and in particular eusociality , using deterministic evolutionary dynamics described by ordinary differential equations . However , stimulated by inclusive fitness thinking , we have sought to understand apparent differences between their results compared to previous models . In every case , we find that their rejection of accepted results is incorrect and that in fact the insights known from inclusive fitness theory also emerge using their method . The claims that relatedness only hastens the spread of eusocial alleles and that relatedness is not causal [13 , 27] are shown by our models to be false . The proposition could not be tested in the Nowak et al . [13] models because they did not examine any low-relatedness case [15 , 20] . We have modeled variable relatedness and shown that , under offspring control , high relatedness broadens the range of conditions allowing eusociality to evolve . Relatedness affects not just speed of selection but whether it is favored at all; when relatedness is zero , eusociality does not evolve even with very high benefits ( increasing queen birthrate 1 , 000-fold ) . This shows that relatedness plays an essential and causal role . Of course , these are not surprising findings because the importance of relatedness was previously well understood from many kinds of models using inclusive fitness [1 , 7] , population genetics [33–35] , quantitative genetics [36–38] , and game theory [39 , 40] , as well as being supported by much empirical evidence [7 , 9 , 41 , 42] . An alternative interpretation of Nowak et al . ’s [13] views is that relatedness is not causal because high relatedness does not always drive the evolution of eusociality . However , this is a rather empty view since no one has ever asserted the contrary and Hamilton’s rule explicitly includes other factors that interact with relatedness . In addition , this view would negate most biological causality of any kind , as no single factor ever completely determines outcomes . Finally , if Nowak et al . agreed that variation in relatedness is an important determinant of eusociality , which is widely regarded as the most important contribution to the topic in 50 years , why did they not say so , instead consistently arguing against its significance ? This pattern extends beyond the Nowak et al . [13] paper to Wilson’s earlier and later papers [24 , 25 , 27] and to work from Nowak’s group purporting to show new pathways to cooperation [43 , 44] that in fact depended critically on relatedness and could be interpreted via inclusive fitness [45–47] . Whatever view of causality is taken , it is important to be clear that the Nowak et al . [13] modeling strategy is just like others in showing that higher relatedness is an important factor promoting higher cooperation . A second claim of Nowak et al . [13] , that workers are robots and simply part of the queen’s reproductive success , cannot be made without testing and contrasting queen and worker decision rules . Nowak et al . [13] tested only offspring control models because the decisions are controlled by genes expressed in workers . It is a longstanding result of inclusive fitness theory that parents and offspring are agents with different interests that can be in conflict [28 , 48] . In particular , in the eusociality context , inclusive fitness predicts that offspring will be selected to help their mothers under a narrower range of conditions than the mothers would favor ( eusociality evolves more readily if mothers control the helping of their offspring ) ( pp . 58–63 of [7] ) . This follows from differences in relatedness . Workers should gain less from helping less-related kin , but queen inclusive fitness improves if she is less related to the workers who pay the fitness cost . To examine this question , one must compare selection of offspring agency ( genes expressed in the offspring determine whether she becomes a worker ) versus maternal agency ( genes expressed in the mother determine whether her offspring become workers ) . We therefore constructed haploid models for maternal control to compare with the results under offspring control . As predicted by inclusive fitness theory , the two cases evolve differently and can be in conflict: mothers favor helping by their offspring under a much broader range of conditions than the offspring themselves favor , except when mothers and offspring are genetically identical ( Fig . 1 , all circles ) . And as predicted , when relatedness is low and eusociality is very difficult to evolve under worker control , it is very easy to evolve if the queen has control , because the queen is unrelated to most of the workers who pay the fitness cost . If queens really were in control from the origin of eusociality , and if they could exert that control on unrelated offspring , that would be the easiest path to eusociality . However , this is contradicted by phylogenetic studies showing that relatedness was always high at the various origins of eusociality [41] . In contrast , the standard kin selection model of worker control predicts this observation . Finally , the claim that eusociality is difficult to evolve [13] is less fundamental than the other two claims and also less wrong because its truth necessarily depends on how one defines “hard to evolve . ” Eusociality has evolved a modest number of times and therefore could be viewed as hard to evolve , but their model does make it appear that eusociality is harder to evolve than has been believed . We show that this result hinges on assumptions that are heavily biased towards that conclusion . Little justification was given for why we should accept these particular assumptions . In particular , assumptions are made that imply that many workers waste their efforts . First , their model assumed that offspring stay with probability q , independent of any information that might be available about the need for workers . One advantage of inclusive fitness thinking is that it induces researchers to think of workers as agents being selected to get better outcomes ( higher inclusive fitness ) using whatever information is available to them . One such piece of information is the number of workers already present on the nest . In the threshold fitness model , there is no inclusive fitness gain to be had from staying above that threshold , unless some workers die , so we asked if there was some obvious better decision rule than stay with probability q . We therefore tested decision rules that have workers staying when the colony is below a threshold size ( not necessarily the same as the fitness threshold ) and leaving when the colony is above that size . Not surprisingly , we find that this class of decision rules makes it easier to evolve eusociality , because fewer workers are making wasteful decisions to stay in large colonies . Such a rule seems well within the capabilities of workers . They need not count adults . They simply need to be able to assess some reasonable correlate of the count , something that even microbes do when using quorum sensing to change their behavior . For social insects , the mechanism might involve the degree of comfort with contacting other adults or the hunger demands of offspring . Similarly , the threshold fitness model assumed by Nowak et al . devalues worker behavior at the other , low , end of colony sizes . In most of their model examples ( though not their general model ) , it was assumed that it was necessary to have two workers to provide any benefit at all to the queen ( m = 3 ) . That means that the first worker to join a colony provides nothing . However , it is easy to envision situations in which the first worker to join would provide real benefits [32] . The simplest is that at this point one individual can guard the nest while the other forages [25] . Empirical evidence suggests that first helpers do provide benefits [49–54] . If we modify the Nowak et al . threshold model to a step model in which each worker below the threshold adds an additional fixed benefit up to the maximum at colony size m , so that the efforts of unjoined first workers are not wasted , eusociality evolves much more easily . Thus , two modifications—the stepped fitness function and the altered worker decision rule—independently make it easier for eusociality to evolve . When we implemented these two rules together so that no workers waste their efforts and assumed workers affect only queen birthrates , eusociality evolved when predicted by inclusive fitness effects on birthrates . We do not know if this is general; the exact correspondence of the two methods may deserve additional study , but our goal here is to address the apparent major discrepancies . The method advocated by Nowak et al . [13] offers the advantage of specifying parameters like birth and death rates explicitly and following their effects over time while allowing some features , like colony size , to change . We expect that these methods can be used to generate interesting results . However , they are more complex and less intuitive than inclusive fitness thinking , so considerable care is needed to fully understand them . The common thread in the three errors pointed out in this paper is overgeneralization from narrow assumptions or particular parameter values . Relatedness was said to be unimportant even though the models did not vary relatedness . The assertion that workers are not independent agents was made in the absence of models that compared decision rules of different agents . Eusociality was said to be difficult to evolve based on specific and questionable assumptions about the fitness function and offspring decision rules . The more complex the model , the easier it is to be misled by particular results that are not general . In this case , the initial Nowak et al . model [13] missed not just minor details but perhaps the most important generalizations known from the last five decades of theory and empirical study: the importance of relatedness and conflict . Apparent lack of agreement with prior results should have triggered more than a quick rejection of inclusive fitness and kin selection; it should have led to a questioning of why the results were , or seemed to be , different . When examined more closely , models of the type advocated by Nowak et al . [13] do not overturn but instead reaffirm principles of social evolution discovered through inclusive fitness . To have multiple theoretical approaches converging on similar results attests to the robustness of social evolution theory .
Our models are all based on the haploid model of Nowak et al . [13] . They modeled the evolution of eusociality with systems of differential equations tracking the number of solitary queens ( x0 ) and eusocial colonies of size i ( xi ) . We use a modified notation because our low-relatedness models require us to also keep track of colonies headed by solitary-genotype queens . We therefore let ei be the number of colonies of size i headed by a eusocial queen ( that is with i – 1 workers ) and si be the number of colonies of size i headed by a solitary queen . With this modified notation , equation set 58 of Nowak et al . [13] can be written as: s˙1= ( b1ϕ−d1 ) s1e˙1=∑i=1∞biϕ ( 1−q ) ei−b1ϕqe1−d1e1+αe2e˙i=bi−1ϕqei−1−biϕqei−diei−α ( i−1 ) ei+αiei+1fori>1 , ( 1 ) where bi and di are the birth and death rates of colonies of size i , q is the probability that an offspring of a eusocial colony stays as a worker ( offspring of solitary colonies never stay ) , α is the worker mortality rate , and ϕ is a density-dependent correction factor equal to 1/ ( 1 + ηX ) , with X being the total population size including workers and η scaling the size of the system . For specific examples , Nowak et al . [13] usually assumed birthrates and death rates were governed by a simple threshold function: below some threshold colony size m , bi = b0 and di = d0 and at or above colony size m , bi = b and di = d . Using two numerical methods ( see below ) , we used Equation 1 to reproduce the results of Figure 4 in Nowak et al . [13] ( see S1 Code ) . The Nowak et al . models all assumed high and fixed relatedness . We modify their haploid model to incorporate a parameterized mixing step , which allows us to vary the degree of relatedness between queens and workers . The mixing occurs before offspring decide to be workers or reproductive queens . We allowed offspring to move to other mothers , eusocial or solitary , with probability 1 – r . Each moving offspring is replaced by a eusocial or a solitary offspring with probabilities fe and fs , which are simply the proportions of such offspring produced in the population: fe=∑ibiei/∑ibi ( si+ei ) fs=∑ibisi/∑ibi ( si+ei ) . After mixing , offspring execute their staying rule ( leave for solitaries and stay with probability q for eusocials ) . r is relatedness to the mother they help because r of the time she is identical , and 1 – r of the time she is genetically random or unrelated . For this offspring decision model , the equations describing changes in colony types are as follows: s˙1=∑i ( biϕ ( 1−r ) fsei ) +∑i ( biϕ ( r+ ( 1−r ) fs ) si ) −d1s1+αs2 s˙i=bi−1ϕ ( 1−r ) feqsi−1−biϕ ( 1−r ) feqsi−disi−α ( i−1 ) si+αisi+1e˙1=∑i ( biϕ ( r+ ( 1−r ) fe ) ( 1−q ) ei ) +∑i ( biϕ ( 1−r ) fe ( 1−q ) si ) −b1ϕ ( r+ ( 1−r ) fe ) qe1−d1e1+αe2e˙i=bi−1ϕ ( r+ ( 1−r ) fe ) qei−1−biϕ ( r+ ( 1−r ) fe ) qei−diei−α ( i−1 ) ei+αiei+1 . ( 2 ) Here ei and si still represent numbers after decision rules are executed and do not reflect numbers in the transient mixing stage . The equations were numerically solved using S3 Code . For maternal control , we implemented the same offspring mixing model but allowed the mother’s genotype to determine whether the offspring in her colony ( some of them resulting from mixing from other colonies ) stay and help . Thus , if the queen is eusocial , her ( mixed ) offspring will become new workers with probability q or new queens with probability 1 – q . If the queen is solitary , then all offspring will become new queens . The equations now become the following: s˙1=b1ϕ ( r+ ( 1−r ) fs ) s1+∑i ( biϕ ( 1−r ) fs ) ( 1−q ) ei ) −d1s1 e˙1=∑i ( biϕ ( r+ ( 1−r ) fe ) ( 1−q ) ei ) +b1ϕ ( 1−r ) fes1−b1ϕqe1−d1e1+αe2e˙i=bi−1ϕqei−1−biϕqei−diei−α ( i−1 ) ei+αiei+1 . ( 3 ) Note that , unlike the worker model , there are no solitary colonies larger than one ( after the transient mixing stage ) because a solitary queen always causes her offspring pool to disperse and become reproductive . The equations were numerically solved using S4 Code . To examine if eusociality is easier to evolve than suggested in Nowak et al . [13] , we tested alternative worker decision rule and fitness functions . First , instead of staying with probability q , eusocial offspring always stay when colony size i < w and always leave when i ≥ w . The equations are as follows: s˙1= ( ϕb1−d1 ) s1e˙1=∑i=w∞ϕbiei−ϕb1e1−d1e1+αe2e˙i=ϕbi−1ei−1−ϕbiei−diei−α ( i−1 ) ei+αiei+1for 1< i < we˙i=ϕbi−1ei−1−diei−α ( i−1 ) ei+αiei+1fori = we˙i=−diei−α ( i−1 ) ei+αiei+1for i > w . ( 4 ) These equations were numerically solved using S5 Code . We also altered the fitness functions from single thresholds to step functions . Now each added worker adds the same amount , up to the maximum b attained at colony size m . The maximum gain in both models is the same , but now each worker up to size m adds something . We can model this with Equation 1: if b0 is the birthrate of a solitary queen and b is the birthrate of a eusocial queen in colony size m , then we let the birthrate of queens in smaller colony sizes 1 < i < m be b0 + ( i −1 ) ( b − b0 ) / ( m −1 ) . Similarly , we let the queen death rate for colony sizes 1 < i < m be d0 + ( i −1 ) ( d − d0 ) / ( m −1 ) . This implementation of the Nowak et al . model was numerically solved using S2 Code . To solve the ordinary differential equations , we used two numerical methods . For Equations 1–4 , Euler's method was used in R to numerically determine the equilibrium population of the system , using a time step of h = 0 . 1 and a maximum colony size of n = 50 and terminating when either E or S population/number of individuals was less than ε = 0 . 1 or after a maximum of 50 , 000 time steps . Equation 1 was also solved with a first-order numerical procedure with the step size 0 . 1 implemented in MATLAB . The procedure was started with equal numbers of solitary females and eusocial queens ( n = 100 ) and was terminated when either the solitary or eusocial populations were extinct ( defined as less than 0 . 05 ) or both the solitary and eusocial populations stabilized at a maximum of 200 , 000 time steps . Both numerical methods successfully reproduced Figure 4 of Nowak et al . [13] . | The evolution of sterile worker castes in social insects has fascinated biologists ever since Darwin; how can selection favor a trait that decreases reproductive fitness ? W . D . Hamilton solved this dilemma in the 1960s with a theory showing that reproductive altruism could evolve if it increased the worker’s inclusive fitness , which included effects that it had on increasing the fitness of its relatives . This solution to a crucial evolutionary problem , sometimes called kin selection , was challenged in a recent paper . The paper generated much controversy , but no one has contested its new theoretical model of the evolution of eusociality , which appeared to overturn much of what was previously thought to be true from kin selection theory . Here we examine this model in greater depth , showing that its apparently novel conclusions are overgeneralized from narrow and often inappropriate assumptions . Instead , this modeling strategy yields results that confirm important insights from kin selection and inclusive fitness , such as the importance of relatedness and the existence of conflicts in social insect colonies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| []
| 2015 | Relatedness, Conflict, and the Evolution of Eusociality |
microRNAs ( miRNAs ) are important post-transcriptional regulators , but the extent of this regulation is uncertain , both with regard to the number of miRNA genes and their targets . Using an algorithm based on intragenomic matching of potential miRNAs and their targets coupled with support vector machine classification of miRNA precursors , we explore the potential for regulation by miRNAs in three plant genomes: Arabidopsis thaliana , Populus trichocarpa , and Oryza sativa . We find that the intragenomic matching in conjunction with a supervised learning approach contains enough information to allow reliable computational prediction of miRNA candidates without requiring conservation across species . Using this method , we identify ∼1 , 200 , ∼2 , 500 , and ∼2 , 100 miRNA candidate genes capable of extensive base-pairing to potential target mRNAs in A . thaliana , P . trichocarpa , and O . sativa , respectively . This is more than five times the number of currently annotated miRNAs in the plants . Many of these candidates are derived from repeat regions , yet they seem to contain the features necessary for correct processing by the miRNA machinery . Conservation analysis indicates that only a few of the candidates are conserved between the species . We conclude that there is a large potential for miRNA-mediated regulatory interactions encoded in the genomes of the investigated plants . We hypothesize that some of these interactions may be realized under special environmental conditions , while others can readily be recruited when organisms diverge and adapt to new niches .
Small RNAs are now accepted as major players in the control of eukaryotic gene expression . Most well known are microRNAs ( miRNAs ) and small interfering RNAs ( siRNAs ) , both of which are derived from the processing of dsRNA molecules by members of the Drosha/Dicer family of endonucleases . In plants , siRNA and miRNA are distinguished mainly by their biogenesis , not by their mechanism of action . MiRNAs arise from stem-loop precursors encoded in the genome , and their major mechanism of action in plants is thought to be post-transcriptional regulation through near-complementary base-pairing to target mRNAs , leading to specific endonucleolytic cleavage and degradation of the target [1] . Most of the initially discovered miRNAs were so highly conserved in evolution that a defining characteristic of a miRNA was that it had to be conserved [2] . This attribute of those miRNAs discovered early has been used successfully by a number of groups to computationally predict new miRNA genes [3–6] . Basically , these methods scan the genome for inverted repeats with the potential to form miRNA precursors . Such scans typically find on the order of hundreds of thousands to millions of hairpins , depending on genome size and search parameters [4] ( plus our own unpublished data ) . This high number is then reduced by only keeping hairpins that are conserved in other species . Another approach is to search only transcribed sequences in the form of expressed sequence tags [7 , 8] . This method works for nonsequenced genomes and efficiently reduces the search space , probably leading to a lower number of false positives , but the method also misses candidates not covered by the expressed sequence tag libraries . In miRBase version 8 . 2 , Arabidopsis thaliana ( Arabidopsis ) has 118 miRNA genes listed , most of which are conserved down to the monocot Oryza sativa ( Oryza ) . However , studies of noncoding RNA have shown that lack of conservation does not necessarily mean lack of function [9] . Potentially , all it takes to evolve a miRNA is for one of the many inverted repeats in the genome to be transcribed and have the necessary structure and sequence features to be recognized and processed by Drosha/Dicer . Indeed , large numbers of more narrowly conserved miRNAs also exist [10] . A recent bioinformatic study in human identified patterns associated with miRNA precursors and suggested that the number of miRNA precursors is larger than 25 , 000 [11] . In plants , a similar situation could exist . A deep sequencing effort in Arabidopsis using the massively parallel signature sequence ( MPSS ) technique has revealed 75 , 000 distinct small RNA species ( not all miRNAs , though ) [12] mapping to a large variety of genomic contexts , including exons , introns , repetitive DNA , and intergenic regions . This is perhaps not surprising considering other studies finding that unexpectedly large fractions of eukaryotic genomes are transcribed also outside and antisense to annotated protein-coding genes [13–15] . A necessary feature of any functional miRNA is that it must target at least one mRNA . In plants , this means that the miRNA must be almost complementary to some part of the spliced mRNA transcript ( not just the 3′ untranslated region as is currently thought to be the main target for animal miRNAs ) . A set of rules allowing mismatches only in certain positions has been suggested based on experimental observations [16] . The requirement for a target has previously been used to predict plant miRNAs [17–19]: instead of ( or in addition to ) relying on phylogenetic conservation ( intergenomic matches ) , these methods have successfully used intragenomic matches with potential target mRNAs to find the hairpins potentially capable of producing miRNAs that can regulate the target ( s ) . Such intragenomic matches will inherently arise from the structure and dynamics of the genome: retrotransposons , formation of pseudogenes , and other duplicative events provide sequences almost ready to regulate the originally copied gene [20]; likewise , the reverse strand of one gene is complementary to other paralogous genes . By not relying on conservation between species , intragenomic matching is capable of more fully charting the potential for post-transcriptional regulation by miRNAs . In an effort to reduce spurious predictions , earlier screens for new miRNAs have removed candidates overlapping existing annotation , such as repeats and protein-coding regions . Although such filters probably increase the signal-to-noise ratio , they also introduce biases assuming that repeat-derived sequences are not functional and that each sequence segment can have only one function . However , transposon-derived conventional miRNAs have been demonstrated in Arabidopsis [21] , and recent work of several groups show that repeat-associated miRNAs are quite common in mammals [22–26] . Borchert et al . point to 50 human miRNAs that are associated with Alu repeats and polymerase III transcription [22] . Piriyapongsa et al . link 55 experimentally characterized human miRNAs to different types of transposable elements [26] . Of these , 18 are conserved in other vertebrate genomes , and the authors predict an additional 85 novel transposable element–derived miRNAs . These observations , along with the evidence of very complex and widespread transcriptional patterns in eukaryotes , including nested transcripts and antisense transcription [27] , underlines the importance of enumerating all possible miRNA/target interactions in order to explore the full potential of miRNA-mediated regulation . In this paper , we develop and apply the miMatcher pipeline to perform intragenomic matching followed by classification of miRNA candidates using support vector machines ( SVMs ) . Using this method in the three plant genomes A . thaliana , O . sativa , and P . trichocarpa , we find species-specific miRNA-like hairpins ( miRNA candidates ) with almost perfect complementarity to mRNA targets . We present indications that many of these are active and hypothesize that the remainder forms a pool of regulators , which can easily be recruited by natural selection on the adapting organisms .
The computational procedure builds on our previously published method [18] that predicted potential miRNA genes in Arabidopsis , most of which are not conserved in Oryza . Three of these previous predictions ( all nonconserved ) have subsequently been confirmed as being expressed and correctly processed into small RNAs ( T . Dezulian , personal communication , unpublished data ) . The miMatcher procedure predicts miRNA candidates and their targets independently in each plant genome . First , we enumerate all intragenomic matches between any mRNA and any other part of the genome , where the genomic part of the match is able to bind complementarily to the mRNA part of the match ( Figure 1 ) . We call such a match a “micromatch . ” The assumption is that the genomic part can be a miRNA gene that targets the mRNA . Looking at micromatches between known Arabidopsis miRNAs and their targets , we have derived a set of rules that the match must fulfill: we start from the observation that targets can be found above noise without using phylogenetic conservation by requiring no more than two mismatches [19] . The match length is required to be between 20 and 25 nucleotides , and we add previously described filters for low complexity [5] and low-binding free energy [18] . Furthermore , for a genomic match to be a potential miRNA gene , it must be part of a sequence that can fold into a stem-loop precursor recognizable by the biosynthetic machinery that makes miRNAs . A necessary ( but not sufficient ) requirement for this is that the match ( potential mature miRNA ) must form base pairs in one direction only; i . e . , the mature miRNA forms base pairs with bases either upstream or downstream . Figure 2 ( step 1 ) shows the number of candidate matches that pass these prefilters for each organism ( see Materials and Methods for a detailed explanation of the filters ) . Not all stem-loop structures can work as dicer substrates . To distinguish those that do work from those that do not , we analyze a range of structural attributes for each candidate . Figure 3 illustrates the structural attributes that we investigated ( see Materials and Methods for details ) . Next , we build a classifier capable of selecting the stem-loops ( having at least one target ) most likely to be true miRNA genes based on the attributes summarized in Figure 3 . To this end , we construct a positive and a negative control set . While the positive controls are simply the known miRBase miRNAs for each plant ( regardless of whether we can find a target for them or not ) , the negative control set is less obvious to construct: we rely on the assumption that all miRNAs that regulate a known target as identified by [28] is already known . Accepting that this assumption is fairly reasonable means that that we can generate a negative control set by running the intragenomic matching ( including prefilters as above ) with the “known targets” as queries and then removing those genomic matches that overlap with already known miRBase miRNAs . Then , for each place in the genome matching a query mRNA , the flanks are extracted and the minimum free energy structure is calculated . The minimum free energy structure is analyzed , and structural features are calculated . For most of the measures , there is a clear separation between the positive and negative control sets ( Figure 3; red and blue traces , respectively ) , but there are still unnegligible overlaps . This shows that if we filter by hard threshold values on each attribute , we will either lose a large portion of the true positives or be forced to allow a large number of false positives to pass through the filters . Instead , we use an SVM [29] to classify based on all the attributes to achieve maximum separation . SVMs have successfully been used for animal miRNA precursor structure classification [30] , but not yet for plants . We train an SVM individually on each species , which is important because some of the input are values for the RNA folding and hybridization , which is strongly influenced by the GC composition of the genomes . Figure 4 shows separation of the miSVM score between positive and negative examples , and Table 1 lists performance estimates using cross-validation ( see Materials and Methods ) . In Arabidopsis , according to the cross-validation , when searching for miRNA candidates targeting a specific mRNA , 93 . 7% of all the positively classified candidates returned ( if any ) will be true positives . This specificity , however , comes at a price: 27% of the Arabidopsis miRBase miRNAs are erroneously classified as non-miRNAs . This remarkably specific identification of the known miRNA genes shows that intragenomic matching according to a strict set of targeting rules followed by classification on the basis of structural features of the precursor is sufficient for prediction of novel miRNA candidates . In the other species , the performance is comparable , albeit slightly less specific . In contrast to other methods , our method does not depend on conservation in other genomes , and is therefore able to predict species-specific miRNA candidates . A summary of the results of applying miMatcher followed by miSVM to three plant genomes is shown in Figure 2 . After classification , the positively classified micromatches are grouped into candidate loci on the basis of the genomic positions and families according to miRNA sequence similarity ( see Materials and Methods ) . We find 1 , 261 , 2 , 613 , and 2 , 148 candidate miRNA loci in Arabidopsis , Populus , and Oryza , respectively ( Datasets S1–S3 ) . The fact that these different genomes despite their genome sizes and structures ( i . e . , Oryza's peculiar repeat genome structure [31] ) have around the same number of candidate miRNAs with targets is striking and supportive of the method . When comparing the classification by miSVM with a recently suggested rule-based classification of Arabidopsis pre-miRNAs [32] , miSVM is much more stringent: the rules-based method accepts 100 out of 107 in our positive examples ( compared to 82 of 107 for miSVM ) , but it fails to reject 224 of 1 , 372 of the negative examples . A recent review [32] questioned some miRBase-registered miRNAs ( ath-MIR413 to 420 and ath-MIR426 ) found in a study relying on miRNA conservation between Arabidopsis and Oryza [6] . These miRNAs seem to lack conservation in organisms outside Arabidopsis and Oryza , and when tested , they gave weak hybridization signals on Northern blots . Moreover , they have less pairing in the miRNA precursor stem than many of the other miRBase miRNAs . Interestingly , these nine miRNAs are not among the predicted miRNAs coming through the miMatcher pipeline steps , and six of them ( ath-MIR413 and ath-MIR417 to 426 ) are among the 25 “false negatives” we get in the above miSVM evaluation . We use two methods to classify candidates as derived from repetitive regions: ( 1 ) RepeatMasker to find known repeats and transposable elements as well as simple low-complexity sequences; but since this relies on the quality of the available repeat libraries , we also ( 2 ) count the copy number of the mature candidate miRNA sequence in the whole genome , regarding candidates with high copy numbers ( >100 ) as repetitive ( see Materials and Methods ) . Following this classification , we find that although underrepresented , there is still a sizeable fraction of the known miRBase miRNA mapping to repeats ( 8%–16% ) . However , since most miRBase miRNAs are located outside repeat and coding regions , we investigated the effect of removing such candidates and found that it reduces the number of candidate miRNAs significantly ( Figure 2 ) . Only about one-fourth of the candidates remain in Arabidopsis and Populus , and in Oryza , the number is reduced to around 10% . While it might be argued that the risk of false positives in the repeat and coding regions is higher , it is striking that there is a very large potential for miRNAs in such regions , and we speculate that the lack of experimental evidence could in part be due to them being actively excluded in previous studies . Because candidates encoded in repetitive or protein-coding segments ( CDS ) of the genome could be qualitatively different from those derived from other regions , we have chosen to focus on the nonrepeat/non-CDS candidates in the following analyses . While conservation is not a requirement for our miRNA candidates , knowing whether a candidate has homologs in other species is useful and does strengthen the reliability of the prediction . To explore the conservation of the miRNA candidates , we compare the candidates predicted by the intragenomic matching in each genome . We consider a candidate to be conserved if there exists a candidate in one of other genomes following the typical miRNA precursor conservation patterns [19 , 33]: ( 1 ) the mature miRNA sequences should be highly similar and should reside on the same arm of the precursor; ( 2 ) the loop region connecting the miRNA and miRNA* should be less conserved than both the miRNA and miRNA* ( see Materials and Methods for details ) . All candidate loci are compared and aggregated into families . We observe that the conserved miRNAs ( including many miRBase miRNAs ) are often members of multilocus families , while 35% of our predicted putative miRNAs are singletons . These loci may be of more recent evolutionary origin , not having undergone as many duplications as the deeply conserved miRNAs . Given that a miRNA candidate is conserved between two species , we investigate whether the conservation extends to a more functional level , namely if the two candidates have orthologous targets . When two orthologous miRNAs have at least one instance of orthologous targets in the two organisms , we call this a “miSquare” ( Figure 5 ) . For the purpose of identifying miSquares , we use an expanded target list based on looser matching criteria as detailed in Materials and Methods and [19] . We note that ∼90% of the candidates with a homolog in another species also share at least one target ( putting them into the miSquare category ) , consistent with conservation of the regulatory function . Consistent with this , 60%–75% of the annotated miRBase miRNAs in each organism participates in at least one miSquare . As can be seen in Figure 6 , conserved miRNAs tend to have more targets than the nonconserved . This fact can be explained by the assertion that compensatory mutations between a miRNA and its target ( s ) are less likely to happen if the miRNA has many targets constraining its sequence . Studying precursor conservation ( miHomology ) between the three species after filtering out candidates overlapping repeat and coding sequence , we find 226 , 410 , and 171 species-specific miRNA candidate families in Arabidopsis , Populus , and Oryza , respectively ( Figure 7A ) . These families cover 272 , 528 , and 183 candidate miRNA loci in the three species . We find 16 miRNA families conserved in all three organisms . In Arabidopsis , all of these 16 conserved candidates are already annotated in miRBase , suggesting that most of the deeply conserved miRNAs are already found . In an evolutionary perspective , one would expect more miRNAs to be common between the two dicots ( Arabidopsis and Populus ) than between a dicot and the monocot ( Oryza ) . Our predictions are fully in agreement with this hypothesis: only a single family is conserved between pairs of Oryza and a dicot , while five families are conserved only between dicots . The picture is more ambiguous when we investigate all the miRNAs in miRBase and use the same family assignment criteria ( Figure 7B ) . Most conserved miRBase miRNA families ( 21 ) are conserved between all three species . Unexpectedly , a high number of miRBase families ( seven ) are only conserved between the dicot Arabidopsis and the monocot Oryza: miR413 , miR414 , miR417–420 , and miR426 ( ath-MIR416 was not part of the analysis , as no targets could be predicted for this miRNA ) . These are miRNAs that do not pass the miMatcher pipeline and whose validity , as mentioned earlier , has been questioned in a recent review [32] . There are only one to two miRBase miRNA families conserved between Populus and one of the other two species . This could be due to the fact that only few studies have looked at conservation in Populus , and no studies have looked at conservation only between Populus and Oryza . Among the predicted miRNA candidates , the conserved ones classified as miSquare miRNAs are most likely to be actively used and have a phenotypic impact . The majority of predictions in this category are identical or overlapping with the already known miRBase miRNAs , because similar criteria have been used before to identify new miRNAs [5 , 19] . We did a manual assessment of the potential novel miSquare candidates that do not overlap other miRBase miRNA precursors or known annotated coding regions or repeats . In Arabidopsis , the two candidates are the miRNA* sequences of MIR172 precursors . Interestingly , Wang et al . have found Northern blot expression evidence of the ath-MIR172b* sequence [6] . In Oryza and Populus , we find no new miSquare families , but three new members of known miRBase families ( oza-MIR399 , ptc-MIR166 , and ptc-MIR395; see Table 2 ) . Both miRBase miRNAs and our predictions are found in many different genomic contexts . Analyzing the genomic context of a miRNA can provide hints to its function . In contrast to animals ( with ∼40% of miRBase human miRNA loci in introns ) , the three plants studied here have the vast majority of the miRBase miRNAs in intergenic regions ( Figure 8 ) . Oryza has the highest fraction ( ∼8% ) of both miRBase miRNAs and predicted miRNA candidates derived from introns in sense direction . miRBase miRNAs contained in protein-coding genes are clearly underrepresented relative to the fraction of the total genome . The conserved and miSquare subsets of our predictions show a similar underrepresentation , whereas the rest of the candidates have a larger fraction overlapping already annotated genes , although still underrepresented in intron and CDS regions compared to the total CDS/intron fraction of the genomes . When on the same strand as another gene , the CDS- , untranslated region– , or intron-mapping candidates are interesting cases , since they could constitute parallel signals that are sent when the “host” is expressed . In contrast to “normal” sense–antisense pairs , supposedly forming dsRNA to trigger the RNAi machinery ( reviewed in [34] ) , miRNAs encoded on the antisense strand to a protein-coding gene suggest an alternative and easily evolvable way of regulating the sense transcript . While the first reports of miRNA targets in plants found that a large proportion of the targets were transcription factors ( TFs ) [28] , subsequent research has suggested that plant miRNA targets are more diverse although still enriched in TFs [6 , 17 , 18] . To test whether the targets for our miRNA candidates are enriched in TFs , we use the Arabidopsis TF database AtTFDB [35] . The enrichment is found as the fraction of predicted targets that are TFs divided by the fraction of all annotated genes that are TFs ( 5 . 8% ) . The results are shown in Figure 9 for different sets of miRNAs with and without repeat/CDS overlapping miRNAs: miRBase , miSVM , miHomology , and miSquare miRNAs . All sets show a high enrichment of TF targets ( miRBase , miHomology , and miSquare of almost identical magnitudes ) . When we filter out miRNAs that overlap repeat/CDS regions , the TF target enrichment rises notably for all sets , indicating a different functional profile of CDS/repeat-derived miRNA candidates . The enrichment tops for non-repeat/CDS miSquare miRNAs , with 40 . 8% ( 2 . 8-fold enrichment ) of the targets being TFs . This high TF target enrichment of conserved miRNAs suggests that miRNA interaction with the core gene regulatory machinery is an important evolutionary feature . Our candidates ( miSVM ) show a lesser but still considerable enrichment compared to miRBase and conserved miRNAs ( both with and without repeat/CDS-overlapping miRNAs ) . This implies that a larger proportion of the nonconserved miRNA candidates have targets outside the core gene regulatory machinery . These observations suggest that a notable fraction of our nonconserved miRNA candidates are functionally different than the conserved miRNA candidates and already known miRNAs . This can be interpreted in at least two ways . It could be that the fraction of estimated false positives has targets spread uniformly throughout the genome and thereby lower the total enrichment of TF targets in our candidate set . On the other hand , it makes biological sense that newly evolved ( or evolving ) miRNAs arise uniformly around the genome with targets uniformly spread on all mRNAs , and only the functionally important ones then being maintained through evolution . Recently , deep sequencing of small RNAs in Arabidopsis using the 454 technology has revealed novel nonconserved miRNAs [36 , 37] . In one study [36] , small RNAs ( 16–28 nt ) were sequenced from libraries made from whole seedlings , rosette leaves , whole flowers , and siliques , resulting in approximately 340 , 000 unique sequences with a perfect match to the genome . Applying very strict filters including a requirement for expression of both the mature miRNA and miRNA* , the authors identified 38 high-confidence novel nonconserved miRNAs among the sequences . The full database of genome-mapped small RNAs from this sequencing study covers 5% of the Arabidopsis genome . A total of 31% ( 104 ) of our 334 candidates overlap with an observed small RNA with 20–23nt . Comparing this overlap frequency to ( 1 ) 22mers randomly chosen from the genome ( 1 . 8% overlap with 454 reads ) , and ( 2 ) miRNA candidates found by intragenomic matching but removed with miSVM ( 4 . 2% overlap ) ( both sets filtered for CDS/repeat overlap ) , it can be seen that both the intragenomic matching and miSVM step improves the frequency of miRNA candidates expressed by small RNAs ( Figure 10 ) . Of the 104 miRNA candidates with read overlap , 74 are already in the new miRBase 9 . 1 ( comprising 184 miRNA precursors , including the findings from Rajagopalan et al . [36] and Fahlgren et al . [37] ) . This leaves us with a short list of 28 novel nonconserved miRNA precursor candidates where the predicted mature miRNA has been observed experimentally ( see Dataset S4 ) . By using intragenomic matching in a single genome followed by hairpin classification , this work demonstrates that miRNA candidates can be found via their targets with high specificity and reasonable sensitivity . Using this approach , we have found surprisingly large numbers of miRNA candidates in the three plants studied . While most of the miRBase miRNAs are conserved along with their targets in other plant species ( although some newly discovered are more species specific , e . g . , [38 , 39] ) , the majority of the candidates found by our approach seem to be specific for each genome . Many of our candidates have a different genomic origin than the known miRNAs: many are encoded in regions annotated as repeats or protein CDS ( both sense and antisense ) . Recently , it has been shown that repeat associated miRNAs are common in animals [22 , 26] . Similarily , in plants we find that a large fraction of the new miRNA candidates derive from repeat regions . This suggests an active role for repeats in the regulation of gene expression . Their functional profile also differs from already known miRNAs in the sense that there is less target overrepresentation among TFs . Recently , deep sequencing of small RNAs in Arabidopsis using the 454 technology has revealed many novel nonconserved miRNAs in Arabidopsis [36–38] . Of our 334 predicted Arabidopsis miRNA candidates outside repeat and protein annotation , we identify 28 novel candidates with experimental support from a small-RNA sequencing project ( see Dataset S4 ) . Together , these observations raise some important questions: how many of the candidates are actually functional ? Do these nonconversed miRNAs play a role in speciation ? Conversely , if they are not functional , we must ask why: does something prevent their transcription or maturation ? For example , in Arabidopsis , we know that many intergenic regions and regions antisense to annotated genes are transcribed [14] . If they are transcribed , what prevents a candidate miRNA from being functional ? We know that their structure looks like that of known miRNAs and that they match at least one target with maximum two mismatches—just like the experimentally confirmed miRNAs . What other unknown features of sequence and structure , if any , are required for a miRNA-like hairpin to be functional ? We hypothesize that the candidates that are not ( yet ) functional form a pool from which functional miRNAs can evolve in relatively few steps , thus facilitating adaptation towards new niches by improving the organisms' evolveability .
Known miRNAs . Sequences were downloaded from miRBase release 8 . 2 [40] . A total of one Populus ( ptc-MIR481a ) and eight Oryza miRBase ( osa-MIR444 , osa-MIR445b/c/e/f/g/h/i ) genes were discarded because their reported precursor sequences could not be mapped to the genome . This leaves us with 118 ( Arabidopsis ) , 212 ( Populus ) , and 174 ( Oryza ) genome-mapped miRNA genes . Requiring nonoverlapping genome loci and at least one predicted target , these numbers are further reduced to 117 ( Arabidopsis ) , 199 ( Populus ) , and 166 ( Oryza ) unique miRNA genes ( see miMatcher procedure and grouping into loci explained below ) . Arabidopsis thaliana genome and annotation TAIR assembly version 6 were downloaded from http://www . arabidopsis . org . We only use RefSeq protein-coding mRNAs as possible miRNA targets . Populus trichocarpa genome assembly and annotation used was kindly provided by Eric Bonnét and is available upon request . The official release of the genome is now available at http://genome . jgi-psf . org/Poptr1_1 . Oryza sativa . TIGR assembly version 4 . 0 and annotation was downloaded from ftp://ftp . tigr . org . This is an improved version of the procedure described in [18] . Finding initial micromatches . For each annotated spliced mRNA , we search the genome for matches of length at least 20 with a maximum of two mismatches ( no gaps or wobbles allowed ) using the suffix array–based program vmatch ( http://www . vmatch . de ) . This is an exhaustive search guaranteed to find all matches . Prefiltering the intragenomic matches . The initial micromatches are filtered by discarding all matches not fulfilling the following criteria . Attributes of the putative mature sequence . Shannon index entropy of the genomic part of the match ( putative mature miRNA sequence ) must be larger than 1 . 7 bits . In addition , the following must hold: ( 1 ) all four bases had to be present at least once; and ( 2 ) at most , 11 of the three most frequent dinucleotides in the sequence were allowed . Length of the genomic part of the match must be 20–25 nt ( both inclusive ) . Attributes of the intragenomic match . Using the program RNAcofold ( Vienna RNA package [41] ) , the free energy change when a miRNA candidate binds to a target site was calculated . The free energy of binding per base must be less than −1 . 4 kcal/mol . Attributes of the precursor structure . In order to predict a possible precursor molecule , two genomic sequences around each micromatch are extracted: one starting 10 bases 5′ of the micromatch and extending 240 bases 3′ of the micromatch , and one with the extension lengths reversed . Each of these is treated independently in the following analysis . First , the potential precursor sequence is folded with RNAfold [41] to find the minimum free energy structure . The complementary part of the miRNA in this stem is denoted miRNA* , and is found as the sequence of nucleotides delimited by the pairing partners of the most 3′ and 5′ bases in the mature sequence . We define the attribute pretty stem to be true if all base pairs involving the mature microRNA and miRNA* are pairing to bases in the same direction opposite to each other . Trimming the precursor . Since all pre-miRNA are not of the same length , we trim down the initially found constant length pre-miRNA structure . We count how far inward toward the loop or outward toward the ends of the RNA sequence the stem extends using the following algorithm: moving out from the terminal base pair between the miRNA and miRNA* , a score of 1 is assigned for each base pair encountered and a score of −1 for each unpaired base . The extension is stopped when the current score is less than 5 lower than the maximum score so far . The last base pair is considered the terminus of the trimmed precursor . Given the predicted minimum free energy secondary structure of the putative miRNA precursor , we calculate the following attributes: pairs to mature miRNA—the number of paired bases in part of the precursor predicted to become the mature miRNA; outer and inner extension—found during the trimming procedure described above; distance between miRNA and miRNA*—the number of nucleotides between the bases participating in the innermost base pair of the mature miRNA; stability of precursor: this is simply calculated by using RNAfold on the trimmed precursor and dividing by the number of bases . This is based on the observation that miRNA precursors are unusually stable [42]; asymmetrically unpaired bases in stem—we count unpaired bases in either the miRNA or miRNA* where there are no corresponding unpaired base on the other side; and 5′ and 3′ stem hybridization—the energy gain calculated by RNAcofold ( Vienna RNA package ) from hybridizing the ten first or last bases of the mature miRNA to miRNA* . It should be noted that the structural attributes are not necessarily strictly independent from each other ( e . g . , a long “inner extension” correlates with the “distance between the miRNA and miRNA*” ) . We used SVM software implemented in the SVMlight package ( downloadable from http://svmlight . joachims . org ) using a radial kernel and double penalization of errors on the ( smaller ) set of positive examples . The input to the SVM is the structural features detailed above . Cross-validation . To avoid overtraining and to get a realistic evaluation of the ability of the SVM to generalize , it is important to reduce redundancy between training and test sets . Because precursors in the same family often have similar structures , we performed “leave-one-family-out” cross-validation to assess generalization across families . The positive examples ( miRBase miRNAs ) were divided into families according to homology ( we used the families provided by miRBase ) . For each family , a training set was constructed from the remaining positive examples , and all but 100 of the negative examples were chosen by random . The SVM was trained on this training set and subsequently tested on the withheld family and negative examples . The final SVM was retrained on the entire dataset and is called miSVM . Given the location ( coordinates and strand ) of the mature part of a miRNA precursor , we assign miRNA candidates into genomic loci by grouping precursors with up to 4 nt overlap of the mature sequence together . In Populus , the 212 miRBase 8 . 2 genome-mapped genes correspond to 200 unique genomic loci; in Oryza , the 174 miRBase genes are reduced to 167 loci . All 118 Arabidopsis miRNAs are correctly mapped to unique loci . Gene models provided by the genome sequencing and annotation groups were downloaded ( see above for sources ) , parsed , and read into database tables indexed by the absolute genomic coordinates . RepeatMasker ( http://www . repeatmasker . org ) was run to identify repeats whose locations were also stored in the database . In addition , we consider a candidate a repeat if it has a copy number ( number of exact genome matches with length 20 allowing two mismatches or indels—corresponding to our miRNA family definition ) greater than 100 . In Arabidopsis , this copy number constraint annotates three miRBase miRNAs ( ath-MIR415 , ath-MIR401 , and ath-MIR414 ) as repeats , two of which were already assigned as repeats by RepeatMasker . Similarly for Oryza , 16 miRBase miRNAs are annotated as repeats ( 15 were already assigned by RepeatMasker ) , and for Populus , 20 miRBase miRNAs are annotated as repeats ( 17 were already assigned by RepeatMasker ) . All candidates where checked against this database to locate overlaps with annotation . When we consider the nonrepeat/CDS overlapping miRNAs , we remove miRNAs overlapping repeat or CDS regions ( regardless of strand ) . All candidate miRNAs were grouped into families on the basis of mature sequence similarity: two candidates were grouped together if they shared at least 20 nucleotides allowing two mismatches or indels . Larger family clusters were constructed using single linkage clustering . In addition , it is required that all members of a family must have the mature miRNA on the same arm of the precursor . These criteria gave us near-perfect recovery of the miRBase-assigned families ( miRBase version 8 . 2 ) . In Arabidopsis , only the miR171 family is divided in two families , and the following miRBase families are pairwise grouped together: MIR319–MIR159 , MIR156–MIR157 , MIR165–MIR166 , and MIR170–MIR171 . To determine if two miRNA precursors from different species are homologous , we require fulfillment of two criteria: ( 1 ) the mature miRNAs must align over a region of minimum 20 bases with a maximum of two mismatches ( gaps count as mismatches ) , and be on the same arm of the precursor; and ( 2 ) no 20mer in the loop region ( connecting the miRNA and miRNA* ) may align better than the miRNA or miRNA* region . We explored the effect of these criteria on a few expected positive and negative miRNA test cases . As a positive case , we classify the three miR172a miRNAs from Arabidopsis , Oryza , and Populus as homologs ( the same is true for miR156a—no other similar cases were explored ) . Testing the Arabidopsis ath-MIR169 family ( 14 members ) , approximately two-thirds could be grouped as homologs: this is as expected , as precursors originating from recent duplications have highly similar loop regions [33] . As a negative test case , we took 21 Arabidopsis “a” precursors ( ath-MIR156a , ath-MIR157a , etc . ) and found only two homologous pairs based on our test: ath-MIR156a–157a and ath-MIR165a–166a . These two pairs are often considered to be from the same miRNA families . We consider two miRNA families from different species as conserved if there exists a precursor in each family with homology ( miHomology ) to a precursor in the other family . Because miRNA families are computationally determined in a genome-dependent manner ( relying on single linkage clustering ) , there can be a minor asymmetry in miRNA family conservation: looking from Arabidopsis , there can be X families conserved in Populus , while looking from Populus , there can be Y families conserved in Arabidopsis . In this paper , we report the larger of these two numbers as the family conservation count . To identify conserved regulatory interactions between a miRNA and target in different species—miSquares—we have two tasks: ( 1 ) determine protein orthology between the species , and ( 2 ) determine the targets of the conserved miRNAs . Protein orthology in the three organisms was determined using the INPARANOID program [43] . The program uses bidirectional best BLAST hits to determine orthologs between two species . In addition , it BLASTs each proteome against itself to determine “inparalogs”—presumed gene duplications after speciation . The program was run using Caenorhabditis elegans ( wormpep157 from Wormbase ) as outgroup , and otherwise default parameters . The intragenomic matching procedure simultaneously finds miRNAs and corresponding targets with up to two mismatches ( no wobbles or gaps allowed ) . According to Jones-Rhoades and Bartel [19] , we can find targets above noise with a weaker matching criterion if we add target homology as a constraint . With the exception that we count wobbles as mismatches , we use the same matching and scoring rules as presented in this paper . Given a miRNA , we find target sequences that align over 20 nucleotides with a score ≤3 according to the scoring scheme: mismatch scores as 1 , gap ( open and extension ) scores as 2 . The original article argues for a cutoff score of 3 . 5 because they score wobbles less restrictively ( score . 5 ) . In other words , our scoring scheme allows for targets with up to three mismatches or a combination of one gap and one mismatch . Based on these target requirements , we cannot find any targets for three miRBase 8 . 2 miRNA genes: ath-MIR416 , ptc-MIR482 , and osa-MIR438 . It should be noted that the miSquare criterion does not require the miRNAs in the two species to target homologous regions in the orthologous target mRNAs . We note , however , that in reality , this is most often the case . We used the full database of sequenced genome-mapped small RNAs from the supplementary data of [36] . Our miRNA candidates were analyzed for overlap with these sequenced small RNAs by requiring a 20–23 nt coordinate overlap with the mature sequence of a candidate . | microRNAs ( miRNAs ) are small RNA molecules that regulate gene expression by complementary basepairing to mRNAs . In plants , this base-pairing is almost perfect along the whole length of miRNAs . This long stretch of complementarity makes it relatively easy to make computational predictions of the targets for known miRNAs . To predict novel miRNA genes , we take advantage of this and reverse the target prediction: instead of predicting targets for known miRNAs , we predict novel miRNA candidates for all known mRNAs . Because matching between target and miRNA candidates is integral to the method , it is possible to achieve good predictions without having to rely on evolutionary conservation , as most other current methods do . This means that we can predict new miRNAs that are specific to an organism . Interestingly , this could help explain the difference between species that have very similar protein-coding genes , but highly different phenotypes . Furthermore , it turns out that many of these new miRNA candidates derive from genomic repeat regions such as transposons , which points to a possible active role for repeats/transposons in the regulation of gene expression . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"populus",
"oryza",
"plant",
"biology",
"arabidopsis",
"computational",
"biology"
]
| 2007 | Intragenomic Matching Reveals a Huge Potential for miRNA-Mediated Regulation in Plants |
Protein interactions play a vital part in the function of a cell . As experimental techniques for detection and validation of protein interactions are time consuming , there is a need for computational methods for this task . Protein interactions appear to form a network with a relatively high degree of local clustering . In this paper we exploit this clustering by suggesting a score based on triplets of observed protein interactions . The score utilises both protein characteristics and network properties . Our score based on triplets is shown to complement existing techniques for predicting protein interactions , outperforming them on data sets which display a high degree of clustering . The predicted interactions score highly against test measures for accuracy . Compared to a similar score derived from pairwise interactions only , the triplet score displays higher sensitivity and specificity . By looking at specific examples , we show how an experimental set of interactions can be enriched and validated . As part of this work we also examine the effect of different prior databases upon the accuracy of prediction and find that the interactions from the same kingdom give better results than from across kingdoms , suggesting that there may be fundamental differences between the networks . These results all emphasize that network structure is important and helps in the accurate prediction of protein interactions . The protein interaction data set and the program used in our analysis , and a list of predictions and validations , are available at http://www . stats . ox . ac . uk/bioinfo/resources/PredictingInteractions .
For understanding the complex activities within an organism , a complete and error-free network of protein interactions which occur in the organism would be a significant step forward . Experimentally , protein interactions can be detected by a number of techniques , and the data is publicly available from several databases such as DIP , Database of Interacting Proteins [1] , and MIPS , Munich Information Center for Protein Sequences [2] . Unfortunately , these experimentally detected interactions show high false negative [3] and high false positive rates [4] , [5] . In this paper we develop a new computational approach to predict interactions and validate experimental data . Computational methods have already been developed for these purposes . For interaction validation , these have mainly centered on the use of expression data [5] , [6] or the co-functionality or co-localisation of the proteins involved [7] , [8] . For prediction of protein interactions in contrast , many methods have been suggested . The majority of these generate lists of proteins with a functional relationship rather than physical interactions [9] , [10] . In terms of physical interaction prediction the available methods can be typified by the two approaches of Deng et al . [11] and Jonsson et al . [12] . In Deng et al . 's method , a domain interaction based approach , a protein interaction is inferred on the basis of domain contacts . If a domain pair is frequently found in observed protein interactions , it is likely that other protein pairs containing this domain pair might also interact . From the observed protein interaction network , the probabilities of domain-domain interactions are estimated . The expectation-maximum algorithm is employed to compute maximum likelihood estimates , assuming that protein interactions occur independently of each other . This likelihood is then used to construct a probability score for a protein pair to interact , it is inferred based on the estimated probabilities of domain interactions within the protein pair . Deng et al . 's prediction is based on a total of 5 , 719 interactions from S . cerevisiae . However , the limited number of known domains may well not be enough to describe the variety of protein interactions . This approach has had further extensions , such as an improved scoring for domain interactions [13] and the inclusion of other biological information [14] . Liu et al . 's model [15] is an extension of Deng et al . 's method which integrates multiple organisms . In addition to S . cerevisiae , two other organisms , C . elegans , D . melanogaster , are included . The second type of approach , as used by Jonsson et al . [12] , is homology-based . It searches for interlogs among protein interactions from other organisms . If an interlog of a protein interaction exists in many other organisms , this protein interaction will score highly . In addition to searching for orthologous interlogs , Mika and Saeed [16] , [17] suggest that paralogous interlogs may provide even more information for inferring interacting protein pairs . In principle , statistical clustering algorithms such as [18] and [19] which identify cliques in the network could be viewed as a prediction method , predicting that all proteins within a clique interact with each other . This interpretation is biologically questionable , and as the focus in the statistical clustering approach is on locating cliques and overlapping modules rather than on predicting individual interactions , we exclude it from our comparisons . Neither Deng et al . 's method nor Jonsson et al . 's method make use of network structure beyond pairwise interactions; interactions are considered as isolated pairs . However these pairs could and should be considered as a network , where the proteins are nodes and their interactions are links [20] , [21] . Topological examination of these networks has revealed many interesting properties , including a clustering tendency [22] , [23] , see also Supporting Information ( Text S1 , Table S1 ) . In our method we exploit the network structure by developing a score which considers triadic patterns of interactions rather than pairs . In this paper we thus take the established idea that the characteristics of a protein ( i . e . , its structure , function and location ) will affect its interactions ( see for example [7] , [21] , [24]–[31] ) alongside the not yet fully explored idea that its network position will also affect its interactions , in order to develop a novel predictive tool . Our goal is to predict ( undirected ) protein interactions of the type x with y , where both x and y interact with a third protein z . Therefore in our approach we particularly focus on two simple three node network structures , triangles and lines . A triangle is a subnet formed by an interacting protein pair with a common neighbour . A line , by contrast , is a subnet formed by an non-interacting protein pair with a common neighbour . We will show that these network structures and the protein characteristics within them help to predict protein interactions . We apply our method to the S . cerevisiae interaction network from the DIP database . During the validation we assume that function and structure are known for all proteins ( fully annotated ) and that the protein interaction network is known for all but one interaction . With triadic interacting patterns , we predict the interaction status of those protein pairs with at least one common neighbour and compare our results with those from three other published scores . We go on to demonstrate that the requirement to have fully annotated proteins can be relaxed to include partially annotated proteins , with a slight drop in the accuracy . The prediction is also compared with simulated networks where all proteins are shuffled while the network structure is maintained , in order to examine whether the specific network structure , triangles and lines , keep useful information in forming protein interaction networks . To measure the true positive rate in a set of protein pairs , Deane et al [5] proposed the expression profile index ( EPR ) , a measure of the true positive rate in a set of protein pairs based on biological relevance . We compare the EPR index to our score , showing that , with a suitable cut-off , our predictions achieve a high true positive rate . We also give examples of validated experimental data and predict new interactions . Our predictive model uses a prior interaction database and for this we use three prior databases , pooling protein interactions collected from prokaryotes , eukaryotes and all interactions . The results from using different prior databases show that the use of interactions from within the same kingdom rather than across kingdoms significantly improves the results , indicating as in [21] that interaction networks may be significantly different between the kingdoms . Comparing our method to three other standard approaches , namely the domain-based approach by Deng et al . and an extension by Liu et al . , and a homology-based approach by Jonsson et al . , we find that our method outperforms the above approaches on the subset of interactions in the DIP Yeast data set which contains enough annotation and connectivity to be included in our analysis . Our method complements the methods by Deng et al . and Liu et al . , as their approaches apply to a rather different subset of potential interactions yielded from the DIP Yeast data set .
Experimental protein interactions of S . cerevisiae , excluding self-interactions , are obtained from DIP ( DIP Yeast ) . Self-interactions ( <3% of all interactions ) are excluded , implying that all triangles and lines are constructed of three different proteins . Three different prior data bases are constructed by pooling interactions considering eukaryotes ( D . melanogaster , C . elegans , S . cerevisiae , M . musculus , H . sapiens ) , prokaryotes ( E . coli and H . pylori ) , or all interactions; the interaction we would like to predict or to validate is always excluded . The proteins in our dataset are classified into the seven SCOP classes [32] using the SUPERFAMILY database [33] , see Supporting Information ( Text S1 , Table S3 ) . Between 61 to 89% of proteins are classified , dependent on organism . In our analysis , a protein is found to be assigned to 1 . 3 classes on average . We use the 24 functional groups from the secondary level of Molecular Function in the Gene Ontology [34] , see Supporting Information ( Text S1 , Table S4 ) as our protein functional categorisation . Molecular Function ontology in GO has 188 secondary level categories , excluding the categories “obsolete” and “unknown” . The 24 groups used are those that are most frequently observed . An annotated protein may be assigned to several nodes in GO , which can be traced back to one or multiple nodes . The protein interaction network is used to build an upcast set of triplets of characteristic vectors as in Figure 1; see also [21] . Here , A , B , C and D denote protein characteristics , whereas different shapes indicate different proteins . A protein may possess more than one characteristic . Our triplets are triangles and lines of three characteristic vectors according to their interacting patterns . A characteristic line is a specific pattern constructed by three vectors with two vector interactions among them . A characteristic triangle is formed by three vectors interacting with each other . Here we abuse the English language; while it would be clearer to say “pair of characteristics” and “triangle of characteristics” we prefer the shorter version “characteristic pair” and “characteristic triangle” for easier reading . To assess our method we also compare it with a score based on characteristic pairs only . In a similar manner to the upcast set of characteristic triplets , we construct an upcast set of characteristic pairs . Here we grasp the opportunity to introduce some notation . For a protein x , its characteristic vector Sc ( x ) contains all its characteristics of a certain type ( e . g . , structure , function ) , and S ( x ) denotes the set of vectors formed using different characteristics . In the case of two protein characteristics , S1 ( x ) and S2 ( x ) are the two respective vectors , and S ( x ) is the set We shall denote the set of all characteristic vectors for all proteins by S; this set may contain a vector va multiple times . A characteristic pair is constructed by two characteristic vectors from two interacting proteins . If two proteins x and y interact , for each pair {νa , νb} with νa ∈ S ( x ) , νb ∈ S ( y ) , we write νa∼νb . If two protein do not interact , the relation between two vectors is denoted by υa≁υb . The upcast set of characteristic pairs is then the collection of all characteristic pairs extracted from the protein interaction network , which may stem from one or from multiple organisms . For our upcast sets to be informative for a protein interaction , an eligible protein pair has to satisfy two conditions: Firstly , the proteins need to have at least one common interacting neighbour; and secondly , the query protein pair and the neighbours have to be at least partially annotated . Among 4 , 931 proteins in the observed interaction network , 2 , 416 ( 49% ) proteins are fully annotated with both characteristics ( structure and function ) and 3 , 808 ( 77% ) are annotated with at least one characteristic . Table 1 gives the number of eligible protein pairs in the Yeast protein interaction network . There are about 90 , 000 eligible fully annotated proteins pairs and around 3% of them are in the experimental data ( DIP Yeast ) . When partially annotated proteins are included , the number of eligible protein pairs is increased by 158% . We derive our triangle rate score from the upcast sets of characteristic triplets . This score thus includes information not only from the query protein pair but also from its neighbours . Therefore , it is a network-based score which goes beyond pairwise interactions . Within the triplet interactions , we assess the odds to observe triangles versus lines around the query protein pair . More formally , let txy be the total frequency of all characteristic triangles around the query protein pair {x , y}; denoting by z ∈ B ( x , y ) the set of all common neighbours of x and y in the protein interaction network , Where f ( νa∼νc∼νb∼νa ) is the frequency of triangle {νa∼νc∼νb∼νa} among all characteristic triangles in the prior data base . Similarly , lxy is the total frequency of all characteristic lines around the query protein pair {x , y} . We define the triangle rate score , tri ( x , y ) for the protein pair {x , y} . as the odds of observing triangles versus lines among triangles and lines in its neighbourhood , ( 1 ) Heuristically , the higher the triangle rate score is , the higher the chance one would observe an interaction between the query protein pair . When multiple characteristics are simultaneously included , the triangle rate score defined above requires the query protein pair and the common neighbour to be fully annotated with multiple characteristics . However , there are many partially annotated proteins in the neighbourhood which may provide useful information . These proteins are particularly important when only a few fully annotated ones are available . In Supporting Information ( Text S1 , C ) , an extended version of the triangle rate score is provided to include partially annotated proteins . To assess whether the triangle rate score significantly improves prediction and validation , we also construct a similar score based on pairwise interactions only , which we call the pair-based score . The details are as follows . Based on the pairwise interactions , we also provide an odds ratio-based score , see also [23] for details , which gives a measure of the relative count of the characteristic pair found between positive and negative protein interactions . We call an interaction “positive” , if it is contained in the database . All potential interactions which are not found in the database are called “negative” . This score can be viewed as a likelihood for a model which assumes that Given a specific characteristic pair {νa , νb} , under the multinomial-binomial model above the maximum likelihood estimate for πab is given bywhere oab is the number of times an interaction has been observed for the characteristic pair {νa , νb} , and nab is the number of times that no interaction was observed for the pair {νa , νb} . With this heuristic we define the pair-based score for a query protein pair {x , y} as ( 2 ) Thus pair ( x , y ) is the average of the estimated probabilities πˆab for all characteristic pairs generated by the query protein pair in the prior data base . Heuristically , the higher the score , the more likely it should be that the two query proteins interact . An extended version of the score is able to cover protein pairs which are only partially annotated , see Supporting Information ( Text S1 , C ) . We note that the triangle rate score and the pair-based score have a slightly different form . While the pair score is the average of all relative frequencies of characteristic pairs , the triangle rate score is the summed frequency of the characteristic triangles over triangles and lines . The different setting here was chosen because around a query protein pair many characteristic triangles might hardly be seen in the observed networks; their counts are too small to be useful . This phenomenon is much less pronounced for the pair patterns , there being rather more triangle patterns than pair patterns; see Supporting Information ( Text S1 , Table S2 ) for the number of observed patterns against all possible patterns . In order to put our scores to work we choose a threshold; all pairs with scores above that threshold would be classified as interacting , while all pairs below that threshold would be classified as non-interacting . The choice of threshold depends on the desired sensitivity and specificity; recall that the sensitivity is the ratio of true positives over ( true positives+false negatives ) and the specificity is the ratio of true negatives over ( true negatives+false positives ) . To assess our scores we first use a Receiver Operating Characteristic ( ROC ) curve , which is a useful technique for examining the performance of a classifier [35]; in our case the classes are “interacting” or “non-interacting” for a pair of proteins . The curve plots sensitivity against ( 1 minus specificity ) . Each point on a ROC curve is generated by selecting a score threshold for a method . We move the cutoff along the range of the score and record different sensitivities and specificities of a method . The closer the curve is to the upper left hand corner ( i . e . , the larger the area under curve ) , indicating that sensitivity and specificity are both high , the better the predictive score . When evaluating performance for a classifier when the test data is unbalanced , such as when there is a disproportionate number of negative versus positive cases , instead of choosing subsamples of the same size as for our tests between two ROC curves , the Precision-Recall Operating Characteristic ( P-ROC ) curve provides an alternative . The precision is the ratio of true positives over ( true positives+false positives ) , whereas the recall is the ratio of true positives over ( true positives+false negatives ) , i . e . the sensitivity . The P-ROC curve plots recall against precision . While there is a tendency for recall and precision to be inversely related , Precision-Recall curves are not necessarily decreasing . An increasing P-ROC curve is an indication for perverse retrieval , in which there is a strong tendency that first the negative interactions are retrieved; only when there are so few of those left that it is almost unavoidable to retrieve positive interactions , these are also covered; see for example [37] for an exposition .
We compare our triangle rate score with three other methods , the two by Deng et al . [11] and Liu et al . [15] being domain-based , and the one Jonsson et al . [12] being homology-based . The two scores by Deng et al . and Liu et al . are downloaded directly from the authors' webpages . Deng et al . 's method predicted 125 , 435 protein pairs . After removal of 5 , 717 interactions , which are the training data in forming the scores , and translating the gene names to ORF names ( to match the reference sets ) , 63 , 013 protein pairs remained . Liu et al . 's method predicted 20 , 088 protein pairs . After the translation of names , 15 , 608 protein pairs remained . Our triangle rate score predicts 87 , 181 protein pairs . The number of predicted pairs using the different methods on the DIP 20060402 data set described above , and the overlap with our pairs , given in Table 2 , illustrates that our method and Deng et al . and Liu et al . 's methods complement each other , as they operate on fairly disjoint sets . In contrast , there is a substantial overlap between the eligible pairs for Jonsson et al . 's score and for the triangle score . Jonsson et al . 's method is implemented in two ways , using orthologs only ( a pooled database of 6 organisms , E . coli , H . pylori , C . elegans , D . melanogaster , M . musculus and H . sapiens from DIP for the search of similar sequences ) , and additionally using orthologs and paralogs ( see Figure 2 and Table 3 ) . In the second case the S . cerevisiae interactions in DIP are also included . The comparison of scores are shown in Figure 2 . The areas under the ROC curve were tested for significant difference; see Table 3 . The results of the z-tests show that our triangle rate score outperforms both the domain-based ( second place ) and homology-based scores , see Table 4 for p-values . Here the comparison with the domain-based methods has to be taken with a pinch of salt , as the amount of overlap between the eligible pairs for those methods and our method is very small . The P-ROC curve in Figure 3 for the comparison between the different methods shows not only that the triangle rate score outperforms the other methods on our data set , but it also reveals that Deng et al . 's score and Liu et al . 's score have marked jumps in recall . The overlap with our data set is so small that these jumps may be artefacts . The number of predictions which overlap with the MIPS-GSP ( 8 , 250 interactions ) is also an indicator of coverage . Our triangle rate score is able to predict 928 of them , which is the largest number of predictions from any of the four sets . Deng et al . and Liu et al . 's scores , based on protein-domain relationships , can only predict 85 and 174 interactions in GSP respectively . Their methods cannot predict protein pairs without domain information , limiting their coverage . Liu et al . 's score , when including information from other organisms , improves the coverage over Deng et al . 's score , but not the overall performance in terms of AUC . Jonsson et al . 's score covers more interactions in GSP ( 390 interactions ) than the domain interaction based approaches , however , it appears to perform worse in terms of AUC , though not significantly . Jonsson et al . 's method is still limited in coverage , however , because only sequences with very high similarity are useful for transferring interactions , and often qualified homologs are not available , see [16] . We also compare our triangle rate score to the pair-based score , thus allowing us to ascertain the effect of network structure on our scoring method . The ROC curves in Figure 4 show that the triangle rate score outperforms its pair-based analog , thus demonstrating that the inclusion of network information beyond pairwise interactions significantly improves prediction . The success of the triangle rate score indicates the importance of network structure ( triangles and lines ) in conjunction with protein characteristics for the understanding of protein interactions . We have also employed a logistic regression model to include pair- and triplet-based statistics , see Supporting Information ( Text S1 , E ) for details . As the preliminary investigation did not show significant improvement over the simple triangle rate score and the full scale leave-one-out validation would be very computation-expensive we did not pursue this model further . The triangle rate score can be used to validate experimentally derived interactions . It is estimated that the false positive rates for high-throughput experiments vary from 35 to 83% dependent on source [3] . At a cut-off score value of 0 . 09 , our prediction reaches 0 . 83 for both sensitivity and the specificity . Of the 2 , 896 DIP Yeast interactions tested by the triangle rate score , 1 , 732 ( 60% ) are validated at the score cut off of 0 . 09 . This gives an estimated false positive rate of around 40% , close to that given by EPR [5] . We also calculate the EPR index ( % correct ) for subsets of our predictions . Figure 5 shows how the EPR index increases with higher ranked prediction sets . As our score cut-off is increased , the EPR index indicates that the quality of our predictions is increasing . The set of the top 14% predictions ( ∼12 , 200 interactions ) shows a higher EPR than the experimentally derived interactions in DIP Yeast . The EPR index estimates the biologically relevant fraction of protein interactions detected in a high throughput screen . As the EPR index is between 70–80% for DIP CORE , we cannot hope for a correct prediction rate ( fraction of true predictions over true positives ) higher than 70–80% . Indeed this upper limit is reflected by a sharp drop-off in the ROC curve ( Figure 2 ) for ( 1- specificity ) between 0 . 2 and 0 . 3 , i . e . specificity between 0 . 7 and 0 . 8 . A second way to assess the accuracy of our predicted set is to consider the overlap between our positive predictions and DIP CORE . DIP CORE includes 5 , 969 high-confidence interactions determined by one or more small scale experiments . As shown in Figure 6 , the percentage of overlap increases with increasing score cut-off values . Both these tests demonstrate that the triangle rate score is a good indicator of interaction prediction quality . The triangle rate score can be extended to gather information from partially annotated proteins; see Supporting Information ( Text S1 , C ) . The inclusion of partially annotated proteins allows more protein pairs to be predicted and more neighbours to be included . Here we compare the prediction using only fully annotated proteins and all ( fully and partially annotated ) proteins . The accuracy is the fraction of correct prediction out of all predictions against each of the 300 reference sets . Again , the 300 reference sets are employed to avoid the bias raised from too many negative pairs , i . e . a high accuracy may arise simply from making no positive prediction . Figure 7 shows the accuracy and the coverage using fully or partially annotated proteins . The inclusion of partially annotated proteins considerably improves the coverage by 158% with an accuracy of 77% ( only a drop of 5% from using fully annotated proteins ) . To explore how different priors affect the prediction , we group protein interactions into prokaryotes , including E . coli and H . pylori , and eukaryotes , including C . elegans , S . cerevisiae , D . melanogaster , M . musculus and H . sapiens , and a final global pooled dataset including all interactions . As a random background , we also generate a simulated interaction network by shuffling the annotation of proteins in the Yeast protein interaction network . Based on the five prior data bases - Yeast , eukaryotes , prokaryotes , all interactions , and a shuffled protein network , we predict protein interactions using the triangle rate score . The AUC for all curves are calculated and tested for differences , see Table 5 and Text S1 and Table S5 ) . The ROC curves show that the prior from Yeast itself gives the best prediction , followed by that from eukaryotes before third , all interactions; see Figure 8 . The prior from prokaryotes gives almost no useful information , suggesting a fundamental difference of protein interaction networks between the two kingdoms . The difference between Yeast and eukaryotes probably arises because Yeast already has a large amount of interaction data , so that the inclusion of data from other similar organisms does not improve prediction . A less well studied organism however may benefit from a larger prior constructed from other close organisms . It is also not a surprise that the prior from eukaryotes performs slightly better , though not significantly , than the prior from all interactions , as the interactions from eukaryotes form the majority of interactions in the pool . The clearly different ROC curves from the eukaryotes prior and the prokaryotes prior suggest that their networks are very different , in terms of the interaction patterns of protein characteristics . We perform a χ2 test of homogeneity for triangles and lines in the two prior data bases . We compare characteristic triangles and lines that are annotated with structure , function and both , and group patterns with counts of at least 5 . All 6 tests suggest a significant difference between eukaryotes and prokaryotes . This difference might arise from evolution and suggests that only priors from close organisms ( within same kingdom ) are helpful . It is not always beneficial to construct a large data base without taking the difference among organisms into account . The ROC curves for predicting interactions from shuffled protein network are close to diagonal , as is expected . Without the information from protein structure and function and the interacting patterns , the prediction is random . The different trends between using real data and simulated data show that the interacting patterns of protein structure and function play important roles in protein interactions . The P-ROC curve in Figure 9 shows a similar pattern in performance for the priors Yeast , eukaryotes , and all interactions , but it also reveals that taking prokaryotes as prior is worse than random shuffling . The figure shows that prokaryotes as prior could lead to perverse retrieval . The different performance of prokaryotic and eukaryotic priors relates to their networks being rather different with respect to their distributions of protein structure and also of protein function . The most striking difference relates to small proteins . While 15% of eukaryote proteins are small proteins , less than 1% of prokaryote proteins are small proteins . Among the 10 most frequently observed structure category interactions , in eukaryote 3 of them ( 23% of all category interactions ) involve small proteins , while in the list of top 10 structure category interactions in prokaryotes small protein related interactions do not appear . Another considerable difference concerns the distributions of the two functions “RNA polymerase II transcription factor activity” and “GTPase regulator activity” . While 4% of the eukaryotic proteins possess one of these two functions , they are not found in the prokaryotic proteins . In addition , in the list of top 10 most frequently observed function category interactions , in eukaryotic networks we observe many function category interactions with “protein binding” proteins , while they do not appear on the list of prokaryotes networks . With the triangle rate score we provide a novel statistical tool for prediction and validation of protein interactions . Our method uses triadic-level statistics , in addition to the traditional dyadic-level statistics arising pairwise interactions . This network-based method is shown to complement the existing domain-based approach , and to outperform the homology-based methods as well as a comparable pair-based method . As our method requires annotated proteins occurring interacting with at least two other proteins , currently the only data set which is large enough to warrant application is that of Yeast , see G in Text S1 and also see Table S6; we anticipate that once more data will become available for many other organisms , our method will be useful in these organisms also . Combining our method with priors from other organisms allows us to compare protein interaction behaviour among kingdoms , from the viewpoint of comparative interactomics . The significant difference in protein interactions networks between eukaryotes and prokaryotes serves not only as a caution to integrate interaction information from only close organisms , but also as encouragement for further , micro-level study between the two upcast sets , hoping for more insight into the biological difference between two kingdoms . | For understanding the complex activities within an organism , a complete and error-free network of protein interactions which occur in the organism would be a significant step forward . The large amount of experimentally derived data now available has provided us with a chance to study the complicated behaviour of protein interactions . The power of such studies , however , has been limited due to the high false positive and false negative rates in the datasets . We propose a network-based method , taking advantage of the tendency of clustering in protein interaction networks , to validate experimental data and to predict unknown interactions . The integration of multiple protein characteristics ( i . e . , structure , function , etc . ) allows our predictive method to significantly outperform two other approaches based on homology and protein-domain relationships on datasets which contain a large amount of interactions , but not much detailed information on the proteins involved in the interactions . In addition , our predictive score based on triadic interaction patterns improves over a pair-wise approach , suggesting the importance of network structure . Moreover , using pooled interactions as prior information , we find evidence for fundamental differences in protein interaction networks between eukaryotes and prokaryotes . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results/Discussion"
]
| [
"computational",
"biology/systems",
"biology",
"computational",
"biology"
]
| 2008 | Predicting and Validating Protein Interactions Using Network Structure |
Controlling for background demographic effects is important for accurately identifying loci that have recently undergone positive selection . To date , the effects of demography have not yet been explicitly considered when identifying loci under selection during dog domestication . To investigate positive selection on the dog lineage early in the domestication , we examined patterns of polymorphism in six canid genomes that were previously used to infer a demographic model of dog domestication . Using an inferred demographic model , we computed false discovery rates ( FDR ) and identified 349 outlier regions consistent with positive selection at a low FDR . The signals in the top 100 regions were frequently centered on candidate genes related to brain function and behavior , including LHFPL3 , CADM2 , GRIK3 , SH3GL2 , MBP , PDE7B , NTAN1 , and GLRA1 . These regions contained significant enrichments in behavioral ontology categories . The 3rd top hit , CCRN4L , plays a major role in lipid metabolism , that is supported by additional metabolism related candidates revealed in our scan , including SCP2D1 and PDXC1 . Comparing our method to an empirical outlier approach that does not directly account for demography , we found only modest overlaps between the two methods , with 60% of empirical outliers having no overlap with our demography-based outlier detection approach . Demography-aware approaches have lower-rates of false discovery . Our top candidates for selection , in addition to expanding the set of neurobehavioral candidate genes , include genes related to lipid metabolism , suggesting a dietary target of selection that was important during the period when proto-dogs hunted and fed alongside hunter-gatherers .
Identifying regions of the genome that have undergone recent positive selection is central to understanding the causes of evolutionary diversification . Nevertheless , developing efficient and statistically robust methods for distinguishing genomic regions under selection from the neutral background expectation remains extremely challenging , particularly under complex , non-equilibrium demographic scenarios . The rapid rise in frequency of a new favorable allele typically leads to a reduced diversity in flanking regions as linked neutral polymorphism accompanies the adaptive substitution in a phenomenon known as genetic hitchhiking [1] . Many methods have been developed to detect such “selective sweep” signatures using genome-wide polymorphism data [2–4] . However , the distortions of the site-frequency spectrum ( SFS ) and/or extended linkage-disequilibrium accompanying episodes of positive selection can be difficult to distinguish from that produced by neutral processes related to a specific demographic history . For example , coalescent trees produced by population bottlenecks or founder events may be indistinguishable from those generated by selection [5 , 6] , and in general , bottlenecks can generate long haplotypes that mimic those observed in selective sweeps [7] . Furthermore , population subdivision can produce counterintuitive and confounding effects [8 , 9] . Consequently , such demographic heterogeneity contributes to the low power and high false positive rates that can occur in genome-wide selection when using contemporary approaches [10–12] . The domestication of dogs from gray wolves is relevant to understanding the broader history of animal domestication and the genetic architecture of rapid phenotypic evolution [13 , 14] . As humans migrated out of Africa , they encountered gray wolves , which served as the founding stock for the domestic dog lineage . Archaeological remains [15–18] , analyses of whole genome sequence data [19] , and mitochondrial genomes of ancient and extant canid lineages [20] jointly support a pre-agricultural origin of dogs which was initiated by association with hunter-gatherers . During this initial interaction , selection for domestication traits was less intentionally directed by humans than it has been with the recent evolution of breed dogs , and instead , was predominantly an incidental by-product of human-wolf-prey interactions [21] . It is likely that dog domestication involved significant genetic changes in response to dietary and behavioral divergence from a wolf ancestor , and comparisons of brain-specific gene expression differences between dogs and wolves support the importance of the latter [22] . Identifying the targets of selection responsible for phenotypic divergence between wolves and dogs is hampered by the demographic complexity of the earliest phases of dog domestication , during which the ancestral dog lineage experienced at least one severe bottleneck and admixture with wolves occurred [19] . Such bottlenecks and admixture can bias selection scans that do not incorporate a demographic model , leading to false positive and negative signals , depending on the circumstances . Despite these potentially confounding effects , of the several studies investigating the genetic basis of phenotypic variation among recently formed breeds and early in domestication [23–28] , none have formally modeled demography to generate a null , neutral expectation for patterns of variation . Recently , we used coalescent-based analysis of whole genome sequence data from dogs and wolves to elucidate the complex demographic history underlying the domestication process . We estimated that domestication entailed a >16-fold reduction in effective population size ( Ne ) for dogs , and a weaker , 3-fold reduction in wolves that began shortly after the initial dog-wolf divergence [19] . By comparison to modern wolves , earlier studies inferred a weaker domestication bottleneck [28–30] , but the ancestral wolf bottleneck had not been previously known , and thus our results showed a greater loss of variation because dogs descended from a more variable ancestral wolf population . We also found evidence for considerable post-divergence admixture , not only between dogs and wolves , but also between wolves and golden jackals , and between golden jackals and the dog-wolf ancestor [19] . Recent admixture between dogs and wolves [31] , and admixture between wolves and coyotes [32] had been previously detected , but the extent to which admixture events may obscure dog origins has only recently been appreciated [18 , 21 , 33 , 34] . This combination of bottlenecks and admixture substantially complicates efforts to distinguish between neutral processes and natural selection . Previous investigations of selection on the dog lineage have taken an approach sometimes referred to as an empirical outlier scan for selection in which putatively selected regions are identified as outliers falling above some arbitrary value [13 , 25 , 27 , 35] . While this approach will detect loci under intense selection , controlling the rate of expected false positives is difficult because the distribution of test statistics under a null demographic model are not taken into account . Similarly , other recent studies of selection in domestic [36 , 37] and wild populations [38] have not accounted for demographic complexity . One difficulty is that a complete demographic model for large genome studies requires a time-consuming investigation of alternative scenarios that is computationally intensive . To investigate positive selection on the dog lineage early in the domestication process and prior to the recent diversification of breeds , we re-examine patterns of polymorphism at 10 million single-nucleotide variant sites using six previously sequenced canid genomes that were used to infer a demographic model of dog domestication [19] . This sample included three wolves from Israel , Croatia , and China; two divergent dog breeds thought to be basal in the dog phylogeny , Basenji and Dingo; and a golden jackal [19] . Specifically , we use our previously inferred demographic model to calibrate a genome-wide scan for signatures of positive selection on the dog lineage and more confidently identify possible targets of recent positive selection while controlling for false positives . Although a recent genomic analysis of a wolf fossil has suggested a slower mutation rate for canids than used in our initial interpretation of our model [39] , the raw parameter estimates from our model are independent of the mutation rate , i . e . our model explains the neutral distribution of polymorphism across our samples , regardless of the well-known uncertainty surrounding mutation rates . Finally , we contrast our findings with a demography-agnostic approach typical of previous studies . Our results expand the catalog of candidate neurobehavioral and dietary genes involved in domestication and provide candidates for future functional studies .
Comparison of our observed data to summary statistics observed in 200 , 000 simulations of 100kb windows under our previously inferred demographic model indicated a general over-dispersion of empirical windows relative to simulated ones . While some of this over-dispersion may be due to heterogeneity in genomic features ( e . g . mutation rate ) and the collective impact of various evolutionary processes , there is a clear excess of extreme values falling in the right-hand tails , outside the distribution of neutrally evolving windows , and consistent with the action of positive selection ( Fig 2 ) . Employing a false discovery rate ( FDR ) of 0 . 01 for Δπ , FST , and Δ TD statistics , we identified 353 , 827 , and 982 windows , respectively , bearing signals consistent with positive selection ( Table 1 ) , for a total of 2081 unique windows . As an alternative approach , we repeated the procedure using null simulations with parameters drawn from the joint posterior distribution rather than fixing them at their mean posterior values ( see Methods ) . The distributions of summary statistics were similar under both approaches ( S2 Fig , Pearson correlations between FDR estimates between each approach >0 . 999 , P < 2 . 2 × 10−16 , 2558 unique windows identified ) . To be conservative , our subsequent analyses focus on the more limited set of 2081 windows found using both approaches . After joining significant windows that were ≤ 200kb apart , both within and across statistics , 349 regions remained in total ( S1 Table ) . These regions overlapped only partially with those identified in previous studies of selection in dogs . Specifically , 53 regions from previous studies were recovered using our approach , and additionally , we detected 296 novel regions . With the 1% threshold , the empirical outlier approach identified 309 outlier regions . The overlap between the FDR-based and empirical outlier methods was low: 59% of the loci based on the FDR-based approach had no overlap with those from the empirical outlier method and 60% of empirical outliers had no overlap with the FDR-based approach ( S3 Fig ) . Two patterns help to explain the low degree of overlap between the methods . First , looking at each summary statistic separately , the vast majority of windows in the top 1% of the empirical distribution have an FDR that would not pass our threshold of 0 . 01 ( S4 Fig ) . A similar pattern is observed with the joint percentile statistics in that the vast majority of windows with an empirical joint percentile in the top 1% have high FDR for individual statistics ( S5 Fig , red points ) , and in many cases more than one statistic . In both cases such outlier windows would be excluded using our FDR-based method . These results suggest that , in the absence of a baseline to assess if signals are consistent with neutral evolution , more than half of outliers in the empirical approach are not supported by an FDR-based approach , and many may actually be false positives . Furthermore , at the gene-level , the FDR and empirical outlier methods identify substantially different sets of genes , with 64% of genes identified in empirical outliers falling outside of FDR-identified regions . This suggests that inferences without demography might lead to mistaken functional interpretation of putative selection signals and gene ontology enrichments ( S6 Fig ) . To rank the putative regions under selection we used a composite-of-multiple signals approach [43] . Specifically , we computed window-specific probabilities of false discovery ( i . e . a false inference of deviation from neutrality ) for our three summary statistics , and then computed the product of 1-FDR across those statistics to obtain a quantity we label CMS1-FDR . As the three summary statistics are not independent within windows , this product does not scale exactly with the weight of evidence for positive selection . Nevertheless , larger values of this statistic should indicate regions that are less likely to have been evolving neutrally . We used the maximum CMS1-FDR statistic observed for any outlier window to rank windows and to localize the selection signal within each region ( Fig 3 ) . This statistic localizes the selection signal within outlier regions more tightly than computing a joint empirical percentile statistic ( S7 Fig ) which does not explicitly incorporate the probability of observing any of the constituent statistics under neutrality . When describing specific candidate genes likely under selection , we employ an additional filter in order to minimize false positives , by considering only genes within the top 100 regions . The joint distribution of summary statistics , joint percentile , and CMS1-FDR for 100kb window highlights the potential problems of not explicitly incorporating demography into selection scan for our set of genomes . To visualize these problems , we classify 100kb windows into four categories . The first category consists of those windows with both a low CMS1-FDR statistic and high FDR for all three summary statistics , falling completely within neutral expectations ( “low CMS , high FDR” in Fig 4A and 4C ) . It is possible for a window to have FDR≥0 . 01 for all three statistics , but still have low enough FDR such that CMS1-FDR is comparable to that observed in outlier regions . We distinguish high CMS1-FDR windows as those with a value for this statistic greater or equal to that observed in the top 100 ranked regions ( i . e . the minimum across those 100 regions of the maximum value observed within a region ) . Thus , the second category consists of sites with FDR≥0 . 01 across all three summary statistics , but CMS1-FDR above this threshold ( “high CMS , high FDR” in Fig 4A and 4C ) . In some cases , windows have FDR≤0 . 01 for at least one summary statistic but there is at least one statistic with high FDR , such that they are classified as deviating from neutrality while having relatively low CMS1-FDR , beneath the threshold defined above ( “low CMS , low FDR in Fig 4A and 4C ) . Finally , there are windows that have consistently low FDR across statistics such that CMS1-FDR is high , owing to consistent signals of selection across statistics ( “high CMS , low FDR in Fig 4A and 4C ) . Based upon this classification of windows , we can distinguish different types of evidence for positive selection ( Fig 4A and 4C ) . In contrast , many of the windows identified by the joint percentile method have high FDR for all three statistics ( contrast blue points in Fig 4A and 4C with red points in Fig 4B and 4D ) , or have high enough FDR for some statistics such that CMS1-FDR is low ( contrast orange points in Fig 4A and 4C with red points in Fig 4B and 4D ) . However , by restricting our analysis to the top 100 windows we exclude regions that would be flagged by such low CMS1-FDR windows that have very low support across all three summary statistics . Key functional changes that derive from selection during the domestication process involve brain function and behavior [27 , 28 , 35] , diet and metabolism [27] , and pigmentation [44] . Consequently , we focus our discussion of the results on genes in regions showing evidence of a selective sweep with the FDR-based approach that are potentially relevant to these phenotypes . We only report genes that either overlap with the peak of the CMS1-FDR statistic within an outlier region , or those that appear most proximate to that peak signal . As a further filter , we evaluated diversity patterns in 500kb intervals surrounding our top 100 outlier regions in a broader panel of 12 diverse breed dogs sequenced to approximately 40x mean coverage ( SRA PRJNA288568 ) . These sequence data include the dingo and basenji used in Freedman et al . [19] and genotypes were called for these data in a manner analogous to [19] . Based on these data , we excluded from further consideration any of the top 100 outlier regions where diversity in the 12-breed panel was greater or equal to that in adjacent non-outlier regions , or where the outlier region was centered on a localized reduction in diversity comparable to those seen in adjacent non-outlier intervals . This confirming data resulted in a reduced set of 68 regions . The filtered set of regions overlapped with only 21 previously identified candidate regions , and contained 47 novel regions ( Fig 5 and S8 Fig ) . In some cases , for any given outlier region , more than one gene may meet our criteria outlined above , such that highlighting particular genes will be ad hoc . Furthermore , it is possible that focusing on particular genes may exclude un-annotated regulatory elements that alter expression of downstream genes more distant from the statistical signal of selection . These caveats aside , we emphasize that our goal is to provide an updated list of candidate genes that can be used as a resource on which to base future investigations and functional assays , rather than to make absolute claims about the importance of any one gene to the domestication process . On a region-by-region basis , we document the extent to which the reported gene is the only one in the putative sweep region or whether it is the gene closest to the peak of the CMS1-FDR statistic . Fig 5 and S8 Fig , provide a summary of the top regions we present given these considerations . Eight of the top 20 candidate regions contain genes that have been implicated in neurological functions in other mammalian species . Our top region is centered on LHFPL3 , a member of the lipoma HMGIC fusion partner family ( Fig 3A ) . Mutations in LHFPL3 have been detected in malignant glioma patients [45] and associated with autism risk [46] . CADM2 is located within the 4th most extreme outlier region ( Fig 3D ) and is a synaptic cell adhesion molecule whose flanking regions show reduced homozygosity in autism patients [47] . GRIK3 is the only gene within the 6th region , and overlaps with the peak in the CMS1-FDR signal . It is a glutamate receptor that has been associated with personality traits such as harm avoidance [48] , schizophrenia and bipolar disorder [49] , and was a neurobehavioral candidate gene in a selection scan of domestic cattle [36] . One cautionary note is that within this region our filters exclude large regions immediately adjacent to it , which raises the possibility that local genomic features might influence the quality of genotype calls . SH3GL2 is the only gene proximate to the peak in the CMS1-FDR within the 8th ranked region and affects synaptic vesicle formation [50] . The peak signal in the 16th ranked region is closest to MBP , a major constituent of the myelin sheath of oligodendrocytes and Schwann cells , and shown to be involved in schizophrenia [51] . PDE7B , which is the only gene overlapping the 17th ranked region , is highly expressed in the brain and is involved in striatal functions related to dopaminergic pathways [52] . Inactivation of NTAN1 ( 19th region ) in mice impairs spatial memory and leads to compensatory gains in non-spatial learning [53 , 54] . However , a RNA polymerase I-specific transcription initiation factor ( RRN3 ) and PDXDC1 , a gene with carboxylase activity associated with diverse phenotypes including renal carcinoma [55] and sensorineural hearing loss [56] were also either proximate to or overlapping the peak in CMS1-FDR signal . GLRA1 ( the only gene in the 20th region ) mediates postsynaptic inhibition in the central nervous system , and mutations have been associated with startle disease [57] . For information on the remaining candidate genes with potential connections to behavior see S1 Text . In our 3rd top outlier , the putative selection signature is most strongly peaked on CCRN4L ( Fig 3B ) . CCRN4L ( also known as Nocturnin ) is expressed in a circadian fashion and studies in mice indicate that CCRN4L activates PPAR-γ , a gene that promotes bone adipogensis as opposed to osteoblast formation and that harbors a known diabetes risk variant in humans [58] . It also is known to regulate the expression of genes involved in lipogenesis and fatty acid binding , and knock-out mice are remarkable in being resistant to diet-induced obesity [58–61] . CCRN4L also suppresses IGF1 , a well-known activator of bone growth [61] that underlies size variation amongst dog breeds [62 , 63] . The direction of these pleiotropic effects of CCR4NL implies a gain-of-function mutation would promote adipocyte formation , alter lipid metabolism , and suppress bone-growth . Within our 9th region , a second peak in CMS1-FDR is centered on SCP2D1 , a paralog of sterol carrier protein 2 ( SCP2 ) , which is highly expressed in genes involved in lipid metabolism , thought to function as an intracellular lipid transfer protein , and for which mice knockouts present altered lipid metabolism [64] . PDXC1 , found within the 19th region in addition to NTAN1 ( see above ) , is associated with plasma phospholipid concentrations and is functionally connected to the glycerophospholipid and sphingolipid pathways [65] . For information on additional candidate genes see S1 Text . The 10th top region was centered on agouti signaling protein ( ASIP ) , a well-known gene influencing pigmentation in mammals [66 , 67] , that has a lesser known role in inhibiting lipolysis [68] . More recently , evidence is emerging that variation at ASIP can influence social behavior , most likely through its antagonistic effects on melanocortin receptors or α-melanocortin stimulating hormone [69 , 70] . Other than a small , predicted gene of unknown function , LYST is the only gene in the 30th region . LYST not only overlaps the peak CMS1-FDR signal , it overlaps the majority of the region as well . LYST has been associated with eye color variation in humans [71] , and mutations can produce lighter skin and hair pigmentation [72] . We found 8883 sites ( 2226 in outlier regions ) containing dog-specific mutations that were at high allele frequency in the 12-breed panel ( S9 Fig ) . Sites fixed between the dog and wolves we sequenced were enriched in outliers with respect to functional class relative to other genomic regions ( χ2 = 23 . 06 , df = 9 , P = 6 . 1 × 10−3 ) . The relative abundance of fixed differences in regions within one kb upstream of the transcriptional start site was twice that of the neutral background . Even so , there were only 12 upstream dog-specific mutations in outlier regions ( S9 Fig ) , representing only 0 . 5% of all fixed sites in outlier regions . In contrast , the majority of dog-fixed sites fall within introns ( 29 . 2% ) and putative intergenic ( 68 . 0% ) regions . Only eight non-synonymous fixed sites were observed in outlier regions , and only five within regions that showed reduced diversity in the12-breed panel . Ensembl’s Variant Effect Predictor tool predicted that , for the transcript annotation displaying the maximum effect , all five variants were mutations of moderate effect . Associated SIFT predictions were as follows: SLK , in 115th ranked region , low-confidence deleterious; two mutations in ACSBG2 , 135th ranked region , deleterious and tolerated , respectively; NOL8 ( uncharacterized protein ) , 292 . 5th ranked region , tolerated; ZNF585B , 292 . 5th ranked region , tolerated . The one high confidence deleterious prediction based upon SIFT is in ACSBG2 , which encodes a protein that is testis and brain-specific , and may play a role in spermatogenesis [73] . Nevertheless , the low frequency of dog-specific non-synonymous fixed sites and their occurrence within relatively low ranked outlier windows suggest coding mutations have been less important in the phenotypic divergence between dogs and wolves . For enrichment analyses , we focused on the top 100 regions ranked by CMS1-FDR , minus those that did not also show reduced diversity in the 12-breed data set . We further filtered the gene set by only considering all genes that fell within 25kb of the peak in CMS1-FDR within those regions . Based upon our requirement that FDR was ≤ 10% , we identified three categories that showed evidence of enrichment in the outlier regions . Notably , we found enrichments for behavior , locomotory behavior , and adult behavior ( Table 2 ) . However , after correction for multiple tests , none of these categories was significant . While it has been suggested that family-wise control of Type I errors is overly conservative for enrichment analyses [74] , we consider our enrichment findings tentative , albeit consistent with the frequent appearance of brain function/behavior genes in our top hit regions .
Extreme population bottlenecks are a hallmark of domestication events , and , in particular , demographic fluctuations and frequent admixture are regarded as important features of the evolutionary history of dogs [18 , 21 , 33] . We present the first effort to control for potential confounding effects of bottlenecks when inferring positive selection on regions of the dog genome , using a robust demographic model constructed from the same set of samples used to perform selection scans . Two categories of genes continually emerged in the top half of our candidate regions list: those influencing behavior , neuropsychiatric disorders and brain function , and genes related to metabolism , in particular lipid metabolism . Genes associated with brain function and behavior are expected , given the dramatic shift from wild to domestic existence . However , genes related to fat metabolism are more surprising , and complement previous evidence for dietary adaptation occurring during domestication , particularly for increased starch metabolism [27] . Our 3rd ranked region is nocturnin ( CCRN4L ) . Evolution at this locus and at other metabolism genes ( e . g . ADRB2 , DIP2C , PLCXD3 ) may have facilitated shifts in lipid content of early domestic dog diets as they scavenged more on carcasses left behind by early humans . In fact , as incipient dogs and early humans began hunting together , prey capture rates may have increased relative to wild wolves and with it , the amount of lipid consumed by the assisting protodogs [75 , 76] . Unique dietary selection pressure may have resulted both from the amount consumed and the shifting composition of tissues that were available to protodogs after humans removed the most desirable parts of the carcass . In addition to genes that may influence behavior and lipid metabolism , our selection scan also identified regions containing genes known to influence pigmentation . The effects of domestication on pigmentation are likely complex , potentially involving a combination of relaxed selection for crypsis , as well as positive selection for particular coat patterns [21] . The classic experiment selecting for tameness in foxes produced piebald and spotted coat color patterns after only 10 generations [44] , suggesting that selection on pigmentation might not be direct but a by-product of selection on behavioral traits . While one investigation found no genetic correlation between coat coloration and tameness in rats [77] , it is possible that in other species these two traits might be functionally coupled . As some of the pigmentation genes in our selection scans influence additional traits , the selection signals we detected may be produced by direct selection on early dog pigmentation phenotypes—the nature of which is not yet clear—or via other traits influenced by such putative pigmentation genes ( e . g . the K locus , Anderson et al . [31] ) . The general trend of reduced fear/aggression in domesticated species raises the possibility that this behavioral shift may have involved selection on the same set of genes in different domesticated species [78] . Although a comprehensive analysis of neurobehavioral candidates across species is beyond the scope of this paper , there are some notably parallelisms . In cats , both glutamate receptor ( GRIA1 ) and protocadherin ( PCDHA1 , PCDH4B ) genes show evidence of positive selection [79] . Similarly , in our top 10 candidate regions , we observe a glutamate receptor ( GRIK3 ) which was also identified in a selection scan in cattle [36] . In contrast , top neurobehavioral candidate genes in rats did not overlap with our candidate gene set [80] . These comparisons suggest that positive selection during domestication may act on particular pathways , such as glutamate receptors , but not necessarily the same genes within those pathways . The simultaneous appearance of multiple traits during domestication , labeled the “domestication syndrome , ” raises fundamental questions concerning the genetic architecture of trait correlations . A recent , as of yet untested , hypothesis is that such correlations are functionally connected early in development during the processes of stem cell proliferation , differentiation , and migration [81] . Additionally , our results also suggest that pleiotropy may play a role in generating trait complexes observed in domestication species . For example , CCRN4L ( our 3rd top hit ) directly influences lipid metabolism , but may indirectly reduce body size through suppressive effects on the well-established growth regulator IGF1 . As an additional example , variation at agouti can influence lipid metabolism and behavior as well as pigmentation . While these examples may point to a mechanism facilitating the domestication syndrome , validation of potential pleiotropic effects among the candidate genes within our outlier regions will require analyses of tissue-specific expression and focused functional studies . Our candidate regions contain a number of potential targets of selection not observed in recent selection scans , and only overlap to a small degree with regions detected by previous studies on dogs ( Fig 5 and S8 Fig ) . While the lack of reproducibility of candidate regions among studies has raised questions concerning their general utility [82] , we attribute discordance with prior studies to several factors . First , we employed a two-level filtering scheme on genotypes that included excluding genotypes intersecting with genome-level features such as copy number variants , where incorrect read mappings will distort allele frequency estimates and summary statistics that rely on those estimates . For example , the filters we used exclude the copy number variable amylase gene that had been reported previously as a crucial target of selection during dog domestication [27] . In that regard , one caveat of our study is that we only will detect adaptations based on copy number variants or structural variation through their effects on linked single nucleotide variation . Second , methods that explicitly incorporate demographic information will likely produce different results from those that do not . This is perhaps most clearly demonstrated by the lack of overlaps between our FDR-based and demography-free joint percentile method ( Fig 5 ) , the latter being characteristic of “empirical” approaches which can potentially miss key targets of selection and falsely identify others [82] . Third , the set of genomes evaluated can have an effect on which regions are identified . From our previous demographic analysis [19] , we determined that admixture between dogs and wolves is geographically structured such that the probability of gene flow is higher for wolf and dog lineages that are geographically proximate . Thus , biased sampling of dogs towards particular breeds may confound selection scan results in dogs by revealing specific features of regional dog breeds and wolves . Interestingly , our candidate regions did show some overlaps with an empirical outlier approach using SNP chip genotyping data when we restricted our sampling to so called “ancient” breeds ( Fig 5 and S8 Fig ) . This overlap presumably occurs because our genomes and the ancient breed panel both retain patterns of polymorphism typical of the earliest dogs . Our model-based assessment of FDR represents the first effort to account for demography in understanding positive selection during dog domestication . While this should reduce false positives among our candidate regions , a few caveats are necessary . First , our analysis is based on a small number of genomes . As a result , our approach should provide sufficient power for sweeps that have been strong , but partial sweeps that have led to less dramatic changes in dog allele frequencies will likely be missed . The challenge for future work is to expand the number of genome sequences analyzed while grounding selection scans with a demographic model that considers the intricacies of population dynamics and inter-lineage admixture . In particular , modeling demography for dozens to hundreds of lineages will pose a substantial inferential and computational challenge . A second caveat is that partial sweeps and soft sweeps may be difficult to detect with the summary statistics used here . It has been recently suggested that soft sweeps are the dominant mode of adaptation in wild populations [83] but there is still considerable uncertainty [84] . A third caveat is that , despite employing a stringent set of filters on both genome and sample level features , we cannot rule out the possibility that clusters of genotyping errors may have occurred in samples from either the dog or wolf lineage , such that some outlier regions may be false positives . Finally , while our use of overlapping , sliding windows allows us to localize the peak signal in outlier regions , it does raise the issue of non-independence and how it affects our approach to controlling for false discovery rate . Previous work indicates that the Benjamini-Hochberg FDR correction should be robust to certain kinds of dependence structure if tests meet the “positive regression dependency on each from a subset” ( PRDS ) criterion [85] . Furthermore , evidence has been presented that linkage mapping and associated tests fulfill PRDS [86] , and the dependency among statistics at SNPs in such cases should very similar to that observed among statistics computed over windows across the genome . Nevertheless , should some features of our genome scans violate PRDS , it would mean that our estimates of FDR would be slightly less conservative ( although certainly more so than empirical outlier approaches ) . We consider this an area worthy of future investigation . Regardless of which mode is dominant , future work will likely uncover additional loci that have undergone positive selection in canids . In particular , future analyses using a larger set of dog and wolf genomes should provide power for assessing potential changes in adaptive substitutions occurring in multiple canine lineages , particularly if a neutral expectation can be calculated using a demographic model inferred for this larger sample . Despite these concerns , our model-based approach identifies a substantial number of new behavioral , metabolic , and pigmentation candidate genes that may contribute to the remarkable success of the oldest domesticated species and the only large carnivore adapted to life with humans .
All sequence alignment , genotyping , and quality-filtering methods were described previously [19] . Genotypes for all six canid genomes in that study were benchmarked against high quality genotypes from the Illumina CanineHD BeadChip , and showed a high degree of concordance with the chip data ( e . g . , 99 . 4% − 99 . 9% of heterozygous genotypes are confirmed by the CanineHD BeadChip ) . Sequence data are available at http://www . ncbi . nlm . nih . gov/bioproject/PRJNA274504 . Vcf files can be obtained via the Dryad data repository at doi:10 . 5061/dryad . sk3p7 . We chose lineages to sequence with the goal of elucidating the timing , demographic context , and geographic origins of dogs . We selected the Basenji and Dingo for sequencing , as they represent two divergent breeds basal on the dog phylogeny [35] . We also utilized the Boxer reference genome as an additional haploid chromosome set . The Chinese , Croatian , and Israeli wolves represent lineages sampled geographically from the three regions from which dogs were previously hypothesized to have originated ( East Asia , Europe , and Middle East , respectively ) . The golden jackal was chosen as an outgroup . This sampling strategy is also informative for understanding selection early in the dog lineage , as it captures the range of variation found in both dogs and wolves , thus minimizing the confusion of selection signals from later , lineage-specific effects , such as might occur were we to bias sampling towards modern breeds of European origin . All genotypes initially generated in CanFam 3 . 0 reference genome coordinates by Freedman et al . [19] were converted to the most current version , CanFam 3 . 1 . For our analysis of the distribution of sites fixed between dogs and wolves , we first identified sites where the Basenji and Dingo were homozygous for the Boxer reference derived allele , and where the three wolves and golden jackal were fixed for an alternative ( i . e . the ancestral ) allele . We then reduced this set of candidate sites by only including sites where the dog derived allele was observed across the 12-breed genome sequences at a frequency ≥ 0 . 75 . We evaluated the functional consequences of dog-specific non-synonymous variants using Ensembl’s Variant Effect Predictor ( http://www . ensembl . org/info/docs/tools/vep/index . html ) . To detect functional enrichment within the genes intersecting our outlier regions , we used the program DAVID [74] , with the Canis lupus gene set as background . We focused on the genes falling within 25kb of the peak CMS1-FDR signal for the top 100 regions , minus those regions that did not also show a reduction in diversity in the 12-breed data set . Because enrichment analyses require a relatively large input set of genes in order to detect enrichment patterns , and given that we already perform statistical inference to identify regions under selection , we report all categories with FDR ≤ 10% . We also report uncorrected P-values as well as P-values corrected for multiple comparisons using Benjamini’s method , although the latter are generally considered to be extremely conservative [74] . | Identification of the genomic regions under selection during dog domestication is extremely challenging because the demographic fluctuations associated with domestication can produce signals in polymorphism data that mimic those imposed by selective sweeps . We perform the first analysis of selection on the dog lineage that explicitly incorporates a demographic model , that by controlling for the rate of false discovery , more robustly identifies targets of selection . To do so , we conduct a selection scan using three wolf genomes representing the putative centers of dog domestication , two basal dog breeds ( Basenji and Dingo ) , and a golden jackal as outgroup , for which we previously inferred a demographic model . We find that our demographically informed analyses filters out many signals that would be otherwise classified as putative selection signals under an empirical outlier approach . We identify 68 regions of the genome that have likely experienced positive selection . Besides identifying a number of new neurobehavioral candidate genes , our candidate regions contain genes related to lipid metabolism , including CCRN4L , which is centered in the 3rd ranked region . This suggests a previously unreported locus of dietary adaptation , potentially due to the change in diet composition as hunting efficiency increased when proto dogs began hunting alongside hunter-gatherers . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"animal",
"types",
"demography",
"domestic",
"animals",
"population",
"genetics",
"wolves",
"vertebrates",
"pets",
"and",
"companion",
"animals",
"dogs",
"mammals",
"animals",
"mammalian",
"genomics",
"population",
"biology",
"zoology",
"animal",
"genomics",
"people",
"and",
"places",
"natural",
"selection",
"evolutionary",
"processes",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"genomics",
"statistics",
"computational",
"biology",
"organisms"
]
| 2016 | Demographically-Based Evaluation of Genomic Regions under Selection in Domestic Dogs |
The frequency of the most common sporadic Apert syndrome mutation ( C755G ) in the human fibroblast growth factor receptor 2 gene ( FGFR2 ) is 100–1 , 000 times higher than expected from average nucleotide substitution rates based on evolutionary studies and the incidence of human genetic diseases . To determine if this increased frequency was due to the nucleotide site having the properties of a mutation hot spot , or some other explanation , we developed a new experimental approach . We examined the spatial distribution of the frequency of the C755G mutation in the germline by dividing four testes from two normal individuals each into several hundred pieces , and , using a highly sensitive PCR assay , we measured the mutation frequency of each piece . We discovered that each testis was characterized by rare foci with mutation frequencies 103 to >104 times higher than the rest of the testis regions . Using a model based on what is known about human germline development forced us to reject ( p < 10−6 ) the idea that the C755G mutation arises more frequently because this nucleotide simply has a higher than average mutation rate ( hot spot model ) . This is true regardless of whether mutation is dependent or independent of cell division . An alternate model was examined where positive selection acts on adult self-renewing Ap spermatogonial cells ( SrAp ) carrying this mutation such that , instead of only replacing themselves , they occasionally produce two SrAp cells . This model could not be rejected given our observed data . Unlike the disease site , similar analysis of C-to-G mutations at a control nucleotide site in one testis pair failed to find any foci with high mutation frequencies . The rejection of the hot spot model and lack of rejection of a selection model for the C755G mutation , along with other data , provides strong support for the proposal that positive selection in the testis can act to increase the frequency of premeiotic germ cells carrying a mutation deleterious to an offspring , thereby unfavorably altering the mutational load in humans . Studying the anatomical distribution of germline mutations can provide new insights into genetic disease and evolutionary change .
Current methods to measure directly the frequency at which human germline nucleotide substitutions arise each generation include the analysis of sporadic cases of human autosomal dominant or sex-linked diseases [1 , 2] and DNA analysis of sperm [3–5] . Studies of the fibroblast growth factor receptor 2 gene ( FGFR2 ) using both approaches [3 , 4 , 6 , 7] have been useful in understanding the origins of Apert syndrome . Individuals born with this condition are characterized by a number of features , including prematurely fused cranial sutures and fused fingers and toes ( see accession numbers ) . This dominantly inherited disease arises with a frequency between 10−5 and 10−6 [6 , 7] virtually always due to a spontaneous germline mutation , . Surprisingly , the high incidence of Apert syndrome cannot be explained by mutations at numerous nucleotide sites in the FGFR2 gene leading to the disease phenotype . In fact , greater than 98% of Apert syndrome cases arise by transversion mutations at only two sites ( C755G and C758G ) . The C755G mutation is located at a CpG [dinucleotide containing a C followed by a G ( 5′ to 3′ ) ] site ( leading to a serine-to-tryptophan substitution at amino acid 252 ) and accounts for two-thirds of the cases [8 , 9] . Based on what is known about the frequency of transversion mutations at neutral CpG sites since humans and chimpanzees last had a common ancestor [10] and mutations at many human disease loci [11] , the C755G mutation frequency is 100–1 , 000-fold greater than expected . One intuitive explanation is that this nucleotide simply has a higher than average chance of undergoing a base substitution ( mutation hot spot model ) . Alternatively , positive selection for diploid germline cells carrying the mutation has also been proposed [4 , 5 , 12 , 13] . In fact , the authors of one study [4] argued that positive selection on mutant spermatogonia was required to explain their results on sperm mutation frequencies . We show , however , that this same data is also compatible with the mutation hot spot model ( Text S1 and Figure S1 ) . With specific regard to human germline selection being responsible for the high C755G mutation frequency , it has been said that “Surprising hypotheses call for unusually strong evidence” [14] . We have therefore developed a new approach to distinguish between the selection and hot spot models that goes beyond sperm analysis .
We studied the spatial distribution of C755G mutations within both testes from each of two normal individuals . The source of the testes , their dissection into small pieces , the purification and quantitation of DNA , and the quantitative mutation detection assay are described in Materials and Methods , Figure 1 , Text S1 , and Figures S2–S4 . For testis 374–1 , the mutation frequencies of 192 individual pieces ranged from <10−6 ( no mutants found among one million genomes ) to as high as 0 . 027 with an average frequency of 3 . 8 × 10−4 ( Table 1 ) . This latter value was close to that observed for an epididymal sperm sample taken from the proximal vas deferens and distal epididymis of the same testis ( 4 . 5 × 10−4 ) . The background mutation frequency of this assay was between 2 . 4 × 10−7 and 6 . 6 × 10−8 ( see Materials and Methods ) . For testis 374–2 , the mutation frequency for the individual pieces varied from <10−6 to 0 . 007 . The average frequency ( 6 . 7 × 10−5 ) was lower than that of testis 374–1 but was consistent with the epididymal sperm sample ( 3 . 9 × 10−5 ) for this testis . The detailed data on the total DNA content and mutation frequency of each piece for these two testes ( and the two additional testes examined below ) are presented in Table S1 . When the data from testes 374–1 and 374–2 are combined , the estimated sperm mutation frequency for this donor is similar to the highest sperm mutation frequencies of normal men of similar age using sperm as a DNA source [3 , 4] ( unpublished data ) . Figure 2A and 2B shows the distribution of the mutation frequencies throughout both testes . Each one is characterized by a very small number of “hot” pieces with very high mutation frequencies compared with the mutation frequencies of the remaining pieces . We also calculated the minimum number of pieces which together contained 95% of the mutant genomes . In testis 374–1 , 12 of the 192 pieces satisfied this criterion . These 12 pieces contain only 5 . 7% of the DNA in this testis , whereas we might have expected that according to a random spatial distribution , many more pieces which together contain 95% of the genomes would have been required . The results for testis 374–2 were quite similar: of the 192 pieces , 95% of the mutant molecules came from five pieces whose total DNA content was only 2 . 6% of the total testis . In many cases , several pieces with high mutation frequencies appear to form foci adjacent to one another in one slice or even between slices ( Figure 2A and 2B ) . We first analyzed the mutation frequency distribution in a portion ( approximately one-quarter ) of testis 854–1 . By dissecting this portion into 192 pieces ( UL in Figure S2 ) , we achieved a resolution four times higher than used for testes 374–1 and −2 . The mutation frequencies of the pieces varied from <2 × 10−6 to 0 . 06 . We found that 98% of the mutants were derived from only three of the 192 pieces ( Figure 2C ) . These pieces contained 1 . 6 % of the genomes in this portion of the testis . The fact that the pieces with the highest mutation frequencies were approximately one-quarter the size of pieces analyzed before suggests that the “hot” pieces from testes 374–1 and −2 studied at the lower resolution might have contained sub-regions with even higher mutation frequencies . The remaining three-quarters of testis 854–1 was analyzed with the same dissection resolution ( Figures 1 and S2 ) used for donor 374 and therefore was cut into 144 pieces ( =0 . 75 × 192 ) . The mutation frequencies for individual pieces ranged from <2 × 10−6 to 0 . 032 . Again , the vast majority of the mutations ( 95% ) came from a few ( six ) pieces that contained only a small fraction of the genomes in this portion ( 5% ) . The mutation distribution is shown in Figure 2D . The total mutation frequency of testis 854–1 calculated using all the data ( 4 . 8 × 10−4 ) was quite similar to the epididymal estimate ( 2 . 6 × 10−4 ) . Finally , we examined testis 854–2 in the same way as 374–1 and 374–2 . This testis had the same features found in the other three: a similar distribution ( Figure 2E ) and range ( Table 1 ) of mutation frequencies and an enrichment of mutant genomes in the hottest pieces ( 95% of the mutants come from 2 . 6% of the testis genomes ) . Surprisingly , the total mutation frequency of this testis was quite different from the epididymal estimate ( Table 1 ) . This epididymal sperm DNA preparation was unique in containing a highly viscous material , suggesting that the epididymis might have been subject to some local pathological process that contaminated the sperm sample with nongermline DNA , thereby explaining the lower mutation frequency estimate . In summary , analysis of the distribution of C755G Apert syndrome mutations in four testes showed a small number of pieces with very high mutation frequencies among a general background of lower mutation frequencies . Some of the pieces had 103- to >104-fold higher mutation frequencies than pieces with the lowest frequencies . In many cases , several pieces surrounding “hot” regions also had relatively high mutation frequencies . To understand whether the unexpectedly high C755G transversion mutation frequency can be explained simply by an inherently higher probability of mutation at this site , we asked whether the frequencies and distribution of the C755G mutations we observed in the testes can be explained by a plausible model of germline mutation ( described in the Materials and Methods section ) . In this model , we make use of data on human germline development and maturation[1 , 15–28] . Our model has two phases ( Figure 3 ) . In the first phase between zygote formation and puberty , called the growth phase , the male germline cells divide symmetrically and thus increase in number exponentially . At each division , there is a probability of a mutation at the disease site . The exponential increase in cell number can magnify the influence of any single mutation on the mutation frequency if it occurs early enough in development . This concept , first noted by Luria and Delbruck [29] in bacteria , can lead to a “mutation jackpot . ” Because the germ cells would be expected to stay in close proximity to their ancestors , any such early mutations will result in regions of the testis with high mutation frequencies . The germ cells of the growth phase eventually form the adult self-renewing Ap spermatogonia ( SrAp ) . In the second phase , called the adult phase , SrAp divide asymmetrically to produce a daughter SrAp ( self-renewal ) and another daughter cell , whose descendants , after only a few additional divisions , will produce sperm . During the adult phase , the number of SrAp remains approximately constant , and each new mutation in this phase produces only one mutated SrAp cell lineage . There are many more SrAp cell divisions during the adult phase than there are SrAp precursor cell divisions during the growth phase; for a 50–60-y-old man , the ratio of mutations expected to occur in the adult phase , compared to the growth phase , is approximately 500 to 1 ( see Materials and Methods ) . We have written a computer program to simulate the model ( Protocol S1 ) . The code can also be downloaded from http://rd . plos . org/pbio . 0050224 . This program allows us to estimate model parameters and test the hypothesis that the data are consistent with this model . A key model parameter is the nucleotide substitution rate per cell division . We infer this value by finding that rate that , in simulations , is most likely to match the overall C755G testis mutation frequency . Even though this frequency is very high relative to data on transversion frequencies at other CpG sites [10 , 11] , the inferred C755G mutation rate per cell division is still low enough that mutations early in the growth phase with the potential to produce a jackpot are unlikely . In those simulations with this inferred rate , almost all of the mutations occur in the adult phase or late in the growth phase , and consequently , there is little variation in the mutation frequency of the testis pieces: 95% of the mutants are found distributed among 95% of the testis pieces . This contrasts sharply with the experimental data where the mutation frequency varies greatly between testis pieces , and 95% of the mutant genomes are found in only a few testis pieces containing between 2 . 6% and 5 . 7% of the total genomes . In simulations , when we increase the mutation rate per cell division so that more mutations occur in the growth phase , there are so many mutations in the adult phase that the overall testis mutation frequency is much greater than that found in the actual data . To test whether the observed distribution of the mutation frequencies among the testis pieces is consistent with the model , we performed both a goodness-of-fit test and a chi-squared test ( see Materials and Methods ) . The mutation hot spot model was rejected ( p < 10−6 ) ( Table 2 ) . We reached the same conclusion when we considered a model with replication-independent mutations ( or a model with both replication-dependent and replication-independent mutations ) , since the adult phase lasts much longer than the growth phase ( see Materials and Methods ) . Our data provide strong support for rejecting the most intuitive explanation for the high frequency of C755G mutations: that the site has an inherently higher probability of undergoing transversion mutations than does an average CpG site in the genome . We conclude that this site is not a mutation hot spot , regardless of the exact molecular mechanism responsible for the C755G mutation event . Incorporating selection as a modification to our model can reproduce both the overall testis mutation frequency and the distribution of mutation frequencies observed in the testis pieces . One possibility is that the C755G mutation promotes rare SrAp symmetric divisions in the adult phase ( in the original model , adult phase divisions were always asymmetric [16–18 , 21 , 24] ) . The occasional symmetric divisions would allow mutated SrAp to grow locally in number over time , and thereby increase the overall mutation frequency in the testis . We add to our first model a selection parameter: at each adult phase generation , with probability p , a mutated SrAp divides symmetrically and with probability 1− p it divides asymmetrically ( after a symmetric division , each daughter SrAp reverts to having asymmetric divisions until the next rare symmetric division occurs; a similar model was independently proposed in a recent publication [12] ) . We infer the probability p and the mutation rate per cell division by fitting both the overall testis mutation frequency and the minimum number of pieces , which together contain 95% of the mutant genomes . The inferred probability p of a symmetric division is approximately 0 . 01 ( on average , one of every hundred divisions is symmetric ) . With this modification , we can no longer reject the model using the chi-squared test ( Table 2 ) : the distribution of frequencies in the testis pieces now matches the data . In the simulations , foci of high mutation frequency emerge and , as in the data , these foci often intersect several adjacent testis pieces . With p near 0 . 01 , simulations with the inferred mutation rate match the observed mutation frequencies of the testes ( 10−4–10−5 ) ; however , simulations with this rate , but with p = 0 ( the no-selection case ) , predict a testis mutation frequency in the 10−8–10−10 range for a 20-y-old male . This lower range is similar to C-to-G mutation frequencies estimated for CpG sites in the studies on neutral and disease mutations [10 , 11] . Another difference between the p = 0 ( no-selection ) and the p > 0 cases is that for the p = 0 case , the model predicts that the mutation frequency increases linearly with the man's age , whereas for the p > 0 case , this increase is exponential . Previously , it was suggested [4] that selection could cause the observed exponential increase in the birth incidence of Apert syndrome with the father's age ( paternal age effect [1 , 12 , 15 , 30 , 31] ) . To match the observed 20-fold increase between fathers who are younger than 24 and fathers who are older than 50 [30] , the inferred p value would have to be approximately 0 . 004 . There are undoubtedly other models with different forms of selection on the mutant cells that are also consistent with the data . A modification to the model that does not include selection , but could reproduce the testis data , is if there was a higher nucleotide substitution rate per cell division in the growth phase than in the adult phase . However , this could not account for the paternal age effect , nor do we know of any evidence to support such a difference . As a control , we also analyzed the C of a CpG site in an intron of the CAV1 gene on Chromosome 7 , which is presumably a neutral site . We reanalyzed testis 374–1 and 374–2 using an assay designed for this site ( Materials and Methods ) . The data are given in Table S2 . The total C-to-G mutation frequency in sperm from epididymis 374–1 was 7 × 10−6 . The mutation frequency that was estimated by summing up the individual pieces of the testis was 3 . 3 × 10−6 . Unlike the disease site , there was a narrow range of frequencies in these individual pieces ( <4 × 10−6 to 2 × 10−5 ) . Ninety-two testis pieces were required to collect 95% of the mutants; these pieces contained 50% of the testis' genomes . Eighty-eight pieces , containing 46% of the genomes , had mutation frequencies less than the background of the assay . Therefore for the testis pieces contributing to the mutation estimate , 88% of the pieces , containing 93% of the genomes , were required to amass 95% of the total mutations . Moreover , selection is not required to explain the data , since we could not reject our model for the neutral mutation case with the selection parameter p fixed at zero . The data from testis 374–2 was similar . The control site mutation frequency is two orders of magnitude less than at the disease site; it is somewhat higher than expected from other estimates [10 , 11] , but those estimates are presumably from younger reproducing-age individuals , whereas donor 374 is 62 y old and has had approximately seven times more adult phase generations to accumulate mutations . The lack of any foci with very high mutation frequencies suggests that Luria and Delbruck [29] jackpot formation is rare in the human testis , and that something is fundamentally different between this CpG site and the Apert 755 disease CpG site .
In theory , heritable nucleotide substitutions can arise in any germline cell during scheduled DNA replication or in nondividing cells by means of error-prone DNA repair ( see [32] ) . Two observations support the importance of the cell division–dependent mutation process . First , neutral germline mutations seem to arise 3–6-fold more frequently in males than females in many organisms , including humans ( male-driven evolution , see [33] ) . Second , a number of genetic conditions including Apert syndrome appear to increase in the offspring of men as they age ( paternal age effect [1 , 12 , 15 , 30 , 31] ) . Both male-driven evolution and the paternal age effect can be explained by the cell division–dependent mutation process , because cell divisions of self-renewing spermatogonia occur throughout a man's life , whereas the cellular precursors of eggs ( oogonia ) cease replication during the fetal life of a female [1 , 15] . We examined the molecular anatomy and frequency of the Apert syndrome C755G mutation in normal human testes to test whether the high mutation frequency was due to an exceptionally high C-to-G transversion mutation rate per cell division . The results show that the observed C755G mutation frequency and distribution within the testes cannot be explained by this hot spot model ( p < 10−6 ) . An alternate hypothesis to explain the high C755G mutation frequency argues that diploid premeiotic cells carrying the C755G mutation have a selective advantage over wild-type cells [4 , 5 , 12 , 13] . In one case [5] , the authors cited the puzzling observation that the magnitude of the sex bias for the C755G Apert syndrome is at least 99-fold greater in the male germline than in the female germline [34 , 35] , whereas estimates of male bias ( male-driven evolution ) using data on neutral mutations at many different sites would indicate only an ∼5-fold male preference [33 , 36 , 37] . Rare patients with multiple mutations in the FGFR2 gene that leads to Apert syndrome were also cited as support for a germline selection model [13] . Finally , in another study [4] , the authors exploited a nearby single nucleotide polymorphism ( SNP ) to argue that selection acted on the C755G mutation ( however , see Text S1 and Figure S1 ) . In our experiments , we were not only able to clearly reject the hot spot model , but also we could show that modifying our model of germline development by incorporating a simple selection scheme led to predictions on mutation frequency and testis distribution consistent with our data . This selection takes place on SrAp cells carrying the C755G mutations that arise at approximately the frequency expected from the existing data on neutral mutations [10 , 11] . The selection model proposes that mutant adult SrAp occasionally divide symmetrically ( inferred rate 1 out of 100 divisions on average , or approximately once every 4 y ) , whereas wild-type SrAp always undergo asymmetric self-renewal divisions . Considering all of the published work [4 , 5 , 12 , 13] as well as our present results , it now seems very likely that positive selection can be a driving force acting to increase the germline mutation frequency in humans above the frequency at which spontaneous nucleotide substitutions arise . We would like to emphasize that the type of selection we are discussing is on diploid premeiotic cells . Previous proposals have suggested that C755G mutation bearing sperm may have a selective advantage over wild-type sperm [3 , 38] . Selection taking the form of competition among sperm is well known in plants and animals [39] , and is even documented in primates [40] . However , for this particular mutation , the testis and epididymal sperm data we collected as well as data on ejaculated sperm ( [3 , 4] and unpublished data ) and Apert syndrome birth data [6 , 7] show similar mutation frequencies . Therefore , while further selection on sperm is possible , we believe that the vast majority of the increase in mutation frequency is due to selection on the diploid premeiotic cells . Why should a mutation that has a distinct selective disadvantage when present in all the cells of an organism have a selective advantage when present only in a small fraction of the germline cells ? It is worth noting here that achondroplasia , the most common cause of dwarfism , has many similarities to that of Apert syndrome ( see [5] ) . Virtually all of the new achondroplasia mutations arise in the male germline at one nucleotide site ( G1138A ) in the fibroblast growth factor receptor 3 ( FGFR3 ) gene and with a mutation frequency even higher than the C755G mutation in FGFR2 . These common characteristics suggest that the G1138A mutation may also increase to such a high frequency by a selective mechanism [4 , 5 , 12 , 13] . It is interesting to note that both FGFR2 and FGFR3 are receptor tyrosine kinases and can influence downstream members of the signal transduction pathway ( for a more detailed discussion see [41 , 42] ) . Especially relevant may be the fact that some mutations in FGFR2 and FGFR3 ( although usually not the specific mutations that cause Apert syndrome or achondroplasia ) have been associated with certain cancers [43 , 44] and cancer susceptibility [45] . Germline selection in diploid germ cells of animals was considered by the population geneticist Ian Hastings [46 , 47] . He examined how mitotic gene conversion and somatic crossing over events in diploid germline cells of animals could lead to loss of heterozygosity of recessive alleles and the possibility of positive or negative selection on such alleles . He calculated that selection against rare germline cells made homozygous for a recessive allele can effectively lower the transmission of the deleterious allele to offspring thereby reducing disadvantageous alleles entering the population and burdening it with reduced viability or fertility . Using plausible models , this reduction in the mutational load could be as large as 100-fold . Similarly , loss of heterozygosity could allow recessive alleles that conferred a germline advantage to be spread more quickly in the population . Experimental literature on germline selection in premeiotic diploid cells in animals is very sparse ( see [48] and references therein ) but in one case , wild-type Drosophila cells were produced by a genetic trick in the germline of females heterozygous for a phenotypically recessive mutation and were found to have a proliferative selective advantage compared to the background heterozygous cells . Hastings' analysis was primarily concerned with recessive alleles that were already polymorphic in the population . But gain-of-function mutations that arise sporadically in the testis would behave in the same way , because a second event leading to loss of heterozygosity is not required for positive and negative selection to be effective . A new gain-of-function mutation with a germline selective advantage will more likely be transmitted to the next generation , because the effective mutation frequency is elevated beyond the level that can be achieved by the mutation process alone . A disadvantageous gain of function mutation would be less likely to be tested in the population if it were selected against in the germline . Finally , Hastings realized that alleles conferring a selective advantage in the germline may be disadvantageous in the adult and might lead to “mitotic drive” systems that could increase the mutational load of a population . Both Apert syndrome and achondroplasia may be examples of such a system , and additional examples of mutations of medical interest may also exist ( see [15 , 30 , 31 , 33] ) . The method we have developed can be used to test this hypothesis at any locus in many different species if a sufficiently sensitive mutation assay can be made available .
Both testes from each of two donors ( 374 , 62 y old , and 854 , 54 y old ) were supplied by the National Disease Research Interchange ( NDRI ) in Philadelphia , PA , United States . The testes were frozen no longer than 10 h post mortem and stored at −80 °C . Procurement of the testes from the NDRI was approved by the Institutional Review Board of the University of Southern California . The epididymis was removed from the testis . Sperm cells were collected from the tail of each epididymis and the first ∼3 cm of the adjacent vas deferens after they were cut into small pieces and incubated in 25 ml of 2 . 2% sodium citrate with shaking for 60 min at room temperature . Sperm cells were concentrated by centrifugation at 6000g . Sperm DNA was isolated using a Puregene DNA purification kit and protocol ( Gentra Systems; http://www1 . qiagen . com/Products/dna . aspx ) except that 40 mM DTT was added to the lysis solution . Each testis was fixed in 70% ethanol at 4 °C for about 3 d . Three of the four testis ( 374–1 , 374–2 , and 854–2 ) were each cut into six slices and each slice further divided into 32 pieces . The dissection scheme and how the position of each piece within the testis was recorded is shown in Figure 1 . Figure S2 shows the slightly different scheme that was used only on testis 854–1 . An introduction to testis anatomy can be found at: http://training . seer . cancer . gov/ss_module11_testis/unit02_sec01_anatomy . html . DNA was extracted using a Puregene kit ( Gentra Systems ) . DNA concentrations were estimated using real-time PCR and primers for a unique sequence on Chromosome 21 ( additional details are found in Text S1 ) . The number of genomes estimated for each testis piece is presented in Table S1 . The assay for C755G mutations resembles an allele-specific PCR ( ASPCR ) that amplifies mutant but not wild-type molecules using a primer whose 3′ end contains a base that is complementary to the mutant [49–51] . The most critical difference between ASPCR and the method we used ( pyrophosphorolysis-activated PCR , or PAP [52 , 53] ) is that both PAP primers are blocked at their 3′ ends with a dideoxynucleoside monophosphate ( ddNMP ) . The PAP primers anneal perfectly to the mutant template but anneal to the wild-type template with 3′ terminal mismatches ( Figure S3 ) . Primer extension by DNA polymerase , and thus PCR , is not expected to occur in either case . However , we used a DNA polymerase ( in our case TMA31FS [54] ) capable of both efficient pyrophosphorolysis of ddNMP terminated primers ( thereby unblocking them ) and primer extension using dNTPs . The pyrophosphorolysis and amplification preferentially occurs using the mutant templates . PAP is far more selective than conventional ASPCR for rare mutation detection . If the TMA31FS polymerase mistakenly removes a mismatched terminal ddNMP , insertion of the correct ( templated ) nucleotide simply results in the generation of another copy of the wild-type template and not loss of selectivity . To measure the mutation frequency of a DNA sample , 25-μl PAP reactions were carried out in 96-well plates [Opticon 2 , MJR ( BioRad; http://www . biorad . com ) ] or 10-μl reactions in 384-well plates [ABI 7900HT ( Applied Biosystems; http://www . appliedbiosystems . com ) or Roche LightCycler 480 ( Roche Applied Science; http://www . roche . com ) ] . Each reaction contained 20 mM HEPES ( pH 7 . 5 ) , 30 mM KCl , 50 μM Na4PPi , 2 mM MgCl2 , 80 μM of each dNTP , 80–200 nM of each primer , 2 μM Rox , 0 . 2 X Syber Green I , 0 . 04 unit/μl TMA31FS DNA polymerase ( Roche Molecular Systems ) , and 25 , 000 copies of testis or epididymal sperm DNA molecules . Research samples of TMA 31FS DNA polymerase may be obtained from Thomas W . Myers , Director , Program in Core Research , Roche Molecular Systems , 1145 Atlantic Avenue , Alameda , CA 95401 , USA ( thomas . myers@roche . com ) . The sequences ( see accession numbers ) of the C755G-specific PAP primers used in these experiments are: 5'-CCCCACTCCTCCTTTCTTCCCTCTCTCCACCAGAGCGAT ( ddG ) and 5'-TTTGCCGGCAGTCCGGCTTGGAGGATGGGCCGGTGAGGC ( ddC ) . . The 79-bp PAP product makes detection possible using quantitative PCR . The primers were synthesized by Biosource International ( http://www . biosource . com ) or Biosearch Technologies ( http://www . biosearchtech . com ) . The PAP primer ending in a ddC was synthesized by conventional methods . The PAP primer ending with ddG required synthesis using 5'-CE phosphoramidites . The ddNTP at the 3' end of each primer is complementary to nucleotide 755 but each to different strands . The PAP cycling conditions included an initial denaturation step of 2 min at 94°C followed by 150 cycles of 6 s at 94 °C and 40 s at 78 °C . Positive reactions could easily be distinguished from negative ones by considering the kinetics of the increase in fluorescence with cycle number and the melting profile of the final PCR product . For sample PAP data , see Figure S4 . For the control , we measured the frequency of a C-to-G transversion at a CpG site on human Chromosome 7 . The C nucleotide is known to be polymorphic in the human population ( see accession numbers ) . Testis donor 374 was genotyped and found to be a C/C homozygote and thus his testis and sperm cells could be examined for the presence of C-to-G transversion mutations at this site . To measure the frequency of mutation , 10 μl PAP reactions were carried out in 384-well plates ( ABI 7900HT , Applied Biosystems ) . Each reaction contained 20 mM HEPES ( pH 7 . 5 ) , 30 mM KCl , 50 μM Na4PPi , 2 mM MgCl2 , 80 μM of each dNTP , 200 nM of each primer , 2 μM Rox , 0 . 2 X Syber Green I , 0 . 04 unit/μl TMA31FS DNA polymerase ( Roche Molecular Systems ) , and 25 , 000 copies of testis or epididymal sperm DNA molecules/reaction . The sequences of the C755G-specific PAP primers are: 5′-TATTAATATAACTTAGTATCTGTCACCCCAAGGGAACCAA ( ddG ) and 5′-TATTAATATAGAGTATTGACTCTTATTCTTGGGCTTCGAC ( ddC ) . Note that the ten 5'-most nucleotides of these primers are not complementary to the mutated target sequence . An 81-bp PAP product results . The PAP cycling conditions included an initial denaturation step of 1 min at 94 °C followed by 134 cycles of 6 s at 94 °C and 40 s at 70 °C . A total of 40 reactions containing 25 , 000 genomes each ( a total of 106 genomes ) were used to estimate the C755G mutation frequency for every piece of testes 374–1 and −2 . In those cases where fewer than 25/40 reactions were positive , we took the number ( after Poisson correction ) as an estimate of the C755G mutation frequency for that piece . If 25 or more reactions were positive , then the experiment was repeated using 10-fold dilutions until a dilution of the DNA from the piece was found to give fewer than 25/40 positives . Based on the mutation frequency results from the first pair of testes , a slightly different strategy was used for testes 854–2 and the upper right and tail half ( UR and TH , respectively ) parts of 854–1 ( Figure S2 ) . First , ten reactions were carried out on every piece ( total of 250 , 000 genomes analyzed ) . In the case of those pieces where four or fewer reactions were positive , we took the number ( after Poisson correction ) as an estimate of the C755G mutation frequency . In those cases where five or more of the ten reactions were positive , we carried out additional experiments using 40 reactions ( 1 , 000 , 000 testis genomes ) and appropriate dilutions in the manner described for the 374 testes . Repeated measurements from the same piece had an average deviation of 32% from the mean value . For all mutation counting , the estimate of the total testis mutation frequency was the average of the frequencies of the pieces weighted by the number of genomes in those pieces . For a positive assay control , ten or 20 Apert C755G mutant molecules from a patient heterozygous for the C755G mutation ( kindly provided by Dr . Mimi Jabs , Johns Hopkins Medical Institutions ) were added to 500 , 000 human genomes and divided into 20 reactions ( an average of 0 . 5 or 1 mutants per reaction ) . Negative assay controls consisted of 25 , 000 genomes of nongermline human blood DNA from normal individuals ( Clontech; http://www . clontech . com ) . Over the course of our study , we observed an assay background in human blood DNA of between 6 . 6 × 10−8 and 2 . 4 × 10−7 , based on experiments using ∼67 , 000 , 000 control genomes . This places an upper limit on the possible mutation frequency in human white blood cells , although it would be lower if false positives can also result from artifacts of the assay itself . The method for counting the Chromosome 7 control C-to-G transversion mutations was identical to that used for testes 854–2 and the UR and TH parts of 854–1 . For a positive control , 20 mutant genomes of blood DNA from a heterozygous ( C/G ) individual were added to 500 , 000 human genomes from individuals homozygous for the C allele and divided into 20 reactions ( an average of one G allele-containing genome per reaction ) . Negative controls consisted of 20 reactions ( a total of 500 , 000 genomes ) of blood DNA from individuals homozygous for the C allele . Classical cytological approaches provide our main source of information about human germline development and maturation [1 , 15–28] . Not long after fertilization , human primordial germ cells arise in the developing human embryo and migrate to the embryonic sites of testis formation ( genital ridges ) . The germ cells become organized into seminiferous tubules , proliferate , and differentiate into gonocytes and fetal spermatogonia . By the beginning of puberty , the seminiferous tubules are composed primarily of spermatogonia and supporting Sertoli cells . The spermatogonia ( including the adult SrAp ) are located in their niche at the periphery of the tubule and in intimate association with Sertoli cells . After puberty , seminiferous tubules also contain B spermatogonia [17] that are derived from the asymmetric division of SrAp . B spermatogonia are a precursor of meiotic cells . Four B spermatogonia and eventually 16 sperm are thought to result from each division of one SrAp [16 , 17 , 21 , 24] . In spermatogenesis , both meiotic and post-meiotic cells gradually move from the periphery toward the lumen of the tubule . Sperm enter the lumen of the tubule ( spermiation ) and are transported to the epididymis , where they undergo further development and are stored . Note that in a seminiferous tubule segment , any mutant SrAp will be in close proximity to its spermatogonial , meiotic , and post-meiotic cell descendants that will carry the same mutation . Although SrAp divide every 16 d , it takes much longer than 16 d for the division products of one to yield sperm [18] . This creates overlapping “generations” of different germline cell types that are found in close proximity to one another and all derived from the same SrAp cell . We model the testis in two phases ( Figure 3 ) . In the first phase between zygote formation and puberty , called the growth phase , self-renewal cells divide symmetrically to produce two self-renewal cells . This continues for g generations . In this phase , the testis grows exponentially from one self-renewal cell to 2 g self-renewal cells . The total number of divisions in the testis in this phase is ( 2 g − 1 ) . This corresponds to a total of ( 2 g+1 − 2 ) daughter cells being produced , and there are this many opportunities for replication-dependent mutations . There is a mutation rate per cell division λ at the disease site . A mutation is shared by all of a cell's descendants , so a mutation early in the growth phase will produce many mutants . To be more specific , a mutation at the kth generation will produce 2 g−k mutant descendants . This concept , first noted by Luria and Delbruck [29] in bacteria , can lead to a “mutation jackpot . ” Since the germ cells would be expected to stay in close proximity to their ancestors , any such early mutations will result in regions of the testis with high mutation frequencies . We will discuss the spatial details in the Spatial Structure section below . Zhengwei et al . measured on average 656 , 000 , 000 type A spermatogonial cells per testis [23] . Using the number of genomes per testis we have measured , the distribution of the different cell types in Table 2 of [23] ( to add sperm , see [20] ) , and the ploidy of these different cell types , we infer that for donor 374 , each testis has approximately 480 , 000 , 000 type A spermatogonial cells and that for donor 854 , each testis has approximately 260 , 000 , 000 type A spermatogonial cells . Because one-half of the type A spermatogonial cells are SrAp [20 , 24 , 55] , we estimate the number of growth phase generations g is 27 or 28 . We have also considered higher numbers from the literature ( 30 [15] or 34 [19] ) . In the second phase , called the adult phase , SrAp divide asymmetrically to produce one SrAp and another cell whose descendants will ultimately produce sperm . In this phase , the number of SrAp remains constant . These cells divide every 16 d [18] , so assuming this phase begins at puberty ( age 13 ) , donor 374 who is age 62 has experienced a = 1 , 127 adult phase generations , and donor 854 who is age 54 has experienced a = 943 adult phase generations . The total number of divisions in the testis in this phase is equal to ( a × 2 g ) . For these donors , there are approximately 500 ( ≈a × 2 g/2 g +1 ) more opportunities for replication-dependent mutation in the adult phase than in the growth phase , and because we assume identical mutation rates per cell division in both the growth and adult phases , there will also be , on average , this many more mutations in the adult phase . These mutations will be randomly spread throughout the testis . Unlike in the growth phase , mutations in the adult phase lead to only a single mutated SrAp ( and approximately 25 other mutated germ cells [23] ) . If the testis mutation frequency is f , then f ÷ a is an upper bound for the mutation rate λ ( this inequality ignores mutations in the growth phase ) . We examine two modifications to this model . First we consider replication-independent mutations . The number of these mutations is proportional to time rather than the number of divisions . The literature is not definitive [27 , 28] , but it appears that most of the growth phase generations occur in approximately 1 y ( the 9 mo of embryonic and fetal development plus possibly several months of infancy ) . There may be several further growth phase generations in the ∼13 y before the adult phase begins; however , the number of such generations is small enough that any mutations in these years , like in the adult phase and unlike in the growth phase , will not produce significant mutation clusters ( in the absence of selection ) . Then , letting z be the age of the donor , the ratio of replication-independent mutations that cannot cause significant mutation clusters to those mutations that will cause such clusters is ( 2 g × z ) / ( 2 g+1/g ) . For our donors , this ratio is at least 700 to 1 , even more extreme than for the replication-dependent mutations . Consequently it would be difficult for a model incorporating replication-independent mutations to explain our testis data . The second modification we consider is selection . We introduce a new parameter into the model: at each adult phase generation , with probability p , a mutated SrAp divides asymmetrically , and with probability 1 − p it divides symmetrically to produce two mutated SrAp . This symmetric division is like that in the growth phase , except that afterward , each daughter SrAp reverts to having asymmetric divisions until the next rare symmetric division occurs; a similar model was independently proposed in a recent publication [12] . The occasional symmetric divisions allow mutated SrAp to grow locally in number over time , and thereby increase the overall mutation frequency in the testis . On average , a mutation in this phase will leave descendants ( if a mutation occurs with x generations remaining , it leaves on average ( 1 + p ) x descendants , the time of a mutation is uniform on [0 , a] ) . Since the mutation probability and p are small , the number of self-renewal cells will still stay roughly constant in the adult phase . Above we discussed the number of mutant cells with regard to our various models , now we will discuss their spatial distribution . For simplicity , we have assumed that the testis is in the shape of a rectangular box , and it is divided into a regular grid of 6 × 8 × 4 = 192 cubic testis pieces ( Figure 1 ) . If a mutation event occurs in the growth phase after testis formation , then there is a mutation cluster where all of the SrAp cells in a cubic region are mutants . Depending on the size of this cluster , it may be contained within one of the testis pieces , intersect parts of several adjacent testis pieces , or contain all of one or more testis pieces and intersect parts of the surrounding testis pieces . The number of mutants in a cluster depends on the cell generation in which the mutation event occurred and is discussed in the previous paragraphs . The size of a cluster is calculated assuming that it has the same density of SrAp cells as the rest of the testis , and the total testis has 2g SrAp cells . We set the units so that 1 is the length of one of the 192 testis pieces . Then a mutation at the kth generation will produce a cubic mutation cluster with length ( 192 × 2−k ) 1/3 . To consider some examples , a mutation at the first generation means half of the testis is mutated , a mutation at the 8th generation produces a cluster with 91% of the length and 75% of the volume of one of the testis pieces , a mutation at the 21st generation produces a cluster with 5% of the length and 0 . 01% of the volume of one of the testis pieces . The location of a cluster is random within the testis . We assume that at the beginning of the adult phase , there is an equal fraction of the 2g SrAp cells ( mutated or not ) in each of the 192 testis pieces . During the adult phase , mutation events occur randomly throughout the testis . If p = 0 , then each such mutation event produces only one mutant SrAp cell . If p > 0 , then each mutation event produces a cluster of mutant SrAp cells . The number of mutants in a cluster is discussed in the previous paragraphs . For simplicity , we assume that the shape of a cluster is a cube . The size of a cluster is calculated assuming that the cluster has the same density of SrAp cells as the rest of the testis . Depending on the cluster's size , it may be contained within one testis piece or intersect several adjacent testis pieces . Unlike the growth-phase clusters , the SrAp cells in the selection clusters are in addition to the nonmutated SrAp cells in the testis pieces . Due to the exponential growth of these selection clusters , there is a relatively narrow range of realistic parameter values for p . For example , a mutation early in the adult phase will produce a cluster approximately 10−5 the size of the testis if p is 0 . 01 , whereas this same cluster will be larger than the rest of the testis if p is 0 . 03 or greater . Multiple mutation clusters may intersect the same testis piece and a given cell may only be mutated once ( there are no reverse mutations ) . We have written a computer program ( Protocol S1 or download from http://rd . plos . org/pbio . 0050224 ) to simulate the model and its modifications . We fix the parameters a = 1 , 127 for donor 374 and a = 943 for donor 854 , and consider both the extremes g = 27 and g = 34 . First we consider the mutation hot spot model ( p = 0 ) . We simulated the model many times to estimate the probability that the simulated mutation frequency is within 5% of the observed mutation frequency as a function of the mutation parameter λ ( mutation rate per cell division ) . The λ that optimizes this quantity is our approximate maximum likelihood estimate ( this technique is sometimes called a rejection sampler ) and standard methods can produce approximate 95% confidence sets . For the selection model ( p > 0 ) , there are two parameters to consider: p and λ . There are an infinite number of pairings of these parameters' values ( a subset of parameter space ) for which the model can match the observed mutation frequency . Consequently , we searched for the parameter values which optimized the following two-criteria probability: ( a ) that the simulated mutation frequency is within √5% of the observed frequency , and ( b ) that the simulated fraction of testis' genomes in the fewest number of testis' pieces necessary to contain 95% of the mutants is within √5% of this observed quantity . Note that for the mutation hot spot model ( p = 0 ) this two-criteria probability is 0 for all values of λ , since no single value of λ can match both criteria . In general , to estimate the empirical likelihood , we simulated the model 1 , 000 times for each set of parameter values . To determine whether the distribution of mutations in the 192 pieces is consistent between the model and the data , we used two hypothesis tests . First we counted the number of pieces with mutation frequencies in each of the eight color categories in Figure 2 ( [0–25/million] , etc . ) . For both the mutation hot spot model ( p = 0 ) and the selection model ( p > 0 ) , we simulated the model under optimal parameters and counted the number of pieces with frequencies in these same categories , and then performed the chi-squared test . For the control site , since the observed frequencies are much lower , we considered the following four categories: [0–4 × 10−6 ) , [4 × 10−6–9 × 10−6 ) , [9 × 10−6–1 . 4 × 10−5 ) , [1 . 4 × 10−5+ ) . The notation [x–y ) means frequencies f such that x ≤ f < y . Because a mutation cluster can overlap adjacent testis pieces , this dependence makes the chi-squared test anticonservative . In order to not reject the selection model ( p > 0 ) , this test is sufficient; however , in order to reject the hot spot model ( p = 0 ) , we have also performed a goodness-of-fit test . For each simulation , we computed the following statistic: the minimum fraction of testis pieces which contain at least 95% of the total mutants . We simulated the hot spot model under the optimal parameters and recorded this statistic for those simulations such that the total testis mutation frequency was within 5% of the real data . We repeated this procedure such that 1 , 000 , 000 simulations met this criterion ( under optimal parameters over 90% of simulations met the criterion ) . The simulated statistic was never lower than 90% , in contrast with the much lower values for the real data in Table 1 ( <6% ) . Therefore an empirical p-value is <10−6 . However , this value is just a function of our patience in running simulations; we contend that for however many simulations are run , this statistic will never be lower than for the data . In order to get a p-value independent of the number of simulations , we fit the distribution of simulated statistics to a t-distribution ( admittedly a poor fit ) and the statistics for the real data were over 300 standard deviations away , making the p-value less than 10−200 . The ( anticonservative ) chi-squared test rejected the hot spot model with p-value less than 2 × 10−16 . Regardless of which p-value is used , there is strong support to reject the hot spot ( p = 0 ) model .
The Online Mendelian Inheritance in Man ( www . ncbi . nlm . nih . gov/omim/ ) accession number for Apert syndrome is 101200 . The National Center for Biotechnology Information ( NCBI; http://www . ncbi . nlm . nih . gov/ ) Nucleotide accession number for the FGFR2 gene is NM_000141 . The Single Nucleotide Polymorphism database ( http://www . ncbi . nlm . nih . gov/projects/SNP/ ) build 127 transversion for the C nucleotide site on human Chromosome 7 is rs3801993 . | Some human disease mutations occur 100–1 , 000 times more frequently than would be predicted from genome wide studies of mutation in different species . In Apert syndrome , for example , two-thirds of all new causal mutations occur at only one base in the affected gene . This unusually high frequency suggests that something about that DNA base or its local surroundings makes it highly susceptible to mutation . We studied this hypothesis by examining the location of cells containing this mutation in the testes of normal men . We found that mutant cells were not uniformly distributed throughout the testes , as would be expected for random mutations . Instead , we found 95% of the mutants in small clusters containing only a few percent of the total testes cells . A higher-than-average mutation rate could not explain the data . We propose that these mutations arise at the expected rate , but that mutated cells gain a selective advantage that allows them to increase their frequency compared to nonmutant cells . Our results—which argue against the mutational hot spot model in favor of a selection model—suggest how germline selection in animals can alter the mutational load of a species . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
]
| [
"developmental",
"biology",
"homo",
"(human)",
"computational",
"biology",
"evolutionary",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
]
| 2007 | The Molecular Anatomy of Spontaneous Germline Mutations in Human Testes |
Diarrheal disease remains among the leading causes of global mortality in children younger than 5 years . Exposure to domestic animals may be a risk factor for diarrheal disease . The objectives of this study were to identify animal-related exposures associated with cases of moderate-to-severe diarrhea ( MSD ) in children in rural western Kenya , and to identify the major zoonotic enteric pathogens present in domestic animals residing in the homesteads of case and control children . We characterized animal-related exposures in a subset of case and control children ( n = 73 pairs matched on age , sex and location ) with reported animal presence at home enrolled in the Global Enteric Multicenter Study in western Kenya , and analysed these for an association with MSD . We identified potentially zoonotic enteric pathogens in pooled fecal specimens collected from domestic animals resident at children’s homesteads . Variables that were associated with decreased risk of MSD were washing hands after animal contact ( matched odds ratio [MOR] = 0 . 2; 95% CI 0 . 08–0 . 7 ) , and presence of adult sheep that were not confined in a pen overnight ( MOR = 0 . 1; 0 . 02–0 . 5 ) . Variables that were associated with increased risk of MSD were increasing number of sheep owned ( MOR = 1 . 2; 1 . 0–1 . 5 ) , frequent observation of fresh rodent excreta ( feces/urine ) outside the house ( MOR = 7 . 5; 1 . 5–37 . 2 ) , and participation of the child in providing water to chickens ( MOR = 3 . 8; 1 . 2–12 . 2 ) . Of 691 pooled specimens collected from 2 , 174 domestic animals , 159 pools ( 23% ) tested positive for one or more potentially zoonotic enteric pathogens ( Campylobacter jejuni , C . coli , non-typhoidal Salmonella , diarrheagenic E . coli , Giardia , Cryptosporidium , or rotavirus ) . We did not find any association between the presence of particular pathogens in household animals , and MSD in children . Public health agencies should continue to promote frequent hand washing , including after animal contact , to reduce the risk of MSD . Future studies should address specific causal relations of MSD with sheep and chicken husbandry practices , and with the presence of rodents .
Diarrheal disease remains among the leading causes of global mortality in children younger than 5 years [1 , 2] . Although the mortality rate due to diarrheal disease in this age group in Africa has decreased by nearly 4% per year since 2000 , it remains unacceptably high: it is estimated that 12% of deaths in children younger than five years in Africa are due to diarrhea , amounting to almost half a million childhood deaths annually [2] . While mortality rates have decreased , the incidence of diarrheal disease in young children in low- and middle-income countries has shown little change , from 3 . 4 episodes/child year in 1990 to 2 . 9 episodes/child year in 2010 [3] . Persistently high incidence rates in these countries are concerning because early childhood diarrhea may have long-term effects on child growth and development [4 , 5] . Data characterising risk factors and etiologies of diarrheal disease in children in these settings are important for focusing interventions to decrease associated morbidity and mortality rates . Many viral , bacterial and protozoal pathogens have been demonstrated as causes of diarrheal disease in children younger than 5 years in developing countries [6] . Contact with domestic animals , including livestock , poultry and companion animals , has been shown to play a role in the epidemiology and transmission to people of a number of these pathogens [7 , 8] including Campylobacter spp . [9–11] , non-typhoidal Salmonella [11 , 12] , diarrheagenic Escherichia coli strains [12 , 13] , Cryptosporidium spp . [12–14] and Giardia duodenalis [15] . In addition , some reports implicate dogs as a possible source of human infections with unusual strains of rotavirus [16 , 17] . Livestock and poultry play a significant role in rural livelihoods in developing countries , providing a variety of benefits to poor households , such as animal-source food ( energy-dense food with high biological-value protein , rich in micronutrients ) , draft power for ploughing and transport , nutrient recycling through manure , income through sale of animals or their products , as well as a form of savings and insurance [18]; however , animal husbandry may also have negative impacts on households , including the transmission of zoonotic and foodborne diseases . In a meta-analysis of demographic health survey data from 30 sub-Saharan African countries examining associations between child health outcomes and household ownership of livestock , Kaur et al [19] found a negative association between livestock and stunting ( an indicator of chronic malnutrition ) , a positive association between livestock and all-cause mortality in children , and no association between livestock and diarrheal illness . In a systematic review and meta-analysis of human diarrhea infections associated with domestic exposure to food-producing animals , Zambrano et al . [20] found consistent evidence of a positive association between exposure and diarrheal illness in people , across a range of animal species and enteric pathogens . Close contact with domestic animals ( such as animals sleeping in the house or room ) is also associated with impaired growth in children [21 , 22] . Considering the potential positive benefits of animal husbandry to rural livelihoods in resource-poor settings , there is a need to identify specific husbandry-related practices associated with diarrheal illness . Such evidence can serve as bases for interventions to reduce transmission of enteric pathogens to household members , especially to children , who are particularly vulnerable to mortality , sequelae and developmental consequences of diarrheal disease . Identifying etiologies of diarrheal illness in household members and concurrent infections in domestic animals may provide further utility for these efforts [23–25] . The Global Enteric Multicenter Study ( GEMS ) , a large-scale case-control study designed to identify the etiology and population-based burden of diarrheal disease in children younger than 5 years in developing countries [6 , 26] , provided an opportunity to study the association between animal-related exposures and diarrheal illness in household children at a rural site in western Kenya . GEMS was a 3-year , prospective , age-stratified , matched case-control study of moderate-to-severe diarrheal illness in children aged 0–59 months , residing in populations under demographic surveillance at four sites in sub-Saharan Africa and three sites in south Asia . The methodology [26–28] and main findings [29] of GEMS have been published . The GEMS Zoonotic Enteric Diseases ( GEMS-ZED ) sub-study was conducted among a subset of case children and their matched controls enrolled at one of the six GEMS sentinel health centers in rural western Kenya . The objectives of the GEMS-ZED study were to identify animal-related exposures associated with cases of moderate-to-severe diarrhea ( MSD ) in children , and to identify the major zoonotic enteric pathogens present in the domestic animals residing in the homesteads of case and control children .
The GEMS sentinel health center for this study was St Elizabeth Mission Hospital in Lwak ( henceforth referred to as Lwak Hospital ) , located in Rarieda sub-county , Siaya County ( formerly Nyanza Province ) in western Kenya . Lwak Hospital is the designated referral facility for population-based infectious disease surveillance ( PBIDS ) conducted in the surrounding 33 villages by the Kenya Medical Research Institute ( KEMRI ) and the U . S . Centers for Disease Control and Prevention ( CDC ) [30] . The area also falls within the KEMRI/CDC health and demographic surveillance system ( HDSS ) site in western Kenya [31] . The HDSS provides general demographic and health information including population age-structure , migration , fertility rates , birth and death rates , verbal autopsy , access and utilization of health care for approximately 220 , 000 inhabitants in 55 , 000 households . The primary economic livelihood is subsistence farming and fishing , and an estimated 70% of the population lived below the poverty line in 2003 [32] . The area is culturally homogeneous , with 95% of people being ethnically Luo [33] . Households in the PBIDS villages are clustered into compounds composed of related family units , with most compounds having between one and five family units [33] . Animal husbandry is common: 89% of compounds own at least one species of livestock or poultry , with 86% owning poultry ( median flock size: 10 ) , 49% cattle ( median herd size: 4 ) , 48% goats ( median herd size: 4 ) and 18% sheep ( median herd size: 3 ) ( KEMRI/CDC HDSS data for 2008 ) . Among compounds that own livestock , approximately one-half also own cats and/or dogs ( International Emerging Infections Program–Zoonoses Project data for 2009 ) . Rodents , including black rats ( Rattus rattus ) , are also commonly present in and around houses in the PBIDS site [34] . From January 31 , 2008 through January 29 , 2011 , children 0–59 months old who sought care at selected sentinel health centers ( including Lwak Hospital ) and belonged to the HDSS population were screened for diarrhea . To be eligible for inclusion in GEMS , the diarrhea episode had to meet the case definition for MSD [29] , which was three or more loose stools within the previous 24 h , with onset within the previous 7 days after a period of at least 7 diarrhea-free days , with one or more of the following: sunken eyes; loss of skin turgor; intravenous rehydration administered or prescribed; dysentery; or hospitalized with diarrhea or dysentery . Each GEMS site restricted enrollment to the first nine eligible cases per age stratum per fortnight . Three age strata were targeted: infants ( 0–11 months ) , toddlers ( 12–23 months ) , and children ( 24–59 months ) . For every enrolled case , one to three children without diarrhea were enrolled as controls . Controls were matched to individual cases by age ( within 2 months of age for patients aged 0–23 months , and within 4 months of age for patients aged 24–59 months ) , sex , and residence ( same or nearby village as patient ) . Potential controls were randomly selected from the KEMRI/CDC HDSS database and enrolled during a home visit within 14 days of the matched case . Potential controls who had diarrhea in the previous 7 days were ineligible . At enrollment , primary caregivers ( parent or other caretaker ) of cases and controls were interviewed to obtain demographic , epidemiological and clinical information . In addition , each case and control provided at least 3 g of fresh stool , which was submitted to the laboratory for identification of enteric pathogens using standard methods as described by Panchalingam et al . [28] . The GEMS-ZED substudy collected and analysed additional data on animal-related factors from a subset of GEMS case and matched control children with reported animal presence at home . From November 4 , 2009 through February 4 , 2011 , all cases enrolled into GEMS at Lwak Hospital were screened for inclusion in the GEMS-ZED study . ( Enrollment into GEMS continued for a short period after the official end date of January 29 , 2011 , during which time 3 case-control pairs were enrolled into GEMS-ZED . Data from the GEMS study [laboratory test results and wealth index] are not available for these 3 pairs . ) Between zero and six cases per fortnight ( median of two ) were enrolled into GEMS at Lwak Hospital during the GEMS-ZED study period . Only cases and controls whose primary caregiver reported presence of animals ( domestic animals as well as peridomestic wild rodents ) at the child’s compound during the GEMS enrollment interview were considered eligible . For each eligible case , the first eligible GEMS-enrolled matched control was identified , resulting in one-to-one matching in the GEMS-ZED dataset . If no eligible child could be identified among the GEMS set of one to three matched controls , then the case was not enrolled into GEMS-ZED . Caregivers of eligible cases and controls were approached for enrollment into the GEMS-ZED study during a separate home visit that took place within 2 weeks of their enrollment into the GEMS study . Written informed consent for participation in the study was sought from the primary caregiver , as well as from the head of the compound of residence of each eligible child; only compounds in which both individuals provided consent were enrolled . Compounds were excluded if the child participating in GEMS had died subsequent to enrollment , or if no domestic animals were found to be resident ( for example , if animals had died or were sold subsequent to GEMS enrollment ) . Following enrollment , both the head of the compound and the child’s caregiver were interviewed using a standard questionnaire . The questionnaire consisted of two parts: the first part dealt with residence and husbandry of domestic animals in the compound ( livestock , poultry , dogs and cats ) , as well as observations relating to the presence of rodents in and around the compound , and was asked of the person in the compound responsible for the management of animals ( typically the head of the compound ) . The second part dealt with information specific to the participating child , relating to exposures to animals and their environment , and was asked of the child’s caregiver . A summary of the items included in the questionnaire is presented in S1 Table . At the enrollment visit , fecal specimens were collected from a convenience sample of domestic animals resident at the compound . Specimens from a single species and age category ( young , unweaned animals vs . older animals ) were pooled together , with specimens from a maximum of five animals collected in a single pool , and a maximum of two pools per species and age category combination ( i . e . a maximum of ten animals per species and age category combination were sampled from a compound ) . A previous study showed good agreement of bacterial culture results between individual and pooled fecal samples of five individuals per pool [35] . Between 3 and 10 g of feces were collected directly from the rectum of larger animals ( cattle , sheep , goats and adult dogs ) . For smaller animals ( cats and young dogs ) , three moistened cotton-tipped swabs were used to collect samples from the animal’s rectum and placed directly into transport media ( two in modified Cary Blair and one in buffered glycerol saline ) ; whole feces were not routinely collected from smaller animals . For poultry , groups of birds of a single species ( chickens or ducks ) were confined overnight on a sheet of thick plastic . Owners were asked to confine approximately five birds per group , and not more than two groups of birds per species . Fecal specimens from a single pool of animals were mixed in a stool cup . Following thorough mixing of the pooled feces , two cotton-tipped swabs were inserted into the feces and then placed in a vial containing modified Cary Blair transport medium . A third swab was placed in a vial containing buffered glycerol saline . All specimen containers were clearly labelled and placed in a sealed bag in a coolbox with icepacks for transport to the laboratory . Identification of potentially zoonotic enteric pathogens in animal specimens ( Campylobacter jejuni , Campylobacter coli , non-typhoidal Salmonella , diarrheagenic E . coli , Cryptosporidium , Giardia , and rotavirus ) was carried out using an identical protocol to that described for the human stool specimens tested in GEMS [28] . Briefly , bacterial agents were isolated and identified using conventional culture techniques . Three putative Escherichia coli colonies of different morphology types were pooled and analysed by multiplex PCR that detect targets for enterotoxigenic ( ETEC ) , enteroaggregative ( EAEC ) , enteropathogenic ( EPEC ) , and enterohaemorrhagic E . coli ( EHEC ) . The following gene targets defined each E . coli pathotype: ETEC ( either eltB for heat-labile toxin [LT] , estA for heat-stable toxin [ST] , or both ) , ST-ETEC ( either eltB and estA , or estA only ) , typical EPEC ( bfpA with or without eae ) , atypical EPEC ( eae without either bfpA , stx1 , or stx2 ) , EAEC ( aatA , aaiC , or both ) , and EHEC ( eae with stx1 , stx2 , or both , and without bfpA ) . Commercial immunoassays were used to detect rotavirus ( ProSpecT Rotavirus kit , Oxoid , Basingstoke , UK ) , Giardia and Cryptosporidium spp . ( TechLab , Inc . , Blacksburg , VA , USA ) . Immunoassays were only performed on whole fecal specimens of adequate volume ( ≥ 3 g ) , and were therefore not completed for the majority of cat specimens , because volumes from this species were often inadequate . To better understand the zoonotic potential , we genotyped Cryptosporidium parasites from immunoassay-positive animal fecal specimens . DNA was extracted from 0 . 5 ml of fecal specimens using a FastDNA SPIN Kit for Soil ( MP Biomedicals , Santa Ana , CA ) . Cryptosporidium species present were differentiated by PCR-restriction fragment length polymorphism ( RFLP ) analysis of the small subunit ( SSU ) rRNA gene , and confirmed by DNA sequencing of the PCR products [36] . Data were analysed using R statistical software version 3 . 1 . 3 [37] . We used conditional logistic regression ( clogit function applying the exact method in R package ‘survival’ [38] ) with one-to-one matching to identify animal-related exposures that were significantly associated with MSD . Exposure variables were screened for inclusion in the multivariable model using univariable conditional logistic regression . As part of the screening process , each exposure variable was evaluated for potential recoding . Husbandry-related variables for which values were conditional upon residence of the species in question were evaluated and recoded if this made biological sense . For example , the question “Do adult sheep enter the cooking area ? ’” was conditional on residence of adult sheep in the compound . If no adult sheep were resident , the response was recoded as “No–no adult sheep present” rather than a missing value , and compared against “No–adult sheep present but do not enter cooking area” and “Yes–adult sheep present and enter cooking area” . For these variables , the null state ( species not resident ) was taken as the reference level . Variables related to exposures of children to animals and their environments were kept as binary variables . For example , the question “Does the child play in an area of the compound where adult sheep defecate ? ” had one of two responses: ‘no’ if no adult sheep were resident in the compound or adult sheep were resident but the child did not play in the area where they defecated , and ‘yes’ if there were adult sheep resident and the child played in the area where they defecated . For categorical variables with four or more categories , we created new binary variables by combining categories based on frequencies . For example , the original four levels for frequency of observation of rodents or their excreta ( never , seldom , often or daily ) were dichotomised to never/seldom vs . often/daily . Both the original and new variables were tested in the univariable analysis . Continuous variables ( e . g . number of chickens owned ) were categorised into three categories [category 1: zero values; category 2: values greater than zero and less than or equal to the median value ( excluding zeros ) ; category 3: values greater than the median value ( excluding zeros ) ]; both the original continuous variable and the new categorical variable were assessed in the univariable analysis . Variables with a significant number of missing values ( >10% of observations ) were discarded . Variables with a Wald test p-value greater than 0 . 2 on univariable analysis were excluded from further analyses . If both the original and recoded variable had a p-value below the threshold of 0 . 2 , the one with the smaller p-value was retained . After the univariable screening , we assessed collinearity between the selected exposure variables using condition indices ( colldiag function in R package ‘perturb’ [39] ) . A condition index is a number ranging from 1 to infinity that is computed from data on a set of exposure variables–the higher the condition index , the greater the amount of collinearity [40] . The condition indices were investigated by calculating the variance decomposition proportion ( VDP ) for each condition index over 30 , beginning with the largest . Exposure variables with a VDP >0 . 5 were considered potentially collinear . In cases where it made biological sense to do so , collinear variables were combined to create a new categorical variable . For example , the collinear variables “Chicken manure used in farm” and “Chicken manure used in the compound” were combined to create a variable “Chicken manure used” . When this did not make biological sense , or when the new variable still exhibited collinearity , the collinear variable with the higher univariable p-value was excluded . Remaining variables were taken forward for consideration in the multivariable conditional logistic regression model . We compared main effects models using Akaike’s information criterion ( AIC ) , whereby models with a smaller AIC are considered more optimal . We used a forward stepwise regression process to select exposure variables to retain in the model . Missing values were handled through multiple imputation ( R package ‘mice’ [41] ) . Building of the main effects model was stopped when the addition of a variable resulted in an increase in the AIC . We assessed interactions between variables in the final main effects model by adding two-way interaction terms to the model and evaluating their effect on the AIC . For evaluation of the final model , we identified outliers and influential pairs , using the transformation method described in [42] and applying a Bonferroni outlier test . We computed leverage values and delta β statistics to identify influential pairs ( in R package ‘car’ [43] ) . To determine if these pairs were having an undue effect on the model , we refit the model with them omitted . In GEMS , a wealth index quintile for households was generated by principle component analysis of thirteen household assets [26 , 44] . The wealth index quintile was forced into the final model as an ordinal variable to evaluate the potential confounding effect of wealth . The GEMS protocol was approved by the KEMRI Scientific and Ethical Review Committee ( protocol no . 1155 ) and the Institutional Review Board at the University of Maryland , School of Medicine , Baltimore , MD , USA . The Centers for Disease Control and Prevention , Atlanta , GA , USA , formally deferred to the IRB at the University of Maryland for review ( protocol no . 5038 ) . Written informed consent was obtained from the parent or primary caretaker of each participant before initiation of study activities . The GEMS-ZED study protocol was approved by the KEMRI Scientific and Ethics Review Unit ( protocol no . 1572 ) and the CDC Institutional Review Board ( protocol no . 5683 ) . Written informed consent for participation in the study was provided by the parent or primary caretaker of each participant , as well as from the head of the compound of residence of each participant . Protocols for animal involvement were reviewed and approved by the KEMRI and CDC Institutional Animal Care and Use Committees ( protocol no . SSC 1572 and 2088OREMULX , respectively ) . CDC IACUC protocols comply with the Animal Welfare Act ( AWA ) regulations promulgated by the United States Department of Agriculture ( USDA ) under Title 9 , Code of Federal Regulations , Parts 1–3 as well as the Public Health Service Policy on Humane Care and Use of Laboratory Animals ( PHS Policy ) administered by the National Institutes of Health ( NIH ) , Office of Laboratory Animal Welfare ( OLAW ) . In Kenya , all vertebrates are protected under Cap 360 ( the Prevention of Cruelty to Animals Act ) ( 1963 , revised 1983 ) .
We collected fecal specimens of acceptable quality for diagnostic testing from 2 , 174 domestic animals of eight species , resulting in a total of 691 pools ( median of 5 and range of 1 to 10 pools per compound ) . Of these , 159 pools ( 23% ) tested positive for one or more potentially zoonotic enteric pathogens ( Campylobacter jejuni , C . coli , non-typhoidal Salmonella , diarrheagenic E . coli , Giardia , Cryptosporidium , or rotavirus ) . Test results for particular pathogens by host species and age group are given in Table 3 . Species with the highest proportion of positive pools for particular pathogens were chickens for C . jejuni [18/231 ( 7 . 8% ) ] and non-typhoidal Salmonella [26/231 ( 11 . 3% ) ]; goats for C . coli [6/106 ( 5 . 7% ) ]; donkeys for diarrheagenic E . coli [1/12 ( 8 . 3% ) ]; dogs for Giardia [19/69 ( 27 . 5% ) ] and Cryptosporidium [4/69 ( 5 . 8% ) ]; and cattle for rotavirus [4/153 ( 2 . 6% ) ] . Domestic animals from 45/73 ( 61% ) compounds at which a child with MSD resided tested positive to one or more pathogens , compared with 44/73 ( 60% ) compounds with a control child . There were no significant associations on univariable conditional logistic regression between the presence of particular pathogens in domestic animals residing in compounds , and MSD in the participating child from the compound ( Table 4 ) . When considering the children’s GEMS laboratory results , we found 21 instances in which the pathogen identified in the child was also identified in one or more species of domestic animals residing in the compound ( Table 5 ) . Nineteen pooled specimens positive for Cryptosporidium spp . by immunoassay were analysed by PCR , including 14 pooled specimens from chickens , 4 from dogs , and 1 from calves . Among them , 7 chicken specimens and the bovine specimen generated the expected PCR products . RFLP analysis indicated the presence of C . meleagridis in 6 chicken specimens , C . bovis in one chicken specimen , and C . parvum in one bovine specimen . None of the canine specimens analysed were positive by PCR .
We identified several animal-related factors associated with MSD in children younger than 5 years from compounds in rural western Kenya in which one or more species of domestic animals were resident . Children who reportedly washed their hands after contact with animals had significantly lower odds of MSD . Water , sanitation , and hygiene ( WASH ) interventions , including hand washing promotion , are shown to significantly reduce the risks of diarrheal illness in less developed countries [45 , 46] , but their effectiveness in reducing pathogen exposure specifically from domestic animals in these settings has not been explored . While the protective effect of hand washing has been demonstrated in outbreaks of enteric diseases associated with exposure to domestic animals in public settings [12 , 13 , 47] , in their review Zambrano et al . [20] could find no studies that focused on WASH as a means of limiting disease transmission following domestic exposure to food-producing animals . Our study may be the first to report evidence of a protective effect of hand washing following exposure to household domestic animals in a developing country context . Hand washing after contact with animals may be a reflection of an overall higher frequency of hand washing in these children , and thus the protective effect may extend beyond ( or be unrelated to ) the risk of diarrheal illness after animal exposure . We recognise that a limitation of our study is reliance on self-reporting of behaviour , including hand washing . Children from compounds that reported frequent observation of fresh rodent excreta outside the house had significantly higher odds of MSD . In a previous study in the area , a number of rodents were trapped in compounds , including a high proportion of black rats [34] . Rodents , and particularly rats , can be infected with pathogens that cause diarrheal illness in humans [48] , including Salmonella Typhimurium [49 , 50] , Shiga-toxin producing E . coli [51] and Cryptosporidium parvum [52 , 53] . Fresh rodent feces in areas of the compound may therefore be a source of exposure of children to these pathogens . Absence of rodent excreta could also be a reflection of better sanitation in these compounds , which may be associated with decreased risk of MSD independent of rodents . Ownership and husbandry of sheep was found to be associated with MSD , but the nature of their role is not clear , with increasing numbers of sheep associated with increased odds , and not confining adult sheep in a pen overnight associated with decreased odds . Distance between children’s sleeping areas and where sheep are kept overnight may also play a role . Sheep are not a common livestock species in the study area , with only 18% of compounds owning sheep ( compared with 49% owning cattle and 48% owning goats ) . Evidence from the literature of a specific role for sheep as risk factors for diarrheal illness in children is scant [54–57] . Consumption of mutton was found to be a risk factor for gastrointestinal illness in children and young adults in Isiolo , eastern Kenya [58] . In our study , we found a low prevalence of potentially zoonotic enteric pathogens in sheep feces ( 0% - 5% ) , with the exception of Giardia ( 21% ) . Giardia infection in children was not associated with MSD in GEMS [29] . Participation of the child in providing water to chickens was identified as a risk factor for MSD . In our study , a relatively high proportion of chicken fecal pools were positive for non-typhoidal Salmonella ( 11 . 3% ) , Campylobacter jejuni ( 7 . 8% ) and diarrheagenic E . coli ( 7 . 6% ) . In their meta-analysis of six studies , Zambrano et al . [20] showed that poultry exposure more than doubled the odds of Campylobacter spp . infections in humans . Limiting exposure to household poultry , by for example corralling poultry , should therefore reduce the incidence of Campylobacter enteritis in children; however , in a randomized study to test this , Oberhelman et al . [59] found that rates of Campylobacter-related diarrhea were in fact significantly higher in children from households in which chickens were corralled , compared to those from households in which chickens were not confined . They speculated that this was due to the effect that corralling had on concentrating infected feces in one area , which would increase the risk of exposure to high doses of Campylobacter in children who entered corrals . Similarly , in our study we speculate that provision of water to chickens will be carried out mainly in situations where chickens are confined rather than free-ranging , increasing exposure of any accompanying children to enteric pathogens in the accumulated feces; however , we lack more detailed information on the nature of the reported exposure to substantiate this supposition . Active ingestion of chicken feces by infants has been observed in a rural African setting [60] , highlighting the risk of zoonotic transmission of enteric pathogens . In general , the prevalence of potentially zoonotic enteric pathogens in chicken feces in our study was lower than those reported in other studies in comparable settings [9 , 24 , 59 , 61 , 62] . Prevalence of zoonotic enteric pathogens in ruminants in our study was also lower when compared with other studies [24 , 25 , 61–65] . While this may be a reflection of differences in the diagnostic methods used , it could also be due to the extensive , subsistence nature of animal husbandry in our study site and the very small herd/flock sizes . We found no evidence of any association between the presence of particular pathogens in domestic animals and MSD in children , or of infection of children with the same pathogen species , although we note this was a pilot study with a small sample size , which may have limited our ability to detect associations . Enteric pathogens are often shed intermittently in the feces of carrier animals , so it is possible that carrier animals may not have been identified at the time of the specimen collection . The sensitivity of the microbiological methods used in children and in animals is low , as shown by a recent reanalysis of GEMS specimens using quantitative molecular diagnostic methods [66] . Even when the same pathogen species are found in children and in domestic animals in close contact , further characterization often shows genotypic differences between human and animal strains [24 , 67 , 68] , although in some instances further subtyping provides support for zoonotic transmission [69] . In our study , most Cryptosporidium species identified from chickens and calves are pathogenic in humans , but further subtyping of species in child and animal specimens is needed to better understand the role of zoonotic transmission in cryptosporidiosis epidemiology . We tested a large number of animal-related variables for an association with MSD in children . We recognise that with this many variables , significant associations may arise by chance , although the use of AIC in model selection should mitigate this . Furthermore , we do not infer a causal relation from the observed associations . We recommend that our results be used to generate hypotheses of causal links that can be tested in specific studies that address causal relations . These could include the role of sheep , chickens and rodents as risk factors for childhood diarrhea , and the application of WASH interventions to reduce risk . These studies should include established predictors of diarrhea in infants and young children , including breastfeeding and HIV status , in their causal models [70] . Future studies might further examine animal-related factors associated with environmental enteric dysfunction , as a number of zoonotic enteric pathogens have been found to be associated with this condition [71] . The use of quantitative molecular diagnostic methods in well-designed case-control and cohort studies of linked human and animal populations will also be important to understand the role of animals in domestic environments as reservoirs of human enteric pathogens . | Diarrheal disease is one of the leading causes of death worldwide in children younger than 5 years . Exposure to animals in homes may be a risk factor for diarrhea in children . To test this , we studied a subset of children in the Global Enteric Multicenter Study ( GEMS ) in rural western Kenya , whose caretakers reported the presence of animals in the children’s homesteads . In GEMS , children with moderate-to-severe diarrhea ( MSD ) were matched with children without MSD , who were of the same sex , similar age and who lived in the same area . We asked questions about the presence and management of animals in the children’s homesteads . We also collected fecal specimens from domestic animals present at homesteads and tested these for microbes that could cause diarrheal disease in children . We found that children who reportedly washed their hands after animal contact , and who lived in a homestead with adult sheep that were not confined to a pen overnight , had a lower risk of MSD . Children who lived in homesteads that owned more adult sheep , or in which fresh rodent droppings were observed frequently , had a higher risk of MSD , as did children who reportedly participated in providing water to chickens in the homestead . We did not find any association between the presence of particular pathogens in household animals , and MSD in children . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"animal",
"types",
"livestock",
"medicine",
"and",
"health",
"sciences",
"zoonotic",
"pathogens",
"animal",
"pathogens",
"pathology",
"and",
"laboratory",
"medicine",
"ruminants",
"pathogens",
"domestic",
"animals",
"vertebrates",
"animals",
"mammals",
"diarrhea",
"signs",
"and",
"symptoms",
"gastroenterology",
"and",
"hepatology",
"zoology",
"infectious",
"diseases",
"zoonoses",
"sheep",
"birds",
"gamefowl",
"fowl",
"poultry",
"agriculture",
"rodents",
"diagnostic",
"medicine",
"biology",
"and",
"life",
"sciences",
"chickens",
"amniotes",
"organisms"
]
| 2017 | Animal-related factors associated with moderate-to-severe diarrhea in children younger than five years in western Kenya: A matched case-control study |
The stress-induced mutagenesis hypothesis postulates that in response to stress , bacteria increase their genome-wide mutation rate , in turn increasing the chances that a descendant is able to better withstand the stress . This has implications for antibiotic treatment: exposure to subinhibitory doses of antibiotics has been reported to increase bacterial mutation rates and thus probably the rate at which resistance mutations appear and lead to treatment failure . More generally , the hypothesis posits that stress increases evolvability ( the ability of a population to generate adaptive genetic diversity ) and thus accelerates evolution . Measuring mutation rates under stress , however , is problematic , because existing methods assume there is no death . Yet subinhibitory stress levels may induce a substantial death rate . Death events need to be compensated by extra replication to reach a given population size , thus providing more opportunities to acquire mutations . We show that ignoring death leads to a systematic overestimation of mutation rates under stress . We developed a system based on plasmid segregation that allows us to measure death and division rates simultaneously in bacterial populations . Using this system , we found that a substantial death rate occurs at the tested subinhibitory concentrations previously reported to increase mutation rate . Taking this death rate into account lowers and sometimes removes the signal for stress-induced mutagenesis . Moreover , even when antibiotics increase mutation rate , we show that subinhibitory treatments do not increase genetic diversity and evolvability , again because of effects of the antibiotics on population dynamics . We conclude that antibiotic-induced mutagenesis is overestimated because of death and that understanding evolvability under stress requires accounting for the effects of stress on population dynamics as much as on mutation rate . Our goal here is dual: we show that population dynamics and , in particular , the numbers of cell divisions are crucial but neglected parameters in the evolvability of a population , and we provide experimental and computational tools and methods to study evolvability under stress , leading to a reassessment of the magnitude and significance of the stress-induced mutagenesis paradigm .
One of the most puzzling and controversial microbial evolution experiments of the 20th century may be the one performed by Cairns and colleagues [1 , 2] in which lac− cells are plated on lactose as the sole carbon source and therefore cannot grow . Revertants toward the lac+ genotype continuously appear after plating at a rate and timing seemingly incompatible with the Darwinian hypothesis of selection of preexisting mutants . In the lac− construct , the lacZ coding sequence is present but nonfunctional , because it is out of frame with the start codon . The lac+ revertants are thus frameshift mutants in which this coding sequence is back in frame with the start codon . Most of the controversy initially came from the question of whether these reversion mutations were Lamarckian , in the sense that they would arise at a higher rate when the cells would “sense” that these mutations would be beneficial [3] . However , many additional experiments quickly suggested that this phenomenon can be explained by more standard Darwinian mechanisms , in which genetic changes are not targeted but occur randomly and are then selected or not . While two seemingly conflicting molecular explanations—the stress-induced mutagenesis model and the gene amplification model—emerged , both are conceptually very similar . In both explanations , mutations occur randomly and independently of their effect on fitness , but the specific conditions of carbon starvation increase the rate at which genetic diversity is generated at the relevant locus ( lacI-lacZ sequence ) . In the stress-induced mutagenesis model [4] , the genome-wide mutation rate is increased as an effect of the stress response triggered by starvation . In the gene amplification model [5 , 6] , random duplications of the lacI-lacZ system happen before plating on lactose and are then selected in presence of lactose because the frameshift mutation is leaky . A small amount of Beta-galactosidase is still synthesized , permitting cryptic growth due to rare expression errors , which compensate the frameshift . This residual expression becomes higher with more copies of the leaky system . As the copy number of the system increases , a reversion mutation in lacI-lacZ becomes more likely because of increased target number . While it is still not clear whether stress-induced mutagenesis is the sole explanation of the phenomenon , the attempts to explain the data presented by Cairns and colleagues have led to a much better understanding of control over mutation rate in response to the environment . An increase in mutation rate under starvation has also been reported in other systems , such as nutrient-limited liquid cultures [7 , 8] and “aging” colonies on agar plates [9 , 10] . However , Wrande and colleagues [11] reported that the accumulation of mutants in aging colonies observed by Bjedov and colleagues [10] can be explained by growth under selection without elevated mutagenesis , because the mutation used by the original authors to infer mutation rate is beneficial in these specific environmental conditions . An emblematic molecular mechanism permitting this regulation of mutation rate is the SOS response ( suggested in 1970 [12 , 13] ) , in which DNA damage is sensed by bacterial cells and leads to the up-regulation of many genes , permitting mutagenic repair and replication of damaged DNA . While the responsible enzymes were unknown at the time , it has indeed been found subsequently that the SOS response increases the dosage of polIV and polV [14 , 15] . These error-prone polymerases are able to replicate damaged DNA that the classical DNA polymerase polIII could not replicate , albeit at the price of a higher error rate [16 , 17] . This strategy , favoring “survival at the price of the mutation , ” is only one side of the story . There is a line of evidence suggesting that this higher error rate is not only an unavoidable trade-off with survival . It is also supposed to be a selected property to increase mutation rate under stressful conditions , increasing the chances that one of the descendants obtains a beneficial mutation that makes it able to better withstand the stress [18 , 19] . The evolution of traits that increase mutation rate under stress needs be considered in the context of second-order selection [20] . Second-order selection relies on the idea that natural selection does not only act on the individual's phenotype and instant fitness but also on its ability to generate fit descendants , leading to selection of properties such as evolvability and mutational robustness [21] . In parallel to the study of environmental control over the mutation rate , genetic determinants of mutation rate have also been studied . It has been shown and is widely accepted that alleles increasing mutation rate ( for example , defective mismatch repair or DNA proofreading ) can be selected when hitchhiking with the beneficial mutations they permit to generate [22 , 23] . On the other hand , the possibility of selection of mechanisms increasing mutation rate under stress but not constitutively has been subject to a more philosophical debate [24 , 25] . While modeling shows such selection is possible [19] , it is hard to distinguish whether an observed increase in mutation rate under a specific stress is ( i ) an evolvability strategy; ( ii ) an unavoidable trade-off of selection for survival , such as replicating damaged DNA to avoid death at the price of making mutations; or ( iii ) a direct effect of the stress and not of the stress-response system [26] . This debate is important for a full understanding of the evolutionary relevance of the phenomenon but does not affect the medical implications concerning the risk of de novo evolution of resistance during antimicrobial treatment [27] . Here , we are interested in the general case of mutation rate in growing stressed populations , and we especially focus on antibiotic stress , although our findings may be valid for other biotic and abiotic stresses . It has been suggested that treatment with subinhibitory doses of antibiotics increases bacterial mutation rate , due to induction of various stress-response pathways [28–32] . Many molecular mechanisms underlying this stress response have been elucidated , including the SOS response [29] or the RpoS regulon [30] . Oxidative damage has also been suggested to play a role in antibiotic-induced mutagenesis [28] and death [33–35] . Although still controversial [36] , these findings link antibiotic stress to the older question of how bacteria deal with oxidative stress and how oxidative damage impacts mutation rates [37] . However , all the evidence for stress-induced mutagenesis relies on accurately measuring mutation rates of bacteria growing in stressful conditions , and comparing them to those of the same strains growing without stress . Computing such mutation rates under stress is harder than it may seem , because stress may change population dynamics and may thus invalidate the assumptions made by the mathematical models used to compute mutation rate . For example , in the case of subinhibitory concentrations of antibiotics in which net population growth is positive , death may nevertheless happen at a considerable rate . Death events , however , are not detected by standard microbiology methods and are not taken into account by the mathematical tools used to compute mutation rate [38–40] . Indeed , such tools take as inputs only the number of observed mutants at a chosen locus and the final population size , making the underlying assumption that there is no death and that population size is thus sufficient information to summarize growth dynamics . The final population size is used to infer the number of DNA divisions leading to the final observed population from a small initial inoculum . If there is death , more divisions are needed to reach this population size , thus giving more opportunities to acquire mutations . The mutation rate will then be overestimated , because the number of DNA replications will be underestimated . In this work , we developed an experimental system to compute death rates in populations growing under stress and a computational method to compute mutation rates from fluctuation assays under stress using the computed death rates . We applied this framework to re-estimate mutation rates of Escherichia coli MG1655 growing under sub- minimal inhibitory concentration ( MIC ) doses of kanamycin ( an aminoglycoside acting on protein synthesis [41] ) , norfloxacin ( a fluoroquinolone acting on the DNA-gyrase complex and potentially leading to the creation of DNA breaks through the cell machinery [42] ) , and hydrogen peroxide ( an oxidizing agent producing reactive oxygen species that directly affect DNA independently of the cell machinery [43 , 44] ) . All these antimicrobials have previously been reported to significantly elevate mutation rate [28 , 31] . We find the same pattern when computing mutation rate without taking death into account . However , for norfloxacin and kanamycin , the estimated increase of mutation rate due to treatment is strongly reduced after conservatively correcting for death . There remains no signal of stress-induced mutagenesis in the case of kanamycin . These findings confirm our suspicion that neglecting death leads to substantial overestimation of mutation rate under stress . We also show that mutation rate estimation does not only present experimental and mathematical challenges; it is also not the most relevant measure of evolvability , meaning the capacity of a population to generate adaptive genetic diversity . Indeed , some of the studied subinhibitory treatments cause a significant drop in population size due to both bactericidal and bacteriostatic effects and thus lead to a smaller genetic diversity despite a higher mutation rate . Ironically , evolvability , approximated as the generation of genetic diversity , can be much more easily estimated from experimental data than mutation rate . In our experiments , antibiotics and hydrogen peroxide have very different effects on evolvability: both subinhibitory norfloxacin and kanamycin treatments significantly reduce it , while hydrogen peroxide treatment strongly increases it .
Subinhibitory treatments are not necessarily sublethal , because minimal inhibitory concentration is defined at population scale . An antimicrobial treatment is subinhibitory if the population grows ( i . e . , colony-forming unit [CFU]/mL increases or , more crudely , culture tubes inoculated at low density are turbid after 24 h ) . However , the death rate can be high , as long as the division rate is higher . Such death events will not be visible to the observer if only population size ( CFU/mL ) is tracked over time ( Fig 1 ) . To reach a given observed final population size , the number of cell divisions has to be higher if there is death . This means that when computing mutation rate using the classical approach ( described in the Materials and methods ) , the number of cell divisions will be underestimated . This is because it is implicitly assumed that there is no death and thus that the final population size is a good approximation for the number of cell divisions . The mutation rate , computed as the number of mutational events divided by the number of cell divisions , will then be systematically overestimated . The above statement—that mutation rates are systematically overestimated when there is death—is the first intuition motivating our work . We explore this intuition more rigorously ( Fig 2 ) using a simulation approach . For an arbitrary chosen value of mutation rate toward a neutral arbitrary phenotype , we simulate the growth of a population of bacteria inoculated from a small number of nonmutant cells with a chosen constant death rate and track the number of mutant and nonmutant cells . We then compute the mutation rate based on the final state of these simulations , using the standard approach ( i . e . , the fluctuation test as described in the Materials and methods ) to test whether we recover the true value of the mutation rate . As shown in Fig 2 , the mutation rate is systematically overestimated when there is death , and the higher the death rate , the higher the overestimation . This result is unchanged when varying other population growth parameters , such as the initial and final population sizes , the mutation rate , and the plating fraction ( underlying data have been uploaded to Zenodo 10 . 5281/zenodo . 1211765 ) . In the previous section , we show that it is necessary to take death into account when computing mutation rate . For this , tracking population size ( and thus net growth rate ) during antibiotic treatment , as classically done by plating and counting colony forming units , is not sufficient . It is not possible to know whether a decreased net growth rate in the treatment compared to the untreated control is due to a purely bacteriostatic effect ( i . e . , the population grows more slowly , but without death ) or to a bactericidal effect ( i . e . , the bacteria keep dividing , potentially at the same rate as without antibiotic , but also die ) . The first scenario will have no effect on the accumulation of mutants as a function of population size , while in the second scenario , turnover implies a higher number of DNA replications and thus more mutants for a given population size , as explained above . To disentangle these two effects , we designed a method allowing us to compute growth rate and death rate simultaneously using a segregative plasmid . The segregation dynamic allows us to estimate the number of bacterial cell divisions . Combining this information with the change in population size allows us to estimate growth rate and death rate , as explained in the Materials and methods . Our ultimate goal is to reliably estimate mutation rates of bacteria treated with subinhibitory doses of antimicrobials . To this end , we quantify population dynamics and compute mutation rates toward a chosen neutral phenotype ( resistance to rifampicin , conferred by substitutions in the gene rpoB ) in populations exposed to subinhibitory doses of other antimicrobials . Our mutagenesis protocol is inspired by the standard fluctuation test with additional measurements of plasmid segregation to compute death rate , as detailed in the Materials and methods . The population dynamics are quantified as a combination of two variables: CFU at various time points ( e . g . , 0 h , 3 h , 6 h , and 24 h after treatment starts ) and relative death rate ( compared to birth rate ) between pairs of two successive time points . We represent these population dynamics in Fig 3 for the chosen subinhibitory antimicrobial treatments . We use kanamycin at 3 ug/mL , norfloxacin at 50 ng/mL , and hydrogen peroxide ( H2O2 ) at 1 mM , allowing direct quantitative comparisons with the data from Kohanski and colleagues [28] . We also include untreated populations , in which we assume there is no death , to fit the plasmid segregation parameters ( see Materials and methods ) . We find that for norfloxacin , there is a strong death rate in all phases of growth and a strong impact of the treatment on final population size . For H2O2 , death is only detectable in stationary phase and the treatment is mostly bacteriostatic during growth . For kanamycin , the dynamics are more complex , because an initially high death rate leads to a strong decline of population size during the first 6 h of growth , followed by a recovery leading to a final population size close to the one reached in untreated controls . During this second phase of growth following the bottleneck at 6 h , death rate is still substantial . This clearly shows that none of the three studied treatments are fully sublethal and thus that the implicit assumption of no death made when using the standard methods of computation of mutation rate ( as done by Kohanski and colleagues [28] ) does not apply . We developed computational tools to quantify mutation rate , taking into account the measured population dynamics and accounting for death . Our software , ATREYU ( Approximate bayesian computing Tentative mutation Rate Estimator that You could Use ) , is described in the Materials and methods . It takes as input any arbitrary population dynamics , described as a list of population sizes ( i . e . , CFU/mL for several time points ) and an associated list of death rates between pairs of consecutive time points . This input is thus exactly what is shown in Fig 3 . We apply this method to analyze the results of our mutagenesis protocol , to quantify whether and by how much subinhibitory treatments with kanamycin , norfloxacin , or H2O2 increase mutation rate . We show the effect of treatment on mutation rate in Fig 4 . We also plot the uncorrected mutation rate estimate , assuming no death as would be obtained by methods such as FALCOR ( Fluctuation AnaLysis CalculatOR ) [39] , bzRates [40] , or rSalvador [38] . Clearly , not taking death into account leads to a strong overestimation of the mutation rate for both kanamycin and norfloxacin . In the case of kanamycin , correctly computing the mutation rate removes all signal for stress-induced mutagenesis . In the case of norfloxacin , this signal is strongly lowered , from a 14-fold to a 6-fold increase . For H2O2 , the signal is less affected , which can be attributed to death rate only being significant in stationary phase . This confirms that neglecting death leads to a systematic overestimation of mutation rates and that taking into account the full population dynamics is necessary and leads to significantly different patterns depending on the antimicrobial and its effect on growth and death . The quantification of mutation rate in different conditions is not sufficient to answer the question of whether subinhibitory antibiotic treatments increase evolvability in general and , in particular , increase the likelihood of emergence of a resistant mutant and thus the probability of treatment failure . Indeed , mutation rate is expressed per DNA division , but , as we have shown in the previous section , antibiotic treatment may significantly change the number of susceptible cells and the number of replications that these cells have undergone . Intuitively , if a treatment multiplies mutation rate by 10 but divides population size by 100 , it is not likely to lead to an increased genetic diversity . This intuition has also been given by Couce and Blázquez ( Fig 2 of [45] ) but has been largely ignored in the literature as it was not the main message of this review . Conversely , a treatment that does not affect mutation rate and only slightly affects carrying capacity but causes death and turnover may result in a significantly increased genetic diversity . We first show the effect of subinhibitory treatment on final population size in Fig 5 . While H2O2 does not affect final population size , there is a strong effect of 1–2 orders of magnitude for norfloxacin and a significant but smaller effect of around 50% reduction for kanamycin . This supports our intuition that at least for norfloxacin , the few-fold increase in mutation rate we report ( Fig 4 ) is probably uncorrelated to any increase in genetic diversity . We expect the generation of genetic diversity to depend on ( i ) the number of cells alive , ( ii ) the population dynamics of these cells , and ( iii ) their mutation rate . Addressing the effect of stress on mutation rate as done in the previous section is necessary for a proper understanding of the bacterial stress response and of DNA repair mechanisms . Nevertheless , mutation rate is not the relevant measure to understand the effect of stress on the generation of genetic diversity and thus on evolvability . As a simple quantification of the generation of genetic diversity and thus an approximation of evolvability , we measure the number of mutants at a neutral locus , here the base-pair substitutions conferring resistance to rifampicin in the gene rpoB . We plot in Fig 6 the absolute number of rifampicin-resistant mutants in the final population for all treatments and for untreated control . Evolvability is reduced by a few-fold by kanamycin treatment ( as expected , since this treatment decreases population size without increasing mutation rate ) . While norfloxacin and H2O2 both induce a small increase in mutation rate , they interestingly have strongly opposite effects on evolvability . Treatment with H2O2 increases evolvability by more than one order of magnitude , while treatment with norfloxacin reduces it by a similar amount . This is due to the very different effects these antimicrobials have on population dynamics: while H2O2 does not affect final population size , norfloxacin causes a strong decrease in population size due to both bactericidal and bacteriostatic effects . So independently of the question whether antibiotics increase mutation rate , we show that the sub-MIC treatments we studied do not in any way increase evolvability . Thus , the standard rationale , that these subinhibitory treatments would increase the risk of emergence of resistance and treatment failure because of a higher generation of genetic diversity [46] , does not hold . This effect is largely due to a strong reduction in population size , which implies a loss in genetic diversity . Population size and mutation rate are not , however , the only factors affecting evolvability . We may also ask how much the measured turnover in our experiments contributes to evolvability . To answer this question , we simulate the same population dynamics as observed for each treatment but without death: Each population reaches the same final population size as measured in our experiments , with the same mutation rate as computed , but with no death . This is similar to what would happen if the antibiotics only had a bacteriostatic effect . For each simulation , we quantify evolvability using the same measure as previously , i . e . , the absolute number of mutants for our phenotype of interest in the final population . We compare this simulated evolvability without turnover with the actual measured evolvability in Fig 7 . For kanamycin and norfloxacin , turnover significantly increases evolvability by a few-fold . In summary , our results show that ( 1 ) mutation rate is systematically overestimated in subinhibitory treatments because of death , ( 2 ) mutation rate is not the only parameter that controls the generation of genetic diversity or evolvability , ( 3 ) population size and turnover play a key role in evolvability , and ( 4 ) treatment with subinhibitory doses of norfloxacin or kanamycin significantly decreases evolvability , measured as the generation of genetic diversity at population scale . These results are in apparent disagreement with the conclusions of previous studies on antibiotic-induced mutagenesis . This discrepancy is due to both miscalculation of mutation rates ( which occurs when one neglects the effect of population dynamics ) and misconceptions about the link between mutation rate and evolvability in these classical papers .
Understanding genetic and environmental control of evolvability is central for the understanding of microbial adaptation to constantly changing environments . Evolvability is defined as the capacity of a population to generate adaptive genetic diversity . This can be decomposed in two variables: the amount of genetic diversity generated by a population ( often inaccurately attributed to the mutation or recombination rate only ) and the fraction of this diversity that is adaptive . We are here interested in the former . Genetic control over the amount of generated genetic diversity has been studied for a long time in the field of mutation rate evolution [20 , 22] . The existence of constitutive mutator alleles in bacteria has been discovered before the mechanisms of DNA replication [47] , and the selection pressures leading to their transient increase in frequency have been elucidated through both theoretical and experimental studies [23 , 48] . Observing the evolution and fixation of such mutator alleles from nonmutator lineages in a long-term evolution experiment [49] plausibly facilitated the acceptance of these theories . On the other hand , plastic , environment-dependent control over the generation of genetic diversity has been a controversial paradigm shift in bacterial evolution [18] . It has been proposed for a long time that various stresses can increase mutation rates in bacteria [9 , 10] , including those triggered by antimicrobial treatments [28–32] . Several molecular pathways have been shown to be implicated in this phenomenon , the emblematic one being the SOS response [12] . In this work , we have shown that the effect of stress on mutation rate can not be computed properly with the existing tools , because the underlying mathematical models make the assumption that there is no stress or , more precisely , that the stress does not affect population dynamics . We develop experimental and computational tools to measure population dynamics and compute mutation rates under stress and apply them to the question of mutagenesis due to antibiotic treatment . We have shown that the intuition that low doses of antibiotics are dangerous because they lead to a higher generation of diversity is based on a misinterpretation of valid experimental data for two reasons: ( 1 ) the increase in mutation rate is overestimated due to overly simplistic assumptions , and ( 2 ) a higher mutation rate does not lead to a higher genetic diversity if population dynamics are affected ( e . g . , if population size is reduced ) . The question of emergence of resistance alleles due to low doses of antibiotics ( reviewed by Andersson and Hughes [50] ) cannot , however , be entirely addressed by measuring the generation of genetic diversity . The study of adaptive evolution can be decomposed in two parts: generation of diversity and natural selection acting on this diversity . While we have shown that treatment with a subinhibitory dose of norfloxacin does not increase but rather strongly decreases the amount of generated genetic diversity , it has also been reported that resistance alleles can be maintained and enriched by selection , even at very low antibiotic concentration [51] . Such selection of preexistent alleles may be a much more valid reason for concern about subinhibitory treatments . However , the literature is not as unanimous regarding bacteria residing within a patient with an immune system , rather than in a test tube [52] . It has , for example , been suggested that treating with a lower dose of antibiotics could slow down the selection of existing resistance alleles by decreasing their fitness advantage compared to the sensitive , wild-type strain , without compromising the success of the treatment [53 , 54] . Combining our results with these papers calls for a reevaluation of the evolution of antibiotic resistance at low doses of antibiotics . The question of the potentially adverse effects of low doses of antibiotics has been of longstanding interest in the medical community , as is evidenced by the famous quote from Alexander Fleming's Nobel lecture [55] , “If you use penicillin , use enough . ” However , given the time of this research ( penicillin was discovered in 1928 and thus 15 years before Luria and Delbrück ) , one should not be surprised that this often cited out-of-context advice relies on a rather Lamarckian reasoning in terms of educating rather than selecting for resistance: Our findings are also relevant outside of the context of evolution during antibiotic treatment . As we mentioned , mutagenesis in bacteria under nutritional stress was a key development in the understanding of the bacterial stress response and DNA repair , with a recent regain of interest [7 , 8] . Our experimental system can a priori not be applied to study starving bacteria , for two reasons: ( 1 ) our plasmid segregation method only gives sufficient signal in nonstationary populations , and ( 2 ) many of the observations on starvation-induced mutagenesis are dependent on the presence of some spatial structure ( for example , bacterial colonies on agar plates [2 , 9 , 10] ) . In this second case , the population dynamics become much more complex and are unlikely to be realistically approximated by a single relative death rate parameter . But the exact same questions remain to be elucidated in this field: How many cell divisions happen in these starving colonies ? In batch cultures , is stationary phase really stationary , or is there some turnover and recycling as recently suggested [56] ? And more importantly , where does death come from: is it an unavoidable , externally caused phenomenon; or is there an internal component , such as an altruistic programmed cell death [57] , or just traits selected in other environments that give a maladaptation to certain stresses [58] ? Stress-induced mutagenesis is of interest for several research fields , with different questions . We showed that the relevant question in a clinical setting is not directly about mutation rate but about evolvability and that the link between both is confunded by the effect of treatment on population dynamics . One the evolutionary side , the central question about stress-induced mutagenesis is “Is it adaptive ? ” Studying the molecular mechanisms of stress response will shed light on one part of the answer: is the increase in mutation rate controlled by the cell , or is it an unavoidable consequence of the stress ? In this regard , the three antimicrobials we study seemingly have very different properties . H2O2 is creating reactive oxygen species that directly damage DNA independently of the cell machinery , iron being the only necessary catalyst [43 , 44] . The way DNA damage leads to mutations is controlled by the cell but is more likely to be a consequence of selection for survival ( “survival at the price of the mutation” ) , rather than selection for evolvability . On the other hand , kanamycin acts on protein synthesis [41] , and any hypothetical mutagenic effect would thus go through the cell machinery . Norfloxacin is in between , because it acts on the DNA-gyrase complex , leading to an arrest of DNA synthesis and , in some conditions , to double strand breaks [42] . Recent findings from J . Collins and colleagues , however , suggest that these different scenarios are not as distant as they may seem , because they suggest that the production of reactive oxygen species is a feature of all bactericidal antibiotics [33–35] . While supporting the idea that antibiotic treatment increases mutation rate and does so in correlation with bactericidal activity , these debated findings would also suggest that such increase in mutation rate does not stem from selection for evolvability . In a nutritional stress scenario , Maharjan and colleagues [8] show that at equal effect on growth rate , limitation of different nutrients has very different effects not only on mutation rate but also on mutational spectrum , again showing the need for a mechanistic understanding of the molecular details and suggesting that the evolutionary outcome is much more complex than a linear increase in mutation rate in response to starvation . We provide tools that may help further developments of these questions . Our software , ATREYU , can be used to compute mutation rates from mutant counts in populations with arbitrary but known birth and death dynamics . The mutant counts are obtained by a protocol similar to the classical fluctuation test . The birth and death dynamics can be obtained by several methods . We used plasmid segregation , but other methods may be possible , such as segregation of engineered self-assembling fluorescent particles [59] , isogenic strain tagging [60] , Carboxyfluorescein succinimidyl ester ( CFSE ) membrane staining [61] , or direct microscopic observations at single-cell resolution . Microscopic observations with cell tracking may give much more precise and less noisy information than other methods but are only suitable when the death rate is sufficiently low , because only a limited number of cells can be tracked . We believe that death and cell turnover are crucial factors in evolutionary microbiology but are often neglected , in part due to the lack of standard methods to measure them . In immunology , in which the population dynamic of lymphocytes has been recognized as a central question [62] , many methods have been developed [63] , including the aforementioned CFSE membrane staining . While our method could be adapted for many nonstandard assumptions other than death ( e . g . , fitness cost of the mutation or partial plating ) , it shares some of the limitations of the more classical systems . Firstly , we suppose that each cell is fully monoploid and has only one chromosome and thus that the number of DNA replications is the number of cell divisions . However , some quinolone antibiotics are known to cause filamentation [64] , increasing the number of chromosomes per cell [42] and potentially changing the evolutionary dynamics [65] . Further complicating the picture , recent work [66] shows that even within a single chromosome , multifork replication may cause different ploidy levels on different loci , affecting mutation rate estimates and evolutionary dynamics . Secondly , we also consider that time does not matter , in the sense that the probability of mutation per division is independent of the growth rate , and that nondividing bacteria do not accumulate mutations , justifying the expression of mutation rate as a quantity of mutations per division event and not per unit of time . Since Luria’s and Delbrück’s experiment , this has been the standard assumption both on the microbiological and mathematical side [67] . However , recent data on fission yeasts suggest that nonreplicating cells may accumulate mutations at a different rate and spectrum than diving cells [68] . Finally , we make the assumption of homogenous behavior in the population , excluding the possibility that different subpopulations have different death and mutation rates . The question of whether a small subpopulation in a different physiological state may contribute most of the mutational supply is still unresolved . Theoretical work [69] shows that such situation could have a large impact on the evolutionary dynamics . Zooming out from evolutionary microbiology , mutagenesis research in bacteria shows an interesting parallel with recent advances in cancer research . For a given cell growth dynamic ( organogenesis , from stem cells to an organized population of differentiated cells ) , a higher mutation rate ( expressed per cell division ) will boost the accumulation of mutations and thus the risks of cancer . This increase in mutation rate can be genetic , such as in the case of hereditary nonpolyposis colon cancer caused by a deficiency of mismatch repair [70] , or environmental , such as exposure to carcinogenic compounds [71–73] . All of this is now part of textbook science on cancer and is similar to an increase in mutation rate in a bacterial population due to genetic ( mutator alleles [47] ) or environmental ( stress-induced [18] or stress-associated [26] mutagenesis ) factors . Tomasetti and Vogelstein [74] recently reported that the number of stem cell divisions is a strong predictor of cancer risk per organ . This is in parallel with our findings , which show that the number of cell divisions is central to predict the generated genetic diversity in a population of cells . Tomasetti and Vogelstein caused a major controversy by concluding that cancers would thus mostly be due to “bad luck” ( i . e . , unavoidable consequences of the large number of cell divisions ) rather than to environmental factors ( e . g . , exposure to mutagenic chemicals ) . We show here that the generation of genetic diversity depends on both mutation rate and cell population dynamic , which is in line with many studies that have criticized the interpretation of the data made by Tomasetti and Vogelstein . The challenge of understanding evolvability in bacterial population is thus strikingly similar to the one of understanding cancer , in the sense that the outcome depends on a complex interplay of extrinsic and intrinsic factors acting at different scales . In the case of bacteria , additional complexity stems from the fact that the same treatments may both impact the number of cell divisions ( death and turnover ) and the mutagenicity of each division . The picture is further complicated by the difficulty of disentangling the direct effects of the drug from the effects of the stress response triggered by the drug . But fortunately , while separating and measuring each factor requires complex experimental methods and mathematical tools , measuring evolvability on neutral loci is simpler , at least in bacteria . We hope that our study will encourage researchers in the field to question more not only the appropriateness of the tools they use for mutation rate estimation and the assumptions implicitely made by using these tools but also the pertinence of the variable they choose to report .
Our mutagenesis protocol is directly inspired by the one used by Kohanski and colleagues [28] ( which is in turn similar to that of Luria and Delbrück [75] ) with the inclusion of a segregative plasmid to compute death rate , as explained further below and graphically represented on Fig 8 . A culture of E . coli MG1655 ( with plasmid pAM34 ) is inoculated from a freezer stock and grown overnight in LB supplemented with 0 . 1 mM IPTG and 100 ug/mL of ampicillin ( to ensure maintenance of pAM34 ) . After the culture reaches stationary phase ( at least 15 h of growth ) , it is washed 3 times in normal saline ( 9 g/L NaCl ) to remove traces of IPTG and then diluted 10 , 000 times in a 500 mL baffled flask containing 50 mL of LB ( to maximize oxygenation ) . After 3 . 5 h of growth , the culture is inoculated at a ratio of 1:3 in 24 culture tubes containing a total volume of 1 mL of LB supplemented with one of the studied antimicrobials at subinhibitory concentration ( 3 ug/mL kanamycin , 50 ng/mL norfloxacin , 1 mM hydrogen peroxide , or untreated control ) . After 24 h of growth at 37°C , the cultures are plated at appropriate dilutions on 3 different LB agar medium: LB only to count the total number of bacteria ( CFU ) , LB supplemented with 100 ug/mL ampicillin + 0 . 1 mM IPTG to count the number of bacteria bearing a copy of the segregative plasmid , and LB supplemented with 100 ug/mL rifampicin ( plated volume: 200 uL ) to count the number of mutants toward the phenotype of interest . Additionally to this 24 h time point , cultures are also plated on LB and LB + ampicillin + IPTG at intermediate time points ( 3 h and 6 h ) to have a more accurate quantification of plasmid segregation dynamics and thus a better time resolution for the estimation of death rate . The plates are incubated between 15 h and 24 h for LB and LB ampicillin IPTG and exactly 48 h for LB rifampicin before counting colonies . Further experimental details are given in S5 Supporting information . pAM34 is a colE1 derivative whose replication depends on a primer RNA put under the control of the inducible promoter pLac [76] . Under the presence of 0 . 1−1mM IPTG ( nonmetabolizable inducer of the lactose operon ) , the plasmid is stably maintained in every cell . When IPTG is removed from the growth medium , the plasmid is not replicating anymore , or not as fast as the cells divide , and thus is stochastically segregated at cell division . The decrease in plasmid frequency between two time points then allows us to compute the number of bacterial cell divisions that occurred between these two time points . Combined with the change in population size , this allows us to compute average death rate and growth rate between these two time points ( see Fig 9 and mathematical explanations below ) . Such segregation measures have been used in a less quantitative way by other researchers [77 , 78] to crudely infer overall population turnover in vivo . pAM34 also carries a betalactamase . The number of plasmid-bearing bacteria can thus be counted by plating an appropriate dilution of the culture on LB supplemented with 0 . 1 mM IPTG ( to ensure maintenance of the plasmid within colonies founded by a plasmid-bearing cell ) and 100 ug/mL ampicillin ( to only permit growth of colonies founded by a plasmid-bearing cell ) . The total number of bacteria is determined by plating an appropriate dilution of the culture on LB . Because mutational dynamic does not depend on time , we chose to compute relative death rate ( ratio of death rate and growth rate as functions of time ) , which is the average number of death events per division event . The link between plasmid segregation , death , and number of divisions between two time points can be expressed mathematically as follows . If we have the following: then the plasmid is diluted/segregated at each division following the equation Ffinal=Finitial× ( 1+res2 ) g So we can estimate g=log2 ( FfinalFinitial ) /log2 ( 1+res2 ) Without any death , we would have gno−death=log2 ( Nfinal/Ninitial ) The difference between the true number of generations g computed from plasmid frequency and this number of generations gno−death computed based on the assumption that there is no death , allows us to estimate relative death rate as follows: Nfinal=Ninitial×2 ( 1−d ) *g This yields d=1−log2 ( Nfinal/Ninitial ) g and thus d=1−log2 ( Nfinal/Ninitial ) log2 ( Ffinal/Finitial ) ×log2 ( 1+res2 ) The only remaining free parameter to estimate is res , which is estimated by performing growth kinetics without antibiotic treatment ( in LB medium ) and thus without ( or with negligible ) death . We then have g=gno−death and thus log2 ( 1+res2 ) =log2 ( Ffinal/Finitial ) log2 ( Nfinal/Ninitial ) from which we can fit the value of the segregation parameter log2 ( 1+res2 ) based on the values of F and N estimated by plating . Further experimental and mathematical details on the plasmid segregation system are given in S2–S5 Supporting Informations . Most modern measures of mutation rate rely on the same standard protocol , the fluctuation test [79] , directly inspired by the Luria and Delbrück experiment [75]: several cultures are inoculated with a small population of nonmutant bacteria , are grown overnight and are then plated on selective media ( to count the number of mutants in the final population ) and on nonselective media ( to count the total number of bacteria in the final population ) . The number of mutants r in the final population ( or , rather , its distribution over several replicate populations ) is used to estimate the number of mutational events m happening during growth . One should note that these two numbers are not equivalent , because one mutational event can lead to several mutants in the final population if it happens early during growth , making this part of the computation complicated for intuition , although good mathematical tools are available . The total number of bacteria N is assumed to be very close to the number of cell divisions ( and thus the number of genome replications ) because the initial number of bacteria is much smaller . Mutation rate can thus be estimated as μ = m/N . The many existing software packages used to compute m from the observed distribution of r use an analytical expression of the probability-generating function ( pgf ) of the number of mutants in the final population [80] . The only free parameter is the number of mutational events ( equivalent to the value of the mutation rate per division when scaled with population size ) . This parameter is estimated from plating data using the maximum likelihood principle . The most used implementation of this idea is FALCOR [39] , available on a webpage: http://www . keshavsingh . org/protocols/FALCOR . html . Other software packages implementing the same ideas have been developed more recently , including , for example , rSalvador [38] and bzRates [40] , which also implement a few alternative assumptions such as fitness impact ( cost or benefit ) of the focal mutation or a more accurate correction for plating efficiency than the one suggested by FALCOR [81] . However , to this day , no available software allows users to compute mutation rate when there is death . Some papers derived analytical expression of the pgf of the number of mutants in the final population in conditions in which there is death [82] , but this has to our knowledge never been applied to real data nor implemented in a software package . In theory , such computations could easily be implemented in a tool similar to FALCOR ( web server ) or rSalvador ( software package ) . However , the basic assumption of the derived formula is that death rate is constant . This assumption is the price to pay for an analytical expression for the pgf and is unfortunately not appropriate in our case , given the observed death kinetics ( see Fig 3 ) . On the other hand , given the computational power available today , we believe that analytical computations are not always necessary . In our case , while the measured population dynamics do not allow us to derive an analytical expression of the pgf , it is straightforward to simulate many times such population dynamics with an arbitrary mutation rate and to obtain an empirical distribution of the number of mutants . Running these simulations for any possible value of the mutation rate parameter then allows Bayesian inference: we look for the simulated mutation rate that gives the closest distribution to the one experimentally observed , as graphically represented in Fig 10 . Such methods are classically referred to as Approximate Bayesian Computing . We implemented such simulations and inference in a Python software package , ATREYU , and use this software as the heart of our data analysis . | The effect of environmental stress on bacterial mutagenesis has been a paradigm-shift discovery . Recent developments include evidence that various antibiotics increase mutation rates in bacteria when used at subinhibitory concentrations . It is therefore suggested that such treatments promote resistance evolution because they increase the generation of genetic variation on which natural selection can act . However , existing methods to compute mutation rate neglect the effect of stress on death and population dynamics . Developing new experimental and computational tools , we find that taking death into account significantly lowers the signal for stress-induced mutagenesis . Moreover , we show that treatments that increase mutation rate do not always lead to increased genetic diversity , which questions the standard paradigm of increased evolvability under stress . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"death",
"rates",
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"population",
"dynamics",
"drugs",
"population",
"genetics",
"cell",
"processes",
"microbiology",
"cell",
"cycle",
"and",
"cell",
"division",
"antibiotics",
"pharmacology",
"population",
"biology",
"ecological",
"metrics",
"species",
"diversity",
"mutagenesis",
"population",
"metrics",
"population",
"size",
"cell",
"biology",
"ecology",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology"
]
| 2018 | Death and population dynamics affect mutation rate estimates and evolvability under stress in bacteria |
The highly pathogenic avian influenza ( HPAI ) H5N1 influenza virus has been a public health concern for more than a decade because of its frequent zoonoses and the high case fatality rate associated with human infections . Severe disease following H5N1 influenza infection is often associated with dysregulated host innate immune response also known as cytokine storm but the virological and cellular basis of these responses has not been clearly described . We rescued a series of 6:2 reassortant viruses that combined a PR8 HA/NA pairing with the internal gene segments from human adapted H1N1 , H3N2 , or avian H5N1 viruses and found that mice infected with the virus with H5N1 internal genes suffered severe weight loss associated with increased lung cytokines but not high viral load . This phenotype did not map to the NS gene segment , and NS1 protein of H5N1 virus functioned as a type I IFN antagonist as efficient as NS1 of H1N1 or H3N2 viruses . Instead we discovered that the internal genes of H5N1 virus supported a much higher level of replication of viral RNAs in myeloid cells in vitro , but not in epithelial cells and that this was associated with high induction of type I IFN in myeloid cells . We also found that in vivo during H5N1 recombinant virus infection cells of haematopoetic origin were infected and produced type I IFN and proinflammatory cytokines . Taken together our data infer that human and avian influenza viruses are differently controlled by host factors in alternative cell types; internal gene segments of avian H5N1 virus uniquely drove high viral replication in myeloid cells , which triggered an excessive cytokine production , resulting in severe immunopathology .
The outcome of infection with an influenza virus can vary widely from asymptomatic infection to death . Although infection outcomes can be influenced by host factors such as age , prior immunity , genetic susceptibility and comorbidities [1–4] , differences in the virus itself undoubtedly contribute to the variation observed . The most devastating human influenza virus in recent recorded history was the ‘Spanish influenza’ virus that caused the 1918 pandemic . Recombinant influenza viruses reconstructed from the sequences of 1918 influenza are rapidly lethal in animal models [5] . H5N1 ‘bird flu’ is a highly pathogenic avian influenza virus that has occasionally infected humans with catastrophic outcome . Of around 856 people infected by this virus , 52 . 8% have died . This contrasts starkly with the outcome of infection in 2009 with the new pH1N1 pandemic virus that was associated with just 0 . 02% case fatality [4 , 6] . One feature often associated with the severe disease following H5N1 infection , and also noted during animal experimental infections with 1918 virus , is a dysregulated host innate immune response known as the cytokine storm . This response is characterized by excessive levels of inflammatory cytokines and chemokines such as type I interferons ( IFNs; IFN-α and β ) , TNF-α , IL-6 , IL-8 , CCL2 , and CXCL10 [7–11] , leading to prolonged fever , lymphopenia , severe pneumonia , and extensive lung damage [12] . Although cytokine storms are not unique to influenza , they are almost always associated with high mortality rates [13–15] . Therefore , it is critical to understand the immunological and virological mechanism behind the activation of these storms , particularly in light of recent evidence that indicates it is possible for H5N1 strains to become airborne transmissible [16–17] . IFNs are important for the antiviral immune response by restricting viral replication . However , excessive type I IFN responses amplify the early pro-inflammatory cytokine response in the lung via type I IFN receptor signaling [18–19] , and have been associated with lung inflammation in severe influenza infection [20] . Previous studies revealed that lung epithelial cells , macrophages , conventional dendritic cells ( cDCs ) and plasmacytoid dendritic cells ( pDCs ) all express type I IFNs to some extent during influenza infection [21–22] . However , the primary source of type I IFNs in response to influenza infection , especially H5N1 , and the virological mechanism behind the cytokine storm remain unknown . Excessive cytokine production during H5N1 influenza infections occurs in spite of the fact that , like all natural influenza viruses , it encodes a well-known interferon antagonist , the NS1 protein . NS1 suppresses the host’s antiviral response by direct inhibition of activation of RIG-I by viral RNAs [23–24] , as well as through dsRNA sequestration , and modification of the expression of induced genes [25–29] . NS1 gene sequences vary between different strains and subtypes of influenza virus and between viruses isolated from different hosts [30] . It is possible the cytokine storm triggered by H5N1 influenza virus in humans is the result of an unadapted avian NS1 protein that does not efficiently antagonize the type I IFN induction pathways in human cells . However , previous reports have shown that NS1 proteins from a variety of different avian influenza virus strains including H5N1 viruses were able to control type I IFN responses in cells of human origin in vitro [31–34] . The severe outcome following HPAI H5N1 influenza virus infection may be , to some extent , dependent on specific genetic features of the H5 HA gene . First , the H5 HA protein harbours a multibasic cleavage site that facilitates multiple cycles of virus replication outside of the respiratory tract . Possession of a multibasic cleavage site in HA determines the highly pathogenic phenotype in poultry hosts [35] and also contributes to the high pathogenicity in the mouse model [36–37] . However , other influenza viruses associated with severe disease including cytokine storm in animal models , such as 1918 H1N1 influenza , do not carry this motif [38] . Other features of the H5 HA that might contribute to H5N1 pathogenicity in mice and humans include a preference to bind α-2 , 3 sialic acid ( SA ) receptors thus targeting the virus to the lung [39] , a pH of fusion that is higher than for human-adapted strains that might enhance entry into endothelial cells [40] , and the ability to trigger acute lung injury through a TLR4 dependent pathway [41] . However , receptor binding specificity and pH stability of H5 HA might be altered if H5 viruses were to gain human transmissibility , so it is important to understand the contribution of internal gene segments to disease severity . In the following research , we investigated the role of internal genes of an H5N1 influenza virus in the activation of a cytokine storm and compared the virus-host interaction with that of human adapted viruses . To avoid the complication that different viral surface proteins , HA and NA , might affect cell tropism and immune responses in vitro and in vivo , we rescued a series of viruses that combined a A/Puerto Rico/8/34 ( PR8 ) HA/NA pairing with internal gene segments from either an H1N1 , H3N2 or H5N1 virus [36 , 39 , 42–45] . Since the PR8 HA and NA genes enable efficient infection of mice , we were able to use mice as a tractable in vivo model to study the outcome of infection with the different RG viruses . Our results reveal that the internal genes of H5N1 facilitate high replication of viral RNAs in haematopoetic cells and drive excessive cytokine production that is the hallmark of infection with these avian influenza viruses .
To address whether the internal gene segments of a highly pathogenic H5N1 avian influenza virus , associated with a fatal human case , contribute to disease severity , we generated a recombinant influenza virus ( 6:2 Tky/05 ) with the six internal gene segments from influenza H5N1 A/turkey/Turkey/05/2005 virus ( Tky/05 ) and the HA and NA genes from the laboratory adapted strain PR8 . The H5N1 Tky/05 virus is highly pathogenic in mice [46] . In addition , we generated similar recombinant viruses with internal gene segments from a seasonal H3N2 influenza virus A/Victoria/3/1975 ( Vic/75 ) or from a prototypic early pH1N1 pandemic virus A/England/195/2009 ( Eng/09 ) , also combined with HA and NA genes from PR8 . Finally to assess the importance of the NS gene segment , we generated a chimeric virus with five internal gene segments from the pH1N1 virus ( PB1 , PB2 , PA , NP and M ) , and the NS segment from the avian H5N1 virus , combined with the PR8 HA and NA genes ( Eng/09:TkyNS ) ( Fig 1A ) . Thus , all four of these viruses had the same tropism determined by the PR8 HA and NA gene segments . The three 6:2 viruses replicated efficiently in MDCK cells . The virus with Tky/05 internal genes replicated slightly faster than the other two , showing higher titres at 12 and 24 hours . The chimeric virus with Tky/NS gene produced titres around one log less than the other viruses at 12 and 24 hours post infection , although it attained a high titre by 36 hours ( Fig 1B ) . To assess the pathogenicity of the four RG ( reverse genetic rescued ) viruses , 6–8 week old BALB/c mice were infected by intranasal inoculation of 104 pfu viruses . All infected mice lost weight over the following 7 days , but weight loss was most dramatic in mice infected with the 6:2 Tky/05 virus ( Fig 2A ) . Mice in this group rapidly lost weight in the first four days after infection , reaching almost 20% weight loss by day 4 . Interestingly , this dramatic weight loss did not correlate with a higher lung viral load ( Fig 2B ) or with increased spread of virus through the lungs ( S1 Fig ) . Indeed , the viral titre in lung homogenates was highest for mice infected with the 6:2 Eng/09 virus on both day 2 and day 7 post infection . Day 2 lung titres were similar between the other three viruses . At the later time point ( day 7 ) two mice surviving in the 6:2 Tky/05 infected group had cleared the virus from the lung but did not regain weight ( Fig 2A and 2B ) . Instead of viral lung titre , the increased weight loss was associated with an early and high level of cytokines in the lungs of the 6:2 Tky/05 virus infected mice ( Fig 2C and S2 Fig ) . Levels of type I IFNs , TNF-α , IL-6 and CXCL1 as well as CCL2 , and IFN-γ were all significantly higher in lungs of 6:2 Tky/05 virus infected mice than in the other three groups at day 2 post infection ( Fig 2C ) . The dynamics of cytokine detection in the lung tissue ( S3 Fig ) showed that type I IFNs were produced earlier than most of the other cytokines . IFN-alpha ( IFN-α ) was detected in the lungs of 6:2 Tky/05 virus-infected mice along with IL-6 at day 2 , whereas levels of TNF-α , CXCL1 , CCL2 and IFN-γ continued to rise through day 3 post infection . The most abundantly infected cell type in the virus infected lung is the epithelial cell [47–49] . We previously showed that different influenza viruses varied in the extent to which they induced a type I IFN response in A549 cells , a human lung epithelial cell line [31] . To assess the induction of type I IFNs by the RG viruses in the present study we utilized a reporter A549 cell line we previously generated that harbors a reporter gene with the IFN-β promoter upstream of luciferase . The reporter cells were infected with equal titres of the four RG viruses and luciferase was measured at 24 hours post infection ( Fig 3A ) . In contrast to what was observed in vivo , infection with the 6:2 Tky/05 virus did not lead to high activation of the IFN-β promoter . Indeed , it was the 6:2 virus with internal genes from Eng/09 ( pH1N1 ) virus that stimulated the highest luciferase signal in the A549 cells , as we previously described [50] . The 5:1:2 chimeric virus that combined polymerase gene segments from Eng/09 with the NS gene segment from Tky/05 did not induce a high luciferase signal from infected A549 reporter cells , implying that the Tky/05 NS1 protein functioned as an effective type I IFN antagonist ( Fig 3A ) . We also measured induction of type I IFNs in primary mouse tracheal epithelial cells , MTEC , infected with equal titres of each virus . All four viruses induced IFN-β in a similar pattern to that seen in the human A549 cell line , with the highest levels induced by the 6:2 Eng/09 virus and the lowest by the 5:1:2 Eng/09:TkyNS virus ( Fig 3B ) . IFN-α levels were lower from MTEC cells but showed a similar pattern ( Fig 3C ) . To compare the ability of the NS1 proteins of each virus to control induction of the IFN-β promoter , we expressed them in the A549 IFN-β reporter cells , and then challenged them with the 6:2 Eng/09 virus , or Newcastle Disease Virus ( NDV ) , to activate the type I IFN induction cascade . We found that the NS1 protein from Tky/05 virus suppressed IFN-β induction efficiently , as did Vic/75 NS1 and both were more effective than the NS1 protein of the Eng/09 virus ( Fig 3D and 3E ) . Indeed a titration of plasmids encoding Vic/75 and Tky/05 NS1 proteins demonstrated no difference in their efficiency to suppress the IFN-β signal ( Fig 3F ) . Although epithelial cells are the primary target of infection and the major producers of progeny virus , dendritic cells and macrophages have also been reported to be infected by influenza virus , and are believed to be the main producers of cytokines and type I IFNs in vivo [10 , 20 , 51–55] . So far there is not a thorough understanding of which is the major cell type that contributes to differences in the extent of type I IFN production induced by infection with different viruses , nor the mechanism that defines such differences . We aimed to find a cell population that could be studied in vitro that reflected the pattern of type I IFNs produced by our panel of viruses in vivo . To this end we propagated bone marrow derived cells in different media to produce populations with phenotypes similar to macrophages ( BMDMs , propagated with L929 conditioned media ) , dendritic cells ( BMDCs propagated in GM-CSF termed GM-DCs ) or plasmacytoid dendritic cells ( BMDCs propagated with Flt3 ligand termed FL-DCs ) . Each cell population was infected at equal multiplicity with the RG influenza viruses . Strikingly , IFN-α and IFN-β production from GM-DCs and IFN-β production from BMDMs propagated using L929 conditioned media , reflected the same pattern as seen in vivo , in that type I IFN levels were significantly higher from cells infected with the 6:2 Tky/05 virus than for any of the other viruses ( Fig 4 ) . In addition to the high type I IFN responses observed , the GM-DCs infected in vitro with 6:2 Tky/05 virus , but not the other viruses , also displayed a high induction of TNF-α and IL-6 mRNAs ( S4 Fig ) . In contrast , FL-DCs produced very low levels of IFN-β upon infection . Levels of IFN-α after infection of FL-DCs were higher but were not different between 6:2 Vic/75 and 6:2 Tky/05 viruses ( S5 Fig ) . We next carried out experiments to understand the basis of the high type I IFN induction observed in vitro in infected GM-DCs . We found that IFN-α production induced by infection with any of the recombinant influenza viruses depended on virus replication since there was no signal following UV inactivation of input virus ( Fig 5A ) . To test the hypothesis that high levels of RNA generated during replication by the 6:2 Tky/05 virus drove the high type I IFN response in myeloid cells , we measured the accumulation of viral RNAs by qRT-PCR in different cell types . In GM-DCs , by 8 hours post infection levels of mRNA and vRNA of the Tky/05 virus were 7 . 7 and 17 . 0 fold higher than that of the human adapted viruses , respectively ( Fig 5B and 5C ) . In contrast , in MTEC levels of viral RNAs were not higher for the Tky/05 virus ( Fig 5D and 5E ) . In other epithelial cells tested , human A549 cells or mouse LA4 ( lung epithelial ) cells , there was also little or no difference between levels of viral RNAs produced ( S6 Fig ) . At 8 hours post infection of LA-4 cells the Tky/05 virus produced just 2 fold more vRNA than the other viruses ( S6 Fig ) . We also generated a mutant virus that would be compromised in replication in mammalian cells by engineering the mutation K627E in the Tky/05 PB2 gene segment . Indeed accumulation of v and mRNAs in GM-DCs infected with this virus were greatly decreased compared to the ‘wild type’ 6:2 Tky/05 virus with the mammalian-adapting 627K motif ( Fig 6A and 6B ) . The mutant virus with PB2 627E no longer induced IFN-α mRNA in infected GM-DCs ( Fig 6C ) . In order to ensure the PR8 HA and NA of the Ty/05 virus was not misrepresenting the ability of an H5N1 virus to infect and replicate in GM-DCs , we generated PR8:TkyHAsbNA , that contained the six internal genes from PR8 and the H5 HA and N1 NA from A/turkey/Turkey/05/2005 virus . To make it biologically safe , the multi-basic cleavage site of H5 HA was removed . We compared the infection of GM-DCs by this virus with infection by whole PR8 virus and found them to be similar ( S7 Fig ) . The m , c , and vRNA accumulation were also similar between PR8 and PR8:TkyHAsbNA infected GM-DCs , and significantly lower than in GM-DCs infected with 6:2 Tky/05 virus ( Fig 6D–6F ) . Accordingly , neither the PR8:TkyHAsbNA virus nor the whole PR8 virus induced high IFN-α expression ( Fig 6G ) . To probe the pathway that led to high type I IFN induction in GM-DCs , we infected GM-DCs derived from Mavs-/- mice and assessed the IFN-α response . In the absence of MAVS , none of the viruses induced a type I IFN response despite there being high levels of viral RNA following infection with the 6:2 Tky/05 virus ( S8 Fig ) . To confirm that cells of haematopoetic origin contribute to the excessive IFN-α in vivo in 6:2 Tky/05 virus infected mice , we FACS purified CD45-positive cells from lungs harvested 2 days post infection and performed qRT-PCR for viral RNAs and for IFN-α and IL-6 mRNAs . We found evidence of vRNA and mRNA in this cell population as well as a type I IFN and IL-6 cytokine response ( S9 Fig ) . To further demonstrate that cells of hematopoetic origin were important in the cytokine response and severe outcome of infection with the 6:2 Tky/05 virus , we engineered a recombinant virus ( NPr142-Tky/05 ) that harboured four copies of a microRNA target sequence in the NP gene for a microRNA specifically expressed in cells of haematopoetic origin , MiR142 , as previously described by Langlois et al [56] . This would result in cell type specific reduction in virus replication since levels of NP protein required to support replication would be specifically reduced in myeloid cells . A control virus contained an inserted sequence at the same location that was not a MiR target ( NPctrl-Tky/05 ) ( Fig 7A ) . NPr142-Tky/05 and NPctrl-Tky/05 viruses showed similar replication in MDCK cells ( Fig 7B ) . However , in GM-DCs , vRNA accumulation following infection with the virus containing MiR142 target sites was significantly reduced compared to control virus ( Fig 7C ) and this led to a decreased level of IFN-α mRNA in these cells ( Fig 7D ) . Mice were infected with 105 pfu of each recombinant virus , and monitored for weight loss , lung titre and IFN-α in lung homogenates at day 2 ( Fig 7E–7G ) . The insertion of the MiR142 target sequence , that reduced the extent of replication in cells of hematopoietic origin , resulted in reduced weight loss that correlated with a lower IFN-α level in the lung ( Fig 7E and 7G ) . Virus titre in the lung was not affected ( Fig 7F ) , suggesting that the majority of infectious virus is produced from epithelial cells , but that replication in cells of haematopoetic origin is the source of the excess type I IFN produced by the 6:2 Tky/05 virus . Finally , we attempted to attribute the high cytokine inducing phenotype to a particular viral polymerase or NP gene by creating a set of recombinant viruses based on the 6:2 Tky/05 virus in which each RNA segment was exchanged for that from the Eng/09 virus ( S10A Fig ) . Although several of these viruses replicated to high titres in mouse lung , none induced rapid weight loss as seen for the 6:2 Tky/05 virus ( S10B and S10C Fig ) . Furthermore , exchanging any one polymerase or NP segment for that of Eng/09 abrogated the high cytokine induction in the lungs of infected mice suggesting that a single gene of the Tky/05 virus is not responsible for the high cytokine phenotype but rather the particular replication activity of the Tky/05 polymerase and NP complex . ( S10D Fig ) .
The severity of the next influenza virus pandemic will be determined by the nature of the virus that emerges from an animal source and acquires an airborne transmissible phenotype . The HPAI H5N1 virus has been a public health concern for more than a decade because of frequent zoonoses and the high case fatality rate associated with human infections [57–59] . At least part of the high virulence of this virus can be explained by features of the H5 HA protein such as a propensity to bind to α-2 , 3-SA receptors in the lungs , an ability to enter endothelial cells , and a multi-basic cleavage site that might enable systemic spread and infection of cell types not usually infected by seasonal influenza viruses [35 , 42 , 60–63] . Indeed recent work by Tundup et al . used a similar approach to that employed here to control virus tropism in vivo by engineering MiR target sites into RG H5N1 virus , and highlighted the importance of endothelial cell infection in H5N1 pathogenesis in mice [37] . However , some features of the H5 HA that support the extended tropism of H5N1 avian influenza would likely be lost if the virus gained airborne transmissibility , and so it is important to understand how other genes of the virus also affect pathogenesis if we are to predict the likely severity of an H5N1 pandemic . Although the ferret model is considered the gold standard for influenza infection and transmission , the HA/NA pairing employed in these studies to normalise viral entry would not facilitate significant viral replication in the ferret respiratory tract due to receptor specific incompatibilities [64] . Therefore , we focused on the mouse model of influenza . In contrast to the ferret model , the mouse model of influenza benefits from a plethora of established protocols and immunological reagents , as well as a fully annotated genome . In particular , the lack of annotation of the ferret genome makes it uncertain as to whether the NPr142-Tky/05 virus would be restricted in hematopoetic cells . A recent article , as well as our own nucleotide BLAST analysis , have revealed a high degree of similarity between mouse miRNA-142 and an unannotated miRNA in the ferret genome , but further research is necessary to understand whether this putative miRNA-142 displays similar cell-type specific effects in ferrets as it does in mice [37] . Here we assessed the contribution of internal gene segments that encode the viral RNA dependent RNA polymerase ( RdRp ) , nucleoprotein , matrix and nonstructural proteins to the outcome of H5N1 influenza virus infection . Using a reverse genetics strategy , we engineered viruses that had an identical ability to bind and enter cells because they encoded the same HA/NA pairing , but differed in their interaction with factors inside the infected cells depending on the human or avian virus origin of the segments encoding the internal genes . This revealed that the internal genes of the H5N1 virus contribute to the severe outcome in vivo . Upon infection of mice with the virus with H5N1 internal genes , dramatic body weight loss was accompanied by high levels of type I IFNs and other inflammatory cytokines in the lungs , although titres of virus were not higher than in mice infected with human adapted viruses . Two previous studies also found that the polymerase genes of H5N1 influenza virus determined the high cytokine response in human macrophages and in mice but did not solve the underlying mechanism [65–66] . Recent microarray analysis has confirmed prior genetics and biochemistry studies , which implicate type I IFNs as the main driver of many other cytokines during influenza infection [67] . However , the cell type that is the primary source of type I IFN , especially during H5N1 infection in vivo was not clear . Previous work has shown that CD11c+ cells , which constitute macrophages , monocytes and dendritic cells , produce the bulk of IFN-β in vivo following infection with a mouse adapted influenza virus [51] . Our data suggest that H5N1 triggers an unusual and excessive cytokine response compared to human adapted viruses in cells of haematopoetic origin ( Figs 4–6 ) . In vitro , using GM-DCs , which are thought to represent a mixture of monocyte derived and conventional DCs [68] , we found that the 6:2 H5N1 virus induced significantly more type I IFNs than the equivalent human adapted viruses . In contrast , in epithelial cells , the 6:2 H5N1 virus was not a potent type I IFN inducer . Conflicting reports in the literature about the propensity of H5N1 influenza viruses to trigger high type I IFNs might be explained by our cell type specific findings: Cheung et al . first reported high cytokines induced by H5N1 viruses using human monocyte derived macrophage populations , whereas Zeng et al . found that H5N1 viruses effectively controlled type I IFNs in human bronchial epithelial cells [10 , 69] . Although GM-derived dendritic cells are not perfect analogues for dendritic cells found in the lung , it is striking that the pattern of type I IFNs production in these cells closely resembled that produced in vivo during infection by the three strains of RG 6:2 influenza viruses . Moreover the decrease in lung type I IFN levels during infection with the Mir142 targeted RG virus confirms that a proportion of the cytokines that contribute to the proinflammatory response during H5N1 virus infection were secreted in vivo by infected haematopoetic cells . One explanation for the high levels of type I IFNs during H5N1 virus infection would be that virally encoded IFN antagonists from the avian derived Tky/05 virus did not control the type I IFN response in mammalian cells . Three pieces of evidence from our study ruled out that the H5N1 NS1 protein was deficient in this regard: first , there were low levels of cytokines in lungs of mice infected with the 5:1:2 Eng/09:TkyNS virus , second , the 6:2 Tky/05 virus effectively controlled type I IFN induction in epithelial cells and third , exogenously expressed Tky/05 NS1 protein could efficiently suppress an IFN stimulus , at least in epithelial cells . Since several other viral gene products are implicated in controlling the innate immune response , including two of the polymerase proteins , PB2 and PA , as well as the accessory proteins PB1-F2 , and PA-X , it may be that one or more of these functions are defunct in the Tky/05 virus , although sequence analysis based on current knowledge does not support this [70] . Perhaps some IFN antagonists of the avian influenza virus do not function in mammalian myeloid cells . We were not able to test directly whether exogenously expressed Tky/05 NS1 protein or any other virally encoded IFN antagonists were able to control type I IFN induction in the GM-DC cells we employed here . However , using a panel of RG viruses in which individual polymerase gene segments were swapped between Tky/05 and Eng09 , we did not find any single viral gene to which we could attribute the high cytokine inducing phenotype ( S10 Fig ) . The alternative and more plausible explanation for our data is that the higher levels of a viral RNA species produced during H5N1 infection in myeloid cells , which are more sensitive to viral pathogen associated molecular patterns ( PAMPs ) than epithelial cells , outweighed the potency of the H5N1 virus-encoded IFN antagonists . We suggest that the PAMP responsible is a replication product of the viral polymerase . We found that viruses with H5N1 internal genes had an unusual propensity to drive high levels of RNA replication in the GM-derived DC cells infected in vitro . The conventional pathway by which influenza virus triggers a type I IFN response is by RIG-I detection of replicated viral RNAs and subsequent signaling through MAVS [71–73] . Infection of GM-DCs derived from Mavs-/- mice did not result in any type I IFN production despite similar levels of virus replication ( S8 Fig ) suggesting the pathways triggered by H5N1 virus are the conventional ones . Type I IFN induction in MAVS positive cells was dependent on replication and correlated with the amount of viral RNAs generated . We hypothesize that one or more host cell factors that are differentially expressed in myeloid cells vs . epithelial cells affect the behavior of the RNA dependent RNA polymerase ( RdRp ) of the avian and the human adapted virus differently . Thus the severe disease resulting from infection with H5N1 virus stems from excessive RNA polymerase activity in mammalian myeloid cells . High viral replication has been previously linked with the clinical outcome of H5N1 infection in mice [36] . The inappropriate early type I IFN response is not dampened by viral IFN antagonists and drives a downstream production of inflammatory cytokines in the lung that leads to severe weight loss . Whether the lack of control of virus replication in myeloid cells is a common feature of other avian influenza viruses associated with severe human infections , whether there is a specific virus signature responsible and the identity of host factors that determine these differences , remain to be established .
LA4 cells were acquired from Prof . Robert Snelgrove ( Leukocyte Biology Section , National Heart and Lung Institute , Imperial College London ) and were originally obtained from American Type Culture Collection ( ATCC ) . LA4 cells were maintained in Ham’s F12 medium conditioned with 2mM L-glutamine and 15% FBS . Human embryonic kidney ( 293T ) ( ATCC ) , human lung adenocarcinoma epithelial cells ( A549 ) ( ATCC ) and Madin-Darby canine kidney ( MDCK ) cells ( ATCC ) cells were maintained in Dulbecco’s modified Eagle’s medium ( DMEM; Gibco , Invitrogen ) supplemented with 10% FBS and 1% penicillin/streptomycin ( Sigma-Aldrich ) . A549-Luc cells were constructed and maintained as previously described [74] . L929 cells were a gift from Caetano Reis e Sousa , The Francis Crick Institute , UK , and were cultured with 10% FBS , 0 . 05mM β-mercaptoethonal and 1% penicillin/streptomycin in conditioned RPMI 1640 medium . Plasmids used in this study to rescue viruses have been previously described [75–76] . NS1 expression plasmids utilized in Fig 3D–3F were described previously[50] . All the viruses used in this research were rescued by reverse genetics . Briefly , 12 plasmids , comprising of 8 polI plasmids encoding the indicated virus segments and 4 helper expression plasmids encoding A/Victoria/3/75 polymerase components and NP expressed by the pCAGGS vector , were transfected into the 293-T cells that were then co-cultured with MDCK cells . Virus stocks were grown on MDCK cells using serum free DMEM supplemented with 1ug/mL of TPCK trypsin . Viruses were stored in -80°C and titrated on MDCK cells by plaque assay . The modified Tky-NP segment was generated by PCR and ligation . We added a BamHI site 62bp upstream of the 3’ terminus of the complimentary sequence by site directed mutagenesis . Utilizing a naturally occurring BstBI restriction site 257bp upstream of the 3’terminus , we were able to insert a synthesized 478bp-length BamHI and BstBI flanked sequence ( GeneArt ) . The sequence contained 231bp of codon optimized open reading frame upstream of the stop codon , four tandem copies of miRNA142 target sequence or scrambled control sequence , and 146bp duplicated segment 5 packaging sequence . Thus we created a plasmid , which conserved both the amino acid composition of the NP protein and the packaging sequence of the segment [56 , 77] . The miRNA142 target was the same as previous described [56] . The scrambled control sequence was provided by Lucy Thorne ( University College London ) : AGACAATACGTCACATATAACGA . The modified NP plasmids and RG virus were sequence verified . MDCK cells were infected with the indicated virus at MOI = 0 . 001 and overlaid with serum free DMEM containing 1μg/ml TPCK trypsin . Supernatants were collected at a given time point after infection and stored at -80°C . Virus was titrated by plaque assay in MDCK cells . Inactivation of virus was performed by UV radiation . Indicated viruses were exposed on ice to short wave UV radiation at 254nm for 10 min at a distance of 15cm . Virus integrity and loss of infectivity was confirmed by Haemagglutination test with chicken erythrocytes ( Envigo RMS ( UK ) Ltd ) and plaque assay , respectively . Plaque assays were performed as previously described [76] . Briefly , 100% confluent MDCK cell monolayers were inoculated with 100 μl of serially diluted virus and overlaid with 2% agarose ( Oxoid ) in supplemented MEM with 2 . 6 μg/ml trypsin ( Worthington ) and incubated at 37°C for 3 days . For Haemagglutination assay , 0 . 5% chicken erythrocytes ( Envigo RMS ( UK ) Ltd ) were mixed with serially diluted virus in a 96-well plate . After half an hour’s incubation on ice , the results were read by eye . A549-IFNβ cells were transfected with NS1-pCAGGS plasmid . Briefly , fully confluent cells were incubated with 1000ng/well of either Eng/09 , Vic/75 , or Tky/05 NS1-pCAGGS plasmid in Opti-MEM and lipofectamine ( Invitrogen ) for 2–4 hours . Opti-MEM was removed and replaced with 10% DMEM overnight . Viral challenge of transfected cells was performed at either an MOI 3 ( influenza infections ) , or with a final dilution of 1:100 ( NDV ) . Eighteen hours post infection , cells were lysed and harvested with 200μl passive lysis buffer ( Promega ) , then assayed for luciferaseusing FLUOstar Omega Plate Reader ( BMG Labtech ) . Six to eight week old female BALB/c or C57B6 mice ( Charles River UK Ltd or Envigo RMS UK Ltd ) were maintained in pathogen-free conditions until used for viral infection or cell isolation . Mavs-/- mice were a kind gift from Professor S . Akira [18 , 78] . Mice were infected intranasally with the indicated PFU of virus or serum free DMEM ( mock ) in a 25ul volume under isofluorane . Animals were monitored and weighed daily . Lungs were harvested on Day 1 , 2 , 3 , 4 , or 7 , or when weight loss dropped below 80% of the original weight on Day 0 . Lungs were suspended in 1ml of PBS and homogenized using 1 . 3mm beads , homogenates were frozen in -80°C before testing the virus titer and cytokine expression . To isolate bone marrow cells , femur and tibia of three-to-six week old female BALB/c mice were excised and cleaned of flesh . Bone marrow cells were flushed out , filtered through a nylon cell strainer ( Falcon , 2350 ) and washed with PBS . Cells were resuspended and differentiated in RPMI 1640 medium ( 10% FBS , 0 . 05mM β-mercaptoethanol and 1% penicillin/streptomycin ) supplemented with GM-CSF ( R&D , cat 415-ML-010: final concentration 40 ng/ml ) , or Flt3-L ( R&D , cat 427-FL-025: final concentration 250 ng/ml ) , or 20% L-929 cell supernatant . On day 3 or 4 of culture , non-adherent cells , which are mostly granulocytes , were removed and fresh medium containing the same concentration of GM-CSF or Flt3-L or L929 cell supernatant was added . On day 7 , cells were harvested for further experimental use [79–81] . Mouse tracheal cells were isolated from three-to-six week old female BALB/c mice by pronase digestion . The cells were seeded in a petri dish and incubated for 3–4 hours to adhere fibroblasts . Then non-adherent cells were collected and reseeded in a 10cm petri dish and cultured with DMEM/F12 medium supplemented with 10% FBS , 15 mM HEPES ( Gibco ) , 0 . 03% NaHCO3 ( Gibco ) , 0 . 01uM Retinoic acid ( Sigma , cat R-2625 ) , Amphotericin B ( Gibco , cat A2942: final concentration 250ng/ml ) , EGF ( BD , cat 354001: final concentration: 25ng/ml ) , D-valine ( Sigma , V1255: final concentration: 0 . 1mg/ml ) , bovine pituitary extract ( Gibco , cat 13028–014: final concentration: 30 ug/ml ) , Cholera Toxin ( Sigma , cat C8052: final concentration: 0 . 1ug/ml ) , Insulin-Transferrin-Selium ( Gibco , cat 41400045: 1:100 ) . Medium was changed every two days . After 14 days incubation and differentiation , the epithelial cells were harvested for experimental use . Bone marrow derived GM-DCs , FL-DCs and macrophages , LA4 , A549 , or MTEC cells were seeded on 96-well plates ( about 1 . 25×105 cells/well ) and infected with virus diluted in serum free DMEM for 1hr at 37°C ( MOI as indicated in the relevant figure legends ) and replaced with culture supplemented with 2% FBS . Cell supernatants were harvested at the indicated time points post-infection . Infected cell lysates were washed with PBS and then used to extract RNA according to the protocol below . Viral RNA was extracted from the infected cells using RNA extraction kits ( QIAGEN , RNeasy Mini Kit , cat . 74106 ) following the manufacturer’s instructions . Complementary DNA ( cDNA ) was synthesized in a reverse transcription step using different polarity specific primers . Primers for generating cDNA from segment 4 ( HA ) vRNA and mRNA were 5’-ACAGCCACAACGGAAAACTATG-3’ and Oligo-dT , respectively . Primers for generating cDNA from segment 7 ( M ) vRNA , cRNA , and mRNA were 5’-CTTGAAGATGTCTTTGCAGG-3’ , 5’-AGCAGAAACAAGGTAGT-3’ , and Oligo-dT , respectively . To quantify the vRNA , cRNA and mRNA levels , real-time quantitative PCR analysis with a gene specific primer pair using SYBR green PCR mix ( Applied Biosystems ) was performed and data was analyzed on the Applied Biosystems ViiATM 7 Real-Time PCR System . For HA vRNA and mRNA analysis , the following primers were used: forward primer , 5’-GGCCCAACCACAACACAAAC-3’ , reverse primer , 5’-AGCCCTCCTTCTCCGTCAGC-3’ . For M vRNA and mRNA analysis , the following primers were used: forward primer , 5’- CCAATCCTGTCACCTCTGAC-3’ , reverse primer , 5’- TGGACAAAGCGTCTACGC-3’ . β-actin was detected as a reference gene using the following primers: Forward primer , 5’-GTACGCCAACACAGTGCTG-3’ , Reverse primer , 5’-CGTCATACTCCTGCTTGCTG-3’ [82] . The gene expression was calculated by normalizing target gene expression to β-actin for each sample and expressed as 2ΔCt . Analyses were performed using 7500 Fast System SDS software ( Applied Biosystems ) . Cytokine quantities for IL-6 , TNF-a , IL-10 , CXCL1 , IL-12p40 , and IFN-g in the lung tissue were determined by the mouse proinflammatory 7-plex tissue culture kit ( Meso Scale Discovery , cat K15012B-1 ) using 25ul of homogenized lung tissue according to manufacturer’s instructions . The concentration of IFN-α and IFN-β from mouse cell supernatant and mouse lung tissue was measured with VeriKine mouse IFN-α ELISA kit ( PBL , cat 42400 ) and VeriKine mouse IFN-β ELISA kit ( PBL , cat 42400 ) , respectively . The mRNA level of TNF-α , IFN-β , IL-6 , and IFN-α in GM-DCs were tested with SYBR green method as described above . Primers used for the cytokine test were the following: TNF-α-F , 5'-GGCAGGTCTACTTTGGAGTCATTG-3’ , TNF-α-R , 5'-ACATTCGAGGCTCCAGTGAATTCGG-3’; IFN-β-F , 5'- AAGAGTTACACTGCCTTTGCCATC-3’ , IFN-β-R , 5'- CACTGTCTGCTGGTGGAGTTCATC-3’; IL-6-F , 5'-GACAAAGCCAGAGTCCTTCAG AGAG-3’ , IL-6-R , 5'-CTAGGTTTGCCGAGTAGATCTC-3’; IFN-α-F , 5'-CGCAGGAGAAGGTGGATGCCCAG-3’ , IFN-α-R , 5'-CAGCACATTGGCAGAGGAAGACAGG-3’ [19] . Six to eight week old BALB/c mice were infected with 104 or 105 PFU of either 6:2 Eng/09 , or 6:2 Tky/05 virus intranasally . Two to three days post infection mice were culled; the lungs inflated with 1mL PBS , and placed in 4% PFA solution overnight . Lungs were embedded in paraffin , and mounted on slides by the Inflammation , Repair , and Development group , NHLI at Imperial College London . For immunohistochemistry , The Francis Crick Experimental Histopathology STP used formalin fixed paraffin embedded sections that were de-waxed in xylene then dehydrated by passage through graded alcohols to water . For antigen retrieval , sections were microwaved in sodium citrate , pH 6 for 15 minutes and then transferred to PBS . Endogenous peroxidase was blocked using 1 . 6% hydrogen peroxide in PBS for 10 minutes followed by washing in distilled water . Biotinylated goat anti-NP antibody was used as primary antibody ( USA biological , cat I7650 ) diluted to 1:100 in 1% BSA and incubated for 1 hour at room temperature . Sections were washed in PBS prior to applying ABC ( Vector Laboratories , cat PK-6100 ) for 30 minutes . Following washing in PBS , DAB solution was applied for 2–5 minutes with development of the colour reaction being monitored microscopically . Slides were washed in tap water , stained with a light haematoxylin , dehydrated , cleared and then mounted . Images were obtained by an Olympus VS120 slide reader , and analyzed by Image J software . GM-DCs were seeded on Poly-L-Lysine treated glass coverslips and incubated for overnight . The cells were infected with the indicated viruses at a MOI = 4 . After 4h and 8h infection , the GM-DCs were fixed for 20 min with 4% paraformaldehyde , permeabilized for 5 min with 0 . 1% TritonX 100 ( Sigma , cat X100RS-5G ) , and blocked with 5% BSA for 1h in room temperature ( RT ) . Cells were incubated with NP-FITC antibody ( ThermoFisher , cat D67J , diluted with 1% BSAS at 1:20 ) for 1h at RT . DNA was stained with 4’ , 6’-diamino-2phenylindole ( DAPI ) for 10 min . After washing with PBST , samples were mounted with Mowiol under a coverslip . Multiple images were obtained for each sample by a Zeiss Axiovert 40CFL , and analyzed by AxioVision SE64 Rel software . Nuclei and FITC positive cells were counted for each image . Mice were sacrificed and the lungs perfused with PBS . To obtain lung leukocytes , lung lobes were collected into a C-Tube ( Miltenyi Biotech ) containing complete DMEM ( cDMEM; supplemented with 10% fetal bovine serum , 2mM L-glutamine , 100U/ml penicillin and 100μg/ml streptomycin ) , Collagenase D ( 1mg/ml; Roche ) and DNase I ( 30μg/ml; Invitrogen ) and processed with a gentleMACS dissociator ( Miltenyi Biotech ) according to the manufacturer’s protocol . Shredded tissue was incubated for 1h at 37°C . After lysis of red blood cells , cells were strained through a 100μm filter ( BD Bioscience ) . For CD45+ lung cell sorting cells were incubated for 20 min with a purified rat IgG2b anti-mouse CD16/CD32 receptor antibody ( BD Bioscience ) to block Fc binding . Cells were then stained with fluorochrome-conjugated antibodies against CD45 ( 30-F11 , eFluor780 , eBiosciences ) in PBS containing 1% BSA and 5mM EDTA for 25 min at 4°C . 1ng/ml of Hoeschst 3358 ( Pentahydrate ( bis-Benzimide ) ; Thermo Scientific ) was added just before running the sample for exclusion of dead cells . Cells were sorted using a standard Becton Dickinson Aria-II and stored in RLT buffer until RNA extraction was performed . The purity of CD45+ cells was >98% . All data are presented as mean ± SD of three or more experiments . For viral replication kinetics and weight loss in Fig 2A and S10B Fig , area under the curve ( AUC ) for each virus was calculated . Difference in the AUC between viruses was analyzed with one-way ANOVA and Bonferroni’s multiple comparisons test . For weight loss of the NP mutant virus infected mice ( Fig 7G ) , two-way ANOVA test with post-tests for multiple comparisons was performed to determine P-value . One-way ANOVA analysis was used for the other comparisons among groups . Pearson correlation test was performed for the correlation analysis . P-value<0 . 05 wasnconsidered significantly different . All data analyses and preparation of all graphs were carried out with GraphPad Prism ( GraphPad Software , San Diego , CA ) . All work with infectious agents was conducted in biosafety level 2 facilities , approved by the Health and Safety Executive of the UK and in accordance with local rules , at Imperial College London , UK . All work was approved by the local genetic manipulation ( GM ) safety committee of Imperial College London , St . Mary’s Campus ( centre number GM77 ) , and the Health and Safety Executive of the United Kingdom and carried out in accordance with the approved guidelines . All animal research described in this study was approved and carried out under a United Kingdom Home Office License , PPL 70/7501 in accordance with the approved guidelines , under the Animals ( Scientific Procedures ) Act 1986 ( ASPA ) . | Some avian influenza viruses , including highly pathogenic H5N1 virus , cause severe disease in humans and in experimental animal models associated with excessive cytokine production . We aimed to understand the virological mechanism behind the cytokine storm , and particularly the contribution of internal gene segments that encode the viral polymerase and the non-structural proteins , since these might be retained in a pandemic virus . We found that the internal genes from an H5N1 avian influenza virus allowed virus to replicate to strikingly higher levels in myeloid cells compared to internal genes of human adapted strains . The higher viral RNA levels did not lead to higher viral load but drove excessive cytokine production and more severe outcome in infected mice . The remarkable difference in viral replication in myeloid cells was not observed in lung epithelial cells , suggesting that cell type specific differences in host factors were responsible . Understanding the molecular basis of excessive viral replication in myeloid cells may guide future therapeutic options for viruses that have recently crossed into humans from birds . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"orthomyxoviruses",
"epithelial",
"cells",
"viruses",
"bone",
"marrow",
"cells",
"developmental",
"biology",
"h5n1",
"rna",
"viruses",
"molecular",
"development",
"infectious",
"diseases",
"influenza",
"a",
"virus",
"zoonoses",
"animal",
"cells",
"medical",
"microbiology",
"microbial",
"pathogens",
"biological",
"tissue",
"viral",
"replication",
"immune",
"system",
"cell",
"biology",
"anatomy",
"influenza",
"viruses",
"virology",
"viral",
"pathogens",
"physiology",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
]
| 2018 | Internal genes of a highly pathogenic H5N1 influenza virus determine high viral replication in myeloid cells and severe outcome of infection in mice |
Persistent non-participation of children in mass drug administration ( MDAs ) for trachoma may reduce program impact . Risk factors that identify families where participation is a problem or program characteristics that foster non-participation are poorly understood . We examined risk factors for households with at least one child who did not participate in two MDAs compared to households where all children participated in both MDAs . We conducted a case control study in 28 Tanzanian communities . Cases included all 152 households with at least one child who did not participate in the 2008 and 2009 MDAs with azithromycin . Controls consisted of a random sample of 460 households where all children participated in both MDAs . A questionnaire was asked of all families . Random-intercept logistic regression models were used to estimate odds ratios ( ORs ) and 95% confidence intervals ( CIs ) , control for clustering , and adjust for community size . In total , 140 case households and 452 control households were included in the analyses . Compared to controls , guardians in case households had higher odds of reporting excellent health ( OR 4 . 12 ( CI 95% 1 . 57–10 . 86 ) ) , reporting a burden due to family health ( OR 3 . 15 ( 95% CI 1 . 35–7 . 35 ) ) , reduced ability to rely on others for assistance ( OR 1 . 66 ( 95% CI 1 . 01–2 . 75 ) ) , being in a two ( versus five ) days distribution program ( OR 3 . 31 ( 95% CI 1 . 68–6 . 50 ) ) and living in a community with <2 community treatment assistants ( CTAs ) /1000 residents ( OR 2 . 07 ( 95% CI 1 . 04–4 . 12 ) . Furthermore , case households were more likely to have more children , younger guardians , unfamiliarity with CTAs , and CTAs with more travel time to their assigned households ( p-values<0 . 05 ) . Compared to full participation households , households with persistent non-participation had a higher burden of familial responsibility and seemed less connected in the community . Additional distribution days and lessening CTAs' travel time to their furthest assigned households may prevent non-participation .
Trachoma is a leading cause of preventable blindness [1] . Nearly 41 million individuals across the globe are estimated to suffer from active trachoma [2] . Of these , the majority are children from impoverished regions [3] , [4] . Virtually all trachoma burden is either concentrated in rural Africa , particularly Ethiopia , Kenya , Niger , Sudan and Tanzania , or parts of Asia [5] . The World Health Organization ( WHO ) advocates mass drug administration ( MDA ) in eligible communities as a key component of the Surgery , Antibiotics , Face-washing , Environmental change ( SAFE ) strategy for treating and preventing trachoma . When a community's prevalence of follicular trachoma ( TF ) is greater than 10% in children less than age ten years , the WHO supports at least three annual mass drug administrations ( MDAs ) . [6] . In Tanzania , azithromycin is provided to the Ministry of Health free of charge through a donation program from Pfizer Inc and International Trachoma Initiative . For each resident over age one year , a single oral dose of azithromycin at 20 mg/kg up to 1 gram , is recommended , and infants one year and younger are treated with topical tetracycline . Data suggest that endemic communities often require multiple rounds of mass treatment for reducing the prevalence of trachoma [7] , [8] . Programs aim for antibiotic coverage goals of at least 80% or more in the entire community [6] . Low treatment coverage with antibiotics in children under ten years is problematic . Young children are a high-risk group for trachoma and infection [9] . Extensive child non-participation in community mass treatments may reduce the effectiveness of trachoma control programs . Untreated children are likely to spread trachoma to other household members and subsequently more individuals in the community [10] , [11] . Furthermore , programs squander resources in having to execute additional MDAs when the community treatment coverage is low . Given the WHO recommendations for multiple MDAs in trachoma-endemic communities , characterizing the households with children who never participate is essential . Understanding households with one or more children who never participate in MDAs may help programs develop strategies for avoiding persistent child non-participation . This study aimed to examine the predisposing and resource risk factors for Tanzanian households with children who never participated in two treatment rounds compared to households where all children participated .
We conducted the study in the Kongwa district of Tanzania . Located in the Dodoma region of Tanzania , approximately 250 , 000 people were residents of Kongwa in 2002 [12] . This study was nested in a larger study [13] of 32 communities which were randomly picked from a list of all communities who met the following criteria: Local government leaders had to provide consent ( all communities who were approached did provide consent ) , and the best-estimated prevalence of trachoma in 2007 was greater than 20% for each community . Due to timing of the MDA , this study was carried out in 28 of the 32 communities . As described in [13] , prior to each round of MDA , trained research staff completed a census of all households in each community by going to each household and enumerating the residents . Demographic information on each household member was collected at this time . The data were used to develop treatment log books for MDAs . This level of precision was required as part of our research program . In 2008 and 2009 , each community received mass treatment . The Kongwa Trachoma Project ( KTP ) team trained a group of CTAs , approximately two to six individuals per 1500 persons in each community . Community leaders assisted in identifying persons in the community who would be trusted to deliver MDA , and the KTP staff interviewed and ultimately chose the CTAs . The CTAs received a one-day program discussing trachoma , the disease and consequences , the SAFE strategy , details on azithromycin and possible side effects and how to record them , instructions on how to administer azithromycin by weight to children under one year , and using the height sticks for children greater than one year . If there was doubt as to age one year or less , and the child was below the smallest level of the height stick , the children were weighed . CTAs delivered MDA in their neighborhoods , as would be done in the national Program . We received ethical approval to treat children from one year to 6 months with oral azithromycin , 20 mg/kg , and those under 6 months were treated with topical tetracycline . In addition , the CTAs received training in recording the observed treatment on treatment logs . They also received modest training in asking about vision problems and recognizing trichiasis , in order to keep a record of all persons in the village who had need of further eye care and surgery . In other districts in Tanzania , there may be modest differences in approaches to MDA; in general the districts provide training to village health workers and community treatment assistants ( CTAs ) on use of height sticks for treating all residents , with those who are adults ( not defined further ) receiving 1 gm . Treatment is recorded in log books , and estimated village populations are used to monitor coverage . Two days at least are allotted for MDA , and the CTAs originally , but not since 2006 , received monetary incentives . All communities in the Kongwa district were mass treated on a rolling basis over a period from June to November 2008 , and again over the same months in 2009 , including communities not in the study . Communities in our study , as part of the larger study were randomly allocated to either a two-day or a five-day distribution program , which began after the census and surveys for the larger study in each community . The June to November time period was chosen because it was after the planting harvest so guardians would be home for mass treatment and to be interviewed . Community treatment assistants offered each resident over six months a single oral dose of azithromycin , 20 mg/kg up to one gram , irrespective of disease status . Oral treatment was directly observed and recorded in a logbook based on the household census . To children less than six months , CTAs gave guardians tetracycline eye ointment to administer topically for four to six weeks . The first dose was instilled but subsequent doses were not directly observed . All communities aimed for treatment coverage greater than or equal to 80% in children under age ten and those in the five day distribution arm were allowed 3 extra treatment days to achieve 90% coverage or greater . At least one member of the KTP staff was in the community each day of MDA , meeting with the CTA and reviewing performance . Each CTA was assigned a certain number of households for which they were responsible . The CTAs administered treatment to residents at a central location , and if necessary , at the household . CTAs were instructed to review their log books after the first day and if necessary schedule a second central location site or go to each non-participant person's home if necessary and treat them directly . The choice of options was up to the individual CTA , but the goal for the entire community was to achieve at least 80% . All data on MDA and treatment verification were entered into customized databases . Standard quality control measures were used to verify coverage . KTP staff went back to a random sample of 5 households per CTA to verify treatment status of all household members . If treatment as recorded in the CTA treatment log was at least 70% concordant with treatment as stated by the family for each member , the CTAs received a small monetary incentive ( 1 , 000 TSH or $0 . 80 ) per day of work . No CTA was found to be under-performing by this criteria . Data were not routinely collected on characteristics of the CTA . Therefore , for this study each CTA completed a survey on their age , sex and marital status . We also asked about past work experience ( e . g . any past MDA experience ) . We used the census and MDA data to identify case and control households with children between six months and nine years at the 2008 census . Our criteria required children to be residents in the households from the 2008 census to the 2009 MDA . Case households included at least one child , between six months to nine years old at the 2008 census , who did not participate in the 2008 and 2009 MDAs . Control households contained children from six months to nine years at the 2008 census who were treated at both MDAs . We did not match or restrict criteria for controls . We interviewed the guardian of the chlld , defined as either the mother or father , or if neither was serving as the guardian , the person in the household who self reported being the guardian . Interviewers surveyed fathers with more than one household in the community only once , and excluded 3 fathers who had already been interviewed . Our sample size calculation included the following assumptions: alpha = 0 . 02 , beta = 0 . 20 , 2∶1 control to case ratio , prevalence of 30% for most risk factors , and odds ratio of 2 . 0 . We conducted a Bonferroni adjustment for 3 comparisons of the same set of risk factors for this and a companion paper on change in participation . If no correction were applied , we would have had a chance of 0 . 1426 ( 14 . 26% ) of finding one or more significant differences in 3 tests . To get an alpha level of 0 . 05 , we lowered the alpha for each test to 0 . 01667∼0 . 017 . Assuming a non-response/ineligible rate of 15% , the necessary sample size was 330 control households and 165 case households . The adjustment was to make sure we had the correct sample , size , and that we could report at alpha = 0 . 5 , Using simple random sampling , we enrolled a random sample of 460 control households from a larger sample of 5375 control households and all households identified as having at least one child who was a persistent non-participant in MDA . We attempted to contact every case and control household in the 28 communities . The primary outcome of interest was a family where at least one child was a continuous non-participant in MDA . We conducted exploratory data analyses , using Pearson's chi-square tests of independent proportions for nominal data , Mann-Whitney tests for ordinal data and t-tests for continuous data . Backwards stepwise logistic regression models assisted in the identification of risk factors with a p-value less than 0 . 10 . One by one , we incorporated each significant risk factor into a random-intercept logistic regression model to evaluate changes in odds ratios and 95% confidence intervals ( CIs ) and adjust for clustering at the community level . We controlled for community size ( small size = bottom 33% communities , medium size = middle 33% communities and large size = top 34% communities ) as a confounder and selected the model with the lowest Akaike's information criterion ( AIC ) . We hypothesized that households where all children never participated were different from household where some children never participated . Thus , we performed subgroup analyses with a multinomial model comparing risk factors between: 1 ) households where every child was a persistent non-participant in both MDAs , 2 ) households in which some children were persistent non-participants , and 3 ) households with all children participating in both MDAs . Using this model , the relative risk ratio ( RRR ) represents the change in the odds of being in the case subgroup versus the control group , where the model can simultaneously estimate the RRR for each case sub-group associated with a one unit change in the independent variable . All analyses were run in STATA ver . 11 ( Stata Corp , College Station , Texas ) .
In this area , our program offered mass treatment twice , and households with children who never participated in two rounds was quite low , occurring in 2% of 6727 households . Our study contacted 612 households , 152 with at least one child who was a persistent non-participant and 460 where all children participated in both rounds . According to our 2008 census , contacted households had 2 , 129 children . The mean community size in the case control study was 1 , 685 people ( standard deviation = 482 ) . Community populations ranged from 750 to 2611 residents in 2008 . The mean number of persons per household was five , and the average number of persons under age ten was two children . Of the 612 households , 596 ( 97% ) households completed the risk factor survey . Twenty households of the original 612 were not included in the analyses . Three households were ineligible ( 2 in case households , 1 in the control households ) . Thirteen households did not respond ( 10 case households , and 3 control households ) , , In four control households , we could not be certain if treatment had taken place for each child following treatment verification . . We had no missing information on mass treatment for any child in the surveyed households . ( Figure 1 ) Our study observed no differences in CTA characteristics and guardian demographics between non-response and response households ( data not shown ) . Some household and program predisposing risk factors were significantly associated with being a household with at least one persistent non-participant ( Tables 1 and 2 ) . The risk increased with each additional child in the household of being a household with a persistent non-participant ( p< . 01 ) . This remained significant after adjusting for other factors ( Odds Ratio ( OR ) = 1 . 70 , 95% Confidence Interval ( CI ) = 1 . 39–2 . 08 ) ( Table 3 ) Guardians in households with a persistent child non-participant had more than a tenfold odds of not rating the assigned CTAs performance ( p = 0 . 02 ) , but this was also correlated with not knowing their assigned CTAs which was also significantly associated with persistent non-participation ( p<0 . 01 ) . Adjusting for other factors , incorrectly naming or being unable to name their assigned CTAs was associated with an increased risk of persistent non-participation ( OR = 1 . 99 ( 95% CI = 1 . 16–3 . 06 ) and 5 . 17 ( 95% CI = 2 . 17–12 . 32 ) , respectively ) . Compared to households with full child participation , households with persistent child non-participation were more likely to be assigned to CTAs living more than one hour from their furthest assigned household in the community . This relationship persisted after adjustment for multiple factors ( OR = 2 . 58 , 95% CI = 1 . 22–5 . 44 ) . A number of guardian and household predisposing factors had no association with household with persistent child non-participation in simple bivariable analyses . Guardian's age , education , perceived health , length of residency , gender , traditional healer use , ethnic group , and attendance at a promotional meeting for mass treatment did not predispose a household to persistent child non-participation . We found no association between persistent non-participation and household predisposing risk factors: reported family health problems , household history of adverse events during the 2008 MDA , and familial possession with malevolent spirits . Our study found associations between households with a child who never participated and guardian and program resource risk factors ( Table 2 ) . Households with a persistent non-participant had low score of social reliance ( not being able to ask anyone for money or for a place to live ) . Compared to households with full child participation , households with persistent child non-participation were more likely to live in a community with a two ( versus five ) days distribution strategy , and more than a threefold odds of being in a community with less than two CTAs per 1000 residents ( p-value<0 . 01 ) . These factors remained significant when adjusted for multiple factors ( Table 3 ) . After controlling for community size , clustering , and the other variables , our final model identified several independent predisposing and resource risk factors for persistent child non-participation ( Table 3 ) . Predisposing factors included younger age and perceived excellent health the week of the 2009 MDA , familial health burden and increasing numbers of children in the family . Resource risk factors included guardians with low scores for social reliance , increased travel time from the assigned CTA's household to the furthest household in the community , less than two CTAs per 1000 residents in the community , and a two days ( versus five ) days distribution strategy . Our case and control guardians had some similarities and differences in their response to the general question of the primary reason why parents in the community did not bring their children for treatment ( Table 4 ) . In case households , the two most common reasons were travel outside the community during mass treatment and perceived negative side effects from drugs . Control households reported negative side effects from drugs as well , but also felt that general lack of knowledge ( stated as “ignorance” ) and lack of education on the part of the guardian were explanations for non-participation . Of the 140 households that completed the risk factor survey , 54 ( 40% ) were households in which all children never participated in both MDAs . The remaining 86 ( 60% ) households contained some children who had participated in one or both rounds as well children who were non-participants in both rounds . We hypothesized differences between these two subgroups , compared to households where all children participated . Common factors for both groups were program predisposing and resource factors: being in a two days distribution program and not being familiar with any assigned CTA ( Table 5 ) . However , compared to households with full child participation , households where all children never participated both times had healthier guardians , expressed a family health burden , and were in a program with less than two CTAs per 1000 residents . Households where some children participated and others did not in both rounds had other risk factors . In comparison to households with all children treated both times , each additional child in the household increased the risk of having a household where some ( but not all ) children were persistent non-participants . In addition , these households had younger guardians , and were assigned CTAs living more than one hour from the furthest assigned household .
Guardians exert a strong influence on their children's healthcare . It is therefore important to ensure that trachoma control programs providing mass treatment address guardian concerns and barriers . Identifying guardian characteristics of households' with persistent child non-participation may help programs target households at-risk . Among the possible guardian predisposing and resource risk factors studied , younger guardian age , perceived excellent health , and decreased ability to rely on others were useful markers of households with persistent child non-participation . Similar to our study , other child health services have found younger guardian age is a risk factor for lower use of child health services [17] . This variable was more important for households where not all children were persistent non participants , which suggests the difficulty young guardians have in bringing all children to MDA . Guardians in households with persistent child non-participation perceived their health as better during the week of mass treatment compared to guardians in households with full child participation . This result is comparable to another program that found people who were healthy tended to not participate in mass treatment [18] . Guardians in households with full child participation may have been less healthy and thus more likely to take their children for MDA because they themselves also wanted to be treated . Also , those who report being healthy were more likely to be guardians of households where all children did not participate , suggesting that there was no perceived need for treatment or low priority was given to participation . We found no difference between the case and control households in perceived risk of trachoma in their children , suggesting that general self-perception of health may be more important than messages about trachoma . Social reliance or the ability to rely on other individuals for money or a place to live was an important guardian resource that households with persistent child non-participation lacked . That ability to rely on others is a key part of kinship systems , systems that continue to thrive in Tanzania [19] . A high degree of reciprocal exchange of goods and services in these systems exists , and it is through this sharing of resources that the groups thrive . Social networks provide an informal social security; research has demonstrated a positive association between larger strong social networks and well-being in low-income countries [19] . Guardians who could not rely on others for money or shelter were likely not as deeply supported as were other guardians in the community . The association was strongest for guardians of households where at least some children participated in one or both rounds . For these guardians , getting help transporting all their children to the central distribution site for treatment in both MDAs was possibly more difficult . Thus , distribution strategies that target marginalized households - guardians who cannot rely on others for money or a place to live- and encourage healthy guardians to bring their children for treatment as a way of keeping them and their family healthy , may prevent persistent child non-participation . An understanding of household barriers may also assist programs in recognizing households at-risk for persistent child non-participation . We found that reported family health problems severe enough to interfere with daily tasks was a strong risk factor for households with persistent child non-participation . This was especially the case for households where no children participated . The family health burden originated from family members ( e . g . husband , grandmother ) not from the guardian , as the guardian perceived him/herself as healthy . Research has linked a stressed guardian to lower use of child health services [20] . Guardians under stress , potentially caused by a health burden within the family , must contend with many hardships that compete with MDA . Multiple young children in the household were a characteristic of households with some children who never participated . Our results were consistent with a previous study that examined households factors associated with azithromycin coverage after a single round [21] , which also found that having more children residing in the household adversely influenced household coverage . Guardians may find the logistics of bringing all children in the household to a treatment center difficult if the number in a household is large . Additional studies have verified that mothers with multiple children are at high risk for having under-vaccinated children and using fewer primary care services for children [20] , [22] , [23] . As noted , these households tend to have younger guardians with less ability to rely on others as well , providing an image of over-burdened young guardians who tend not to participate . Social mobilization programs , involving local groups , have been useful in targeting marginalized households at-risk for non-participation in child health services [24] . In Dhaka , Bangladesh , an intervention package including a supportive group for social mobilization was valuable in moving from 43% children fully immunized to 99% children fully immunized . Since we now know characteristics of households with children who persistently do not participate , we could use a similar strategy to identify and assist households with many young children who have less ability to rely on others , or describe having a family member with an illness . Modifying some treatment program characteristics related to visibility , access and organization might reduce persistent child non-participation . Inability to name the CTAs was one program factor in our study that could be targeted . Other mass treatment programs have observed non-familiarity with CTAs as a risk factor for individual non-participation [25] , [26] . In these MDAs , community members did not trust CTAs because they were unknown and not part of their community . However , as most CTAs in our study were from the community , this is not likely the problem and may reflect the fact that if the household did not participate , they did not meet the CTA . However , the CTA was supposed to travel to the household to offer MDA , and this finding suggests that this was not always the case . Future MDAs should ensure that in the case of non-participation the CTA visit the household . One program feature was related to less accessibility . Community treatment assistants living more than one hour from the furthest assigned household were characteristic of households with persistent child non-participation . Ivermectin MDAs for onchocerciasis also observed further distance from the CTA's household to the furthest assigned households was an issue . The CTAs working within one km were more likely to attain 90% treatment coverage in the community [27] . With greater travel time in a community , CTAs have less motivation to return several times to treat non-respondents , especially if there are only a few in a household that otherwise participated . This supposition is supported by our finding that this risk factor is more important for households where some , but not all , children were persistent non-participants . Programs seeking to stop persistent child non-participation could also address accessibility by increasing the number of distribution days and improve organization by increasing the number of CTAs per 1000 residents . In our study , supplemental treatment distribution days appeared to provide parents with more flexibility; Guardians could bring their children for treatment on days that were convenient for them . Past research in child immunization programs verified that shorter distribution time was associated with non-participation [28] , [29] . Modifications in the schedule allowed more guardians to attend a location , particularly working mothers . In addition , more assigned CTAs at the central distribution site cut the treatment lines , helped the drug administration process run more efficiently , and allowed CTAs time to visit households on more than one occasion . However , case and control households both resided in communities that had two and five days distribution programs so just increasing days alone is not the only factor . Given that the research provided a small incentive for CTA time doing MDAs , the cost per additional coverage needs to be evaluated . Factors related to the MDA delivery system ( good training , community government support , CTA incentives ) are liable to influence the effectiveness of treatment assistants positively , and this program contained all of these elements . An experienced non-government organization , KTP , supervised CTAs during the course of the MDA through daily observations . Furthermore , the community leadership recommended and supported CTAs . In addition , most CTAs were residents in their communities , so other residents in mass treatment programs would likely be familiar with their CTAs , even if they did not know they had taken on that responsibility . Following treatment verification of their work quality , the program offered CTAs an incentive for completing high coverage . Thus , we could not measure the effect of lack of incentives , or CTAs chosen by other mechanisms or lack of supervision as possible additional program factors . Increasing distribution time and number of CTAs in programs alone may not always result in improved performance , as the other training components for CTAs , CTA incentives , and support from local community are likely important overall factors as well . Nevertheless , with such additional factors in mind , programs like ours that have these elements in place should consider allotting funds for increased distribution time and more local personnel to improve participation . Finally , programs need specific interventions for households where all children never participated and households where some children never participated . Our study found each group had guardians strained in different ways . Strategies for encouraging households where some children never participated could include providing CTAs with bikes to travel to families , and working with local groups to reach out to younger guardians and those with multiple young children . For households where all children never participated , CTAs could work with local groups to identify households with guardians caring for sick family members , develop a protocol for “mop-up” treatment , and assist these guardians in getting their children treated . Hiring more than two CTAs for every 1000 residents may also enable the program to reach households where children never participated . The strengths in this study include minimal misclassification of cases and controls due to direct observation and recording of treatment , and the high participation among cases and controls . Treatment was directly observed by the CTA at the time of distribution . CTAs were spot-checked by KTP staff during the implementation of MDA , and treatment verification was carried out to ensure that records were maintained correctly . Therefore , we are confident that reporting errors were rare . Community treatment assistants could have over-reported compliance . However , treatment verification for the 2008 and 2009 were exceptional . Our study found misclassification in less than 1% of households in our study . We had very high response rate to the survey , 92% case households and 98% of control households . We found no differences in any CTA and census characteristics for case response households and case non-response households . Thus , we were confident that the risk factors found in our study have minimal bias due to non-response . Case control designs have limitations , notably the problem of recall bias . We retrospectively collected guardian time-dependent risk factor data three to six weeks after the 2009 mass treatment . Data may not be accurate if parents did not recall the information correctly , such as the state of their health or the other members of the family . We attempted to improve guardian recall by providing guardians with the exact dates of mass treatment during the field interview . Guardians were prompted with the number of weeks since mass treatment for mass treatment questions in the survey . Since recall bias might be in any direction it is difficult to predict how this might impact findings . Second , our study may have missed additional important factors , especially as related to the first , 2008 , MDA . We did not ask about factors related to the first MDA as it was over a year ago , but instead factors related to the second MDA . However , conducting a prospective study , with data collection immediately prior to each mass treatment , was not possible . We also recognize that the non –participation studied here is in the context of a generally high participation rate using CTA's that were motivated and incentivized . Thus , in other settings with lower coverage or different MDA program design , program factors may emerge as even more important than what we observed . Nevertheless , the guardian and program factors we did find are likely to have good generalizability to other settings similar to the communities sampled in this study . Trachoma remains the most common infectious cause of blindness and is likely to remain so in the near future . Although treatment participation for trachoma was high in this study , at both rounds , there were households with children who never participated . We identified predisposing and resource risk factors for these households that programs may address with more effort . Program designs should consider targeting marginalized households ( e . g . multiple children in household , family health burden ) through social mobilization programs . By increasing the number of distribution days to improve program access and by increasing the number of treatment assistants to help program organization , in addition to proper training , incentives , and record keeping , the program may decrease persistent child non-participation . Moreover , partnerships and buy-in with community leadership and liaison groups will no doubt be vital for the long-term successful trachoma control program . We dispelled the notion that households were passive recipients or apathetic non-participants of mass treatment programs . In actuality , households judge the treatment's value and relevance against their children's needs and factors related to the guardian , household , and program . In their design , programs can target these risk factors . The recognition of risk factors for households with persistent child non-participation is a critical step for implementing programs encouraging full child participation over time . | The World Health Organization advocates at least three mass drug administrations ( MDAs ) with antibiotics when the prevalence of follicular trachoma ( TF ) is greater than 10% in children under age ten . Full child participation is necessary for maximizing the impact of trachoma control programs . The present paper identifies guardian , household , and program risk factors for households with a child who never participated in two annual rounds of MDAs with azithromycin . In comparison to households with full child participation , guardians with at least one child who never participated had a higher burden of familial responsibility , as represented by reporting ill family members , more children , and were younger in age . In addition , guardians of persistent non-participants seemed less well connected in the community , in terms of reliance on others and not knowing who their assigned community treatment assistants ( CTAs ) were . These guardians were assigned to CTAs who had a wide geographic dispersion of their assigned households . By developing programs with local groups to find and encourage participation in at-risk households , program managers may have the greatest impact on preventing persistent child non-participation . Increasing the number of distribution days and reducing CTAs' travel time may further prevent non-participation . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"global",
"health",
"ophthalmology",
"neglected",
"tropical",
"diseases",
"public",
"health"
]
| 2012 | Azithromycin Mass Treatment for Trachoma Control: Risk Factors for Non-Participation of Children in Two Treatment Rounds |
Nipah virus ( NiV ) is a member of the genus Henipavirus ( family Paramyxoviridae ) that causes severe and often lethal respiratory illness and encephalitis in humans with high mortality rates ( up to 92% ) . NiV can cause Acute Lung Injury ( ALI ) in humans , and human-to-human transmission has been observed in recent outbreaks of NiV . While the exact route of transmission to humans is not known , we have previously shown that NiV can efficiently infect human respiratory epithelial cells . The molecular mechanisms of NiV-associated ALI in the human respiratory tract are unknown . Thus , there is an urgent need for models of henipavirus infection of the human respiratory tract to study the pathogenesis and understand the host responses . Here , we describe a novel human lung xenograft model in mice to study the pathogenesis of NiV . Following transplantation , human fetal lung xenografts rapidly graft and develop mature structures of adult lungs including cartilage , vascular vessels , ciliated pseudostratified columnar epithelium , and primitive “air” spaces filled with mucus and lined by cuboidal to flat epithelium . Following infection , NiV grows to high titers ( 107 TCID50/gram lung tissue ) as early as 3 days post infection ( pi ) . NiV targets both the endothelium as well as respiratory epithelium in the human lung tissues , and results in syncytia formation . NiV infection in the human lung results in the production of several cytokines and chemokines including IL-6 , IP-10 , eotaxin , G-CSF and GM-CSF on days 5 and 7 pi . In conclusion , this study demonstrates that NiV can replicate to high titers in a novel in vivo model of the human respiratory tract , resulting in a robust inflammatory response , which is known to be associated with ALI . This model will facilitate progress in the fundamental understanding of henipavirus pathogenesis and virus-host interactions; it will also provide biologically relevant models for other respiratory viruses .
Nipah virus ( NiV ) is a member of the genus Henipavirus ( family Paramyxoviridae ) that causes severe and often lethal respiratory illness and encephalitis in humans resulting in case fatality rates of up to 92% [1] . The first human cases of NiV infection were identified during an outbreak of severe febrile encephalitis in Malaysia and Singapore in 1998–1999 [2] , [3] . More recently , outbreaks have occurred in Bangladesh and India almost yearly since 2001 [1] , [4] . NiV can cause Acute Lung Injury ( ALI ) in humans , and human-to-human transmission has been observed in recent outbreaks of NiV [5] , [6] , [7] . Data on the histopathology of the lungs of NiV cases is limited to necropsy findings in the respiratory tract of NiV infected cases and include hemorrhage , necrosis and inflammation in the epithelium of the small airways but not in the bronchi [8] . Endothelial cells have been identified as a major target for NiV and most studies have focused on the role of the endothelium in NiV pathogenesis [9] , [10] , [11] . However very limited data is available on the host responses following NiV infection in the human lung . While the endothelium plays an important role in the terminal stages of NiV infection , the role of the respiratory epithelium in the early stages of infection is critical; however , it remains largely unexplored . The specific sites of henipavirus infection in the human respiratory tract are still unknown as well as the molecular mechanism by which these viruses cause disease in humans . We have previously shown that NiV can efficiently infect human respiratory epithelial cells from the trachea , bronchi and small airways resulting in the induction of key inflammatory mediators that have been implicated in leukocyte recruitment and ALI [12] . Similarly , in animal models ( hamster , ferret and African green monkey ) , NiV can replicate to high titers in the lungs of these animals and cause acute and severe respiratory distress [13] , [14] , [15] , [16] , [17] . Human xenograft mouse models have previously been used to study tissue development and cancer as well as the pathogenesis of infectious agents [18] , [19] , [20] . The majority of viral pathogenesis studies involving human xenograft mice focus on Human Immunodeficiency Virus ( HIV ) or Human Cytomegalovirus ( HCMV ) in humanized mice that have been grafted with human hematopoietic stem cells and thymus [21] , [22] . Here we report the first characterization of NiV infection of the human respiratory tract using a human lung xenograft model to gain further insight into the mechanisms of NiV pathogenesis in humans . Our results showed that NiV replicates to high titers in the human lung and that infection results in the induction of a robust host response .
Severely immunodeficient NSG mice served as hosts to support the successful engraftment of human lung xenografts . We implanted 6 small fragments of human fetal lung in the dorsal subcutaneous space , 3 on each side of the spine . Following transplantation , human fetal lung xenografts typically increased in size by 2–10 fold over 3 months . The human fetal lung xenografts rapidly grafted and developed mature structures similar to those seen in adult lung ( Figure 1A ) . These mature structures included bronchi that were partially surrounded by cartilage ( Figure 1B ) , lined with ciliated pseudostratified columnar epithelium ( Figure 1C ) and surrounded by longitudinal elastic fibers . The bronchi divide into bronchioles and terminal bronchioles lined by cuboidal to flat epithelium ( Figure 1D ) . The distal respiratory tract comprises of primitive alveolar spaces that are lined with both cells that have flat ( type 1 ) and larger rounded ( type 2 ) pneumocyte morphology ( Figure 1E ) . The alveolar walls of xenografts were thicker than those of normal adult human lungs . The human graft was well vascularized with the presence of arteries , veins , and capillaries ( Figure 1F ) . Finally , the expression patterns of ephrin B2 , the receptor of NiV , was similar to that seen in normal human lung tissue [23] ( Figure 1G ) with expression on bronchial epithelium , alveolar cells and vasculatures ( Figure 1H ) . Natural NiV infection involves exposure to the virus through the respiratory epithelium . To mimic this route in human lung xenografts that lack air exchange , tissues were directly injected with NiV . Following direct intragraft injection of the 3 lung tissues on the left side ( primary infection ) within each mouse , NSG mice did not show any signs of morbidity or mortality during our observation period of 10 days . In addition , two non-grafted NSG mice that were challenged intradermally as controls , with the same dose as xenografts , did not develop any clinical signs . Primary infection of the human lung xenografts resulted in detection of infectious NiV as early as 1 day post infection ( Figure 2A ) and NiV replicated to high titers ( 107 TCID50/gram tissue ) by day 3 post infection . NiV titers remained high until the end of the experiment at 10 days post infection . Importantly , high titers of NiV were also detected in the other 3 lung tissues ( on the right side of each mouse ) that were not initially infected through direct intragraft injection as early as 3 days post infection . This finding clearly demonstrates that the virus can spread from infected human lung grafts to uninfected grafts in the same mouse , most likely through viremia ( secondary infection; Figure 2A ) . The presence of viremia was further supported by the observation that infectious NiV was detected , albeit at lower levels , in several mouse tissues including lung , brain , heart , spleen and kidney at various time points post infection ( Figure 2B ) . In fact , viremia was detected in a blood sample of 1 animal on day 10 post infection in which a low level ( 300 TCID50/mL ) of infectious NiV was determined ( Table 1 ) . Interestingly , virus was not detected in organs from non-grafted NSG mice that were challenged intradermally with the same dose ( Table 1 ) , suggesting the NSG mouse tissues are probably not intrinsically susceptible to NiV . In order to confirm that the human lung xenografts could be infected via the hematogenous route , we next challenged 2 lung-engrafted NSG mice with NiV via the IP route . The IP challenge with NiV in this model confirmed that infection resulted in detectable viremia in 1 animal with virus spreading to the human lung xenografts in both , replicating to high titers and resulting in histopathological changes similar to those observed with intragraft challenge ( Table 1 ) . Together , these data suggest that following intragraft infection , the human lung is highly susceptible to NiV infection and results in viremia and subsequent spread to other organs in the absence of disease . In order to study the histopathological changes associated with NiV infection in the human lung , tissue sections were stained with hematoxylin and eosin ( H&E ) . No gross pathologic lesions were observed in the human lung grafts . Since NSG mice exhibit multiple defects in innate and adaptive immunity [24] , NiV infection in human lung grafts did not result in significant inflammation . Histopathological changes in the human lung tissues following NiV infection were independent of the route of infection ( intragraft , indirect or intraperitoneal ) and included small focal areas with syncytia and necrosis as early as day 3 pi ( Figure 3A ) . These areas rapidly expanded to large areas with hemorrhages and significant loss of architecture of the small airways by day 10 pi ( Figure 3B ) . The main histopathological features of NiV infection in these tissues were the characteristic syncytia formation ( Figure 3C ) and areas of necrosis ( Figure 3D ) . Syncytia formation could be observed in bronchial epithelium ( Figure 3C ) , alveolar epithelium ( Figure 3E ) and vascular endothelium ( Figure 3F ) . In addition , fibrinoid necrosis was observed in some of the vasculature as well as recruitment of granulocytes ( Figure 3F ) . In agreement with the absence of clinical signs , NiV infection did not result in histopathological changes in any visceral mouse tissue ( Table 1 ) . In order to identify the cells targeted by NiV in the human lung , viral nucleocapsid protein ( N ) expression in human lung grafts was examined with immunohistochemistry . Expression of NiV N coincided with the focal areas of histopathological changes on day 3 pi and showed intense staining ( Figure 4A ) . By day 10 , widespread expression of NiV N was observed throughout the human lung tissues ( Figure 4B ) . NiV primarily targeted the respiratory epithelium of the bronchi and bronchioles , interstitial mesenchymal cells ( Figure 4C ) , and the small airways ( Figure 4D ) . Cells targeted in the small airways were primarily cuboidal , which is consistent with type-2 pneumocyte morphology , although cells with type-1 pneumocyte morphology also showed reactivity . In addition to the respiratory epithelium , NiV replication also involved the vasculature ( Figure 4E ) . In agreement with the observation that low levels of infectious virus were detectable in several mouse tissues , small focal areas of viral antigen primarily focused in small airway epithelium could be detected in mouse lungs ( Figure 4F ) but not in other organs tested ( Table 1 ) . Tropism of viral antigen in mouse lung was similar between tissues infected by direct injection or following IP challenge ( data not shown ) . Although focal areas were generally not centered around vessels ( Figure 5A ) during the early stages of infection , when NiV infection involved the vasculature , CD31-positive endothelial cells were a specific target of infection ( Figure 5B ) . Similar findings were observed in animals challenged via the IP route ( Table 1 ) . In order to elucidate the host responses following NiV infection in the human lung , the expression of several cytokines and chemokines was determined in homogenates of human lung xenografts following direct infection with NiV ( Figure 6 ) . Since human immune cells were absent in this model , any expression of human cytokines or chemokines was primarily the result from NiV infection of human epithelial and endothelial cells . NiV infection in human lung resulted in the expression of several cytokines/chemokines , including eotaxin-1 , G-CSF , GM-CSF , TNFα , VEGF , IP-10 , IL-1β and IL-6 starting by day 5 pi ( Figure 6 ) . Expression of GM-CSF , TNFα , IP10 and IL-1β peaked on day 5 post infection and gradually declined over time . IL-6 and eotaxin-1 expression peaked at day 7 pi , whereas G-CSF initially peaked on day 5 but remained high throughout infection . Interestingly , VEGF expression continued to increase over time concomitant with the increased hemorrhaging and remodeling of the lung . The cytokine and chemokines profiles were similar between lung xenografts following primary ( direct ) or secondary ( indirect ) infection ( data not shown ) .
Nipah virus is an emerging zoonotic virus that can cause severe respiratory distress and encephalitis in humans [1] . Despite intensive studies in vitro and in animal models , little is known about the mechanisms governing the development of NiV-related respiratory disease in humans; this is due to difficulties in obtaining human samples where the disease is endemic . To address this important limitation , the goal of the present study was to characterize a novel human lung xenograft model to study the pathogenesis of NiV infection in human lung in vivo . Studies on the molecular mechanisms of NiV-mediated pathogenesis have been hampered by the lack of biologically relevant in vitro models for studying the initial host responses to NiV infection in the human lung [7] , [13] , [15] . To fill this gap , we recently showed that NiV can efficiently replicate in primary epithelial cells from the human respiratory tract [12] . While this is an attractive model to study the early steps of NiV entry in the host , it lacks the complexity of the microenvironment in the lung . In the current model , we show that human fetal lung tissues grafted on an immunocompromised mouse develop into more mature human lung tissues within 3 months after implantation . Transplanted lung tissues rapidly vascularized and developed bronchioles , lined with columnar epithelium , and alveolar-like spaces closely resembling those seen in normal human lung tissue . The prototype strain of NiV ( Malaysia ) was used in this study . While the outbreaks in Malaysia and Singapore have primarily been associated with the development of severe febrile encephalitis with a case fatality rate of 38% , respiratory symptoms were observed ( 40% of lethal cases ) [8] . Interestingly , the more recent outbreaks in Bangladesh and India are associated with a higher prevalence of respiratory disease as well as a significantly higher case fatality rate of 67% to 92% [1] , [6] . It is currently unknown whether differences in respiratory involvement are due to genetic difference between the Malaysia and Bangladesh strains of NiV or whether confounding factors are involved , however both NiV strains can replicate efficiently and cause respiratory distress in animals [25] , [26] . In addition , no histopathological data is available for human cases of the Bangladesh strain of NiV; therefore , the Malaysia strain was used in this study to allow for comparisons of histopathology and viral tropism . In humans , vasculitis and fibrinoid necrosis in the lungs was observed in the majority of fatal cases of NiV infection [8] . Multinucleated giant cells were occasionally observed in alveolar spaces and showed prominent immunostaining for viral antigen , along with alveolar hemorrhage , edema and pneumonia . Bronchial epithelium rarely showed histopathological changes . Nipah viral antigen could also be observed in the vasculature and rarely in bronchiolar epithelium [8] . We believe that the lack of viral antigen in the bronchial epithelium in fatal human cases is most likely due to timing of sampling . We previously showed , using a hamster model , that the bronchial epithelium is initially targeted by NiV early on during infection followed by rapid spread to the interstitium and involvement of pulmonary vessels [13] . In the present model , NiV replicated to high titers following intragraft injection , and virus was found to primarily replicate in respiratory epithelium of the bronchi and small airways . This is consistent with our previous finding that human respiratory epithelium is highly susceptible to NiV infection [12] . In animal models , the lung is the primary target organ of NiV infection following intranasal challenge [13] , [14] . In addition to the respiratory epithelium , NiV replication was also found in the endothelium , a type of cell that has been identified as an important target for NiV [11] . The infection of the vascular system is thought to occur in the late stages of disease and lead to systemic spread of these viruses to other organs , including brain and kidney [13] , [27] . In our model , systemic spread of the virus was indicated by similar titers and replication kinetics of NiV in directly inoculated lung grafts and grafts not directly injected with virus . This suggests that following infection of the lung , NiV quickly becomes viremic and spreads to other organs . In addition , systemic infection through intraperitoneal injection with NiV also resulted in infection of the human lung grafts , thus confirming hematogenous spread of the virus . Interestingly , NiV infection in NSG mice engrafted with human lung tissues did not result in clinical signs despite evidence of replication in mouse organs , including lung and brain . Previous studies have shown that NiV infection in type I IFN receptor knock-out mice and aged mice is lethal [28] , [29] . Aged mice have been shown to mount an aberrant IFN response [30] . While the NSG mice are immunocompromised , they can mount an IFN response , which seems sufficient to protect against lethal disease in our studies . Alternatively , it is possible that NSG mice are not susceptible to NiV and that the virus measured in mouse organs is only found in the blood . Also , the small foci of viral antigen detected in mouse lungs could correspond to emboli of infected human cells that slough off from the infected grafts . Several cytokines and especially IL-6 , IP-10 and VEGF were upregulated during NiV infection in the human lung . Interestingly , the levels of cytokines observed in our xenograft lung model are similar to those observed in the lungs of fatal cases of influenza virus A ( H1N1 ) [31] . Upregulation of inflammatory mediators such as TNF-α , IP-10 , IL-1β and IL-6 in the lungs was previously shown to play a role in the pathogenesis of lethal NiV in hamsters [13] , as well as in the development of ALI with other respiratory virus infections , including SARS-CoV and influenza virus ( H5N1 ) [30] , [32] , [33] . VEGF has an important role in ALI pathogenesis by acting as a growth factor and increasing vascular permeability [34] . We previously showed that VEGF is expressed by human respiratory epithelial cells during NiV infection [12] . This suggests that VEGF may be partly responsible for the increased pulmonary hemorrhage , endothelial destruction , and alveolar remodeling in an emphysema-like phenotype as observed in our model . Since these inflammatory mediators also play an important role in the recruitment of immune cell , our data suggests that inflammation could be observed in this model when human immune cells are present . In addition to implanting human lung xenografts , the NSG mice have been used to engraft the human hematopoietic system to study hematopoiesis , immunity , inflammatory disease and human-specific pathogens . This humanized NSG mouse model routinely contains >25% human CD45+ cells in the peripheral blood 12 weeks post engraftment of hematopoietic stem cells [35] . Many of the inflammatory mediators expressed in the current study play an important role in immune cell recruitment [36] , [37] . The ability to engraft human immune cells will allow us to study the effect of these mediators on specific immune cells populations . Future studies will make use of the fully humanized lung xenograft model to study the role of the inflammatory response in the pathogenesis of the different henipavirus strains in the human lung . In conclusion , these data confirm that the human lung is highly susceptible to NiV infection . NiV is capable of replicating to high titers in the human lung and targets both respiratory epithelium and endothelium . Infection results in the characteristic syncytial formation and extensive lung damage . Key inflammatory mediators such as IL-6 , IP-10 , G-CSF and GM-CSF are expressed during infection . This model will allow for more detailed studies of the pathogenesis of respiratory disease caused by henipavirus infection . Furthermore , these data point to several inflammatory mediators that potentially play critical roles in henipavirus pathogenesis , which may be valuable as candidates for future studies of the mechanism of henipavirus pathogenesis and as potential targets for treatment .
Approval for animal experiments was obtained from the Institutional Animal Care and Use Committee , University of Texas Medical Branch ( protocol number 0905041 ) . Animal work was performed by certified staff in an Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) approved facility . Animal housing , care and experimental protocols were in accordance with NIH guidelines of the Office of Laboratory Animal Welfare . Discarded tissue from deceased human fetuses was obtained via a non-profit partner ( Advanced Bioscience Resources , Alameda , CA ) as approved under exemption 4 in the HHS regulations ( 45 CFR Part 46 ) . Need for informed consent was waived by the UTMB Institutional Review Board . NiV ( Malaysia strain ) was kindly provided by the Special Pathogens Branch of the Centers for Disease Control and Prevention , Atlanta , Georgia , United States . The virus were propagated on Vero cells in Dulbecco's minimal essential medium supplemented with 10% fetal calf serum ( Hyclone , Logan , UT ) , L-glutamine , penicillin and streptomycin at 37°C in a humidified CO2 incubator ( 5% ) . All infectious work was performed in a class II biological safety cabinet in a biosafety level 4 laboratory ( BSL4 ) at the Galveston National Laboratory . NOD . Cg-Prkdcscid Il2rgtm1Wjl/SzJ mice , also known as NOD/SCID/γcnull or NSG mice ( Jackson Laboratories ) , 3–5 weeks of age , were housed in a sterile microisolator environment . Mice were engrafted with human lung tissue ( Advanced Bioscience Resources , Alameda , CA ) . Six fragments of human fetal lung from the same donor ( 15–19 weeks of age ) were sutured to muscle fascia in the dorsal subcutaneous space in each mouse ( ∼0 . 5 cm from the spine , three on each side ) . Animals received appropriate post-surgery treatment including antibiotics and analgesics . Twelve weeks post-engraftment , animals were transferred to an ABSL-4 facility . Prior to infection , animals were anesthetized by chamber induction ( 5 Liters/min 100% O2 and 3–5% isoflurane ) . Three of the six lung tissues were inoculated via intragraft injection of 105 TCID50 NiV in a 50 µl volume . Animals were monitored daily for weight loss and clinical signs . Groups of 3 animals were euthanized on days 1 , 3 , 5 , 7 and 10 post infection , and samples for virus isolation and histological examination were procured from whole blood ( EDTA vacutainer ) , human lung tissues , and mouse liver , spleen , kidney , lung , heart and brain . In a separate experiment , 2 animals were injected via the intraperitoneal route with 105 TCID50 NiV and euthanized 10 days post infection . Control groups were NSG mice without a lung xenograft and challenged via the intradermal route with 105 TCID50 NiV on the back of the mouse at the same location the human lung xenografts would be . Whole blood was tested for presence of infectious virus by 10-fold diltutions as described below . Tissue samples were weighed and homogenized in 10 equivalent volumes of DMEM to generate a 10% solution . The solution was centrifuged at 10 , 000 rpm under aerosol containment in a table top centrifuge for 5 min to pellet insoluble parts . Virus titration was performed using a TCID50 assay on 96-well plates ( 1×104 Vero cells per well ) with 100 µL inocula ( cleared homogenate or whole blood ) from 10-fold serial dilutions . Plates were incubated for 3 days at 37°C , and wells were scored for cytopathic effect ( CPE ) . Virus concentrations were calculated as TCID50 per gram of tissue . All tissue samples were immersion-fixed in 10% neutral buffered formalin for at least 7 days under BSL4 conditions . Prior to removal from the BSL4 laboratory , formalin was changed and specimens were processed under BSL2 conditions by conventional methods , either embedded in paraffin , sectioned at 5 µm thickness and stained with hematoxylin and eosin ( H&E ) or embedded in Tissue Tek and frozen sections cut at 3–8 µm thickness and used for immunofluorescent ( IF ) staining . Tissues for immunohistochemistry ( IHC ) were stained as previously described using a rabbit anti-NiV-nucleoprotein ( N ) antibody ( kindly provided by Dr . C . Broder , Uniformed Services University , Bethesda , MD ) [13] . Tissues for IF were stained with a rabbit anti-NiV-N antibody , a biotinylated anti-CD31 ( eBioscience ) , anti-collagen IV labeled with Alexa 647 ( eBioscience ) , anti ephrin B2 ( Santa Cruz Biotechnology ) or ephrin B3 ( R&D Systems ) and Hoechst for nuclear staining . NiV N in mouse tissue could only be detected following immunofluorescent staining , likely due to the limit of detection by conventional IHC . An Alexa 546 labeled secondary antibody ( Life Technologies ) was used for detection of the anti NiV N antibody as well as anti ephrin B2 and B3 antibodies and an Alexa 488 conjugated streptavidin ( R&D Systems ) was used for detection of the anti-CD31 antibody . Cytokine/chemokine concentrations in the homogenates of NiV infected human lung tissues were determined using a Milliplex Human Cytokine PREMIXED 28 Plex Immunoassay Kit ( Millipore , Billerica , USA ) . Prior analysis , samples were inactivated on dry ice by gamma-radiation ( 5 MRad ) . The assay was performed according to the manufacturer's instructions . The concentration of the following 28 cytokines were determined using the Bio-Plex 200 system ( BioRad ) : Epidermal Growth Factor ( EGF ) , Granulocyte-Colony Stimulating Factor ( G-CSF ) , Granulocyte Macrophage-Colony Stimulating Factor ( GM-CSF ) , interferon ( IFN ) -α2 , IFNγ , Interleukin ( IL ) -1α , IL-1ß , IL-2 , IL-3 , IL-4 , IL-5 , IL-6 , IL-7 , IL-8 , IL-10 , IL-12 ( p40 ) , IL-12 ( p70 ) , IL-13 , IL-15 , IL-17A , chemokine ligand 3-like 1 ( CCL3L1 or MIP-1α ) , chemokine ligand 4 ( CCL4 or MIP-1ß ) , chemokine ligand 10 ( IP-10 or CXCL10 ) , chemokine ligand 11 ( CCL11 or Eotaxin-1 ) , chemokine ligand 13 ( CCL13 or MCP-1 ) , Tumor Necrosis Factor ( TNF-α ) , Lymphotoxin alpha ( TNFß ) and Vascular Endothelial Growth Factor A ( VEGF ) . | Nipah virus ( NiV ) is a highly pathogenic zoonotic virus that causes fatal disease in humans and a variety of other mammalian hosts including pigs . Given the lack of effective therapeutics and vaccines , this virus is considered a public health and agricultural concern , and listed as category C priority pathogen for biodefense research by the National Institute of Allergy and Infectious Diseases . Both animal-to-human and human-to-human transmission has been observed . Studies on the molecular mechanisms of NiV-mediated pathogenesis have been hampered by the lack of biologically relevant in vivo models for studying the initial host responses to NiV infection in the human lung . We show here a new small animal model in which we transplant human lung tissue for studying the pathogenesis of NiV . We showed that NiV can replicate to high levels in the human lung . NiV causes extensive damage to the lung tissue and induces important regulators of the inflammatory response . This study is the first to use a human lung transplant for studying infectious diseases , a powerful model for studying the pathogenesis of NiV infection , and will open up new possibilities for studying virus-host interactions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"animal",
"models",
"of",
"infection",
"infectious",
"diseases",
"veterinary",
"diseases",
"medicine",
"and",
"health",
"sciences",
"emerging",
"infectious",
"diseases",
"pathology",
"and",
"laboratory",
"medicine",
"host-pathogen",
"interactions",
"veterinary",
"virology",
"virology",
"emerging",
"viral",
"diseases",
"biology",
"and",
"life",
"sciences",
"microbiology",
"pathogenesis",
"veterinary",
"science"
]
| 2014 | A Human Lung Xenograft Mouse Model of Nipah Virus Infection |
The dynamics of growth of bacterial populations has been extensively studied for planktonic cells in well-agitated liquid culture , in which all cells have equal access to nutrients . In the real world , bacteria are more likely to live in physically structured habitats as colonies , within which individual cells vary in their access to nutrients . The dynamics of bacterial growth in such conditions is poorly understood , and , unlike that for liquid culture , there is not a standard broadly used mathematical model for bacterial populations growing in colonies in three dimensions ( 3-d ) . By extending the classic Monod model of resource-limited population growth to allow for spatial heterogeneity in the bacterial access to nutrients , we develop a 3-d model of colonies , in which bacteria consume diffusing nutrients in their vicinity . By following the changes in density of E . coli in liquid and embedded in glucose-limited soft agar , we evaluate the fit of this model to experimental data . The model accounts for the experimentally observed presence of a sub-exponential , diffusion-limited growth regime in colonies , which is absent in liquid cultures . The model predicts and our experiments confirm that , as a consequence of inter-colony competition for the diffusing nutrients and of cell death , there is a non-monotonic relationship between total number of colonies within the habitat and the total number of individual cells in all of these colonies . This combined theoretical-experimental study reveals that , within 3-d colonies , E . coli cells are loosely packed , and colonies produce about 2 . 5 times as many cells as the liquid culture from the same amount of nutrients . We verify that this is because cells in liquid culture are larger than in colonies . Our model provides a baseline description of bacterial growth in 3-d , deviations from which can be used to identify phenotypic heterogeneities and inter-cellular interactions that further contribute to the structure of bacterial communities .
In 1942 , Jacques Monod developed a mathematical model of bacterial growth in a liquid culture , where the bacterial cells and nutrient molecules were homogeneously distributed [1 , 2] . A simple ordinary differential equation was accurate enough to account for the exponential growth of bacteria and their ascent into stationary phase following the exhaustion of the limiting resource . The model has proven to be long-lived since most experimental studies of the population dynamics of bacteria are in liquid culture [3 , 4] . In contrast , outside the tubes , flasks , and chemostats of laboratory culture , bacteria most commonly live in physically structured habitats as colonies or microcolonies . Such colonies are heterogeneous; at a minimum , cells vary in their access to nutrients depending on their position within the colony and thereby divide at different rates . The majority of research directed at understanding structured bacterial population growth has been confined to two dimensional ( 2-d ) surfaces [5–10] , including studying the interplay of evolution and the physical structure [11 , 12] , or analyzing effects of mechanical interactions in an expanding colony [13–15] . However , diffusion in two dimensions is different from three , making it easier to form diffusion-limited instabilities [5 , 16 , 17] . In 3-d , work has been done to understand nutrient shielding of the interior of a colony by the microbes on the surface , treating them as individual agents [18] . In addition , in the context of modeling biofilms , there exist many complex models that account for mechanical stresses , adhesion properties , fluid flows , fluxes of multiple metabolites and waste product , and so on ( see Refs . [19–26] for just a few examples ) . These models typically involve many free parameters , not all of them experimentally constrained . Their complexity prevents analytical treatments , so that most such models are formulated in terms of agent-based or cellular automata approaches . As a result of this complexity , very few of these models have provided analytical insights , or have been compared to experiments quantitatively . In summary , we are not aware of 3-d models of colony growth that account for the spatially varying density of nutrients and bacteria , explain the observed experimental phenomenology of bacterial growth in such colonies , and do so in a relatively simple coarse-grained ( PDE ) Monod-style manner , rather than relying on complex agent-based simulations of individual bacteria . Here we develop such a model that treats the growth rate heterogeneity due to the non-uniform nutrient distribution produced self-consistently by consumption of a nutrient by the bacteria . We explore its fit experimentally with the growth of E . coli maintained and growing as colonies embedded in 3-d matrix of soft agar with an initially uniform spatial distribution of a limiting carbon source , glucose ( Fig 1 ) . We compare dynamics of growth of bacteria in colonies with that of planktonic cells in liquid culture with the same concentration of limiting glucose . In our model , we assume that the colony is essentially unconstrained by the soft agar and is free to expand , and the bacteria within it are non-motile . This combined theoretical-experimental study reveals two surprising features of bacterial populations growing as colonies: ( i ) the bacteria within these structures exist as loosely packed viable cells , and ( ii ) the viable cell densities of bacteria produced in colonies is more than two-fold greater than that in liquid cultures with the same concentration of the limiting glucose .
We use population growth of E . coli in minimal medium as the basis for developing the model . To control the amount of nutrients available to the bacteria , we use glucose at the initial concentration 0 . 2 mg/ml , at which it limits the stationary phase density of E . coli produced as planktonic cells and as colonies in soft agar . We grow bacteria either in liquid cultures or as three-dimensional colonies embedded in soft agar ( Fig 1 ) . Unless otherwise noted , 3-d colonies are grown in 3 ml of soft agar , inoculated with approximately 50 bacteria/ml . Under these conditions , each colony has an access to a nutrient subvolume of v ∼ 1/50 ml , or , on average , a nutrient sphere of radius R = ( 3v/4π ) 1/3 ≈ 1 . 7 mm . For the liquid and the 3-d growth , we estimate the density of viable cells , N ( t ) , at different times diluting and plating and then counting the number of resulting colonies ( colony-forming units , or CFU ) , see Methods for details . For each time point , we obtain 6 independent replicates of CFU density estimates , and each experiment was replicated at least 3 times . The results of these population growth experiments are shown in Fig 2 ( data points ) . In liquid , the density of the population increases exponentially , and then abruptly stops and begins to decline at a low rate , presumably because the bacteria consume the available glucose and enter the stationary phase , at which time the rate of cell mortality exceeds that of division . In contrast , in 3-d colonies , the exponential growth and the stationary phase are separated by a gradual decline in the net rate of growth . We expect that this is because the growth of the population here is limited by the speed with which diffusion brings glucose from the periphery of the available nutrient volume to the colony , where it is consumed by the bacteria . Surprisingly , the maximum density of bacteria growing as colonies is substantially greater than that in liquid , despite the concentration of the limiting glucose being equal for liquid and the soft agar . To understand these findings quantitatively , we now develop a simple ( minimalist ) mathematical model of resource-limited bacterial growth in liquid and as spatially structured colonies . Our liquid culture model of bacterial growth is a variant of that of Monod [2] . In this model , all bacteria have the same resource ( glucose ) concentration dependent growth rate g ( ρ ) = gmax ρ/ ( ρ + K ) , where ρ is the concentration of glucose , gmax is the maximum growth rate , and K , the Monod constant , is the concentration of the resource when the growth rate is half its maximum value gmax/2 . With these parameters , the rate of change of the density of the bacterial population n = N/v is given by d n d t = n Θ ( t - τ lag ) g max ρ ρ + K - n m , ( 1 ) d ρ d t = - 1 a l n Θ ( t - τ lag ) g max ρ ρ + K . ( 2 ) Here v is the volume of the liquid where the culture grows , and al is the liquid yield , which measures the number of bacteria produced by a microgram of the nutrient . Further , Θ ( t − τlag ) is the Heaviside Θ-function , which is equal to zero for t < τlag , and to unity otherwise . It represents the lag phase before the growth starts after a transfer to a new environment . Note that Eqs ( 1 ) and ( 2 ) differ slightly from the standard Monod model . Specifically , we added a small constant death rate m to account for the decrease of the population in the liquid culture after the saturation ( Fig 2 ) . Thus the population has a zero net growth at a critical nutrient density of ρm = mK/ ( gmax − m ) , which represents the minimum nutrient concentration needed to sustain life without growth [27] . We fit the five growth parameters ( gmax , K , al , τlag , and m ) to the experimental data using nonlinear least squares fitting , and estimate uncertainties of the fit using bootstrapping ( see Materials and methods , and also Table 1 ) . As seen in Fig 2 ( blue curve ) , after the lag phase , the population increases exponentially before it saturates abruptly when all the cells in the colony run out of food at the same time . The excellent agreement between the experiments and the model is encouraging . It allows us to use the Monod model with death as the basis for 3-d studies . To develop the minimal model of 3-d growth , we assume that the bacteria within the colony are physiologically identical , but depending on their position , vary in their access to the diffusing carbon source . Thus all cells grow according to the Monod model , differing only by the local availability of the limiting nutrient , glucose , ρ ( x , y , z ) . For 3-d colonies , the dynamics of the bacterial density n ( x , y , z ) is a result of a complex interplay between mechanical properties of the extracellular colony matrix and the substrate , in which the colony grows , the stiffness of the bacterial wall , and the growth properties of bacteria themselves . Mathematical models that account for these complexities are typically over-parameterized , making it hard to make precise quantitative predictions [23 , 24] . In contrast , here we aim at building the simplest possible model consistent with data . We assume that soft agar is too soft to provide mechanical resistance to the colony , but sufficiently dense to keep cells from moving . Thus the colony would grow in density until bacteria get as close-packed , on average , as possible given the amount of the extracellular matrix they secrete . We denote this maximum cell number density as μ . Having reached the maximum density , any new cellular divisions must expand the colony into the soft agar , with the same fixed maximum density in the volume occupied by the colony , save for possibly smaller density at the colony edge . Further , as seen in Fig 1 , the colony spreads spherically-symmetrically , so that the density of the cells and the nutrient concentration are functions of the radius and the time only , n ( r , t ) and ρ ( r , t ) . Thus we have ∂ n ( r , t ) ∂ t = n ( r , t ) [ Θ ( t - τ lag ) g max ρ ( r , t ) ρ ( r , t ) + K - m ] , ( 3 ) ∂ ρ ( r , t ) ∂ t = D ∇ 2 ρ - 1 a c n ( r , t ) Θ ( t - τ lag ) g max ρ ( r , t ) ρ ( r , t ) + K , ( 4 ) with the initial uniform spatial concentration of the nutrient ρ ( r , 0 ) = ρ0 at time t = 0 , and a single bacterium starting the colony at r = 0 . In these equations , D is the nutrient ( glucose ) diffusion coefficient . Further , we allow for the yield in the colony ac to be different from the liquid yield al to account for the different saturation values in Fig 2 , as further discussed below . Importantly , since the agar is more than 99% liquid , the four other growth parameters gmax , K , τlag , and m are taken to be the same in both media . Bacterial density in Eq ( 3 ) would grow to infinity with time , which is physically unrealistic . Thus we need to bound the cell density from above by a maximum value , corresponding to maximally packed cells ( and extracellular matrix ) that can be compressed no further . To do this , and to establish such maximum cellular packing density μ in Eq ( 3 ) , we now impose that the overall increase in cell number beyond the maximum density leads to the proportionate growth of the colony radius rc , so that N ≡ 4 π ∫ d r r 2 n ( r , t ) = ( 4 / 3 ) π r c 3 μ . In other words , at each point in time , we impose the condition that n ( r , t ) = { μ , 0 < r ≤ r c = ( 3 N / 4 π μ ) 1 / 3 , 0 , r c < r ≤ R , ( 5 ) where R = ( 3v/4π ) 1/3 is the radius of the nutrient subvolume accessible to the colony . To reconcile Eqs ( 3 ) and ( 5 ) , we say that all new growth is accounted for by the expansion of the colony edge , rc ( t ) , while the death results in a decrease in the cell density locally ( see Materials and methods for description of the algorithm for simulating this growth model ) . This is reminiscent of earlier hybrid differential-discrete simulations [28 , 29] . However , we note that , in our model , the biomass changes differentially: the cell density in each spherical shell is a real number , and it is not necessarily equal to μ in either the largest shell ( due to growth ) or in all other shells ( due to cell death ) . In fact , we emphasize that one should not view the colony growth as biomass transfer , but rather as growth in the inner shells creating pressure that expands the colony radius continuously . It might be possible to write this mass dynamics as an integro-differential equation . However , for the purpose of solving the model and comparing to experiments , an integro-differential equation will not be more useful than the explicit dynamical rules that we have provided . Although the assumption of redistribution of new growth into a spherically symmetric front of a growing colony will likely be violated for a generic colony or biofilm , it is clearly satisfied for colonies in our experiments ( cf . Fig 1 ) . One would expect violations of the symmetry if there are nearby colonies competing for the same nutrients , and thus partially shielding each other . However , in our experiments , colonies are either well separated and hence weakly interacting ( small initial bacterial densities ) , or there are many colonies in arbitrary directions from each other ( large bacterial densities ) ; in both cases , the approximate spherical symmetry is restored ( especially since we average over multiple colonies before counting CFUs ) . Thus there are no obvious reasons to go beyond the spherical symmetry assumption . In fact , we will see that predictions of this simple model will be verified against new experimental data , further confirming the spherical symmetry assumption a posteriori . As mentioned above , there are a lot of models in the literature describing growth of bacterial communities in spatially structured environments . [19 , 23 , 24] . However , we have not found any readily available 3-d spherically symmetric soft agar colony models , where every included biological process or physical feature is essential to the population biology of the colony . This necessitated development of our model in this section . To illustrate the behavior of the 3-d model of bacterial growth as colonies , we plot numerical solutions of Eqs ( 3 ) – ( 5 ) for different values of the nutrient diffusion coefficient in Fig 3 ( A ) . Especially at small D , two different growth regimes are clearly visible after the lag but before the ultimate saturation and the slow cell death . The first is the fast exponential growth based on local , immediately accessible resources . This regime is indistinguishable from the growth in liquid . When the local nutrients are depleted at a certain time τ1 following the start of the growth at τlag , new nutrients must be brought from afar by diffusion . This is slow , resulting in a slower diffusion-limited growth regime . Here the overall colony growth rate is an average over cells growing at different rates due to different concentrations of the locally accessible nutrient . Our numerical solutions suggest that , in this regime , the nutrient concentration at the colony edge decays exponentially fast , in agreement with Ref . [18] , cf . Fig 3 ( B ) . The nutrient penetration depth is only a few μm , or a few cell layers . Therefore , in the diffusion-limited regime , there are , essentially , no nutrients deep inside a colony , and only cells at the periphery can grow . In the absence of resource storage [30] , nutrient sharing from the outer cells , or cannibalism ( we model none of these ) , interior cells would not grow at all and will eventually die . The diffusion-limited growth regime finally ends with saturation and slow death when most of the nutrients in the accessible subvolume are depleted at time τ2 after τlag . The onset of the saturation takes longer than in liquid since small ( but larger than ρm ) amounts of the nutrient linger at the far edges of the nutrient subvolume for a long time . Analytical expressions for τ1 , τ2 , and the growth dynamics can be obtained from the following arguments . First , in the exponential growth regime , the population grows as N ∼ e g max t . This requires e g max t / a c of the nutrient mass , which must come from the volume immediately accessible by diffusion , equal to ∼ ρ 0 ( D t ) 3 . Equating the two expressions gives , to the leading order , τ 1 ∼ g max - 1 log [ ρ 0 a c ( D / g max ) 3 / 2 ] . When local resources are exhausted , growth is limited by nutrients diffusing in from the volume ∼ ( D t ) 3 . However , because the encounter probability for a 3-d random walk is less than one [31] , most of the nutrient molecules coming from afar will not be immediately absorbed . In fact , since the box-counting dimension of a diffusive process is two , only ∼ ρ 0 ( D t ) 2 r c nutrient molecules will be captured in time t , resulting in N ∼ ρ0 Dtrc ac . On the other hand , the radius of the colony grows as rc = ( 3/4π ) 1/3 ( N/μ ) 1/3 . Combining these expressions gives N ∼ [ ( ac ρ0 D ) 3/μ]1/2 t3/2 in the diffusion-limited regime . Finally , the total amount of nutrients available to the colony is ∼ρ0 R3 , and so the diffusion-limited growth will saturate , and the cells will start dying with the rate of m when the colony grows to N ∼ ac ρ0 R3 , which occurs at τ2 ∼ ( μ/ac ρ0 ) 1/3 R2/D . Altogether , we find N ∼ { const , t < τ lag , e g max t , t − τ lag ≪ τ 1 ∼ log [ ρ 0 a c ( D g max ) 3 / 2 ] g max , [ ( a c ρ 0 D ) 3 μ ] 1 2 t 3 / 2 , τ 1 ≪ t − τ lag ≪ τ 2 ∼ ( μ a c ρ 0 ) 1 3 R 2 D , a c ρ 0 R 3 e − m t , τ 2 ≪ t − τ lag , ( 6 ) These analytical estimates are supported by the numerical solutions in Fig 3 ( A ) . We note that in one or two dimensions , the diffusion limited growth would scale as N ∝ td/2 for dimension d , independently of the ( small ) colony radius , or even for a point colony , since the random walk encounter probability there is one [31] . In contrast , our three-dimensional result depends critically on knowing how the radius of the colony scales with the number of growing bacteria . In particular , here we cannot model the colony as a point-like object . Thus the exponent of the power law scaling is not universal in 3-d , and it may change for heterogeneous colonies with varying cell size and cell density . To determine the extent to which our minimal model accounts for the dynamics of growth of bacteria in colonies , we fit the model to data using nonlinear least squares fitting , similar to the liquid case . We keep the parameters al , K , gmax , m , and τlag equal to the values inferred for liquid , and only optimize D , μ , and ac for the 3-d culture data . See Materials and methods for the details of the fits , including estimation of the prediction uncertainty using bootstrapping . Table 1 shows fitted parameter values with the corresponding nominal values from the literature . The fitted parameters are consistent with the nominal values where the latter are known . A possible exception is the value of the glucose diffusion coefficient D , which is lower than those reported in previous publications ( though the confidence interval on our fits is rather large ) . This could be a result of the previous measurements done in hydrogels , rather than in 0 . 35% agar preparation used in this study . Further , the best fit curve shows an excellent agreement with data ( cf . Fig 2 , red ) , and the prediction confidence bands are very narrow ( cf . Fig 4 ) . This suggests that nutrient diffusion and the ensuing geometric heterogeneity of growth are sufficient to explain the population dynamics of these E . coli colonies in 3-d at our experimental precision , and consideration of additional phenotypic inhomogeneities is not needed . Our analysis also provides estimates of two previously unknown parameters , μ ( the maximum packing density of viable cells ) and ac ( yield in 3-d colonies ) . The inferred packing density is μ = 3 . 0 ⋅ 10−2 CFU/μm3 , with the 80% confidence interval of [1 . 7 , 4 . 2] ⋅ 10−2 CFU/μm3 . Since an E . coli cell has a volume of between 0 . 5 and 2 μm3 [32 , 33] , this suggests that only about ∼3% of all space in a colony is occupied by viable cells . This is a surprising finding , and it requires an independent corroboration . Towards this end , we measure radii of large colonies and calculate their packing densities by diving colony volumes by the average CFUs per colony . This gives μ = 1 . 5 ± 0 . 08 × 10−2 CFU/μm3 , consistent with our estimation of μ from the fitted growth model . In other words , in our experiments , viable E . coli cells are sparsely packed . Notice that here we only say that , in large colonies , there is a small density of viable cells that grow into visible , countable colonies when plated . There could be many other cells , which , for whatever reason , do not grow into large colonies after plating . Without additional investigations , we cannot make the distinction . Also notice that we can only make this claim for large , old colonies which is when most cells in growing colonies emerge . In particular , the low density claim is not valid during early growth , when each colony starts with an individual cell , which by definition takes 100% of the colony volume . The second inferred parameter is ac . We find that the yield as measured by the ratio of the CFU estimated stationary phase density and the quantity of glucose in 3-d is 2 to 3 times higher than that in liquid culture , ac > al ( cf . Table 1 ) . This implies that , at saturation , colonies produce more CFUs than liquid cultures , which is directly apparent from Fig 2 . This is a surprising result , since in the colony the bacteria grow more slowly and there is more time for cell death . Nonetheless , similar results have been reported for colonies growing on surfaces [34] . Here this effect is likely a direct consequence of the growth dynamics during the diffusion-limited regime . Indeed , E . coli cells growing at a rate of >1 hr−1 grow to be 2 to 3 times larger than cells growing at a rate of <0 . 1 hr−1 [35] . While the diffusion limited regime lasts only for a few hours ( cf . Fig 2 ) , more than 90% of all cells emerge at that time , so that the majority of cells in the colony are smaller than in liquid , yielding more cells from the same nutrient amount . As an independent test of the developed 3-d growth model , we use it to predict results of experiments distinct from those used for fitting the model . Specifically , we investigate how the population size depends on the density of bacteria used to inoculate the soft agar . At a long measurement time ( 72 hrs ) , our model predicts a non-monotonic dependence of the population size on the inoculation density ( cf . Fig 5 , dashed line ) . This is because , at very low densities , each colony has access to a large nutrient subvolume , and the colony cannot clear this subvolume by diffusion in just 72 hrs . As a result , at the end of the experiment , there are still nutrients in the media , and the colony does not reach its maximum size . In contrast , at very high inoculating densities , colonies rapidly exhaust their small available nutrient subvolumes , the cell death becomes important throughout much of the experiment duration , and the population is smaller again . Thus the population reaches its maximum at intermediate densities , where these two effects balance . We test this prediction by experimentally measuring population sizes at 72 hrs for E . coli growing in soft agar at inoculums varying from 101 to 105 cells/ml As seen in Fig 5 , the experimental data agree with the prediction within errors and , in particular , exhibit the expected non-monotonicity . We emphasize that no additional fitting was done for this figure , and yet the agreement between the experiment and the theory is very good . Our simplest 3-d spherically symmetric colony growth model has been able to fit bacterial population dynamics data remarkably well . To further challenge the model and to suggest its possible future improvements , we now use microscopy ( see Materials and methods ) to collect additional data that is of a very different nature compared to the data used for building the model . We use these new experiments to further investigate the most salient prediction of the model , namely the difference in the liquid vs . colony yield . According to our fits , the bacterial yield per microgram of glucose in 3-d colonies is 2 to 3 times higher than that in the liquid culture . We proposed that this is because the cells in liquid grow faster ( and hence are larger ) than those in colonies . To verify this directly , we measured the cell size ( length ) in these different growth conditions as a function of the time since inoculation ( see Materials and methods ) . For liquid , the mean cell length at 6 hours post-inoculation was 1 . 9 ± 0 . 7 μm . For older cultures , we have no way of distinguishing young and old cells , and so the distribution of cell sizes includes both cells that were born in earlier stages of the experiment , as well as recently . Crucially , as the cultures grew older , long , filamentous cells emerged ( for old cultures , the longest cells were > 100 μm ) . While the fraction of such extremely long cells was small , this tail of the cell size probability distribution [42 , 43] had a pronounced effect on the mean cell length , increasing it to ∼5 μm for the oldest cultures ( Fig 6 ) . To account for this long tail , we report both the mean and the median cell sizes , as well as the fraction of cells that remained non-filamentous ( defined as < 5 μm in length ) . As seen in Fig 6 , the number of short cells stabilizes near 60 − 70% for the oldest cultures . At the same time , the median cell size in liquid does not depend on the culture age , hovering around 2 μm . In contrast , the distribution of cell sizes in colonies is much less skewed . Less than 1% of the sampled cells become filamentous at long times ( Fig 6 ) , so that the mean and the median cell lengths are nearly equal . The average cell size drops when the diffusion-limited growth starts , and it saturates near 1 . 5 μm for very old colonies . Combining these measurements , the size ratio of cells grown in the liquid and in colonies is between ∼1 . 6 ( for the median length ) and ∼3 . 4 ( for the mean length ) , in agreement with the population biology estimate above , again validating our model . Crucially , these experiments also suggest that the immediate next modification of our growth model should not be inclusion of mechanical stresses and more complicated nutrient and waste product fluxes , but rather the growth-speed dependence of the yield , a = a ( g ( ρ ) ) , which would replace the two parameters al and ac with a single function and unite the two growth models .
To our knowledge , the model developed here is the first continuous , rather than agent-based , model to explicitly study bacterial growth as colonies . We consider this the minimal model because it assumes spherical symmetry and that the availability of nutrients ( a carbon source ) is the sole factor determining the rate of cell division within colonies . In reality , the cellular growth , division , and death rates would also depend on cell-to-cell interactions of various sorts , on the enrichment and deterioration of the environment due to the buildup of secondary metabolites and waste , on cell-environment mechanical interactions , and on diverse cellular phenotypic commitments . The model we developed and experimentally tested here only accounts for the spatial heterogeneity in access to the diffusing nutrient and assumes no such additional effects [30 , 44 , 45] . Nevertheless , despite these limitations , with five parameters describing the growth in liquid , and three additional parameters specific to the 3-d spherically symmetric colony growth , this model provides an impressively accurate description of growth of populations of E . coli as colonies in soft agar as well as planktonic cells in liquid . Unlike the anticipated and observed nearly precipitous termination of growth in liquid culture as nutrients become depleted , our 3-d model accounts for the experimentally observed gradual reduction in net rate of replication as diffusion of the resource increasingly limits colony growth with time . With no additional fitting , the model also correctly predicted the non-monotonic , upside-down U shaped dependence of the population size on the inoculating bacterial density . Moreover , all of the best-fit parameters inferred from the data agreed with prior estimates in the literature , where these are available ( see Table 1 and references therein ) , indicating high-quality fits without overfitting . Our study has revealed and/or confirmed several intriguing observations about bacteria growing in colonies . First , the growth in colonies yielded substantially greater viable cell densities than obtained in liquid culture with the same concentrations of limiting carbon source . We proposed that this was a direct consequence of the diffusion-limited growth , which happens at a slower division rate . In turn , slow division is correlated with smaller size of bacterial cells [35] , resulting in more bacteria for the same nutrient amount . This slowing down is very important phenotypically—according to our model , over 90% of all bacteria in the colony are formed at such decreased growth rate , and the yield ac is an average over yields at different stages of the slowing . We have directly verified this hypothesis by measuring cell sizes in liquid and colony cultures as a function of the time since inoculation . These new data provided a confirmation of our population biology estimates of the yields and of the hypothesis behind their difference . The experiments also suggest a natural future extension of our model , which would come from measuring the dependence of the cell size and the yield on the growth rate and then verifying if both the liquid and the colony growth can be described by the same dependence a = a ( g ( ρ ) ) . Our second intriguing observation , which is supported by two independent sets of measurements , is that the packing density inside colonies is very low , μ ∼ 0 . 03 CFU/μm3 , so that the vast majority of a volume of a colony is not occupied by viable cells . The accuracy of this observation depends strongly on whether , when the 3-d cultures are liquified and plated for counting , cells get perfectly separated from each other , and the counted colonies start from individual cells . We notice that , when we put liquified cultures under a microscope to produce data for Fig 6 , we observe that the cells are well-separated . Further , if low μ was a result of us underestimating the number of cells in the colony , it would mean that ac must be even larger than our current estimate . Thus one would be able to make even more cells from the same nutrient amount in 3-d cultures . This is extremely unlikely since we additionally measured the cell lengths in liquid and in 3-d colonies ( Fig 6 ) , with the ratio of cell lengths agreeing with the independently estimated ratio al/ac . Thus we are confident in our estimate of μ . What could be the reasons for its small value ? It is possible that the colonies are , indeed , largely void of viable cells , with extracellular fluids and matrix fibers filling in the gaps . Another possibility is that cells deep inside the colony are dead or dormant due to the absence of nutrients , or due to other effects , such as mechanical stresses , so that the viable cells that we measure are a minority of all the bacterial cells that existed . Our experiments show no evidence for such deviations from the minimal growth model , but it is clear that additional studies , including direct imaging of the colony structure , must be done in the future . One interpretation of the close fit between the predictions of this minimal model and the results of our soft-agar experiments is that heterogeneities beyond nutrient access contribute little to the growth dynamics of bacteria in colonies . It remains to be tested how general this result is . Is the E . coli in glucose-limited minimal medium used in this experiment exceptional ? Is the spherically symmetric growth special ? Will the results hold for other bacterial species and for complex media , like broth ? We propose that the minimal model developed here be used as a baseline to address such questions of generality with other bacteria , geometries , and media . Models are most useful when they do not fit data and thus point to other factors contributing to the studied dynamics . For growth of bacteria in colonies , such factors can be mechanical or other stresses , cell-cell interactions , and others . From an evolutionary perspective particularly intriguing would be studies of growth of bacteria in colonies initiated with multiple cells of different genotypes ( or even species ) , where deviations from the model could signal such important phenomena as clonal competition or cooperation within a clone .
We used Escherichia coli B , ara rpsL T6 r-m- , originally obtained from Seymour Lederberg [38] . We employ one of the evolved strains , REL1976 , generously provided by Richard Lenski [46] . Overnight cultures were grown in Lysogeny Broth ( LB ) , Becton Dickinson ( Franklin Lakes , NJ , USA ) , diluted in 0 . 85% saline and introduced into in liquid or into 0 . 35% Bacto agar with Davis minimal salts [47] supplemented with 0 . 2mg/ml glucose as the sole and limiting carbon source . The liquid cultures were maintained with 10 ml of the medium in 50 ml flasks . For the 3-d colony experiments , 3ml of bacteria suspended in soft agar were put into the wells of 6-well Costar Macrotiter plates , set in a tray with distilled water to reduce the rate of evaporation . To measure the cell sizes in liquid cultures at a certain time point , 5 ml of the experiment culture was centrifuged at 4 , 000 rpm for 4 min . Excess supernant was then disposed of , so that the cell density of the resuspension of the rest of the culture was at least 107 cells/ml . Cells were then put on a microscope slide ( Leica ( Wetzlar , Germany ) TCS SP8 inverted confocal microscope with live-cell chamber at 60x ( HC PL APO 63x/1 . 40 oil CS2 WD 0 . 14 mm ) ) , photographed , and their length in the pictures was measured manually using ImageJ ( National Institutes of Health , Bethesda , MD ) . For colony cultures , centrifuge would not separate the agar and the cells . Thus we first dispersed the colony culture and then directly sampled from the mixture . In order to have enough cells per slide to image , the 6-hour old cultures were inoculated with 106 cells/ml . The 14-hour old cultures were inoculated with 1000 cells/ml . The older cultures were all inoculated with 50 cells/ml . The well-mixed Monod model , Eqs ( 1 ) and ( 2 ) , was solve using ode15s MatLab routine . To solve the growth equations Eqs ( 3 ) – ( 5 ) numerically , we rewrote the equations in spherically symmetric coordinates , and then discretize the space into concentric shells so that the partial differential equations become sets of coupled ordinary differential equations describing dynamics within each shell . These were then solved again using Matlab’s ode15s , with an additional constraint that redistributed the total number of bacteria N into a bacterial colony with the constant cell packing density , as in Eq ( 5 ) , at every time step . That is , each discretized shell of the space had a maximum cellular capacity given by the packing density and the shell volume . The constraint redistributed those cells that overflowed each shell’s capacity to the colony’s edge , but we did not shrink the inner shells when the cells in them started dying . Newly grown cells are first filled in the colony’s current edge shell . If the current edge shell is overflowed , the extra cells are filled in the next shell , and so on . We first fitted the five parameters of the Monod model for the growth in the liquid culture , Eqs ( 1 ) and ( 2 ) . For this we defined the loss function L = ∑ i ( N i - N ( t i ; g max , m , K , τ lag , a l ) ) 2 , where Ni was the population size ( in CFU/ml ) in the i’th measurement , and N ( ti; gmax , m , K , τlag , al ) was the model prediction for the same time and for given parameter values . Note that we did not average measurements at the same t , but incorporated all individual observations into the loss function , cf . Fig 4 . We optimized L over the five parameters using MatLab’s fmincon . For K and m , which are small and have large uncertainties , we optimized w . r . t . their logarithms , thus enforcing their positivity ( the other parameters were sufficiently constrained by data away from zero even without this reparameterization ) . The optimization was performed with ten different random initial conditions for the parameters , and the best values from among all the runs were chosen , resulting in the best-fit parameters g ¯ max , m ¯ , K ¯ , τ ¯ l , a ¯ l , which we report in Table 1 . To estimate the confidence intervals for these inferences , we bootstrapped the data 1000 times [48] . When re-sampling with replacements for bootstrapping , we resampled separately from the exponential growth region ( t ≤ 22 hrs ) and the saturated region ( t > 22 hrs ) , so that the number of data points in each of the regions was fixed in all resampled datasets . We refitted the five growth parameters for each of the resampled data sets . The middle 80% of the best-fit parameter realizations are reported in Table 1 as confidence intervals , and the covariances among the bootstrapped best-fit values are reported in Table 2 . Since the sensitivities to the parameters vary widely , and L near its minimum is badly approximated by a quadratic form , we additionally report confidence intervals directly on the model predictions , rather than just the parameters . For this , for each of the 1000 resampled datasets , we calculated the population growth with the best-fit parameters , and the middle 80% of these growth curves are shown as the colored band in Fig 4 ( top ) . For fitting the 3-d growth model , Eqs ( 3 ) – ( 5 ) , we write the loss function L = ∑ i ( N i - N ( t i ; g ¯ max , m ¯ , K ¯ , τ ¯ l , a c , D , μ ) ) 2 . This is minimized as above over ac , D , μ , with the first four parameters inherited from the optimizations for liquid data . Results of the optimization are shown in Fig 4 ( bottom ) . To establish confidence intervals , we bootstrap the entire analysis pipeline 30 times ( the number is limited since parameter optimizations for PDEs describing the nutrient dynamics are computationally costly ) , resampling both the liquid and the 3-d colony data . While resampling the colony data , we keep the number of data points in each of the three regions constant ( exponential , t < 24 , diffusion-limited , 24 ≤ t < 48 , and saturated , t ≥ 48 ) . Confidence intervals on parameters and model predictions in Fig 4 ( bottom ) and Fig 5 are then done as explained above . We use the same bootstrapped data sets to estimate the covariances and correlations of the parameters ( Table 2 ) . These are evaluated as empirical covariances and correlation coefficients of the best-fit values for the bootstrapped data sets . | The vast majority of theoretical and experimental studies assume that bacteria exist as planktonic cells in well-mixed liquid cultures , all with equal access to nutrients , wastes , toxins , antibiotics , bacterial viruses , and each other . However , in the real world , bacteria are more often found in physically structured , spatially heterogeneous habitats as colonies and micro-colonies . While one can experimentally explore the population and evolutionary dynamics of bacteria in such physically structured habitats , there is dearth of mathematical models to generate hypotheses for and to interpret results of these experiments . As a step towards the construction of a theory of the population dynamics of bacteria in physically structured habitats , we develop and experientially explore the simplest such model of the dynamics of bacterial growth in 3-d structured colonies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"death",
"rates",
"classical",
"mechanics",
"chemical",
"compounds",
"fluid",
"mechanics",
"demography",
"microbiology",
"carbohydrates",
"geometry",
"organic",
"compounds",
"glucose",
"mathematical",
"models",
"developmental",
"biology",
"mathematics",
"statistics",
"(mathematics)",
"population",
"biology",
"microbial",
"growth",
"and",
"development",
"research",
"and",
"analysis",
"methods",
"packing",
"density",
"microbial",
"physiology",
"mathematical",
"and",
"statistical",
"techniques",
"chemistry",
"fluid",
"dynamics",
"bacterial",
"growth",
"continuum",
"mechanics",
"physics",
"people",
"and",
"places",
"population",
"metrics",
"organic",
"chemistry",
"confidence",
"intervals",
"monosaccharides",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"population",
"density"
]
| 2017 | Growth of bacteria in 3-d colonies |
Holoprosencephaly ( HPE ) is a severe human genetic disease affecting craniofacial development , with an incidence of up to 1/250 human conceptions and 1 . 3 per 10 , 000 live births . Mutations in the Sonic Hedgehog ( SHH ) gene result in HPE in humans and mice , and the Shh pathway is targeted by other mutations that cause HPE . However , at least 12 loci are associated with HPE in humans , suggesting that defects in other pathways contribute to this disease . Although the TGIF1 ( TG-interacting factor ) gene maps to the HPE4 locus , and heterozygous loss of function TGIF1 mutations are associated with HPE , mouse models have not yet explained how loss of Tgif1 causes HPE . Using a conditional Tgif1 allele , we show that mouse embryos lacking both Tgif1 and the related Tgif2 have HPE-like phenotypes reminiscent of Shh null embryos . Eye and nasal field separation is defective , and forebrain patterning is disrupted in embryos lacking both Tgifs . Early anterior patterning is relatively normal , but expression of Shh is reduced in the forebrain , and Gli3 expression is up-regulated throughout the neural tube . Gli3 acts primarily as an antagonist of Shh function , and the introduction of a heterozygous Gli3 mutation into embryos lacking both Tgif genes partially rescues Shh signaling , nasal field separation , and HPE . Tgif1 and Tgif2 are transcriptional repressors that limit Transforming Growth Factor β/Nodal signaling , and we show that reducing Nodal signaling in embryos lacking both Tgifs reduces the severity of HPE and partially restores the output of Shh signaling . Together , these results support a model in which Tgif function limits Nodal signaling to maintain the appropriate output of the Shh pathway in the forebrain . These data show for the first time that Tgif1 mutation in mouse contributes to HPE pathogenesis and provide evidence that this is due to disruption of the Shh pathway .
Holoprosencephaly ( HPE ) is a prevalent human disorder affecting forebrain and craniofacial development , with an incidence of up to 1∶250 during embryogenesis , and a high frequency of intrauterine lethality [1] , [2] . Recent estimates of the frequency of HPE live births are as high as 1 . 3 per 10 , 000 [3] , and many children born with severe HPE phenotypes die soon after birth [4] , [5] . The primary defect in HPE is a failure of ventral forebrain development with concomitant defects in midline facial structures [6] , [7] . In its most severe form ( alobar HPE ) the forebrain fails to divide , resulting in a single brain ventricle . Less devastating forms of HPE allow near or complete separation of left and right hemispheres [8] , [9] . At least 12 genetic loci have been implicated in HPE by mapping of the minimal chromosomal regions deleted in affected families [10]–[12] . Perhaps the best studied HPE gene , Sonic hedgehog ( SHH ) , maps to the HPE3 locus [13] . In humans heterozygous SHH loss of function mutations account for 17% of familial HPE and 3 . 7% of sporadic cases [13]–[15] , suggesting a loss of function haploinsufficient phenotype [16] , [17] . The genes encoding the transcription factors TGIF1 , Six3 and Zic2 have been identified as the affected genes at other HPE loci [18]–[20] . Interestingly , recent work has shown that Six3 specifically activates expression of Shh in the forebrain , and in mice Shh and Six3 mutations synergize to cause HPE , further emphasizing the importance of the Shh pathway [21] , [22] . To establish forebrain dorsoventral patterning , the proper output of the Shh signaling pathway is essential in prechordal plate ( PrCP ) , a primitive streak-derived axial tissue . In mouse embryos at 7 . 75 dpc , Shh expression is seen in the PrCP underlying the forebrain precursor tissue . Shh expression in the PrCP is essential for activating Shh expression in the overlying ventral diencephalon tissue by 9 . 0 dpc , where Shh specifies ventral identity [1] , [23] . Gli3 , a zinc-finger transcription factor that primarily acts as a repressor of Shh signaling , has been shown to play a crucial role in forebrain dorsoventral patterning . In the developing neural tissue , Gli3 is expressed in a gradient that is higher dorsally , and Gli3 homozygous null embryos have a forebrain with dorsally expanded ventral tissue , that lacks dorsal identity [24]–[26] . It has been shown that the proper balance between Gli3 and the ventralizing Shh is critical during forebrain patterning [25] , [27] . The lack of ventral identity seen in Shh null embryos is partially rescued when the dose of Gli3 is reduced genetically , suggesting that the mutual antagonism of these two factors is critical for forebrain dorso-ventral patterning . However , since the forebrain develops relatively normally in the absence of both Shh and Gli3 , there must be additional pathways that specify telencephalon development , which likely depend on Foxg1 and FGF signaling [28] . Disruption of FGF signaling in the anterior by deletion of the Fgfr1 and Fgfr2 genes results in defective ventral telencephalon development , without disruption of the Shh signaling pathway [29] . TGIF1 ( Thymine/Guanine-Interacting Factor ) is a homeodomain protein , which binds directly to DNA via a thymine/guanine-containing consensus site , or interacts with Transforming Growth Factor ( TGF ) β-activated Smad proteins [30] , [31] . In response to binding of a TGFβ family ligand to its receptors , the receptor complex phosphorylates and activates specific receptor Smad ( R-Smad ) proteins: Smad2 or Smad3 in the case of TGFβ , Nodal and Activin [32] , [33] . Activated R-Smads complex with the co-Smad , Smad4 , translocate to the nucleus and activate target gene expression via direct binding to DNA , or by interactions with other sequence specific DNA binding proteins [33] . Once recruited to DNA , a Smad complex activates transcription in part through interactions with general coactivators , such as p300/CBP [33] . The presence of specific Smad transcriptional corepressors , such as TGIF1 , limits the transcriptional response by competing with coactivators and by recruiting general corepressor complexes to the Smads [31] , [34] . The more recently identified TGIF2 is homologous to TGIF1 and functions similarly . TGIF2 interacts directly with DNA , or with TGFβ activated Smads and represses gene expression via the mSin3/HDAC complex , but unlike TGIF1 , it does not interact with CtBP corepressors [35]–[37] . Thus overall Tgif function ( TGIF1 and TGIF2 ) limits the magnitude of the transcriptional response to TGFβ family ligands . In addition to regulating TGFβ signaling , TGIF1 can also repress gene expression via the RXR retinoid receptor [30] , [38] , [39] . The TGIF1 gene lies within the minimal HPE4 locus , and TGIF1 sequences were shown to be absent from individuals affected with HPE [20] . In addition to the more common deletions of TGIF1 , single amino acid miss-sense mutations have been identified , some of which reduce transcriptional repression by TGIF1 [20] , [40]–[42] . Heterozygous loss of TGIF1 causes HPE in humans , suggesting a haploinsufficient phenotype [20] . While there is no evidence for mutations in the human TGIF2 gene being associated with HPE , it is clearly possible that these two related proteins share overlapping functions during embryogenesis [42] . In mice , loss of Tgif1 does not have severe phenotypic consequences , at least in a mixed strain background [38] , [43]–45 . In a more pure C57BL/6 strain background placental defects and reduced viability are associated with loss of Tgif1 , and an intragenic mutation in Tgif1 that may result in expression of a truncated polypeptide caused some anterior defects [46] , [47] . As with Tgif1 , Tgif2 null mice are normal on a mixed strain background , but the combination of both mutations results in early embryonic lethality with gastrulation defects in all embryos that are homozygous null for both genes [48] . Genetically reducing Nodal signaling in these embryos reduces the severity of the gastrulation defects , consistent with an inhibitory role for Tgifs in the TGFβ/Nodal pathway . While this demonstrates an essential role for TGIF function early in embryogenesis , the function of Tgifs after gastrulation is less well understood . Here , we investigated the role of Tgif1 and Tgif2 during forebrain development . We demonstrate that loss of Tgif function is indeed important in HPE pathogenesis , and that Tgif1 and Tgif2 play overlapping essential roles during ventral forebrain development by regulating Shh signaling . Conditional loss of function of Tgif1 in the background of a Tgif2 null mutation causes HPE . Furthermore , we show that the HPE phenotype is partially rescued when the dose of Gli3 is reduced . Additionally , we show that reducing Nodal signaling reduces the severity of the HPE phenotype , and partially restores the output of the Shh pathway . This provides the first evidence that Tgifs are required for proper Shh signaling during ventral forebrain development , and verifies that TGIF1 is a bona fide HPE gene .
We have previously shown that loss of both Tgif1 and Tgif2 results in a failure of gastrulation [48] . Conditional deletion of Tgif1 in the epiblast , using a loxP flanked Tgif1 allele [45] and the Sox2-Cre transgene , which is expressed in the epiblast after 5 . 5 dpc [49] , in the background of a Tgif2 null mutation allows these embryos ( which we refer to as cdKO , for conditional double knock-out ) to complete gastrulation . However , most cdKO embryos do not survive past 11 . 0 dpc , have left-right asymmetry defects , and have severe anterior defects . Scanning electron microscopy ( SEM ) analysis of frontal forebrain structure revealed that the ventral lips of the cephalic folds are fused in cdKO embryos at 8 . 25 dpc , as seen in Shh null embryos ( asterisks , Figure 1A ) . It has been shown previously that the normally separated cephalic neural tube is fused in mouse mutants with HPE , including Shh null embryos [50] , [51] . Additional SEM analysis at later stages shows that the midbrain neural tube fails to close in cdKO embryos even at 9 . 25 dpc ( Figure 1B ) . Since human TGIF1 mutations are associated with HPE , we next analyzed the forebrain morphology of control and cdKO embryos to determine whether there was additional morphological evidence to suggest that cdKO embryos have HPE . Whole-mount morphology of the cdKO forebrain at 9 . 0 dpc showed that overall forebrain size and morphology were relatively normal compared to the control . H&E staining showed that neuroepithelium and surface ectoderm were present , but that the neuroepithelium is thinner and lacks any indication of ventral morphology of the control ( Figure 1C ) . By 10 . 0 dpc the cdKO forebrain was clearly abnormal , and was significantly smaller than the control ( Figure 1D ) . Further analysis of forebrain structure by H&E staining showed that ventral forebrain morphology was defective , and that cdKO embryos appeared to have a single thickened layer of surface ectoderm in the ventral forebrain , suggesting that the nasal field has not separated by 10 . 0 dpc ( Figure 1D ) . Since classic HPE phenotypes , such as cyclopia , are more apparent after 11 . 0 dpc , we analyzed a large number of embryos at 12 . 5 dpc in an attempt to identify any cdKO embryos that survive to this stage . Although the most of the cdKO embryos die by 11 . 0 dpc , we were able to identify two cdKO embryos that had survived to 12 . 5 dpc . For this analysis , we dissected a total of 117 embryos at 12 . 5 dpc , 76 ( 65% ) of which appeared normal , 39 ( 33% ) were in the process of being resorbed , and only two were doubly homozygous null for Tgif1 and Tgif2 . Both cdKO embryos showed cyclopia , and one of the two had developed a proboscis , similar to that in an equivalent stage Shh null embryo ( Figure 1E ) . H&E staining of coronal sections through the brain tissue clearly showed that only one nasal epithelium structure was present in the proboscis tissue of both cdKO and Shh null embryos ( Figure 1E , i ) , and that only one eye field was present in cdKO and Shh null embryos ( Figure 1 , ii ) . Thus , the morphological abnormalities in the cdKO forebrain appeared to be quite similar to those seen in Shh null embryos , suggesting that cdKO embryos exhibit a classic form of HPE . The defects in forebrain structure led us to test whether anterior patterning is defective in cdKO embryos . The expression of Six3 , a transcription factor that activates the Shh gene in ventral forebrain [21] , [22] , was seen in forebrain in both control and cdKO embryos ( Figure 2A ) . Foxg1 , a transcription factor that is required for proper forebrain patterning [52] , is expressed in approximately the appropriate pattern in cdKO embryos ( Figure 2A ) . Although there was no major change in the expression pattern , the expression levels of Six3 and Foxg1 were slightly increased in cdKO embryos . In addition , the expression of Fgf8 was clearly increased in the cdKO forebrain , but was still present in approximately the same region as in control embryos ( Figure 2A ) . Consistent with these observations , Fgf8 has been shown to be a FoxH1/Smad2 target gene in the anterior , so may be up-regulated in the absence of Tgifs due to derepression of Smad dependent transcription [53] . Hesx1 , which is a highly specific marker for ventral diencephalon [54] , shows the appropriate expression pattern in the cdKO ventral diencephalon tissue at 9 . 0 dpc , suggesting that the midline of the ventral diencephalon is formed in cdKO embryos ( Figure 2B ) . Emx2 , a transcription factor that is required for dorsal forebrain patterning [55] , was slightly decreased , but was present in a similar domain as in the control ( Figure 2B ) . We next analyzed prospective forebrain tissue in younger embryos . At 7 . 25 dpc Hesx1 was expressed in the anterior of both control and cdKO embryos ( Figure 2C ) . We have shown previously that the forebrain marker , Otx2 , was expressed in cdKO embryos at 7 . 5 dpc [48] , and Six3 was also expressed in the prospective forebrain tissue of cdKO embryos at early head fold ( EHF ) stage ( Figure 2C ) . Taken together these results suggest that forebrain tissue is for the most part correctly patterned in cdKO embryos . In the mouse , forebrain induction and patterning is mediated by primitive streak-derived anterior midline tissue , which includes anterior definitive endoderm ( ADE ) and PrCP [23] , [56] . At 7 . 25 dpc the expression of Hex , a transcription factor that is essential for endoderm development [57] , was seen in both control and cdKO embryos in anterior visceral endoderm and also in the ADE migrating out of the primitive streak at this stage ( Figure 2D ) . By 7 . 5 dpc , Hex expression in anteriorly migrated ADE tissue was present , and did not appear to be significantly different between control and cdKO embryos ( Figure 2D ) . A member of the Forkhead transcription factor family , Foxa2 , which is normally expressed in axial tissue [58] , was expressed in midline tissue of cdKO embryos at the EHF stage ( Figure 2E ) . The PrCP can be identified by expression of Gsc and Dkk1 at late head fold ( LHF ) stage and at 8 . 0 dpc [56] , [59] . Appropriate expression of both Gsc and Dkk1 was seen in cdKO embryos ( Figure 2E and [48] ) , suggesting that the PrCP is present in the absence of Tgifs . This analysis suggests that anterior structures are initially patterned relatively normally in cdKO embryos . While there are clearly some phenotypic differences , such as the failure of the midbrain to close in cdKO embryos , the similarities between cdKO and Shh null embryos raised the possibility that HPE in cdKO embryos may be due to defects in the Shh signaling pathway . At 9 . 5 dpc , Shh was expressed throughout the neural tube in the floor plate , including the midline of the ventral diencephalon of control embryos ( Figure 3A ) . However , Shh expression was clearly reduced in the ventral diencephalon of cdKO embryos . Similarly , in cdKO embryos Shh expression was reduced in the anterior midline at 8 . 25 dpc ( Figure 3B ) . By 8 . 75 dpc Shh expression was present in the ventral forebrain in the control , whereas expression was clearly reduced in the cdKO ventral forebrain tissue ( Figure 3B ) . Transverse sections showed that Shh expression is present but is reduced in the midline tissue including the PrCP ( arrows , Figure 3B ) , and that Shh expression is not detected in the ventral forebrain ( Figure 3B ) . We next analyzed the expression pattern of Shh signaling components at 9 . 0 dpc . Ptch1 encodes a 12 transmembrane Shh receptor , and Gli1 , a transcription factor that mediates Shh signaling [60] . Both genes are direct downstream targets of Shh signaling and are normally expressed strongly in the ventral diencephalon . In cdKO embryos the expression of Gli1 was clearly reduced primarily in the ventral forebrain , while expression was more normal throughout the neural tube up to the forebrain-midbrain boundary ( Figure 3C ) . Ptch1 expression was more similar between cdKO and control embryos , although there was a slight decrease in expression in the anterior in cdKO embryos ( brackets , Figure 3C ) . Together , these results suggest that forebrain patterning is relatively normal , but that the Shh signaling pathway is defective specifically in the ventral forebrain and PrCP . Thus it appears that Tgif function may be required for normal Shh signaling in anterior tissues . The transcription factor , Gli3 , acts as a potent repressor of the Shh signaling pathway . In the absence of Shh , it has been shown that there is some increase in Gli3 expression [24] , and the HPE phenotype in Shh null embryos is partially rescued when Gli3 gene dosage is reduced , suggesting that the proper balance of dorsalizing and ventralizing signals is critical during forebrain development [27] , [61] . We , therefore , analyzed the expression level of Gli3 in control and cdKO embryos . Strikingly , Gli3 expression was clearly increased throughout the neural tube including the forebrain in cdKO embryos ( Figure 4A ) . We also performed WISH for Gli3 in Shh null embryos and compared the level of Gli3 expression with cdKO embryos . Surprisingly , Gli3 expression was higher in cdKO embryos than in Shh null embryos ( Figure 4A ) , suggesting that there may be an additional Tgif-mediated mechanism , distinct from the reduction in Shh expression , that regulates Gli3 expression . To determine whether the increased level of Gli3 contributes to defective Shh signaling in the absence of Tgif function , we performed a genetic rescue experiment by introducing a Gli3 mutant allele into the cdKO background . The Gli3 allele has exon 8 flanked by loxP sites such that Cre-mediated recombination creates a null allele [62] , which is referred to here as Gli3r . In Gli3+/r;cdKO embryos , Gli3 expression was significantly reduced , to below the expression level seen in Shh null embryos ( Figure 4A ) . In contrast , Shh expression was not restored in cdKO embryos that were heterozygous for Gli3 , or in cdKO embryos that were homozygous null for the Gli3 gene ( Gli3r/r;cdKO ) , suggesting that the reduction in Shh expression is at least partially independent of Gli3 activity in cdKO embryos ( Figure 4B ) . We then analyzed the expression of Nkx2 . 1 , a downstream target gene of Shh signaling in the forebrain 1 , 23 , in control and a series of mutant embryos . At 9 . 0 dpc , the expression of Nkx2 . 1 was seen in the ventral diencephalon in control embryos , whereas , Nkx2 . 1 expression was not detected in cdKO or Shh null embryos ( Figure 4C ) . In Gli3+/r;cdKO embryos , Nkx2 . 1 expression was clearly restored while Nkx2 . 1 expression in the ventral diencephalon was not rescued in Gli3+/r;Shhr/r embryos ( Figure 4C ) . These results suggest that a reduction in the excess Gli3 expression partially restores the output of the Shh signaling pathway in cdKO embryos , without affecting Shh expression itself . Initial observation of Gli3 heterozygous cdKO embryos suggests that there may be some phenotypic rescue of the cdKO phenotype . Instead of the round forebrain morphology seen in cdKO embryos , a more structured forebrain vesicle was observed in Gli3+/r;cdKO embryos at 10 . 0 dpc ( Figure 5A ) . To further determine the degree of phenotypic rescue , we H&E stained coronal sections through the forebrain vesicle of control , cdKO and Gli3 heterozygous cdKO embryos . Gli3+/r;cdKO embryos clearly had a more organized forebrain neuroepithelium morphology , and the neuroepithelium appeared to have initiated division of the nasal placode ( arrows , Figure 5A ) , suggesting that the altered balance between Gli3 and Shh expression in cdKO embryos does contribute to the HPE phenotype . In addition , SEM analysis of Gli3 heterozygous cdKO embryos at 8 . 25 dpc shows a partial rescue of the forebrain structure , such that the Gli3 heterozygous forebrain appears to be less disorganized than the cdKO , and the ventral lips of the cephalic folds appear to be partially separated in the Gli3+/r;cdKO ( arrows , Figure 5B ) . Thus , it appears that reducing Gli3 levels results in some rescue of the cdKO phenotype . To address this further , we tested for changes in proliferation and examined forebrain patterning . Since the anterior of the cdKO is clearly reduced in size by 10 . 0 dpc , we tested whether the apparent morphological rescue by Gli3 heterozygosity might be due to a restoration of proliferation . Antibody staining for cleaved caspase 3 , which is a marker of apoptotic cells , identified very few apoptotic cells in either control or cdKO forebrain at 9 . 0 dpc ( Figure 5C ) . Although the cdKO embryos were still alive at 10 . 0 dpc , cells that were positive for cleaved caspase were present throughout the cdKO forebrain neuroepithelium , but were rarely seen in the control ( Figure 5C ) . Consistent with this , TUNEL analysis showed increased apoptosis in the cdKO forebrain at 10 . 0 dpc ( Figure 5C ) . To determine whether proliferation is reduced in cdKO embryos , we stained multiple coronal sections of control and cdKO forebrains at 9 . 0 and 10 . 0 dpc with an antibody to Histone H3 , phosphorylated on serine 10 ( pHH3 ) , which is a marker for cells in late G2 and mitosis . Mitotic cells were seen throughout neuroepithelium for both control and cdKO at 9 . 0 dpc ( Figure 5D ) . Quantification of the proportion of mitotic cells in the neuroepithelium showed that there was a significant reduction in proliferation at 9 . 0 dpc , that was more pronounced by 10 . 0 dpc ( Figure 5E and 5F ) . These results suggest that cdKO embryos have proliferation defects in the forebrain neuroepithelium , and that the reduced proliferation is seen prior to any increase in apoptosis . We next tested whether the apparent rescue of forebrain morphology in Gli3+/r;cdKO embryos was accompanied by a restoration of normal levels of proliferation . However , in Gli3+/r;cdKO embryos , proliferation levels were not different from the cdKO at 10 . 0 dpc ( Figure 5E and 5F ) . This suggests that the phenotypic rescue in Gli3+/r;cdKO embryos is independent of changes in proliferation , and that the morphological defects in the cdKO are not solely due to reduced proliferation . To further characterize ventral structure , we analyzed the expression pattern of Pax7 , a nasal field marker , as well as the eye field marker , Pax2 [61] . Normally by 10 . 0 dpc , the nasal field is well separated as evidenced by the position of the ventral neuroepithelium clearly separating the facial field ( see Figure 1D , for example ) . In Shh null embryos , Pax7 expression is present in a single central region suggesting that the nasal field is not fully separated , whereas when the dose of Gli3 is reduced in Shh null embryos Pax7 expression becomes separated to the two nasal fields [61] . In cdKO embryos , Pax7 expression was observed as a single continuous band , suggesting that nasal field separation is defective ( Figure 6A ) . In Gli3+/r;cdKO embryos , Pax7 expression was clearly well separated and was more similar to that seen in controls , suggesting that the nasal field separation defect is partially rescued in Gli3 heterozygous cdKO embryos ( Figure 6A ) . Similarly , Pax2 expression was reduced and was seen as a single continuous band in cdKO embryos , suggesting that eye field separation is defective ( Figure 6B ) . In Gli3+/r;cdKO embryos , the Pax2 expression level was increased , and appeared as less of a continuous band with distinct eye fields on both sides of the forebrain ( Figure 6B ) . These results suggest that the increase in Gli3 expression , and the altered balance between Gli3 and Shh contribute to the HPE phenotype seen in cdKO embryos resulting in a disruption of the separation of facial primordia . The TGFβ/Nodal signaling pathway has been linked to HPE pathogenesis . For example , HPE has been reported in mouse mutants that result in reduced TGFβ/Nodal signaling , such as Nodal;Smad2 double heterozygotes [63] . Since mutations in these genes result in a reduction in the output of TGFβ/Nodal signaling , rather than the expected increase in cdKO embryos , we generated mice that are heterozygous for both Nodal and Smad2 genes for comparison to our cdKOs . The Smad2 null allele is referred to here as ‘r’ and the Nodal null allele as ‘z’ ( see Materials and Methods for a full explanation ) . Of 41 Nodal;Smad2 double heterozygotes analyzed between 10 . 5 and 12 . 5 dpc only one had HPE , although an additional 15 of the 41 double heterozygotes had anterior truncations or a severe growth delay . The Nodal;Smad2 double heterozygous embryo with HPE had a proboscis and a partial failure to separate the eyes , but was significantly larger than cdKO and Shh null embryos ( Figure S1 ) . H&E staining of sections through the nasal structure showed a single nasal epithelium that appears structurally similar to that of cdKO and Shh null embryos ( Figure S1 , i ) . H&E staining of sections through the eye field showed that a laterally elongated , large optic structure containing two distinct eyes had begun to form , while cdKO and Shh null embryos had only one small pigmented eye field vesicle ( Figure S1 , ii ) . Thus , in contrast to the cdKO embryos , it appears that in embryos with reduced Nodal pathway activity HPE is relatively rare . Our previous analysis of Tgif1;Tgif2 double null mutants showed that Tgifs limit Nodal signaling [48] . To test whether the HPE phenotypes in cdKO embryos were due to increased Nodal signaling , we generated cdKO embryos that carry a Nodal heterozygous mutation . Initial examination of the Nodal heterozygous cdKO embryos suggests that there may be some rescue of the HPE phenotype ( Figure 7A ) . From 317 embryos dissected at 10 . 0 dpc we identified 38 Nodal heterozygous cdKO embryos , representing 12% of the total , which compares well to the expected 12 . 5% from these crosses . Other than two severely delayed embryos , and a small proportion ( less than 10% ) that had severe anterior truncations , the Nodal heterozygous cdKO embryos could be divided into two main phenotypic classes . Around one quarter of the total showed a partial rescue of the cdKO phenotype , such that the forebrain vesicle was better organized and larger in size compared to the cdKO ( Figure 7A ) . Additionally , it appears that there is some improvement in the morphogenesis of the ventral neuroepithelium in these embryos ( arrowhead , Figure 7A ) . The other major phenotype , seen in almost two thirds of Nodal heterozygous cdKO embryos was a reduction in the forebrain . Nodal+/z;cdKO embryos with a reduced forebrain also had a highly disorganized neuroepithelium ( Figure 7A ) . These results suggest that the HPE phenotype seen in cdKO embryos can be at least partially rescued by Nodal heterozygosity , consistent with the defects being due to increased activity of the Nodal/Smad pathway . To confirm that the Nodal heterozygous mutation was reducing expression of Smad2 target genes , we analyzed expression of Fgf8 , which is a direct Smad2/FoxH1 target [53] . As shown earlier , Fgf8 expression is increased in cdKO embryos ( Figure 2A ) , whereas , Fgf8 expression was significantly reduced in the forebrain of Nodal+/z;cdKO embryos , consistent with a reduction in Nodal signaling to Smad2 ( Figure 7B ) . In order to determine whether reducing Nodal signaling in cdKO embryos could affect the output of the Shh signaling pathway , we analyzed the expression level of Nkx2 . 1 , a target of Shh signaling in the forebrain at 9 . 0 dpc . Strikingly , Nkx2 . 1 expression was restored in the ventral forebrain of Nodal+/z;cdKO while Nkx2 . 1 expression was clearly reduced in cdKO embryos ( Figure 7C ) . Taken together , these results suggest that Nodal signaling plays a role in regulating Shh signaling during forebrain development , and that unchecked Nodal signaling in the absence of Tgifs is responsible , at least partially , for disrupting Shh signaling in cdKO embryos . Fgf8 plays a role in coordinating multiple patterning centers during forebrain development [64] , [65] . In the telencephalon , Fgf8 is a target of TGFβ/Nodal signaling , and is also negatively regulated by Gli3 , a potent inhibitory factor of Shh signaling , during early forebrain development [24] . Analysis of Fgf8 expression in Shh null embryos at 8 . 5 dpc showed that Fgf8 was expressed in the ventral forebrain ( Figure 8A ) . However , consistent with previous work [64] , Fgf8 expression was reduced in the telencephalon of Shh null embryos at 8 . 5 dpc and effectively absent by 9 . 0 dpc ( Figure 8A ) . In contrast to the reduction of Fgf8 expression in the Shh null embryos , the cdKO forebrain at 9 . 0 dpc showed increased expression of Fgf8 , most likely due to increased Nodal signaling ( Figure 2A and Figure 7B ) . Interestingly , however , analysis at 9 . 5 dpc revealed that Fgf8 expression was not maintained in cdKO embryos , while Fgf8 expression was clearly restored in Gli3+/r;cdKO embryos ( Figure 8B ) . This result suggests that , by 9 . 5 dpc , Fgf8 expression is no longer maintained by Nodal signaling and that the excess Gli3 in the cdKO limits Fgf8 expression . We next analyzed the expression pattern of Foxg1 , a target of Fgf8 signaling at 9 . 5 dpc . Foxg1 expression was increased in the cdKO forebrain tissue at 9 . 0 dpc consistent with the increased expression of Fgf8 ( see Figure 2A ) . At 9 . 5 dpc , Foxg1 expression in the telencephalon was clearly reduced in the cdKO , whereas , the level of Foxg1 expression was restored to levels similar to that in controls in Gli3+/r;cdKO embryos ( Figure 8C ) . Analysis at 10 . 0 dpc also revealed that Foxg1 expression was reduced in the neuroepithelium , but was partially restored in Gli3+/r;cdKO embryos . The expression of Foxg1 in the optic vesicle was reduced and was seen as a continuous band in the cdKO ( Figure 8C ) . Although in Gli3+/r;cdKO embryos Foxg1 expression was lower than in controls in the optic vesicle , the expression domains were clearly better separated than in cdKO embryos , providing further evidence for a partial rescue of eye field separation ( arrowheads , Figure 8C ) . Taken together , these results suggest that , at 9 . 0 dpc Fgf8 expression is dependent on TGFβ/Nodal signaling , whereas , by 9 . 5 dpc the effect of TGFβ/Nodal signaling decreases and repression of Fgf8 by Gli3 becomes more pronounced .
Of the 12 genetic loci associated with HPE in humans , the best characterized ( SHH , SIX3 and ZIC2 ) are all linked to the Shh pathway . In contrast , while mutations in the TGIF1 gene , which encodes a corepressor for TGFβ/Nodal signaling , are associated with HPE pathogenesis , the underlying role of Tgif function in forebrain development has remained unclear . We now demonstrate that all embryos with a conditional epiblast-specific double knock-out of Tgif1 and Tgif2 exhibit early HPE-like phenotypes that are reminiscent of those seen in Shh null embryos . Our results provide strong evidence that a major function of Tgifs in the forebrain is to maintain the proper balance between Shh and its antagonist , Gli3 , by limiting Nodal signaling . These results resolve the conundrum of how Tgif function is associated with HPE , and identify novel points of coordination between the Shh , Nodal and FGF signaling pathways during anterior development ( Figure 9 ) . SHH , SIX3 , ZIC2 and TGIF1 , are the four genes that are most commonly screened as a part of the genetic evaluation of human HPE patients [66] . Mice homozygous for a Shh null allele exhibit defects in midline facial features including cyclopia and proboscis that are typically seen in severe cases of human HPE , suggesting that SHH mutations do contribute to HPE in humans [50] . Recent work showed that the transcription factor Six3 is directly linked to Shh signaling by acting as a transcriptional activator of the Shh gene , specifically in the ventral forebrain [21] , [22] . ZIC2 , encodes a zinc-finger containing transcription factor , that has been shown to be important for forebrain patterning and Shh signaling [67] , [68] . Thus , the best characterized HPE mutations appear to target the Shh signaling pathway . In contrast , the role in HPE pathogenesis of mutations in TGIF1 , which encodes a corepressor for TGFβ/Nodal signaling , has long remained unclear . Loss of function mutations in the Tgif1 gene in mice have no severe phenotypes in a mixed strain background , although an intragenic mutation in Tgif1 , which may create a hypomorphic allele , has been shown to cause anterior defects in a strain specific manner [47] . However , HPE phenotypes have not been seen in Tgif1 or Tgif2 mutants , and these analyses have not yet shed light on any potential role in HPE pathogenesis . Tgif2 , a closely related Tgif1 paralog present in mouse and human , shares conserved functions with Tgif1 [69] . Both Tgif1 and Tgif2 show ubiquitous expression in the embryo proper from at least 6 . 0 dpc , consistent with the possibility of overlapping function during early development . As with Tgif1 mutations , mice that carry a homozygous Tgif2 mutation do not show appreciable phenotypes in a mixed strain background . Mice with both Tgif1 and Tgif2 mutations , with at least one wild-type allele of either Tgif1 or Tgif2 , are also viable and fertile in a mixed strain background [48] . In contrast , embryos with homozygous mutation of both Tgif1 and Tgif2 fail to gastrulate , providing strong evidence that Tgif1 and Tgif2 perform essential overlapping functions during embryogenesis . Thus , although there is no evidence suggesting that human TGIF2 is associated with HPE [42] , it is possible at least in mice , that both proteins share overlapping functions in anterior development . We generated embryos with Sox2-Cre mediated conditional deletion of Tgif1 in the background of a Tgif2 null , which allows the resulting embryos to undergo gastrulation successfully . At 10 . 0 dpc , the cdKO embryos have an HPE-like forebrain and neuroepithelium morphology , and the expression patterns of Pax2 and Pax7 suggest that separation of midline facial features is defective . Moreover , SEM analysis shows that separation of the ventral lips of the cephalic neural fold is defective , consistent with the failure to divide midline facial features . These phenotypes are typical of early HPE mouse mutants such as Shh null embryos , clearly demonstrating that Tgif1 and Tgif2 share redundant functions and together are essential players in normal forebrain development . Although the majority of cdKO embryos fail to survive past 11 . 0 dpc , from an analysis of 117 embryos where approximately 30 were expected to be cdKO , we were able to identify two embryos lacking both Tgif1 and Tgif2 at 12 . 5 dpc , which had presumably survived to this point due to a slight delay in recombination of the conditional Tgif1 allele . Interestingly , these two embryos also showed remarkable similarity to Shh null embryos at the same stage . Specifically , one had a proboscis and both had cyclopia , further reinforcing the idea that the early phenotypes analyzed in detail here are clear precursors of later HPE . While the fact that relatively few embryos survive past 11 . 0 dpc limits our ability to analyze later HPE phenotypes in detail , those cdKO embryos that do survive to 12 . 5 dpc have classic HPE phenotypes . Despite the similarity of the HPE-like phenotypes , it should be noted that there are some differences between our cdKO and Shh null embryos . Such differences include the failure of the midbrain neural tube to close , which is not seen in Shh nulls , and the fact that the majority of cdKO embryos die by 11 . 0 dpc , whereas most Shh null embryos survive to late gestation . These differences aside , this work provides the first clear evidence from mouse models for a role for loss of Tgif function in HPE pathogenesis . Our data suggest that Tgif function is required for appropriate Shh signaling during forebrain development . In cdKO embryos , Shh expression is present but reduced in the PrCP , and is undetectable in the neuroepithelium , suggesting that Shh is transcriptionally activated but that its expression is not properly maintained . In addition to the defective Shh expression in the forebrain , the expression of downstream targets of Shh signaling is significantly reduced in the forebrain . Expression of Gli3 , which encodes a repressor for the Shh signaling pathway in the forebrain , is up-regulated in Shh null embryos , and the HPE phenotype of Shh null embryos is partially rescued when the genetic dose of Gli3 is reduced [27] , [61] . Similarly , cdKO embryos showed an increased level of Gli3 expression in the forebrain . Intriguingly , the increase in Gli3 expression in cdKO was clearly higher than in Shh null embryos , suggesting that there is an additional , Shh-independent , Tgif-dependent mechanism that regulates Gli3 gene expression . In cdKO embryos with a reduced dose of Gli3 , there was a phenotypic rescue in the morphology of the forebrain neuroepithelium and also of the craniofacial features . Additionally , Nkx2 . 1 expression was restored in the ventral diencephalon of cdKO embryos carrying a Gli3 heterozygous mutation , while , in agreement with previous work , there was no rescue of Nkx2 . 1 expression in the diencephalon of Shh null embryos with a Gli3 heterozygous mutation [61] . This suggests that some level of Shh expression is required for Nkx2 . 1 expression , and also suggests that sufficient Shh expression is present to activate Nkx2 . 1 in the ventral diencephalon of cdKO embryos . However , it should be noted that Shh expression was not rescued in the ventral forebrain of Gli3 mutant cdKO embryos . Although many mutations that cause HPE may do so by affecting the Shh pathway , and specifically the balance between Shh and Gli3 , it is worth pointing out that Gli3 heterozygosity does not rescue all mouse models of HPE . For example , the phenotype of Fgfr1;Fgfr2 double mutant embryos is not rescued by Gli3 mutation , suggesting that there is some specificity to the rescue by Gli3 mutations [29] . Taken together , these data provide strong evidence that Tgifs play a critical role in regulating Shh signaling during forebrain development , and that the loss of Tgif-mediated regulation of the Shh pathway is important for HPE pathogenesis . Studies in humans and mice have implicated both the retinoic acid and TGFβ/Nodal pathways in HPE pathogenesis . Retinoic acid mediated teratogenesis in humans is known to contribute to CNS anomalies such as hydrocephalus , and in a few rare cases , HPE , and in mice in utero administration of retinoic acid to pregnant females on gestational day 7 leads to embryos with severe craniofacial phenotypes including HPE [70] , [71] . However , mutations in genes associated with retinoic acid signaling have not been identified in HPE patients . Mutations that likely reduce the output of the TGFβ/Nodal pathway have been found in human patients with HPE or laterality defects . Mutations in TDGF1 ( also referred to as CRIPTO ) , an EGF-CFC family member that acts as a co-factor for the NODAL ligand , and in the gene encoding the forkhead transcription factor FOXH1 ( also known as FAST1 ) , which complexes with SMAD2 and SMAD4 to mediate TGFβ/NODAL signaling , have been identified [72] , [73] . However , these mutations are found very rarely in HPE , and in general are not complete loss of function alleles . Studies in Tdgf1 null and Foxh1 null embryos show that these genes are required for the activity of the early organizing centers during gastrulation [74] , [75] . In Tdgf1 null embryos , marker analysis shows that expression of organizer genes including Brachyury , Cerl1 and Lhx1 is defective . Similarly in Foxh1 null embryos , expression of organizer genes such as Foxa2 and Goosecoid , is reduced , and analysis of forebrain markers such as Six3 , Hesx1 and Fgf8 shows that the forebrain tissue is significantly reduced , exhibiting a mild anterior truncation phenotype [75] . It has also been suggested that Nodal;Smad2 double heterozygous mutations can result in HPE , again indicating that a reduction in TGFβ/Nodal signaling is important in HPE pathogenesis [1] . However , the morphology of these embryos suggests that in most cases forebrain tissue is reduced or missing , rather than exhibiting a clear HPE phenotype as seen in Shh null embryos , for example . Thus it appears that , at least in mice , a reduction in the TGFβ/Nodal signaling pathway primarily results in defective early organizing centers , leading to phenotypes such as a small or truncated forebrain . In contrast , in our cdKO embryos , marker analysis shows that the organizing centers are formed , and that the forebrain does not show an anterior truncation phenotype . In addition , the forebrain morphology shows an HPE phenotype that is similar in many respects to that seen in Shh null embryos , and forebrain markers show relatively normal expression patterns , suggesting that the forebrain is reasonably formed in cdKO embryos . Our own analysis of embryos that are heterozygous for both Smad2 and Nodal is in agreement with the idea that HPE is relatively rare in this genetic combination – only one out of 41 double heterozygotes analyzed at 10 . 5–12 . 5 dpc had HPE , with an additional 15 showing severe growth delays or anterior truncations . Additionally , it is interesting to note that the comparison of cdKO , Shh null and Smad2/Nodal double heterozygous embryos with HPE at 12 . 5 dpc suggests that , at least superficially , the Shh null and cdKO are more similar to each other than to the Smad2/Nodal double heterozygote . Thus the loss of Tgif1 and Tgif2 causes a classic HPE phenotype , rather than the predominance of anterior truncations that are seen in embryos with reduced activity of the TGFβ/Nodal pathway . Our results , together with evidence from mouse mutants with reduced Nodal activity , support a model in which decreased Nodal signaling primarily results in a truncation of anterior tissues , whereas increased Nodal signaling ( as in our cdKO embryos ) causes classic HPE phenotypes . One alternate interpretation of this difference between the HPE phenotype in cdKO embryos and other TGFβ/Nodal mouse mutants is that the effects of loss of Tgif function are independent of TGFβ/Nodal signaling during forebrain development . However , we have shown that embryos that are homozygous null for both Tgif1 and Tgif2 fail gastrulation , and that the gastrulation defect is dependent on increased TGFβ/Nodal signaling . Similarly , left-right asymmetry defects in cdKO embryos can be partially rescued by reducing the dose of Nodal [48] . Here we show that at 9 . 0 dpc , Fgf8 expression is increased in the cdKO , consistent with the derepression of a Smad/Foxh1 target gene [53] . Importantly , this excess Fgf8 expression is reduced in the Nodal heterozygote . Reducing the dose of Nodal also results in a partial rescue of the HPE phenotypes in a proportion of cdKO embryos . Most of the remaining Nodal heterozygous cdKO embryos have a mild anterior truncation , which might indicate that there are additional Nodal and Tgif specific phenotypes , but could also reflect the effect of mutating multiple components of the Nodal pathway . However , with the restoration of Nkx2 . 1 expression in the Nodal heterozygous cdKO forebrain , this is clearly consistent with a model in which Tgifs limit Nodal signaling and that the absence of this restraint causes disruption of the Shh pathway and HPE . It should , however , be noted that we have not yet exhaustively analyzed the Shh signaling pathway in Nodal heterozygous cdKO embryos , and it will clearly be of interest in the future to determine precisely how Nodal heterozygosity rescues Nkx2 . 1 expression and forebrain morphology . One attractive candidate for the Nodal target would be the Gli3 gene , given its striking upregulation in the cdKO . However , this remains to be tested and potential effects of other pathways , such as FGF signaling , that specify forebrain patterning should also be considered . On balance , it is reasonable at this point to suggest that the HPE phenotype seen in cdKO embryos is dependent on excessive TGFβ/Nodal signaling due to the loss of Tgif-mediated repression , and that disruption of the Shh pathway makes a major contribution to the phenotype . The increased Fgf8 expression seen at 9 . 0 dpc in cdKO embryos is consistent with an increase in Nodal signaling , and is in fact reduced in the Nodal heterozygote . However , this also appears to be somewhat at odds with the increased Gli3 expression seen in cdKO embryos , since Gli3 represses Fgf8 expression in the anterior . However , by 9 . 5 dpc , we show that Fgf8 expression in the cdKO telencephalon is essentially lost , consistent with increased repression by Gli3 . It is likely that by this stage the effect of Nodal signaling is diminishing , even in the cdKO , and so the excess Gli3 predominates . In support of this , Gli3 heterozygosity restores some Fgf8 expression and restores expression of Foxg1 , which is a downstream target of FGF signals in the anterior [65] . Analysis of Fgf8 expression in Shh null embryos reveals that expression is already lost by 9 . 0 dpc , while at this stage in the cdKO it is increased . However , as with the Gli3 heterozygous cdKO at 9 . 5 dpc , the loss of Fgf8 expression in Shh null embryos can be rescued by Gli3 heterozygosity [24] , [61] . Thus the loss of Fgf8 expression in the anterior may contribute to the HPE phenotypes seen in both Shh null and cdKO embryos , and the difference in timing of the loss of expression may also be in part responsible for some of the differences between these two models . Given that loss of Fgf8 expression is common to the Shh null and cdKO HPE models , it is tempting to speculate that in the small proportion of Smad2/Nodal double heterozygous mutants with the HPE phenotype is in part due to a failure to fully activate Fgf8 expression . Taken together , our data suggest a model in which Tgifs limit the activity of the Nodal-Smad2 pathway , which is required for full activation of Smad/Foxh1 targets , such as Fgf8 ( Figure 9 ) . In addition we provide evidence that regulation of Nodal signaling by Tgifs is required to maintain the appropriate balance between Shh and Gli3 levels in the forebrain . However , it should be noted that we do not yet know whether this occurs via direct regulation of Gli3 or Shh expression ( dashed lines in Figure 9 ) , or whether the regulation is less direct . An additional possibility is that at least some of the regulation of the Shh pathway by Tgifs is independent of Nodal/Smad2 . For example , Gli3 might be a direct target of Tgif repression , although the rescue of Nkx2 . 1 expression in the Nodal heterozygotes is consistent with a Nodal dependent regulation of the Shh pathway . In summary , this work provides the first clear evidence for a role for loss of Tgif function in HPE pathogenesis , and suggests that Tgifs regulate Shh signaling pathway activity . We propose that Tgif function limits Gli3 expression , and that by a mechanism that is independent of changes in Gli3 levels , Tgifs are required for full Shh expression in the PrCP and neuroepithelium . Thus , the Tgifs have significant contributions to HPE pathogenesis by functioning as key regulators of Shh signaling during forebrain development , most likely by limiting Nodal signaling .
All animal procedures were approved by the Animal Care and Use Committee of the University of Virginia , which is fully accredited by the AAALAC . The loxP flanked Tgif allele [45] , Tgif2 null [48] , loxP flanked Gli3 allele [62] , Nodal mutants [76] , loxP flanked Smad2 allele [77] , and the Sox2-Cre line [49] have been described previously . Conditional Shh mice were obtained from Jackson labs ( stock 4293; [78] ) . The Gli3 , Shh and Smad2 alleles each contain loxP flanked exons , which when recombined result in null alleles , and are referred to here as ‘r’ for recombined ( null ) . The Nodal null allele is referred to as ‘z’ , for an introduced lacZ reporter . All mouse lines were maintained on a mixed C57BL/6J×129Sv/J background . Genomic DNA for PCR genotype analysis was purified from ear punch , at post-natal day 21 ( P21 ) , or yolk sac ( 7 . 0–10 . 0 dpc ) by HotShot [79] . Whole-mount in situ hybridization was performed on 7 . 5–10 . 0 dpc embryos with digoxigenin-labeled riboprobes , as described [80] . Stained embryos were processed for sectioning and histology as described [58] . All images are representative of at least three embryos analyzed . Embryos were fixed overnight in 4% paraformaldehyde at 4°C , dehydrated through an ethanol series ( 70% , 90% , 95% , 100% ×2 for 30 minutes each ) , incubated in xylene twice for 60 minutes and 1∶1 xylene/paraffin for 60 minutes at 60°C , then embedded in paraffin wax , and sectioned at 7 µm . For Hematoxylin and Eosin ( H&E ) histological analysis , sections were de-paraffinized with xylene and stained with H&E . Multiple sections per embryo were incubated with primary antibodies for pHH3 or active caspase 3 as described [48] . For IHC , antibody staining was detected using Vectastain ABC ( Vector Laboratories ) and developed with Impact DAB ( Vector Laboratories ) . For H&E and IHC images were captured using an Olympus BX51 microscope and either an Olympus SZX12 or DP70 digital camera , and manipulated in Adobe Photoshop . Images of 7 . 0–10 . 0 dpc embryos were captured using a Leica MZ16 stereomicroscope and QImaging 5 . 0 RTV digital camera . Embryos were fixed overnight in 4% paraformaldehyde at 4°C , and then fixed with osmium tetraoxide for 30 min and dehydrated through an ethanol series ( 40% , 60% , 80% and 100% ×2 for 15 minutes each ) . Dehydrated samples were further processed in an Autosamdri-815 ( Tousimis Research Corporation ) and were gold coated by using a SCD005 Sputter Coater ( Bal-Tec ) . Images were captured using a JSM-6400 Scanning Electron Microscope ( JEOL ) . | Holoprosencephaly ( HPE ) is a devastating genetic disease affecting human brain development . HPE affects more than 1/8 , 000 live births and up to 1/250 conceptions . Several genetic loci are associated with HPE , and the mutated genes have been identified at some . We have analyzed the role of the TGIF1 gene , which is present at one of these loci ( the HPE4 locus ) and is mutated in a subset of human HPE patients . We show that Tgif1 mutations in mice cause HPE when combined with a mutation in the closely related Tgif2 gene . This provides the first evidence from model organisms that TGIF1 is in fact the gene at the HPE4 locus that causes HPE when mutated . The Sonic Hedgehog signaling pathway is the best understood pathway in the pathogenesis of HPE , and mutation of the Sonic Hedgehog gene in both humans and mice causes HPE . We show that mutations in Tgif1 and Tgif2 in mice cause HPE by disrupting the Sonic Hedgehog signaling pathway , further emphasizing the importance of this pathway for normal brain development . Thus we confirm TGIF1 as an HPE gene and provide genetic evidence that Tgif1 mutations cause HPE by disrupting the interplay of the Nodal and Sonic Hedgehog pathways . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"animal",
"models",
"medicine",
"developmental",
"biology",
"model",
"organisms",
"molecular",
"development",
"genetics",
"biology",
"human",
"genetics",
"morphogenesis",
"genetics",
"and",
"genomics",
"clinical",
"genetics"
]
| 2012 | Loss of Tgif Function Causes Holoprosencephaly by Disrupting the Shh Signaling Pathway |
Coregulator proteins ( CoRegs ) are part of multi-protein complexes that transiently assemble with transcription factors and chromatin modifiers to regulate gene expression . In this study we analyzed data from 3 , 290 immuno-precipitations ( IP ) followed by mass spectrometry ( MS ) applied to human cell lines aimed at identifying CoRegs complexes . Using the semi-quantitative spectral counts , we scored binary protein-protein and domain-domain associations with several equations . Unlike previous applications , our methods scored prey-prey protein-protein interactions regardless of the baits used . We also predicted domain-domain interactions underlying predicted protein-protein interactions . The quality of predicted protein-protein and domain-domain interactions was evaluated using known binary interactions from the literature , whereas one protein-protein interaction , between STRN and CTTNBP2NL , was validated experimentally; and one domain-domain interaction , between the HEAT domain of PPP2R1A and the Pkinase domain of STK25 , was validated using molecular docking simulations . The scoring schemes presented here recovered known , and predicted many new , complexes , protein-protein , and domain-domain interactions . The networks that resulted from the predictions are provided as a web-based interactive application at http://maayanlab . net/HT-IP-MS-2-PPI-DDI/ .
CoRegs are members of multi-protein complexes transiently assembled for regulation of gene expression [1] . Assembly of these complexes is affected by ligands that bind to nuclear receptors ( NRs ) , such as steroids , retinoids , and glucocorticoids [2]–[5] . CoRegs complexes exist in many combinations that are determined by post-translational modifications ( PTMs ) and presence of accessory proteins [6] , [7] . To date , over 300 CoRegs have been characterized in mammalian cells [8] and it has been shown that CoRegs complexes control a multitude of cellular processes , including metabolism , cell growth , homeostasis and stress responses [6] , [9] , [10] . Many CoRegs complexes are considered master regulators of cell differentiation during embryonic and post-developmental stages [10] , [11] , and evidence suggests that malfunction of these proteins can lead to the pathogenesis of endocrine-related cancers [3] , [12] and diabetes [13] . Importantly , it is believed that development of better chemical modulators of CoRegs will lead to a ‘new generation’ of drugs with higher efficacy and selectivity [14] , [15] . To accelerate research in the area of CoRegs signaling , the Nuclear Receptor Signaling Atlas ( NURSA ) [16] have been applying systematic proteomic and genomic profiling related to CoRegs [17] , [18] . Recently , the NURSA consortium released a massive high-throughput ( HT ) IP/MS study reporting results from 3 , 290 related sets of proteomics pull-down experiments [19] . The results from these experiments are protein identifications with semi-quantitative spectral count measurements , which can be used to approximate protein enrichment in individual IPs . Multiple IP experiments that sample different protein complex subunits can be integrated to gain a global picture of protein complex composition [20]–[22] . Several prior studies applied to human cells have proposed strategies to reconstruct protein complexes by combining results from HT-IP/MS [23]–[28] . Some of the results from such studies have been processed by algorithms that probabilistically predict binary protein-protein interactions ( PPIs ) . In some cases , such predictions were validated using known PPIs from the literature , where in few cases predicted interactions were further validated experimentally . For example , Washburn and colleagues implemented the multidimensional protein identification technology ( MudPIT ) method to pull down complexes using 27 bait proteins from the Mediator complex to suggest 557 probabilistic interactions between the baits and their pulled preys [23] . They used the Jaccard distance to integrate protein co-occurrence in the different experiments , and compared their ‘high-confidence’ interactions with those listed in a literature-based database , the human protein reference database ( HPRD ) [29] . Experimentally , the study validated few predicted interactions using co-IP and western blots . In a follow up study , different clustering approaches to extract sub-complexes from related affinity purification ( AP ) -MS experiments using three distance measures: Manhattan , Euclidian , and Correlation Coefficient for clustering are described [30] . The aforementioned work , and other similar prior studies , ranked predicted associations and provided probabilities for interactions between baits and preys , building on the explicit nature of bait-prey relationship in epitope-based purifications . However , due to secondary cross-reacting proteins , bait-prey relationships are rarely explicit in IPs carried out with primary antibodies . Hence , here we developed and compared different ways , coded into mathematical functions , to score prey-prey interactions from a large , recently published , HT-IP/MS dataset . The equations predict direct protein-protein interactions between prey proteins without considering the specific baits . We also used the same equations to predict domain-domain interactions underlying the protein-protein interactions . We evaluated the performance of these equations using known protein-protein and domain-domain interactions from the literature and validated one protein-protein interaction experimentally , and one domain-domain interaction using computational docking . By combining the data from the 3 , 290 IP-MS experiments collected by NURSA we predicted binary interactions between prey proteins and their domains . We offer a global view of CoRegs complexes in human cells , and provide the predicted networks for exploration on the web through a web-based application with downloadable tables freely available at http://maayanlab . net/HT-IP-MS-2-PPI-DDI/ .
A detailed description of the IP-MS procedure can be found in references [19] , [26] and the list of experiments in Dataset S1 . The data we analyzed is provided as supporting material tables for reference [19] . These supporting tables contain GeneIDs for identified protein products , as well as the spectral count ( SPC ) measurements , and ‘abundance’ values , defined as SPCs/MW , where MW is the molecular weight for the largest isoform of the gene product . The latter normalization approximately accounts for the number of peptides expected from a protein . Abundance is logically similar to the normalized spectral abundance factor ( NSAF ) scores previously proposed [30] , except the values are not scaled per experiment . To score prey-prey interactions from the HT-IP/MS data table , containing the ranks of proteins from the 3 , 290 IP-MS experiments , we evaluated existing and developed new equations implemented as algorithms in MATLAB and Java . Sørensen similarity coefficient ( Sor ) provides a symmetric similarity coefficient for comparing two finite sets . The coefficient ranges between 0 and 1 , where 0 denotes no similarity , and 1 denotes identical sets . The Sørensen coefficient is calculated as the ratio of the cardinality of shared members between two sets and the sum of the cardinalities of the same sets . ( 1 ) The Sørenson coefficient was applied to determine the likelihood that proteins A and B directly interact . MA and MB are the sets of all experiments that reported either protein A , B or both as present in the lists of pulled prey proteins . MA , B are lists where both A and B are present . Pearson's Correlation coefficient ( Pr ) characterizes the linear dependency of two variables . Here we used the Pearson's Correlation coefficient to quantify the correlation the SPC scores of two proteins across all IP/MS experiments . ( 2 ) ρA , B is the Pearson's Correlation coefficient between proteins A and B where Q denotes the reported ‘abundance’ which is SPC/MW ( MW , molecular weight ) . and are the column vectors of Q at indices and . is the covariance and and are the standard deviations of and . Equation 3 ( E3 ) was developed through an intuitive manual symbolic search for functions that perform well , based on benchmarking , using known protein-protein interactions . E3 calculates a ratio between the sum of the SPC scores in experiment j ( ) and the difference between the ranks of protein pairs based on their SPC scores in the same experiment . The average E3 scores across all experiments is the final score that is used to quantify the likelihood that two prey proteins interact . The rationale behind the E3 equation is to reward pairs of proteins that have similar SPC scores and similar ranks across all experiments , rewarding pairs of proteins with high SPC scores that appear in the same complexes . ( 3 ) The AB correlation was also developed through an intuitive manual symbolic search for functions that perform well based on benchmarking using known protein-protein interactions . The AB correlation computes the mean of the product of SPC scores normalized by dividing by the sum of mean SPC scores across all experiments . ( 4 ) The AB method also rewards pairs of proteins that have higher SPC scores in the same subset of experiments . To evaluate the predicted prey-prey protein interactions using the four equations , we used an updated version of the human literature-based protein-protein interactome we developed for the program Genes2Networks [31] . The PPIs are from 12 databases: HPRD [29] , MINT [32] , DIP [33] , MIPS [34] , PDZBase [35] , PPID [36] , BIND [37] , Reactome [38] , BioGRID [39] , SNAVI [40] , Stelzl et al . [41] , and Vidal and co-workers [42] . These databases contain direct physical interactions for mouse , rat , and human proteins containing 11 , 438 proteins connected through 84 , 047 interactions extracted manually from publications . We converted all IDs to human IDs using homologene ( http://www . ncbi . nlm . nih . gov/homologene ) . To identify domains for proteins , we used the Pfam domain database release 24 . 0 . The file ‘Pfam-A . full . gz’ was downloaded from: ftp://ftp . sanger . ac . uk/pub/databases/Pfam/releases/Pfam24 . 0/on November 1st 2010 . Domain-domain interactions ( DDI ) were obtained from the Domine database [43] . The Domine database contains 26 , 219 domain-domain interactions . Among these domain-domain interactions , 6 , 634 were inferred from the protein data bank ( PDB ) and 21 , 620 were computationally predicted by one or more of 13 prediction methods . In order to score domain-domain interactions , we developed a prediction vector containing a combined score for all predicted PPIs that contain domain-pairs at each side of a scored PPI . We assigned the score of the predicted PPI to the DDI score . Antibodies for STRN , also called Striatin , are polyclonal rabbit , and were purchased from Millipore Corp . Antibodies for CTTNBP2NL were purchased from GeneTex . MCF-7 cells were lysed in immunopreciptation buffer containing Hepes ( 50 mM , pH 7 . 4 ) , NaCl ( 150 mM ) , EDTA ( 1 mM ) , Tween-20 ( 0 . 1% ) , glycerol ( 10% ) and protease inhibitors . The lysates were pre-cleared in the presence of rabbit IgG and protein A beads . The input sample was collected after pre-clearing . Samples were rotated overnight with IgG or Striatin antibody and subsequently incubated for two hours with Protein-A beads . The washed protein-containing beads were denatured and analyzed by Western blot . The MolSoft ICM software was used to perform the domain-domain docking simulation . ICM uses a two-step method: pseudo-Brownian rigid-body docking followed by biased probability Monte Carlo minimization of the ligand side-chains , to sample conformational space in order to identify the global energy minimum for a given interaction [44] . For this specific simulation , the protein PPP2R1A ( PDB ID: 1B3U ) , the receptor , was kept rigid , while conformations of the ligand STK25 ( PDB ID: 2XIK ) were sampled around the receptor and corresponding docking scores were retrieved . Domains were then examined for interactions based on these scores .
We analyzed the experimental data from 3 , 290 IP-MS experiments targeting 1 , 083 antigens ( bait proteins ) using 1 , 796 different antibodies . These experiments detected 11 , 485 non-redundant proteins ( Dataset S1 ) . Some of the baits were pulled-down with several different antibodies . Some of the experiments with the same baits and antibodies were repeated several times but conducted under different conditions , i . e . , stimulated/un-stimulated cells , or different cell types . Complexes are mostly isolated from nuclear fractions but some experiments use cytosolic fractions . Summary of the experimental conditions , cell types , antibodies and baits used , counts of normalized peptides identified in each experiment per protein , and size of the lists of proteins identified in each experiment can be directly obtained from the primary publication provided as reference [19] . IP-MS proteomics profiling have several known experimental challenges that need to be considered when applying functional global analyses on such data . First , it is well established that the proteins identified in such experiments are enriched for highly abundant and “sticky” proteins . This results in numerous proteins appearing frequently in almost all pull-downs regardless of the cell type , cellular fraction or experimental conditions . To address this we used a list of “non-specific” proteins to filter protein identifications that appear frequently in many pull-downs ( Dataset S1 ) . For all further analyses we removed these proteins from the results . Such a “non-specific” protein list can be useful as a guideline for filtering other IP-MS proteomics data applied to human cells . However , it should be noted that the concept of filtering IP-MS proteomics data based on a “non-specific” list is only meant as a guide . The sticky non-relevant proteins may play an important biological role that would be missed by removing them . In general , proteins that appear in the list are enriched in heat shock , ribosomal , and heterogeneous nuclear ribonucleoproteins ( hnRNPs ) . Also , the majority of proteins on the non-specific list were selected based on the purifications from nuclear extracts , so some abundant cytosolic proteins may be over represented in the protein-protein and domain-domain interaction predictions since these may not have been removed . In order to integrate and visualize the results from the 3 , 290 IP-MS experiments , we first used the Jaccard Distance ( JD ) to construct a CoRegs complex similarity graph were nodes represent protein lists from each experiment and links represent overlap between experiments ( Fig . S1 ) . Nodes and links are preserved in the network if the similarity is greater than the Jaccard distance of 0 . 7 . This retained 491 experiments and 2233 links between them , which are a small portion of all possible experiments and their similarities ( Fig . S2A ) . On average , pull-down experiments reported the identification of ∼30–200 proteins but the distribution has a heavy tail with few experiments identifying over 1000 proteins ( Fig . S2B ) . Our aim in this study is to assign confidence scores to binary prey-prey protein-protein and domain-domain interactions by integrating information from the 3 , 290 IP-MS experiments . The rationale for this approach is that the experiments , reporting lists of ∼30–200 proteins for each pull-down , taken together , provide enough information to reconstruct high-fidelity , small-sized complexes and potentially enough to recover direct physical interactions between pairs of proteins and domains . We reasoned that if we use all the information across all experiments to score each pair of proteins for potential direct interaction , we will be able to identify novel associations in addition to recovering known interactions better than by chance . In contrast with most prior methods that focused on scoring bait-prey interactions , our equations predict interactions between prey proteins that commonly reappear together in different pull-downs . Although the data collected for this study was aimed at the recovery of interactions between the intended antigens ( baits ) and other proteins , the majority of primary antibodies cross-react with multiple secondary antigens and those antigens interact with other proteins . This complicates bait-prey scoring of HT-IP/MS data . Yet , logically , if two proteins reappear together at the top of lists in many different pull-downs , we can guess that they may physically interact regardless of which baits were used to pull them down , making it possible to predict likely binary interactions by utilizing the spectral counts , not just co-occurrence . To encode such logic into mathematical functions we devised four scoring schemes , each attempting to address the problem in a slightly different way . To evaluate the performance of the four scoring schemes we used known PPIs we consolidated from online databases [31] . The overall schema for this approach is depicted in Fig . 1 . To compare the performance of the different scoring methods we visualized the results as either receiver operator curve ( ROC ) ( Fig . S3 ) , random walks ( Fig . S4 ) , or a sliding window ( Fig . S5 ) . Visualization of overlap between a ranked list and a gene set using a random walk was borrowed from the popular Gene-Set Enrichment Analysis method [45] . The three equations AB , E3 , and Pr can be combined with the Sørenson coefficient to slightly improve the predictions by the AB and E3 equations , and significantly improve the predictions made with the Pr equation . AB and E3 perform best when combined with the Sørenson coefficient because these equations take into account the quantitative levels of the peptides , rewarding interactions that appear on top of the same pull-downs and penalizing potential interactions where the two proteins are not present in the same pull-down , or when one protein appears at the top and the other at the bottom . The different methods recover different sets of interactions and in some cases complement each other , suggesting perhaps that a combined weighted score may provide better results than using a single equation ( Fig . S6 , Dataset S2 ) . Next , we used ball-and-stick diagrams to visualize the results across all experiments . We first visualized all overlapping interactions listed in the top 10% of predicted protein-protein interactions by each method ( AB , E3 and Pr combined with Sor ) . This resulted in a network made of 2 , 509 proteins ( nodes ) and 28 , 886 interactions ( edges ) ( Fig . 2 ) . Using Cytoscape's organic visualization algorithm , the hubs of this network self-organize into an interesting hierarchical structure that may reflect their complex formation relationship . This network provides a global view of the CoRegs interactome , allowing zoom-in to view the identity of high confidence predicted protein-protein interactions and the complexes that these interactions form . To accomplish this zoom-in view , we increased the threshold to only include interactions from the top 1% of predicted interactions by all three scoring methods and include only three-node cliques . Three-node cliques are triangles in the network topology where three proteins are connected to each other with a maximum of three links . The resultant network contains 543 proteins and 1 , 893 interactions organized into 63 tightly connected protein complexes containing 3 to 25 proteins ( Fig . 3 ) . Many of the interactions and complexes that emerged are already known from low-throughput protein-protein interactions studies . However , some of the complexes within this network and many of the predicted protein interactions are novel . As a proof of concept , we focused on one predicted complex where most of the members of the complex were exclusively prey proteins in all experiments , and most interactions in the complex were not previously known ( Fig . 4A ) . The complex contains ten densely connected proteins with the protein STRN in the center , predicted to interact with all other nine members . STRN , STRN3 and STRN4 are scaffolding proteins with a calmodulin binding domain . Interestingly CTTNBP2NL has been previously reported with STRN and STRN3 in another IP/MS study [46] . To experimentally validate one of the interactions within this complex we used IP and western blotting to demonstrate a direct interaction between STRN and CTTNBP2NL which is another member of the predicted complex ( Fig . 4B ) . We chose this interaction based on antibody availability . Our experiment clearly shows that the two proteins interact . Such a demonstration of physical interaction experimentally does not prove that our prediction method works well , but it demonstrates how predicted interactions can be further validated experimentally . To prove that the predictions are of high quality , many such experiments need to be performed with appropriate controls to show statistically that the combined equations can predict , with high fidelity , physical interactions . Before analyzing all of the 3 , 290 IP-MS experiments published by Malovannaya et al [19] , we had access to a subset of the data before it was published . Therefore , we developed our analysis methods on a subset of 114 IP-MS experiments that are a fraction of the entire set of the 3 , 290 IP-MS experiments . In order to integrate and visualize the results from these 114 IP-MS experiments , similarly to the network shown in Fig . S1 , we created the Jaccard Distance ( JD ) CoRegs complex similarity graph ( Fig . S7 ) . Most of these initial 114 experiments used Estrogen Receptor α ( ESR1 ) and nuclear receptor co-activator 3 ( NCOA3 ) , also called SRC3 , as baits in different cellular conditions . Both proteins play an important role in breast cancer , where SRC3 serves as the main co-activator of estradiol-dependent ESR1 [47] , [48] . The experiments that used ESR1 and NCOA3 as baits resulted in similar protein lists ( clusters in the subnetwork in Fig . S7 ) compared with the other pull-downs . Using the same prediction combined scores with the three equations , with lower thresholds , we identified five distinct high confidence complexes we named: SMARC , CSTF , RCOR , MBD , and SIN3A ( Fig . S8 ) . These five complexes have been previously reported in the Corum database [49] and some have been functionally characterized ( Fig . S9 ) . Specifically , the SMARC complex highly overlaps with complex IDs 238 , 714 , 803 , and 806 in Corum , a database of reported protein complexes [49] . The CSTF complex is listed as complex number 1147 in Corum , RCOR is listed as 626 , and MBD and SIN3A have associated IDs with highly overlapping entries for complexes in Corum . The SMARC and CSTF complexes were recovered mostly from ESR1 pull-down experiments , while the other three complexes are formed by combinations of many other types of baits . Notably , the SMARC and CSTF complexes are nearly mutually exclusive to two different antibodies targeting ESR1 , and are recovered in the control experiment from HeLa cells that do not express ESR1 . Thus , one antibody is likely cross-reacting with a member of the SMARC complex , whereas the other antibody cross-reacts with a member of the CSTF complex ( Fig . S10 ) . This result highlights the importance of protein complex reconstruction from HT-IP/MS based on prey-prey co-occurrence alone , independently of the intended baits . Since PPIs are often the result of interactions between the structural domains of the interacting proteins , and since we know most of those domains for all pulled prey proteins based on their amino-acid sequences , we can use the scores for PPIs to also score and rank domain-domain interactions ( DDIs ) . The scoring of domain interactions is slightly more complex since most proteins have several different domains and the domains can appear more than once within the same protein . To resolve this we used the score for PPIs containing domains between all possible domain pairs from each side of the PPI and normalized the score across all the domains ( see methods ) . The aggregated score for all DDIs was accumulated across and within all 3 , 290 IP-MS experiments . The idea of predicting DDIs from PPIs is not new [50]–[52] . DDIs can also be predicted using structural biology methods or by evolutionary conservation of sequences across organisms [53] . To evaluate which PPI scoring method works best to predict DDIs , we compared the predicted scores for DDIs with reported DDIs from the Domine database . The Domine database contains both structurally observed and computationally predicted DDIs [43] . ROC curves and random-walk plots were used to evaluate DDI predictions , similarly to how we evaluated the PPI prediction methods ( Fig . S11 and S12 , Dataset S3 ) . The plots show that we can reliably recover known and predicted DDIs . In addition to the four equations used to score PPIs , we introduced another scoring scheme , λ , for scoring DDIs . λ is an index that counts the number of times two predicted interacting prey proteins have a domain on each side of the PPI . Such an index improves DDI predictions . In addition to the type of analysis we did for PPIs , we also attempted to further combine different prediction methods to optimize DDI predictions . Finally we visualize our predicted DDIs with known DDIs as a network diagram to visually explore interactions among all domains ( Fig . S13 ) and within the STRN centered complex identified by the PPIs predictions ( Fig . 5A ) . To further validate one of the predicted DDIs we pursued a computational structural biology approach . We attempted to dock the PKinase domain of STK25 to the HEAT domain of PPP2R1A . We chose these two proteins because they had a crystal structure in PDB . Although the DDI is already listed in Domine , the prediction of this DDI interaction is based on sequence and homology . Hence there is no direct evidence of such interaction between these two proteins and their domains . Using the Molsoft ICM software we obtained a docking score of −46 . 75 kcal/mol . This score is considered high and as such confirms the interaction . By examining the confirmation of this interaction it appears that the Pkinase domain of the STK25 protein binds to the HEAT domain of PPP21RA . The energy gap of approximately 2 kcal/mol ( ICM score units ) between the best obtained and next consecutive docking score clearly suggests strong recognition of the HEAT domain by the Pkinase domain ( Fig . 5B–D ) .
In this study we combined results from 3 , 290 experiments that identified nuclear protein complexes in human cells using IP-MS . We implemented and evaluated four different equations assessing their ability to predict direct physical PPIs from the aggregated proteomics data using known PPIs from the literature . The highest scoring predictions were visualized as networks with many densely connected clusters that are likely made of real protein complexes . The prediction scores for potential interactions could be considered as surrogates to real affinity constants . However , since we do not know the exact quantities of proteins , it is not possible to compute exact dissociation constants . Such binding constants can be useful for dynamical simulations where we could stochastically trace the transient dynamics of CoRegs complex formation in-silico . Scoring PPIs by only using the prey measurements may become more robust as more IP-MS experiments are published . However , careful attention should be given to weighting the repetitiveness of experiments so interactions from similar pull-downs , if repeated , are not mistakenly given higher scores . Regardless of possible limitations , the ability to recover direct PPIs based on such a massive dataset is an important step toward utilizing HT/IP-MS datasets for reconstructing networks and generating hypotheses . In addition , we show that the equations can be extended to predict interactions between structural domains . We also demonstrated two ways to further validate predicted PPIs and DDIs , using experimental and computational approaches . In summary , our analyses explored new methodologies for scoring PPIs and DDIs using data from related IP-MS experiments , providing many hypotheses about mammalian CoRegs complexes formation , and allowing users to explore novel complexes , PPIs and DDIs online at http://maayanlab . net/HT-IP-MS-2-PPI-DDI/ . This resource can help us advance the catalogue of transcriptional regulation by CoRegs in normal and diseased mammalian cells . | In response to various extracellular stimuli , protein complexes are transiently assembled within the nucleus of cells to regulate gene transcription in a context dependent manner . Here we analyzed data from 3 , 290 proteomics experiments that used as bait different member proteins from regulatory complexes with different antibodies . Such proteomics experiments attempt to characterize complex membership for other proteins that associate with bait proteins . However , the experiments are noisy and aggregation of the data from many pull-down experiments is computationally challenging . To this end we developed and evaluated several equations that score pair-wise interactions based on co-occurrence in different but related pull-down experiments . We compared and evaluated the scoring methods and combined them to recover known , and discover new , complexes and protein-protein interactions . We also applied the same equations to predict domain-domain interactions that might underlie the protein interactions and complex formation . As a proof of concept , we experimentally validated one predicted protein-protein interaction and one predicted domain-domain interaction using different methods . Such rich information about binary interactions between proteins and domains should advance our knowledge of transcriptional regulation by CoRegs in normal and diseased human cells . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"systems",
"biology",
"spectrometric",
"identification",
"of",
"proteins",
"biochemistry",
"mathematics",
"protein",
"interactions",
"protein",
"abundance",
"statistics",
"regulatory",
"networks",
"biology",
"computational",
"biology",
"signaling",
"networks",
"proteomics",
"biostatistics"
]
| 2011 | Recovering Protein-Protein and Domain-Domain Interactions from Aggregation of IP-MS Proteomics of Coregulator Complexes |
In an inflammatory setting , macrophages can be polarized to an inflammatory M1 phenotype or to an anti-inflammatory M2 phenotype , as well as existing on a spectrum between these two extremes . Dysfunction of this phenotypic switch can result in a population imbalance that leads to chronic wounds or disease due to unresolved inflammation . Therapeutic interventions that target macrophages have therefore been proposed and implemented in diseases that feature chronic inflammation such as diabetes mellitus and atherosclerosis . We have developed a model for the sequential influx of immune cells in the peritoneal cavity in response to a bacterial stimulus that includes macrophage polarization , with the simplifying assumption that macrophages can be classified as M1 or M2 . With this model , we were able to reproduce the expected timing of sequential influx of immune cells and mediators in a general inflammatory setting . We then fit this model to in vivo experimental data obtained from a mouse peritonitis model of inflammation , which is widely used to evaluate endogenous processes in response to an inflammatory stimulus . Model robustness is explored with local structural and practical identifiability of the proposed model a posteriori . Additionally , we perform sensitivity analysis that identifies the population of apoptotic neutrophils as a key driver of the inflammatory process . Finally , we simulate a selection of proposed therapies including points of intervention in the case of delayed neutrophil apoptosis , which our model predicts will result in a sustained inflammatory response . Our model can therefore provide hypothesis testing for therapeutic interventions that target macrophage phenotype and predict outcomes to be validated by subsequent experimentation .
Macrophages play an essential role in both the progression and the resolution of inflammation . These contradictory roles may be explained by the idea of a spectrum of macrophage phenotypes , ranging from the inflammatory M1 phenotype to the anti-inflammatory M2 phenotype at either extreme , with diverse subpopulations of macrophages in between [1–4] . Another possible explanation is that macrophages exhibit typically M1 or M2 type functions to varying degrees at various points in time , or there are portions of each type present at each of the different phases of inflammation [5] . While this duality of purpose is not fully understood , it is known that an imbalance between pro- and anti-inflammatory macrophage activities has been linked to disordered healing and implicated in many inflammatory diseases . For example , overpopulation of M1 macrophages can induce tissue injury [1] , and the accumulation of M1s in adipose tissue which secrete pro-inflammatory cytokines can lead to insulin resistance , diabetes , and atherosclerosis [6 , 7] . Even M2 macrophages , which are thought of as resolving inflammation , can cause disorders such as allergies , asthma , fibrosis , and excessive scarring when present in large numbers [4] . There is also an increased association of M2 polarized macrophages with solid tumor formation [8 , 9] . All macrophages begin life as monocytes circulating in the bloodstream and , upon settling into tissues and organs in the body , will adapt to their local environment . At an inflamed site , monocytes are triggered to differentiate into macrophages in response to stimuli such as chemokines and cytokines in the environment , phagocytosis of apoptotic cells or debris , or the presence of pathogen [1 , 4 , 7 , 10 , 11] . These first invading macrophages primarily activate to a more M1 phenotype but , under normal conditions , M2 macrophages producing anti-inflammatory cytokines will eventually dominate , suppressing the inflammatory and Th1 adaptive immune response , while promoting a Th2 response [4] . In response to infection or presence of pathogens , neutrophils are the first immune cell to appear to facilitate removal . Subsequent macrophage infiltration is essential for the removal of apoptotic neutrophils and continued secretion of cytokines to further limit the effects of the invading pathogens [12] . This timely recruitment and egress of immune cells is central to the mounting of an appropriate immune response that resolves to restore tissue homeostasis . Dysfunction or disruption of this response is the cause of essentially all chronic inflammatory diseases . Appropriate switching of phenotype of the overall macrophage population from initial M1 to M2 phenotype is critical for a balanced response . Knowledge of which subpopulations of macrophages to modulate is therefore necessary for the development of interventions that can aid in the resolution of inflammation . Mathematical modeling has been extensively applied to the problem of inflammation in a variety of contexts such as wound healing [13–24] and atherosclerosis [25–32] . Deterministic ordinary differential equations ( ODEs ) in particular have been used when the primary interest is capturing time course and/or qualitative behavior at the cellular level . Reynolds et al . in 2006 [14] modeled the innate immune response to pathogen including activated phagocytes , level of pathogen , tissue damage , and anti-inflammatory mediators and this model was modified to apply to a local wound with the inclusion of fibroblast activity and the effect of tissue oxygen levels in Menke et al . [17] . The work was further extended by Segal et al . in 2012 [20] , adding collagen accumulation as a means of tracking the healing progress . Cooper et al . [23] next tracked macrophages and neutrophils specifically rather than a single variable representing immune response . Phagocytosis of apoptotic neutrophils was considered a key driver of the resolution of inflammation in models developed by Dunster et al . [22] . In a study analyzing macrophage polarization following myocardial infarction , Wang et al . [21] tracked both M1 and M2 macrophages as well as pro- and anti-inflammatory mediators . Recent work by Lee et al . [33] models M1 and M2 macrophage response to respiratory viral infection along with epithelial cells , cytokines , and enzymes . In this manuscript , we draw on the work done in these previous models to develop a new computational model of inflammation that seeks , in part , to explain the relationship between macrophage polarization and neutrophils . To our knowledge , our model is the first to include both inflammatory M1s and resolving M2s that is fit to in vivo experimental data . We first use ODEs to develop a computational model of the sequential influx of immune cells in response to an external trigger to permit a system-level analysis of the processes . We then parametrize the model by fitting to cell count data for neutrophils , M1 macrophages , and M2 macrophages obtained from a mouse model of peritonitis , a well-accepted model to assess inflammatory responses in vivo that is also widely used to evaluate the efficacy of targeted anti-inflammatory interventions . This step entails finding a subset of identifiable parameters to estimate and fixing those that were unidentifiable , a process that has many approaches across a wide application area [34–39] . Once a final parameter set is estimated , we conduct a local sensitivity analysis of the fitted model in order to gain an understanding of the primary controls of the system . The results support the dependence of macrophage polarization on neutrophils that has been hypothesized in the literature [1 , 3 , 40–42] . Finally , we use the model to test several macrophage-targeted treatment scenarios that are hypothesized to dampen inflammation . The resulting predictions could have implications in the development of treatment strategies for chronic inflammation .
The use of animals for this study was approved by VCU IACUC and the approved protocol number is AM10346 with an approval date of 4-14-18 and this approval will expire on 3-13-2021 . Isofluorane inhalation was used for euthanasia . Induction of peritonitis by intraperitoneal injection of thioglycollate broth , which will facilitate the rapid growth of bacteria in the peritoneal cavity , is a well-suited platform to monitor the influx of immune cells and also permits easy characterization of the infiltrating cells in a time dependent manner . Peritoneal exudates were harvested from mice at 10 different time-points over 7 days after a single intraperitoneal injection of 3% thioglycollate broth . The peritoneal cavity was flushed with serum free RPMI medium . The cells were collected by brief centrifugation , re-suspended , and then stained with fluorescently conjugated antibodies to CD45 , CD11b , Ly6G ( Gr-1 ) , F4/80 and Ly6C and analyzed by flow cytometry to determine the distribution of neutrophils , macrophages and Ly6CHi ( M1 ) or Ly6CLo ( M2 ) polarization [43] . While all leukocytes are CD45+ , neutrophils and macrophages can be distinguished by the presence of specific markers , namely Ly6G or Gr1 and CD11b or F4/80 on neutrophils and macrophages , respectively . The macrophages in the peritoneal exudates can further be differentiated into resident ( CD11bHi and F4/80Hi ) , inflammatory M1 ( CD11b+Ly6CHi ) and anti-inflammatory M2 ( CD11b+Ly6CLo ) phenotypes . The gating strategy and representative dot plots and histograms used to identify individual cell populations are shown in Fig 1 . Flow cytometry data was analyzed using the FlowJo software and percent distribution of individual cell type determined as described earlier [43] . The data collected from these experiments is used to calibrate the model parameters ( see Supporting Information ) . The model developed in this manuscript tracks the signaling and resulting immune response within the peritoneal cavity . We do not explicitly model the blood component and all variables represent local levels . To create this model , previous models of immune response to a wound [14 , 20 , 23] have been adapted to include polarization of macrophages between phenotypes M1 and M2 , transition of neutrophils to the apoptotic state , and the injection of nutrient broth to induce growth of pathogen and stimulate immune response . System variables include cell populations given by M1 ( M1 macrophages ) , M2 ( M2 macrophages ) , N ( neutrophils ) , and AN ( apoptotic neutrophils ) as well as P ( pathogen ) and B ( inflammatory stimulus ) . We track the total cells for each population with units of 107 cells . Model parameters for rates of activation , transition , decay , and interactions are specified in Table 1 . Units for many of the model parameters are given in terms of their associated variable , since they are representative of immune functions such as cell signaling and mediators for which units cannot be determined . The model is summarized in Fig 2 and described by Eqs 1–6 . Macrophages: dM1dt=smrRM1 ( P , N , M1 , AN ) μmr+RM1 ( P , N , M1 , AN ) +RM2 ( M2 ) ︷activation/influxrate−km1m2kanm1ANfi ( M1 , N ) ︷switchtoM2fromM1perphagocytizedAN+km2m1M2︷switchtoM1fromM2−μm1M1︷decay ( 1 ) dM2dt=smrRM2 ( M2 ) μmr+RM1 ( P , N , M1 , AN ) +RM2 ( M2 ) ︷activation/influxrate+km1m2kanm1ANfi ( M1 , N ) ︷switchratefromM1perphagocytizedAN−km2m1M2︷switchfromM2toM1−μm2M2︷decay ( 2 ) where the activation/influx rates for M1 and M2 are given by RM1=km1pP︷activationbyP+km1nN︷activationbybyproductsofN+km1m1M1︷activationbyM1sandtheircytokines+km1anμanAN︷activationbynecroticANRM2=km2m2M2︷activationbyM2sandtheircytokines+kc︷backgroundanti-inflammatorycytokines Neutrophils: dNdt=snrRN ( P , AN ) μnr+RN ( P , AN ) ︷activationrate−kanN︷apoptosis ( 3 ) dANdt=kanN︷apoptotisofN−kanm1ANfi ( M1 , N ) ︷removalbyM1−kanm2ANfi ( M2 , N ) ︷removalbyM2−kannN︷removalbyN−μanAN︷secondarynecrosis ( 4 ) where the activation rate for neutrophils is RN=knpP︷activationbyP+knanμanAN︷activationbynecroticAN Inflammatory Stimulus: dPdt=kpgP ( 1−PP∞+B ) ︷logisticbroth-dependentgrowth−kpnPN︷removalbyN−kpmPfi ( M1 , N ) ︷removalbyM1−kpmPfi ( M2 , N ) ︷removalbyM2 ( 5 ) dBdt=−kbBP︷consumptionbyP ( 6 ) Inhibition function: fi ( x , N ) =x1+ ( Nn∞ ) 2 The healthy peritoneal cavity is impermeable and is assumed to be nearly sterile prior to inflammatory stimulus , with very low levels of pathogen , and so has no immune cell influx . Therefore , all of our immune cell variables have an initial condition of zero . The injection of nutrient broth is assumed to stimulate a very rapid increase in pathogen growth that quickly subsides as broth is consumed and pathogen is removed by macrophages and neutrophils . This brief spike in pathogen modeled by Eqs 5 and 6 initiates the subsequent immune cell response . As in Cooper et al . [23] , immune cells are assumed to activate and influx into the local environment rapidly compared to other dynamics , so the quasi-steady state assumption is used . This gives rise to Michaelis-Menten type activation and influx terms in Eqs 1–3 . In addition , we do not explicitly model cytokines but instead allow the production of immune cells to act as an indicator of associated cytokine level . Resting neutrophils are the first immune cells to arrive at the site of infection , rapidly becoming activated by pathogen and the debris formed by apoptotic neutrophils at the rate RN ( P , AN ) . As neutrophils become laden with bacteria , they undergo apoptosis at rate kan . Apoptotic neutrophils are then removed by M1s at rate kanm1 , M2s at rate kanm2 , and by active neutrophils at rate kann . We have chosen kann to be much smaller than both kanm1 and kanm2 as appropriate for the case when both macrophages and neutrophils are present , but in the absence of macrophages , the clearance of apoptotic cells by neutrophils may take on greater importance [44 , 45] . Apoptotic neutrophils that are not cleared undergo secondary necrosis at rate μan , contributing to the positive feedback described in the neutrophil activation term RN . Resting monocytes ( MR ) are next to arrive . The majority of these first monocytes differentiate to an inflammatory M1 phenotype in response to pathogen , byproducts of neutrophils , M1s and their cytokines , and cytokines spilled by necrotic apoptotic neutrophils at rate RM1 ( P , N , M1 , AN ) . Background levels of anti-inflammatory cytokines , kc ( related to the anti-inflammatory source term in Reynolds et al . [14] ) , cause a small portion of monocytes to differentiate to an M2 phenotype . Intrinsic decay is assumed to occur at rate μm1 in M1s and at rate μm2 in M2s . M1s are assumed to be able to switch to M2s at rate km1m2 , and this switch is assumed to be promoted by the phagocytosis of apoptotic cells [1 , 3 , 42 , 46] . Plasticity of macrophage phenotype is not fully understood , therefore , we allow for the possibility of a transition from M2 to M1 in Eq 1 at rate km2m1 as well . Late arriving monocytes are assumed to be able to activate to the M2 phenotype in response to anti-inflammatory cytokines produced by M2s at rate RM2 ( M2 ) . The inhibition term fi ( x , N ) models the inhibition of macrophage activity by neutrophils due to oxidation of the environment . The same parameter , n∞ , is used to determine the level at which the presence of neutrophils inhibit the macrophages regardless of phenotype ( M1 or M2 ) and what they are phagocytosing ( pathogen or apoptotic neutrophils ) . This is due to the simplifying assumption that all macrophages are inhibited the same by the oxidative stress in the local environment . Cell count data is given in units of 107 cells . The model given by Eqs 1–6 was fit to experimental data using the trust region method within PottersWheel , a MATLAB toolbox for parameter estimation [47] . The trust region approach uses the lsqnonlin algorithm of MATLAB’s optimization toolbox , which allows for the specification of bounds on the parameter space to be searched . Bounds for each parameter are given in Table 1 . The fitting procedure was then performed iteratively via weighted least squares with merit function χ2 ( p→ ) =∑i=1n ( yi−y ( ti;p→ ) σi¯ ) 2 ( 7 ) with p→ the vector of estimated parameters , yi the observations , y ( ti;p→ ) the model predictions given the parameter estimates , σi¯ the standard errors , and n equal to the total number of observations over all response variables . Minimizing χ2 ( p→ ) /2 is equivalent to maximizing the log-likelihood logL ( p→|ydata ) =−∑i ( yidata−yimodel ) 22σi2−Nlog2π−∑ilogσi since only the first term is parameter-dependant [47] . Fitting was performed in logarithmic parameter space since some parameter bounds span several orders of magnitude . This local optimization routine seeks parameters that minimize the sum of squared errors between the data and model predictions while accounting for variance . Since each observable N , M1 , and M2 has high standard deviations for measurements taken at time points near the maximum , weighting by these standard deviations would result in compliance with many models . We chose instead to use error model σi = 0 . 05yi + 0 . 1max ( y ) , assuming 5% uncertainty at each time point and 10% overall uncertainty relative to the maximum of each observable . At each step of the fitting process , parameter estimations were performed iteratively to ensure minimization of the merit function . Results at each step were analyzed to determine free and fixed parameters and to narrow the search for an identifiable subset of parameters as described in the Results section . Under the assumption that residuals between the data and model predictions are Gaussian distributed , the log-likelihood is distributed like a χ2 distribution with N − M degrees of freedom , with N data points and M parameters being estimated [47] . PottersWheel calculates a χ2 p-value after each fit for the null hypothesis that ( 1 ) the model sufficiently explains the data , ( 2 ) true standard deviations do not exceed standard deviation estimates , and ( 3 ) the residuals are normally distributed [47] . PottersWheel also calculates the Akaike Information Criterion AIC=−2logL+2p for a model with p parameters [48] . Given two models under consideration , the one with the lowest AIC value is preferred .
Structural identifiability ( SI ) is a prerequisite for model prediction [49] , while numerical or practical identifiability is required to determine confidence intervals around parameter estimates and ensure that the connection between the dynamic model and the data model is sufficiently strong for prediction . Determining which parameters can be uniquely determined , or at least limited to a finite range of possible values , is also a critical step in informing further experimentation . This process includes selecting parameters that significantly impact model outputs as well as defining interactions between parameters that can influence parameter estimates obtained during fitting . In this section , we analyze local parameter identifiability as outlined in the steps below and in Fig 3: Estimate all parameters . Use the fitted model to generate the discretized sensitivity matrix S . Fix insensitive parameters . Use S to rank parameters by sensitivity . Set a threshold such that parameters with sensitivity below the threshold ( insensitive ) are fixed and parameters with sensitivity above the threshold ( sensitive ) are analyzed in Step 3 . Select low collinearity group of parameters as identifiable ( ID ) subset . Check for pairwise correlations between parameters by deriving an approximate correlation matrix from S . Check for collinearity between groups of parameters with a collinearity index ( CI ) measure . Set a threshold such that groups of parameters with CI above the threshold are considered collinear . Groups of parameters with CI below the threshold are considered identifiable subsets . Estimate identifiable ( ID ) subset of parameters . One identifiable subset of parameters is selected to be estimated . The remaining parameters are fixed . Model parameters were estimated using a maximum likelihood equivalent criterion and trust region search algorithm as described ( see Materials and methods ) . Since reducing parameters to be estimated can be considered a form of model reduction [50] , we refer to our final model with 6 estimated parameters as the “identifiable” model versus the “full” model with all 24 parameters estimated in the comparisons below . First , we performed parameter estimation on the full model . For all three observable model outputs ( N , M1 , and M2 ) sampled at 10 time points with 24 model parameters , a 30 × 24 discretized sensitivity matrix S is produced . To test structural identifiability of the model a posteriori , we generated these matrices at a variety of locations in parameter space within the bounds given in Table 1 and found the rank and the singular values for each . Since each of these matrices was determined to have full column rank and no zero singular values , we concluded that the model is locally SI [51] within the bounds we had set for parameter estimation . Next , we ranked the impact of each parameter on all three observable model outputs ( N , M1 , and M2 ) by calculating a root mean square sensitivity measure , as defined in Brun et al . [34] , for each column j of the normalized sensitivity matrix as RMSj=1n∑i=1n ( pjyi∂yi∂pj ) 2 . Parameter j is deemed insensitive if RMSj is less than 5% of the value of the maximum RMS value calculated over all parameters . By this measure , 8 parameters were deemed insensitive , as shown in Fig 4 , and fixed at their nominal values . We had determined that all singular values were greater than zero , indicating SI , but only 6 of the 24 singular values obtained had values with order of magnitude greater than zero . If we consider the very small singular values essentially zero for the purpose of rank calculation ( in order to reduce problems with numerical identifiability ) this gives rank ( S ) = 6 , and since rank ( S ) can be used to identify the number of parameters that can be included in an identifiable subset [36 , 50] , a subset of size 6 is suggested . The parameter estimation problem was therefore reduced to finding identifiable subsets of size 6 out of the 16 sensitive parameters . The estimated correlation matrix for the sensitive subset of parameters , shown in Fig 5 , shows a large number of dependencies between pairs of parameters . Effects of nearly linearly dependent parameters on output are pairwise indistinguishable and cannot be reliably estimated , due to compensating effects by changes in other parameters in the group . In addition to discovering pairwise parameter relationships , we sought a minimally correlated group of 6 parameters . A measure that applies to parameter subsets of any size is the collinearity index defined by Brun et al . [34] as CI=1λk where λk is the smallest eigenvalue of S¯kTS¯k and S¯ is a submatrix of S containing the sensitivity vectors for parameters in subset K . In practical terms , changes in model output caused by a change in parameter pj can be compensated for by other parameters by up to 1CI ( e . g . , for CI = 20 a change in output caused by a change in pj can be compensated for up to 5% by other parameters in subset K ) [34] . A cutoff of CI = 20 was used to select subsets of parameters with low collinearity . Collinearity indices were calculated for parameter subsets of increasing size as described in Brun et al . [34] , using code in the VisId MATLAB toolbox [35] . Thirteen parameter pairs that were found highly correlated by this measure are shown in Table 2; others are not shown due to the large number of collinear groups ( for example , there were 68 highly collinear parameter subsets of size 3 ) . No subsets of size greater than 6 met our criteria for low collinearity between parameters . In all , 25 parameter subsets of size 6 met our criteria , involving 10 different parameters ( shown in Table 3 ) . In selecting one of these parameter subsets to be estimated in an identifiable model , we considered several factors . First , from a practical standpoint , it was desirable to choose parameters that may be reasonably estimated from currently available data and also that we hope to vary in future simulated experiments . Next , we sought to both minimize the CI and maximize the sum of the RMS sensitivity measures over all of the parameters in the subset . Minimizing the CI reduces the likelihood of parameter dependencies interfering with optimization , while choosing the subset with the most sensitive parameters should require the smallest adjustment to their values [50] . The chosen identifiable subset of 6 parameters is shown in Table 4 along with pointwise 95% confidence intervals calculated based on the approximate Hessian matrix of the objective function given in Eq 7 , as described in Maiwald et al . [47] . The fit of the identifiable model to M1 , M2 , and neutrophil data along with state variable predictions for pathogen , nutrient broth , and apoptotic neutrophils are shown in Fig 6 . A plot of the differences between model predictions and observations is available in the Supporting Information . The full model , with 24 parameters estimated , and the identifiable model , with 6 parameters estimated , are compared with respect to goodness-of-fit using the Akaike information criterion ( AIC ) and χ2 test ( see Methods ) in Table 5 . By these measures , the data is best explained by the identifiable model even though the difference in χ2 metric value between models is small . There is close agreement between model predictions and observations achieved with our obtained parameters , however , we remark that there is some dependency between fixed and estimated parameters and that there are inherent limitations in estimating parameters with limited experimental data . Therefore , these estimates should be taken as conditional , and we can determine which fixed parameters they may be conditioned on by viewing the profile likelihood [37 , 52 , 53] . The profile likelihood approach for analyzing identifiability fixes a parameter pi at values over a specified range , re-estimating all other parameters at each point [52]: χPL2 ( pi ) =minpj≠i[χ2 ( p ) ] . Using the profile likelihood , it is possible to trace out the functional form of identifiable combinations of parameters , and this information can be used in re-parametrization [36 , 37] . However , this requires reducing extra degrees of freedom in the estimated parameters in order to avoid compensation effects . [37] . Even with collinearities present , it is possible to get an idea of compensation effects between parameters during fitting by observing how estimated parameters change over the profiled parameter . This can be important in determining whether estimated parameters are conditional on parameters that were fixed prior to fitting [34] . We have plotted the profile likelihoods of parameters in the identifiable subset versus other parameters that change significantly over the profile likelihood ( see Supporting Information ) . The impact that both fixed and estimated parameters have on predictions for M1 and M2 macrophages was analyzed with one-at-a-time sensitivity analysis . We focused our analysis on these two observable outputs since our goal is to identify drivers of population level phenotype switch in macrophages . In applying this method , we increased each parameter by a factor of 1 . 001 of its baseline value while holding all other parameters at their baseline values to determine the effects on the M1 and M2 characteristics shown in Figs 7 and 8 . The sensitivity of characteristic f with respect to parameter p is then estimated as s = ( f ( 1 . 001 * p ) − f ( p ) ) / ( 1 . 001 * p − p ) * p/f using the PottersWheel MATLAB toolbox [47] . The parameter is then reset to its baseline value and the process is repeated for the next parameter , until sensitivity of all parameters is analyzed . Baseline values for parameters that were fixed during fitting are given in Table 1 and baseline estimated parameter values are given in Table 4 . Baseline characteristics of each cell type are shown in Figs 7 and 8 , along with sensitivities of each characteristic to variations in each parameter . Since parameters are varied individually , this analysis does not take into account interactions between variables that may influence model results in unexpected ways if more than one parameter is varied simultaneously . Taken with the above caution , however , we can gain some insight into which factors may drive macrophage phenotype balance . The most influential parameters on M1 behavior are snr and smr ( availability of resting neutrophils and monocytes ) , kpg ( behavior of inflammatory stimulus ) , km1p and knp ( response of M1s and neutrophils to inflammatory stimulus ) , and uan ( rate of secondary necrosis of neutrophils ) . In the present context , M1s are primarily activated by initial inflammatory stimulus and necrosis of apoptotic neutrophils that have not been phagocytosed . This supports the hypothesis that effective clearance of apoptotic cells is important in the resolution of inflammation [1 , 40 , 46 , 54–59] . If our parameter estimates had been obtained by fitting to data from chronic inflammation , feedback from existing M1s and the pro-inflammatory byproducts of existing neutrophils would likely be greater contributors to M1 response . Negatively related to magnitude of M1 response are parameters μm1 ( decay or efflux rate of M1s ) and n∞ ( the level of neutrophils required to inhibit macrophage activity by 50% ) . As the threshold for inhibition of M1s increases , the magnitude of the M1 population decreases because less M1s are required to mount an adequate response . The importance of neutrophils and neutrophil apoptosis in mounting a timely and sufficient M2 response is evidenced by the high sensitivity of M2 peak timing and amplitude to neutrophil-associated parameters snr , uan , kan , knp , unr , n∞ , kanm1 , kanm2 , and knan . The magnitude of the M2 population peak is also strongly positively associated with km1m2 ( switch rate from M1s ) and km2m2 ( feedback from existing M2s ) . Increasing rates of decay or efflux for resting monocytes ( μmr ) and resting neutrophils ( unr ) diminishes M2 population magnitude , as does reduced M1 activation by pathogen ( km1p ) , indicating M2 dependence on the population size of other immune cells . Our objective in this work is to identify key drivers of macrophage phenotype balance during the inflammatory response , in order to identify potential clinical targets . Therefore we now perturb parameters from fitted values in order to view effects on model behavior and simulate therapeutic targeting of macrophages for intervention in the early inflammatory process critical to disease progression , as has been proposed [60–62] . One proposed strategy to dampen inflammation is to directly polarize M1 macrophages to an M2 phenotype [60] . To evaluate the effects of varying the transition rate of M1 to M2 , we varied parameter km1m2 over 10 linearly spaced values within a factor of 1 ± . 3 of its baseline value . The model predicts that increasing km1m2 has a small effect on M1 magnitude of response while increasing the magnitude of M2 response , which is expected . However , the time course of both macrophage populations is predicted to be shortened due to a higher transition rate; whether this results in faster resolution of inflammation or an insufficient M2 population for a subsequent proliferation or repair phase may depend on the nature and magnitude of the inflammatory stimulus . Next , we simulated a change in the apoptosis rate of neutrophils , kan , based on our hypothesis that efferocytosis ( phagocytic removal of apoptotic and necrotic cells ) is a key driver of macrophage phenotype change and that this requires a sufficiently sized population of apoptotic cells [1 , 3 , 40–42] . Dysregulation of neutrophil population level and turnover is known to be a direct contributor to human inflammatory and autoimmune diseases such as coronary artery disease , rheumatoid arthritis , acute arterial occlusions , gout , asthma , and many others [63 , 64] . Macrophages themselves are known to modulate neutrophil lifespan by releasing cytokines that can delay apoptosis [65] and some microbial pathogens delay or accelerate neutrophil apoptosis to promote their own growth [63] . From the results in Fig 9 , we note that modulating the size kan has some interesting effects . In the biologically unlikely case where kan = 0 and there is no population of apoptotic neutrophils available for efferocytosis , neutrophils remain the dominant immune cell . For low values of kan , sustained inflammation appears to be the result of too many inflammatory neutrophil byproducts and the low M2 population levels . Midrange kan values were determined during fitting to produce a normal response , while higher kan levels seem to produce faster resolution similar to increasing the transition rate km1m2 . This is unsurprising given the dependency of the second term of Eq 1 on km1m2 , kanm1 , and AN , which tracks the size of the apoptotic neutrophil population . Yet the magnitude of the effects of modulating kan versus acting on transition directly via km1m2 are predicted to diverge for lower values , with the former providing more dramatic changes . To explore points of intervention in the case of delayed neutrophil apoptosis , we set kan = 5 . 56 . This results in sustained inflammation as shown in Fig 9 , and changes in sensitivity to parameters across this bistability is also shown in Fig 10 . For example , with delayed neutrophil apoptosis ( unhealthy case ) , the number of M1s remaining at day 7 becomes strongly positively associated with parameter snr ( influx rate of resting neutrophils ) and the number of M2s remaining at day 7 becomes negatively associated with increased μm2 . By changing these as shown in Fig 11 we are able to resolve inflammation in spite of impaired neutrophil apoptosis . Modulating resting neutrophils by either reducing influx ( simulated by lowering the value of snr ) or increasing decay or efflux ( simulated by increasing the value of unr ) returns all immune cell populations to homeostasis . However , reducing decay or efflux of M2s ( by lowering the value of μm2 ) led to a resolution of inflammation but a sustained M2 population that could potentially be problematic . Finally , we simulated reducing availability of monocytes for recruitment by reducing monocyte source parameter , smr , by 1/2 at early versus late time points ( 16 hours or 5 days ) to compare effects as shown in Fig 12 . Resulting predictions support what has been demonstrated experimentally: that intervening at early timepoints to block or reduce monocyte recruitment and their subsequent differentiation to inflammatory macrophages can actually impair resolution of inflammation [60 , 66 , 67] .
Modulating macrophage subpopulations has been proposed as a strategy to resolve inflammation [60–62 , 68] , but the mechanisms driving macrophage phenotypic switch are not well understood . In this work we have developed a model that includes macrophage polarization during inflammation . To our knowledge , it is the first model of its kind to be fit to in vivo experimental data . Our model allows some insight into key drivers of macrophage population shift over the time course of inflammation and allows us to predict the effects of therapies targeting macrophages . The experimental data used to fit this mathematical model was obtained from the widely studied peritonitis model of inflammation . In addition to recapitulating the influx and egress of inflammatory cells in response to stimulus-induced inflammation , this model is also extensively used to assess the involvement of endogenous processes in mounting as well resolving the inflammatory processes . In recent studies , the pro-inflammatory role of human proteinase 3 ( PR3 ) during acute inflammatory responses by modulating neutrophil accumulation and the underlying mechanisms were almost entirely determined using a zymosan-induced peritonitis model [69] . Extending the investigations into endogenously produced pro-resolving lipid mediators , Ramon et al . not only identified PCTR1 , a member of the protectin family as a potent monocyte/macrophage agonist but also established the therapeutic potential of PCTR1 supplementation in resolving inflammation using microbial-induced peritonitis in mice [70] . Similarly , Juhas et al . confirmed the ability of RX-207 to reduce neutrophil migration using thioglycollate-induced peritonitis [71] . These examples not only underscore the importance of developing a mathematical model based on experimental data from mouse peritonitis , but also provide the rationale and future application of such a model for evaluating and predicting outcomes to be validated by subsequent experimentation . The process of parameter selection is fully elucidated ( see Results ) . Parameter estimation was carefully conducted such that unidentifiable parameters were fixed and the confounding effects of parameter interactions were reduced in order to obtain an identifiable subset of parameters of interest for estimation . We also stipulate that other , equally viable , identifiable subsets could have been estimated ( see Table 3 ) and that estimated parameters may be conditional on parameters that were previously fixed . It is important to acknowledge that parameters chosen for estimation will depend on the experimental context and available measurements . It is hypothesized that efferocytosis of apoptotic cells is an important determinant of macrophage phenotype [1 , 3 , 40–42] , and our sensitivity analysis supports the dependence of macrophage behavior on neutrophils . Our analysis indicates that timing and magnitude of the M2 response in particular is closely related to neutrophil dynamics . We simulated several treatment scenarios targeting macrophages both directly and indirectly . We compared the effects of targeting macrophage transition rate directly ( in the model via parameter km1m2 ) versus varying neutrophil apoptosis rate , kan , in order to increase the population of apoptotic cells available for macrophage efferocytosis . A shorter time course of both M1 and M2 response is predicted in both cases; whether this indicates fast resolution or introduces the possibility of an insufficient M2 population given a sustained pathogen insult or injury requires further examination . Our model predicts that timing may be critical in blocking or reducing availability of monocytes in order to reduce the inflammatory M1 response , as has been proposed , and that this could lead to chronic inflammation . These effects have been observed in an experimental setting as well [60 , 66 , 67] . Since pro- and anti-inflammatory mediators could not be measured experimentally , we instead used cellular feedback loops to describe their contribution to inflammatory processes . The future addition of parameters such as local production/levels of pro- or anti-inflammatory mediators that likely influence the function of infiltrated immune cells will further fine-tune this model . It is noteworthy that using the mouse model of peritonitis , Dequine et al . demonstrated that local TNFR1 signaling modulated neutrophils for increased cytokine production with implications on neutrophil recruitment and egress [72] . Further experimentation is also likely to allow a larger identifiable subset of parameters , especially if cytokines associated with the various cell types are explicitly measured , giving a stronger connection between available data and feedback loop components in the model . In future work , this peritonitis model will be extended to the case of early atherosclerosis . In addition to the routinely monitored changes in serum lipid profiles , changes in monocytosis as well as increased circulation of pro-inflammatory mediators are also causally related to atherogenesis and chronic unresolved inflammation is recognized as an underlying cause of multiple metabolic diseases . It is noteworthy that Angsana et al . reported a positive correlation between delayed clearance of macrophages from the peritoneal cavity and atherosclerotic plaque burden [73] and Feige et al . showed that a small molecule lecinoxoid ( VB-201 ) which reduced monocyte migration in a peritonitis model , also reduced atheroma development [74] . These studies underscore the predictive value of computational models based on cellular influx/egress from the peritoneal cavity . Chronic inflammatory diseases in general require timely peaks and ebbs in immune cell response in order for homeostastis to be restored; particularly in macrophages , which include subpopulations that either contribute to or resolve inflammation . In the case of atherosclerosis , this phenotype switching is believed to be critical to a balanced response to hyperlipidemia . Our extended model will be able to provide hypothesis testing for points of intervention in atherosclerosis that target macrophage phenotype . Jacinto et al . have recently demonstrated the importance of extra-arterial contributors such as functionality of monocytes in aggravation of atherosclerosis under normocholesterolemic conditions emphasizing the need for the inclusion of such measures into predictive models [75] . This work could also be extended to other disease systems that feature chronic inflammation , and the modeling of variables pathogen and nutrient broth could be replaced by an inflammatory stimulus input function f ( t ) that is more general and applicable to pathogen insult or injury . In conclusion , data presented herein describes the development of a computational model of the sequential influx of immune cells in response to an external trigger and fitting this model to experimental data obtained from a well-established in vivo model of inflammatory response namely peritonitis . Fine tuning this model with inclusion of other systemic parameters related to inflammation will permit the future application to chronic inflammatory diseases with dysfunctional resolution of inflammation . | Using experimental data and mathematical analysis , we develop a model for the inflammatory response that includes macrophage polarization between M1 and M2 phenotypes . Dysfunction of this phenotypic switch can disrupt the timely influx and egress of immune cells during the healing process and lead to chronic wounds or disease . The modulation of macrophage population has been suggested as a strategy to dampen inflammation in diseases that feature chronic inflammation , such as diabetes and atherosclerosis . It is therefore important that we learn more about which components of the system drive the population level switch in phenotype . Our model is able to reproduce the expected timing of sequential influx of neutrophils and macrophages in response to an inflammatory stimulus . Model parameters were estimated with weighted least squares fitting to in vivo experimental data from a mouse model of peritonitis while considering identifiability of parameter sets . We perform sensitivity analysis that identifies primary drivers of the system , and predict the effects of variations in these key parameters on immune cell populations . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"blood",
"cells",
"inflammatory",
"diseases",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"cell",
"processes",
"signs",
"and",
"symptoms",
"neutrophils",
"white",
"blood",
"cells",
"inflammation",
"animal",
"cells",
"immune",
"response",
"diagnostic",
"medicine",
"cell",
"biology",
"monocytes",
"apoptosis",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages"
]
| 2019 | Identifying important parameters in the inflammatory process with a mathematical model of immune cell influx and macrophage polarization |
A recently published transcriptional oscillator associated with the yeast cell cycle provides clues and raises questions about the mechanisms underlying autonomous cyclic processes in cells . Unlike other biological and synthetic oscillatory networks in the literature , this one does not seem to rely on a constitutive signal or positive auto-regulation , but rather to operate through stable transmission of a pulse on a slow positive feedback loop that determines its period . We construct a continuous-time Boolean model of this network , which permits the modeling of noise through small fluctuations in the timing of events , and show that it can sustain stable oscillations . Analysis of simpler network models shows how a few building blocks can be arranged to provide stability against fluctuations . Our findings suggest that the transcriptional oscillator in yeast belongs to a new class of biological oscillators .
Cells have to operate reliably under internal and external noise in order to survive . Their robustness is partially a result of various signal-processing sub-networks called “motifs , ” embedded in the transcriptional network of the cell that controls gene expression [1]–[5] . Such motifs are employed by the cell to produce reliable responses to internal and external signals: a negative auto-regulation motif decreases response time and increases robustness to noise [3] , [6] , [7]; a positive feedback generates bistability and thus can act as a switch [8]–[10]; a coherent feed-forward loop with OR logic acts like a capacitor , sustaining a high output when the input signal is transiently lost [11]; and an incoherent feed-forward loop allows adaptation to a sustained input signal [12] . It is known that combinations of some motifs such as positive and negative feedback loops , can generate stable cyclic behavior [1] , [10] , [13]–[19] . The exact mechanism underlying the oscillations may vary [20]–[22] . Two examples have been particularly well studied . In a negative feedback oscillator ( ) , a sufficiently long time delay in the negative feedback loop makes the system repeatedly overshoot an unstable steady state [10] , [13] . In an activator-inhibitor oscillator ( ) , a positive feedback loop creates bistability and a negative feedback loop causes oscillations due to hysteresis [10] , [13] , [15] , [16] . An important feature in these examples is the spontaneous activation of , which is required to avoid collapse to a quiescent state . In a transcriptional oscillator , this corresponds to a constant input signal ( due , for example , to a constitutive promoter ) or positive auto-regulation sufficiently strong to cause levels of to rise to an active state as long as the inhibitor is not present . To our knowledge , all models of biological oscillatory networks described in the literature , such as cyclin-cdc2 oscillations [23] , [24] , or circadian oscillations in Drosophila [25] , require spontaneous activation to sustain the oscillations [1] , [13] , [20] , [21] . This is also true for synthetic examples such as the repressilator [26] , ( in which all three genes have constitutive but repressible promoters ) , the E . coli predator-prey system [27] , and the synthetic gene-metabolic oscillator [28] . The recently published transcriptional yeast ( Saccharomyces cerevisiae ) cell-cycle oscillator [29] , however , does not seem to share this feature . The gene expression data suggest that this oscillator relies mainly on a sequence of activations on a long , slow positive feedback loop [29]–[31] . There does not appear to be an element in this transcriptional network that is activated spontaneously . Expression profiles also indicate that the period of the oscillator is very close , if not identical , to the time it takes for the wave of activations to cycle around the long positive feedback loop . Here , we show how it is possible to maintain stable oscillations within this architecture . We demonstrate that a slow positive feedback loop coupled to certain stabilizing motifs can sustain oscillations , and that a model of the transcriptional oscillator associated with the yeast cell-cycle works in this fashion . Oscillator stability is conventionally studied in the context of a differential equation model [20] , [21] . On the other hand , the essential organizing logic of regulatory networks can be studied much more easily using Boolean models [29] , [32]–[40] . A drawback of the standard synchronous Boolean approach is that it does not permit the implementation of small perturbations , i . e . , noise , of the type that would result from stochastic fluctuations of the number of molecules of a given species or the rates of production of the various species involved . Indeed , synchronous Boolean models are known to produce many cyclic attractors that represent only marginally stable behavior , which disappear in the presence of noise [41] , [42] . Here we take an intermediate approach that emphasizes the essential Boolean logic of the system within a continuous-time updating scheme that allows the modeling of small perturbations [41] , [43]–[46] . We associate a time delay with each link in the network of regulatory interactions that determines the timing of activation and deactivation events . The stochastic fluctuations thus appear in our model as deviations of the delay times from their nominal values . Such models have been termed autonomous Boolean networks [47] , [48] to distinguish them both from models based on synchronous or random asynchronous timing of updates and from Boolean Delay Equations [49] , [50] that do not account for finite response times . The results presented here apply as well to appropriately constructed ordinary differential equation ( ODE ) models [46] . Regulatory networks based on the cyclin/CDK-centered view of the cell cycle [51] in S . cerevisiae [38] and Schizosaccharomyces pombe [52] have been studied previously using a synchronous Boolean framework . In those models , the intrinsic dynamics is not cyclic and the transition sequence corresponding to the cell cycle must be triggered by an external signal . We emphasize that the network we study is based on the recent experiments [29] , [53] suggesting the existence of a self-sustaining transcriptional oscillator in yeast . The rest of the paper is organized as follows . We first define the autonomous Boolean formalism and discuss the necessity for it . We then demonstrate that it is possible to construct a stable autonomous Boolean oscillator consisting of a long positive feedback loop with two stabilizing motifs added . This toy oscillator has topological features resembling the yeast cell-cycle oscillator . We then describe numerical experiments demonstrating that these features are the source of stability in the autonomous Boolean version of the network of Orlando et al . [29] . We close with a discussion of the implications of these findings . The details of the computer simulations are provided in the Methods section .
In an autonomous Boolean network ( ABN ) , each node takes one of only two values at any given time: or . Updates are executed in continuous time as follows . When a node , , changes its state , it signals all the downstream nodes directly connected to its outputs . Each downstream node , , receives the signal after a time delay , , which is a real ( not necessarily integer ) value . When the signal is received , reevaluates its state according to its assigned Boolean function and adopts the resulting value , . If the new value is different from its present value , a new signal is sent to its own downstream targets . Nodes do not update at externally dictated times , as in the synchronous model or various asynchronous versions . The update dynamics is determined by the timing of events , delays , and the topology of the network . In principle , delays associated with activation ( switch-on events ) , , can be different from the ones associated with deactivation ( switch-off events ) , , because of the different physical processes involved . The former characterizes multiple processes , including transcription , and translation , folding , post-translational modification , and spatial transport , while the latter can be attributed to degradation of mRNAs and transcription factors . The difference between and can cause a change in the duration of a pulse of transcriptional activity as it propagates down a chain of nodes [46] . Consider , for example , a simple cascade with two nodes , where output of regulates . Suppose we turn on manually at and turn it off at , forming a pulse of width , as shown in Figure 1 . The rising edge of this pulse arrives at at and the falling edge arrives at . When , the initial pulse grows as it propagates ( Figure 1 ) and if , it shrinks . Small perturbations due to stochastic fluctuations , or noise , can significantly alter the dynamics of a network and can be used as a mathematical tool for analyzing the stability of cycles . Noise is incorporated by taking the time delay associated with a switching event to be , where the noise term , for each propagating signal is drawn at random from a uniform distribution on with . For present purposes , we take the intrinsic delays and to be equal , allowing the noise to play a dominant role in determining which cycles are stable . The choice of corresponds to the regime in which the asymmetry in propagation times is small compared to , so that pulses grow or shrink according to the relative values of chosen for the leading and trailing edges . In certain cases , the noisy dynamics can generate a pulse of negligibly small width , which we call a spike [41] , [47] , [48] . In the present context , a spike would correspond to arbitrarily fast build-up and degradation of transcripts and therefore is not realistic . We employ a short-pulse rejection mechanism in the simulations , discarding both pulses and dips with widths less than time unit . The Methods section below provides details of our computer simulation of ABNs . As mentioned above , the backbone of the oscillator in the network of interest is a positive feedback loop , also known as a loop of copiers or a simple loop because each node simply assumes the value of its input after some specified time delay . To demonstrate the need for a stabilization mechanism , we consider first the simple case of a loop of two copiers . We can assume without loss of generality that the two links have identical delays , . The network cycles between the 01 and 10 states when one node is initialized with a pulse of sufficiently large width . Setting that width equal to and setting the noise level to zero reproduces the dynamics of the synchronous Boolean case . To test the stability of the cycle , we apply arbitrarily small random perturbations: each time a signal propagates across a link , the delay is taken to be , where is a random number drawn from a distribution that is symmetric around zero . Each perturbation causes the pulse width to grow or shrink as explained above , so that the oscillation eventually collapses to either the or fixed point ( stationary state ) . Thus the cycle is only marginally stable in the autonomous model and its apparent stability under synchronous updating is an artifact of that scheme . We identify two classes of motifs , [1] , [2] , [4] , which we call rectifiers and growers , that can correct small perturbations to the timing of the updates and stabilize cycles on an autonomous loop of copiers . A rectifier imposes an upper limit on the width of the pulse traveling on the positive feedback loop . The simplest example of a rectifier is auto-repression ( Figure 2A ) , which cuts long pulses down to a width equal to the delay on the auto-repressive link , , and lets short pulses pass through unaffected [46] . Small perturbations that cause the pulse width to exceed will be filtered by this motif as seen in Figure 2C . An incoherent feed-forward loop of type 1 ( I1-FFL in the notation of [1] ) , and a negative feedback containing more than one node can also function as rectifiers . Grower motifs increase the duration of a pulse by a constant amount , but do not adjust them to a particular value . One example is the coherent feed-forward loop with OR logic ( C1-FFL-OR [1] , [2] ) shown in Figure 2B . This motif grows pulses by transmitting the input pulse of width from to through two paths with time delays that differ by . The slower path sustains the output , producing a pulse of width , assuming . ( If the condition is not met , two pulses will be generated . ) A diamond motif [1] with OR logic , in which both paths connecting the input to the output contain an intermediate node , functions in the same manner . We also note that both C1-FFL and the diamond motifs function as shrinkers when their output is an AND gate , shrinking the input pulse by or destroying it completely . A rectifier cannot prevent the collapse to the all-OFF state and a grower alone inserted in a loop will keep growing the pulse until the all-ON attractor is reached . The two motifs working in tandem ( Figure 2B ) , however , can act as a stabilizing module for cyclic attractors , as seen in Figure 2D: both pulse-growing and pulse-shrinking perturbations are filtered because the grower-rectifier combination resets the pulse width to after each cycle . Such a network can sustain stable oscillations that have been started with an external signal . The two motifs will be incompatible if because the grower will generate two pulses from each rectified pulse . We note that there is no simple motif that acts as a low-pass rectifier , allowing long pulses to pass unaffected while boosting short pulse widths up to a specified value . Thus the shrinker motif is of limited use for stabilizing oscillations . Furthermore , a grower-shrinker combination cannot be a stabilizer as it simply acts either as an overall grower or an overall shrinker . If one allows and to be different , a pulse may grow or shrink as it travels around a simple loop . When for the links in the loop , we have a source of “intrinsic growth” that may render a grower motif unnecessary , or just assist the grower in restoring pulse widths more rapidly . In fact , it has been shown using an ODE model with time delays that when switch-on events propagate faster than switch-off events , an auto-repressive link can by itself create a stable cycle on a loop of copiers [46] . Similarly , when along the loop , a pulse will shrink as it propagates . Stabilization in the presence of intrinsic shrinkage requires a grower regardless of the noise level . We do not consider intrinsic growth or shrinkage here , focusing instead on cases where stochastic effects ( noise ) dominate over the intrinsic effects . Also , we consider only the stabilization of single-pulse cycles , in which each node along the loop ( through ) turns on and off exactly once per cycle time , which we define as the time required for a single signal to propagate around the loop once . For a simple loop , the cycle time is equal to the sum of the delays , but for more complex circuits , it can depend on the pulse width . A crucial feature of the oscillator architecture under consideration here ( Figure 2B ) is that it does not rely on any constitutive input or positive auto-regulation . Consider , for example , the model of circadian oscillations in Drosophila [13] , [21] , [25] , which contains one protein , PER , whose biphosphorylated form represses its own transcription . It is assumed that Per mRNA is transcribed at the maximum rate in the absence of biphosphorylated nuclear PER , thereby building up spontaneously . Such an oscillator can be represented as a simple negative feedback loop , PerPER Per with a long time delay on the repressive link . A Boolean model of the oscillator can be constructed by assigning a NOT function to Per indicating that it builds up spontaneously , but only in the absence of PER; and a COPY function to PER as it is produced only in the presence of Per . This model has a cycle containing all four states of the circuit , . From the Boolean perspective , the underlying principle for these oscillations is the impossibility of satisfying all the Boolean functions simultaneously , as the combination of an inverter and a copier creates frustration [42] . For this reason , we refer to the Boolean versions of such oscillators , which have no fixed points , as frustration oscillators . The oscillator we propose in Figure 2B , however , has the all-OFF fixed point attractor; there is no frustration in its logic . It therefore belongs to a different class that involves a stable transmission of a pulse on a loop of copiers , i . e . , a positive feedback loop . We refer to these as transmission oscillators . The recently published cell-cycle oscillator network in yeast consists of nineteen interactions between eight transcription factors and one cyclin , CLN3 , which was used as a proxy for currently unidentified transcription factors that complete the circuit ( Figure 3A ) [29] , [30] , [53] . The regulatory logic functions of the multi-input nodes are not known . This oscillator was studied using a synchronous Boolean model with eight different “biologically interpretable” logic configurations for the network given in Figure 3A and Table 1 [29] . Each logic configuration was found to support at most two out of the three possible cycles in addition to the all-OFF fixed point . All three cycles match the sequential order of the expression of the transcription factors . We emphasize here , however , that these features may only be artifacts of the synchronous update scheme and their stability requires further investigation . This version of the yeast cell-cycle oscillator is a complex network that does not seem to be a frustration oscillator . Expression profiles of transcription factors suggest that sequential activations are triggered by immediate upstream regulators in the network [29] . Therefore , the oscillations are unlikely to be driven by a frustration oscillator that is either a part of or coupled to the circuit . Several intertwined feed-forward and negative feedback motifs in the network suggest that a grower-rectifier combination may be at play in stabilizing the oscillations . Specifically , we hypothesize that this network is a simple loop consisting of CLN3 , SBF , SFF , and ACE2 or SWI5 ( since this is the loop of copiers with the least number of links ) , and all other nodes conspire to provide stabilizing motifs . We use computer simulations to test this hypothesis . Briefly , we assign random delays to each link and start the network by manually turning CLN3 on then off . The distribution we choose for the delays roughly captures the variation in delays seen in the experiments [29] . A broader distribution would not qualitatively change the results . The details of the simulations are described in Methods .
We have shown using an autonomous Boolean model , that a long positive feedback loop can be turned into a stable oscillator with the addition of two stabilizing motifs that can correct fluctuations in the pulse width ( the duration of activity of each node in the network ) : a rectifier involving a repressor that limits the width of the traveling pulse , and a grower that lengthens the duration of a pulse so that it cannot shrink and disappear . In combination , a grower and a rectifier ensure that the pulse width returns to the same value after each cycle . The recently published yeast cell-cycle oscillator [29] has a structure built around a long positive feedback loop , on which waves of activation events propagate . Numerical simulations of eight different logic configurations and multiple realizations of randomly assigned time delays revealed the presence of grower and rectifier functions in this network . To our knowledge , there is no other biological oscillator model described in the literature that relies on a long , slow positive feedback loop . We note that a proposed cell cycle network for Caulobacter crescentus [54] has a structure reminiscent of that of yeast , but no dynamical model of it has yet been reported . Previous synchronous Boolean models of Drosophila segmentation network [39] , or cyclin/CDK-based cell-cycle networks of S . pombe [52] and S . cerevisiae [38] predicted essential features of the robust dynamics of these networks [55] . We have demonstrated that the autonomous Boolean framework can be used to further study such problems , since it addresses important elements of the regulatory dynamics associated with the timing of updates and the effects of stochastic fluctuations . We note that ABNs have also been used recently for analyzing chaos and the stability of periodic orbits in digital electronic oscillators [47] , [48] . Our results also point to a drawback of fully asynchronous Boolean models: a stable cycle in a continuous-time system such as that of Figure 2D would not be observed in an asynchronous model . In asynchronous models , cycles generated by loops containing an even number of inverters cannot be sustained [42] because there always exists a sequence of updates that leads to the fixed point state . We have shown , however , that when appropriate motifs are present , the autonomous rules for determining the order in which nodes are updated never permit evolution to the fixed point even in the presence of a substantial level of noise . In analyzing the dynamics of gene networks containing feedback loops , it is therefore important to take into account timing information associated with signal propagation . For gene networks containing feedback loops , results from discrete-time Boolean models ( both synchronous and asynchronous ) should be interpreted with care . The stability of the oscillations we have observed is not an artifact of the autonomous Boolean model . The presented results are qualitatively compatible with ODE analogs involving explicit time delays [56] . An ODE model of a similar system with explicit time delays has already been shown to exhibit stable oscillations very similar to our Boolean idealization when synthesis rates , Hill coefficients , and time delays are large enough [46] . Our own preliminary studies indicate that it is also possible to construct an ODE model of a transmission oscillator without explicit time delays by selecting appropriate parameters for the stabilizing motifs .
To simulate the dynamics of an autonomous Boolean network , we use an event-driven code . A time-ordered event queue is established , in which each event represents the switching of an input at a specified node . Each time an event is processed that results in the switching of a node , events are added to the queue according to the time delay associated with each output link from that node . After each update of a node , we check to see whether it creates a short pulse that should be rejected . If so , the queue is purged of all events derived from the leading and trailing edges of the pulse . To avoid causality problems coming from propagation of a switching event that is later rejected , we choose the maximum noise amplitude , , to be less than half of the short-pulse rejection time ( time unit ) . To reveal the structure of the yeast cell-cycle oscillator , we study numerical simulations of autonomous Boolean versions of the network with the logic choices in Reference [29] and different randomly selected sets of time delays . For each logic configuration , we generate an ensemble of 10000 networks with quenched random delays on each link . Delays were chosen from a uniform distribution between and 2 time units . The system was initialized by turning CLN3 on at and turning it off at , while other nodes were OFF . All nodes were assumed to be OFF for . To simulate noise , a random value selected from a uniform distribution on the interval , was added to the delay associated with each update . We are interested in the stability of a particular cycle , so we have chosen an initial condition that is very likely to lie in the basin of attraction of that cycle ( if the cycle exists ) . A different initial condition , turning CLN3 on at and letting a repressor turn it off , yields roughly the same statistics reported in Table 1 . We do not test the network's robustness to general changes in the initial condition [38] , [39] , [57] , [58] . We run simulations up to 125000 updates with noise turned on between the 800th and 90000th updates in order to eliminate marginally stable oscillations . For single-pulse oscillations , this typically translates into a runtime of time units under noise . Periodic single-pulse oscillations that survive this long with noise present are highly likely to be stable attractors . An oscillation is considered to be PSP if pulse widths on two consecutive cycles differ by less than time unit on each node . We do not check whether all nodes turn on and off once per cycle time , i . e . , whether the cycle is a single-pulse or a dual-pulse with identical pulse widths . However , we never observed the latter in the inspected realizations and believe that it is very unlikely to occur in this circuit . The numbers of oscillating realizations differ in the different logic configurations for two reasons . First , an FFL or diamond motif operates as a grower with OR logic and as a shrinker with AND logic . When two logic configurations differ only by the selection of or , the one with OR logic always has a larger number of oscillating networks . In configurations 1 and 5 , both SFF and CLN3 are AND gates , so the motifs they belong to will act as shrinkers . The existence of two shrinkers in the network should make oscillations very unlikely , and indeed we find that all realizations in both configurations collapse on the all-OFF attractor . On the other hand , configurations that contain a larger number of AND logic for and generate mostly periodic single-pulse oscillations and fewer complex ones . The second reason for the difference in the number of oscillating networks is the logic , , of the repressors . The case gives a smaller total number of oscillating realizations than . | Technologies such as gene arrays enable acquisition of large amounts of data on gene expression variations , which reveal the structures of gene regulatory networks that govern the metabolic and developmental machinery in the cell . We study a model of an oscillatory gene regulatory network that has been recently suggested to play an integral role in maintaining the cell cycle in yeast . The oscillator differs from other known biological and synthetic oscillatory networks in that it seems to rely on a long positive feedback loop . We show that the presence of certain stabilizing sub-networks can account for the robustness and the unusual architecture of this oscillator . Our modeling approach elucidates both the logical structure of the system and the importance of the timing of update events . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"computational",
"biology/systems",
"biology",
"physics/interdisciplinary",
"physics",
"computational",
"biology/transcriptional",
"regulation"
]
| 2010 | Reliability of Transcriptional Cycles and the Yeast Cell-Cycle Oscillator |
T-cell proliferation and generation of protective memory during chronic infections depend on Interleukin-7 ( IL-7 ) availability and receptivity . Regulation of IL-7 receptor ( IL-7R ) expression and signalling are key for IL-7-modulated T-cell functions . Aberrant expression of soluble ( s ) and membrane-associated ( m ) IL-7R molecules is associated with development of autoimmunity and immune failure in acquired immune deficiency syndrome ( AIDS ) patients . Here we investigated the role of IL-7/IL-7R on T-cell immunity in human tuberculosis . We performed two independent case-control studies comparing tuberculosis patients and healthy contacts . This was combined with follow-up examinations for a subgroup of tuberculosis patients under therapy and recovery . Blood plasma and T cells were characterised for IL-7/sIL-7R and mIL-7R expression , respectively . IL-7-dependent T-cell functions were determined by analysing STAT5 phosphorylation , antigen-specific cytokine release and by analysing markers of T-cell exhaustion and inflammation . Tuberculosis patients had lower soluble IL-7R ( p < 0 . 001 ) and higher IL-7 ( p < 0 . 001 ) plasma concentrations as compared to healthy contacts . Both markers were largely independent and aberrant expression normalised during therapy and recovery . Furthermore , tuberculosis patients had lower levels of mIL-7R in T cells caused by post-transcriptional mechanisms . Functional in vitro tests indicated diminished IL-7-induced STAT5 phosphorylation and impaired IL-7-promoted cytokine release of Mycobacterium tuberculosis-specific CD4+ T cells from tuberculosis patients . Finally , we determined T-cell exhaustion markers PD-1 and SOCS3 and detected increased SOCS3 expression during therapy . Only moderate correlation of PD-1 and SOCS3 with IL-7 expression was observed . We conclude that diminished soluble IL-7R and increased IL-7 plasma concentrations , as well as decreased membrane-associated IL-7R expression in T cells , reflect impaired T-cell sensitivity to IL-7 in tuberculosis patients . These findings show similarities to pathognomonic features of impaired T-cell functions and immune failure described in AIDS patients .
T cells are crucial for protection against Mycobacterium ( M . ) tuberculosis infection but biomarkers that characterise T-cell failure and progression towards tuberculosis disease are not available [1] . CD4+ T cells are key to anti-mycobacterial immune protection [2] and CD4+ T-cell deficiency , e . g . of AIDS patients , results in increased susceptibility against tuberculosis [3–5] . There is growing evidence that impaired CD4+ T-cell functions play a role in tuberculosis [6] . Recent studies identified T-cell exhaustion as a feature of tuberculosis [7 , 8] . T-cell exhaustion impairs immunity against chronic viral infections and harms memory T-cell potential [9] . IL-7 is central for generation of memory T cells and was shown to revert T-cell exhaustion in chronic viral infections [10] . Notably , IL-7 induced T-cell memory was hampered in the presence of persistent antigen and inflammation as seen for chronic viral infections [11] . In AIDS patients , failure of immune reconstitution is accompanied by a dysfunctional T-cell response that showed features of senescence and exhaustion [12–14] . Recently , persistent inflammation characterised e . g . by increased IL-6 serum concentrations from AIDS patients were found to correlate with T-cell exhaustion/senescence and impaired T-cell response to IL-7 [14 , 15] . High IL-7 plasma levels as well as decreased membrane-associated ( m ) IL-7R expression of T cells were found in AIDS patients with immune failure [16 , 17] . Concomitantly impaired T-cell response to IL-7 was detected in immune failure patients [13–15 , 18–20] . The regulation of IL-7R expression is central for control of IL-7-mediated effects on T cells [21] . On IL-7 binding , the mIL-7R assembles as a heterodimer ( comprising the IL-7Rα ( CD127 ) and the common γ-chain ( CD132 ) ) and induces signalling cascades mainly via the Jak/STAT pathway . Jak1 and Jak3 are involved in IL-7R signalling , and STAT5 gets phosphorylated and initiates multiple transcription events [22] . As part of IL-7 signalling , the mIL-7R is rapidly internalised , becomes partly degraded or recycles to the cell surface [23] . Regulation of IL-7R expression is also controlled on the transcriptional level and IL-7 and other cytokines were shown to suppress IL-7R mRNA expression [24] . Alternative splicing of the IL7RA gene generates a soluble IL-7R ( sIL-7R ) variant [25] . The sIL-7R variant binds IL-7 although with lower affinity as compared to the mIL-7R heterodimer and is present in blood plasma at high molar excess relative to IL-7 [26] . The exact role of the sIL-7R for IL-7 metabolism remains elusive . Competitive inhibition of IL-7 uptake as well as IL-7 reservoir functions have been described [26–28] . Differential sIL-7R plasma concentrations are found in immune pathologies , e . g . autoimmune diseases [26 , 29 , 30] and AIDS [28 , 31] . In addition , a functional IL7RA polymorphism ( rs6897932 ) that interferes with IL-7R alternative splicing and thereby leads to reduced sIL-7R levels in plasma was found to be associated with autoimmune diseases [32 , 33] and to affect immune reconstitution in AIDS patients [34–36] . Initial results indicating a role of IL-7 during T-cell immunity against tuberculosis were derived from animal models . Increased IL-7 and soluble IL-7R expression in pulmonary tissue of primates with tuberculosis was found , indicating a possible role of IL-7 metabolism in tuberculosis pathogenesis [37 , 38] . Furthermore IL-7 was shown to promote survival and to improve BCG vaccination efficacy in M . tuberculosis-infected mice [39 , 40] . However , a comprehensive understanding of the possible role of IL-7 or IL-7R functions in human tuberculosis has not yet been developed . This present study aimed to elucidate a possible role of IL-7 modulated T-cell responses in human tuberculosis . We determined sIL-7R and IL-7 plasma concentrations and mIL-7R expression of T cells from tuberculosis patients—before , during , and after chemotherapy—and compared these to healthy contacts . Since results resembled pattern seen in AIDS patients with impaired T-cell response to IL-7 , we then performed functional T-cell assays in a second set of tuberculosis patients and healthy contacts to determine IL-7-mediated signalling and promoted cytokine release on M . tuberculosis-specific T-cell activation . Finally , mRNA expression of exhaustion markers was compared in CD4+ T cells between the cohorts to evaluate a possible causative role of T-cell exhaustion for impaired IL-7 response in tuberculosis .
Aberrant sIL-7R plasma levels indicate pathologic T-cell immunity in autoimmune , inflammatory , and chronic viral diseases . Hence , we determined sIL-7R plasma concentrations in individuals infected with M . tuberculosis . Patients with active tuberculosis ( n = 57 ) and healthy contacts ( n = 151 ) were included . Tuberculosis patients had significantly lower sIL-7R concentrations as compared to healthy contacts ( p < 0 . 001 ) ( Fig 1a ) . Since study groups differed in gender distributions ( tuberculosis: 30% females; contacts: 56% females; Table 1 ) , we compared sIL-7R between male and female subgroups . Female patients with tuberculosis showed moderately lower sIL-7R concentrations as compared to male patients , whereas no differences were detected for healthy contacts ( S1 Fig ) . Therefore , differences in plasma sIL-7R were not due to gender differences . Next we determined the influence of anti-tuberculosis therapy and recovery on plasma sIL-7R in tuberculosis patients ( i . e . 2 months and 6 months after therapy onset ) . Analyses revealed significantly increased sIL-7R plasma levels after 2 months ( p = 0 . 03 ) and after recovery ( p = 0 . 009 ) ( Fig 1b ) . sIL-7R plasma concentrations of recovered tuberculosis patients were comparable to healthy contacts ( Fig 1b ) . To determine if changes in sIL-7R under therapy were dependent on sIL-7R concentrations prior to treatment , we compared initial sIL-7R concentrations with changes of sIL-7R expression between 0 and 6 months . Absolute differences and ratios were calculated . Absolute differences ( month 6 –month 0 ) showed only moderate negative correlation with initial sIL-7R levels ( rho = -0 . 26; p = 0 . 13 ) ( S2a Fig ) , but changes of ratios ( month 6 / month 0 ) were strongly associated with sIL-7R levels prior to treatment ( rho = -0 . 61 , p < 0 . 001 ) ( S2a Fig ) . Therefore , especially tuberculosis patients with low sIL-7R concentrations prior to treatment showed increased sIL-7R levels after recovery and a relative gain of sIL-7R plasma concentration was detected . A functional single nucleotide polymorphism ( SNP , rs6897932C>T ) in exon 6 of the IL7RA gene interferes with splicing and impairs sIL-7R expression [32] . Therefore , we determined the rs6897932 minor T allele ( rs6897932T ) frequency in tuberculosis patients and healthy contacts . Tuberculosis patients had a marginally higher MAF proportion ( 7 . 3% ) as compared to healthy contacts ( 5 . 6% ) . No homozygous rs6897932T/T carriers were identified in the study groups . As expected , lower levels of plasma sIL-7R were detected for rs6897932C/T healthy contacts as compared to rs6897932C/C wild type healthy contacts ( p = 0 . 02 ) , and the same tendency was seen for the tuberculosis patients ( p = 0 . 06 ) ( Fig 1c ) . However , stratification for SNP genotypes confirmed lower plasma sIL-7R among tuberculosis patients when compared to healthy contacts ( p < 0 . 001 ) . We concluded that increased frequencies of IL-7R rs6897932T alleles in tuberculosis patients contributed to differential sIL-7R levels but did not account for lower sIL-7R plasma concentrations of tuberculosis patients . We hypothesised that differential sIL-7R plasma levels would affect IL-7 consumption . Consequently we next determined IL-7 plasma concentrations in tuberculosis patients and healthy contacts . Tuberculosis patients showed significantly increased IL-7 concentrations prior to therapy as compared to healthy contacts ( p < 0 . 001 ) ( Fig 2a ) . IL-7 concentrations decreased under therapy and recovery ( 0 vs . 6 months , p < 0 . 001 ) and reached levels comparable to healthy contacts ( Fig 2b ) . Higher initial IL-7 levels were associated with stronger decrease rates until month 6 ( rho = -0 . 58 , p < 0 . 001; S2b Fig ) . Notably , and in contrast to sIL-7R results , also absolute differences between month 0 and 6 correlated strongly with IL-7 levels prior to therapy ( rho = -0 . 79 , p < 0 . 001; S2b Fig ) . This indicated different mechanisms involved in IL-7 and sIL-7R regulation during tuberculosis . In accordance , no dependency was detected between IL-7 and sIL-7R plasma concentrations for tuberculosis patients or healthy contacts ( Fig 2c ) . These results suggested that IL-7 and sIL-7R could be useful as biomarkers for diagnosis of tuberculosis patients . Comparison of tuberculosis patients and healthy contacts revealed moderate discrimination capacity for both sIL-7R ( AUC = 0 . 67 ) and IL-7 ( AUC = 0 . 73 ) using Receiver Operating Characteristic ( ROC ) analysis ( Fig 2d ) . Independency of IL-7 and sIL-7R plasma levels ( Fig 2c ) prompted us to calculate the combined efficacy of both markers using Random Forest analysis ( for details see Methods section ) . Correct prediction of tuberculosis patients and healthy contacts was achieved for 73% of all donors , and IL-7 was about two times more influential on prediction than sIL-7R . These results indicated that IL-7 and sIL-7R plasma concentrations were largely independent and may contribute to tuberculosis diagnosis . Increased IL-7 plasma concentrations are likely caused by decreased T-cell consumption of IL-7 . Low T-cell numbers or impaired T-cell receptivity of IL-7 may account for this . Hence we compared mIL-7R protein expression for subgroups of tuberculosis patients and healthy contacts by flow cytometry . We detected lower mean mIL-7R expression for CD8+ T cells ( p = 0 . 02 ) and a tendency for CD4+ T cells ( p = 0 . 05 ) ( Fig 3a ) . Analysis of mIL-7R on T-cell subpopulations revealed increased proportions of mIL-7Rlow CD4+ ( p = 0 . 006 ) and CD8+ T cells ( p = 0 . 02 ) from tuberculosis patients as compared to healthy contacts ( Fig 3b ) . To confirm these observations , we performed mIL-7R analysis in a second independent cohort study including additionally recruited tuberculosis patients ( n = 22 ) and healthy contacts ( n = 24 ) . Due to restriction in the number of flow cytometry parameters , CD4+ and CD4- T cells were analysed for mIL-7R protein expression . Tuberculosis patients showed significantly decreased mIL-7R expression for both CD4+ ( p = 0 . 01 ) and CD4- ( p = 0 . 006 ) T cells ( S4 Fig ) . This confirmed initial results and led us to the conclusion that impaired mIL-7R expression of T cells resulted in increased proportions of mIL-7Rlow CD4+ and CD8+ T cells in tuberculosis patients . Differential mIL-7R expression may be affected by plasma IL-7 and sIL-7R levels . We determined correlation between these parameters to identify possible interactions . A tendency of positive correlation between mIL-7R expression and sIL-7R plasma ( rho = 0 . 42 , p = 0 . 06 ) was found only in the group of contacts , whereas mIL-7R and IL-7 showed a marginal negative correlation ( rho = -0 . 38 , p = 0 . 10 ) ( S1 Table ) in this study group . No correlation between any parameters was found for tuberculosis patients ( S1 Table ) . High IL-7 plasma levels and low mIL-7R expression of T cells have previously been described for HIV/AIDS patients [16 , 17 , 41 , 42] . In AIDS patients these differences are accompanied with mIL-7R regulatory dysfunctions [43] . Therefore we questioned whether aberrant expression of IL-7R variants in tuberculosis patients is caused by differential regulation on the transcriptional or post-transcriptional level . Hence , we analysed IL-7R mRNA transcripts of purified CD4+ T cells from tuberculosis patients and healthy contacts . Three IL-7R transcripts coding for the mIL-7R ( all 8 exons included; H20 ) and a sIL-7R ( H6 and H5-6; for details see Methods section [25] ) were measured . None of the IL-7R variants were differentially expressed on the mRNA level of CD4+ T cells between tuberculosis patients and healthy contacts ( Fig 4a ) . Also relative expression of sIL-7R vs . mIL-7R transcripts was similar between study groups ( Fig 4b ) . These results indicated that differential IL-7R mRNA expression is not the cause for aberrant sIL-7R and mIL-7R expression in tuberculosis patients and render causative post-transcriptional mechanisms likely . Impaired IL-7 signalling has been associated with diminished IL-7Rlow expression of T cells from AIDS patients , but different mechanisms about the role of STAT5 were described [18 , 44 , 45] . To evaluate the effect of IL-7 signalling , we recruited a second cohort of tuberculosis patients ( n = 22 ) or healthy contacts ( n = 24 ) ( Table 1 ) . A lower surface level of mIL-7R on T cells from tuberculosis patients was confirmed in this cohort ( S4 Fig ) . Next , we measured IL-7-induced STAT5 phosphorylation and detected decreased phosphorylated STAT5 in CD4+ T cells from tuberculosis patients as compared to healthy contacts ( p = 0 . 04 ) ( Fig 5a ) . Since IL-7 was shown to enhance specific T-cell cytokine release [46] , we determined intracellular cytokines after M . tuberculosis antigen ( PPD ) in vitro stimulation in the presence or absence of IL-7 . No differences were detected for PPD-specific T cells co-expressing IFNγ and CD40L when comparing tuberculosis patients and healthy contacts ( Fig 5b ) . However , co-stimulation with IL-7 induced increased proportions IFNγ-producing T cells solely in the study group of healthy contacts ( p = 0 . 003 ) , but not in tuberculosis patients ( p = 0 . 94 ) ( Fig 5b ) . Next , IL-7-specific effects were quantified by calculating the difference of PPD induced T cells with or without IL-7 for each individual ( Fig 5c ) . We found a significantly stronger effect of IL-7 on cytokine release in healthy contacts as compared to tuberculosis patients ( p = 0 . 02 ) . These results suggested impaired T-cell responses to IL-7 in patients with tuberculosis . Chronic inflammation and increased IL-6 serum concentrations were found in AIDS patients with impaired T-cell immunity to IL-7 [14 , 15] . One study found a direct inhibitory effect of IL-6 on IL-7-mediated T-cell functions [15] . Since increased IL-6 plasma levels were described in tuberculosis previously [47] , we measured plasma IL-6 levels and detected increased IL-6 concentrations in tuberculosis patients as compared to healthy contacts ( p < 0 . 001 ) ( Fig 5d ) . The distribution of IL-6 plasma concentrations indicated two subgroups of tuberculosis patients . Hence we set an arbitrary threshold ( 15 pg/ml ) and compared IL-6high and IL-6low tuberculosis patients for IL-7-promoted T-cell responses . No significant differences in IL-7-induced STAT5 phosphorylation or IL-7 co-stimulated IFNγ/CD40L expression was found between the two IL-6high and IL-6low subgroups of tuberculosis patients ( Fig 5e and 5f ) . Therefore differential IL-6 serum levels were not associated with impaired IL-7-promoted T-cell responses in tuberculosis patients . Programmed cell death ( PD ) -1 , a marker of T-cell exhaustion and senescence was recently found to be expressed on T cells with impaired response to IL-7 [14] . We determined PD-1 mRNA expression of purified CD4+ T cells and found similar PD-1 expression among healthy contacts and tuberculosis patients prior to therapy ( Fig 6a ) . Under therapy , a decrease of PD-1 expression was found for tuberculosis patients ( p = 0 . 007 ) followed by an increase until recovery ( p < 0 . 001 ) . PD-1 levels in recovered tuberculosis patients were even higher as compared to healthy contacts ( p = 0 . 04 ) . We found a moderate but significant positive correlation of PD-1 ( rho = 0 . 22 , p = 0 . 005 ) with IL-7 ( Fig 6b ) . Previously , we identified SOCS3 as a marker of CD4+ T cells in tuberculosis [48] , and others described SOCS3 as a central regulator of T-cell exhaustion and target of IL-7 in chronic viral infections [10] . Therefore we determined SOCS3 mRNA expression of CD4+ T cells . Marginal increased SOCS3 expression was detected in tuberculosis patients prior to therapy ( p < 0 . 16 ) , and significantly increased SOCS3 levels were detected at two months under therapy ( p < 0 . 001 ) and after six months ( p = 0 . 04 ) as compared to healthy contacts ( Fig 6c ) . As for PD-1 , a moderate positive correlation between SOCS3 expression and IL-7 concentrations was found ( rho = 0 . 22 , p = 0 . 005 ) ( Fig 6d ) . We concluded that expression of T-cell exhaustion marker SOCS3 was increased in tuberculosis patients during therapy but was only moderately associated with aberrant IL-7 plasma concentrations . These observations indicated similarities and differences of aberrant IL-7 pathway features in tuberculosis patients as compared to AIDS patients .
In the presented study , we identified alterations in the IL-7 pathway and impaired T-cell response to IL-7 co-stimulation in tuberculosis patients . First , we detected higher IL-7 plasma concentrations in tuberculosis patients that decreased during therapy and recovery . Lymphopenia may cause high IL-7 plasma levels [49 , 50] and few reports indicated a role of lymphopenia in tuberculosis [51–53] , but this has not been verified by others [54] . We did not determine lymphocyte counts in the present study and cannot prove or refute lymphopenia as a possible cause for high IL-7 levels . However , there is evidence that IL-7 serum concentrations are affected only at very low CD4+ T-cell numbers in AIDS patients [41 , 55] , and these levels are far below lymphopenia described in tuberculosis [49 , 50] . Another possible explanation for higher IL-7 plasma concentrations is impaired receptivity/consumption of IL-7 by T cells [21] . Our investigations provide evidence for reduced mIL-7R expression and impaired IL-7 co-stimulatory effects on T cells from tuberculosis patients . Strong evidence for impaired IL-7 regulation and T-cell function was found for chronic viral infections , especially AIDS [56] . In AIDS patients increased IL-7 plasma levels and decreased mIL-7R expression of T cells were described [17 , 42 , 43 , 57–59] . Furthermore , impaired T-cell response to IL-7 in AIDS patients was shown to affect immune reconstitution during anti-retroviral therapy [13 , 60 , 61] . In order to determine possible dependencies between mIL-7R expression on T cells and IL-7/sIL-7R plasma concentrations , we performed correlation analyses . For contacts there was a tendency of positive correlation between mIL-7R and sIL-7 levels , whereas IL-7 plasma levels showed a marginal negative correlation with mIL-7R expression . Given the described regulatory influence of IL-7/sIL-7R on mIL-7R expression [21] , we speculate that IL-7 and sIL-7R plasma level alterations caused by tuberculosis disrupted this dependency that indicates the homeostatic balance in healthy individuals . The low number of samples included for mIL-7R analyses restricted the validity of these results . In addition , analyses of mIL-7R during disease course and after recovery are needed to confirm this thesis . Several mechanisms and T-cell phenotypes were described to play a role in impaired IL-7 functions of AIDS patients . Chronic inflammation and increased serum concentrations of IL-6 were found in HIV/AIDS [14 , 15] , and functional in vitro assays indicated inhibitory effects of the pro-inflammatory cytokines IL-6 and IL-1β on IL-7-mediated signal transduction [15] . Higher IL-6 plasma concentrations were described for tuberculosis [47 , 62] , and we confirmed higher IL-6 plasma concentrations in a subgroup of tuberculosis patients in the present study . However , we did not detect IL-7 response differences between IL-6high and IL-6low subgroups among tuberculosis patients ( Fig 5e and 5f ) . Hence there was no indication for an association between IL-6 plasma concentrations and impaired IL-7 T-cell response of tuberculosis patients . T-cell exhaustion was found in AIDS patients [12 , 13 , 60 , 63–65] and was associated with decreased IL-7R expression [60 , 64] and impaired IL-7 response [13] . Initial studies indicated a role of T-cell exhaustion in tuberculosis animal models [7 , 8] . Our results on SOCS3 and PD-1 expression did not support a major role of T-cell exhaustion in human tuberculosis and this is in accordance with a previous study [66] . These differences might at least partly be due to the fact that exhaustion is poorly defined for CD4+ T cells in contrast to CD8+ T cells [67] . Therefore , other marker molecules may be indicative for exhaustion in CD4+ T cells . We were not able to study the phenotype of CD8+ T cells in detail in the present study but decreased mIL-7R expression ( Fig 3a ) may indicate exhaustion of CD8+ T cells in tuberculosis patients . Impaired mIL-7R signalling was described for T cells from AIDS patients [13 , 14 , 18 , 68] . We detected lower STAT5 phosphorylation and showed also impaired IL-7 promoted cytokine release in T cells from tuberculosis patients . The capacity of IL-7 to promote IFNγ-expressing T cells for detection of M . tuberculosis infection has been shown before [46] . Here we provide first evidence that IL-7 mediated increased sensitivity of T cells to stimulation ( e . g . by decreasing the T-cell receptor activation threshold [69] ) was impaired in tuberculosis patients . One may therefore speculate that impaired IL-7 response not only hampered generation of effective memory but also effector T-cell response against acute tuberculosis . This raised the question if impaired T-cell response to IL-7 can be interpreted as a feature of T-cell anergy . Anergy is defined as unresponsiveness of T cells to their cognate antigen and anergy against PPD—measured by tuberculin skin test—has been described for tuberculosis patients before [70] . We did not detect differences in the PPD response of CD4+ T cells between tuberculosis patients and healthy contacts in the present study ( Fig 5b ) . However , we would speculate that impaired T-cell responses to IL-7 contributed to the phenomenon of diminished tuberculin reactivity in tuberculosis patients as this in vivo test would be better reflected by IL-7-supplemented PPD stimulation in our in vitro assay . Since IL-7 effects on T-cell function include a decreased T-cell receptor activation threshold [69] , impaired mIL-7R signaling may contribute to diminished T-cell receptor signaling characteristic for T-cell anergy [71] . Therefore impaired mIL-7R signaling may contribute to tuberculin skin test anergy described for tuberculosis patients but additional studies are needed to further clarify the exact role of IL-7 . We also detected lower sIL-7R plasma concentrations in tuberculosis patients and normalisation during therapy and recovery . sIL-7R levels were previously shown to affect IL-7-availability for T cells , but the role of aberrant sIL-7R levels in immune pathologies is a matter of controversy [26–28] . Crawley et al . detected increased sIL-7R concentrations in plasma samples from AIDS patients and described sIL-7R-Fc chimera-mediated inhibition of IL-7 bioactivity [28] . They hypothesised that increased sIL-7R concentrations limited availability of IL-7 for T cells [28] . In contrast , Rose et al . found decreased sIL-7R plasma concentrations in HIV/AIDS patients as compared to controls [31] . sIL-7R plasma concentrations of this study were similar to the present study and 5 to 10 times lower for both study groups as compared to the study published by Crawley et al . [28] . Recently , Lundstrom et al . proposed an alternative model of IL-7 storage provided by the sIL-7R [26] . They demonstrated that sIL-7R even potentiates the bioactivity of IL-7 by forming a reservoir of accessible IL-7 [26] . In accordance , high sIL-7R as well as IL-7 plasma concentrations were associated with multiple sclerosis , and sIL-7R had potentiating effects on exacerbation of experimental autoimmune encephalomyelitis [26] . From this , they concluded that increased plasma concentrations of sIL-7R supported generation of autoimmunity by promoting IL-7-dependent T cells [26] . Since IL-7 serum levels are predominantly regulated by T-cell consumption [21] , both restriction and reservoir hypotheses suggest dependency of IL-7 on sIL-7R levels . In the present study , we did not detect a correlation between IL-7 and sIL-7R plasma levels in tuberculosis patients or healthy contacts , although both factors were affected during tuberculosis pathogenesis . It is therefore tempting to speculate that sIL-7R has either no regulatory activities on IL-7 , or that additional factors influence sIL-7R and/or IL-7 serum levels . In accordance , the proposed regulatory function of sIL-7R on IL-7 has been questioned by others [72] . We evaluated the utility of IL-7 and sIL-7R plasma concentrations as biomarkers for diagnosis of active tuberculosis using ROC curve and Random Forest-based statistics . Both markers showed moderate classification capacity and the combined efficacy of both markers revealed correct prediction for 73% of all donors . Since normalization of low sIL-7R and high IL-7 plasma concentrations during recovery from tuberculosis was found , these parameters may qualify as biomarker candidates for successful tuberculosis chemotherapy . This study was not designed to evaluate markers for the efficacy of tuberculosis therapy but future studies may address this important question . Immunomodulatory therapies of tuberculosis gained increasing interest during recent years to complement antibiotic therapy that is periled e . g . by multi-drug resistant mycobacteria [73] . IL-7 is a promising candidate for immunotherapies and is already applied in clinical trials against chronic viral infections [74 , 75] . However , the mechanisms underlying impaired IL-7 signalling pathways during chronic infections may antagonise IL-7-based novel therapy strategies . Our study contributed to the characterisation of impaired IL-7 T-cell response that may indeed counteract IL-7 treatment in tuberculosis . We provide initial evidence that IL-7-availability is not critical during tuberculosis . Instead , T-cell functions in response to IL-7 are impaired , and therefore approaches targeting T-cell abnormalities—causative for reduced IL-7 response—may be helpful . Since IL-7 availability is a crucial factor for the development of memory T-cell induction [76] , such an approach might also aim at improving protection against recurrent M . tuberculosis infection and disease .
In this hospital-based observational study , we recruited adult tuberculosis patients ( n = 57; Table 1 ) and exposed but healthy household contacts ( healthy contacts ) ( n = 151 ) . Tuberculosis patients were recruited at the Komfo Anokye Teaching Hospital ( KATH ) , the Kumasi South Hospital ( KSH ) , and the Kwame Nkrumah University of Science and Technology ( KNUST ) Hospital , Ghana , in 2011–2012 . Diagnosis of tuberculosis was based on patient history , chest X-ray , and sputum smear test . For sputum smear negative cases , laboratory confirmation by M . tuberculosis sputum culture was performed . Tuberculosis patients with a known history of HIV infection were excluded from this study . Chemotherapy according to the Ghanaian guidelines was initiated immediately after the first blood sample was taken . For the patient study group , peripheral heparinised blood was taken consecutively ( i . e . prior to treatment , under treatment ( at 2 months ) , and after recovery ( at 6 months ) ) . Only a subgroup of tuberculosis patients ( n = 36 ) completed the study procedure . Twenty-one tuberculosis patients were not included at all time points , including nine patients included only prior to treatment; six patients prior to treatment and under treatment; two patients prior to treatment and after recovery , and four patients during treatment and after recovery . Healthy tuberculosis patient contacts ( short: healthy contacts ) were recruited at the homes of tuberculosis index cases and showed no clinical symptoms of tuberculosis . A subgroup of healthy contacts ( n = 19 ) and tuberculosis patients ( n = 32 ) was tested for M . tuberculosis PPD-specific immune response before and showed significant IFNγ expression [77] . We took heparinised blood ( up to 30 ml ) from each donor . Not all samples were included for all experiments , and the respective numbers of samples included are given in the figure legends . A second cohort of tuberculosis patients ( n = 22 ) and healthy contacts ( n = 24 ) were recruited in the period of October 2015 to March 2016 . HIV-positive individuals were excluded from the analysis ( First Response HIV 1–2 . 0 Card Test , Premier Medical Corporation ) . All study participants were adults who gave written informed consent . All participants were free to drop out at any time of the study . The studies were approved by the Committee on Human Research , Publication and Ethics ( CHRPE ) at the School of Medical Sciences ( SMS ) at the Kwame Nkrumah University of Science and technology ( KNUST ) in Kumasi , Ghana . Peripheral blood mononuclear cells ( PBMCs ) were isolated from heparinised whole blood ( diluted 1:1 in PBS ) by density centrifugation ( Ficoll , Biochrom ) according to manufacturer’s instructions . PBMCs were cryopreserved in DMSO/FCS ( each 10% ) containing medium . The plasma layer ( diluted 1:1 in PBS ) were collected and frozen at -80°C until processing . Diluted plasma samples were thawed in parallel and analysed for sIL-7R expression . Quantification of sIL-7R was performed according to the protocol of Faucher et al . [78] with minor modifications . In brief , we applied cytometric bead array ( CBA ) ( Bead A4 , BD Biosciences ) . Conjugation of beads with polyclonal goat anti-human CD127 ( IL-7Rα ) antibody ( R&D Systems , AF306 ) was done according to manufacturer’s instructions . Biotinylated mouse anti-human CD127 ( clone HIL-7R-M21 , BD Biosciences ) was used as detection antibody . Samples were incubated with labelled beads in PBS for 1 hour at room temperature and then the detection antibody ( 5 μl ) was added for overnight incubation in the fridge . Afterwards , Streptavidin-PE ( 1 μl ) ( Southern Biotech ) was added and incubated for 30 min at room temperature . Finally the beads were washed twice in PBS . For analyses , the bead pellets were resuspended in 80 μl PBS and analysed using a BD LSRFortessa flow cytometer ( BD Biosciences ) and the FCS Express 4 ( De Novo Software ) software . For absolute quantification , the assay was calibrated with dilutions of rhIL-7R alpha-Fc chimera ( R&D Systems ) . sIL-7R concentrations were calculated using the non-linear regression tool of GraphPad Prism 6 ( Graphpad Software Inc . ) . Possible effects of IL-7 on sIL-7R measure were excluded by Faucher et al . [78] . IL-6 and IL-7 was determined in duplicate for diluted plasma samples using Human IL-6 ELISA Ready-SET-Go ! ( eBioscience ) and Human IL-7 Quantikine HS ELISA kit ( R&D Systems ) , respectively , according to manufacturer’s instructions . Samples were measured using the Infinite M200 ELISA reader ( Tecan ) . Concentrations were calculated from the respective standard curves by applying 4-parametric logistic regression . Samples outside the detection range were set to the corresponding lower or upper range value . CD4+ cells were isolated from freshly isolated PBMCs ( 1 . 5 x 107 cells ) using anti-human CD4 magnetic particles ( BD Biosciences ) according to manufacturer’s recommendations . Cell purity was evaluated by flow cytometry and was generally higher than 95% . miRNA was isolated from at least 5 x 106 enriched CD4+ cells using mirVanaTM miRNA Isolation Kit ( Life Technologies ) following manufacturer’s instructions . cDNA was generated by Maxima H Minus First Strand cDNA Synthesis kit ( Thermo Scientific ) , while RT-PCR was performed with the QuantiTect SYBR Green PCR kit ( Qiagen ) for full-length IL-7R ( H20: forward 5’-AATAATAGCTCAGGGGAGATGG-3’ , reverse 5’-ATGACCAACAGAGCGACAGAG-3’ ) , IL-7R lacking exon 6 ( H6: forward 5’-GATCAATAATAGCTCAGGATTAAGC-3’ , reverse 5’-AAGATGTTCCAGAGTCTTCTTATG-3’ ) , and IL-7R lacking exon 5–6 ( H5-6: forward 5’-ATGAAAACAAATGGACGGATTAAGC-3’ , reverse 5’-AAGATGTTCCAGAGTCTTCTTATG-3’ ) , PD-1 ( forward 5’-CTCAGGGTGACAGAGAGAAG-3’ , reverse 5’-GACACCAACCACCAGGGTTT-3’ ) , SOCS3 ( forward 5’-GACCAGCGCCACTTCTTCAC-3’ , reverse 5’-CTGGATGCGCAGGTTCTTG-3’ ) using glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) as housekeeping control gene ( forward 5’-CACCATCTTCCAGGAGCGAG-3’ , reverse 5’-GACTCCACGACGTACTCAGC-3’ ) . The reaction with a final volume of 25 μl was run 2 min . at 50°C , 10 min . at 95°C , 45 cycles of 15 s at 95°C , 30 s at 53°C and 30 s at 72°C , followed by a melt curve sequence of 15 s at 95°C , 60 s at 60°C with a slow gradient to 95°C and finally 15 s at 60°C . Data from duplicate reactions was evaluated using the 2-ΔCt method . A 7500 Real-Time PCR machine ( Applied Biosystems ) was used for quantitative PCR analyses . DNA was isolated from PBMCs using QIAamp DNA Mini Mini kit ( Qiagen ) followed by rs6897932C>T genotyping using a predesigned TaqMan SNP Genotyping Assay ( Applied Biosystems ) following manufacturer’s instructions . Frozen PBMCs were thawed and washed with RPMI 1640 supplemented with 10% foetal calf serum ( FCS ) , 2 mM L-Glutamine , 10 mM HEPES , and 50 U/ml Penicillin-Streptomycin ( all from Thermo Fisher ) . Cells were stained with Viability Dye eFluor 780 ( eBioscience ) and antibodies against CD3 ( PE-labelled , clone HIT3a , BD Biosciences ) , CD4 ( BrilliantViolet510-labelled , clone OKT4 , BioLegend ) , CD8 ( PerCP-Cy5 . 5-labelled , clone HIT8a , BioLegend ) , CD25 ( PE-Cy7-labelled , clone 2A3 , BD Biosciences ) and CD127 ( AlexaFluor647-labelled , clone HIL-7R-M21 , BD Biosciences ) . After cell wash , PBMCs were fixed with Fixation Buffer ( BioLegend ) and subsequently analysed using a BD LSRFortessa flow cytometer ( BD Biosciences ) . Gating procedures are depicted in S3 Fig . For detection of mIL-7R in the second independent cohort of tuberculosis patients and healthy contacts we used the CD127 antibody clone A019D5 ( BioLegend ) . Comparison of both antibody clones revealed similar T-cell binding pattern as well as percentages of mIL-7Rhigh and mIL-7Rlow T cells . Freshly isolated PBMCs were stained for CD4 ( AlexaFluor488 , clone RPTA-4 , BioLegend ) followed by addition of 100 μl pre-warmed X-VIVO 15 medium ( Lonza ) added 50 U/ml Penicillin-Streptomycin with or without human recombinant IL-7 . The concentration of IL-7 was titrated prior to the study and a concentration of 10 ng/ml was sufficient to induce pSTAT5 in 94% of the T cells ( S5 Fig ) . Higher IL-7 concentrations ( 25 or 50 ng/ml ) did not further increase STAT5 phosphorylation ( S5 Fig ) . Therefore we cultured the samples with and w/o 10 ng/ml of recombinant human IL-7 in this study . After 15 min incubation at 37°C , 5% CO2 , cells were fixed for 15 min . with 100 μl 1x True-Nuclear Transcription Factor buffer ( BioLegend ) . Subsequently , cells were permeabilised with 100% methanol , washed in PBS/10% FCS and stained for p-STAT5 Y694 ( PE , clone SRBCZX , eBioscience ) . Analysis was performed on a BD Accuri C6 flow cytometer . Gating procedure is shown in S5 Fig . Heparinised blood was diluted 1:2 in RPMI 1640 supplemented with 2 mM L-Glutamine and 50 U/ml Penicillin-Streptomycin in a 96-well U bottom plate . Cells were stimulated with 10 μg/ml PPD ( Statens Serum Institute ) and/or 10 ng/ml recombinant human IL-7 ( BioLegend ) , or left unstimulated . After 2 . 5 hours of stimulation at 37°C , 5% CO2 , Brefeldin A ( Sigma Aldrich ) was added at a concentration of 3 . 75 μg/ml followed by 16 hours of incubation . Erythrocytes were subsequently lysed in two rounds by resuspending pelleted cells in 100 μl RBC Lysis Buffer ( Roche ) followed by 10 min incubation at room temperature . Next , cells were fixed and permabilised ( BioLegend ) and stained with antibody against CD4 ( AlexaFluor488 , clone RPTA-4 , BioLegend ) , IFNγ ( PE , clone 25723 . 11 , BD Biosciences ) and CD154 ( APC , clone 24 . 31 , BioLegend ) . Cells were analysed using a BD Accuri C6 flow cytometer ( BD Biosciences ) . Gating procedure is shown S6 Fig . Statistical analyses were performed using R version 3 . 3 . 0 , applying Exact Mann-Whitney U test from the package coin for comparison between groups and Wilcoxon signed-rank test for evaluation of repeated measurements . Spearman correlation was used to evaluate association between continuous variables , while Receiver Operating Characteristic ( ROC ) was performed using the package ROCR . Random forest analysis was performed with the package ranger , applying 105 random trees and adjusting the importance measure by permutation . Plots were generated in R and GraphPad Prism version 6 . 07 . | IL-7 is important for the development and homeostasis of T cells and promotes antigen-specific T-cell responses . Aberrant expression of plasma IL-7 and soluble IL-7R are found in autoimmune diseases and chronic viral infections . In AIDS patients—especially those who fail to reconstitute T-cell numbers during therapy—impaired IL-7-promoted T-cell functions indicated T-cell exhaustion/senescence . In order to evaluate the potential impact of IL-7 on tuberculosis , we characterised various parameters involved in the IL-7-response of tuberculosis patients and healthy contacts . Despite IL-7 being available at higher plasma levels among tuberculosis patients , the T-cell response to IL-7 was impaired when compared to healthy contacts . Soluble IL-7R levels were aberrantly low in plasma during acute tuberculosis but did not account for impaired IL-7 usage . Chronic inflammation in tuberculosis patients—reflected by increased IL-6 plasma levels—did not account for dysfunctional T-cell responses and analysed T-cell exhaustion markers were only moderately correlated . Our findings demonstrate that availability of IL-7 alone is not sufficient to promote protective T-cell immunity against tuberculosis . We describe aberrant IL-7/soluble IL-7R expression and impaired IL-7-mediated T-cell functions in tuberculosis patients with similarities and differences to described IL-7 dysregulation seen in patients with AIDS . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"blood",
"cells",
"phosphorylation",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"body",
"fluids",
"immune",
"physiology",
"cytokines",
"immunology",
"tropical",
"diseases",
"cloning",
"bacterial",
"diseases",
"developmental",
"biology",
"molecular",
"development",
"molecular",
"biology",
"techniques",
"cytotoxic",
"t",
"cells",
"bacteria",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"tuberculosis",
"proteins",
"t",
"cells",
"actinobacteria",
"molecular",
"biology",
"blood",
"plasma",
"immune",
"system",
"biochemistry",
"blood",
"cell",
"biology",
"anatomy",
"mycobacterium",
"tuberculosis",
"post-translational",
"modification",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
]
| 2017 | Aberrant plasma IL-7 and soluble IL-7 receptor levels indicate impaired T-cell response to IL-7 in human tuberculosis |
Uropathogenic Escherichia coli ( UPEC ) is a leading etiological agent of bacteremia in humans . Virulence mechanisms of UPEC in the context of urinary tract infections have been subjected to extensive research . However , understanding of the fitness mechanisms used by UPEC during bacteremia and systemic infection is limited . A forward genetic screen was utilized to detect transposon insertion mutants with fitness defects during colonization of mouse spleens . An inoculum comprised of 360 , 000 transposon mutants in the UPEC strain CFT073 , cultured from the blood of a patient with pyelonephritis , was used to inoculate mice intravenously . Transposon insertion sites in the inoculum ( input ) and bacteria colonizing the spleen ( output ) were identified using high-throughput sequencing of transposon-chromosome junctions . Using frequencies of representation of each insertion mutant in the input and output samples , 242 candidate fitness genes were identified . Co-infection experiments with each of 11 defined mutants and the wild-type strain demonstrated that 82% ( 9 of 11 ) of the tested candidate fitness genes were required for optimal fitness in a mouse model of systemic infection . Genes involved in biosynthesis of poly-N-acetyl glucosamine ( pgaABCD ) , major and minor pilin of a type IV pilus ( c2394 and c2395 ) , oligopeptide uptake periplasmic-binding protein ( oppA ) , sensitive to antimicrobial peptides ( sapABCDF ) , putative outer membrane receptor ( yddB ) , zinc metallopeptidase ( pqqL ) , a shikimate pathway gene ( c1220 ) and autotransporter serine proteases ( pic and vat ) were further characterized . Here , we report the first genome-wide identification of genes that contribute to fitness in UPEC during systemic infection in a mammalian host . These fitness factors may represent targets for developing novel therapeutics against UPEC .
Uropathogenic Escherichia coli ( UPEC ) , one of the most common bacterial pathogens infecting humans , is the primary etiological agent of urinary tract infections ( UTI ) in otherwise healthy individuals [1] . UPEC is a subset of extraintestinal pathogenic E . coli ( ExPEC ) , which causes a broad spectrum of conditions including colibacillosis in poultry , and UTIs , bacteremia , and neonatal meningitis in humans [2] . A subset of patients with UTI develops pyelonephritis and is at risk for developing bacteremia that may result in life threatening sepsis . UTI is the source of E . coli in >70% of both young and elderly patients with bloodstream infections [3] , [4] . E . coli strains isolated from the bloodstream are becoming increasingly resistant to trimethoprim/sulfamethoxazole and ciprofloxacin , two first line antibiotics used to treat bacterial UTIs [5] . Despite the prevalence of these infections and potential difficulties in treatment , little is known about the fitness and virulence mechanisms employed by E . coli to establish a systemic infection . The marriage between transposon mutagenesis and high-throughput ( HT ) sequencing has resulted in the emergence of powerful techniques that can be harnessed for global functional genomic studies [6] . Here , we utilize an adaptation of transposon directed insertion-site sequencing ( TraDIS ) [7] to identify genes required for optimal fitness of UPEC during colonization and survival in a murine model of bacteremia . Recently , such approaches were used to determine virulence and fitness factors in Yersinia pseudotuberculosis [8] and Salmonella enterica serovar Typhimurium [9] utilizing animal models of infection and colonization . Genes that encode microbial proteins and organelles that specifically aid in pathogenesis are known as virulence genes . Bacterial pathogens are adept at co-opting genes that are otherwise used in non-pathogenesis related roles for gaining fitness advantage during infection . In this context , fitness refers to enhanced survival and growth within a given niche . Genes that promote colonization and survival of UPEC within murine hosts are referred to as fitness factors in this manuscript . A subset of the fitness factors reported here , represent virulence factors that meet the criteria defined by molecular Koch's postulates [10] . This report , to our knowledge , represents the first global functional genomic screen aimed at identification of in vivo fitness factors in a pathogenic E . coli strain involving a targeted-sequencing approach . In this study , a murine model of invasive UPEC infection , previously developed in our laboratory [11] , was used in conjunction with transposon mutagenesis to identify bacterial fitness mechanisms involved in establishing systemic infection . Mice were inoculated intravenously with an inoculum derived from a saturating transposon mutant library of a clinical bacteremia isolate , E . coli CFT073 . Transposon mutants that colonized and survived in mouse spleens ( output ) were isolated . Transposon insertion sites in the input and output samples were mapped to the genome of the UPEC strain CFT073 . 242 candidate fitness genes that are required for optimal survival in the spleen were identified in the primary screen . Genetically defined mutants were constructed and tested for in vivo and in vitro fitness phenotypes using assays relevant to the infection biology of UPEC . A subset of these fitness factors are also involved in the development of UTI in a mouse model and suggests the existence of shared fitness mechanisms used at these disparate body sites . In summary , we present a comprehensive study of fitness factors that augment the survival of UPEC during systemic disseminated infection in a mammalian host .
E . coli CFT073 , isolated from the urine and blood of a patient hospitalized with pyelonephritis and bacteremia [12] , was used to construct a high-density transposon mutant library . An estimated 48 , 174 transposon mutants are required to obtain a 99 . 99% saturation of the CFT073 genome [13] , which is 5 . 2 Mbp in length [14] . A genome-supersaturating Tn5 transposon mutant library , containing 360 , 000 kanamycin-resistant transformants , was generated for this study . The library was passaged three times in lysogeny broth ( LB ) to enrich for mutants that did not exhibit a fitness defect in vitro . This enriched mutant pool was used as the inoculum for infection experiments . A murine model of systemic disseminated UPEC infection [11] was used to determine the highest dose of wild-type CFT073 that consistently resulted in non-lethal infection . Three doses ( 106 , 107 and 108 CFU/mouse ) were compared in the CBA/J mouse model of systemic disseminated infection . Mice were inoculated via tail vein , and livers and spleens were collected 24 h post inoculation ( hpi ) . An inoculum of 107 CFU resulted in consistent colonization without causing distress in the inoculated animals ( Fig . 1A ) . Inoculation with 106 CFU led to poor colonization , whereas a dose of 108 CFU resulted in 20% mortality ( Fig . 1A ) . Twenty mice were inoculated with 107 CFU of transposon-insertion mutants ( input ) and euthanized 24 hpi ( Fig . 1B ) . As a major reticuloendothelial organ , the spleen is a critical site of active bacterial killing during systemic infection [15] . Therefore , a bacterium that successfully survives in the spleen should contain the full complement of fitness factors that are critical for survival in that niche , including the ability to overcome host defenses activated during systemic bacterial infection . Bacteria that grew from splenic homogenates were harvested ( output ) and used to isolate genomic DNA for Illumina sequencing . Transposon insertion sites in the inoculum ( input ) and bacteria colonizing the spleens ( output ) were determined using transposon directed insertion-site sequencing ( TraDIS ) , a HT sequencing-based approach [7] . Genomic DNA from the input and output samples were used to generate TruSeq sequencing libraries ( Illumina ) . Libraries were amplified using a transposon-specific forward primer and a custom adapter-specific reverse primer ( Table S1 ) . Resulting amplicons were used for cluster generation and each library was sequenced with a Tn-specific primer ( Table S1 ) in an IIlumina HiSeq 2000 sequencer . Fifty nucleotide single-end reads in FASTQ format were aligned to the E . coli CFT073 genome [14] using the short read aligner , BOWTIE [16] . The number of reads processed was 75 , 935 , 499 and 87 , 030 , 926 for the input and output samples , respectively . 76 . 7% ( 58 , 209 , 557 ) of the reads from input and 77 . 9% ( 67 , 810 , 765 ) of the reads from output were aligned unambiguously to the CFT073 genome . TFAST [17] was used to determine the exact genomic location of the Tn-insertion and the frequency of reads that map to a given insertion site . Reads for each transposon insertion site were normalized to the total number of reads obtained from that sample and a fitness factor was calculated for each Tn-insertion mutant as the ratio of normalized frequency of reads in the input to that of the output ( Fig . 2A , Table 1 and Table S2 ) . Therefore , a fitness factor ≥1 indicates that a given mutant is underrepresented in the output pool . For example , the sensitive to antimicrobial peptide ( sap ) gene cluster in E . coli CFT073 is depicted along with the frequency of representation of transposon-insertion mutants in the input and the output pools ( Fig . 2A ) . Transformants containing an insertion in the sap genes are less well represented in the output pool compared to the input pool . A total of 6732 unique Tn5 insertion sites were mapped in the CFT073 genome ( Table S2 ) with a mean fitness factor of 3 . 27±1 . 57 . The insertion sites were distributed throughout the length of the CFT073 genome . At least a single transposon insertion site was observed in 3020 genes and an additional 843 intergenic regions , which could exert polar effects of downstream genes . A relatively short region in the genome ( 30 Kbp in length ) reveals the presence of transposon insertion mutants with a broad range of fitness factors ( Fig . 2B ) . The median distance between independent insertion sites was 561 bp . A total of 372 transposon insertion mutants , resulting in inactivation of 242 genes , exhibited fitness factors >6 . 41 ( mean+2 standard deviations ) and were considered as candidate fitness genes . 50 transposon mutants with the highest in vivo fitness defect phenotype are listed in Table 1 . Seven ( 1 . 9% ) of 372 candidate transposon mutants were previously designated as essential genes in a laboratory strain of E . coli K-12 [18] . These genes encode NrdB , an aerobic ribonucleotide reductase; MviN , a peptidoglycan lipid II flippase; LolD , lipoprotein releasing system ATP-binding protein; MinD , septum site-determining protein; GapA , glyceraldehyde-3-phosphate dehydrogenase; FabG , 3-ketoacyl-acyl carrier protein reductase; and MsbA , lipid flippase . Genes involved in nucleotide metabolism were previously described to play a critical role in growth of a non-pathogenic strain of E . coli in human blood [19] . The following genes were identified in our primary screen: guaB , inosine monophosphate dehydrogenase involved in guanosine monophosphate biosynthesis; ntpA , dATP pyrophosphohydrolase involved in degradation of dATP; pyrC , dihydroorotase catalyzes the conversion of carbamoylaspartate to dihydrooratate; and yeiA , dihydropyrimidine dehydrogenase catalyzes first step in degradation of uracil and thymine ( Table S2 ) . Several genes encoding surface structures that could potentially be involved in direct interaction with host cells were identified in our screen ( Table S2 ) . Periplasmic murein-peptide binding protein precursor gene ( mppA ) and a periplasmic protease that processes penicillin-binding protein 3 ( prc , c2239 ) [20] are peptidoglycan biosynthetic genes that were identified in our primary screen . arnT and yfbH genes involved in resistance to polymyxin B , a peptide antibiotic that mimics the activity of host-derived cationic antimicrobial peptides , were identified in our primary screen . ArnT reduces the negative charge on the lipopolysaccharide due to its 4-amino-4-deoxy-L-arabinose transferase activity [21] . YfbH is a homolog of PmrJ , a deacetylase involved in biosynthesis of amino-arabinose-modified lipid A [22] . Two outer membrane porins , ompC and ompG were also identified in our screen . Colanic acid is a surface polysaccharide that is associated with biofilm formation in E . coli [23] . Two genes involved in colonic acid biosynthesis , wcaM and wcaL , [24] were identified as candidate fitness genes in the mouse bacteremia model . pgaABD , c2394-95 and yddB are other genes associated with surface structures identified in our screen and were subjected to further investigation . Mammalian hosts actively limit the bioavailability of iron to hamper the growth of invading pathogens . Multiple genes involved in distinct iron acquisition systems were identified in our screen ( Table S2 ) . The sitC gene harbored on a bacteriophage , part of the SitABCD system involved in manganese and iron transport [25] , was among the candidate fitness genes identified in this study ( Table S2 ) . A chuA hma double mutant , lacking two heme receptors , was previously found to exhibit fitness defect during bacteremia [11] . In the current study , mutation in hma alone reveals a fitness defect ( Table S2 ) suggesting that heme is a major source for iron during systemic infection in a mammalian host . Multiple insertion sites were found within the genes involved in enterobactin biosynthesis , export , uptake and utilization with a mean fitness factor of 3 . 2 . Salmochelin is a glycosylated derivative of enterobactin that evades chelation by host protein lipocalin-2 [26] . Inactivation of genes involved in salmochelin biosynthesis ( iroB ) and uptake ( iroN ) also resulted in attenuated fitness ( Table S2 ) . Insertion mutants in yersiniabactin biosynthesis and uptake genes also revealed a minor fitness defect ( Table S2 ) . Yersiniabactin , however , is not produced by E . coli CFT073 due to a previous insertion event at this locus . Coprogen and hydroxamate siderophore receptor gene , fhuE , was found among the candidate fitness genes ( Table S2 ) . Our results are consistent with the previously established role of iron acquisition genes in fitness of UPEC during systemic infection . Tn-insertions leading to fitness defect in multiple genes within an operon/cluster and genes that were previously not known to affect fitness of UPEC during systemic infection were selected for further validation . Additionally , we tested the role of pic and vat in fitness primarily to establish that we have utilized a conservative threshold to delineate fitness genes . Co-infection experiments were performed by inoculating mice intravenously with equal numbers of both wild-type and mutant bacteria lacking select genes identified in the primary screen . Since most cases of bacteremia caused by UPEC are a result of ascending UTIs , we also tested the ability of a subset of these mutants to colonize murine urinary tract . Growth kinetics of all the mutants used in the experiments described in the following sections is indistinguishable from that of the wild-type strain ( Fig . S1 ) . Homogenates of organs ( spleen and liver from bacteremia model; urinary bladder and kidneys from ascending UTI model ) were plated on plain and selective media . Differential plate counts were used to determine bacterial loads of wild-type and mutant strain in each tissue . Competitive indices ( CI ) were calculated using colony counts as: [mutant CFU/wild-type CFU ( output ) ]/[mutant CFU/wild-type CFU ( input ) ] . CI values less than 0 ( log10 scale ) indicate a comparative fitness defect for the mutant with respect to wild-type strain ( Fig . 3A and 3B , 4A and 5A ) . Nine of the 11 mutants ( 82% ) were out-competed by the wild-type strain during co-infection indicating that a high proportion of the candidate fitness genes identified in the primary screen indeed function as fitness factors during systemic infection ( Fig . 3A and 3B , 4A and 5A ) . After successful validation of the primary screen , we probed the function of select candidate fitness genes in UPEC . Transposon insertions within pgaA , pgaB , and pgaD resulted in reduced fitness , corresponding to fitness factors of 7 . 43 , 9 . 41 , and 4 . 32 , respectively ( Tables 1 and S2 ) . The pgaABCD operon is involved in the biosynthesis and export of an extracellular polysaccharide , poly-N-acetyl glucosamine ( PNAG ) , in E . coli [27] . Loss of PNAG biosynthetic operon resulted in a fitness defect in a mouse model of bacteremia ( spleen , P = 0 . 002; Fig . 4A ) . The pgaABCD mutant was out-competed by the wild-type strain , ∼10-fold , both within the urinary bladder and the kidneys demonstrating that PNAG acts as a fitness factor in vivo within the murine urinary tract ( Fig . 4A ) . The pgaA gene was upregulated 32-fold and pgaC transcript was detected by RT-PCR in urine collected from mice infected with E . coli CFT073 , indicating that these genes are highly expressed during UTI . Furthermore , transcriptome analysis of UPEC CFT073 revealed that the pgaABCD genes are upregulated ( ∼2-fold ) during culture in human urine compared to LB ( unpublished results ) . Since PNAG is involved in biofilm formation in a non-pathogenic strain of E . coli [27] , we tested the contribution of PNAG to biofilm formation in the UPEC strain CFT073 . Biofilm-forming ability of wild-type and pgaABCD mutant was tested using a crystal violet binding assay . Loss of PNAG did not affect biofilm formation on polystyrene ( Fig . 4B ) or glass surface ( data not shown ) . Since UPEC is decorated with several surface structures , including multiple fimbriae and autotransporter adhesins , which might compensate for the loss of PNAG-dependent adhesion , the effect of overexpression of the pgaABCD operon on biofilm formation was also tested . Full-length pgaABCD operon including the native promoter was cloned into a multi-copy vector ( pSS1 ) ; PNAG could be readily detected , by immunoblot analysis , in the overexpression strain but not in the vector control ( Fig . 4C inset ) . Upon overexpression , PNAG promotes robust biofilm formation in UPEC strain CFT073 ( Fig . 4B ) . E . coli K-12 strain MG1655 also displayed a profound , PNAG-dependent increase in biofilm formation , indicating that PNAG promotes biofilm formation in E . coli using a non strain-specific mechanism ( Fig . 4B ) . Factors involved in adherence are known to affect motility in UPEC [28] . In strain CFT073 , loss of PNAG production results in a significant increase in motility ( Fig . 4C ) that is accompanied by a 4-fold increase in the expression of fliC ( data not shown ) . A higher level of fliC expression ( encoding flagellin , the major structural subunit of flagella ) explains the increased motility observed in the pgaABCD mutant . Conversely , overexpression of PNAG diminishes motility ( Fig . 4C ) suggesting that PNAG production and motility could be controlled in a reciprocal manner . Overexpression of PNAG also resulted in decreased motility in E . coli K-12 ( Fig . 4C ) . Taken together , motility is adversely affected during PNAG overexpression in a non strain-specific manner . Additionally , known repressors of flagellar motility , PapX and FocX , are not involved in this crosstalk between PNAG levels and swimming motility ( data not shown ) . Intact epithelial surface precludes the access of pathogens to ECM proteins; however , inflammation-associated mucosal denudation results in contact with ECM proteins . A plate-based adherence assay was used to determine whether PNAG is involved in adherence to common ECM proteins collagen I , collagen IV , fibronectin and laminin . PNAG overexpression resulted in significantly higher adherence of UPEC strain CFT073 to collagen I , collagen IV and laminin ( Fig . 4D ) compared to vector control . PNAG does not affect binding to fibronectin under the assay conditions tested . To determine if PNAG protects UPEC from killing by macrophages , survival of wild-type , pgaABCD , wild-type ( pSS1 ) , and wild-type ( pTopo ) within the murine macrophage cell line RAW264 . 7 was assessed . Under our experimental conditions , PNAG did not contribute to adherence or intracellular survival ( data not shown ) . c2394 encoding PilV was identified in the primary screen as a putative fitness gene ( fitness factor = 6 . 7 , Table S2 ) . pilV ( c2394 ) and pilS ( c2395 ) , encoding pilin subunits of type IV pilus two , are highly associated with UPEC strains compared to fecal E . coli isolates [29] . Additionally , these genes are more prevalent in E . coli isolated from humans than from animals [29] . Co-infection experiments revealed that the mutant strain lacking pilV and pilS genes was significantly out-competed ( P<0 . 05 ) by wild-type E . coli CFT073 in the bladder , kidneys , spleen and liver ( Fig . 5A ) . Our data demonstrate that these putative type IV pilin subunit genes are involved in colonization during both systemic infection and UTIs . The ability of the isogenic mutant , c2394-95 ( pGEN ) , and the complemented strain , c2394-95 ( pGEN- c2394-95 ) to adhere to the immortalized epithelial cell lines UMUC-3 ( human bladder ) , HEK293 ( human embryonic kidney ) , VERO ( green monkey kidney ) , and MM55K ( mouse kidney ) was compared to that of wild-type ( pGEN ) strain . Compared to wild-type , c2394-95 mutant was less adherent to UM-UC-3 ( P = 0 . 032 ) , HEK293 ( P = 0 . 031 ) , and VERO ( P = 0 . 012 ) ( Fig . 5B ) cells . However , no significant difference was observed on MM55K ( P = 0 . 675 ) cells . Complementation restored adherence to wild-type levels on all cell lines ( Fig . 5B ) . This suggests that c2394-95 encode proteins involved in adherence to uroepithelial cells and the receptor for type IV pilin is likely expressed by both bladder ( human ) and kidney ( human and monkey ) epithelial cells , but not by the mouse kidney cell line . Electron microscopy was used to determine if the type IV pilus is indeed found on the cell surface . Wild-type and complemented mutant cells are densely piliated compared to the mutant strain that is sparsely piliated ( Fig . 5C ) . A c2394-95 mutant had a swimming diameter of 44 . 9±7 . 7 mm , significantly lower than that of wild-type E . coli CFT073 ( P = 0 . 005 ) , which swam 59 . 8±4 . 3 mm . Motility was not restored to wild-type levels by complementation ( 38 . 4±6 . 0 mm ) with c2394-95 in trans; instead , the motility defect was increased upon expression of the pilin genes , which suggests that there may be a decrease in motility due to the level of expression from a multi-copy plasmid . Deletion of c2394-95 does not appear to affect cell aggregation , biofilm formation or invasion of kidney epithelial cells in E . coli CFT073 ( data not shown ) . Multiple peptide uptake genes ( oppABD , sapACF and tppB ) were identified as candidate fitness genes in the primary screen ( Tables 1 and S2 ) . Mutants lacking the sap gene cluster or oppA , but not the tppB were found to exhibit fitness defects in spleen and liver during co-challenge experiments , compared to the wild-type strain ( Fig . 3 ) . The opp gene cluster harbors the genes involved in oligopeptide uptake and multiple transposon insertion sites were observed within these genes ( Table S2 ) . This observation suggests that the ability to utilize oligopeptides as a source of carbon and nitrogen is critical for UPEC survival in murine spleens . Transposon insertions and corresponding fitness factors for the sap genes are depicted in Fig . 2A . Cationic antimicrobial peptides represent a major antimicrobial defense system that aids in clearing invading pathogens . Polymyxin B ( PB ) is a peptide antibiotic that emulates the activity of host-derived cationic antimicrobial peptides [30] . The role of peptide uptake systems in resistance of UPEC to PB was tested . A mutant defective in dipeptide uptake ( dppA ) [31] , not identified in our primary screen , was also used to determine if multiple peptide uptake systems are involved in PB resistance . Bacterial cultures in exponential phase of growth were exposed to PB and percent survival was calculated using colony counts from PB-treated and control cultures . Fold-change in resistance was calculated as the ratio of survival percentage of a given mutant to that of wild-type strain ( Fig . 6A ) . Compared to the wild-type strain , the sapR mutant , that lacks the sapABCDF genes , exhibited increased sensitivity to PB ( Fig . 6A , P<0 . 0001 ) . The oppA mutant exhibited decreased sensitivity to PB compared to wild-type strain ( Fig . 6A , P = 0 . 03 ) and the dppA mutant also showed a trend towards decreased sensitivity to PB ( Fig . 6A ) . Gentamicin protection assay was used to determine if the peptide uptake systems contributed to intracellular survival of UPEC in murine macrophage cells ( RAW 264 . 7 ) . Plate counts were used to determine the number of bacteria that entered and survived within RAW264 . 7 cells for 2 h . Ratio of killing percentages were determined and values >1 indicate that a given mutant was defective in intracellular survival within RAW 264 . 7 cells ( Fig . 6B ) . A modest , but statistically significant reduction ( P = 0 . 03 ) in intracellular fitness of the sapR mutant was observed ( Fig . 6B ) , whereas the oppA and dppA mutants were not defective in intracellular survival compared to wild-type strain . Although tppB was identified as a fitness gene in the primary screen , a co-challenge experiment revealed no role for this gene in fitness ( Fig . 3 ) . This discrepancy could be due to the differences in the nature of competition during infection with the transposon mutant library in the primary screen versus one-to-one competition between wild-type and mutant strains in our secondary validation experiments . In the E . coli CFT073 genome , yddA , yddB and pqqL encode an ABC transporter ATPase , an outer membrane β-barrel protein and an inner membrane-associated zinc metallopeptidase , respectively . yddB and pqqL were identified as fitness genes in our primary screen and median fitness factors for multiple insertion mutants in these genes are depicted in Fig . 7A . A BLAST search revealed that this gene cluster is found only among E . coli and Shigella strains . yddB and pqqL genes are involved in fitness during systemic infection in a mammalian host ( Fig . 3 ) . RT-PCR experiments revealed that these genes are indeed co-transcribed as a single mRNA ( Fig . 7B ) . YddB exhibits a high degree of sequence similarity to ligand-gated outer membrane β-barrel proteins such as ferrienterobactin receptor , FepA in E . coli . Since outer membrane β-barrel proteins are usually involved in iron uptake and a putative Fur box ( GGGAATGGTTATCATTAG ) is found overlapping the start codon of yddA , we tested whether these genes are differentially expressed during culture in human urine , an iron limited milieu . RNA was extracted from CFT073 bacterial cells cultured to mid-exponential phase in either LB or filter-sterilized human urine and gene expression was quantified using RNAseq ( unpublished results ) . The yddA , yddB , and pqqL genes are highly upregulated ( >30-fold ) during growth in human urine compared to LB ( Fig . 7C ) . We also tested whether iron levels directly regulate the expression of yddA-yddB-pqqL genes . Transcript levels were determined in wild-type strain cultured in LB , LB supplemented with an iron chelator ( Dipyridyl ) or additional iron . Iron levels alone do not affect the expression of these genes in UPEC ( Fig . S2A ) . However , yddA , yddB , and pqqL genes are upregulated in the Δfur mutant that lacks ferric uptake regulator ( Fur ) , compared to the wild-type strain ( Fig . S2B ) . Taken together , our data indicate that these genes are upregulated during growth in human urine but not subject to regulation by iron levels alone . Proteins in the SPATE ( serine protease autotransporter proteins of Enterobacteriaceae ) family have previously been implicated in the pathobiology of UPEC [32] . Genes encoding members of SPATE family , protease involved in colonization ( Pic ) and vacuolating autotransporter toxin ( Vat , previously known as Tsh ) were identified in our primary screen ( Table S2 ) . The pic and vat transposon insertion mutants exhibited a fitness factor of 5 . 2 and 4 . 4 , respectively . Co-infection experiments were performed with these genes to test whether a conservative threshold was used to delineate fitness genes . Competitive indices reveal that both pic and vat , which exhibit fitness factors lower than the cutoff used to delineate fitness genes , play a role in the fitness of UPEC during systemic infection in mice ( Fig . 3 ) . 3-deoxy-D-arabino-heptulosonic acid-7-phosphate synthase ( DAHPS ) , encoded by c1220 , catalyzes the formation of 3-deoxy-D-arabino-heptulosonate 7-phosphate ( DAHP ) from phosphoenolpyruvate and erythrose 4-phosphate , an early step in shikimate biosynthesis [33] . In CFT073 , c1220 is located on the serX pathogenicity island [34] . DAHPS encoded by c1220 is the fourth isozyme , in addition to aroF , aroG and aroH that catalyzes the production of DAHP in E . coli CFT073 . Two transposon insertion sites mapped to this gene resulted in reduced fitness during survival in spleen . Co-infection experiments with a c1220 mutant and wild-type strain confirmed that the mutant has a fitness defect in spleen and liver during systemic infection in mice ( Fig . 3 ) . A gene encoding an EAL domain protein , ycgF , was identified in our primary screen . EAL domain proteins are usually associated with phosphodiesterase activity that reduces the intracellular levels of an important intracellular messenger , cyclic-di-GMP [35] . YcgF has been designated as an inactive phosphodiesterase that nevertheless positively regulates swimming motility by increasing flagellin levels in CFT073 [35] . Although ycgF was identified as a candidate fitness gene , co-infection experiments failed to reveal a role for ycgF in fitness in a mouse model of systemic infection ( Fig . 3 ) .
Uropathogenic Escherichia coli ( UPEC ) is a major cause of bacteremia in humans , yet , there is limited understanding of the fitness mechanisms used by this important pathogen during bacteremia and systemic infection . Here , we describe screening transposon mutants of E . coli CFT073 in a murine model of systemic disseminated infection and identifying 242 candidate fitness genes . Specific mutations were introduced in 11 candidate fitness genes and the contribution of the following nine gene or gene clusters in fitness was confirmed: pgaABCD ( biosynthesis and export of poly-N-acetyl glucosamine ) , c2394-95 ( major and minor pilin of type IV pilus two ) , oppA ( oligopeptide uptake periplasmic-binding protein ) , sapABCDF ( sensitive to antimicrobial peptide ) , yddB ( putative outer membrane receptor ) , pqqL ( zinc metallopeptidase ) , c1220 ( a shikimate pathway gene ) , and pic and vat ( autotransporter serine proteases ) . 82% of the specific mutants in representative candidate fitness genes were significantly outcompeted by the wild-type strain , validating the TraDIS approach in our murine model of systemic infection . Transposon mutagenesis has been an indispensable tool in unraveling gene function . The complex nature of experiments involving transposon mutant pools , including bottlenecks when screening signature-tagged mutants in animal models of infection [36] , has resulted in screens with fewer mutants than required to achieve genome saturation . Recently , HT sequencing and bioinformatic analyses have been used in tandem to identify transposon insertion sites in genome-saturating transposon mutant pools [6] . Several variants of this approach include HITS , high-throughput insertion tracking by deep sequencing [37]; INSeq , insertion sequencing [38]; Tn-Seq , transposon sequencing [39]; and TraDIS , transposon directed insertion-site sequencing [7] . These techniques utilize chromosomal regions flanking the transposons as unique tags in lieu of the synthetic tags used in signature-tagged mutagenesis . Availability of a large number of bacterial genome sequences combined with cost-effective HT sequencing is poised to make these approaches a staple of functional genomic studies in the near future . Understanding the fitness strategies employed by UPEC during infection of a mammalian host has the potential to identify targets for novel intervention strategies . Here , we describe the first comprehensive identification of fitness factors involved in systemic infection by an ExPEC strain , CFT073 , in a mammalian host . The original work describing TraDIS catalogued the essential genes in Salmonella enterica subsp . enterica serovar Typhi [7] . Since the primary objective of our study was to identify in vivo fitness factors , the mutant pool was passaged in LB to deplete mutants with in vitro fitness defects from the inoculum . This might have led to a reduction in the diversity of the mutant pool used for infection , compared to the original pool comprising of 360 , 000 transformants and may explain the fact that we identified only 6732 independent transposon insertion sites . Notwithstanding the reduced complexity of the input pool , we have identified novel fitness factors from this study . TFAST was previously developed in our laboratory to determine the transcription factor binding sites [17] and facilitated successful identification of PapX binding site in the flhDC promoter [40] . Here , TFAST was applied to determine the chromosomal location and the frequency of detection of a given transposon mutant . Potential fitness genes identified in this study could be an underestimate because genes pic and vat did not meet the threshold ( mean+2 standard deviations ) but were confirmed as fitness genes in the co-infection experiments . The EZ-Tn5 transposon used for random mutagenesis was not modified to incorporate promoter regions at either end; therefore , the transposon insertion mutations could exert polar effects on co-transcribed genes . Additionally , random events could result in the loss of a transposon mutant during infection and could result in misinterpretation as a fitness gene . In co-infection experiments , nine of the 11 ( 82% ) tested mutants revealed a fitness defect confirming the validity of our primary screen . Seven of 372 predicted transposon mutants with a fitness defect ( 1 . 9% ) were found in another study as essential genes in a laboratory strain of E . coli [18] . Studies on gene essentiality in E . coli have been conducted primarily on non-pathogenic , laboratory-adapted strains . Genomes of UPEC are usually larger than these laboratory strains . For instance , genome of UPEC CFT073 is ∼590 Kbp longer than the widely studied E . coli K-12 strain MG1655 [14] . Depending on the transposon insertion site and growth conditions , it is possible that transposon insertions could be tolerated in some essential genes . For instance , degS is designated as an essential gene in E . coli [18] . However , a degS mutant has been successfully constructed in E . coli CFT073 and used to demonstrate that DegS , likely by modulating members of Sigma E regulon , affects the fitness of UPEC during peritonitis as well as during UTI in a mouse model of infection [41] . Another possible explanation is the emergence of suppressor mutations that negate the effects of original mutation . It is also possible that these are artifacts due to sequence similarity to parts of other non-essential genes or gene duplication events . Potentially , some of the essential genes involved in non-structural components could be complemented by other transformants within the mutant pool . These genes constitute only a small fraction of all the fitness genes unraveled in this study . A previous study in ExPEC during systemic infection in a mammalian host , led to the identification of type 1 pilus; P fimbria; Hma and ChuA , heme receptors; TonB , iron uptake energy transducer; Ksl , K2 capsule biogenesis; and NanA , N-acetylneuraminate aldolase as fitness determinants [11] . This study has greatly expanded the potential bacteremia fitness determinants in UPEC and offers evidence for the role of nine of these novel fitness determinants in a murine model of systemic infection . Furthermore , 81 ( 33 . 5% ) of the candidate fitness genes are predicted to encode hypothetical proteins and constitute a unique resource that can be exploited to identify previously unknown fitness determinants . Biosynthetic mutants defective in either salmochelin or both salmochelin and enterobactin production revealed reduced fitness in the primary screen . Since these mutants retain the ability to utilize catecholate siderophores synthesized by other transformants , it is likely that the observed fitness defect emerges from iron uptake-independent roles . Recently , catecholate siderophore biosynthesis , but not uptake-alone , was demonstrated to mitigate the effects of oxidative stress in both Salmonella Typhimurium and E . coli [42] . It is , therefore , plausible that UPEC utilizes catecholate siderophore biosynthesis not only for canonical iron acquisition functions but also for protection against oxidative stress encountered during systemic infection . PNAG , an extracellular polysaccharide , has been associated with the virulence in a broad spectrum of bacterial pathogens , including Aggregatibacter actinomycetemcomitans [43] , Bordetella pertussis [44] , Staphylococcus aureus [45] , and Yersinia pestis [46] . Antibodies raised against S . aureus-derived PNAG confer passive protection against systemic infection with a clinical UPEC strain [47] . Here , we provide evidence that biosynthesis of PNAG is required for optimal fitness of UPEC during both UTI and systemic infection ( Fig . 4 ) . Type IV pili are filamentous organelles found at the bacterial surface that affect adherence and motility in several bacterial species , including enteropathogenic E . coli [48] . We found that the genes encoding predicted major and minor type IV pilins ( c2394-95 ) are critical for fitness during both bacteremia and UTIs . Although the mutant did not exhibit reduced adherence to MM55K cells , an immortalized cell line derived from the kidneys of AKR strain mice , the mutant revealed colonization defect in murine kidneys in a mouse model of ascending UTI . Differences in the expression of surface receptors on MM55K cells compared to those found within the nephrons of live CBA/J mice used for infection experiments could account for the discrepancy between adherence phenotypes observed for the type IV pilus mutant during in vitro and in vivo assays . Electron micrographs revealed reduced number of pili in the mutant compared to wild-type and complemented mutant strain . However , UPEC strain CFT073 produces multiple fimbria [29]; therefore , this observation must be verified with immunostaining to enable specificity . UPEC also harbors a locus similar to that encoding type IV pilus in E . coli K-12 and has been demonstrated to affect the fitness in mouse urinary tract [49] . Mutants in both type IV pilus loci exhibit fitness defects independent of each other and here we demonstrate that type IV pilus two is a novel fitness factor in UPEC . Oligopeptide uptake system gene oppA was previously shown to be critical for fitness of UPEC in the urinary tract [31] . Gene clusters involved in peptide uptake , opp and sap , were found to contribute to the fitness of UPEC during systemic invasion in the current study . The sapABCDF gene cluster contains homologs of genes involved in sensitivity to antimicrobial peptides in Salmonella enterica subspecies Typhimurium [50] and non-typeable Haemophilus influenzae [51] . Our data support a model in which the sap gene cluster , but neither oppA nor dppA , is required for optimal protection against polymyxin B and intracellular survival in murine macrophages ( Fig . 6 ) . Targets of Fur in E . coli MC4100 were detected using a macroarray and yddABpqqL was determined as a Fur-regulated gene cluster in E . coli [52] . A transposon mutant screen in E . coli strain CC118 , revealed that yddA and yddB are required for optimal growth in rich medium at 37°C [53] . However , UPEC CFT073 mutants lacking yddB or pqqL genes exhibited no difference in growth rate compared to wild-type strain in vitro ( Fig . S1 ) . yddA acts as a colonization factor in enterohemorrhagic E . coli O26:H− in a calf model of intestinal colonization [54] . Enhanced expression in urine ( Fig . 7C ) and the high degree of identity of YddB protein to ligand-gated outer membrane siderophore receptors allowed us to speculate that these genes could be involved in iron uptake . Although these genes are regulated by Fur ( Fig . S2B ) , they do not appear to be regulated by iron levels alone in CFT073 ( Fig . S2A ) . Cues , other than reduced iron levels , sensed by UPEC in human urine likely govern the regulation of yddABpqqL genes . pqqL from E . coli has been previously shown to complement pyrolloquinoline quinone ( PQQ ) biosynthetic genes pqqE and pqqF in Methylobacterium organophilum [55] . PQQ is a cofactor for quinoproteins , including glucose dehydrogenase in E . coli . It must be noted that E . coli is not capable of PQQ biosynthesis [55] . Studies are in progress to address whether this gene cluster is involved in uptake and processing of PQQ . We have identified Pic and Vat , autotransporter serine proteases , to be involved in fitness during bacteremia ( Fig . 3 ) . In E . coli CFT073 , pic was upregulated during UTI in a murine host and Pic exhibited serine protease activity in vitro [32] . On the contrary , Vat ( referred to as Tsh in [32] ) did not exhibit a detectable serine protease activity and both of these genes did not appear to affect fitness of UPEC during UTI . Key human leukocyte adhesion molecules such as CD43 , CD44 , CD45 and CD93 , are targeted by Pic resulting in deregulation of leukocyte migration and inflammation [56] . Therefore , it is possible that reduced fitness of Pic mutants during systemic infection could emerge from its role in modulating inflammatory response to systemic infection with UPEC . Shikimate is a critical intermediary molecule in chorismate biosynthetic pathway and chorismate is a precursor for the biosynthesis of aromatic amino acids , catecholate siderophores , folate , menaquinone and ubiquinone in bacteria [33] . Biosynthesis of aromatic amino acids has been associated with virulence in several bacterial species in various animal models of infection , including Neisseria meningitidis [57] , Proteus mirabilis [58] , Salmonella enterica subspecies enterica serovar Typhimurium [36] , and Staphylococcus aureus [59] . Taken together , these findings indicate that aromatic amino acids and other compounds derived from the chorismate pathway are critical for optimal fitness of multiple bacterial pathogens during infection . In summary , a combination of transposon mutagenesis and HT sequencing was used to determine genes in UPEC that contribute to fitness in a mouse model of systemic infection . The role of multiple candidate fitness genes was confirmed by independent experiments using a mouse model of infection and in vitro assays . Further characterization of the fitness genes unraveled in this study has the potential to identify targets for developing novel intervention strategies against bacteremia caused by UPEC .
All animal experiments were performed in accordance to the protocol ( 08999-3 ) approved by the University Committee on Use and Care of Animals at the University of Michigan . This protocol is in complete compliance with the guidelines for humane use and care of laboratory animals mandated by the National Institutes of Health . E . coli CFT073 , a prototypical uropathogenic strain that caused bacteremia of urinary tract origin [12] was used to generate a saturating Tn5 insertion mutant library . Strains and plasmids used in this study are listed in Table 2 . Bacterial strains were cultured in LB containing 0 . 05% NaCl , unless otherwise noted . Tn5 transformants were cultured in LB containing kanamycin ( 12 . 5 µg/ml ) . Lambda red recombineering was used to introduce specific mutations in strain CFT073 [60] . Genetically defined mutants used in this study were cultured in LB containing either kanamycin ( 25 µg/ml ) or chloramphenicol ( 20 µg/ml ) . Oligonucleotide primers used in this study are listed in Table S1 . Growth kinetics of the wild-type and mutant strains were determined using a Bioscreen C system ( Growth Curves USA ) . Overnight cultures were diluted 1∶100 in LB and incubated at 37°C . Optical density values were recorded at 600 nm , every 15 min , for 22 h and included three biological replicates , comprised of two technical replicates . Tn5 insertion mutants were generated in E . coli CFT073 using the EZ-Tn5 transposome kit ( Epicentre ) . Briefly , transposome complexes were electroporated into E . coli CFT073 and bacteria were allowed to recover in SOC broth for 50 min prior to plating on LB agar containing kanamycin using an automated plater ( Spiral Biotech ) . Plates were incubated overnight at 37°C and CFUs were enumerated using a Qcount colony counter ( Spiral Biotech ) . A total of 360 , 000 transformants were generated for this study and archived in pools of 1800 CFUs . The entire Tn5 mutant collection was passaged three times in LB for 16 h at 37°C and the resulting pool was used as the inoculum for experimental infections . CBA/J mice ( 6–7 week old , Harlan Laboratories ) were inoculated with 106 ( n = 5 ) , 107 ( n = 10 ) or 108 ( n = 5 ) CFU of CFT073 bacteria via tail vein . Mice were euthanized after 24 h and livers and spleens were harvested . Homogenates of these organs were plated on LB plates containing kanamycin and bacterial burden was determined . Mice were inoculated with 107 ( n = 20 ) CFU of CFT073::Tn5 mutants . After 24 h , mice were euthanized to collect spleens . Homogenates of spleens were plated in their entirety , as described above and the bacterial burden was calculated . Colonies from splenic cultures were harvested and pooled from all 20 mice before archiving bacterial pellets at −80°C . For co-infection experiments , wild-type and specific mutants , cultured overnight , were resuspended in PBS to yield 2×108 CFU/ml , containing equal number of wild-type and mutant CFUs . Inoculum ( 100 µl ) was administered via the tail vein and mice were euthanized 24 h pi . For the ascending UTI model , female mice were inoculated intravesically via a transurethral catheter with 50 µl of the inoculum containing 109 CFU/ml ( equal number of wild-type and mutant CFUs ) and animals were euthanized after 48 hpi . Homogenates of spleen and liver ( intravenous infection ) or urinary bladder and kidneys ( intravesical infection ) were plated on plain and antibiotic-containing plates . Both wild-type and mutant strains grow on LB plates , whereas only a mutant strain can grow on antibiotic containing LB plates . Plate counts were used to calculate the number of wild-type and mutant bacteria surviving in vivo . Competitive indices ( CI ) were calculated as the ratio of mutant over wild-type in tissues to the ratio of mutant over wild-type in the inoculum . Urine was collected from a group of 5 mice infected with CFT073 , periodically over 48 hours and immediately stabilized with RNAprotect ( Qiagen ) prior to RNA extraction . Genomic DNA was isolated from the inoculum used for infections ( input ) and from cultures derived from infected spleens ( output ) using DNeasy blood and tissue DNA extraction kit ( Qiagen ) . Genomic DNA ( 5 µg ) was sheared to yield ∼300 bp fragments ( Covaris ) . Illumina Truseq adapters were ligated to DNA fragments and used for Tn-specific amplification . A Tn-specific primer composed of the flowcell binding region of the Truseq adapter and Tn-specific region was used in conjunction with the Truseq adapter-specific primer to amplify transposon-chromosome insertion junctions ( Table S1 ) . Briefly , 25 ng of the TruSeq library was used as template for 30 cycles of amplification . Amplicons were further processed for Illumina sequencing ( cluster generation ) according to manufacturer's recommendations and sequenced using a Tn-specific primer ( Table S1 ) . Libraries from input and output samples were sequenced in two separate lanes of a single sequencing run in an Illumina HiSeq2000 sequencer . Library preparation and sequencing were performed at the University of Michigan DNA core facility . Reads from the input and output libraries , in FASTQ format , were aligned to the genome of E . coli CFT073 ( NCBI accession no . NC_004431 ) [14] using the short read aligner BOWTIE [16] . The alignment files , in SAM format , were then used in the TFAST [17] program to determine the number of reads that map to a given chromosomal location in the input and output libraries . To assess biofilm production , strains were cultured in tryptic soy broth containing 1% glucose in 96-well tissue culture-treated polystyrene plates for 24 h , at 37°C . Supernatants were aspirated and plates were washed three times with water and stained with 0 . 3% crystal violet solution for 10 min . Unbound crystal violet was removed by three additional washes with water . Biofilm-bound crystal violet was dissolved in 200 µl of 33% acetic acid and absorbance was measured at 590 nm ( μQuant , Bio Tek instruments , Inc . ) . This experiment was repeated at least three times , independently . The protocol described by Cerca et al . [47] was adapted . Cultures , incubated overnight in tryptic soy broth containing 1% glucose , were adjusted to an OD600 of 1 . 5 . Cell pellets from 1 ml samples were resuspended in 300 µL of 0 . 5M EDTA ( pH 8 . 0 ) and boiled for 5 min . Samples were centrifuged at 13 , 000 RPM for 6 min . Supernatants were treated with Proteinase K ( 2 µg/µL ) , heat inactivated and diluted 3-fold in Tris-buffered saline ( TBS; 20 mM Tris-HCl , 150 mM NaCl , pH 7 . 4 ) . Extracts ( 200 µl/sample ) were immobilized on nitrocellulose membranes and blocked with 5% skim milk in TBST ( TBS containing 0 . 1% Tween20 ) for 2 h . Blots were incubated for 2 h with an affinity-purified anti S . aureus PNAG antibody ( 1∶2000 ) raised in rabbits [61] . Horseradish peroxidase-conjugated secondary anti-rabbit IgG antibody ( 1∶20 , 000 ) was used in conjunction with ECL Plus enhanced chemiluminescence detection system ( GE Healthcare ) to determine the presence of PNAG . These experiments were repeated at least three times , independently . Agar ( 0 . 25% ) plates containing NaCl ( 0 . 5% ) and tryptone ( 1% ) were used to measure swimming motility . Ampicillin ( 100 µg/µl ) was added for plasmid maintenance , when required . Cultures were stab-inoculated and incubated at 30°C for 16 h . Diameters ( mm ) of swimming zone were determined . Four independent experiments were performed with at least two technical replicates . E . coli CFT073 ( pTopo ) and CFT073 ( pSS1 ) were cultured overnight in TSB with 1% glucose and ampicillin ( 100 µg/ml ) . Bacteria , washed and resuspended in PBS to an OD600 of 1 , were incubated in ECM protein coated plates ( Biocoat plates , Becton Dickinson ) for 2 h at 37°C . The number of bacteria in the inoculum and the number of bacteria that remain in the supernatant ( non-adherent ) were determined by plate counts and used to calculate percentage of adherent bacteria . Fold-change in adherence was calculated as the ratio of adherence percentages of CFT073 ( pSS1 ) over CFT073 ( pTopo ) . The following immortalized cell lines were used in adherence assays: human bladder epithelium , UM-UC-3 ( ATCC CRL-1749 ) ; murine kidney , MM55 . K ( ATCC CRL-6436 ) ; green monkey kidney , VERO ( ATCC CCL-81 ) ; and human embryonic kidney , HEK293 ( ATCC CRL-1573 ) . Cells were cultured to confluence in 24-well cell culture plates ( Corning ) in Dulbecco's Modified Eagle Medium supplemented with 10% fetal bovine serum , penicillin ( 100 U/ml ) , streptomycin ( 100 µg/ml ) and L-glutamine ( 0 . 3 mg/ml ) , referred to as DMEM-PSG , at 37°C in a humidified atmosphere with 5% CO2 . Epithelial cell cultures were washed once with PBS , and inoculated with a 250 µl suspension containing 1×108 CFU of E . coli CFT073 ( wild-type ) , c2394-95 mutant , or the complemented mutant in DMEM without antibiotics . Infected epithelial cells were incubated at 37°C with 5% CO2 for 30 min , and then washed three times with PBS . Epithelial cells , along with any adherent bacteria , were lifted by incubation in 1 ml sterile distilled water containing 5 mM EDTA . Colony counts were used to enumerate CFUs in the inocula and cell-associated bacteria . Adherence was expressed as cell-associated CFU/initial CFU . Adherence of each mutant was normalized to the wild-type control; assays were performed in triplicate , each with three technical replicates . E . coli CFT073 ( wild-type ) , c2394-95 mutant , and the complemented mutant were cultured for 3 h at 37°C . Samples were swirled gently and 10 µl of the culture was dropped onto formvar carbon support film on TEM specimen grids ( Electron Microscopy Sciences ) . Grids were incubated at room temperature for 5 min , and excess medium was wicked off with filter paper . Grids were washed once with 10 µl of deionized water , and then stained for 2 min with 10 µl of 1% phosphotungstic acid ( pH 6 . 8 ) . Excess stain was removed; grids were washed immediately with deionized water and dried . Grids were visualized using a Philips CM-100 transmission electron microscope . Overnight cultures , diluted 1∶100 in fresh medium , were incubated at 37°C for 2 h . Cultures were exposed to polymyxin B ( 4 µg/mL ) for 30 min . Colony counts were determined by plating and percent survival upon exposure to Polymyxin B was calculated as the ratio of CFU in the treated samples to untreated controls . The experiment was repeated at least three times , independently . RAW 264 . 7 cells were cultured in RPMI1640-PSG supplemented with 10% fetal bovine serum and seeded in 24-well tissue culture plates . CFT073 and mutant strains , cultured overnight in LB , were washed in PBS and resuspended in RPMI1640 to an OD600 of 0 . 004 . Cells were washed with PBS , overlaid with the inoculum at an MOI of 10 and incubated for 30 min . Two identical plates , for 0 h ( T0 ) and 2 h ( T2 ) , were set up during each experiment . Supernatants were aspirated and cells were washed three times with PBS . Fresh RPMI1640 supplemented with gentamicin ( 200 µg/ml ) was added and incubation was continued . The T0 plate was removed at 15 min post gentamicin addition and cells were lysed with saponin ( 10% , w/v in water ) . Lysates were plated to determine the number of intracellular bacteria . T2 plates were processed as described here at 2 h post gentamicin addition . Percent killing was calculated as the percent of intracellular bacteria that were killed within RAW 264 . 7 cells . Killing percentages of mutants were compared to that of wild-type bacteria to determine the comparative fitness of a given mutant during survival within RAW 264 . 7 cells . The experiment was repeated three times with three technical replicates per strain . RNA was extracted from E . coli CFT073 cultured in LB or in filter sterilized human urine to mid-exponential phase or from cells harvested from urine of mice infected with wild-type strain using RNeasy mini kit ( Qiagen ) . Contaminating DNA was removed by incubation with DNase ( Turbo DNA-free , Ambion ) and reverse transcribed using SS RT III ( Invitrogen ) . To determine co-transcription , cDNA , genomic DNA and RNA samples were used as templates in standard PCR reaction ( primers listed in Table S1 ) . The entire experiment was repeated twice , independently . Overnight cultures of E . coli CFT073 were diluted 1∶100 in LB or LB with 300 µM dipyridyl ( Sigma ) or LB with 10 µM ferric chloride and incubated for 2 h at 37°C . RNA extraction and cDNA synthesis were performed as described above . Expression of yddA , yddB and pqqL transcripts was determined by qPCR using SYBR green chemistry ( Agilent Technologies ) in a Stratagene Mx3000P thermal cycler ( Stratagene ) . Transcripts were normalized to gapA mRNA ( Table S1 ) and relative quantification was performed using CFT073 cultured in LB as the calibrator . Overnight cultures of E . coli CFT073 and Δfur mutant were diluted 1∶100 in LB and incubated for 2 h . RNA extraction , cDNA synthesis and qPCR were performed as described above . Relative quantification was performed using CFT073 as the calibrator . Both qPCR experiments were repeated three times with two technical replicates . DNase-treated RNA from mouse UTI urine was used to determine levels of pgaA transcript by qPCR , essentially as described above . Mid-exponential phase cells from LB were used as calibrator and all samples were normalized to gapA levels . Statistical tests were performed using Prism 5 ( www . graphpad . com ) . Data were analyzed using the following tests: co-infection experiments , Wilcoxon signed-rank test against a theoretical median of 0; biofilm assay , swimming motility assay and adherence to epithelial cells and ECM proteins , two-way ANOVA with Bonferroni's multiple comparison test; polymyxin B resistance assay and intracellular survival assay , student's t test . P<0 . 05 was considered as a statistically significant difference . Error bars in the figures represent standard error of the mean . The raw reads can be accessed under the accession number , SRP027190 in NCBI SRA . | Uropathogenic E . coli is a major cause of bacterial bloodstream infections in humans . Dissemination of E . coli into the bloodstream during urinary tract infections may lead to potentially fatal complications . This pathogen is becoming increasingly resistant to currently used antibiotics . To develop additional tools to treat such infections , a thorough understanding of the mechanism of pathogenesis is required . Here , we report major progress towards that goal by identifying bacterial genes that are critical for the ability of this pathogen to cause bloodstream infections using a mouse model of infection . This study sheds light on the conditions encountered by E . coli during systemic infection . Further research on the genes identified in this study may reveal bacterial targets that can be used to develop novel therapeutics against bloodstream infections caused by E . coli . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | Genome-Wide Detection of Fitness Genes in Uropathogenic Escherichia coli during Systemic Infection |
Meiotic recombination normally takes place between allelic sequences on homologs . This process can also occur between non-allelic homologous sequences . Such ectopic interaction events can lead to chromosome rearrangements and are normally avoided . However , much remains unknown about how these ectopic interaction events are sensed and eliminated . In this study , using a screen in rice , we characterized a homolog of HUS1 and explored its function in meiotic recombination . In Oshus1 mutants , in conjunction with nearly normal homologous pairing and synapsis , vigorous , aberrant ectopic interactions occurred between nonhomologous chromosomes , leading to multivalent formation and subsequent chromosome fragmentation . These ectopic interactions relied on programed meiotic double strand breaks and were formed in a manner independent of the OsMER3-mediated interference-sensitive crossover pathway . Although early homologous recombination events occurred normally , the number of interference-sensitive crossovers was reduced in the absence of OsHUS1 . Together , our results indicate that OsHUS1 might be involved in regulating ectopic interactions during meiosis , probably by forming the canonical RAD9-RAD1-HUS1 ( 9-1-1 ) complex .
Meiosis is a highly dynamic process in which chromosomes undergo dramatic structural changes and movements [1] , [2] . During the course of meiosis , intimate interactions develop between homologous chromosomes . Among these interactions , homologous recombination ( HR ) and pairing are the core events that occur during the production of functional gametes [3] , [4] . Meiotic recombination is a powerful determinant that creates genetic diversity and provides mechanical stability for the accurate separation of homologous chromosomes . Therefore , meiotic recombination has a strong bias towards homologous chromosomes rather than sister chromatids and is mediated by a complex mechanism [5] , [6] . After DNA replication in the premeiotic S phase , a proteinaceous axis is assembled between two chromatids . Homologous recombination then occurs along the chromosomes , beginning with the formation of programmed double strand breaks ( DSBs ) . In conjunction with the initiation of recombination , homologous chromosomes begin to align in pairs . Studies have shown that for most species , homologous pairing depends on homologous recombination [7] . However , recombination not only occurs between allelic DNA sequences on homologs , but it also frequently occurs between dispersed non-allelic DNA segments that share high sequence similarity [8] . The latter recombination pattern is usually referred to as ectopic recombination ( ER , also known as non-allelic homologous recombination ) . As eukaryotic genomes are rich in repeated DNA sequences , ER can produce chromosomal rearrangements , which in humans result in numerous genomic disorders [9] , [10] . Despite the adverse impact of ER on genome integrity , ER occurs relatively frequently; in budding yeast , the frequency of ER is roughly on par with that of allelic recombination [11] , [12] . To avoid the deleterious consequences of ER , cells have evolved multiple strategies to suppress ER formation [13] . One strategy is preventing DSB formation in or near DNA repeats . In budding yeast ( Saccharomyces cerevisiae ) , suppression of DSBs in rDNA repeats depends strongly on silent information regulator 2 ( Sir2 ) , which encodes a histone deacetylase that promotes the formation of a closed , compact chromatin structure in the rDNA and other regions . Sir2 may suppress DSBs in rDNA in part through the formation of a nucleosomal conformation that is not permissive for SPO11 activity [14] . The second strategy is preventing the use of non-allelic homologous templates for recombination and/or favoring the use of allelic templates . In budding yeast , homologous alignment and synapsis restrict the ability of ectopically located sequences to find each other and recombine [15] . There are also reports on the competition between normal allelic recombination and ER [16] . As both mechanisms involve preventing ectopic interaction intermediate formation , we classified these events as ectopic interaction preventing mechanisms in this study . The frequent occurrence of ER in yeast suggests that these mechanisms cannot totally prevent all ER initiation [17]–[19] . Once a non-allelic partner is used as the template and ER intermediates are built , a mechanism monitors and resolves those ER intermediates into non-crossovers . This ER-eliminating mechanism can be classified as a surveillance mechanism . Studies in S . cerevisiae have shown that the mismatch repair proteins Pms1 and Msh2 are likely to be involved in this mechanism , although direct evidence for this is still lacking . Several DNA helicases , including Sgs1 in yeast and BLM in humans , may also possess anti-crossover activities that are potentially involved in preventing deleterious outcomes of meiotic ER [13] . Rad9 , Hus1 , and Rad1 ( in S . cerevisiae: Ddc1 , Mec3 , and Rad17 ) interact in a heterotrimeric complex ( dubbed the 9-1-1 complex ) , which resembles a PCNA-like sliding clamp [20] , [21] . In response to genotoxic damage , the toroidal 9-1-1 complex is loaded around damage sites , collaborating with ATM and ATR to carry out its best known function of activating the DNA damage checkpoint [22] . In addition , studies have revealed functional interactions between 9-1-1 and multiple partners , most notably translation polymerases , base excision repair enzymes , and mismatch repair factors [23] , [24] . This evidence implies that 9-1-1 also plays a direct role in DNA repair . In addition to its role in the conventional somatic DNA damage response , 9-1-1 also plays roles in meiosis . In S . cerevisiae , it is evident that Ddc1 colocalizes with Rad51 on meiotic chromosomes and is required for the pachytene checkpoint [25] . The rad17 mutants exhibit aberrant synapsis and increased rates of ER during meiosis [26] . In mouse , RAD1 was found to be associated with both synapsed and unsynapsed chromosomes during prophase I [27] . Recently , the HUS1 homologs in Drosophila and mouse were reported to be essential for meiotic DSB repair [28] , [29] . The function of the 9-1-1 complex in suppressing meiotic ER was first suggested in yeast [26] . However , this function has not been reported in higher organisms , likely due to a lack of direct cytological evidence . In this study , we aimed to isolate genes that are involved in ectopic interaction suppression . Several mutants showing normal homologous pairing at pachytene and the appearance of ectopic interactions at diakinesis were isolated . Among these , two allelic mutants were characterized in detail . These mutants were found to be mutated in the functional homolog of fission yeast and mammalian HUS1 . In the Oshus1 mutants , meiotic homologous pairing took place normally during prophase I , while nonhomologous chromosomes interacted vigorously as well . Multivalents were frequently found to be arranged on the equatorial plate at metaphase I . Chromosome bridges and fragments occurred at anaphase I and telophase I , rendering the mutants completely sterile . These results suggest that OsHUS1 might specifically function in sensing and removing aberrant associations between non-allelic sequences during meiosis , probably via the 9-1-1 complex .
Among our rice sterile mutant libraries , 16 lines with phenotypes meeting the criteria mentioned above were isolated . One of the mutant lines , S7678 , which was derived from Nipponbare ( a japonica cultivar ) tissue culture , was selected for further study . Based on information about its mutation ( see below ) , the mutant was named Oshus1-1 . The Oshus1-1 plants did not exhibit defects in vegetative growth under natural growth conditions , except for total male sterility ( Figure S1 ) . Fertile plants and sterile plants from the progeny of Oshus1-1+/− produced a 3∶1 segregation ratio ( fertile , 214; sterile , 66 ) , which established this mutant as a single recessive mutant ( χ2 = 0 . 30; P>0 . 05 ) . When we pollinated the mutant flowers with wild-type pollen , the mutant did not set seed , indicating that female fertility is also affected in this mutant . We isolated OsHUS1 by map-based cloning . A mapping population was constructed by crossing Oshus1-1+/− plants to Nanjing 11 ( an indica cultivar ) plants . The mutant gene locus was mapped to a physical region of approximately 100 kb on the long arm of chromosome 4 . According to information obtained from the public database ( Rice Genome Annotation Project , http://rice . plantbiology . msu . edu ) , we sequenced several genes in this region . As a result , a point mutation ( A to T ) was found in the gene Os04g44620 , which introduced a stop codon ( AAG to TAG ) in the second exon . We named the mutant Oshus1-1 based on the homology of the protein sequence . Next , we isolated another mutant from Huanghuazhan ( an indica cultivar ) , which has the same phenotype as that of Oshus1-1 . Using map-based cloning and DNA sequencing , we found that this mutant carries a ten-nucleotide deletion in the fourth exon of OsHUS1 , causing frame shift and premature stop codon formation . We named this allele Oshus1-2 . The chromosome behavior in Oshus1-2 meiocytes was the same as Oshus1-1 ( Figure S2A ) . We generated a gene-specific p35S OsHUS1-RNAi construct and used it to transform Yandao 8 ( a japonica cultivar ) rice . Most OsHUS1-RNAi lines showed a severe reduction in fertility ( 93% , n = 30 ) , and the chromosome behavior in the male meiocytes of these lines mirrored that of Oshus1 ( Figure S2B ) . From these results , we conclude that the mutation in the OsHUS1 gene led to the sterility phenotype . There are three full-length cDNA sequences of Os04g44620 published in the Rice Genome Annotation Project website , including AK107445 , AK101159 , and AK064120 . Using RT-PCR and RACE ( rapid amplification of cDNA ends ) on young panicles , we found that AK064120 is the correct sequence for this gene . Alignment of the cDNA sequence with the genomic sequence revealed that OsHUS1 is composed of six exons and five introns ( Figure S3 ) . The open reading frame of OsHUS1 has a length of 981 bp , encoding a 326 amino-acid peptide . Using BLASTp , we found that OsHUS1 shares some similarity ( approximately 25% identity and 45% similarity ) with the HUS1 protein in mammals and fission yeast ( Figure S4 ) . Reciprocal BLAST searches further confirmed that the isolated protein is the closest relative of HUS1 in rice ( Figure S5 ) . As shown above , there were several defects during meiosis in Oshus1 . We then examined the spatial and temporal expression patterns of OsHUS1 . Using quantitative RT-PCR , we found that OsHUS1 could be detected as early as the seedling stage . In adult-stage rice , OsHUS1 was expressed not only in young panicles but also in vegetative organs such as leaves , roots , and internodes ( Figure S6 ) , with the highest expression observed in leaf blades . The behavior of meiotic chromosomes was revealed by 4′6-diamidino-2-phenylindole ( DAPI ) staining . In wild-type pollen mother cells ( PMCs ) , meiosis began with chromosome condensation and the appearance of chromosomes as thin , thread-like structures at leptotene ( Figure 1A ) . As zygotene progressing , homologous chromosomes underwent pairing and synapsis ( Figure 1B ) . During pachytene , homologous pairing culminated in the formation of synaptonemal complexes ( SCs; Figure 1C ) . After the disassembly of the SC at diplotene , the resulting 12 bivalents were further condensed , revealing the presence of chiasmata at diakinesis ( Figure 1D ) . At metaphase I , the bivalents were aligned in the middle of the cell ( Figure 1E ) . Homologous chromosomes separated and migrated toward opposite poles at anaphase I and telophase I ( Figure 1F and 1G ) , generating dyads at the end of meiosis I ( Figure 1H ) . Then , the dyads underwent meiosis II and finally produced tetrads ( Figure 1I ) . In Oshus1-1 PMCs , the chromosome behaviors appeared the same as those observed in wild-type from leptotene to zygotene ( Figure 2A and 2B ) . However , anomalies began to be manifested at early pachytene . At first glance , almost all homologous chromosomes aligned well . However , upon careful examination , we found that some regions of the chromosomes could not complete close alignment perfectly and exhibited “bubble-like” structures ( Figure 2C ) . During middle pachytene stage , associations between nonhomologous chromosome were observed in all PMCs ( n = 252 ) , which caused the chromosomes to stick to each other ( Figure 2D ) . At late pachytene , this type of association became more prominent ( Figure 2E–H ) . At diakinesis , multivalents were detected in all PMCs ( n = 521 ) . These multivalents ranged in size from associations of four chromosomes to the extreme case of 24 chromosomes ( Figure 3A , 3E ) ; the average number of bivalents per cell was only 1 . 6 . At metaphase I , multivalents and bivalents were located on the equatorial plate due to the drag force exerted on centromeres by spindle fibers ( Figure 3B , 3F ) . During anaphase I , the multivalents and bivalents fell apart , and extensive chromosome bridges and fragments were observed ( Figure 3C , 3G ) . At telophase I , two masses of chromosomes arrived at opposite poles of the nuclei , and several distinct dot-like chromosome fragments still remained on the equatorial plate ( Figure 3D , 3H ) . In a few cells ( 4% , n = 881 ) , up to 10 or 11 homolog pairs could be individualized at diakinesis and metaphase I . We also found some cells with a small amount of chromosome bridges and fragments at anaphase I and telophase I ( Figure 3E , F ) . These types of defects were maintained during meiosis II , and no normal tetrad was produced . By performing DAPI staining of pachytene chromosomes , we found that in Oshus1-1 , homologous chromosomes could pair normally . To validate whether normal SCs were affected by the mutation of OsHUS1 , we performed immunofluorescent examination using antibodies against ZEP1 in Oshus1-1 PMCs . ZEP1 is the transverse filament protein of SC in rice and hence , a perfect tool to mark the course of synapsis [30] . In leptotenic and zygotenic Oshus1-1 PMCs , the ZEP1 patterns appeared as dots and short fragments , which were identical to those of the wild-type ( Figure 4A and 4B ) . During pachytene , only approximately 10% ( n = 300 ) of the meiocytes showed full-length ZEP1 signals along the homologous chromosomes ( Figure 4C ) . In the remaining 90% of meiocytes , linear ZEP1 signals extended and could be detected along almost the entire chromosomes , with the exception of a few discontinuities/gaps , some of which exhibited the “bubble-like” structures mentioned above ( Figure 4D and 4E ) . The discontinuities/gaps of ZEP1 signals indicate that the SC integrity might be slightly affected by the mutation of OsHUS1 . Many incidents during meiosis are believed to be interdependent , e . g . , pairing is recombination-dependent in mammals and higher plants . The nearly normal SC assembly observed in Oshus1 meiocytes is reminiscent of the proper loading of important factors involved in SC assembling and homologous recombination . To further verify the relationship between OsHUS1 and several other meiotic recombination factors , immunodetection was carried out in Oshus1-1 using antibodies against PAIR3 , PAIR2 , OsZIP4 , OsMER3 , and HEI10 . PAIR2 is the rice homolog of yeast HOP1 and Arabidopsis ASY1 , which associates with unpaired chromosome axes at early meiosis I . PAIR3 is also an axis-associated protein that can bind both unpaired chromosomes and paired chromosomes . Both PAIR2 and PAIR3 are usually utilized to mark the meiotic chromosome axis , and they also play fundamental roles in the recombination process [31]–[33] . OsMER3 and OsZIP4 are members of the ZMM protein family and are essential for early meiotic HR in rice [34] , [35] . In the Oshus1-1 mutant , PAIR2 appeared as foci at leptotene and associated with the chromosome axis as linear signals at early zygotene ( Figure 5A ) . PAIR3 signals were first observed as dots at early leptotene and then elongated gradually along the entire lengths of the chromosomes during zygotene ( Figure 5B ) . The appearance of OsMER3 and OsZIP4 commenced at early leptotene , and the number of OsMER3 foci ( average 257±15 , n = 44 , range 221–281 ) and OsZIP4 foci ( average 299±22 , n = 35 , range 289–328 ) reached its peak at early zygotene ( Figure 5C and 5D ) ; similar results were obtained in the wild-type . At pachytene , both OsMER3 and OsZIP4 decreased rapidly and no signals were found in the later stages in the wild-type and Oshus1-1 . The normal loading patterns of these four meiotic factors showed that early HR in Oshus1-1 is not disturbed . Previous studies suggest that the interference-sensitive pathway accounts for most of the crossovers ( COs ) in rice [34]–[36] . We thus wanted to know whether interference-sensitive COs were affected by the mutation of OsHUS1 . The HEI10 prominent foci correspond to the interference-sensitive CO sites in rice [36] . We counted the number of HEI10 foci ( average 16 . 9±1 . 9 , n = 17 , range 13–20 ) in Oshus1-1 ( Figure 5E ) and compared that with the corresponding data for the wild-type ( average 24 . 5±1 . 8 , n = 30 , range 22–28 ) . We found that the mean number of HEI10 bright foci of Oshus1-1 was significantly reduced compared with that of the wild-type ( t[45] = 13 . 8 , P<0 . 01 ) . Therefore , the number of interference-sensitive COs is reduced in Oshus1-1 due to the loss of OsHUS1 . Meiotic recombination is initiated by the formation of DSBs , which is catalyzed by SPO11 proteins; these proteins have been identified in budding yeast , Arabidopsis , and animals [37] . However , to date , no spo11 mutants have been isolated in rice [38] . Recently , three new proteins that are also implicated in DSB formation were reported in Arabidopsis , i . e . , PRD1 , PRD2 , and PRD3 [39] , [40] . Among these , PRD3 is thought to be the homolog of rice PAIR1 . Furthermore , the phenotype of the pair1 mutant ( asynaptic , with no bivalent formation ) is reminiscent of the phenotype observed in a mutant lacking DSBs [41] . We isolated an asynaptic mutant ( Figure 6A–D ) , and it was proven to be a new allele of pair1 . Then , pair1 Oshus1-1 double mutants were generated using this new pair1 allele . The double mutants showed a typical pair1 phenotype , i . e . , an absence of bivalents and lack of chromosome fragments at anaphase I ( Figure 6E–H ) . Therefore , ectopic interactions , as well as chromosome fragmentations in Oshus1-1 , require the formation of DSBs . To learn whether OsHUS1 is involved in DSB repair pathway in rice meiosis , we generated Osrad51c Oshus1-1 double mutants . OsRAD51C , like its functional homolog AtRAD51C , is essential for meiotic DSB repair [42]–[44] . In the Osrad51c mutant , homologous pairing and synapsis were defective at zygotene and pachytene , and univalents were observed at diakinesis and metaphase I ( Figure 6I–K ) . In anaphase I , all of the univalents broke into fragments without any chromosome associations and scattered randomly in the nucleus ( Figure 6L ) . These defects are consistent with the role of Osrad51c in meiotic DSB repair . In the Osrad51c Oshus1-1 double mutant , a cumulative effect of the two single mutations was detected; homologous pairing was disrupted , and ectopic chromosome associations were detected in all meiocytes observed . ( Figure 6M–O; n = 322 ) . At anaphase I , extensive chromosome fragments were also produced ( Figure 6P ) . Therefore , the occurrence of ectopic interactions in Osrad51c Oshus1-1 suggests that ectopic interactions between nonhomologous chromosomes do not require OsRad51C . OsCOM1 functions both in promoting homologous recombination and in resolving chromosome entanglements [45] . In the Oscom1 mutant , both homologous pairing and synapsis were abolished at pachytene ( Figure 6Q ) , and aberrant nonhomologous associations were detected . From diakinesis to metaphase I , the most obvious phenotype was an entangled chromosome mass ( Figure 6R , S ) . At anaphase I , chromosome fragments were generated ( Figure 6T ) . We also generated Oscom1 Oshus1-1 double mutants . The phenotype of the Oscom1 Oshus1-1 double mutant could not be distinguished from that of the Oscom1 single mutant ( Figure 6U–X ) , suggesting that OsHUS1 might function after OsCOM1 during meiosis . Of course , we cannot exclude the possibility that the ectopic interaction phenotype of Oshus1 might be hidden by the severe chromosome entanglement of Oscom1 . Since most COs in rice are derived from the interference-sensitive pathway , we set out to study the relationship between ectopic interactions and interference-sensitive COs . To this aim , the Osmer3 Oshus1-1 double mutant was generated , and its chromosome behaviors were investigated . In Osmer3 , fully aligned chromosomes were detected during pachytene ( Figure 7A ) , indicating the homologous pairing is not affected by the mutation of OsMER3 . However , during diakinesis and metaphase I , the mutant cells showed a mixture of both univalent and bivalent chromosomes ( Figure 7B , C ) . In anaphase I , the bivalents separated normally but the scattered univalents segregated randomly ( Figure 7D ) . Intriguingly , in the Osmer3 Oshus1-1 mutant , homologous pairing was not observed at pachytene stage ( Figure 7E ) . FISH experiments further confirmed that homologous pairing was disrupted in Osmer3 Oshus1-1 meiocytes ( n = 101 , Figure S7 ) . In diakinesis and metaphase I , both multivalents with ectopic interactions and univalents were detected in all meiocytes ( n = 122 , Figure 7F , G ) . The multivalents contained an average of 7 . 0 associated chromosomes ( ranging from 2 to 22 ) ; the average number of univalents per cell was 8 . 2 ( ranging from 0 to 16 ) . At anaphase I , both univalents and multivalents were pulled toward two poles of the nucleus . Additionally , chromosome bridges and fragments were also found at this stage ( Figure 7H ) . These results suggest that ectopic interactions in Oshus1 arise independently from the OsMER3-mediated pathway . To determine whether the defects in Oshus1 are affected by synapsis , we also generated the zep1 Oshus1-1 double mutant . In the zep1 mutant , synapsis was totally disrupted , but 12 bivalents were present at metaphase I and segregated normally at anaphase I ( Figure 7I–L ) . In the zep1 Oshus1-1 double mutant , homologous chromosomes aligned along the entire length of the chromosome , but the SC was not assembled ( Figure 7M ) . However , ectopic interactions were still clearly observed in all meiocytes ( n = 298 , Figure 7N–P ) . These results indicate that ectopic interactions are likely independent of synapsis in Oshus1-1 . To further elucidate the role of OsHUS1 in meiosis , we prepared polyclonal antibodies in mice against the entire length of recombinant , His-tagged OsHUS1 . Using antibodies against OsREC8 and OsHUS1 , we performed dual immunofluorescence staining in rice PMCs . OsREC8 , the cohesin protein in rice , was used to indicate the meiotic chromosome axes in this study [34] , [46] . During leptotene , OsHUS1 proteins appeared as discrete foci in the nuclei and were loaded on the chromosome axes , as indicated by their full colocalization with OsREC8 ( Figure 8A ) . The intensity of OsHUS1 then reached its peak at early zygotene , but this protein still appeared as foci rather than short lines ( Figure 8B ) . At late zygotene , the number of OsHUS1 foci decreased , and many of them fell off the chromosomes ( Figure 8C ) . At pachytene , the OsHUS1 immunostaining signal was completely absent in the nuclei ( Figure 8D ) . No OsHUS1 signal was observed in male meiocytes of Oshus1-1 , which confirmed the specificity of the OsHUS1 antibody ( Figure 8E ) . To further investigate the function of OsHUS1 protein , the immunolocalization pattern of OsHUS1 was investigated in Osmer3 , zep1 , and pair1 mutants . The localization pattern of OsHUS1 was not obviously affected in Osmer3 or zep1 ( Figure S8A , B ) . This result is consistent with the observation that no ectopic interaction was found in either of the mutants . On the contrary , in the pair1 mutant , we failed to detect any OsHUS1 signals ( Figure S8C ) , implying that the function of OsHUS1 depends on the formation of DSBs . In yeast and mammals , HUS1 protein is implicated in various DNA damage response pathways [47]–[51] . In rice , OsHUS1 has the highest expression abundance in leaves , suggesting that this protein , like its counterparts in yeast and mammals , is potentially involved in the mitotic DNA damage response . To address this possibility , we tested whether Oshus1 plants showed higher sensitivity to mitomycin C ( MMC ) , a DNA cross-link agent , than wild-type plants . Surface-sterilized seeds from wild-type and Oshus1-1+/− plants were sown on solid 1/2 MS medium containing 0 or 20 µg/ml MMC . When planted on medium lacking MMC , the development of wild-type seedlings was identical to that of progeny derived from an Oshus1-1+/− plant . However , when treated with MMC , the development of wild-type seedlings was only slightly suppressed , while approximately one-quarter of the progeny derived from the Oshus1-1+/− plants showed severe growth retardation ( Figure 9 ) . Using a PCR genotyping assay , we determined that all of the severely growth-retarded seedlings were Oshus1-1−/− ( n = 20 ) . These data demonstrate that Oshus1-1 rice is hypersensitive to MMC , indicating that OsHUS1 plays an important role in somatic DNA damage repair .
HUS1 is thought to form a PCNA-like complex with its two partners , RAD9 and RAD1 [23] . HUS1 has been intensively investigated in yeast and mammals , with studies primarily focusing on the mitotic DNA damage response . A mutation in MEC3 ( the HUS1 counterpart in budding yeast ) results in delayed entry into the S phase and slow DNA replication in response to DNA damage-inducing agents [52] . Fission yeast lacking HUS1 also fails to arrest the cell cycle after DNA damage or the blocking of DNA synthesis [53] . Targeted disruption of mouse HUS1 causes embryonic lethality due to the accumulation of chromosome breaks [49] . In this study , we found that rice hus1 seedlings were hypersensitive to the genotoxin MMC , suggesting that OsHUS1 has a conserved function in somatic DNA repair . Expression data for OsHUS1 show high accumulation of its transcript in somatic tissues , which further supports the somatic role of this protein . These findings are also in agreement with the hypothesis that OsHUS1 in rice is the functional homolog of fungal and animal HUS1 . By performing a BLASTp search , we found that the homologs of S . pombe RAD9 and RAD1 also exist in rice . In addition , RAD9 is also involved in the regulation of DNA damage repair in the model plant Arabidopsis [54] . Therefore , it is highly possible that OsHUS1 in rice , like its yeast and animal counterparts , also participates in somatic DNA repair responses by forming the 9-1-1 complex . Studies in yeast and humans have revealed parallels between meiotic ER and allelic recombination , such as the observation that both processes occur during prophase I and are initiated by programmed DSBs . ER also results in crossover formation , which can affect genome stability during gametogenesis [12] , [13] , [55] . Therefore , ER should be inhibited , and/or its intermediates must be quickly eliminated , to ensure accurate homolog segregation during meiosis . The function of the 9-1-1 complex in suppressing ER was first suggested in yeast [26] . However , to our knowledge , this function has not been reported in higher organisms , likely due to the lack of cytological evidence . Here , in Oshus1 meiocytes , we noticed that at late pachytene , one homolog pair frequently adhered or fused to another homolog pair at several sites , forming cross-like shapes . At the pachytene to diplotene transition ( in which homologous pairs began to separate partially due to SC disassembly ) , the associations became more pronounced . The most remarkable defects observed in Oshus1 meiosis were multivalents at metaphase I and subsequent chromosome fragmentation . The chromosome behaviors observed in the pair1 Oshus1-1 double mutant indicate that ectopic interactions rely on meiotic DSBs in Oshus1-1 , which supports the notion that ectopic and allelic interactions share a common mechanism [56] . DSB formation is essential for homologous chromosome pairing in meiosis [3] . Here , although strong ectopic interactions occurred in Oshus1 , homologous pairing took place normally . The nearly perfect ZEP1 signals along the entire lengths of chromosomes at pachytene indicated that synapsis was not severely disturbed in Oshus1 . In addition , OsZIP4 and OsMER3 localized normally in Oshus1 . It is likely that the early ectopic intermediate-preventing system may function well , and excessive ectopic interaction initiations are prevented in a timely manner in Oshus1 . Intriguingly , unlike the Oshus1 and Osmer3 single mutants , the Oshus1 Osmer3 double mutant exhibited disrupted homologous pairing . In light of the competition between allelic and ectopic recombination [57] and the important roles they play during homologous pairing [3] , it is attractive to consider that the increase in ectopic interactions and the decrease in allelic associations reduce the chance of homolog recognition and subsequent homolog alignment . Since homolog alignment mainly occurs at zygotene stage , it is reasonable to postulate that ectopic interactions initiate during or prior to zygotene . This hypothesis is consistent with the view that in yeast , ectopic recombination occurs concurrently with allelic recombination during meiosis [13] . Studies in budding yeast have revealed that ER occurs frequently during meiosis [11] , [12] . However , it remains unknown whether ER also occurs frequently during plant meiosis . Here , we observed that all meiocytes showed the presence of multivalents in Oshus1 . We therefore propose that in wild-type meiocytes , early ectopic interactions , accompanied by allelic interactions , may inevitably occur during homolog searching and homolog recognition . Once homolog recognition is accomplished , those ectopic interaction intermediates might be quickly detected and resolved by the surveillance mechanism . OsHUS1 is likely to be an important component of the surveillance mechanism that specifically eliminates ectopic interaction intermediates during meiosis . In budding yeast , a physical assay revealed that levels of ER increase from 1% in wild-type to 3–5% in rad17 , rad24 , and mec1-1 single mutants . HR is also reduced approximately two-fold in these mutants , from 25–30% in wild-type to 15% in rad17 , rad24 , and mec1-1 . These data indicate that the increase in ER does not quantitatively account for the decrease in HR . Therefore , ER and HR likely occur via different pathways [26] . Here , we demonstrated that the loss of OsMER3 function did not affect ectopic interactions ( through characterization of Osmer3 Oshus1-1 ) , implying that these ectopic interactions do not arise from the interference-sensitive crossover formation pathway . In this study , we also observed that the average number of bright HEI10 foci was reduced in the Oshus1-1 mutant , showing that the number of interference-sensitive COs was reduced in the absence of OsHUS1 . Thus , the similar alterations in ectopic and allelic interactions between yeast and rice imply that the function of HUS1 may be conserved among different organisms . Interference plays a role in both controlling and constraining the final distribution of COs . Although the mechanism underlying these processes remains unclear , it has been postulated that spreading interference signals are transmitted along the length of the chromosome axes [58] . Therefore , one possible explanation for the decrease in interference-sensitive CO number is that the spreading interference signals may also be transmitted through associated nonhomologous chromosome axes in Oshus1 . Alternatively , it is possible that partial allelic interactions are redirected into ectopic interactions or resolved toward sister chromatids in the absence of OsHUS1 . Studies in yeast and mammals have shown that the 9-1-1 complex is involved in multiple DNA repair courses by binding to numerous partners , including base excision repair proteins and mismatch repair proteins [23] . Among these , the mismatch repair protein MSH2 is postulated to be involved in the intermediate elimination of ER [13] . In yeast and humans , MSH2-MSH6 heterodimer ( MutSα ) and MSH2-MSH3 heterodimer ( MutSβ ) are mismatch recognition factors that function in the mismatch repair pathway . Recent studies have revealed that each subunit of the 9-1-1 complex can interact with both the MSH2/MSH3 and MSH2/MSH6 complexes . In addition , the 9-1-1 complex can also stimulate the DNA binding activity of MutSα [59] . The biochemical properties of the 9-1-1 complex are likely similar during mitosis and meiosis . We therefore postulate that OsHUS1 may also function as a component of the 9-1-1 complex to sense ectopic interaction and further recruit MutS to eliminate ectopic interaction intermediates . The characterization of RAD9 , RAD1 , and MSH2 homologs in rice will deepen our understanding of the ER-eliminating mechanism .
Oshus1-1 was derived from Nipponbare ( a japonica cultivar ) induced by tissue culture . Oshus1-2 was derived from Huanghuazhan ( an indica cultivar ) induced by 6°Co∼γ ray radiation . The new pair1 mutant allele was obtained from Nipponbare through tissue culture . In this allele , a Tos17 retrotransposon was inserted in the 7th exon of PAIR1 . The new Osrad51c allele was derived from an indica rice variety Zhongxian 3037 , induced by 6°Co∼γ ray radiation and found to have a premature stop codon in the 9th exon of OsRAD51C . The Oscom1 and zep1 alleles employed in this study is Oscom1-3 and zep1-1 , respectively [30] , [45] . Nipponbare was used as the wild type in the related experiments . STS markers were developed based on sequence differences between japonica variety Nipponbare and indica variety 9311 , which were used for map-based cloning of OsHUS1 . Primers sequences were listed in Supporting information , Table S1 . The cDNA sequence for OsHUS1 was verified by 3′RACE . Total RNA was extracted from rice young panicles ( 6–8 cm ) using TRIZOL reagent ( Invitrogen ) . A measure of 3 µg RNA was reverse-transcribed with Oligo-Adaptor primer ( CTGATCTAGAGGTACCGGATCC-d ( T ) 16 ) using the superscript III RNaseH reverse transcriptase ( Invitrogen ) . Two rounds of PCRs were carried out using Adaptor primer ( CTGATCTAGAGGTACCGGATCC ) , gene specific primers RACE1F ( TGTACCTTCTATGGTATTTC ) and RACE2F ( CTAGACTGACGGACAAGTCC ) . The product was cloned into pMD19-T vector ( TaKaRa ) and sequenced . A 261bp fragment from the exons of OsHUS1 was amplified by PCR with the primer pair OsHUS1RNAiF ( AAGGATCCCTGACAGTAGCTGTTACTC ) and OsHUS1RNAiR ( AGGTCGACACCATAGAAGGTACAGTCGG ) . The product was introduced into the BamHI-SalI and Bg II-XhoI sites of the pUCCRNAi vector in an inverted repeat orientation . The stem-loop fragment was finally cloned into the pCAMBIA 1300 vector . The OsHUS1-RNAi construct was introduced into Agrobacterium tumefaciens strain EHA105 and transformed the japonica cultivar Yandao 8 . Total RNA was extracted from the internode , leaf , root , panicle and seedling of Nipponbare , and was reverse-transcribed into cDNA . Quantitative RT-PCR analysis was performed using the CFX96 Real Time system ( Bio-Rad ) and Eva Green ( Biotium ) . The primer pair OsHUS1RTF ( CTTGGTGTTCGTGCAACC ) and OsHUS1RTR ( ACCACCAGGAGAAATACC ) was used . The standard control UBIQUITIN gene was examined with the primers UBI-RTF ( CAAGATGATCTGCCGCAAATGC ) and UBI-RTR ( TTTAACCAGTCCATGAACCCG ) . Husked seeds from the wild-type plants and the heterozygous Oshus1+/− plants were surface sterilized . Then they were sown on solid 1/2 MS medium containing 20 µg/ml MMC ( Solarbio ) in a light incubator . Genotype and phenotype assays of the seedlings were assayed 14 days later . To generate the antibody against OsHUS1 , the coding region of it was amplified from Nipponbare leaf cDNA with primer pair OsHUS1PETF ( ATGGATCCATGAAGTTCAAGGCCTTC ) and OsHUS1PETR ( ATCTCGAGACTGCCAGGGTCAAGGAC ) , and then ligated to the BamHI-XhoI site of the expression vector pET-30a ( Novagen ) . The expression vector was transformed into Escherichia coli strain BL21 ( DE3 ) and was induced for 3 h at 37°C by addition of 0 . 3 mM IPTG . His-tagged OsHUS1 were accumulated in the inclusion bodies and they were washed and subjected to SDS-PAGE . The main band of His-tagged OsHUS1 on the gel was cut off and powdered and used as an antigen against mice . The OsREC8 , PAIR2 , PAIR3 , OsMER3 , OsZIP4 , HEI10 , and ZEP1 polyclonal antibodies were used as described before [30] , [32] , [34] , [35] . Young panicles of at meiosis stage were harvested and fixed in Carnoy's solution ( ethanol:glacial acetic acid = 3∶1 ) for chromosome spreading . Meiotic chromosome preparation and immunofluorescence were performed as previously described [34] . The FISH procedure was performed as described [60] . Microscopy was conducted using a ZEISS A2 fluorescence microscope with a microCCD camera . Image capture and analysis was carried out using IPLab software ( BD Biosciences ) . | Meiosis is a special type of cell division that generates gametes for sexual reproduction . During meiosis , recombination not only occurs between allelic sequences on homologs , but also between non-allelic homologous sequences at dispersed loci . Such ectopic recombination is the main cause of chromosomal alterations and accounts for numerous genomic disorders in humans . To ensure genomic integrity , those ectopic recombinations must be quickly resolved . Despite the importance of ectopic recombination suppression , the mechanism underlying this process still remains largely unknown . Here , using rice as a model system , we identified the rice HUS1 homolog , a member of the RAD9-RAD1-HUS1 ( 9-1-1 ) complex , and elucidated its roles in meiotic recombination . In Oshus1 , vigorous ectopic interactions occur between nonhomologous chromosomes , and the number of crossovers is reduced . We suspect that OsHUS1 participates in regulating ectopic interactions during meiosis , probably by forming the canonical RAD9-RAD1-HUS1 ( 9-1-1 ) complex . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biology",
"and",
"life",
"sciences"
]
| 2014 | OsHUS1 Facilitates Accurate Meiotic Recombination in Rice |
CD8+ T cells play a key role in the in vivo control of HIV-1 replication via their cytolytic activity as well as their ability to secrete non-lytic soluble suppressive factors . Although the chemokines that naturally bind CCR5 ( CCL3/MIP-1α , CCL4/MIP- 1β , CCL5/RANTES ) are major components of the CD8-derived anti-HIV activity , evidence indicates the existence of additional , still undefined , CD8-derived HIV-suppressive factors . Here , we report the characterization of a novel anti-HIV chemokine , XCL1/lymphotactin , a member of the C-chemokine family that is produced primarily by activated CD8+ T cells and behaves as a metamorphic protein , interconverting between two structurally distinct conformations ( classic and alternative ) . We found that XCL1 inhibits a broad spectrum of HIV-1 isolates , irrespective of their coreceptor-usage phenotype . Experiments with stabilized variants of XCL1 demonstrated that HIV-1 inhibition requires access to the alternative , all-β conformation , which interacts with proteoglycans but does not bind/activate the specific XCR1 receptor , while the classic XCL1 conformation is inactive . HIV-1 inhibition by XCL1 was shown to occur at an early stage of infection , via blockade of viral attachment and entry into host cells . Analogous to the recently described anti-HIV effect of the CXC chemokine CXCL4/PF4 , XCL1-mediated inhibition is associated with direct interaction of the chemokine with the HIV-1 envelope . These results may open new perspectives for understanding the mechanisms of HIV-1 control and reveal new molecular targets for the design of effective therapeutic and preventive strategies against HIV-1 .
The replication of HIV-1 is regulated in vivo by a complex network of cytokines and chemokines expressed by immune and inflammatory cells . Key players in the mechanisms of HIV-1 control are CD8+ T cells , which , in addition to their cytolytic activity , secrete soluble factors that suppress HIV-1 in a non-lytic fashion [1]–[5] . Following the initial observation of this latter phenomenon in 1986 by Walker and colleagues [4] , subsequent studies demonstrated HIV-1 inhibition in co-cultures of CD8+∶CD4+ T cells separated by a semi-permeable membrane , as well as in cell-free supernatants from activated CD8+ T cells [1] , [3] , thus , ruling out the need for cell-to-cell contact . Moreover , the ability of CD8+ T cells to suppress HIV-1 replication was shown to correlate with the clinical stage of HIV-1 infection , suggesting a potential in vivo protective effect of this non-lytic CD8+ T cell activity [5] . Approximately 10 years after the initial description of soluble CD8+ T cell-derived inhibition of HIV replication , three chemokines of the CC ( β ) chemokine family ( CCL3/MIP-1α , CCL4/MIP- 1β , CCL5/RANTES ) were identified as major components of the soluble CD8+ T cell-derived anti-HIV activity [2] . These three chemokines act via a redundant mechanism of binding and downmodulating CCR5 to block entry of viruses with CCR5 coreceptor tropism . However , multiple lines of evidence indicate the existence of additional , still undefined , CD8-derived factors that can suppress HIV-1 infection . In particular , the observation that CD8+ T-cell culture supernatants can also inhibit CC-chemokine-resistant HIV-1 strains , such as those restricted to CXCR4 coreceptor usage [6] , [7] , substantiates a role for new , still uncharacterized anti-HIV factors produced by CD8+ T cells . Additionally , several reports have documented a suppressive effect of these factors at the transcriptional level [8]–[10] , whereas CCR5-binding chemokines act at the level of viral entry/fusion . In addition to CD8+ T cells , other cells of the immunohematological system can produce soluble HIV-suppressive factors , including CD4+ T cells , γ/δ T cells , NK cells , cells of the mononuclear phagocytic system , and platelets [11]–[13] . Recently , we identified a novel antiviral chemokine , CXCL4/PF4 , which is mainly produced by megakaryocytes and platelets . CXCL4 was shown to inhibit a broad spectrum of HIV-1 isolates , irrespective of their coreceptor usage and genetic subtype; it acts at the level of viral entry via a novel mechanism mediated by direct interaction with the viral envelope [14] . In this study , we report the characterization of a novel anti-HIV-1 C-chemokine , XCL1 , which exhibits a broad spectrum of activity against different biological variants of HIV-1 . We present evidence that this chemokine blocks infection at an early step of the viral life cycle , namely , viral attachment and entry into host cells . Similar to our previous work with CXCL4/PF4 , we found that XCL1 acts through an unconventional mechanism mediated by direct interaction with the HIV-1 envelope . Moreover , we investigated the correlation between the unique metamorphic properties of XCL1 [15] and its antiviral activity , showing that the alternatively-folded ( all β-sheet ) structure of XCL1 is the specific conformation responsible for HIV-1 blockade . These results offer insights into pathogenesis and provide new molecular targets for HIV-1 therapy and vaccine development .
As multiple lines of evidence indicated the existence of still uncharacterized HIV-suppressive factors produced by CD8+ T cells [16] , we used a wide-platform cytokine array ( RayBio ) , which evaluates in a semi-quantitative fashion 507 soluble factors , to screen culture supernatants from activated primary human CD8+ T cells . Among the top 25% most expressed proteins , we identified the C-chemokine XCL1/lymphotactin , in addition to other previously reported anti-HIV chemokines ( Guzzo et al . , in preparation ) . We focused our attention on XCL1 because it is produced preferentially by CD8+ T cells [17] , [18] . Production of XCL1 by primary human CD8+ T cells was confirmed by ELISA in culture supernatants from CD8+ T cells activated ex vivo with either PHA , PMA plus ionomycin , or anti-CD3/CD28 antibodies ( Figure 1 ) . PMA plus ionomycin was the most potent stimulation for XCL1 production , followed by anti-CD3/CD28 antibodies , while PHA elicited the secretion of markedly lower levels . To assess the ability of XCL1 to inhibit HIV-1 infection , acute infection assays were initially performed with recombinant human XCL1 ( Peprotech ) in primary human PBMC infected with laboratory-adapted viral strains . XCL1 potently inhibited HIV-1 infection irrespective of coreceptor specificity , as it was equally effective on strains specific for CXCR4 ( X4; IIIB ) and CCR5 ( R5; BaL ) ( Figure 2A ) . To further examine the breadth of XCL1-mediated inhibition , we evaluated the ability of recombinant XCL1 to inhibit infection of primary PBMC with a panel of primary HIV-1 isolates with different coreceptor-usage specificity . As seen with the laboratory-adapted HIV-1 strains , XCL1 was equally effective on CCR5-specific and CXCR4-using primary isolates ( Figure 2B ) . Of note , XCL1 did not reach 90% inhibition on two R5 isolates even at the highest dose used ( 1 µM ) . Treatment with XCL1 did not reduce cell viability , indicating that the HIV-inhibitory effect of XCL1 was not due to toxic or anti-metabolic effects on the cells ( data not shown ) . The anti-HIV-1 activity of XCL1 was also confirmed in the engineered cell line TZM-bl [19]–[21] , a HeLa cell derivative expressing CD4 , CXCR4 , and CCR5 ( data not shown ) . XCL1 is a metamorphic chemokine that interconverts in solution between two distinctly folded structures , namely the canonical chemokine fold ( three-stranded anti-parallel β-sheet and C-terminal helix ) , which has been reported to bind and activate the specific XCL1 receptor ( XCR1 ) , and an alternative , four-stranded sheet that forms a dimeric β sandwich , which was reported to bind glycosaminoglycans ( GAGs ) , but not XCR1 [15] . Thus , we tested two stabilized recombinant XCL1 variants produced in E . coli: CC3 , a variant locked in the chemokine-folded structure [22] , and W55D , a variant that preferentially adopts the alternatively-folded dimeric structure [15] . As a control , we also tested a full-length , wild type ( WT ) XCL1 , which retains the ability to interconvert between protein folds . A dual-tropic primary HIV-1 isolate sensitive to XCL1-mediated inhibition , 92HT599 , was used for acute infection assays to evaluate inhibition with XCL1 variants . As shown in Figure 3 , we observed striking differences in the inhibition exhibited by the locked XCL1 variants , CC3 and W55D . Indeed , the W55D variant inhibited HIV-1 with similar potency , as did the WT chemokine , while the CC3 variant showed no appreciable inhibition at the doses used . We observed similar results in infection assays performed with both CXCR4-tropic ( IIIB ) and CCR5-tropic ( BaL ) HIV-1 isolates , with the all-β , alternatively-folded conformation ( W55D ) conferring inhibition , and the chemokine-folded conformation ( CC3 ) showing minimal , if any , activity ( Figure S1 ) . These results suggested that the antiviral activity of XCL1 is dependent on the protein conformation and appears to be unrelated to activation of the XCR1 receptor , which is specifically bound and activated by the CC3 variant , but not by the W55D variant . In line with this observation , we were unable to detect XCR1 expression by flow cytometry in the HIV-1 target cells used in our study ( data not shown ) . Since we established that the anti-HIV activity of XCL1 is associated with XCL1 adopting the all-β ( alternatively-folded ) conformation , which does not bind and activate the XCR1 receptor but is capable of interacting with GAGs [23] , we focused on the extracellular events in the HIV-1 replication cycle and assessed the ability of XCL1 to interfere with viral attachment and entry . To achieve a high level of consistency of these assays , experiments were performed using TZM-bl cells infected with the primary dual-tropic isolate , 92HT599 , which is highly sensitive to XCL1-mediated inhibition . As seen in Figure 4A and B , the WT chemokine and the W55D variant effectively blocked viral attachment and entry , while the CC3 variant had no appreciable inhibitory effect , reflecting the pattern of inhibition seen in the infection assays ( Figures 2 and 3 ) . In parallel , we examined the inhibitory effects of two other anti-HIV chemokines , namely CXCL4/PF4 and CCL5/RANTES . In accordance with our previous work , CXCL4 effectively inhibited viral attachment and entry [14] , whilst CCL5 had enhancing effects , as previously observed with other CXCR4-tropic HIV-1 strains [24] . Additional controls included T-20 , a well-characterized fusion inhibitor [25] , which showed no effect on viral attachment , but a significant reduction in viral entry , and an anti-CD4 monoclonal antibody ( mAb ) , which showed only a slight reduction in attachment but a very marked inhibition of viral entry ( Figure 4A , B ) . Since we documented an inhibitory effect of XCL1 at the level of HIV-1 attachment and entry , we examined the ability of XCL1 to downmodulate the main HIV-1 cellular receptors , namely , CD4 , CXCR4 and CCR5 . Flow cytometry did not reveal any change in surface expression of these receptors after XCL1 treatment for 24 h ( data not shown ) . Furthermore , we also investigated interactions between XCL1 and CD4 via binding assays with anti-CD4 antibodies targeting different domains of CD4 ( D1 , D2 and D3–4 ) ; we did not observe any modification in CD4 staining , suggesting that XCL1 does not interact with cell surface-expressed CD4 ( data not shown ) . In view of the data described above , we investigated the possibility that XCL1 may interact directly with HIV-1 virions , as we previously demonstrated for CXCL4/PF4 [14] . To investigate this hypothesis , we performed a virion capture assay by which immunomagnetic beads were armed with different XCL1 variants ( WT , W55D , or CC3 ) as molecular “baits” to capture whole HIV-1 virions , as previously described [14] . Figure 5A shows that both WT and W55D efficiently captured HIV-1 virions . The specificity of this interaction was validated upon observation of reduced capture when XCL1-armed beads were pre-incubated with anti-XCL1 mAb or polyclonal antibody ( pAb ) prior to exposure to the virus . In agreement with our infection and attachment/entry assays , we did not observe any appreciable virion capture when the beads were armed with the CC3 XCL1 variant . Our data demonstrate that XCL1 can directly interact with HIV-1 virions , and that the all-β ( alternatively-folded ) XCL1 conformation ( W55D ) mediates this interaction , while the classic chemokine-folded conformation ( CC3 ) does not . To support the relevance of these data to the antiviral activity of XCL1 , we found that the same anti-XCL1 pAb that was used to block HIV-1 capture reversed the antiviral activity of XCL1 in PBMC infection experiments ( Figure S2 ) . To demonstrate that XCL1 can interact directly with the external viral envelope glycoprotein , gp120 , we performed co-immunoprecipitation studies with biotin-conjugated XCL1 WT and variants . As seen in Figure 5B , XCL1 WT and W55D were able to specifically co-immunoprecipitate gp120 , while the CC3 variant did not . Importantly , the same anti-XCL1 pAb that prevented virion capture abrogated gp120 co-immunoprecipitation ( Figure 5B ) . Taken together , these data support a mechanism of HIV-1 inhibition whereby XCL1 interacts with viral particles via direct binding to the external viral envelope glycoprotein , gp120 . Furthermore , these data confirm the dependency of the anti-HIV-1 activity of XCL1 on the all-β ( alternatively-folded ) conformation . As an additional test for specificity of the interaction between XCL1 and the HIV-1 envelope glycoprotein ( gp120 ) , we performed both virion-capture and infection assays using VSV-G pseudotyped virions , which contain an HIV-1 core surrounded by the VSV envelope . Figure 6A shows that XCL1 was unable to capture VSV-G pseudotyped virions , indicating the HIV-1 capture observed in Figure 5A required the presence of the HIV-1 envelope . To verify the relevance of these observations to the antiviral activity of XCL1 , we performed acute infection assays with VSV-G pseudotyped virions in primary PBMC . We did not observe any inhibitory effect of XCL1 , as evidenced by measuring both the absolute numbers of infected cells ( Figure 6B ) , and the levels of reporter gene ( GFP ) expression within the gated population of infected cells ( Figure S3 ) . Additionally , we observed no inhibition of VSV-G pseudotyped virus attachment or entry in TZM-bl cells ( data not shown ) . Altogether , these results further validate that the mechanism of XCL1 inhibition is via direct interaction with the HIV-1 envelope . Since we demonstrated that the antiviral activity of XCL1 depends on the all-β conformation ( W55D ) , previously shown to bind GAGs with high affinity [15] , we examined the inhibitory activity of XCL1 in PBMC infected with X4- or R5-tropic HIV-1 following digestion of cell-surface GAGs with heparitinase . As seen in Figure 7 , we observed that both WT and W55D XCL1 were equally effective at blocking HIV-1 infection in heparitinase-treated and -untreated cells , while in contrast the CC3 variant remained inactive in both conditions . The efficiency of GAG removal was evaluated by ELISA using two different anti-GAG mAbs ( Figure S4 ) . These data provide further evidence for an antiviral mechanism mediated by direct interaction of XCL1 with the viral envelope , irrespective of it's binding to GAGs and/or other structures expressed on the target cell surface .
In this study , we report the characterization of the CD8+ T cell-derived C-chemokine , XCL1 , as a novel , broad-spectrum inhibitor of HIV-1 infection . XCL1 is primarily produced by activated CD8+ T cells and NK cells [17] , and recruits T lymphocytes and dendritic cells via binding to and activation of a specific cellular receptor , XCR1 [26] , [27] . A possible link between XCL1 and HIV-1 was previously suggested in two wide-screening studies of chemokines and chemokine receptors: the first reported low-level inhibition of HIV-1 replication by XCL1 [28] , while the second identified a small subset of HIV-1 isolates that could use the XCL1 receptor , XCR1 , as a coreceptor in cells transfected in vitro [29] . However , these data were not subsequently validated in further studies , nor were the potential underlying mechanisms investigated . Our work provides a thorough characterization of the anti-HIV-1 activity of XCL1 , showing no apparent relationship between the antiviral action of XCL1 and the putative function of XCR1 as a coreceptor , although these findings do not exclude that XCR1 may serve as a minor HIV-1 coreceptor in specific cells or anatomical sites . XCL1 is a unique metamorphic chemokine that can interconvert between two different conformational folds: the conserved chemokine fold ( monomer ) , which was shown to bind to and activate XCR1 , and an alternatively-folded ( all β-sheet ) dimeric conformation which does not activate XCR1 , but instead binds to glycosaminoglycans ( GAGs ) with high affinity [15] , [23] . Using XCL1 variants designed to predominantly fold into one of the two conformations , we found that only the alternatively-folded ( all β-sheet ) molecule elicited anti-HIV activity , while the chemokine locked in the classical , XCR1-interacting fold was inactive . In line with this observation , we were unable to detect XCR1 expression by flow cytometry in the HIV-1 target cells used in our study; in fact , the ability of CD4+ T cells to express XCR1 is controversial [26] , [30] . At this stage , it is uncertain whether and to what extent the inherent tendency of alternatively-folded XCL1 to dimerize plays any role in HIV-1 inhibition , as a monomeric form of the alternatively-folded XCL1 is not available for testing . Regardless , since the alternatively-folded molecule binds cell-surface GAGs and not XCR1 , our results led us to investigate the early events in the viral infectious cycle that take place at the target cell surface . Indeed , we documented XCL1-mediated blockade of HIV-1 at an early stage of infection , namely , viral attachment and entry . Moreover , we provide multiple lines of evidence that XCL1 inhibits HIV-1 through an unconventional mechanism mediated by direct interaction with the viral envelope , similar to that previously reported for the α-chemokine CXCL4 [14] . We showed that XCL1 efficiently captures infectious HIV-1 virions and binds to the external viral envelope glycoprotein , gp120 , and that both of these interactions depend on the alternatively-folded XCL1 structure . Furthermore , the same polyclonal antibody that antagonized virion capture and gp120 binding by XCL1 also neutralized the antiviral activity of XCL1 . In line with the proposed antiviral mechanism , we demonstrated that binding to cell-surface GAGs was not required for the antiviral activity of XCL1 , despite the dependence on the alternative , GAG-binding conformation for HIV-1 inhibition . These findings indicate that gp120 is another selective target of the alternative XCL1 conformation in addition to GAGs . The fact that the biologically active conformation of XCL1 against HIV-1 is the high-affinity GAG-binding structure raises several mechanistic considerations . Foremost , the amount , complexity and variability of the glycan shield that decorates the surface of gp120 most likely influences the ability of XCL1 to block HIV-1 infection , since nearly half the molecular mass of gp120 is comprised of N-linked and O-linked glycans , and changes to these carbohydrate moieties result in altered neutralization sensitivity [31]–[36] . Indeed , it is possible that XCL1 interacts with a negatively-charged domain on the surface of gp120 [37] . Future structure-function studies with mutagenized XCL1 will help delineate key domains of the chemokine that are responsible for HIV-1 inhibition . Whether and to what extent endogenous XCL1 contributes to the mechanisms of virus control during the course of HIV-1 infection is presently unknown . Although we found that XCL1 is a broad-spectrum HIV-1 inhibitor , we observed some variability in sensitivity among HIV-1 isolates . Different degrees of sensitivity have been documented for a wide range of antiviral biomolecules , including neutralizing antibodies [38]–[42] , in line with the remarkable variability of the HIV-1 envelope , which we identified as the primary target for XCL1 antiviral activity . Another unresolved question is the discrepancy ( ∼2-log ) between the XCL1 concentrations required for HIV-1 blockade and the levels released by activated CD8+ T cells cultured in vitro . However , it is important to emphasize that our data were obtained with E . coli-produced recombinant XCL1 , leading to a significant underestimation of the potency of this chemokine . In fact , the C-terminus of XCL1 is a 22-amino acid mucin-like domain containing a cluster of O-glycosylated serine and threonine residues , and previous work has demonstrated that mammalian cell-produced , fully glycosylated XCL1 exhibits approximately 2-log higher biological activity compared with non-glycosylated XCL1 produced in prokaryotic cells [43] . Currently , there are no commercial sources of mammalian cell-produced XCL1 , and efforts are underway in our laboratory to produce glycosylated XCL1 . In addition , it is conceivable that in vivo-activated CD8+ T cells may release larger amounts of XCL1 into the local microenvironment , particularly within secondary lymphoid tissues . We are currently investigating if CD8+ T cells derived from asymptomatic HIV-infected subjects produce higher concentrations of XCL1 than CD8+ T cells from uninfected subjects , in line with their reported higher production of crude antiviral factor activity [5] . The identification of the first HIV-suppressive chemokines ( CCL5/RANTES , CCL3/MIP-1α and CCL4/MIP-1β ) has led not only to novel insights into endogenous host defenses against HIV-1 , but also to the definition of new molecular targets for antiviral drugs [44]–[47] and genetic markers of innate HIV-1 resistance [48] , [49] . In a similar manner , this study could be a first step toward determining the potential physiological role of XCL1 in HIV-1 infection . Analysis of XCL1 expression in subjects that are naturally protected from HIV infection ( exposed-uninfected ) or from disease progression ( long-term nonprogressors ) may offer new insights on mechanisms of natural resistance to HIV . Furthermore , a precise identification of the XCL1-interactive surface on the viral envelope may lead to the development of novel HIV-1 entry inhibitors , as well as new molecular targets for vaccine design .
Recombinant human XCL1/lymphotactin was obtained from Peprotech ( Rocky Hills , NJ ) ; recombinant XCL1 WT and variants ( CC3 and W55D ) were cloned and produced by two of the authors ( JF , BFV ) at the Medical College of Wisconsin , Milwaukee , WI , as previously described [15]; and recombinant RANTES/CCL5 and CXCL4/PF4 were purchased from R&D Systems ( Minneapolis , MN ) . Molar values were calculated based on the molecular weight of the monomeric chemokines . PBMC from healthy donors were activated with phytohemagglutinin ( PHA; Sigma , St . Louis , MO ) and recombinant human IL-2 ( Roche Applied Science , Mannheim , Germany ) in complete RPMI medium ( Invitrogen , Carlsbad , CA ) , containing 10% fetal bovine serum ( FBS , Hyclone , Thermo Scientific , Waltham , MA ) , glutamine at 2 mM , streptomycin at 50 µg/mL , and penicillin at 100 U/mL for 72 hr prior to HIV-1 infection . Cell surface glycosaminoglycan ( heparin sulfate ) digestion was performed by incubating PBMC ( 1×106 cells/mL ) with heparitinase ( Heparinase III , Sigma ) at 2 U/mL for 2 hr at 37°C in recommended buffer ( 20 mM Tris-HCl pH 7 . 5 containing 1% FBS and 4 mM CaCl2 ) . Digested PBMC were washed once in complete RPMI and then used in acute infection assays as described . TZM-bl cells ( NIH AIDS Research and Reference Reagent Program , Germantown , MD ) were maintained in DMEM ( Invitrogen , Carlsbad , CA ) containing 10% fetal bovine serum . CD8+ T cells were enriched via negative selection from PBMC with the EasySep enrichment kit ( Stem Cell Technologies , Vancouver , Canada ) and activated by either PHA ( 20 µg/mL , Sigma ) , PMA ( 0 . 05 µg/mL , Sigma ) plus ionomycin ( 1 µg/mL , Sigma ) , or anti-CD3/CD28 antibody-loaded beads ( T Cell Activation/Expansion Kit , Miltenyi Biotec , Auburn , CA ) , all in the presence of 50 U/mL of IL-2 . After 3 days of activation , the cells were washed to remove stimuli , medium was replaced with complete RPMI supplemented with IL-2 at a cell density of 1×106 cells/mL , and the culture supernatants were harvested at day 5 and 7 post-stimulation . CD8+ T-cell culture supernatants ( after 3 days of activation and washing ) were tested for XCL1 production using the Human XCL1/Lymphotactin DuoSet ELISA ( R&D Systems ) . To confirm the efficiency of GAG removal following heparitinase digestion , a cell-based ELISA was performed according to a previously established protocol with some modifications [50] . Briefly , activated PBMC were washed in PBS and seeded at 5×104 cells/well in 50 µL of PBS , and dried by evaporation at room temperature overnight . Plates were washed once with PBS and immediately fixed in ice-cold 2% paraformaldehyde at 4°C for 20 min , followed by washing in PBS . The wells were blocked in 0 . 2% casein-PBS buffer for 1 hr at 37°C , washed once with PBS , and incubated with 10 µg/mL of anti-heparan sulfate mAbs , clone 10E4 ( AMSBIO , Lake Forest , CA ) and clone T320 . 11 ( EMD Millipore , Temecula , CA ) for 2 hr at RT . An anti-CD4 mAb ( RPA-T4 , BD Biosciences ) was used as a control for the non-specific effects of digestion on cell-surface protein expression ( 2 hr at RT ) . After 3 washes with PBS , the wells were incubated with polyclonal HRP-conjugated anti-mouse antibodies ( Thermo Fisher Scientific , Rockford , IL ) . After 3 washes with PBS , wells were incubated with substrate solution until color development and immediate incubation with stop solution ( R&D Systems ) , followed by reading optical density at 450 nm . Background measurements obtained with secondary antibody alone were subtracted from all readings . The ability of XCL1 to downmodulate cell surface expression of CD4 , CXCR4 , and CCR5 was investigated by flow cytometry . Briefly , CD4+ T cells were cultured in the presence/absence of XCL1 at 20 µg/mL for 24 hours . Cells were then washed and stained for receptor expression using anti-CD4 , CXCR4 , and CCR5 mAbs ( BD Biosciences , San Jose , CA ) . To further determine possible interactions between XCL1 and CD4 , we assessed the ability of XCL1 to interfere with binding of different anti-CD4 mAbs targeted to different domains of CD4 . Six fluorochrome-conjugated antibodies were used: OKT4 ( eBioscience , San Diego , CA ) , 13B8 ( Beckman Coulter , Inc . , Indianapolis , IN ) , VIT4 ( Miltenyi Biotec ) , RPA-T4 , Leu3A/SK3 , and L200 ( all 3 from BD Biosciences ) . Two unlabeled mouse mAbs were used , DB-81 [51] and 9H5A8 ( Novus Biologicals , Littleton , CO ) , followed by subsequent anti-mouse-R-phycoerythrin staining ( Sigma ) . Enriched CD4+ T cells were incubated with XCL1 at 20 µg/mL or with PBS for 30 minutes at 4°C , without washing , cells were then stained with the various anti-CD4 mAbs listed . All data were acquired on a BD FACS Canto flow cytometer ( San Jose , CA ) and analyzed with FlowJo software version 9 . 5 . 2 for Macintosh ( TreeStar , San Carlos , CA ) . The HIV-1 isolates used in this study included two laboratory strains [IIIB ( X4 ) and BaL ( R5 ) ] , the dual-tropic primary isolate , 92HT599 ( X4R5 ) , and a set of primary isolates derived in our laboratory and minimally passaged ex vivo ( 98USSG , 07USLD , 07USPC , 08USSE , 97IT6366 , 08USKD ) , obtained by culture of PBMC from chronically infected individuals . Acute cell-free HIV-1 infection was performed by addition of the viral stocks ( 50–100 pg of p24 Gag antigen per well ) to duplicate cultures of activated PBMCs ( PHA+IL-2 for 72 h ) in round-bottom 96-well plates seeded at 2×105 cells per well in RPMI+10% FBS+20 U/mL of IL-2 , or to TZM-bl cells seeded in 24-well plates overnight at 5×104 per well in DMEM+10% FBS for infection . Infected cells were cultured in the presence/absence of XCL1 at doses ranging from 0 . 06–1 . 5 µM . The levels of HIV-1 replication were assessed by measuring the extracellular release of p24 Gag protein in cell-free culture supernatants taken daily between days 3 and 7 post-infection using a highly sensitive Alpha ( Amplified Luminescent Proximity Homogeneous Assay ) technology immunoassay ( AlphaLISA HIV p24 Research Immunoassay Kit , PerkinElmer , Waltham , MA ) . On day 7 of infection , cells were harvested for viability testing via absolute counting by flow cytometry . Cell viability was determined by normalization of the total live-gated cell counts in XCL1-treated wells to the number of cells recovered from control wells ( untreated with XCL1 ) . To show the physiological relevance of our infection data , we performed an infection assay whereby HIV-1 IIIB was pre-incubated ( prior to addition to target cells ) with XCL1 alone ( 1 µM ) or with XCL1 combined with anti-XCL1 pAb ( 10 µg/mL of the same pAb used to block HIV-1 capture by XCL1 ) . To control for non-specific effects of the pAb we also included control wells ( no XCL1 treatment ) with pAb alone . As a test for specificity of the interaction between XCL1 and the HIV-1 envelope glycoprotein ( gp120 ) , we performed infection assays with GFP-expressing VSV-G pseudotyped virus provided by Michael P . Marino ( CBER/FDA , Bethesda , MD , USA ) . PBMC ( 5×104 ) were seeded in 96-well round bottom plates and infected in a 50 µL volume of pseudotyped virus ( MOI of 10 ) in the presence or absence of XCL1 WT overnight , with each condition in quadruplicate wells . Wells were supplemented with an additional 50 µL of complete RPMI at 24 hr post-infection to yield a final well volume of 100 µL . Individual wells were harvested 48 hr post-infection for flow cytometry detection of GFP-positive ( infected ) cells . To supplement the data counting the absolute numbers of infected cells , mean fluorescence intensity was also determined to indicate the amount of virus infection within each GFP-positive event ( infected cell ) . The HIV-1 attachment and entry assays were performed on TZM-bl cells with the primary , dual-tropic HIV-1 isolate 92HT599 . TZM-bl cells ( 106 per replicate; two replicates per treatment ) were seeded in 12-well plates overnight to achieve a confluent cell monolayer . Without disturbing the monolayer , cells were washed with PBS to remove media , followed by pre-incubation for 15 min at room temperature with XCL1 diluted in PBS , and then exposed to 500 µL of the undiluted viral stock ( 124 ng/mL of p24 ) for 4 h ( attachment ) or 6 h ( entry ) at 37°C , in the continuous presence of XCL1 . Two wells of untreated cells were incubated for 4 hr with virus at 4°C to determine the background signal level ( trypsin-insensitive despite low-temperature conditions preventing virus entry ) . As specificity controls , replicate wells were pretreated with known inhibitors/inducers of viral attachment/entry prior to virus incubation: CXCL4/PF4 at 15 µg/mL ( R&D Systems ) , peptide T-20 at 50 µg/mL ( NIH AIDS Research and Reference Reagent Program , Germantown , MD ) , an anti-CD4 mAb at 20 µg/mL ( azide-free RPA-T4 , eBioscience , San Diego , CA ) , or CCL5/RANTES at 15 µg/mL ( R&D Systems ) . After incubation , the cells were washed with PBS to remove unbound virus , without disturbing the cell monolayer . Entry assay wells were treated with pre-warmed bovine trypsin ( Sigma ) for 5 min at 37°C , followed by trypsin inactivation with cold DMEM medium containing 10% ( vol/vol ) FBS . Both trypsin-treated ( entry assay ) and untreated ( attachment assay ) cells were then washed two times with cold PBS , and lysed with 100 µL of 0 . 5% ( wt/vol ) Triton X-100 to quantify the amount of cell-associated p24 protein . The specific signal was calculated by subtracting the background p24 levels measured in wells incubated at 4°C ( treated with trypsin ) from the p24 levels measured in each test sample . The virion capture assay was performed as previously described [14] . Briefly , immunomagnetic beads ( 4×104 per tube ) covalently linked to a polyclonal antiserum to rabbit IgG ( Invitrogen ) were incubated with a polyclonal rabbit IgG antibody to human XCL1 ( Peprotech ) , washed with PBS containing 0 . 05% ( wt/vol ) bovine casein and then loaded with recombinant human XCL1 ( 2 . 5 µg per reaction ) . After removing unbound XCL1 by repeated PBS washes , chemokine-armed beads were incubated with 0 . 5 mL of the viral stock ( HIV-1 IIIB ( X4 ) ; 20 ng of p24 Gag protein/test ) . To test the specificity of XCL1 interaction with the virus , the XCL1-armed immunomagnetic beads were pre-incubated with monoclonal ( mAb ) and polyclonal ( pAb ) anti-XCL1 antibodies ( R&D Systems , 20 µg/mL ) for 10 minutes at room temperature prior to virus addition . After incubation with virus for 1 h at room temperature , the beads were washed to remove unbound virus particles and treated with 0 . 5% Triton X-100 to lyse the captured virions . The amount of captured p24 Gag protein was quantified by AlphaLISA® . As an additional measure for the exclusive interaction between XCL1 and the HIV-1 envelope glycoprotein ( gp120 ) , we assessed the ability of XCL1 to capture VSV-G pseudotyped virus . Anti-mouse immunomagnetic beads were armed with monoclonal anti-VSV-G antibody ( KeraFAST Inc . , Boston , MA ) to show the efficiency of VSV-G pseudotyped virus capture in our experimental design . In parallel , anti-rabbit immunomagnetic beads armed with both anti-XCL1 pAb and subsequent XCL1 WT were tested for the ability to capture VSV-G-pseudotyped virus . For accurate comparison of capture between beads armed with anti-VSV-G mAb and beads armed with XCL1 , equal amounts of VSV-G-pseudotyped virus was added to all capture reactions . To evaluate the direct interaction between XCL1 and the gp120 external envelope glycoprotein , we performed co-immunoprecipitation experiments using purified YU2 gp120 protein . To assess this interaction , XCL1 WT , W55D and CC3 proteins were biotinylated using the LYNX Rapid Conjugation Kit ( AbD Serotec , Kidlington , UK ) . In two conditions we assessed the specificity of XCL1-gp120 interactions via pre-incubation of biotinylated XCL1 WT with 5 µg of anti-XCL1 pAb ( R&D Sytems ) or goat IgG ( R&D Systems ) , as a control , for 1 h in 100 µL of PBS . Following the presence or absence of antibody pre-incubation , a mixture of biotinylated XCL1 WT , W55D , or CC3 ( 2 µg ) was incubated with gp120 ( 2 µg ) in 100 µL of PBS+0 . 2% casein for 3 h at room temperature with constant rotation . Subsequent incubation with 50 µL of streptavidin-coated magnetic beads ( Invitrogen ) in 200 µL of RIPA buffer was performed for an additional 10 min incubation . The samples were washed three times with RIPA buffer , dissolved in SDS loading buffer , and loaded on 12% polyacrylamide gels and resolved by SDS gel electrophoresis . Protein was transferred onto nitrocellulose membranes and blotted with an anti-gp120 mAb ( b24; a gift from George K . Lewis , University of Maryland , Baltimore , MD ) . Anonymized samples of peripheral blood were obtained from healthy volunteer donors at the NIH Blood Bank under a protocol approved by the NIH IRB . | Although HIV , the causative agent of AIDS , establishes a lifelong infection that cannot be eradicated even with effective treatment , the host immune system has the ability to contain its replication for many years in which the disease remains asymptomatic . Key players in HIV control are CD8+ T cells , specialized immune cells that can not only destroy infected cells , but also secrete soluble factors that suppress the virus without killing infected cells . CD8+ T cells produce multiple HIV-suppressive factors , including certain chemokines ( soluble proteins that attract immune cells ) , which block the virus even before it can gain access to its target cells . In the present study , we characterize a new anti-HIV chemokine , XCL1 or lymphotactin , which is primarily produced by CD8+ T cells . A unique feature of XCL1 is that , unlike other antiviral chemokines , it has a very broad spectrum of activity against different variants of HIV-1 and directly binds the virus outer coat , rather than blocking specific receptors on the target cell . Also unique is that fact that XCL1 adopts two possible conformations , and only one of them is capable of HIV inhibition . These findings may open new avenues for the design of effective drugs or vaccines against HIV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | The CD8-Derived Chemokine XCL1/Lymphotactin Is a Conformation-Dependent, Broad-Spectrum Inhibitor of HIV-1 |
Dengue is one of the most serious mosquito-borne infectious diseases in the world . Aedes albopictus is the most invasive mosquito and one of the primary vectors of dengue . Vector control using insecticides is the only viable strategy to prevent dengue virus transmission . In Guangzhou , after the 2014 pandemic , massive insecticides have been implemented . Massive insecticide use may lead to the development of resistance , but few reports are available on the status of insecticide resistance in Guangzhou after 2014 . In this study , Ae . albopictus were collected from four districts with varied dengue virus transmission intensity in Guangzhou from 2015 to 2017 . Adult Ae . albopictus insecticide susceptibility to deltamethrin ( 0 . 03% ) , permethrin ( 0 . 25% ) , DDT ( 4% ) , malathion ( 0 . 8% ) and bendiocarb ( 0 . 1% ) was determined by the standard WHO tube test , and larval resistance bioassays were conducted using temephos , Bacillus thuringiensis israelensis ( Bti ) , pyriproxyfen ( PPF ) and hexaflumuron . Mutations at the voltage-gated sodium channel ( VGSC ) gene and acetylcholinesterase ( AChE ) gene were analyzed . The effect of cytochrome P450s on the resistance of Ae . albopictus to deltamethrin was tested using the synergistic agent piperonyl butoxide ( PBO ) . The results showed that Ae . albopictus populations have rapidly developed very high resistances to multiple commonly used insecticides at all study areas except malathion , Bti and hexaflumuron . We found 1534 codon mutations in the VGSC gene that were significantly correlated with the resistance to pyrethroids and DDT , and 11 synonymous mutations were also found in the gene . The resistance to deltamethrin can be significantly reduced by PBO but may generated cross-resistance to PPF . Fast emerging resistance in Ae . albopictus may affect mosquito management and threaten the prevention and control of dengue , similar to the resistance in Anopheles mosquitoes has prevented the elimination of malaria and call for timely and guided insecticide management .
Dengue is one of the most rapidly spreading mosquito-borne diseases in the world . Currently , 3 . 9 billion people in 128 countries or regions are at risk of dengue fever [1–3] . Guangzhou , the largest city in southern China and the capital of Guangdong Province , has become the epicenter of dengue outbreaks in China . In Guangzhou , the number of dengue cases accounted for 50% of the national incidence between 1978–2011 and an epidemic had occurred once every 3–4 years since the 1990s [4 , 5] . In particular in 2014 , a pandemic of dengue broke out in Guangzhou with more than 37 , 000 cases reported [6] . Aedes albopictus is the most invasive mosquito and is widely distributed in China , from Hainan in the south to Dalian in the north , while Aedes aegypti is only distributed in Hainan , Yunnan and a small area of the southernmost part of Guangdong Province [7] . Ae . albopictus is the main vector for dengue virus in China and in Guangzhou Ae . albopictus is the sole vector of dengue virus [8 , 9] . Currently , due to the lack of effective drugs and vaccines against dengue , vector management is the main strategy to prevent and control mosquito-borne diseases , including dengue [10–13] . In China , chemical control through the use of insecticides is one of the major tools for the control of vector mosquitoes [14 , 15] . During the outbreak of dengue in Guangzhou in 2014 , more than 27 , 000 kg of pyrethroids were used for ultralow-volume ( ULV ) spraying to control adult Ae . albopictus , and a large amount of temephos , an organophosphate larvicide , was used for larval control . Chemical insecticides were also frequently used for focal hot-spot control of sporadic dengue transmission in Guangzhou [16 , 17] . At the same time , agricultural insecticide usage in rural areas and residential insecticide usage in the city affected the resistance of Ae . albopictus in Guangzhou , although insecticide use is greater in the public health field . The government also regularly organized the patriotic health campaign to clean up aquatic mosquito habitats in Guangzhou . With the extensive use of insecticides , insecticide resistance has become a threat . Since 2014 , resistance to some insecticides has been reported in Ae . albopictus in limited regions of Guangzhou . Li et al . reported that the Ae . albopictus adult population in Yuexiu had developed resistance to dichlorodiphenyltrichloroethane ( DDT ) , and deltamethrin [18] . The reports of insecticide resistance in Guangzhou raise serious concerns about the efficacy of chemical insecticides against Ae . albopictus and the dengue transmission control policy in China . Current research on the resistance mechanism of Ae . albopictus mainly focuses on target-site insensitivity and increased metabolic detoxification of insecticides[19] . Non-synonymous mutations in the voltage-gated sodium channel ( VGSC ) gene that cause resistance to pyrethroids and DDT insecticides are known as knockdown resistance ( kdr ) [18 , 20 , 21] and mutations ( ace-1 ) in the acetylcholinesterase ( AChE ) gene cause a resistance to carbamates and organophosphates [22 , 23] . However Grigoraki et al . reported that the resistance of Ae . albopictus to temephos is associated with elevated carboxylesterases ( CCEs ) which is caused by up-regulation of CCEae3a gene [24] , and no difference was detected between resistant and susceptible CCEae3a_aeg variants [25] . Detoxification pathways are very complex and can be divided into three major gene families , monooxygenases ( P450s ) , carboxylesterases ( COEs ) , and glutathione S-transferases ( GSTs ) [26] . P450s are related to pyrethroid resistance in Ae . albopictus [27] . Determine the insecticide resistance status and mechanisms of Ae . albopictus in Guangzhou is very important for local vector control . In this study , Ae . albopictus was collected from four districts in Guangzhou during 2015–2017 and the resistances to the currently used insecticides was comparatively analyzed through a series of experiments . The aim was to characterize the spatial distribution , temporal changes , and mechanism of insecticide resistance in Guangzhou , and provide guidance for monitoring and controlling vector mosquitoes and mosquito-borne diseases .
The study was conducted in four districts in Guangzhou , Guangdong Province , China , from 2015 to 2017: 1 ) Yuexiu district is located in the old downtown area , 2 ) Tianhe district is located in the new downtown area , 3 ) Baiyun district is located in the suburban area , and 4 ) Conghua district is located in the rural area . The reported dengue incidence varied among the four districts . The study sites and dengue incidence rates was marked in Fig 1 , which was created by ArcGIS 10 . 2 . The research site is a subtropical area with a monsoon climate . The annual average annual temperature is 20–22°C , the average relative humidity is 77% , and the annual rainfall is approximately 1720 mm . The Foshan strain of Ae . albopictus was used as a control in this study , which was collected from Foshan City in 1983 , and kept in the laboratory without insecticide exposure since then . Ae . albopictus larvae were collected from three localities in each of the four districts , with representatives samples collected from parks , schools and residential areas , and all collection was done on public land ( S1 Table ) . The larvae were housed in a steel tank with a size of 23 cm*29 cm*6 . 5 cm , and 1 . 5–2 L of dechlorinated tap water and small turtle food for feeding were added to the tank . Adult mosquitoes were housed in 20 cm*45 cm*30 cm yarn cages and fed with 10% glucose water . The female mosquitoes were bloodfed from an anesthetized mouse for spawning . The larvae were reared in the laboratory until adulthood . In the laboratory , the temperature was maintained at 26 ± 2°C , the relative humidity was 70 ± 10% , and the light: dark cycle was 14 h: 10 h . Non-blood-fed F1-generation female mosquitoes aged 3–5 days were used for the resistance test . R24 is a laboratory resistant strain selected with deltamethrin for 24 generations from susceptible Ae . albopictus populations . Selection was performed by exposing each generation of fourth-stage larvae for 24 h to a 50% lethal concentration ( LC50 ) of deltamethrin . The LC50 was determined by a larval bioassay . After 24 generations , the LC50 of deltamethrin increased from 0 . 001 mg/L to 0 . 033 mg/L . The adult resistance bioassays were performed using five insecticides , including the four major classes of insecticides currently used which were recommended by WHO Pesticides Evaluation Scheme ( WHOPES ) [28] , i . e . , type II pyrethroid: deltamethrin; type I pyrethroid: permethrin; organophosphates: malathion; organochlorine: DDT; and carbamate: bendiocarb , following the standard WHO tube test protocol [29 , 30] . We used deltamethrin ( 0 . 03% ) , permethrin ( 0 . 25% ) , malathion ( 0 . 8% ) and bendiocarb ( 0 . 1% ) test for 1h and DDT ( 4% ) test for 0 . 5h by the standard WHO tube test . Testing kits and insecticide-impregnated papers with standard diagnostic doses were provided by the Universiti Sains Malaysia , Penang , Malaysia . In each holding tube , 25 adult female mosquitoes were tested with five replicates of field mosquitoes and two replicates of controls . The number of adult mosquitoes knocked down was recoded every ten minutes and used to calculate the value of 50% knockdown times ( KDT50 ) . After 1 h of exposure , the mosquitoes were transferred to holding tubes and fed on a 10% sucrose solution for 24 h . Mortality was scored after 24 h of recovery to determine the susceptibility status . After the bioassay , the dead and live mosquitoes were separated and stored individually in 95% alcohol for subsequent DNA analysis . The effect of cytochrome P450s on the resistance of Ae . albopictus to deltamethrin was tested using the synergistic agent piperonyl butoxide ( PBO ) following WHO guidelines [29 , 30] . Test paper with 4% PBO was prepared from 95% PBO ( Yien Co . Ltd , Shanghai , China ) . Experiments on sets of 25 field-collected female mosquitoes were performed separately for 1 ) exposure to PBO alone for 1 h; 2 ) exposure to PBO for 1 h followed by exposure to deltamethrin for 1 h; and 3 ) exposure to deltamethrin alone for 1 h; and 4 ) control: no exposure to any agent . After the 1 h experiments , experimental mosquitoes were transferred to holding tubes , and mortality at 24 h was documented . This process was repeated five times . Mosquito larval resistance bioassays were conducted using four insecticides which were recommended by WHOPES [31]: 1 ) organophosphate , temephos; 2 ) microbial bacterial toxin , Bti; 3 ) hormonal insect growth regulators , pyriproxyfen ( PPF ) ; 4 ) the chitin biosynthesis inhibitor , hexaflumuron; following WHO guidelines [32] . Industrial grade temephos ( 87 . 4% ) and PPF ( 98 . 3% ) were provided by the Chinese Centers for Disease Control and Prevention . Bti ( 7000 ITU/mg ) was provided by Wuhan Nature’s Favour Bioengineering . Hexaflumuron ( 99 . 0% ) was provided by Shanghai Yien . Twenty-five 3-4-instar Ae . albopictus larvae were added to 99 mL of dechlorinated tap water and 1 mL of different concentrations of insecticide solution . Nine concentration gradients for each insecticide were tested during the experiment , with concentrations ranging between 10% and 90% mortality , three replicates per concentration . For temephos and Bti , the number of dead larvae was counted 24 h after the experiment , and the LC50 was calculated . For PPF and hexaflumuron , emergence inhibition was measured daily until complete mortality or adult emergence , and IE50 was calculated . Genomic DNA was extracted from individual mosquitoes using the Extract-N-Amp Tissue PCR Kit ( Sigma Aldrich ) following the manufacturer’s protocol . Extracted DNA was stored at 4°C or used immediately for PCR . For each insecticide , 48 surviving individuals and 20 dead individuals were used to extract genomic DNA for mutation detection of target genes for insecticide resistance . DNA was extracted from individual mosquito and this is only done for adults which tested by the standard WHO tube test . Samples exposed to deltamethrin , permethrin , and DDT were genotyped at the voltage-gated sodium channel ( VGSC ) gene to detect mutations within domains II , III and IV , following the protocol by Kasai et al . , 2011 [33] . Samples exposed to bendiocarb were genotyped at the ace-1 gene to detect mutations within G119 following the protocol by M . Weill et al . , 2004 [34] . The details of the primers and PCR conditions are given in S2 Table . A total of 204 samples were sequenced for the kdr gene , and 68 samples were sequenced for the ace-1 gene . The sequences were aligned and analyzed using BioEdit ( http://www . mbio . ncsu . edu/BioEdit/bioedit . html ) . A survey on the use of pesticides in public regions was carried out via questionnaires from March to September of 2017 in the Yuexiu , Tianhe , Baiyun and Conghua districts of Guangzhou . The survey sites were selected near the collection sites of the Ae . albopictus samples . The surveys targeted employees of Pest Control Organizations ( PCOs ) and the local street administrators responsible for mosquito control . The contents of the survey included the species targeted and the frequency of insecticides used to control adult and larval mosquitoes . Each site completed 30 questionnaires . LC50 and KDT50 were estimated using the log-probit models . For larvae bioassays , the resistant status was measured by the resistant ratio ( RR50 ) , i . e . , the ratio of LC50 ( or IE50 ) for the field population over LC50 ( or IE50 ) for the laboratory-susceptible strain . Larval resistance status was defined as susceptible if RR50 < 5 , moderately resistant if 5 < RR50 < 10 , and highly resistant if RR50 > 10 [28] . Post hoc Tukey’s HSD test of analysis of variance ( ANOVA ) was used to compare differences in RR50 among different study sites . For adult bioassays , resistant status was defined by mortality rate: Resistant if mortality < 90% , probably resistant if mortality was between 90 and 98% , and susceptible if mortality > 98% [27 , 28] . The relationship between nonsynonymous mutations and resistance was verified by Fisher's exact test or the χ2-test ( when all n >5 ) , and the odds ratio ( OR ) was calculated for each mutation . The χ2-test was used to compare differences in adult mortalities between deltamethrin and deltamethrin + PBO groups at different study sites .
Ae . albopictus adult populations in the four districts were all resistant to four insecticides ( deltamethrin , permethrin , DDT and bendiocarb ) ( mortality<90% ) except malathion ( mortality >98% ) ( Fig 2A , S3 Table ) . The lowest mortality rate against deltamethrin and bendiocarb was 5 . 6% and 42 . 4% in the populations from Yuexiu , respectively , 42 . 1% to permethrin in Tianhe , and 25 . 2% to DDT in Conghua ( Fig 2A , S3 Table ) . Ae . albopictus larvae from all four districts were still sensitive to Bti and hexaflumuron ( RR50<5 ) , but displayed high resistance to temephos and pyriproxyfen ( RR50 > 10 ) , and moderate resistance to pyriproxyfen ( 5 ≤ RR50 ≤ 10 ) in the Conghua population ( Fig 2B , S4 Table ) . Comparatively , resistance to temephos and PPF were significantly higher in the urban areas , i . e . , Yuexiu , Tianhe and Baiyun , than in the rural areas , Conghua . Pyrethroids are the most commonly used insecticides to control the adult Ae . albopictus in Guangzhou . In 2015 , Ae . albopictus populations was susceptible to deltamethrin in Conghua and possible resistant in Tianhe and Yuexiu ( Fig 3A ) . In 2016 , Ae . albopictus populations in all four districts became resistant to deltamethrin ( Fig 3A ) . In 2017 , Ae . albopictus mortality against deltamethrin was significantly decreased at all study sites compared to 2016 , and the mortality in Yuexiu was only 5 . 6% ( Fig 3A ) . The average mortality against deltamethrin was 95 . 3% in 2015 . It decreased to 63 . 6% in 2016 and dropped again to 31 . 2% in 2017 ( ANOVA , all p < 0 . 05 ) , with a 30% decrease every year ( Fig 3A ) . The decrease in Ae . albopictus mortality against permethrin was also very fast from 2016 ( average 95 . 6% ) to 2017 ( 55 . 8% ) , with a 40% decrease in one year ( Fig 3B ) . Sequences of domains II ( 480 bp ) , III ( 346 bp ) and IV ( 280 bp ) of the VGSC gene were obtained from resistant and susceptible mosquitoes after deltamethrin , permethrin and DDT adult bioassays . Three synonymous mutations in domain II , 5 synonymous mutations in domain III , and 3 synonymous mutations in domain IV were detected ( S5 Table ) . In domain III , non-synonymous mutations were detected at codon 1534 , where wild-type TTC ( Phe ) was changed to either TCC ( Ser ) or CTC ( Leu ) . Data analysis showed that the F1534S and F1534L mutations were significantly associated with the resistance to deltamethrin , permethrin and DDT ( p<0 . 05 ) ( Table 1 ) . The 194-bp fragment in exon 5 of the AChE gene was obtained for ace-1 mutation detection , but no amino acid mutation was found in the G119 site in the 33 successfully sequenced samples . Exposing mosquitoes to PBO before exposing them to deltamethrin significantly increased Ae . albopictus mortality compared to directly exposing them to deltamethrin ( χ2-test , all p < 0 . 01 ) ( Fig 4 ) , indicating that PBO can reduce the resistance of Ae . albopictus to deltamethrin through anti-P450s activity . PPF is an insect growth regulator and has never been used previously in mosquito control in Guangzhou . However , resistance to PPF was detected in Ae . albopictus collected in all four study sites ( Fig 2B ) . Comparative analysis on Ae . albopictus larval resistance to deltamethrin from four field populations and one laboratory selected resistant strain , R24 , showed that all of them were resistant to deltamethrin as well as PPF ( Table 2 ) . Because R24 resistance to deltamethrin was artificially selected by exposing fully susceptible laboratory Ae . albopictus larvae to deltamethrin after 24 generations , this strain has never been exposed to any other insecticide; therefore , Ae . albopictus resistance to PPF could be explained by cross resistance to pyrethroids . Our survey found that four major classes of insecticides ( pyrethroids , organophosphates , organochlorine and carbamate ) were currently used in Guangzhou ( Table 3 ) . Pyrethroids ( mainly type I permethrin S-biomethrin , and type II beta-cypermethrin ) were the most commonly used adulticides , while organophosphates ( mainly temephos and fenthion ) were the most commonly used larvicides ( Table 3 ) . The frequency of insecticide usage in urban areas was more frequent than that in the rural district of Conghua ( Table 3 ) . Conghua in the rural area did not use the insecticide routinely . The survey also found that Bti and nicotine ( imidacloprid ) have gradually become the new choices for larvae control ( Table 3 ) . Adulticides were more frequently used than larvicides .
In this study , we characterized the current insecticide resistance in Ae . albopictus in Guangzhou , China in the following ways: ( 1 ) Increasing . In 2017 , Ae . albopictus populations in four districts were all resistant to the four tested insecticides with sensitivity only to malathion ( Fig 2 ) , whereas in 2014 , all were susceptibility but had only low or moderate resistance to some insecticide in very limited areas of Guangzhou [18 , 35] . In 2017 , all four tested districts were resistant ( Fig 2 ) . ( 2 ) Rapid . The resistance of Ae . albopictus to pyrethroids has changed from susceptibility or moderately resistant to resistant within three years , 2015–2017 ( Fig 3 ) . ( 3 ) Multiresistance . In 2014 , Ae . albopictus was resistant only to limited pyrethroids and DDT , whereas in 2017 , it was resistant to all four types of insecticides except malathion[11 , 19] . In 2016–2017 , the Mosquito and Oviposition Positive Index ( MOI ) in Guangzhou is greater than 5 ( MOI>5 is the risk of dengue virus transmission ) [36] . At the same time , in 2015 , dengue fever affected 31 towns in 8 districts in Guangzhou and 108 cases were reported; in 2016 , dengue fever expanded to 61 towns in 10 districts , and 253 cases were reported; in 2017 , dengue fever expanded to 122 towns in 11 districts and cases increased to 950 , only less than 2013 and 2014 in the past 10 years [6] . The quick generation , wide distribution , and increasing insecticide resistance to multiple agents in Ae . albopictus is bound to affect mosquito management and threaten the prevention and control of dengue , similar to the resistance in Anopheles mosquitoes has prevented the elimination of malaria [37 , 38] . Mutations in the VGSC gene have been correlated with the resistance of vector mosquitoes including Ae . albopictus [33 , 39 , 40] . The present study also found that F1534S and F1534L mutations were significantly correlated with the resistance of Ae . albopictus to deltamethrin , permethrin and DDT ( p<0 . 05 ) ( Table 1 ) . At present , pyrethroids are the most commonly used insecticides in China; therefore , sensitive and specific techniques based on the detection of 1534 mutations must be developed to monitor the resistance of Ae . albopictus to pyrethroids . However , the mechanism for insecticide resistance in mosquitoes is very complicated . In addition to the correlation of kdr mutations with resistance , the increased expression and enhanced activity of P450s have also been proven to be associated with mosquito resistance to pyrethroids in the present study and in other studies [41–43] . Only P450s changes have been reported in the resistant mosquitoes , and no kdr mutations were reported [44] . Therefore , clarifying the mechanism of insecticide resistance in vector mosquitoes and developing suitable monitoring systems , especially those easily used in the field , remain challenging . In the era of rapidly emerging and widely distributed insecticide resistance , it is important to make suitable and updated guidelines for insecticide usage . PBO is an inhibitor of monooxygenase , such as P450s [26 , 45] . The present study proved that PBO can significantly reduce the resistance of Ae . albopictus to deltamethrin by anti-P450s ( Fig 4 ) , which could be used as a synergistic agent to enhance the effect of pyrethroids . PPF is an insect growth regulator and has been used as an automatically disseminated insecticide to control habitats , especially for Aedes mosquitoes [46 , 47] . Recently , PPF has also been used to sterilize adult Anopheles mosquitoes by reducing their fecundity and longevity [48] . Unexpectedly , in the present study , the higher resistance to PPF was widely detected in Ae . albopictus populations in Guangzhou probably because of the cross resistance to pyrethroids ( Fig 2B ) . Yunta C et al . also proved that PPF is metabolized by P450s and associated with pyrethroid resistance in Anopheles gambiae [41] . Considering the higher and widely distributed resistance to pyrethroids , they should be cautious when using PPF as the larvicide or as the synergistic agent for adulticide . In mosquito incense and aerosol insecticides used daily among residents , the active ingredient is primiarly pyrethroid ( such as transfluthrin and s- bioallethrin ) . More and more residents are using mosquito nets , electric mosquito swatters or mosquito killer lamps to prevent mosquito bites . In Conghua , vegetable farmers use DDVP and beta-cypermethrin to prevent Plutella xylostella . According to our research , the Ae . albopictus populations were still sensitive to malathion , hexaflumuron and Bti . Similarly , in Brazil where pyrethroid and temephos resistance has developed , local health authorities recommend the use of malathion against adult mosquitoes and chitin synthesis inhibitors against larvae ( Controle de vetores . http://www . saude . gov . br/vigilancia-em-saude/controle-de-vetores ) . Therefore , we suggest useing malathion against adult mosquitoes and hexaflumuron or Bti against larvae for dengue vector control in Guangzhou . Timely monitoring of resistance is critical for the proper management of insecticides . Additionally , every year at the end of February or early in March , the Guangzhou government launches a patriotic health campaign to focus on cleaning up aquatic mosquito habitats for one month , which is important for cleaning the over winter eggs of Ae . albopictus and reducing the population density of mosquitoes . In conclusion , increased and more widespread insecticide resistance to multiple agents has been rapidly developing in Ae . albopictus , the primary dengue virus vector in China . Extensive applications and inapposite applications of insecticides were likely one reason for development of resistance generation . The 1534 codon mutations in the VGSC gene were significantly correlated with resistance to pyrethroids and possibly used as a biomarker to monitor insecticide resistance . PBO can significantly reduce the resistance of Ae . albopictus to deltamethrin an act as a synergistic agent of pyrethroids . Gravid Ae . albopictus does not oviposit all eggs into one place , although they used to lay eggs in different breeding sites . By treatment of a breeding site with an insecticide such pyriproxyfen ( PPF ) , gravid mosquitoes could be contacted and contaminated with PPF when they oviposit the eggs . Then , when they fly to neighborhood breeding sites to lay the remaining eggs , the contaminated PPF would be transferred automatically to cryptic habitats [49 , 50] . PPF may display cross resistance to deltamethrin , and the concentration should be cautiously considered when used as an automatically disseminated insecticide . Fast emerging resistance in Ae . albopictus raises the alarm for dengue vector control and calls for timely and guided insecticide management . | Guangzhou is the most epidemic area of dengue in China . Massive insecticides have been used to control the vector mosquito Ae . albopictus , as no specific vaccines are available for dengue . Regular monitoring of insecticide susceptibility is essential for insecticide resistance management . In this study , the insecticide resistances of Ae . albopictus in Guangzhou were comparatively analyzed from 2015 to 2017 . The results displayed that Ae . albopictus had rapidly generated high resistance to the most commonly used adult insecticide pyrethroid ( deltamethrin and permethrin ) and larvicide organophosphate ( temephos ) . The combination of malathion for adult mosquitoes and Bti or hexaflumuron for larvae might be a better choice for vector control . Resistance to deltamethrin can be significantly reduced by PBO but may generated cross-resistance to PPF . F1534S and F1534L mutations in the VGSC gene were significantly correlated with resistance to pyrethroids . This study indicated that the insecticide resistances had been generated in Ae . albopictus in Guangzhou which was correlated with the dengue epidemic responses . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"death",
"rates",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"malathion",
"animals",
"organic",
"compounds",
"developmental",
"biology",
"organophosphates",
"ddt",
"mutation",
"population",
"biology",
"insect",
"vectors",
"agrochemicals",
"infectious",
"diseases",
"life",
"cycles",
"chemistry",
"disease",
"vectors",
"insects",
"agriculture",
"arthropoda",
"insecticides",
"population",
"metrics",
"mosquitoes",
"eukaryota",
"organic",
"chemistry",
"ethers",
"gene",
"identification",
"and",
"analysis",
"genetics",
"mutation",
"detection",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"physical",
"sciences",
"larvae",
"organisms"
]
| 2019 | Fast emerging insecticide resistance in Aedes albopictus in Guangzhou, China: Alarm to the dengue epidemic |
An ultimate goal of genetic research is to understand the connection between genotype and phenotype in order to improve the diagnosis and treatment of diseases . The quantitative genetics field has developed a suite of statistical methods to associate genetic loci with diseases and phenotypes , including quantitative trait loci ( QTL ) linkage mapping and genome-wide association studies ( GWAS ) . However , each of these approaches have technical and biological shortcomings . For example , the amount of heritable variation explained by GWAS is often surprisingly small and the resolution of many QTL linkage mapping studies is poor . The predictive power and interpretation of QTL and GWAS results are consequently limited . In this study , we propose a complementary approach to quantitative genetics by interrogating the vast amount of high-throughput genomic data in model organisms to functionally associate genes with phenotypes and diseases . Our algorithm combines the genome-wide functional relationship network for the laboratory mouse and a state-of-the-art machine learning method . We demonstrate the superior accuracy of this algorithm through predicting genes associated with each of 1157 diverse phenotype ontology terms . Comparison between our prediction results and a meta-analysis of quantitative genetic studies reveals both overlapping candidates and distinct , accurate predictions uniquely identified by our approach . Focusing on bone mineral density ( BMD ) , a phenotype related to osteoporotic fracture , we experimentally validated two of our novel predictions ( not observed in any previous GWAS/QTL studies ) and found significant bone density defects for both Timp2 and Abcg8 deficient mice . Our results suggest that the integration of functional genomics data into networks , which itself is informative of protein function and interactions , can successfully be utilized as a complementary approach to quantitative genetics to predict disease risks . All supplementary material is available at http://cbfg . jax . org/phenotype .
Understanding the genetic bases of human disease has been an overarching goal of biology since the foundation of genetics as a scientific discipline . Efforts in quantitative genetics have utilized new laboratory technology to quickly genotype and phenotype large populations in order to determine which sequence features are most related to specific phenotypes . There are currently two major quantitative genetics approaches used to identify these genotype-phenotype associations [1] . First , linkage mapping examines genetically well-characterized populations , such as the progeny of the crosses of reference strains or individuals related through a known pedigree , to identify quantitative trait loci ( QTL ) that contain causal mutations . Second , genome-wide association studies ( GWAS ) can be performed on a more arbitrary population to identify common genetic factors associated with a phenotype . Hundreds of GWAS and QTL studies have been performed in humans and in model organisms , resulting in the identification of thousands of loci associated with phenotypes and diseases . Despite promising results , each of these approaches for quantitative genetics have common and unique unresolved issues that limits their utility . Both QTL and GWAS approaches can suffer from sampling biases . Population structure and proper selection of representative case and control groups are challenges for many GWAS , while linkage disequilibrium and limited genetic diversity are challenges for many QTL studies [1]–[4] . Further , many linkage mapping QTL studies lack the statistical power to narrowly define a causal loci , often resulting in regions spanning entire chromosomes that contain hundreds of candidate genes [5] . While these QTL regions are often broad , they can typically explain a large fraction of phenotypic variation . In contrast , GWAS typically define narrow regions of interest , but the amount of heritable variation explained by these loci tends to be small , possibly due to epistatic effects , rare alleles , or sampling biases [6] . For example , a meta-analysis GWAS of bone mineral density ( BMD ) based on nearly 20 , 000 genotyped and phenotyped individuals can only account for less than 3% of the observed heritability of BMD [7] . Thus , there is a strong need for complementary approaches to quantitative genetic techniques that are independent from the biases inherent in linkage mapping QTLs and GWAS . Many of the shortcomings of quantitative genetics could be attenuated by considering the functional roles of proteins and by evaluating existing large-scale experimental evidence . As such , we propose a complementary , alternative approach to discovering gene-phenotype associations that applies machine learning techniques to functional genomic measurements of the activities of genes and proteins ( e . g . expression , interactions , etc . ) to identify candidate genes that may be involved in a phenotypic outcome . Recent efforts have undertaken the task of summarizing the entirety of the experimental genomic literature into functional networks of genes . These networks typically encode genes as nodes , and contain experimental evidence of the relationships between genes as edges between nodes . Computationally exploring diverse functional genomics data through networks has been intensively studied , with the purpose of elucidating the functional roles of genes [8]–[11] , predicting physical interactions [12] , [13] , and determining pathway structures [14] . Functional networks have been produced in several organisms , including yeast [9] , [15] , worm [11] , [16] , plant [17] , mouse [10] , and human [18] . These networks have the advantage of being able to efficiently handle very-high dimensional data and allow for visual analysis of results . However , attempts to extract phenotypic information from these functional networks are limited , with only the most naïve summarization of links being applied in model organisms [11] , [17] , [19] as well as human [20] . In order to utilize functional networks to identify phenotypically important genes in higher organisms , where we face complex biology and increased data heterogeneity , more sophisticated computational approaches must be developed . Historically , the fields of quantitative genetics and functional genomics have been largely isolated from each other . A major exception to this observation is the recent development of genetical genomics and expression QTL studies [21] , [22] . These approaches use the mapping populations of traditional quantitative genetics , but utilize gene expression measurements as the phenotype to map , rather than direct physiological or disease phenotypes . These studies have begun to illuminate the regulatory programs of gene networks and have quantified the effects of genetic diversity on gene expression . However , these efforts suffer from the same technical problems as other QTL studies , and the interpretation of results is currently limited and is often disconnected from more clinically relevant phenotypes and analyses [23] . We propose that functional genomics approaches can complement the potential shortcomings of quantitative genetics results in two ways: first , by identifying candidates that may have been missed due to biases in sampling or low allele frequency; and second , by prioritizing candidates in loci containing many genes due to limited mapping resolution . Here , we adopt a state-of-the-art machine learning algorithm ( support vector machine ) to analyze the functional network of the laboratory mouse to identify genes involved in phenotypes and diseases . We show that our approach significantly out-performs previous naïve functional genomic methods used to predict phenotypes . Further , we demonstrate that our results are complimentary to quantitative genetics methods since a statistically significant number of our predictions fall within QTLs or GWAS loci , but several of our most confident predictions fall outside of these regions as well . We have experimentally validated a phenotypic role for two genes predicted to be involved in bone mineral density ( BMD ) , a risk factor for osteoporotic fracture , which were not identified by any previous quantitative genetics study ( Timp2 and Abcg8 ) . Our results concretely demonstrate that the combination of quantitative genetics and functional genomics approaches can more comprehensively associate genes with phenotypes or diseases , which may aid in identifying risk factors and potential drug targets .
Integrated functional relationship networks have the advantage of summarizing multiple complex datasets into a concise , visually interpretable graph representation where genes are nodes and connections between them represent the probability that two genes work together [8]–[11] , [15] , [17]–[19] . We applied a Bayesian network approach for supervised data integration , which assesses the conditional probability that individual data sources contain evidence for gene relationships based on a training set of positive examples ( e . g . gene pairs known to interact ) and negative examples ( e . g . gene pairs not known to interact ) . Given these learned conditional probabilities for each data source and prior probabilities for gene relationships , Bayesian inference is used to generate a network defined by pair-wise posterior probabilities of functional relationships between all genes . In order to use this network to predict genotype-phenotype associations , we must exclude phenotypic data to avoid circularity . Therefore we gathered diverse genomic data for inputs , including protein-protein physical interactions [24]–[27] , phylogenetic profiles [28] , [29] , homologous functional relationship predictions in yeast [9] , and expression and tissue localization data [30]–[32] to construct a functional relationship network for the laboratory mouse ( Figure 1; input data sources and integration methods are available in Supplemental Text S1 ) . We integrated these diverse data using a Bayesian network trained using a gold standard derived from the Mouse Genome Informatics ( MGI ) [33] annotations to the Gene Ontology ( GO ) [34] biological process branch as described previously [10] and in Supplemental Text S1 . This procedure resulted in a probabilistic functional relationship network for the laboratory mouse that combines diverse knowledge and data . Given this integrated functional relationship network , we applied two methodologies to generate hypotheses about genotype-phenotype associations . Each method is supervised , and thus requires a starting set of genes known to be associated with each phenotype or disease of interest . Here , we used the mammalian phenotype ( MP ) ontology [35] annotations to create our training sets of gene-phenotype associations . For each phenotype examined , genes were considered positive examples if any allele is annotated to the phenotype and all other genes were considered negative examples . First , our new approach treats the connection weights from the integrated network ( i . e . the inferred probabilities that genes are functionally related ) as features for support vector machine ( SVM ) classification [36] . In order to reduce the size of the feature space , only connection weights to known genes associated with a phenotype ( i . e . positive examples ) are used for training . We then apply the trained SVM to classify all genes for each phenotype ( Figure 1 ) . For ease of interpretation , the raw SVM scores were normalized to represent probabilities of gene-phenotype association as described in the Methods section . The result of this approach is used in all of our later analyses , and is denoted simply as “SVM” below . Second , in order to establish a baseline for comparison , we also explored a naïve method previously used in other model organisms [11] , [17] , [19] , as well as human [20] , which assigns a score to an unknown gene by its summed connection weights to all known genes associated with a phenotype in the integrated network . For this method , genes were ranked according to their summed connection weight , which we denote as “summed weight” throughout the text . ( In addition to these two approaches , we also trained SVMs directly on the input data used to create the functional relationship network . As detailed in Supplemental Text S2 , this approach induced a dramatic increase in feature space and running time , and was always outperformed by our method . ) Each of these approaches was applied to a set of 1157 diverse phenotypes defined by the mammalian phenotype ( MP ) ontology [35] . Predictions were computationally evaluated through bootstrapping for each method and phenotype . Performance summary statistics were calculated , including the area under the precision-recall curve ( AUPRC ) and precision at n% recall . To establish general performance measures , we first focused on 30 MP terms from the first level of the ontology , which represent a wide sampling of “high-level , ” well-characterized phenotypic areas [37] . Cross-validation performance for these 30 high-level terms revealed significant improvement in performance for our new SVM-based approach compared to the summed weight method ( Figure 2 ) . The median AUPRC for these terms using SVM is roughly 1 . 8 fold greater than for the summed weight approach ( results for a sampling of three representative MP terms are shown Figure 2B; full results for all 30 phenotypes are available in Supplemental Figure S1 ) . This performance improvement is especially apparent for our most confident predictions ( i . e . at the low-recall , high-precision end ) , which is most important for subsequent biological validation where only a handful of candidates can be reasonably examined . At 1% recall ( roughly 200 predictions ) , our SVM approach achieved a median of 75% precision , compared to 43% for the summed weight method; and at 10% recall ( roughly 2000 predictions ) our SVM approach outperforms summed weight by 40% to 15% . The comparisons of precisions at multiple levels of recall confirm the overall improved quality of our algorithm over naïve methods ( Figure 2C ) . Prediction algorithms often show drastic differences in baseline performance related to the number of training examples ( i . e . the number of genes annotated to each MP term ) [38] , which is an important factor in fully evaluating the strength of algorithms . We therefore assembled all phenotype terms into groups of 30–50 , 50–100 , 100–200 , and 200–300 annotated genes to assess the impact of term size on results . Both the summed weight and the SVM method achieved better performance than random , regardless of term size ( Figure 3A ) . However , our SVM-based method demonstrated a more significant improvement for reasonably large terms ( with more than 100 genes annotated , or >0 . 5% of the genome; shown in Figure 3B ) . For example , in the 200–300 annotation group , our SVM approach achieved an average improvement of 1 . 78 fold over summed weight , and in the 100–200 annotation group we observed a 1 . 67 fold improvement . The superior performance of our SVM-based method implies that more sophisticated machine learning techniques are better able to fully extract phenotypic information from functional networks than previous simpler approaches . As discussed above , there is a significant effect on prediction accuracy based on the number of known gene annotations to each phenotype . In addition to this size effect , there appears to be a strong correlation between prediction accuracy and the ability of a phenotype to be accurately and reproducibly measured . For example , among the phenotypes most accurately predicted by our approach are “decreased IgE level” ( MP:0002492 ) and “decreased circulating free fatty acid level” ( MP:0002702 ) . Each of these is a concrete phenotype that is measurable using an unbiased metric such as “concentration” or “width” . In contrast , among the most poorly predicted phenotypes are “head bobbing” ( MP:0001410 ) , “disheveled coat” ( MP:0001511 ) , and “lethargy” ( MP:0005202 ) , which are more qualitative in nature ( i . e . presence/absence calls ) or which are measured based on a subjective “severity score” . While it is difficult to concretely assess the notion of “phenotypic specificity , ” we generally observe in our overall results that more concrete phenotypes tend to perform more accurately . In order to quantify this effect , we conducted a small , blinded survey of 19 laboratory biologists . Survey respondents ranked phenotypes on a scale from 1 ( not specific , qualitative ) to 5 ( highly specific , quantitative ) . Based on these results , we found a small , but significant , difference between the top and bottom 20 phenotypes ranked by overall precision ( 3 . 6 versus 3 . 3; p = 0 . 008; full survey results in online supplement ) , which confirms our observation that more quantitative phenotypes tend to perform better . This phenomenon is not surprising in that functional genomic evidence is more likely to be informative for well-defined phenotypes . However , this effect is also promising in that better defined phenotypes are more likely to reflect specific molecular-level changes that may be more relevant from a drug target or clinical diagnosis perspective . Our prediction approach is based on an integrated functional relationship network , rather than pure genotype and phenotype information , and thus potentially avoids several of the caveats of quantitative genetics methods . While we expect that our approach and GWAS/QTL studies will share many candidates , since the underlying assumptions of functional genomics and quantitative genetics are very different , we also expect to obtain predictions unique to each method . We evaluated the utility of our functional genomic approach both by comparing our predictions to previous quantitative genetics loci , and by experimentally validating predictions unique to our approach . We selected bone mineral density ( BMD ) as an example to evaluate our approach since this is an extensively studied heritable trait in both human populations and mammalian model organisms [5] . We compared the predictions for the phenotype “abnormal bone mineralization” ( MP:0002896 ) from our functional genomic approach against a comprehensive list of mouse linkage QTLs and human GWAS results examining BMD [5] . Due to the limited resolution of some mouse QTL studies , fully 83% of mouse genes lie under the confidence interval of at least one BMD QTL reported in the literature ( Figure 4 ) . Despite this lack of specificity , we still observe a significant overlap between our top predictions and these QTLs as 93 of our top 100 predictions are contained within a QTL confidence region ( hypergeometric p-value = 0 . 002 ) . Similarly , genomic regions within 5cM of a QTL peak contain 16% of mouse genes , but contain 71 of our top 100 predictions ( hypergeometric p-value = 7×10−32 ) . While this overlap is significant , a large number of our most confident predictions ( ∼30% of the top 100 ) do not fall within 5cM of a QTL peak and are thus not likely candidates from previous studies . We consider these to be candidates likely missed by prior quantitative genetics studies due to sampling , population biases , epistasis , or other circumstances . In fact , two of our most confident predictions for genes associated with BMD are not candidates from previous quantitative genetics studies , but both have been experimentally verified as described below . For experimental validation , we selected two of our genes predicted for association with BMD that are not candidates from any previously reported linkage QTL or GWAS regions [5]: Timp2 and Abcg8 . These genes are our two most confident predictions for involvement in BMD that are not quantitative genetics candidates and that have existing , live knockout strains available for testing ( see Supplemental Table S3 for all top BMD predictions ) . In the laboratory mouse , the Timp2 gene falls on the distal end of chromosome ( Chr ) 11 , outside of all previous QTLs identified on Chr 11 . The Abcg8 gene lies on the distal end of Chr 17 . One previous QTL study found a BMD QTL on Chr 17 that contains over 100 genes , including Abcg8 [39] . However , Abcg8 was not considered as a possible candidate gene in this study because there is no known polymorphism within 20 kb of Abcg8 in the two strains crossed for linkage mapping ( NZB/B1NJ and RF/J ) . None of the 20 BMD loci identified in human GWAS lie near either Timp2 or Abcg8 ( Figure 4 ) . We have high confidence in the accuracy of these candidates for two reasons . First , our approach produced accurate cross-validation results for many osteoporosis and BMD related phenotypes such as “abnormal bone density” ( MP:0005007 ) and “abnormal bone structure” ( MP:0003795 ) ( Figure 5A–B ) . Second , in our functional network , Timp2 and Abcg8 are linked to several genes known to be related to BMD and bone diseases in the Online Mendelian Inheritance in Man ( OMIM ) database ( Figure 5C and Supplemental Tables S1 and S2 ) . For example , Timp2 is directly linked to Mmp2 , which is a known player in hereditary osteolysis ( OMIM # 259600 ) . It is also linked to the metalloproteinases Mmp8 and Mmp14 , the collagen Col1a1 , and the glycoprotein Sparc , all of which have been associated with bone defects in the literature [40] , [41] . Among the top interactors of Abcg8 are the collagens Col1a2 and Col1a1 , which are involved in osteoporosis ( OMIM #166710 ) . Additional interactors of Abcg8 include the bone related genes Sparc and the proteoglycan Bgn . All of these connections are supported by multiple evidence sources , including expression , physical interactions and phylogenetic profiles ( Supplemental Table S1 ) . Despite the connections apparent in our functional network , to our knowledge no bone related phenotypes have been reported for either Timp2 or Abcg8 knockout mouse strains . However , there is limited evidence in the literature that these genes may play a role in bone biology . Polymorphisms in the human Timp2 gene were found to be weakly associated with increased risk of non-vertebral osteoporotic fracture in a small study of post-menopausal women [42] . Abcg8 is involved in cholesterol absorption and serum cholesterol levels [43] , which are processes that have been related to bone homeostasis through other genes [44] , [45] . These two cases show that our integrative approach is able to draw implicit information from a variety of high throughput data to confidently associate these two proteins with BMD defects . We examined femoral volumetric BMD ( vBMD ) in male Timp2 and Abcg8 knockout mice at 16 weeks of age . Animals homozygous for deletion were compared to heterozygous littermate controls . As shown in Figure 6A , we observed a significant decrease of roughly 8% of vBMD in Timp2−/− male mice ( p-value = 0 . 033 ) and a significant increase of roughly 6 . 5% was observed in Abcg8−/− male mice ( p-value = 0 . 044 ) . Most individual mouse QTLs account for a 3–6% change in vBMD [5] , which places our observed differences in the high end of this range . We also examined phenotype results from the Deltagen and Lexicon collections of over 200 knockout mouse strains ( http://www . informatics . jax . org/external/ko/ ) to assess the likelihood of identifying genes with a significant affect on BMD . While these strains were not randomly selected , we can use these results to gain an estimate of how often single gene deletions affect bone density . Of 206 strains tested , only 20 exhibited a significant alteration in BMD , indicating a roughly 10% background rate . Thus , our confirmation of 2 out of 2 predictions is well above the expected result by chance . Furthermore , we observed morphological defects in the bones of both strains , including an increase in periosteal circumference in the Timp2−/− mice ( p-value = 0 . 0105 ) and an increase in cortical thickness in the Abcg8−/− mice ( p-value = 0 . 032 ) as shown in Figure 6BC . Osteoporotic fracture risk increases with decreasing bone mass , but morphologic factors such as bone shape also contribute to fracture risk [46] . The decrease in bone density seen in Timp2−/− and the increase density seen in Abcg8−/− , along with the noted differences in bone morphology , indicate that these genes are likely related to osteoporotic fractures . Neither of these genes is a candidate from any previous quantitative genetics study of BMD , which indicates that our approach produces results that are complementary to GWAS and QTL studies . We have developed a novel SVM-based classification method to predict genotype-phenotype associations based on a probabilistic functional relationship network integrated from diverse data sources . Through bootstrapping and cross validation we confirmed its superior performance compared to previous approaches . Using osteoporosis related phenotypes as an example , we have computationally demonstrated and experimentally validated how integration of functional genomics data can facilitate disease gene identification in a complementary manner to quantitative genetics approaches . This study demonstrates the potential for integrative functional relationship networks to be applied in new ways , especially when successfully combined with sophisticated machine learning techniques . Functional networks have been intensively studied during recent years , resulting in multiple networks available for several model organisms [8]–[11] , [15]–[19] as well as for humans [18] , [20] . In addition to the original applications of these networks , our results demonstrate that functional relationship networks can be used to accurately predict gene-phenotype associations , and that supervised machine learning approaches outperform the simple fusion methods previously applied to this problem [11] , [17] , [19] , [20] . By combining two computational learning methodologies ( Bayesian network integration of diverse data into a probabilistic relationship network and SVM classification ) , we were able to utilize the complementary advantages of each method to produce accurate results . We anticipate that such an approach could be a prototype for other forms of network-assisted prediction methods in cases where gold standard positive and negative examples are available for training . We have also shown that our integration of functional genomics data is able to identify potential disease genes not yet identified by any quantitative genetics screens . The caveats of quantitative genetics , including sampling biases , rare allele effects , epistasis , and potentially limited explanatory power have been recognized for years [1]–[4] . Our approach suggests a complementary new avenue to address some of these limitations through the analysis of existing genomics data in model organisms , which is generally not included in quantitative genetics studies . We expect that a combination of complementary approaches will be required to realize the ultimate goals of improved genetic diagnosis , treatment , and prevention that form the basis of personalized medicine .
To avoid circularity in the phenotype prediction process , we created a functional relationship network that excludes phenotypic data . We pre-computed this network following the Bayesian method described in [10] by integrating diverse sources of genome-scale data . This results in a network with genes as nodes and links between them representing the probability that the pair participates in the same biological processes . The input data used to generate this network is listed in Supplemental Text S1 and the complete network is downloadable at http://cbfg . jax . org/phenotype . We utilized the annotations to the mammalian phenotype ( MP ) ontology [35] as the gold standard gene sets . The MP ontology is organized into a multi-level hierarchy , with broader terms describing more general phenotypes at the top and more specific terms describing detailed phenotypes toward the bottom . Within this hierarchy , any annotation to a child node implies annotation to all of its parent nodes . Therefore , for each MP term , positive examples were taken as the genes annotated directly to this term or to any descendent of this term . Negative examples were assumed to be all other genes . We obtained the phenotype annotations for mouse from MGI [33] in Jan 2009 . These included 134722 entries , containing alleles for 11382 genes in total ( ∼55% of the mouse proteome ) . If any allele of a gene was annotated to a phenotype , we associated that gene with the phenotype . For each term , the number of positive examples np , corresponds to all genes known to be associated with the phenotype or its descendents in the MP hierarchy . All other genes were considered as negative examples; the total number of negative examples is denoted as nn . A summed weight approach has been applied in previous studies [11] , [17] , [19] , [20] to predict phenotypes of uncharacterized mutants in other model organisms . Variants of this approach have been used for function prediction and other forms of analysis [18] . For each phenotype , a score , f , is calculated for each gene , x , as the sum of all links between the gene and all positive examples from the gold standard:where K ( xi , x ) represents the connection weight between the genes xi and x in the probabilistic functional network . Predicting phenotypic effects by summing connection weights to positive examples has achieved satisfactory performance [11] , [17] , [19] , [20] . Although straightforward , this approach does not fully explore the predictive potential of functional networks that can be achieved by applying more principled machine learning techniques . We therefore designed a new method combining Support Vector Machine ( SVM ) classifiers with a functional relationship network to predict phenotypes associated with each mouse gene . For each phenotype , we constructed a specific feature space consisting of the network connection weights to all positive examples of the phenotype . Therefore the number of features varies across different phenotypes and is equal to np , the number of genes positively associated with that phenotype . These features were used as input vectors for a set of linear SVMs [36]:where xi is the feature vector for gene i ( i . e . the connection weight to the positive examples ) , yi equals to 1 or −1 depending on whether gene i is annotated to the phenotype term or not , p is any gene annotated to the term in study , and n is any of the other genes . Alternately , we could use the connection weight to all examples as the input vectors , regardless of whether they are positive or negative examples . However , this construction is excessively time-consuming for SVM calculations . For example , in a Lagrangian SVM , the running time is approximately O ( nN2 ) , where n is the number of examples and N is the number of features . In comparison , our approach takes only an average of 6 minutes ( approximately 100 fold reduction in time ) for a 25-round bootstrap validation to generate the prediction results for each phenotype term . Another intuitive algorithm is to directly input all original data into a linear SVM [38] . However , this approach is both time consuming due to the large number of features involved and performed less accurately than our approach ( see Supplementary Text S2 ) . In order to limit over-fitting , and because each phenotype is only associated with a limited set of genes , we consider bootstrap cross-validation to be an ideal method for estimating error rates [47] . We therefore applied bootstrap aggregation to predict genes associated with each phenotype and to estimate accuracy . Intuitively , this method trains models on a subset of genes and tests it on a different subset of genes repeatedly , thus minimizing the possibility of over-fitting and the effects of potentially mis-annotated genes . Specifically , for each iteration , genes were sampled with replacement to form a training set , and all the remaining genes form a test set . Classification values were only recorded for the test set during each iteration . The final outputs were calculated as the median of the out-of-bag values across 25 independent bootstraps , and the precision-recall curves were derived from these median values . The outputs from the SVMs represent the distances from the examples to the separating hyperplane [36] , which are not intuitive to understand . To make the value of these outputs more comprehensible , we estimated the probability of being annotated to a phenotype by fitting the output distribution of positive and negative examples with two normal distributions . According to Bayes' rule , wherewhere X is the raw output value for a gene , y represents positive examples , n represents negative examples , σy and σn are the standard deviations of the raw outputs for positive and negative examples , and μy and μn are the mean of the raw outputs for positive and negative examples . Based on these distributions we estimate the probability of a gene being associated with a phenotype as p ( y|X ) , given its observed value X . Outputs with a value lower than the average of negative examples are assigned as zero . Transforming to probability does not affect the ordering of results , and consequently has no effect on our performance evaluation metrics . We provide the complete list of predictions in terms of probability on our supporting website http://cbfg . jax . org/phenotype . To assess the performance of the phenotype predictions , we obtained precision-recall curves and summary statistics for each phenotype . We computed the precision at various recall rates as previously described [48] . Precision is defined as the number of genes correctly classified as having a certain phenotype ( true positives , TP ) divided by the total number of genes classified as having that phenotype ( TP and false positives , FP ) : Recall is defined as the percentage of genes annotated to a given phenotype that were classified as having that phenotype:where FN represents the number of false negative predictions . There is a trade off between precision and recall in that the most confident predictions are more likely to be accurate , whereas in order to achieve high levels of recall , we must accept a lower level of precision . As such , precision values are measured at many levels of recall to produce a curve . In order to produce single number summary statistics from these curves , we use the area under the precision-recall curve ( AUPRC ) as well as the precision at fixed levels of recall , including 1% , 10% , 20% and 50% . All studies and procedures were approved by the Institutional ACUC of The Jackson Laboratory . The B6 . 129S4-Timp2tm1Pds/J ( Stock Number 0008120 ) and B6 . 129-Abcg8tm1Elk/J ( Stock Number 008763 ) mice were originally purchased from the resource colonies of The Jackson Laboratory ( Bar Harbor , ME ) and colonies were maintained by pair mating heterozygous mice . After weaning , mice were maintained in groups of 3–5 in polycarbonate boxes ( 130 cm2 ) on bedding of sterilized white pine shavings under conditions of 12 hours light; 12 hours darkness . All mice used in this study had free access to water and diet for the duration of the study . Total femoral volumetric BMD ( vBMD ) and femoral geometry was assessed ex vivo by peripheral quantitative computed tomography ( pQCT ) . Specifically , mice were killed at 16 weeks of age and femurs were isolated and fixed in 95% ethanol for 14 days . Femurs were measured for density using an SA Plus pQCT densitometer ( Orthometrics , Stratec SA Plus Research Unit , White Plains , NY ) as previously described [49] . Daily quality control of the SA Plus instrument's operation was checked with a manufacturer supplied phantom . The bone scans were analyzed with threshold settings to separate bone from soft tissue and to separate cortical from sub-cortical bone . Precision of the SA Plus for repeated measurements of a single femur was previously found to be 1 . 2–1 . 4% . Isolated femurs were scanned at 7 locations at 2 mm intervals , beginning 0 . 8 mm from the distal ends of the epiphyseal condyles . Total vBMD values were calculated by dividing the total mineral content by the total bone volume and expressed as mg/mm3 . Periosteal circumference and cortical thickness measures were made at the exact midshaft of the femur . | Many recent efforts to understand the genetic origins of complex diseases utilize statistical approaches to analyze phenotypic traits measured in genetically well-characterized populations . While these quantitative genetics methods are powerful , their success is limited by sampling biases and other confounding factors , and the biological interpretation of results can be challenging since these methods are not based on any functional information for candidate loci . On the other hand , the functional genomics field has greatly expanded in past years , both in terms of experimental approaches and analytical algorithms . However , functional approaches have been applied to understanding phenotypes in only the most basic ways . In this study , we demonstrate that functional genomics can complement traditional quantitative genetics by analytically extracting protein function information from large collections of high throughput data , which can then be used to predict genotype-phenotype associations . We applied our prediction methodology to the laboratory mouse , and we experimentally confirmed a role in osteoporosis for two of our predictions that were not candidates from any previous quantitative genetics study . The ability of our approach to produce accurate and unique predictions implies that functional genomics can complement quantitative genetics and can help address previous limitations in identifying disease genes . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
]
| [
"genetics",
"and",
"genomics/genomics",
"genetics",
"and",
"genomics/functional",
"genomics",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"genetics",
"and",
"genomics/gene",
"function",
"computational",
"biology/genomics"
]
| 2010 | Functional Genomics Complements Quantitative Genetics in Identifying Disease-Gene Associations |
GDNF signaling through the Ret receptor tyrosine kinase ( RTK ) is required for ureteric bud ( UB ) branching morphogenesis during kidney development in mice and humans . Furthermore , many other mutant genes that cause renal agenesis exert their effects via the GDNF/RET pathway . Therefore , RET signaling is believed to play a central role in renal organogenesis . Here , we re-examine the extent to which the functions of Gdnf and Ret are unique , by seeking conditions in which a kidney can develop in their absence . We find that in the absence of the negative regulator Spry1 , Gdnf , and Ret are no longer required for extensive kidney development . Gdnf−/−;Spry1−/− or Ret−/−;Spry1−/− double mutants develop large kidneys with normal ureters , highly branched collecting ducts , extensive nephrogenesis , and normal histoarchitecture . However , despite extensive branching , the UB displays alterations in branch spacing , angle , and frequency . UB branching in the absence of Gdnf and Spry1 requires Fgf10 ( which normally plays a minor role ) , as removal of even one copy of Fgf10 in Gdnf−/−;Spry1−/− mutants causes a complete failure of ureter and kidney development . In contrast to Gdnf or Ret mutations , renal agenesis caused by concomitant lack of the transcription factors ETV4 and ETV5 is not rescued by removing Spry1 , consistent with their role downstream of both RET and FGFRs . This shows that , for many aspects of renal development , the balance between positive signaling by RTKs and negative regulation of this signaling by SPRY1 is more critical than the specific role of GDNF . Other signals , including FGF10 , can perform many of the functions of GDNF , when SPRY1 is absent . But GDNF/RET signaling has an apparently unique function in determining normal branching pattern . In contrast to GDNF or FGF10 , Etv4 and Etv5 represent a critical node in the RTK signaling network that cannot by bypassed by reducing the negative regulation of upstream signals .
Signaling by the secreted protein GDNF through the RET receptor tyrosine kinase ( RTK ) and the GFRα1 co-receptor plays a central role in the initiating event of kidney development , the outgrowth of the ureteric bud ( UB ) from the Wolffian duct ( WD ) into the metanephric mesenchyme ( MM ) . They are also important for the subsequent growth and branching of the UB to form the renal collecting duct system . This is apparent not only from the lack of UB development in Gdnf , Ret , and Gfra1 mutants in mice [1]–[5] and humans [6] , but also from the observation that most of the other genes whose absence causes renal agenesis are upstream regulators of Gdnf or Ret expression [7] . We have recently reported that expression in the UB of the ETS transcription factors ETV4 and ETV5 is upregulated by GDNF/RET signaling , and that Etv4−/−;Etv5−/− double homozygous mice fail to develop kidneys . Thus , the effects of GDNF/RET signaling on UB branching morphogenesis are largely transduced via ETV4 and ETV5 [8] . The mechanism by which GDNF/RET signaling induces epithelial branching remains to be fully elucidated . In the WD , it initially promotes cell movements that precede and lead to the formation of the UB [9] , and it then induces UB outgrowth from the duct [10] , [11] . In the UB tips , it increases cell proliferation [12] , [13] , a likely prerequisite for branching . Furthermore , because GDNF is capable of acting as a chemoattractant for cultured kidney cells [14] , [15] , it has been suggested that GDNF may act as a chemoattractant for UB tips in vivo , thereby promoting and patterning their branching [10] , [11] , [16] . Here , we have further investigated the role of GDNF/RET signaling by identifying conditions under which the kidney can develop in the absence of either GDNF or RET . To achieve this , we employed a null allele of Sprouty1 ( Spry1 ) , a negative feedback inhibitor of RTK signaling , which modulates the response to GDNF during kidney development . Spry1−/− mutants show a pervasive defect in the development of the ureteric tree , including formation of supernumerary buds from the WD , which develop into multiplex ureters and kidneys , and an increase in the number and diameter of UB branches in the developing kidneys [17] , [18] . The molecular mechanism of Sprouty protein function is incompletely understood . Engineered expression of Sprouty in cells leads to inhibition of signaling through the MAP kinase ( MAPK ) pathway , but effects have also been observed on the PI3K and PLCγ signaling pathways downstream of RTKs [19] , [20] . It was previously found that removing one Spry1 allele corrected the renal hypoplasia in Gdnf+/− heterozygous mice , and that removing one Gdnf allele corrected the abnormal UB branching in Spry1−/− mice . These findings demonstrated that the balance between GDNF and SPRY1 levels is critical for normal kidney development [17] , [18] . We have now further tested this idea by examining the consequences of eliminating Gdnf and Spry1 ( or all Ret and Spry1 ) . Surprisingly , such doubly homozygous mutant mice developed two large and well-formed kidneys , each with a single , normally-positioned ureter . Thus , in the absence of GDNF/RET signaling , other factors must be able to support normal UB outgrowth and extensive UB branching , but only when SPRY1 is absent . We provide in vivo , genetic evidence that FGF10 is one such factor , consistent with the previous observation that exogenous FGFs are capable of inducing budding by the Wolffian duct in organ culture [21] . However , our data also reveal that the specific pattern of UB branching is abnormal in Ret−/−;Spry1−/− and Gdnf−/−;Spry1−/− double mutant kidneys . Therefore , although endogenous FGF10 and perhaps other factors can promote extensive UB branching , GDNF appears to serve a unique role in the patterning of UB branching morphogenesis . Finally , we show that , unlike the rescue of Gdnf or Ret mutations , the lack of both Etv4 and Etv5 cannot be overcome by removing Spry1 . Thus , these two transcription factors represent a critical link in a signaling network downstream of Ret and other RTKs .
Gdnf−/− newborn ( P0 ) mice display ∼80% renal agenesis and ∼20% severe renal hypodysplasia [1]–[3] ( n = 28 ) ( Figure 1A and 1B ) ( for statistical purposes , we count each of the two potential kidneys as a separate sample [22] ) . In contrast , we found that newborn Gdnf−/−;Spry1−/− ( abbreviated GGSS ) mice displayed only 11% renal agenesis ( n = 18 ) and 89% of kidneys were normally shaped and only slightly smaller than controls ( cross-sectional area 70±11% of wild-type ) ( Figure 1E ) . Unlike Spry1−/− mice ( Figure 1D ) , GGSS newborns never showed hydroureter , although the bladder was often filled with urine ( Figure 1E ) , indicating that the ureters were correctly connected to the bladder , and thus suggesting that the site of outgrowth of the UB from the WD had been normal [23] , [24] . These observations raised the possibility that in the absence of Gdnf and Spry1 , kidney development was supported by another GDNF-family ligand , such as Neurturin , which is expressed in the developing kidney [25] . To investigate this possibility we generated Ret−/−;Spry−/− ( RRSS ) newborn mice , as RET is the common signaling receptor for all GDNF family ligands[26] . Whereas Ret−/− newborn mice display renal agenesis ( ∼70% ) or severe renal hypodysplasia ( ∼30% ) ( Figure 1C ) , 88% of the RRSS double mutants ( n = 26 ) developed fairly large and well-formed kidneys ( cross-sectional area 73±15% of wild-type ) with apparently normal ureters ( Figure 1H ) . Histological analysis of GGSS and RRSS kidneys indicated that the overall organization into renal papilla , medulla , cortex , and nephrogenic zone was essentially normal , with well differentiated collecting ducts , nephron epithelia and glomeruli in both double mutant genotypes ( Figure 1F , 1G , 1I , and 1J and data not shown ) . Consistent with these findings , many podocalyxin-positive glomeruli were observed in the cortex of double mutant kidneys , although they were reduced in number ( 52±6% of wild-type in P0 RRSS mutants ) ( Figure 2A and 2B ) . Despite their apparently functional kidneys , GGSS and RRSS mice did not survive beyond 3–4 days after birth , presumably because removing Spry1 does not correct the multiple defects in the nervous system caused by lack of Gdnf or Ret [26] . As GDNF/RET signaling is important for UB growth and branching , we crossed into the mutant backgrounds a Hoxb7/myrVenus transgene , which fluorescently labels the WD and UB lineage [27] , to visualize UB branching in vivo or in cultured kidneys . In P0 wild-type kidneys , branching UB tips are numerous and regularly spaced over the kidney surface ( Figure 3A ) . In Spry1−/− kidneys , the UB tips are likewise evenly spaced , but abnormally swollen ( Figure 3D ) . In contrast , although there were numerous UB tips on the surface of GGSS and RRSS kidneys , indicating that the UB had branched very extensively even in the absence of Gdnf or Ret , the tips were irregularly and less densely arrayed , elongated , and abnormally shaped ( Figure 3B and 3C ) . We also examined the kidneys at E15 . 5 , when branching is less complex and the kidneys are small enough to image by confocal microscopy and perform 3D reconstruction ( Figure 3E–3P ) . Volume rendering of the Hoxb7/myrVenus-positive UB tree showed that the double mutants had extensively branched , but the spacing and the branching geometry of UB tips was irregular ( Figure 3F and 3G ) . In such samples , the points where UB tips connect to nephrons could be mapped in three dimensions , and were found to be essentially normal in GGSS mutants ( Figure 2C–2J ) , indicating that the double mutant UB tips produce the factors necessary to connect to the nephrons . Higher magnification 3D reconstructions revealed a characteristic pattern of UB branching in wild-type kidneys ( Figure 3I–3M ) , where successive branch generations occur at regular intervals , mostly at right angles to the parental branches ( yellow dashed lines ) . In contrast , this regular pattern was rarely observed in GGSS or RRSS kidneys , where instead the UB tips were highly irregular in shape , orientation , and branching frequency ( Figure 3J , 3K , 3N , and 3O ) . Spry1−/− UB tips resembled the wild type , except for an increased tip diameter ( Figure 3L and 3P ) , indicating that the branching abnormalities in RRSS and GGSS are not due simply to lack of Spry1 . To examine the initial branching events , we explanted the WD , ureter and kidney at E12 . 5 . Consistent with what was observed in newborn GGSS and RRSS mutants , the ureter and kidney were nearly always present ( 88% , n = 26 and 100% , n = 16 , respectively ) . In contrast , few Gdnf−/− or Ret−/− mutants had ureter and kidney at this stage ( 20% , n = 30 and 8% , n = 12 , respectively ) . In none of the GGSS or RRSS mutants were duplicated ureters present , as they are in many Spry1−/− mutants . UB branching was somewhat delayed in the GGSS and RRSS kidneys compared to controls ( Figure 4A versus Figure 4D , 0 hours; and data not shown ) . Several of the wild type , Spry1−/− , and GGSS E12 . 5 kidneys were cultured to examine the subsequent branching events . While the GGSS kidneys branched extensively in culture , some of the tips elongated abnormally without branching ( Figure 4D , asterisks ) and some tips grew too slowly ( Figure 4D , arrowheads ) , resulting in an irregularly patterned tree . Thus , while GDNF/RET signaling is not required for the UB to undergo extensive growth and branching when Spry1 is also absent , it is necessary to impose a regular pattern on UB branching . The UB tips and trunks maintain different patterns of gene expression throughout kidney development , with many genes expressed specifically in one domain or the other [12] , [28] , [29] . Many tip-specific genes can be upregulated by exogenous GDNF , and their expression is reduced in a Ret hypomorphic mutant [8] , [12] , [30] , suggesting that GDNF/RET signaling may be required to maintain the tip-specific pattern . However , we found that three tip-specific markers , Ret , Wnt11 , and Etv4 , all of which normally require wild-type levels of GDNF/RET signaling for expression in the UB , continued to be expressed in a tip-specific pattern in GGSS or RRSS double mutants ( Figure 5A–5F ) . The trunk-specific marker Wnt7b [31] also retained its normal expression pattern in GGSS double mutants ( Figure 5G and 5H ) , indicating that the lack of Wnt7b expression in the UB tip does not require GDNF/RET signaling . Therefore , there must be other , Ret-independent mechanisms that can establish and maintain tip/trunk differences in gene expression . We next sought to determine what signaling molecule ( s ) support ureteric bud outgrowth from the WD , and subsequent growth and branching , in the absence of Gdnf/Ret and Spry1 . The observation that kidney development is rescued in Gdnf−/− or Ret−/− embryos only when Spry1 is absent suggests that the signaling responsible for the rescue must itself be negatively regulated by Spry1 . Since Sprouty genes are negative regulators of RTK signaling , the rescue most likely occurs through a RTK . According to this reasoning , FGF signaling is a strong candidate . Genetic studies in the mouse have identified FGF7 and FGF10 , signaling through FGFR2 , as important factors for normal UB branching [32] , [33]; however , the effects of Fgf7 or Fgf10 knockouts ( KOs ) are far less severe than those caused by loss of Gdnf or Ret , indicating that these FGFs play a secondary role under normal conditions . Fgf7 mRNA was not detected in the kidney before E14 . 5 [34] , whereas Fgf10 , like Gdnf , is expressed in the MM at least as early as E10 . 5 ( Figure 6A–6D ) , making Fgf10 a good candidate to participate in UB outgrowth and early branching morphogenesis . Fgf10−/− mice [35] have small kidneys at birth [33] , and we found this to be reflected in reduced UB branching during kidney development ( Figure 6E and 6F ) . The reduction in UB branching was comparable to that in kidneys lacking Fgfr2 ( or both Fgfr1 and Fgfr2 ) in the UB lineage [36] , suggesting that FGF10 is the major FGF signaling through FGFR2 in the UB . Furthermore , this defect could be corrected by deletion of one Spry1 allele ( Figure 6G ) , indicating that Spry1 negatively regulates FGF10 ( as well as GDNF ) signaling . To examine the relationship between FGF10 and GDNF in kidney development , we performed gain- and loss-of-function studies . FGF10-soaked beads placed next to the WD of E10 . 5 embryos induced the formation of multiple ectopic buds ( Figure 6H and 6I ) , as do GDNF beads [10] . To test whether FGF10 induced the ectopic buds indirectly , by up-regulating Gdnf , we performed the same experiment in Gdnf−/− embryos , but the result was similar ( Figure 6J ) . Therefore , FGF10 is capable of inducing UB outgrowth , presumably by acting directly on the WD . The role of Fgf10 was also examined by performing genetic crosses between Fgf10 and Gdnf KO mice , and examining UB formation at early stages ( E11 . 5–12 . 5 ) and kidney development in late fetal or newborn mice . Fgf10 heterozygotes always had normal ureters and kidneys ( Figure 7A and 7B ) , whereas Gdnf heterozygotes had a low frequency ( 7–10% ) of defective UB outgrowth or renal agenesis ( Figure 7A ) . However , in Fgf10+/−;Gdnf+/− double heterozygotes , 81% of the UBs were missing or severely delayed at E11 . 5–E12 . 5 ( e . g . , Figure 7A and 7C–7E ) , and 58% of kidneys were absent at E17 . 5–P0 ( e . g . , Figure 7A , 7F , and 7G ) , roughly equivalent to what is observed in Gdnf null homozygotes with normal Fgf10 dosage ( Figure 7A ) . Furthermore , although renal agenesis was rare in Fgf10 homozygotes ( 15% ) , removing one Gdnf allele ( Fgf10−/−;Gdnf+/− ) caused 100% agenesis ( e . g . , Figure 7A and 7H ) . Thus , while the consequences of deleting both Fgf10 alleles in a wild-type background are relatively mild , in a Gdnf+/− background the loss of even one Fgf10 allele causes more severe defects , and loss of both Fgf10 alleles is catastrophic , indicating that Fgf10 and Gdnf normally cooperate to promote UB outgrowth from the WD . To ask if it is FGF10 that rescues kidney development in Gdnf−/−;Spry1−/− mice , we next examined Gdnf−/−;Spry1−/− mice in which Fgf10 gene dosage was reduced . We found that removal of either one or both Fgf10 alleles resulted in 100% renal agenesis ( Figure 8 ) . These data conclusively demonstrate that FGF10 supports the extensive kidney development that occurs in Gdnf−/−;Spry1−/− mice . Expression of the ETS transcription factors ETV4 and ETV5 in the UB in vivo requires normal levels of GDNF/RET signaling , and they can also be upregulated in kidney cultures by exogenous FGF10 , suggesting that they function downstream of both RET and FGFR2 [8] . If ETV4 and ETV5 are needed to transduce both GDNF and FGF10 signals , removing Spry1 should be unable to rescue kidney development in Etv4−/−;Etv5−/− mice . In accordance with this prediction , the three triple mutant ( Etv4−/−;Etv5−/−;Spry1−/− ) mice obtained also lacked both kidneys , like Etv4−/−;Etv5−/− mice ( Figure 9 ) .
To investigate the roles of GDNF/RET signaling and negative regulation by Sprouty1 in branching morphogenesis of the Wolffian duct and ureteric bud during ureter and kidney development , we generated mice that lacked Spry1 and either Gdnf or Ret . We found , unexpectedly , that nearly all the double homozygous mutants developed two large , well organized kidneys , with normal ureters , a highly branched collecting duct system , and extensive nephrogenesis . Thus , it appears that for many aspects of ureter and kidney development , the balance between positive signaling via GDNF/RET and negative regulation via SPRY1 is more critical than the specific role of GDNF . These observations suggested that other signaling molecules , whose activity like that of GDNF is negatively regulated by SPRY1 , must be able to perform many of the functions of GDNF , but only when SPRY1 is absent . We identified FGF10 as one such factor; although knockout of Fgf10 normally has relatively minor effects on kidney development , FGF10 plays a critical role when GDNF/RET signaling is reduced or absent . Close examination of Gdnf−/−;Spry1−/− and Ret−/−;Spry1−/− double mutant kidneys revealed that while the UB branches extensively , and proximal-distal UB patterning is retained , the characteristic branching pattern is significantly disrupted . Thus , although GDNF/RET signaling is not required for UB growth or branching per se ( when SPRY1 is also absent ) , it has an apparently unique role in determining the normal branching pattern . In a wild-type background ( i . e . , in the presence of SPRY1 ) , GDNF/RET signaling is essential for the positioning and normal outgrowth of the UB from the WD . Not only does the UB usually fail to emerge in Ret−/− or Gdnf−/− mice , but when it does , its position is often abnormal , resulting in the lack of a normal connection to the bladder [37] . Furthermore , ectopic expression of Gdnf causes ectopic UBs to form along the WD [38]–[40] . This led to the model that the specific domain of Gdnf expression in the nephrogenic cord is critical for positioning the UB in the correct location [5] , [16] . However , in the absence of SPRY1 , mice lacking GDNF or RET make a normal UB that develops into a normal ureter connected to the bladder , as indicated by the absence of hydroureter . Therefore , signaling via another ligand/receptor that is also negatively regulated by SPRY1 must be able to properly position and induce outgrowth of the UB in the absence of GDNF or RET . We found that FGF10 , presumably signaling via FGFR2 , is an essential component of this alternative signaling , as removing either one or two Fgf10 alleles in a Gdnf−/−;Spry1−/− background caused failure of UB emergence , leading to renal agenesis . Moreover we showed that this FGF10/FGFR2 signaling is not dependent on GDNF signaling because FGF10 induces WD budding of the cultured mouse urogenital system , even in a Gdnf−/− embryo . The possibility remains that other factors ( other FGFs , or other signaling molecules ) are also involved in this process . It has been shown that several FGFs can induce UB outgrowth from cultured rat WD , and in this assay FGF10 had relatively weak activity whereas FGF7 and other FGFs were more active [21] . The ability of FGF7 to cooperate with or replace GDNF in this process remains to be tested genetically . Other mechanisms , such as the local inhibition of BMP4 by Gremlin1 [41] , [42] , also contribute to the normal positioning of the UB . Unlike Gdnf−/− or Ret−/− ureteric buds on a wild-type ( Spry1+/+ ) background , which grow and branch minimally if at all , the double mutant UBs ( GGSS or RRSS ) grew and branched extensively , leading to a kidney that was often close to normal in size , with an extensive collecting duct system , normal overall histoarchitecture and large numbers of nephrons connected to the collecting ducts . Therefore , GDNF/RET signaling does not have a unique ability to induce UB branching , including the predominant terminal bifurcations , nor is it required for the UB tips to induce nephrogenesis . As in the case of UB outgrowth from the WD , it appears that other factors are potentially redundant with GDNF in their ability to promote UB branching . Since loss of Fgf10 in a Gdnf−/−;Spry1−/− double mutant background eliminated initial UB outgrowth , it could not be determined to what extent FGF10 contributes to later UB branching in the absence of GDNF . However , the reduced UB branching in Fgf10−/− kidneys shows that FGF10 normally contributes significantly to UB branching , and is likely to be at least one of the factors that can promote this process in GGSS or RRSS double mutant mice . Other factors that might also be involved include HGF and EGF [43] . It was recently reported that the effects of a Ret-Y1062F point mutation , which causes renal agenesis or hypodysplasia similar to that observed in Ret knockout mice , can be rescued by removal of Spry1 [44] . The double mutant mice had kidneys of normal size , with normal glomerular number . The Y1062F mutation abolishes signaling through the PI3K-AKT and RAS-MAPK pathways , but does not affect signaling through PLC-γ or other pathways that potentially act downstream of RET ( e . g . , SRC ) [45] . The authors speculated that in the double mutants , the ERK MAPK pathway might be activated by RET via an alternative pathway involving PLC-γ , allowing kidney development to proceed normally , and they did not suggest that other signaling molecules might substitute for GDNF under these conditions . However , in our Ret−/−;Spry1−/− double null mutant mice , the ability of RET to signal through alternative pathways was eliminated , which revealed the ability of other signaling molecules , including FGF10 , to support kidney development in the absence of RET or GDNF . Based on our findings , we propose a model ( Figure 10 ) in which GDNF , FGF10 and probably other signaling molecules expressed in the MM signal through their cognate receptor tyrosine kinases in the UB epithelium to collectively promote budding from the Wolffian duct and subsequent growth and branching during kidney development . RET and FGFR2 ( and probably other RTKs ) activate a series of shared downstream signaling pathways , including RAS-MAPK , PI3K-AKT and PLC-γ-Ca++ [46] , which together support UB branching morphogenesis . Spry1 expression is upregulated by these signals , and SPRY1 then provides negative feedback by regulating one or more of the shared signaling pathways downstream of RET and FGFR . In early kidney development , GDNF is the predominant signal , while FGF10 is much weaker ( presumably due to lower expression ) ( Figure 10A ) . Expression of Etv4 and Etv5 is upregulated by these signals , thus controlling transcription of downstream genes required for UB growth and branching . Loss of Gdnf ( Figure 10B ) causes renal agenesis because in the presence of SPRY1 the level of FGF10 signaling via FGFR2 is not sufficient to produce the necessary responses , such as an appropriate level of Etv4 and Etv5 expression . Normally , loss of Fgf10 has relatively mild consequences because of the high level of GDNF signaling . When Spry1 is absent there is no brake on signaling via FGFR2 ( Figure 10C ) , and GDNF can be removed without causing renal agenesis , due ( at least in part ) to the effects of FGF10 , and to the restoration of Etv4/Etv5 expression; however , UB branching pattern is abnormal . If Fgf10 is also removed ( Figure 10D ) any remaining factors are insufficient to rescue kidney development , resulting in renal agenesis . In the absence of Etv4 and Etv5 , removal of Spry1 is unable to rescue kidney development ( Figure 10E ) . This suggests that Etv4 and Etv5 normally mediate the combined effects of several RTKs ( RET , FGFR2 and probably others ) , and therefore elevated RTK signaling due to lack of SPRY1 cannot bypass the requirement for these transcription factors . The main abnormality observed in the GGSS and RRSS double mutant kidneys was in the specific pattern of branching . Instead of the regular terminal bifurcations in wild-type kidneys , which typically occur at right angles to the previous branching event , the double mutant branching UB tips were heterogeneous in shape , spacing , orientation , branch angle and frequency of branching . The defects were distinct from those caused by loss of Spry1 alone , which causes the UB tips to swell but does not alter branch orientation or tip spacing . Therefore , it appears that these specific defects in branching pattern are a consequence of the loss of GDNF/RET signaling , and reflect a function that cannot be replaced by FGF10 or other factors present in the double mutant kidneys . How may GDNF/RET signaling influence the specific pattern of UB branching ? One possibility is that GDNF in the metanephric mesenchyme acts as a chemoattractant to direct the growth of the UB tips toward local foci of GDNF expression [10] , [14] , [16] , similar to the way in which FGF10 is thought to direct the branching of the developing lung epithelium [47] , [48] . We have previously argued against such a model for several reasons [40] . First , the distribution of Gdnf mRNA in the MM is extremely diffuse; however , it remains possible that the protein is more limited in its spatial distribution than the mRNA . Second , we found that kidneys developed rather normally in Gdnf null mice in which Gdnf was misexpressed in the UB epithelium , suggesting that it is the presence , but not the location , of GDNF that is important [40] . However , the specific pattern of UB branching was not closely examined in those mutant/transgenic mice , and it remains possible that they had subtle branching defects similar to the GGSS and RRSS double mutant kidneys . Methods to locally and precisely manipulate the pattern of Gdnf expression will be needed to better test this model . If not through chemoattraction , then GDNF/RET signaling must in some other manner influence the specific pattern of growth and branching of the UB tips .
All work on animals was conducted under PHS guidelines and approved by the relevant Institutional Animal Care and Use Committees . Ret [4] , Gdnf [3] , Spry1 [17] , Fgf10 [35] , Etv4 [49] , Etv5 [50] and HoxB7/myrVenus [27] mutant mice have been described . These mice were maintained on a mixed background ( 129S1/SvmJ:C57BL/6 ) . Embryo stage was estimated by considering noon of the day of the vaginal plug as embryonic day ( E ) 0 . 5 , and more accurate staging was determined by counting somites . PCR genotyping of mice and embryos was done as described previously [3] , [4] , [8] , [17] , [35] . Whole-mount and section RNA in situ hybridization and detection of β-galactosidase activity were performed as described previously [38] , [51] using digoxigenin-UTP-labeled anti-sense riboprobes . Newborn mice were sacrificed according to Institutional and NIH guidelines . Whole kidneys and urogenital tracts were dissected in PBS . Kidney cross-sectional area was determined from whole-mount photographs of 28 wt , 16 GGSS and 12 RRSS P0 kidneys using ImageJ . For histological analysis , 7–10 µm sections were prepared from paraffin-embedded samples fixed in 4% paraformaldehyde ( PFA ) . De-waxed sections were stained with either haematoxylin and eosin ( H&E ) or Periodic Acid Schiff ( PAS ) . To count glomeruli , five evenly-spaced sections across each kidney ( two wild-type and four RRSS mutants ) were stained with podocalyxin , and the number of glomeruli per section was averaged for each kidney . Intermediate mesoderm or metanephric kidneys were dissected from E10 . 5 to E14 . 5 embryos in PBS +Ca +Mg ( Invitrogen ) . Explants were cultured at 37°C in DMEM/F12 ( Invitrogen ) supplemented with Glutamax , 100 U/ml penicillin , 100 µg/ml streptomycin , and 10% fetal bovine serum in a 5% CO2 , humidified atmosphere at the medium-air interface on Costar Transwell filters ( 0 . 4 µm ) . After culture , explants were fixed in 4% PFA . For immunostaining , explants were incubated with goat anti-podocalyxin antibodies ( R&D Systems ) , followed by Cy2 or Cy3 anti-goat Ig ( Jackson ImmunoResearch ) . Images were captured on a Zeiss Axio Observer Z1 . For FGF bead experiments , posterior intermediate mesoderm was dissected from embryos at the 29 to 35 somite stage in Hanks Balanced Salt Solution , with 1% FBS . Two heparin acrylic beads ( Sigma ) soaked in FGF10 ( R&D Systems ) , reconstituted in PBS at 1mg/ml ) or in PBS were inserted between the WDs , and rudiments were cultured on Whatman Nuclepore Track-Etch Membrane filters ( 8 micron pore size ) in 44% F12 , 44% DMEM , 10% FBS , 1% glutamine , 1% Penstrep at the air-liquid interphase for 48–55 hours . Samples were fixed in cold 100% methanol and stained with anti-pan cytokeratin antibody ( Sigma C9687 ) . E15 . 5 metanephric kidneys were dissected in PBS and fixed overnight in 4% PFA . After clearing using FocusClear ( CelExplorer ) , kidneys were mounted in MountClear ( CelExplorer ) and scanned using a Leica LS5 confocal microscope , and 3D rendering was performed using Volocity software . | Kidney development requires the secreted protein GDNF , which signals via its cellular receptor RET to promote growth and branching of the ureteric bud , the progenitor of the collecting duct system . The transcription factors ETV4 and ETV5 regulate gene expression in response to GDNF . We report that deleting Spry1 , a feedback inhibitor downstream of RET , largely rescues kidney development in mice lacking GDNF or RET , although not in those lacking ETV4 and ETV5 . Thus , GDNF and RET become dispensable in the absence of SPRY1 , when their roles can be largely assumed by other signals and receptors , while ETV4 and ETV5 remain indispensible . We identify FGF10 as the signal responsible for kidney development in the combined absence of GDNF/RET signaling and SPRY1 negative regulation . But while the ureteric bud branches extensively in Gdnf−/−;Spry1−/− and Ret−/−;Spry1−/− kidneys , its pattern of branching is severely perturbed . This points to a unique function of GDNF in ureteric bud patterning . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"genetics",
"and",
"genomics/animal",
"genetics",
"developmental",
"biology",
"developmental",
"biology/morphogenesis",
"and",
"cell",
"biology",
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/gene",
"function",
"developmental",
"biology/organogenesis",
"developmental",
"biology/developmental",
"molecular",
"mechanisms"
]
| 2010 | Kidney Development in the Absence of Gdnf and Spry1 Requires Fgf10 |
The unfolded protein response ( UPR ) is a conserved mechanism that mitigates accumulation of unfolded proteins in the ER . The yeast UPR is subject to intricate post-transcriptional regulation , involving recruitment of the RNA encoding the Hac1 transcription factor to the ER and its unconventional splicing . To investigate the mechanisms underlying regulation of the UPR , we screened the yeast proteome for proteins that specifically interact with HAC1 RNA . Protein microarray experiments revealed that HAC1 interacts specifically with small ras GTPases of the Ypt family . We characterized the interaction of HAC1 RNA with one of these proteins , the yeast Rab1 homolog Ypt1 . We found that Ypt1 protein specifically associated in vivo with unspliced HAC1 RNA . This association was disrupted by conditions that impaired protein folding in the ER and induced the UPR . Also , the Ypt1-HAC1 interaction depended on IRE1 and ADA5 , the two genes critical for UPR activation . Decreasing expression of the Ypt1 protein resulted in a reduced rate of HAC1 RNA decay , leading to significantly increased levels of both unspliced and spliced HAC1 RNA , and delayed attenuation of the UPR , when ER stress was relieved . Our findings establish that Ypt1 contributes to regulation of UPR signaling dynamics by promoting the decay of HAC1 RNA , suggesting a potential regulatory mechanism for linking vesicle trafficking to the UPR and ER homeostasis .
In eukaryotes , folding and assembly of most membrane-bound and secreted proteins takes place in the endoplasmic reticulum ( ER ) . When proper folding of proteins in the ER is disrupted , cells turn on a protective mechanism known as the unfolded protein response ( UPR ) . In a cascade conserved from yeast to humans , the UPR is activated through the ER-resident transmembrane kinase/endoribonuclease Ire1 . In yeast , Ire1 cleaves the precursor to the RNA encoding the Hac1 transcription factor , HAC1u ( ‘u’ for ‘uninduced’ ) , at the exon-intron junctions . In subsequent steps , the 5′ and 3′ terminal cleavage products , non-canonical exons , are ligated by the tRNA ligase Rlg1 to produce the mature RNA isoform , HAC1i ( ‘i’ for ‘induced’ ) [1] , [2 and others] . Once translated , the Hac1 protein is translocated into the nucleus , where it activates the transcription of a set of genes encoding proteins important for alleviating ER stress [3] . To prevent UPR activation under normal conditions , the HAC1 intron forms a stable secondary structure by base-pairing to the 5′UTR , rendering the unspliced RNA translationally inactive [4] , [5] . This intron-5′UTR base-pairing , along with a conserved sequence in the 3′UTR , are both necessary , and together they are sufficient , for proper localization of HAC1 RNA to the ER and for Ire1 cleavage during activation of the UPR [6] . Clearly , UPR activation is tightly regulated post-transcriptionally , but the non-canonical splicing of HAC1 RNA may not be the only important control point . However , relatively few factors that interact with HAC1 RNA have been identified . We used an in vitro proteomic assay for RNA-binding to identify several novel HAC1-interacting proteins in Saccharomyces cerevisiae . The top HAC1-binding protein was Ypt1 , an essential small Rab-family GTPase and a central regulator of ER-to-Golgi transport in the secretory pathway [7] . We further found that Ypt1 associates in vivo with unspliced HAC1 in a UPR-dependent manner . YPT1 knockdown resulted in elevated HAC1 RNA levels under normal growth conditions , by reducing the rate of HAC1 RNA decay . We found that normal Ypt1 expression was required for proper attenuation of the UPR upon recovery from ER stress . Extensive genetic interactions have previously established an important functional relationship between the UPR and vesicle trafficking pathways [8] , [9] , [10] , [11] . Our results uncover a novel regulatory mechanism by which Ypt1 , a key regulator of vesicle trafficking , controls the post-transcriptional fate of HAC1 , the major transcription factor for the UPR , providing a regulatory link between these two critical pathways .
We recently described an unbiased approach to identify RNA-protein interactions in vitro on a genome-wide scale by binding RNA to protein microarrays that represent over 80% of the currently annotated S . cerevisiae proteome [12] . We used this strategy to screen for proteins that selectively bind to HAC1 in preference to total yeast RNA ( “Reference” ) . The greater the ratio of HAC1:Reference RNA fluorescent signal for a particular protein on the microarray , the greater the inferred specificity of that protein for HAC1 relative to total RNA . To avoid potential biases associated with the fluorescent dye label , we performed multiple replicate experiments , swapping the dyes used to label the HAC1 RNA and the Reference , respectively . We found five proteins with strong , consistent evidence of specific binding to HAC1 RNA , based on significantly elevated Log2 HAC1:Reference ratios ( p<10−4 , combined p-values based on triplicate data; see “Materials and Methods” for details ) . This statistical threshold represents a stringent significance cutoff of p<10−5 ( after correcting for multiple hypothesis testing ) based on a null model of independent Gaussian distributions . Two proteins , Rlg1 and Ire1 , were already known to interact with HAC1 ( Table S1 ) . Rlg1 was not represented on the microarrays we used in this study , and we did not detect any fluorescent signal from the Ire1 protein spots . We suspect that the batch purification procedure we used to prepare the protein microarrays [12] may have failed to isolate the transmembrane protein Ire1 in its functional form . The five proteins that reproducibly and specifically associated with HAC1 RNA were Ypt1 , Ypt7 , Ypt32 , Rho3 and Gis2 ( Table S2 ) . Of the five , only Gis2- a putative zinc-finger containing protein [13] , had been reported to bind to RNA [12] , [14] , [15] ( Table S1 ) . We previously found that Gis2 associates with ∼150 RNAs in vivo , including HAC1 RNA [12] . The remaining four proteins ( the three Ypt proteins and Rho3 ) , are small ras-family GTPases with roles in endo- and exocytosis [13] and no previous evidence for an RNA-binding function ( Table S1 ) . These results add to increasing evidence that RNAs might use the cell's trafficking machinery for selective , targeted delivery to specific parts of the cell [12] , [16] , [17] , [18] , [19] . Such targeted localization could in turn provide a mechanism for linking an RNA's stability and translation to the activity of a cellular system . Ypt1 , which associates with ER and Golgi membranes to control vectorial vesicle trafficking between the ER and the Golgi [20] , [21] , a process disrupted by ER stress , is a promising candidate regulator of HAC1 expression in response to ER stress . We thus chose to test whether Ypt1 indeed interacts with HAC1 RNA in vivo and if so , to investigate the nature and consequences of that interaction . To identify RNAs associated with Ypt1 in vivo , we used a yeast strain expressing Ypt1 fused to glutathione S-transferase ( GST ) at its N-terminus , to avoid disrupting C-terminal prenylation , which is necessary for proper folding and function of the essential Ypt1 protein [22] . To confirm the functionality of this fusion protein , we transformed a yeast strain carrying the ypt1-3 temperature-sensitive mutation , which displays a defect in ER-to-Golgi transport [20] , [23] , with the GST-Ypt1 plasmid . We found that even short-term induction of GST-Ypt1 expression ( 15 min ) was sufficient to rescue the ypt1-3 defect in carboxypeptidase Y ( CPY ) export from the ER ( a marker for functional ER trafficking- [24] ) ( Figure S1a , lane 10 versus lane 5 for example ) . Because expression of the GST-fusion protein was induced by galactose , we also confirmed that induction of Ypt1 expression does not appreciably affect either abundance or splicing of HAC1 RNA ( Figure S1b and S1c ) . To test whether Ypt1 associated with specific RNAs in vivo , we affinity purified the tagged protein , then used DNA microarrays to identify any co-purifying transcripts [25] . “Mock” IPs done with lysates from the isogenic untagged parental strain provided a negative control . Under normal growth conditions , the RNA most reproducibly enriched in association with Ypt1 was the unspliced isoform of HAC1 RNA ( HAC1u ) ( Figure 1a , Table S3 ) ( 5-fold enrichment computed by SAM [26] , p = 1 . 6×10−4 by one-tailed t-test ) . This result confirmed the interaction we had observed in vitro using protein microarrays . In the absence of ER stress , most HAC1 RNA is unspliced [4]; the absence of detectable signal corresponding to the HAC1 exon-exon junction thus still left open the possibility that Ypt1 might also bind to the spliced RNA . We therefore also evaluated the association under UPR-inducing conditions , in cells treated with DTT . To our surprise , the Ypt1-HAC1u association was lost under these stress conditions - a significant change from its behavior under normal growth conditions ( ∼6 . 5-fold change computed by SAM [26] , p = 1 . 3×10−2 by one-tailed t-test ) ( Figure 1b , Table S3 ) . Thus , Ypt1 associates with unspliced HAC1 RNA only in the absence of ER stress . How does this novel in vivo interaction relate to known mechanisms of HAC1 RNA regulation ? Specifically , we wondered if Ypt1 would associate with HAC1 in cells defective in UPR activation . Ire1 and Ada5 are necessary for the initial processing of unspliced HAC1 RNA upon ER stress induction . Ire1 is the nuclease that cleaves HAC1 RNA , while Ada5 , a component of the SAGA ( Spt-Ada-Gcn5-acetyltransferase ) histone acetylation complex [27] , interacts directly with Ire1 and is also required for the proper splicing of HAC1i [28] ( Table S1 ) . We investigated if deleting either protein would have an effect on the Ypt1-HAC1 interaction by purifying Ypt1 from ire1Δ or ada5Δ cells . In contrast to the results seen in wild-type cells , we found no evidence for enrichment of HAC1 RNA in association with Ypt1 in these mutant cells . The absence of HAC1 RNA enrichment in each mutant was in significant contrast to the enrichment observed in wild-type cells: ∼18-fold less ( p = <4 . 7×10−3 ) or ∼11-fold less ( p = 1 . 6×10−2 ) in ire1Δ and ada5Δ cells , respectively ( p-values computed by SAM [26] ) - Table S4 . Loss of the Ypt1-HAC1 interaction in ire1Δ and ada5Δ strains let us to consider if the two proteins physically interact with Ypt1 and/or HAC1 . We tested if Ypt1 bound Ire1 or Ada5 by immunopurification of Ypt1 and Western blotting for Ire1 or Ada5 ( see “Materials and Methods” ) . In formaldehyde-crosslinked cells , tagged Ypt1 was associated with Ada5 , but not with Ire1 ( Figure S2b ) . The Ada5-Ypt1 association was also detectable , albeit to a considerably lesser extent in the absence of crosslinking ( Figure S2c ) . This raised the possibility that Ada5 could be a component of the Ypt1-HAC1 complex . However , when we purified Ada5 protein from cells and tested for interaction with HAC1 RNA , we found no significant HAC1 enrichment ( Table S5 ) . Our results , therefore , are inconsistent with Ada5 or Ire1 stably associating with Ypt1-bound HAC1 . This suggests that Ire1 and Ada5 may function to recruit Ypt1 and HAC1 RNA in proximity to each other , and this recruitment may be required for proper formation of the GTPase-RNA complex ( see “Discussion” section ) . What role , if any , does the association between Ypt1 and HAC1 RNA play in regulating HAC1 expression ? YPT1 is an essential gene [7] . We therefore used a strain from the yeast hypomorphic allele ( DAmP ) library to determine whether reduced expression of YPT1 has any effect on HAC1 RNA levels [29] . We compared global RNA expression patterns in ypt1-DAmP and wild type strains using DNA microarrays . As expected , YPT1 transcript levels were significantly reduced in the ypt1-DAmP mutant ( ∼2 . 5-fold reduction , p = 7 . 4×10−8 by t-test ) , confirming successful knockdown ( Table S6 ) . Notably , levels of both unspliced and spliced HAC1 transcripts were significantly higher ( ∼2-fold increase , p = 1 . 2×10−6 by t-test ) in the knockdown strain , pointing to a functional role for Ypt1 in regulation of HAC1 ( Figure 2a , Table S6 ) . In the absence of ER stress , however , this increase in spliced HAC1 was not sufficient to induce the UPR , as reflected by the expression of key UPR target genes relative to all genes ( p = 0 . 1 by t-test ) . This could be explained if the HAC1i mRNA is not translated as efficiently under these circumstances , or if the overall increase in HAC1i levels is unequally shared among the different cells in the population , such that induction of UPR targets in a subset of cells is obscured by the absence of UPR activation in the rest of the population . Alternatively , the ∼2-fold increase in HAC1i expression may not be enough to surpass a threshold level for eliciting UPR . Increased HAC1 transcript levels in the ypt1-DAmP cells could reflect a direct effect of Ypt1 on expression of the HAC1 RNA or an indirect effect: impaired export of proteins from the ER , leading to activation of the UPR . Indeed , when we examined the ability of CPY to transit from the ER to the vacuole by Western blotting , we found a higher fraction of CPY precursor in ypt1-DAmP compared to wild type cells ( Figure S3 , lanes 1 and 3 ) , a hallmark of defect in ER export . In order to assess the possibility of indirect effect of trafficking block on HAC1 RNA levels , we compared global RNA expression patterns in ypt1-DAmP and another ER export mutant , sec12-DAmP . Sec12 is a guanine nucleotide exchange factor , required for COPII vesicle formation at the ER [30] , [31] ( Table S1 ) . Similar to Ypt1 , intact Sec12 is essential for proper ER-to-Golgi transport [30] , which we verified by measuring CPY precursor accumulation in sec12-DAmP mutant cells ( Figure S3 , lane 2 ) . The sec12-knockdown strain exhibited a more severe block in ER export ( Figure S3 , compare lanes 1 and 2 ) that also led to UPR activation ( Table S6 and Figure 2b , p = 1 . 1×10−2 by t-test ) . However , in contrast to their significant elevation in ypt1-DAmP strains , levels of HAC1u remained unchanged in sec12-DAmP cells ( Figure 2b , p = 0 . 1 by t-test ) . Thus , the difference in phenotype between the two DAmP strains suggests that the changes in HAC1u/i expression in the ypt1-DAmP strain were due specifically to impairing the Ypt1-HAC1 association . Our next goal was to determine what accounts for the effect of Ypt1 on HAC1 transcript levels . Elevated HAC1 RNA levels observed in the ypt1-knockdown strain could stem from an increase in transcription , enhanced RNA stability , or both . Previous reports have shown that HAC1 transcription is positively autoregulated by the Hac1 protein: Hac1p binds to UPR elements ( UPRE ) present in its own promoter [32] . Despite the relatively elevated levels of HAC1 RNA , the basal concentration of Hac1p was not increased in the ypt1-DAmP strain ( p = 0 . 4 by t-test ) ( Figure S4 ) , consistent with the lack of UPR induction under normal conditions in the knockdown ( Figure 2 ) . It is still conceivable that HAC1 transcription could be affected by Ypt1-dependent factors other than Hac1p itself . Indeed , independent mechanisms allow restricted cell survival under ER stress in mutants lacking HAC1 UPRE sites [32] . Thus , a Hac1p-independent change in promoter activity was still a feasible explanation for the difference in RNA expression in the YPT1 mutant strain . To measure the activity of the HAC1 promoter , we generated a HAC1 transcription reporter gene by fusing the HAC1 promoter region to the green fluorescent protein ( GFP ) coding sequence . We did not observe significant differences in the amounts of GFP RNA produced between ypt1-DAmP and wild type ( p = 0 . 9 by t-test ) - Figure 3a . Although we cannot exclude the possibility that HAC1 transcription may be influenced by intragenic or distal regulatory elements not present in the construct , the data suggest that increased transcription is unlikely to account for increased HAC1u abundance in the mutant strain . Another possibility to explain the overall increase in HAC1u expression was an altered decay rate . To evaluate RNA stability , we treated mutant and wild type cells with thiolutin to inhibit transcription [33] , , and measured RNA levels before and after drug addition by quantitative RT-PCR . We estimated half-life by comparing HAC1u levels after 30 min treatment with thiolutin to its pre-treatment abundance . We found that both unspliced and spliced HAC1 isoforms decayed more slowly in the YPT1 mutant ( Figure 3b and 3c ) . We estimated the half-life of HAC1u in wild type cells to be 19 min , consistent with previous reports [10] , [35] . In the DAmP strain , however , the rate of HAC1u RNA decay was markedly reduced ( p = 1 . 7×10−2 by one-tailed t-test ) to an estimated half-life of 37 min . This two-fold difference in HAC1u RNA stability is sufficient to account for the ∼2-fold difference we had measured in steady-state RNA expression levels ( Table S6 ) . Thiolutin treatment triggers ER stress , which probably accounts for the increase in HAC1i levels following drug addition in both the wild type and ypt1-DAmP strains ( Figure 3c ) . If Ypt1 normally promotes HAC1 RNA decay , abolishing the Ypt1-HAC1 interaction should impair HAC1 RNA stability . Since we found that the HAC1 RNA-Ypt1 association was lost in ire1Δ and ada5Δ strains ( Table S4 ) , HAC1 RNA might be expected to be more stable in these mutants . Indeed , based on assaying decay following thiolutin treatment , we found that the half-life of HAC1 RNA was 1 . 6 and 2 . 7 times longer in the ada5Δ and ire1Δ strains , respectively , than in the corresponding wild-type strain ( p = 5 . 0×10−2 and 1 . 3×10−2 by t-test , respectively ) - Figure S5 . These results imply that Ypt1 protein controls HAC1 expression by accelerating the decay of HAC1 RNA . To begin to assess the physiological significance of the Ypt1-HAC1 interaction , we evaluated the effect of a partial loss of Ypt1 on the dynamics of the ER stress response . First , we tested whether the ypt1-DAmP mutation impaired growth in the presence of tunicamycin ( a drug that impairs glycosylation and thus proper protein folding in the ER ) . We found no appreciable growth defect in the mutant compared to wild type ( Figure S6a ) . To more sensitively evaluate the potential role of Ypt1 in the dynamics of the UPR , we tracked the mRNA dynamics of four canonical UPR targets - KAR2 , ERO1 , PDI1 and HAC1i ( Table S1 ) - in DTT-treated cells , by quantitative RT-PCR . The levels of KAR2 , ERO1 and PDI1 did not differ significantly between the YPT1 mutant and wild type cells in the first 20 min after DTT addition ( Figure S6b–S6d ) . The expression of spliced HAC1 RNA , however , was 1 . 7-fold higher in the DAmP strain than in the wild-type cells after 20 min of DTT exposure ( p = 4 . 1×10−3 by t-test ) ( Figure S6e ) . Hac1 protein levels as well were ∼1 . 5 times higher in the mutant compared to wild-type cells after 20 min of DTT exposure ( p = 6×10−2 by t-test ) , based on quantitative Western blot analysis ( Figure S6f ) . Thus , although the HAC1 RNA-Ypt1 association did not appear to impair initiation of the UPR ( up to 10 min ) , the later phases ( beyond 10 min after UPR induction ) - possibly including the kinetics of recovery - differed between mutant and wild type . To examine the role of Ypt1 in attenuating the ER stress response , we induced UPR with DTT for one hour in Ypt1 knockdown or wild-type cells , respectively , then transferred cells to fresh media and monitored levels of HAC1i and KAR2 RNA for another hour by quantitative RT-PCR . Transcripts of both genes persisted at significantly higher levels in the ypt1-knockdown strain than in the wild type cells - HAC1i abundance levels were ∼3 times higher ( p = 2 . 2×10−5 by t-test ) and those of KAR2 RNA were ∼1 . 3 times higher ( p = 9 . 0×10−3 by t-test ) in the mutant , an hour after transfer to non-inducing media ( Figure 4a and 4b , respectively ) . By 3 hours after removal of DTT , expression of HAC1i had returned to near-basal levels in both wild-type and mutant cells ( Figure S7 ) . We speculate that delayed attenuation of the UPR in ypt1-knockdown cells ( Figure 4 ) could point to a key role played by Ypt1 in aiding cellular recovery from ER stress .
UPR activation is a tightly regulated process [6] , [36] , [37] , in which recruitment of HAC1 RNA to the ER followed by non-canonical splicing/ligation [1] , [38] , [39] is required for proper cascade initiation . We identified five proteins that specifically associated with HAC1 RNA in vitro by a proteomic assay . Other proteins that consistently ranked highly for HAC1 RNA association in this assay ( Table S2 ) , but that we have not yet investigated further , may also have functionally significant interactions with HAC1 RNA . Remarkably , three of the five proteins we found to interact with HAC1 RNA were ras-superfamily GTPases with roles in endocytosis and exocytosis [40] , [41] , [42] , [43]; none had previously been implicated as components of RNA-protein complexes . Recent reports show that a number of enzymes , a significant fraction of which participate in vesicle trafficking , have “moonlighting” roles as RBPs [12] , [15] , [44] , [45] , [46] , [47] . In light of these findings , the possibility that Rab GTPases might moonlight as RBPs to regulate HAC1 is intriguing . As we focused the present study on Ypt1 , it remains to be determined whether and how Ypt7 and Ypt32 ( Table S2 ) might affect HAC1 expression . We confirmed that the in vitro binding screen reflected an in vivo association between Ypt1 and HAC1 RNA . The Ypt1-HAC1 interaction was disrupted by conditions that trigger ER stress ( Figure 1 ) . The interaction had functional consequences– knocking down expression of Ypt1 led to reduced HAC1 RNA decay and , consequently , higher HAC1 RNA expression levels ( Figure 2 and Figure 3 ) . Moreover , we found that UPR kinetics were distinctly abnormal in YPT1-deficient cells ( Figure 4 ) , establishing a physiologically significant role for Ypt1 in the regulation of this critical stress response . Even though Ypt1 consistently and significantly interacted with HAC1 RNA both in vitro and in vivo , it remains possible that the Ypt1-HAC1 interaction could be indirect . Even in the in vitro protein microarray experiment , a distinct HAC1 RNA-binding protein could have remained associated with Ypt1 in the high-throughput purification procedure used to prepare the protein microarrays [12] . The in vivo IPs required the use of chemical crosslinking , which might preserve protein complexes responsible for the association . We evaluated the possibility that Ire1 or Ada5 could be RBP adaptors present in the Ypt1-HAC1 complex . However , Ire1 did not physically associate with Ypt1 ( Figure S2 ) , and Ada5 did not interact with HAC1 RNA ( Table S5 ) . It is also possible that the Ypt1-HAC1 interaction is direct , but weak and/or transient . For example , complex formation with HAC1 RNA could be localization-dependent ( see below ) . Because Ypt1 can toggle between two nucleotide states- GDP or GTP-bound , it is feasible that only one of these conformations binds RNA . A direct but transient interaction , and/or one that is nucleotide- or localization-dependent could be hard to detect reproducibly in the absence of crosslinking . Given the importance of subcellular localization to the functions of both Ypt1 and HAC1 RNA , there are compelling reasons to consider whether their interaction may occur specifically at the ER . First , previous reports have shown that a substantial fraction of Ypt1 is localized on the ER membrane [20] . Second , we found that an intact HAC1 3′UTR , which is required for proper ER localization [6] , is also necessary ( Table S4 ) but not sufficient ( data not shown ) for the Ypt1-HAC1 interaction . Last , our data establish that IRE1 and ADA5 are essential for the association . Ada5 physically binds to Ire1 [28] , and Ire1 is an integral ER protein that is necessary for proper ER-localization of HAC1 RNA [6] , [28] . Therefore , we propose that Ire1 may recruit unspliced HAC1 RNA to the ER , and Ada5 may recruit Ypt1 in proximity to Ire1 , thus enabling Ypt1-binding to HAC1 ( Figure 5 ) . What characteristics of Ypt1 make it suitable for its newly identified role in UPR regulation ? An association between HAC1u and ER-Golgi transport vesicles ( and perhaps also transport vesicles further downstream in the secretory pathway ) could provide an efficient way to recruit the HAC1 RNA away from the ER-localized UPR splicing machinery in the absence of stress ( Figure 5 ) . Since Ypt1 orchestrates multiple steps of ER-to-Golgi transport , including budding of ER vesicles [48] and their subsequent fusion with the Golgi [20] , [23] , as well as generation of vesicles at the Golgi and their docking at the ER [49] , it would be a natural candidate for modulating a potential interaction between HAC1 and ER-Golgi transport vesicles to divert any ER-proximal HAC1 RNA away from Ire1 during normal growth ( Figure 5 ) . The idea of the vesicle machinery playing a role in RNA localization is not far-fetched: studies done in two diverse systems have implicated the Rab11 GTPase- important for recycling of cell surface proteins [18] , [50]- in RNA localization , and anchoring [16] , [17] , [19] . Ypt1-dependent regulation of HAC1 RNA stability establishes a regulatory link that may underlie known functional relationships between the UPR and vesicle trafficking pathways . Past studies have shown that defects in the secretory pathway induce the UPR and that a functional UPR is required for cell survival under these conditions [8] , [10] . Furthermore , constitutive activation of the UPR can rescue growth defects of vesicle trafficking mutants [8] , [9] , [11] . These important findings have been largely interpreted as reflecting a reactive process , whereby disruption of ER export causes accumulation of unfolded proteins , which in turn triggers the UPR [11] . Our results raise the possibility that deficiencies in ER trafficking may also interfere with Ypt1-mediated control of HAC1 expression , thus potentiating the UPR via a proactive regulatory mechanism . Ultimately , active regulatory events that enable communication between the UPR and vesicle trafficking pathways may contribute to proper cellular homeostasis in response to ER stress . The present study describes a novel mode of post-transcriptional regulation of the HAC1 RNA through association with a Rab-family GTPase . The regulatory logic and molecular mechanisms of Ypt1-dependent decay and the specific role of each identified component ( including Ada5 , Ire1 , Ypt1 , and HAC1 ) in mediating crosstalk between the UPR and vesicle trafficking systems are new avenues for further investigation .
Yeast strains and their corresponding genotypes are listed in Table S1 . All commercially available Saccharomyces cerevisiae strains ( GST-tagged , knockdowns and knockouts ) were purchased from Open Biosystems . The ypt1-3 and the corresponding wild-type strains were a gift from Charles Barlowe ( Hanover , NH ) . Lithium acetate transformations were performed according to standard protocols [51] . For generating strains used in experiments testing ADA5 and IRE1 requirement , GST-YPT1 plasmid was isolated from the Open Biosystems strain and transformed into ire1Δ or ada5Δ . For experiments testing HAC1 promoter activity , the ∼1000 bases upstream of the annotated start site were fused to the GFP coding sequence and the ACT1 3′UTR and ligated into pRS315 [52] . For experiments measuring HAC1 decay in knockout mutants , HAC1 splice junction mutant ( HAC1-G1137C ) [53] was ligated to plasmid pRS406 and integrated into the Ura locus of ada5Δ , ire1Δ , and hac1Δ strains . For Ire1-Ypt1 and Ada5-Ypt1 co-purification experiments , we initially wished to use the commercially available TAP-tagged strains , but found that , consistent with previous observations [6] , Ire1 did not tolerate a large tag on its C-terminus ( data not shown ) . Therefore , we generated HA-tagged strains for the proteins ( IRE1 , ADA5 , MRS6 , PUF5 ) . The 3xHA tag sequence was PCR-amplified from the pYM1 plasmid [54] and integrated into each corresponding locus . Strains were subsequently transformed with the GST-YPT1 plasmid . HA-tagged Ire1 was functional ( Figure S2a ) . All microarrays were scanned with GenePix Pro 6 . 0 ( Molecular Devices ) . Data were deposited on Stanford Microarray Database ( http://smd . stanford . edu/ ) and GEO ( www . ncbi . nlm . nih . gov/geo ) under accession number GSE33751 . Samples were prepared and binding reactions were performed as described in [12] . Full length HAC1 RNA , including UTRs ( annotated by [61] ) and intron , was transcribed in vitro using the Epicentre AmpliScribe T7 Aminoallyl-RNA Transcription Kit ( #AA5025 ) and then Cy-dye labeled ( GE Healthcare , Cat# RPN5661 ) . Total yeast mRNA labeled with a different Cy-dye was used as a “Reference” [12] . All RNA samples were refolded by incubating at 70°C for 5 min and then cooling on ice prior to protein microarray binding . For each protein , a mean-normalized Log2 ( HAC1/Total mRNA ) ratio was calculated in order to identify specific HAC1-interactors . Six replicate experiments were performed , including three replicates using Cy5-labeled HAC1 RNA vs Cy3-labeled Reference RNA , and three additional “dye-swap” replicates with Cy3-HAC1 vs Cy5-Reference . For each experiment , proteins that failed to give detectable fluorescent signal above 1 . 5 times the background in either the Cy5 or Cy3 channels were filtered out and excluded from further analysis . Each Log2 ratio for the remaining proteins was converted into a ‘z-score’ by subtracting the mean value and dividing by the standard deviation . The computed z-scores were used to calculate p-values reflecting the significance of the specific interaction with HAC1 RNA for each protein based on the Gaussian distribution . For each set of “dye-swap” experiments , a single combined p-value was computed as the product of the p-values for its three replicates . Proteins with combined p-values satisfying p<10−4 for both sets of dye-swap experiments were classified as high-confidence HAC1 interactors ( Table S2 ) . We acknowledge that proteins that did not meet our stringent threshold criteria for specific HAC1-binding , but still ranked at the top in Table S2 , may also represent real HAC1-interacting proteins . Open Biosystems haploid yeast strain containing GST-tagged Ypt1 was grown first in SC-Ura overnight and then diluted into SC-Ura/2% raffinose . Cells were grown to OD600∼0 . 4 and protein expression was induced with 4% galactose for 2 cell divisions . Formaldehyde was added for the last 5 min to a final concentration of 1% . For experiments , in which UPR was induced , DTT was added to a final concentration of 10 mM for the last 50 min of growth , while formaldehyde was added for the last 5 min . Cells were washed twice with Buffer A ( 50 mM Tris pH 7 . 4 , 140 mM NaCl , 1 . 8 mM MgCl2 , 0 . 1% NP-40 ) and lysed using a Cryogenic grinder ( Retsch ) . Lysed cells were resuspended in Buffer B ( Buffer A supplemented with 0 . 5 mM DTT , 40 U/ml RNase Inhibitor , 1 mM PMSF , 0 . 2 mg/mL heparin , protease inhibitor complete tablet from Roche ) and sonicated . Lysates were cleared by centrifugation for 10 min at 8 , 000 rpm/4°C . Anti-rabbit Dynabeads M-280 from Invitrogen ( Cat# 11203D ) were prepared by washing three times for 3 min each at RT with 1 volume 1X PBS/0 . 1% BSA . Beads were resuspended in 1 volume 1X PBS/0 . 1% BSA . Rabbit anti-GST antibody from Open Biosystems ( Cat# CAB4169 ) was added at a ratio 100 ug antibody per 10 mgs beads and incubated overnight at 4°C on a rotator . Excess antibody was removed by washing with 1 volume 1X PBS/0 . 1% BSA three times for 15 min each at 4°C . Beads were resuspended to original volume in Buffer B . Lysates were incubated with 500 ul anti-GST conjugated beads per 1L original cells for 2 hrs at 4°C on a rotator . A fraction ( ≤300 ul ) of the depleted supernatant served as a reference ( see below ) . We took advantage of the formaldehyde crosslinking , which allows for more stringent washes . Different salt and detergent concentrations were tested to find optimal conditions that would minimize indirect interactors without affecting GST tag folding . Beads were washed twice in 1 . 5 ml Buffer B for 10 min/4°C/rotator each , then once in 1 . 5 ml “high salt” ( 2M NaCl ) for 10 min/4°C/rotator , once in 1 . 5 ml “high detergent” ( 2M urea ) for 10 min/4°C/rotator , and finally twice in 1 . 5 ml Buffer C ( 50 mM Tris pH 7 . 4 , 140 mM NaCl , 1 . 8 mM MgCl2 , 0 . 01% NP-40 , 10% glycerol , 1 mM PMSF , 0 . 5 mM DTT , 40 U/ml RNase Inhibitor , protease inhibitor complete tablet ) for 10 min/4°C/rotator each . After washing , beads were resuspended in 300 ul Buffer C supplemented with 1% final SDS and heated at 70°C for 45 min with constant mixing to de-crosslink samples and elute antibody-bound protein-RNA complexes from the beads . Cells for “Mock IPs” from isogenic parental untagged strain BY4741 were grown in synthetic media with 2% raffinose , treated with galactose , crosslinked , lysed , and incubated with anti-GST beads following the same procedure . The same protocol was used for all other GST-Ypt1 containing strains and respective mocks . RNA from both depleted supernatant ( “Reference” ) and eluted fraction ( “IP” ) was isolated using Purelink RNA Mini Kit from Invitrogen ( Cat # 12183018A ) . Up to 5 ug of RNA were amplified with Ambion's Aminoallyl MessageAmp II aRNA Kit ( Cat# AM1751 or AM1819 ) and labeled with Cy5 ( “IP” ) or Cy3 ( “Reference” ) according to manufacturer recommendations ( GE Healthcare , Cat# RPN5661 ) . Samples were prepared and bound to oligonucleotide microarrays as described previously [55] . For growth under normal conditions ( i . e . in the absence of UPR ) , we performed a total of 6 GST-Ypt1 IPs and 6 corresponding Mocks ( total 12 ) . Data were median-centered on an array-by-array basis and “Log2Ypt1-IP Enrichment” for each gene was calculated as:We filtered for genes that had signal 1 . 5 times greater than background in either Cy3 or Cy5 channel in ≥9 of 12 experiments and ranked genes based on their “Ypt1 Enrichment” signal . Probes spanning the unspliced isoform of HAC1 gave consistently high signal with p = 1 . 6×10−4 by unpaired t-test ( Table S3 ) . For UPR experiments and IPs with knockout strains , 2 replicates of each Ypt1 IP and Mock IP were performed . “Log2Ypt1-IP Enrichment” was calculated as described above for genes that had signal in >50% of the replicates . p-values for under-enrichment in the UPR IP were obtained by one-tailed t-test comparing “Log2Ypt1-IP Enrichment” values for the HAC1 probes in the UPR to the normal IPs . p-values for under-enrichment for mutant strains were calculated based on FDR values from pairwise SAM analysis [26] of wild-type versus each of the knockout IPs . For experiments testing growth phenotype on tunicamycin plates , dilution series of ypt1-DAmP [29] , ire1Δ and BY4741 strains were plated on YPD or YPD+0 . 5 ug/ml TM according to standard plating assay protocols . For UPR induction time course experiments , ypt1-DAmP strain and isogenic wild-type BY4741 strain , respectively , were grown in YPD to OD600∼0 . 7 . A sample was taken out ( “uninduced” ) , and DTT was added to a final concentration of 10 mM to the remaining cells . Samples were taken out at 5 , 10 , and 20 min , cells were collected by vacuum filtration and quick-frozen . RNA was isolated with hot phenol [56] and reverse transcribed with a mix of oligo ( dT ) and a random nine-mer primer . qPCR was performed with primers for KAR2 , ERO1 , PDI1 , spliced HAC1 , and ACT1 ( normalization control ) . Experiments were performed in duplicate , actin-normalized data were averaged for each strain , and one-tailed unpaired t-test analysis was performed to compare the measurements at each time-point between strains . For UPR attenuation time course experiments , ypt1-DAmP strain and isogenic wild-type BY4741 strain were grown in YPD to OD600∼0 . 7 . A sample was taken out ( “uninduced” ) , and 10 mM DTT was added to remaining culture . One hour after induction , a second sample was taken out ( “pre-wash” ) , and the rest of the culture was re-suspended in fresh YPD media . Samples were collected at indicated times , RNA was extracted with hot phenol and qPCR was performed and analyzed analogous to UPR induction experiments ( see above ) . Experiments were done in quadruplicate for each strain . For ypt1-DAmP and sec12-DAmP experiments , 50 ml of each DAmP and BY4741 were grown in YPD to OD600∼0 . 7 . Cells were collected , lysed and RNA was extracted with hot phenol . Samples for microarray analysis were prepared as described before [55] using 30 ug total RNA as starting material . “Uninduced” DAmP cDNA was labeled with Cy5 and “uninduced” BY4741 cDNA was labeled with Cy3 . YPT1 mutant and parental BY4741 cells grown to OD600∼0 . 7 in duplicate . For quantifications done under normal condition , a portion of the cells was harvested by centrifugation prior to drug addition . For UPR expreriments , the remaining cells at OD600∼0 . 7 were treated with 10 mM DTT for 20 min and harvested by centrifugation . Lysates were boiled in 4X Sample Buffer ( Biorad , Cat# 161-0791 ) and loaded on a gel . Anti-Hac1 antibody , a generous gift from Dr . Peter Walter ( San Francisco , CA ) , and anti-Gapdh ( Abcam , Cat# ab93378 ) were used at 1∶2 , 000 dilutions . Staining was done overnight at 4°C and appropriate secondary HRP-conjugated antibodies were used to detect protein levels . ImageJ [57] was used for precise quantification; p-values were determined by one-tailed unpaired t-test . To evaluate potential effects of knocking down Ypt1 on HAC1 transcription , ypt1-DAmP or isogenic BY4741 wild type were transformed with pRS315 bearing GFP under the control of the HAC1 promoter ( see “Yeast strains” above ) . Cells were grown in YPD to OD600 = 0 . 6–0 . 7 and harvested . Total RNA was extracted with hot phenol [56] and reverse transcribed with a mix of oligo ( dT ) and a random nine-mer primer . qPCR was performed with primers for GFP or ACT1 ( control ) . Experiments were performed in triplicate , actin-normalized data were averaged for each strain , and unpaired t-test analysis was performed to compare the measurements between strains . Immediately prior to drug addition , samples were removed ( “Untreated” ) . Thiolutin ( from fresh 1 mg/ml stock in DMSO ) was added to 3 ug/ml final concentration to cells at OD600∼0 . 8 for 30 min ( “Treated” ) . Cells were collected by rapid filtration and RNA was extracted with hot phenol . Samples for quantitative RT-PCR were prepared by reverse transcription of total RNA with oligo ( dT ) /random nine-mer primer mix followed by qPCR with Taqman probe specific for spliced HAC1 or primers recognizing the unspliced HAC1 isoform . GAPDH levels were used to normalize data and unpaired t-test analysis was performed to compare the measurements . RNA half-lives were determined with the formula: Half-life = ( t1−t0 ) /Log2 ( X ( t1 ) /X ( t0 ) ) , where X ( t ) is the expression level at time t following thiolutin treatment . Since HAC1 splicing is abolished in the ADA5 and IRE1 knockout mutants [6] , [28] , we generated ada5Δ , ire1Δ , and hac1Δ strains with integrated “unspliceable” HAC1 variant ( HAC1-G1137C ) ( see “Yeast strains” subsection above and Table S1 ) . The purpose was to rule out any non-specific effects of HAC1 splicing and UPR induction post-thiolutin treatment on HAC1 stability . | The unfolded protein response ( UPR ) , which allows eukaryotic cells to cope with stresses that impair their ability to properly fold and assemble their membrane and secreted proteins , is implicated in many human diseases such as diabetes , neurodegeneration , and cancer . In yeast , the HAC1 gene encodes a transcription factor that plays a central role in regulating the UPR . By using protein microarrays to screen the yeast proteome , we discovered that Ypt1 , a member of the Rab family of small regulatory GTPases , specifically interacts with the HAC1 RNA . Further investigation revealed that Ypt1 associates with HAC1 RNA under normal conditions , but not when the UPR is activated . The interaction with Ypt1 regulates the stability of HAC1 RNA and plays a significant role in shaping the kinetics of the UPR . These findings provide new insight into a system with a critical role in defending cells against stress . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cellular",
"stress",
"responses",
"mechanisms",
"of",
"signal",
"transduction",
"gene",
"regulation",
"rna",
"stability",
"gene",
"function",
"crosstalk",
"biological",
"transport",
"molecular",
"genetics",
"stress",
"signaling",
"cascade",
"gene",
"expression",
"biology",
"molecular",
"biology",
"systems",
"biology",
"biochemistry",
"rna",
"signal",
"transduction",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"molecular",
"cell",
"biology",
"metabolism",
"genetics",
"and",
"genomics",
"signaling",
"cascades"
]
| 2012 | The Yeast Rab GTPase Ypt1 Modulates Unfolded Protein Response Dynamics by Regulating the Stability of HAC1 RNA |
Spatial arrangement of neurite branching is instructed by both attractive and repulsive cues . Here we show that in C . elegans , the Wnt family of secreted glycoproteins specify neurite branching sites in the PLM mechanosensory neurons . Wnts function through MIG-1/Frizzled and the planar cell polarity protein ( PCP ) VANG-1/Strabismus/Vangl2 to restrict the formation of F-actin patches , which mark branching sites in nascent neurites . We find that VANG-1 promotes Wnt signaling by facilitating Frizzled endocytosis and genetically acts in a common pathway with arr-1/β-arrestin , whose mutation results in defective PLM branching and F-actin patterns similar to those in the Wnt , mig-1 or vang-1 mutants . On the other hand , the UNC-6/Netrin pathway intersects orthogonally with Wnt-PCP signaling to guide PLM branch growth along the dorsal-ventral axis . Our study provides insights for how attractive and repulsive signals coordinate to sculpt neurite branching patterns , which are critical for circuit connectivity .
Branching of the axon or dendrite expands the connectivity of neural circuits and is critical for the functions of the nervous system . Transcription factors have been shown to specify the morphology of neuronal branch arbors as part of cell fate determination [1] . In addition , diffusible , secreted cues also regulate neurite branching . For example , locally applied nerve growth factor ( NGF ) or Netrin-1 induce de novo interstitial branch formation in cultured cortical neurons [2 , 3] . Neurite branching is also patterned by inhibitory signals . In the amphibian and vertebrate visual systems , repulsive ephrin-Eph signaling shapes topographic innervation of tectal neurons by preventing ectopic branching of retinal ganglion cells ( RGC ) beyond the target zones [4–6] . Furthermore , graded Wnt glycoproteins repel the chick RGC axon branches in the tectum [7] . These studies highlight the importance of inhibitory cues in instructing neurite branching patterns . How extracellular signals remodel neuronal cytoskeleton to generate branches at specific locations is incompletely understood . Previous studies suggest that focal enrichment of filamentous actin ( F-actin ) is an early molecular signature for axon branch formation , which precedes the development of protrusive membrane activity and subsequent branch outgrowth [8–11] . Adhesion receptors instruct axon branches of the hermaphrodite-specific neuron ( HSN ) in C . elegans by locally promoting F-actin assembly [12] . A recent study in Drosophila suggests that inter-neuronal interaction of transmembrane protein Dscam1 specifies dendrite branching sites by regulating F-actin dynamics through kinases such as DOCK and Pak [13] . These studies provide a link between attractive cues and F-actin assembly in defining axon branching sites [14] . It is less clear how the repulsive signals engage neuronal cytoskeleton to pattern neurite branching . In the present study , we uncover a role for secreted Wnt glycoproteins in specifying the stereotyped branching pattern of the PLM mechanosensory neurons in C . elegans . In the mutants of Wnts , Frizzled receptors or the planar cell polarity ( PCP ) gene vang-1/Strabismus/Vangl2 , the PLM branch develops at ectopic sites preceded by aberrant F-actin distribution , suggesting that Wnt-Frizzled/PCP signaling spatially patterns F-actin assembly to instruct branching sites . Our results suggest that VANG-1 promotes Wnt signaling by facilitating Frizzled endocytosis , and that endosomal localization of Frizzled is crucial for patterning PLM branching .
The PLMs are bilaterally symmetric touch mechanosensory neurons in C . elegans , with a single collateral branch extending from the long anterior process that forms chemical synapses with interneurons in the ventral nerve cord ( Fig 1A ) . The development of the PLM branch begins at late embryonic stages and is complete by 12 hours post-hatching in the wild-type animals [15] . The PLM branching sites were remarkably predictable in wild-type animals at the fourth larval ( L4 ) stage when visualized with PLM-specific transgenes or by staining the animals with an antibody for acetylated microtubules ( Fig 1B ) . This pattern was preserved throughout larval development ( S1A Fig ) . We observed one and only one branch for each neuron at all time points , implying that PLM branching pattern is not shaped by pruning of unwanted branches ( S1B Fig ) , as recently reported in the mammalian cortex [16] . We set out to determine how branch outgrowth is initiated at well-defined locations along the anterior-posterior ( A-P ) body axis . In search for molecular signatures that predict future branching sites in the unbranched PLM process , we expressed the F-actin binding protein COR-1/Coronin-1 or the actin binding domain of VAB-10B fused with fluorescent mCherry to label F-actin . We found that F-actin transitioned from a diffuse distribution along the PLM process to a more restricted localization in the distal region ( Fig 2A–2C ) . Importantly , the locations of the F-actin patches in nascent , unbranched neurites strongly correlated with the PLM branching sites at L1 stage ( Fig 2D ) , suggesting that focal enrichment of F-actin in nascent neurites defines future branching sites . To test this , we performed time-lapse imaging to trace F-actin distribution during PLM branch outgrowth ( Fig 2E and 2F ) . We observed that initially COR-1::GFP puncta were broadly distributed along the nascent , unbranched PLM neurite , followed by progressively restricted localization that eventually formed a bright F-actin patch ( Fig 2E ) . Robust local protrusive activity of the neurite membrane was initiated after formation of the F-actin patch , and was subsequently stabilized with one collateral branch extending among several slender filopodia ( Fig 2E and 2F ) . This nascent branch later reached the ventral midline and formed a presynaptic varicosity that connected with the ventral nerve cord . Of seven animals that we successfully recorded and generated a mature PLM branch tipped with a presynaptic varicosity , the PLM branch always developed from the brightest F-actin patch . This observation strongly indicates that the focal F-actin patch instructs PLM branch outgrowth . With these observations , we hypothesize that extracellular signals instruct placement of the PLM branch by regulating the pattern of F-actin assembly in the PLM neurite . Wnt signaling had been shown to be important in cell fate determination , axon guidance and synapse formation . In C . elegans , Wnts functions as repulsive cues for neuronal migration , neurite extension and synapse formation [17–19] . To test whether Wnts pattern PLM branches , we first examined animals with a hypomorphic mutation in mig-14/Wntless , which encodes a transmembrane protein essential for Wnt secretion [20 , 21] . This mig-14 mutation significantly depleted extracellular Wnts and caused defective PLM morphology in around 30% of the animals [22 , 23] . For the remaining PLM neurons with normal morphology , branching sites became random along the PLM process . By contrast , the length of the PLM neurites with ectopic branching was indistinguishable from that of the wild-type PLM ( Fig 3A and 3B and S2A Fig ) . This result suggests that normal Wnt trafficking is required for PLM branching pattern . We next examined PLM branching patterns in individual Wnt mutants ( cwn-1 , egl-20 , cwn-2 , mom-2 ) , except for lin-44 , the mutation of which resulted in severe PLM polarity defects [17] . While the cwn-1 and the egl-20 single mutants had no overtly defects in PLM branching patterns , distribution of branching sites in the cwn-1; egl-20 double mutant expanded significantly , especially proximal to the cell body , indicating that cwn-1 and egl-20 together control PLM branching sites ( Fig 3A and 3B ) . A mutation in the Wnt cwn-2 did not alter PLM branching pattern . However , the cwn-2 mutation suppressed the proximal branching phenotype of the cwn-1; egl-20 double mutant ( Fig 3B ) , suggesting that cwn-2 prevents distal branching . Taken together , these data indicate that combinatorial effects of spatially distinct Wnts ensure that the PLM branch forms at a fairly predictable anterior-posterior position . In around 5% of cwn-1; egl-20 animals , we observed two PLM branches distant to each other in the same neuron during L1 stage ( S2B Fig ) . This phenotype was transient , as none of the cwn-1; egl-20 animals had multiple PLM branches when examined at L4 , suggesting that unknown mechanisms exist to select one from these two PLM branches and eliminate the other . We found that , in contrast to the single , highly localized F-actin patch in the wild-type unbranched PLM neurite 3 hours post-hatching , F-actin became dispersed and showed increased intensity in the cwn-1; egl-20 mutants ( Fig 3C ) , which persisted as late as 7 hours post-hatching ( S2C Fig ) . The high F-actin activity may explain in part the transient multiple-branch phenotype in the cwn-1; egl-20 mutant . Overall , these observations suggest that Wnt signals restrict F-actin distribution to mark the future branching sites . To test whether Wnts instruct PLM branching sites , we first induced widespread CWN-1 and EGL-20 expression by heat shock . Widespread CWN-1 , but not EGL-20 , expression shifted PLM branches both proximally and distally ( Fig 3D–3F ) . This result suggests that CWN-1 functions instructively and EGL-20 does not . Indeed , anterior EGL-20 expression from the cwn-2 promoter rescued PLM branching pattern of the cwn-1; egl-20 double mutant to the same level as done by EGL-20 expression from the endogenous egl-20 promoter expressed in the posterior ( Fig 3F ) . By contrast , while posterior expression of CWN-1 from the egl-20 promoter mimicked the endogenous CWN-1 gradient and rescued defective branching patterns of the cwn-1; egl-20 mutant , anterior CWN-1 expression failed to rescue ( Fig 3E ) . Moreover , the mild proximal branching phenotypes of the cwn-1 mutant became more severe when CWN-1 was expressed in the anterior ( Fig 3E ) . These data indicate that cwn-1 is a repulsive cue for PLM branch placement . In our screen of the Wnt pathway mutants for defective PLM branching patterns , we found that mutations in the Frizzled receptor mig-1 , the PCP transmembrane protein vang-1 and the β-catenin bar-1 caused abnormal PLM branching patterns ( Fig 4A–4C ) . While Wnts , the Frizzled receptor and the β-catenin are all required for PLM branching patterns , we only observed mild phenotypes in dsh-2 , one of the three Dishevelled mutants , likely due to functional redundancy between these genes . We focused on mig-1 and vang-1 because they showed comparable penetrance of PLM branching abnormality to that in the cwn-1; egl-20 mutant . The proximal branching phenotype was not further enhanced in the mig-1; cwn-1; egl-20 triple mutant compared to that of the mig-1 or the cwn-1; egl-20 mutant ( Fig 4D ) , indicating that mig-1 acts in a common pathway with cwn-1 and egl-20 and is likely the receptor for these two Wnts . Likewise , defects in PLM branching sites were not enhanced in the mig-1; vang-1 double mutant compared to either single mutants , suggesting that mig-1 and vang-1 also function in the same pathway ( Fig 4D ) . Similar to that in the cwn-1; egl-20 double mutants , F-actin in the unbranched PLM neurite was increased and became dispersed in the mig-1 and the vang-1 mutants ( Fig 4E ) . We also observed transient multiple-branch phenotypes in around 10% of mig-1 or vang-1 mutant animals , consistent with high , ectopic F-actin activity promoting branch outgrowth . Touch neurons-specific expression of mig-1 or vang-1 rescued the PLM branching defects as well as the aberrant F-actin patterns ( Fig 4E ) , confirming that they act cell-autonomously in the PLM to pattern F-actin assembly and neurite branching . Increased F-actin activity in the mig-1 , vang-1 or cwn-1; egl-20 double mutants prompted us to examine the Rho and Rac small GTPases , which are important F-actin regulators and had been implicated in Wnt-PCP signaling [24] . We found that mutations in the Rac small GTPases mig-2 and ced-10 significantly suppressed defective branching patterns of the mig-1 and vang-1 mutants ( Fig 4F ) . Importantly , the single mig-2 and ced-10 mutants displayed normal PLM branching patterns ( Fig 4F ) . Moreover , the mig-2 ( gm103 ) gain-of-function mutant displayed proximal branching phenotypes as well as aberrant F-actin activity reminiscent of those in the Wnt , mig-1 or vang-1 mutant ( Fig 4E ) . Expression of constitutively active RHO-1 in the touch neurons also caused defective PLM branching patterns ( Fig 4F ) . These results suggest that the repulsive activity of Wnts in F-actin assembly and subsequent neurite branching in part acts through inhibition of Rac and Rho small GTPases . By expressing MIG-1::GFP or MIG-1::mCherry chimeric proteins in the touch neurons , we found that MIG-1 , mostly in punctate forms , was localized to both the plasma membrane and cytosol , as well as the proximal segment of the PLM neurite ( Fig 5A and 5B ) . This highly polarized subcellular distribution of MIG-1 required Wnts , as in the cwn-1; egl-20 double mutant , MIG-1 signal became diffuse over the plasma membrane and in the PLM process ( Fig 5C ) . Clustering of MIG-1 on the membrane required the cysteine-rich domain ( CRD ) that binds Wnts , but not the short cytosolic tail ( Fig 5C and 5D ) , while both were essential for transducing Wnt signaling to define the PLM branch patterns ( Fig 5E ) . As our genetic experiments suggest that EGL-20 acts permissively and CWN-1 functions as an instructive signal , we investigated whether EGL-20 and CWN-1 have distinct effects on the MIG-1 receptor . In the egl-20 mutant , the punctate MIG-1 signals were markedly reduced , and MIG-1 became diffuse on the PLM membrane ( S3A Fig ) . By contrast , punctate MIG-1 signals were not affected in the cwn-1 mutant ( S3A Fig ) . Expression of EGL-20 in the posterior fully restored the punctate MIG-1 signals , and anterior EGL-20 expression partially restored punctate MIG-1 distribution on the PLM membrane ( S3A and S3B Fig ) . These data suggest that EGL-20 is important for clustering of MIG-1 on the plasma membrane , while CWN-1 may regulate MIG-1 in a different way . We found that GFP::VANG-1 was also in punctate forms but was evenly distributed along the PLM neurite , the cell membrane and in the cytosol ( Fig 5B ) . When co-expressed , VANG-1 and MIG-1 puncta partially colocalized in the PLM neuron ( Fig 5B ) . To test whether MIG-1 and VANG-1 form protein complexes , we ectopically expressed MIG-1 and VANG-1 in mammalian HEK293 cells and performed coimmunoprecipitation . We found that MIG-1 and VANG-1 coimmunoprecipitated each other , suggesting that they form protein complexes ( Fig 6A ) . Of note , interaction with VANG-1 in the coimmunopreciptation experiments was independent of the CRD domain or the C-terminus of MIG-1 ( S4 Fig ) . A fraction of cytosolic MIG-1::GFP colocalized with the early endosome marker RAB-5 , which was reduced in the vang-1 mutant ( Fig 6B and 6C ) . To further analyze MIG-1 distribution , we quantified MIG-1::GFP signals in series of single confocal optical sections through the thickness of the PLM soma , with myristoylated mCherry to label the plasma membrane ( Fig 6D and 6E; see Materials and methods for detail of quantification ) . Elimination of the C-terminus domain of MIG-1 ( MIG-1ΔC ) resulted in its accumulation at the plasma membrane , confirming that the C-terminus is required for MIG-1 endocytosis ( Fig 6D and 6E ) . MIG-1 localization to the plasma membrane was also increased in the vang-1 mutant ( Fig 6D and 6E ) . By contrast , VANG-1 overexpression resulted in increased MIG-1 cytosolic distribution and diminished membrane localization ( Fig 6D and 6E ) , suggesting that VANG-1 controls MIG-1 endocytosis and subsequent localization in the early endosomes . We found that animals with a mutation in arr-1/β-arrestin 2 , which is essential for Frizzled endocytosis [25] , had defective branching and F-actin patterns similar to those in the mig-1 , vang-1 or cwn-1; egl-20 double mutants ( Fig 6F and 6G ) . Our genetic experiments further suggested that arr-1 and mig-1 acted in a common pathway and arr-1 functioned in the PLM neuron ( Fig 6F ) . MIG-1::GFP was more enriched on the plasma membrane in the arr-1 mutant , similar to what was observed with MIG-1 that lacked the C-terminus tail ( Fig 6D and 6E ) . In addition , mutations in RABS-5 , a regulator of RAB-5-dependent endosomal trafficking , showed defective branching similar to that of the mig-1 mutant , suggesting that endosomal trafficking ( presumably of MIG-1 ) is required for proper PLM branching ( Fig 6F ) . Based on these results , we propose that MIG-1 endocytosis is essential for transducing Wnt signals to pattern PLM branching locations , and that VANG-1 and ARR-1 control MIG-1 endocytosis . In the mig-1 , vang-1 and cwn-1; egl-20 mutants , misplaced PLM branches still extend successfully to the ventral cord axons . Furthermore , we found that synaptic contacts were not disrupted in the mig-1 mutant , judged by the GRASP ( GFP Reconstitution Across Synaptic Partners ) technique ( S5 Fig ) [26] . These observations suggest that mechanisms independent of Wnt signaling control PLM branch development along the dorsal-ventral ( D-V ) axis and later PLM synaptogenesis . Mutations in the axon guidance cue unc-6/Netrin and its receptor unc-40/Deleted in Colorectal Cancer ( DCC ) disrupted PLM branch growth along the dorsal-ventral ( D-V ) axis ( Fig 7A ) . We found that F-actin activity in the nascent PLM neurite before branch outgrowth was significantly reduced in the unc-40 but not the unc-6 mutant ( Fig 7B and 7C ) . This finding is consistent with the role of UNC-40 in regulating branch outgrowth by promoting F-actin assembly [27] . While a significant percentage of the unc-6 and unc-40 animals lost their PLM branch , for those that developed branches , the branches were at wild-type locations along the PLM neurite ( Fig 7D ) . This observation suggests that Wnts and ventrally-derived Netrin act orthogonally to pattern the PLM branch , with Wnts instructing the A-P position of the branch and Netrin promoting its growth along the D-V axis . To test this , we first analyzed mig-1; unc-6 animals , and found that they showed comparable penetrance of missing and misplaced PLM branch to that in the unc-6 and mig-1 single mutants , respectively ( Fig 7A and 7D ) . This suggests that mig-1 and unc-6 function independently . When unc-6 was ectopically expressed in the dorsal musculature from a fragment of the unc-129 promoter , the PLM branches grew at normal A-P locations but were sometimes misrouted dorsally ( Fig 7E ) , suggesting that unc-6 instructs D-V growth of the PLM branch . When unc-6 was expressed dorsally in the mig-1 mutant , we found that dorsally-routed PLM branches developed at ectopic locations along the A-P axis ( Fig 7D ) . Together these experiments support the model that Wnts along the A-P axis and Netrin along the D-V axis interact orthogonally to pattern branch outgrowth in the PLM neuron ( Fig 8 ) .
Unlike the mammalian and avian nervous systems where pruning of excessive collateral branches shapes the final axon arbors [5 , 16 , 29] , the C . elegans PLM neuron generates one and only one neurite branch at remarkably defined locations , suggesting that instructive mechanisms are involved . Previous studies indicate that Wnt signals function as inhibitory cues for neuronal growth cone migration and synapse formation [18 , 19 , 30 , 31] . Our findings demonstrate that Wnt signals instruct PLM neurite branching by inhibiting F-actin assembly , expanding the roles of repulsive Wnt signaling in wiring the C . elegans nervous system connectivity . It would be interesting to test whether inhibition of synapse formation by Wnts also engages F-actin modulation , a mechanism recently shown to be critical for instructing synapse formation by adhesion receptors in the C . elegans hermaphrodite-specific neurons ( HSN ) [12] . Functional redundancy among different Wnt molecules is well-established , but whether they signal in discrete ways at the molecular level is not adequately addressed . Our results suggest that CWN-1 is an instructive signal , while EGL-20 acts permissively to regulate PLM branching patterns . Permissive EGL-20 functions had been demonstrated in the polarization of the V5 seam cell or the posterior migration of the left Q neuroblast descendants [32 , 33] . By contrast , EGL-20 functions as a repulsive signal in the migration of the hermaphrodite-specific neuron ( HSN ) , projection of the AVM and PVM neurites and synapse formation in the DA motor neurons [19 , 30] . Our genetic experiments imply that the Frizzled MIG-1 is likely a shared receptor for both CWN-1 and EGL-20 . At the subcellular level , we found that EGL-20 , but not CWN-1 , clustered membrane MIG-1 receptors . In the absence of EGL-20 , MIG-1 was delocalized and became diffuse on the membrane . Consistent with EGL-20 being a permissive signal for PLM branching pattern , clustering of MIG-1 could also be achieved in part by ectopic EGL-20 expression , although we cannot exclude the possibility that the anterior-expressing Pcwn-2::EGL-20 transgene also expressed EGL-20 in the posterior at lower level . Clustering of MIG-1 per se does not seem to be essential for proper PLM branching , as ectopic PLM branching was minimal in the egl-20 single mutant . Based on these observations , we speculate that EGL-20-dependent MIG-1 clustering facilitates activation of MIG-1 by the directional CWN-1 signal . In the cwn-1 single mutant , other Wnts ( such as MOM-2 and LIN-44 ) could compensate for the loss of CWN-1 and correctly pattern PLM branching . In the egl-20 mutant , activation of MIG-1 by CWN-1 is less efficient but is still sufficient to correctly pattern PLM branching in most animals . When both CWN-1 and EGL-20 are removed , signaling through MIG-1 is largely abolished , resulting in significant ectopic branching . We reason that the use of both instructive and permissive signaling mechanisms in a functionally redundant manner insulates PLM branching pattern from perturbation in any individual signaling axis . Endocytosis is a key step in the transduction of Wnt signals in planar cell polarization , growth cone guidance and synaptogenesis [34–38] . We found that the PCP protein VANG-1 is both necessary and sufficient to promote MIG-1 internalization , which extends previous studies of Frizzled endocytosis in neuronal development [25 , 37 , 39 , 40] . A previous study reported that in the commissural axons of the mouse spinal cord , Vangl2 promoted Frizzled3 localization to the intracellular vesicles by antagonizing Dvl1-dependent Frizzled phosphorylation [36] . By contrast , we show that MIG-1 and VANG-1 were all necessary for proper PLM branching , suggesting that they act in the same direction . As the C-terminus of MIG-1 was dispensable for VANG-1 binding , we hypothesize that VANG-1 interact with MIG-1 through other domains . Development of neurite branches in the 2D or 3D space requires sophisticated coordination between signaling cues distributed along distinct body axes . Our findings that Wnts and Netrin independently specify the position and trajectory of the PLM branch along A-P and D-V axes , respectively , illustrate how orthogonal cues intersect to generate precise neurite branching patterns . While Wnt signaling inhibits and the Netrin-DCC signaling promotes F-actin assembly , both pathways leverage F-actin assembly to control PLM branch development . UNC-40/DCC had been previously shown to directly interact with CED-10/Rac [41] , and our data indicate that genetically Wnt-PCP signaling inhibits CED-10/Rac and MIG-2/Rac . It is tempting to speculate that the Rac and Rho small GTPases serve as a shared regulatory step at which signals from orthogonal patterning cues converge to shape neurite branching patterns . An important but unanswered question is how endosomal signaling generates asymmetry in F-actin assembly , the elucidation of which will shed light on the mechanisms that translate directional signaling cues into polarized cytoskeletal activity and compartmentalized morphogenesis patterns of the neurons .
Strains were cultured and maintained as described [42] . Strains and transgenes used in this study are: LG I: unc-54 ( e190 ) [43] , mig-1 ( e1787 ) [19] , unc-40 ( n324 ) ; LG II: mig-14 ( ga62 ) [20] , cwn-1 ( ok546 ) [44] , cam-1 ( gm122 ) [45] , dsh-2 ( ok2162 ) /mIn1 , dsh-1 ( ok1445 ) , mig-5 ( rh147 ) [46]; LG IV: egl-20 ( n585 ) [47] , cwn-2 ( ok895 ) [44] , prkl-1 ( ok3182 ) , rabs-5 ( ok1513 ) , ced-10 ( n1993 ) [48]; LG V: fmi-1 ( hd121 ) [49] , mom-2 ( ok591 ) /nT1; LG X: vang-1 ( tm1422 ) [50] , lin-18 ( e620 ) [51] , bar-1 ( ga80 ) [52] , arr-1 ( ok401 ) , mig-2 ( mu28 ) , mig-2 ( gm103gf ) [53] , unc-6 ( ev400 ) [54] , zdIs5 ( Pmec-4::GFP ) , muIs42 ( Pmec-7::GFP ) , jsIs973 ( Pmec-7::RFP ) , twnEx110 ( Pmec-7::COR-1::GFP , Pttx-3::GFP ) , twnEx195 ( Pmec-7::COR-1::mCherry , Pgcy-8::mCherry ) , twnEx202 ( Pmec-7::ΔCRD::MIG-1::GFP , Pdpy-30::NLS::dsRed ) , twnEx205 ( Phsp::CWN-1 , Pttx-3::GFP ) , twnEx209 ( Pmec-7::MIG-1::GFP , Pdpy-30::NLS::dsRed ) , twnEx230 ( Pmec-7::GFP::VANG-1 , Pdpy-30::NLS::dsRed ) , twnEx231 ( Pmec-7::MIG-1::GFP , Pmec-7::myr::mCherry , Pgcy-8::mCherry ) , twnEx232 ( Pmec-7::GFP::VANG-1 , Pmec-7::myr::mCherry , Pgcy-8::mCherry ) , twnEx254 ( Pmec-7::MIG-1ΔC::GFP , Pmec-7::myr::mCherry , Pgcy-8::mCherry ) , twnEx256 ( Pmec-7::VAB-10B-ABD , Pttx-3::GFP ) , twnEx262 ( Pegl-20::CWN-1::Venus , Pttx-3::GFP ) , twnEx263 ( Punc-129::UNC-6 , Pdpy-30::NLS::dsRed ) , twnEx266 ( Pegl-20::EGL-20::Venus , Pttx-3::GFP ) , twnEx265 ( Phsp::EGL-20 , Pttx-3::GFP ) , twnEx267 ( Pcwn-2::CWN-1::Venus , Pgcy-8::mCherry ) , twnEx268 ( Pcwn-2::EGL-20::Venus , Pgcy-8::mCherry ) , twnEx269 ( Pcwn-2::EGL-20 , Pcwn-2::CWN-1 , Pgcy-8::mCherry ) , twnEx272 ( Pmec-7::MIG-1ΔC::GFP , Pdpy-30::NLS::dsRed ) , twnEx273 ( Pmec-7::GFP::VANG-1 , Pmec-3::MIG-1::mCherry , Pttx-3::GFP ) , twnEx275 ( Pmec-3::MIG-1::mCherry , Pmec-7::GFP::RAB-5 , Pttx-3::GFP ) , twnEx276 ( Pmec-7::ARR-1::mCherry , Pdpy-30::NLS::dsRed ) , twnEx337 ( Pmec-7::RHO-1 ( G14V ) , Pdpy-30::NLS::dsRed ) , twnEx351 ( Prig-3::mCherry , Prig-3::CD4::GFP ( 1–10 ) , Pmec-7::mCherry , Pmec-7::CD4::GFP ( 11 ) , Pgcy-8::GFP ) , twnEx378 ( Pmec-7::GFP::VANG-1 , Pmec-7::COR-1::mCherry , Pgcy-8::mCherry ) , twnEx379 ( Pmec-3::MIG-1::GFP , Pmec-7::COR-1::mCherry , Pgcy-8::mCherry ) . The transgene twnEx351 was used in the GRASP experiment ( S5 Fig ) . For detailed information regarding the strains and transgenes used in individual Figures , please see S1 Table . For rescue experiments with transgenes , we examined at least two independent lines of transgenic animals to confirm that the results were reproducible and consistent between these lines . We then used the one that was easier in manipulation and maintenance ( higher transmission rate in most cases ) for subsequent data acquisition and analyses . Animals with zdIs5 ( Pmec-4::GFP ) or jsIs973 ( Pmec-7::RFP ) were anesthetized by 1% sodium azide , and the PLMs were imaged under the 10X objective of an AxioImager M2 system ( Carl Zeiss ) . The length of the PLM process and the distance between the PLM branch and the cell soma were analyzed using ImageJ [55] . Each animal was only scored for left or right PLM based on which side is clearly visible in the image . There was no side-to-side difference in PLM branch positions , so data from left and right PLMs were pooled for analyses . We define normal PLM branch locations as those that fall within 99% ( mean ± 3 standard deviations ) of PLM branch sites visualized by zdIs5 ( Pmec-4::GFP ) in the wild type , which ranges from 0 . 46 to 0 . 88 . The percentage of both proximally and distally mislocalized PLM branches in indicated genotypes was calculated and presented along with the scatter plots . The molecular cloning and plasmid construction were performed by standard molecular biology techniques . All of expression constructs used to generate the twnEx series of transgenes were in the pPD95 . 77 Fire vector backbone , including transgenes induced by heat shock ( through the hsp-16 . 2 promoter ) . Site-direct mutagenesis was performed with QuickChange kit . Detailed information , including primer sequences , for cloning cor-1 , vab-10b-ABD , cwn-1 , egl-20 , mig-1 , vang-1 , rab-5 , and arr-1 are available upon request . For heat shock experiments , animals with heat-shock inducible transgenes were grown on NGM plates at 20°C , transferred to 34°C for 30 minutes at early L1 stage , and recovered at 20°C for another 24 hours before the branch locations were scored . HEK293 cells were transfected by lipofectamine ( Invitrogen ) and then lysed in lysis buffer ( 50 mM Tris , 150 mM NaCl , 2 mM EDTA [pH 8 . 0] , 0 . 5% sodium deoxylcholate , 10 mM phenylmethylsulfonyl fluoride , and 1M dithiothreitol ) with 1% NP-40 . For co-immunoprecipitation , cell lysates were immunoprecipitated by anti-HA ( Invitrogen ) or anti-FLAG ( Sigma ) beads . Immuno-complexes or samples for western blot analysis were electrophoresed in a SDS-polyacryalamide gel , transferred onto PVDF membrane and probed with anti-HA Y-11 ( 1:4000 , Santa Cruz ) or anti-FLAG ( 1:5000 , Sigma ) antibodies . To immobilize worms without interfering with animal development , we used the unc-54 ( e190 ) mutation to genetically paralyze the animals . The unc-54 mutation did not affect PLM development . We placed early L1 larva and 2–4 μl polystyrene beads ( 0 . 1 μm , Polysciences Inc . ) on 5% gel pad . The cover slip was sealed by vaseline to prevent desiccation . z-stack maximum projection images were acquired using the Zeiss LSM 700 Confocal Imaging System ( Carl Zeiss ) . Pixel-wise colocalization of GFP and mCherry fluorescence signal was quantified using the Zeiss Zen imaging software . For the quantification of MIG-1 or VANG-1 cytosolic localization , myristolated mCherry was expressed to label the cell membrane of the touch neurons . Using Zeiss Zen software , total and cytosolic MIG-1::GFP signal intensity was derived by first quantifying each single optical sections of the z-stack confocal images that span most of the thickness of the PLM cell body , followed by summation of data from individual optical sections to calculate the cytosolic/total signal intensity ratio . Heat maps of COR-1::mCherry intensity in the PLM process were generated from z-stack maximum projection images using ImageJ . Ten individual PLMs were assembled into a single aligned heat map of COR-1 intensity for each genotype . The ANOVA , Student’s t test , Mann-Whitney U test and two proportion z test were conducted in MS Office Excel or Prism for experiments indicated in the Figure Legends , with Bonferroni correction for multiple comparisons . Error bars represent standard error of means ( S . E . M . ) . This study does not involve any human subject , non-human primates and other vertebrates . | Extrinsic cues instruct neurite branching patterns through cytoskeletal remodeling at precise locations . We show that the Wnt glycoproteins signal through the Frizzled receptor and the Planar Cell Polarity ( PCP ) protein VANG-1 to instruct neurite branching in the nematode C . elegans , by restricting F-actin assembly to positions along the nascent neurite where future branches emerge . VANG-1 facilitates Frizzled endocytosis and its subsequent signaling in the endosome , which is critical for shaping F-actin assembly . This Wnt-PCP-endosomal pathway intersects orthogonally with Netrin signaling to pattern neurite branching in the two-dimensional space . Our study uncovers a novel role for repulsive Wnt signaling in patterning neurite branches and illustrates how instructive cues along orthogonal body axes coordinate to define neural circuit connectivity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"caenorhabditis",
"nervous",
"system",
"cell",
"processes",
"electrophysiology",
"neurites",
"neuroscience",
"animals",
"animal",
"models",
"caenorhabditis",
"elegans",
"signal",
"inhibition",
"model",
"organisms",
"experimental",
"organism",
"systems",
"cellular",
"structures",
"and",
"organelles",
"neuronal",
"dendrites",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"cell",
"membranes",
"endocytosis",
"signal",
"transduction",
"cellular",
"neuroscience",
"cell",
"biology",
"secretory",
"pathway",
"anatomy",
"synapses",
"physiology",
"neurons",
"nematoda",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"wnt",
"signaling",
"cascade",
"cell",
"signaling",
"neurophysiology",
"organisms",
"signaling",
"cascades"
]
| 2017 | A Wnt-planar polarity pathway instructs neurite branching by restricting F-actin assembly through endosomal signaling |
The intestinal immune system must be able to respond to a wide variety of infectious organisms while maintaining tolerance to non-pathogenic microbes and food antigens . The Vitamin A metabolite all-trans-retinoic acid ( atRA ) has been implicated in the regulation of this balance , partially by regulating innate lymphoid cell ( ILC ) responses in the intestine . However , the molecular mechanisms of atRA-dependent intestinal immunity and homeostasis remain elusive . Here we define a role for the transcriptional repressor Hypermethylated in cancer 1 ( HIC1 , ZBTB29 ) in the regulation of ILC responses in the intestine . Intestinal ILCs express HIC1 in a vitamin A-dependent manner . In the absence of HIC1 , group 3 ILCs ( ILC3s ) that produce IL-22 are lost , resulting in increased susceptibility to infection with the bacterial pathogen Citrobacter rodentium . Thus , atRA-dependent expression of HIC1 in ILC3s regulates intestinal homeostasis and protective immunity .
The intestinal immune system is held in a tightly regulated balance between immune activation in response to potential pathogens and the maintenance of tolerance to innocuous antigens , such as food and commensal flora . Disruption of this balance can lead to the development of serious inflammatory disorders , such as food allergy or inflammatory bowel disease ( IBD ) . A complex network of different immune cell types including dendritic cells ( DCs ) , macrophages , innate lymphoid cells ( ILCs ) , and T cells , are essential for both the induction of active immunity and the maintenance of intestinal homeostasis . The vitamin A metabolite all-trans-retinoic acid ( atRA ) plays an important role in shaping intestinal immunity by regulating both the innate and adaptive immune systems . atRA that is generated by the metabolism of Vitamin A by intestinal epithelial cells ( IECs ) and a subset of CD103-expressing intestinal dendritic cells ( CD103+ DCs ) has been shown to directly affect the localization and function of lymphocytes . For example , atRA has been shown to induce expression of chemokine receptors ( CCR9 ) and integrins ( α4 and β7 ) that are associated with homing to , and retention in , the intestinal microenvironment [1–3] . In addition , atRA has been shown to control the balance of regulatory T ( Treg ) cells and CD4+ T helper 17 ( TH17 ) cells in the intestine by promoting Treg cell differentiation and inhibiting TH17 cell development [4–9] . Similarly , atRA controls the development of ILC subsets in the intestine , as mice raised on a Vitamin A-deficient ( VAD ) diet display reduced numbers of ILC3s [10 , 11] , with one study showing a concomitant increase in ILC2 numbers and enhanced type 2 immunity within the intestine [10] . In addition , intestinal DC differentiation is influenced by atRA as mice raised on a VAD diet display reduced numbers of CD103+ CD11b+ DCs [12 , 13] . Thus , atRA-dependent processes are central to the function of intestinal TH cells , ILCs , and DCs in vivo . However , the molecular mechanisms downstream of atRA signaling that control immune cell function and homeostasis remain unknown . Hypermethylated in cancer 1 ( HIC1 , ZBTB29 ) is a transcriptional factor that was first identified as a gene that is epigenetically silenced in a variety of human cancers [14 , 15] . HIC1 has been shown to regulate cellular proliferation , survival and quiescence in multiple normal and tumour cell lines [16–19] . HIC1 is a member of the POZ and Kruppel/Zinc Finger and BTB ( POK/ZBTB ) family of transcription factors that consists of regulators of gene expression that are critical in a variety of biological processes [20] . Importantly , several members of the POK/ZBTB family are key regulators in immune cell differentiation and function , including: BCL6 , PLZF and ThPOK [21–25] . Recently , we identified HIC1 as an atRA responsive gene in intestinal TH cells and demonstrated a T cell-intrinsic role for HIC1 in the regulation of intestinal homeostasis as well as in development of several models of intestinal inflammation [26] . In this study , we show that deletion of HIC1 in hematopoietic cells results in a significant reduction in the number of αβ and γδ T cells , CD11b+ CD103+ DCs , and ILC3s in the intestine , resulting in susceptibility to infection with the bacterial pathogen Citrobacter rodentium . Although loss of HIC1 expression in T cells or CD11c+ cells had no effect on immunity to Citrobacter , deletion of HIC1 in RORγt-expressing ILC3s resulted in susceptibility to infection , due to a reduction in IL-22 production . These results identify a central role for atRA-dependent expression of HIC1 in ILC3s in the regulation of intestinal immune responses .
We have previously shown that HIC1 is expressed in a wide variety of immune cells in the intestinal microenvironment , and that deletion of HIC1 specifically in T cells resulted in a severe reduction in the frequency and number of CD4+ and CD8+ T cells in the intestine [26] . To test if HIC1 played a general role in intestinal immune cell homeostasis , we generated mice with a hematopoietic cell-specific deletion of Hic1 ( Hic1Vav mice ) by crossing mice with loxP sites flanking the Hic1 gene ( Hic1fl/fl mice ) with mice that express the Cre recombinase under control of the Vav promoter ( Vav-Cre mice ) . Hematopoietic cell-specific deletion of HIC1 resulted in ~50% reduction in the number of CD45+ cells in the intestinal lamina propria ( LP ) ( Fig 1A and 1B ) . Consistent with our previous study [26] , we found reduced frequencies and numbers of γδ and αβ T cells in the LP of Hic1Vav mice ( Fig 1C and 1D ) . Further , analysis of macrophage and DC populations in the LP of Hic1fl/fl and Hic1Vav mice revealed a specific requirement for HIC1 in CD103+ CD11b+ DCs ( Fig 1E and 1F ) , which is consistent with previous studies identifying a role for atRA in the regulation of this DC subset [12 , 13] . We also found a specific reduction in the frequency and number of ILC3s in the absence of HIC1 , while number of ILC2s were unaffected by the loss of HIC1 in the hematopoietic cell compartment ( Fig 1G and 1H ) . Thus , HIC1 expression is critical for regulation of specific immune cell populations in the LP . To directly test the role of hematopoietic cell-specific deletion of HIC1 , we infected Hic1fl/fl and Hic1Vav mice with attaching and effacing intestinal bacterial pathogen Citrobacter rodentium . Following infection with C . rodentium , Hic1Vav mice exhibited enhanced weight loss and significantly higher bacterial burdens in the feces compared to Hic1fl/fl controls ( Fig 2A and 2B ) . Furthermore , infected Hic1Vav mice–but not Hic1fl/fl mice–had dissemination of bacteria to the liver ( Fig 2C ) , demonstrating a significant impairment in the intestinal barrier following infection . Associated with impaired bacterial containment and clearance were reduced levels of transcripts for the cytokines Il17a and Il22 , as well as the intestinal antimicrobial peptide Reg3g ( Fig 2D ) . Thus , expression of HIC1 within hematopoietic cells is critical to mount a proper immune response against C . rodentium . As T cells , CD103+ CD11b+ DCs and ILC3s are all important in initiating and propagating ILC3/TH17 responses in the intestine [27–30] and these population are perturbed in Hic1Vav mice , we next sought to determine the effect of HIC1 deficiency in these specific cell populations during infection C . rodentium . We crossed Hic1fl/fl mice with mice expressing Cre under the control of either the Cd4 promoter or Itgax ( CD11c ) promoter to generate T cell-specific ( Hic1CD4 mice ) and dendritic cell-specific ( Hic1CD11c mice ) HIC1-deficient mice . Both Hic1CD4 mice ( Fig 3A–3C ) and Hic1CD11c mice ( Fig 3D–3F ) were as resistant to infection with C . rodentium as control Hic1fl/fl mice , with equivalent weight loss , fecal bacterial burdens and expression of cytokines and antimicrobial peptide mRNA in the intestine . Thus , these results demonstrate that expression of HIC1 in T cells or CD11c-expressing cells is not required for immunity to bacterial infection and suggests loss of HIC1 in another cell population is responsible for the phenotype observed in Hic1Vav mice . ILC3s have been shown to play a significant role in resistance to infection with C . rodentium [31 , 32] . To determine the role of HIC1 expression in RORγt+ ILC3s during infection with C . rodentium , we crossed Hic1fl/fl mice with mice expressing Cre recombinase under the control of the Rorc promoter ( Hic1Rorc mice ) . Following infection with C . rodentium , and similar to what we observed in the Hic1Vav mice , Hic1Rorc mice displayed increased weight loss , higher fecal bacterial burdens and increased bacterial dissemination than control Hic1fl/fl mice ( Fig 4A–4C ) . Associated with increased susceptibility was reduced expression of Il17a , Il22 and Reg3g in intestinal tissues ( Fig 4D ) . We observed significant inflammation and tissue destruction in the intestine of infected Hic1Rorc mice ( Fig 4E ) , as well as inflammatory foci in the liver of Hic1Rorc mice ( Fig 4F ) . Thus , these results demonstrated that expression of HIC1 in RORγt+ cells is critical for immunity to C . rodentium . In addition to ILC3s , Cre expression in RORγt+ cells will drive deletion in TH17 cells . To remove any potential contribution of CD4+ T cells in the phenotype observed , we treated both Hic1fl/fl and Hic1Rorc mice with a depleting antibody against CD4 prior to infection with C . rodentium ( Fig 5A ) . The absence of CD4+ cells had no significant effects on the differences observed during infection of Hic1fl/fl and Hic1Rorc mice , including weight loss ( Fig 5B ) , fecal bacterial burden ( Fig 5C ) , and bacterial dissemination and inflammation ( Fig 5D–5G ) . Thus , the absence of HIC1 in RORγt+ ILC3s renders mice susceptible to C . rodentium infection . Our results suggest that HIC1 expression in ILC3s is critically important for immunity to intestinal bacterial infection . Using mice with a fluorescent reporter gene inserted in the Hic1 locus ( Hic1Citrine mice ) [33] we determined that in addition to previously identified populations including T cells , dendritic cells and macrophages [26] , lineage-negative ( linneg ) CD90 . 2+ CD127+ ILCs isolated from the intestinal LP express HIC1 ( Fig 6A ) , which was dependent on the availability of atRA , as Hic1Citrine mice reared on a VAD diet did not express HIC1 in ILCs within the LP ( Fig 6B ) . Loss of HIC1 in RORγt+ cells ( in Hic1Rorc mice ) resulted in a specific change in ILC populations in the LP . In the steady state , we observed significantly fewer RORγt+ ILCs ( ILC3s ) in the LP of Hic1Rorc mice , with a significant reduction in the number of RORγt+ TBET+ ILC3s ( Fig 6C and 6D ) . We found no change in the number of CD4+ ILC3s ( also known as lymphoid tissue inducer ( LTi ) cells ) nor in numbers of the canonical GATA3+ ILC ( ILC2 ) population ( Fig 6C and 6D ) . As we observed a significant reduction of ILC3s in the LP in the absence of HIC1 , we next tested whether the lack of HIC1 affected the upstream development of ILC precursors in the bone marrow . ILCs develop in the bone marrow through a lineage pathway that begins with a common lymphoid progenitor ( CLP ) and progresses through an α4β7-expressing lymphoid progenitor ( αLP ) , a common progenitor to all helper-like ILCs ( ChILP ) and , in the case of ILC2s , an ILC2 precursor ( ILC2p ) [34] . Analysis of surface marker expression on lineage-negative , CD45+ bone marrow cells showed that HIC1 was not required for the development of CLP , αLP , ChILP , or ILC2p populations ( Fig 7A and 7B ) . Thus , the reduced number of ILC3s in the LP is not due to a reduced frequency of ILC precursors and suggests that HIC1 is required for ILC3 homeostasis in the periphery . IL-22 production by innate immune cells is critically important for immunity to C . rodentium [35] . We observed a significant reduction in IL-22-producing ILC3s in naïve Hic1Rorc mice ( Fig 8A and 8B ) and infection with C . rodentium failed to expand the small number of ILC3s in Hic1Rorc mice ( Fig 8C and 8D ) . We hypothesized that the reduced levels of IL-22 were responsible for susceptibility to infection . Treatment of Hic1Rorc mice with recombinant IL-22 on days -2 , -1 , 0 , 1 , 3 , 5 and 7 during C . rodentium infection resulted in significant protection from infection , as measured by reduced weight loss ( Fig 8E ) , less intestinal pathology ( Fig 8F ) and a lack of bacterial dissemination to the liver ( Fig 8G ) . Thus , the RA–HIC1 axis is critical for immunity to intestinal bacterial infection by regulating IL-22-producing ILC3s in the intestine .
Our results demonstrate that in the steady state , HIC1 is expressed by intestinal ILCs in a Vitamin A-dependent manner . In the absence of HIC1 , we observed a dramatic decrease in intestinal ILC3 numbers , which was associated with a failure to clear C . rodentium infection . Together , these results highlight an important role for HIC1 not only in regulating intestinal immune homeostasis but also in mounting proper immune responses to an intestinal bacterial infection . In the absence of HIC1 , we found a significant reduction in the number of ILC3s with no effect on ILC2s in the intestine . Specifically , there were reduced numbers of RORγt+ TBET+ ILC3s that produce IL-22 with no difference in CD4+ ILC3s ( LTi ) . This is consistent with studies that have demonstrated that these two lineages have distinct developmental pathways; LTi cells develop in the fetus while TBET+ ILC3s develop postnatally and rely on environmental signals [10 , 36 , 37] . Interestingly , it has been shown that atRA signalling is also important for generation of LTi cells in the fetus [38] . However , our results suggest that HIC1 is not involved in fetal LTi formation , as we find no differences in LTi numbers or lymphoid structures in the absence of HIC1 . Further , the development of ILC progenitor cells in the bone marrow is not perturbed by loss of HIC1 , suggesting that the primary role of HIC1 is to regulate the development and function of adult cells in the periphery . Although we have yet to define the precise molecular mechanisms of HIC1-dependent regulation of intestinal ILC3 function , the RA-HIC1 axis may be acting to control cellular quiescence . atRA has been shown to control hematopoietic stem cell dormancy , a profound state of quiescence , that is counteracted by activation of Myc-dependent proliferation [39 , 40] . As we have previously observed in intestinal T cells , HIC1 is dispensable for the expression of intestinal homing markers and migration to the intestine [26] . Thus , our current working hypothesis is that RA induces expression of HIC1 to promote cellular quiescence/dormancy in the intestinal microenvironment , possibly by regulating Myc-dependent processes including metabolism and proliferation . Future studies will examine if HIC1 has a similar role in ILC3s . Resistance to intestinal infection with C . rodentium is mediated by IL-22 , and ILC3s are the predominant IL-22-producing cell population during the first week of infection [32 , 41] . There are contradictory studies on which ILC3 populations are key for resistance to C . rodentium with both CD4+ LTis and natural cytotoxicity receptor ( NCR ) + ILC3s each being described as either individually critical or redundant [32 , 42 , 43] . Another study looking at TBET+ ILC3s ( which include NCR+ ILC3s ) demonstrated that TBET expression in a subset of ILC3s is critical for resistance to C . rodentium infection [44] . Our results are consistent with a role for NCR+ or TBET+ ILC3s in immunity to C . rodentium as Hic1CD4 mice ( deficient for HIC1 in T cells and LTi ) are resistant to infection while Hic1Rorc mice ( deficient for HIC1 in TH17 cells and all ILC3s ) are susceptible . Further , depletion of CD4-expressing cells in Hic1Rorc mice had no effect on disease development . Thus , HIC1 expression in ILC3s is critical for immunity to C . rodentium . Taken together , these results establish a role for the transcriptional repressor HIC1 as an atRA-responsive cell-intrinsic regulator of ILC3 cell function in the intestine , and identify a potential regulatory pathway that could be targeted to modulate ILC3 responses in the intestine .
Experiments were approved by the University of British Columbia Animal Care Committee ( Protocol A13-0010 ) and were in accordance with the Canadian Guidelines for Animal Research . The generation of Hic1Citrine mice has been described [33] and Hic1fl/fl mice will be described elsewhere ( manuscript in preparation ) . Cd4-Cre mice were obtained from Taconic , Vav-Cre mice were obtained from T . Graf ( Centre for Genomic Regulation , Barcelona , Spain ) and CD11c-Cre ( B6 . Cg-Tg ( Itgax-cre ) 1-1Reiz/J ) and RORc-Cre ( B6 . FVB-Tg ( RORc-cre ) 1Litt/J ) mice were obtained from the Jackson Laboratory ( Bar Harbor , ME , USA ) . Animals were maintained in a specific pathogen-free environment at the UBC Biomedical Research Centre animal facility . Vitamin A-deficient ( TD . 09838 ) diet was purchased from Harlan Teklad Diets . At day 14 . 5 of gestation , pregnant females were administered the vitamin A-deficient diet and maintained on diet until weaning of litter . Upon weaning , females were returned to standard chow , whereas weanlings were maintained on special diet until use . Peyer’s patches were removed from the small intestine , which was cut open longitudinally , briefly washed with ice-cold PBS and cut into 1 . 5 cm pieces . Epithelium was stripped by incubation in 2mM EDTA PBS for 15 minutes at 37°C and extensively vortexed . Remaining tissue was digested with Collagenase/Dispase ( Roche ) ( 0 . 5 mg/mL ) on a shaker at 250 rpm , 37°C , for 60 minutes , extensively vortexed and filtered through a 70μm cell strainer . The flow-through cell suspension was centrifuged at 1500rpm for 5 min . The cell pellet was resuspended in 30% Percoll solution and centrifuged for 10 minutes at 1200 rpm . The pellet was collected and used as lamina propria lymphocytes . Absolute numbers of cells were determined via hemocytometer or with latex beads for LP samples . Intracellular cytokine ( IC ) staining was performed by stimulating cells with 50 ng/ml phorbol 12-myristate 13-acetate ( PMA ) , 750 μg/ml ionomycin , and 10 μg/ml Brefeldin-A ( Sigma , St . Louis , MO ) for 4 hours and fixing/permeabilizing cells using the eBioscience IC buffer kit . All antibody dilutions and cell staining were done with phosphate-buffered saline ( PBS ) containing 2% fetal calf serum ( FCS ) , 1 mM Ethylenediaminetetraacetic acid ( EDTA ) , and 0 . 05% sodium azide . Fixable Viability Dye eFluor 506 was purchased from eBioscience ( San Diego , CA ) to exclude dead cells from analyses . Prior to staining , samples were Fc-blocked with buffer containing anti-CD16/32 ( 93 , eBioscience ) and 1% rat serum to prevent non-specific antibody binding . Cells were stained with fluorescent conjugated anti-CD11b ( M1/70 ) , anti-CD11c ( N418 ) , anti-CD19 ( ID3 ) , anti-CD5 ( 53–7 . 3 ) , anti-CD8 ( 53 . 67 ) , anti-CD3 ( KT3 ) ( 2C11 ) , anti-NK1 . 1 ( PK136 ) , anti-B220 ( atRA-6B2 ) , anti-Ter119 ( Ter119 ) , anti-Gr1 ( RB6-8C5 ) produced in house , anti-CD4 ( GK1 . 5 ) , anti-CD45 ( 30-F11 ) , anti-CD90 . 2 ( 53–2 . 1 ) , anti-GATA3 ( TWAJ ) , anti-RORγt ( B2D ) , anti-TBET ( eBio4B10 ) , anti-FLT3 ( A2F10 ) , anti-CKIT ( ACK2 ) , anti-TCRβ ( H57-597 ) , anti-TCRγδ ( eBioGL3 ) , anti-MHCII ( I-A/I-E ) ( M5/114 . 15 . 2 ) , anti-F4/80 ( BM8 ) , anti-α4β7 ( DATK32 ) , anti-IL-22 ( IL22JOP ) , anti-Ki67 ( SolA15 ) , anti-CCR9 ( eBioCW-1 . 2 ) purchased from eBioscience , anti-CD127 ( 5B/199 ) , anti-CD64 ( X54 . 5/7 . 1 . 1 ) purchased from BD Biosciences . Data were acquired on an LSR II flow cytometer ( BD Biosciences ) and analysed with FlowJo software ( TreeStar ) . Mice were infected by oral gavage with 0 . 1 ml of an overnight culture of Luria-Bertani ( LB ) broth grown at 37°C with shaking ( 200 rpm ) containing 2 . 5 x 108 cfu of C . rodentium ( strain DBS100 ) ( provided by B . Vallance , University of British Columbia , Vancouver , British Columbia , Canada ) . Mice were monitored and weighed daily throughout the experiment and sacrificed at various time points . For enumeration of C . rodentium , fecal pellets or livers were collected in pre-weighed 2 . 0 ml microtubes containing 1 . 0 ml of PBS and a 5 . 0 mm steel bead ( Qiagen ) . Tubes containing pellets or livers were weighed , and then homogenized in a TissueLyser ( Retche ) for a total of 6 mins at 20 Hz at room temperature . Homogenates were serially diluted in PBS and plated onto LB agar plates containing 100 mg/ml streptomycin , incubated overnight at 37°C , and bacterial colonies were enumerated the following day , normalizing them to the tissue or fecal pellet weight ( per gram ) . Colon tissues were fixed overnight in 10% buffered formalin and paraffin-embedded . A total of 5-μm-thick tissue sections were stained with hematoxylin and eosin ( H&E ) for histological analysis . In some cases , mice were treated with 400 ng recombinant mouse IL-22 ( Biolegend ) by i . p . injection daily for 4 days starting 2 days prior to infection . Injections with rmIL-22 continued every other day following day 1 post infection . In other cases , mice were injected i . p . on days -1 , 2 , 5 and 8 post infection with 500 μg of anti-CD4 ( GK1 . 5 ) ( produced in-house by AbLabBiologics , UBC ( Vancouver , BC ) ) , constituted in sterile PBS . Tissues were mechanically homogenized and RNA was extracted using the TRIzol method according to the manufacturer's instructions ( Ambion ) . cDNA was generated using High Capacity cDNA reverse transcription kits ( Applied Biosystems ) . Quantitative PCR was performed using SYBR FAST ( Kapa Biosystems ) and SYBR green-optimized primer sets run on an ABI 7900 real-time PCR machine ( Applied Biosystems ) . Cycle threshold ( CT ) values were normalized relative to beta-actin ( Actb ) gene expression . The primers used were synthesized de novo: Data are presented as mean ± S . E . M . A two-tailed Mann-Whitney test using GraphPad Prism 5 software determined statistical significance . Results were considered statistically significant with P < 0 . 05 . | Innate lymphoid cells ( ILCs ) are emerging as important regulators of immune responses at barrier sites such as the intestine . However , the molecular mechanisms that control this are not well described . In the intestine , the Vitamin A metabolite all-trans-retinoic acid ( atRA ) has been shown to be an important component of the homeostatic mechanisms . In this manuscript , we show that the atRA-dependent transcription factor Hypermethylated in cancer 1 ( HIC1 , ZBTB29 ) is required for ILC homeostasis and function in the steady state as well as following infection with the bacterial pathogen Citrobacter rodentium . Thus , HIC1 links RA signalling to intestinal immune responses . Further , our results identify HIC1 as a potential target to modulate ILC responses in vivo in health and disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immunology",
"cell-mediated",
"immunity",
"diet",
"physiological",
"processes",
"bacterial",
"diseases",
"infectious",
"disease",
"immunology",
"clinical",
"medicine",
"nutrition",
"homeostasis",
"digestive",
"system",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"t",
"cells",
"gastrointestinal",
"tract",
"anatomy",
"cell",
"biology",
"immunity",
"clinical",
"immunology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"colon"
]
| 2018 | HIC1 links retinoic acid signalling to group 3 innate lymphoid cell-dependent regulation of intestinal immunity and homeostasis |
Oomycetes in the class Saprolegniomycetidae of the Eukaryotic kingdom Stramenopila have evolved as severe pathogens of amphibians , crustaceans , fish and insects , resulting in major losses in aquaculture and damage to aquatic ecosystems . We have sequenced the 63 Mb genome of the fresh water fish pathogen , Saprolegnia parasitica . Approximately 1/3 of the assembled genome exhibits loss of heterozygosity , indicating an efficient mechanism for revealing new variation . Comparison of S . parasitica with plant pathogenic oomycetes suggests that during evolution the host cellular environment has driven distinct patterns of gene expansion and loss in the genomes of plant and animal pathogens . S . parasitica possesses one of the largest repertoires of proteases ( 270 ) among eukaryotes that are deployed in waves at different points during infection as determined from RNA-Seq data . In contrast , despite being capable of living saprotrophically , parasitism has led to loss of inorganic nitrogen and sulfur assimilation pathways , strikingly similar to losses in obligate plant pathogenic oomycetes and fungi . The large gene families that are hallmarks of plant pathogenic oomycetes such as Phytophthora appear to be lacking in S . parasitica , including those encoding RXLR effectors , Crinkler's , and Necrosis Inducing-Like Proteins ( NLP ) . S . parasitica also has a very large kinome of 543 kinases , 10% of which is induced upon infection . Moreover , S . parasitica encodes several genes typical of animals or animal-pathogens and lacking from other oomycetes , including disintegrins and galactose-binding lectins , whose expression and evolutionary origins implicate horizontal gene transfer in the evolution of animal pathogenesis in S . parasitica .
Saprolegnia species are watermolds or oomycetes that are endemic to probably all fresh water ecosystems . These understudied pathogens can cause destructive diseases of amphibians , crustaceans , fish and insects in aquaculture and in natural environments worldwide [1] , [2] . With the rise of fish as a principal source of animal protein , and the decline of wild fish stocks , aquaculture production has increased on average by 11% per year worldwide over the past ten years ( FAO Fishery Information ) . Intensive aquatic farming practices have resulted in explosive growth in pathogen populations , which has been exacerbated by the ban of malachite green as a pesticide . Losses due to microbial , parasitic and viral infections are the largest problem in fish farms nowadays , and have a significant effect on animal welfare and sustainability of the industry . The salmon farming industry is particularly affected by Saprolegnia parasitica . This pathogen causes Saprolegniosis ( also known as Saprolegniasis ) , a disease characterized by visible grey or white patches of mycelium on skin and fins , and subsequent penetration of mycelium into muscles and blood vessels [1] , [3] . It is estimated that 10% of all hatched salmon succumb to Saprolegnia infections and losses are estimated at tens of millions of dollars annually [2] . In addition to the damage to the aquaculture industry , the declines of natural salmonid populations have also been attributed to Saprolegnia infections [1] . More in-depth knowledge of the epidemiology , biology and pathology of the pathogen is urgently needed . A draft genome sequence of S . parasitica provides an excellent starting point to investigate the disease process at the molecular and cellular level and may lead to novel avenues for sustainable control of Saprolegniosis . Animal pathogens have evolved independently multiple times in lineages such as Stramenopila , Alveolata , Amebozoa , Euglenozoa and Mycota , as well as in numerous bacterial lineages . Oomycetes such as Saprolegnia belong to the kingdom Stramenopila ( Patterson , 1989 ) , syn . Straminipila ( Dick , 2001 ) , that includes photosynthetic algae such as kelp and diatoms , ubiquitous saprotrophic flagellates such as Cafeteria roenbergensis , and obligate mammalian parasites such as Blastocystis [4] , [5] . Although many Saprolegnia and related species are capable of causing diseases on a wide range of animal hosts including humans , relatively little is known about their mechanisms of pathogenicity . Among the oomycetes , most animal pathogens including S . parasitica belong to the class Saprolegniomycetidae ( Figure 1A ) . The oomycetes also include many plant pathogens and these are mainly concentrated within the class Peronosporomycetidae . There are a small number of interesting exceptions to this otherwise sharp dichotomy , including the mammalian pathogen Pythium insidiosum ( Peronosporomycetidae ) and the plant pathogens Aphanomyces euteiches and Aphanomyces cochlioides ( Saprolegniomycetidae ) [1] , [6] , [7] . Plant pathogens in the class Peronosporomycetidae include Phytophthora and Pythium species , downy mildew pathogens and white rusts ( Figure 1A ) . Well-known examples are the potato late blight pathogen Phytophthora infestans that caused the great Irish famine in the 1840s [8] , the soybean root rot pathogen Phytophthora sojae [9] , and the sudden oak death pathogen Phytophthora ramorum [10] . Sequenced and assembled genomes have been generated for P . sojae and P . ramorum [11] , P . infestans [12] , and for several relatives including the broad host range plant pathogens Phytophthora capsici [13] and Pythium ultimum [14] , the downy mildew pathogen of Arabidopsis , Hyaloperonospora arabidopsidis [15] , and the white blister rust of Brassicaceae , Albugo candida [16] . Genome analyses have revealed a bi-partite genome organization in these pathogens , in which gene dense regions containing clusters of orthologs with well-conserved sequences and gene order are interspersed with repeat-rich regions containing rapidly evolving families of virulence genes and numerous transposons [12] , [17] , [18] . The virulence gene families of plant pathogenic oomycetes encode numerous hydrolytic enzymes for degradation of plant carbohydrates , extracellular toxins such as NLP and PcF toxins , and at least three families of cell-entering effector proteins , RXLR effectors , CHXC effectors and Crinkler proteins [7] , [11] , [19] , [20] . These classes of effector proteins contain amino acid sequence motifs ( RXLR , CHXC , and LFLAK respectively ) that are involved in entry into the host plant cell [12] , [21] , [22] , [23] , [24] , [25] and effector domains that target diverse host physiological processes to suppress immunity and promote infection [17] . In the genome of the broad host-range necrotrophic oomycete Py . ultimum , which primarily targets immature or stressed plant tissues , enzyme families are expanded that enable degradation of readily accessible carbohydrates such as pectins , starch , and sucrose , while RXLR effectors appear to be completely absent [14] . In the obligate biotroph H . arabidopsidis , most gene families are smaller , especially those encoding hydrolytic enzymes [15] . Interestingly , EST libraries from the saprolegniomycete plant pathogen A . euteiches revealed the presence of Crinkler effectors ( but neither NLP toxins nor RXLR effectors ) [26] , suggesting that Crinklers are ancestral to oomycete pathogens [23] . So far , only limited genomic resources are available for animal pathogenic oomycetes . Analysis of small sets of EST data of S . parasitica [27] , [28] and Py . insidiosum [29] , revealed the presence of secreted protein families with potential roles in virulence such as glycosyl hydrolases , proteases , and protease inhibitors , as well as proteins involved in protection against oxidative stress . The S . parasitica data set included a host-targeting protein SpHtp1 ( S . parasitica host-targeting protein 1 ) that was subsequently demonstrated to enter fish cells through binding to a tyrosine-O-sulfated fish cell surface ligand [30] . Here we report the genome sequence and transcriptome analysis of S . parasitica , the first genome sequenced from an animal pathogenic oomycete . We compared the genome to those of plant pathogenic oomycetes , revealing distinctive genome expansions and adaptations tailored to the physiology of the respective hosts . This study adds to our understanding of mechanisms for invasion and colonization of animal host cells by eukaryotic pathogens .
The strategy of whole genome shotgun sequencing was applied to S . parasitica strain CBS223 . 65 , which is a strain isolated from infected pike ( Esox lucius ) . A combination of 454 ( fragment and 3 kb jumping libraries ) and Sanger ( Fosmid library ) sequencing data ( ∼50-fold average read coverage ) were used to assemble the genome , yielding an initial assembly length of 53 Mb and a scaffold N50 length of 281 kb ( see Table S1 for additional assembly statistics ) . Read coverage and rates of polymorphism were computed based on alignments of Illumina data ( generated for polymorphism discovery ) to the genome assembly . Based on the distribution of coverage and polymorphisms ( Figure S1A and S1B ) , we conclude that the assembly represents a composition of regions of diploid consensus ( 76% of the assembly ) with the remainder corresponding to separately assembled haplotypes , resulting from the high polymorphism rate with a peak of 2 . 6% ( Table S2 and Figure S1C ) . Taking into account the 24% of the assembly corresponding to separately-assembled haplotypes , the total assembled haploid genome size was adjusted to 42 . 3 Mb ( see Text S1 ) . Based on read coverage we estimated the total genome size to be 62 . 8 Mb . The remaining 20 . 4 Mb of genomic sequence were estimated to correspond to collapsed tandem repeat content in the genome assembly ( Text S1 ) . Read coverage analysis of the separated haplotypes and of regions showing loss of heterozygosity ( see below ) produced an overall genome size estimate of 63 Mb ( Text S1 ) , consistent with our effective assembled genome size estimate . The difference between the assembly size and the read coverage estimate likely results from tandem repeats collapsed in the assembly and uncertainties in read alignments , as observed in other oomycete genome sequence assemblies ( Table 1 ) . More than 98% of Trinity [31] de novo assembled transcripts from in vitro growth RNA-Seq data ( described in Material and Methods ) mapped to the genome assembly , indicating that the expressed gene content is very well represented by the assembled genome . The genome size of S . parasitica is consistent with the size range of most previously sequenced oomycete genomes , which rank from 45 Mb ( Py . ultimum ) [14] to 65 Mb ( P . ramorum ) [11] , far below the outlier of 240 Mb ( P . infestans ) [12] ( Table 1 ) . Approximately one-third of the assembled S . parasitica genome was found to correspond to regions exhibiting loss of heterozygosity ( LOH ) ( Figure S1C and Text S1 ) . LOH resulting from mitotic instability has been observed in other oomycete genomes [13] , [32] , [33] , and provides a potential adaptive mechanism that promotes expression of genetic diversity within a clonal pathogen population [13] . The prevalence of LOH within a S . parasitica population and the role LOH plays in S . parasitica evolution remain to be determined . A total of 20 , 113 coding gene models were computationally predicted , with 3 , 048 pairs of coding genes assigned as alleles within separately assembled haplotype regions ( Table S3 ) , yielding an adjusted coding gene count of 17 , 065 , similar in magnitude to counts of genes identified in other sequenced oomycete genomes ( Table 1 ) . There are 5 , 291 coding genes ( 31% of total ) found to reside within regions of LOH ( Table S3 ) . No obvious enrichment or depletion of biologically relevant gene functions could be detected within the defined regions of LOH that would suggest that the LOH observed had resulted from selection in the individual strain sequenced . The gene density found in S . parasitica is one of the highest reported so far for oomycetes , with one gene per 2 . 6 kb ( Table 1 ) . This is slightly denser than in Py . ultimum ( one gene per 2 . 9 kb ) and A . candida ( 2 . 9 kb ) , and much denser than in P . infestans ( one gene per 10 . 7 kb ) . Many of the S . parasitica genes are novel; only 40% of the S . parasitica predicted proteins have homologs with more than 50% sequence similarity to those from other organisms , including oomycetes . A conserved core-proteome of 4 , 215 proteins can be identified for plant pathogenic peronosporomycetes based on genomes from multiple Phytophthora species , H . arabidopsidis and Py . ultimum [14] . Of these 4215 , S . parasitica shares only 3518 orthologs ( Figure S2A ) , of which only about 40% show strong sequence similarity ( >50% ) . 20% of the core set is not detectable in the S . parasitica proteome ( Figure S2B ) . Although the genomes of the peronosporomycetes show substantial conservation of gene order ( synteny ) , little of this synteny is preserved in S . parasitica , as was observed for A . candida [16] . Interestingly , compared to other oomycetes , S . parasitica genes contain a larger number of introns ( Figure 1B , Table 1 ) . More than 73% of the S . parasitica genes contain at least one intron , compared to 50–60% in other oomycete species ( Table 1 ) . Among 4008 orthologs shared between S . parasitica and three Phytophthora species , the majority of the genes have different numbers of introns . For example , more than half of the S . parasitica genes have more introns than their orthologs in Phytophthora , and 15% of the S . parasitica genes have 5 or more additional exons compared to their Phytophthora orthologs ( Figure 1C ) . The intron abundance in S . parasitica potentially more closely matches the ancestral state , assuming a trend of intron reduction as found in animal and fungal lineages [34] . The S . parasitica genome has very few known mobile elements , which is consistent with its smaller size compared to the transposon-rich Phytophthora genomes . Of the 160 repeat families identified among all sequenced Phytophthora species , only one LTR retrotransposon family was found in the S . parasitica genome ( Figure S3A ) . This group of LTR elements , which occur at low copy numbers ( <20 ) in known oomycete genomes ( Figure S3B ) , is thus ancient . The largest transposon family in S . parasitica ( approx . 50 copies in the assembled sequence , and an estimated a few hundred copies in the genome ) belongs to the LINE repeat group ( Text S1 ) . LINE elements are abundant in animal genomes and play roles in genome evolution and modulation of gene expression [35] . Curiously , the S . parasitica LINE element family is absent from the Phytophthora genomes but shares sequence similarity with the LINE elements from animal genomes ( Figure S3C ) , raising the possibility that this family was acquired from an animal host . Eukaryotic protein kinases ( ePKs ) regulate a myriad of cellular activities by phosphorylating target proteins in response to internal or external signals . S . parasitica has one of the largest kinomes that have been identified to date , with 543 predicted ePKs . For example , S . parasitica has 65 more ePKs than the human kinome by the same prediction criteria ( Figure S4A ) . Eukaryotic protein kinases have been classified into eight major groups based on sequence similarity . We classified members of the S . parasitica ePK superfamily using a set of HMMs based on previously identified kinases ( Figure S4A ) . S . parasitica has a large expansion of tyrosine kinase-like proteins with 298 members and a large number of unclassified kinases ( 114 ) , suggesting novel functions performed by the S . parasitica kinome . Interestingly , the S . parasitica kinome contains several kinase families that have typically not been seen outside the metazoan clade , for example CAMK2 , NUAK , SNRK , and PHK from the Ca2+ calmodulin-dependent kinase group . Some of these “metazoan” kinases are shared with plant pathogenic oomycetes [36] , substantially pushing back the evolutionary origin of these kinases . We found 131 kinases containing predicted transmembrane helices ( Figure S4B ) , suggesting that S . parasitica has a large number of protein kinases that may function as cell surface receptors with roles in signaling . Such a large receptor repertoire may facilitate the recognition of stimuli from extracellular abiotic and biotic environments . The cell wall of a pathogen plays a central role at the host-pathogen interface . In particular , cell wall related proteins and polysaccharides are a large source of PAMPs ( Pathogen Associated Molecular Patterns ) of both animal and plant pathogens [37] , [38] . In addition , cell wall carbohydrate biosynthetic enzymes represent a potential target of antimicrobial compounds when similar activities are not encountered in the host . This is illustrated by the demonstration that the specific inhibition of chitin synthase ( CHS ) in Saprolegnia leads to cell death although chitin represents no more than 1% of the total cell wall carbohydrates of the pathogen [39] , [40] . Chitin is a structural crystalline polymer typically associated with the fungal cell wall . Historically , the absence of chitin in Phytophthora [41] has led to the general concept that oomycetes are devoid of chitin and that cellulose , the major load-bearing structural polysaccharide in oomycete walls , is a key feature distinguishing oomycetes from true fungi . However , the occurrence of chitin has since been demonstrated in various oomycete species belonging to the Saprolegniales [39] , [40] , [42] . In addition , GlcNAc-based carbohydrates that do not seem to correspond to crystalline chitin , but whose biosynthesis is most likely performed by putative chitin synthase gene products are present in the walls of Aphanomyces euteiches [43] . The S . parasitica genome appears to contain genes that encode enzymes involved in chitin biosynthesis , modification and degradation ( Table S4 ) . Out of these , six genes correspond to putative chitin synthases ( SPRG_02074 , SPRG_02554 , SPRG_04151 , SPRG_09812 , SPRG_06131 and SPRG_19383 ) . Interestingly , this number is higher than in other oomycetes where only one or two CHS putative genes have been detected [39] , [43] ( Table S4 ) . The only oomycete CHS gene product for which the function has been unequivocally demonstrated through heterologous expression and in vitro biochemical assays is CHS2 from Saprolegnia monoica [39] . Thus , as for most CHS genes from other oomycetes , the function of the S . parasitica genes annotated here as CHS remains to be demonstrated . As reported earlier in S . monoica [39] , two of the newly identified S . parasitica CHS gene products ( SPRG_09812 and SPRG_04151 ) contain a so-called MIT ( Microtubule Interacting and Trafficking ( or Transport ) ) domain . These MIT domains are possibly involved in the trafficking and delivery of the corresponding CHS at the apex of the hyphal cells , as previously suggested for S . monoica [39] . There are also a large number ( 14 ) of chitinase genes in S . parasitica compared to other oomycetes , which could represent many ancestral forms of oomycete chitinases ( Figure 1D ) . There are major structural and physiological differences between plant and animal cells , and thus the metabolisms of plant and animal pathogens have likely adapted accordingly to the respective host cellular environments . Pectin is a major constituent of plant cell walls and a target for extracellular enzymes produced by pathogenic and saprophytic microorganisms . Plant pathogenic fungi and oomycetes produce a large array of enzymes to degrade pectin , including polygalacturonase ( PG ) , pectin and pectate lyases ( PL ) , and pectin methylesterases ( PME ) . Animal cells lack a cell wall , and as might be expected , the pathogen S . parasitica encodes very few cell wall degrading enzymes in its genome . Genes encoding hydrolytic enzymes such as cutinase and pectin methyl esterases appear to be absent , and PL and PG genes are greatly reduced in numbers as compared to plant pathogenic oomycetes ( Table 2 ) . The remaining small numbers of PLs and PGs may play a role in the saprophytic life stage of S . parasitica in the aquatic environment outside of fish hosts . S . parasitica proliferates in host tissue rich in proteins and ammonium . Concomitantly , its pathways involved in inorganic nitrogen and sulfur assimilation have degenerated ( Figure 2A ) . The loss of these metabolic capabilities has occurred independently in the obligate oomycete plant pathogen H . arabidopsidis [15] , as well as within several lineages of obligate fungal plant pathogens , presumably due to the high level of parasitic adaptation in these organisms . Strikingly the same physical clusters of genes have been lost in each lineage , namely the genes encoding nitrate reductase , nitrite reductase , sulfite reductase and nitrate transporters ( Table S5 ) . Also in line with a protein-rich environment that is a major source of both carbon and nitrogen , the S . parasitica genome contains 56 genes predicted to encode amino acid transporters . Most of the S . parasitica transporters appear to be novel because less than 20 of the predicted amino acid transporters have orthologs or closely related paralogs in other oomycete genomes . Phylogenetic analysis shows there are lineage-specific expansions of amino acid transporter genes in the different oomycete genomes , with recently duplicated S . parasitica genes forming the largest group ( Figure 2B ) . The gene for phospholipase C ( PLC ) is absent in all of the sequenced peronosporomycete plant pathogens , but is present in S . parasitica ( SPRG_04373 ) . Phylogenetic analysis groups the S . parasitica PLC gene with that of other heterokont species ( Figure S5 , Table S6 ) . This shows that the S . parasitica PLC is most likely to be ancestral and that the absence of PLC in other oomycetes is due to gene loss . Peronosporomycete plant pathogens are sterol auxotrophs and their genomes are missing most genes encoding enzymes involved in sterol biosynthesis [44] . In contrast , analysis of the EST collection from A . euteiches and the S . parasitica genome predicts the existence of enzymes that function in a novel sterol biosynthetic pathway [26] which has been shown to lead to the synthesis of fucosterol in A . euteiches [45] . Importantly , one of the genes SPRG_09493 encodes a CYP51 sterol-demethylase ( Figure S6 ) , a major target of antifungal chemicals that could perhaps also be used to combat Saprolegniomycetes . Like plant pathogens , S . parasitica presumably secretes a battery of virulence proteins to promote infection . Due to co-evolution with the host , virulence proteins are typically rapidly evolving and may appear to be unique to the species , or encoded by recently expanded gene families [17] , [19] . The S . parasitica genome contains a large number of genes ( 11 , 825 ) that are not orthologous to any known genes in other species ( Figure S2A and S2B ) , and many recently expanded gene families . There are at least 87 pfam domains that are either unique or show recent expansions in S . parasitica as compared to other oomycete species ( Table S7 ) . An estimated 970 proteins ( Table S8 ) were predicted to be extracellular based on previously established bioinformatics criteria [11] , [12] , such as the presence of a eukaryotic signal peptide , and lack of targeting signals to organelles or membranes . Many of the expanded families appear to function at the exterior or cell surface of the pathogens , such as proteins containing CBM1 ( Carbohydrate Binding Module Family I according to the CAZy database ( http://www . cazy . org/; [46] ) , ricin B lectin , Notch domains , and also numerous peptidases . Among the proteins that are unique to S . parasitica compared to plant pathogenic oomycetes , the largest families have similarities to animal-pathogenesis-associated proteins , such as disintegrins , ricin-like galactose-binding lectins and bacterial toxin-like proteins ( haemolysin E ) . Oomycetes contain an unusually large number of proteins with novel domain combinations , recruited from common metabolic , regulatory and signaling domains [47] , [48] . S . parasitica contains in total 169 novel domain combinations that are specific to this pathogen ( Table S9 ) . As described above , some of the lineage-expanded domains such as CBM and ricin are used for novel combinations to form composite proteins . Additional domains used for novel combinations are the cytochrome p450 and tyrosinase domains . Proteins carrying S . parasitica-specific domain combinations are significantly enriched ( hypergeometric test , p<0 . 001 ) in predicted secreted proteins ( 3 . 6% of secretome ) , whereas only 1 . 2% of total proteins have S . parasitica-specific novel combinations . The enrichment in secreted proteins is strongly suggestive of a role for the novel domain combinations in pathogenesis . There are about 1000 proteins that are predicted to be secreted by S . parasitica , based on criteria used for secretome prediction in other oomycetes [11] , [12] . Two groups of proteins dominate the secretome of S . parasitica: proteases and lectins ( Figure 3A , Table S8 ) . There are over a hundred members in the each of the two groups . In the proteome , S . parasitica has one of the largest repertoires of proteases ( 270 ) known to date compared to most other single cell or filamentous eukaryotic pathogens ( such as most sequenced fungi species ) that typically have between 70 and 150 proteases . In almost every family of proteases , including cysteine- , serine- and metallo-proteases ( Table 2 ) , there is an expansion in S . parasitica compared to P . sojae ( Figure 2C ) . The most relatively abundant family of proteases is the papain-like peptidase_C1 proteins , comprising 48 proteins; the other oomycetes contain only about 20 proteins . The majority of papain-like peptidase_C1 genes ( 80% ) have been generated by S . parasitica lineage-specific gene duplications and form a lineage-specific clade in the phylogenetic reconstruction ( Figure 2D ) . Amongst the cell-surface associated proteins , ricin_B_lectin-like proteins and CBM1 domain-containing proteins are most abundant ( Table 2 ) . There are 40 ricin-like and 40 CBM1 genes in S . parasitica , a large expansion compared to other known oomycetes . In some cases , these domains are fused to other secreted protein domains having catalytic activities ( protease and cellulose ) to form novel proteins unique to oomycetes ( Figure 3B ) . However , typical peronosporomycete proteins were not found , such as CBEL ( Cellulose Binding Elicitor Lectin ) , which contains a CBM1-PAN domain association and mediates the binding of mycelium to cellulosic substrates [49] . The saprolegniomycete plant pathogen Aphanomyces euteiches also lacked CBEL proteins [26] . We investigated whether culture filtrates containing secreted proteins from S . parasitica could degrade trout immunoglobulin M ( IgM ) as previously found for bacterial fish pathogens [50] . Although no effect was observed when supernatants of two-day old cultures were incubated with an IgM enriched fraction ( data not shown ) , the 7-day post-inoculation supernatant degraded the IgM protein fraction within several hours ( Figure 4A ) . No degradation of trout IgM was detected when heat-treated 7-day post-inoculation supernatant was used . The protease inhibitors EDTA ( a metalloproteinase inhibitor ) and PMSF ( serine protease inhibitor ) showed partial inhibition of the IgM-degrading activity while E-64 ( a cysteine protease inhibitor ) did not show any inhibition . The combination of EDTA and PMSF prevented IgM degradation and the detection was similar to the pea broth control . These results suggest that secreted proteases from S . parasitica could degrade fish IgM and that metalloproteinases and serine proteases may be the classes involved in this process . To further characterize the IgM-degrading properties of S . parasitica proteases , a serine protease ( SPRG_14567 ) was selected from the S . parasitica genome based on our observation that this protease possesses a secretory signal peptide ( Figure 4B ) and is highly expressed ( RNA-Seq data ) ( Figure 4C ) . Interestingly , SPRG_14567 showed strong degrading activity towards trout IgM while no activity was detected when control E . coli soluble proteins were used ( Figure 4B and 4E ) . This indicates that this serine protease is capable of degrading fish IgM and could be a virulence factor that combats the activity of fish immunoglobulins against Saprolegnia . To establish a successful infection , pathogens often deliver effector proteins and toxins into the host cytoplasm to manipulate host immunity [7] , [24] , [51] . Analogous to bacterial Type II secreted toxins that enter host cells via lipid-receptor-mediated endocytosis , plant pathogenic oomycetes utilize a host-targeting domain to deliver effectors into plant cells . Hundreds of effectors carrying the host targeting motifs of RXLR and LFLAK have been identified in plant pathogenic oomycetes [11] , [12] , [22] . However , these large families of Crinkler and RXLR effectors appear to be absent in S . parasitica ( Table 2 ) . Using sensitive BLAST and HMM searches based on the RXLR domain and C-terminal domains of effectors , no plant pathogen-like RXLR effectors could be detected in the genome . Bioinformatic searches with the de novo motif-finding program MEME did not identify other putative host-targeting motifs . Despite the absence of large RXLR effector families , S . parasitica does have a small family of host targeting proteins related to SpHtp1 , which do contain an N-terminal RXLR sequence . Interestingly one of these proteins was shown to translocate into fish cells [28] and entry required the N-terminal leader sequence [30] . The lack of sequence similarity between any part of SpHtp1 and RXLR-proteins from plant pathogen oomycetes , except for the three proximally located amino acid residues , suggests that the presence of the RXLR-sequence in SpHtp1 is currently unclear . No significant matches were found to Crinkler effectors , suggesting they are absent from S . parasitica . Using search criteria that do not rely on sequence homology , such as induction of expression during the pre-infection stage , presence of secretion signals , and signatures of fast evolution , several candidates ( Table S10 ) for host-targeting proteins ( including SpHtp1 ) were identified in S . parasitica . None of these candidates have homology to known proteins , and their functions are currently unknown . The commonalities between animal and plant pathogenic oomycetes are highlighted by their shared PAMPs . Proteinaceous PAMPs found in plant pathogenic oomycetes , namely CBM1 [37] , elicitins , and Cys-rich-family-3 proteins [14] are found in S . parasitica ( Figure S7A ) . Elicitins are extracellular lipid transfer proteins that elicit defense responses in some species of plants , especially Nicotiana [52] . There are 29 elicitin-like proteins in S . parasitica . Phylogenetic reconstruction shows that the majority of the S . parasitica elicitins form three lineage-specific clades , distant from the canonical elicitin group [53] ( Figure S7B ) . We also detected six YxSL[RK] containing secreted proteins , previously identified as candidate effectors [14] , that also show sequence divergence from known members of this family . It is an intriguing question whether animal innate immune systems can detect these potential PAMPs , as does the plant innate immune system . For example it seems unlikely that CBM1 can act as immune elicitor for animal cells since binding to the plant cell wall cellulose is required to induce immune responses in plants [37] . To study the polymorphism and evolutionary rate of S . parasitica genes , we sequenced a related strain , VI-02736 , which was isolated from an infected Atlantic salmon , for comparison . More than of 90% of the CBS223 . 65 reference sequence was covered by VI-02736 reads , allowing us to identify 1 , 467 , 567 SNPs between the two strains , giving an average SNP rate of 3 . 3% ( Text S1 , Table S11 and Figure S8A and S8B ) . We have also determined that LOH is unique to the CBS strain , as it is apparently absent in the sequenced VI-02736 strain ( Text S1 , Figure S8C and S8D ) . The Ka/Ks ratio ( Ka - number of non-synonymous substitutions per non-synonymous site , Ks - number of synonymous substitutions per synonymous site ) was calculated for the annotated S . parasitica gene set . The set of 3518 oomycete core ortholog genes gives a median Ka/Ks ratio of 0 . 05 , a rate comparable to previously published Ka/Ks rate of the conserved gene dense region of Phytophthora sibling species and related strains [54] . Non-parametric Z tests between the core ortholog group and a particular gene family ( listed in Table 2 ) were performed to determine which family shows elevated substitution rates . Among the pathogenesis-related gene families , we identified four groups , namely elicitins , disintegrins , host targeting proteins and haemolysins , that have significantly elevated median Ka/Ks ratios as compared to the core-ortholog groups ( Figure S9A ) . Between the two S . parasitica strains , the highest Ka/Ks ratio was observed in the haemolysin E family , a group of toxin-like genes that were possibly horizontally acquired from bacteria ( see below ) . Interestingly , similar patterns of elevated Ka/Ks ratios were also found between haplotypes of the reference strain ( Figure S9B ) . Three families , the disintegrins , elicitins and haemolysins , show significant elevations in Ka/Ks as compared to the core ortholog group ( Figure S9B ) . Horizontal gene transfer events from bacteria and archaea appear to have contributed to some of the novel biosynthetic pathways found in the oomycetes [47] . By utilizing codon usage and protein domain classification ( see Methods ) , at least 100 genes could be identified with a potential phylogenetic origin outside the super-kingdom of Chromalveolates by using the program Alien_hunter [55] and by interrogating Pfam search results . Among these , we further identified around 40 genes belonging to 5 families that could potentially be associated with pathogenesis ( Table 3 ) . Many of these genes appear to have been acquired from bacterial species , in particular from Proteobacteria . Some potential acquisitions may have occurred relatively recently as they only occur in Saprolegnia and seem to have nucleotide compositions predictive of foreign acquisition ( Alien_hunter score >50 ) . A caveat for our analysis is that S . parasitica is the only available genome so far outside of the plant pathogenic oomycetes . There are a great variety of basal oomycetes [56] , [57] that do not have genome sequence information and have not been investigated . Therefore , it could be that these HGT events could have occurred in some ancestral oomycetes . Several groups of extracellular enzymes were potentially acquired from bacteria; for example , the CHAP ( cysteine , histidine-dependent amidohydrolases/peptidases ) family and a family of secreted nucleases ( Table 3 ) . The presence of the CHAP family ( pfam hit E value<1e-5 ) is unexpected in S . parasitica , because it is commonly associated with bacterial physiology and metabolism [58] . These genes have undergone repeated duplications in S . parasitica , resulting in an expansion of gene numbers . Another potential bacterial acquisition is a family of toxin-like proteins similar to haemolysin ( HlyE ) , a pore-forming toxin from enterobacteria such as Salmonella [59] . These genes have also undergone recent duplication resulting in nine copies in S . parasitica . Two members ( SPRG_03140 , SPRG_20514 ) of the HlyE family are expressed during infection . Another distinctive feature of the S . parasitica secretome is the presence of animal-like surface proteins . The phylogenetic affinities of the two groups , distinguished by gal_lectin-like and disintegrin-like domains , suggest a possible origin via HGT , but from different sources . Gal_lectin refers to the D-galactoside binding lectin initially purified from the eggs of sea urchin [59] , [60] , [61] , representing a group of lectins that occurs widely on fish eggs and skins [62] . Phylogenetic analysis shows that the closest homologs are animal gal_lectins ( Figure 3C ) . The S . parasitica gal_lectin genes are among the most highly induced and highly expressed genes in pre-infection and infection stages , suggesting that gal_lectin may facilitate adhesion and invasion of fish cells . The gal_lectin genes contain an unusually large number of introns ( 7 introns ) , as do mammalian and fish gal_lectin genes , further suggesting a common origin . The S . parasitica genes have a codon usage similar to core orthologs , in contrast to other candidate HGT genes , suggesting they have been in the S . parasitica genome long enough to be largely assimilated . The S . parasitica disintegrin genes were potentially acquired from bacteria ( Figure 3C , Table 3 ) and have since expanded . Disintegrins were initially identified as proteins preventing blood clotting in viper venoms [63] . In animals , disintegrins inhibit aggregation of the platelets by binding to the integrin/glycoprotein IIb-IIIa receptor . The crucial amino acid motif CRxxxxxCDxxExC , mediating ligand binding , is conserved in the S . parasitica disintegrins . All 16 disintegrin genes in S . parasitica are expressed in the pre-infection stages ( see below ) , with several members belonging to the top 1% most highly expressed genes , suggesting that they may play a role interacting with animal hosts . However , we have not been able to experimentally demonstrate the role of disintegrins in pathogenesis so far ( Text S1 and Figure S10 ) . S . parasitica has , like most other oomycetes , clearly defined life stages , including motile zoospores that are able to swim , encyst and germinate upon attachment to its host tissue . We performed strand-specific Illumina RNA-Seq analysis of 4 developmental stages ( mycelium , sporulating mycelium , cysts , and germinating cysts [3–5 hours] ) of S . parasitica , as well as a time course analysis ( 0 , 8 and 24 hours ) of a rainbow trout fibroblast cell-line ( RTG-2 ) challenged with cysts of S . parasitica . RNA-Seq reads were mapped to the S . parasitica genome and gene annotations , and transcript abundance values were quantified as RPKM ( reads per kilobase transcript length per million reads mapped ) . Large numbers of genes were differentially expressed in the various life stages and conditions . Cyst and germinating-cyst stages showed similar expression profiles ( linear correlation R2 = 0 . 93 , p<0 . 001 ) , with less than 6% of all the genes showing >4-fold expression differences ( p<0 . 001; Figure 5A ) . In contrast , the other developmental life stages and infection time course showed somewhat larger differences ( R2 of 0 . 84–0 . 88 ) , with up to 27% of genes differentially expressed between cysts and mycelia ( > = 4-fold difference; p< = 0 . 001 ) . The time course experiment shows the relative abundance of the host and pathogen transcripts changing as infection progressed ( Figure 5B ) . At the 0 hour and 8 hours time points , very few pathogen transcripts were detected ( less than 1% and 3% , respectively ) whereas at 24 hours , 63% of the transcripts were derived from the pathogen . At 24 hours , the transcript profile of the pathogen closely resembled that of in vitro-grown mycelia , with only 3 . 5% of all genes differentially induced between the two stages ( >4 fold differences , p<0 . 001 ) . Compared to the cysts used as inoculum , 7 . 2% of genes were induced after 8 hours of infection and 10% of genes were induced by 24 hours ( >4 fold differences , p<0 . 001 ) ( Figure S11A ) . We have defined the stage prior to host tissue colonization ( germinating cyst ) as the pre-infection stage . In plant pathogenic oomycetes , germinating cysts express many pathogenesis-associated genes [12] , [64] , [65] , [66] , [67] . The previously characterized S . parasitica gene encoding the host targeting protein SpHtp1 [28] is induced more than 100-fold in the pre-infection stage as compared to other stages . At the pre-infection stage , 10% of all genes were induced as compared to the mycelial stage ( >4 fold differences , p<0 . 001 ) ( Figure 5C; Figure S11B; Table S12 ) . The profile of the germinating cysts - resembled that of the 8 hours fish cell infection ( R2 = 0 . 88 , p<0 . 001 ) , with only 4% of genes showing induction ( >4-fold differences , p<0 . 001 ) . For genes that encode proteins specific to or expanded in S . parasitica as compared to other oomycetes , around 25% were induced in germinating cysts compared to mycelia ( >4-fold differences , p<0 . 001 ) ( Figure S11C , Table S13 ) . A total of 14 protein families showed both lineage-specific domain expansion and up-regulation in germinating cysts ( Figure 5D ) . The largest groups are proteins carrying ankyrin domains or lectin domains , which suggests the importance of protein-protein and protein-carbohydrate interactions in the initial stages of host colonization . Another group of proteins belonging to this category are transporters such as ion transporters and sodium symporters , suggesting that metabolic exchange processes are active during the establishment of early infection . The RNA-Seq data also suggest that the very large S . parasitica kinome plays an important role during the infection process . Around 10% of all kinase genes , spanning all major kinase classification groups , showed 4-fold or more induction in germinating cysts compared to mycelia ( >4-fold differences , p<0 . 001 ) ( Figure S11C ) . Many of the numerous S . parasitica proteases were expressed in a specific life stage or distinct point during the infection process ( Figure 5D/E , Table S14 ) . One group of peptidases ( 6% ) was expressed in germinating cysts . These include subtilase proteins ( SPRG_15005 ) that carry ricin lectin domains and are highly expressed in germinating cysts . A large group of peptidases ( 19% ) were induced during the interaction with fish cells . There are 7 protease-inhibitors encoded in the genome; and they also showed patterns of differential expression ( Figure 5C ) . The Kazal peptidase inhibitor ( SPRG_09563 ) was most highly expressed in cysts and germinating cysts . To confirm the RNA-seq data , we have also performed qPCR for a set of disintegrin genes , showing expression in cysts and germinating cysts , but no detection in mycelium and sporulating mycelium as seen for the RNA-seq data ( Figures S11 C , D ) . Taken together , the RNA-seq data reveal that members of the kinases , proteases , disintegrins and gal_lectins show upregulated expression in the pre-infection stage ( Table S13 and S14 ) , which warrants using these candidate genes for future pathogenesis related studies .
The genome sequence of S . parasitica , together with transcriptome and polymorphism data , reveals a predicted core proteome very similar to those of plant pathogenic oomycetes , but an adapted proteome strongly aligned to its animal-pathogenic lifestyle . From the genome sequence and the RNA-Seq data we could identify several groups of proteins predicted to be secreted at the pre-infection stages that might facilitate early interactions with the hosts of S . parasitica . Several of these proteins seem to be unique to S . parasitica and may have evolved to interact specifically with fish cells . They are predicted to be targeted to the extracellular environment or incorporated into the exterior surface of the pathogen . Several of the proteins contain CBM domains ( fungal cellulose binding domain ) , ricin-like domains , Notch-like domains and/or various peptidase domains . Most putative early interaction proteins have potential roles in pathogenesis , such as animal-cell-surface-like proteins ( disintegrin , gal_lectin ) and haemolysin E toxin-like proteins . Lectins could help cysts to bind to host skin . Since the lectins are highly expressed during the initial and later infection stages , we hypothesize that they play an important role in cell-cell contact throughout the interaction . This would also suggest that an intimate contact with the host cell is required for pathogenesis , suppression of host defence processes , and/or nutrient uptake . Following attachment , the pathogen engages a large arsenal of potential virulence factors in the form of proteases to attack the host tissue . Interestingly , S . parasitica has one of the largest numbers of proteases found in any organism . At least one protease ( SPRG_14567 ) was found to be able to degrade IgM , suggesting an active role in suppressing initial immune responses , as fish IgM's are able to bind to infection related antigens , even in the absence of prior immunization . In comparison to plant pathogenic oomycetes , S . parasitica has no canonical RXLR or Crinkler effector genes , nor NLP toxin genes in its genome . Nevertheless , one small protein family has been found for which one member was shown to translocate inside trout cells [2] , [30] , which suggests that the interaction of S . parasitica with its hosts is more subtle than a simple necrotrophic interaction based on secretion of toxins and protein degradation . In fact , we speculate that the initial stages of the interaction may involve a more ‘biotrophic’ approach by the pathogen , whereby the immune response of the host is avoided or even suppressed during initial colonization via , for example , proteases or effector proteins . Following this biotrophic stage , the host tissue is bombarded with proteases , lipases , and lysing enzymes . If S . parasitica was a plant pathogen , it would thus have been classified as a ‘hemi-biotroph’ and not a saprotroph as its name would suggest . Further experiments are required to demonstrate that this is indeed the case . Pathogenicity towards animals has evolved independently in both the fungi and oomycetes . It has also evolved at least three additional times within the kingdom Stramenopila: within the Pythiales ( Figure 1A ) , within the genus Aphanomyces [6] and in the non-oomycete Blastocystis . Analyses of the genomes or sequenced ESTs of these other pathogens reveals some interesting parallels . For example , based on only a small set of ESTs of the oomycete human pathogen Py . insidiosum [29] , an identified expressed lectin CBM-encoded transcript has been implicated in the process of pathogenesis . Blastocystis hominis is a Stramenopile human pathogen with a very small genome of 19 Mb encoding only 6020 genes , and is phylogenetically very divergent from oomycetes [4] . Despite the large evolutionary distance , the genome of Blastocystis shows reduction of nitrogen and sulfite metabolic pathways , similar to what is seen in obligate plant pathogens . In addition , similar to the acquired lectins in S . parasitica , Blastocystis has also horizontally transferred genes with animal like features , such as genes encoding Ig domains [4] . The fungal pathogen Batrachochytrium dendrobatidis that causes global amphibian decline , inhabits an aquatic environment like S . parasitica , and also causes diseases on animal skin and tissues . Analysis of its genome revealed patterns of expansion of protease families [68] . Although the particular families are different , B . dendrobatidis possesses several families of lineage expanded proteases such as metallopeptidase ( M36 ) . The extensive LOH observed in S . parasitica , covering approximately one-third of the genome and also the high rate of polymorphism ( 2 . 6% ) in the remainder of the genome are similar to recent observations in the genome of the oomycete plant pathogen Phytophthora capsici , where LOH was found to be associated with changes in mating type and pathogenicity [13] . We speculate that , as in P . capsici , LOH may provide a mechanism in S . parasitica for rapidly expressing diversity within a population , fixing alleles , and enabling rapid adaptation to its environment . The evolution of plant and animal pathogenesis in oomycetes has been associated with several major molecular events ( Figure 6 ) . Since both Phytophthora and Saprolegnia have large kinomes , the expansion of kinases is likely a relatively early event in oomycete evolution . The comparison of the animal pathogen S . parasitica with plant pathogens with different lifestyles has shown that surviving on ammonium rich tissue has led to a reduction of metabolic pathways independently in S . parasitica and obligate oomycete plant pathogens . Similar reductions have also occurred in obligate fungal pathogens [69] and in the distantly related stramenopile human pathogen Blastocystis [4] . The evolution of plant pathogenicity has been associated with a series of reduction events such as intron loss , chitin loss and sterol loss . Some of these losses may be due to evasion of plant immunity; for example chitin can act as a PAMP to trigger plant defense responses . The evolution of plant pathogens has been accompanied by expansions of large repertoires of effectors , which have been shown to modulate plant host physiology . In contrast , the development of animal pathogenesis has been facilitated by expansion of proteases and horizontally acquired lectins and toxins .
Saprolegnia parasitica isolate CBS223 . 65 was isolated from young pike ( Esox lucius ) in 1965 and obtained from Centraal Bureau voor Schimmelcultures ( CBS ) , the Netherlands . Saprolegnia parasitica isolate VI-02736 ( N12 ) was obtained from parr of Atlantic salmon in Scotland in 2002 ( Lochailort ) [70] and kindly provided by Dr . Ida Skaar ( Norwegian Veterinary Institute ) . Both isolates were maintained on potato dextrose agar ( Fluka ) . For genomic DNA isolation , S . parasitica was grown for three days at 24°C in pea broth ( 125 g L−1 frozen peas , autoclaved , filtered through cheese cloth , volume adjusted to 1 L and autoclaved again ) . Genomic DNA extraction was adapted from Haas et al [12] . Fresh mycelia ( ∼1 g ) were ground to a fine powder under liquid nitrogen , mixed with 10 mL of extraction buffer ( 0 . 2 M TrisHCl pH 8 . 5 , 0 . 25 M NaCl , 25 mM EDTA , 0 . 5% SDS ) , 7 mL Tris-equilibrated phenol and 3 mL of chloroform∶isoamyl alcohol ( 24∶1 ) , incubated at room temperature for 1 hr and centrifuged at 6 , 000 g for 30 m . The aqueous phase was extracted with equal volume of chloroform∶isoamyl alcohol ( 24∶1 ) and centrifuged at 10 , 000 g for 15 min . 50 µl of 10 mg/ml RNase A was added to the aqueous phase and incubated for 30 min at 37°C . Isopropanol ( 0 . 6 volumes ) was added , mixed gently , and then the DNA was precipitated on ice for 30 min . DNA was collected by centrifugation at 10 , 000 g for 20 min , washed with 70% ethanol , dried and resuspended in DNase/RNase-free water . DNA was checked for quality and RNA contamination by gel electrophoresis using 0 . 8% agarose . The life stages of S . parasitica were harvested as described by van West et al . [28] . Zoospores and cysts were collected by pouring the culture filtrate through a 40–70 µm cell strainer and concentrated by centrifugation for 5 min at 1500 g . RNA was isolated from mycelia and sporulating mycelia using the RNeasy kit ( Invitrogen ) according to the manufacturer's protocol . RNA from zoospores , cysts and germinating cysts was resuspended in TRIzol ( Invitrogen ) and aliquoted as 1 ml portions in 2-ml screw-cap tubes containing 10–35 glass 1 mm diameter beads ( BioSpec ) and immediately frozen in liquid nitrogen . Frozen cells were processed with a FastPrep machine ( ThermoSavant ) and shaken several times at a speed of 5 . 0 for 45 sec until defrosted and homogenized . TRIzol RNA isolation was then performed according to manufacturer's recommendations ( Invitrogen ) . Host-induced gene expression was measured in rainbow trout ( Oncorhynchus mykiss ) gonadal tissue continuous cell line RTG-2 , obtained from the American Type Culture Collection ( ATCC CCL-55 ) [71] . RTG-2 cells were grown to a monolayer at 24°C ( no CO2 ) and washed twice with Hank's Balanced Salt Solution ( HBSS ) . Cells were harvested in approximately 5 ml of HBSS and S . parasitica cysts were added ( 5×104 cysts per 75 cm flask ) and the mixture was then incubated at room temperature for 15 min . At this time , the time 0 hr ( T0 ) stage sample was collected , by gently pouring off medium ( and also loose/non-attached cysts ) and then adding TRIzol reagent for RNA extraction . Other samples were incubated for 8 hr ( T8 ) or 24 hr ( T24 ) at 24°C ( no CO2 ) in Leibovitz ( L-15 ) growth medium ( Gibco ) supplemented with 10% foetal calf serum ( Biosera ) , 200 U ml−1 penicillin and 200 µg ml−1 streptomycin ( Fisher ) . Cells were harvested for RNA extraction by pouring off medium and adding TRIzol . Cells were loosened using a cell scraper ( Fisher ) and the suspension was aliquoted as 1 ml portions into 2-ml screw-cap tubes containing 10–35 glass 1 mm diameter beads ( Biospec ) and immediately frozen in liquid nitrogen . Frozen cells were processed using a FastPrep machine and shaken several times at a speed of 5 . 0 for 45 sec until defrosted and homogenized . TRIzol RNA isolation was then performed according to manufacturer's recommendations . Infected and uninfected fish tissue was collected from a trout ( O . mykiss , caught in a commercial Scottish hatchery ) that showed lesions caused by S . parasitica infection . Tissue was collected from the lesion ( body ) and from an infected anal fin . As an uninfected control , similar tissue not showing symptoms of infection was collected from the same fish . Tissue was ground up in liquid N2 and RNA was isolated using the RNeasy kit ( QIAGEN ) according to manufacturer's recommendations . Infected and uninfected Atlantic salmon ( Salmo salar ) egg samples were collected from a commercial Scottish hatchery . Five eggs per sample were ruptured using a needle and placed into 5 ml of TRIzol and RNA was isolated according to manufacturer's recommendations with an additional phenol-chloroform extraction step . Multiple sample batches were pooled to obtain sufficient material for RNA-Seq library construction . All RNA samples were resuspended in DNase/RNase-free water and treated with Turbo DNA-free DNase ( Ambion ) according to manufacturer's recommendations . RNA was checked for quantity and purity using a Nanodrop spectrophotometer ( Thermo Scientific ) and 1% agarose gel electrophoresis . Samples were stored at −80°C and were not defrosted until use . 454 fragment and 3 kb jumping whole genome shotgun libraries were generated for isolate CBS 223 . 65 as previously described [72] . Libraries were sequenced with ∼400 base single end reads ( Titanium chemistry ) using a 454 GS FLX sequencer following the manufacturer's recommendations ( 454 Life Sciences/Roche ) . Approximately a total of 22-fold sequence coverage was generated from fragment and 3 kb jumping libraries combined . A 40 kb insert Fosmid whole genome shotgun library from isolate CBS 223 . 65 was generated using the EpiFOS fosmid cloning system following manufacturer's recommendations ( Epicentre ) . The Fosmid library was end-sequenced with ∼700 bp reads using Sanger technology to approximately 0 . 3-fold coverage using a 3730xl DNA analyzer following manufacturer's recommendations ( Applied Biosystems/Life Technologies ) . For variant calling , Illumina whole genome shotgun fragment libraries were generated for isolates CBS 223 . 65 and VI-02736 as previously described [73] and sequenced with 76 base paired-end reads to a minimum of 70-fold sequence coverage using an Illumina Genome Analyzer II ( Illumina ) following the manufacturer's recommendations . Illumina strand-specific dUTP RNA-Seq libraries were generated for all RNA samples as previously described [74] with the following modifications . The mRNA was processed using the Dynabeads mRNA purification kit ( Invitrogen ) and incubated with RNA fragmentation buffer ( Affymetrix ) at 80°C for 1 . 5 to 3 . 5 minutes depending sample quality . Indexed adaptors for Illumina sequencing were ligated onto end-repaired , A-tailed cDNA fragments by incubation with 4 , 000 units of T4 DNA ligase ( New England Biolabs ) in a 20 ml reaction overnight at 16°C . 16 to 21 cycles of PCR were used to amplify sequencing libraries . Libraries were purified using 1–3 rounds of AMPure beads ( Beckman Coulter Genomics ) following manufacturer's recommendations . Libraries were sequenced with 76 base paired-end reads using an Illumina Genome Analyzer II following manufacturer's recommendations ( Illumina ) generating a total of 124 million reads . Sanger Fosmid paired-end reads were quality trimmed using the ARACHNE ‘Assemblez’ module ( http://www . broadinstitute . org/crd/wiki/index . php/Assemblez ) . Read headers were amended to contain template and pairing information . Sanger , 454 fragment and 3 kb jumping reads were assembled using 454's Newbler assembler version 04292009 ( 454 Life Sciences/Roche ) . The resulting assembly was further processed using the ARACHNE ‘HybridAssemble’ module ( http://www . broadinstitute . org/crd/wiki/index . php/HybridAssemble ) using the 454 assembly , Sanger and 454 read data as input with option ‘RecycleBadContigs’ turned off allowing for extra copies of repeat sequences to be assembled . The Arachne ‘AddReadsAsContigs’ module was run to allow assembly of additional repetitive sequence . The assembly was screened for known sequencing vector sequences using BLAST and contigs with hits to known sequence vectors were removed . The assembly was further post-processed by removing contigs and scaffolds less than 200 bp and 2 kb in length respectively . Illumina shotgun sequence data were used to identify polymorphisms in two S . parasitica strains ( CBS 223 . 65 and VI-02736 ) . Illumina paired-end fragment reads from strains CBS 223 . 65 and VI-02736 were independently aligned to the S . parasitica CBS 223 . 65 reference assembly using the BWA aligner [75] using default settings . Read alignments were sorted by scaffold and position along the reference assembly . SNP calling was performed using the GATK Unified Genotyper module [76] . Variant Call Format ( VCF ) files containing SNP calls were filters for low quality using parameters: AB>0 . 75 && DP>40 ∥ DP>500 ∥ MQ0>40 ∥ SB>−0 . 10 . SNP calls for CBS 223 . 65 and VI-02736 can be retrieved from the Broad Institute Saprolegnia parasitica genome database website ( http://www . broadinstitute . org/annotation/genome/Saprolegnia_parasitica/MultiDownloads . html ) . SNP calls and depth of read coverage information were parsed from the VCF file described above and analyzed in non-overlapping 5 kb windows ( Figure S1 ) . Using this information genome segments were partitioned into three groups: separated haplotypes ( coverage depth ranging 20–55-fold and SNP rate <1% ) , diploid homozygous involving LOH ( coverage depth ranging 56–90-fold and SNP rate <1% ) , and diploid heterozygous ( coverage depth ranging 40–90-fold and SNP rate > = 1% ) . Coverage of separated haplotype regions peaks at ∼40-fold and the regions are mostly devoid of SNPs . The region corresponding to the diploid consensus exhibits ∼60-fold coverage and nearly a 3% SNP rate . The peaks in Figure S1B corresponding to the separated haplotype and consensus diploid regions are connected by a small ridge , which correspond to windows spanning boundaries between the different kinds of regions . The coverage for the diploid consensus regions is not exactly double as compared to the predicted separately assembled haplotype regions , and is less than the diploid homozygous regions; most likely this results from the relative difficulty of aligning short Illumina reads to diploid consensus sequences in the context of the high polymorphism rate observed . Individual genes located in haplotype contigs were assigned as likely allelic pairs based on SNP rate , depth of coverage , and taking into consideration best reciprocal blast matches and synteny between separately assembled haplotype contigs . Gene finding used both evidence-based ( including EST , RNA-Seq and homology data ) and ab initio methods . Gene-finding algorithms FGenesH , GeneID and GeneMark were trained for S . parasitica using existing gene and EST datasets . Then a statistical sampling of gene calls as well as genes of interest were manually curated , and the results were used to validate gene calls and fine-tune the gene caller . RNA-Seq data was incorporated into gene structure annotations using PASA [77] as described in Rhind et al . [78] . Subsequently , the annotated total gene set was subjected to Pfam domain analysis , OrthoMCL clustering analysis and KEGG metabolic pathway analysis . Illumina RNA-Seq data was processed as follows . Sequencing adaptors were identified and removed from reads by exact match to adaptor sequences . Reads were aligned to S . parasitica gene transcripts using Bowtie ( allowing up to 2 mismatches per read , and up to 20 alignments per read ) . Transcript levels were calculated as FPKM ( fragments per kilobase cDNA per million fragments mapped ) . The program EdgeR [79] was used to identify differentially expressed transcripts . Transcripts with significantly different levels ( p< = 0 . 001 and over 4-fold difference ) were identified , and p-values were adjusted for multiple testing by using the Benjamini & Hochberg [80] correction . The predicted proteomes of S . parasitica and representative plant pathogenic oomycetes were annotated by mapping against a reference set of metabolic pathways from KEGG ( Kyoto Encyclopedia of Genes and genomes ) [81] . The method used , KAAS ( KEGG Automated Annotation Server ) , utilizes bidirectional best hits to assign pathways . Subsequently , the metabolic genes and pathways were manually annotated and compared to other oomycete pathogens . Known genes involved in nitrogen and sulfur metabolism were used to search the S . parasitica genome by TBLASTN search; genes were considered to be candidates if a positive hit was found ( E value<1e-5 ) . For chitin metabolism analysis , genes were annotated based on Pfam homology ( E value<1e-5 ) with the exception of chitin synthase genes , which were identified after Blastp analysis against a set of oomycete and fungal CHS . A total of twelve GH18 genes were identified , of which six are arranged in small clusters of two paralogous genes . Two different approaches were used to screen S . parasitica genes for candidate HGT origins . In the first approach , the genome sequence was screened with the program Alien_hunter [55] . The program utilizes an interpolated variable order motif method to determine horizontally transferred events , purely based on compositional difference between a region and the whole genome framework . Because the methodology is independent of any existing datasets , we used it to examine the S . parasitica genome . Genomic regions were identified as alien when the Alien_hunter score was above 50 . Out of 1442 S . parasitica supercontigs , 206 supercontigs had distinct regions marked as alien after running Alien_hunter . Subsequently , the 1616 gene models that lay within the candidate alien regions were extracted and compared with other oomycete genomes ( P . sojae , P . ramorum , P . infestans , H . arabidopsidis and Py . ultimum ) . In the second approach , the entire proteome of each oomycete was scanned for homology to Pfam-A protein families using the hmmscan algorithm from Hmmer 3 . 0 applied to the Hidden Markov Model dataset ( Pfam-A . hmm v . 24 ) . A cut-off e-value threshold of 1e-3 was applied . From the Pfam domain analysis we obtained 307 sequences that had distinct domains not found in any of the Phytophthora species , and 31 of the candidates derived from the Pfam analysis overlapped with the results from Alien_hunter . We then blast-searched all 1616 genes from Alien_hunter and the 307 genes from the domain analysis against the NCBI non-redundant database ( nr ) to obtain the primary functions . Phylogenetic analysis using neighbor joining was then performed on the final set of genes . S . parasitica CBS 223 . 65 mycelium plugs were grown in pea broth for 2 or 7 days . The culture supernatants were harvested , centrifuged at 5000× g for 10 min ( 4°C ) and the soluble fraction used as a source of secreted proteases . The ammonium sulfate precipitated fraction from rainbow trout serum [82] was used as source of fish IgM ( 10 mg/mL of total protein concentration ) . For protease activity experiments , 50 µL of the culture supernatants were incubated overnight at 10°C with 5 µL of the IgM enriched fraction . Pea broth was used as a negative control . Heat-inactivated supernatant was obtained by incubating the culture supernatants for 15 min at 95°C . The protease inhibitors EDTA ( 10 mM ) , PMSF ( 1 mM ) and E-64 ( 10 µM ) were used to identify classes of proteases involved in IgM degradation . Two samples volumes ( 2 . 5 and 5 µL ) were spotted onto nitrocellulose membranes , allowed to dry for 45 min , blocked with milk-PBS ( 5% dry-milk ) and remaining IgM was detected using a monoclonal anti-trout/salmon HRP conjugated antibody ( Aquatic Diagnostics ) . SPRG_14567 was cloned from S . parasitica CBS223 . 65 cDNA into the pET21B vector . E . coli Rossetta-gami B competent cells were transformed with the expression vector by heat-shock . Transformed cells were grown in modified LB media ( 100 mM NaHPO4 , pH 7 . 4; 2 mM MgSO4 , glucose 0 . 05% w/v; and NaCl 0 . 5% w/v ) . When E . coli cells reached an OD600 of 0 . 8 , cultures were induced with IPTG ( 10 mM ) and incubated at 200 rpm at 37°C for 12 hr . The soluble fraction was purified from French press supernatant by tandem ion exchange ( on SO3− ) and nickel affinity columns . Chromatography fractions were submitted tor 1D SDS-gel electrophoresis ( NuPAGE Electrophoresis System , Invitrogen ) and then transferred to nitrocellulose membranes ( XCell II Blot module , Invitrogen ) . Membranes were blocked with milk-PBS-T ( 0 . 02% v/v Tween , 10% dry milk ) for 20 min . Anti- ( His ) 5 HRP conjugate ( QIAGEN ) was added to a final dilution of 1∶20 , 000 and incubated at room temp for 1 hr for detection of the recombinant protein . Chromatography fractions were tested for IgM degrading activity as previously described [50] . Briefly , 50 µL ( 5 µg/mL of total protein concentration ) of the fractions were incubated overnight at 10°C with 5 µL of the IgM enriched fraction . Untransformed E . coli soluble proteins were used as a control . Remaining IgM was then detected in a dot-blot as described above . Gene models were analyzed using TMHMM and ClustalX . For phylogeny reconstruction , sequences were aligned using ClustalX and aligned sequences were subjected to phylogenetic analysis ( NJ ) using PAUP . The predicted proteome of S . parasitica was compared to that of six pathogenic oomycetes together with sixty-four other eukaryotic species covering all major groups of the eukaryotic tree of life , as previously used by Seidl et al . [48] . We used hmmer-3 [http://hmmer . org] and a local Pfam-A database to predict 1 , 798 , 601 domains in 862 , 909 proteins of which 19 , 896 domains in 10 , 887 proteins are found in S . parasitica . The architectures of multi-domain proteins were analyzed from the N- to the C-terminus , which identified 18 , 512 domain combinations consisting of two contiguous domains . Of these there were 1120 oomycete-specific domain combinations encoded by 2286 proteins . The great majority of combinations were specific to a single species , and only 58 combinations were found in more than one oomycete species . S . parasitica contained 338 domain combinations that are specific for oomycetes including 169 domain combinations encoded by 215 genes that are specific for this species ( Table S9 ) . Saprolegnia parasitica CBS 223 . 65: Sanger sequence data were submitted to the NCBI Trace Archive ( http://www . ncbi . nlm . nih . gov/Traces/trace . cgi ) and can be retrieved using query: CENTER_NAME = “BI” and CENTER_PROJECT = “G1848” . 454 and Illumina sequence data were submitted to the NCBI Short Read Archive ( SRA ) ( http://www . ncbi . nlm . nih . gov/sra ) and can be retrieved using the following accession numbers: 454 fragment reads ( SRX007896 , SRX007901 , SRX007898 , SRX007895 , SRX005344 , SRX007971 , SRX007958 , SRX005346 ) ; 454 3 kb jumping reads ( SRX007902 , SRX007899 , SRX007937 , SRX007574 , SRX007903 ) ; Illumina fragment reads ( SRX022535 ) ; Illumina RNA-Seq reads ( BioProject 164643: mycelium SRX155934 , sporulation mycelium SRX155933 , cysts SRX155932 , germinating cysts SRX155938 , infected fish cell-line t = 0 SRX155937 , infected fish cell-line t = 8 SRX155936 and infected fish cell-line t = 24 SRX155935 . BioProject 167986: infected fish tissue SRX155944 , uninfected fish tissue SRX155942 , infected salmon eggs SRX155943 , SRX155940 and uninfected salmon eggs SRX155941 ) . Draft genome assembly sequence was submitted to GenBank ( BioProject ID 36583 , accession ADCG00000000 ) . Saprolegnia parasitica VI-02736 : Illumina sequence data were submitted to the SRA ( BioProject 164645 , accession SRX155939 ) . SNP calls for strains CBS 223 . 65 and VI-02736 can be downloaded from the Broad Institute Saprolegnia parasitica genome database website ( http://www . broadinstitute . org/annotation/genome/Saprolegnia_parasitica/MultiDownloads . html ) . | Fish are an increasingly important source of animal protein globally , with aquaculture production rising dramatically over the past decade . Saprolegnia is a fungal-like oomycete and one of the most destructive fish pathogens , causing millions of dollars in losses to the aquaculture industry annually . Saprolegnia has also been linked to a worldwide decline in wild fish and amphibian populations . Here we describe the genome sequence of the first animal pathogenic oomycete and compare the genome content with the available plant pathogenic oomycetes . We found that Saprolegnia lacks the large effector families that are hallmarks of plant pathogenic oomycetes , showing evolutionary adaptation to the host . Moreover , Saprolegnia harbors pathogenesis-related genes that were derived by lateral gene transfer from the host and other animal pathogens . The retrotransposon LINE family also appears to be acquired from animal lineages . By transcriptome analysis we show a high rate of allelic variation , which reveals rapidly evolving genes and potentially adaptive evolutionary mechanisms coupled to selective pressures exerted by the animal host . The genome and transcriptome data , as well as subsequent biochemical analyses , provided us with insight in the disease process of Saprolegnia at a molecular and cellular level , providing us with targets for sustainable control of Saprolegnia . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"genome",
"sequencing",
"genome",
"expression",
"analysis",
"genome",
"databases",
"emerging",
"infectious",
"diseases",
"microbial",
"pathogens",
"comparative",
"genomics",
"biology",
"genomics",
"microbiology",
"host-pathogen",
"interaction",
"parasitology",
"pathogenesis"
]
| 2013 | Distinctive Expansion of Potential Virulence Genes in the Genome of the Oomycete Fish Pathogen Saprolegnia parasitica |
The visual system of a particular species is highly adapted to convey detailed ecological and behavioral information essential for survival . The consequences of structural mutations of opsins upon spectral sensitivity and environmental adaptation have been studied in great detail , but lacking is knowledge of the potential influence of alterations in gene regulatory networks upon the diversity of cone subtypes and the variation in the ratio of rods and cones observed in numerous diurnal and nocturnal species . Exploiting photoreceptor patterning in cone-dominated zebrafish , we uncovered two independent mechanisms by which the sine oculis homeobox homolog 7 ( six7 ) regulates photoreceptor development . In a genetic screen , we isolated the lots-of-rods-junior ( ljrp23ahub ) mutation that resulted in an increased number and uniform distribution of rods in otherwise normal appearing larvae . Sequence analysis , genome editing using TALENs and knockdown strategies confirm ljrp23ahub as a hypomorphic allele of six7 , a teleost orthologue of six3 , with known roles in forebrain patterning and expression of opsins . Based on the lack of predicted protein-coding changes and a deletion of a conserved element upstream of the transcription start site , a cis-regulatory mutation is proposed as the basis of the reduced expression of six7 in ljrp23ahub . Comparison of the phenotypes of the hypomorphic and knock-out alleles provides evidence of two independent roles in photoreceptor development . EdU and PH3 labeling show that the increase in rod number is associated with extended mitosis of photoreceptor progenitors , and TUNEL suggests that the lack of green-sensitive cones is the result of cell death of the cone precursor . These data add six7 to the small but growing list of essential genes for specification and patterning of photoreceptors in non-mammalian vertebrates , and highlight alterations in transcriptional regulation as a potential source of photoreceptor variation across species .
Sensory systems provide a critical link for an animal to its ever changing and complex environment . Retinal photoreceptors are the highly specialized neurons that transduce light into the chemical and electrical signals of the nervous system . Representatives from nearly all classes of extant vertebrates possess a duplex retina with two distinct types of photoreceptors: rods , which are highly sensitive to light , mediate scotopic or dim light vision; and cones , which function under daylight or bright light conditions , are responsible for color vision . The spectral sensitivity of cones is dependent upon the expression of one of four different visual pigments or opsins with peak sensitivity to ultraviolet or violet ( SWS1 ) , blue ( SWS2 ) , green ( RH2 ) , or red ( LWS ) wavelengths of light . Rods express rhodopsin ( RH1 ) which is most sensitive to green light [1] , [2] . Detailed phylogenetic and functional analyses of structural mutations affecting spectral sensitivity provide much insight about the evolution of the visual system and adaptation to different lighting environments [1] , [2] , [3] , [4] , [5] , [6] . Nevertheless , the molecular mechanisms leading to the major evolutionary changes in photoreceptor composition among vertebrate species remain unclear . Electrophysiological data provide compelling evidence that the first jawless vertebrates already possessed a duplex retina containing four cone subtypes as well as cells adapted to dim light conditions [7] , [8] , [9] , [10] , [11] . A cone rich architecture is still present in many extant species of teleosts , amphibians , reptiles , and birds [12] , [13] , [14] , [15]; in stark contrast , retinas of nocturnal animals are typically rod-dominated and possess only one or two cone subtypes [16] . For example , the high number of rods , relatively few cones and eye shape reflect the prevailing view that Mesozoic ancestors of extant mammals were adapted to a nocturnal environment [17] , [18] , [19] , [20] , [21] . Today , the remaining cones in marsupials and eutherian mammals express LWS and SWS1 opsins , and in monotremes express functional LWS and SWS2 opsins [22] , [23] , [24] . The absence of RH2 and SWS2 , but preservation of RH1 , LWS and SWS1 opsins in the basal lineage of modern snakes is an example of convergent evolution to maintain short- and long-wavelength sensitivity in nocturnal or burrowing species; yet continued adaptation is observed in more recent gene losses , adaptation of additional sensory modalities or regain of trichromacy [24] , [25] , [26] , [27] [28] . Retinal development proceeds in a highly conserved order with cones generated in the first wave of neurogenesis and rod generated later in development [29] , [30] . The temporal difference is thought to represent a change in the competency of retinal progenitors over time [31] , [32] . Phylogenetic analysis and experimental data support the notion that shifts in the timing of mitosis ( heterochrony ) are associated with alterations in the proportion of neuronal subtypes produced during retinogenesis [33] . For example , the greatly increased numbers of rod and bipolar cells in the nocturnal owl monkey ( Aotus azarae ) are associated with shifts in mitosis to later stages of development compared to a closely related diurnal capuchin monkey ( Cebus apella ) [34] . By comparison , analysis of mouse mutations and human diseases show that alterations in the photoreceptor gene-regulatory network leads to dramatic changes in the types and numbers of photoreceptors generated during development . The specification of photoreceptor precursors and subsequent expression of rod and cone specific genes requires the expression of the homeobox transcription factor CRX [35] , [36] , [37] , [38] . Subsequently , TRβ2 regulates the specification of the LWS cone [39] , [40] , and the transcription factor NRL and its downstream target NR2E3 act synergistically with CRX to specify the rod fate and drive rod gene expression [41] , [42] while repressing expression of cone genes [43] , [44] , [45] , [46] , [47] . The roles of these transcription factors are highly conserved yet studies have failed to find evidence that alteration of this gene regulatory network drives adaptation of the visual system in different classes of vertebrates . In fact , little is known about the factors that generate the greater diversity of cone subtypes in non-mammalian vertebrates or the mechanisms underlying the wide range of rod to cone ratios in diurnal and nocturnal species [48] , [49] . The spatial patterning of zebrafish photoreceptors combined with classical genetics and emerging gene-targeting technologies , offer unprecedented opportunities to investigate photoreceptor biology in a diurnal species [50] , [51] , [52] , [53] , [54] . Larval zebrafish retina contains four cone subtypes , which outnumber the far fewer , sparsely distributed rods . Previously , in a genetic screen , we identified a novel role for the transcription factor tbx2b in photoreceptor development . Mutations of tbx2b , a co-orthologue of TBX2 , results in a cell fate switch of the SWS1 cone precursors into rods [53] . These data supported the conservation of the ontological relationship between the SWS1-expressing cones and rods in mammals and zebrafish , but challenged the notion of a default photoreceptor fate among species . Here , we report the characterization of a second mutation called lots-of-rods junior ( ljrp23ahub ) that results in an increased number and uniform distribution of rods in larvae but with little affect upon cones . Data provide strong evidence that ljrp23ahub is a mutation in a cis-regulatory element of six7 , a teleost member of the sine oculis family of homeobox transcription factors [55] , [56] . Our previous data showed that knockdown of six7 led to an increased number of rods , and Ogawa et al . , ( 2015 ) reported increased rod gene expression and altered cone opsin expression in a six7 knockout line [57] , [58] . In this study , our genetic analysis using the hypomorphic allele and novel loss-of-function alleles reveal that six7 independently regulates mitosis of photoreceptor progenitors in a dosage dependent manner and survival of green-sensitive cone precursors . In addition to expanding our knowledge of genes essential for maintenance of photoreceptor diversity in a diurnal species , the developmental variation in rod and cone numbers provide insight that may be informative for pursuing evolutionary steps from a cone-rich towards a rod-dominated retina .
In our previously published genetic screen [53] , [59] to isolate loci that regulate rod development and spatial patterning in zebrafish , we identified a mutation , lots-of-rods-junior ( ljrp23ahub ) that results in an increased number and uniform distribution of rods across the retina in otherwise normal appearing larvae . In wild-type ( WT ) larvae , rods are asymmetrically patterned along the dorsal-ventral axis of the eye , with the highest density in the ventral retina , few in a belt spanning the central retina , and sporadic , yet non-random labeling across the dorsal retina [60] . In homozygous ljrp23ahub larvae , immunolabeling for rods results in a uniform distribution typical of the cone pattern ( Fig 1A and 1B ) . For all mutant samples analyzed , conformity ratios ( CR ) were significantly different from random based on Cook’s criteria ( p<0 . 05 ) , and Nearest Neighbor Dispersion Analysis ( NNDA ) indicates the rods are arrange in a uniform pattern ( p<0 . 05 ) [61] . Previously , we reported that mutations in tbx2b ( lorp25bbtl ) result in an increased number and uniform distribution of rods through a cell fate switch of SWS1-cone precursors into rods [53] ( Fig 1A ) . Mating of homozygous ljrp23ahub adults to homozygous tbx2bp25bbtl adults revealed that ljrp23ahub complements tbx2bp25bbtl; all larvae displayed a rod pattern and SWS1 cone number typical of WT larvae . Co-immunolabeled larvae from mating of the double heterozygous ljrp23ahub/tbx2bp25bbtl adults revealed that double homozygous larvae for ljrp23ahub/tbx2bp25bbtl displayed an additive number of rods and few UV-sensitive cones suggesting that these two genes regulate photoreceptor cell patterning through distinct mechanisms ( Fig 1A , 1B and 1C ) . Furthermore , ljrp23ahub is semi-dominant; heterozygous ljrp23ahub larvae display a rod number between those of WT and homozygous ljrp23ahub mutants ( Fig 1D ) . Whole mount ljrp23ahub larvae and WT controls were immunolabeled for Zpr-1 monoclonal antibody , which recognizes Arr3a , a selective marker for the red- and green-sensitive cones [62] . In confocal images of flat mount WT retinas , Zpr-1-labeling appears as rows of cells with more intensely fluorescent red-sensitive cones alternating with less intense labeling of the green-sensitive cones ( Fig 1E ) . No significant change in the number of Zpr-1 labeled cones in ljrp23ahub mutant was observed when compared to WT ( Fig 1F ) . Immunolabeling of serial sections of WT and ljrp23ahub mutant larvae with polyclonal antisera to RH2 , SWS2 , and SWS1 revealed no differences in their expression levels ( S1A Fig ) . No differences were observed for cell specific markers of amacrine , bipolar , ganglion or Muller cells in ljrp23ahub compared to WT , suggesting a very specific retinal phenotype ( S1B Fig ) . Genetic linkage analysis positioned the ljrp23ahub locus to a 0 . 3 Mb interval on chromosome 7 encompassing seven genes ( S2A Fig ) : Ras and Rab interactor 1 ( Rin1 ) , actin related protein ( Arp2 ) , six7 , beta-1 , 4 glucoronyltransferase 1 , prefoldin subunit 2 , nitrilase , and chloride channel 3 . Genes lacking eye specific expression ( Zfin ) or with housekeeping functions were excluded from further consideration leaving six7 as the most plausible candidate . In order to more comprehensively examine for molecular lesions associated with the ljrp23ahub we performed whole-genome sequencing of DNA pooled from 118 homozygous ljrp23ahub larvae . Genome-wide SNP frequencies of reads were calculated against the Tuebingen ( zv9 ) reference genome to identify regions depleted of SNPs that were associated with the mutant genetic background . Large , contiguous depletions in SNPs were present on chromosomes 3 , 7 , 9 , and 20 . A 5-megabase depletion of SNPs present on chr7 was centered directly over the mapping interval based on genetic linkage analysis ( Fig 2A ) . Further examination of this interval revealed two regions , 14 kb and 40 kb upstream of six7 , devoid of uniquely-aligning reads that were not associated with assembly gaps . The region 40 kb upstream of six7 , but not the region 14 kb upstream , was also associated with sequence conservation among six7 genes from 4 other fish species , but not with Six3 from frog , mouse , or human . This region also displayed ChIP-seq signal for H3K4me1 in 24 hours-post-fertilization ( hpf ) embryos [63] . This mark is associated with distal enhancers in mammals [64] . To verify that the deletion was associated with the ljrp23ahub mutation , DNA samples from fin clips of homozygous ljrp23ahub adults and wildtype AB and TL strains of zebrafish were subjected to PCR using primer pairs targeting a DNA sequence unique to the distal deletion or the first exon of six7 as a control . PCR confirmed the co-segregation of the genomic deletion upstream of six7 with ljrp23ahub ( Fig 2A ) ; the upstream region was only amplified in the TL and AB DNA . six7 is a teleost specific member of the sine oculis homeobox family of transcription factors , which have important roles in eye and forebrain patterning [55] , [65] , [66] , [67] , [68] . In zebrafish , six7 expression co-localized with six3a/b transcripts in the anterior region of the forebrain and optic vesicles [55] , [69] with no detectable expression by RT-PCR at 24 hpf [55] . However , we and others observed by in situ hybridization that by 48 hpf , six7 expression is detectable in the retina , specifically in neuroblasts and the differentiating outer nuclear layer ( ONL ) coincident with photoreceptor cell genesis [58] ( Fig 2B ) . More precisely , between 48 hpf and 52 hpf more general labeling of the retinal neuroblasts gives way to robust six7 expression that follows the temporal and spatial wave of photoreceptor genesis spreading from the ventral to the nasal and temporal retina [70] , [71] . qRT-PCR for six7 transcripts at developmental stages from 10 hpf to 52 hpf mirrored the in situ hybridization; in ljrp23ahub mutant embryos , greater expression was observed at 10 hpf compared with WT , and expression was absent in ljrp23ahub mutant embryos at 18 hpf and from both groups at 24 hpf . Lower levels in ljrp23ahub mutant embryos at 52 hpf were observed ( Fig 2C ) . Thus , the spatial and temporal pattern of six7 expression and the changes observed in ljrp23ahub embryos are consistent with previously described roles in photoreceptor development [57] , [58] . To test the candidacy of six7 as the mutated gene in ljrp23ahub , two antisense morpholinos targeting either the 5’UTR region ( MO1 ) [69] or the donor splice site in the first intron ( MO3 ) of six7 were injected into one-cell-stage WT embryos ( S2B Fig ) . Injection of either morpholino phenocopied ljrp23ahub mutants ( Fig 2D ) . Morphants showed increased rod immunolabeling in a dosage dependent manner ( Fig 2E ) , and did not demonstrate any obvious morphological defects . To confirm the efficiency of the splice blocking morpholino , RNA was isolated from un-injected and MO3-injected embryos , and the region spanning exon 1 and exon 2 amplified by PCR from resulting cDNA . Sequence data revealed that the MO3-injections resulted in alternative splicing upstream of the initiation codon deleting the majority of the exon 1 of six7 mRNA including the SIX domain ( S2C Fig ) . Previous studies reported highly conserved roles for six3/six6 family members in patterning of the forebrain and eye field , and in zebrafish six3b and six7 appear to be functionally redundant [58] , [73] , [74] . To study the hypothesis that six3a/b could be a target of six7 , the levels of expression of six3a/six3b were determined by qRT-PCR . No difference in expression levels for six3a/six3b transcripts were detected between WT and ljrp23ahub mutant ( S2D Fig ) , suggesting that six7 is not regulating the expression of six3a/six3b . Consistent with the expression of six7 in ljrp23ahub during forebrain patterning , analysis of 128 embryos from inbreeding of double heterozygous adults for ljrp23ahub and six3bvu87 did not result in any embryos displaying a greatly reduced or absent eye phenotype as previously observed for six7-morpholino knockdown on the six3b mutant background or in the recently reported double mutant harboring deletions of six7 and six3b [69] , [58] . Based upon the genetic analysis , whole genome sequencing and gene expression changes , we propose that ljrp23ahub is a hypomorphic allele of six7 that affects a regulatory element controlling expression during photoreceptor genesis . To further characterize the increased labeling for rods , cell counts from methylene blue stained plastic sections of 4-days-post-fertilization ( dpf ) six7-MO1 embryos revealed a modest yet significant increased number of cells in the ONL compared with WT retinas , consistent with the observed increased rod number ( S2E Fig ) . However , no changes were detected in the number of nuclei in the inner nuclear layer ( INL ) or the ganglion cell layer ( GCL ) , arguing against a general increase in neurogenesis across the retina ( χ2 , p> 0 . 05 ) . Additionally , at 6 dpf , electron microscopy of the ONL in the central retina , where the highest change in rod numbers were detected , showed that rods in six7-MO1 were characterized by an outer segment composed of stacks of discs enclosed within the plasma membrane , a vitread located nucleus and a single invaginating synapse at the terminal ( S2F Fig ) . These results suggest that knockdown of six7 led to an increased number of retinal cells specifically in the ONL with gene expression and morphological characteristics consistent with rods . The increased cell number in the ONL and lack of changes in cone numbers open the possibility that six7 regulates mitosis during photoreceptor development . Proliferation was assayed by EdU incorporation or phospho-histone 3 ( PH3 ) immunolabeling . In zebrafish , neurogenesis occurs in three distinct waves; postmitotic cells appear first in the GCL , followed by the INL , and finally the ONL . At 48 hpf , a time coincident with photoreceptor cell genesis [75] , co-labeling for EdU incorporation and in situ hybridization for six7 showed complex patterns of labeling . In histological sections , EdU labeling was most abundant in wedged-shaped clusters of highly proliferative cells near to the dorsal and ventral ciliary marginal zone ( CMZ ) , and to a lesser degree in the developing ONL . In contrast , six7 expression near the CMZ was opposite of the EdU labeling; highest proximal to the CMZ where the neuroblasts had taken on a more salt and pepper EdU-labeling pattern ( arrows ) , and nearly absent from the highly proliferative CMZ . In the central retina , more robust labeling for six7 coincided with reduced EdU incorporation ( Fig 3A ) . These data show that six7 is expressed in photoreceptor precursors at or near the time of terminal mitosis . In a second set of experiments , un-injected , six7 MO1- , and control morpholino-injected embryos were incubated with EdU at 48 or 52 hpf and immediately processed for labeling . At 48 hpf , all treatment groups showed EdU labeling in the ONL . However , at 52 hpf , six7 MO1-injected embryos showed nuclear EdU labeling in the ONL but none was observed in the un-injected and 5 base mismatch control morpholino-injected embryos ( Fig 3B ) . To determine if the EdU-labeled cells in the six7-morphant retinas differentiate specifically as rods , embryos were labeled with EdU at 48 hpf and maintained until 72 hpf , then fixed and processed for immunolabeling with a rod specific marker . EdU positive cells in the central retina co-labeled preferentially with a rod marker in six7-morphant retinas . However , some EdU positive cells differentiated as cones , consistent with the coincident timing of rod and cone differentiation in un-injected embryos ( Fig 3C ) . We next immunolabeled retinas with anti PH3 to verify that the increased number of cells labeled with EdU was reflected by changes in the level of mitosis ( Fig 3D ) . At 48 hpf retinas from six7-morphant embryos showed significant increases in the PH3 immunolabeling of the ONL compared to un-injected embryos ( Fig 3E; Student t-test; p<0 . 001 ) . No significant changes were observed in labeling of the INL . To test if the six7-depleted cells are biased to differentiate as rods , genetic chimeras were generated by transplanting cells from six7 MO1-injected or WT donor embryos into WT hosts . In histological sections , six7-MO1 donors cells immunolabeled for a rod specific marker at a rate three times higher than WT transplanted cells ( p<0 . 05; S3A Fig ) . The spatial and temporal appearance of the additional mitoses and bias to form rods led us to test the identity of the proliferative cells . WT and six7-knockdown retinas were labeled by in situ hybridizations with molecular markers for retinal progenitors , rx1 and pax6a [76] , [77] or two transcription factors expressed by developing photoreceptors , crx and neurod [78] , [79] , [80] . By 48 hpf in the WT and morphant larvae , the expression of the retinal progenitor marker rx1 was restricted to the CMZ and the pax6a gene was expressed by neuroblasts of the CMZ and in neurons located in the GCL and the proximal portion of the INL [81] . None was observed in the ONL ( S3B Fig ) . Probes for crx and neurod strongly label the forming ONL and to lesser extent cells of the INL , with no differences observed between WT and mutant larvae ( S3B Fig ) . The data suggest that the mitotic cells are photoreceptor progenitors . In lower vertebrates , such as zebrafish , cell death in the retina can trigger proliferation of Muller glia cells and photoreceptor regeneration [82] , [83] . To exclude apoptosis-induced proliferation as the mechanism leading to the increase in rod number , sections from control , ljrp23ahub , and six7-morphant retinas were subjected to transferase-mediated dUTP nick end labeling ( TUNEL ) assay . Few TUNEL positive cells were observed in WT and control-morpholino retinas . A modest increased in TUNEL positive cells was observed in the inner retina and GCL in the six7-MO1 retina or ljrp23ahub compared with control embryos ( Fig 3F–3I ) , however none was observed in the ONL ruling out cell death-induced regeneration as the mechanism triggering the increase in mitosis and rod number . Lastly , previous studies in chick embryos have shown that ablation of the dorsal retina results in expansion of ventral domain and increased rod number in the central retina [84] . Given the high number of rods in the ventral patch of the zebrafish retina , we tested for expansion of ventral markers and loss of dorsal markers in ljrp23ahub mutants . However , in situ hybridization for the dorsal marker tbx2b , the midline marker cyp26c1 , and the ventral marker vax2 showed no difference in labeling of WT and mutant embryos ( S3C Fig ) , decreasing the likelihood that the increased rod number resulted from alteration of dorsal-ventral patterning of the optic cup . Based upon the sequencing data , morpholino phenotypes and gene expression , we designed TALENs ( transcription activator-like effector nucleases ) to target six7 . The homeodomain and the SIX domain are two evolutionarily conserved domains in the SIX proteins involved in DNA-protein or protein-protein interactions respectively [85] , and mutations in the SIX domain in SIX3 are associated with congenital brain and eye defects [86] . Therefore a TALENs pair was designed to target 18-bp and 20-bp flanking a 14-bp spacer sequence of the first exon of six7 , which corresponds to the SIX domain ( Fig 4A ) . mRNAs encoding for the TALENs pair were co-injected into one-cell stage zebrafish embryos . Surviving embryos were grown to adulthood and mated to WT adults . Disruption of the six7 locus in the F1 larvae was detected by the loss of the HaeIII restriction site in the spacer region ( Fig 4B ) . Fifty-two percent of the founders transmitted TALEN-induced mutations to the F1 ( Fig 4C ) . F1 progeny were grown to adults and heterozygous carriers identified by fin clip analysis . F2 carriers of the following alleles were used in subsequent studies: c . 217_229del CAGGTGGCCCGAG , p . ( Q11Cfs*39 ) , from now on ( six7fl4 ) ; p . ( E10Ifs*50 ) ; p . ( F7Lfs*44 ) , ( all predicted to result in frameshift mutations and premature termination of the Six7 protein ) . Approximately , one quarter of the F2 progeny from inter-crosses between carriers demonstrated an increased number ( t-test , p<0 . 001 ) and uniform distribution of rods ( Cook’s CR , p<0 . 05; NNDA , p<0 . 05 ) as observed in ljrp23ahub mutants ( Fig 4D ) . Genotyping of the F2 larvae revealed that one quarter of the embryos with the lots-of-rods phenotype were heterozygous for the six7fl4 mutation , consistent with the semi-dominance previously observed in ljrp23ahub mutants . In mating between carriers or homozygous mutant adults , six7fl4 failed to complement ljrp23ahub mutants ( Figs 4D and S4A; One-way ANOVA , Tukey’s follow-up test , p<0 . 0001 ) ; no significant difference in rod number was observed between six7fl4/six7fl4 , six7fl4//ljrp23ahub , and ljrp23ahub/ljrp23ahub larvae ( One-way ANOVA , p>0 . 5 ) providing genetic evidence that ljr23ahub is indeed an allele of six7 , and ljrp23ahub shall be referred to as six7p23ahub . Lastly , as previously observed in six7-knockdown embryos , immunolabeling for PH3 showed significant difference in sections from six7fl4 homozygous embryos and WT siblings ( S4B Fig ) . But unlike the six7p23ahub allele , six7fl4 homozygous animals showed significantly reduced viability ( χ2 , p<0 . 05 ) , but the few adults recovered were fertile . Ogawa et al . ( 2015 ) recently reported altered cone opsin expression in six7 knock-out animals with RH2 expression nearly absent and SWS2 significantly reduced [58] . However , we observed few alterations with green-sensitive cone opsin expression in six7p23ahub homozygous and six7 morphant larvae and no changes in SWS2 immunolabeling . Therefore , we tested for differences in cone photoreceptor cell phenotypes between the six7p23ahub and six7fl4 alleles . Immunolabeling and confocal analyses revealed a significant decline in the number of Zpr1/Arr3a positive photoreceptors in homozygous six7fl4 larvae consistent with the lack of RH2 expression reported previously ( Fig 4E ) . Immunolabeled cells were evenly distributed across the retina , but clearly isolated from their neighbors as opposed to the mosaic of alternating red- and green-sensitive cones observed in the WT and six7p23ahub mutant larvae ( Fig 4E ) . In situ hybridization using an RH2 probe or immunolabeling for green-sensitive opsin demonstrated that six7fl4/fl4 larvae were mostly devoid of green-sensitive opsin expression with only a few pairs of labeled cells in either eye ( Fig 4F and inset ) ; immunolabeling of histological sections confirmed the phenotype ( S4 Fig ) . By comparison , homozygous six7p23ahub mutants show variable penetrance of the loss of green-sensitive opsin expression phenotype; 83% of six7p23ahub homozygous larvae showed a WT labeling pattern and only 17% showed labeling similar to the six7fl4 mutants ( Fig 4F ) . Co-labeling with a rhodopsin antibody showed that all six7p23ahub mutants retained the increased rod number regardless of the presence or absence of the green-sensitive opsin labeling ( Fig 4G ) , and green-sensitive opsin positive photoreceptors do not co-label with several different rod markers , suggesting that the extra rods in the six7p23ahub mutant retinas are not a rod-cone hybrid . The similarities and differences in phenotypes of the hypomorphic and knock-out alleles from our lab and previously reported [58] suggest that six7 regulates two distinct processes in photoreceptor cell genesis: terminal mitosis and differentiation or survival of green-sensitive cones precursors . Consistent with our hypothesis , confocal images of DAPI-labeled retinas from homozygous six7fl4 larvae showed gaps in the photoreceptor mosaic and small intensely labeled structures consistent with nuclei of dead or dying cells ( Fig 4H ) . TUNEL was performed on six7fl4 retinas at 56 hpf , a time when expression of the all opsin subtypes should be detected [70] , [87] , and at 96 hpf , when all of the cones are mature . Few apoptotic nuclei were detected in WT or mutant retinas at 56 hpf . However , as development progressed to 96 hpf , considerable labeling of TUNEL positive cells was observed in the ONL of six7fl4 retinas ( Fig 4H ) . Labeling was also observed in the INL and in fibers extending across both plexiform layers to the inner limiting membrane . The morphology and labeling pattern are consistent with that of Muller glial cells which have been shown to become TUNEL positive from phagocytosis of cellular debris following photoreceptor degeneration [88] , [89] , although we cannot rule out the possibility that a small population of cells in the inner retina is also dying . Together , the data show that six7 is essential for development of green-sensitive cone precursors in the zebrafish retina and in its absence the precursors die . Genetic chimeras were generated to further test the cell autonomy of six7 in photoreceptor biology . At blastula stage , cells were transplanted from rhodamine dextran-injected six7fl4 mutant donors into equivalent stage WT hosts . As control , WT cells were transplanted into WT embryos at the same developmental stage . The fate of WT vs six7fl4 donor cells showed statistically significant differences based upon co-immunolabeling host embryos for green-sensitive opsin and a rod marker ( χ2 , p<0 . 0001 ) . Of 96 six7fl4 rhodamine positive donor cells in the ONL of 3 hosts , 49 cells ( 51% ) co-labeled with a rod specific marker consistent with the data from the MO-injected genetic mosaics , but only 2 cells ( 2% ) immunolabeled for green-sensitive opsin . However , the neighboring host cells frequently labeled for green-sensitive opsin ( Fig 5A ) . In stark contrast , 16% of 106 rhodamine-labeled WT-donor cells co-labeled for the green-sensitive opsin , but only 5 . 6% for the rod-specific marker . These data are consistent with a cell-autonomous role of six7 in regulating rod number and green-sensitive cone precursor differentiation or survival .
Taking advantage of photoreceptor patterning in the cone-rich , larval zebrafish retina , we characterize two independent roles for the transcription factor six7 in photoreceptor development: six7 regulates proliferation affecting the number and distribution of rods; six7 is essential for survival of the green-sensitive cone precursor . We show that the increased number and uniform distribution of rods are associated with increased mitosis , and independent of and do not account for the loss of green -sensitive cones . Furthermore , the six7p23ahub and tbx2bp25bbtl double mutants show an additive number of rods indicative of activity in different pathways . Our research has identified genes essential for maintenance of a cone-dominated retina , and based upon the mutant phenotypes provide a context for understanding how alteration of cis-regulatory elements could drive developmental changes transitioning a cone-dominated retina to a rod-dominated retina . The changes in rod number suggest that six7 has dosage-dependent affects upon mitosis . The heterozygous and homozygous mutant larvae displayed varying degrees of increased numbers of rods , and increased mitosis was observed in the ONL of mutant and six7 knockdown embryos . In WT embryos six7 is expressed in the ONL at 48 hpf , when few progenitors co-labeled for markers of proliferation , consistent with roles in the photoreceptor progenitors at or near the time of terminal mitosis . In zebrafish , gene expression studies , time lapse imaging and cell transplantation show that photoreceptor specification occurs prior to or coincident with cell cycle exit [58] , [82] , [90] . In six7 mutants , the expression of crx , neurod , pax6 and rx1 were unchanged compared to WT animals suggesting that six7 functions in mitotic photoreceptor progenitors downstream of crx and neurod . The proliferation phenotype in six7 morphants and mutants was distinct from that observed in the lep/ptc2 mutant larva which is characterized by a proportional increase in the number of neurons in each retinal layer [91] . Rather , the increase in rod number in six7 mutants is consistent with the hypothesis that selective alterations in the timing of cell-cycle exit can vary the proportion of the retinal cell types produced [34] , [92] , [93] . The effects on mitosis are surprisingly different from those observed for Six3 , the closest homologue for which data are available . In murine cortical progenitors , mis-expression of Six3 caused clonal expansion , but the fate of cells could not be identified as the progenitors failed to differentiate [94] . Similarly , in the rat retina , retroviral-mediated ectopic expression of Six3 led to an increased number of infected cells in the ONL relative to controls , though again the cells failed to mature properly . However , over-expression of a Six3 variant that alters the protein binding domain resulted in nearly exclusive generation of differentiated rods . These data are consistent with a role for Six3/6/7 family members in cell cycle regulation but antagonistic to differentiation [95] , The dissimilarities between six7 and Six3 may reflect inherent differences in the two protein , their binding partners , protein-protein interactions , or changes in the competency of the neural progenitors . We initially reported that knockdown of six7 resulted in an increased number of rods in the larval zebrafish retina , but no change in cone number was observed [57] . More recently , Ogawa et al . ( 2015 ) reported that TALENs-mediated knock-out of six7 , in addition to increased expression of rod genes , resulted in loss of expression of RH2 and lower expression of SWS2 , however no mechanisms underlying these changes were identified [57] [58] . We show that these functions of six7 in photoreceptor development are cell-autonomous . However , the differences in the phenotypes observed between the hypomorphic allele and loss-of-function alleles distinguish separate functions underlying the rod and cone phenotypes . In six7fl4 larvae , the virtual lack of labeling for RH2 , gaps in the red- and green-sensitive cone mosaic and the presence of numerous TUNEL positive cells in the ONL are consistent with failure of the green-sensitive cone precursors to express markers of terminal differentiation and cell death . A few six7fl4 heterozygous larvae failed to label for RH2 , and the phenotype was partially penetrant in larvae homozygous for the six7p23ahub allele . This all-or-none labeling pattern suggests that a small but reproducible number of animals is sensitized to modest changes in the level of six7 expression opening up the potential for genome sequencing data to reveal potential modifiers of the cone phenotype . In contrast , the observation of similar increased numbers and uniform spacing of rod in six7p23ahub and six7fl4 larvae and the weaker phenotype observed in six7p23ahub and six7fl4 heterozygous larvae suggests that rod number is quantitatively sensitive to changes in gene dosage . The ability to genetically separate the rod phenotype from the cone phenotype suggests that six7 functions independently in the two populations of photoreceptor progenitors . six7 is the second gene we have identified which regulates rod and cone development in the zebrafish retina . We initially report a role for tbx2b in the specification of UV-sensitive cones . Although our results support the conservation of the ontological relationship between the UV-sensitive cones and rods observed in mammalian retinas , the identification of a novel role for tbx2b challenged the notion of a default photoreceptor phenotype . The subsequent identification of expression of TBX2 in SWS1-expressing cones in chick suggests a conserved role in cone-dominated retinas [96] . The isolation of alleles of tbx2b and six7 that show no change in coding sequence , but altered expression , provides insight into the potential for modulation of cis-regulatory elements as an underlying feature in varying the number and types of photoreceptors in some species [97] . Ogawa et al . ( 2015 ) speculated that genomic rearrangement led to six7 acquiring a role in RH2 expression in teleosts . Their phylogenic analysis predicts an early duplication event leading to the six7-subfamily but subsequent loss in birds and mammals . The evidence for a reptilian Six7 opens the possibility for a broader role for SIX7 as many lizards also express a functional RH2 [58] . Cis-regulatory alleles are considered unique players in phenotypic evolution [98] , [99] , [100] , [101] . A basic tenant of the field of evolutionary developmental biology ( Evo-Devo ) is that small spatial or temporal changes in gene expression during development can have a dramatic effect upon morphology [102] . Cis-regulatory mutations are often co-dominant where natural selection operates more efficiently; heterozygous organisms express a new trait immediately rather than postponed until brought to homozygosity in the population [103] , [104] . Frequently , mutations in cis-regulatory sequences are modular in their effect , leading to alleles with reduced pleiotropy , which would be favored over structural changes in individual proteins which would risk loss of essential functions in the intermediate phenotypes . Lastly , selection would necessitate that the output of the system is sensitive to variations in the level of expression of the factors . The alleles we recovered show many if not all of these features . We propose that changes in the photoreceptor gene-regulatory network are one potential source for adaptive changes in rod and cone numbers in evolution . This and our previous study of tbx2b identified distinct mechanisms for maintaining the cone-dominated retina in a diurnal species . The mutant phenotypes are consistent with the previously proposed evolutionary trajectories that may have been associated with the adaptation to a nocturnal environment although the precise mechanisms remain to be discovered . Based upon phylogenetic analysis and environmental considerations , Davies et al . , ( 2012 ) proposed a series of structural mutations in opsins associated with adaptation to the present day , rod-dominated phenotype of extant mammals [23] . Similarly , the loss of RH2 and SWS2 are observed in the basal lineage of snakes [25] . However , diurnal or nocturnal vision is not merely defined by the expression of a specific opsin , but rather by the coordinated expression of signal transduction genes , metabolic function and structural elements to maximize sensitivity or spatial and temporal resolution . Additively , increased mitosis of late stage progenitors , selective loss of specific opsin subtypes , and mutations of cis-regulatory enhancers could dramatically alter the types and ratios of photoreceptors in the retina . It is worth mentioning that evidence suggests modification of a rod into a middle wavelength-sensitive cone-like photoreceptor in the recent evolution of the all-cone retina of garter snake [105] . Thus , a few genetic changes could result in a significant shift of photoreceptor composition to retinas better adapted for a novel environment . Characterizing the cis-regulatory elements and trans-acting factors are essential steps towards a more complete understanding of the mechanisms regulating the variations in photoreceptor numbers in zebrafish , and how the potential conservation or loss of these mechanisms shapes photoreceptor patterning in other species . Regardless of the exact mechanism , our study clearly indicates the potential of a small number of genotypic changes in a gene regulatory network provide substantive developmental alterations in photoreceptor genesis .
Zebrafish ( Danio rerio ) were reared , bred and staged according to standard methods [106] . ljrp23ahub was isolated from a three-generation screening of N-ethyl-N-nitrosurea-mutagenized zebrafish immunolabeled for rods at 5 dpf [88] . Mutagenesis was performed at the University of Pennsylvania as previously described [107] . The lorp25bbtl mutant was previously characterized [53] . The six3bvu87 mutant was previously described [69] and was generously provided by Dr . Solnica-Krezel ( Washington University , St . Louis , MO ) . All animal procedures were approved by the Florida State University ( FSU ) Institutional Animal Care and Use Committee , ACUC Protocol #1421 . Animals were anesthetized using MS222 and euthanized in ice water . ljrp23ahub mutant embryos from mating heterozygous adults were identified by immunolabeling as previously described [53] . Linkage mapping was performed at the Zebrafish Mapping Facility at the University of Louisville from DNA isolated from 100 ljrp23ahub mutant- and 100 WT sibling embryos using simple sequence-length polymorphism markers . Fine resolution mapping was performed with 463 ljrp23ahub mutant embryos [108] . Genomic DNA from 118 ljrp23ahub mutant embryos was isolated ( DNeasy Blood 7 Tissue Kit; Quiagen , Valencia , CA , USA ) and used for Illumina sequencing at the University of Texas Genomic Sequencing and Analysis Facility as previously described [109] . Reads were aligned to the zv9 Zebrafish genome assembly ( ensembl ) with BWA [110] using default parameters . Reads with alignment quality of at least 30 were used identify SNPs against the zv9 genome assembly using samtools mpileup and bcftools 0 . 1 . 19 [111] . SNP densities were calculated using bedtools2 [112] . Data was visualized using the UCSC genome Browser [113] , [72] . Genomic DNA was isolated from tail-clip of adult zebrafish and the candidate deleted region was confirmed by PCR in: ljrp23ahub mutants ( n = 9 ) and WT embryos from: AB genetic background ( n = 6 ) , TL genetic background ( n = 6 ) using the primers listed in S1 Table . PCR of six7 fragment was used as positive control using primers listed in S1 Table . One of three different morpholinos ( MO ) were injected into one-cell stage WT embryos ( Gene Tools , LLC , Philomath , OR ) : mispaired-control MO , 5’-CGAACGCCATTCCGAGTCTGACTAAC-3’; antisense nucleotide targeting six7 5’-UTR ( MO1 ) , 5’-CCAACGGCATTCCAGTGTGAGTAAC-3’ [73]; and six7 splice-blocking MO ( MO3 ) , 5’-GTACTTTTTGGTCTCACCTTAAAGC-3’ . Unless otherwise stated , embryos were injected with 0 . 87 ng of the indicated MO . To confirm the efficiency of MO3 , RNA was isolated from un-injected and MO3-injected embryos and the region spanning from exon 1 to exon 2 of the six7-transcript was amplified by PCR using primers listed in S1 Table . The truncated six7-transcript was sequenced using Applied Biosystems 3730 Genetic Analyzer with Capillary Electrophoresis ( Foster City , CA ) . TALEN expression vectors were constructed in the Mutation Generation and Detection Core , University of Utah to target the exon 1 of six7 transcript . DNA plasmids were linearized by NotI ( Invitrogen , Carlsbad , CA ) and used as templates for TALEN mRNA synthesis with SP6 mMESSAGE mMACHINE Kit ( Ambion , Austin , TX ) . To target the six7 genomic sequence , 50–200 pg of the pair of TALEN mRNAs were injected in one-cell stage zebrafish embryos . Injected embryos were raised to adulthood and crossed to WT animals to generate the F1 . DNA was extracted from either F1 embryos ( groups of three to six embryos ) from the outcross of founders or tail clips from adult F1 fish . To screen for insertions and deletions ( indels ) , DNA was extracted and used as the PCR template to amplify the six7-TALENs targeted region using primers listed in S1 Table . The DNA fragment was subjected to restriction fragment length polymorphism ( RFLP ) assay . Indels were tracked by loss of HaeIII ( Invitrogen ) restriction enzyme site in the targeted region . PCR products were sequenced to characterize the indels . The F1 embryos of positive founders were intercrossed to generate the F2 generation . F1 and F2 embryos were fixed in 4% paraformaldehyde in 80% phosphate-buffered saline ( PFA/PBS ) and processed for whole-mount rod immunolabeling as described ( see Immunohisto-and immunocytochemistry ) . RNA extraction was performed in TRIzol ( Invitrogen ) from pool of whole embryos ( n = 30 ) at 10 hpf , 18 hpf , 24 hpf and 52 hpf . Transcription into cDNA was performed using SuperScript™ II Reverse Transcriptase ( Invitrogen ) . Real time quantitative PCR ( RT-qPCR ) was carried out using a 7500 Real-Time PCR Systems ( Applied Biosystems ) with SRBY-Green PCR Master Mix ( Applied Biosystems ) and the primers listed in S1 Table . Three biological replicates were performed for each developmental time and were duplicated for each cDNA sample for six7 qRT-PCR . The fold expression change was normalized to β-actin using the 2-∆∆CT ( Livak ) method [114] . Student’s t test was applied for comparison between groups at each developmental time . Immunolabeling of larvae whole mount or cryosections ( 10 μm ) was performed as previously described [88] . Sections and enucleated eyes from whole-mounted immunolabeled larvae were imaged using either a Zeiss Axiovert S100 fluorescent microscope ( Carl Zeiss Inc . , Thornwood , NY ) or a LSM 510 or LSM 710 ( Carl Zeiss ) Laser Confocal equipped with a 40x C-Apochromat water immersion objective ( N . A . 1 . 2 ) . The following primary antibodies were used: monoclonal antibody 4C12 that labels rods ( 1:200 , [115] ) , a monoclonal antibody zpr1 that labels double cone cells ( arr3a ) ( 1:20 , [62] ) , a monoclonal antibody 1D1 against rhodopsin [115] a polyclonal antibody against zebrafish blue- , red- , green- or UV-sensitive cone opsin ( 1:200 , [12] ) , Zn8 which recognizes ganglion cells ( 1/10 , ZIRC ) , 5E11 that labels amacrine cells ( [115] , PKCα that labels bipolar cells ( 1/100 , [57] ) , CAZ which recognizes Muller glia ( 1/100 , [116] ) and polyclonal PH3 antibody that labels mitosis marker phospho-Histone 3 ( 1:500; Cat . No . 06–570 , Millipore , Billerica , MA ) . Host-specific , Alexa fluor-conjugated secondary antibodies ( Invitrogen ) were used at a dilution of 1:200 . Sections were counterstained with 4′ , 6-diamidino-2-phenylindole , dihydrochloride ( DAPI , 1:15000; Sigma-Aldrich ) . Proliferation was assessed by incubation of 48 hpf and 52 hpf embryos in fish water with 1 . 5 mM EdU ( 5-ethynil-2’-dexyuridine ) during 30 minutes and subsequently fixed in 4% paraformaldehyde ( PFA/PBS ) . The EdU labeling was processed by the Click-iT EdU Alexa Fluor 546 Imaging kit ( Invitrogen ) following the manufacturer’s instructions . For lineage tracing cell experiments , the EdU was washed with fish water and the embryos were incubated until 4 dpf and subjected to immunohistochemistry . Larvae were euthanized with tricaine and processed according to the protocol by [117] , with slight modifications . Briefly , larvae were fixed overnight with 1% glutaraldehyde and 1% osmium tetroxide in 0 . 1 M cacodylate buffer . The eyes were then washed 3 times and dehydrated through a graded ( 70% , 75% , 90% , 100% , 100% , 100% ) acetone-water series . The tissue was infiltrated overnight at room temperature in a 1:1 mixture of epoxy resin and 100% acetone . Larvae were then embedded in epoxy resin and placed in a 60°C oven for 22 hours . 1μm sections were cut with a microtome and mounted on glass slides and stained with 1% methylene blue in 1% sodium borate . Photomicrographs were taken with a Zeiss Axiovert microscope , and images were captured by the Zeiss Axiocam Digital Camera and processed using the Axiovision software . For electron microscopy , sections were collected on copper grids and viewed with a FEM CM 120 transmission electron microscope . Electron micrographs were taken using a Tietz Tem-Cam F224 slow scan CCD camera , prior to being imported into Photoshop version 8 . 0 ( Adobe Systems ) . Confocal images from whole eyes immunolabeled for UV-sensitive cones and rods were analyzed with the Scion Image Software ( Scion Corp , Frederick , MA ) . Areas of 3500 μm2 located dorsal to the optic nerve [53] were counted for rods and UV-sensitive opsin expressing cones in WT ( n = 5 ) ; lorp25bbtl ( n = 5 ) ; ljrp23ahub ( n = 5 ) ; double mutant lorp25bbtl/ljrp23ahub ( n = 5 ) at 4 dpf . Quantification of rods was conducted similarly as for WT ( n = 5 ) ; ljrp23ahub ( n = 5 ) and ljrp23ahub/+ ( n = 6 ) an independent experiment . In addition , rods were similarly counted in WT ( n = 8 ) ; six7 fl4/fl4 ( n = 7 ) , six7 fl4/p23ahub ( n = 4 ) . The number of red-green-sensitive cones was quantified in WT ( n = 6 ) ; ljrp23ahub ( n = 6 ) ; and in independent experiment WT ( n = 3 ) ; six7 p23ahub ( n = 6 ) ; six7 fl4 ( n = 3 ) . When possible two of 3500 μm2 retinal areas were counted . six7-MO1 injected retinas ( n = 5–6 retinas/each MO1 dose ) were imaged and rods were quantified at 4 dpf for 3500 μm2 area . Un-injected ( n = 4 ) retinas were used as controls . The average number of UV- , Arr3a-positive or rod-positive cells per unit area and the standard deviation ( SD ) were reported . One-way ANOVA with Tukey’s post-hoc test was used to compare means of rods or UV-sensitive cones between different genotypes . Student t test was applied to compare two sample data . The number of PH3 positive cells in the ONL and INL was quantified using 10-μm-thick-retina sections of 21 000 μm2 area per section that excluded the CMZ . The following strains and number ( n ) of 48-hpf retinas were analyzed: WT ( n = 5 ) , six7-MO1 injected embryos ( n = 4 ) . A Student’s t-test was conducted to compare the number of PH3 positive cells in ONL and INL between WT and six7-knockdown retinas . Same procedure was used to count PH3 positive cells in the ONL and INL of six7 fl4/f+ ( n = 5 ) and six7 fl4/fl4 ( n = 11 ) . Un-paired Student t test with Welch’s correction was used for statistical analysis . Quantitative analysis of photoreceptor pattern was performed as described [59] . Nearest Neighbor Dispersion Analysis ( NNDA ) was determined using Biotas and the conformity ratio was calculated and analyzed for randomness using the Ready-Reckoner Chart of Cook [61] . Fluorescent structures were assigned ( x , y ) coordinate using ImageJ software ( National Institutes of Health Windows version , ( http://rsbweb . nih . gov/ij/index . html ) . For each point in the field Nearest Neighbor Distance ( NND ) was calculated using Biotas ( Version 1 . 02; Ecological software Solutions ) . as previously described [59] . Whole-mount in situ hybridizations were performed as previously described [118] using pools of 25 embryos at 28 hpf and between 46–52 hpf . The antisense riboprobes were: six7 ( this study ) , vax2 ( this study ) , cyp26c1 ( this study ) and tbx2b [53] . The plasmids containing the probes for neurod [80] , crx [76] , pax6a [119] , rx1 [120] were kindly provided by A . C . Morris ( University of Kentucky , Lexington , KY ) . To prepare a probe for six7 , a 444 bp fragment of six7 gene was amplified from a cDNA fragment obtained from 10 hpf embryos , using primers listed in S1 Table and cloned into the vector PCR2 . 1-TOPO ( Invitrogen ) . Antisense RNA probe was synthesized with a digoxigenin RNA-labeling kit ( Roche , Indianapolis , IN ) by in vitro transcription with T7 RNA polymerase , according to the manufacturer's instructions . A 620 bp of vax2-cDNA fragment and a 692 bp of cyp26c1 were amplified using primers listed in S1 Table . The antisense probes were prepared as described above . The hybridized probe was detected with alkaline phosphatase coupled with anti-digoxigenin antibodies and NBT/X-phosphate substrate ( Roche ) . Labeled embryos were cleared in a graded series of glycerol and viewed on a Zeiss Axiovert S100 microscope . Images were captured by Carl Zeiss Axiocam Color Microscope camera and processed with Axiovision SE64 Rel 4 . 9 . 1 and Photoshop 5 . 5 ( Adobe , Mountain View , CA ) software . Terminal deoxynucleotide transferase ( TdT ) -mediated dUTP nick labeling ( TUNEL ) was performed on 3 dpf retinal cryosections using the ApopTag Red In Situ Apoptosis Detection Kit ( Millipore , Temecula , CA ) according the manufacturer’s instructions and co-labeled for rods ( 4C12 ) to identify mutants . TUNEL positive cells were counted in the ONL , INL and GCL from: WT ( n = 7 ) ; ljr ( n = 9 ) ; control ( n = 7 ) and six7-MO1 ( n = 6 ) , one section for individual embryo . TUNEL positive cell counts were transformed ( log Y+1 ) before student t test was conducted . TUNEL assay was performed in six7fl4 mutants and WT embryos at 56 hpf and 4 dpf . Tail-clip genotyping was used to identify mutants at 56 hpf . The following strains and number ( n ) of 56 hpf were analyzed: WT ( n = 5 ) , six7fl4 ( n = 7 ) and at 4 dpf: WT ( n = 3 ) , six7fl4 ( n = 6 ) . Genetic chimeras were generated as previously described [121] . Donor embryos were injected at the 1- and 2-cell stage with the lysine-fixable , dextran-conjugated Alexa Fluor 594 ( Invitrogen ) . Donor blastulae cells were transferred to unlabeled host cells . At 4 dpf the chimeras were fixed with 4% PFA/PBS and immunolabeled for rods and green-opsin as described above . Imaging of the whole-dissected eyes was performed by confocal microscopy ( WT into WT , n = 3; six7fl4 into WT n = 3; WT into WT , n = 5; six7-MO1 into WT , n = 6 ) . The number of rhodamine-dextran labeled cells , 4C12/dextran ( rod from donor cells ) -labeled and green-opsin/dextran labeled cells were quantified . The percentage number of donor cells differentiated as rod photoreceptor or green-sensitive cones were compared for six7fl4 mutants vs WT transplants into WT background . Numbers of donor cells were counted across the retinal layers from retinal sections of six7-MO into WT . Statistical analysis was performed by chi-square test . | Vision begins when an image is focused on the neural retina where rod and cone photoreceptors convert light into the electrical signals of the brain . The 4 cone subtypes in retinas of the majority of fishes , lizards and birds , provide rich color vision . In contrast , retinas of most mammals are better adapted for dim light conditions with rods vastly outnumbering the sparse and less diverse cone subtypes . However , our understanding of photoreceptor development largely based on findings from mammalian models fails to explain the tremendous diversity of cone subtypes and variation of rod and cone numbers across the majority of vertebrate species . Taking advantage of the cone-rich zebrafish retina , we identified a nuclear factor that suppresses the number of rods and is essential for the development of a cone subtype not present in mammals . Combined with prior studies , the findings provide insight into adaptive mechanisms underlying maintenance of a cone-dominated retina . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"ocular",
"anatomy",
"social",
"sciences",
"vertebrates",
"neuroscience",
"animals",
"alleles",
"cell",
"differentiation",
"animal",
"models",
"osteichthyes",
"mitosis",
"developmental",
"biology",
"model",
"organisms",
"embryos",
"research",
"and",
"analysis",
"methods",
"embryology",
"fishes",
"animal",
"cells",
"chromosome",
"biology",
"sensory",
"receptors",
"genetic",
"loci",
"signal",
"transduction",
"zebrafish",
"cellular",
"neuroscience",
"psychology",
"retina",
"anatomy",
"cell",
"biology",
"neurons",
"genetics",
"photoreceptors",
"ocular",
"system",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"afferent",
"neurons",
"sensory",
"perception",
"metamorphosis",
"larvae",
"organisms"
]
| 2016 | Genetic Dissection of Dual Roles for the Transcription Factor six7 in Photoreceptor Development and Patterning in Zebrafish |
Heme is a cofactor in proteins that function in almost all sub-cellular compartments and in many diverse biological processes . Heme is produced by a conserved biosynthetic pathway that is highly regulated to prevent the accumulation of heme—a cytotoxic , hydrophobic tetrapyrrole . Caenorhabditis elegans and related parasitic nematodes do not synthesize heme , but instead require environmental heme to grow and develop . Heme homeostasis in these auxotrophs is , therefore , regulated in accordance with available dietary heme . We have capitalized on this auxotrophy in C . elegans to study gene expression changes associated with precisely controlled dietary heme concentrations . RNA was isolated from cultures containing 4 , 20 , or 500 µM heme; derived cDNA probes were hybridized to Affymetrix C . elegans expression arrays . We identified 288 heme-responsive genes ( hrgs ) that were differentially expressed under these conditions . Of these genes , 42% had putative homologs in humans , while genomes of medically relevant heme auxotrophs revealed homologs for 12% in both Trypanosoma and Leishmania and 24% in parasitic nematodes . Depletion of each of the 288 hrgs by RNA–mediated interference ( RNAi ) in a transgenic heme-sensor worm strain identified six genes that regulated heme homeostasis . In addition , seven membrane-spanning transporters involved in heme uptake were identified by RNAi knockdown studies using a toxic heme analog . Comparison of genes that were positive in both of the RNAi screens resulted in the identification of three genes in common that were vital for organismal heme homeostasis in C . elegans . Collectively , our results provide a catalog of genes that are essential for metazoan heme homeostasis and demonstrate the power of C . elegans as a genetic animal model to dissect the regulatory circuits which mediate heme trafficking in both vertebrate hosts and their parasites , which depend on environmental heme for survival .
From a nutritional perspective , heme is a readily bioavailable source of iron for human consumption [1] , [2] . From a cellular perspective , heme is an iron-containing porphyrin which serves as a prosthetic group in diverse biological processes ranging from gas-sensing to microRNA processing [3] . In most eukaryotes , heme is synthesized in the mitochondrial matrix by a defined biosynthetic pathway and subsequently exported as needed for heme-containing proteins that are found in the cytoplasm and membrane-bound organelles [3] . Given the hydrophobicity and cytotoxicity associated with free heme , it is likely that specific intracellular transport pathways exist to deliver heme for assimilation into hemoproteins found in various subcellular compartments [4] . Although the pathway and intermediates for heme biosynthesis and degradation have been well defined , the intracellular networks that mediate heme homeostasis in eukaryotes remain poorly understood [4] . Heme transport molecules in animals are likely to be divergent from bacterial and yeast proteins at the genetic level; bacterial and yeast heme-binding proteins have no obvious orthologs in mammals [5]–[8] . This is demonstrated by the identification of a heme exporter , the feline leukemia virus subgroup C cellular receptor ( FLVCR ) , which does not show any obvious similarities to known bacterial heme transport proteins [9] , [10] . Genetic ablation of FLVCR in mice resulted in severe macrocytic anemia with proerythroblast maturation arrest . That a viral receptor could be a potential heme exporter in developing erythroid cells underscores the divergence among heme transport proteins and emphasizes the importance of implementing unbiased genetic approaches to elucidate the heme homeostasis pathways in tractable model systems . Progress in understanding heme homeostasis in most eukaryotic systems is hampered by the inability to separate heme biosynthesis from downstream intracellular transport pathways . To circumvent this issue , we established the genetically tractable nematode Caenorhabditis elegans as an animal model ideally suited in which to conduct heme studies . We have previously demonstrated that this roundworm does not synthesize heme but instead relies on environmental heme for survival [11] . Moreover , analyses of available genomes from related parasitic nematodes suggest that these helminths are also heme auxotrophs [11] . The C . elegans genome encodes a repertoire of hemoproteins that have vertebrate orthologs . It is likely that the pathways for heme trafficking and incorporation are conserved in C . elegans , parasitic worms , and vertebrates [4] . The validity of the C . elegans model system was recently underscored by the discovery of HRG-1 proteins that transport heme [12] . We identified C . elegans hrg-1 and its paralog hrg-4 from microarray experiments as genes that were highly upregulated by low heme [12] . Expression of these genes and their human homolog , HRG-1 , in Xenopus oocytes resulted in strong heme-induced electrophysiological currents – an indication that the corresponding proteins were heme transporters . Additionally , depletion of hrg-1 in worms led to aberrant heme homeostasis . Transient knockdown of hrg-1 in zebrafish caused severe impairment in erythropoiesis along with brain and skeletal defects; these phenotypes were fully rescued by worm hrg-1 [12] . Collectively , these studies further validated the advantage of C . elegans as a model par excellence to dissect the pathways responsible for heme transport and homeostasis in mammals . Moreover , C . elegans bridges the evolutionary divide to heme auxotrophic parasitic species and provides insight into helminthic-specific vulnerabilities in heme uptake and utilization that can be exploited for drug design [13] , [14] . The current study specifically seeks to explain and draw conclusions from the genomic data that was generated from our microarray analysis . This expression array analysis using C . elegans wild-type worms grown in an axenic liquid medium at three different concentrations of heme was performed as a first step in the genome-wide identification of genes involved in heme homeostasis . Our results have identified several hundred heme-responsive genes ( hrgs ) , some of which are evolutionarily conserved across metazoa while others are found only in nematodes . We anticipate that results from our genomic studies may be universally applicable and result in the discovery of heme homeostasis pathways in other metazoans .
C . elegans lacks the highly conserved genes of heme biosynthesis but acquires heme from the environment for growth and development [11] . Worms cultured in axenic liquid mCeHR-2 medium in the presence of different amounts of heme revealed a characteristic growth curve [11] . The optimal concentration for worm growth and reproduction was found to be 20 µM heme , although animals grew and reproduced at concentrations ranging from ≥1 . 5 µM to <800 µM heme . Worms grown in the absence of exogenous heme arrested at the L4 larval stage , whereas concentrations of heme ≥800 µM caused the worms to arrest at the L2/L3 larval stages , possibly due to heme cytotoxicity . These results are consistent with metabolic labelling experiments in which the fluorescent heme analog , zinc mesoporphyrin IX ( ZnMP ) , was used to demonstrate that the heme uptake system is regulated in C . elegans [12] . To determine if there were transcriptionally regulated components of heme uptake , wild-type N2 worms were grown at 4 , 20 , or 500 µM heme in axenic liquid mCeHR-2 medium; 20 µM served as the reference sample . We chose 4 and 500 µM heme because these concentrations were on either side of the biphasic growth curve . More importantly , although worms grown at these heme concentrations exhibited a 16 h growth delay , they were morphologically indistinguishable from worms grown at 20 µM heme . In order to reduce variability due to carryover of maternal heme from the P0 hermaphrodites , worms were grown in their respective heme concentrations for two successive generations ( Figure 1 ) . Synchronized , late L4 larvae from the F2 generation were harvested for RNA isolation , and corresponding cDNA probes were generated and hybridized to Affymetrix C . elegans expression microarray chips . Three biological replicates were prepared for each heme concentration . Statistical analyses of the microarray data were initially performed using the Affymetrix MAS 5 . 0 suite software ( see Materials and Methods ) . Of the 22 , 627 probe sets , 835 probe sets revealed changes at either 4 or 500 µM heme compared to the control data from 20 µM heme . We identified 288 genes with a ≥1 . 6-fold change in expression . To improve and augment these analyses , we also subjected the microarray results to the Robust Multichip Average method ( RMA from R package ) with the goal of combining the results with those obtained by MAS 5 . 0 . The RMA analysis ( minimum change in expression ≥1 . 2 fold ) identified an additional 82 hrgs . The MAS 5 . 0 and RMA analyses yielded a total of 370 candidate genes . Subsequently , duplicate genes were eliminated , the minimum cut-off value for RMA analysis was increased to ≥1 . 6 fold , and the average of the fold-change values was calculated for the replicates . This resulted in a list of candidate genes consisting of 266 genes identified using MAS 5 . 0 and 22 genes selected using the RMA method . The expression of these 288 genes , eight of which were previously identified as germline genes [15] , revealed a ≥1 . 6-fold change at either 4 or 500 µM heme compared to the 20 µM controls . Consequently , all 288 genes were classified as hrgs ( Table S1 ) . Normalized signal intensity values can be graphed to visualize the quality of microarray data generated by each replicate ( Figure 2A ) . The value at which the colored lines cross each thin vertical line is the value of the normalized signal for that replicate . Accurate replicates should have nearly horizontal lines ( all values approximately equal ) within each condition that may then decrease or increase in the next condition if there is a change . In this experiment , analysis of each of the 288 hrgs revealed that individual biological replicates had nearly equal values with little variation within a particular heme concentration , indicating that changes in heme-dependent gene expression were uniform . A principal components analysis ( PCA ) for the hrgs showed that , with one exception , the quality of the microarray data was consistent across biological replicates for all three heme concentrations . The data obtained from one of the 4 µM heme replicates showed an inconsistent global gene expression pattern when compared to the other two replicates and was , therefore , excluded from further analysis ( Figure S1 ) . The 288 hrgs were assigned to one of eight categories based on whether the gene expression was upregulated , downregulated , or unchanged in samples obtained from worms grown in 4 or 500 µM heme and compared to the 20 µM reference samples ( Figure 3 ) . Eighty genes were upregulated at 4 µM heme ( Table S2 ) . Seventy-five genes were upregulated at 500 µM heme ( Table S3 ) . Quantitative real-time PCR analysis ( qRT-PCR ) of three representative genes from each of the eight categories was performed to ensure that the changes observed in the microarray were reproducible . As determined by the significance ( P<0 . 0001 ) and the Pearson's correlation coefficient , the qRT-PCR confirmed that the changes observed with the microarray results were consistent and , therefore , reliable ( Figure 2B; Table S4 ) . Since identification of the hrgs common to both C . elegans and mammals might provide unique insights into the evolutionary conservation of heme homeostasis pathways in metazoans , we performed reciprocal BLAST searches to identify putative human orthologs of each of the 288 genes ( Figure 3 ) . Searches using protein sequences revealed that there were 121 putative human orthologs ( minimum E-value = 10−4 ) of C . elegans hrgs . The hrgs with human homologs were present among those upregulated in both extreme heme concentrations . Forty-four were upregulated at 4 µM heme and 42 were upregulated at 500 µM heme , while 28 were downregulated at 4 µM heme and 36 were downregulated at 500 µM heme ( Table S1 ) . We have previously demonstrated by biochemical enzyme assays and genomic analyses that several of the parasitic nematodes with sequenced genomes lack the genes for heme synthesis enzymes and , therefore , likely rely on environmental heme to sustain growth and development [11] . Similarly , the genomes of Trypanosoma and Leishmania appear to lack most of the genes for heme synthesis [16] , [17] . This suggests that these protozoa may also acquire heme from their parasitized host . Figure 3 identifies the hrg homologs in Trypanosoma brucei , Trypanosoma cruzi , and Leishmania major . Of the 288 hrgs , only 12 genes were exclusive to these heme auxotrophs . Thirty-seven genes had homologs only in humans , and 84 genes were found in both human and parasitic genomes ( Figure 4A ) . These results indicate that heme-regulated genes in C . elegans may have commonality with humans that are heme prototrophs and protozoan parasites which rely on environmental heme . A small percentage of the 288 hrgs had homologs in parasitic nematodes ( Figure 4B ) . To date , draft genomes of several parasitic nematodes have become available [18]–[21] , in addition to the partial genomes available for over 30 parasitic species . For a summary of available genomes , see [22] . Using all available sequence data divided into taxonomically distinct clades [23] , we identified homologs for 62 of the 288 hrgs in the clade V nematodes ( C . elegans belongs to clade V ) and homologs to only 10 genes in the clade I nematodes ( where the basal nematode , the zoonotic parasite Trichinella spiralis , resides ) . While the number of identified putative orthologs was much higher for the crown lineages than in the basal nematodes that reside at the root of the nematode evolutionary tree , two of the eight categories ( categories 1 and 3 ) had no homologs in any of the parasitic species . Categories 1 and 3 are represented by 13 and 10 sequences in C . elegans , respectively . Gene ontology ( GO ) analysis [24] indicated that the hrgs identified from our microarray study were involved in processes as varied as embryonic development , electron transport , lipid metabolism , and iron-sulfur cluster assembly . Of the 288 genes in the study , 115 were annotated with a biological process ( Table S5 ) . Using the Fisher's exact test , a hierarchical graph was constructed with the most significant GO terms and their associated parent terms [25] . Highly significant GO terms ( P<0 . 005 ) associated with the subset of genes that were upregulated at 4 µM heme were ‘embryonic development’ , ‘lipid transport’ , and ‘proteolysis’ ( Figure 4C; Table S6 ) ; ‘responses to stress’ and environmental stimuli' were associated with genes that were downregulated at 4 µM heme ( Figure S2 and Tables S7 , S8 , S9 ) . The Kyoto Encyclopedia of Genes and Genomes ( KEGG ) is also frequently used to analyze complex microarray data and make functional predictions [26] . Only 10 hrgs ( ∼3% ) have been mapped to KEGG pathways ( Table S10 ) . These hits included genes for transporters and also for metabolism of sugars , an amino acid , and fatty acids . A majority of hrgs that we identified were uncharacterized with no assigned biological pathway . Genome sequencing has demonstrated that chromosomes I , II , III , IV , and X in C . elegans each contain roughly equivalent numbers of genes ( 13–17% ) , whereas chromosome V has the most genes ( 25% ) [27] . Furthermore , co-regulated or functionally related genes , especially those essential for interactions with the environment , tend to reside in local clusters on the chromosome [27] . We found that Chr I and Chr III each contained just 6% of the hrgs , but 35% of all hrgs were found on Chr V ( Figure 5 ) . Additionally , of the 129 hrgs on Chr V , 43 genes were upregulated at 4 µM heme while 41 genes were upregulated at 500 µM heme ( Figure 6 ) . Our analysis suggests that the genomic distribution of hrgs was non-random , reveals gene clustering , and indicates a common biological response to an environmental stimulus such as heme . If the genomic distribution of hrgs is purposeful , we reasoned that perhaps the response of the promoters of the hrgs is directed by a cis-acting element within a cluster or elements that are common to all hrgs in a specific category . First we analyzed Categories 1 and 2 for overrepresented transcription factor binding sites using all sequences against a control set of random promoter sequences but failed to identify common cis elements . We reiterated our search to encompass the presumptive promoters ( ≥2 kb upstream ) of all 288 hrgs using TRANSFAC [28] . Once again , no common elements were identified . A number of genome-wide RNA-mediated interference ( RNAi ) experiments have been performed in C . elegans , and the data from all these experiments are available on Wormbase ( http://www . wormbase . org/ ) . Forty-six hrgs ( 16% ) had a reported RNAi phenotype ( Table S11 ) . RNAi knockdown of these genes most often resulted in developmental defects such as sterility and embryonic lethality . These phenotypes were expected because heme is essential for growth and reproduction of C . elegans . The relatively small fraction of genes yielding a reported RNAi phenotype probably reflects redundancy of function among some of the hrgs and the limited phenotypic assays performed to date . We have previously reported that transgenic worms expressing the hrg-1::gfp transcriptional fusion ( strain IQ6011 ) specifically respond to heme in the growth medium . Thus , strain IQ6011 can be used as a whole animal heme sensor to interrogate changes in organismal heme homeostasis [12] . To identify the function of the hrgs in heme homeostasis , we established a functional RNAi screen using IQ6011 ( see Materials and Methods for details ) . First , we generated a sequence-confirmed hrg mini-library in the E . coli feeding strain HT115 ( DE3 ) that expressed double-stranded RNA ( dsRNA ) against each of the 288 hrgs . Second , we established a sensitive GFP-based assay that conditionally screened for genetic modulators of heme homeostasis simultaneously in the presence of low ( 5 µM ) or high ( 25 µM ) heme . Third , we verified the positive candidate genes with a secondary screen to eliminate false positives using a vha-6::gfp transgenic worm that does not respond to heme and served as a negative control . Fourth , we confirmed the authenticity of each candidate gene by simultaneously measuring the GFP fluorescence intensity in IQ6011 and vha-6::gfp with a COPAS Biosort instrument that sorts each worm by its time of flight ( axial length of object ) and extinction ( optical density of object ) . Synchronized IQ6011 worms were grown in mCeHR-2 medium supplemented with 10 µM heme to repress GFP and subsequently transferred to NGM agar plates for exposure to dsRNA produced by E . coli grown either in the presence of 5 µM or 25 µM heme on NGM agar plates . These experiments were performed in duplicate , and GFP levels and patterns in worms fed bacteria expressing each of the 288 hrgs were analyzed by eye . RNAi depletion of the 288 hrgs resulted in the identification of 32 genes that specifically upregulated or downregulated GFP expression in the IQ6011 heme-sensor strain but not in the vha-6::gfp control strain . These 32 genes were selected for further analysis by the COPAS BioSort . We identified six hrgs which caused either a two-fold increase or a two-fold decrease in GFP expression , ( Figure 7A ) . A significant upregulation of GFP was observed at 5 µM when five hrgs that encoded either putative membrane-spanning proteins ( F36H1 . 5/HRG-4 , F14F4 . 3/MRP-5 , F58G6 . 3/CTR-1 , and F22B5 . 4/unnamed protein ) or a putative lysosomal cysteine protease ( F32H5 . 1/cathepsin-L ) were depleted . In contrast , GFP was downregulated only when F46E10 . 11 , which encodes an uncharacterized protein proposed to bind metals through cysteine residues , was depleted . To identify potential heme transporters , we identified hrgs which encoded for proteins with transmembrane domains ( TMD ) . TMHMM analysis predicted that 41 of the 288 hrgs encoded for proteins with at least one putative TMD ( Table S12 ) . Among these 41 genes were those encoding aquaglyceroporin-related proteins ( aqp-1 and aqp-8 with six and four TMD , respectively ) that transport small molecules such as glycerol , urea , and water; cyp-33C9 ( one TMD ) which belongs to the cytochrome P450 family of heme binding proteins; heme permeases ( hrg-1 and hrg-4 with 4 TMD ) ; and ABC transporters ( mrp-5 and pgp-1 with ≥12 TMD ) . To narrow the list of candidate heme transporters , we used RNAi to deplete the 41 hrgs which encoded TMD proteins and exposed the worms to gallium protoporphyrin IX ( GaPP ) , a toxic heme analog that causes severe defects in worm growth and development [11] . We reasoned that knockdown of a putative heme transporter in the presence of GaPP would result in a concomitant increase in worm survival due to a reduced ability to transport toxic GaPP [12] . We identified seven hrgs which , when depleted by RNAi , revealed greater survival of the F1 progeny at 1 . 5 µM GaPP , a concentration that is lethal to wild-type worms ( Figure 7B ) . These seven genes included F36H1 . 5/hrg-4 , F14F4 . 3/mrp-5 , K08E7 . 9/pgp-1 , Y51A2D . 4/hmit1 . 1 , Y37A1A . 2 , T21C9 . 1 , and F22B5 . 4 . Three genes – hrg-4 , mrp-5 , and F22B5 . 4 – were positive in both of the RNAi screens ( Figure 7C ) . Taken together , our genomic studies identified a small subset of genes that are not only regulated by heme at the mRNA level but are also essential for heme transport and homeostasis at the functional level . To better understand the role of hrg-4 , mrp-5 , and F22B5 . 4 in heme homeostasis , we determined their mRNA expression in response to heme and their ability to transport heme as a function of ZnMP accumulation [12] . qRT-PCR results indicated that all three genes were upregulated by heme but the magnitude of change in mRNA expression at 4 µM heme was significantly greater for hrg-4 than mrp-5 or F22B5 . 4 ( 8 . 5-fold versus 4 . 5- and 2-fold ) ( Figure 8A ) . Heme uptake assays with ZnMP revealed that hrg-4 RNAi resulted in abrogation of ZnMP accumulation in the worm intestine compared to wild-type control worms ( Figure 8B ) , a result consistent with our previous studies [12] . By contrast , ZnMP accumulation was dramatically increased by the knockdown of both mrp-5 and F22B5 . 4 . Although HRG-4 has been implicated in intestinal heme transport in C . elegans , no function has been attributed to either F22B5 . 4 or MRP-5 in WormBase . Membrane topology algorithms predicted that , unlike F22B5 . 4 , which is predicted to contain a single TMD , HRG-4 and MRP-5 contain four and twelve TMD respectively , a characteristic feature of membrane transporters . To correlate the intestinal ZnMP uptake studies with membrane transport , we examined the gene expression pattern of hrg-4 and mrp-5 . We generated transgenic worms that expressed hrg-4::gfp and mrp-5::gfp transcriptional fusions . hrg-4::gfp was expressed specifically in the intestinal cells of larvae and adults ( Figure 8C ) , and was regulated by exogenous heme ( not shown ) . Unlike hrg-4::gfp , we found that mrp-5::gfp was expressed in almost all worm tissues examined . Altogether , these studies identify HRG-4 and MRP-5 as membrane transporters that are essential for intestinal heme homeostasis in C . elegans .
A major impediment to the identification of heme uptake and transport pathways has been the inability to disassociate the tightly regulated process of heme synthesis from the downstream pathways for heme transport [3] . C . elegans is unique among the model organisms because it does not synthesize heme but , instead , relies solely on exogenous heme for normal growth and development [11] . Thus , worms allow the study of heme homeostatic mechanisms in response to fluctuations in environmental heme within the context of an intact , live animal . Our study of the genome-wide transcriptional changes associated with heme availability represents , to the best of our knowledge , the first study of nutrient-gene interactions in C . elegans exploiting axenic liquid growth medium . The mCeHR-2 medium permits fine control of organismal heme levels as a function of heme in the growth medium , allowing us to identify 288 heme-responsive genes ( hrgs ) . Some of the genes we identified were predictable , because they encode either known heme-binding proteins or permeases for transport of other small molecules . Other genes made sense in retrospect , such as glutathione transferases ( GST ) . A recent proteomic analysis of C . elegans identified GST-19 as a highly abundant protein that was proposed to sequester heme when intracellular heme is in excess [29] . GSTs have also been shown to bind heme in helminths such as hookworms and Barber pole worms [30]–[32] . Our genomic analysis indicates that gst-22 and gst-16 were upregulated at 500 µM heme . Whether these GST proteins also bind heme remains to be determined . GO and KEGG pathway analyses reveal that hrgs represent the full spectrum of biological processes . Interestingly , only a few hrgs are enzymes or proteins that are known to bind heme . We speculate that the transcriptional regulation by heme primarily targets the cellular pathways involved in heme homeostasis , including uptake and sequestration , rather than the genes which encode target hemoproteins . The vast majority of hrgs have no known function and , therefore , do not have any biological processes or pathways attributed to them . Furthermore , phenotypes from RNAi studies involving the 288 hrgs reported growth and developmental defects , plausibly because disruption of heme homeostasis will affect hemoprotein function in diverse biological pathways ranging from miRNA processing ( DGCR8 ) to gas sensing ( soluble guanylyl cyclases ) to circadian clock control ( Rev-erbα ) [33]–[35] . The 288 hrgs we identified also provide the first insight into metazoan heme regulation . The fact that >40% of hrgs have human homologs suggests that our study may provide genetic insights into mammalian heme regulation . This is underscored by the presence of human homologs for genes that were positive in our functional RNAi screen . Indeed , recent studies using C . elegans as a model system have led to the identification of HRG-1 as the first bona fide metazoan heme importer that is conserved in vertebrates [12] . Analysis of the presumptive promoters of all 288 hrgs in eight categories identified no common cis elements [28] . A more detailed analysis of the 67 genes in Category 2 , to which hrg-1 and hrg-4 belong , found no overrepresented transcription factor binding sites using all sequences against a control set of random promoter sequences . These in silico results corroborate our experimental studies ( Chen , Sinclair , and Hamza , unpublished results ) and further support the concept that regulation of organismal heme homeostasis is complex , multi-tiered , and effected by diverse cellular modulators . Studies have demonstrated that the infectivity of hookworms , which feed on the blood of the host , is significantly lower in severely anemic hamsters fed a low-iron diet [13] . Furthermore , filarial nematodes , such as the causative agent of elephantiasis , harbor Wolbachia – an intracellular bacterial symbiont that contains the intact heme biosynthesis pathway [14] , [36] . Thus , nematodes may have adapted to heme auxotrophy by evolving pathways to acquire heme either from the host ( extracellular ) or from the symbiotic relationship with the bacteria ( intracellular ) . This auxotrophy can be exploited to develop drugs that block parasite-specific heme uptake or utilization . Indeed , genome database searches of heme auxotroph parasites led us to identify 12 hrg homologs in protozoans and 62 hrgs in clade V nematodes . This finding is significant because these genes may encode proteins involved in heme uptake and sequestration from the parasitized host . Further studies aimed at elucidating the role of these hrgs in heme metabolism may validate them as novel anti-parasitic drug targets . We found that seven of the 41 hrgs that encode for proteins which contain putative TMDs showed different levels of resistance against GaPP toxicity . Among these were a heme permease ( HRG-4 ) , ABC transporters ( PGP-1 and MRP-5 ) , and Major Facilitator Superfamily transporters ( HMIT-1 . 1 and Y37A1A . 2 ) . The remaining 247 hrgs encoded proteins without any predicted TMDs . These proteins may encode soluble effectors for heme transport such as chaperones or sequestering proteins . In support of this concept , cellular iron is stored in ferritin , a cytosolic multi-subunit protein; cytoplasmic copper is delivered to membrane bound P-type ATPases in the secretory pathway by the copper chaperone Atox1 [37] , [38] . We propose that a similar network for trafficking intracellular heme and maintaining homeostasis is likely to exist in C . elegans and most metazoa [3] . Interestingly , HRG-4 , MRP-5 , and F22B5 . 4 were the only positive candidates identified in both the heme-sensor and GaPP functional RNAi screens . RNAi studies have implicated HRG-4 as a heme transporter in the C . elegans intestine [12] , while the function of MRP-5 and the protein encoded by F22B5 . 4 are currently unknown . We speculate that MRP-5 , a member of a family of membrane effluxers [39] , may export heme from the intestinal cells to extra-intestinal cells . These results are consistent , in part , with the ubiquitous expression of mrp-5::gfp in worm tissues , and with the RNAi studies which show that mrp-5 depletion results in accumulation of ZnMP in the worm intestine and resistance to GaPP toxicity . Unlike HRG-4 and MRP-5 which are transporters with multiple TMD , F22B5 . 4 encodes a predicted Type II membrane protein with a single TMD . Although our results clearly implicate a role for F22B5 . 4 as an essential component of heme homeostasis in C . elegans , it is unclear how this protein may function in heme homeostasis . Excitingly , microarray and RNAi studies identified F22B5 . 4 as a gene that is highly upregulated by the hypoxia-inducible factor ( HIF ) transcription complex , a master regulator of hypoxia response [40]–[42] . HIF is regulated by degradation through hydroxylation of proline residues , a process which requires the presence of oxygen , 2-oxoglutarate , and iron [43] . Given the dependence of C . elegans on heme for oxygen binding and sensing [44]–[46] and as a nutritional source of iron [11] , it is conceivable that F22B5 . 4 may play an important role in coordinating heme transport and availability with oxygen metabolism . In the current study we have identified a novel catalog of genes that are responsive to heme in C . elegans . Although it is unclear mechanistically how worms respond to heme at the mRNA level , a thorough study to identify the cis regulatory elements and the corresponding trans acting factors will significantly accelerate our understanding of how C . elegans adapts to environmental and nutritional changes . Using the facile and genetically tractable C . elegans model system , the RNAi screen with the hrg mini-library can be easily adapted for whole genome screens to identify regulatory pathways which govern how metazoans sense and respond to heme at an organismal level .
C . elegans wild-type N2 strain worms were grown either in an axenic liquid mCeHR-2 medium [47] or on NGM agar plates spotted with E . coli OP50 or HT115 ( DE3 ) strains [48] . Synchronized , L1 larvae were obtained by bleaching P0 gravid worms grown in mCeHR-2 medium supplemented with hemin chloride [11] . Hemin chloride and gallium protoporphyrin IX were purchased from Frontier Scientific , Inc ( Logan , UT ) . Plasmids for cloning and injecting into worms were part of the Fire Vector Kit ( Addgene , Cambridge , MA ) . Primers designed to PCR amplify worm open reading frames were based on Wormbase predictions and ordered from IDT ( Coralville , IA ) . The PCR products were TA cloned into the L4440 plasmid . Equal numbers of F1 larvae in the L1 stage were inoculated in mCeHR-2 medium with 4 , 20 , or 500 µM hemin chloride and grown with gentle shaking at 20°C . Synchronized , F2 larvae in the L1 stage were obtained by hatching the eggs obtained from F1 gravid adults in M9 buffer containing 4 , 20 , or 500 µM hemin . Equal numbers of F2 larvae in the L1 stage were inoculated in mCeHR-2 medium supplemented with 4 , 20 , or 500 µM hemin . The F2 worms were allowed to develop to the late L4 stage , harvested , flash frozen in liquid nitrogen , and stored at −80°C . Frozen worm pellets were ground into a fine powder , and total RNA was extracted using Trizol ( Invitrogen , Carlsbad , CA ) . RNA thus obtained was subjected to RNase-free DNase treatment for 1 h at 37°C and purified using the RNeasy kit ( Qiagen , Germantown , MD ) . Total RNA from three biological replicates was used to make cDNA , which was then hybridized to C . elegans Whole Genome Arrays ( Affymetrix , Santa Clara , CA ) . First strand cDNA was synthesized using 2 µg of total RNA using a Superscript II First Strand cDNA synthesis kit ( Invitrogen ) . For quantitative real-time PCR ( qRT-PCR ) , primers spanning at least one intron were designed using Primer Express ( Applied Biosystems ) and Beacon designer 4 ( Premier Biosoft ) programs . PCR was performed using the iCycler iQ Real-time PCR Detection System ( BioRad ) with 0 . 12 U/µl Taq DNA polymerase , 40 nM fluorescein ( Invitrogen ) , and SYBR Green I Nucleic Acid Gel Stain ( Invitrogen ) diluted 1:10 . The PCR amplification was run for 40 cycles . The PCR products were between 150 and 200 bp in length . Quality of the PCR products was determined by dissociation curve analysis and gel electrophoresis . Each experiment was performed in triplicate . Average CT values were used for 2−ΔΔCt calculations of relative fold changes in gene expression [49] . Expression data were normalized and analyzed using MAS 5 . 0 suite software ( Affymetrix ) . Data from worms grown in mCeHR-2 medium with 4 and 500 µM hemin were compared to data from worms grown in medium containing 20 µM hemin ( baseline samples ) . Microarray data were verified with the Robust Multichip Average Method ( RMA , R package ) . Quantile normalization and background corrections were performed using perfect match probe intensities . Using an initial cut-off of ≥1 . 2-fold change in mRNA expression for RMA and a ≥1 . 6-fold change for MAS 5 . 0 resulted in the identification of 370 genes . Increasing the stringency to ≥1 . 6-fold change for both RMA and MAS 5 . 0 reduced the number of genes identified as heme responsive to 288 genes . To identify putative human orthologs , worm protein sequences were used to query human genome databases at NCBI by reciprocal BLAST analysis with an E-value cut-off ≥10−4 . Sequences for each of these 288 genes were obtained from WormBase and further analyzed for topology ( TMHMM 2 . 0 , SOSUI ) , motifs ( ELM , BLOCKS , Pfam ) , and pathway classification ( GO and KEGG ) . The Ahringer and Vidal feeding libraries were replicated to individual 96-well plates [50] , [51] . Thirty-four clones in the initial list of 370 hrgs were absent from both libraries . To complete the hrg mini-library , we PCR amplified the missing genes from N2 worm genomic DNA and cloned the PCR fragments by TA cloning into the RNAi feeding vector pL4440 . Only 19 of the 34 RNAi clones were in the final list of 288 hrgs . DNA for all 288 hrgs was sequenced to confirm authenticity . NGM agar plates containing IPTG , carbenicillin , and tetracycline were seeded with HT115 ( DE3 ) bacteria expressing double-stranded RNA ( dsRNA ) against each clone in the hrg mini-library . Duplicate bacterial cultures of each clone had been grown for 5 . 5 h in LB containing carbenicillin and tetracycline and 5 µM or 25 µM heme . Plates were seeded with a lawn of bacteria and dsRNA induction occurred for ≈20 h at room temperature . Subsequently , forty L1 larvae from gravid IQ6011 worms which had been grown in liquid media supplemented with 10 µM heme were added to each well of the 12-well plates . Each 12-well plate had 10 wells seeded with experimental clones and one well seeded with each of the control clones – vector and hrg-4 . The plates were incubated at 15°C overnight and then incubated at 20°C for three additional days . The GFP levels in gravid adults were observed visually using a Leica Microsystems MZ16FA stereoscope . The intensity and pattern of GFP in gravid worms feeding on bacteria producing dsRNA against each hrg was compared to the intensity and pattern of GFP in same-stage worms feeding on bacteria transformed with the empty vector . Worms that displayed altered GFP in both replicates were designated as potential modulators . Potential modulators were screened in a strain that produced GFP under the control of a promoter that was not responsive to heme ( vha-6::gfp ) . Any clone that altered GFP levels in the vha-6::gfp strain worms was removed from the list of modulators , since the change in GFP was not in response to heme . A COPAS BioSort worm sorter ( Union Biometrica , Holliston , MA ) was used to measure GFP levels in live worms . Plates , bacteria , and worms were prepared and treated as described in the previous section . After 84 h on RNAi plates , P0 gravid and F1 L1-stage worms were washed from each well with 600 µL of M9 buffer containing 0 . 01% Tween-20 , transferred to a 1 . 5-mL microcentrifuge tube , and allowed to settle for 5 min . The supernatant was removed and discarded . Each worm pellet was transferred to an individual well of a 96-well plate . Duplicate samples were transferred to successive wells in the 96-well plate and were separated from other samples by an empty well , which served to flush the flow cell where worms are analyzed and prevent contamination of subsequent samples . The contents of each well were washed , aspirated , and analyzed by a COPAS BioSort worm sorter . The GFP gain was set to 2 , and the GFP PMT setting was 400 . Using highly synchronized worms in the gravid stage , we had previously defined the gate settings in order to ensure that the data obtained from P0 gravid animals would be easily and quickly separated from the data obtained from worms in other developmental stages . Text file data was imported into Microsoft Excel and sorted based on the gating parameters recorded in the “Status Select” column . The worm sorter records a fluorescence profile of each worm in the form of a curve , which reflects the intensity of GFP from the mouth to the tail . The “Green” column recorded the GFP value of the area under the curve , reduced by a factor of 40 , 000 . The background levels of GFP were subtracted from all values used to generate Figure 7A . The background level of GFP was equal to the GFP levels in IQ6011 worms feeding on HT115 ( DE3 ) bacteria transformed with the gfp RNAi vector . The COPAS BioSort detects very low levels of GFP in these worms . The mean of all values for each sample was determined , and the average of each duplicate was calculated . This mean was normalized to the average value for the GFP obtained from the vector-only sample , and reported in arbitrary units ± SEM for each clone analyzed . Synchronized , F1 wild-type worms in the L1 larval stage were obtained from P0 worms grown in mCeHR-2 containing 1 . 5 µM hemin . Equal numbers of these F1 worms were placed on NGM agar plates containing 2 mM IPTG , 50 µg/mL carbenicillin , 12 µg/mL IPTG and plated with a lawn of HT115 ( DE3 ) RNAi feeding bacteria harboring the respective L4440 plasmid that had been grown in LB broth with carbenicillin and tetracycline [12] . Worms were fed on the RNAi bacteria for ≈60 h and allowed to develop to the late L4 stage . At this point , worms were transferred to fresh RNAi plates containing 1 . 5 µM GaPP . Worms developed to the gravid stage and laid eggs . After 24 h of egg-laying , the P0 worms ( all in the gravid stage ) were discarded in order to prevent additional eggs from being laid . On day 5 , both the total number of surviving larvae and the number of unhatched eggs were counted . P values for statistical significance were calculated by using a one-way ANOVA with Student–Newman–Keuls multiple comparisons test by using GraphPad InStat v . 3 . 06 ( GraphPad , San Diego , CA ) . Equal numbers of synchronized N2 L1 larvae obtained from P0 worms grown in mCeHR-2 plus 2 µM hemin were exposed to the RNAi bacteria on NGM plates containing 2 mM IPTG for 72 h . This was followed by exposure to 5 µM ZnMP plus 1 . 5 µM hemin chloride for 16 h in mCeHR-2 medium . ZnMP fluorescence intensity was measured as described previously [12] . GFP reporter fusion constructs were created using the Gateway cloning system ( Invitrogen , Frederick , MD ) . The promoter of interest , gfp gene , and the 3′ untranslated region of the unc-54 gene were cloned by recombination into the entry vectors pDONR P4-P1R , pDONR 221 , and pDONR P2R-P3 , respectively , using the Gateway BP Clonase kit . Sequence verified entry clones were then recombined into a destination vector pDEST R4-R3 using the Gateway LR Clonase II plus enzyme kit to produce the final recombinant plasmid . For microparticle bombardment , ≈5×106 unc-119 ( ed3 ) gravid worms were co-bombarded with 10 µg of Gateway reporter construct and 5 µg of unc-119 rescue plasmid ( pDM016B ) using the PDS-1000 particle delivery system ( Bio-Rad , Hercules , CA ) . Worms were washed from bombardment plates and transferred to plates seeded with a lawn of E . coli strain JM109 . After two-weeks at 25°C , multiple wild-type F2 worms were screened for gene integration either by PCR or transgene expression . Individual transgenic lines were isolated and transferred to axenic liquid mCeHR-2 medium supplemented with antibiotics . After two weeks of serial passages , worms were bleached and maintained as transgenic strains in axenic liquid mCeHR-2 medium . The microarray data was submitted to GEO on Aug 6 , 2007 . The GEO accession number is GSE8696 and available at http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE8696 . | Heme is an iron-containing cofactor for proteins involved in many critical cellular processes . However , free heme is toxic to cells , suggesting that heme synthesis , acquisition , and transport is highly regulated . Efforts to understand heme trafficking in multicellular organisms have failed primarily due to the inability to separate the processes of endogenous heme synthesis from heme uptake and transport . Caenorhabditis elegans is unique among model organisms because it cannot synthesize heme but instead eats environmental heme to grow and develop normally . Thus , worms are an ideal genetic animal model to study heme homeostasis . This work identifies a novel list of 288 heme-responsive genes ( hrgs ) in C . elegans and a number of related genes in humans and medically relevant parasites . Knocking down the function of each of these hrgs reveals roles for several in heme uptake , transport , and detection within the organism . Our study provides insights into metazoan regulation of organismal heme homeostasis . The identification of parasite-specific hrg homologs may permit the selective design and screening of drugs that specifically target heme uptake pathways in parasites without affecting the host . Thus , this work has therapeutic implications for the treatment of human iron deficiency , one of the top ten mortality factors world-wide . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/comparative",
"genomics",
"genetics",
"and",
"genomics/gene",
"expression",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"biochemistry/chemical",
"biology",
"of",
"the",
"cell",
"cell",
"biology/membranes",
"and",
"sorting",
"infectious",
"diseases/helminth",
"infections",
"hematology/anemias",
"nutrition/malnutrition",
"infectious",
"diseases/protozoal",
"infections",
"nutrition/deficiencies",
"cell",
"biology/chemical",
"biology",
"of",
"the",
"cell",
"cell",
"biology/gene",
"expression",
"chemical",
"biology/chemical",
"biology",
"of",
"the",
"cell"
]
| 2010 | Genome-Wide Analysis Reveals Novel Genes Essential for Heme Homeostasis in Caenorhabditis elegans |
Transcriptional/translational feedback loops drive daily cycles of expression in clock genes and clock-controlled genes , which ultimately underlie many of the overt circadian rhythms manifested by organisms . Moreover , phosphorylation of clock proteins plays crucial roles in the temporal regulation of clock protein activity , stability and subcellular localization . dCLOCK ( dCLK ) , the master transcription factor driving cyclical gene expression and the rate-limiting component in the Drosophila circadian clock , undergoes daily changes in phosphorylation . However , the physiological role of dCLK phosphorylation is not clear . Using a Drosophila tissue culture system , we identified multiple phosphorylation sites on dCLK . Expression of a mutated version of dCLK where all the mapped phospho-sites were switched to alanine ( dCLK-15A ) rescues the arrythmicity of Clkout flies , yet with an approximately 1 . 5 hr shorter period . The dCLK-15A protein attains substantially higher levels in flies compared to the control situation , and also appears to have enhanced transcriptional activity , consistent with the observed higher peak values and amplitudes in the mRNA rhythms of several core clock genes . Surprisingly , the clock-controlled daily activity rhythm in dCLK-15A expressing flies does not synchronize properly to daily temperature cycles , although there is no defect in aligning to light/dark cycles . Our findings suggest a novel role for clock protein phosphorylation in governing the relative strengths of entraining modalities by adjusting the dynamics of circadian gene expression .
A large variety of life forms manifest circadian ( ≅24 hr ) rhythms in behavior and physiology that are driven by endogenous cellular clocks or pacemakers [1] , [2] . Perhaps the most biologically relevant property of circadian clocks is that they can be synchronized ( entrained ) to local time by external time cues , a feature that endows organisms with the ability to anticipate environmental changes and hence perform activities at optimal times during the day . The main environmental synchronizing agents of circadian clocks in nature are the daily cycles in light/dark and ambient temperature . In general , photic cues are the most potent synchronizing agent for organisms , whereas thermal entrainment is less powerful [3] , [4] . Work in the last 20 years using a variety of model organisms has revealed that the molecular logic underlying circadian clock mechanisms is highly conserved [2] . Circadian clocks are based on intracellular mechanisms that involve a core transcriptional translational feedback loop ( TTFL ) composed of central clock proteins that drive daily oscillations in their own gene expression as well as downstream clock-controlled genes ( ccgs ) . Daily oscillations in the transcript levels of ccgs ultimately drive many of the rhythmic behaviors and physiologies manifested by organisms . The rate-limiting component of the main TTFL in Drosophila is the basic-helix-loop-helix ( bHLH ) PAS domain containing transcription factor termed dCLOCK ( Drosophila CLOCK; dCLK ) [5] , which forms a heterodimer with CYCLE ( CYC ) , another bHLH-PAS containing clock transcription factor [6] . The dCLK-CYC heterodimer binds to E box DNA elements inducing the expression of the clock genes period ( per ) and timeless ( tim ) , in addition to other clock genes and ccgs . Subsequently , the PER and TIM proteins interact in the cytoplasm and after a time-delay translocate to the nucleus where they function with other factors to inhibit the transcriptional activity of dCLK-CYC . Eventually , the levels of PER and TIM decline in the nucleus , facilitating another round of dCLK-CYC-mediated transcription . In a “secondary” stabilizing TTFL , the dCLK-CYC heterodimer induces the expression of PAR domain protein 1ε ( pdp1ε ) and vrille ( vri ) , whose protein products ( i . e . , PDP1ε and VRI ) in turn activate and repress the expression of dClk , respectively , leading to daily cycles in dClk mRNA levels [7] , [8] . Mammalian clocks also use a CLOCK-based transcription factor in their main TTFL , which involves a heterodimer comprised of mCLOCK ( mammalian CLOCK; mCLK ) and BMAL1 ( homolog of CYC ) that governs rhythmic expression of the negative regulators Per1-3 , in addition to other clock genes and ccgs [9] . Although TTFLs constitute a major molecular framework for the oscillatory behavior of cellular clocks , posttranslational modifications of clock proteins are central to maintain proper timekeeping functions by regulating clock protein stability , sub-cellular localization and activity [10]–[14] . A well-studied example of clock protein phosphorylation is the progressive phosphorylation of PER , which has a critical role in setting the pace of the clock and controlling temporal changes in dCLK-CYC-mediated transcription by regulating PER stability , timing of nuclear entry and how long it persists in the nucleus [15]–[22] . Newly synthesized PER is present as non-to-hypo-phosphorylated isoforms in the late day/early night and undergoes progressive increases in the extent of phosphorylation , culminating in the appearance of mostly or exclusively hyper-phosphorylated isoforms in the late night/early day that are recognized for rapid degradation by the 26S proteasome [10] . Numerous PER-relevant kinases have been identified , with DOUBLETIME [DBT; homolog of vertebrate casein kinase Iδ/ε ( CKIδ/ε ) ] [21] , [23] operating as the major kinase regulating temporal changes in the stability of PER . Other kinases include SHAGGY [SGG; homolog of vertebrate glycogen synthase kinase 3β ( GSK3β ) ] [24] , casein kinase 2 ( CK2 ) [25] , [26] and NEMO [15] , [27] . dCLK also undergoes circadian changes in phosphorylation state , but in a manner different from that of PER [28] , [29] . dCLK is present in a mostly intermediate phosphorylated state throughout the day , converting to largely hyper-phosphorylated isoforms in the late night/early day . DBT stably interacts with PER throughout most of its daily life-cycle and this association likely facilitates the ability of DBT to regulate dCLK [28]–[31] . Although the role ( s ) of dCLK phosphorylation is not clear it appears that hyper-phosphorylated isoforms have decreased stability and possibly reduced transcriptional activity [28]–[30] . In addition to DBT , several kinases such as protein kinase A ( PKA ) , CaMKII , MAPK , and NMO have been implicated in regulating the activity and/or levels of dCLK [27] , [32] . More recently CK2 was reported to act directly on dCLK , stabilizing it while reducing its activity [33] . The mammalian CLOCK protein also manifests circadian oscillations in phosphorylation in vivo [34] , [35] , which is triggered by hetero-dimeric complex formation with BMAL1 [34] , [35] . Mass spectrometric analysis of purified mCLK from the mouse liver identified Ser38 , Ser42 , and Ser427 as sites phosphorylated in vivo [36] . Ser38 and Ser42 are located in the bHLH region and phosphorylation of those residues down-regulates transcriptional activity of mCLK via decreasing binding activity to E box element [36] . Phosphorylation of Ser427 is reported as being dependent on GSK-3β activity and relevant for degradation of mCLK [37] . PKG and PKC have been implicated as mCLK kinases regulating phase resetting [38] , [39] . Despite these advances using several animal model systems , it is still unclear how CLOCK phosphorylation impacts the function of circadian timing systems at the organismal level . In this study , we used a simplified Drosophila S2 cell culture system in combination with mass spectrometry to map phosphorylation sites on dCLK . Our results indicate that dCLK is highly phosphorylated ( at least 14 phospho-sites ) . In S2 cells , mutated versions of dCLK where all the mapped Ser/Thr sites were switched to Ala ( herein referred to as dCLK-15A ) manifested increased E box dependent transcriptional activity without affecting interactions with other core clock partners such as CYC and PER . In flies , dCLK-15A protein is exclusively hypo-phosphorylated suggesting that we identified , at the very least , a major portion of the total phosphorylation sites found on dCLK in flies . Expression of dCLK-15A rescues the arrhythmicity of Clkout flies yet with an approximately 1 . 5 hr shorter period . Consistent with a role in regulating protein stability , the levels of dCLK15A are substantially higher compared to the control situation , which along with increases in transcriptional activity likely explains the faster pace of the clock . The daily peak levels in per/tim mRNA and protein reached higher values in dCLK-15A expressing flies , further supporting the notion that dCLK levels are normally rate-limiting in the clock mechanism . Surprisingly , the clock-controlled daily activity rhythm in dCLK-15A mutant flies fail to maintain synchrony with daily temperature cycles , although there is no defect in aligning to light/dark cycles . Together , our findings indicate that in animal systems , the post-translational modification of a master circadian transcription factor plays a critical role in setting the pace of the clock and regulating circadian entrainment .
As an initial attempt to better understand the role ( s ) of dCLK phosphorylation we sought to map phosphorylation sites using recombinant protein production in cultured Drosophila S2 cells . This simplified cell culture system was successfully used to identify physiologically relevant phosphorylation sites on Drosophila PER [15] , [16] , [18] , [40] . Prior work showed that production of recombinant dCLK in S2 cells leads to significant shifts in electrophoretic mobility that are due to phosphorylation [29] . Thus , we established S2 cell lines stably expressing HA-dClk-V5 under the inducible metallothionein promoter ( pMT ) . Total cell lysates were subjected to immunoprecipitation with anti-V5 antibody , followed by multi-protease digestion , titansphere nanocolumn phosphopeptide enrichment , and tandem mass spectrometry , as previously described [16] , [41] . We identified 14 phosphorylation sites on dCLK , all of which are at Ser residues , with the possible exception of Tyr607 ( Table 1 ) . Many of the identified phosphorylation sites are in the C-terminal half of dCLK , which contains several Q-rich regions that might function in transcriptional activation ( Figure 1A ) . Phosphorylation was also detected at sites close to the N- and C-terminus of the dCLK protein . Interestingly , no phosphorylation sites were found in any of the known functional domains of dCLK; e . g . bHLH , PAS domains and Q-rich regions ( Figure 1A and Table 1 ) . In preliminary studies we individually mutated each of the identified phosphorylated Ser residues to Ala residues but did not see major effects on dCLK electrophoretic mobility , except for the S859A mutant version of dCLK , which manifested slightly faster electrophoretic mobility . ( Figure S1A and B ) . The transcriptional activities of most single site mutants were somewhat increased ( ≤2 fold ) , except for the S924A mutant version of dCLK , which manifested a slight but reproducible decrease ( Figure S1C ) . Overall , our initial studies in S2 cells were not able to identify whether certain individual phospho-sites are particularly significant in regulating dCLK metabolism or activity . While ongoing work is aimed at better understanding the roles of individual phospho-sites , in this study we focused on more global aspects of dCLK phosphorylation by generating a mutant version wherein all the Ser phospho-acceptor sites identified by mass spectrometry were switched to Ala . Since the mass spectrometry data did not unambiguously identify which Ser among amino acids 209–211 is phosphorylated , we switched all 3 Ser to Ala . In addition , although Tyr 607 or Ser 611 is phosphorylated , to focus on Ser phosphorylation , we only mutated Ser 611 to an Ala . By using site-directed mutagenesis , we serially mutated the aforementioned 16 Ser to Ala ( dCLK-16A ) . The electrophoretic mobility of dCLK-16A is indistinguishable from that of λ-phosphatase treated wild-type dCLK and was not altered by λ-phosphatase treatment ( Figure 1B ) , indicating that we mapped most , if not all , the sites on dCLK phosphorylated by endogenous kinases in S2 cells . dCLK-16A interacts with either CYC or PER proteins to a similar extent as that observed for the wild-type version ( dCLK-WT ) , demonstrating that dCLK-16A is not grossly misfolded ( Figure 1C ) . In addition , our findings suggest that the phosphorylated state of dCLK is not a major signal regulating interactions with core clock partners . Consistent with prior work , we observed increases in non/hypo-phosphorylated isoforms of dCLK when dPER is co-expressed ( Figure 1C , compare lane 1 and 6 ) [31] . With regards to transcriptional activity , dCLK-16A is more potent compared to dCLK-WT in stimulating E-box dependent transcription ( Figure 1D ) , while still maintaining its sensitivity to inhibition by dPER ( Figure 1E ) . Earlier findings showed that hyper-phosphorylated dCLK is less stable and that DBT might contribute to this instability , although the exact role of DBT is not clear [28]–[30] . We compared the stabilities of dCLK-16A and dCLK-WT under a variety of conditions , including overexpressing DBT , but did not detect a significant difference ( Figure S2A and B ) , suggesting we did not map one or more phosphorylation sites critical for regulating dCLK stability and/or the pathway for dCLK degradation in S2 cells is not identical to that in flies ( see below ) . Taken together , the results obtained using well-established S2 cell based assays indicate that dCLK-16A retains key clock-relevant biochemical functions and suggest that global phosphorylation of dCLK reduces its transactivation potential . To investigate whether the dCLK phosphorylation sites we identified play a physiological role in the Drosophila circadian timing system , we first evaluated the ability of a novel wildtype dClk transgene to rescue behavioral rhythms in the arrhythmic Clkout genetic background ( herein , termed as p{dClk-WT};Clkout ) . Clkout is a newly described arrhythmic null mutant that does not produce dCLK protein ( Mahesh et al . , submitted ) . The transgene was constructed with a 13 . 9 kb genomic fragment that contains the dClk gene , which we modified by introducing a V5 epitope tag at the C-terminus of the dClk open reading frame for enhanced protein surveillance . Flies were exposed to standard entraining conditions of 12 hr light∶12 hr dark cycles [LD; where zeitgeber time 0 ( ZT0 ) is defined as lights-on] at 25°C , followed by several days in constant dark conditions ( DD ) to measure free-running behavioral periods . In the behavioral analysis , p{dClk-WT};Clkout flies manifested robust locomotor activity rhythms with normal ∼24 hr periods ( Table 2 , Mahesh et al . , submitted ) , indicating that the circadian clock system functions properly in these flies . Next , we sought to generate transgenic flies harboring a dCLK-16A version of the dClk rescue transgene . However , because of technical difficulties in generating a version that also included replacing Ser5 with Ala , we made a dClk version wherein the other 15 Ser residues were switched to Ala , termed dClk-15A . In S2 cells , dCLK-15A behaves similar to dCLK-16A , including no observable effect of phosphatase treatment on electrophoretic mobility and enhanced E-box dependent transcriptional activity ( Figure S3 ) . Although phosphorylation of Ser5 might affect dCLK function in a manner that is not revealed in the S2 cell based assays we used , the CLK-15A protein contains the majority of phosphorylation sites and should address if global phosphorylation of dCLK plays an important role in the circadian timing system . Two independent lines of transgenic flies harboring the dClk-15A transgene were obtained and circadian behavior was monitored in the Clkout genetic background ( referred to as p{dClk-15A} , 2M; Clkout and p{dClk-15A} , 6M; Clkout ) . In sharp contrast to flies harboring the control version of dClk , the two independent lines of p{dClk-15A};Clkout flies manifested generally weaker behavioral rhythms that are approximately 1 . 5 hr shorter compared to their wild-type counterparts ( Table 2 ) . Under standard conditions of LD at 25°C , D . melanogaster exhibits a bimodal distribution of activity with a “morning” and “evening” bout of activity centered around ZT0 and ZT12 , respectively . Although p{dClk-15A};Clkout flies manifest the typical bimodal distribution of locomotor activity , the onset of both the morning and evening bouts of activity were earlier ( Figure 2 , compare panels B and C to A ) , consistent with the shorter free-running periods . The Clkout flies only showed a “startle” response to the lights-on transition but no rhythmic behavior ( Figure 2E ) . In constant dark conditions , the downswing in evening activity is clearly earlier in p{dClk-15A};Clkout flies , in agreement with the shorter free-running period ( Table 2 , Figure 2 and S4 ) . We also examined the locomotor behavior of flies harboring the dClk-WT transgene in a wild-type genetic background , resulting in flies with four copies of the dClk gene ( herein referred to as p{dClk-WT};+/+ ) . The circadian period was shortened to approximately 23 hr ( Table 2 and Figure 2D ) , which is well correlated with previous reports demonstrating that increasing the copy number of dClk shortens the circadian period of behavioral rhythms [42] , [43] . A hallmark property of circadian rhythms is that the period length is very constant over a wide range of physiologically relevant temperatures , termed temperature compensation [44] . To investigate whether phosphorylation of dCLK might have a role in temperature compensation , we analyzed behavioral rhythms at three standard temperatures ( i . e . , 18° , 25° and 29°C ) . Although we noted a decrease in rhythmicity for dClk-15A;Clkout flies at 29°C , the periods were quite similar over the temperature range tested ( Table 2 ) , suggesting that global phosphorylation of dCLK does not play a major role in temperature compensation . We examined the temporal profiles of dCLK protein by analyzing head extracts prepared from p{dClk-WT};Clkout and p{dClk-15A};Clkout flies in LD conditions ( Figure 3A and C ) . dCLK-WT protein undergoes daily changes in phosphorylation that are consistent with earlier results probing endogenously produced dCLK; namely , hypo- to medium- phosphorylated isoforms present during the mid-day/early night ( e . g . , ZT 8 to ZT 16 ) and mostly hyper-phosphorylated isoforms present in the late night/early day ( e . g . , ZT20 to ZT4 ) ( Figure 3A ) [28] , [29] . However , the mobility of dCLK-15A was similar throughout a daily cycle ( Figure 3A ) , and co-migrated with λ phosphatase treated dCLK-WT ( Figure 3B ) . Thus , similar to results in S2 cells , dCLK-15A exhibits little to no phosphorylation in vivo , suggesting that the phospho-sites we identified by mass spectrometry comprise , at the very least , a major portion of the total phosphorylation sites found on dCLK in flies ( it is also possible that one or more of the phospho-sites we mutated are required for phosphorylation at other sites , but this would still result in a mainly hypo-phosphorylated dCLK protein ) . Intriguingly , the levels of dCLK-15A were substantially higher compared to dCLK-WT throughout a daily cycle . Quantification of immunoblots indicated that the average daily levels of dCLK-15A are about 2 . 5 times more than those of dCLK-WT ( Figure 3C ) . To examine whether the high levels of dCLK-15A proteins results from elevated mRNA abundance , we measured dClk transcript levels . As reported previously , although the overall daily abundance of dCLK-WT protein is essentially constant throughout a daily cycle , dClk-WT mRNA levels oscillate with peak amounts attained during the late night-to-early day and reaching trough values around ZT12 [45] , [46] . The daily oscillation in dClk-15A mRNA abundance is similar to the wild-type situation and even seemed to have lower peak levels ( Figure 3D ) . These results indicate that in general global phosphorylation of dCLK decreases its stability in vivo , consistent with prior findings using S2 cells [28] , [29] . To further examine the status of the clockworks , we measured the daily profiles in per and tim transcripts and protein levels . Both per and tim mRNA levels in p{dClk-15A};Clkout flies were reproducibly higher , especially during the daily upswing that occurs between ZT4 – 12 ( Figure 4A and B ) . These result further support the notion that dCLK levels are normally rate-limiting for circadian transcription and suggest that despite the increased abundance of dCLK-15A there is sufficient PER to engage in normal repression of dCLK-15A/CYC activity . Indeed , PER protein levels were reproducibly higher in p{dClk-15A};Clkout flies ( Figure 4C and D ) , consistent with the increased transcript levels . In p{dClk-15A};Clkout flies , TIM protein levels were slightly but reproducibly increased ( Figure 4E and F ) . The increased daily upswing in tim mRNA levels in p{dClk-15A};Clkout flies might have a smaller effect on overall TIM protein levels because light induces the rapid degradation of TIM [47] , possibly limiting the ability of TIM to accumulate during the early night prior to the start of transcriptional feedback repression . Taken together , we show that the stability of dCLK in flies is strongly increased by blocking phosphorylation at one or more sites . Moreover , augmenting the total abundance of dCLK accelerates the daily accumulation of per/tim transcripts and increases their peak levels , indicative of higher overall dCLK-CYC-mediated transcription . In addition , increased in vivo transcriptional activity of dCLK-15A may also contribute to higher dCLK-CYC-mediated transcription , as is the case in S2 cells ( Figure 1D ) . These results demonstrate that dCLK phosphorylation plays a key role in setting the amplitudes of the per mRNA and protein rhythms , molecular oscillations that are central to the primary TTFL and circadian speed control in Drosophila [16] , [48] . Besides light-dark cycles , daily changes in temperature can also synchronize ( entrain ) circadian rhythms in a wide variety or organisms [49] . Prior work showed that D . melanogaster can entrain to daily cycles of alternating 12 hr ‘warm’/12 hr ‘cold’ cycles that differ by as little as 2–3°C [50]–[52] . To determine if flies expressing dCLK-15A have a defect in entraining to temperature cycle , flies were kept in constant darkness , exposed to 12 hr∶12 hr temperature cycles of 24°C∶29°C ( TC ) and locomotor activity rhythms analyzed ( Table 3 and Figure 5 ) . The daily distribution of activity in p{dClk-15A};Clkout flies is strikingly different compared to the wild-type control . As previously observed for wildtype strains of D . melanogaster entrained to daily temperature cycles [50]–[52] , p{dClk-WT};Clkout flies exhibit the classic “anticipatory” rise in activity just prior to the low-to-high and high-to-low temperature transitions , similar to what is observed in light-dark cycles around ZT0 and ZT12 ( Figure 5 and S5; there is a “startle” response at the transition from low-to-high temperature that is also observed in Clkout flies , analogous to the transient burst in activity at lights-on in a LD cycle ) . In sharp contrast , during the beginning of the temperature entrainment regime although p{dClk-15A};Clkout flies also manifest two activity peaks , they occur much earlier at around the middle of the warm- and cryo-phases ( Figure 5B ) . Interestingly , while the timing of the “startle” response at the transition from low-to-high temperature remained constant in p{dClk-15A};Clkout flies , the timing of the major activity peak occurring during the mid-warm phase appeared to progressively advance on subsequent days in TC ( Figure 5B ) . Analysis of individual activity records also confirmed this trend ( Figure 5E ) . The abnormal behavioral pattern under temperature cycles for p{dClk-15A};Clkout flies was also observed when flies were exposed to a temperature cycle after first treating them with constant light for 6 days to abolish the circadian timing system ( Figure S6 ) . Thus , the defective entrainment of p{dClk-15A};Clkout flies to TC is not dependent on the status of the clock at the time that the temperature entrainment was initiated . Furthermore , although the main activity bout in p{dClk-15A};Clkout flies advanced on subsequent days during TC , the rate of advancement was clearly greater during free-running conditions following TC ( Figure S5 ) , suggesting partial entrainment during TC . Following entrainment to TC , the free-running period of dCLK-15A producing flies is about 1 . 5 hr shorter compared to the wild type dCLK-WT control ( Table 3 and Figure S5 ) . The faster running clock in p{dClk-15A};Clkout flies during free-running conditions after exposure to TC is consistent with results obtained following entrainment to LD ( Table 2 ) . Thus , it appears that when exposed to daily temperature cycles p{dClk-15A};Clkout flies can adopt some alignment with the entraining conditions , albeit without a normal phase relationship , but that this entrainment is weak and the flies partially free-run at their shorter endogenous periods , leading to progressive advances in their behavioral rhythm relative to the 24 hr entraining regime . Although not as dramatic , the timing of the warm-phase activity bout in p{dClk-WT};+/+ flies also advanced during TC ( Figure 5F ) , whereas this was not the case for p{dClk-WT};Clkout flies ( Figure 5D ) . In addition , the clock in p{dClk-WT};+/+ runs about 1 hr faster than the control situation , strongly suggesting that augmenting dCLK levels ( Figure S7 ) impairs the ability of the circadian timing system to entrain to daily temperature cycles . Temperature cycles can entrain behavioral rhythms in Drosophila exposed to constant light ( LL ) despite the fact that LL normally abolishes circadian rhythms [51] , [53] , [54] . Intriguingly , constant light exposure rescues the ability of the p{dClk-15A};Clkout flies to maintain a more stable 24-hr phase relationship with the temperature cycle ( Figure 5 , panels G–J ) , further supporting the notion that entrainment to temperature but not light is specifically impaired in these flies . Taken together , these data suggest that in the absence of light , the dCLK phosphorylation program is required for the proper entrainment of behavioral rhythms to daily temperature cycles and reveal an unanticipated role for a central clock transcription factor in modality-specific entrainment . In p{dClk-WT};Clkout flies , hypo/intermediate-phosphorylated dCLK isoforms are present throughout the thermo phase in TC ( Figure 6A , lane 2 and 3 ) , while hyper-phosphorylated dCLK isoforms are only observed during the latter half of the cryo phase ( Figure 6A , lane 5 and 6 ) . This temporal pattern in dCLK phosphorylation is similar to that observed in LD cycles and is consistent with prior work showing that the circadian clock mechanism in Drosophila can be synchronized by daily temperature cycles [50] , [51] . As expected and similar to results using LD cycles , dCLK-15A attains higher overall daily levels and does not exhibit significant phosphorylation in p{dClk-15A};Clkout flies exposed to temperature cycles ( Figure 6A ) . Since p{dClk-15A};Clkout flies display altered entrainment to TC cycles that becomes progressively more abnormal with prolonged duration , we tested whether the molecular clock might also exhibit a more defective status with increasing time by measuring the levels of the tim mRNA at both early ( e . g . , day 3 ) and later ( e . g . , day 6 ) days of exposure to TC . We chose tim mRNA levels as a surrogate marker for clock dynamics because it normally has a robust high amplitude rhythm ( Figure 4A , B; [55] ) , facilitating measuring changes in molecular oscillations over the course of several days . Although the daily average levels in tim mRNA were higher in p{dClk-15A};Clkout flies on day three of TC compared to the wild-type situation ( Figure 6B ) , consistent with findings in LD ( Figure 4B ) , both genotypes showed similar and robust cycling patterns . However , by day six of TC , the tim mRNA oscillation pattern in p{dClk-15A};Clkout flies became significantly different from that observed for p{dClk-WT};Clkout flies ( Figure 6C ) . Most notably , while tim mRNA cycling still manifested high-amplitude cycling in p{dClk-WT};Clkout flies on day six of TC , tim mRNA levels during the upswing phase were significantly higher in p{dClk-15A};Clkout flies , resulting in an abnormal cycling pattern . Although we did not establish a causal relationship between the observed loss in normal tim mRNA cycling and the defective behavioral entrainment in p{dClk-15A};Clkout flies during TC , the results clearly show that prolonged exposure to TC is not only associated with increasingly altered phasing of rhythms at the behavioral level ( Figure 5 ) but also at the molecular level . As with behavioral rhythms prior work showed that circadian molecular cycles can be synchronized to TC in the presence of constant light [51] , [53] . In agreement with the observation that constant light exposure enabled p{dClk-15A};Clkout flies to more robustly synchronize to temperature cycles ( Figure 5 ) , daily rhythms in the levels of tim mRNA for both genotypes were quite similar even after six days in constant light during TC ( Figure 6D and E ) , indicating the clock in p{dClk-15A};Clkout flies is functioning in a more wild-type manner under these conditions . Taken together , while this molecular analysis is of limited scope , it suggests that constant light exposure facilitates the ability of p{dClk-15A};Clkout flies to entrain to TC by enhancing normal clock function .
Phosphorylation of clock proteins plays diverse roles in circadian oscillatory mechanisms by regulating numerous aspects of clock protein metabolism/activity , including time-of-day dependent changes in stability , transcriptional activity and subcellular localization [10]–[12] . Although dCLK , the master transcription factor in the Drosophila circadian clock [43] , [56] , undergoes daily changes in phosphorylation , the physiological role of dCLK phosphorylation was not clear . As a means to address this issue , we first identified phosphorylation sites on dCLK purified from cultured Drosophila S2 cells ( Table 1 and Figure 1A ) . To examine the in vivo significance of dCLK phosphorylation , we generated transgenic flies expressing dCLK-15A wherein 15 serine residues that were identified as sites ( or possible sites ) of phosphorylation were switched to alanine , and examined circadian behavior in a Clkout genetic background . Our results indicate that global phosphorylation of dCLK is an important aspect of setting clock speed by regulating the daily levels and/or activity of dCLK . This is consistent with earlier work suggesting dCLK is the rate-limiting component in the central transcriptional/translational feedback loop ( TTFL ) in the Drosophila clock mechanism , and that increasing the levels of dCLK lead to shorter behavioral periods [5] , [42] , [43] . A surprising finding is that entrainment to daily temperature cycles but not light-dark cycles are highly dependent on dCLK phosphorylation . These results suggest a novel role for phosphorylation in circadian timing systems; namely , the effective strength of an entraining cue can be modulated by adjusting the dynamics of the TTFL via controlling the levels/activity of a master circadian transcription factor ( see below ) . In this study , we show that dCLK undergoes multi-site phosphorylation . Among the phospho-sites identified , seven serine residues are situated immediately N-terminal to a proline , indicating a major role for the CMGC group of kinases . Indeed , studies using cultured S2 cells suggested that dCLK is a potential target of several distinct CMGC kinases [32] . More recent work also identified the pro-directed kinase NEMO as a dCLK-relevant kinase [27] . Ongoing work is aimed at identifying the kinases responsible for targeting the different phospho-sites on dCLK . It should be noted that in this study we mapped phosphorylation sites on dCLK expressed in S2 cells , which when resolved by SDS-polyacrylamide gel electrophoresis is mainly observed as two major electrophoretic mobility bands corresponding to non/hypo-phosphorylated isoforms and an ‘intermediate’ more highly phosphorylated slower migrating species [29] . Although DBT is endogenously expressed in S2 cells , the addition of exogenous DBT and/or the inhibition of protein phosphatases leads to the detection of hyper-phosphorylated isoforms of dCLK in S2 cells [29] . Thus , it is likely that we did not identify all the phospho-sites on dCLK . However , we cannot rule out the possibility that there were minor levels of hyper-phosphorylated dCLK in our preparations that were above the detection limit for phospho-site mapping by mass spectrometry . Irrespective , the phospho-sites that we identified in S2 cultured cells make a clear contribution to the daily dCLK phosphorylation program in flies and contribute to the circadian timing system . Elimination of phosphorylation sites from dCLK ( dCLK-15A ) leads to significant increases in the overall daily levels of dCLK in flies , which is well correlated with previous reports in S2 cells showing that hyper-phosphorylated dCLK is sensitive to degradation [28] , [29] . In general , global phosphorylation appears to reduce the stabilities of clock proteins by generating one or more phospho-degrons that are recognized by E3 ubiquitin ligases , which ultimately leads to the accelerated degradation of the phosphorylated isoforms via the proteasome pathway [13] . The E3 ligase termed CTRIP appears to directly regulate the levels of dCLK ( and possibly PER ) , although the role of dCLK phosphorylation in this mechanism , if any , is not clear [57] . When assayed in S2 cells the stability of dCLK-16A was similar to that of dCLK-WT ( e . g . , Figure 1 and Figure S2 ) . Because differences in transcript levels cannot explain the significantly higher levels of dCLK-15A in flies compared to dCLK-WT ( Figure 3 ) , it is almost certain that dCLK-15A is a more stable protein in clock cells . Thus , it appears that S2 cells do not fully recapitulate the in vivo role of phosphorylation on dCLK degradation . If we did miss mapping some sites on hyper-phosphorylated dCLK that are critical for regulating stability it is possible that these sites can still be phosphorylated on dCLK-15A expressed in S2 cells but not in flies . For example , hyper-phosphorylation of dCLK might depend on prior phosphorylation at one or more of the 15 phospho-sites we identified , and this dependency might be more strict in flies compared to the S2 cell over-expression system . Hierarchical phosphorylation has been demonstrated for other clock proteins , such as Drosophila PER and mammalian CLK [15] , [37] , [40] . Future work will be required to determine if there are other phospho-sites besides those we identified that regulate dCLK stability in flies . Besides regulating the stability of core clock transcription factors , phosphorylation modulates trans-activation potential [36] , [37] , [58]–[61] . dCLK-15A expressed in S2 cells exhibited normal binding to CYC ( and PER ) but exhibits more potent transcriptional activity , at least in the context of a simple E-box driven expression ( Figure 1D ) . Consistent with this , the levels of dper and tim mRNAs in p{dClk-15A};Clkout flies are higher compared to the control situation ( Figure 4A , B ) . Of course , phosphorylation also affects the levels of dCLK-15A in flies , so at this stage it is not possible to determine how much the increased per/tim transcript levels are due to changes in the levels or activity of dCLK-15A . Nonetheless , our results strongly suggest that in wild-type flies the levels and/or activity of dCLK act in a rate-limiting fashion during the daily accumulation phase of per/tim transcripts and possibly other targets . In addition , the phospho-sites that we identified do not seem to be play a major determinant in feedback repression by PER and associated factors . Strong repression was observed in S2 cells for the dCLK-15A version ( Figure 1E ) and the normal daily downswing in per/tim levels occurred in p{dClk-15A};Clkout flies ( Figure 4A and B ) . However , it is possible that we missed some phospho-sites that more specifically regulate the transcriptional activity of dCLK . At the behavioral level , p{dClk-15A};Clkout flies exhibit short period rhythms , consistent with prior work showing that increasing the dosage of dClk quickens the pace of the clock [42] , [43] . In light-dark cycles , p{dClk-15A};Clkout flies maintain a stable phase relationship with the entraining environment , displaying the typical anticipatory bimodal activity pattern ( Figure 2 ) . Moreover , in a daily light-dark cycle the timing of the morning and especially evening peak of activity is shifted in flies with different endogenous periods , appearing earlier in fast clocks and later in slow clocks [52] . Indeed , the p{dClk-15A};Clkout flies follows this trend as the evening ( and morning ) bout of activity in LD is earlier compared to control flies ( Figure 2 ) . Together , these results indicate that although global phosphorylation of dCLK is an important determinant in setting clock speed , it plays little to no role in photic entrainment . Surprisingly , the elimination of phosphorylation sites on dCLK strongly influences circadian behavior in daily temperature cycles ( Figure 5 ) . Temperature cycles with amplitudes of only 2° to 3°C robustly synchronizes circadian rhythms in Drosophila and other organisms [51] , [52] , [62]–[66] . When exposed to temperature cycles of 24°C/29°C , control p{dClk-WT};Clkout flies manifested the typical bimodal activity pattern with bouts of activity anticipating the two temperature transition points , similar to that occurring during entrainment to LD cycles ( Figure 5A and S5A ) . However , even during the first days in TC , p{dClk-15A};Clkout flies already exhibit a very abnormal phase alignment with ‘morning’ and ‘evening’ bouts of activity that occur much earlier , around the middle of the cryo- and thermal-phases , respectively ( Figure 5B and S5B , C ) . The advanced timing of the morning and evening bouts of activity is much earlier than would be expected based solely on the 1 . 5 hr shorter circadian period in p{dClk-15A};Clkout flies ( Table 3 ) . That entrainment to TC is highly defective in p{dClk-15A};Clkout flies is even more dramatically underscored by the progressive advances in the evening component of activity on subsequent days ( Figure 5 ) . Although not as apparent , flies with increased dosage of dClk ( p{dClk-WT};+/+ flies ) also showed progressively earlier evening activity bouts in thermal cycles ( Figure 5C and F ) but not LD cycles , further suggesting that increased levels/activity of dCLK are causally linked to the inability of maintaining a stable phase relationship with TC . Because the timing of the evening activity in both p{dClk-15A};Clkout and p{dClk-WT};+/+ flies occurs progressively earlier during TC , our results strongly suggest that these flies are only weakly synchronized to TC and are partially free-running at their faster endogenous periods . In trying to determine why p{dClk-15A};Clkout flies might exhibit a defect in temperature entrainment but not photic entrainment , it is important to note that several lines of evidence support the notion that light is a more potent synchronizer of the clock in D . melanogaster compared to temperature entrainment , including the use of out-of-phase light/dark and temperature cycles [4] . In addition , lowering the levels/function of the key photic entrainment photoreceptor CRYPTOCHROME ( CRY ) increases the ability to synchronize to TC [67] , suggesting the dominance of light input under normal conditions . Also , it takes many more days to shift the phase of the clock via TC compared to LD cycles [50] . The overall strength of light in D . melanogaster entrainment is not surprising given the ability of light pulses to evoke the rapid degradation of TIM and the great sensitivity of Drosophila CRY/TIM to light [68] . Indeed , constant light rescues the ability of TC to stably entrain behavioral rhythms in p{dClk-15A};Clkout ( Figure 5 , G–J ) , presumably by maintaining the clock in a more normal state ( Figure 6 ) . Intriguingly , prior work showed a similar pattern for the classic perS and perL mutants that display short ( 19 hr ) and long ( 29 hr ) endogenous rhythms , respectively [66] . That is , while wild-type flies entrain to TC in DD or LL , but perS and perL flies only entrain to TC in LL [66] . This suggests that alterations in the PER protein rhythm might preferentially disrupt thermal entrainment . In the case of p{dClk-15A};Clkout flies the amplitude of the PER abundance cycle is increased reaching higher peak values ( Figure 4 ) . Clocks with higher amplitudes are more resistant to entrainment by weak zeitgebers [69]–[71] . Relevant to this discussion , reducing CLOCK activity in mice decreased the amplitude of the circadian pacemaker and per gene expression , enhancing the ability to evoke phase shifts in behavioral rhythms [72] , [73] . Thus , a simple model for our results is that the increased per mRNA and protein rhythms in p{dClk-15A};Clkout flies leads to an increase in pacemaker amplitude minimizing their ability to synchronize to weaker entraining signals such as TC . However , it should be noted that higher amplitude rhythms of cycling mRNAs are highly suggestive but not definitive proof of an increase in oscillator . A standard approach to infer the relative amplitude of a clock is to increase the strength of the entraining signal , which should enhance its entrainment potential [69] , [70] , [74] . Although a change in the amplitude of the clock in p{dClk-15A};Clkout flies offers a plausible explanation for the preferential defect in temperature entrainment , there are other possibilities . For example , CRY-positive clock cells are more important for entraining to LD cycles , whereas CRY-negative clock cells are more important for TC entrainment [4] . Thus , dCLK-15A could have preferential effects in CRY-negative cells to lessen their contribution , impairing TC entrainment . Another more speculative idea is that the phosphorylation of dCLK can act as a thermal sensor , although this would be specific to temperature entrainment as temperature compensation appears normal in the p{dClk-15A};Clkout flies ( Table 2 ) . Clearly , future studies will be required to better address the mechanism underlying the impaired synchronization of p{dClk-15A};Clkout flies to temperature cycles . However , our findings reveal that phosphorylation of a key rate-limiting circadian transcription factor is critical for entrainment to daily temperature cycles . Indeed , the CLOCK protein in zebrafish [65] was shown to be regulated by temperature , suggesting a universal role for CLOCK in the adaptation of animal circadian clocks to thermal cues .
The pMT-dClk-V5 , pMT-HA-dClk-V5 , pMT-HA-dClk , pAct-per , pAct-per-V5 and pMT-dbt-V5 plasmids were described previously [20] , [29] , [31] . pMT-dClk15A-V5 and pMT-dClk16A-V5 were generated by serially changing codons for Ser to those of Ala by using a Quick Change site-directed mutagenesis kit ( Stratagene ) . All final constructs were verified by DNA sequencing . Hygromycin-resistant stable Schneider 2 ( S2 ) cell lines expressing pMT-HA-dClk-V5 were established for dCLK purification . dClk expression was induced by adding 500 µM CuSO4 to the medium and cells were harvested 24 hr post-induction . 200 ml of culture ( 3×106 cells/ml ) was used and harvested cells were lysed using modified-RIPA buffer ( 50 mM Tris-HCl [pH 7 . 5] , 150 mM NaCl , 1% NP-40 , 0 . 25% Sodium deoxycholate ) with the addition of a protease inhibitor cocktail ( Roche ) containing 1 mM EDTA , 25 mM NaF , and 1 mM Na2VO3 . To extracts , anti-V5 antibody ( Invitrogen ) was added and incubated overnight with gentle rotation at 4°C followed by the addition of Dynabeads Protein A ( Invitrogen ) with a further overnight incubation . Beads were collected using DynalMPC . dCLK was eluted with modified Laemmli buffer ( 150 mM Tris-HCl [pH 6 . 8] , 6 mM EDTA , 3% SDS , 30% Glycerol ) supplemented with 50 mM reducing agent TCEP ( Calbiochem ) at 65°C for 20 min . Alkylation was performed by adding 0 . 5M IAA ( iodoacetamide ) for 20 min at room temperature in the dark . The eluate was resolved using 8% SDS-PAGE , and all the detectable dCLK bands of differing electrophoretic mobility excised ( which under the conditions used was mainly the ‘intermediate’ phosphorylated band ) , subjected to protease digestion and analyzed by mass spectrometry . Mass spectrometry was performed as described in Schlosser et al . 2005 . Data analysis was performed as described previously [41] . S2 cells were obtained from Invitrogen and transfected using effectene following the manufacturer's protocol ( Qiagen ) . Luciferase ( luc ) reporter assay was performed as described previously [29] , [75] . Briefly , S2 cells were placed in 24-well plates and co-transfected with 0–100 ng pMT-dClk-V5 and pMT-dClk-16A-V5 along with 10 ng of perEluc , 30 ng of pAct-β-gal-V5/His as indicated . dPER mediated repression of dCLK dependent transactivation was measured by transfecting 0–20 ng of pAct-dper together with 2 ng of pMT-dClk-V5 or pMT-dClk-16A-V5 . One day after transfection , dClk expression was induced with 500 µM CuSO4 ( final in the media ) , and after another day cells were washed in phosphate buffered saline ( PBS ) , followed by lysis in 300 µl of Reporter Lysis Buffer ( Promega ) . Aliquots of cell extracts were assayed for β-galactosidase and luciferase activities using the Luciferase Assay System and protocols supplied by the manufacturer ( Promega ) . Clkout flies were generated in one of our laboratories ( P . E . H . ) as follows: 5 . 2 kb deletion of dClk exon 1 and upstream sequences was generated by FLP-mediated recombination between FRT sites in the pBac Clk[f06808] and pBac Clk [f03095] [76] , [77] . Flippase ( FLP ) -induced recombination was induced by a daily 1 h heat-shock at 37°C given to hsFLP;;f06808/f03095 larvae and pupae . Three recombinants were recovered , and each produced a deletion rather than a duplication of intervening dClk sequences . The remaining pBac insert was excised via pBac transposase induced transposition resulting in white-eyed flies harboring the deletion [78] . A DNA fragment containing the deleted sequences was amplified using primers situated upstream of the f03095 insertion site ( 5′ CGGAATATTGGACAACAAACAG 3′ ) and downstream of the f06808 insertion site ( 5′CAGCAGTGGAATCTTAATACAG 3′ ) , and sequenced to confirm the endpoints of the deletion . This new dClk deletion allele was named Clkout . To generate transgenic flies that produce wild-type dCLK tagged with V5 at the C-terminus , dClk-containing P[acman] transgene was generated using recombineering-mediated gap repair [79] . To prepare the P[acman] vector , homology arms were amplified from genomic DNA with primers clkLA-f ( ATGTGGCGCGCCGCCCCAAAAATCCATAAATGCT ) and clkLA-r ( GTGTTGGATCCAGGGGTGTTATAGAGAGGGACA ) for the left arm and clkRA-f ( GTGTGGATCCGCAGAGTGAAACCTGTGCAA ) and clkRA-r ( ATATATGTGCGGCCGCTCCCGGTTATGAGTTTTTCG ) for the right arm via PCR , and cloned as AscI-BamHI and BamHI-NotI fragments into AscI and NotI digested attB-P[acman]-ApR vector ( modified to remove the SphI site ) to form attB-P[acman]ClkLARA . Recombination-competent SW102 cells harboring BAC clone RP98 5K6 ( BACPAC Resource Center , Oakland , Ca , USA ) , which contains the dClk genomic region , were transformed with the attB-P[acman]ClkLARA vector ( linearized with BamHI ) . Recombinants containing 15 . 5 kb of genomic sequence beginning ∼8 kb upstream of the dClk translation start and ending ∼2 . 5 kb downstream of the dClk stop codon were verified by PCR and sequencing and termed attB-P[acman]-Clk . To introduce a V5 epitope tag at the C-terminus of the dClk open reading frame ( ORF ) , a 3′ genomic fragment of dClk ( from 351 bp upstream to 1580 bp downstream of the translation stop ) was cloned into pGEM-T vector ( Promega , Madison , WI ) . Sequences encoding V5 were introduced in-frame immediately upstream of the dClk stop codon using the Quickchange site directed mutagenesis kit ( Stratagene , La Jolla , CA ) to create pGEM-T-dClk3′V5 . The 3′ dClk genomic fragment in attB-P[acman]- dClk was swapped with the 3′ fragment in pGEM-T-dClk3′V5 using SphI and NotI to form attB-P[acman]- dClkV5 . This transgene was inserted into the VK00018 attP site on chromosome 2 via PhiC31-mediated transgenesis [79] , [80] . Transformation vector containing a genomic dClk wherein the codons for the 15 identified phospho-serine were switched to those for alanine was generated in multiple stages as follows: A genomic dClk sub-fragment from NheI to SphI site was isolated from P[acman]-dClk-V5 and subcloned into pSP72 vector where the multi-cloning sites were mutagenized to introduce a NheI site , and named this plasmid as pSP72-dClk ( NheI/SphI ) . Next , we obtained a dClk sub-fragment spanning from the NcoI to SphI sites by restriction digestion of pSP72-dClk ( NheI/SphI ) , subcloned the released fragment into pSP72 where the multi-cloning sites were mutagenized to introduce a NcoI site , and named this plasmid as pSP72-dClk ( NcoI/SphI ) . We performed serial site directed mutagenesis with pSP72-dClk ( NcoI/SphI ) and finally made pSP72-dClk ( NCoI/SphI ) -S11A wherein codons for the serine residues at amino acids 209 , 210 , 211 , 444 , 450 , 487 , 504 , 611 , 645 , 859 , 902 on dCLK were all switched to those for alanine residues [ ( GenBank accession number NP_001014576 ) ] . We purified the dClk ( NCoI/SphI ) -S11A insert by restriction enzyme digestion of pSP72-dClk ( NCoI/SphI ) -S11A and replaced the wild-type dClk ( NcoI/SphI ) insert , generating pSP72-dClk ( NheI/SphI ) -S11A . Next , a more 3′ genomic dClk sub-fragment from the SphI to NotI sites was subcloned into pSP72 vector where the multi-cloning sites were mutagenized to include NotI and NheI sites , and named this plasmid as pSP72-dClk ( SphI/NotI ) . We performed serial site directed mutagenesis with pSP72-dClk ( SphI/NotI ) and made pSP72-dClk ( SphI/NotI ) -S4A wherein codons for the serine residues at amino acids 924 , 934 , 938 , 1018 were switched to those for alanine . Finally , the genomic dClk ( SphI/NotI ) -S4A fragment was ligated with pSP72-dClk ( NheI/SphI ) -S11A generating pSP72-dClk ( NheI/NotI ) -S15A , and then dClk ( NheI/NotI ) -S15A fragment was switched with wild-type dClk ( NheI/NotI ) fragment in pacman-dClk-V5 plasmid yielding P[acman]-dClk-15A-V5 . Transgenic flies were generated by BestGene Inc . ( CA , USA ) . P[acman]-dClk-15A-V5 transformation vector was injected into flies carrying the VK00018 attP docking site on the second chromosome for site-specific integration [79] . Two independent germ-line transformants bearing the dClk-15A-V5 transgene in a wild-type background were obtained and then crossed into a Clkout genetic background to yield dClk-15A-V5;Clkout . The locomotor activities of individual flies were measured as previously described using the Drosophila Activity Monitoring system from Trikinetics ( Waltham , MA ) . Young adult flies were used for the analysis and exposed to 4 days of 12 h light followed by 12 h dark [where zeitgeber time 0 ( ZT0 ) is defined as the time when the light phase begins] at 25°C and subsequently kept in constant dark conditions ( DD ) for 7 days . Temperature entrainment ( temperature cycle , TC ) was performed in constant dark condition and in some cases , in the presence of constant light ( >2000lux ) . Temperature cycles were 12 h of 24°C ( cryo phase ) followed by 12 h of 29°C ( thermal phase ) ( where ZT0 is defined as the time when the cryo phase begins ) for 4 days and subsequently kept at 24°C for 7 days . The locomotor activity data for each individual fly was analyzed using the FaasX software ( Fly Activity Analysis Suite for MacOSX ) , which was generously provided by F . Rouyer ( CNRS , France ) . Periods were calculated for each individual fly using chi-square periodogram analysis and pooled to obtain a group average for each independent transgenic line or genotype . Power is a quantification of the relative strength of the rhythm during DD . Individual flies with a power ≥10 and a ‘width’ value of 2 or more ( denotes number of peaks in 30-min increments above the periodogram 95% confidence line ) were considered rhythmic . Actogram represents the locomotor activity data throughout the experimental period . Vertical bars in the actogram represent absolute activity levels for each 30 min intervals averaged for each given genotypes of flies . The strength of this measurement can be manipulated by using the function called hash density , which represent the number of times fly need to make beam crossing to be registered as one vertical bar . The hash density of the actogram was varied for better comparison depending on the activity levels of given genotypes of flies . Protein extracts from S2 cells were prepared as previously described [31] . Briefly , the cells were lysed using modified-RIPA buffer ( 50 mM Tris-HCl [pH 7 . 5] , 150 mM NaCl , 1% NP-40 , 0 . 25% Sodium deoxycholate ) with the addition of protease inhibitor cocktail ( GeneDEPOT ) and phosphatase inhibitor cocktail ( GeneDEPOT ) . For detection of dCLK recombinant protein , extracts were obtained using RIPA buffer 25 mM Tris-HCl [pH 7 . 5] , 50 mM NaCl , 0 . 5% Sodium deoxycholate , 0 . 5% NP40 , 0 . 1% SDS ) and were sonicated briefly as previously described [29] . Flies were collected by freezing at the indicated times in light-dark ( LD ) or temperature cycles ( TC ) and total fly head extracts prepared using modified-RIPA buffer or RIPA buffer with sonication ( for dCLK ) . Extracts were resolved by 5% polyacrylamide gels or by 3–8% Tris-acetate Criterion gel ( Bio-Rad ) in some case for dCLK , transferred to PVDF membrane ( Immobilon-P , Millipore ) , and immunoblots were treated with chemiluminescence ( ECL , Thermo ) . Primary antibodies were used at the following dilutions; anti-V5 ( Invitrogen ) , 1∶5000; anti-HA ( 12CA5 , Roche ) , 1∶2000; anti-OGT ( H-300 , Santa Cruz ) , 1∶3000; anti-PER , ( Rb1 ) 1∶3000; anti-TIM ( TR3 ) , 1∶3000; anti-dCLK ( GP208 ) 1∶3000 . Quantification of band intensity was performed using image J software . For immunoprecipitation , cell extracts from S2 cells were prepared and 3 µl of anti-HA ( 12CA5 ) or anti-V5 antibody was added depending on the target protein sought , and incubated for overnight at 4°C with gentle rotation . The next day , 20 µl of Gammabind-sephase bead ( GE healthcare ) was added with a further incubation of 3 hr at 4°C . The immune complexes were eluted with 1X SDS-PAGE sample buffer . For λ- phosphatase treatment , the purified immune complexes were resuspended in λ protein phosphatase buffer ( 50 mM Tris-HCl [pH 7 . 5] , 0 . 1 mM EDTA , 5 mM DTT , 0 . 01% Triton X-100 , 2 mM MnCl2 , and 0 . 1 mg/ml bovine serum albumin ) , divided into two equal aliquots . One aliquot of bead was treated with 200 units of λ protein phosphatase ( NEB ) and no addition was made to the other aliquot . Both aliquots were incubated for 30 min at 30°C with occasional shaking , and immune complexes analyzed by immunoblotting . Total RNA was isolated from frozen heads using QIAzol lysis reagent ( QIAGEN ) . 1 µg of total RNA was reverse transcribed with oligo-dT primer using Prime Script reverse transcriptase ( TAKARA ) and real-time PCR was performed in Corbett Rotor Gene 6000 ( Corbett Life Science ) using Quantitect SYBR Green PCR kit ( Qiagen ) . Primer sequences used here are as follows; dper forward: 5′-GACCGAATCCCTGCTCAATA-3′; dper reverse: 5′-GTGTCATTGGCGGACTTCTT-3′; tim forward: 5′-CCCTTATACCCGAGGTGGAT-3′; tim reverse: 5′-TGATCGAGTTGCAGTGCTTC-3′; dClk forward: 5′-CAGCCGCAATTCAATCAGTA-3′; dClk reverse: 5′-GCAACTGTGAGTGGCTCTGA-3′ . We also included primers for the noncycling mRNA coding for CBP20 as previously described , and sequences are as follows; cbp20 forward: 5′-GTCTGATTCGTGTGGACTGG-3′; cbp20 reverse: 5′-CAACAGTTTGCCATAACCCC-3′ . Results were analyzed with software associated with Rotor Gene 6000 , and relative mRNA levels were quantitated using the 2−ΔΔCt method . | Circadian clocks are synchronized to local time by daily cycles in light-dark and temperature . Although light is generally thought to be the most dominant entraining cue in nature , daily cycles in temperature are sufficient to synchronize clocks in a large range of organisms . In Drosophila , dCLOCK is a master circadian transcription factor that drives cyclical gene expression and is likely the rate-limiting component in the transcriptional/translational feedback loops that underlie the timekeeping mechanism . dCLOCK undergoes temporal changes in phosphorylation throughout a day , which is also observed for mammalian CLOCK . However , the role of CLOCK phosphorylation at the organismal level is still unclear . Using mass-spectrometry , we identified more than a dozen phosphorylation sites on dCLOCK . Blocking global phosphorylation of dCLOCK by mutating phospho-acceptor sites to alanine increases its abundance and transcriptional activity , leading to higher peak values and amplitudes in the mRNA rhythms of core clock genes , which likely explains the accelerated clock speed . Surprisingly , the clock-controlled daily activity rhythm fails to maintain synchrony with daily temperature cycles , although there is no observable defect in aligning to light/dark cycles . Our findings suggest a novel role for clock protein phosphorylation in governing the effective strengths of entraining modalities by adjusting clock amplitude . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"biochemistry",
"biology",
"and",
"life",
"sciences",
"chronobiology"
]
| 2014 | Phosphorylation of a Central Clock Transcription Factor Is Required for Thermal but Not Photic Entrainment |
Schistosomiasis is a major disease of the developing world for which no vaccine has been successfully commercialized . While numerous Schistosoma mansoni worm antigens have been identified that elicit antibody responses during natural infections , little is known as to the identities of the schistosome antigens that are most prominently recognized by antibodies generated through natural infection . Non-reducing western blots probed with serum from schistosome-infected mice , rats and humans on total extracts of larval or adult schistosomes revealed that a small number of antigen bands predominate in all cases . Recognition of each of these major bands was lost when the blots were run under reducing condition . We expressed a rationally selected group of schistosome tegumental membrane antigens in insect host cells , and used the membrane extracts of these cells to unambiguously identify the major antigens recognized by S . mansoni infected mouse , rat and human serum . These results revealed that a limited number of dominant , reduction-sensitive conformational epitopes on five major tegumental surface membrane proteins: SmTsp2 , Sm23 , Sm29 , SmLy6B and SmLy6F , are primary targets of mouse , rat and human S . mansoni infection sera antibodies . We conclude that , Schistosoma mansoni infection of both permissive ( mouse ) and non-permissive ( rat ) rodent models , as well as humans , elicit a dominant antibody response recognizing a limited number of conformational epitopes on the same five tegumental membrane proteins . Thus it appears that neither infecting schistosomula nor mature adult schistosomes are substantively impacted by the robust circulating anti-tegumental antibody response they elicit to these antigens . Importantly , our data suggest a need to re-evaluate host immune responses to many schistosome antigens and has important implications regarding schistosome immune evasion mechanisms and schistosomiasis vaccine development .
Schistosomiasis is a disease affecting more than 200 million individuals living mostly in underdeveloped tropical and subtropical regions ( http://www . who . int/mediacentre/factsheets/fs115/en/ ) . The disease is caused by infection with schistosome blood flukes which can survive , if untreated , for decades inside the vascular system of immune competent permissive hosts . Illness is primarily a consequence of immunopathology from schistosome eggs trapped in tissues ( reviewed by [1] ) . Despite the long term presence of adult worms in their vascular system , permissive hosts , by definition , do not typically develop immune responses directed at juvenile or adult worms [2] that are capable of preventing new infections or eliminating all adult worms . This is not to imply that anti-worm immune responses are completely ineffective , as a measure of protective immunity following multiple rounds of reinfection has been documented [3 , 4] . The humoral response to schistosome infection has been extensively characterized in efforts to identify antigens for use in the diagnosis of infection or that might elicit a response that protects vaccinated individuals from infection . To seek protective vaccine candidates , researchers have sought to identify antigens that are recognized more intensely by serum antibodies from animal models with demonstrated resistance to infection . Examples include comparisons in schistosome antigen recognition by: 1 ) non-permissive rats vs permissive mice models [5 , 6]; 2 ) mice +/- vaccination by irradiated cercariae [7] , and; 3 ) humans displaying susceptibility vs putative resistance to infection [8 , 9] . Through these efforts , dozens of schistosome antigens have been found to be recognized by antibodies in serum from infected humans or rodents ( reviewed by [10–12] ) . While these efforts have identified antigens recognized by various infection sera , it remains unclear as to which antigens are most abundantly targeted by antibodies in these sera . This information could aid in understanding the nature of the normal humoral immune response to schistosome infection and provide important guidance to efforts seeking improved diagnostic tools and new vaccines . In an effort to identify which proteins on larval and adult schistosomes are most abundantly recognized by antibodies in serum of infected rodents and humans , we began by probing western blots with these sera on total schistosome antigen extracts . To better preserve antigen conformation , we did not destroy protein disulfide bridges by chemical reduction prior to resolution on gels . Highly diluted sera from schistosome infected rats , mice and humans ( 1:2000 ) all , surprisingly , produced similar banding patterns on blots of total extracts of schistosomula and adults , revealing a remarkably small number of antigen bands that predominate . When the antigen samples were reduced , these major bands were no longer observed by the same diluted sera . Based on a variety of inferences , explained below , we selected a set of candidate schistosome antigens which were then expressed within insect cells . Here we use extracts of these cells to identify two tetraspanins: SmTsp2 and Sm23 and three Ly6 family proteins: Sm29 , SmLy6B , and SmLy6F , as the major schistosome antigens recognized by the antibodies in dilute serum of infected mice , rats and humans . All of these proteins are reported to be present within the host-interactive surface tegument of schistosomes , and two of the antigens , SmTsp2 and Sm29 , are currently strong candidates as vaccine immunogens to prevent schistosomiasis [8 , 13 , 14] . We show that the infection serum antibodies recognizing these two antigens can be largely blocked by two previously reported monoclonal single-chain Fv domain antibodies ( scFvs Teg1 and Teg4 [15] ) , which suggests that the serum primarily recognizes limited epitopes on these antigens . These findings have significant implications regarding both the mechanisms by which schistosomes evade immune damage and the development of schistosomiasis vaccines , which are discussed .
Total antigen extractions were performed on larval and adult schistosomes by boiling the freshly frozen parasites in an SDS gel loading buffer under non-reducing conditions . Serum was typically diluted to 1:2000 so as to preferentially reveal the antigens recognized by the more prevalent anti-schistosome antibodies . Fig 1A shows that similar patterns are recognized on Western blots prepared from either juvenile ( 5 day cultured schistosomula ) or adult worm extracts when probed with pooled sera prepared from S . mansoni-infected mice , rats and humans ( ‘infection sera’ ) . All three infection sera recognize antigens of the same apparent MW , though with differences in intensity , on both juvenile and adult worm extracts . The predominant staining species were estimated to be about 11 , 15 , 20 , 29 and 37 kDa ( arrows , Fig 1A ) by comparison to molecular weight standards . Surprisingly , recognition of these major bands is largely lost from the infection sera when the same schistosome extracts are resolved on western blots under reducing conditions at 0 . 1% β-mercaptoethanol ( βME ) ( Fig 1B ) . A similar dramatic reduction in western blot signals produced by mouse , rat or human infection sera was also observed when the ( non-reducing gel ) filters were pre-washed with 0 . 1% βME prior to serum incubation ( not shown ) . These results demonstrate that the major epitopes on the dominant schistosome antigens present in both larval and adult schistosome extracts are reduction-sensitive , and thus likely to be conformational epitopes dependent on disulfide cross-linkages . Studies were also performed to test the consistency of the infection sera western blot data when using independent pools of serum , and to compare sera from human patients having different infection histories . First , during the course of our studies we generated a second pool of mouse infection sera and found antigen recognition within juvenile and adult schistosome extracts to be virtually indistinguishable and both sera were used interchangeably ( not shown ) . Next we compared four different serum pools from patients with different schistosome infection histories and different apparent susceptibilities to infection ( see [13] ) . Two of the pools were obtained from patients that had active schistosome infections ( I , infected; SR , susceptible to reinfection ) and two pools from patients that had frequent exposures to cercariae and yet were currently uninfected ( RR , resistant to reinfection; NR , natural resistant ) . Unlike the I and NR groups , the SR and RR patients had been recently treated with praziquantel . A fifth pool of serum ( NI , noninfected ) came from people having no history of schistosome exposure . Each group contained both male and female donors . As shown in Fig 1C , while the pools produced variable signal intensities , all pools produced similar banding patterns . Non-infected patient sera completely lack any signal , demonstrating the absence of background under the conditions used . Overall , these results suggest that the schistosome antigen recognition patterns reported here with infection sera , when resolved on gels under non-reducing conditions , are almost certainly representative of infection sera from other rodent and human individuals , even those having differences in their infection histories . Thus , we conclude that the dominant recognition of these five schistosome antigen bands appear to be generally consistent across various host genetic backgrounds and parasite exposure histories . We previously reported [15] the isolation of four recombinant single-chain Fv proteins ( scFvs ) recognizing surface antigens on live schistosomes that we obtained from the B cells of schistosome infected rats , three of which recognize living schistosomula ( Table 1 ) . Two of the scFvs were shown to recognize known tegumental antigens; Teg1 recognizes the tetraspanin SmTsp2 [16] , Teg4 recognizes Sm29 [17] ( also called SmLy6D [18] ) . A third scFv ( Teg5 ) was shown to weakly recognize living schistosomes , and an scFv called S3 had been obtained earlier from the same source [19] which recognizes the schistosome tetraspanin antigen , Sm23 . When schistosome extracts resolved by non-reducing gels were probed by western blotting with these scFvs , several produced strong signals ( Fig 2 ) . These signals were dramatically diminished or undetectable when the extracts were resolved on reducing gels ( not shown ) . Of interest , each of the scFvs recognized antigens having the same mobility as antigens recognized by the various ( mouse , rat , human ) infection sera depicted in Fig 1 . Teg1 and S3 recognized their tetraspanin targets ( Tsp2 and Sm23 ) as 20 and 37 kDa species ( Fig 2 , arrows ) . The presence of the larger species suggests that the ~20 kDa monomers become partially cross-linked into ~37 kDa dimeric forms under the non-reducing conditions . Teg4 strongly recognized the 29 kDa Sm29 while Teg5 recognized a minor antigen of about 20 kDa . To identify the reduction-sensitive schistosome antigens from larval and adult worm extracts that are recognized by dilute schistosome infection sera on Western blots , we sought to produce conformationally-native recombinant versions of a set of selected candidate antigens and test their ability to specifically block infection sera antibodies recognizing one or more of the major bands . Three obvious candidate antigens were SmTsp2 , Sm23 and Sm29 based on the fact that our prior studies had identified them as targets of scFvs that were obtained from B cells of schistosome-infected rats [15] , and the fact that the gel mobility of their target antigens matched those of the three larger antigens ( 20 , 29 , 37 kDa ) recognized by infection sera ( Figs 1 and 2 ) . The two other major bands recognized by schistosome infection sera had apparent MWs of 11 or 15 kDa . This small size is similar to that of most members of the SmLy6 family of membrane proteins [18 , 20] . The Ly6 proteins are GPI-anchored proteins , some of which are known to be present in the schistosome tegument [20 , 21] , and all are expected to have conserved , reduction-sensitive , intra-chain disulfide linkages within their extracellular domains [22] . Several of the smaller SmLy6 proteins are also known to be well-expressed by mammalian stage schistosomes [20] . In fact , Sm29 is also a Ly6 family protein ( SmLy6D ) containing two linked Ly6 domains [18] . Based on their small size , sensitivity to reduction and available expression data , we selected four additional members of the SmLy6 family: SmLy6A , SmLy6B , SmLy6C and SmLy6F ( nomenclature from [18] ) as strong candidates to be major antigens recognized by schistosome infection sera . These four SmLy6 proteins were also called SmCD59 . 1 , SmCD59 . 2 , SmCD59 . 3 and SmCD59 . 4 , respectively , in another recent report on putative schistosome membrane antigens [20] . Early efforts to produce recombinant schistosome SmTsp2 and Sm29 proteins were performed in E . coli host cells transformed with expression vectors containing the extracellular regions of these proteins fused to E . coli thioredoxin ( Trx ) to facilitate protein folding . Western blots with these purified recombinant proteins were poorly recognized by rat infection sera or the scFvs , probably because the reduction-sensitive epitopes were not reproduced well when expressed in E . coli cytosol . We thus began employing the baculovirus expression system within insect cell hosts in an effort to generate recombinant membrane protein antigens that we expected would better retain the native conformations of these proteins on schistosomes . The seven rationally selected schistosome membrane protein candidates ( above ) were each expressed in recombinant insect cells as described in Materials and Methods . The SmLy6 proteins ( SmLy6A , SmLyB , SmLy6C and SmLy6F ) were expressed with an insect secretory leader and each contained a FLAG-tag epitope ( DYKDDDDK ) that remained at their amino ends following signal processing . The tetraspanin proteins , SmTsp2 and Sm23 were expressed using their complete S . mansoni coding DNA . Sm29 was also expressed with its natural signal peptide and using full-size coding DNA . Recombinant insect cells expressing the schistosome antigens were harvested and the membrane proteins solubilized in a lysis buffer containing 1% Triton X-100 and no reducing agents . Western blots were prepared from insect cell extracts containing the four small recombinant SmLy6 proteins ( SmLy6A , B , C , F ) and then probed with anti-FLAG antibody ( Ab ) . As shown in Fig 3A , under non-reducing conditions the insect cell extracts expressing SmLy6A , B and F contained FLAG-tagged proteins in the expected size range ( ~12 kDa ) . For each of the extracts , one or two larger FLAG protein species were also present in the 15–20 kDa size range . Unexpectedly , insect SmLy6C extracts did not contain a FLAG protein of the expected size , but instead contained several bands clustered about 19 kDa . Higher molecular weight forms were visible with all four SmLy6 proteins . When the SmLy6 extracts were resolved on gels under reducing conditions , the lower MW bands remained while most of the 15–20 kDa bands and higher MW forms were no longer visible , suggesting that the higher MW forms are multimers stabilized by disulfide linkages . The same four recombinant SmLy6 insect cell extracts were also probed with pooled S . mansoni rat infection sera ( Fig 3B ) . All four extracts contained a protein species recognized by the rat sera that was a subset of bands recognized by anti-FLAG , indicating that only some of the recombinant protein had folded to a conformation recognized by the infection sera . Surprisingly , for SmLy6A , C and F , the bands primarily recognized by immune rat sera were the MW species ~15–20 kDa . These bands are larger than expected and may be glycosylated . Only the SmLy6B extract contained a band of ~11 kDa , the expected size of these SmLy6 proteins based on coding sequence , which was clearly recognized by the rat infection sera . As when probing the schistosome extracts ( Fig 1 ) , rat infection sera recognition of the four recombinant SmLy6 proteins was lost when the gel was run under reducing conditions ( Fig 3B , right panel ) . Insect extracts expressing SmTsp2 , Sm23 and Sm29 , which lack a FLAG tag , were resolved on non-reducing or reducing gels and probed with the appropriate scFvs ( Teg1 , S3 , Teg4 respectively ) or with rat infection sera ( Fig 4 ) . As expected , the three scFvs recognized their insect cell expressed target antigens . Rat infection sera produced a virtually identical recognition pattern as the scFvs on replicate non-reducing blots , indicating , as has been previously reported , that schistosome infection sera contain antibodies recognizing SmTsp2 , Sm23 and Sm29 [8 , 9 , 23] . Most or all recognition of these proteins by the scFvs or the infection serum was lost when these recombinant insect extracts were resolved under reducing conditions . We also tested Teg5 scFv for recognition of the insect cell extracts containing the various S . mansoni membrane proteins . As shown in Fig 5 , Teg5 clearly recognizes a reduction-sensitive epitope on SmLy6C . Since Teg5 was shown to recognize living worms [15] , these results shows that SmLy6C can be expressed on the outside surface of larval and adult schistosomes . Thus Teg5 , like Teg1 and Teg4 , recognize conformational epitopes on schistosome tegumental antigen targets , consistent with the finding that rat infection sera predominantly recognize reduction-sensitive epitopes on schistosome antigens . Each of the insect extracts expressing the selected schistosome membrane proteins were characterized by non-reducing Western blots for their recognition by serum from infected rats , mice and humans . Blots with the five SmLy6 proteins are shown in Fig 6A and reveal that rat , mouse and human infection sera recognize each of the proteins , though with variable intensity . In general , mice and rats recognize SmLy6A and SmLy6B more intensely than SmLy6C and SmLy6F , while the reverse was true with the human infection serum pool . All infection sera recognized Sm29 to a similar extent . Results with the two tetraspanin proteins , SmTsp2 and Sm23 , were more variable ( Fig 6B ) . When quantified by imaging , the rat infection serum pools recognized both proteins about equally , while the mouse pool recognized Sm23 with >10x greater intensity than SmTsp2 and the human serum pool recognized SmTsp2 with >10x greater intensity than Sm23 . Mild detergent extracts of insect cells expressing a single recombinant schistosome antigen were next tested for their ability to block recognition of specific bands produced by various infection sera on western blots of SDS-PAGE resolved juvenile and adult schistosome extracts ( Figs 7–10 ) . The insect cell extracts from all of the seven rationally selected schistosome membrane antigens were separately incubated with mouse , rat or human infection sera prior to and during the incubation of the sera with western blots of total extracts of juvenile and adult schistosomes . All the results shown were repeated at least three times , and compared two different pools of mouse infection sera and four different pools of human infection sera over the course of these studies , and yet produced consistent results . In some cases , we were unable to completely block recognition of a specific band through pre-incubation with an insect cell extract , and it is possible in these cases that the insect cell recombinant protein lacks a conformer of the target antigen which is present in schistosome extracts , or that a second , unidentified minor antigen having the same gel mobility co-migrates with the major antigen species . Figs 7 and 8 show that pre-incubation of rodent or human infection serum pools with insect cell extracts containing recombinant SmLy6B or SmLy6F specifically absorbed antibodies that recognize the 11 kDa or 15 kDa bands , respectively . Insect SmLy6B blocked the strong recognition of an 11 kDa band by infection sera from both of the rodent sera ( Fig 7 ) . A band at 11 kDa was also weakly recognized by human infection sera and this band was reproducibly blocked by SmLy6B in similar blots , though it is not clear in the example shown . Insect extracts containing SmLy6F consistently blocked recognition of the 15 kDa band that is well recognized by rodent and human infection sera ( Fig 8 ) . This antigen was not expressed well by insect cells and the blocking studies were less robust , yet we consistently observed reduction in the 15 kDa signals produced by all the infection sera . These results show that 11 kDa and 15 kDa schistosome antigens prominently recognized by infection sera on western blots of schistosome total extracts are SmLy6B and SmLy6F . We were repeatedly unable to observe a reduction in the recognition of any of the antigens when the three infection sera pools were pre-incubated with extracts containing SmLy6A or SmLy6C ( not shown ) . Negative results in these studies though are inconclusive due to uncertainty as to whether sufficient insect antigen , in its properly folded state , is present in the insect cell extracts to successfully block the infection serum antibodies . Nevertheless , since the rat infection sera recognize recombinant SmLy6A and SmLy6C ( Fig 3B ) , these results suggest that these antigens may not be sufficiently well expressed by larval or adult schistosomes to produce major bands on Western blots of total worm extracts . Serum blocking studies with Sm29 , shown in Fig 9 , were particularly dramatic and robust . The major 29 kDa antigen , which is present in both schistosomula and adult schistosomes and recognized by mouse , rat and human infection sera , is almost quantitatively quenched when the sera are pre-incubated with recombinant Sm29 extracts . Antibody recognition of Sm29 has proven to be extremely sensitive to reduction and other denaturing conditions ( not shown ) , yet when care is taken in the preparation of schistosome extracts , Sm29 has consistently been observed as one of the two or three most dominant antigens recognized by infection sera from mice , rats and humans . Similar blocking studies were also performed with insect extracts expressing SmTsp2 or Sm23 ( Fig 10 ) . The major 20 kDa band recognized by all three infection sera was variably blocked by the presence of these extracts , likely because of the large variation in the titers of anti-SmTsp2 and anti-Sm23 antibodies in mouse vs human infection sera ( Fig 6B ) . While both extracts seemed to consistently diminish the 20 kDa signal produced by rat infection sera , only the Sm23 extracts had this effect on mouse infection sera . The human infection serum recognition of the 20 kDa band was largely and consistently blocked by incubation with SmTsp2 extracts . Sm23 extracts were not tested due to the negligible titer of Sm23 antibodies in human sera ( Fig 6B ) . We conclude that the major 20 kDa species recognized by the three infection sera is likely a mixture of both SmTsp2 and Sm23 , and perhaps other proteins of this size such as other known tetraspanins [24] . We next tested whether our four anti-schistosome scFvs ( Teg1 , S3 , Teg4 and Teg5 ) could effectively compete with antibodies in infection sera for binding to their antigen targets . In Fig 11A , we individually tested our monoclonal scFvs for their ability to interfere with rat infection sera binding to their targets expressed by insect cells . Surprisingly , Teg5 , Teg4 , and Teg1 , were able to almost completely block rat infection sera recognition of SmLy6C , SmTsp2 and Sm29 respectively , while pre-incubation of the filters with a different scFv had no effect . Only S3 scFv was unable to appreciably diminish recognition of its target , Sm23 , by the rat sera . We next tested the ability of the scFvs to block rat infection sera recognition of specific bands within the juvenile and adult schistosome total extract ( Fig 11B ) . The Teg1 and S3 scFvs both appeared to diminish recognition of the 20 kDa species in juvenile schistosomula , while only Teg1 had this effect in adult extracts . This is consistent with the results in Fig 2 showing that Sm23 is much more abundant than SmTsp2 in schistosomula than in adults . The results suggest that the Teg1 monoclonal scFv blocks most or all rat infection sera recognition of SmTsp2 . This blocking effect was even more obvious with Teg4 pre-incubation which produced a major reduction in the rat sera recognition of the 29 kDa band identified as Sm29 . Together these results suggest that the Teg1 and Teg4 scFvs bind epitopes that represent the dominant epitope for rat infection sera binding to either SmTsp2 or Sm29 . Fig 12 shows a Western blot comparing rat infection sera recognition of adult worm total extracts with an adult worm extract that is highly enriched in tegumental proteins [15 , 25] . The figure summarizes the identities of the major antigens in total worm extracts recognized by infection sera ( Figs 7–10 ) . The identities of these proteins in tegumental extracts were confirmed by antigen blocking studies ( not shown ) . These results show that Sm29 and SmTsp2 ( both monomer and putative dimer ) are particularly enriched in the tegumental extracts . In contrast , SmLy6F , and particularly SmLy6B , are less enriched in the tegument extracts . This result is consistent with localization studies showing that SmLy6B is widely distributed throughout schistosomes , including the tegument ( SmCD59 . 2 [20] ) .
The use of non-reducing SDS-PAGE for the resolution of schistosome antigens substantially changes the nature of western blot data generated on filters probed by the sera of infected mice , rats and humans when compared to standard reducing gels . Studies reported here reveal that a major portion of serum antibodies from S . mansoni infected mice , rats and humans recognize reduction sensitive epitopes on a limited number of antigens in total extracts of schistosomula and adult worms . Using recombinant schistosome antigens expressed in eukaryotic insect host cells , we unambiguously identified dominant antigens as three Ly6 family proteins- SmLy6B , SmLy6F and Sm29; and two tetraspanins- Sm23 and SmTsp2 . Of course it remains possible that other major conformational antigens are present in the schistosome extracts , but not observed on western blots because their epitopes are denatured by SDS-PAGE , perhaps because their conformations are not stabilized by the presence of disulfide linkages . While many other antigens not observed in our studies are known to be recognized by schistosome infection sera , the titers of these antibodies must be relatively low and/or the antigens present in low abundance in total worm extracts such that their signals are low or undetectable under the conditions we used . The parasite may even benefit by somehow biasing the host response away from some antigens , for example nutrient transporters or sensing receptors , against which a robust antibody response could be harmful to parasite survival . All of the major antigens identified in this study are predicted to be membrane proteins and all except SmLy6F have been reported to be associated with the schistosome tegument in proteomic studies [26 , 27] . Notably , SmLy6F has not been identified in any schistosome fractions through proteomics despite its predicted abundance [20] , perhaps because it is small , with several predicted N-linked glycosylation sites . Two of these antigens , SmTsp2 and Sm29 , are currently strong candidates as vaccine immunogens to prevent schistosomiasis [8 , 13 , 14] and we previously showed that both of these antigens have epitopes exposed on the surface of living schistosomes [15] . SmLy6B was shown to be a surface tegumental GPI-anchored protein due to its susceptibility to worm surface cleavage by phospholipase C [21] . These observations seem to suggest that host antibody responses during schistosome infection are targeted largely to a limited number of conformational epitopes present on tegumental membrane proteins , particularly those present at the host/parasite interface . The presence of glycosylation on some of these antigens may help limit epitope availability . Finding two tetraspanins , Sm23 and SmTsp2 , as major antigens recognized by S . mansoni infection serum , was not surprising . Both of these proteins have long been identified as important schistosome antigens [8 , 28] , and SmTsp2 is a particularly promising vaccine antigen [8] . More surprising was the observation that the prominent epitopes are reduction-sensitive and infection sera recognition of SmTsp2 can be largely blocked with a single scFv ( Teg4 ) . These results suggest that tetraspanin recognition by infection sera involves a limited number of conformational epitopes . The functions of Sm23 and SmTsp2 remain unknown , although a role for SmTsp2 in tegumental turnover has been inferred [29] . The identification of three Ly6 family proteins as dominant schistosome antigens was unexpected as , prior to this study , only Sm29 had been identified as an antigen during schistosome infections [9] . The two other Ly6 family proteins identified as major antigens in these studies are SmLy6B and SmLy6F . Expression of both proteins is dramatically upregulated following transformation of cercariae to schistosomula [20 , 30] . These proteins are both related to the complement inhibiting protein CD59 , though SmLy6B was tested and shown not to possess this activity when expressed on recombinant mammalian cells [20] , while SmLy6D remains untested . To date , no evidence as to the functions of these proteins has been experimentally demonstrated despite their immunogenicity during schistosome infection and the identification of at least eight other schistosome Ly6 proteins encoded in the genome [18] . The results of this study , and our previous study [15] , clearly show that schistosome infections result in a high titer of antibodies capable of binding to the living schistosomes . In the prior study , we found that infection sera from both mouse and rat clearly stained the surface of living schistosomula and lung worms . The staining was apparent only with infection sera , not pre-infection sera , showing that this staining was not non-specific binding to the antibody Fc domain [31] . In this study , we find high titers of antibodies that recognize known tegumental surface antigens , SmLy6B , SmTsp2 and Sm29 , and recognition of SmTsp2 and Sm29 can be largely blocked by pre-treatment with Teg1 and Teg4 , scFvs known to recognize living schistosomes [15] . These findings seem to contradict earlier reports that schistosomes bound little host antibody [32 , 33] , although results would be affected by any denaturants that destroy conformational epitopes . Our findings also seem inconsistent with multiple literature reports showing that humoral responses to tegumental antigens can be protective [5 , 6 , 33–35] . We find that a very similar host humoral response to the schistosome tegument occurs in both permissive ( humans and mice ) and non-permissive hosts ( rats ) , suggesting that the presence of these anti-tegument antibodies have little demonstrable impact on the progression of schistosomiasis . We conclude that differences in susceptibility between rats , mice and humans are unlikely due to differences in their humoral responses to major tegumental antigens although differences in isotype production in different species might be important . In addition , we cannot exclude that humoral responses to less abundant surface antigens could be protective , particularly neutralizing antibodies targeting essential parasite functions . The results reported here have significant implications for schistosomiasis vaccine development . Most obviously , the results imply that induction of an antibody response to the major surface tegument proteins through vaccination may not be protective on its own , because a response of this type is already present following natural infection within permissive hosts such as mice and humans . A similar humoral response to these major conformational antigens was found among human patients that display evidence of both susceptibility and resistance to schistosome infection , suggesting that these antibodies provide no obvious protective benefit to the host . Yet several reports show that two of the major antigens we find to be recognized by serum antibodies following schistosome infection , SmTsp2 and Sm29 [8 , 13 , 14] , are also among the most promising vaccine antigens . One likely explanation is that these vaccines are effective because they elicit a cell-based immune effector response targeting the host-interactive schistosome tegument . These vaccine immunogens are typically produced in prokaryotic hosts which often do not reproduce the conformation of the native proteins and thus the humoral response they elicit is likely poor at binding to live schistosome in vivo . These same vaccine immunogens , though , are known to be capable of eliciting a Th1 response that can be harmful to schistosomes [13 , 14] , and it is likely that this is primarily responsible for the reduction in parasite load following vaccination . It will be interesting to compare the potency of SmTsp2 or Sm29 vaccines prepared in prokaryotic host cells vs conformationally native antigens produced in a eukaryotic host . The results reported here seem to suggest that efforts to identify schistosome vaccine candidates that promote antibody-mediated damage to worms should focus on antigens that are not among the five antigens we identify as targets of major conformation-dependent antibodies within infection sera . These findings also suggest that antibody binding to the tegument is , in itself , insufficient to result in host immune damage to the parasites and protection from infections . Thus , our findings seem to best support the use of an ‘Achilles Heel’ ( also called "Waksman's postulate" ) approach [36 , 37] in which the vaccine target would be a host-exposed functional antigen , essential to worm survival , that is not naturally immunogenic . Antibodies generated to such targets through vaccination should block the function of this antigen , thereby damaging the parasite and protecting the host from future schistosome infections . A striking finding of these studies is the surprising predominance of antibodies in schistosome infection serum recognizing a limited set of reduction-sensitive epitopes , particularly those on SmTsp2 and Sm29 . These epitopes proved to be poorly reproduced when these antigens were expressed by E . coli . For example , Supplemental S1 Fig shows that the major antigens in schistosome extracts generate strong signals on western blots with rat infection serum despite loadings too low for detection by conventional protein staining . In contrast , a recombinant version of SmLy6-2 produces a barely detectable , reduction-sensitive band on an infection serum blot despite an easily detected protein band . Similar results were obtained with the other SmLy6 proteins , including Sm29 , and the recombinant proteins themselves were found to largely self-aggregate ( not shown ) . An exception to this was SmTsp2 in which the E . coli recombinant protein did produce a strong , reduction-sensitive band on a western blot with infection serum . As another example , we attempted to express Sm29 in insect cells by replacing the schistosome leader sequence with a leader sequence from insect melittin , and even this minor change resulted in a protein that was unrecognized by the Teg4 scFv and much more poorly recognized by infection sera ( not shown ) . These results highlight the importance of using conformationally-native immunogens when preparing antibodies for detection of schistosome tegumental antigens under natural conditions , such as for in situ localization or live worm staining . It is interesting to speculate whether our surprising finding of a strong apparent bias to a limited subset of reduction sensitive tegumental epitopes in the humoral response during schistosome infection has implications with regard to schistosome host immune evasion . The subject of immune evasion by schistosomes has been extensively investigated ( reviewed by [38] ) and the studies have suggested that schistosomes employ evasion strategies such as minimal protein exposure at the tegument surface , host antigen masking , epitope concealment by carbohydrates , poor immunogenicity of essential antigens , induction of blocking antibodies , secretion of immune modulators and rapid tegument turnover . Our results suggest that schistosomes may somehow elicit a strong antibody response to a limited set of surface antigens on which antibody binding is well-tolerated and perhaps even somehow beneficial to the parasite . For example , these anti-tegumental antibodies could serve as ‘blocking Abs’ [39] that prevent induction of a damaging antibody response . It is worth noting that any agent that induces a minor conformational change to the epitope targeted by these anti-tegumental antibodies will result in the dissociation of the antibody , and schistosomes may somehow exploit this feature to evade immune damage .
All studies followed the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and were approved by the Tufts University Institutional Animal Care and Use Committee ( IACUC ) under Protocol # G2015-113 . This protocol adheres to the National Institutes of Health’s Public Health Service Policy on Humane Care and Use of Laboratory Animals . The protocol for the use of human sera was previously approved by the Ethical Committee of the Federal University of Minas Gerais and the patients or their legal guardians provided written informed consent after receiving an explanation of the protocol . Horseradish peroxidase HRP-conjugated anti-E-tag antibodies were purchased from Bethyl Labs; HRP-conjugated anti-rat IgG , HRP-conjugated anti-mouse IgG , and HRP-conjugated anti-human IgG were purchased from Santa Cruz . The rat infection sera were obtained by performing two S . mansoni infections of non-permissive Fischer rats , each with 1 , 000 cercariae , performed 8 weeks apart as previously described [15] , and the serum was drawn four weeks after the second infection . Two preparations of mouse infection sera were employed , each obtained from two groups of 10 Swiss-Webster mice infected with 120 cercariae nine weeks earlier . All of the mice survived and contained adult worms at the time the serum was obtained . The pools of serum were prepared by mixing equal amounts of serum from each infected mouse . All human infection sera were obtained from individuals living in two different endemic areas for schistosomiasis ( “Melquiades” and “Côrrego do Onça” , Minas Gerais , Brazil ) . Four different pools of serum ( previously reported in [13] ) were prepared that each included serum from eight patients . Patients in the pools included sixteen that had egg counts at the time of infection ( eight that were recently treated with praziquantel ) , and sixteen that were stool negative despite known exposure to water contaminated with cercariae ( eight there were recently treated with praziquantel ) . These infected individuals were examined for S . mansoni infection using the Kato–Katz technique and were negative for other helminthic infections as previously described [13] . Five day old schistosomula were obtained from the Biomedical Research Institute ( Rockville , MD , USA ) . The schistosomula were produced by in vitro transformation of cercariae and five days of tissue culture following the procedure of [40] . Schistosomula were homogenized in PBS , briefly sonicated to shear the DNA , and diluted in non-reducing SDS buffer . Adult worms were recovered by perfusing infected Swiss Webster mice at 6–7 weeks post infection . Parasites’ extracts were prepared by homogenizing the adult parasites in PBS on ice . Extracts were briefly sonicated to shear the DNA then diluted in non-reducing SDS buffer . Preliminary Western blot experiments were run to determine the appropriate amount of extract to be used in subsequent experiments . The rat scFv proteins [15 , 19] were expressed in E . coli host cells as fusions with an amino terminal E . coli thioredoxin to facilitate folding , with a hexa-histidine tag for purification and a carboxyl terminal E-tag epitope for detection . The expression , purification and quantification were performed as previously described for VHH antibody proteins [41] . Synthetic DNA encoding SmLy6A , B , C , F , SmTsp-2 , Sm23 and Sm29 was ordered from GenScript Inc in which the codon usage is optimized by the manufacturer for insect expression . Restriction sites were included to facilitate simple ligation into the pFastBac vector ( Invitrogen ) . For SmLy6 proteins , the natural schistosome signal sequence ( predicted by SignalP ) was replaced with the insect melittin signal sequence ( MKFLVNVALVFMVVYISYIYA ) . Immediately downstream from the signal sequence , we included the following spacer coding sequence ( AADYKDDDDKGGGGS ) which includes a Not1 cloning site , FLAG epitope and an enterokinase cleavage site . The schistosome mature protein coding sequences followed through the carboxyl terminus . For SmTsp-2 , Sm23 and Sm29 , the synthetic coding DNA included the complete schistosome protein including leader sequence . We made an additional construction in which the natural signal peptide for Sm29 was replaced with the melittin signal sequence and the spacer sequence described above . The synthetic coding DNAs were ligated into pFastBac and engineered for insect cell expression using the Bac-to-Bac system ( Invitrogen ) following the procedures recommended by the manufacturer . Typically , the first virus supernatant was used to infect Sf9 cells at an estimated MOI of 0 . 1 and cells were harvested 3–5 days post-infection . Recombinant insect cell pellets were either solubilized directly into non-reducing SDS Laemmli buffer ( BioRad ) , or for use in western blot competition studies , we employed a non-reducing insect cell lysis buffer ( 1% Triton-X-100 , 10mM Tris pH 8 . 0 , 140 mM NaCl , 1x Sigma protease inhibitor cocktail ) . DNA encoding the predicted extracellular domains of SmLy6B , C , F , SmTsp-2 , and Sm29 was amplified by PCR and ligated into a pET32 E . coli expression plasmid in frame with the E . coli thioredoxin ( Trx ) coding DNA and containing hexahistidine and epitope tags . Coding DNA cloning , expression and purification of these recombinant schistosome proteins were all performed by standard methods as previously reported [41] . Proteins were resolved by SDS-PAGE using AnyKd mini-PROTEAN TGX gels from BioRad under reducing ( 1 . 5% βME unless otherwise specified ) or non-reducing conditions . Proteins were transferred from gels to Immobilon-P PVDF membranes ( Fisher ) using a BioRad semi-dry transfer apparatus . NEB Color Prestained Protein Standard , Broad Range ( 11–245 kDa ) , which contains only trace reducing agent , was used for estimating molecular weight and to guide the excision of filter strips . Membranes were blocked with 5% non-fat dry milk , 0 . 1% Tween-20 in PBS ( mPBS ) for 30 minutes at room temperature or overnight at 4°C . Primary antibody incubations ( dilutions indicated in figure legends ) were overnight at 4°C in fresh mPBS , and secondary antibody incubations ( as recommended by manufacturers ) were for one hour at room temperature , all followed by 3 washes , 5–10 minutes each , in PBS , 0 . 1% Tween-20 ( PBST ) . In some cases , the filters were incubated for 1 hour at room temperature in mPBS 0 . 1% β-mercaptoethanol . The resulting immunoblots were developed using the ECL Western Blotting detection system from GE Healthcare , as recommended by the manufacturer , and imaged on a BioRad ChemiDoc Touch instrument . | Schistosomiasis is caused by blood flukes residing in the veins of infected individuals and afflicts millions of people in the developing world . The schistosome worms can remain healthy in the bloodstream for more than 10 years , implying an extraordinary ability to evade host immune damage . Scientists are seeking to understand immune evasion so as to find weaknesses in defenses that can be exploited in the development of effective vaccines . Here we investigate the normal antibody response to schistosomes during infection of mice , rats and humans , and show for the first time that this response is highly skewed to the recognition of a small number of proteins present at the worm surface . Surprisingly , these abundant antibodies recognize their targets only when the proteins retain their native conformations , stabilized by the presence of intramolecular disulfide bridges . Because of this conformational-dependence , these antibodies have remained undetected in prior studies in which antibody binding assays were routinely performed in a reducing environment that destroys disulfide bridges . The routine presence of these antibodies within the serum of schistosome infected patients and animals raises new and interesting questions as to their possible role in immune evasion , and has significant implications for schistosomiasis vaccine development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"helminths",
"immunology",
"animals",
"antibodies",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"protein",
"extraction",
"proteins",
"recombinant",
"proteins",
"insects",
"arthropoda",
"biochemistry",
"antigen",
"processing",
"and",
"recognition",
"physiology",
"biology",
"and",
"life",
"sciences",
"serum",
"proteins",
"organisms"
]
| 2017 | Schistosoma mansoni Infection of Mice, Rats and Humans Elicits a Strong Antibody Response to a Limited Number of Reduction-Sensitive Epitopes on Five Major Tegumental Membrane Proteins |
The Gambia’s National Eye Health Programme has made a concerted effort to reduce the prevalence of trachoma . The present study had two objectives . The first was to conduct surveillance following mass drug administrations to determine whether The Gambia has reached the World Health Organization’s ( WHO ) criteria for trachoma elimination , namely a prevalence of trachomatous inflammation—follicular ( TF ) of less than 5% in children aged 1 to 9 years . The second was to determine the prevalence of trichiasis ( TT ) cases unknown to the programme and evaluate whether these meet the WHO criteria of less than 0 . 1% in the total population . Three cross-sectional surveys were conducted between 2011 and 2013 to determine the prevalence of TF and TT in each of nine surveillance zones . Each zone was of similar size , with a population of 60 , 000 to 90 , 000 , once urban settlements were excluded . Trachoma grading was carried out according to the WHO’s simplified trachoma grading system . The prevalence of TF in children aged 1 to 9 years was less than 5% in each surveillance zone at each of the three surveys . The prevalence of TT cases varied by zone from 0 to 1 . 7% of adults greater than 14 years while the prevalence of TT cases unknown to the country’s National Eye Health Programme was estimated at 0 . 15% total population . The Gambia has reached the elimination threshold for TF in children . Further work is needed to bring the number of unknown TT cases below the elimination threshold .
Trachoma , caused by the obligate intracellular bacterium Chlamydia trachomatis , is believed to be endemic in at least fifty countries , primarily in Africa and Asia [1] . The disease is characterized by repeated episodes of follicular conjunctivitis in childhood , resulting in progressive scarring of the upper eyelid , in-turning of the eyelashes ( trichiasis; TT ) and corneal opacification in later life [2] . Recognising that concerted action to control the disease was needed , the World Health Organization ( WHO ) , in 1996 , proposed the ‘SAFE’ strategy of trachoma control: Surgery to correct trichiasis , Antibiotics to treat ocular chlamydial infection , and the promotion of Facial cleanliness and Environmental Improvement to address risk factors associated with the transmission of the infection [3] . The Gambia , situated in the Sahel of West Africa , has done much to implement the SAFE strategy to control trachoma within its borders . In 1986 , following a national survey that determined trachoma was the second leading cause of blindness in the country [4] , the National Eye Health Programme ( NEHP ) was formed and a network of community ophthalmic nurses was trained to screen communities for active trachoma and to conduct trichiasis surgery . Public health initiatives with a focus on preventative eye care and facial cleanliness have been targeted to school children and rural communities while urban centres have benefitted from a targeted programme designed to meet the eye health needs of marginalised populations . In 2007 , a mass drug administration ( MDA ) programme , which distributed more than 400 , 000 doses of azithromycin , was rolled out in twenty-three priority health districts . Most recently , trichiasis case-hunting and surgery camps have been carried out across the rural regions of the country . In 1996 , a second survey of blindness found the prevalence of active trachoma in children aged 0 to 9 was approximately 7% nationally [5] . A decade later , population-based surveys conducted in two regions in advance of the MDA campaign found greater than 10% prevalence of trachomatous inflammation—follicular ( TF ) in the same age group [6] . Based on these data , eleven rural districts with predicted prevalence greater than 10% were earmarked for three rounds of MDA with azithromycin according to the WHO criteria . Treatment began in 2007 and finished in 2010 . In another twelve districts where the prevalence of TF in children was projected to be between 5 and 10% , a screen and treat strategy was adopted . Under this scheme , community-based screening of all children aged 1 to 9 years was first carried out . If the prevalence of TF in 1 to 9 year olds was greater than 10% , the community was selected for three rounds of MDA . If the prevalence was less than 10% , households of active cases were treated with one round of MDA . In 2011 , the Partnership for the Rapid Elimination of Trachoma , or PRET study , surveyed four formerly endemic districts by conducting screening in villages that had received either one round of MDA three years previously or three rounds of MDA completed one year previously . The results demonstrated very low levels of TF and ocular chlamydial infection in all communities surveyed and suggested the country might have reached the elimination benchmark of TF prevalence less than 5% in children [7] . The WHO has convened global scientific meetings on post-endemic surveillance for trachoma and these have established criteria for the elimination of trachoma as a public health problem [8 , 9] . These criteria indicate that , following three years of surveillance , countries would be required to demonstrate , through appropriately powered surveys of rural areas , that ( i ) the prevalence of TF among 1–9 year olds was less than 5% and ( ii ) that the prevalence of ‘unknown’ cases of trichiasis ( those that have never been offered surgery ) was less than one per thousand ( 0 . 1% ) of the total population in each district . A district was defined notionally as a unit of 250 , 000 population . With this advice in mind , the NEHP conducted a rolling programme of rural trachoma surveys over three years designed to show that TF prevalence remained low and that elimination criteria had been met . TT prevalence was also assessed and the results generated were used to calculate the number of TT cases unknown to the NEHP throughout the country .
The regions of The Gambia were divided , on the basis of their population in the 2003 national census , into nine zones ( A-I; Fig 1 ) . Gambian administrative districts are significantly smaller than those in many other countries and the zones were constructed to be functionally equivalent to ‘sub-districts’ by amalgamating multiple contiguous districts to give similar sized units of 60 , 000–90 , 000 population . Urban settlements , as designated by The Gambia Statistics Department [10] , were excluded as per the WHO recommendations . A sample size of 1 , 448 individuals for each population unit of 60 , 000 to 90 , 000 ( zones A to I ) was deemed necessary to estimate a prevalence of TF , in children aged 1 to 9 years , of 3% within a precision of +/- 2% given a 95% confidence limit and a design effect of 4 . As sampling was intended to proceed over 3 years , this equated to 483 individuals aged 1 to 9 to be examined per zone , per year . The average Gambian household was estimated to contain 9 people , of whom 3 . 5 are in the age range 1–9; to achieve the required sample size , 138 households would need to be screened in each zone , each year . We therefore chose to conservatively target 160 households annually in each of the surveillance zones . This was achieved by selecting 16 settlements in each zone by probability proportional to size , conducting a census to list all households in each settlement and then screening ten randomly selected households from the list in each settlement for ocular examinations . A further two households were randomly selected as reserve households in the case of selected households being unavailable or choosing not to participate . In the case of large settlements ( >1 , 000 population ) , segments of the village , defined on the basis of landmarks by the national census data , were selected for screening . Based on the 2011 and 2012 results , the sample size was doubled in 2013 in order to have TF prevalence data in all zones from a single survey adequately powered to demonstrate that TF was less than 5% in each zone and thus fulfil the criteria for elimination . Data collection took place from July 2011 to June 2013 . Prior to the onset of field-work , a workshop was held for all community ophthalmic nurses and healthcare workers involved in the surveys . Training included quality control exercises of trachoma grading from slides in a classroom setting followed by practical exercises in the field . Graders were required to achieve agreement ( kappa coefficients of 0 . 8 or more ) for TF with a senior grader ( RB ) for the slide grading exercise , which consisted of a variety of images from The Gambia , Niger and Tanzania and which were used in the validation system for the PRET study [7] . During the surveys , all household members , defined as those who slept in the household the night before , were eligible for screening . Consenting individuals were examined for clinical signs of trachoma using a 2 . 5× magnifying loupe and adequate sun or torch-light . Trachoma grading was carried out according to the WHO simplified grading system [11]; TF was defined as the presence of five or more follicles , greater than 0 . 5mm diameter , on the upper tarsal conjunctiva and TT was defined as at least one lash touching the globe , or evidence of epilation . Household data were summarised from paper survey forms as the number of children aged 1–9 , the number of cases of TF in children aged 1–9 , the number of individuals aged more than 14 and the number of TT cases in those over 14 in the household . These summary data were entered into a Microsoft Access database by NEHP staff and subsequently analyzed in Stata version 12 . 1 ( StataCorp LP , College Station , USA ) . Confidence intervals were calculated using the Wilson score interval [12] . All cases of TF along with their household contacts were treated with a single oral dose of azithromycin as is NEHP policy . Pregnant women and infants under the age of 6 months were instead offered tetracycline eye ointment . Individuals with TT were offered Trabut surgery free of charge . The study adhered to the tenants of the Declaration of Helsinki and was approved by The Gambia’s Ministry of Health and Social Welfare and by The Gambia Government/Medical Research Council Unit , The Gambia Joint Ethics Committee . Written , informed consent was obtained from all participants; in the case of minors , informed consent was obtained from the parent or guardian .
Over the course of the study period , we examined 19 , 205 children aged 1 to 9 years and 27 , 921 adults over the age of 14 years for trachoma . The prevalence of TF in children varied by zone in each of the three years ( Table 1 ) , ranging from 0 to 3 . 8% [95% Confidence Interval ( CI ) , 2 . 5–5 . 9] . In 2013 , the final year of the survey , TF prevalence ranged from 0 . 2% ( 95% CI 0 . 1–0 . 9 ) to 3 . 2% ( 95% CI 2 . 3–4 . 4 ) Prevalence of TT varied by year and by zone from 0 to 1 . 7% ( 95% 1 . 0–3 . 0 ) ( Table 2 ) . In the final year of surveillance , TT prevalence ranged , by zone , from 0 to 0 . 6% ( 95% CI , 0 . 3–1 . 1 ) . The projected number of trichiasis cases in each zone was estimated by first calculating the average TT prevalence in individuals aged over 14 and who were available to survey in each zone over the course of the three-year study . The average prevalence per zone was then multiplied by 60% of the rural population in each zone [60% being the estimated proportion of the Gambian population over 14 years of age and assuming that TT in young children is negligible ( no cases were observed among 19 , 205 children examined in the survey ) ] to determine an approximate number of projected cases living in each zone ( Table 3 ) . Using this method , the total number of projected cases across each of the nine zones was estimated to be 1 , 594 . In an effort to determine the number of trichiasis cases known to the NEHP , the trichiasis registries held at seven of the eight major health facilities with dedicated eye health units were examined . One registry was unavailable . 570 TT cases were recorded in these registries . We therefore estimate the total number of ‘unknown’ cases to be 1 , 024 ( 1 , 594 minus 570 ) , which represents 0 . 15% of the total rural population .
Data collected during three years of rolling surveys indicate The Gambia has achieved the elimination criteria for TF in children , representing a significant public health achievement . Further efforts in identifying and a more robust system of documenting TT cases would help reduce the number of unknown TT cases from the current figure of 0 . 15% total population to below the 0 . 1% population elimination threshold . | Trachoma , the world’s leading infectious cause of blindness , is caused by ocular infection with the bacterium Chlamydia trachomatis . The Gambia , situated in West Africa , has implemented all facets of the World Health Organization-recommended SAFE strategy for trachoma control including surgery to correct the in-turning of eyelashes ( trichiasis ) , mass drug administration with antibiotics , promotion of facial hygiene and environmental improvements . In 2011 , The Gambia’s National Eye Health Programme began three years of rolling surveys to determine the prevalence of trachoma in the country and to evaluate whether trachoma elimination has been reached . The results suggest the country has reached the elimination threshold for trachoma in children ( less than 5% prevalence ) but that more work needs to be done to reduce the prevalence of trichiasis in adults . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"geographical",
"locations",
"census",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"bacterial",
"diseases",
"research",
"design",
"eye",
"diseases",
"sexually",
"transmitted",
"diseases",
"global",
"health",
"neglected",
"tropical",
"diseases",
"eyes",
"africa",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"chlamydia",
"infection",
"head",
"people",
"and",
"places",
"survey",
"research",
"anatomy",
"gambia",
"ophthalmology",
"biology",
"and",
"life",
"sciences",
"ocular",
"system",
"trachoma",
"ophthalmic",
"procedures"
]
| 2016 | Cross-Sectional Surveys of the Prevalence of Follicular Trachoma and Trichiasis in The Gambia: Has Elimination Been Reached? |
The four Rep proteins of adeno-associated virus ( AAV ) orchestrate all aspects of its viral life cycle , including transcription regulation , DNA replication , virus assembly , and site-specific integration of the viral genome into the human chromosome 19 . All Rep proteins share a central SF3 superfamily helicase domain . In other SF3 members this domain is sufficient to induce oligomerization . However , the helicase domain in AAV Rep proteins ( i . e . Rep40/Rep52 ) as shown by its monomeric characteristic , is not able to mediate stable oligomerization . This observation led us to hypothesize the existence of an as yet undefined structural determinant that regulates Rep oligomerization . In this document , we described a detailed structural comparison between the helicase domains of AAV-2 Rep proteins and those of the other SF3 members . This analysis shows a major structural difference residing in the small oligomerization sub-domain ( OD ) of Rep helicase domain . In addition , secondary structure prediction of the linker connecting the helicase domain to the origin-binding domain ( OBD ) indicates the potential to form α-helices . We demonstrate that mutant Rep40 constructs containing different lengths of the linker are able to form dimers , and in the presence of ATP/ADP , larger oligomers . We further identified an aromatic linker residue ( Y224 ) that is critical for oligomerization , establishing it as a conserved signature motif in SF3 helicases . Mutation of this residue critically affects oligomerization as well as completely abolishes the ability to produce infectious virus . Taken together , our data support a model where the linker residues preceding the helicase domain fold into an α-helix that becomes an integral part of the helicase domain and is critical for the oligomerization and function of Rep68/78 proteins through cooperative interaction with the OBD and helicase domains .
The four adeno-associated virus ( AAV ) Rep proteins are generated from a single open reading frame by the transcriptional use of two different promoters ( p5 and p19 ) and subsequent alternative splicing mechanisms [1] , [2] , [3] . These reactions produce proteins that share three functional domains: an origin binding domain ( OBD ) , a SF3 helicase domain and a putative zinc-finger domain [4] , [5] . The combination of these domains imparts these proteins with striking multifunctionality . In particular , the larger proteins Rep78 and Rep68 function as initiators of DNA replication , transcriptional regulators , DNA helicases and as key factors in site-specific integration [6] . The smaller Rep proteins Rep40 and Rep52 , play a critical role during packaging of viral DNA into preformed empty capsids , where they are thought to be part of the packaging motor complex [7] , [8] , [9] . Although in terms of domain architecture the AAV Rep proteins resemble other members of the SF3 protein family , the peculiar OBD with its additional nuclease activity and the complex character of their oligomeric properties , set them apart from other SF3 helicases such as simian virus 40 large T antigen ( SV40-LTag ) and papilloma virus E1 ( PV-E1 ) proteins [10] , [11] , [12] , [13] . In both of these proteins , the minimal SF3 helicase domain assembles into a hexameric ring in a process that can be induced by the presence of ATP and/or single-stranded DNA [14] , [15] . In contrast , Rep40 containing only the helicase domain and Rep52 with an additional Zn-finger domain , appear to be monomeric [16] , [17] . This indicates that oligomerization of AAV Rep proteins requires the presence of both the OBD domain and the helicase domain . This combination imparts both Rep68 and Rep78 with a complex and dynamic oligomeric behavior in-vitro that is modulated in large part by the nature of the DNA substrate [18] . The monomeric behavior of both Rep40 and Rep52 is striking in that they appear to contain the required structural features that are present in other SF3 helicase members . The X-ray structures of both SV40-LTag and PV-E1 show that their helicase domains assemble as hexameric rings and that the oligomerization interface is bipartite [15] , [19] . One interface is formed by the interaction of neighbouring N-terminal oligomerization domains ( OD ) . The second interface is formed by the interaction of the C-terminal AAA+ domains and is further stabilized by the presence of nucleotides [11] , [15] . In order to understand the structural features that promote AAV Rep oligomerization , we pursued in this study a detailed structural comparison of SF3 helicases . We show that the OD domain in Rep40/52 has been hindered in its ability to oligomerize by the transcriptional use of the p19 promoter . This event generates proteins with a smaller OD domain as compared to other SF3 helicases . More importantly , we show that in the context of Rep68/78 the required oligomerization is supported by the interdomain linker which is directly involved in oligomerization interface and we provide evidence that the tyrosine residue preceding the start of Rep40/52 ( Y224 ) is critical in the oligomerization and therefore activity of the large AAV Rep proteins . Taken together , our results support a model where oligomerization of Rep68/78 is mediated by a composite oligomerization interface formed by the OBD , helicase and linker domains , with the latter playing an essential role in the inducing the oligomerization process .
As a first step in our attempt to determine the structural features that promote oligomerization in AAV Rep proteins , we analyzed the oligomeric interface of SF3 family members SV40-LTag and PV-E1 . As previously described , the helicase domain contains two subdomains: a N-terminal helical bundle of four α-helices known as the oligomerization domain ( OD ) and the C-terminal AAA+ subdomain ( Figure 1A ) . In PV-E1 the oligomerization interface spans both subdomains forming two extended surfaces at opposite faces of the proteins . In the AAA+ subdomain , one face comprises all the catalytic residues , including: the P-loop , its subsequent helix , the β-strands with the associated Walker B residues , sensor 1 motif , and one side of the β-hairpin ( Figure 1B ) . The neighboring subunit interacts through areas that are located in the α-helices “behind” the β-sheet and on the opposite side of the β-hairpin ( Figure 1B ) . Overall , about 20% of the solvent accessible area takes part in the interface and includes about 34% of all residues . In PV-E1 , the OD domain consists of 68 residues forming a four helical bundle . The oligomeric interface comes from interaction of residues located in helices 1 and 4 in one monomer , with residues in helices 2 , 3 and part of helix 4 in the other subunit ( Figure 1B ) . Most of the interface is hydrophobic with many tyrosine and isoleucine residues . Similar types of interactions are seen in the interface formed by the SV40-LTag OD domains . This domain is a lot bulkier , spanning 89 residues that form a five-helix bundle . The extra helix originates from an additional Zn-finger motif . Significantly , the OD of Rep40 , on the other hand , has only 52 aminoacids and , thus , is significantly shorter than PV-E1 and SV40-LTag OD domains . The direct result of this difference is a decrease in the total accessible surface area by more than 1000 Å2 . In addition , the packing of the helices is less compact , producing a more dynamic structure ( Figure 1C ) . We hypothesize that the smaller OD domain of AAV Rep proteins imparts these proteins unique oligomeric properties where the smaller Rep40/52 are mostly monomeric while Rep68/78 -with the additional OBD domain- form oligomers . However , the measurable ATPase activity in all Rep proteins , suggest that Rep40/52 should oligomerize in the presence of nucleotides [20] . To determine if the presence of nucleotides can induce oligomerization of Rep40 -containing the minimal helicase domain- , we carried out sedimentation velocity experiments in the presence and absence of nucleotides at different concentrations . The sedimentation velocity profiles offer a complete characterization of the number and type of oligomers in solution . The data were analyzed using the program sedfit [21] , [22] . Figure 2A shows plots of the c ( s ) distribution against the sedimentation coefficient ( s ) for two concentrations of Rep40 in the absence of nucleotides . A single peak whose s20 , w increases slightly with increasing concentrations is observed . The slight but significant increase in s and calculated molar mass is consistent with a weak and transient dimerization ( for hydrodynamic reasons , s is expected to decrease with increasing concentrations of an ideal solute ) . The data where also fitted using the program sedphat to a monomer-dimer association were the process is in rapid exchange on the time scale of the centrifuge [22] . Table 1 shows that the dissociation constant in the absence of nucleotides is ∼10−3 M , which is at the upper end of detection by sedimentation velocity . Similar distributions of Rep40 ( at 36 µM ) in the presence of either 5 mM ATP or ADP are shown in Figure 2B and 2C . Here an increase is observed in the width of these peaks if compared to those for Rep40 alone . This is a well-understood behavior for a associating system whose exchange kinetics are neither slow of fast on the time scale of the centrifuge , thus , broadening the c ( s ) distribution peak [23] . The presence of a small shoulder suggest that dimer formation is occurring here as well , although perhaps its rate of dissociation is slower than for Rep40 alone . The s-value of the shoulder is consistent with a transient Rep40 dimer that represents ∼0 . 2% of the total amount of protein . The relatively low ATPase activity of Rep40 reported in the literature supports our model of transient dimerization promoted by the binding and/or hydrolysis of ATP [20] . In order to assess whether the interdomain linker connecting the OBD domain and the helicase domains contains additional regions of distinct structure that may play a role in promoting oligomerization , we first carried out secondary structure prediction analysis to determined if the linker contains additional regions of structure . The results suggest that the region from residue 215 to 224 has the potential to form an α-helix ( Figure 3A ) . We hypothesized that this region could extend the first helix of the OD domain ( Figure 3A ) and the ensuing increase in surface accessible area may be sufficient to drive oligomerization . To test this hypothesis , we designed a new Rep construct beginning at the start of the linker region and extending to aminoacid 536 ( a truncated version of Rep68 without the OBD domain , Rep68Δ200 ) , and performed sedimentation velocity and cross-linking studies in order to characterize its oligomerization properties . The sedimentation profile of Rep68ΔN200 shows the presence of two peaks , one corresponding to the monomeric species ( ∼2 . 53S ) and the other to a dimer ( ∼3 . 71S ) . The amount of formed dimer increases at higher concentrations as expected from a monomer-dimer equilibrium system ( Figure 3B ) . Formation of dimers was also observed when we performed cross-linking experiments . Figure 3C shows that the amount of dimeric species has significantly increased in Rep68ΔN200 as compared to Rep40wt . We calculated the dimerization constants of Rep40wt and Rep68ΔN200 from a global fitting of the sedimentation velocity data to a monomer-dimer model ( Table 1 ) . In summary , we determined that the presence of the linker region increases the strength of dimerization by about 10-fold relative to that of Rep40 . Next , we sought to determine the minimal length of linker that is needed to promote oligomerization . We generated three additional constructs , named Rep68ΔN209 , Rep68ΔN214 and Rep68ΔN219 and tested their ability to oligomerize ( Figure 4 ) . Our results indicate that Rep68ΔN214 contains the minimal length of linker that is required to promote detectible oligomerization , although with the shorter construct Rep68ΔN219 , a small shoulder is seen at higher concentration ( data not shown ) . These results confirm that the linker region from 215 to 224 may fold into a α-helix , resulting in an increase of the surface accessible area of the OD domain that mediates oligomerization . This increase , however , is not sufficient to produce higher order oligomers . In order to determine the contribution of ATP and ADP to the oligomerization of the extended linker Rep linker constructs , we performed sedimentation velocity studies in the presence of nucleotides . Our hypothesis was that if oligomerization reflects the functional state of these proteins , the addition of nucleotides should support and induce further oligomerization . Figure 5 shows that the presence of ATP and ADP induces the formation of higher order oligomers . Formation of dimeric species at this concentration can be seen with Rep68Δ214 as well as the longer constructs RepΔN209 and RepΔN200 . In the later two , ADP produces two main populations sedimenting at ∼3S and ∼7S with additional intermediate oligomers . ATP on the other hand , seems to generate more stable species at ∼7S . Again , these data show that the presence of the linker region induces oligomerization of the Rep constructs and that the addition of nucleotides , in particular ATP , induces formation of larger oligomers , possibly through the stabilization of the interface formed by the AAA+ domains . This finding is in good agreement with the unique characteristics of the AAV Rep nucleotide binding pocket , which , based on its open conformation together with the presence of an arginine finger predicts the nucleotide contribution to oligomerization [24] . To determine if the linker is critical for the oligomerization of Rep68 , we replaced it with an unrelated sequence and examined its effect on oligomerization using sedimentation velocity . The only prerequisite for the substitute linker were a lack of structure and no impact on the native structures of the connected domains . We chose a sequence from the transcription factor Oct-1 . This transcription factor has two DNA binding domains connected by a linker of 29 residues . The X-ray structure of this protein shows that the linker is unstructured and flexible . In addition , it has been used to connect different protein domains without affecting their properties [25] , [26] . We generated a Rep68 mutant protein ( Rep68octlink ) , where residues 206 to 224 were replaced with 18 residues from the Oct-1 linker and tested its ability to oligomerize . The sedimentation profile of Rep68 typically shows two populations with sedimentation coefficients of ∼3S and ∼13S ( Figure 6A ) . We have determined that the 13S peak corresponds to a mixture of oligomeric rings ( data not shown ) . Figure 6B shows that the replacement of the linker completely abolishes the oligomerization of the mutant protein Rep68octlink . We can detect formation of dimeric species only at the highest concentration tested and in the presence of ATP , ( Figure 6C ) . These results show that replacement of the linker produces a Rep68 protein whose ability to oligomerize has been severely affected . The above findings indicate that the linker region plays a central role in the oligomerization of AAV Rep proteins . To confirm that the linker region has an intrinsic property to induce oligomerization , we generated a construct that spans the OBD domain and the linker region ( OBD-linker residues 1–224 ) and measured its ability to oligomerize . We first analyzed the OBD domain ( 1–208 ) to determine any oligomerization up to concentrations of 1 mg/ml ( 43 µM ) . Our results show that while OBD is a monomer ( Figure 7A ) , the OBD-linker protein construct displays formation of dimers at increasing protein concentrations ( Figure 7B ) . These results support the hypothesis that the linker region has an intrinsic property to induce oligomerization We generated a model of the Rep68ΔN214 construct using the X-ray structure of Rep40 ( residues 225–490 ) and 9 residues of the linker ( 215–224 ) that were added as a helical extension to the N-terminus . The model of the α-helix was generated using Robetta [27] . Figure 8A and 8B shows the structural alignment of the OD domain of the Rep68ΔN214 model with the OD domains of PV-E1 and SV40-LTag . The alignment shows that residue Y224 superimposes with aromatic residues F313 and W270 located at the beginning of helix 1 in the OD domains of PV-E1 and SV40-LTag respectively . Analysis of the structures of both proteins reveals that these aromatic residues play a critical role in forming and stabilizing the oligomerization interface . They pack against both the N-terminal end of helix 4 of the same subunit and the C-terminus end of helix 4 of the neighboring subunit . In order to test the hypothesis that Y224 plays an equivalent role in AAV Rep proteins , we mutated it to alanine and tested its effect on the oligomerization of Rep68ΔN200 . Mutation to the smaller residue alanine should have a direct effect in the oligomerization of this protein because of the significant reduction of surface exposed area . Figure 8C shows the sedimentation profile of this mutant protein showing that it completely abolishes the formation of dimers . To confirm that residue Y224 plays an important role in the oligomerization of AAV Rep proteins , we generated a Rep68Y224A mutant and compared its ability to form oligomers with respect to wild type Rep68 . Analysis of the Rep68Y224A mutant reveals that at low concentration the protein is mostly found as a monomer with a sedimentation coefficient of ∼3S . At higher concentrations , we observed the appearance of multiple peaks that correspond to dimers , trimers and larger oligomers; nevertheless , the majority of the protein is present as a monomer . The presence of ATP induces a small degree of stability to the dimeric species at 5 µM and both the 5S and 11S species at 10 µM . However , the 13S complex observed with the wild type Rep68 is not formed and most of the protein is still found as a monomer ( Figure 8E ) . These results indicate that residue Y224 is critical for the oligomerization of AAV Rep proteins . To assess if the disruption of oligomerization observed with the Rep68Y224A mutant has any consequences on the AAV viral life cycle , we produced recombinant AAV2 particles expressing the GFP gene in presence of a helper virus containing the Y224A mutation in the Rep ORF . The cells were harvested and lysed , and the crude lysate ( treated with an endonuclease ) was used to infect Hela cells . Strikingly , the crude lysate from cells transfected with the mutant helper plasmid didn't contain any infectious rAAV2-GFP particles , as determined by FACS analysis of GFP positive cells ( Figure 9 ) . These results show that the residue Y224 of AAV Rep proteins , and the oligomeric properties it confers to these proteins , have a crucial role during the AAV life cycle .
In this study we report that the interdomain linker present in the larger AAV Rep68/78 proteins is an integral part of their oligomerization interface . We showed that the linker region is in fact an extension of the OD domain of AAV Rep proteins . Our results have shown that Rep40 constructs containing either a complete or half linker have the ability to oligomerize . This effect is enhanced in presence of ATP or ADP . We hypothesized that the linker region from residues 215 to 224 forms a α-helix that is connected to the first α-helix of the SF3 helicase domain . Secondary structure prediction and modeling of the linker region supports this argument ( Figure 3A and 8B ) . Furthermore , we have identified a critical aromatic residue ( Y224 ) located at the end of the linker region that is conserved in Rep proteins from all AAV serotypes . The bulky nature of this aromatic residue appears to be a conserved feature in SF3 helicases ( Figure 8A ) . Structural alignment of the OD domain of a Rep40 model with an extended helical linker and those of SV40-LTag and PV-E1 shows that residue Y224 aligns with equivalent aromatic residues Trp270 and Phe313 respectively ( Figure 8A , 8B ) . A detailed analysis of the oligomeric interface of these proteins shows that these aromatic residues have a dual role: they stabilize the hydrophobic core of the OD domain helical bundle , and are part of the oligomerization interface between neighboring subunits . Our results reveal the critical role of the OD domain in the formation of stable oligomers in SF3 helicases . The larger OD domains of SV40-Tag and PV-E1 proteins in cooperation with the AAA+ motor domain generate a helicase domain that forms stable hexamers . Constructs of SV40-LTag and PV-E1 without the OD domain fail to oligomerize [14] , [19] . Another example that shows the fundamental role of the OD domain in oligomerization comes from the study of the evolutionary related proteins involved in rolling circle replication ( RCR ) of plasmids . The protein RepB from streptococcal RCR plasmid pMV158 is a hexameric protein that initiates replication of plasmid DNA and has a domain structure that resembles SF3 helicases but lacks the AAA+ subdomain [28] . Its N-terminal OBD domain is structurally and functionally related to the OBD from AAV Rep proteins due to the presence of the HUH motif critical for DNA nicking . Its C-terminal domain only consists of a 4 helical bundle that is similar to the OD domains of SF3 helicases and is responsible for hexamerization . Structural alignment shows that RepB has an aromatic residue ( Phe143 ) equivalent to residue Y224 in AAV Rep68/78 . We hypothesize that the role of this residue has been conserved throughout evolution to serve as a modulator of oligomerization in SF3 helicases and related RCR proteins . The smaller AAV Rep proteins Rep40/52 with truncated OD domains are missing the Y224 residue and thus are not able to sustain a stable oligomerization interface and are mostly monomeric . Consequently , the stable oligomerization of AAV Rep proteins requires the cooperative interaction of the OBD domain , the linker and the helicase domain . In this context , the OD sub-domain , and in particular the aromatic residue at the C-terminus of linker , appear to be the triggering element required for the oligomerization of AAV Rep proteins . The critical role of residue Y224 in the overall AAV-2 viral life cycle is illustrated by the complete abolishment of production of infectious particles from AAV-2 vector constructs produced in the context of Rep carrying the Y224A mutation ( Figure 9 ) . This result prompts the question of which specific functions are affected by this mutation . We think that most of the biochemical activities of Rep68/78 will be affected due to the impairment in oligomerization . Remarkably , an earlier report by Walker et al . on the identification of residues necessary for site-specific endonuclease activity showed that a Y224 mutant was defective in AAV hairpin/DNA binding , trs endonuclease , DNA helicase and ATPase activity [29] , suggesting that correct oligomerization of Rep proteins may be important in all of these functions . In agreement with our results , a recent report has shown that the presence of the linker in an AAV5 Rep40 construct induces oligomerization in presence of DNA . However , the authors concluded that the linker effect is primarily due to its interaction with DNA [30] . As we demonstrated in this report , the oligomerization effect is an intrinsic property of the linker due to its critical role in the formation of an oligomerization interface as part of the OD domain . The presence of DNA induces further oligomerization as seen with all helicases [13] . However , it appears that the linker also plays an additional role in protein-DNA interaction that may be important during the assembly of Rep68/78 on DNA substrates such as the AAV origin of replication and AAVS1 integration site . The use of alternative gene promoters is a common mechanism to generate protein diversity and flexibility in gene expression . At the same time it allows to obtain multiple functions from a limited number of genes , thus optimizing the size of the genome . It is clear that in the case of the Rep proteins from the AAV virus , nature has generated two sets of proteins that differ primarily in their ability to oligomerize . Rep proteins obtained from the AAV P19 promoter generate Rep40 and Rep52 with truncated OD domains and are thus unable to oligomerize . Both proteins play a critical role during DNA packaging into capsids; however , the mechanism of action of monomeric Rep40/52 during packaging remains elusive . Rep proteins generated from the P5 promoter , on the other hand , require the cooperative interaction of three different oligomeric interfaces produced by the OBD domain , the linker and the helicase domain . This feature potentially provides an additional dimension for the regulation of the diverse Rep activities when compared to the related proteins from SV40 and PV . We suggest that the cooperative interactions and the modulation of these interfaces – in particular in the presence of various specific DNA substrates – orchestrate the variety of functions performed by Rep68/Rep78 proteins and may thus represent a key to our understanding of the underlying mechanisms . Finally , our report introduces the possibility of two distinct helicase modes for the biological functions supported by AAV Rep proteins . In the context of the large Rep proteins , a complete OD domain directs the formation of stable oligomers with a DNA unwinding mode likely to resemble that of the related viral proteins SV40-Tag and E1 . The small Rep proteins , however , appear to utilize an incomplete OD domain that retains Rep40/52 in a monomeric state with formation of transitional dimeric complexes required for ATP hydrolysis . It is intriguing to speculate that this unique arrangement allows AAV to utilize two distinct motor activities with a single AAA+ domain . As Rep40/52 have been demonstrated to be required for genome packaging it is feasible to address the question whether this process requires a Rep40/52–mediated dimeric DNA helicase activity by a mechanism that is as yet undiscovered or whether further oligomerization is induced by interaction with capsid proteins .
All mutant proteins were generated using the pHisRep68/15b plasmid , which contains the AAV2 Rep68 ORF subcloned in vector PET-15b ( Novagen ) . Site-directed mutagenesis for mutants Y224A was generated using the QuickChange mutagenesis kit ( Stratagene ) . Rep constructs with different linker extensions were generated by PCR with primers designed to encompass the particular protein region . Primers included restriction enzyme sites NdeI and XhoI , and the sequence of the TEV protease site . The Rep68 protein used in these studies contained a Cys to Ser mutation that prevented aggregation but was functionally identical to the wild type protein ( data not shown ) . The Rep68octlink construct was generated by substitution of residues 206 to 224 of AAV2 Rep68 with the mouse Oct-1 linker residues 328–346 ( GeneBank CAA49791 ) using the gene synthesis services from GeneScript . The sequences of all constructs were confirmed by DNA sequencing ( GeneWiz ) . All proteins were expressed using the pET-15b vector , expressed in E . coli BL21 ( DE3 ) cells ( Novagen ) , and purified as described before [18] . The final buffer contains ( 25 mM Tris-HCl [pH 8 . 0] , 200 mM NaCl , and 2 mM TCEP ) . His6-PreScission Protease ( PP ) was expressed in BL21 ( DE3 ) -pLysS at 37°C for 3 h , in LB medium containing 1 mM IPTG . Cell pellets were lysed in Ni-Buffer A ( 20 mM Tris-HCl [pH 7 . 9 at 4°C] , 500 mM NaCl , 5 mM Imidazole , 10% glycerol , 0 . 2% CHAPS , and 1 mM TCEP ) . After five 10-s cycles of sonication , the fusion protein was purified using a Ni-column – equilibrated in Ni-buffer A . Protein eluted was desalted using buffer A and a HiPrep™ 26/10 desalting column ( GE Healthcare ) . His-PP tag was removed by PreScission protease treatment using 150 µg PP/mg His-PP-Rep68 . After overnight incubation at 4°C , buffer was exchanged using the same desalting column and Ni-Buffer A . Subsequent Ni-column chromatography using the buffer B ( same as buffer A but with 1 M imidazole ) , was performed to remove the uncleaved fusion protein , and untagged Rep68 was eluted with 30 mM imidazole . Rep68 was finally purified by gel filtration chromatography using a HiLoad Superdex 200 16/60 column ( GE Healthcare ) and Size Exclusion buffer . N-terminus His6-tagged WT and mutant Rep68 proteins were concentrated to 10 mg/ml , flash-frozen in liquid N2 , and kept at −80°C until use . The cross-linking reactions for Rep40 and Rep68ΔN200 were made according to an adapted protocol from Packman and Perham [31] . The reaction mixture was in cross-linking buffer ( 25 mM HEPES , 200 mM of NaCl , pH 8 . 0 ) and protein concentration was 2 mg/ml . A 30 fold molar excess of 100 mM DMP ( dimethyl pimelimidate dihydrochloride , MP Biomedicals , LLC ) was added to the reaction and incubated 60 min at room temperature . The reaction was quenched by addition of 1 M Tris , pH 7 . 5 to a final concentration of 50 mM . The samples were analyzed in an 8% SDS-PAGE . Hek 293T cells were triple transfected using polyethylenimine ( PEI ) with an AAV2 ITR-containing plasmid including the GFP gene , a helper plasmid expressing AAV2 Rep ( wt or Y224A cloned from the pHisRep68Y224A/15b ) and Cap , and a third construct containing the adenovirus helper functions ( pXX6 , University of North Carolina Vector Core Facility ) . The presence of the Y224A mutation was confirmed by sequencing ( Eurofins ) . After 72 h , the cells were harvested and lysed in 150 mM NaCl , 50 mM Tris at pH 8 . 5 , followed by three freeze - thaw cycles . The lysate was treated for 30 minutes at 37°C with 150 units/ml of benzonase endonuclease ( Sigma ) . HeLa cells were infected with increasing amounts of crude lysate , and the percentage of GFP-positive cells was determined three days post-infection . Sedimentation velocity experiments were carried out using a Beckman Optima XL-I analytical ultracentrifuge ( Beckman Coulter Inc . ) equipped with a four and eight-position AN-60Ti rotor . Rep protein samples were loaded in the cells , using in all cases buffer used in the final purification step . Samples in double sector cells were centrifuged at 25 , 000 rpm for Rep68 proteins ( Rep68 and Rep68Y224A ) . For Rep40 and linker constructs sedimentation was performed at 40 , 000 rpm . In all experiments , temperature was kept at 20°C . Sedimentation profiles were recorded using UV absorption ( 280 nm ) and interference scanning optics . For the analysis of the results the program Sedfit was used to calculate sedimentation coefficient distribution profiles using the Lamm [21] . Structures of AAV-2 Rep40 ( 1S9H ) , Bovine papillomavirus E1 protein ( 2GXA ) , Simian virus 40 T large antigen ( 1SVO ) and plasmid pMV158 RepB ( 3DKY ) were analyzed using the programs COOT [32] , PYMOL [33] and CHIMERA [34] . Structural alignment was done using the DALI server [35] . Secondary structure prediction was performed using PredictProtein [36] . Modeling of the linker region was done using ROBETTA [27] . | Viruses have to optimize the limited size of their genomes in order to generate the proteins required for infection and replication . Several mechanisms are used to accomplish this including the use of multiple promoters and alternative splicing . These processes generate gene products with diverse functions through the combinatorial assembly of a small number of protein domains . The small genome of the adeno-associated virus has two major open reading frames that generate seven proteins , four non-structural Rep proteins and three capsid proteins . The non-structural Rep proteins share a motor domain that uses hydrolysis of ATP to generate the conformational changes that drive DNA replication , transcriptional regulation , site-specific integration and the packing of viral genome into capsids . These functions depend upon the oligomerization of Rep proteins on specific DNA sites through the cooperation of the N-terminal origin binding domain and the C-terminal helicase domain . We provide evidence that the linker that connects the two domains is an integral feature of the helicase domain and contains a conserved aromatic residue that is critical for oligomerization . This residue emerges to be a signature motif of SF3 helicases and is also present in a subset of bacterial Rep proteins that support rolling circle replication mechanism . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biomacromolecule-ligand",
"interactions",
"enzymes",
"macromolecular",
"assemblies",
"viral",
"enzymes",
"microbiology",
"viral",
"preintegration",
"complex",
"proteins",
"biology",
"viral",
"replication",
"complex",
"viral",
"replication",
"biophysics",
"biochemistry",
"virology",
"computational",
"biology",
"macromolecular",
"structure",
"analysis"
]
| 2012 | The Interdomain Linker of AAV-2 Rep68 Is an Integral Part of Its Oligomerization Domain: Role of a Conserved SF3 Helicase Residue in Oligomerization |
Leptospirosis is an important zoonotic disease worldwide . Humans usually present a mild non-specific febrile illness , but a proportion of them develop more severe outcomes , such as multi-organ failure , lung hemorrhage and death . Such complications are thought to depend on several factors , including the host immunity . Protective immunity is associated with humoral immune response , but little is known about the immune response mounted during naturally-acquired Leptospira infection . Here , we used protein microarray chip to profile the antibody responses of patients with severe and mild leptospirosis against the complete Leptospira interrogans serovar Copenhageni predicted ORFeome . We discovered a limited number of immunodominant antigens , with 36 antigens specific to patients , of which 11 were potential serodiagnostic antigens , identified at acute phase , and 33 were potential subunit vaccine targets , detected after recovery . Moreover , we found distinct antibody profiles in patients with different clinical outcomes: in the severe group , overall IgM responses do not change and IgG responses increase over time , while both IgM and IgG responses remain stable in the mild patient group . Analyses of individual patients’ responses showed that >74% of patients in the severe group had significant IgG increases over time compared to 29% of patients in the mild group . Additionally , 90% of IgM responses did not change over time in the mild group , compared to ~51% in the severe group . In the present study , we detected antibody profiles associated with disease severity and speculate that patients with mild disease were protected from severe outcomes due to pre-existing antibodies , while patients with severe leptospirosis demonstrated an antibody profile typical of first exposure . Our findings represent a significant advance in the understanding of the humoral immune response to Leptospira infection , and we have identified new targets for the development of subunit vaccines and diagnostic tests .
Leptospirosis causes over one million cases and nearly 60 , 000 deaths annually , with the greatest disease burden in urban slums in tropical and subtropical countries [1–3] . Ten pathogenic Leptospira species , over 200 serovars , and a large number of mammalian reservoirs , including rats , have facilitated the emergence of leptospirosis as a major , global public health problem . Humans typically become infected through direct contact with reservoir urine-contaminated soil or water , and develop a broad spectrum of clinical manifestations , including hepato-renal failure and pulmonary hemorrhage syndrome in severe cases , which have high mortality rates [2 , 4–6] . The factors contributing to disease severity remain poorly understood , but bacterial virulence , inoculum dose and the host immune response are thought to play important roles in development of severe outcomes [2 , 4] . Experimental animal models of Leptospira infection have provided a majority of evidence that antibodies play a key role in protection against and clearance of Leptospira infection [7–9] . Passive transfer of whole cell leptospiral vaccine and specific anti-leptospiral antibodies ( Ligs ) are protective against homologous infection in animal models , demonstrating antibodies are sufficient for immunity against experimental homologous infection [10–13] . Additionally , antibodies against LPS are serovar-specific , are correlated with agglutinating antibody titers , and confer limited cross-protection against other serovars [14 , 15] . Several studies have shown that leptospirosis patients develop a robust antibody response during infection , especially anti-LPS antibodies , which correspond to the majority of the antibodies produced [12 , 16 , 17] . The large number of pathogenic Leptospira serovars and poor cross-protection observed for anti-LPS antibodies , have made the identification of anti-Leptospira protein antibodies a high priority for vaccine and diagnostic test development [18 , 19] . In support of this , immunization with an LPS-deficient Leptospira strain in experimental animal models conferred cross-protection , implicating anti-protein and other immune responses in protection against infection . [19] Additionally , our group has applied a protein microarray methodology to evaluate the antibody repertoire generated in natural Leptospira infection and identified strong antibody responses in healthy exposed individuals as well as several IgG serodiagnostic antigens specific to patients [20 , 21] . Analyses of antibody immune responses against infectious agents are essential not only for diagnostic and vaccine development , but also to providing insight in the mechanisms involved in pathogenicity [22] . Protein arrays are an excellent platform that allow for the screening of antibody protein targets in a high-throughput manner , with high sensitivity and high specificity [22–24] . These elements facilitate the assessment of many analytes simultaneously and allow for the identification , quantification and comparison of individual antigenic responses following exposure to microorganisms . Our group has efficiently employed high-density proteome arrays in the characterization of antibody signatures against several infectious agents of human and veterinary importance [25–30] , including Leptospira interrogans and other spirochetes [21 , 31] . In the current study , we used a whole genome proteome microarray approach to describe the first comprehensive profile of the human antibody response to symptomatic Leptospira infection . We probed 192 serum samples including patients with different clinical outcomes and healthy controls , and compared their antibody profiles against L . interrogans serovar Copenhageni proteins , the serovar associated with >90% of the urban leptospirosis cases in Salvador , Brazil [32 , 33] . We identified promising candidates for the development of new diagnostic tests and subunit vaccines and discovered different antibody profiles , which associated with disease severity . Lastly , the antibody kinetics suggest a majority of patients with severe leptospirosis likely have a primary infection , while those with milder disease have evidence of a secondary infection . Our results provide novel insights into the complexity of the immunity in naturally-acquired leptospirosis as well as new diagnostic test candidates .
The study protocol was approved by the institutional review board committees of Yale University and Oswaldo Cruz Foundation prior to study initiation . All participants provided written informed consent in their native language prior to sample and data collection . All samples were anonymized before research use . All 61 patient samples were collected during active surveillance for acute leptospirosis at the Hospital Couto Maia ( 31 severe group patients ) and the São Marcos Emergency Clinic ( 30 mild group outpatients ) in Salvador , Brazil between years 2005–2011 . Laboratory confirmation was defined as positive microagglutination test ( seroconversion , four-four rise in titer , or single titer ≥ 1:800 ) and/or positive ELISA and/or positive PCR for Leptospira DNA as previously described [32] . Serum samples from patients with mild or severe leptospirosis were collected twice: ( i ) acute sample , collected at patient admittance at the health care unit and ( ii ) convalescent sample , collected 5–276 days after the first sampling . Controls consisted of ( i ) 37 sera from healthy Leptospira-unexposed ( naïve ) volunteers from California/US and ( ii ) 37 sera from healthy participants enrolled in a cohort study in a high risk urban slum community in Salvador , endemic for leptospirosis . The entire ORFeome of Leptospira interrogans serovar Copenhageni strain Fiocruz L1-130 was amplified by PCR and cloned into pXI vector using a high-throughput PCR recombination cloning method developed by our group [34] . In this strategy , cloned ORFs were expressed with C-terminal hemaglutinin ( HA ) tag and N-terminal poly-histidine ( His ) tag . Genes larger than 3kb were cloned as smaller segments as described previously [20 , 21] and the ligA and ligB genes ( LIC10465 and LIC10464 , respectively ) were fragmented according to the repeated Big domains present in the structure of each protein ( LigB Repeats 7–12 , LigA Repeats 7–13 and LigA/B Repeats 1–6 ) [35] . After identifying the seroreactive antigens on the microarrays , the inserts in the corresponding plasmids were confirmed by nucleotide sequencing by the Sanger method . Microarray fabrication was performed as described previously [20 , 21] . Briefly , purified mini-preparations of DNA were used for expression in E . coli in vitro based transcription-translation ( IVTT ) reaction system ( RTS Kit , Roche ) , following the manufacturer´s instructions . Negative control reactions were those performed in the absence of DNA template ( “NoDNA” controls ) . Protease inhibitor mixture ( Complete , Roche ) and Tween-20 ( 0 . 5% v/v final concentration ) were added to the reactions , which were then printed onto nitrocellulose coated glass FAST slides using an Omni Grid 100 microarray printer ( Genomic Solutions ) . Multiple negative control reactions and positive control spots of an IgG mix containing mouse , rat and human IgG and IgM ( Jackson Immuno Research ) were added to the arrays . Protein expression was verified by probing the array with monoclonal anti-polyhistidine ( Sigma Aldrich ) and anti-hemaglutinin ( Roche Applied Science ) as previously described [20 , 21] . Human sera samples were diluted 1/100 in Protein Array Blocking Buffer ( Whatman ) supplemented with 10% v/v E . coli lysate 10mg/mL ( McLab ) and incubated 30 min at room temperature ( RT ) with constant mixing prior to addition to the microarray . Arrays were blocked for 30 min with Protein Array Blocking Buffer and then incubated with diluted samples overnight at 4°C , with gentle rocking . Washes and incubation with conjugate antibodies were performed as described previously [20 , 21] . Slides were scanned in a Perkin Elmer ScanArray confocal laser and intensities were quantified using QuantArray package . Selected ORFs were cloned into pET100-TOPO plasmid ( Invitrogen ) for His-tagged recombinant protein expression in BL21 ( DE3 ) Star E . coli cells , according to the manufacturer’s recommendations . Recombinant protein expression was performed with EnPresso B system ( Biosilta ) . Briefly , pre-cultured cells were inoculated 1/100 into 3 . 5 mL of EnPresso B medium supplemented with Ampicilin 100 μg/mL Reagent A 1 . 5 U/μL and grown shaking ( 160 rpm ) at 30°C for 16–18 hs in 24-well culture blocks . Expression was induced by the addition of 350 μL of the booster reagent supplemented with 15U/μL Reagent A and 100 mM IPTG , for 24 h at 30°C under 160 rpm shaking . Cells were then harvested and lysed with 0 . 05 g of Cellytic Express ( Sigma ) for each mL of final culture , for 30 min at RT . Lysates were applied to a Ni2+-charged resin ( Qiagen ) and recombinant proteins were manually purified using 20mM Tris ( pH 8 . 0 ) buffers with increasing concentrations of Imidazole . Washes varied from 5 mM to 40 mM Imidazole , depending on the protein , and elution was performed with 500 mM or 1M Imidazole . Imidazole was removed by dialysis ( Thermo Scientific dialysis cassettes ) and the purified proteins were checked for homogeneity in 12 . 5% SDS-PAGE . Protein concentration was determined by the BCA method ( Thermo Scientific ) according to the manufacturer's recommendations . The assay was performed as described previously [23] . Briefly , 100 ng of each purified protein was immobilized on a nitrocellulose membrane strip . A semi-automatic micro-aerolization device was used to generate parallel bands with no visible marks . The membrane was cut into 0 . 5 cm wide strips perpendicularly to the antigen bands . The strips were blocked for 90 min with 4% reduced-fat bovine milk diluted in PBST ( PBS + 0 . 5% Tween 20 ) and then incubated for 1 h at RT with individual serum samples diluted 1:200 in PBST 0 . 25% BSA and 5% v/v E . coli lysate 20mg/mL . After 3 washes with PBST , the strips were incubated for 1 hour with alkaline phosphatase–labeled anti-human IgG antibody ( Sigma-Aldrich ) diluted 1:30 . 000 in PBST 0 . 25% BSA . The strips were then washed 3 times with PBST and revealed with Western Blue Stabilized Substrate for Alkaline Phosphatase ( Bio-Rad ) for 10 min . The reaction was stopped with distilled water . Strips were air-dried and scanned images were converted to gray scale before band intensity quantification with ImageJ software ( found at http://rsbweb . nih . gov/ij/ ) . Array signal intensity was quantified using QuantArray software . Spots intensity raw data were obtained as the mean pixel signal intensity with automatic correction for spot-specific background . Data was normalized by dividing the raw signal for each IVTT protein spot by the median of the sample-specific IVTT control spots ( fold-over control [FOC] ) and then taking the base-2 logarithm of the ratio ( log2 FOC ) . Conceptually , a normalized signal of 0 . 0 is equal to control spot signal , and a normalized signal of 1 . 0 is 2-fold higher than control spot signal . When evaluating a protein spot as reactive or non-reactive , normalized signals >1 . 0 were considered reactive . These designations were used to evaluate response frequency and to identify a subset of sero-reactive proteins for further analysis . A given protein on the array was considered sero-reactive if it was reactive in at least 60% of the samples in one or more of the following groups: severe disease , acute sample ( n = 30 ) ; severe disease , convalescent sample ( n = 30 ) ; mild disease , acute sample ( n = 30 ) ; mild disease , convalescent sample ( n = 30 ) ; endemic controls ( n = 30 ) ; naïve controls ( n = 30 ) . Sero-reactive proteins were identified separately using IgG and IgM responses . For each sero-reactive protein , sample groups were compared using t-tests [R stats package] and the area under receiver operator characteristic curve ( AUC ) [R rocr package] . Proteins with t-test p-value < 0 . 05 after correction for false discovery [36] and AUC > 0 . 70 were identified as differentially reactive . Clinical features of the leptospirosis patients participating in this study were described using frequencies and medians with interquartile ( IQR ) ranges calculated in Excel ( Table 1 ) . The Fisher Exact test or the Mann-Whitney test were used to compare clinical presentations of patients with mild or severe disease using GraphPad Prism 5 . 02 software . The raw and normalized array data used in this study have been deposited in the Gene Expression Omnibus archive ( www . ncbi . nlm . nih . gov/geo/ ) , accession number GSE86630 .
To identify antigens associated with symptomatic leptospirosis and severe disease ( requiring hospitalization ) , we enrolled 31 patients hospitalized with suspected leptospirosis , 30 individuals treated at an urgent care facility for suspected leptospirosis , 30 individuals living in the same communities as enrolled patients ( hyperendemic controls ) , and 30 unexposed controls ( naïve controls ) . All patients survived and provided paired acute and convalescent sera samples . Table 1 describes patient characteristics for clinical and biochemical tests performed during hospitalization or outpatient treatment . Hospitalized patients presented with more severe disease: 77 . 4% had oliguric renal failure , 6 . 5% had respiratory failure , and 22 . 5% required ICU admission , while none of these outcomes were observed in outpatients . Additionally , the agglutinating antibody titers for hospitalized patients were significantly higher during acute illness and convalescence compared to patients with mild leptospirosis ( p = 0 . 011; p = <0 . 0001 ) . However , while hospitalized patients ( severe disease ) were older ( p = 0 . 039 ) and predominantly male ( p = 0 . 03 ) , there were no significant differences in days of symptoms at acute or convalescent sample collections between patients with mild and severe leptospirosis ( acute p = 0 . 085; convalescent p = 0 . 681 ) . Therefore , any differences observed in outcomes were not due to duration of illness or sampling times . In order to determine whether there is an antibody signature specific to symptomatic disease , we probed the protein arrays with a collection of 192 sera samples , including leptospirosis patients and healthy individuals living in areas with or without endemic transmission of leptospirosis . IgM and IgG probing revealed a set of 478 reactive antigens for both acute and convalescent phases , corresponding to 12 . 5% of all 3819 proteins and segments included on the arrays . Of these , 255 were specific for IgM , 128 were specific for IgG and 95 were recognized by both antibodies ( Fig 1A ) . Interestingly , we detected a majority of the IgM and IgG antigens in patients with mild disease ( Fig 1B and 1C ) . To identify antigens specific to patients with confirmed leptospirosis ( serodiagnostic antigens ) , we then compared antigens from the sera of patients with those from healthy individuals and found 36 antigens with significantly higher IgG reactivity in leptospirosis patients than in healthy volunteers from United States or healthy individuals living in a highly endemic area in Brazil . Of these , 12 ( 33% ) were identified during acute leptospirosis ( S2 Table ) and 33 ( 92% ) during convalescence ( S3 Table ) . Early antigen detection during infection is critical for the development of a new diagnostic test for leptospirosis . Therefore , we first focused on serodiagnostic antigens identified during acute phase in patients with mild or severe disease . Surprisingly , we found only a limited subset of all the seroreactive antigens were significantly recognized by IgGs in patients relative to endemic and naïve control volunteers: 11 of the 128 in the mild patient group and 28 of the 55 in the severe group ( Fig 2A ) . Of these only 5 of the 11 and 9 of the 28 were present during acute illnesss . For the mild group , the Lig proteins were the antigens with highest accuracy , especially LigA/B 1–6 , with 90% sensitivity , 86% specificity and AUC of 0 . 916 . To determine whether we could increase both sensitivity and specificity by combining the antigens , we constructed Receiver Operating Characteristic ( ROC ) curves for combinations of the 5 antigens to assess antigens diagnostic performance ( Fig 2A ) . We found that combining the top two antigens LigA/B 1–6 and LigA 8–13 yielded slightly higher sensitivity ( 86% ) and specificity ( 91% ) than the other combinations ( Fig 2A ) . We performed similar analyses for the 9 antigens specific to the severe group . Again , the best diagnostic accuracy was achieved with LigA/B 1–6 ( AUC = 0 . 935 , 87% sensitivity , 100% specificity ) followed by LIC20276 ( AUC = 0 . 901 , 84% sensitivity , 92% specificity ) . When we combined both antigens , sensitivity reached 94% , and specificity was 100% ( Fig 2B ) . For the remaining antigens , sensitivity ranged from 77% to 90% and specificity ranged from 77% to 92% . Again , other combinations did not yield better combined sensitivity and specificity ( Fig 2B ) . Our results indicate that we have identified candidates for new leptospirosis diagnostic tests and have discovered that there may be a limited dominant antigen antibody response to Leptospira infection . We analyzed the responses from convalescent sera to determine whether there were major shifts in antibody responses to specific antigens with time . Patients recovering from mild disease had significantly higher IgG titers for 10 antigens compared to endemic controls , while the number of antigens nearly tripled for patients with severe clinical presentation ( S3 Table ) . Antigens identified at convalescent phase accounted for ~92% of all diagnostic antigens ( 33 in 36 total IgG antigens ) and LigA/B 1–6 and LigB 8–12 were the antigens with best diagnostic performance for patients with severe and mild disease , respectively . While these antigens do not have diagnostic potential , they do represent possible subunit vaccine candidates as robust antibody responses were generated over the duration of illness . To confirm the diagnostic and subunit vaccine potential of the sero-reactive antigens detected on the microarray chips , we purified six proteins from E . coli BL21 in vitro ( Fig 3B ) , and printed onto nitrocellulose membranes . We probed the immunostrips with serum from 8 endemic controls and 20 acute-phase patients , of which 10 had mild disease and 10 had severe disease . Serum from leptospirosis patients showed greater reactivity than serum from controls , especially serum from severe patients at convalescent phase ( Fig 3A ) . To assess the ability of these six antigens to distinguish between patients and controls , a multi-antigen ROC curve was generated ( Fig 3C ) , and demonstrated that the six selected antigens yielded a specificity of 100% and a sensitivity of 60% for acute mild group and 90% for the remaining groups . As there is limited knowledge of the factors contributing to leptospirosis severe disease outcomes , we compared the antibody kinetics of patients to determine whether there are differences in antibody responses based on disease severity . We first compared the global IgG and IgM reactivities against all 478 reactive antigens identified in the microarrays by comparing the summed average signal intensities for each antigen during acute illness with that at convalescence . We detected a trending increase in IgG reactivity in patients with severe leptospirosis , which reached statistical significance when we analyzed the signals from the 36 patient-specific antigens ( p<0 . 05 ) ( S2 Fig ) . We did not observe this trend in patients with mild disease . For IgM-specific antigens , we observed no significant differences for either patient group or antigen set ( S2 Fig ) . Thus , we identified significant IgG responses increases only in the severe patient group over time . To understand the differences in antibody kinetics in patients in more detail , we next compared the antibody responses to the 36 differentially reactive antigens at the acute and convalescent time points for each individual by two way t-test . Based on the results of each t-test the individuals were categorized as: ( i ) increasing , when average response to the 36 differentially reactive antigens was higher at convalescent time point than acute , and p-value < 0 . 05 ) , ( ii ) no change ( p-value > 0 . 05 ) or ( iii ) decreasing , when average response to the 36 differentially reactive antigens was lower at convalescent time point than acute , and p-value < 0 . 05 . This comparison yielded vastly different profiles for patients with mild disease and severe disease . When analyzing IgG responses , we categorized 74 . 4% of patients with in the severe group as “increasing” versus only 29 . 6% of patients in the mild group ( Fig 4A ) . When analyzing IgM responses , we categorized 32 . 3% of patients in the severe group as “increasing” versus only 3 . 3% in the mild group ( Fig 4B ) . Additionally , 90 . 0% of IgM responses did not change over time in the mild group , compared to 51 . 6% in the severe group . Altogether , these data clearly demonstrate that leptospirosis patients with different clinical presentations generate distinct antibody profiles . In our kinetic antibody analyses , we enrolled five patients with mild leptospirosis , which had antibody profiles that resembled those of patients with severe leptospirosis: all had increases in IgG levels over time for 10 antigens ( Fig 4C and S2 ) . Though these five patients clearly developed an antibody response more representative of patients with severe disease ( S3 Fig ) , including a higher convalescent agglutinating antibody titer ( 400–12800 ) , they did not present with any severe clinical outcomes we measured . All other clinical and laboratory features were similar to the 25 patients with mild leptospirosis ( S4 Table ) .
Leptospirosis is a disease with a broad spectrum of clinical manifestations ranging from asymptomatic and nonspecific acute febrile illnesses to life-threatening renal failure or pulmonary hemorrhage syndrome [2 , 37] . Over a million cases of severe leptospirosis occur every year . This figure represents only a faction ( potentially 5–15% ) of the total mild leptospirosis cases , which usually are not identified by surveillance systems . The mechanisms involved in poor disease progression remain poorly defined , but pathogen related and host factors likely contribute to this heterogeneity [2 , 4] . Here , we identified 12 specific IgG antigens that differentiate acute symptomatic disease from uninfected individuals in endemic regions and therefore represent promising diagnostic candidates for an early laboratory test for the diagnosis of leptospirosis . We also identified patient-specific antigens during convalescence , which are putative subunit vaccine candidates . Lastly , we showed that patients with different clinical presentations generate distinct antibody kinetic profiles , and we hypothesize that since antibodies are protective , disease severity and the antibody signatures may indicate primary and secondary infections . We identified 12 IgG serodiganostic antigens for acute leptospirosis . Among them are the well-known sero-reactive proteins LigA/B 1–6 , LigA 8–13 , LigB 8–12 and LIC10973 ( OmpL1 ) . Several published studies used the Ligs as diagnostic markers for leptospirosis [35 , 38–41] as well as OmpL1 , especially in combination with LipL21 , LipL32 or LipL41 [42] . Our group has previously identified LIC10486 ( hypothetical protein ) and LIC12544 ( DNA binding protein ) using the protein microarray platform [21] . The remaining 6 proteins LIC10024 ( adenylate/guanylate cyclase ) , LIC11591 ( exodeoxyribonuclease VII large subunit ) , LIC20077 ( polysaccharide deacetylase ) and the hypothetical proteins LIC11274 , LIC20276 and LIC12731 are promising newly identified serodiagnostic antigens , especially LIC20276 , which improved diagnostic performance for severe disease in combination with LigA/B 1–6 . Interestingly , patients showed antibody reactivity against several proteins annotated as hypothetical proteins , not only at acute disease , but also during convalescence . These results indicate that even though these proteins have not been assigned any function , they are indeed expressed by the bacteria and might play an important role in host infection . Further studies should be done in order to evaluate these antigens performance in different diagnostic platforms , such as ELISA and rapid tests . For diagnostic purposes , a complete validation study needs to be performed , including the probing of a more extensive sample collection , comprising more leptospirosis patients as well as healthy controls and patients with other febrile illness , such as dengue , sifilis and hepatitis A . The results presented here are consistent with our previous findings [21] . We detected 13 of the 24 IgG antigens previously found in hospitalized patients , strengthening the diagnostic potential of those antigens and validating the protein microarray antigen discovery platform . The inclusion of 39% of L . interrogans predicted ORFeome , however , did not provide significant advantage in diagnostic antigen discovery , since only 3 out of the 1489 proteins and segments added to the microarray were serodiagnostic , indicating that the algorithm used by our group to select the proteins included in the partial microarray was effective . Indeed , 32 out of the 36 diagnostic antigens identified here fall in at least one of the enrichment categories described by our group for antibody recognition [20 , 43 , 44] . Leptospirosis patients and healthy controls reacted against 12% of the L . interrogans predicted ORFeome . The majority of the imunodominant antigens were IgM specific , which corresponded to >50% of the sero-reactive proteins . The high number of IgM antigens may reflect the broad and low-affinity antigen-antibody interaction typical of IgM antibodies [45 , 46] . These features usually make IgM a hard indicator of reliable diagnostic tests and might have hindered the identification of IgM diagnostic targets , as they usually account for lower specificity in IgM-based serological tests and high background reactivity in negative samples [45] . Here , we had great success in detecting IgG antigens with potential use as diagnostic or vaccine targets , but further studies are needed to identify IgM antigens . In our previous work , we have shown that healthy individuals who live in areas with endemic transmission of leptospirosis have a background IgG reactivity against leptospiral protein antigens , possibly due to the constant exposure to the pathogen [21] . As it is well known that antibodies are one of the main immune mechanisms in naturally-acquired leptospirosis [16] , the presence of high IgG levels in such individuals suggests that those antibodies might play an important role in protection against the development of clinical leptospirosis . Despite this background IgG reactivity , we were able to identify antigens for which IgG levels were even higher among hospitalized leptospirosis patients , especially at the patient's convalescent sample [21] . Indeed , most of the 36 serodiagnostic antigens identified in the present study were detected in the convalescent sample of patients with severe disease . A considerably smaller number of antigens was detected in patients with the mild form , suggesting that their IgG antibody response is more similar to healthy individuals living in the same area . The distinct antibody profiles associated with each group were not due to differences in days of symptoms . We hypothesize that patients with mild leptospirosis had a background IgG reactivity that protected them from severe clinical manifestations while the lack of such IgG response might have favored the development of severe outcomes in hospitalized patients . In general , the first contact with an infectious agent is serologically characterized by a gradual increase in IgM , with a peak on days 7–10 after pathogen exposure , followed by an increase in IgG on days 10–14 . In a secondary infection , however , a robust IgG response is rapidly mounted as a consequence of the activation of memory B cells generated during the primary infection [47–49] . In the light of this , the fact that patients with mild leptospirosis maintained their IgG levels at acute serum sample , collected approximately 5 days after the onset of symptoms , and at convalescent sample , collected at least 13 days later , suggests that they mounted an anamnestic response due to a secondary leptospiral infection . In contrast , patients with the severe form showed an antibody response typical of a primary infection , with an increase in IgG levels from acute to convalescent phases . Our results indicate that the presence of antibodies anti-leptospiral proteins may be protective against clinical severe leptospirosis and that patients with mild disease might have had previous leptospiral infection ( s ) . However , numerous aspects can affect the host immune response against an infectious agent , including the inoculum size . Patients with severe clinical presentations might have been infected with a higher bacterial load than patients who presented the mild form , developing thereby a more intense immune response . In addition , we can't affirm that any of the patients enrolled in the present study had never been exposed to leptospira before since leptospirosis is highly endemic in their community . Nonetheless , there is a need of studies of this kind to help elucidate the immune response associated with naturally-acquired leptospirosis and we believe our work brings relevant information to the field . | Leptospirosis is zoonotic disease of global importance , with over a million cases and nearly 60 , 000 deaths annually . Symptomatic disease presentation ranges from a mild febrile disease with non-specific symptoms to severe forms , characterized by multi-organ failure , lung hemorrhage , and death . Factors driving severe outcomes remain unclear , but the host immune response likely plays an important role . In the present study , we applied high throughput techniques to identify the antibody profiles of patients with severe and mild leptospirosis . We discovered a limited number of immunodominant antigens , specific to patients . Surprisingly , we found the antibody repertoire varies in patients with different clinical outcomes and hypothesized that patients with mild symptoms were protected from severe disease due to pre-existing antibodies , while the profile of patients with severe outcomes was representative of a first exposure . These findings represent a substantial step forward in the knowledge of the humoral immune response to Leptospira infection , and we have identified new targets for vaccine and diagnostic test development . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"leptospira",
"pathology",
"and",
"laboratory",
"medicine",
"acute",
"phase",
"proteins",
"pathogens",
"immunology",
"tropical",
"diseases",
"microbiology",
"bacterial",
"diseases",
"neglected",
"tropical",
"diseases",
"bioassays",
"and",
"physiological",
"analysis",
"antibodies",
"bacteria",
"bacterial",
"pathogens",
"antibody",
"response",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"zoonoses",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"leptospirosis",
"immune",
"response",
"microarrays",
"biochemistry",
"diagnostic",
"medicine",
"physiology",
"biology",
"and",
"life",
"sciences",
"leptospira",
"interrogans",
"organisms"
]
| 2017 | Distinct antibody responses of patients with mild and severe leptospirosis determined by whole proteome microarray analysis |
Studying aneuploidy during organism development has strong limitations because chronic mitotic perturbations used to generate aneuploidy usually result in lethality . We developed a genetic tool to induce aneuploidy in an acute and time-controlled manner during Drosophila development . This is achieved by reversible depletion of cohesin , a key molecule controlling mitotic fidelity . Larvae challenged with aneuploidy hatch into adults with severe motor defects shortening their life span . Neural stem cells , despite being aneuploid , display a delayed stress response and continue proliferating , resulting in the rapid appearance of chromosomal instability , a complex array of karyotypes , and cellular abnormalities . Notably , when other brain-cell lineages are forced to self-renew , aneuploidy-associated stress response is significantly delayed . Protecting only the developing brain from induced aneuploidy is sufficient to rescue motor defects and adult life span , suggesting that neural tissue is the most ill-equipped to deal with developmental aneuploidy .
Aneuploidy , a state of chromosome imbalance , was observed over a century ago . Since then , numerous studies have shown that aneuploidy is largely detrimental both at cellular and organism level . In multicellular organisms , chromosome gain or loss results in lethality or developmental defects [1 , 2] . At the cellular level , studies in yeast and cell culture have demonstrated that aneuploidy has a high fitness cost for the cell because unbalanced karyotypes lead to the activation of multiple stress-response pathways , resulting in reduced proliferation , cell-cycle arrest , or cell death ( reviewed in [3] ) . Cellular stress induced by aneuploidy seems at odds with the hypothesized role of aneuploidy in promoting malignancy , as well as its reported role as a driving force of yeast fitness and evolution [4–6] . Ninety percent of solid tumors harbor whole-chromosome gains and/or losses [7] . Therefore , aneuploidy and its effects on cell fitness and proliferation are context dependent , which emphasizes our need for a better understanding of the immediate and ultimate consequences of this abnormal cellular condition in metazoan tissues and through development . Study of aneuploidy in vivo is challenging because somatic aneuploidy is a rare event , difficult to capture and trace in real time because of several constraints: i ) cells are equipped with surveillance mechanisms that prevent chromosome mis-segregation , making naturally occurring aneuploidy events virtually impossible to evaluate; ii ) experimentally induced aneuploidy , by compromising mitotic fidelity , is often of low prevalence , as has been demonstrated for several mammalian [8 , 9] and Drosophila tissues [10–13]; and iii ) induction of somatic or constitutional aneuploidy in metazoans relies on chronic mitotic perturbation ( listed in [14] ) , which usually causes embryonic lethality ( reviewed in [15] ) as a result of progressive accumulation of damage in the developing organism . Thus , from these studies , it is impossible to disentangle short- and long-term consequences of aneuploidy or to examine the kinetics of the aneuploid state response during development . To circumvent these limitations , we generated a genetic system with the power to induce aneuploidy in an acute and time-controlled manner in all the dividing tissues of the developing Drosophila . The tool is based on reversible depletion of cohesin , a key molecule regulating mitotic fidelity [16 , 17] . Cohesin is a tripartite ring complex , composed of Structural Maintenance of Chromosome subunits ( SMCs ) SMC 1 and SMC3 and the bridging kleisin subunit Double-strand-break repair protein rad21 homolog ( RAD21 ) [18 , 19] . The primary mitotic role of cohesin is to mediate sister chromatid cohesion by topologically entrapping DNA fibers from neighboring chromatids [20 , 21] . Cells entering mitosis with premature loss of cohesin and sister chromatid separation activate the Spindle Assembly Checkpoint ( SAC ) , resulting in prolonged mitosis [17 , 22] . During this SAC-dependent mitotic delay , chromosomes are shuffled from one cell pole to the other by the mitotic spindle [22 , 23] . Consequently , chromosome shuffling induces genome randomization and aneuploidy upon mitotic exit with a theoretical rate of nearly 100% . Our engineered system enables a quick restoration of cohesin shortly after its inactivation , thereby restricting mitotic abnormalities to a short time frame concomitantly with the generation of high levels of aneuploidy . The acute and controlled nature of the tool allows us to dissect the kinetics of aneuploidy response across various tissue types and developmental stages .
To induce aneuploidy in an acute and time-controlled manner , we developed a genetic system based on rapid removal of cohesin complex , the molecular glue that holds sister chromatids together . To prevent chronic cohesin depletion and restrict mitotic failure to a single cell cycle , cohesin depletion is followed by subsequent cohesin rescue . The system relies on the artificial cleavage of a modified version of the RAD21 cohesin subunit that contains Tobacco Etch Virus ( TEV ) protease cleavage sites ( RAD21-TEV ) [24] . As previously shown , this system is very efficient at inactivating cohesin upon synthesis of the exogenous TEV protease induced by a heat-shock promoter ( HSprom ) , resulting in long-term inactivation of this complex ( >24 hours ) [22–24] . To restrict cohesin depletion , we modified this system by promptly rescuing cohesin integrity through the expression of a TEV-resistant RAD21 protein , RAD21-wild type [WT] , right after TEV-mediated inactivation . For this purpose , RAD21-WT expression is under the control of upstream activating sequence ( UAS ) promoter ( UAS-Rad21-wt-myc ) that is induced by the yeast transcription activator protein Gal4 ( Gal4 ) induced concomitantly with the TEV protease ( also under a HSprom ) ( Fig 1A ) . Given that the TEV protease is under the direct control of HSprom , whereas RAD21-WT relies on a dual expression system ( Gal4-UAS ) , we anticipated that the temporal delay in RAD21-WT expression relative to the induction of TEV protease would lead to a short time window of cohesin inactivation ( RAD21 cleavage ) ( Fig 1A ) . To test this , we probed for the kinetics of TEV-mediated cleavage of RAD21-TEV and synthesis of RAD21-WT in different tissues of the developing larvae . After heat shock , both Drosophila larvae brains and wing discs showed similar kinetics of the TEV-sensitive RAD21 disappearance , followed by the appearance of RAD21-WT ( Fig 1B and 1C ) . The timing of protein depletion/re-establishment differs slightly among different tissues or developmental stages but leads on average to a period of approximately 1 hour without cohesin ( Fig 1B , 1C and S1D Fig ) . The cohesive function of cohesin is established in the Synthesis ( S ) -phase , concomitantly with DNA replication . Once stabilized on the replicated genome , cohesive cohesin complexes do not turn over [25] . As such , loss of cohesin using our system will affect sister chromatid cohesion in all cells that are in S/Gap 2/Mitosis ( S/G2/M ) phases during the short period between TEV protease presence and synthesis of RAD21-WT ( Fig 1A ) . In addition to its canonical cohesive function , cohesin has also been implicated in other interphase functions , including regulation of gene expression [26] . In contrast to the cohesive pool , these cohesin molecules are known to be highly dynamic [25 , 27] . Moreover , recent studies report that cohesin-mediated loops are quickly restored upon cohesin re-establishment [28] . We therefore anticipated that this function should not be severely affected by our system . In sharp contrast , mitotic errors induced upon cohesin cleavage are irreversible because there is no way to restore cellular ploidy after a compromised round of mitosis . Whereas canonical chronic mitotic perturbations lead to several rounds of mitotic failures , our novel genetic system should lead to cohesion defects only in the first mitosis following the heat-shock because the expression of RAD21-WT should be able to rescue cohesion in the subsequent cell cycle ( Fig 1A ) . To confirm that our genetic system works as anticipated , we focused our analysis on two different cycling tissues from the larva: the developing brain and the epithelial wing discs . The developing brain of Drosophila is an excellent model to study the consequences of developmental aneuploidy . The well-characterized cell lineages , in combination with our tractable system to induce mis-segregation of chromosomes , offer a unique opportunity to trace the fate of aneuploid cells in real time and analyze their effect on the nervous system development . Through larval development , approximately 100 large neural stem cells called Neuroblasts ( Nbs ) [29] located in the central brain ( CB ) region divide asymmetrically to self-renew and generate distinct neuronal lineages via differentiating progeny [30] . We evaluated , by live-cell imaging , mitotic fidelity in these Nbs using two independent criteria to estimate the state of sister chromatid cohesion: i ) the presence of prematurely separated sister chromatids ( single sisters ) as opposed to metaphase chromosome alignment and ii ) the time cells spend in mitosis , given that premature loss of sister chromatid cohesion is known to activate the SAC and delay mitotic exit [22] . As expected , the first division after the heat shock results in full cohesin cleavage in Nbs , followed by cohesin rescue in subsequent divisions ( S1 and S2 Movies ) . The fast cell cycle of Nbs , coupled with continued proliferation of these cells despite their abnormal genome content ( further discussed below ) , enables analysis of mitotic fidelity throughout several consecutive divisions in great detail . Consistently , in the first mitosis after heat-shock induction ( AHS ) , 95% of Nbs contain single sisters and exhibit mitotic delay and chromosome shuffling ( Fig 2A and 2B ) . However , in the subsequent mitosis , cohesion is restored in approximately 80% ( 60% fully rescued , 20% intermediate rescued ) of the Nbs , with clear metaphases and a shorter mitotic delay ( Fig 2A , 2B and 2C ) . Finally , during the third cell division AHS , the mitotic timing and the cohesive state of Nbs are comparable to heat-shocked controls ( Fig 2A , 2B and 2C ) . Similar results were obtained for larvae heat-shocked at earlier stages of development ( S1A , S1B , and S1C Fig ) . In contrast to the Nbs , in the epithelial cells of the wing disc , we observe the presence of single sisters even at 48 hours AHS , despite high levels of RAD21-WT protein ( Figs 1C , 2D; 2E and 2F ) . These findings are consistent with the heterogeneity in cell-cycle duration of the wing disc cells , with some cells exhibiting a cell cycle of approximately 48 hours [31 , 32] . The high frequency of cells in S/G2 phases in this tissue , quantified by the fly Fluorescence Ubiquitination Cell Cycle Indicator ( FUCCI ) system [33] ( S3B Fig ) , further supports that a high number of cells are affected despite the short time of cohesin inactivation . Note that cohesion establishment is restricted to replication , and consequently , any cell in which RAD21 was depleted and rescued postreplication is unable to “rescue” cohesion despite the presence of the WT RAD21 protein . To validate the ability of our tool to induce aneuploidy in an acute manner in epithelial tissues , we examined the events following the reversible loss of cohesin . In Drosophila epithelial cells , multiple cellular insults , including aneuploidy , can trigger the apoptotic cascade [10 , 34] . In agreement with these studies , 24 hours AHS in the wing disc , Cleaved Caspase 3 ( CC3 ) staining reveals a large population of dying cells ( S2A and S2B Fig ) . However , at 48 hours AHS , the number of dying cells decreases significantly if cohesin is brought back , but not if the cohesin depletion by TEV is long-term ( S2A and S2B Fig ) . These results suggest that tissue recovery is limited and only possible if the mitotic disruption is restricted in time ( or number of cell cycles ) . Although quantitative analysis was performed exclusively for wing disc epithelial cells and brain Nbs , analysis of other epithelial dividing tissues of the Drosophila larvae reveal a similar high incidence of single sisters 3 hours AHS , implying that our system is able to induce a reversible whole-organism loss of cohesin ( S3A Fig ) . To understand how the organism responds to such high degree of aneuploidy , we traced the larvae through development after cohesin cleavage . For comparative analysis , we monitored eclosion rates for organisms with the long-term TEV protease cleavage system ( inducing cohesin removal for >24 hours ) and our newly developed system with reversible inactivation of cohesin . Both systems represent a strong injury for all the dividing tissues of the larva; therefore , we expected them to be lethal in the pupa-to-adult transition . However , in contrast to several studies using chronic mitotic perturbations [11 , 35] , flies challenged with aneuploidy using our reversible mitotic perturbation ecloded into adult flies at high frequency ( Fig 3A and S3 Movie ) . Eclosion rates of adults were dependent on the developmental stage at which cohesin was reversibly cleaved ( Fig 3A ) . Early induction of aneuploidy at 48 hours after egg laying ( AEL ) resulted in eclosion both with and without cohesin rescue . However , with 72-hours–AEL heat shock , there was almost no eclosion if the RAD21 protein subunit was not brought back ( Fig 3A ) . If the larvae were heat-shocked 96 hours AEL , no cohesin rescue resulted in dead pupae , while cohesin rescue resulted in flies trying to escape the pupa but unable to do so ( “head-out pupae” ) ( Fig 3A and 3B ) . These differences in developmental response to aneuploidy are likely due to increase of cell proliferation during larval development [11 , 36] ( S1A–S1C Fig ) . Regardless of the developmental stage , all flies that ecloded into adults after the aneuploidy challenge were unable to fly or move normally , even when showing serviceable wings and appendages ( Fig 3D and S3 and S4 Movies ) . Consequently , these flies exhibited markedly shorter lifespans than their control counterparts ( Fig 3C ) . Notably , even when aneuploidy is induced at earlier stages ( 48 hours AEL ) , thereby affecting fewer Nbs ( S1A Fig ) , adults displayed a “lethargic” behavior ( S5 Movie ) while having an otherwise healthy adult morphology . We hypothesized that the severe motor defects in the newly hatched flies are a direct consequence of aneuploidy in the developing larva brain . Recently , it has been proposed that neural stem cells with unwanted karyotypes are eliminated [11 , 35] , yet Nbs were previously shown to be resistant to large variations in ploidy [37 , 38] . Furthermore , previous studies also reported fly eclosion despite mitotic perturbations in the Nbs [13 , 39] . To gain insight on the potential fates of the aneuploid Nbs , first we analyzed a possible change in their number after acute aneuploidy induction , using the Nb marker Deadpan ( DPN ) . We quantified all the nuclei with Nb morphology ( Nb-like cells ) based on their size and location in the CB area . The analysis indicates that there is a gradual decline of the Nb number after the induction of aneuploidy from 12 hours AHS onwards , but never a complete loss of the aneuploid Nb population ( Fig 4A and 4B ) . The slow kinetics and incomplete elimination of the stem-cell population were quite surprising given the extreme levels of aneuploidy generated upon cohesin disruption ( approximately 100% ) . Premature differentiation and apoptosis were suggested as the main mechanisms of aneuploid Nb elimination , reported in two recent studies [11 , 35] . However , after acute aneuploidy induction in the entire Nb population , we found a very low frequency of cells undergoing premature differentiation or cell death ( S4A and S4B Fig ) . As a proxy for premature differentiation events , we quantified Nb-like cells that had either lost the DPN marker or abnormally exhibit the differentiation marker Prospero ( Pros ) with or without coexpression of DPN ( S4A and S4B Fig , arrowheads and dashed circles ) . Pros is the key factor acting as a switch for the transition from stem cell self-renewal to terminal differentiation [40]; therefore , this marker should not be present in Nbs . ( S4A Fig ) . These findings suggest that premature differentiation , although still taking place , is unlikely to be the only form of Nb elimination . To quantify the levels of apoptosis , we evaluated cells positive for cell death markers like CC3 and Death Caspase-1 ( DCP1 ) . We found a significant increase in CC3-positive cells in aneuploid brains ( S4C and S4D Fig ) , indicating that apoptosis may also contribute to the elimination of aneuploid cells , as recently proposed [11] . However , CC3 and DCP1 signals rarely correspond to Nb-like cells ( S4C , S4D and S4E Fig , arrowheads and dashed circles ) , suggesting that apoptosis may not be the major cause for Nb elimination . Thus , loss of stem-cell identity and/or cell death are more likely potential consequences of genome randomization rather than specific mechanisms controlling aneuploidy in the neural stem-cell population . Supporting this idea , inhibition of apoptosis by overexpression of the baculovirus protein P35 does not rescue Nb number per brain lobe 24 hours after induction of aneuploidy ( S4F Fig ) . To dissect the kinetics of the aneuploidy outcomes , we took advantage of the temporal resolution of our system to examine fate of aneuploid cells in real time . We restricted our analysis to third-instar wandering larvae because at this stage , all Nbs are engaged into active cell divisions [30] . Induction of aneuploidy at this developmental stage affects the entire Nb population , which facilitates cell-fate analysis . We observed a significant number of Nbs proliferating for several days and displaying a tendency for chromosome accumulation over time ( Fig 4C and 4D ) . To analyze the number of chromosomes in each dividing Nb , we performed chromosome spreads and counted the number of centromeres per mitotic figure ( each chromosome contains two centromere dots in mitosis ) . A single round of mitosis upon premature loss of sister chromatid cohesion should result in a maximum of 16 chromatids per Nbs , in the rare cases of complete asymmetric segregation ( 0 . 0013% ) . However , chromatid numbers can reach over 32 chromatids per cell , 24 hours and 72 hours after loss of cohesin , at a much higher frequency ( Fig 4D ) . This analysis suggests that chromosome accumulation does not solely result from the initial loss of cohesin . To investigate this further , we characterized the mitotic fidelity of aneuploid Nbs . As described above , mitotic divisions that immediately follow the initial loss of cohesin do not display major mitotic errors , and most of the defects observed are cohesin-related ( as expected from our experimental setup ) ( Figs 2B , 4E and 4F ) . However , 16 hours AHS , aneuploid cells start changing their behavior , and a variety of mitotic defects appear , becoming more frequent over time ( Fig 4E and 4G ) . Detailed characterization of the mitotic defects arising 16 hours after the induction of aneuploidy revealed that the majority of them ( approximately 60% ) are mild , consisting of either a prolonged metaphase or a lagging chromosome . Yet , the remaining approximately 40% consisted of cytokinesis defects , tripolar spindles , and sister chromatid cohesion defects , which are serious abnormalities that can drastically alter numerical ploidy ( Fig 4G and 4H ) . As an alternative way to estimate mitotic errors , we analyzed micronuclei formation after reversible loss of cohesin and consequent aneuploidy . Micronuclei were assessed by evaluating DNA signal together with LAMIN immunofluorescence in spreads from brain tissues at 8 hours and 24 hours AHS . Aneuploid cells exhibited a higher percentage of micronuclei but only 24 hours AHS , reinforcing our observations that severe mitotic abnormalities appear several hours ( >16 hours ) after aneuploidy was induced ( S5A and S5B Fig ) . These results reveal that a few hours are enough for the stable divisions of aneuploid karyotypes to become unstable in vivo , leading to further randomization of the genome , as previously shown in yeast and tissue culture [41 , 42] . This chromosomal instability contributes to Nb number decline because catastrophic mitotic errors can result in complete loss of Nb morphology and positioning ( Fig 4I ) . All together , these results indicate that transient cohesin loss and aneuploidy induction in the Nbs seem to induce a large spectrum of phenotypes contributing to the gradual Nb number decrease over time rather than specific mechanisms for elimination of abnormal karyotypes , as previously postulated [11 , 35] . To test whether there is a selection of specific karyotypes in the population of dividing aneuploid Nbs , we preformed Fluorescence In Situ Hybridization ( FISH ) analysis at 8 hours and 24 hours after aneuploidy was induced . To estimate the predicted frequency of specific chromosomes upon cohesin cleavage , we first modeled this process , assuming full random chromosome segregation in a single round , followed by a second round of random segregation in approximately 20% of the cases ( based on our experimental observations , see Fig 2B ) . FISH profiles were then compared with the statistical predictions ( Fig 5A and 5B ) . The FISH profiles confirmed the propensity for chromosome accumulation over time ( Fig 5C and 5E ) . Additionally , this analysis revealed that the karyotypes that can be tolerated by dividing Nbs are restricted to those containing at least one of the major three chromosomes , II , III , or X . The rate of complete loss of these chromosomes in the aneuploidy Nbs population was comparable to the control and thus likely a consequence of experimental error of the FISH ( Fig 5D and 5F ) . We concluded that , although dividing aneuploid Nbs can persist in the tissue , their karyotypes have restrictions , as complete loss of any of the major three chromosomes prevents their proliferation in the developing brain . In contrast , other aneuploid chromosome combinations are compatible with continued proliferation , particularly when chromosomes are gained . Our findings reveal that aneuploid cells are not promptly eliminated but instead continue to proliferate within certain karyotype restrictions . This should not only lead to the maintenance of aneuploid stem cells ( because of Nb self-renewal ) but also to the accumulation of differentiated aneuploid progeny ( note that each Nb divides approximately every 2 hours [36] ) . Therefore , we examined how such increase in aneuploidy within the tissue could affect cellular physiology and influence tissue development . Several aneuploidy-associated stresses that include oxidative , metabolic , and proteotoxic stress are likely to alter cellular homeostasis [3] , leading to p53 activation , cell-cycle arrest/senescence , and in some cases , programed cell death [3 , 43 , 44] . We took advantage of system to acutely induce aneuploidy and examine whether abnormal karyotypes trigger a stress response in the developing Drosophila brain and , if so , what the kinetics of such response is . We assessed , by immunohistochemistry , the presence of P53 and the senescence marker Dacapo ( DAP , a p21/p27 homologue [45 , 46] ) after the loss of cohesin and consequent aneuploidy . We observe that both P53 and the DAP accumulation start to be evident at 12 hours AHS , but only at 24 hours AHS are a significant number of different cell types labeled with these markers observed ( Fig 6A , 6B and 6C ) . Furthermore , the large majority of cells that appeared positive for these markers are not Nb-like cells ( mainly based on size and shape of the staining , Fig 6A ) . Nb-like cells stained with these markers are noticeable only at 48 hours AHS ( Fig 6A , arrowheads and dashed circles ) , suggesting that despite their aneuploid state , neural stem cells are delayed at displaying an evident stress response . We confirmed this observation by quantifying specifically the appearance of cells costained with these markers ( P53 and DAP ) and the Nb marker DPN through time ( S6A , S6B and S6C Fig ) . In addition to mitotic fidelity , cohesin has also been implicated in DNA damage response [47] and recent studies show that silencing RAD21 leads to the accumulation of gamma histone H2A X variant ( γH2AX ) foci [48] . Although it is impossible to dissociate mitotic loss of cohesin and the ensuing mitotic defects , abnormal karyotypes , and DNA damage , we reasoned that the ability of our tool to restore cohesin function in the next G1/S-phase would minimize the effect of RAD21 depletion on promoting double strand breaks ( DSBs ) . To test this , we compared the presence of Histone H2A variant H2Av foci ( H2AX homolog in Drosophila ) after induced RAD21 depletion with and without restoring RAD21 function . Consistent with our hypothesis , restoring RAD21 activity hours after its depletion keeps the amount of H2Av foci not significantly different from the controls . Yet , if depletion of RAD21 is long-term , a substantial increase in the number of H2Av foci is observed at 8 and 12 hours AHS ( S7A and S7B Fig ) . Although we cannot exclude DNA damage ( or even the presence of micronuclei; see S5A and S5B Fig ) as a contributing factor to the stress response observed in the tissue , it does not seem sufficient to explain the degree of Nb loss and stress markers observed in the tissue mainly at 24 hours AHS . The delayed stress response ( i . e . , approximately 48 hours after induction of aneuploidy ) in the neural stem-cell pool may imply a selective aneuploidy tolerance of Nbs when compared to the other cell types of the developing brain . To investigate this possibility , we took advantage of the Brain tumor ( brat ) mutant condition [49] . In brat mutant larvae brains , each Nb divides into two daughter cells that retain Nb-like properties as they continue self-renewing , leading to the formation of a tumor-like neoplasm [50 , 51] . We reasoned that if indeed Nbs are more resistant to aneuploidy , the complete occupancy of the developing brain by Nb-like cells observed in the brat mutant phenotype should be sufficient to prevent the stress response observed at 24 hours AHS . To test this , we combined our system for acute induction of aneuploidy with brat mutations and analyzed the presence of stress markers at 24 and 48 hours AHS . In brat mutants , the Nb marker DPN stains almost all the cells in the brain , demonstrating the Nb-like state of the entire tissue ( Fig 6D and 6E ) . As predicted , DAP appearance was significantly delayed in aneuploid brat mutants when compared to aneuploid brains alone ( Fig 6D and 6E ) . The same result is observed for P53 staining ( S8A and S8B Fig ) . These results suggest that Nbs are uniquely resistant to aneuploidy-associated stresses . Such a delayed response has the drawback that it enables continuous proliferation , as reported in our data and previous studies [37 , 38] . The continued proliferation , in turn , allows for further brain growth despite the presence of aneuploidy . Indeed , our results show that induced aneuploidy during larval development has no significant impact in the length of the brain and optic lobes in adult flies ( S9A–S9D Fig ) , unlike in chronic disruption states that result in microcephaly [13 , 38] . Upon aneuploidy challenge , we observed a striking difference across analyzed Drosophila tissues: whereas epithelial tissues like wing discs are able to regenerate from this injury ( S2A and S2B Fig ) , about half of neural stem cells are lost , while the remaining half continues proliferating and becomes highly chromosomally unstable ( Fig 4A and 4E ) . These findings , together with the fact that flies that survive the developmental aneuploidy induction show severe motor defects in otherwise healthy adult morphology , led us to hypothesize that the brain is the only tissue restricting aneuploid fly development . To test this hypothesis , we devised a system to protect the brain from cohesin removal and consequent aneuploidy . To achieve this , we complemented our reversible cohesin cleavage system with brain-specific expression of RAD21-WT throughout the course of the experiment ( Fig 7A ) . In this way , TEV presence should lead to cohesin loss in all larval tissues that survive solely on RAD21-TEV at the time of heat shock . In contrast , neural stem cells should be resistant to this challenge because they express both RAD21-TEV and RAD21-WT ( Fig 7B ) . Nb-specific expression of RAD21-WT was achieved by the use of inscutable-Gal4 ( insc-Gal4 ) or worniu-Gal4 ( wor-Gal4 ) drivers to constitutively express UAS-Rad21-wt-myc in the developing brain ( Fig 7B ) . As expected , constitutive presence of RAD21-WT in the brain prevents any cohesin defects in third-instar larvae Nbs ( Fig 7C ) . To confirm that the rescue of sister chromatid cohesion occurs exclusively in the brain , we performed parallel characterization of the first mitotic division AHS in multiple imaginal discs derived from the same larvae . As anticipated , full cohesin cleavage was observed in all the dividing epithelial tissues ( Fig 7C and 7D ) . Notably , protecting only the brain from developmental aneuploidy fully rescued the severe motor defects of the ecloded flies from the 72 hours AEL heat-shock , as demonstrated by mobility essays ( Fig 7F and S6 Movie ) . The brain protection was enough to rescue the life span of approximately 70% of the adult flies affected by organism-wide aneuploidy during development , demonstrating that the brain is indeed the most sensitive tissue when challenged with aneuploidy ( Fig 7E ) .
We developed a novel genetic tool in Drosophila to study aneuploidy in vivo . This tool enables the induction of a controlled pulse of aneuploidy at the developmental stage of choice . The outcomes of using reversible perturbation are significantly different from the ones resulting from chronic disruption of mitotic fidelity . The long-term survival after aneuploidy challenge coupled with the reversibility of the mitotic perturbation overcomes one of the major limitations present in other metazoan models: we were able the study the kinetics of aneuploidy and its consequences across different tissues/developmental stages . Cohesin loss and induction of aneuploidy is tolerated better by the organism if induced early in development , as observed by comparing the rates of eclosion . The developing larvae are progressively scaling mitotic machines , with each consecutive stage containing more divisions than the previous one [36] . This implies that the heat shock at the first- and third-instar larvae are not the same because they affect different number of dividing cells , thus generating different numbers of aneuploid progeny . Although the more parsimonious explanation for aneuploidy tolerance in early development would be a quantitative one , it is also important to mention that a developmental delay is observed after aneuploidy induction ( e . g . , delayed pupariation stage ) . It is well known that delayed development allows the organism to adjust their growth programs after disturbances [52 , 53] . This induced delay is a development-stage dependent response because some perturbations only appear to retard pupariation when induced at or before a certain stage in larval development , as , e . g . , beginning of the third instar [54–56] . Nbs have been used as a system to study aneuploidy response in previous studies [11 , 35 , 38 , 57] . Recent studies postulate two different but not mutually exclusive mechanisms of response to induced aneuploidy: premature differentiation [35] and cell death by apoptosis [11] . We reasoned that if these are the only mechanisms of response to aneuploidy in neural stem cells , they should be detectable in high frequency after the aneuploidy induction by our acute approach . Contrary to that notion , both premature differentiation and cell death were detected at a very low frequency , even days after cells became aneuploid . It is important to note that the degree of aneuploidy in the Nbs upon cohesin loss should be around 98% because of the extensive genome shuffling prior to mitotic exit . Therefore , the finding that aneuploidy does not eliminate the entire Nb population strongly argues against the existence of specific , active mechanisms controlling the integrity of the neural stem-cell genome . The more plausible explanation is that the Nb elimination due to aneuploidy stems from a wide spectrum of abnormalities due to a randomized genome , documented throughout this study . Supporting this idea , recent studies in yeast have shown that the same aneuploid karyotypes can have different outcomes [58] Examination of Nbs in real time after aneuploidy induction further revealed that aneuploidy is sufficient to induce chromosomal instability within a short time period ( approximately 16 hours ) , which was shown in yeast and tissue culture but never in a metazoan system [41 , 42] . The appearance of chromosomal instability , characterized by a wide range of mitotic defects , takes several cell cycles after cohesin has been restored , which strongly supports the notion that chromosomal instability is a consequence of the abnormal karyotype and not the mitotic disruption initially applied . Overall , we observe a selection towards the accumulation of chromosomes , generating huge Nbs that keep proliferating despite their increased ploidy [35 , 38] . This clearly demonstrates that just a single round of chromosome mis-segregation in these cells is enough to originate a complex array of karyotypes , which can lead to mitotic abnormalities . During the last years , studies in tissue culture and yeast cells have collected solid evidence on how aneuploid karyotypes can alter physiology of eukaryotic cells ( reviewed in [59] ) . They can lead to different aneuploidy-associated responses that include oxidative , metabolic , replication , and proteotoxic stress , which likely contribute to p53 activation and , in most cases , cell senescence and/or cell death [43] . However , our understanding of how aneuploidy-induced stress at the cellular level influences development of tissues is very limited . Our time-course assessment of classical stress response marker P53 and the Drosophila senescence marker DAP [45 , 46] clearly showed that the response to induced aneuploidy is not immediate and takes several hours ( 12 to 24 hours AHS ) . This delayed stress response is in agreement with recent observations in tissue culture , in which it has been shown that chromosome mis-segregation did not lead to the arrest in the following G1 in the vast majority of aneuploid daughter cells [60 , 61] . Our results highlight that cell identity might determine the kinetics of this stress response . Aneuploidy response is specifically delayed in the neural stem-cell pool ( displayed mainly at approximately 48 hours AHS ) compared to the rest of the tissue , which exhibits it considerably earlier . Forcing cells to self-renew using Brat mutations is sufficient to delay the appearance of P53 and DAP markers in the entire tissue , suggesting that uncontrolled self-renewal makes cells less sensitive to trigger a stress response . Quite fittingly , it was recently reported that aneuploidy in Drosophila Intestine Stem Cells ( ISCs ) results in increased stem-cell proliferation . Interestingly , ISCs do not activate apoptotic pathways in response to aneuploidy , suggesting that this type of cell is somewhat resistant to genomic imbalance [62] . Unusual resistance to altered ploidy was also observed in human and mouse embryonic stem cells ( ESCs ) , mostly achieved by relaxing the cell-cycle control and uncoupling the spindle checkpoint from apoptosis [63] . The ability of neural stem cells to continue dividing despite the aneuploid karyotype dubbed them as aneuploidy “tolerant” [11] . Yet , based on our findings , it is clear that keeping these aneuploid cells is catastrophic for normal tissue architecture and development . Thus , aneuploidy may be “tolerated” better in Nbs , but the tissue as a whole is unable to be functional . In contrast , the “sensitivity” of epithelial cells enables the tissue to clean up and regrow properly . Chromosomal aberrations have been long associated with neurological disorders [64] . However , their impact on brain development and function remains poorly understood , partially due to limitations of available experimental approaches . Previous studies in Drosophila have shown that the mitotic disruption in larvae Nbs generates a reduction of their brain size , reinforcing the idea about a link between aneuploidy and microcephaly [11 , 13 , 35 , 65] . However , our results showed that induced acute aneuploidy has no significant impact in the size of the adult brain . These findings suggest that the continued proliferation of neural stem cells , caused by incomplete cell elimination and delayed aneuploidy stress response , is sufficient to support the development of an apparently normal-sized organ . It is conceivable that the observed normal size reflects a sample selection because this analysis was restricted to flies that survived the aneuploid challenge ( approximately 70% ) . Supporting this possibility , a screening performed to isolate anatomical brain mutants of Drosophila has shown that mutant strains showing altered brain shape and particularly small brains are very weak , having mostly the mutations that are lethal at pupa stage [66] . Despite the unaltered shape and size of the adult brains , we reasoned that the neural circuits are likely impaired in those brains , giving rise to the adult phenotype observed in all the surviving flies . In accordance with the notion of the brain as the tissue most sensitive to aneuploidy , we show that preventing aneuploidy exclusively in the brain is sufficient to rescue the behavioral and life span defects promoted by developmental aneuploidy , suggesting that neural tissue is the most ill-equipped to deal with aneuploidy during development and imposes a significant cost for the organism . Several pathophysiological chromosomal disorders in humans—including trisomy 21 , trisomy 18 , and trisomy 13 , as well as the mosaic disorder mosaic variegated aneuploidy ( MVA , characterized by the presence of a different number of chromosomes in some cells ) —are well known to result in intellectual disability [64] , yet the impact of the aneuploid condition on brain development is still unclear [67 , 68] . Additionally , there is a lively debate about whether aneuploidy exists in metazoan brains and whether aneuploid neurons could be functional [69] . Usage of an inducible , in vivo system such as ours opens up the way for the exploration of this conundrum . Future work should also aim at elucidating the molecular mechanisms underlying the physiological changes in stem/somatic cells generated by aneuploidy and its implications on tissue development and homeostasis .
Flies were raised using standard techniques at room temperature ( 20°C–22°C ) . All stocks used in this study are summarized in Table 1 . We established both chronic and the acute inactivation of cohesin complex by crossing the following genotypes: w;hspr-nlsV5TEV;Rad21 ( ex3 ) /TM6B with w;;tubpr-Rad21 ( 550-3TEV ) -EGFP , Rad21 ( ex15 ) , polyubiq-His-RFP and w;hspr-nlsV5TEV;Rad21 ( ex3 ) , hspr-Gal4 , UAS-Rad21 ( wt ) -myc/TM6B with w;;tubpr-Rad21 ( 550-3TEV ) -EGFP , Rad21 ( ex15 ) , polyubiq-His-RFP , respectively . The progeny were then heat-shocked once at 37°C for 45 min at the desired developmental stage . The correct genotype larvae were selected based on the absence of the “tubby” phenotype; the heat-shocked “tubby” larvae were used as negative controls ( control heat shock ) . As genetic control , we used the same genotypes for the induction of aneuploidy but without performing the heat shock . To determine the proportion of adult eclosion , the crosses mentioned were raised in cages to monitor the time of egg collection . After 6 hours of collection , the plates were removed from the cages , the number of eggs was counted , and the plates were kept until larvae hatched . The plates were then heat-shocked at 37°C for 45 min at different larvae developmental time ( approximately 48 hours AEL , approximately 72 hours AEL , approximately 96 hours AEL , and approximately 120 hours AEL ( ± 6 hours ) ) and placed in a new clean plastic cage . Once they reached pupae stage ( “yellow body” ) , the pupae were gently removed with a wet brush and separated into “tubby” ( control heat shock ) and “no tubby” phenotypes ( conditions ) . The different batches of pupae were placed over agar plates covered with two layers of absorbent paper to maintain the humidity and counted . The plates with the pupae were kept at room temperature until flies ecloded , and the proportion of eclosion was calculated . To combine the induction of aneuploidy ( acute cohesin inactivation ) and the brat mutant genetic background , we generated the following stocks: w;brat1/CyO-CTG;Rad21 ( ex3 ) , hspr-Gal4 , UAS-Rad21 ( wt ) -myc/TM6B and w;hspr-nlsV5TEV , bratTS/CyO-CTG;tubpr-Rad21 ( 550-3TEV ) -EGFP , Rad21 ( ex15 ) , polyubiq-His-RFP . These stocks were crossed , the progeny were heat-shocked once at 37°C for 45 min at the developmental stage desired , and the genotype w;brat1/hspr-nlsV5TEV , bratTS;Rad21 ( ex3 ) , hspr-Gal4 , UAS-Rad21 ( wt ) -myc/tubpr-Rad21 ( 550-3TEV ) -EGFP , Rad21 ( ex15 ) , polyubiq-His-RFP was selected at larva stage based on the absent of both GFP signal and “tubby” phenotype . To inhibit apoptosis , we induced the overexpression of the baculovirus p35 in the context of the genetic background for acute inactivation of cohesin complex . To achieve this purpose , we generated the following stock: w;UAS-P35;tubpr-Rad21 ( 550-3TEV ) -EGFP3 , Rad21 ( ex15 ) , polyubiq-His-RFP to be crossed with w;hspr-nlsV5TEV;Rad21 ( ex3 ) , hspr-Gal4 , UAS-Rad21 ( wt ) -myc/TM6B . The progeny were then heat-shocked once at 37°C for 45 min at the developmental stage desired . Finally , for the “brain rescue” experimental setup , we generated the following stocks: w;insc-Gal4;tubpr-Rad21 ( 550-3TEV ) -EGFP , Rad21 ( ex15 ) , polyubiqpr-His-RFP and w;wor-Gal4;tubpr-Rad21 ( 550-3TEV ) -EGFP , Rad21 ( ex15 ) , polyubiqpr-His-RFP . These stocks were crossed with the w;hspr-nlsV5TEV;Rad21 ( ex3 ) , hspr-Gal4 , UAS-Rad21 ( wt ) -myc/TM6B stock . The crosses and the progeny were raised and treated as described above for the determination of the eclosion proportion . Life span was measured at room temperature according to standard protocols . In brief , newly ecloded animals ( 0 to 3 days ) were collected ( 50 per genotype: “control , ” “Aneuploidy , ” and “Aneuploidy + brain rescue” ) and then placed in vials ( up to 10 per vial ) and transferred to fresh vials every two days . Survival was recorded for each vial . Because of the reduced mobility of the aneuploidy genotypes , we scored flies stacked in the food as death events in all the vials analyzed . We created survival curves with Prism 5 . 00 for Windows ( GraphPad Software , San Diego , CA , USA ) using the method of Kaplan and Meier . For the climbing assay , flies were anesthetized with CO2 , separated into groups of around 20 adults ( 3 replicas for each genotype ) , and allowed to recover for 2 hours before being subjected to a climbing assay . Briefly , the groups of over 20 flies were placed in an empty climbing vial and then tapped down to the bottom . They were allowed to climb past the halfway point from the bottom of the vial for 30 seconds ( 10 cm ) . The number of flies above the 10 cm mark was recorded as the percentage of flies able to climb . Briefly , flies were anesthetized with CO2 and then were placed gently in agarose blocks to immobilize them and prevent any damage to the head or eyes . The agarose blocks with the flies were immersed in Carnoy fixative overnight at 4°C . The next day , the Carnoy solution was removed , and three 70% ethanol washes were performed . Immediately after , the flies were decapitated , and the heads were oriented one by one in melted 2% agarose to guarantee similar orientation of the tissue sections . Agarose blocks were then processed and embedded , and the whole head was sectioned into 5-μm-thick sequential sections and stained with hematoxylin–eosin . The histology was performed in the Histopathology unit at Instituto Gulbenkian de Ciência , and the slides were analyzed by a pathologist with a DMLB2 microscope ( Leica , Wetzlar , Germany ) . Images were acquired with a DFC320 camera ( Leica ) and NanoZoomer-SQ Digital slide scanner ( Hamamatsu , Hamamatsu City , Japan ) . Larvae third-instar brains were dissected in Schneider medium supplemented with 10% FBS , and intact brains were mounted on a glass-bottomed dish ( MatTek In Vitro Life Science Laboratories , Bratislava , Slovak Republic ) , covered with an oxygen-permeable membrane ( Xylem Analytics UK , Tunbridge Wells , United Kingdom ) , and sealed with Voltalef oil 10S ( VWR , Radnor , PA , USA ) . This procedure allowed long-term imaging of brains for periods of up to 10 hours . For imaging of imaginal discs and early instar larvae brains , tissues were dissected in Schneider medium with 10% FBS . Dissected discs were placed and oriented in a 200 μl drop of medium at the bottom of a glass-bottomed dish ( MatTek ) . Live imaging was performed on a spinning disc confocal using a Revolution XD microscope ( Andor , Belfast , UK ) equipped with a 60× glycerol-immersion 1 . 30 NA objective ( Leica Microsystems ) and an iXon Ultra 888 1 , 024 × 1 , 024 EMCCD ( Andor ) . 25–35 Z-series optical sections were acquired 0 . 5–1 μm apart . For brain spreads and immunofluorescence , third-instar larvae brains were dissected in PBS , incubated with 100 μM colchicine for one hour , hypotonic shocked in 0 . 5% sodium citrate for 2–3 min , and fixed on a 5-μl drop of fixative ( 3 . 7% formaldehyde , 0 . 1% Triton-X100 in PBS ) placed on top of a siliconized coverslip . After 30 seconds , the brains were squashed between the coverslip and a slide , allowed to fix for an additional 1 min , and then placed in liquid nitrogen . Slides were further extracted with 0 . 1% Triton X-100 in PBS for 10 min and used for immunofluorescence following standard protocols . Primary antibodies were rat anti-CID ( gift from Claudio E . Sunkel ) used at 1:2 , 000; cleaved Drosophila Dcp-1 ( Asp216 ) antibody ( 1:300 ) #1679578S ( Cell Signaling Technology , Danvers , MA , USA ) ; CC3 ( Asp175 ) antibody #9661 ( 1:300 ) ( Cell Signaling Technology ) ; anti-DPN antibody #ab195173 ( 1:1 , 500 ) ( Abcam , Cambridge , UK ) ; anti-P53 ( 1:150 ) p53 25F4 , which was deposited to the DSHB by G . M . Rubin ( DSHB Hybridoma Product p53 25F4 ) ; anti-histone H2AvD pS137 rabbit ( 1:1 , 000 ) ( Rockland , Limerick , PA , USA ) ; anti-lamin ( 1:1 , 000 ) ADL84 . 12 , which was deposited to the DSHB by P . A . Fisher ( DSHB Hybridoma Product ADL84 . 12 ) ; and anti-DAP ( 1:50 ) NP1 , which was deposited to the DSHB by I . Hariharan ( DSHB Hybridoma Product NP1 ) . Secondary antibodies conjugated with fluorescent dyes from Alexa series ( Invitrogen , Carlsbad , CA , USA ) were used according to the manufacturer's instructions . Third-instar wing imaginal disc fixation and staining , as well as immunofluorescence of whole brains , was performed using standard procedures [72] . Briefly , third-instar larvae wing disc tissue ( still attached to the larva body ) was fixed on ice for 30 min . The fixative consisted of 4% formaldehyde ( Polysciences , Warrington , PA , USA ) in 1× PEM buffer solution . Following this , tissues were washed by gentle agitation three times for 20 min in 1× PBS + 0 . 1% Triton X-100 ( PBS-T ) . Primary antibody incubation was performed overnight at 4 °C in PBS-T supplemented with 1% BSA and 1% donkey serum . The following day , the tissues were washed again and incubated for 2 hours at room temperature with the appropriate secondary antibodies diluted in PBS-T solution . Finally , after the wash of secondary antibodies , wing discs were mounted in Vectashield ( Vector Laboratories , Burlingame , CA , USA ) . Fluorescence images were acquired with a ×40 HCX PL APO CS oil immersion objective ( numerical aperture: 1 . 25–0 . 75 ) on a Leica SP5 confocal microscope . Brains from third-instar larvae were dissected in PBS , incubated with 100 μM colchicine for one hour , and transferred to 0 . 5% sodium citrate solution for 3–4 min . Then , the brains were transferred to a fixative containing 11:11:2 methanol/acetic acid/MQ water for 30 seconds before being placed in a droplet of 45% acetic acid for 2 min , squashed , and transferred to liquid nitrogen . Then , the coverslip was removed and the slide incubated in absolute ethanol for 10 min at −20 °C ( freezer incubation ) . The slides were air dried at 4 °C ( 20 min ) . The slides were dehydrated at room temperature in 70% , 90% , and absolute ethanol for 3 min prior to DNA denaturation in 70% formamide–2× SCC solution for 2 min at 70 °C . This was done on a thermomixer set at 70°C with a formamide solution heated to 70 °C . Then , the slides were transferred to cold 70% ethanol ( −20 °C ) and dehydrated at room temperature in 90% and absolute ethanol for 3 min . FISH probes were denatured in the hybridization buffer at 92°C for 3 min . Hybridization was done overnight at 37 °C using 30 μl of FISH hybridization buffer/probe mix per slide . Hybridization buffer: 20% dextran sulfate in 2× SCCT/50% formamide/0 . 5 mg/ml salmon sperm DNA . Then , slides were washed 3 × 5 min in 50% formamide–2× SCC at 42 °C and 3 × 5 min in 0 . 1× SCC at 60 °C . These steps were done on the thermomixer , with the solutions previously heated to desired temperatures . Finally , the slides were washed in PBS and mounted in Vecta shield with DAPI . The probes were used in the final concentration of 70 Nm in hybridization buffer . Probes used were Chr_X ( 359-bp satellite DNA ) A546-GGGATCGTTAGCACTGGTAATTAGCTGC and Ch_3 ( dodeca satellite DNA ) Cy5-ACGGGACCAGTACGG DNA probes , Chr_2 A488- ( AACAC ) . To analyze RAD21 protein amounts , Drosophila tissues were dissected in PBS and homogenized with a pestle in sample buffer . Samples were centrifuged and boiled for 5 min in 2× sample buffer . Samples were loaded on a 13% SDS-gel for electrophoresis and then transferred to nitrocellulose membranes . Western blot analysis was performed according to standard protocols using the following antibodies: anti-α-tubulin ( 1:50 , 000 , DM1A , Sigma-Aldrich Cat# T9026; Sigma-Aldrich , St . Louis , MO , USA ) , guinea pig anti-Rad21 [73] , and V5 Tag mouse monoclonal antibody ( Novex , Thermo Fisher Scientific , Waltham , MA , USA ) . Imaging analysis was performed using FIJI software [74] . For z projections , slices were stacked into maximum intensity ( 10 frames , 2 μm each ) . Some pictures were rotated and/or flipped to orient them in the same way . Statistical analysis and graphic representations were performed using Prism 5 . 00 for Windows ( GraphPad Software , San Diego , CA , USA ) . Unpaired t test or one-way ANOVA ( using the Bonferroni’s multiple comparison ) were applied depending on the measurements analyzed in the corresponding experiment . Sample size details are included in the respective plotted graphs . | Aneuploidy—the presence of an abnormal number of chromosomes in a cell—is a hallmark of cancer and developmental disorders . However , it is notoriously difficult to study in a living organism . We have thus developed a new genetic tool that allows for the inducible generation of aneuploidy in the fruit fly Drosophila melanogaster at any developmental stage . The tool is based on reversible depletion of the protein cohesin , a major regulator of fidelity of chromosome segregation during cell division . Contrary to our expectations , when larvae are challenged with organism-wide mosaic aneuploidy , they still hatch into adult flies , albeit with severe motor defects and reduced life span . While most of the developing epithelial tissues respond to aneuploidy by inducing cell death , eliminating the cells with an abnormal number of chromosomes , the developing brain does not . Most of the aneuploid neural stem cells can keep proliferating despite their abnormal chromosomal number and chromosomal instability , suggesting that these cells are uniquely resistant to aneuploidy-associated stresses . The differential tissue response led us to hypothesize that the brain is the limiting tissue in response to developmental aneuploidy . To test it , we modified the aneuploidy induction system to protect the brain from aneuploidy while the rest of the tissues were affected . As a result , we observed a complete rescue of previous motor defects and life span reduction , demonstrating that the developing brain is the tissue most susceptible to aneuploidy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"cell",
"death",
"invertebrates",
"cellular",
"stress",
"responses",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"dna-binding",
"proteins",
"chromatids",
"departures",
"from",
"diploidy",
"animals",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"drosophila",
"research",
"and",
"analysis",
"methods",
"aneuploidy",
"animal",
"studies",
"proteins",
"chromosome",
"biology",
"life",
"cycles",
"insects",
"arthropoda",
"biochemistry",
"eukaryota",
"cell",
"biology",
"apoptosis",
"genetics",
"biology",
"and",
"life",
"sciences",
"larvae",
"organisms",
"chromosomes"
]
| 2019 | Induced aneuploidy in neural stem cells triggers a delayed stress response and impairs adult life span in flies |
The determinants of transcriptional regulation in malaria parasites remain elusive . The presence of a well-characterized gene expression cascade shared by different Plasmodium falciparum strains could imply that transcriptional regulation and its natural variation do not contribute significantly to the evolution of parasite drug resistance . To clarify the role of transcriptional variation as a source of stain-specific diversity in the most deadly malaria species and to find genetic loci that dictate variations in gene expression , we examined genome-wide expression level polymorphisms ( ELPs ) in a genetic cross between phenotypically distinct parasite clones . Significant variation in gene expression is observed through direct co-hybridizations of RNA from different P . falciparum clones . Nearly 18% of genes were regulated by a significant expression quantitative trait locus . The genetic determinants of most of these ELPs resided in hotspots that are physically distant from their targets . The most prominent regulatory locus , influencing 269 transcripts , coincided with a Chromosome 5 amplification event carrying the drug resistance gene , pfmdr1 , and 13 other genes . Drug selection pressure in the Dd2 parental clone lineage led not only to a copy number change in the pfmdr1 gene but also to an increased copy number of putative neighboring regulatory factors that , in turn , broadly influence the transcriptional network . Previously unrecognized transcriptional variation , controlled by polymorphic regulatory genes and possibly master regulators within large copy number variants , contributes to sweeping phenotypic evolution in drug-resistant malaria parasites .
Plasmodium falciparum is an apicomplexan parasite that causes the most severe and lethal form of human malaria . Parasites isolated from patients across the globe exhibit a wide range of phenotypic variation , including drug responses , growth rates , and a variety of virulence factors . Until recently , inter-strain variation had been studied primarily at the DNA sequence and phenotype levels . Since the P . falciparum genome was fully sequenced [1] , several large-scale gene expression studies [2–8] have provided the malaria research community with detailed insights into gene expression across the parasite's life cycle . The P . falciparum transcriptome is expressed as an unusual continuous cascade across the distinct stages of the parasite's erythrocytic cycle [4] . However , the lack of traditional DNA-binding proteins raises important questions regarding the nature of transcriptional regulation , characteristically dictated by specific transcription factors in other eukaryotic systems [9 , 10] . While malaria parasites must have a complex regulatory architecture to control the precise waves of gene expression during erythrocytic development , it is not known whether natural allelic diversity in this species includes variations in the regulatory network itself . P . falciparum is not amenable to many of the standard tools employed to study model organisms . However , combinations of the few available genome-wide methods , including classical genetics , offer novel opportunities to dissect layers of regulatory complexities such as DNA copy number variation ( CNV ) and transcription . Transcription in malaria parasites is rigidly programmed through the erythrocytic cycle and largely unresponsive to specific perturbation [11 , 12] . Recently , it was concluded that the parasite “lacks ubiquitous heritable transcriptional variation” [13] , based primarily on recent work comparing gene expression profiles between three unrelated , lab-adapted parasite strains ( i . e . , parasite clones ) : 3D7 , HB3 , and Dd2 [6] . If true , it follows that divergent phenotypes between strains , such as drug resistance , do not result from variation in the transcriptional profile . This interpretation runs contrary to small datasets on direct co-hybridization of cDNA from malaria parasites [3] , recent evidence that malaria parasites display distinct physiological states in their in vivo transcriptional profiles [14] , and numerous observations that modifications in transcriptional regulation underpin complex adaptations in human , fly , worm , and yeast clonal populations [15–18] . Genetic mapping of genome-wide RNA levels as traits [19 , 20] has been applied to a variety of organisms to map regulators of transcription [16 , 21–27] . As has been observed for other classical phenotypes , transcript levels are inherited as complex traits that can be mapped to their causal genetic polymorphisms and , consequently , can broaden our understanding of the regulatory mechanisms underpinning adaptation to fluctuating environments . Mapping the regulatory determinants of gene expression can be particularly useful in malaria parasites . The P . falciparum genome encodes much of the basal eukaryotic transcriptional machinery , including RNA polymerase II [1] , and putative orthologs to general transcription factors have been identified [28] . However , for the most part , the functional roles of even these standard transcriptional regulatory components have not been experimentally confirmed . Comprehensive searches of Plasmodium spp . proteomes for specific transcription factors have yielded few candidates [29] , with the exception of the identification of members of a candidate transcription factor ApiAP2 gene family [30] . A recent study by De Silva et al . ( 2008 ) [31] provides empirical support for ApiAP2 transcription factors and their cognate binding sequences as a potential source of developmental regulation in this species . The extremely AT-rich ( > 80% ) P . falciparum genome may obscure signatures of upstream or downstream regulatory motifs ( e . g . , [32] ) , as well as regulatory proteins , increasing the difficulty of identifying regulatory determinants in the genome . To compliment in silico searches for potential regulatory domains , an unbiased genetic approach can reveal components of the seemingly unique transcriptional regulation network in this important eukaryotic pathogen . By mapping genome-wide variations in transcript levels , it is possible to describe a broad architecture of regulatory variation . Mapping transcript level traits to gene expression quantitative trait loci ( eQTL ) identifies “local” or “distant” genetic contributions for which a regulatory polymorphism resides near the target transcript's gene , or the regulatory variation is displaced from the gene's position , respectively [13] . On the genome scale , many different transcripts mapping to a single distant eQTL suggests a transcriptional regulator with multiple targets . Multiple expression traits that map to a common local eQTL point to either a local cis mechanism ( e . g . , polycistronic-like transcription ) or a local structural determinant of gene expression ( e . g . , chromatin organization or a CNV such as a chromosomal amplification or deletion that alters the dosage of genes in a given locus ) . Traditional quantitative trait loci ( QTL ) mapping in the HB3 × Dd2 genetic cross of malaria parasites identified the major candidate gene responsible for chloroquine resistance [33] and multiple QTL and candidate genes contributing to quinine susceptibility [34] . HB3 , isolated from Central America ( Honduras ) , represents a “wild-type” parasite sensitive to the quinoline line of drugs [35] , while Dd2 was derived from a Laotian patient in whom chloroquine ( CQ ) therapy failed [36] . Dd2 is also resistant to pyrimethamine , as is typical of multidrug resistant ( MDR ) parasites . Moreover , prior to its use in the genetic cross , Dd2 was further selected in the laboratory for resistance to mefloquine ( MQ ) [37 , 38]; consequently the Dd2 genome has been reshaped by sequential drug selections and carries a genetic signature of , and effectively models , southeast Asian MDR parasites . The progeny from the HB3 × Dd2 genetic cross present a unique population in which to study the effects of drug selection on the parasite genome and its impact on gene expression . Understanding how a parasite's drug selection history has impacted its genetic plasticity has been the focus of persistent research [39] to devise sustainable drug therapies . Here , we use genome-wide expression profiling and linkage analysis in the segregating population of P . falciparum derived from the Dd2 × HB3 genetic cross to locate regions of the genome contributing to heritable levels of transcriptional variation that distinguish parasite strains . We show that transcript level variation is strongly influenced by parasite genotype , and the controlling eQTLs are distributed throughout the genome . Regulatory loci are observed both proximal to and distant from the genes they regulate . They reveal the profound influence of structural chromosomal polymorphisms underlying the identified eQTLs . Overall , we have taken the first step toward understanding how groups of genes are co-regulated . By combining genomic methods with classical genetics , we illuminate previously unrecognized transcriptional complexity and variation in P . falciparum , including a prominent role for drug selection and CNVs in shaping the regulation of transcript levels and their downstream constituents .
Using DNA microarrays , relative gene expression levels were measured across a genetic dimension in 34 progeny from a genetic cross of two laboratory adapted parasite strains . A high-resolution linkage map is available for this cross , and previous efforts have led to the mapping of genes involved in simple and multi-gene drug resistances [33 , 35 , 40] . For each progeny clone , relative transcript levels were determined for 7 , 665 probes representing 5 , 150 putative genes ( open reading frames [ORFs] ) . Experimental precision derives from nested biological replication: each parental allele for each probe was represented on approximately 18 different microarrays given the predominant 1:1 mendelian segregation of markers genome-wide [40] . Malaria parasites are haploid during the erythrocytic stages and the impact of genotype on phenotype is more direct than for organisms in which influence from heterozygous loci must be taken into consideration . Extensive expression level polymorphisms ( ELPs ) appeared to be segregating in this population based on the range of relative transcript level variation observed for the vast majority of genes ( Figure 1A ) . Relative gene expression levels , reported as log2 ( test/HB3 reference ) , for all 34 progeny clones were used as expression traits for mapping eQTL . Transcript levels that vary due to nongenetic sources , such as biological or experimental noise , would not generate significant eQTL; however , performing the large number of tests required in eQTL studies increases the chance of obtaining a type I error . To account for the 329 microsatellite markers tested in the linkage analysis , 1 , 000 permutations were conducted for each eQTL scan ( see Methods ) to establish corrected genome-wide significance levels of p ≤ 0 . 05 and p ≤ 0 . 01 for each expression trait . All nominal and corrected p-values are provided in Table S1 . A total of 874 ORFs ( 981 probes ) was differentially regulated by 1 , 063 significant eQTLs at a genome-wide significance level of p ≤ 0 . 05 ( Tables 1 and S1 ) ; approximately 18% of all P . falciparum genes displayed a significant genetic component leading to variation among the progeny at 18 h post-erythrocyte invasion ( hpi ) . We also considered a genome-wide significance level of p ≤ 0 . 01 and found 315 expression traits with significant linkage ( Table 1 ) . In addition to correcting for multiple tests across the large number of markers for each expression trait , we computed false discovery rates ( FDRs ) associated with the genome-wide ( corrected ) p-value thresholds to account for the testing of 7 , 665 expression traits ( Figure S1 ) . The 981 ( p ≤ 0 . 05 ) and 315 ( p ≤ 0 . 01 ) expression traits regulated by eQTLs corresponded to FDRs of 24% and 14% , respectively . Although this study was performed in a non-model organism with relatively few progeny , the observed FDRs are consistent with observations from model genetic systems ( e . g . , [25 , 27] ) . Nine hundred seven expression traits mapped to a single locus , and 74 traits mapped to multiple loci . Mapping in a small progeny population of 34 individuals might be expected to have limited power to detect multiple QTL for a given trait; however , this methodology was previously implemented for these same progeny to map five QTL contributing to quinine sensitivity and genetic effects accounting for as little as 10 . 5% of the phenotypic variation [34] . Nevertheless , the 74 traits for which multiple loci were identified could be disproportionately represented in the pool of expected false positives . In the present study , the detection of eQTLs was not obviously biased for particular phenotype characteristics . For example , eQTLs were detected across a range of relative expression levels among the progeny and span the range of transcript abundances ( Figure 1A and 1B , blue squares ) . This illustrates the strength of the segregation filter and the high degree of nested replication among these progeny . Indeed , we detected loci for expression traits that varied by as little as 1 . 65-fold between the lowest and highest expressing progeny . Most eQTLs were detected for traits whose relative expression levels in the progeny varied 1 ≥ log2 ( test/HB3 reference ) ≥ 2 ( see Methods ) . As observed for the range in relative expression levels in the progeny , a large majority of eQTLs was derived from low abundant transcripts , and there was no skewing of eQTL detection toward the higher-abundance transcripts ( Figure 1B , inset ) . This observation suggests that technical issues , such as labeling efficiency bias , were not influencing the global picture of eQTLs . As expected , traits with the widest range of variation among the progeny were more likely to have a significant eQTL ( Figure 1C ) . By considering eQTL numbers , positions , and effect sizes , it was possible to gain a sense of the potential regulatory complexity controlling expression traits . Local regulatory variation , wherein the causal locus of differential gene expression overlapped with the gene being regulated , accounted for 23 . 6% of the observed eQTLs ( Table 1 ) . The remaining majority ( 76 . 4% ) of the eQTLs denoted mutations that regulated distant transcripts . A comprehensive list of genes and their eQTLs detected at different thresholds is provided in Table S2 . Of the traits mapping to a single eQTL , 24% were local effects , commensurate with the overall eQTL pool . These were also the strongest genetic effects , as discerned by examination of the p ≤ 0 . 01 genome-wide eQTL threshold for detection ( Table 1 ) . Conversely , of the 74 probes that mapped to multiple loci , the most common ( 62% of the 74 probes ) mapped to at least two distant eQTLs . It should be noted that such patterns of multi-“trans” factor regulation is much more difficult to detect in a small mapping population; therefore , while our data suggest complex “trans” regulators , we were limited to detecting only the largest genetic effects , no doubt underestimating the genetic complexity of transcriptional regulation in this species . In addition to numbers and effect sizes , the genome-wide distribution of eQTLs can identify the regulatory architecture driving expression variation . Regulatory loci resided on each chromosome , ranging from as few as 17 eQTLs on Chr 6 to 513 on Chr 5 ( Figure 2; Table 2 ) . Of the 329 informative positions in the genome defined by recombination in the combined progeny pool [40] , 203 loci harbored at least one eQTL , and 122 had multiple eQTLs ( Figure 2 ) . The 81 genome positions with a single eQTL ( singletons ) influenced expression of 32 local genes and 49 distant genes . Local effects were more common than distant effects in the singleton group ( 40% versus 24% of the total eQTLs , respectively ) . This could be because local eQTLs contributed to larger genetic effects and were thus more readily detected . Several loci influenced the expression of a very large number of genes across the P . falciparum genome . Using permutation tests ( p ≤ 0 . 05; n = 1 , 000 ) , we identified 12 regulatory hotspots ( see Methods ) , each of which drove transcriptional changes of as few as 14 genes or as many as 182 genes ( Figure 2; Table 3 ) . These hotspots accounted for 63% of the detected eQTLs . More than half of the eQTL hotspots ( seven of the 12 ) were found on Chr 5 ( Table 3 ) . These included the two largest eQTL hotspots that mapped to adjacent positions on Chr 5 ( 68 . 8 cM and 65 . 9 cM , respectively ) . The remaining eQTL hotspots were on Chr 3 ( one hotspot ) , Chr 7 ( two ) , Chr 9 ( one ) , and Chr 12 ( one ) . eQTL positions are defined with respect to their nearest independently mapped microsatellite marker; however , for any one expression trait , the resolution of a locus is a function of the genetic recombination resolution ( i . e . , cM distance ) . Although the high recombination rate ( 15 kb/cM ) in P . falciparum [40] can facilitate physical mapping resolution to within tens of kilobases , it was not possible to know the exact degree to which neighboring loci overlap physically . Some overlap was expected , and this overlap influenced the designation of the total number of hotspots but did not influence the overall numbers of transcripts regulated by the adjacent segments . Consequently , under different criteria , neighboring loci would combine to generate fewer regulatory hotspots with more associated expression traits per hotspot . For example , when regulatory hotspots at adjacent markers were grouped together , the number of hotspots decreased to six , with hotspots 5_0 . 0 and 5_5 . 7 ( Chr number_cM distance of genetic marker ) , 5_11 . 4 and 5_20 . 0 , and 7_20 . 2 and 7_28 . 9 coalescing to three hotspots . Because this distribution of hotspots was critical to our interpretation of novel regulatory features of P . falciparum , we further compared the distribution of all eQTLs identified at the p ≤ 0 . 05 genome-wide significance threshold to those at identified at p ≤ 0 . 01and found a highly similar pattern of genome-wide eQTL clusters ( R2 = 0 . 9229 ) and retained seven of the 12 total eQTL hotspots ( 3_0 . 0 , 5_0 . 0 , 5_11 . 4 , 5_20 . 0 , 5_65 . 9 , 5_68 . 8 , and 12_103 . 3 ) . Notably , some eQTL hotspots coincided with structural copy number variations that define their probable physical limits ( described below ) . eQTL hotspots consisting of distant regulatory effects point to pleiotropic regulators [16] . Alternatively , eQTL hotspots composed of many local eQTLs likely result from chromosomal structural events , such as sequence amplifications or deletions that impact the expression of resident genes . Cis- and trans-acting hotspots may coincide if , for example , a deletion includes transcription factors that act at distant sites , i . e . , the genes within the deletion will have reduced transcription as will the distant genes they regulate . The vast majority ( 86% ) of the “linked” eQTLs in P . falciparum hotspots regulated the transcription of unlinked genes . Only a single hotspot ( Chr 12 ) was found to correspond to mostly local regulation of transcripts . Furthermore , when considering a reduced eQTL hotspot threshold ( ten or more co-mapping expression traits ) , we uncovered one additional locus with a high proportion of local genes ( Chr 2 ) , compared to an additional five loci corresponding to distant gene regulation ( Figure 2 , inset ) . It was of interest to compare the eQTL hotspots with previous comparative genome hybridization studies that identified chromosomal amplifications and deletions in the parental P . falciparum clones . The two eQTL clusters in the present study that predominantly regulated local gene expression ( Figure 2 , inset , see asterisks ) included a Chr 12 hotspot coinciding with a CNV in the underlying DNA segments [41] . Three hotspots aligned with sequence amplification events ( eQTL hotspots 5_65 . 9 , 5_68 . 8 , and 12_103 . 3 ) [41–43] and one hotspot ( 9_97 . 7 ) corresponded to a deleted segment from the HB3 parent [44] . Four hotspots were linked to loci rich with cytoadherence and highly polymorphic surface antigen genes such as cytoadherence linked asexual protein ( CLAG ) genes , rifin genes , and var genes ( 3_0 . 0 , 5_0 . 0 , 5_5 . 7 , and 7_28 . 9 , respectively ) [1] , of which three were located in the sub-telomeres and one ( 7_28 . 9 ) at an internal ( chromosomal ) var cluster [45 , 46] . One hotspot ( 7_20 . 2 ) was observed near the pfcrt drug-resistance locus , and the remaining three ( 5_11 . 4 , 5_20 . 0 , and 5_48 . 7 ) did not coincide with a previously identified CNV , drug resistance locus , or other highly polymorphic region of the genome . To assess the genetic basis for transcriptional variation mapping to the Chr 5 amplicon ( markers 5_65 . 9 and 5_68 . 8 ) , we evaluated the relative gene expression levels within the progeny of the HB3 × Dd2 cross ( Figure 3A ) . These loci span the previously reported amplification event involving the multiple drug resistance gene , pfmdr1 ( PFE1150w ) , found in Dd2 but not in HB3 [35 , 47] . We merged the genes whose transcripts map to these prominent eQTLs into a single set of genes for further analysis , because the posterior probability for the probes mapping to each of these markers generally spanned both markers and the entire amplicon . Of the 269 transcripts mapping to the amplicon , 85% ( 228 ) were expressed at higher levels in the Dd2 parent compared to the HB3 parent; furthermore , these genes were expressed at higher levels in the progeny that inherited the Dd2 alleles at these loci , as expected for an overexpressed positive transcriptional regulator located within the amplicon . The remaining 15% ( 41 ) of the transcripts regulated by these hotspots were generally expressed higher in those individual genotypes inheriting the HB3 alleles at these loci . In the simplest scenario , this would argue for a trans-acting , negative regulator located in the amplified region . Clear segregation of the two allelic effects was evident among these transcripts ( Figure 3A ) , demonstrating strong regulation associated with the number of copies of this amplicon carried by individual progeny . We were surprised to observe clear subsets of up-regulated and down-regulated genes due to this amplification . While there is no need to presume that the underlying mechanisms regulating these genes would necessarily lead to solely up-regulation , we considered the possibility that the down-regulated genes were more likely false positives . We observed no statistical difference between the p-values associated with genes regulated in each direction , and also found the same proportion of genes up- and down-regulated at the highly significant p ≤ 0 . 01 genome-wide eQTLs ( unpublished data; see Table S1 ) , supporting the validity of these two regulatory directions of transcription . In Dd2 , the pfmdr1 amplicon contains 14 ORFs [41] and is repeated three times [47] ( Figure 3B ) . This eQTL hotspot indicates that a polymorphism ( s ) in or linked to the amplicon is regulating genes across the genome; therefore , we closely examined genes in the amplicon for possible transcriptional regulators ( Figure 3C ) . Of the 14 genes , nine are of unknown function , i . e . , hypothetical proteins . We used the predicted amino acid sequences for each of the nine hypothetical proteins to search the PfamA database [48] for protein domains characteristic of DNA binding function , a domain function common to transcription factors . Three genes with significant hits to known protein domains were identified: PFE1130w had a hit to a Duf803 domain common to the drug metabolite transporter superfamily of genes ( E-value = 0 . 0001 ) , PFE1135w had a hit to a domain common to iron-sulfur cluster biosynthesis genes ( E-value = 2 . 9 × 10−24 ) , and PFE1145w had hits to two tandem zinc-finger CCCH-type domains ( E-value = 0 . 18 and 0 . 00084 ) , spaced ten amino acids apart . Two additional genes with unknown function ( PFE1095c and PFE1110w ) were identified with weaker hits to nucleic acid binding domains ( YL1 nuclear protein and transcription initiation factor TFIID 23–30 kDa subunit , respectively ) . PFE1110w previously has been hypothesized to be a P . falciparum ortholog for TFIID TAF10 , a basal transcription factor associated with RNA polymerase II activity in other eukaryotic systems [28] . These three genes , whose proteins encode putative DNA binding domains ( PFE1110w , PFE1110w , and PFE1145w ) , are primary candidate genes with a role in transcriptional regulation requiring experimental validation . Distant regulatory variation accounted for most of the transcripts mapping to the eQTL hotspots , suggesting possible master regulators residing within the hotspots . To ascertain co-regulation of possible functionally related genes , we sought Gene Ontology ( GO ) enrichment ( Table 3 ) . Focusing on the Chr 5 pfmdr1 amplification hotspot ( 5_65 . 9 and 5_68 . 8 ) , we found that protein modification via the proteasome complex was a prominent , differentially regulated GO category within our experimental population ( Table 4 ) . Notably , four additional hotspots contained differentially regulated genes involved in protein modification ( eQTL hotspots at 5_48 . 7 , 7_28 . 9 , 9_97 . 7 and 12_103 . 3 ) , implying that transcriptional regulators of post-transcriptional machinery may play a prominent role in phenotypic differences between HB3 and Dd2 ( Tables 3 and S2 ) . This strong presence of genes involved in protein modification highlights a biological pathway that is alternatively regulated in the parental parasites' genetic backgrounds . Complete results for calculated GO enrichment values for each of the eQTL hotspots are provided ( Table S3 ) . HB3 and Dd2 are known to diverge in the duration of their erythrocytic cell cycle [6 , 49]; consequently , we evaluated the possibility that eQTLs corresponded to events of the cell cycle . We first binned genes represented on the microarray by their peak expression time in the life cycle using the same microarray platform [4 , 6] . We then compared these peak-time bins with genes regulated by eQTLs ( Figure 4 ) . No bias toward stage-specific gene expression was present in the eQTL pool .
An emerging theme in evolutionary biology recognizes that mutations in regulatory sequences can account for major physiological differences between strains even when coding genes are relatively unchanged [18] . While in numerous species variation in gene expression serves as a storehouse for phenotypic variation [16 , 21–23 , 50] , it has been argued that P . falciparum is unusual in exhibiting remarkably little heritable , strain-to-strain variation [13] . This conclusion was based on a study by Llinas et al . that compared broad expression “cascades” across the complete erythrocytic cell cycle for three different parasite clones [6] , including HB3 and Dd2 . Their approach assessed transcriptional profiles by measuring a parasite's transcript levels against pooled samples from the same parasite clone and was not designed to directly assess the relative abundance of transcripts between strains . In contrast , the current study measured relative transcript abundance of the Dd2 parent and each progeny clone against a common reference , HB3 , the other parent from the genetic cross . The two studies are not necessarily contradictory , but rather illuminate different features of transcription in this lethal malaria parasite species emphasizing both the robust transcriptional program that has been so well characterized in this species and the subtle but abundant variation that exists between strains . Viewing transcription across a genetic rather than a developmental dimension allows us to tease out variations in transcriptional regulation that could have important implications for the Plasmodium regulatory network and its role in adaptive evolution . Until now , no study of malaria parasites ( to our knowledge ) has specifically examined the genetic inheritance of transcriptional variation . More generally , our work illustrates the concept that genome-wide readouts of segregating natural variation point to polymorphic regulatory loci , a finding particularly relevant in light of recent observations by Daily et al . ( 2007 ) [14] that transcriptional profiles associated with distinct metabolic states in blood stage forms of malaria are observed in parasites isolated directly from patient blood; these metabolic states are proposed to influence the course of infection and virulence . In the present study involving cultured parasites under controlled laboratory conditions , we find heritable variation in expression levels to be as extensive as that reported for other organisms . We also find transcription levels to be regulated by few , predominantly distant , eQTL hotspots . These co-regulated genes underpin altered biological processes of the regulatory network and provide an evolutionary path to phenotypic change that is distinct from potentially deleterious coding mutations that alter protein function and would thus be poorly tolerated in haploid parasites . Given the relatively small number of progeny available from the HB3 × Dd2 cross , the eQTLs presented probably capture only the largest genetic effects and certainly underestimate the total number and complexity of regulatory polymorphisms . We find that a large proportion ( 76 . 4% ) of the genes with variable transcript levels is influenced by distant regulators , dominated by several regulatory hotspots . Distant regulatory variation is often associated with trans-acting transcription factors . While P . falciparum has been described as having a paucity of transcription factors [1] , De Silva et al . ( 2008 ) recently experimentally validated a family of specific transcription factors and their DNA binding sites [31] , suggesting that the machinery for complex regulation in P . falciparum is present yet difficult to discern by standard homology searches . Our data support the presence of at least a few additional potent regulatory factors , and eQTLs can be dissected to locate candidate transcriptional regulating genes in apicomplexans [30] . Regulatory loci can be identified through this method irrespective of the specific biochemical functions ( e . g . , DNA binding , nuclear localization , or protein phosphorylation ) and is therefore uniquely suited to identify loci harboring genetic determinants that act as traditional and atypical or unique modes of genome regulation . Although most eQTLs identified in this study are trans-acting , the strongest genetic effects are due to local regulatory polymorphisms ( Table 1 ) . These polymorphisms are more likely to occur individually than in clusters . Local linkages arise from classical cis-acting mechanisms , e . g . , a polymorphism in the gene regulatory region; various other scenarios are also possible , including CNVs , splicing , mRNA decay , regional chromatin structure [23] , or even unconventional autologous protein–nucleic acid interactions [51] . Such cis elements would be expected to have strong genetic effects due to the direct molecular control of the transcript levels . We are aware that eQTL mapping has the potential to overestimate local regulatory variation if substantial sequence variation is present between alleles due to reduced hybridization efficiency , effectively mimicking lower gene expression levels [52] . However , for the divergence between HB3 and Dd2 , less than 10% of the probes ( expression traits ) with eQTLs arose from the highly polymorphic antigenic gene families including vars , rifins , and stevors , nearly the same proportion of these genes represented in the overall set of 7 , 665 probes ( 10% versus 8% , respectively ) . In fact , we detected a relative paucity of local eQTLs compared to a major role for trans-acting mechanisms controlling the majority of the observed ELPs in the HB3 × Dd2 genetic cross . We find that genetic effects due to distant regulatory variation are smaller and more likely partnered , indicative of genetic complexity of the regulatory network . Regulatory hotspots have been hypothesized to contain “master regulators” with the effective mutations having pleiotropic effects [24 , 26] , wherein one DNA sequence variant impacts multiple expression traits with potential broad impact on downstream classical phenotypes . Another striking feature of the regulatory architecture is the prominent role of CNVs in directing transcription variation , particularly from regions previously associated with drug resistance traits . Originally , the HB3 × Dd2 cross was generated to study chloroquine resistance in malaria parasites [35]; fortuitously , the differing geographic and drug selection histories of these genomes represent independent solutions for survival in their respective environments . It is also likely that adaptations to intensive , long-term drug selection by CQ , and subsequent selections by pyrimethamine and MQ , broadly affect biological processes , perhaps pointing to physiological compensation of mutations in drug resistance genes . Our analysis demonstrates GO term enrichment for genes regulated by each of the eQTL hotspots . Genes related to post-transcriptional protein modification are , intriguingly , enriched in six of the 12 eQTL hotspots , including the two hotspots coinciding with the pfmdr1 amplicon on Chr 5 ( Tables 3 and 4 ) , suggesting that physiological differences between the parental parasite clones , HB3 and Dd2 , may be at least partially due to post-translational modification/regulation of proteins . This is interesting given the recent findings in yeast that protein levels were unchanged in individuals with increased aneuploidy despite increased gene expression levels , suggesting active regulation at the protein level [53] . These functional differences are prime candidate processes driving phenotypic differences derived from the genetic adaptations associated with multiple drug resistance in the Dd2 parent line and may shed light on post-transcriptional regulation of the genome key to adaptation in the malaria parasite . DNA sequence amplification events , occurring as multiple tandem copies , can influence gene expression both locally and distantly through dosage effects . Previous laboratory drug selection studies demonstrated that parasites exhibited amplification of the genomic region carrying pfmdr1 on Chr 5 that coincided with a loss of sensitivity to quinine and MQ [54] . Additionally , work performed on parasite isolates from MDR populations confirmed amplification of the pfmdr1 locus is the primary contributor for resistance to MQ [55] . The focus of previous studies on this locus has been the pfmdr1 gene itself , and the roles of coding mutations and gene dosage in influencing drug sensitivities [54–56] . Our data emphasize a potentially critical role of neighboring genes , e . g . , increased copy numbers of the entire pfmdr1-containing locus not only coincides with decreased sensitivities to common antimalarial compounds but significantly alters gene expression throughout the genome . In addition , genes regulated by the Chr 5 amplicon are both positively and negatively regulated , potentially indicating multiple regulators or multiple mechanisms for gene regulation contained in this locus . Figure 3C illustrates transcriptional regulator candidates , PFE1095c , PFE1110w , and PFE1145w , residing on the pfmdr1 amplicon , including a tandem zinc-finger domain protein ( PFE1145w ) . Notably , eQTL hotspots are not confined to transcription factors but could also point to novel regulatory mechanisms . The degree to which drug selection on malaria parasites can impact the expression of genes across the genome is evidenced by the sweeping affects the Chr 5 pfmdr1-containing amplicon has on genome-wide gene expression . In general , Chr 5 has been a primary target for genomic adaptations in the HB3 and Dd2 parental parasites; determining whether Chr 5 is unique to this parasite population or whether it is a key player in all natural parasite adaptations is relevant to understanding the propensity of certain parasite clones to rapidly become resistant to multiple drugs . Identifying the mechanisms of transcriptional regulation and the functional relationships among the co-regulated genes will provide crucial information regarding potential antimalarial targets . It is likely that the drug-selection history of Dd2 affecting relatively few “resistance” genes has had a broader impact with significant implications for parasite biology in the form of drug sensitivity modulators , compensatory mechanism in physiology , and/or simple hitchhiking effects that could themselves acquire an adaptive role in subsequent selection . In light of historical selection bottlenecks and its impact on multi-drug resistance and virulence , even “simple” resistance mechanisms , e . g . , the point mutation in the chloroquine resistance transporter gene pfcrt , conferring resistance to CQ , may elicit a complex expression signature . Here , we illustrated how eQTL scans can uncover nontraditional patterns of transcriptional regulation underlying strain-level variation and regulatory hotspots in P . falciparum . This study bolsters a significant and unexpected role for divergent transcription as a source of phenotypic variation and evolution in malaria parasites and elevates a role for structural changes ( e . g . , CNVs ) as having potentially prominent consequences in the cell , perhaps contributing to adaptive evolution .
Parent and progeny parasites ( i . e . , clones ) of the HB3 × Dd2 genetic cross were obtained from the original cloned stocks [35] . The HB3 × Dd2 genetic cross consists of 35 haploid progeny ( 34 of which were available for this study ) , mimicking , in effect , recombinant inbred lines for linkage analysis . Each progeny was previously genotyped for the 329 informative microsatellite markers spanning the 14 chromosomes ( 23 Mb ) at a resolution of approximately 17 kb/cM [40] . Parasites were cultivated in human erythrocytes ( RBCs ) by standard methods [57 , 58] utilizing leukocyte-free human RBCs ( Indiana Regional Blood Center , Indianapolis , Indiana ) suspended in complete medium ( CM ) [RPMI 1640 with L-glutamine ( Invitrogen , Carlsbad , California ) , 50 mg/l hypoxanthine ( Sigma-Aldrich , St . Louis , Missouri ) , and 25 mM HEPES ( Calbiochem , San Diego , California ) ; 0 . 5% Albumax II ( Invitrogen ) , 10 mg/l gentamicin ( Invitrogen ) , and 0 . 225% NaHCO3 ( Invitrogen ) ] at 5% hematocrit . Cultures were maintained independently in sealed flasks at 37 °C under an atmosphere of 5% CO2 , 5% O2 , and 90% N2 . Parasitemia was monitored and maintained at 5%–7% . Parasites were synchronized using two consecutive 5% sorbitol treatments for two generations , followed by one cell cycle without treatment . The 18 hpi time point was determined by light microscopy . Samples were flash frozen in liquid nitrogen prior to extraction of total RNA . Total RNA was isolated as previously described using TRIzol reagent [4] . cDNA was synthesized using the Ovation Aminoallyl RNA Amplification and Labeling Kit ( cat . # 2101–12; NuGEN Technologies , San Carlos , California ) and prepared for hybridization to microarrays as previously described [4] . Microarrays contained 7 , 665 70mer oligonucleotide probes representing 5 , 150 Plasmodium falciparum ORFs kindly provided by Dr . Joseph DeRisi ( University of California , San Francisco , San Francisco , California ) . The majority of the oligonucleotides for printing the array corresponded to the Qiagen Operon set ( Operon Biotechnologies , Huntsville , Alabama ) , but some additional sequences matched ORFs in the original P . falciparum sequence reads . The oligonucleotides were printed on polylysine-coated slides using a new generation , ultra fast , linear servo driven DeRisi microarrayer . Slides were post-processed and hybridized in 3× SSC at 63 °C for 12 h as previously described [4] . Slides were scanned in an Axon GenePix 4000B microarray scanner ( Axon Instruments , Union City , California ) with 532 nm ( 17 mW ) and 635 nm ( 10 mW ) lasers . Data were collected as an image file , gridded , and converted into a text file using Genepix 3 . 0 software ( Axon Instruments ) . Each parental and progeny cDNA ( labeled with Cy5 ) were hybridized to a common reference sample of the HB3 parental clone ( labeled with Cy3 ) , allowing for direct comparisons between the results from each microarray; we tested 34 progeny clones , for a total of 36 hybridizations . Lowess smoothing of the raw expression data was used to normalize across all microarray slides using TIGR's MultiExperiment Viewer ( MeV ) v4 . 0 and Microarray Data Analysis System ( MIDAS ) v2 . 19 software ( http://www . tm4 . org/ ) . Gene expression values are reported as log2 ( sample/HB3 reference ) of each normalized fluorescent signal . The DecisionSite software ( Spotfire , Somerville , Massachusetts ) was used to generate relative gene expression heat maps . For each expression trait , interval mapping linkage analysis was performed using the Bayesian approach implemented in the Pseudomarker v2 . 03 package [59] executed in MATLAB ( The MathWorks , Natick , Massachusetts ) . Marker genotypes were used directly ( i . e . , imputation of “pseudomarkers” was not necessary given the high-resolution , uniform marker coverage of the P . falciparum genome [40] ) , and the marker corresponding to the highest LOD score for each expression trait was retained as the eQTL position . The empirical genome-wide significance was determined for each trait by permutations [60 , 61] in which progeny haplotypes were randomly associated with expression trait values and linkage analysis performed 1 , 000 times . The nominal p-values for each trait were converted to genome-wide corrected p-values using the permutation distribution of the maximum LOD score for each trait . This approach corrects the nominal p-values for the multiple testing of 329 markers . LOD score thresholds corresponding to p ≤ 0 . 05 and p ≤ 0 . 01 genome-wide significance thresholds were determined for each expression trait . To account for the multiple testing of 7 , 665 expression traits , we followed the work of Petretto et al . ( 2006 ) [27] in which we use the lowest corrected p-value for each expression trait to calculate the FDR , using the q value method of Storey and Tibshirani ( 2003 ) [62] . This approach was motivated by the fact that the underlying procedure to estimate the FDR works well under “weak” dependence of the considered features . To be conservative , an eQTL was classified as “local” if the gene was located on the same chromosome as the linkage or as due to “distant” regulatory variation if located elsewhere in the genome . To locate regulatory hotspots within the genome , the total number of expression trait mapping to eQTL at the p ≤ 0 . 05 genome-wide significance threshold , specifically 1 , 063 , were randomly assigned to the markers ( loci ) in the genome . The number of linkages observed at the locus with the most linkages was retained and this process was repeated 1 , 000 times creating a random distribution . From this distribution , linkage hotspots were identified as the number of traits mapping to a given locus exceeding the 95% distribution frequency derived from the 1 , 000 permutations ( i . e . , at p ≤ 0 . 05 for n = 1 , 000 ) . Genetic markers harboring 14 or more traits with an eQTL were not expected by chance . Genes comprising the groups of genes at the eQTL hotspots were sifted for enriched GO terms to identify possible co-regulated genes contributing to the same biological processes . To obtain statistically significant measurements of overrepresented GO terms , we applied a Fisher's exact test as implemented in GOStat [63] . GO terms significant at p < 0 . 01 for each regulatory hotspot were retained and reported . The peak hours of gene expression for all genes represented on our microarrays were collected from prior microarray experiments using the same platform ( DeRisi Lab Malaria Transcriptome Database; http://malaria . ucsf . edu/ ) . Genes were binned at 5 h intervals based on their respective peak hour of expression across the HB3 and Dd2 parasites' life cycles . The percentage of genes whose probes showed significant linkage within each of the 5 h bins was calculated . Regression analyses were performed using GraphPad Prism v4 . 0 ( GraphPad Software , La Jolla , California ) . Relative transcript abundance was qualitatively estimated for each probe on our microarray by using the sum of median ( SOM ) intensity from each array , similar to a previous study using the same microarray platform [6] . For each probe , the highest 18 SOM intensity values across all progeny were retained and the median was calculated . Each probe was then binned according to its corresponding relative transcript abundance . The percentage of probes showing significant linkage within each of the relative abundance bins was then calculated . The range of relative gene expression levels for each transcript was calculated from the progeny with the highest and lowest relative expression levels compared to the reference HB3 samples , excluding outliers . Outliers were determined as those progeny clones expressing at levels ± 2 standard deviations from the mean , calculated for each probe independently . Probes were then binned according to their relative expression level range in the progeny . The percentage of probes with significant linkage within each of the expression range bins was then calculated . | Development of the malaria parasite , Plasmodium falciparum , in the blood is driven by a number of different genes expressed at different times and at different levels . Exactly what influences such transcriptional changes remains elusive , particularly in regard to important phenotypes like drug resistance . Using cDNA microarray hybridizations from the progeny of a Plasmodium genetic cross , we mapped gene expression quantitative trait loci ( eQTLs ) in an experimental population of malaria parasites . Each gene's transcript level was used as a segregating phenotype to identify regions of the Plasmodium genome dictating transcriptional variation . Several regulatory hotspots controlled the majority of gene expression variation , mostly via trans-acting mechanisms . One , influencing the largest number of transcripts , coincided with an amplified region of the genome traditionally associated with multiple drug resistance ( MDR ) . Overall , integration of two functional genomic tools ( gene mapping and transcript quantitation ) has revealed a system-wide rewiring of the parasite transcription network: pleiotropic phenotypic variation , driven by drug selection on genome structure that may be attributed in large part to adaptive copy number polymorphisms in the parasite . These structural variants alter the expression of genes within the amplicon as well as many genes scattered across the Plasmodium genome . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"infectious",
"diseases",
"genetics",
"and",
"genomics"
]
| 2008 | Regulatory Hotspots in the Malaria Parasite Genome Dictate Transcriptional Variation |
Autism spectrum disorders ( ASD ) are neurodevelopmental disorders with phenotypic and genetic heterogeneity . Recent studies have reported rare and de novo mutations in ASD , but the allelic architecture of ASD remains unclear . To assess the role of common and rare variations in ASD , we constructed a gene co-expression network based on a widespread survey of gene expression in the human brain . We identified modules associated with specific cell types and processes . By integrating known rare mutations and the results of an ASD genome-wide association study ( GWAS ) , we identified two neuronal modules that are perturbed by both rare and common variations . These modules contain highly connected genes that are involved in synaptic and neuronal plasticity and that are expressed in areas associated with learning and memory and sensory perception . The enrichment of common risk variants was replicated in two additional samples which include both simplex and multiplex families . An analysis of the combined contribution of common variants in the neuronal modules revealed a polygenic component to the risk of ASD . The results of this study point toward contribution of minor and major perturbations in the two sub-networks of neuronal genes to ASD risk .
Autism is the most severe end of a group of neurodevelopmental disorders referred to as autism spectrum disorders ( ASDs ) . ASD is a heterogeneous genetic syndrome characterized by social deficits , language impairments and repetitive behaviors . Although it is known that ASD has a genetic basis [1]–[3] , its genetic architecture is unclear . Previous studies have identified both common and rare variants , including de novo mutations , as risk factors for ASD [4] , [5] . However , how much of the genetic risk can be attributed to rare versus common alleles is unknown . Since ASD is relatively common with a complex pattern of inheritance it was previously suggested to be caused by multiple common variants [4] , [5] , where each of the common variants only makes a small contribution to the risk of disease . The principal methods for discovering common variations related to ASD include association studies of candidate genes , and more recently genome-wide association studies ( GWAS ) [6] , [7] . Despite major efforts to identify common variants associated with ASD , the success so far has been limited [6] , [7] . At the same time , an increasing number of studies have shown that rare and de novo mutations contribute to ASD [8]–[12] . These rare variants include mutations causing single-gene disorders , cytogenetically visible chromosomal abnormalities , and more recently the identification of rare and de-novo copy number variations ( CNVs ) [8]–[10] , [12] . The genes already known to be disrupted by rare variants still account for only a small proportion of the cases , because many of them have only been found in one or very few individuals [13] . Other findings that further complicate the interpretation and utilization of rare variants is the fact that many of the same variants have been found in patients with distinct illnesses ( such as schizophrenia , epilepsy , and intellectual disability ) , as well as in healthy family members or controls [14] . This genetic heterogeneity constitutes a considerable obstacle to establishing a thorough understanding of the etiology of ASD . One promising avenue of exploration is to find key molecular pathways and apply system-wide approaches to determine the function of the genes disrupted in ASD . Delineating these pathways will not only lead to insights into the molecular basis of ASD , but may ultimately lead to potential treatments . Most attempts so far have concentrated on determining the functional connection between genes affected by CNVs . These studies showed that many of the genes are related to synapse development , cellular proliferation , neuronal migration and projection [15] , [16] . Another way to identify the connection between autism susceptibility genes is based on studying protein interactions for genes mutated in syndromes associated with autism [17] . This study suggested that shared molecular pathways are implicated in different ASD associated syndromes [17] . A different approach to identify key molecular pathways is based on gene expression , and relies on the assumption that co-expressed genes are functionally related [18] . A weighted gene co-expression network analysis ( WGCNA ) of specific human brain regions ( cerebral cortex , cerebellum and caudate nucleus ) demonstrated that the transcriptome of the human brain is organized into modules of co-expressed genes that reflect different neural cell types [19] . Recently , this type of analysis was also applied to compare the expression profiles of three brain regions from autistic and control individuals . This network analysis led to the identification of specific co-expression modules that are differentially expressed in ASD and controls [20] . These included a neuronal module that was enriched for genes with low GWAS P-values , suggesting that the differential expression of this module between cases and controls reflects a causal relationship [20] . In the current study , we constructed a gene network using a WGCNA approach based on a widespread survey of gene expression undertaken by the Allen Human Brain Atlas project ( http://www . brain-map . org ) . This survey of gene expression provides unprecedented coverage across different brain regions . We found modules which are associated with specific neural cell types , and modules with highly significant enrichment for specific cellular processes . We used the gene network to address several fundamental questions regarding the genetic architecture of autism . First , can we identify gene networks that are perturbed by rare variations that in turn lead to ASD ? Second , can we identify gene networks that are perturbed by common variations ? Third , do rare and common variations converge on the same molecular pathways or do they represent diverse biological etiologies ? Lastly , can we integrate the gene network with GWAS results to predict potential genes associated with ASD ? To answer these questions we integrated the co-expression network with the results of autism GWAS and with known rare mutations . We identified specific modules that are enriched for both rare and common variations that are potentially associated with ASD risk . We replicated the enrichment in two additional samples . The modules showing the highest enrichment for rare and common variants in ASD included highly connected genes that are involved in synaptic and neuronal plasticity , and are expressed in areas associated with learning and memory and sensory perception . Additionally , we found that a genetic risk score based on these modules significantly predicts ASD risk . Taken together , these results suggest a common role for rare and common variations in autism , and illustrate how rare and de novo mutations , in conjunction with common variations , can act together to perturb gene networks involved in neuronal processes , and specifically neuronal plasticity . Furthermore , the modules found in this study may serve as starting points for designing potential therapeutic interventions for ASD .
In order to construct a robust network of the human brain transcriptome we used the Allen Brain Atlas RNA microarray data , which to the best of our knowledge , is one of the most comprehensive expression profiling of different regions of the human brain . The Allen Brain Atlas RNA microarray data includes 1340 measurements from two individuals , representing the entirety of the adult human brain . We generated a network based on a combined dataset , as the two individuals exhibited high correlations in trends of expression and connectivity ( Figure S1 ) . The network included 19 modules of varying sizes , from 38 to 7385 genes ( Figure 1A , Table S1 ) . The different modules are color-coded for presentation purposes and referred to hereafter based on these colors ( Figure 1A ) . To study the modules specificity to brain areas , we plotted the modules eigengenes across different anatomical regions , and observed that none of the modules were specific to one anatomical region ( Figure S2 , Table S2 ) . We hypothesized that the modules may correspond to cell types or subcellular compartments , which are distributed in different densities across different brain areas . We thus tested the modules for enrichment of specific neural cell populations based on gene expression levels in neurons , astrocytes and oligodendrocytes , as found in a survey performed on mouse brain cells [21] . One module , Magenta , stood out as showing a very high enrichment for genes up-regulated in astrocytes ( relative risk [RR] = 3 . 93 , P<0 . 0001 ) ( Figure 1B ) . Three other modules showed enrichment for genes up-regulated in neurons ( Figure 1B ) . Of these , Salmon showed the highest enrichment signal ( RR = 3 . 18 P<0 . 0001 ) , in addition to two other modules , Lightgreen and Grey60 , that also showed substantial enrichment for neuronal genes ( RR = 2 . 75 and RR = 2 . 16 , respectively , with P<0 . 0001 in both ) . Enrichment for genes up-regulated in oligodendrocytes was found in the Blue module ( RR = 3 . 04 , P<0 . 0001 ) , and the Greenyellow module ( RR = 1 . 92 , P<0 . 0001 ) ( Figure 1B ) . To test whether the modules were specifically enriched for the most representative genes of each cell type , we used a score of the relative expression in a particular cell type relative to other cells ( Figure S3 ) . Notably , the modules with the strongest enrichment for genes expressed in neurons , astrocytes and oligodendrocytes showed specific enrichment for the most up-regulated genes in the corresponding cell types ( Figure S3 ) . We tested the degree of overlap between these cell type-specific modules and ones that were discovered in a previous study that constructed a gene co-expression network that was largely based on differences between individuals rather than between brain areas [19] . The general comparison of the two networks is described in Table S3 . A significant overlap in gene content between the studies was observed for an oligodendrocyte module ( Blue , RR = 5 . 29 , P<2×10−5 ) and the astrocyte module ( Magenta , RR = 5 . 62 , P<2×10−5 ) . Similarly , the Salmon module significantly overlapped with a previously identified cortical module ( RR = 9 . 73 , P<2×10−5 ) and the Grey60 module showed a high overlap with a parvalbumin-expressing cortical interneuron module ( RR = 83 . 44 P<2×10−5 ) . The module Lightgreen had no significant overlap with any of the previously identified modules . To further characterize the different modules we used gene ontology ( GO ) analysis ( Table S4 ) . The Salmon module was enriched for genes active in the synapse ( P = 2 . 2×10−6 ) and involved in synaptic transmission ( P = 4 . 8×10−3 ) , as well as for genes in the calmodulin-binding pathway ( P = 9 . 9×10−4 ) . The Lightgreen module was also enriched for genes active in the synapse ( P = 1 . 6×10−5 ) . The GO analysis also showed a different module , Black , to be highly enriched for genes in the nucleosome core ( P = 1×10−31 ) . The representative of the gene expression profile of the Black module ( the module eigengene ) had the highest values in the corpus callosum and cingulum bundle , suggesting that this module may represent enrichment for cell bodies of glia cells ( Table S5 ) . In the Red module the genes having a positive relationship with the module eigengene were enriched for mitochondrion ( P = 2 . 9×10−40 ) , and the genes having a negative relationship were enriched for DNA binding ( P = 6 . 6×10−23 ) and regulation of transcription ( P = 2 . 2×10−21 ) . Another module , Pink , was highly enriched for genes containing a Kruppel-Associated Box domain ( P = 2 . 2×10−46 ) . This group of zinc finger transcription factors has been recognized as transcriptional repressors [22] . The Tan module was highly enriched for genes involved in the G-protein-coupled receptor pathway ( P = 2 . 6×10−50 ) , as well for genes involved in olfactory receptor activity ( P = 2 . 2×10−42 ) , hormonal activity ( P = 1 . 1×10−28 ) and HOX genes ( 1 . 6×10−11 ) . Another way to infer the function of the modules is based on the known function of highly connected genes with central positions within the modules ( “hub” genes ) . We explored the strongest connections in each module using Cytoscape software [23] ( Figure S4 ) . In the Magenta module , which was found to be highly enriched for genes up-regulated in astrocytes , the most connected gene was FGFR3 , which was reported to mark astrocytes and their neuroepithelial precursors in the CNS [24] ( Figure 1C ) . The Yellow module , which was highly significantly enriched for genes involved in protein translation ( P = 7 . 4×10−98 ) , presented as two separate sub-networks of genes ( Figure 1D ) . One group of highly connected genes is involved in protein translation , and the other group contains genes related to the function of microglia ( Figure 1D ) . The central components of the microglia sub-network include TYROBP , AIF1 , RGS10 , CX3CR1 , as well as other genes which are known to be involved in microglia function and regulation [25]–[27] . These results suggest that the module is representative of microglia which also show high protein translation associated with their high proliferation rate . Consistent with this observation , the module eigengene of the Yellow module was most highly expressed in the corpus callosum , where immature microglial progenitor cells accumulate [28] , [29] . Given that our analysis highlighted three groups of neuronal genes , the next step was to determine whether they represent three different types of neurons . To that end , we visualized the top connections in the three modules ( Figure 2A ) , and highlighted the brain areas showing the highest values for the first principal component of each module ( the module eigengene ) ( Figure 2B ) . The top connections in one module ( Grey60 ) included the genes KCNC1 , SCN1B , PVALB and HAPLN4 ( Figure 2A ) . These genes have been shown to be highly expressed in a group of fast-spiking , parvalbumin-expressing cortical interneurons [30] , [31] . The module was most expressed in the superior temporal gyrus , an area that receives auditory signals from the cochlea [32] , the dentate nucleus , which is a structure linking the cerebellum to the rest of the brain [33] , and the dorsal lateral geniculate nucleus , which is the primary relay center for visual information [34] . The eigengene of the Lightgreen module was most expressed in brain regions involved in sensory processes , including the inferior occipital gyrus and the lingual gyrus of the occipital lobe ( Figure 2B ) , which are involved in processing visual information [35] , [36] , and the post central gyrus , which contains the primary somatosensory cortex [37] . The module Lightgreen harbors highly connected genes involved in clathrin-dependent endocytosis in the synapse . These include SNAP91 ( also known as AP180 ) , VSNL1 ( also known as VILIP-1 ) , SYN1 and and STXBP1 [38]–[41] . The Salmon module included several highly connected genes ( FOXG1 , LHX2 , MKL2 , CDH9 and genes of the protocadherin family ) , which are all known to be involved in neurogenesis and neuronal plasticity in the developing brain [42]–[47] . FOXG1 , MKL2 and PCDH20 have also been shown to be involved in structural and functional plasticity of neurons in the adult brain [48]–[50] . Similarly , the eigengene of the Salmon module was most expressed in brain regions that are involved in learning and memory , including the hippocampus ( dentate gyrus and CA1 field ) and the dorsal striatum ( tail of the caudate nucleus and putamen ) ( Figure 2B ) . We sought to test whether autism genes affected by rare or spontaneous mutations are associated with specific modules . A list of 246 autism susceptibility genes was compiled using the SFARI gene database ( https://sfari . org/sfari-gene ) , and was restricted to the 121 genes with reported rare mutations in autism . Of these , 91% ( 109 genes ) were represented in our network . Genes on the list exhibited a significantly skewed distribution between the modules ( P = 0 . 025 , Fisher's test ) . Specifically , three modules showing up-regulation in neurons also showed the highest enrichment for autism risk genes . The most enriched module was the Salmon ( RR = 2 . 92 ) , followed by Lightgreen ( RR = 2 . 19 ) and Grey60 ( RR = 1 . 89 ) ( Figure 3A ) . To test whether CNVs also tended to be distributed in a non-random way among modules , we assembled a list of de-novo CNV events from a recent study [10] , and calculated enrichment to specific modules . As larger genes can be expected to harbor more CNVs by chance , and since neuronal specific genes are larger than average [51] , we corrected for gene size in our analysis ( see Materials and Methods ) . However , none of the modules showed significant enrichment for CNV events after correcting for gene size . Subsequently , we tested the distribution across modules of genes affected by common variants , as reflected by low P-values in a GWAS for autism , previously performed [6] on multiplex families ( with more than one member of the family with ASD ) from the Autism Genetic Resource Exchange ( AGRE ) ( Figure 3A ) . Notably , two of the three neuronal modules ( Salmon and Lightgreen ) , which also showed the highest enrichment for genes affected by rare variants , were also found to be significantly enriched for genes affected by common variants ( Salmon , P = 0 . 000030; Lightgreen , P = 0 . 0019; Bonferroni corrected P<0 . 05 ) . The enrichment in the third neuronal module ( Grey60 ) was not significant after correcting for multiple tests ( nominal P = 0 . 005 , Bonferroni corrected P = 0 . 095 ) . In addition to the neuronal modules , significant enrichment was found in the astrocyte-associated Magenta module ( P<0 . 00001 ) and the oligodendrocyte associated Blue module ( nominal P = 0 . 0008 , Bonferroni corrected P = 0 . 015 ) . We next examined the correlation between the degree of enrichment of rare and common variants for the different modules . Strikingly , the overall propensity to harbor genes with common variants enriched in autism , and the overall propensity to harbor genes with rare mutations linked to autism , were significantly correlated ( Pearson correlation r = 0 . 69 , P = 0 . 0010 ) ( Figure 3A , 3B ) . Specifically , two of the three modules representing neuronal genes ( Lightgreen and Salmon ) were significantly enriched for genes affected by both rare and common variations , with the highest overall evidence for association in the Salmon module . As can be seen in Figure 3C , the genes affected by common and rare variants within the Lightgreen and Salmon modules are highly interconnected . Differences in transcriptome organization between autistic and normal brain have been recently reported , including a neuronal module associated with ASD [20] . To study how the enrichment of rare and common variants corresponded to this study , we tested the overlap between the neuronal modules obtained in our study and the neuronal module that was previously shown to be differentially expressed between cases and controls [20] . Interestingly , the highest overlap was observed with the Grey60 module ( RR = 6 . 18 ) , followed by Lightgreen ( RR = 4 . 59 ) , but there was only relatively minor overlap with the Salmon module ( RR = 1 . 84 ) . To test the robustness of the enrichment of GWAS low p-values in specific modules we first applied the same analysis on GWAS data for type-2-diabetes [52] . The analysis with type-2-diabetes did not reveal any association with the modules . Next , we attempted to replicate the results in two additional GWAS of ASD . The first is a previously reported [53] GWAS from the Autism Genome Project ( AGP ) , which includes both multiplex and simplex families ( around 40% of families had two or more ASD children ) . The second is based on genotyping data of simplex families ( with a single child with ASD ) from the Simons Simplex Collection ( SSC ) . Inherited and de novo CNVs were previously reported for this sample [10] , but no genome-wide association for common variants was reported . We performed a genome-wide association using the transmission disequilibrium test ( TDT ) . To reduce the genetic heterogeneity , in both datasets we focused on families with European ancestry ( Figure S5A ) . Quantile-quantile ( Q-Q ) plots showed that there was minimal inflation of the test statistics ( genomic control inflation factor for AGP λGC = 1 . 0268 , for SSC λGC = 1 . 0013 ) ( Figure S5B ) . None of the SNPs in the SSC cohort were genome-wide significant ( P<5×10−8 ) . The 10 most significant SNPs in the SSC GWAS are shown in Table S6 . We also examined the 29 SNPs that were proposed as possible ASD risk variants by previous genome-wide studies [6] , [7] , [53] ( Table S7 ) . Of these , 22 were either available in our data or had a proxy SNP with an R2>0 . 8 . None of the 22 SNPs were associated in the SSC cohort ( all P>0 . 05 ) . Despite the limited results when testing single SNPs by association , the enrichment of low p-values in specific modules was replicated across different GWAS . The enrichment in the neuronal modules , Salmon and Lightgreen , was replicated both in the AGP ( Salmon , P = 0 . 012; Lightgreen , P = 0 . 000057 ) and in the SSC GWAS ( Salmon , P = 0 . 033; Lightgreen , P = 0 . 0026 ) . The combined p-value for low p-values enrichment , across the three studies , was 2 . 2×10−6 for the Salmon module , and 7 . 3×10−8 for the Lightgreen module . In addition , a replication of the enrichment of low p-values was obtained for the Blue and Magenta modules using the results of the AGP GWAS ( Blue , P = 0 . 014; Magenta , P = 0 . 0011 ) , but not with the SSC ( Blue , P = 0 . 062; Magenta , P = 0 . 43 ) . Based on the three genome-wide studies the most enriched module for common risk variants is the Lightgreen module , while the Salmon is the module most enriched for rare variants ( Figure 3A ) . However , the correlation between the enrichment of rare variants and common variants ( based on the three GWAS together ) remained significant ( r = 0 . 72 , P = 5×10−4 ) ( Figure 3B ) . To identify candidate genes central to the enrichment for common variants , we calculated a gene-wide P-value for association with ASD for all genes that contributed to the enrichment score in the three samples ( showing overrepresentation of low GWAS P-values in the modules ) . Eighty five genes passed a cutoff of 0 . 05 for gene-wide significance in one of the studies ( Table S8 ) . Out of these 85 genes , SNPs in four genes ( DMD , ATP2B2 , MACROD2 and MKL2 ) were previously found to be associated with ASD [53]–[56] . The replicated enrichment suggests that multiple common variants , particularly in sub-networks of neuronal genes , contribute collectively to ASD risk . This raised the possibility that common variants within the two neuronal modules may specifically predict ASD risk . If the observed enrichment is specific to ASD , one would expect that a score that incorporates the effect of multiple SNPs would be a significant predictor of ASD risk . To test this , we performed a genetic risk score analysis based on 79 , 079 tag SNPs ( as previously reported [57] ) . The AGRE dataset served as the discovery sample . We selected SNPs at different thresholds of association p-values ( PT ) , and based on whether they belong to the neuronal modules , Salmon or Lightgreen . Based on the GWAS results in the AGRE , we calculated a genetic score for each individual in the AGP or SSC samples and tested whether the score can predict ASD status . While a marginally significant correlation was observed between the score and diseases status with genome-wide data ( PT<0 . 3 , AGP , P = 0 . 029; SSC , P = 0 . 0085 ) , the score based on the neuronal modules was highly significant ( Figure 4 ) . The score based on SNPs in the Lightgreen module had increased association with ASD risk with more liberal thresholds in both AGP and SSC samples , with 0 . 66–0 . 5% ( respectively ) of the variance explained at the threshold of PT<0 . 5 ( AGP , P = 1 . 1×10−5; SSC , P = 0 . 0017 ) . Strikingly , a very different pattern was observed for the Salmon module: the strongest association , in both AGP and SSC , was with the strictest threshold of PT<0 . 1 ( AGP , P = 4 . 0×10−5; SSC , P = 0 . 0040 ) . The Salmon module , which is one of the enriched modules for ASD risk variants , includes genes that are known to be expressed in both the developing and the adult brain . This raises the question of whether this module represents pathways that are mainly involved in neuronal plasticity in the adult brain , or whether it represents genes that operate mainly in the developing brain . To address this question we examined gene expression profiles of brain samples from different developmental stages , using data from the BrainSpan database . For each of the neuronal modules we calculated the average expression for the 50 most connected genes across different brain areas , and plotted this as a function of developmental stages ( Figure 5 ) . In all three neuronal modules there was a relatively low expression during fetal brain development that increased with fetal age . In the Salmon module the highly connected genes showed the highest expression during infancy ( Figure 5A ) . In contrast , in the Grey60 module , which represents genes expressing in cortical interneurons , there was a continuous increase into adulthood ( Figure 5B ) . The most connected genes in the Lightgreen module had , on average , a relatively stable expression from childhood to adulthood ( Figure 5C ) . A flat temporal pattern was observed for the entire dataset of genes ( Figure 5D ) .
We constructed a gene co-expression network based on comprehensive expression profiling of the human brain . The network was based on the variation in expression between different brain regions . Similar to the findings of a previous study [19] , modules in the network corresponded to specific cell types . The vastness of the data allowed us to detail various cell-types , and even , in the case of neurons and oligodendrocytes , to identify modules corresponding to sub-populations of cells . Furthermore , functional annotation of the modules allowed us to characterize genes related to specific cellular processes and molecular functions in the brain , which in many cases ( but not all ) are also related to specific cell populations . These modules , and the hierarchy of the genes within them ( especially the “hub” genes ) , can be used to predict the function of yet uncharacterized genes and learn about new biological phenomena . An intriguing example is the observed coupling between two sub-networks within the Yellow module . One sub-network corresponds to genes involved in protein translation and the other to microglia function and regulation . This module's eigengene can be used to estimate the relative distribution of microglia in the brain . Another example is the identification of three separate neuronal modules , suggesting that the neurons in the brain could be roughly divided into three main types based on their gene expression profiles . One module corresponds to fast-spiking Parvalbumin-expressing interneurons . These interneurons have been shown to be of importance in the generation of gamma-oscillations [58] , which are required for speech perception and production [59] , [60] , consistent with the strong signal of the module eigengene in the temporal cortex . The observed increase in the expression of the genes in this module with age is consistent with previous reports in human and rats [61] , [62] . The second module is involved in sensory perception . Accordingly , it is highly expressed in the visual and somatosensory cortices and enriched with synaptic genes . The third module includes genes implicated in neuronal plasticity , and is highly expressed in brain areas responsible for learning and memory . Similarly , we identified two modules that are enriched for genes up-regulated in oligodendrocytes , the Blue and Yellowgreen modules . The Yellowgreen module was also found to be enriched for genes involved in mitosis and the cell cycle . We suggest that the Blue module may represent mature oligodendrocytes , whereas the Yellowgreen module might represent immature dividing cells . An important route in utilizing this network is as a framework to explore the functional aspects of genetic variations in brain related phenotypes . Because the network is based on measurements from control individuals alone , it can only shed light on diseases where specific aspects of brain functionality are involved . Our focus in this study was ASD , as this is a heterogeneous syndrome with a diverse genetic contribution . Although the genetic architecture of ASD is still under debate , we found enrichment of genes affected by both common and rare variants within specific neuronal modules . The enrichment of genes affected by common variants was replicated in two additional samples . Furthermore , we found a genetic risk score based on the two neuronal modules to be significantly associated with ASD status in the two target samples . The replication was evident despite the fact that the discovery sample consists mainly of multiplex families and the target samples of only simplex families or a mix of both . GWAS for ASD have had limited success so far; however , our study suggests a polygeneic component of ASD risk that is shared by multiplex and simplex families . This implies that a GWAS with larger samples should further contribute to the identification of ASD susceptibility genes . The effect of multiple common variants with very low effect size perturbs neuronal sub-networks , which are also affected by rare variants . With this in mind , it is tempting to speculate that both common and rare variants contribute to perturbations of the same neuronal pathways , which in turn lead to ASD . Unlike genome-wide studies of SNPs and CNVs that aim to identify specific genes associated with ASD , the approach used here seeks to identify sub-networks that have a causal relationship with ASD . Nevertheless , by integrating the network and the GWAS data , we were able to elucidate genes within the modules that are more likely to be responsible for the observed enrichment , and are thus likelier candidates for association with ASD , needing further validations . One of the sub-networks that are enriched for rare and common variants ( the Salmon module ) represents genes that are expressed in neurons , and are related to neuronal plasticity and neurogenesis . Accordingly , the expression of genes in this module was highest in the dentate gyrus , the CA1 field of the hippocampus , and the dorsal striatum . By examining expression levels during different developmental stages , we found that the highest expression of the most connected genes in the Salmon module was during infancy . The other associated module ( Lightgreen ) is enriched with synaptic genes , specifically genes involved in clathrin-dependent endocytosis , with the highest expression in cortical areas involved in sensory processes . While this could reflect the involvement of these regions with ASD etiology , it is important to note that our findings could reflect the distribution of specific cell types in the adult brain , and not necessarily the brain areas affected in ASD . The results of this study are in line with previous findings that connected rare mutations in autism with neuronal activity-dependent genes [63] . The hypothesis is that these genes are highly expressed during critical periods of infancy and early childhood as they are influenced by neural activity , which is dependent on inputs from the environment [64] . Perturbation by common and rare variants in these genes and pathways that are involved in learning and memory of social cues during postnatal stages may increase the risk of developing ASD . The potential involvement of postnatal neuronal plasticity in ASD gives hope that these pathways may be amenable to treatment long after symptom onset , as has been suggested by animal studies on various neurodevelopmental syndromes [65] , [66] . In summary , we constructed a gene network based on comprehensive expression profiling of the human brain . This network can be used as a framework to study multiple questions including ones related to disease mechanisms , but also to normal functions of genes in the brain . It could also be integrated with other functional assays of the brain or other datasets . In our current study we focused on ASD as a case study . We integrated the gene co-expression network with genetic variations associated with ASD . The results support the notion that common and rare variants contribute to ASD by perturbation of common neuronal networks . Further integration of genetic and molecular data with the network has the potential to reveal a more detailed picture of the particular molecular features depicted in the network that contribute to ASD . Such knowledge is essential , first for providing insights into the molecular functionality related to the etiology of ASD , and also for the development of diagnostic tools and effective therapies .
Microarray data were acquired from the Allen Brain Atlas ( http://human . brain-map . org/well_data_files ) , and included a total of 1340 microarray profiles from donors H0351 . 2001 and H0351 . 2002 , encompassing the different regions of the human brain . Donors were 24 and 39 years old , respectively , with no known psychopathologies . For donor H0351 . 2001 a total of 921 microarray profiles were available , and for H0351 . 2002 a total of 419 microarray profiles were available . A detailed description of the donor individuals , including available medical profile and post-mortem analyses performed is available at the following link:http://help . brain-map . org/download/attachments/2818165/CaseQual_and_DonorProfiles_WhitePaper . pdf . A detailed description of the regions measured by microarray in each donor is available in Table S9 . Statistical analysis was done using the R project for statistical computing ( http://www . r-project . org ) . Network construction deployed the WGCNA R-package [67] , and followed closely the tutorials available on the authors' website . First , the correlation between both individuals was tested by correlating first the mean rank of the expression values in each gene , and then by correlating the mean connectivity values in each gene [68] . For genes with at least 3 available probes , the connectivity for each of the probes was calculated , and the probe with the highest connectivity was chosen for the network analysis . For genes with 2 probes , the one with the highest mean was chosen . Probes not corresponding to refSeq genes were removed , leaving a total of 16 , 298 probes used in the network . The network was assembled following previously published parameters [69] . An adjacency matrix was calculated by raising the correlation matrix by a power of 6 ( determined to be optimal for scale free topology in our dataset ) , and a TOM matrix was generated [20] . To determine the modules , hierarchical clustering was performed , and the tree was cut using the cutreeHybrid function in the WGCNA R package , with the minimum module size set to 30 genes , and parameter deepSplit set to 2 [67] . The resultant modules were merged using the mergeCloseModules function with cutHeight set to 0 . 3 . The module eigengenes were derived by taking the 1st principal component in a PCA analysis for the expression values in each module . To visualize the modules , the 150 strongest connections were drawn in the Cytoscape software [23] . For presentation purposes , the nodes were ordered based on their degree of connectivity , and their number was restricted to 50 nodes in each module . The enrichment analysis was based on a dataset of genes enriched in mouse neurons , oligodendrocytes and astrocytes [21] . First , the number cell-type enriched genes in each module ( ) was calculated , as well as the total number of cell-type enriched genes ( ) appearing in the entire network . Subsequently , a Relative Risk ( RR ) measure was calculated for each module and for each cell type , , with as the number of genes in each module , and the total number of genes in the network . P-values were obtained by permutation testing , whereby a module of the same size was randomly selected and the RR calculated . Standard error was calculated for each module using bootstrap analysis . To determine whether the observed overall enrichment was specific to the more up-regulated genes in the cell types , the distribution of rank fold-change for the cell-type enriched genes in each module was plotted . The correlation between the median of the bins in the histogram and the number of genes in the bins was tested . A strong negative correlation indicates a substantial enrichment of the higher ranked cell type specific genes . Lists of the genes in each module were tested with the DAVID bioinformatics tool [70] . For background , the complete list of the genes in the network was used . For the module red , due to its size , the 3000 genes with the highest correlation with the module eigengene ( see above ) and the 3000 genes with the lowest correlation ( most negative ) with the module eigengene were used separately for enrichment testing . A list of autism susceptibility genes was compiled using the SFARI gene database ( https://sfari . org/sfari-gene ) , downloaded on the 23/6/2011 . The list was restricted to genes with reported rare mutations in autism . Fisher's exact test was used to test the distribution of ASD genes within the modules with 10 , 000 permutations . The number of risk genes with rare variants ( from the SFARI Gene ) in each module ( ) , and the total number of risk genes in the network ( ) , were used to calculate the RR , similar to the method described above: . For the CNV analysis , gene length was used instead of gene count , to correct for biases arising from differences in gene lengths between the modules . For each module , the total length in base pairs covered by CNV ( ) and the total length of the module ( ) , were used along with the total length covered by CNV ( ) and the total length of the genes in the network ( ) , in the following formula: . Testing for the enrichment of low GWAS P-values was performed using the discovery cohort of a previously published GWAS [6] , which included 943 ASDs families . The analysis incorporated a previously published method [71] , in a manner previously described [20] . Briefly , the minimum P-value for each gene was used in an enrichment score similar to the Kolmogorov-Smirnov statistic [71] . Gene boundaries included the 20 kb upstream and 10 kb downstream of each gene . To arrive at a P-value corrected for the size of the genes , the gene labels were permuted . Permutations were run until either reaching 20 instances of the higher enrichment score , or 100 , 000 permutations . For each of the three neuronal modules found to be enriched for rare and common variations in ASD , a list of genes that contributed positively to the enrichment score of the module was obtained . As the enrichment for low GWAS p-values was tested using a running sum statistic over a sorted gene list , all genes above the point where the statistic reached the maximum were taken . Gene-wide P-values were determined by taking the SNP with the minimum P-value in each gene and correcting for the number of SNPs in the gene using a Bonferroni correction . All GWAS analyses were performed using the PLINK software by Shaun Purcell [72] . SNP genotyping data was acquired from the Simons Simplex Collection ( SSC ) and the Autism Genome Project ( AGP ) . The SSC cohort included 734 nuclear families with an autistic proband and an unaffected sibling , along with two parents , genotyped using the Illumina 1M platform . The AGP cohort included 1369 nuclear families with an autistic proband and two parents , genotyped using the Illumina 1M platform . To determine divergent ancestry , each sample separately was combined with data from The HapMap Phase III , following a previously published procedure [73] . Multidimensional scaling analysis to four dimensions was then performed in PLINK , followed by clustering to four groups using the R Package Mclust [74] . After removing individuals who did not cluster with the Hapmap CEU cohort , 588 families remained in the SSC cohort , and 1165 families remained in the AGP cohort . On these , TDT was performed , limiting the analysis to SNPs with a minor allele frequency of over 10% , in Hardy-Weinberg Equilibrium ( P>0 . 001 in an exact test ) , with more than 90% genotyping rate , and with less than 10% rate of mendelian errors . This left 788010 SNPs in the SSC and 668221 in the AGP . Families with 5% mendelian errors were set to be removed , but none crossed that threshold . Q-Q plots were generated by plotting the observed −log10P against the expected distribution , and visualized using a function available online . ( http://gettinggeneticsdone . blogspot . com/2011/04/annotated-manhattan-plots-and-qq-plots . html ) . To estimate the contribution of common variation to autism , we followed a previously published paradigm [57] . First , a list of tag SNPs was compiled wherein no two SNPs had an r2>0 . 25 in a combined SSC and AGP sample . For these SNPs , the z-score of the reference allele for association in the AGRE cohort was used to calculate a score in Plink for each individual , which was defined as the sum across all SNPs of the number of reference allele multiplied by the z-score . The predictive value of the score was tested by fitting a logistic regression model with ASD status as the explained variable and individual score as the predictor , and calculating both a Wald's test p-value and a Nagelkerke's pseudo-r2 . To test the Salmon and Lightgreen modules , the list of tag SNPs was further pruned for SNPs in genes in these modules , and the same analysis was performed . Gene expression microarray profiles of the brain from individuals of different ages were retrieved from the BrainSpan database ( http://developinghumanbrain . org/ ) . The data included 492 microarray measurements from a total of 35 individuals of 28 different ages , ranging from 8 weeks post-conception to 40 years of age ( full sample information is available on the BrainSpan website ) . We first accounted for global differences between the different array samples and between the different genes . For each measurement ( a ) of a gene ( i ) in each array ( j ) , the following compound z-score was calculated: . As several array measurements from different brain regions existed for each age , the mean normalized score was used in the final analysis . For each module tested , the mean score of the 50 genes with the most connections out of the top 150 connections was plotted . A smoothed signal was calculated using the cubic smoothing spline algorithm implemented in the R function smooth . spline , using default parameters . | Autism spectrum disorders ( ASD ) are neurodevelopmental syndromes with a strong genetic basis , but are influenced by many different genes . Recent studies have identified multiple genetic risk factors , including rare mutations and genetic variations common in the population . To identify possible connections between different genetic risk factors , we constructed a network based on the expression pattern of genes across different brain areas . We identified groups of genes that are expressed in a similar pattern across the brain , suggesting that they are involved in the same processes or types of cells . We found that the genetic risk factors were enriched in specific groups of connected genes . Of these , the strongest enrichment was discovered in a group of neuronal genes that are involved in processes of learning and memory , and are highly expressed during infancy . Further study of this group of genes has the potential to reveal a more detailed picture of the neuronal mechanisms leading to ASD and to provide knowledge required for developing diagnostic tools and effective therapies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"systems",
"biology",
"medicine",
"mental",
"health",
"genetics",
"biology",
"genomics",
"neuroscience",
"genetics",
"and",
"genomics"
]
| 2012 | Networks of Neuronal Genes Affected by Common and Rare Variants in Autism Spectrum Disorders |
The stabilization of the replisome complex is essential in order to achieve highly processive DNA replication and preserve genomic integrity . Conversely , it would also be advantageous for the cell to abrogate replisome functions to prevent inappropriate replication when fork progression is adversely perturbed . However , such mechanisms remain elusive . Here we report that replicative DNA polymerases and helicases , the major components of the replisome , are degraded in concert in the absence of Swi1 , a subunit of the replication fork protection complex . In sharp contrast , ORC and PCNA , which are also required for DNA replication , were stably maintained . We demonstrate that this degradation of DNA polymerases and helicases is dependent on the ubiquitin-proteasome system , in which the SCFPof3 ubiquitin ligase is involved . Consistently , we show that Pof3 interacts with DNA polymerase ε . Remarkably , forced accumulation of replisome components leads to abnormal DNA replication and mitotic catastrophes in the absence of Swi1 . Swi1 is known to prevent fork collapse at natural replication block sites throughout the genome . Therefore , our results suggest that the cell elicits a program to degrade replisomes upon replication stress in the absence of Swi1 . We also suggest that this program prevents inappropriate duplication of the genome , which in turn contributes to the preservation of genomic integrity .
Initiation of DNA replication is directed by the formation of the pre-replication complex ( pre-RC ) at the origin of replication [1] . The pre-RC includes a number of essential replication proteins such as origin recognition complex ( ORC ) , Cdc6 , Cdt1 , and the mini-chromosome maintenance ( MCM ) DNA helicase complex . However , to initiate actual DNA synthesis , additional factors are needed to facilitate the unwinding of origins and generation of replication forks . These factors include Cdc45 , go-ichi-ni-san ( GINS ) , replication protein A ( RPA ) , proliferating cell nuclear antigen ( PCNA ) , and other accessory factors prior to the loading of DNA polymerases . Together , these factors form the replisome complex at the replication fork [1] . However , how the cell maintains the integrity of the replisome is not well understood . In response to replication stress , cells activate the DNA replication checkpoint to allow time for DNA repair . Central to this system are protein kinases such as human ATM and ATR , fission yeast Rad3 , and budding yeast Mec1 [2]–[6] . These kinases are required for activation of downstream effector kinases by phosphorylation . In the fission yeast Schizosaccharomyces pombe , Rad3 activates Cds1 and Chk1 kinases in response to replication stress or DNA damage , facilitating DNA repair and recombination pathways [2] , [4] , [7] . Another essential function of the replication checkpoint is to stabilize replication forks by maintaining proper assembly of replisome components and preserving DNA structures during DNA replication problems [8]–[12] . Recent studies found that ancillary factors that are not essential for DNA synthesis but are important for DNA replication accuracy also travel with moving replication forks . Such factors include fission yeast Swi1 and Swi3 , which together form the replication fork protection complex ( FPC ) and are required for efficient activation of the replication checkpoint kinase Cds1 and for the stabilization of replication forks [13]–[16] . In the absence of Swi1 or Swi3 , cells accumulate abnormal fork structures that lead to Rad22 ( the Rad52 orthologue ) DNA repair foci formation and accumulation of recombination structures during S phase [16] , [17] . The functions of the Swi1–Swi3 complex appear to be conserved among eukaryotes [14] , [15] , [18]–[20] . Studies show that Swi1–Swi3 orthologues ( Tof1–Csm3 in budding yeast , and Timeless–Tipin in vertebrates ) are components of the replisome , are involved in fork stabilization , and regulate the intra-S phase checkpoint [18] , [21]–[28] . Furthermore , genetic studies in yeast also suggest that the Swi1–Swi3 FPC has roles in coordinating leading- and lagging-strand DNA synthesis and in coupling DNA polymerase and helicase activities at the replication fork [15] , [16] , [29] . However , how the FPC protects moving replication forks and coordinates with multiple genome maintenance processes at the replication fork is not well understood . Replication checkpoint studies have typically used chemical agents to stall replication forks . However , emerging evidence indicates that there are a number of chromosome regions that present obstacles to DNA replication . These include programmed fork blocking sites , DNA-binding proteins such as the transcription machinery , and DNA secondary structures caused by repeat sequences . These sites are considered to be difficult to replicate , causing arrest of replication forks or even fork breakage [30]–[35] . Fork arrest at difficult-to-replicate genome sites can promote both genome instability and stability depending on the circumstances . For example , polar fork pausing at rDNA loci stimulates recombination-dependent rDNA repeat expansion and contraction , which can lead to rDNA instability . On the contrary , this polar fork pausing is also required to coordinate directionality of replication and transcription at rDNA loci , preventing genome instability due to head-on collisions of the replisome and transcriptional machinery [36] , [37] . Interestingly , studies found that FPC-related proteins are required for a number of fork arrest events , which are mediated by DNA–protein complexes . These include fork pausing at the rDNA loci , the fission yeast mating-type locus , tRNA loci , and highly transcribed RNA polymerase II genes [14] , [29] , [38]–[41] . At rDNA loci , loss of FPC causes hyper recombination , leading to contraction of rDNA repeats [17] , [40] , [42] . Similarly , the high rate of transcription and the presence of DNA-binding factors increase the chances of the replisome colliding with a transcription fork . Indeed , studies in fission yeast revealed that Swi1 is required to prevent DNA damage and hyper recombination activity at these natural obstacles scattered throughout the genome [43]–[45] . In addition to these DNA–protein complex-mediated fork barriers , repeat DNA sequences themselves also cause genome instability in the absence of FPC-related proteins . At these sites , instead of promoting fork stalling , FPC appears to prevent or reduce the rate of fork stalling when the fork encounters DNA secondary structures caused by repeat sequences . Therefore , in the absence of FPC , fork stalling results in elevated levels of ssDNA exposed at the replication fork , which appear to cause genome instability due to expansion and contraction at DNA structure-based impediments [46]–[51] . Thus , the mechanisms of the FPC-dependent fork regulation at repeat regions and at DNA–protein complex-mediated fork barriers are different . However , accumulated evidence indicates that FPC proteins are required for smooth passage of replication forks and for suppression of replication stresses at these natural impediments [14] . Therefore , in this study , we used swi1Δ as a model to understand replication stress response mechanisms . Strikingly , we have found that replicative DNA polymerases and helicases are highly unstable in the absence of Swi1 . Our investigation revealed that this degradation is mediated by the ubiquitin-proteasome system , in which the SCFPof3 ( Skp1/Cul1/F-box ) ubiquitin ligase complex is involved . In the absence of Pof3 , swi1Δ cells undergo mitotic catastrophes , suggesting the importance of proteasome-dependent replisome regulation in preserving genomic integrity . Considering that swi1Δ cells accumulate replication stress at difficult-to-replicate regions throughout the genome , our findings suggest that ubiquitin-dependent degradation of replisome components play a critical role in genome duplication in response to replication stresses . It is widely understood that checkpoint proteins stabilize replication forks and replisomes in response to replication stress . However , our findings suggest an alternative mechanism that cells abrogate replisome functions when the fork encounters obstacles . Therefore , our study proves new mechanistic insights into the understanding of the replication stress response . In addition , although a number of studies have focused on the processes of replication initiation and regulation of fork progression , how the replisome itself is regulated is still largely unknown . Therefore , our findings also fill the knowledge gap in the regulation of replisome components in the DNA replication program .
Recent studies have shown that fork progression is impaired in the absence of FPC orthologues [24] , [27] , [28] , [38] , [52] . We also found a similar defect in S . pombe swi1Δ cells ( Figure S1 ) , suggesting that FPC might regulate replisome stability . To test this possibility , we investigated the stability of various replication proteins in cells treated with cycloheximide ( CHX ) , a compound that blocks the synthesis of new proteins and allows for the examination of protein stability . First , we examined the stability of the catalytic subunits of major essential replicative DNA polymerases . For this purpose , we employed S . pombe cells expressing Pol2-FLAG ( the catalytic subunit of DNA polymerase ε , required for leading-strand synthesis ) [53] and Pol3-FLAG ( the catalytic subunit of DNA polymerase δ , required for lagging-strand synthesis ) [54] from their genomic loci . Pol2-FLAG showed significant degradation in wild-type cells , whereas , Pol3-FLAG was relatively stable ( Figure 1A and 1B ) . Intriguingly , Pol2 displayed even faster degradation when swi1 was deleted . In addition , Pol3 showed dramatic instability in swi1Δ cells ( Figure 1A and 1B ) . Next , we examined the stability of MCM helicase components . S . pombe cells expressing Mcm2-GFP or Mcm6-GFP from their genomic loci were used , and Mcm4 was detected by the anti-Mcm4 antibody . In wild-type cells , Mcm2-GFP , Mcm4 , and Mcm6-GFP were stable and did not undergo significant degradation throughout the CHX treatment ( Figure 1C and 1D ) . In contrast , these helicase subunits were rapidly degraded in swi1Δ cells ( Figure 1C and 1D ) . To determine whether such degradation is specific to certain replication proteins , we also assessed the stability of Orc1 ( an ORC subunit ) , Mrc1 ( a mediator of S-phase checkpoints ) , and PCNA . Although the steady-state levels of Orc1-FLAG and PCNA before the addition of CHX were somewhat lower in swi1Δ cells , their cellular amounts were maintained throughout the 4 h of CHX treatment in both wild-type and swi1Δ cells ( Figure 1C ) . As previously reported [55] , Mrc1 was unstable and shows rapid degradation in the presence of CHX , although this degradation was not strengthened by the deletion of swi1 ( Figure 1C and 1D ) . Thus , we concluded that Swi1 is involved in preventing rapid degradation of Pol2 and Pol3 , as well as helicase components . Since Swi1 is involved in the suppression of fork collapse at difficult-to-replicate regions in fission yeast [43]–[45] , it is possible that chromatin-bound replisome components are susceptible to degradation . Therefore , we fractionated cells into Triton-X-100-soluble ( cytosol and nucleoplasm ) and Triton-X-100-insoluble ( enriched with chromatin- and nuclear matrix-bound proteins ) fractions as described in previous studies ( Figure 1E ) [55] , [56] . Tubulin and histone H3 were exclusively fractionated into the Triton-soluble and Triton-insoluble fractions , respectively , indicating that fractionation was successful . Pol2 was mainly fractionated into the Triton-insoluble fraction , while approximately 20% and 30% of Pol3 and Mcm4 were recovered into the Triton-insoluble fraction , respectively . Importantly , degradation of Pol2 , Pol3 and Mcm4 was observed in the Triton-insoluble fraction when cells were treated with CHX , suggesting that the chromatin fraction of replisome components undergoes degradation ( Figure 1E ) . Therefore , our results are consistent with the notion that cells promote a fast turnover of replisome components bound to chromatin in response to the accumulation of fork collapse . To understand the mechanisms of replisome degradation in response to unstable forks in the absence of Swi1 , we determined whether the proteasome is responsible for degradation of DNA helicases and polymerases . The mts3-1 temperature-sensitive allele , which has a mutation in a subunit of the 26S proteasome machinery , was used to inactivate the proteasome [57] , [58] . It is estimated that proteasome activity of mts3-1 cells is about 50% and 30% of the wild-type enzyme at 25°C and 35°C , respectively [58] . Cells were grown at 25°C or 35°C for 2 h , and then treated with CHX for 2 to 4 h . Strikingly , degradation of Pol2 was substantially inhibited in mts3-1 and swi1Δ mts3-1 cells even at 25°C ( Figure 2A and 2C ) . We observed similar stabilization of Pol3 and Mcm6 in mts3-1 and swi1Δ mts3-1 cells ( Figure 2B and 2C ) . At 35°C , degradation of these replisome components was accelerated both in wild type and swi1Δ cells probably due to increased cell metabolism ( Figure 2A and 2B ) . However , degradation of these replisome components was abolished in mts3-1 and swi1Δ mts3-1 cells at 35°C ( Figure 2A and 2B ) . Thus , our data indicate that Swi1 prevents proteasome-dependent degradation of replisome components . Ubiquitin moieties ( Ub ) are conjugated to most of the proteins degraded by the proteasome [59] , . Therefore , aforementioned data suggest that replisome core components ( polymerases and helicases ) are ubiquitinated . To test this possibility and further understand the mechanism of replisome degradation , we investigated whether replisome components were ubiquitinated . Cells harboring FLAG-tagged versions of Pol2 and Pol3 were engineered to express hexahistidine-fused ubiquitin ( 6xHis–Ub peptide ) under the control of the thiamine ( B1 ) -repressible nmt1 promoter . They were first cultured in the presence of thiamin ( B1 ) to repress the nmt1 promoter and then grown in the absence of thiamine for 22 h at 25°C , allowing cells to express 6xHis–Ub peptide . After the 22 h incubation , cultures were divided and further incubated at 25°C or 35°C for 2 h . Ubiquitinated proteins were purified with nickel agarose beads and analyzed by immunoblotting using antibodies against the FLAG-tag ( Figure 2B ) . As shown in Figure 2D , Pol2-FLAG species with slower gel mobility were clearly detected in both wild type and swi1Δ cells , indicating that Pol2 is ubiquitinated . We also observed precipitation of non-ubiquitinated Pol2 with nickel agarose as previously reported for other proteins [61] . In addition , multiple Pol2 bands , which are probably products of degraded Pol2 , were detected in swi1Δ cells ( Figure 2D ) , suggesting that Pol2 is more susceptible to degradation in the absence of Swi1 . Similarly , ubiquitinated forms of Pol3-FLAG were detected in wild type and swi1Δ cells ( Figure 2D ) . However , with our methods , we were not able to observe ubiquitinated forms of Mcm proteins ( data not shown ) . Considering that Mcm proteins are stabilized in mts3-1 cells ( Figure 2A ) , it is possible that the ubiquitination and degradation processes of Mcm proteins are too rapid to be detected . Swi1 and its orthologues are involved in DNA replication , and their defects cause replication stress at difficult-to-replicate genome regions [14] , [15] , [18] , [21] , [22] , [24] , [25] , [27] , [28] , [43]–[45] . Thus , our results suggest that replisome core degradation occurs during S phase in the absence of Swi1 . To test this possibility , wild type and swi1Δ cells were synchronized at the G1/S boundary in the presence of 12 mM hydroxyurea ( HU ) and released into S phase in fresh medium supplemented with CHX . FACS analysis showed that the addition of CHX did not perturb cell cycle progression through S phase after the removal of HU ( Figure 3A ) . There was no significant Pol2 degradation in both wild type and swi1Δ cells in the absence of CHX . In contrast , the level of Pol2-FLAG dramatically dropped between 30 and 45 min after CHX addition in the absence of Swi1 ( Figure 3B and 3C ) , when cells are in S phase ( Figure 3A ) . In contrast , wild-type cells displayed only a mild decrease in the level of Pol2-FLAG ( Figure 3B and 3C ) . We also used the cdc25-22 temperature sensitive allele to synchronize cells at the G2/M boundary at the restrictive temperature ( 35°C ) , and cells were released into the cell cycle at permissive temperature ( 25°C ) . As determined by the increase of septation index , cells synchronously entered S phase after the release in the absence of CHX ( Figure S2A ) . In this condition , Pol2 levels were maintained throughout the experiments in both cdc25-22 and cdc25-22 swi1Δ cells ( Figure S2B and S2C ) . When cells were treated with CHX , Pol2-FLAG levels gradually decreased in cdc25-22 swi1Δ cells but not in cdc25-22 cells ( Figure S2B and S2C ) . This mild degradation is probably because cells were unable to synchronously progress through S phase in the presence of CHX ( Figure S2A ) , although our data indicate that Pol2-FLAG is unstable in swi1Δ cells . Interestingly , Mcm4 showed rapid degradation as cdc25-22 swi1Δ cells progress through S phase in the absence of CHX ( Figure S2B and S2C ) , indicating that Mcm4 is degraded during replication . Mcm4 degradation in cdc25-22 swi1Δ cells was further exacerbated in the presence of CHX . In contrast , there is no significant Mcm4 degradation in cdc25-22 cells with or without CHX treatment ( Figure S2B and S2C ) . Taken together , we concluded that degradation of replisome core components occurs during DNA replication in the absence of Swi1 . Since SCF ubiquitin ligases are often involved in protein degradation during S phase [62] , we examined the stability of Pol2 in skp1-94 temperature-sensitive cells , which have a mutation in Skp1 , a major component of SCF ubiquitin ligases in S . pombe [63] . Strikingly , Pol2-FLAG was significantly stabilized when skp1-94 cells were incubated at 35°C , indicating the involvement of SCF ubiquitin ligases in Pol2 degradation ( Figure 3D; Figure S3B ) . SCF ligases contain F-box subunits , which are responsible for substrate specificity . Therefore , we examined Pol2 stability in a series of mutants defective for F-box proteins ( Figure S3 ) . Among the eleven F-box mutants we tested , we found that Pol2 becomes most stable in the absence of Pof3 ( Figure 3E; Figure S3 ) , an F-box protein that has been suggested to be involved in the preservation of genomic integrity [64] . Thus , our data suggest that Pol2 degradation is in part mediated by the SCFPof3 ubiquitin ligase . To further understand the mechanism of Pol2 degradation , we examined whether Pof3 interacts with Pol2 , using immunoprecipitation assays . Cells expressing Pol2-FLAG proteins were engineered to express Pof3-Myc from its genomic locus . As shown in Figure 3F , Pol2-FLAG coprecipitated with Pof3-Myc in the absence of Swi1 , indicating that SCFPof3 interacts with Pol2 . The Pol2–Pof3 interaction was not detectable in wild-type cells even in the presence of a protein crosslinker dithio-bis succinimidyl propionate ( DSP ) ( Figure 3F ) . Therefore , our data suggest that SCFPof3–Pol2 interaction is transient in wild-type cells but is enhanced when Pol2 degradation is accelerated in the absence of Swi1 . SCFPof3 has been shown to interact with fission yeast Mcl1 , a DNA polymerase α accessory factor related to budding yeast Ctf4 [64]– . Moreover , in budding yeast , Dia2 ( Pof3-related protein ) is recruited to the replication fork [67] , [68] and is involved in the ubiquitination of Mrc1 [69] . Therefore , SCFPof3-dependent Pol2 degradation suggests that SCFPof3 may also target other replisome components for degradation . We first sought to determine whether Pof3 is also involved in degradation of Mrc1 in S . pombe . As shown in Figure 4 , Mrc1 became highly stable in pof3Δ cells under CHX treatment ( Figure 4A and 4C ) . We then examined whether Mcm4 degradation in swi1Δ cells is inhibited by the inactivation of SCFPof3 ( Figure 4B and 4D ) . Intriguingly , Mcm4 was significantly more stable in pof3Δ swi1Δ cells than in swi1Δ cells after CHX treatment . This result suggests that SCFPof3 also targets Mcm4 for proteasome-dependent degradation in response to replication stress provoked by swi1 deletion . Taken together , our results are consistent with the notion that SCFPof3 is involved in degradation of multiple replisome components . In order to understand the physiological importance of replisome core degradation in the absence of Swi1 , we investigated cellular phenotypes of swi1Δ , pof3Δ and swi1Δ pof3Δ double mutant cells . For this purpose , cells were stained with DAPI to visualize nuclear DNA . As shown in Figure 5A and 5B , swi1Δ and pof3Δ cells displayed an increased level of mitotic catastrophes ( including chromosome missegregation , aneuploidy , fragmented nuclei , hypercondensed nuclei , “cut” and other aberrant phenotypes , which are shown by arrows ) compared to wild-type cells . Importantly , this phenotype was further exacerbated in swi1Δ pof3Δ cells even in the absence of exogenous genotoxic agents ( Figure 5A and 5B ) . We then used HU and camptothecin ( CPT ) to introduce S phase specific genotoxic stress . HU depletes the dNTP pool and causes an arrest of replication fork progression , while CPT traps topoisomerase I on DNA and induces replication fork breakage . HU or CPT treatment further enhanced the aberrant mitotic phenotypes ( Figure 5A and 5B ) . Consistently , swi1Δ pof3Δ cells were more sensitive to HU and CPT than either single mutant ( Figure 5C ) . In the presence of HU or CPT , swi1Δ cells accumulate DNA damage due to failure in the completion of DNA replication , which causes activation of the DNA damage checkpoint , leading to abnormal cell cycle arrest and a cell elongation phenotype [17] , [70] . As expected , HU or CPT treatment caused cell elongation in swi1Δ cells ( Figure 5A ) . However , this elongation phenotype was abolished in swi1Δ pof3Δ cells ( Figure 5A ) , suggesting that the stabilization of replisome components attenuated cell cycle arrest in swi1Δ pof3Δ cells . This attenuation of cell cycle arrest could have caused a growth advantage , leading to the rather weak increase in the HU and CPT sensitivity of swi1Δ pof3Δ cells ( Figure 5C ) , although these cells show strong mitotic catastrophes ( Figure 5A ) . Next , we examined the ability of cells to recover DNA replication after CPT-dependent replication fork breakage . Exponentially growing cells ( Log ) were exposed to a low dose of CPT ( 5 µM ) for 3 h and returned to fresh medium for 2 and 4 h ( Figure 5D ) . Chromosome samples were then analyzed by pulsed-field gel electrophoresis ( PFGE ) , which permits only fully replicated chromosomes to migrate into the gel . In contrast , chromosomes with replication intermediates stay in the well of the gel , allowing us to determine the rate of replication recovery . There was no detectable DNA replication defect in wild-type cells throughout the experiment ( Figure 5D , Log , CPT , 2 h ) , indicating that the low dose of CPT used in this assay did not cause major replication problems in wild-type cells . Although chromosomes from swi1Δ cells migrated into the gel immediately after the CPT exposure ( CPT ) , we observed a marked reduction in chromosome intensity at 2 h after CPT treatment ( Figure 5D ) . This result indicates that the low dose of CPT caused replication problems in swi1Δ cells , which is consistent with previous studies [70] . However , there was a significant recovery at 4 h after the CPT removal due to the completion of DNA synthesis . A similar replication recovery was also observed in pof3Δ cells . In contrast , there was no DNA replication recovery in swi1Δ pof3Δ cells during the course of the experiment ( Figure 5D ) , indicating that these cells experience further difficulties in replication and/or repair of broken replication forks when treated with a replication-stressing agent . Interestingly , we repeatedly observed much less appearance of chromosomes I and II in the gel for swi1Δ pof3Δ cells ( Figure 5D; Figure S4 ) , suggesting that these cells experience major problems in DNA replication and chromosome maintenance . Consistently , there was an increased level of mitotic catastrophes in these cells ( Figure 5A and 5B ) . Considering that pof3 deletion stabilizes replisome components ( Figure 3E; Figure 4 ) , our results suggest that programmed replisome degradation represents a mechanism to prevent catastrophic DNA replication in response to replication stress caused by swi1 deletion ( Figure 6 ) . Similar replication and mitotic phenotypes were observed in swi1Δ mts3-1 cells , which are defective in proteasome functions ( Figure S5 ) , strengthening the idea that replisome degradation plays a critical role in the maintenance of genomic integrity .
Swi1 and its orthologues are known to be involved in the stabilization of replication forks to prevent genetic instability during DNA replication . Genetic analyses have suggested that FPC is involved in coordinating leading- and lagging-strand DNA synthesis . It is also suggested that the FPC couples polymerase and helicase activities at stalled forks [14] , [15] . Thus , the functions of Swi1 would become even more important to maintain the integrity of the replication fork when it encounters difficult-to-replicate sites or programmed fork pausing sites that are scattered throughout the genome . Consistently , FPC plays a critical role in programmed fork pausing and replication termination events near the mating-type ( mat1 ) locus and at fork pausing sites in rDNA repeats and tRNA loci in yeast [16] , [17] , [21] , [29] , [38] , [39] . Importantly , recent studies indicated that swi1Δ cells experience fork collapse at these difficult-to-replicate regions [43]–[45] . Therefore , inactivation of Swi1 causes defects in replication fork stabilization at natural impediments , leading to general replication stress at the replication fork . It is well known that Cdt1 and Cdc6 undergo rapid proteasome-dependent degradation to restrict replication licensing once per cell cycle [71] , [72] . However , how replisome degradation contributes to DNA replication process is largely unknown . In this report , we show that DNA polymerases and helicases undergo rapid degradation upon replication stress in the absence of Swi1 ( Figure 1 ) . This degradation is dependent on the ubiquitin-proteasome system ( Figure 2 ) . In the absence of Swi1 , cells experience unstable replication forks that lead to an increased level of replication-dependent DNA damage and hyper-recombination [16] , [17] , [43] . Such replication stress appears to cause replisome degradation in order to prevent abnormal DNA replication and mitotic catastrophes ( Figure 5; Figure S5 ) . These results suggest that replisome degradation functions to maintain genomic integrity during DNA replication in response to replication stress ( Figure 6 ) . Similar mechanisms have been described in the transcription-coupled DNA repair ( TCR ) , which is activated by transcription blockage in response to genotoxic agents [73] , [74] . In this mechanism , the Cockayne syndrome B protein ( budding yeast Rad26 ) interacts with Def1 to regulate ubiquitination of Rpb1 , the large subunit of RNA polymerase II ( RNAPII ) , which results in proteasome-dependent degradation of RNAPII [75] , [76] . Ubiquitination of Rbp1 is achieved by the Rsp5/Nedd4 ubiquitin ligase , which promotes DNA-damage induced degradation of RNAPII in budding yeast and human cells [77]–[79] . RNAPII degradation appears to be an alternative mechanism to TCR . Studies indicate that the loss of both TCR and RNAPII degradation pathways renders cells hypersensitive to DNA damage , thus Def1 promotes proteolysis of RNAPII when the lesion cannot be rapidly repaired by TCR [75] , [80]–[82] . Therefore , analogous to the DNA damage-induced RNAPII degradation pathway , our present findings suggest that the cell elicits a replisome degradation program when the replication fork is adversely blocked . We speculate that , depending on the degree of replication problems , re-building and re-loading new replisomes might be advantageous to the cell , rather than re-using existing replisome components that are compromised . Therefore , we suggest that replisome degradation is an alternative mechanism to replisome stabilization and prevents DNA synthesis by compromised replisomes . We also found that Pol2 ( Polε ) is significantly unstable even in wild-type cells ( Figure 1A and 1B ) , while Pol3 ( Polδ ) is relatively stable ( Figure 1A and 1B ) . The high rate of Pol2 turnover may suggest that Pol2 needs to be re-loaded during leading-strand synthesis . Since Pol2 is suggested to work continuously on the leading-strand [53] , one might think that the high turnover of Pol2 poses a disadvantage to the cells . However , it is possible that the polymerases fall off the chromatin every time the fork arrives at programmed pausing sites or difficult-to-replicate regions . In addition , Pol2 may undergo degradation once it falls off the chromatin . Such a degradation mechanism would also be advantageous for the cell to refresh Pol2 enzymes by efficiently reloading newly synthesized Pol2 at the moving replication fork . On the other hand , the discontinuous nature of Pol3-dependent lagging-strand synthesis would be sufficient to keep Pol3 refreshed at the fork in order to avoid replication-dependent errors or mutations . Another possibility is that this mechanism might simply maintain the coupling of leading- and lagging-strand synthesis . Thus , in addition to the role of replisome degradation in preventing genomic instability described above , polymerase degradation may function to eliminate non-functional replisomes and serve as a mechanism to maintain active DNA polymerases at the replication fork . Our investigation also revealed that Pol2 and Mcm4 undergo rapid degradation in the presence of CPT ( Figure S6 ) , which breaks replication forks . However , in this condition , the Mcm6 level was maintained ( Figure S6 ) , although it was highly unstable in swi1Δ cells ( Figure 1C and D ) . It is possible that some replisome components remain stable on the chromatin in the presence of CPT . Interestingly , Trenz et al . reported that polymerases fall off the chromatin in response to CPT , whereas Mcm7 persists [83] . Therefore , swi1 deletion generates a situation distinct from a simple mechanical breakage of the fork caused by DNA damaging agents , where the replisome cannot continue replicating DNA . Importantly , Swi1 functions as an ancillary component of the replisome by interacting with various replisome components , coupling polymerase and helicase activities and coordinating semi-discontinuous DNA synthesis [14] , [15] . It is also reported that Swi1 protects replication forks at difficult-to-replicate sites [43]–[45] . Therefore , we suggest that the loss of Swi1 results in unstable replisome structures at the moving replication fork during ongoing DNA synthesis , allowing us to examine replisome degradation pathways during DNA replication . The FPC moves with the replication fork and interacts with replisome components [16] , [18] , [21]–[25] , [27] , [28] , [84]–[87] . Surprisingly , Pol2 , Pol3 , and MCM subunits are rapidly degraded in swi1Δ cells ( Figure 1 ) . Consistently , replication fork progression is compromised in FPC deficient cells ( Figure S1 ) [27] , [52] . These results suggest that Swi1 prevents degradation of replisome components to maintain efficient progression of replication forks . In wild-type cells , multiple activities required for DNA synthesis are coupled to form a large replisome complex , resulting in efficient progression of the replication fork . However , in the absence of Swi1 , DNA replication-related activities are probably uncoupled especially at naturally difficult-to-replicate regions . This uncoupling generates unstable replisome structures , which may expose degradation signals of various replisome components to a ubiquitin ligase ( s ) associated with the replication fork . Importantly , swi1Δ pof3Δ double mutants showed catastrophic DNA replication and mitosis , suggesting that Pof3-dependent degradation of replisome components prevents genomic instability . However , we cannot exclude the possibility that mitotic catastrophe phenotypes are caused by stabilization of other Pof3 targets . For example , Pof3-dependent proteolysis of Ams2 is responsible for cell cycle-dependent transcriptional activation of core histone genes in S . pombe [88] . Indeed , defects in Ams2 degradation leads to accumulation of histones and alteration of centromere structures [88] . Such dysregulation of histone homeostasis during S phase could also lead to abnormal DNA replication , leading to mitotic problems . However , Dia2 , a Pof3-related F-box protein , is associated with the replisome and regulates replication forks in budding yeast . Dia2 is involved in ubiquitination of budding yeast Mrc1 , which is a component of the replisome [67]–[69] . Moreover , Tof1 ( Swi1 orthologue ) collaborates with Dia2 to maintain genomic integrity [89] . These findings suggest that Pof3/Dia2 acts as a part of the replisome . Consistently , we found in fission yeast that SCFPof3 is largely responsible for degradation of some replisome components ( Figure 3 and Figure 4 ) . Therefore , Pof3-mediated ubiquitination of replisome components may be prevented by Swi1-dependent replisome stabilization , which may mask potential degradation signals of multiple replisome components ( Figure 6 ) . Since many SCF ubiquitin ligases are known to recognize phosphorylated degradation signals ( phospho-degrons ) , it is also possible that replisome components undergo phosphorylation in the absence of Swi1 . Therefore , Swi1 might have direct functions in inhibiting SCFPof3 ligase possibly by inhibiting Pof3 or inhibiting potential kinases . In this regard , it is interesting to note that Mrc1 contains a potential phospho-degron , and that the Hsk1 kinase is required for efficient degradation of Mrc1 [55] . Consistently , our present study shows that Pof3 is involved in Mrc1 degradation ( Figure 4 ) . Therefore , it is possible that Hsk1-dependent phosphorylation creates Pof3-targeted phospho-degrons on multiple replisome components . However , Mrc1 degradation is independent of replication stress ( Figure 1C ) , raising the possibility that other kinases are responsible for replisome degradation upon replication stress . Further investigation of proteasome-dependent replisome degradation would identify detailed pathways in the regulation of the replisome .
The methods used for genetic and biochemical analyses of fission yeast have been described previously [90] , [91] . Drug sensitivity assays , Western blotting , pulsed-field gel electrophoresis ( PFGE ) and 4′ , 6-diamidino-2-phenylindole ( DAPI ) staining of nuclear DNA were performed as described [70] , [92] . Flow cytometry of DNA content has been described [93] , [94] . S . pombe strains used in this study were constructed using standard techniques [91] , and their genotypes and sources are listed in Table S1 . swi1Δ ( swi1::hphMX6 and swi1::natMX6 ) and pof3Δ ( pof3::ura4MX6 ) were generated by a two-step PCR method [95] , to replace swi1 and pof3 open reading frames with selection marker genes . The two-step PCR method was also used to construct a GFP or 13Myc tag at the C terminus of mcm2 , mcm6 and pof3 , generating mcm2-GFP:hphMX6 ( mcm2-GFP ) , mcm6-GFP:hphMX6 ( mcm6-GFP ) , and pof3-13Myc:hphMX6 ( pof3-13Myc ) , respectively . Oligonucleotide primers used in the two-step PCR method described above are listed in Table S2 . A temperature-sensitive skp1-94 mutation was isolated using error-prone PCR methods [63] . Mutations and epitope-tagged genes have been described for orc1-5FLAG [96] , pol2-5FLAG , pol3-5FLAG [97] , swi1::kanMX6 [17] , cdc45-5FLAG [98] , cdc25-22 [99] and mts3-1 [57] . mcm2-GFP , mcm6-GFP , pof3-13MYC , orc1-5FLAG , pol2-5FLAG , pol3-5FLAG , and cdc45-5FLAG cells show normal growth phenotype and were not abnormally sensitive to HU , CPT and MMS , indicating that the tagged version of these proteins are functional . To examine protein stability , exponentially growing cells were treated with 0 . 1 mg/ml of cycloheximide ( CHX ) for the indicated times and collected . Whole-cell extracts were prepared as described [100] . Briefly , cells were washed in STOP buffer ( 150 mM NaCl , 50 mM NaF , 10 mM EDTA , and 1 mM NaN3 ) and lysed by glass beads in lysis buffer U ( 50 mM Tris-HCl pH 6 . 8 , 2% SDS , 2 mM EDTA , 10% glycerol , and 4 M urea ) using a FastPrep Cell disruptor ( Qbiogene , Irvine , CA ) for 40 seconds at speed 6 . Protein extract was clarified by centrifugation at 13 , 000 rpm in an Eppendorf microcentrifuge 5415R for 10 min at 4°C , and the protein concentration was determined using BCA protein Assay Reagent ( Thermo Fisher Scientific , Waltham , MA ) . Immediately after the protein concentration assay , protein extracts were boiled in the presence of 5% beta-mercaptoethanol and stored at −20°C . For immunoblotting , Myc , GFP , and FLAG fusion proteins were probed with the anti-c-Myc 9E10 antibody ( Covance , Princeton , NJ ) , anti-GFP antibody ( Roche , Indianapolis , IN ) , and anti-FLAG M2 ( Sigma-Aldrich ) antibody , respectively . The anti-tubulin TAT-1 ( gift from Dr . K . Gull ) , anti-Mcm4 ( gift from Drs . S . Kearsey , Z . Lygerou , and H . Nishitani ) , anti-Mrc1 ( gift from Dr . K . Tanaka ) , and anti-PCNA ( gift from Dr . T . Tsurimoto ) antibodies were used to detect the corresponding proteins . Quantification of protein bands was performed using NIH ImageJ software . Cell fractionation was performed as described elsewhere [55] , [56] with modifications . Exponentially growing cells were harvested in 0 . 01% sodium azide by centrifugation and washed sequentially with STOP buffer , water , and 1 . 2 M sorbitol , at 4°C . Cells were resuspended in CB1 buffer ( 50 mM sodium citrate , 40 mM EDTA , 1 . 2 M sorbitol ) and treated with 2 . 5 mg/ml of Zymolyase for approximately 20 min at 32°C . When cell lysis reached approximately 95% , cell wall digestion by Zymolyase was terminated by adding equal volume of ice-cold CB2 buffer ( 1 . 2 M sorbitol , 10 mM Tris-HCl ph7 . 5 ) , and resulting spheroplasts were washed twice with 1 . 2 M Sorbitol . Spheroplasts were then incubated in Lysis buffer T ( 50 mM potassium acetate , 2 mM MgCl2 , 20 mM HEPES-KOH pH 7 . 4 , 10 mM EDTA , 1 M Sorbitol , 1% Triton X-100 ) supplemented with Halt protease inhibitor cocktail ( Thermo Fisher Scientific ) for 10 min at 4°C . Subsequently , extracts were fractionated into soluble and pellet fractions by centrifugation for 10 min at 4°C . Supernatants ( Triton X-100-soluble fraction ) were removed , boiled with a one-third volume of 3× SDS-PAGE loading buffer ( 150 mM Tris-HCl pH 6 . 8 , 6% SDS , 6 mM EDTA , 30% glycerol , 15% beta-mercaptoethanol ) , and stored at −20°C . The pellets ( Triton X-100-insoluble fraction ) were washed once with Lysis buffer ( without Triton X-100 ) , suspended in Lysis buffer , boiled with a one-third volume of 3× SDS-PAGE loading buffer , and stored at −20°C . Immunoprecipitation was performed using the anti-myc 9E10 ( Covance ) antibody with protein G sepharose beads as described [70] . Proteins associated with the anti-myc antibody were analyzed by Western blotting . For detection of ubiquitinated protein , S . pombe cells expressing a hexahistidine-ubiquitin ( 6xHis-Ub ) peptide [61] were lysed in lysis buffer G ( 6 M guanidine hydrochloride , 100 mM sodium phosphate pH 8 . 0 , and 50 mM Tris-HCl pH 8 . 0 ) . Hexahistidine-ubiquitinated proteins were purified with Ni-NTA agarose beads ( Qiagen , Valencia , CA ) , eluted in the presence of 4 M urea , and analyzed by Western blotting . | Replication stress interferes with the normal progression of the replication fork . Under these conditions , cells activate the replication checkpoint to coordinate DNA repair with cell cycle arrest . The current understanding is that , in response to replication block , this checkpoint stabilizes replication forks and replisome structures to achieve accurate DNA replication . However , it would also be advantageous for the cell to stop DNA replication and reorganize the replisome structures when conditions are not ideal , but such mechanisms have not been explored . In this study , we describe a mechanism that regulates replisome stability in response to replication stress . We found that replisome components become highly unstable and degraded when replication forks are perturbed in the absence of Swi1 , a subunit of replication fork protection complex . We demonstrate that replisome degradation is dependent on the SCFPof3 ubiquitin ligase complex . Strikingly , when we forced cells to stabilize replisome components , cells underwent abnormal DNA replication , leading to mitotic catastrophes . Thus , our study provides novel mechanistic insights into understanding how the replication machinery is regulated to achieve faithful duplication of the genome upon replication stress . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"cell",
"division",
"dna",
"replication",
"nucleic",
"acids",
"cell",
"biology",
"genetics",
"dna",
"biology",
"molecular",
"cell",
"biology",
"dna",
"repair",
"molecular",
"biology",
"genetics",
"and",
"genomics"
]
| 2013 | Coordinated Degradation of Replisome Components Ensures Genome Stability upon Replication Stress in the Absence of the Replication Fork Protection Complex |
Arbovirus infections are a serious concern in tropical countries due to their high levels of transmission and morbidity . With the outbreaks of chikungunya ( CHIKV ) in surrounding regions in recent years and the fact that the environment in Vietnam is suitable for the vectors of CHIKV , the possibility of transmission of CHIKV in Vietnam is of great interest . However , information about CHIKV activity in Vietnam remains limited . In order to address this question , we performed a systematic review of CHIKV in Vietnam and a CHIKV seroprevalence survey . The seroprevalence survey tested for CHIKV IgG in population serum samples from individuals of all ages in 2015 from four locations in Vietnam . The four locations were An Giang province ( n = 137 ) , Ho Chi Minh City ( n = 136 ) , Dak Lak province ( n = 137 ) , and Hue City ( n = 136 ) . The findings give us evidence of some CHIKV activity: 73/546 of overall samples were seropositive ( 13 . 4% ) . The age-adjusted seroprevalences were 12 . 30% ( 6 . 58–18 . 02 ) , 13 . 42% ( 7 . 16–19 . 68 ) , 7 . 97% ( 3 . 56–12 . 38 ) , and 3 . 72% ( 1 . 75–5 . 69 ) in An Giang province , Ho Chi Minh City , Dak Lak province , and Hue City respectively . However , the age-stratified seroprevalence suggests that the last transmission ended around 30 years ago , consistent with results from the systematic review . We see no evidence for on-going transmission in three of the locations , though with some evidence of recent exposure in Dak Lak , most likely due to transmission in neighbouring countries . Before the 1980s , when transmission was occurring , we estimate on average 2–4% of the population were infected each year in HCMC and An Giang and Hue ( though transmision ended earlier in Hue ) . We estimate lower transmission in Dak Lak , with around 1% of the population infected each year . In conclusion , we find evidence of past CHIKV transmission in central and southern Vietnam , but no evidence of recent sustained transmission . When transmission of CHIKV did occur , it appeared to be widespread and affect a geographically diverse population . The estimated susceptibility of the population to chikungunya is continually increasing , therefore the possibility of future CHIKV transmission in Vietnam remains .
Chikungunya virus ( CHIKV ) belongs to alphavirus of the family Togaviridae . The name chikungunya is derived from the East African language Makonde from the root verb kungunyala which means “that which bends up” , describing the stooped posture in CHIKV cases caused by swelling , stiff joints , and muscle pains . The most prominent symptom of CHIKV infection is high fever and joint pain in the acute phase . Symptoms include headache , diffuse back pain , myalgia , nausea , vomiting , polyarthritis , rash , and conjunctivitis . Due to the similarity of symptoms , CHIKV can be misdiagnosed as dengue fever , especially in the acute phase . The laboratory tests to differentiate dengue virus ( DENV ) from CHIKV infection only work days after symptom onset and are not commonly recommended , though an algorithm for testing for Zika virus ( ZIKV ) , CHIKV , and DENV has recently been published by CDC [1] . For some individuals , rheumatic symptoms can occur 2–3 months after the acute phase [2 , 3] . The primary vectors are the Aedes aegypti and Aedes albopictus mosquitoes . These vectors also spread DENV , which explains why the distributions of these two viruses overlap . Once infected , as far as is known , antibody may last for life and individuals will be protected against reinfection for a long time [4] . The proportion of infections that are asymptomatic varies widely and has been reported to be 3 . 2% , 18% , or even 82 . 1% [5–7] . The first clinical report of CHIKV fever was as early as the 1770s [8] and the first serological test of CHIKV infection was validated after an epidemic in Tanzania in 1952–1953 [9] . After the Tanzania outbreak , the virus began to be detected throughout sub-Saharan Africa , India , and countries in Southeast Asia , leading to numerous epidemic reports in subsequent years [10] . The first outbreak in Asia was reported from Bangkok in 1958 [11] and was thought to lead to the initial epidemics in Cambodia , Vietnam , Malaysia , and Taiwan . In a 2005–2006 outbreak on La Re'union Island in the Indian Ocean , CHIKV started to present with very complicated manifestations , primarily associated with encephalopathy and hemorrhagic fever [12 , 13] . This epidemic also marked the first CHIKV deaths , as well as the first cases of peripartum mother-to-child transmission [14–16] . Particularly significant was the extent of this outbreak , with 220 , 000 people , nearly 40% of the population , infected . The second recent CHIKV reemergence caused outbreaks in Malaysia ( 2006 ) and various Pacific islands with ongoing circulation since 2011 [17–19] . In 2013 , CHIKV transmission was detected in St . Martin in the Caribbean [20] , later spreading to 45 countries and territories in the Americas . The outbreaks in the Americas caused about 1 . 1 million cases in a year [21] . From December 2013 to March 2017 , there have been over 2 . 4 million cases reported , including severe cases and deaths [22] . Little is known about CHIKV transmission in Vietnam , where dengue is endemic and Aedes mosquitoes are abundant . Since CHIKV is easily misdiagnosed as DENV because of the similar symptoms , it is possible that CHIKV transmission in Vietnam could be masked by ongoing DENV transmission . Previous studies have used serology to study CHIKV activity in Brazil , India , Caribbean islands , the Union of Comoros and Tanzania [23–27] . In order to test for past exposure of infections in the population , IgG ELISA test is often used as it is commercially available and relatively easy to perfom [26–30] . These studies can provide information about the extent of present and past transmission and can help us quantify parameters of transmission . In this study , we investigated the past and current transmission of CHIKV in Vietnam by undertaking a systematic review and performing a population seroprevalence survey using IgG ELISA at four locations across the country . With the results of the serosurvey we used a mathematical model to infer parameters of transmission in the four locations across Vietnam .
The Scientific and Ethical Committee of the Hospital for Tropical Diseases in Ho Chi Minh City and the Oxford Tropical Research Ethics Committee at the University of Oxford approved the study . We performed a systematic review by searching for information in international journals , national Vietnamese journals , and online news . The key word “chikungunya Vietnam” was used in the search engines: PubMed , Web of Science , ProMED and Google news search . For the Google news search engine , the search results only from Vietnam in the period 01/01/2000 to 03/21/2017 were used . The key word “chikungunya” was used in local search engines for papers and theses: Vietnam Journal of Preventive Medicine—a peer-reviewed journal in Vietnam [31] and a National Library of Vietnam–Ph . D . thesis storage , which contains electronic versions of about 21 . 300 Ph . D . theses ( full-texts or summaries ) [32] . As part of a large ongoing study of serial seroepidemiology , residual serum samples are collected from four hospital laboratories in southern Vietnam: the Hospital for Tropical Diseases in Ho Chi Minh City , An Giang General Hospital , Dak Lak General Hospital , and Hue Central Hospital [33–36] . For this study we selected age stratified samples from those collected in 2015 . Geographic locations are shown in Fig 1 . The patients came to those hospitals as inpatients or outpatients; admitting ward for inpatients is recorded , but diagnosis and reason for visit are not . All samples were anonymized . In each location , we tested 136 or 137 samples and aimed for at least 15 samples in each of the following age groups: 1–10 years , 11–20 , 21–30 , 31–40 , 41–50 , 51–60 , 61–70 , 71+ years . In some of the older age groups , there were fewer than 15 samples in the serum bank . However , each age group contained at least 10 samples . The serum samples were tested for the presence of CHIKV IgG antibody using the ELISA kit for CHIKV–NovaLisa [38 , 39] . The absorbance values of the samples were translated to NovaTec Units ( NTU ) . According to the manufacturer's guidelines , samples were defined as positive with > 11 NTU , negative with <9 NTU , and inconclusive with 9–11 NTU . We assessed whether there were any differences in proportion positive by gender , sample locations , and original admission wards . The binomial 95% confidence intervals for each age group were calculated using the Wilson score interval method . We also calculate the age-adjusted seropositivity for each location using the direct standardization method with the Vietnam standard population from the World Population: the 2015 Revision online dataset , conducted by the United Nations [40] . All analyses were performed in R software version 3 . 3 . 3 [37] . The force of infection ( FOI ) is defined as the per capita rate at which susceptible individuals are infected by an infectious disease . In this study , we used Muench’s catalytic model [41] to estimate the FOI varying in time but not with age . The ELISA results of seropositives and seronegatives were stratified into 15 age groups , each spanning five years . The likelihood of a single positive individual in age group n is: Lnpos=1−exp ( −5∑i=1n−1λi−2 . 5λn ) ( 1 ) The likelihood of a single negative individual in age group n is: Lnneg=exp ( −5∑i=1n−1λi−2 . 5λn ) ( 2 ) With λi ( i = 1 , 2 , … , 14 ) the average annual FOI in respective period 2011–2015 , 2006–2010 , 2001–2005 through to 1946–1950 . λ15 represents for the annual FOI estimated for the period 1931–1945 . Further details of the annual FOI estimation from 1931 to 2015 can be found in S1 Text . We introduce index iend to the model to infer the last time period that transmission occurred . The index is defined so that every FOI before it ( which means FOI values of more recent years ) are fixed at near 0 . For other time periods , a flat uninformative prior was used . The model fit was run for each index value and the 16 different models were compared using the DIC values . We fit the model using RStan package [42] . Further details of this model can be found in S1 Text . From the average annual FOI , we can derive the susceptible proportion over time . Our model assumes that the chance of being seropositive results only from transmission in that location , and that there was no inward migration or individuals acquiring infection elsewhere . From 1930 to 2015 , the proportions of susceptible individuals in each age group were calculated for each iteration of the model by a formula: Sj=∑k=080+aj , ke−λj , kaff ( 3 ) Sj is the proportion of the susceptibles in each year j = 1930 , 1931 , … , 2015 . aj , k is the age-specific proportion of year j for each age group k = 0 , 1 , 2 , … , 79 , 80 + . For the time periods before we are able to estimate FOI from our data , we consider two scenarios . The “no endemic scenario” hypothesizes there was no transmission prior to 1930 , while the “endemic scenario” takes the FOI prior to 1930 as equal to the FOI estimated for 1931 . λj , kaff is the FOI that affects the age group k in the population in year j . Full descriptions of λj , kaff definition with the 2 endemic scenarios are in S1 Text . From 1950 to 2015 , the demographic data comes from the World Population Prospects: the 2015 Revision online dataset , conducted by the United Nations as above [40] . Another paper provided the estimated age composition data of 1929–1944 every 5 years [43] . The years in between were inferred using linear interpolation .
Forty papers were found in different search engines and reviewed ( see Fig 2 ) . Further details of all abstracts are in S1 Table . One paper [44] was in PubMed but it was not possible to find the abstract . There were 3 duplicates between PubMed and Web of Science . After further ruling out 20 irrelevant abstracts and 2 unavailable abstracts , the full-text of 15 abstracts were inspected . There are two references with information on fever of unknown origin in United States soldiers in Vietnam [45 , 46] , which was already thoroughly recorded in another paper [47] . In the process of full-text reviewing , one more relevant Vietnamese paper [48] was found in the reference of another paper [49] . We also include a recent study of CHIKV from our research institution [50] . Finally , 15 papers were included . There were 251 articles from Google News search and one report from ProMED search . Most articles mentioned CHIKV only briefly in the context of recent ZIKV management in Vietnam . There were only 3 relevant unique articles from Google News and a ProMED report included . From the systematic searches we were able to assemble information about the past and current transmission of CHIKV in Vietnam . In August 1963 , there was an outbreak of hemorrhagic fever disease in children in the Mekong River Delta ( southern Vietnam ) [51] , characterized by shock and a high mortality rate [52 , 53] . Research on this outbreak showed that several children from An Giang province ( rural area ) and Ho Chi Minh city ( urban area ) with acute hemorrhagic fever had evidence of CHIKV infection ( high hemagglutination-inhibition ( HI ) and complement-fixation ( CF ) antibodies titres to CHIKV ) [51] . In that same study , 22 of 75 serum samples from healthy children in Ho Chi Minh City had detectable CHIKV HI antibody , suggesting past CHIKV transmission in Ho Chi Minh City [51] . Serosurveys on a larger scale from consecutive years ( 1963–1965 ) also reported the proportion of seropositive people in Ho Chi Minh City as 31 . 36% by CF in 1963 and 10 . 26% by HI in 1964–65 [54 , 55] . In the following years , there were scattered reports of CHIKV from 1963 to 1982 . In summary , in 1966 , 10 out of 110 American soldiers with fever of unknown origin in southern Vietnam tested positive by virus isolation to CHIKV [47] . In 1972 , in a large serosurvey in three locations in southern Vietnam , 21 out of 130 suspected cases of DENV had positive CHIKV PRNT [56] . Finally , a multi-year study reporting hemorrhagic fever cases in southern Vietnam , found 12 cases with positive CHIKV virus isolation from 1978 to 1982 [57] . From 1983 to 1986 , all tests in this survey were negative and after 1987 the survey was stopped . This suggests transmission of CHIKV may have ended in 1982 . There were no papers reporting testing for CHIKV between 1986 and 2005 . More recently , a retrospective study of samples from febrile individuals in 6 countries in South East Asia and Fiji from 2005 to 2006 found IgM antibody positive cases of CHIKV in some countries but none in Vietnam . However , 11 of 44 samples from febrile patients tested in Hanoi ( northern Vietnam ) were positive by IgG ELISA and confirmed by neutralization assay [58] . In 2009 , there was a 14-fold increase in cases with dengue-like symptoms compared to previous years and 60 percent of patients with classic dengue-like symptoms tested negative for DENV [59] . Of the acute hemorrhagic fever patients tested , 4/50 were PCR positive to CHIKV and genetic sequencing showed > 93% similarity with the CHIKV derived from Africa ( S27 subtype ) [48] . In 2010 , the National Institute of Hygiene and Epidemiology ( NIHE ) announced 15 patients with hemorrhagic fever who were negative for DENV and positive to CHIKV [60] . From September 2010 to June 2011 , a cohort study conducted in My Tho city in southern Vietnam reported that a total of 19 cases out of 32 febrile children tested were positive by ELISA IgM CHIKV [61 , 62] . After 2011 , due to the CHIKV outbreaks in Laos and Cambodia [63 , 64] , several studies were conducted in the border locations in Vietnam but showed no activity of the virus [49 , 65–67] . A recent study tested 8015 febrile children in southern Vietnam and revealed four positive cases of CHIKV by RT-PCR [50] . The positive samples were found between August and November 2012 , with one case from Ho Chi Minh City and three from Binh Duong province ( about 20 kilometers away from the Cambodian border ) . Phylogenetic analysis indicated those four strains were closely related to the Cambodia strain which caused the outbreaks in 2012 [63] . In early 2016 , a news report stated that of 83 people with Zika-like symptoms tested across eight provinces in Vietnam , nine individuals from Can Tho City tested positive to CHIKV[65] . However , the test used was not reported . In April 2016 , NIHE found 56 ( 0 . 29% ) A . aegypti that were positive to ZIKV , 29 ( 0 . 12% ) positive to DENV but none positive to CHIKV [68] . In this study , 546 individuals were tested for CHIKV IgG using the NovaTec CHIKV ELISA . Individuals were defined as positive , negative , or borderline depending on the NovaTec titres as defined in the methods . The full dataset is in S1 Data . Across all ages , there were 21 , 22 , 16 , 14 positive individuals in An Giang , Ho Chi Minh , Dak Lak , and Hue respectively . Fig 3 shows the numbers of positive , negative and borderline cases by 5 year age group for each location ( more detail in S2 Table ) . The direct age-adjusted seropositivity percentages were 12 . 30% ( 6 . 58–18 . 02 ) , 13 . 42% ( 7 . 16–19 . 68 ) , 7 . 97% ( 3 . 56–12 . 38 ) , and 3 . 72% ( 1 . 75–5 . 69 ) for An Giang province , Ho Chi Minh City , Dak Lak province , and Hue City respectively . For males the percentage positive was 18% ( 9 . 8–30 . 8 ) , 19 . 6% ( 11 . 0–32 . 5 ) , 10 . 2% ( 6 . 2–16 . 4 ) , and 14 . 5% ( 7 . 6–26 . 2 ) ; and for females they were 14 . 5% ( 8 . 5–23 . 6 ) , 15 . 2% ( 8 . 9–24 . 7 ) , 10 . 2% ( 4 . 7–20 . 5 ) , and 10 . 1% ( 5 . 2–18 . 7 ) , for each province respectively . There were no statistical differences between the proportion positive in males and females . Samples in this survey come from people admitted to different hospital wards and we saw no difference in positivity by ward of admission ( full details in the S2 Table ) . It is noteworthy that almost all the positive tests were in those 30 years old and above , except for Dak Lak where three children were positive ( two 10-year-olds and one 13-year-old , more detail in S3 Table ) . The proportion of positive individuals and binomial 95% confidence intervals by age groups ( excluding borderline results ) are shown in Fig 4 . The DIC of the different models with transmission ending in each of the 5 year periods are shown in Table 1 . In An Giang , the best model ( smallest DIC ) is with transmission ending in 1986–1990 . In Ho Chi Minh City , the model ranking supports endemic transmission ending in 1976–1980 . The transmission in Dak Lak is estimated to be more recent and Hue , on the other hand , is estimated to have the longest time without CHIKV transmission—since 1966–1970 . There are large changes in DIC values for each location , hence clearly suggesting transmission before 1980 , 1975 , 2000 , and 1960 in An Giang , Ho Chi Minh , Dak Lak , and Hue respectively . Based on [70 , 71] , if we choose models that have DIC differences < 3 compared to the best model , we estimate that transmission ended in the periods 1981–2000 , 1976–1995 , 2001–2015 , and 1961–1980 in An Giang , Ho Chi Minh , Dak Lak , and Hue respectively . Though ( with a DIC difference between 3 and 5 ) we cannot rule out slightly more recent transmission in An Giang , Ho Chi Minh , and Hue , the DIC values clearly suggest no on-going transmission in these locations . FOI estimates and model fits for 3 best models in each location shown in Figure A and B in S2 Text . Annual FOI estimates by time period from the model with the smallest DIC value for each location are shown in Fig 5 ( full posterior distributions shown in Figure A in S1 Text ) . We estimate An Giang to have the highest transmission in 1931 to 1955 with a peak of FOI of 0 . 067 ( 0 . 001–0 . 222 ) in 1946–1950 . In Ho Chi Minh City there is a similar peak in transmission estimated in 1946–1950 of 0 . 040 ( 0 . 001–0 . 136 ) , but transmission is estimated to be higher than An Giang in the later years from 1966 to 1980 . In Dak Lak , we estimate low transmission with almost no mean FOI estimated larger than 0 . 02 in any period ) . The peak of FOI in 2006–2010 hints a recent activity of CHIKV in this area though the model does not capture the rest of the data well . In Hue , the magnitudes of the past FOI are comparable to Ho Chi Minh City . The model output and the seroprevalence data for each location is shown in Fig 6 . We estimated the FOI as an average over five year periods . In order to assess how sensitive our results were to this assumption we estimated the FOI for models with 1–5 year groups ( S1 Text ) . The results suggest that our data may not have enough resolution for FOI estimation in 1–2 years however the 5 year group models fit similarly to the 3–4 year group models and estimate transmission ending at broadly similar times for all locations . Based on the FOI estimates , we were able to infer changes in the susceptible proportion over time as shown in Fig 7 . Depending on our assumption about previous transmission before 1930 , there are differences in our estimates of the susceptible proportion in the early years . However , the discrepancy becomes negligible in later years . The higher FOI in Hue , HCMC and An Giang means we estimate the greatest decrease in the susceptible proportions in these locations . In Dak Lak , by contrast , the susceptible proportion remains high at all times . These results are consistent across the top three models for all locations ( see Figure C and D in S2 Text ) . Due to the estimated lack of transmission in recent years , we see a steady increase in the susceptible proportion in recent years , up to over 85% in all locations in 2015 .
Using two complementary methods , a systematic review and a seroprevalence study , we have been able to provide a description of past and present CHIKV transmission in central and southern Vietnam . We conclude that there has been little to no recent transmission in these locations in Vietnam , but that CHIV was last circulating widely in the early 1980s . In the locations where there was past literature on CHIKV cases ( HCMC and An Giang ) results provided solid evidence of CHIKV transmission in the periods 1963–1966 and 1978–1982 . The results from the force of infection estimates , estimate transmission in the intervening years ( with increases in the proportion seropositive in the relevant age groups leading to estimates of transmission in these time periods ( Fig 5 ) ) . Therefore we conclude consistent transmission before 1980s , though cannot rule out multiple smaller outbreaks rather than one large outbreak . With the serological results , we were able to reconstruct the magnitude of CHIKV transmission that occurred before the 1980s . In two of the areas we surveyed , HCMC and An Giang , we estimated the highest and most consistent transmission , with an average of 2–4% of the population being infected each year . The results from previous serosurveys in 1963–1965 estimated between 10 and 30% seropositivity depending on the methods used [54 , 55] . For the same time period we estimated between 10 and 40% ( depending on the scenario used for past transmission ) , which is consistent with this estimate . In Dak Lak , three seropositive individuals less than 30 years old caused the estimates of the FOI in recent years to be slightly greater than 0 . As the model does not incorporate external introductions , we cannot determine if these seropositive individuals were caused by introductions or by low level local transmission . Supporting the external introduction theory , Dak Lak borders Cambodia and there could be limited transmission from Cambodia into Dak Lak province; alternatively , some of those found to be positive could have traveled to Cambodia and been infected there . In the systematic review , we did not find any information about past transmission in Dak Lak . Hue had the lowest transmission , and Hue’s transmission appeared to end earlier than in Dak Lak , An Giang , or HCMC . We found no information about testing for historical transmission in Hue , however , in a study from 2012 to 2014 , no CHIKV activity was found in this location [66] . We can consider the differences we estimate in chikungunya transmission intensity across locations in the context of climate and transmission patterns of DENV , as they are transmitted by the same mosquitoes . An Giang and HCMC in the south are generally hot and have ongoing DENV transmission throughout the year , but with higher numbers of DENV cases in the wet season [39 , 72] . The DENV transmission intensity is thought to be higher in these two locations than in the other two we tested [39] . Dak Lak , by contrast , is cooler and has generally lower levels of DENV transmission , however with some years with large numbers of cases . Hue , farther north , has clearer seasons and generally lower DENV case numbers . These patterns across the areas in DENV transmission are broadly consistent with the forces of infections and length of transmission we observed for CHIKV in our study . It is worth noting however that although the same mosquitoes can transmit CHIKV and DENV , the way climate impacts the virus transmission may differ [73] . It is useful to compare our findings on force of infection and the proportion seropositive after CHIKV outbreaks to those in other settings . A recent CHIKV seroprevalence study in Chennai , India found steady consistent seropositive proportions across 5–40 years old , suggesting epidemic but not endemic transmission [29] . Our results are comparable to those in Cebu in Philippines which estimated the susceptible population to be above 50% at all times despite large outbreaks [70] . It would be interesting to see estimates of the FOI and the proportion that remain susceptible after the recent outbreaks in South and Central America . There are limitations to our study . Our samples are from individuals who have attended hospital for some reason , so are not fully representative of the population . However , for the younger individuals , we would assume this would bias the study towards finding positive individuals as in Berto et al [35] , which we did not . For older individuals , this may be more representative of the population as a whole and as the exposure is assumed to be over 30 years ago , it is reasonable to assume this hospital visit is unrelated to the CHIKV infection exposure . It is also reasonable to assume that CHIKV did not lead to death in a significant proportion of individuals [10] , therefore , our seroprevalence estimates are unbiased . We did not have samples from Hanoi , despite reports of recent chikungunya transmission there . Future serological studies in Hanoi would be of interest . The three hospitals in An Giang , Dak Lak , and Hue province , are the largest hospitals in the province and therefore are likely to geographically represent the population in the respective areas . However , for the hospital in HCMC , patients may come from across the south of Vietnam and not just from the city . To assess this possible bias , we presented analysis for HCMC and An Giang together in S1 Text , and the findings are not different from the results for each location . In addition , by stratifying the serological results to 5-age groups , our model might lose some finer details from the exact age of each individual; however we show that the trend of the results is consistent across models with different age groupings . Another limitation with any serological survey is concern about cross-reactivity . The manufacturers state that the test has no cross-reactivity to DENV virus , tick-borne encephalitis , CMV , EBV and Helicobacter pylori and that the diagnostic specificity and sensitivity are > 90% [38] . Cross-reactivity with O'Nyong Nyong virus is not excluded , however , there are no reports of O’Nyong Nyong virus in Vietnam , and the virus is thought to be limited geographically to East Africa [74–76] . However we cannot rule out cross-reactivity with other unknown alphaviruses . In summary , through a systematic review and seroprevalence study , we have provided information on the past and current transmission of CHIKV in southern and central Vietnam . We saw evidence of widespread transmission of CHIKV over 30 years ago , with variation across locations . It is not clear why transmission ended , but there may be lessons here for what we could expect to occur in South and Central America after a number of years of widespread transmission . We estimate that there is high and increasing susceptibility to CHIKV in southern and central Vietnam suggesting that the population is vulnerable to a CHIKV outbreak should the virus be re-introduced to Vietnam , therefore public health agencies should be vigilant for chikungunya cases . | In recent years , the reemergence of chikungunya has gained global attention and threatened to become a global outbreak . Although the epidemiology of chikungunya is known globally , the viral activity in Vietnam has not been thoroughly investigated . In this paper , we used information from a systematic review and serological survey to understand the past activity of chikungunya in Vietnam . In the serological survey , we tested age -stratified population serum samples in the central and southern parts of Vietnam from 2015 for chikungunya IgG , which is indicative of past exposure to chikungunya . Our results show almost no positive individuals below the age of 30 years old , indicating very low to no virus transmission for the last 30 years . In individuals over 30 years old , there were more positive individuals , and we estimate widespread transmission with an average of 2–4% of the population infected each year . Because of the lack of recent transmission in these locations , the susceptible population is increasing and is now high , suggesting Vietnam is vulnerable to future outbreaks if chikungunya is reintroduced . Hence , chikungunya surveillance along with surveillance for other arboviruses , such as dengue , should be conducted in Vietnam . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"chikungunya",
"infection",
"pathogens",
"geographical",
"locations",
"microbiology",
"tropical",
"diseases",
"alphaviruses",
"viruses",
"age",
"groups",
"chikungunya",
"virus",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"zoonoses",
"serology",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"vietnam",
"research",
"assessment",
"people",
"and",
"places",
"hemorrhagic",
"fevers",
"asia",
"viral",
"pathogens",
"systematic",
"reviews",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"viral",
"diseases",
"organisms"
]
| 2018 | Evidence of previous but not current transmission of chikungunya virus in southern and central Vietnam: Results from a systematic review and a seroprevalence study in four locations |
Gene order is not random in eukaryotic chromosomes , and co-regulated genes tend to be clustered . The mechanisms that determine co-regulation of large regions of the genome and its connection with chromatin three-dimensional ( 3D ) organization are still unclear however . Here we have adapted a recently described method for identifying chromatin topologically associating domains ( TADs ) to identify coexpression domains ( which we term “CODs” ) . Using human normal breast and breast cancer RNA-seq data , we have identified approximately 500 CODs . CODs in the normal and breast cancer genomes share similar characteristics but differ in their gene composition . COD genes have a greater tendency to be coexpressed with genes that reside in other CODs than with non-COD genes . Such inter-COD coexpression is maintained over large chromosomal distances in the normal genome but is partially lost in the cancer genome . Analyzing the relationship between CODs and chromatin 3D organization using Hi-C contact data , we find that CODs do not correspond to TADs . In fact , intra-TAD gene coexpression is the same as random for most chromosomes . However , the contact profile is similar between gene pairs that reside either in the same COD or in coexpressed CODs . These data indicate that co-regulated genes in the genome present similar patterns of contacts irrespective of the frequency of physical chromatin contacts between them .
Genome-wide expression studies have shown that gene order is not random in eukaryotes , and that genes with similar expression patterns are often linked ( reviewed in [1 , 2] ) . For instance , early work in yeast showed that neighboring pairs of genes ( adjacent and nonadjacent-but-nearby ) have correlated expression that is independent of their orientation [3] . Furthermore , Cho et al . showed that 25% of genes with cell cycle–regulated expression patterns were linked to genes induced in the same cell cycle phase [4] . Since then , clusters of genes with similar expression patterns and related functions have been identified in several model organisms . For example , muscle expressed genes have been shown to cluster in Caenorhabditis elegans [5] and in humans [6] . Clustering of other tissue-specific genes has been reported in Drosophila [7] , mouse [8 , 9] , and human [10 , 11] . Lee and Sonnhameer showed that genes involved in the same KEGG pathway tend to be clustered in several eukaryotic genomes [12] . Kosak and Groudine defined tandem gene arrays as contiguous stretches of genes differentially expressed in the same way during cellular development or differentiation [13] . Tandem gene arrays , mostly constituted by two or three genes co-regulated during hematopoiesis or myogenesis have been described [14 , 15] . Other works have suggested the existence of clusters of coexpressed genes with little or absence of co-functionality [16 , 17] . Finally , clusters of housekeeping genes [18] or of highly expressed genes [19] have been also reported . The mechanisms responsible for coexpression of gene clusters are unknown . Coexpression of gene pairs can be explained by bidirectional promoters or by defects in transcription termination . In fact , polycistronic transcripts have been detected in some eukaryotes . For example , C . elegans contains around 1000 operons that are 2–8 genes long [20] . However , polycistronic transcripts are not common in mammals [21] . It has been proposed that long-range cis regulatory elements , such as enhancers and local control regions , may be responsible to a certain extent for coexpression of gene clusters [2 , 22] . So far , this has only been demonstrated for some well-known examples , such as the globin locus or the Hox genes . The existence of nuclear and chromatin physical three-dimensional ( 3D ) domains [23 , 24] appears an obvious possible mechanism to explain coexpression clusters . Microscopic techniques have demonstrated that chromosomes occupy specific regions within the nucleus , with larger chromosomes positioned closer to the nuclear envelope , and smaller chromosomes generally located in the center of the nucleus [25–27] . In addition , chromosome organization has been demonstrated to be nonrandom as a function of cellular differentiation [28] . At the gene level , several reports have demonstrated that inactive genes are often located close to the nuclear lamina , and that gene activation triggers the movement of loci to the center of the nucleus [29–31] . Furthermore , some microscopy data indicate the existence of transcription factories , in which distant genes appear closely linked [32 , 33] . Chromosome conformation capture ( 3C ) and 3C-based technologies ( 4C , 5C , and Hi-C ) have revealed the existence of a vast number of chromatin interactions along the genome ( recently reviewed in [23 , 34] ) . Most of the chromatin contacts occur within proximal regions; however , intra-chromosomal long-range contacts and , to a lesser extent , inter-chromosomal contacts also occur [35–37] . Contacts between promoter and transcription termination regions , and between promoters and distant regulatory regions , have been well characterized and occur at the kilobase ( kb ) to megabase ( Mb ) scale . However , the functional implications of long-distance interactions are less well understood and have been associated with transcription factories [38 , 39] . The existence of regions of high-density of internal contacts surrounded by regions of relatively low density of contacts has suggested that the genome is organized into modular domains , called topologically associating domains ( TADs ) or contact domains [40–42] . TADs usually show specific patterns of histone marks and , in some cases , a coordinated regulation of their genes [15 , 41 , 43] . However , the relationship at the genomic level between coexpression gene clusters and TADs has not been studied . Here we have computationally identified and characterized human coexpression domains ( CODs ) of normal breast and cancer breast samples . Interestingly , we find that COD genes tend also to be coexpressed with other genes localized in different CODs , indicating an inter-COD co-regulation . As compared to the genome of normal breast tissue , the breast cancer genome had a similar number of CODs but a different COD gene composition , with less co-regulation between CODs , suggesting less structured expression patterns . CODs are not coincident with TADs or contacts domains . However , we observed a similar profile of long-range chromatin contacts between co-regulated CODs , indicating that co-regulated CODs interact with similar regions of the genome .
We obtained RNA-seq expression data of 20 , 502 genes from 100 normal breast tissue samples from The Cancer Genome Atlas ( TCGA ) ( see Methods ) . Genes were divided into 23 groups according to its chromosome location . Twenty-three correlation matrices , one for each chromosome , were then constructed in which the matrix entry mij was the Pearson correlation coefficient between the expressions of gene i ( in the i-th row ) and j ( in the j-th column ) in the 100 samples . Coexpression between the genes of each chromosome was visualized in heat maps , whereby genes are ordered in the 5′–3′ chromosomal order and color intensities indicate correlation coefficient . The heat maps showed regions of positive and negative correlations and , in some chromosomes , a very sharp plaid pattern ( Fig 1A for chromosome 16 , see S1 Fig for all chromosomes ) . Close analysis of the diagonals of the maps revealed regions containing highly coexpressed collinear genes . We called these regions Coexpression Domains ( CODs ) ( Fig 1B ) . Physical gene maps of two CODs are shown in S2 Fig . To systematically identify all CODs in the genome , we adapted a method recently designed for TAD identification , TopDom [44] . Following their method , first we calculated an average coexpression value between a window of four genes upstream and downstream around a gene i ( binsignal ( i ) ) . As shown in Fig 1C and 1D , the value of binsignal ( i ) is relatively high inside CODs but low between CODs . COD boundaries were determined as regions where the binsignal ( i ) value changes significantly ( p < 0 . 05 ) . CODs were defined as regions with a high binsignal ( i ) value ( above average binsignal ( i ) of the genome ) delimited by statistically significant boundaries ( see Methods ) . CODs with less than four genes were discarded . Randomization of the positions of the genes dramatically altered the distribution of the binsignal ( i ) values , such that CODs could not be identified ( S3 Fig ) . Using this system , we identified 524 CODs in the human genome that were distributed proportionally to the chromosome gene number ( S4A and S4B Fig ) . Of the 20 , 502 genes analyzed , 7666 ( 37 . 4% ) were found within CODs , with an average of approximately 14 . 6 genes per COD and a median of 10 genes per COD ( S5A Fig ) . This distribution was remarkably similar in most chromosomes ( S4C Fig ) . Median length was around 0 . 9 Mb per COD ( S5B Fig ) . We then compute the average intra-COD coexpression as the average of Pearson coefficients of coexpression among all pairs of genes within the COD . Average intra-COD coexpression values of real CODs were significantly higher than average intra-COD coexpression of randomized CODs of the same size and separated by the same distance as real CODs ( p = 5 . 0x10-74 , Bonferroni-corrected Mann-Whitney test ) ( Fig 1E ) . S1 Table lists all identified CODs , their gene composition , the average intra-COD coexpression value , and the p-value of the boundaries . We next tested whether CODs correlate to previously-described gene clusters . Caron et al . described clusters of highly expressed genes ( called RIDGEs , Regions of Increased Gene Expression ) in the human genome [19] . COD genes did not display higher levels of expression than non-COD genes ( Fig 1F ) , indicating that CODs are not exclusively formed by highly expressed genes . Furthermore , COD genes were not enriched in genes located in RIDGEs ( S2 Table ) ( Fig 1G ) , indicating that CODs do not correspond to RIDGEs . Further , Lercher et al . reported clustering of housekeeping genes on the human genome [18] . We investigated the presence of housekeeping genes in CODs . Only 21 . 7% ( 1662 ) of the 7666 genes found in CODs were housekeeping genes , as defined by Eisenberg and Levanon [45] ( Fig 1H ) . While these data indicate a significant enrichment of housekeeping genes in CODs ( hypergeometric test p = 2 . 2 × 10−17 ) , they also imply that most CODs genes ( 78 . 3% ) are not housekeeping genes . Tandem duplicated genes are often coexpressed because they use to have similar promoters [2 , 46] . Of the 148 gene clusters of more than 4 genes previously defined in the human genome [47] , 65 were found in CODs . For example , COD 36 of chromosome 1 is formed by 13 members of the S100A gene family; CODs 3 and 5 of chromosome 6 contain 29 and 15 canonical histone coding genes , respectively; and COD 16 of chromosome 5 contains protocadherin alpha and beta clusters , among other genes ( S1 Table ) . We performed gene ontology ( GO ) analysis of all genes included in CODs . Interestingly , COD genes were slightly but significantly enriched in categories involved in DNA and RNA metabolism , such as nucleic acid metabolic processes ( Bonferroni-corrected hypergeometric test p = 1 . 7 × 10−9 ) , gene expression ( p = 7 . 06 × 10−8 ) , and RNA biosynthetic processes ( p = 7 . 09 × 10−8 ) ; this is consistent with the slight enrichment in housekeeping genes . As expected , random COD genes were not enriched in any GO category . Interestingly , we also observed that genes located in CODs are often coexpressed with other genes also located in CODs , creating the plaid pattern in the heat maps ( Fig 1A , 1B , 1C and 1D; S1 Fig ) . Coexpressions between pairs of genes located in the same COD ( intra-COD ) are significantly higher than coexpressions between pairs of genes located in different CODs ( inter-COD ) , and both of these are higher than the rest of pairwise gene coexpressions ( Fig 2A ) . Randomization of the CODs dramatically decreased these differences ( S6 Fig ) . We then defined an average inter-COD coexpression between two CODs ( containing i and j genes , respectively ) as the average of Pearson coefficients of coexpression among all the i–j pairs of genes . The distribution of average inter-COD coexpressions from real CODs was very significantly different from the values obtained using randomized CODs of the same size ( p < 10−300 , Bonferroni-corrected Mann-Whitney test ) ( Fig 2B ) . We selected an |average Inter-CODs coexpression| ≥ 0 . 2 as strongly significant for future calculations . Thus , 44 , 85% of the possible intra-chromosomal pairs of CODs are positively co-regulated ( average inter-CODs coexpression ≥ 0 . 2 ) , and 0 . 53% are negatively co-regulated ( average inter-CODs coexpression ≤ –0 . 2 ) . Examples of positive coexpression , no coexpression , or negative coexpression inter-CODs are shown in Fig 1C and 1D , and S7 Fig , respectively . These data indicate the existence of mechanisms of co-regulation of gene expression that operate over long distances . We also observed that the average inter-COD coexpression is maintained along very large distances , and even between CODs placed in different arms of the same chromosome . In fact , inter-COD coexpression was higher than expected by chance for all distances ( Fig 2C , Mann-Whitney test p < 0 . 01 for all bins ) . In summary , our data suggest existence of a level of co-regulation inter-CODs along the chromosomes . We next compare the previously identified CODs from normal breast samples with CODs identified in breast cancer samples . For that , we collected RNA-seq expression data of 20 , 502 genes from 369 breast cancer tumors from TCGA ( see Methods ) and used it to construct correlation matrices encompassing all pairs of genes of each chromosome . After applying our COD identification method , we obtained 504 CODs in the breast cancer genome ( S1 Table ) , which is similar to the 524 found in normal tissue . Likewise , the number of CODs per chromosome ( S4D and S4E Fig ) , and the number of genes per COD ( S4F and S5C Figs ) , were also similar between the breast cancer samples and normal samples . Furthermore , distribution of average intra-COD coexpression values from cancer samples was not significantly different from the normal tissue distribution ( Bonferroni-corrected Mann-Whitney test > 0 . 01 ) ( S5E Fig ) . These data suggest that gene expression is also organized in CODs in the tumor genome . We then compared the overlap between normal and cancer CODs . We found that 59 . 1% of the genes found in CODs in normal breast were also found in CODs in breast cancer genomes ( Fig 3A ) . In sharp contrast , only 9 . 5% of the CODs overlapped in more than 80% of the genes ( Fig 3B ) , indicating a strong reorganization of CODs from the normal to the cancer state . Interestingly , about 22 . 5% of the COD boundaries coincided between normal and cancer samples ( Fig 3C ) , which is significantly higher than the coincidence obtained between random normal CODs and real cancer CODs ( 14 . 6% ) ( p < 0 . 0001 , χ2 test ) . These data suggest that often breast cancer CODs results from fusion and division of normal CODs or by shifting of one of the borders but not the other . Some of the cancer-specific CODs are related to known breast cancer amplified regions . For example , CODs 5 , 6 and 7 of chromosome 8 , COD 13 and 14 of chromosome 11 , COD 15 of chromosome 17 , and COD 6 of chromosome 20 are absent in normal breast tissue and correspond to the well-known breast cancer–associated amplicons in 8p11-p12 , 11q13 , 17q12 , and 20q13 ( S8 Fig ) ( [48–50] and data obtained from TCGA ) . Other regions with strong reorganization of CODs between normal and breast cancer samples are not associated to genomic reorganizations . For example , the cancer-specific COD 6 of chromosome 7 ( Fig 3D ) encompasses 21 genes , of which 10 have been implicated in cancer . Particularly interesting are the genes PGAM2 [51] , GCK [52] , and OGDH [53] , which encode enzymes of the glycolysis and the tricarboxylic acid cycle , as well as POLM [54] and POLD2 [55] , which encode DNA polymerase μ and a subunit of DNA polymerase δ , respectively , suggesting a possible co-regulation of energetic metabolism and DNA replication genes in breast cancer . Fading of normal breast CODs in cancer was also observed ( Fig 3E ) . We next compared average inter-CODs coexpression . Strikingly , we observed a very significant decrease of inter-CODs coexpression in breast cancer with respect to normal breast samples ( Bonferroni-corrected Mann-Whitney test p < 10−300 ) ( Fig 3F and 3G ) . Using the same criterion that we used for normal tissue ( |average inter-CODs coexpression| ≥ 0 . 2 ) , only 5 . 1% ( 283 ) positive , and 0 . 06% ( 4 ) negative , average inter-COD coexpression values were found . Loss of inter-COD coexpression of a region of chromosome 1 is shown in Fig 3G . In clear contrast to normal tissue , average inter-CODs coexpression decreased very significantly with chromosomal distance in breast cancer samples ( Fig 3H ) . We also observed strong changes of long-range coexpression patterns associated with genomic reorganizations . Thus , the heat map of coexpression of chromosome 8 in cancer changes drastically with respect to the normal breast tissue , probably due to the frequent amplifications of the q arm of this chromosome ( S8 Fig ) . Taken together , these data suggest that long-distance gene co-regulation is impaired in the cancer genome . We next investigated whether specific CODs can be associated with specific clinicopathological tumor characteristics . For that , we divided the 369 breast tumor collection into two subtypes , according to the presence or absence of metastasis in lymph node of the patients ( N0 , non-metastatic , versus N1-3 , metastatic nodes ) . The group of patients with metastasis in nodes presented a poor prognosis respect to the group of patients without invaded nodes ( Log rack test p-value = 0 . 0048 ) ( S9A Fig ) . Expression data of these two sets of tumors were used to determine CODs . We found about 34% ( 182 ) of CODs coincidence between N1-3 and N0 tumors ( S9B Fig ) . Among the non-coincident CODs , divisions , fusions , limits shifting and completely different CODs were found . S9C Fig shows an example of a N1-3-specific highly coexpressed COD containing nine genes: VPS26A , SUPV3L1 , HKDC1 , HK1 , TACR2 , TSPAN15 , NEUROG3 , C10ORF35 , COL13A1 that is not present neither in the N0 tumors nor in the normal breast . High expression of some of these genes has been previously linked to metastasis and poor prognosis [56–58] . The coexpression of these nine genes in tumors with metastatic nodes suggests the existence of mechanisms specific for this type of tumors that coordinate expression of all the genes of this COD . Recent observations suggest that TADs represent fundamental features of chromatin organization [23 , 34] . Furthermore , two studies have shown that in some cases , genes lying within the same TAD are co-regulated [41 , 43] , although other studies have found no correlation between TADs and gene expression [59] . Therefore , we investigated whether CODs correspond to TADs . For this , we used two published datasets of breast cell lines corresponding to two different levels of resolution for TAD analysis: TADs of a median size of about 1 Mb were obtained from the mammary epithelial T47D cell line [43] , and high-resolution contact domains of a median size of about 185 kb were obtained from human mammary epithelial cells ( HMEC ) [37] . First , we calculated the coincidence between CODs and TADs . Only 7 . 8% and 2 . 6% of the CODs match ( coincidence in at least 80% of the length ) with TADs or with contact domains , respectively ( Fig 4A and 4B , upper panels ) . These values were similar ( 7 . 6% and 1 . 7% , respectively ) to the coincidence between randomized CODs of the same size and TADs or contact domains ( Fig 4A and 4B , lower panels ) . Likewise , no significant coincidence higher than random was observed when TAD and COD boundaries were compared ( Fig 4C ) . Examples of comparison of CODs with TADs and contact domains distributions are shown in Fig 4D and 4E . In summary , our data demonstrate that CODs do not correspond to TADs or contact domains . In order to clarify whether TADs are , to some extent , related to co-regulation of gene expression , we compared coexpression between gene pairs positioned in the same TAD ( intra-TAD ) with coexpression of gene pairs placed in different TADs ( inter-TADs ) ( Fig 4F ) . Intra-TAD coexpression was higher than inter-TAD coexpression . Similar results were obtained with randomized TADs of the same size , probably due to the fact that linked genes tend to be co-regulated [1 , 2 , 60 , 61] . However , in the chromosomes 8 , 12 , 13 , 14 , 15 , 17 , 19 , 22 , and X , TAD randomization significantly decreased coexpression with respect to real TADs , indicating that , at least in some chromosomes , gene pairs localized in the same TAD tend to be coexpressed more often than other randomly-selected pairs of close genes . In summary , our data suggest that while TADs do not correspond to the main level of organization of coexpression , they might play a role in gene co-regulation . After establishing that CODs do not correspond to TADs , we investigated the relationship between coexpression data of normal breast cells and physical chromatin contacts . For this , we used intra-chromosomal Hi-C data at 100 kb resolution from HMEC cells [37] . Since our coexpression matrix is gene-based , and the Hi-C matrices contain pairwise contact frequencies between 100 kb genomic segments , we first assigned connectivity values to each gene pair based on the Hi-C interactions between the corresponding segments in which the genes reside . The frequency of contacts is strongly influenced by sequence proximity and is not a good parameter to be correlated with coexpression of genes situated at very different distances . Lieberman-Aiden et al . defined the correlation interaction profile of a pair of loci as the correlation between distance-normalized contact frequencies of these two loci with the rest of the loci of the chromosome [35] . They assumed that if two loci are close in the 3D volume of the nucleus , they would have highly correlated interaction profiles . We thus compared interaction profiles with gene coexpression . Interestingly , we found that the similarity of interaction profiles of two genes increases with their coexpression in most of the chromosomes ( S10 Fig ) , suggesting that coexpressed genes display similar chromatin contacts . Positive and negative correlation of interaction profiles have been associated to two different compartments—termed A and B—that have different chromatin characteristics [35] , which have been recently subdivided into three [62] or five [37] compartments . A detailed analysis of the correlation plots shown in S10 Fig revealed the existence of two or three different linear behaviors in some chromosomes , depending on the range of interaction profile values . The sharp changes of correlation line slope around 0 and 0 . 2 ( see for example chromosomes 7 , 12 , 16 , 19 , and 20 ) suggest that the relationship between coexpression and chromatin contacts is different in different compartments . We then analyzed the relationship between interaction profiles and CODs . We found that the profile of contacts between gene pairs in the same COD was more similar than the profile of contacts for the rest of gene pairs ( Fig 5A ) . Randomization of CODs significantly decreased the similarity of the contact profiles in 2/3 of the genome ( 15 chromosomes ) being the effect not significant for chromosomes 4 , 5 , 10 , 12 , 13 , 17 , 18 , and 21 . These data suggest that in most of the cases , genes within the same COD have similar contact profiles . Next , we analyzed the relationship between the coexpression of genes from different CODs and the pattern of chromatin contacts . We found that the pattern of contacts between two genes placed in different coexpressed CODs ( with significant average inter-COD coexpression ) is significantly more similar than the contact profile of other pairs of genes . This effect decreased upon randomization of CODs ( and therefore , inter-CODs ) for all chromosomes except chromosome 10 and the low gene density chromosome 18 ( Fig 5B ) . These data indicate that two genes located in distant coexpressed CODs display similar physical contacts .
Clustering of co-expressed human genes was reported several years ago ( reviewed in [1 , 2] ) . Different strategies have been used since then to define these clusters , with the expression data from different tissue and from cancerous samples often pooled [16 , 18 , 19] . This pooling drastically reduces the number and size of statistically significant clusters [63] and creates large discrepancies between studies regarding the size and location of these clusters ( see for example [18] and [63] ) . Here we have used RNA-seq expression data from one single normal tissue ( breast ) or from one single type of tumor ( breast tumor ) to determine coexpression domains ( CODs ) by using a method based in TopDom [44] , a software initially designed for TAD identification . We found a very significant organization of the genome into CODs of about 0 . 9 Mb in size , with a median of about 10 genes per COD . In contrast to other studies [14 , 16 , 18 , 19] , we did not consider clusters of less than 4 genes , which rules out bidirectional promoters and deficient termination as possible mechanisms behind COD coexpression . In contrast to the RIDGE domains defined by Caron et al . [19] , COD genes are not expressed at higher level than non-CODs genes . Further , while CODs are enriched for housekeeping genes , most COD genes are not housekeeping , differentiating our CODs from the housekeeping gene clusters defined by Lercher et al . [18] . We compared normal breast tissue CODs with breast cancer CODs . As expected , due to the very different pattern of gene expression , the gene composition and the distribution of CODs changes drastically in breast cancer with respect to normal breast tissue , suggesting that CODs are tissue-specific . We also noticed that cancer CODs are strongly influenced by copy number amplification and deletions typically observed in cancer samples , which may also hinder COD identification when normal and cancer samples are mixed . Interestingly , we also found that COD genes tend to be coexpressed with other COD genes of the same chromosome , suggesting a long-range intra-chromosomal co-regulation of CODs . Which mechanisms are responsible for intra-COD coexpression ? One possibility is that genes inside CODs are regulated by cis-regulatory elements , such as enhancers , that affect most or all of the genes of the COD . Enhancers are typically placed in the vicinity of their regulated genes , although some enhancers may act at Mb-scale distances [64 , 65] . How many genes can be regulated by one enhancer , and how far away an enhancer can be to still function , are interesting questions that remain to be clarified [64] . Specific for our study is the question of whether about ten genes within a COD can be controlled by the same enhancer ( within the same COD ) . Two recent studies support a highly dynamic interaction between an enhancer and two different promoters [66 , 67] , suggesting the existence of multi-target enhancers . Much less clear is how to explain inter-CODs coexpression . In other words—how can different CODs separated by dozens of Mb be coordinately regulated ? One possibility is that a number of enhancers of different regions ( different CODs ) cooperate to coordinately control the expression of distant CODs of the genome , as 3D super-enhancers . In fact , spatial enhancer clustering has been reported [68] . This coordinated enhancer function may result in a transient and local high concentration of RNA polymerases and associated factors around a region of the genome , which would coordinately regulate transcription of large sets of genes from different CODs . This hypothesis is consistent with the transcription factories concept developed by Cook and others [69–71] . Interestingly , “specialized” factories in which tissue-specific and/or pathway-specific co-regulated genes are transcribed have been reported , supporting a role of factories in coexpression [33 , 72] . The mechanisms previously discussed imply the existence of short- and long-range contacts between different regions of the chromatin . Hi-C studies have shown that chromatin is segmented into self-interacting domains called topological-associated domains ( TADs ) , with median sizes ranging from 185 kb to one Mb , depending on the resolution of the study [37 , 40–42] . It is accepted that enhancer activity is limited to the genes that fall within the same TAD [64] , and contacts between different TADs have been reported [37] . Therefore , one logical initial possibility was that CODs correspond to TADs . However , our results indicate that CODs do not match with TADs in either normal or cancer breast cell genomes . This is in agreement with the fact that most TADs are invariant between cell types or between cancer and normal tissue , in clear contrast to CODs . To what extend genes inside TADs are co-regulated is unclear . Two studies have shown that under particular conditions ( progesterone treatment and genes of X-chromosome involved in differentiation of ESC into epiblast ) genes lying within the same TAD are significantly co-regulated [41 , 43] . Neems et al . , have recently reported TADs enriched for muscle-specific genes [15] . However , most myogenesis-specific genes ( 87 . 3% ) were located in TADs non-enriched for myogenesis-specific genes . In fact , in many cases genes in the same TAD display different expression patterns [64] . For example , Barutcu et al . did not find gene expression changes in the genes contained in the small number of TADs that change between the non-tumorigenic mammary epithelial MCF-10A and the breast cancer MCF-7 cell lines [59] . Correspondingly , we did not find intra-TAD coexpression higher than random in most of the chromosomes . However , a small , but higher than random , intra-TAD coexpression was observed for chromosomes 8 , 12 , 13 , 14 , 15 , 17 , 19 , 22 , and X . Definition of TADs is based on the frequency of chromatin contacts [40] . Consistent with the absence of coincidence between CODs and TADs , genes of the same COD did not display a higher frequency of contacts among themselves than with genes of different CODs . However , we found that genes in the same COD displayed a similar ( correlated ) profile of contacts in most of the chromosomes . Interestingly , a similar profile of contacts was also found between genes of different but co-regulated CODs . These data are again consistent with the transcription factories model , in which different co-regulated genes can interact with one or a few factories without necessarily having to have interactions between themselves . A recent study using in situ hybridization data from 4 , 345 genes of the mouse brain reached similar conclusions about the correlation between coexpression and connectivity data [73] . However , that study did not analyze coexpression clusters , maybe due to the small coverage of the genome obtained ( with only 4 , 345 genes ) . The fact that two loci display a correlated profile of contacts suggests that they are close in space [35] . Indeed , intra-chromosomal coexpression is much higher than inter-chromosomal coexpression [74] , which is also consistent with the existence of chromosomal territories [25–27] . Nonetheless , coexpression of genes from different chromosomes also occurs . Whether CODs of different chromosomes can be coexpressed , and whether this also implies that similar inter-chromosomes chromatin contacts occur , remains to be investigated . Cancer samples showed an overall lower coexpression than normal samples . Since average intra-CODs coexpression was very similar between normal and cancer samples most reduction come from the inter-COD coexpression , suggesting that cancer CODs have , at least partially , lost the capacity of co-regulation with other CODs . Interestingly , the number of transcription factories range from 200 in a primary cells [75] to 2000 in HeLa cells [76] . Although preliminary , due to the small number of cell types analyzed , these data suggest a higher number of factories in cancer cells than in primary cells . If factories are hubs for coexpression , a high number of factories implies less coexpression , which would fit with the lower capacity of coordination between CODs that we observed in cancer cells . Our data are also in agreement with data from Barutcu et al . , who have recently reported a decrease in the frequency of inter-chromosomal interactions between small chromosomes , and intra-chromosomal interactions particularly in telomeric and subtelomeric regions of the genome , in breast cancer cells with respect to epithelial non-transformed cells [59] . In summary , our data support that genes are organized into highly coexpressed regions—CODs—that have similar profiles of physical interactions . CODs can also be coexpressed with other CODs , and these also have similar profiles of chromatin contacts . It is tempting to speculate that common physical contacts are the mechanism that determine coexpression . However , whether physical contacts are the cause or consequence of co-regulation requires further investigation .
Gene expression ( RSEM normalized RNA-seq V2 data ) of the 20 , 502 genes available ( based on hg19 UCSC Gene standard track ( December 2009 version ) from 100 normal breast tissue samples and 369 breast tumor samples were collected from TCGA ( https://cancergenome . nih . gov/ ) . GDC manifest files for identification of normal and cancer samples used are provided in S1 Appendix and S2 Appendix , respectively . First , twenty-three different expression vectors–one for each chromosome–were constructed , and genes ( gene i ( gi ) ) were sorted according to their 5′ to 3′ chromosomal order , using the assembly hg19 of the human genome . With 1≤ i , j ≤ 20502 With 1≤n≤23 , where n is the number of the chromosome . Then , twenty-three correlation matrices , C , containing the Pearson correlation coefficient between the expression profiles of every pair of genes ( gene i ( gi ) , gene j ( gj ) ) was constructed in R ( http://www . rproject . org ) for each chromosome of each set of data ( normal breast and breast tumor ) , using the cor function of the stats package . With 1≤n≤23 , where n is the number of the chromosome . Coexpression matrices heat maps were visualized using Gitools 2 . 3 . 1 version [77] . In all heat maps genes are arranged in the chromosomal order . Centromere coordinates were obtained from the USCS Genome Browser through the Table Browser ( http://genome . ucsc . edu/cgi-bin/hgTables ) . We designed a method to determine coexpression domains ( CODs ) based on methods used to identify TADs , such as the directionality index [40] and the TopDom methods [44] . The input data are the coexpression matrices of each chromosome , where each position contains Pearson coefficient values between any two genes . For each gene , we computed an average coexpression signal between upstream and downstream regions around its position as previously defined in the TopDom method . binsignal ( i ) =1/w2∑l=1w∑m=1wcoexpressionvalue ( Ui ( l ) , Di ( m ) ) where Ui = {i-w+1 , …i-1 , i} , Di = {i+1 , i+2 , . . i+w} , and w is the size of the window around i . As shown in Fig 1C and 1D , this parameter is high for genes around the center of the CODs and decreases at COD boundaries and at chromosomal regions between CODs . We defined CODs as regions of binsignal ( i ) ≥ 0 . 15 delimited by 5′ and 3′ significant boundaries . We selected 0 . 15 because it is the average binsignal ( i ) of the genome . Boundaries are defined as regions larger than three genes with binsignal ( i ) < 0 . 15 that delimit regions with significantly different binsignal ( i ) ( p < 0 . 05 ) . We determine boundaries to be statistical significance by computing a Student´s t-test between the four upstream ( i-w+1 , … i-2 , i-1 , i ) , and the four downstream ( i+1 , i+2 , … , i+w ) binsignal ( i ) values for each gene i . Less than two consecutive genes with binsignal ( i ) < 0 . 15 inside a COD are allowed if these segments are not boundaries . A script for CODs determination ( CODfinder ) was written in R and deposited in the GitHub repository ( https://github . com/joseguem/CODfinder . git ) . To determine the best gene window size w , we run the program using different w values for chromosome 1 and calculated the average intra-COD coexpression of the CODs identified . Similar average intra-COD coexpressions were obtained with w between 3 and 6 genes , indicating the robustness of the system of identification of CODs ( S11A Fig ) . We selected w = 4 for the rest of the study because is the window for which average intra-COD coexpressions was maximum . S11B and S11C Fig show variation in the number and size of CODs depending on w value . For randomization of gene order along chromosome 1 in S3 Fig , gene positions were shuffled using the sample function available in R base package . Genomic coordinates of CODs , according to human genome assembly hg19 , were specified by using the first nucleotide of the first COD gene and the last nucleotide of the last COD gene , irrespective of gene orientation . These coordinates were used to determines COD length in bp . To compare expression of genes in CODs with those outside of CODs , the average of RSEM normalized data of expression of every gene in the 100 normal breast samples was computed . COD genes were compared to a list of housekeeping genes obtained from https://www . tau . ac . il/~elieis/HKG/ [45] . Venn diagrams were constructed using the Venn Diagram Generator ( http://www . pangloss . com/seidel/Protocols/venn . cgi ) . The hypergeometric tests were performed in R using the dhyper function from the stats package . For box plots of Fig 2A , we called intra-CODs coexpressions to the Pearson coexpression coefficients of all pairwise combinations between the genes inside the same COD . We called inter-CODs coexpressions to the Pearson coexpression coefficients of all pairwise combinations between the genes from two different CODs within the same chromosome . Pearson coexpression coefficients of all pairwise combinations between the genes that are not in CODs or between a COD gene and a non-COD gene were considered the rest of the intra-chromosomal coexpressions . These data were extracted from each chromosomal coexpression matrix ( Cnormal ( i , j ) _chr_n or Ctumo ( i , j ) _chr_n ) . Gene ontology functional categories and pathway enrichment were analyzed using WebGestalt software packages ( http://www . webgestalt . org/ ) [78] . Bonferroni-adjusted p-values of the hypergeometric tests were used to determine enrichment significance . For comparison of normal and cancer CODs , coincidence of at least 80% of normal COD genes was required . For determination of CODs boundaries coincidence between normal and cancer samples a discrepancy of 10% of the number of normal CODs genes was allowed . Gene copy number data ( relative linear copy number from Affymetrix SNP6 ) corresponding to the 369 analyzed breast tumor samples were obtained from TCGA through cBioPortal ( http://www . cbioportal . org/ ) [79] . The gene copy number profiles in S8 Fig correspond to the average gene copy number values plotted according to the chromosomal gene order . http://dgd . genouest . org/ . Clinicopathological data ( N stage and survival ) of patients in the breast tumors cohorts were obtained from TGCA . Cancer population was subdivided into two according to N stage . The N grade indicates whether lymph nodes have metastasis N1 , N2 , N3 ) or not ( N0 ) , respectively . No subdivisions of N stages were used ( e . g . , N1a , N1b , and N3 were considered as T1-3 ) . Kaplan-Meier survival plots were constructed using Prism 5 ( GraphPad ) . Significance of the difference between groups was comutes using the Log-rank test . We used our average gene expression data from 100 normal breast samples to determine RIDGEs , according to the method described by Caron et al . [19] with some modifications . For this , genes were ordered according to the chromosomal order . Then , for each gene i the median expression of a moving window of 39 genes ( 19 upstream and 19 downstream of i ) was calculated . We found 45 regions with 10 or more consecutive moving medians higher than twice the genomic median . These regions were considered RIDGEs . Genes and coordinates of identified RIDGEs are listed in S2 Table . Genomic coordinates , according to human genome assembly hg19 , of Topologically Associating Domains ( TADs ) from the cell line T47D [43] were provided by M . Marti-Renom ( CRG , Barcelona ) . Genomic coordinates of contact domains from HMEC cells [37] were collected from the Gene Expression Omnibus ( GEO ) at NCBI ( accession number GSE63525 ) . CODs and TADs with at least 80% coincidence of the length were considered as coincident . To determine COD-TAD coincidence , 5′ and 3′ boundary coordinates of every COD of each chromosome were compared with the 5′ and 3′ boundary coordinates of every TAD of the same chromosome . To estimate boundary overlap , every COD boundary was compared with every TAD boundary ( irrespectively of the 5′ or 3′ position ) , with a discrepancy of less than 10% of the COD length allowed . Contact domains were calculated similarly . In order to investigate whether gene pairs residing within the same TAD had higher coexpression than pairs of genes residing in different TADs or in non-TADs , genes were first assigned to TADs . For that genomic coordinates of genes obtained from the UCSC Genome Browser through the Table Browser ( http://genome . ucsc . edu/cgi-bin/hgTables ) . Genes were then assigned to TADs of the corresponding chromosomes according to their midpoint coordinate . After that , pairwise coexpression of genes ( Pearson coefficient correlation ) within the same TAD was compared with that of genes located in different TADs , and the significance of the differences was estimated using the Bonferroni-corrected Mann-Whitney test . To investigate whether the significantly higher coexpression of genes placed in the same TAD is a consequence of spatial proximity , TAD positions were randomized but keeping exactly the same TAD size . For randomization , TAD genomic coordinates were inverted 3′ to 5′ in each chromosome . Similar results were obtained when TAD borders were shifted 100 , 000 bp upstream or downstream . Intra-chromosomal Hi-C data at 100 kb resolution of HMEC cells were collected from GEO at NCBI ( accession number GSE63525 ) [37] . Contact matrices of each chromosome were normalized using KR-Normalization [80 , 81] as described [37] . The O/E ( “observed over expected” ) matrices , which correct for the increased number of contacts due to sequence proximity , and the Pearson correlation matrices of the O/E , which identifies spatial relationships between loci by looking for correlations in their contact patterns , were constructed as described [35] . Next , genes were assigned to the 100 kb anchors , and connectivity data were assigned to pairs of genes based on the corresponding anchor pairs in which the genes reside . Vectors were then constructed that assign to every possible pair of genes of a chromosome their corresponding coexpression ( Pearson correlation between expressions ) and connectivity ( normalized contacts , or Pearson correlation matrices of the O/E ) . For S10 Fig , connectivity data ( Pearson correlation matrices of the O/E ) were ranked and grouped into 20 bins with the same number of elements . The average value of connectivity of each bin was then represented against the average coexpression of the corresponding pairs of genes . For Fig 5 , box plots of connectivity values ( Pearson correlation matrices of the O/E ) of pairs of genes that reside in the same COD ( intra-COD ) or in different CODs ( inter-CODs ) were compared with the same parameter as the rest of gene pairs of each chromosome . To estimate the random values in each chromosome , COD borders were randomized as described below . Robust randomization of the COD borders was performed by inverting the chromosomal coordinates from 5′–3′ to 3′–5′ on each chromosome , similarly as described in [43] . This method allows the COD size , the distance between CODs , the genomic context , and the gene proximity to be maintained while changing the gene composition of the CODs . Similar results were obtained when randomization was performed by shifting the COD borders 100 , 000 bp upstream or downstream . Student´s t-test and Mann-Whitney test with confidence interval 95% were computed in R using t . test and wilcox . test functions from stats package . To test significance of overlapping in Venn diagrams , the hypergeometric tests were performed in R , using the dhyper function from the stats package . Genomic coordinates of contact domains and Intra-chromosomal Hi-C data at 100 kb resolution , from HMEC cells [37] were collected from the Gene Expression Omnibus ( GEO ) at NCBI ( accession number GSE63525 ) . | Prokaryotic operons normally comprise functionally related genes whose expression is coordinated . Even though operons do not exist in most eukaryotes , results from the last fifteen years indicate that gene order is nonetheless not random in eukaryotes , and that coexpressed genes tend to be grouped in the genome . We identify here about 500 coexpression domain ( CODs ) in normal breast tissue . Interestingly , we find that genes within CODs often are coexpressed with other genes that reside in other CODs placed very far away in the same chromosome , which is indicative of long-range inter-COD co-regulation . Furthermore , we find that coexpressed genes within CODs or within co-regulated CODs display similar three-dimensional chromatin contacts , suggesting a spatial coordination of CODs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"cancer",
"genomics",
"medicine",
"and",
"health",
"sciences",
"breast",
"tumors",
"reproductive",
"system",
"cancers",
"and",
"neoplasms",
"basic",
"cancer",
"research",
"human",
"genomics",
"oncology",
"chromosome",
"mapping",
"molecular",
"biology",
"techniques",
"epigenetics",
"chromatin",
"research",
"and",
"analysis",
"methods",
"gene",
"mapping",
"chromosome",
"biology",
"gene",
"expression",
"breast",
"cancer",
"molecular",
"biology",
"breast",
"tissue",
"cell",
"biology",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"genomic",
"medicine"
]
| 2017 | Analysis of the relationship between coexpression domains and chromatin 3D organization |
Intellectual disability and seizures are frequently associated with hypomagnesemia and have an important genetic component . However , to find the genetic origin of intellectual disability and seizures often remains challenging because of considerable genetic heterogeneity and clinical variability . In this study , we have identified new mutations in CNNM2 in five families suffering from mental retardation , seizures , and hypomagnesemia . For the first time , a recessive mode of inheritance of CNNM2 mutations was observed . Importantly , patients with recessive CNNM2 mutations suffer from brain malformations and severe intellectual disability . Additionally , three patients with moderate mental disability were shown to carry de novo heterozygous missense mutations in the CNNM2 gene . To elucidate the physiological role of CNNM2 and explain the pathomechanisms of disease , we studied CNNM2 function combining in vitro activity assays and the zebrafish knockdown model system . Using stable Mg2+ isotopes , we demonstrated that CNNM2 increases cellular Mg2+ uptake in HEK293 cells and that this process occurs through regulation of the Mg2+-permeable cation channel TRPM7 . In contrast , cells expressing mutated CNNM2 proteins did not show increased Mg2+ uptake . Knockdown of cnnm2 isoforms in zebrafish resulted in disturbed brain development including neurodevelopmental impairments such as increased embryonic spontaneous contractions and weak touch-evoked escape behaviour , and reduced body Mg content , indicative of impaired renal Mg2+ absorption . These phenotypes were rescued by injection of mammalian wild-type Cnnm2 cRNA , whereas mammalian mutant Cnnm2 cRNA did not improve the zebrafish knockdown phenotypes . We therefore concluded that CNNM2 is fundamental for brain development , neurological functioning and Mg2+ homeostasis . By establishing the loss-of-function zebrafish model for CNNM2 genetic disease , we provide a unique system for testing therapeutic drugs targeting CNNM2 and for monitoring their effects on the brain and kidney phenotype .
Brain defects including seizures , migraine , depression and intellectual disability are frequently associated with hypomagnesemia [1] . Indeed , low Mg2+ concentrations may cause epileptiform activity during development [2] . Specifically , the Mg2+ channel transient receptor potential melastatin 7 ( TRPM7 ) is essential for brain function and development [3] . Interestingly , patients with genetic defects in TRPM6 , a close homologue of TRPM7 , may have neurological complications [4] . Although TRPM6 and TRPM7 share similar Mg2+ transporting properties , they are differentially expressed and regulated [5] . TRPM7 is a ubiquitously expressed protein regulating intracellular Mg2+ levels in a broad range of cells , whereas TRPM6 is localized in the luminal membrane of renal and intestinal epithelia involved in Mg2+ absorption [1] , [6]–[8] . Recently , we have identified mutations in the gene encoding cyclin M2 ( CNNM2 ) in two unrelated families with dominant isolated hypomagnesemia ( CNNM2 [MIM 607803] ) [9] . Patients suffered from symptoms associated with low serum Mg2+ levels ( 0 . 3–0 . 5 mM ) such as tremors , headaches and muscle weakness . The role of CNNM2 in the kidney for the maintenance of serum Mg2+ levels can be traced to the distal convoluted tubule ( DCT ) , where also TRPM6 is expressed . Here , CNNM2 is present in the basolateral membrane of DCT cells and its expression is regulated by dietary Mg2+ availability [9]–[10] . Although CNNM2 has been proposed as a Mg2+ transporter in overexpression studies in Xenopus oocytes [11] , Mg2+ transport could not be directly measured in mammalian cells using patch clamp analysis [9] . On the other hand , modelling of the CNNM2 cystathionine β-synthase ( CBS ) domain resulted in the identification of a Mg2+-ATP binding site , suggesting a role in Mg2+ sensing within the cell [12] . Consequently , the molecular mechanism explaining the role of CNNM2 in DCT-mediated Mg2+ transport remains to be elucidated . The CNNM2 gene is ubiquitously expressed in mammalian tissues , most prominently in kidney , brain and lung [12]–[13] . Although the role of CNNM2 beyond the kidney has never been studied , genome wide association studies have related the CNNM2 locus to blood pressure , coronary artery disease and schizophrenia , suggesting an important role of CNNM2 in the cardiovascular system and brain [14]–[15] . CNNM2 is widely conserved among species . In zebrafish ( Danio rerio ) , a frequently used model for ion homeostasis and human genetic diseases in general [16]–[17] , the cnnm2 gene is duplicated and two paralogues , cnnm2a and cnnm2b , are described [18] . Both paralogues share a high conservation with human CNNM2 ( 79% amino acid identity ) . In detail , transcripts are abundantly expressed in zebrafish brain and in ionoregulatory organs such as kidney and gills , which act as a pseudokidney in fish [18] . Consistent with the regulation of CNNM2 transcripts in mammals [11] , the expression of cnnm2a and cnnm2b is regulated by Mg2+ in vivo [18] . In the present study , we aim to elucidate the function of CNNM2 in brain and kidney . Hence , we can demonstrate the genetic origin of symptoms in five unrelated families suffering from a distinct phenotype of mental retardation , seizures and hypomagnesemia , where we have identified novel mutations in the CNNM2 gene . By combining functional analyses and a loss-of-function approach in the zebrafish model , we provide functional evidence for a key role of CNNM2 in brain development , neurological activity and renal Mg2+ handling .
Patients F1 . 1 and F1 . 2 presented in the neonatal period with cerebral convulsions . Serum Mg2+ levels at manifestation were found to be 0 . 5 mM in both patients ( Table 1 ) . Convulsions were refractory to conventional antiepileptic medications . Intravenous Mg2+ supplementation with ∼1 mmol/kg body weight/day was initiated after oral Mg2+ failed to correct serum Mg2+ levels . However , seizure activity continued even in face of normomagnesemia . An extensive analysis for infectious causes or inborn errors of metabolism did not yield any positive results . Ultrasound examination of the kidneys did not reveal nephrocalcinosis , whereas basal ganglia calcifications were noted in early central nervous system ( CNS ) sonographies . During follow-up , severe developmental delay was noted accompanied by microcephaly ( head circumference below third percentile for age and sex in both patients ) . A magnetic resonance imaging ( MRI ) at 5 . 5 years of age in patient F1 . 1 showed wide supratentorial outer cerebrospinal liquor spaces with failure of opercularization together with a significantly reduced myelinization of the white matter tract ( Figure 1C–D ) . The severe degree of intellectual disability , which became apparent with increasing age comprised major deficits in cognitive function , the inability to verbally communicate , and severely limited motor skills . Both children are not able to perform main activities of daily living and require full-time care by an attendant . Seizure activity is sufficiently controlled in the older brother by valproate and lamotrigine , electroencephalography ( EEGs ) merely shows generalized slowing , but no epileptic activity . In contrast , the younger sister suffers from ongoing generalized , myoclonic seizures despite antiepileptic treatment with valproate and levetiracetame . Laboratory investigations during follow-up demonstrated persistent hypomagnesemia of ∼0 . 6 mM despite oral Mg2+ supplementation . Patients F2 . 1 , F3 . 1 , and F4 . 1 presented with seizures during infancy ( between 4 and 12 months of age ) . Laboratory evaluation yielded isolated hypomagnesemia of ∼0 . 5 mM ( Table 1 ) . Urine analyses demonstrated inappropriate fractional excretion for Mg2+ in face of persistent hypomagnesemia . In addition , the renal Mg2+ leak was verified by Mg2+ loading tests in patients F2 . 1 and F4 . 1 as described before [19] . Urinary calcium excretion rates were normal , renal ultrasound excluded the presence of nephrocalcinosis . After acute therapy with intravenous Mg2+ , the patients received a continuous oral Mg2+ supplementation of 0 . 5 to 1 mmol/kg body weight/day of elemental Mg2+ . This oral therapy however failed to correct the hypomagnesemia , serum Mg2+ remained in the subnormal range in all three children . Because of recurrent cerebral seizures , patients F2 . 1 to F4 . 1 received diverse antiepileptic medications . Currently only patient F4 . 1 is still treated with clobazam . In all three patients ( F2 . 1 to F4 . 1 ) , a significant degree of intellectual disability was already noted in early childhood with delayed speech development , but also impaired motor as well as cognitive skills . In addition , patient F4 . 1 was noted to exhibit disturbed social interaction , abnormal verbal and non-verbal communication , as well as stereotyped behaviour and finally received the formal diagnosis of early onset autism . All three patients F2 . 1 to F4 . 1 were not able to attend regular schools . Standardized intelligence testing in patients F2 . 1 and F3 . 1 revealed a significant degree of mental retardation ( see Table 1 ) . While patients F2 . 1 and F3 . 1 are currently living with their parents , patient F4 . 1 is placed in a home for children with mental illness because of episodes of violence and destructive behaviour . The parents of all three children ( F2 . 1–F4 . 1 ) had normal serum Mg2+ levels and no signs of intellectual disability . Finally , patient F5 . 1 presented with muscle spasms and dysesthesia in adolescence . Serum Mg2+ levels were found to be low ( ∼0 . 6 mM ) . Because of concomitant borderline hypokalemia , she was suspected to have Gitelman syndrome ( [MIM 263800] ) and received oral Mg2+ and K+ supplements . Also this patient exhibited a mild degree of intellectual disability . Unfortunately , she was not available for further examination . Common genetic causes of mental retardation were excluded in patients F1 . 1 and F2 . 1 by array CGH ( comparative genomic hybridization ) . The presence of two affected siblings together with the suspected parental consanguinity in family F1 suggested an autosomal-recessive pattern of inheritance . Therefore , we subjected patients F1 . 1 and F1 . 2 to homozygosity mapping which , at a cut-off size of >1 . 7 megabases ( Mb ) , yielded eleven critical intervals on autosomes with a cumulative size of 62 Mb . The gene list generated from these loci included 322 RefSeq genes and putative transcripts . CNNM2 in a critical interval of 7 . 1 Mb on chromosome 10 emerged as the most promising candidate gene because of its known role in Mg2+ metabolism [9] , [11]–[12] , [20] . Conventional Sanger sequencing of the complete coding region of the CNNM2 gene revealed a homozygous mutation , c . 364G>A , leading to a non-conservative amino acid substitution of glutamate to lysine at position 122 of the CNNM2 protein ( p . Glu122Lys , Figure 1A–B ) . The mutation was present in heterozygous state in both parents . After discovery of this homozygous mutation in patients F1 . 1 and F1 . 2 , a larger cohort ( n = 34 ) of patients with Mg2+ deficiency of unknown origin was screened for mutations in the CNNM2 gene . Mutations in heterozygous state were discovered in patients F2 . 1 to F5 . 1 ( Table 1 ) . However , sequencing of the complete coding region and adjacent exon-intron boundaries did not reveal a second pathogenic allele . Next , we examined the CNNM2 gene in parents and unaffected siblings of families F2 to F4 . The mutations previously identified in our patients were not detected in either of the parents pointing to de novo mutational events . Interestingly , patients F2 . 1 and F4 . 1 exhibited the same mutation , p . Glu357Lys ( c . 1069G>A ) , affecting a highly conserved amino acid residue in the 2nd membrane-spanning domain of the CNNM2 protein . Also the p . Ser269Trp ( c . 806C>G ) mutation detected in patient F3 . 1 affects a highly conserved residue located in the 1st transmembrane domain . All three mutations were ranked “probably damaging” by Polyphen-2 when tested for functional consequences of the mutations ( p . Glu122Lys , p . Ser269Trp and p . Glu357Lys with scores of 0 . 981 , 1 . 000 and 1 . 000 , respectively ) . Finally , in patient F5 . 1 with a late manifestation and putatively milder phenotype , the variant p . Leu330Phe ( c . 988C>T ) was identified in heterozygous state . This variant affects an amino acid residue conserved among mammals , however a phenylalanine appears at this position in certain fish species . The variant is predicted to be possibly damaging by Polyphen-2 with a score of 0 . 711 . None of the identified variants was detected in 204 controls or present in publically available exome data . To clarify the function of CNNM2 , Human Embryonic Kidney ( HEK293 ) cells were transiently transfected with mouse Cnnm2 or mock constructs and examined for Mg2+ transport capacity using the stable 25Mg2+ isotope . At baseline , approximately 10% of the total intracellular Mg2+ content consists of 25Mg2+ , which is the natural abundance of 25Mg2+ [21] . By incubating the cells in a physiological buffer containing pure 25Mg2+ , the intracellular 25Mg2+ concentration increases over time . Interestingly , Cnnm2 expressing cells displayed a higher 25Mg2+ uptake compared to mock cells ( Figure 2A ) . After 5 minutes , Cnnm2-expressing cells had approximately 2 times more 25Mg2+ uptake than mock-transfected cells ( Figure 2B ) . All further experiments were performed at the 5 minutes time point to cover the exponential phase of the uptake . To reduce the background in 25Mg2+ uptake , inhibitors of known Mg2+ channels and transporters were added during the uptake process; 2-APB to inhibit TRPM7 [22] , Ouabain to block the Na+-K+-ATPase [23] , Quinidin for SLC41A1 [24] and Nitrendipin [25] for silencing MagT1 activity ( Figure 2C ) . Only 2-APB was capable of significantly inhibiting 25Mg2+ uptake in HEK293 cells . Moreover , 2-APB inhibition also abolished the CNNM2-dependent increase in 25Mg2+ uptake . Dose-response experiments confirmed that the IC50 of 2-APB inhibition is 22 µM ( Figure 2D ) . CNNM2-dependent 25Mg2+ uptake was found to be independent of Na+ and Cl− availability , when uptakes were performed in N-methyl-d-glucamine ( NMDG ) or Gluconate buffers ( Figure 2E ) . Interestingly , the highest CNNM2-dependent 25Mg2+ uptake was measured between 1–2 mM , suggesting a Km in the physiological range of approximately 0 . 5 mM ( Figure 2F ) . At high Mg2+ concentrations ( 5 mM ) , 25Mg2+ uptake was inhibited . When subjected to 24 hours 25Mg2+ loading , Cnnm2-expressing cells showed a significantly higher 25Mg2+ content baseline . Subsequently , 15 minutes extrusion of Cnnm2-expressing cells demonstrated no difference in Mg2+ extrusion rate , compared to mock-transfected cells ( Figure 2G ) . To characterize the effect of the CNNM2 mutations identified in our hypomagnesemic patients , 25Mg2+ uptake was determined in HEK293 cells expressing mutant CNNM2 proteins . Of all missense mutations that are identified to date , only p . Leu330Phe was capable of increasing 25Mg2+ uptake to a similar extent as wild-type CNNM2 ( Figure 3A ) . All other CNNM2 mutants exhibited severely decreased 25Mg2+ uptake or had lost their ability to increase 25Mg2+ uptake completely . To examine whether CNNM2 dysfunction can be explained by a reduced plasma membrane expression , all mutants were subjected to cell surface biotinylation analysis . Indeed , p . Glu122Lys CNNM2 membrane expression was significantly reduced compared with wild-type CNNM2 ( 66% decrease , P<0 . 05 ) and p . Ser269Trp CNNM2 showed a trend towards reduction ( 46% decrease , Figure 3B ) . Patients with mutations in CNNM2 suffer from hypomagnesemia . Therefore , zebrafish cnnm2 morphants were tested for disruptions of their Mg2+ homeostasis . Extraction of serum from zebrafish embryos is technically not feasible . Thus , total body Mg contents of controls and morphant larvae were examined at 5 days post-fertilization ( dpf ) . During these 5 days of zebrafish development , intestinal absorption of Mg2+ does not take place since larvae do not eat and drink . For that reason , Mg2+ homeostasis is the result of the balance between Mg2+ excretion , passive Mg2+ uptake from the yolk , Mg2+ reabsorption in the kidney , and Mg2+ uptake in the integument , where ionocytes are analogous to renal tubular cells in terms of function and transporter and channel expression [16] . Therefore , when knocking down a gene involved in active epithelial Mg2+ uptake , disturbances in total body Mg content reliably represent disturbances in active Mg2+ reabsorption and/or uptake , through pronephric ( renal ) tubular cells and/or their analogous in the skin , respectively . The cnnm2a gene , one of the two zebrafish cnnm2 paralogues , is expressed during early development ( Figure 4A ) . Injection in embryos of higher doses than 2 ng of morpholino ( MO ) blocking cnnm2a translation resulted in a significantly reduced survival compared to controls at 5 dpf ( Figure 4B ) . At non-lethal doses of MO ( when mortality caused by the cnnm2a-MO does not differ significantly from mortality in controls ) , knockdown of cnnm2a resulted in morphological phenotypes characterized by enlarged pericardial cavities and notochord defects ( Figure 4C–D ) . The biochemical equivalency between mammalian CNNM2 and its zebrafish orthologue cnnm2a was demonstrated by the fact that co-injection of cnnm2a-MO with mouse wild-type Cnnm2 cRNA induced a rescue of all phenotypes observed ( Fig . 4E ) . Conversely , co-injection with mouse mutant Cnnm2 cRNA did not result in any rescue . In line with the symptom of hypomagnesemia in patients with mutated CNNM2 , cnnm2a morphants exhibited significantly reduced levels of Mg compared to controls when increasing doses of MO were injected ( Figure 4F ) . Total Mg content in cnnm2a morphants was restored to control levels when cnnm2a-MO was co-injected with cRNA encoding for mouse wild-type CNNM2 and not with mutant Cnnm2 ( Figure 4G ) . This demonstrated that the defects observed in morphants were indeed caused by dysfunctional cnnm2a and not by toxic off-target effects . Zebrafish cnnm2b is also expressed during early development ( Figure 5A ) . Survival in cnnm2b morphants was not affected by the knockdown ( Figure 5B ) . The cnnm2b morphants were characterized by enlarged pericardial cavities , kidney cysts and , in agreement with the morphological brain abnormalities observed in the F1 . 1 patient , by accumulation of cerebrospinal fluid in the cerebrum ( Figure 5C–D ) . Interestingly , most cnnm2b morphants were morphologically normal at the dose of 2 ng MO/embryo ( Figure 5D ) . All morphological phenotypes were rescued by co-injection of cnnm2b-MO with mouse wild-type Cnnm2 ( Figure 5E ) . As for cnnm2a , cnnm2b was demonstrated to reduce Mg levels when knocked down ( Figure 5F ) and to be functionally equivalent to mammalian CNNM2 in cRNA rescue experiments ( Figure 5G ) . Phenotype rescue with mouse wild-type Cnnm2 demonstrated the absence of toxic off-target effects and the specificity of the cnnm2b-MO to produce defects attributable to impaired CNNM2 function . Patients with CNNM2 mutations suffer from mental retardation and seizures . As the severe neurological phenotype in patients F1 . 1 and F1 . 2 was diagnosed early after birth ( Table 1 ) , we hypothesized that the deleterious effects of mutant CNNM2 could result from early developmental defects in brain primordia . In zebrafish , the segmental organization of the brain rudiment , and morphologically visible boundaries and primordia are established at 25 hours post-fertilization ( hpf ) . At this stage , maldevelopment of the midbrain hindbrain boundary ( MHB ) is observed in cnnm2a morphant embryos ( Figure 6A–B ) . Interestingly , these phenotypes could not be rescued by exposure to media with high Mg2+ concentrations ( Figure 6B ) , even though these media significantly increased the Mg content of morphant embryos ( Figure 6C ) . More importantly , phenotypes were rescued by co-injection with the mouse orthologue cRNA and not by co-injection with the mutant transcript ( Figure 6D ) . In addition to brain developmental defects , the frequency of spontaneous embryonic contractions was increased in cnnm2a morphants compared to controls ( Figure 6E ) , which could indicate that ( motor ) neurons are hyperexcitable [26] . This phenotype was not rescued by exposure to high Mg2+ concentrations in the medium ( Figure 6E ) . In contrast , co-injection of the cnnm2a-MO with mouse wild-type Cnnm2 cRNA did result in a rescue of the neurological functioning ( Figure 6F ) . Conspicuously , co-injection with the mutant Cnnm2 cRNA even worsened this motor neuronal phenotype by increasing the number of spontaneous contractions significantly compared to embryos injected only with cnnm2a-MO ( Figure 6F; Movies S1 , S2 , S3 ) . In the case of cnnm2b morphants , enlarged tectums were also present in a 30% of morphants in addition to the defects in the MHB , phenotypes that were not rescued by exposure to high Mg2+ concentrations ( Figure 7A–C ) but by co-injection of cnnm2b-MO with mouse wild-type Cnnm2 cRNA ( Figure 7D ) . Spontaneous contraction frequency was increased in cnnm2b morphants ( Figure 7E ) , restored to control levels with overexpression of mouse wild-type Cnnm2 ( Figure 7F ) , and 4-fold increased with overexpression of mouse mutant Cnnm2 compared to cnnm2b morphants injected solely with cnnm2b-MO ( Figure 7F; Movies S4 , S5 , S6 ) . The validated cRNA rescue controls proved the specificity of the brain defects observed , attributable to dysfunctional orthologues of CNNM2 for both translation blocking MOs used . As our in vitro data pointed to a putative interaction between CNNM2 and TRPM7 and zebrafish morphants presented brain developmental defects , the touch-evoked escape behaviour in zebrafish was evaluated , a parameter largely dependent on TRPM7 activity in sensory neurons and/or brain development [27]–[29] . Indeed , in cnnm2a and cnnm2b morphants ( at 5 dpf ) , touch-evoked escape behaviour was significantly weaker than that in controls ( Figures S1 , S2 ) . Additionally , this phenotype was rescued by co-injection of the MO with wild-type Cnnm2 cRNA and not by mutant Cnnm2 cRNA . As for the other phenotypes , cRNA rescues proved the causality between the weak touch-evoked escape behaviour in morphants and dysfunctional cnnm2 paralogues .
In the present study , a severe brain phenotype consisting of cerebral seizures , mental retardation and brain malformations in patients with hypomagnesemia was shown to be caused by mutations in CNNM2 . Our experiments established CNNM2 as a new essential gene in brain development , neurological functioning and Mg2+ homeostasis . This notion is supported by the following observations; i ) hypomagnesemic patients with CNNM2 mutations suffer from seizures , mental disability , and if mutations are present in recessive state , brain malformations are observed in addition; ii ) Mg2+ supplementation does not improve the neurological phenotype of the patients; iii ) CNNM2 increases Mg2+ uptake in HEK293 cells , whereas mutant CNNM2 does not; iv ) knockdown of CNNM2 orthologues in zebrafish results in impaired development of the brain , abnormal neurodevelopmental phenotypes manifested as altered locomotor and touch-evoke escape behaviours , and Mg wasting; v ) the zebrafish phenotype can be rescued by injection of mouse Cnnm2 cRNA . In addition to the previously reported dominant mode of inheritance [9] , the genetic findings in our patients support heterogenous patterns of inheritance . In family F1 , a recessive mode of CNNM2 inheritance was observed . The homozygous CNNM2 p . Glu122Lys mutation in this family resulted in the manifestation of a neonatal onset and a considerably more severe cerebral involvement than in the remaining patients . Yet , a central nervous system ( CNS ) phenotype with seizures and intellectual disability , which was not reported previously , represented the cardinal clinical symptom in all of our patients . Seizures constituted the major symptom at manifestation coinciding with hypomagnesemia , but were also seen during follow-up despite Mg2+ supplementation . Pronounced Mg2+ deficiency reflected by severely low serum Mg2+ levels clearly represents a promotive element in the development of seizures . However , the persistence of seizure activity despite Mg2+ supplementation might point to a genuine disturbance in brain function caused by defective CNNM2 . Accordingly , the extent of hypomagnesemia found in the two siblings with the recessive mutation , F1 . 1 and F1 . 2 , was identical to other patients ( F2 . 1–F5 . 1 ) with heterozygous CNNM2 mutations and a milder neurological phenotype . Continuous oral Mg2+ supplementation stabilized serum Mg2+ levels in the subnormal range , however a complete normalization of Mg2+ metabolism could not be achieved in any of the patients . In three out of five families ( F2 , F3 and F4 ) , the mutation of the patient was not present in the parents . This finding supports the recent advancements evidencing that de novo mutations provide an important mechanism in the development of mental disability disorders [30] . Remarkably , the same de novo p . Glu357Lys mutation was identified in two unrelated individuals . Although the mutation rate along the human genome varies significantly [31] , the chance to observe an identical de novo base pair change in two individuals is extremely small , supporting causality of this mutation . The clinical and genetic findings observed in patient F5 . 1 should be interpreted with caution . Though the p . Leu330Phe variant was not present in controls or exome variant databases , the presence of an innocuous polymorphism cannot be completely excluded . The clinical phenotype with a milder degree of intellectual disability and a later manifestation with hypomagnesemic symptoms during adolescence support a partial loss of CNNM2 function caused by the p . Leu330Phe variant . Unfortunately , the patient was not available for clinical re-evaluation . Over the recent years , the function of CNNM2 in the context of Mg2+ handling has been heavily debated [9] , [11]–[13] , [20] . Therefore , an important question is how CNNM2 mutations cause impaired Mg2+ metabolism and lead to CNS dysfunction . Functional studies in HEK293 cells demonstrated a putative role in cellular Mg2+ transport . Overexpression of CNNM2 increased cellular Mg2+ uptake , which was abrogated by introduction of the CNNM2 mutants identified in our patients . The p . Ser269Trp and p . Glu357Lys mutants as well as the previously published p . Thr568Ile mutant [9] , all identified in heterozygous state , failed to enhance the cellular Mg2+ uptake , indicating a loss-of-function in mutated CNNM2 . The recessively inherited p . Glu122Lys mutant identified in patients F1 . 1 and F1 . 2 displayed a small but significant residual function , while Mg2+ uptake was almost completely retained for mutant p . Leu330Phe . Biotinylation experiments demonstrated a trafficking defect of p . Glu122Lys and p . Ser269Trp mutants supporting a loss-of-function nature of CNNM2 mutations . Together , these findings argue for distinct degrees of severity of the disease depending on the number of affected alleles . Furthermore , a small residual function of p . Glu122Lys is in line with the lack of a clinical phenotype in the parents of family F1 . The parents , however , declined a thorough evaluation of their Mg2+ status . To further analyze the relevance of CNNM2 for brain and Mg2+ metabolism deduced from the human disease model , the translation of orthologues of CNNM2 ( cnnm2a and cnnm2b ) was knocked down in zebrafish . In line with the human disease , the concentration of total body Mg was decreased in zebrafish cnnm2a and cnnm2b morphants when compared to controls . The decrease in total body Mg content is interpreted as a decrease in the renal absorption and/or skin uptake ( through ionocytes analogous to renal tubular cells ) of the ionic fraction , Mg2+ , since only Mg2+ is transported transcellularly and no intestinal Mg2+ uptake takes place in zebrafish larvae . Additionally , Mg losses observed in morphant larvae were rescued by expression of wild-type Cnnm2 , but not by expression of mutant Cnnm2 . This demonstrates the specificity of our MO antisense oligos , as well as the functional equivalence between mammalian CNNM2 and its zebrafish orthologues . Consistent with the human pathology , knockdown of cnnm2a or cnnm2b induced brain malformations . Specifically , the brain phenotype observed in cnnm2b morphants resembles that found in patient F1 . 1 showing enlarged outer cerebrospinal liquor spaces . This provides further consistency to link CNNM2 dysfunction with the brain morphological defects found in this homozygous patient . The absence of outer cerebrospinal liquor spaces in the cerebrum of cnnm2a morphants shows that cnnm2 paralogues in zebrafish are a case of subfunctionalization at the level of the cerebrum . CNS malformations were rescued in morphants by co-injection with mouse wild-type Cnnm2 . In homozygous patients , the neurological defects became evident early after birth . In line with a developmental role for CNNM2 within the CNS , gene expression of zebrafish cnnm2a and cnnm2b peaked within the first 24 hpf . In addition , in situ hybridization located cnnm2a expression specifically in the MHB [32] , an organizing center in the neural tube that determines neural fate and differentiation in the CNS during development [33]–[34] . Indeed , the most striking brain developmental defect in our study is maldevelopment of the MHB . These defects were rescued with Cnnm2 cRNA . Interestingly , the brain phenotypes observed in both zebrafish and patients were independent of Mg2+ , as Mg2+ supplementation was unsuccessful to rescue the phenotypes . Thus , our findings suggest that a brain-specific CNNM2 function is crucial for the development of constitutive regions of the CNS , which in the zebrafish model is illustrated by defects in the MHB . At 25 hpf , a time point in which the locomotor behaviour is unaffected by the brain and only depends on signals from the spinal cord [27] , zebrafish morphant embryos displayed an increased frequency of spontaneous contractions , especially when the MOs were co-injected with mutant Cnnm2 . This hyperexcitability of motor neurons suggests a function of zebrafish Cnnm2 proteins in the regulation of the activity of the neurological network in the spinal cord or in the synaptic junctions with muscle fibbers . Consistent with these findings , patients with mutations in CNNM2 presented impaired motor skills , which were severe in the case of homozygous patients . In the CNS , TRPM7 is essential during early development [35] , as it modulates neurotransmitter release in sensory neurons [36]–[37] . Specifically , when using 2-APB , an inhibitor of TRPM7 [22] , CNNM2-dependent Mg2+ transport was abolished in HEK293 cells . Remarkably , in a similar fashion to trpm7 mutants in zebrafish [28]–[29] , cnnm2a or cnnm2b morphants showed weaker touch-evoked escape behaviour compared to controls . In 5 dpf larvae , and unlike in 25 hpf embryos , locomotor behaviours elicited by touch require the involvement of high brain structures [27] . Therefore , it is reasoned that CNNM2 conditions locomotor behaviour with an etiology that can be related to lack of excitation of sensory neurons via TRPM7 and/or to the defects in early brain development observed in zebrafish morphant embryos . In kidney , where CNNM2 is expressed at the basolateral membrane in DCT , specific regulation of TRPM7 Mg2+ reabsorption is unlikely , since TRPM6 is the main Mg2+ transporter in this segment . TRPM7 is a ubiquitously expressed gene regulating cellular Mg2+ metabolism , which is for instance involved in regulation of brain Mg2+ levels [6] . Therefore , one could hypothesize that CNNM2 may regulate other proteins in addition to TRPM7 in kidney for the control of Mg2+ reabsorption , which remain to be identified . In conclusion , our findings of CNNM2 mutations in patients with hypomagnesemia and severe neurological impairment widen the clinical spectrum of CNNM2-related disease . By establishing a zebrafish CNNM2 loss-of-function model of the genetic disease , we provide a unique model for the testing of novel therapeutic drugs targeting CNNM2 .
All genetic studies were approved by the ethics committee of the Westfälische Wilhelms University , Münster . All patients or their parents provided written informed consent in accordance to the Declaration of Helsinki . All animal experiments were performed in agreement with European , National and Institutional regulations . Animal experimentation and analysis was restricted to the first five days post-fertilization ( dpf ) . We studied a cohort of six patients from five families with hypomagnesemia and mental retardation . Patients F1 . 1 to F4 . 1 are followed in secondary or tertiary care neuropediatric centres . Neuroimaging was performed in F1 . 1 , F2 . 1 , F3 . 1 , and F4 . 1 by cranial MRI ( magnetic resonance imaging ) . Psychological diagnostic evaluation in patients F2 . 1 and F3 . 1 was performed using Snijders Oomen Non-Verbal ( SON ) Intelligence Test ( revised ) 5 . 5–17 years . Copy number variations ( CNVs ) associated with neurodevelopmental delay and intellectual disability were excluded in patients F1 . 1 and F2 . 1 by array CGH ( comparative genomic hybridization ) using the Sureprint G3 Human CGH Microarray kit ( Agilent Technologies , Boeblingen , Germany ) in patient F1 . 1 and using the Affymetrix Cytogenetics Whole-Genome 2 . 7 Array in patient F2 . 1 . Genomic DNA of affected individuals and available family members was extracted from whole blood using standard methods . A genome scan for shared homozygous regions was performed in the two affected children F1 . 1 and F1 . 2 with suspected parental consanguinity . Samples were genotyped on an Illumina human 660W Quad beadchip SNP array ( Illumina , Eindhoven , The Netherlands ) . Merlin 1 . 1 . 2 ( University of Michigan , Ann Arbor , MI , USA ) was used to determine homozygous regions by linkage analysis . As exact information on pedigree structure was missing , we used a 1 . 7 Mb threshold for regions identical by descent that is very rarely crossed by non-consanguineous samples , but allows to identify most of the true homozygosity regions if parental consanguinity is present [38] . A list of candidate genes within the identified homozygous intervals was generated including known Refseq genes as well as novel transcripts using Ensembl Genome assembly GRCh37 via biomart ( www . ensembl . org ) . At a cut-off size of >1 . 7 Mb , eleven critical intervals were yielded on autosomes with a cumulative size of 62 Mb . The gene list generated from these loci included 322 RefSeq genes and putative transcripts , including CNNM2 in a critical interval of 7 . 1 Mb on chromosome 10 . The entire coding region and splice-sites of the most promising candidate gene CNNM2 were sequenced from both strands ( Genbank: NM_017649 . 4 , Uniprot: Q9H8M5 ) . After discovery of a homozygous mutation in the index family F1 , the mutational screening was extended to patients with hypomagnesemia without mutations in known genes involved in hereditary magnesium wasting . The presence of newly identified CNNM2 sequence variations was tested in at least 204 ethnically matched control alleles and compared to publically available exome data ( www . 1000genomes . org; evs . gs . washington . edu ) . Additionally , all identified mutations were ranked for potential damage on protein function using Polyphen-2 ( genetics . bwh . harvard . edu/cgi-bin/ggi/ggi2 . cgi ) . Mouse wild-type Cnnm2 construct was cloned into the pCINeo HA IRES GFP vector as described previously [12] . Cnnm2 mutations were inserted in the construct using the QuikChange site-directed mutagenesis kit ( Stratagene , La Jolla , CA , USA ) according to the manufacturer's protocol . All constructs were verified by sequence analysis . Primer sequences used for cloning or mutagenesis PCR are reported in Table S1 . HEK293 cells were grown in Dulbecco's modified eagle's medium ( DMEM , Bio Whittaker-Europe , Verviers , Belgium ) containing 10% ( v/v ) fetal calf serum ( PAA , Liz , Austria ) , 2 mM L-glutamine and 10 µg/mL non-essential amino acids , at 37°C in a humidity-controlled incubator with 5% ( v/v ) CO2 . The cells were transiently transfected with the respective DNA constructs using Lipofectamin 2000 ( Invitrogen , Breda , The Netherlands ) at 1∶2 DNA∶Lipofectamin ratio for 48 hours unless otherwise stated . HEK293 cells were transfected with wild-type and mutant CNNM2 constructs for 48 hours . Subsequently , cell surface proteins were biotinylated as described previously [39] . Briefly , cell surface proteins were biotinylated for 30 min at 4°C in 0 . 5 mg/mL sulfo-NHS-LC-LC-biotin ( Pierce , Rockford , IL , USA ) . Cells were washed and lysed in lysis buffer ( 150 mM NaCl , 5 mM EGTA , Triton 1% ( v/v ) , 1 µg/ml pepstatin , 1 mM PMSF , 5 µg/ml leupeptin , 5 µg/ml aproptin , 50 mM Tris/HCl , pH 7 . 5 ) . 10% ( v/v ) of the sample was taken as input control and the rest of the protein lysates were incubated overnight with NeutrAvidin-agarose beads ( Pierce , Rockford , IL , USA ) at 4°C . The next day , unbound protein was discarded by washing the beads 5 times with lysis buffer . The remaining protein lysates were denatured in Laemmli containing 100 mM DTT for 30 min at 37°C and subsequently subjected to SDS-PAGE . Then , immunoblots were incubated with mouse anti-HA 1∶5 , 000 primary antibodies ( Cell Signaling Technology , Danvers , MA , USA ) and peroxidase conjugated sheep anti-mouse secondary antibodies 1∶10 , 000 ( Jackson Immunoresearch , Suffolk , UK ) . HEK293 cells were transfected with wild-type and mutant CNNM2 constructs for 48 hours and seeded on poly-L-lysine ( Sigma , St Louis , MO , USA ) coated 12-well plates . Mg2+ uptake was determined using a stable 25Mg2+ isotope ( Cortecnet , Voisins Le Bretonneux , France ) , which has a natural abundance of ±10% . Cells were washed with basic uptake buffer ( 125 mM NaCl , 5 mM KCl , 0 . 5 mM CaCl2 , 0 . 5 mM Na2HPO4 , 0 . 5 mM Na2SO4 , 15 mM HEPES/NaOH , pH 7 . 5 ) and subsequently placed in basic uptake buffer containing 1 mM 25Mg2+ ( purity ±98% ) for 5 minutes unless stated differently . After washing three times with ice-cold PBS , the cells were lysed in HNO3 ( ≥65% , Sigma ) and subjected to ICP-MS ( inductively coupled plasma mass spectrometry ) analysis . For extrusion experiments , cells were transfected with wild-type or mutant CNNM2 constructs for 24 hours . After 24 hours , cells were placed in culture medium containing 1 mM 25Mg2+ ( purity ±98% ) for an additional 48 hours . Before the start of the experiment , the cells were briefly washed in basic uptake buffer and subsequently placed in basic uptake buffer containing 0 . 5 mM Mg2+ ( containing ±10% 25Mg2+ ) for 5 minutes . After washing three times with ice-cold PBS , the cells were lysed in nitric acid and subjected to ICP-MS analysis . Wild-type Tupfel long-fin zebrafish were bred and raised under standard conditions ( 28 . 5°C and 14 h of light: 10 h of dark cycle ) in accordance with international and institutional guidelines . Zebrafish eggs were obtained from natural spawning . The following antisense oligonucleotides ( MOs ) were raised against the translational start site of cnnm2a and cnnm2b , along with the standard mismatch control MO ( Gene Tools , Philomath , OR , USA ) : cnnm2a , 5′-GCGGTCCATTGCTCTGCCATGTTGA-3′; cnnm2b , 5′-ACCGACGGTTCTGCCATGTTGATAA-3′; and the negative control ( standard mismatch MO ) , directed against a human β-globin intron mutation , 5′-CCTCTTACCTCAGTTACAATTTATA . The underlined areas indicate the complementary sequences to the initial methionines of cnnm2a and cnnm2b . MOs were diluted in deionized , sterile water supplemented with 0 . 5% ( w/v ) phenol red and injected in a volume of 1 nl into the yolk of one- to two-cell stage embryos using a Pneumatic PicoPump pv280 ( World Precision Instruments , Sarasota , FL , USA ) . Wild-type ( uninjected ) embryos were also included in the experiments to control for the effects of the injection procedure per se . To determine the most effective dose of the cnnm2a- and cnnm2b-MO , 2 , 4 and 8 ng were injected . In these experiments , control embryos were injected with 8 ng of the standard mismatch control MO ( the highest dose ) . After injection , embryos from the same experimental condition were placed in 3 Petri dishes ( at a maximum density of 45 embryos/dish , allowing statistical comparisons between survivals in the different experimental conditions ) and cultured at 28 . 5°C in E3 embryo medium ( 5 mM NaCl , 0 . 17 mM KCl , 0 . 33 mM CaCl2 , 0 . 33 mM MgSO4 ) , which was refreshed daily . As criteria for subsequent experiments , the dose of MO that caused major effects and induced a percentage of mortality non-significantly different from controls was injected ( 2 ng for cnnm2a-MO and 8 ng for cnnm2b-MO ) . In experiments that implied exposure to high Mg2+ concentrations , the high Mg2+ medium had a composition of 5 mM NaCl , 0 . 17 mM KCl , 0 . 33 mM CaCl2 and 25 mM MgSO4 . In order to control for the specificity of the MOs blocking the translation of cnnm2a and cnnm2b , as well as for toxic off-target effects , in vivo cRNA rescue experiments were performed [40] . For these experiments , mouse wild-type CNNM2 and mutant ( p . Glu357Lys ) CNNM2 cRNAs were prepared using the mMESSAGE mMACHINE Kit ( Ambion , Austin , TX , USA ) according to the manufacturer's instructions . The cRNAs , in an amount of 50 pg , as based on other studies [26] , were ( co ) injected together with MOs as described above . Zebrafish embryos and larvae were phenotyped at 25 hpf or 5 dpf , respectively . For the analyses of brain phenotypes , the brain rudiment of zebrafish embryos at 25 hpf was observed for morphological changes under a Leica MZFLIII microscope ( Leica Microsystems Ltd , Heerburgg , Germany ) . Morphological phenotypes , which also included the brain , were also analysed in larvae at 5 dpf . Embryos or larvae were classified into different classes of phenotypes on the basis of comparisons with stage-matched control embryos of the same clutch . In cnnm2a morphant embryos ( 25 hpf ) , two phenotype classes were distinguished: class I , normal; and class II , embryos with underdeveloped MHB . In cnnm2a morphant larvae ( 5 dpf ) , the following classes were distinguished: class I , normal; class II , normal with non-inflated swim bladder; class III , larvae with enlarged pericardial cavity ( edema ) and non-inflated swim bladder; class IV , larvae with notochord defects , enlarged pericardial cavity and non-inflated swim bladder; and class V , larvae with severely enlarged pericardial cavity , notochord defects and non-inflated swim bladder . In cnnm2b morphant embryos ( 25 hpf ) , three different phenotypes were distinguished: class I , normal; class II , embryos with underdeveloped MHB; and class III , embryos with underdeveloped MHB and enlarged tectum . In cnnm2b morphant larvae ( 5 dpf ) , the number of different phenotypes distinguished were four: class I , normal; class II , normal with non-inflated swim bladder; class III , larvae with enlarged pericardial cavity ( edema ) and non-inflated swim bladder; and class IV , larvae with widened outer cerebrospinal fluid spaces , kidney cysts , severely enlarged pericardial cavity and non-inflated swim bladder . Representative images were obtained with a DFC450C camera ( Leica Microsystems Ltd ) after anaesthetising embryos or larvae with tricaine/Tris pH 7 . 0 solution . Prior to anaesthesia and image acquisition , zebrafish embryos were manually dechorionated . Zebrafish embryos or larvae were anesthetized with tricaine/Tris pH 7 . 0 solution and 5–7 individuals were pooled as one sample . Samples were then snap frozen in liquid nitrogen and stored at −80°C in order to ensure euthanasia of animals and remained at these storage conditions until the beginning of the analytical procedures . Analytical procedures started by quickly washing the samples with nanopure water in order to avoid contamination of remaining waterborne Mg2+ . The washing procedure was repeated twice . Fish were then dried at 65°C for 1 . 5 hours , at which time 2 . 5 µl of HNO3 ( ≥65% , Sigma ) was added to each tube . Samples were digested at 65°C during 1 . 5 hours . After , digested samples were diluted 1∶10 with 22 . 5 µl nanopure water . The total Mg content in each sample was determined with a commercial colorimetric assay ( Roche Diagnostics , Woerden , The Netherlands ) following the manufacturer's protocol . Blanks ( HNO3 diluted 1∶10 with nanopure water ) were added during assays and values were equal to zero . Within-run precision and accuracy was controlled by means of an internal control Precinorm ( Roche Diagnostics ) . Samples from embryos exposed to 25 mM Mg2+ were further diluted ( 1∶20 ) . Furthermore , samples were normalized by protein content , which was determined in 1∶20 ( embryos at 25 hpf ) or 1∶50 ( larvae at 5 dpf ) diluted samples using the Pierce BCA protein assay kit ( Pierce Biotechnology , Rockford , IL , USA ) . Zebrafish embryos or larvae at specific developmental times ( 6 , 12 , 24 , 48 , 72 , 96 and 120 hpf ) were anaesthetised with tricaine/Tris pH 7 solution and 10 individuals were pooled as one sample . RNA isolation , cDNA synthesis and quantitative real-time PCR ( RT-qPCR ) measurements were carried out as previously described using validated cnnm2a and cnnm2b primers [18] . Samples were normalized to the expression level of the housekeeping gene elongation factor-1α ( elf1α ) [18] . Relative mRNA expression was analysed using the Livak method ( 2−ΔΔCt ) , where results are expressed relative to the gene expression at 6 hpf ( time point chosen as calibrator ) . At 25 hpf , 10 zebrafish embryos per Petri dish ( n = 30 per experimental condition ) were randomly selected . The number of complete body contractions each zebrafish made in 30 second period was counted and was used as indicative of motor neuron activity [27] . Representative videos of each experimental condition were taken using Leica Application Suite ( Leica Microsystems Ltd ) and a Leica MZFLIII microscope ( Leica Microsystems Ltd ) equipped with a DFC450C camera ( Leica Microsystems Ltd ) . For the analysis of the touch-evoked escape behaviour , 10 zebrafish larvae per Petri dish ( n = 30 per experimental condition ) were randomly selected . Touch-evoked escape behaviours were elicited by touching a larva in the tail up to 6 times with a pair of forceps at 5 dpf . Three categories were distinguished , responders , late responders and non-responders , to which the following scores were given: 3 points for responders: fish quickly react ( swimming or flicking the tail ) to the stimuli after 1 or 2 twitches; 2 points for late responders: fish react ( swimming or flicking the tail ) to the stimuli after 3 , 4 or 5 twitches; and 1 point for non-responders: fish do not react to the stimuli after more than 5 twitches . Representative videos were recorded with the system described above . All results are depicted as mean ± standard error of the mean ( SEM ) . Statistical analyses were conducted by one- or two-way ( for experiments where zebrafish embryos were exposed to different Mg2+ concentrations , then two factors of variance appear: Mg2+ concentration and treatment ) ANOVA . Where appropriate , data were logarithmically transformed to fulfil the requirements for ANOVA , but all data are shown in their decimal values for clarity . When data did not comply with the premises of the parametric ANOVA , data were analyzed using a Kruskal-Wallis ANOVA on ranks . Tukey's post-test was used to identify significantly different groups . Statistical significance was accepted at P<0 . 05 . | Mental retardation affects 1–3% of the population and has a strong genetic etiology . Consequently , early identification of the genetic causes of mental retardation is of significant importance in the diagnosis of the disease , as predictor of the progress of the disease and for the determination of treatment . In this study , we identify mutations in the gene encoding for cyclin M2 ( CNNM2 ) to be causative for mental retardation and seizures in patients with hypomagnesemia . Particularly , in patients with a recessive mode of inheritance , the intellectual disability caused by dysfunctional CNNM2 is dramatically severe and is accompanied by severely limited motor skills and brain malformations suggestive of impaired early brain development . Although hypomagnesemia has been associated to several neurological diseases , Mg2+ status is not regularly assessed in patients with seizures and mental disability . Our findings establish CNNM2 as an important protein for renal magnesium handling , brain development and neurological functioning , thus explaining the physiology of human disease caused by ( dysfunctional ) mutations in CNNM2 . CNNM2 mutations should be taken into account in patients with seizures and mental disability , specifically in combination with hypomagnesemia . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"mineral",
"metabolism",
"and",
"the",
"kidney",
"epilepsy",
"medicine",
"and",
"health",
"sciences",
"neurology",
"developmental",
"and",
"pediatric",
"neurology",
"nephrology"
]
| 2014 | CNNM2 Mutations Cause Impaired Brain Development and Seizures in Patients with Hypomagnesemia |
Convergent morphologies have arisen in plants multiple times . In non-vascular and vascular land plants , convergent morphology in the form of roots , stems , and leaves arose . The morphology of some green algae includes an anchoring holdfast , stipe , and leaf-like fronds . Such morphology occurs in the absence of multicellularity in the siphonous algae , which are single cells . Morphogenesis is separate from cellular division in the land plants , which although are multicellular , have been argued to exhibit properties similar to single celled organisms . Within the single , macroscopic cell of a siphonous alga , how are transcripts partitioned , and what can this tell us about the development of similar convergent structures in land plants ? Here , we present a de novo assembled , intracellular transcriptomic atlas for the giant coenocyte Caulerpa taxifolia . Transcripts show a global , basal-apical pattern of distribution from the holdfast to the frond apex in which transcript identities roughly follow the flow of genetic information in the cell , transcription-to-translation . The analysis of the intersection of transcriptomic atlases of a land plant and Caulerpa suggests the recurrent recruitment of transcript accumulation patterns to organs over large evolutionary distances . Our results not only provide an intracellular atlas of transcript localization , but also demonstrate the contribution of transcript partitioning to morphology , independent from multicellularity , in plants .
Convergent morphologies have arisen multiple times in plants ( Viridiplantae ) . Diverse cellular architectures underlie these moprhologies , with varying relationships between the number of nuclei per cell and the number of cells within an organism . Within the Chlorophyta , Acetabularia possesses an anchoring rhizoid , supporting stalk , and photosynthetic cap , but is , during most of its life cycle , a unicellular organism reaching heights of up to 10 cm with a single nucleus [1]–[3] . Another green alga , Caulerpa , is one of the largest known single-celled organisms , with stolons ( up to meters in length ) producing fronds and holdfasts [4]–[9] ( Fig . 1A–C ) . Unlike Acetabularia , which is a single-celled organism , Caulerpa is coenocytic , with numerous nuclei . Siphonocladous chlorophytes have a chambered body plan compartmentalizing variable numbers of nuclei , as in Cladophora . Land plants ( Embryophyta ) are multicellular organisms , in which organs are composed of tissues and distinct cell types . Developmental biology in land plants was historically influenced by cell theory and studies in animals , in which organismal level morphology is an emergent property of cell division and histogenesis [10] , [11] . Animal development is a poor example for land plants , in which morphogenesis is dissociated from histogenesis because cellular lineages and division patterns are largely independent from organ morphology . For the above reasons , it has been argued [11] that cell theory is not as applicable to plants as in animals with respect to explaining how complex morphologies arise . In its place , Kaplan and Hagemann [11] argued for organismal theory , which they define as “[interpreting] the living protoplasmic mass as a whole , rather than considering its constituent cells as the basic unit . ” In other words , the morphology in plants arises at the organismal level rather than as an emergent cellular property . Kaplan and Hagemann assert that “higher plants are also siphonous , but at a subtler , microscopic level . ” Some of the siphonous features they argue land plants possess include: 1 ) cell division through a phragmoplast , 2 ) plasmodesmata , 3 ) the symplasm , 4 ) a multinucleate endosperm and megagametophytes , 5 ) distinct cytohistological zonations of the shoot apical meristem throughout the Embryophyta , 6 ) that cell lineage is often independent of morphology ( e . g . , in leaves ) , 7 ) and convergent morphology in multicellular red algae and land plants with different cell lineage patterns . That land plants are truly siphonous is false: cell walls are a prominent features of land plants upon which morphology is dependent and land plants are multicellular organisms . However , it is useful to think about development in land plants from this unique perspective . That organ growth and morphogenesis are separate from cell division reduces the importance of cell-type specific transcript accumulation in these organisms . Transcriptomics and phylogenetics provide a mechanism to test hypotheses of cell versus organismal theory in siphonous green algae and land plants . Do the accumulation patterns of transcripts differ between single-celled and multicellular organisms with convergent morphology ? Are groups of transcripts recurrently recruited to organs across large evolutionary distances regardless of whether an organism is multicellular ? Here , we provide a transcriptome of the giant coenocyte Caulerpa taxifolia . We detect a strong apical-basal gradient of transcript accumulation within the cell . Groups of transcripts with distinct functionalities accumulate in relevant pseudo-organs ( morphological structures equivalent to a multicellular organ but not comprised of tissues or cells ) . Cell compartmentalization is partitioned in Caulerpa , despite its polynucleate condition , and transcripts are patterned according to the flow of genetic information , from transcription-to-translation in a basal-to-apical fashion . An analysis of the intersection of the transcriptomic atlases of a land plant ( tomato , Solanum lycopersicum ) and Caulerpa demonstrates a limited , recurrent recruitment of genes with similar functions to morphological structures . Our results provide a broad , evolutionary context for the relationship between the cell and organismal morphology at a molecular level within plants , confirming and expanding upon the organismal theory originally proposed by Kaplan and Hagemann [11] .
To develop a resource to address how organismal morphology can arise in the absence of multicellularity , we sequenced transcripts from multiple pseudo-organs and de novo assembled the intracellular transcriptome of Caulerpa taxifolia ( see sequence submission information and S1–S4 Datasets ) . Caulerpa taxifolia stolons , upwards of 1 m in length , bear fronds ( typically 15–30 cm long at maturity ) with pinnately-arranged , tapered pinnules . The pinnules arise from active growth at the frond apex , which superficially resembles , in form and function , the apical cells and meristems of embryophytes ( Fig . 1B ) . Caulerpa taxifolia is anchored by holdfasts , which take up phosphorus , nitrogen , and carbon from the substrate , and harbor both ecto- and endosymbiont bacteria that aid this process [9] . We sampled five replicates each of 1 ) the frond apex , 2 ) rachis , 3 ) pinnules , 4 ) the lower third of the frond base , 5 ) stolon , and 6 ) holdfast regions ( Fig . 1C ) . One holdfast sample was lost when thawing for library preparation , reducing holdfast sampling to four replicates . The sample we sequenced was clonal in origin , having proceeded through numerous rounds of asexual reproduction . In its vegetative phase , Caulerpa taxifolia is a haplophasic diploid . Caulerpa taxifolia has one of the smallest genome sizes in its genus ( ∼100 Mbp , approximately the size of the Arabidopsis thaliana genome ) and unlike other Caulerpa species does not exhibit extensive endopolyploidy [12] , [13] . The transcriptome of Caulerpa taxifolia is dominated by patterning along the apical-basal axis . Throughout this manuscript , we use the terms “accumulation” and “abundance” in a relative sense to describe transcript accumulation patterns . Transcript accumulation in replicates derived from basal regions ( holdfast , stolon , frond base ) is highly similar and distinct from apical regions ( frond apex , rachis , pinnules ) , as shown in a Principal Component Analysis ( PCA ) performed on replicates ( Fig . 1D; S5–S20 Datasets ) . The growing frond apex in particular exhibits a unique transcriptomic signature , perhaps indicative of the “meristemplasm” found in this region , as previously described [5] , [14] . A Self-Organizing Map ( SOM ) was used to partition transcripts into six clusters ( nodes ) , each with a distinct accumulation pattern ( Figs . 2 , S1; S21 Dataset ) . These nodes explain prominent densities of transcripts with similar accumulation patterns across organs , as visualized using a PCA ( Fig . 2A–B ) . The nodes are roughly organized along the apical-basal axis ( Fig . 2C ) . For example , Node 3 transcripts exhibit high frond apex accumulation , and progressing basally to Node 5 transcripts which accumulate at high levels in the holdfast , nodes with intermediate accumulation patterns along the apical-basal axis are observed . The overall patterns of transcript accumulation , visualized using the combination of SOMs and PCA , can be explored for a random subset of genes in an interactive graphic we have prepared ( http://danchitwood . github . io/CaulerpaGeneExpression/ , for use with a Google Chrome web browser ) . Such strong apical-basal , intracellular partitioning of transcript accumulation is not surprising considering the influence of gravitropism on regeneration and anchoring [6] , circadian movements of chloroplasts into and out of the apex , and cytoplasmic streaming along the frond and pinnule lengths [5] . Transcripts belonging to each node are highly enriched for associated Gene Ontology ( GO ) terms , often specific to cellular functions and organelles ( S22–S28 Datasets ) . For example , Node 2 transcripts , which accumulate at high levels in the pinnules and rachis , are enriched for photosynthetic GOs , but additionally those associated with mitochondria , respiration , the electron transport chain and ATP synthesis , as well as the production of secondary metabolites . Node 3 transcripts with high abundances in the frond apex are enriched for COPI/II vesicle coat proteins and kinases . Most surprising is the overwhelming concentration of transcripts associated with nuclear functions—DNA replication and damage , chromatin , RNAi , and even the subunits of DNA polymerase II—in the frond base , stolon , and holdfast . Multicellular land plants possess an inherent constraint at the cellular level . Generally , every cell must have a nucleus , plastids , mitochondria , and cytoskeletal components to carry out basic metabolism , cell division , and differentiation , although numerous exceptions exist . But how is cellular compartmentalization distributed over similar morphology in Caulerpa ? One hypothesis is that because morphogenesis is decoupled from multicellularity in Caulerpa , the distribution of different cell compartment identities might consolidate within distinct organs . That is , each cell is subcompartmentalized in a multicellular land plant , whereas the siphonous body plan of Caulerpa may maintain compartmentalization in pseudo-organs . Indeed , GO enrichment analysis reveals that transcript identity loosely follows the flow of genetic information progressing in a basal to apical direction in Caulerpa ( Fig . 2 , S22–S28 Datasets ) . Transcripts associated with transcriptional gene regulation accumulate at high levels in the holdfasts , stolon , and frond base , whereas those associated with translation are more abundant in the pinnules . Vesicular trafficking and kinase activity , associated with the cytoplasm and plasma membrane , are enriched within the frond apex . To explore the fundamental relationship between cell compartmentalization and organism morphology , we selected all transcripts belonging to significantly enriched GO terms related to transcriptional gene regulation , translation , and other important organellar and cell biological functions ( Fig . 3 , S29 Dataset ) . Transcripts encoding RNA polymerase II subunits are highly abundant in the holdfast . Those encoding numerous chromatin , epigenetic , DNA recombination , repair , and replication , and RNAi machinery components are highly abundant in the stolon . Transcripts related to translation accumulate at high levels in the photosynthetic tissues , mostly in the pinnules and somewhat in the rachis . Proteolysis transcripts are found in these regions too , but additionally in the frond apex where translational components accumulate at lower levels . COPI/II coat proteins and numerous kinases are highly abundant in the frond apex . The association of COPI/II trafficking with the apex , an active growth region containing white “meristemplasm , ” is consistent with the previously reported enrichment of rough endoplasmic reticulum and Golgi bodies in this region [4] , [5] , [14] . The overall transcriptomic signature in Caulerpa—a single cell—is striking . From the holdfast to the frond apex , transcript accumulation loosely follows a nuclear-to-cytoplasm and transcriptional-to-translational pattern of identity ( Fig . 3 ) . The Caulerpa body plan is compartmentalized as if a single land plant cell , and different cellular compartments in Caulerpa are associated with different types of morphogenesis . Land plant morphology , and the numerous and diverse morphologies of various chlorophytes , are derived from the monophyletic inheritance of a core gene set . In some instances , as between land plants and Caulerpa , convergent structures with related functions ( for example , leaves and fronds , and roots and holdfasts ) have evolved using these genes . If land plant morphology is viewed from the perspective of organismal theory proposed by Kaplan and Hagemann [11] , and land plants are even considered to be siphonous and cellularization patterns arbitrary , then the accumulation of transcripts throughout the organism can be compared to detect molecular homology . To what degree have similar transcript accumulation profiles been recruited to morphological structures in land plants and Caulerpa ? To answer this question , we analyzed the intersection of the Caulerpa transcriptomic atlas ( Figs . 1–3 ) with a transcriptomic atlas from a land plant ( tomato , Solanum lycopersicum cv . M82 ) that was derived using similar molecular and bioinformatics methods as presented here [15]–[17] . Putative homologous transcripts from tomato ( see Materials and Methods ) were used to assign Caulepra transcripts to a corresponding tomato self-organizing map node [17] ( Fig . 4A ) . The distribution of genes from each Caulepra node across tomato nodes was then compared to the expected distribution using a χ2 test . A higher than expected enrichment of genes assigned to a particular tomato node indicates that a group of Caulerpa genes with similar accumulation patterns are associated with a specific accumulation profile in tomato ( Fig . 4B ) . For example , Caulerpa Node 2 genes , which are highly abundant in the photosynthetically active pinnules and rachis ( Fig . 2 ) , are associated with tomato genes belonging to tomato Node 3 , which are highly abundant in leaf , seedling , and vegetative apex samples ( note: tomato and Caulerpa nodes are distinct and should not be confused with each other ) ( Fig . 4B , C ) . The intersection of Caulerpa Node 2 with tomato Node 3 is predominately photosynthetic genes ( S30 Dataset ) . Although the molecular correspondence between photosynthetic structures is expected , other associations are less so . Caulepra Node 3 genes are highly abundant in the frond apex and are associated with tomato Node 8 genes that accumulate at high levels in the root and stem ( Fig . 4D ) . Consistent with enriched GO terms in both Caulerpa and tomato , these genes are associated with vesicular trafficking ( particularly COP cotamers ) and vacuolar transporters ( S30 Dataset ) . Interestingly , the Caulerpa Node 6 genes with high stolon abundance are associated with tomato Node 1 genes with high abundance in the inflorescence and relatively high abundance in the vegetative apex , both meristematic organs ( Fig . 4E ) . The genes intersecting both nodes ( S30 Dataset ) are members of RNAi , chromatin , and DNA recombination , repair , and replication pathways , suggesting a molecular association between the stolon and meristems of land plants . The ability of the stolon to repetitively produce pseudo-organs ( both fronds and holdfasts ) and the enrichment of nuclear replication-associated transcripts indicates meristem-like identity at the molecular level . The association between transcript accumulation profiles in Caulerpa and a land plant suggests , to a limited extent , molecular homology underlying morphology . Kaplan and Hagemann [11] argued that land plants , like Caulerpa , are siphonous . While the statement is extreme and not technically correct , reevaluating land plant development from this perspective is insightful , with respect to the role cells play in determining morphology . Morphogenesis and key patterning events in land plants rely on non-cell autonomous , symplastic movement of transcription factors and small RNAs , that transcend cell division patterns [18]–[21] . Spatially restricted transcripts in a siphonous organism , and their correspondence with land plant morphology , demonstrate that the plant form is achievable without cells and questions the centrality of cell division patterns in determining plant morphology .
RNA-seq libraries were prepared from at least four replicates of the frond apex , rachis , pinnules , the frond base , stolon , and holdfast , sampling a prolifically growing Caulerpa taxifolia strain obtained from an aquarium in St . Louis , MO . The sampled strain was entrained to a circadian cycle using aquarium lighting roughly synchronized with the outside light-dark cycle . Sampling occurred mid-afternoon , at a time when chloroplasts were enriched in the frond apex ( an important consideration , given the nightly retreat of chloroplasts into the frond base and stolon ) [5] , [22] . Large , intact fragments consisting of fronds , stolons , and holdfasts were removed from the marine aquarium and cleaned in synthetically prepared seawater for approximately 5 seconds to help reduce levels of outside contamination . Different samples corresponded to separately growing clones in the same aquarium . Samples were immediately immersed in liquid nitrogen after cleaning . The samples were then removed from the liquid nitrogen , dissected before they thawed , and placed into microcentrifuge tubes that were immersed again in liquid nitrogen . Samples were then stored at −80°C until library preparation . During thawing before library preparation , one holdfast microcentrifuge sample exploded and was removed from analysis , reducing the holdfast sample number to four . All five samples from other pseudo-organs were successfully processed . Libraries were created using a custom high-throughput method for Illumina RNA-seq with a poly-A enrichment step [15] , and sequenced in 100 bp paired-end format at the UC Berkeley Genomics Sequencing Laboratory on two lanes of HiSeq 2000 platform ( Illumina Inc . San Diego , CA , USA ) . Library making was undertaken exactly as published in Kumar et al . [15] without modification of the protocol . We believe that the freezing step during sample preparation is important to bypass the Caulerpa wounding response for successful RNA isolation . Reads were preprocessed using custom perl scripts that involved removal of low quality reads with average Phred quality score <20 , trimming of low-quality bases with Phred score <20 from the 3′ ends of the reads , and removal of adapter/primer contamination . In addition identical reads , which originated during the PCR enrichment step of the library preparation , were collapsed into a single read using a custom perl script in order to reduce the computational resources required for transcriptome assembly . However , PCR-duplicated reads were retained for downstream quantification of transcript abundances . The pre-processed reads were sorted into individual samples based on barcodes using fastx_barcode_splitter and barcodes were trimmed using fastx_trimmer from Fastx_toolkit ( http://hannonlab . cshl . edu/fastx_toolkit/ ) . A total of 420 million reads ( 210 million paired-end 100×100 ) , obtained after preprocessing , were used for transcriptome assembly . De novo transcriptome assembly was carried in a similar fashion as Ranjan et al . [23] , but is described here again in detail . The Trinity software package ( version r2013-02-16 ) was used to assemble , de novo , a Caulerpa taxifolia transcriptome using preprocessed RNA-seq reads [24] . The assembly was performed , using 24 large-memory computing nodes , at The Lonestar Linux Cluster at Texas Advance Computing Center ( TACC , University of Texas ) . “Trinity . pl —seqType fq —JM 1000G —left reads-1 . fq —right reads-2 . fq —min_contig_length 200 —CPU 24 —bflyHeapSpaceMax 7G” was the command line used for assembly . Subsequently , assembly statistics and downstream analysis were performed in the iPlant atmosphere and Discovery computing atmosphere [25] . A total of 77 , 285 contigs with N50 ( N50 is defined as the largest contig length such that using equal or longer contigs produces half the bases of the transcriptome ) of 1243 bp , mean length of 807 bp and median of 433 bp , were assembled . In order to remove redundant and/or highly similar contigs , the contigs were then clustered using the CD-HIT-EST program from the CD-HIT suite at a sequence similarity threshold value ( -c ) of 0 . 95 and word-length ( n ) of 8 , leaving other parameters at default [26] . This resulted in the final Caulerpa transcriptome assembly with 57 , 118 contigs and N50 of 813 bp , mean length of 632 bp and median of 381 bp ( see sequence submission information ) . The prediction of likely coding sequences from 57 , 118 clustered contigs , using TransDecoder ( http://transdecoder . sourceforge . net/ ) , resulted in 35 , 827 putative open reading frames ( ORFs ) /coding sequences ( CDS ) ( see sequence submission information ) . The contigs from the final Caulerpa transcriptome assembly were compared to the NCBI nr ( nonredundant ) database ( ftp://ftp . ncbi . nlm . nih . gov/blast/db/FASTA/nr . gz ) , Arabidopsis protein database ( ftp://ftp . arabidopsis . org/home/tair/Sequences/blast_datasets/TAIR10_blastsets/TAIR10_pep_20110103_representative_gene_model_updated ) and tomato ( Solanum lycopersicum ) ITAG2 . 3 protein database ( ftp://ftp . solgenomics . net/tomato_genome/annotation/ITAG2 . 3_release/ITAG2 . 3_proteins . fasta ) using BLASTX with an e-value threshold of 1e-5 ( S1 Dataset ) [27] . BLAST searches against the nr database resulted in annotation of 24 , 589 contigs ( 43% of clustered contigs ) of which 20 , 146 contigs had reasonably stringent e-value of less than 1−e10 . 17427 ( 13698 with e-value <1e−10 ) and 17392 ( 13274 with e-value <1e−10 ) contigs were annotated against Arabidopsis and tomato protein databases . When BLASTX comparison was performed against the nr database , more than 49% of annotated C . taxifolia contigs found top BLAST hits against the sequences from members of Cholorophyta , such as Volvox carteri , Chlamydomonas reinhardtii , Chlorella variabilis , and Coccomyxa subellipsoidea . Almost half of the top BLAST hits to cholorophytes confirm high sequence similarity between C . taxifolia and chlorophytes . The Caulerpa contigs were , further , compared explicitly against protein sequences of the chlorophytes C . reinhardtii and V . carteri using BLASTx , and vice-versa using tBLASTn with an e-value threshold of 1e−5 [27] . Chlamydomonas reinhardtii and V . carteri protein sequences were downloaded from Phytozome v10 ( http://phytozome . jgi . doe . gov/pz/portal . html ) . BLAST searches against V . carteri and C . reinhardtii sequences found a hit for 19048 ( 33% ) and 20377 ( 36% ) of assembled Caulerpa contigs , respectively ( S2 Dataset ) . Reciprocal BLAST searches of V . carteri and C . reinhardtii sequences against C . taxifolia contigs found homologs for 42% of sequences of each species . The BLASTX output generated against the NCBI nr database , with maximum twenty hits for each contig , was used for Blast2GO analysis to annotate the contigs with GO terms describing biological processes , molecular functions , and cellular components [28] . Blast2GO performs GO annotation by applying an annotation rule ( AR ) on the found ontology terms from the BLAST-hits , which is based on annotation score . The default e-Value hit filter ( 1e−6 ) and annotation cut-off ( 55 ) was used to calculate the annotation score . Our gene ontology annotations are , by necessity , based in part on the inclusion of hits using a relatively low e-value threshold . It will be critical in the future to validate these assignments using functional analysis . After the Blast2GO mapping process , proper GO terms were generated followed by use of ANNEX and GO Slim , which are integrated in the Blast2GO software , to enrich the annotation ( S3 , S4 Datasets ) . Sequence descriptions were also generated from Blast2GO , which are arbitrary nomenclature based upon degrees of similarities identified in the nr database according to e-value and identity to blasted genes ( S1 , S2 Datasets ) . BLASTX against the nr database resulted in annotation of 24 , 589 contigs ( 43% of clustered contigs ) of which only 14 , 206 ( 25% of clustered contigs ) were assigned GO-terms . Given the problems associated with the de novo transcriptome assembly algorithms and lack of functional tools in Caulerpa , BLASTX annotation of 43% of clustered contigs and GO annotation of only 25% of clustered contigs is not surprising . Similar functional annotation for only a fraction of assembled contigs has been noted for other de novo assembled plant transcriptomes [23] , [29] , [30] . These non-annotated contigs likely correspond to 3′ or 5′ untranslated regions , non-coding RNAs , or short sequences not containing a known protein domain , some of which might represent potential Caulerpa-specific genes . RSEM ( RNAseq by expectation maximization ) , which allows for an assessment of transcript abundances based on the mapping of RNA-seq reads to the assembled transcriptome , was used for transcript abundance estimation of the de novo assembled transcripts [31] . Due to read mapping ambiguity among de novo assembled transcripts , it is common to have the same read mapped to multiple contigs . RSEM models the reads mapped at multiple contigs taking into account length of target contigs , number of mismatches , sequencing errors , etc . , and generates an estimated read count for each contig . Single end reads , retaining the PCR-duplicated reads , from individual libraries of each Caulerpa sample were mapped to clustered contigs using the perl script run_RSEM_align_n_estimate . pl that employs RSEM , followed by joining RSEM-estimated abundance values for each sample using merge_RSEM_frag_counts_single_table . pl , generating raw estimated counts for each contig from each Caulerpa sample ( S5 Dataset ) . Subsequently , differential expression analysis for each organ pair was carried out using run_DE_analysis . pl , which involves the Bioconductor package EdgeR in the R statistical environment [32] . Contigs that had RSEM-estimated counts ≥30 for all samples combined were used for transcript abundance estimates . Normalization factors were calculated using the trimmed mean of M-values method to obtain normalized read count per million for each contig of a sample . This normalized reads per million was then used for the pair-wise differential expression analysis for each organ pair using EdgeR . The lists of significant differentially expressed contigs ( FDR<0 . 05 ) for each organ-pair comparison are presented in S6–S20 Datasets . All the Perl scripts used for read mapping , generating read counts and differential expression analysis are documented with Trinity software suite [33] . Those transcripts differentially expressed between at least one organ pair ( S6–S20 Datasets ) were subsequently used to find clusters of genes with similar transcript accumulation patterns using Self Organizing Maps ( SOMs ) [34] . Differentially expressed transcripts were averaged across replicates for each pseudo-organ sample . Averaged transcript abundance values were then scaled across pseudo-organs to arrive at scaled transcript accumulation patterns which were used to assign cluster membership . Scaling was performed using the scale ( ) function in the base package using default settings , such that the average transcript abundance value across pseudo-organs was 0 and the variance equal to 1 . To cluster transcripts across organs , a 3×2 hexagonal SOM was implemented , using the Kohonen package in R [35] , [36] . 100 training iterations were used during clustering , over which the α-learning rate decreased from 0 . 05 to 0 . 01 . Mean distance of transcript accumulation patterns to their closest unit stabilized after approximately 15 iterations of training . A decision to use six clusters was arrived at by first analyzing the results of a Principal Component Analysis ( PCA ) on scaled transcript accumulation across tissues , using the prcomp ( ) function in R with default settings . Average and scaled transcript accumulation levels across organs , as well as SOM cluster membership and PC values are provided ( S21 Dataset ) . Four main densities in the variance attributable to accumulation patterns were discernable ( arrows , Fig . 2A ) , and the results of a 4 cluster SOM largely overlap with the densities . Variance in accumulation among transcripts across organs was large , however , and the decision to specify 6 SOM clusters not only produced clusters with unique accumulation patterns and lower variance in abundance ( Fig . 2C ) , but also yielded clusters with more interpretable GO enrichment categories ( that is , significant GO enrichment consistent with known biology , such as photosynthetic GOs enriched in nodes with high pinnule transcript abundance ) . To verify that 6 GOs was indeed the maximum cluster number specifying unique transcript accumulation profiles without redundancy , we undertook partitioning of the PCA space into a variable number of SOM clusters over 100 iterations for each node number with random seeds . Linear Discriminant Analysis ( LDA ) was performed on genes maximizing separation of cluster identity using PCs 1–5 ( PC6 explained negligible amounts of variance and could not be incorporated into the LDA ) using the lda function from the MASS package [37] . The predict function ( stats package ) and table function ( base package ) were used to reallocate genes to predicted clusters ( within MASS ) using the linear discriminants . A high fraction of a node's originally assigned transcripts by SOM being reassigned correctly indicates little redundancy in node transcript accumulation patterns . Starting with 6 nodes , reassignment using LDA begins to drop before reaching a plateau of low reassignment rates , indicating that choosing 6 nodes maximizes the number of unique accumulation profiles represented by clusters without redundancy ( see S1 Fig . for results ) . Analysis of intersection between tomato and Caulerpa transcriptomic atlases was undertaken using data from Chitwood et al . [17] . Best BLASTX hits of Caulerpa transcripts to tomato ( Solanum lycopersicum ) ( see “Functional annotation of transcriptome” above and S1 , S2 Datasets ) were used to assign tomato transcript accumulation patterns , across a number of organs , to Caulerpa transcript accumulation patterns . The distribution of tomato transcripts assigned to tomato SOM nodes was taken as the null distribution and compared to the number of Caulerpa transcripts assigned to each tomato node . p values , indicating the degree of significant difference between the two distributions , were obtained from χ2 values using the chisq . test function ( stats package ) . Clusters of transcripts were analyzed for GO enrichment terms at a 0 . 05 FDR cut-off using the “goseq” package in Bioconductor ( S22–S28 Datasets ) [38] . Unless otherwise specified , all statistical analyses on transcript accumulation were performed using R [36] and data visualization using the ggplot2 package [39] . The quality filtered , barcode-sorted and trimmed short read dataset , which was used for transcriptome assembly , was deposited to the NCBI Short Read Archive under accessions SRR1228213–SRR1228223 , SRR1228225–SRR1228232 , SRR1228234–SRR1228238 and . SRR1228240–SRR1228244 . All assembled contigs have been deposited at DDBJ/EMBL/GenBank under the accession GBCY00000000 . The version described in this paper is the first version , GBCY01000000 . Sequences of all contigs of Caulerpa_final_transcriptome , obtained after clustering of transcripts , can be downloaded as a FASTA file at http://de . iplantcollaborative . org/dl/d/80CF0D47-5A80-4CE7-B6DF-F4A7ED803493/Caulerpa_final_transcriptome . fasta . The contigs were named as Ctaxi_contig plus a serial number with the Trinity identifiers . Sequences of all predicted ORFs from the Caulerpa transcriptome assembly can be downloaded as a FASTA file at http://de . iplantcollaborative . org/dl/d/40273882-35DD-4930-9DBD-6D60CEAA7890/Caulerpa_predicted_ORFs . fasta . The ORFs were named as Ctaxi_predicted_CDS plus a serial number . | Plants include both the green algae and land plants . Multiple times , root , stem , and leaf-like structures arose independently in plant lineages . In some instances , such as the siphonous algae , these structures arose in the absence of multicellularity . It has been argued by some that the morphology of multicellular land plant organs similarly arises independently of cell division patterns . Here , we explore the partitioning of gene transcripts within what is debatably the largest single-celled organism in the world , the siphonous alga Caulerpa taxifolia . We find that within this giant cell specific transcripts localize within pseudo-organs ( morphological structures that are not comprised of cells or tissue ) . The overall pattern of transcript accumulation follows an apical-basal pattern within the cell . Moreover , transcripts related to different cellular processes , such as transcription and translation , localize to specific regions . Analyzing the signatures of transcript accumulation in land plant organs and the pseudo-organs of Caulerpa , we find that groups of transcripts accumulate together in morphological structures across evolution at rates higher than expected by chance . Together , our results demonstrate a relationship between transcript partitioning and organism morphology , independent from multicellularity , throughout diverse plant lineages . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
]
| [
"organismal",
"evolution",
"plant",
"growth",
"and",
"development",
"plant",
"cell",
"biology",
"marine",
"biology",
"developmental",
"biology",
"plant",
"science",
"plant",
"morphology",
"molecular",
"cell",
"biology",
"organism",
"development",
"morphogenesis",
"cellular",
"structures",
"and",
"organelles",
"pattern",
"formation",
"phycology",
"plant",
"development",
"subcellular",
"localization",
"organogenesis",
"molecular",
"biology",
"plant",
"cells",
"cell",
"biology",
"plant",
"evolution",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"evolutionary",
"biology",
"evolutionary",
"developmental",
"biology"
]
| 2015 | An Intracellular Transcriptomic Atlas of the Giant Coenocyte Caulerpa taxifolia |
Cellular senescence is a driver of various aging-associated disorders , including osteoarthritis . Here , we identified a critical role for Yes-associated protein ( YAP ) , a major effector of Hippo signaling , in maintaining a younger state of human mesenchymal stem cells ( hMSCs ) and ameliorating osteoarthritis in mice . Clustered regularly interspaced short palindromic repeat ( CRISPR ) /CRISPR associated protein 9 nuclease ( Cas9 ) -mediated knockout ( KO ) of YAP in hMSCs resulted in premature cellular senescence . Mechanistically , YAP cooperated with TEA domain transcriptional factor ( TEAD ) to activate the expression of forkhead box D1 ( FOXD1 ) , a geroprotective protein . YAP deficiency led to the down-regulation of FOXD1 . In turn , overexpression of YAP or FOXD1 rejuvenated aged hMSCs . Moreover , intra-articular administration of lentiviral vector encoding YAP or FOXD1 attenuated the development of osteoarthritis in mice . Collectively , our findings reveal YAP–FOXD1 , a novel aging-associated regulatory axis , as a potential target for gene therapy to alleviate osteoarthritis .
Mesenchymal stem cells ( MSCs ) are widely distributed in adult tissues and have the capacities of self-renewal and differentiation into multiple cell lineages , such as chondrocytes , osteoblasts , and adipocytes [1] . MSCs are involved in tissue repair and homeostatic maintenance [2 , 3] . Over time , MSCs exhibit an age-associated decline in their number and function [4–6] , namely , MSC senescence , which may be implicated in the loss of tissue homeostatic maintenance and leads to organ failure and degenerative diseases [7–10] . Therefore , an understanding of the mechanisms underlying MSC senescence will likely reveal novel therapeutic targets for ameliorating degenerative diseases . Osteoarthritis is a prevalent aging-associated disorder that is characterized by the progressive deterioration of articular cartilage [11 , 12] . In osteoarthritis joints , degenerative changes start with cellular disorganization , gradual stiffening , and irregular surface of superficial zone followed by loss of matrix , clefts , and osteophyte formation in the deep articular cartilage [13 , 14] . Accordingly , disruption of the superficial zone of cartilage is an onset of osteoarthritis . Previous reports have demonstrated that cells isolated from the superficial zone of mouse and human articular cartilage express MSC markers , including cluster of differentiation ( CD ) 105 , CD166 , CD29 , and exhibit MSC characteristics [15–20] . Cell death induced by oxidative stress or wound occurs primarily at the surface zone of cartilage [21 , 22] . When such cell death is inhibited by chemicals , cartilage disorganization and matrix loss are greatly reduced [23] . Therefore , MSCs or chondrocyte progenitor cells residing in the superficial zone of cartilage may be a critical target for the prevention of osteoarthritis . Although the transplantation of ex vivo cultures of MSCs into the osteoarthritic joint has been shown to improve the symptoms [24–26] , the rejuvenation of endogenous senescent MSCs may also be a therapeutic option for osteoarthritis . The localized nature of osteoarthritis , which has no major extra-articular or systemic manifestations , makes it an ideal candidate for local , intra-articular gene therapy [27 , 28] . However , gene therapy strategies aiming at alleviating senescence , particularly MSC senescence , for treating osteoarthritis have not yet been reported . Yes-associated protein ( YAP ) and transcriptional coactivator with PDZ-binding motif ( TAZ ) are primary targets of the Hippo signaling pathway , which plays important roles in the regulation of development , homeostasis , regeneration , and so forth [29–31] . The Hippo kinase cascade phosphorylates YAP and TAZ , resulting in their cytoplasmic retention and proteolytic degradation . When the Hippo pathway is inactive , YAP and TAZ translocate into the nucleus and interact with transcription factors to regulate the expression of target genes [32] . YAP and TAZ , as paralogs , have been demonstrated as key regulators in organ size control [33] and essential transducers of mechanical signals [34] . Here , we identified a critical role for YAP , but not TAZ , in regulating human MSC ( hMSC ) senescence . YAP exerted a geroprotective effect on hMSCs through the transcriptional activation of forkhead box D1 ( FOXD1 ) in a TEA domain transcription factor ( TEAD ) -dependent manner . Gene therapy with lentiviral vectors encoding YAP or FOXD1 prevented cellular aging and attenuated osteoarthritis in mice . Our data suggest that YAP and its downstream target FOXD1 are novel suppressors of hMSC senescence and that the YAP–FOXD1 regulatory axis represents a potential therapeutic target for osteoarthritis .
We first used Clustered regularly interspaced short palindromic repeat ( CRISPR ) /CRISPR-associated protein 9 nuclease ( Cas9 ) -mediated gene editing [35] to generate isogenic human embryonic stem cells ( hESCs ) lacking YAP or TAZ to study the functions of YAP and TAZ in regulating human stem cell homeostasis ( S1A , S1B and S1G Fig ) . Successful gene targeting at the YAP or TAZ locus were verified by genomic polymerase chain reaction ( PCR ) and reverse transcription quantitative PCR ( RT-qPCR ) ( S1C and S1D Fig ) . Western blot and immunofluorescence further confirmed the complete ablation of YAP and TAZ protein in YAP−/− and TAZ−/− hESCs , respectively , with no detectable compensations between YAP and TAZ ( Fig 1A , 1B and 1C ) . Both YAP−/− and TAZ−/− hESCs maintained normal pluripotency ( Fig 1D ) and cell cycle kinetics ( Fig 1E ) and were able to differentiate into tissues composed of all 3 germ layers in vivo ( S1E Fig ) . Karyotype and genome-wide copy number variation ( CNV ) analyses demonstrated that genomic integrity was maintained in YAP−/− and TAZ−/− hESCs after more than 30 passages ( Figs 1F and S1F ) . Moreover , YAP−/− and TAZ−/− hESCs displayed transcriptional profiles that were highly similar to wild type ( WT ) hESCs ( Fig 1G ) . We next differentiated WT , YAP−/− , and TAZ−/− hESCs into hMSCs ( Fig 2A ) [35–38] . The derived hMSCs expressed a series of hMSC markers including CD105 , CD166 , CD29 , CD90 , CD73 , CD44 , CD13 , and human leukocyte antigens , A , B and C ( HLA-ABC ) and were negative for hematopoietic or skeletal lineage markers CD34 , CD43 , CD45 , CD14 , CD19 , podoplanin ( PDPN ) , and CD164 , resembling the resident CD105+ , CD166+ , and CD29+ MSCs in the superficial zone of articular cartilage ( Figs 2A and S2A ) [39–41] . Whereas WT hMSCs were able to differentiate into chondrocytes , osteoblasts , and adipocytes , YAP−/− and TAZ−/− hMSCs exhibited compromised differentiation abilities into osteoblasts and chondrocytes ( S2B–S2E Fig ) . Additionally , the growth rates of WT , YAP−/− , and TAZ−/− hMSCs were analyzed through in vitro serial passaging . Compared with WT and TAZ−/− hMSCs , YAP-deficient hMSCs exhibited early-onset aging characteristics and arrested at passage 6 ( Fig 2B ) . Increased levels of senescence-associated–β-galactosidase ( SA-β-gal ) activity ( Fig 2C ) and increased expression of P16 , P53 , and P21 ( Fig 2D ) were detected at as early as passage 4 . Concomitantly , YAP−/− hMSCs showed a series of premature phenotypes , including ( 1 ) a decreased percentage of cells in synthesis ( S ) phase and increased percentages of cells in gap phases ( G0 , G1 , and G2 ) and mitosis ( M ) phase ( Fig 2E ) , ( 2 ) a lower percentage of Ki67-positive cells ( S2F and S2G Fig ) , ( 3 ) decreased levels of the nuclear lamina-associated protein 2 ( LAP2; S2F and S2G Fig ) , ( 4 ) reduced levels of heterochromatin protein 1 alpha ( HP1α ) and heterochromatin protein 1 gamma ( HP1γ; S2F and S2G Fig ) , and ( 5 ) higher levels of reactive oxygen species ( ROS; S3A Fig ) , compared with WT and TAZ−/− hMSCs . We subsequently examined whether the YAP deficiency resulted in stem cell attrition in vivo . WT , YAP−/− , and TAZ−/− hMSCs were transduced with a lentiviral vector expressing luciferase ( Luc ) and injected into the tibialis anterior ( TA ) muscles of immunodeficient mice . Consistent with the in vitro observations , YAP−/− hMSCs , but not TAZ−/− or WT hMSCs , exhibited an accelerated functional decay after transplantation in vivo ( Figs 2F and 2G and S3B and S3C ) . To further evaluate the effect of YAP in hMSCs , WT hMSCs were transduced with lentiviruses encoding a single guide RNA ( sgRNA ) targeting YAP or a non-targeting control ( NTC ) sgRNA , as well as CRISPR/Cas9 [42 , 43] . Phenotypic characterizations revealed that the down-regulation of YAP in hMSCs also resulted in a similar premature aging phenotype ( S3D and S3E Fig ) . By contrast , ectopic expression of YAP rescued the premature senescence observed in YAP−/− hMSCs , as evidenced by the reduced percentage of SA-β-gal–positive cells ( Fig 2H ) , enhanced growth rate and clonal expansion ability ( S3F and S3G Fig ) , decreased expression of P16 and P21 ( S3H Fig ) , lower levels of ROS ( S3I Fig ) , and slower in vivo decay after engraftment ( Figs 2I and S3J ) . Taken together , these data suggest that YAP , but not TAZ , plays an essential role in protecting hMSCs from premature senescence . Given that YAP and TAZ displayed distinct functions in regulating hMSC senescence , we next examined whether there were differences in the subcellular localizations of YAP and TAZ . In hMSCs , YAP was predominantly located in the nucleus , whereas TAZ was in the cytoplasm ( Fig 3A ) . It has been shown that nuclear YAP binds transcription factors , including TEAD family of transcription factors ( TEAD1 , 2 , 3 , and 4 ) , as a transcriptional coactivator to induce target gene expression and thus regulate a series of cellular processes [44] . To test whether the nuclear YAP acted in conjunction with TEAD to regulate hMSC senescence , we blocked the activities of all the members of TEAD family in hMSCs as confirmed by immunoblotting analysis ( referred to as TEADs knockdown [KD] and KO hMSCs , TEADs KD/KO hMSCs; Fig 3B ) . Similar to YAP-deficient hMSCs , TEADs KD/KO hMSCs also showed major phenotypes of premature senescence , such as an increased number of SA-β-gal–positive cells ( Fig 3C ) , compromised clonal expansion abilities ( S4A Fig ) , and up-regulation of P16 and P21 ( S4B Fig ) . These observations suggest that YAP safeguards hMSCs from premature senescence in a TEAD-dependent manner . To elucidate the molecular mechanism underlying YAP–TEAD regulation of hMSC senescence , RNA sequencing ( RNA-seq ) analyses of WT , YAP−/− , and TAZ−/− hMSCs were performed ( S4C Fig ) . TAZ−/− hMSCs displayed comparable transcriptional features compared to those of WT hMSCs , whereas YAP−/− hMSCs exhibited a substantial number of differentially expressed genes ( Figs 3D and S4E and S2 and S3 Datas ) . We observed few overlaps between differentially expressed genes in YAP KO and TAZ KO hMSCs compared to WT cells ( S4D Fig ) , consistent with differential subcellular localization patterns of YAP and TAZ in hMSCs . Searching for TEAD binding motifs in the genome [45 , 46] identified 476 ( 55% ) of the 862 down-regulated genes in YAP−/− hMSCs as potential TEAD targets ( P < 1 . 0 × 10−4; Fig 3E ) . Among them , FOXD1 was the most significantly down-regulated gene in YAP−/− hMSCs ( S4 Data ) . Western blotting verified the down-regulation of FOXD1 expression in YAP−/− hMSCs ( Fig 3F ) as well as its up-regulation upon the reintroduction of YAP ( Fig 3G ) , suggesting that FOXD1 was transcriptionally controlled by YAP . We examined the FOXD1 promoter region , including 1 , 500 bp upstream of the transcriptional start site ( TSS ) and identified 4 putative TEAD binding sites between −1 , 500 and −1 , 000 bp and 1 between −1 , 000 bp and the TSS ( Fig 3H ) . Accordingly , we detected these 2 regions followed by chromatin immunoprecipitation ( ChIP ) using YAP and TEAD4 antibodies , revealing that YAP and TEAD4 bound predominantly within 1 , 000 bp upstream of the FOXD1 TSS , where there was a putative TEAD binding site ( Fig 3I and 3J ) . Next , we cloned this promoter region ( −1 , 000 bp to the TSS ) as a transcriptional element upstream of a basic Luc reporter . Reporter activity was lower in YAP−/− hMSCs than in WT cells ( Fig 3K ) and was increased upon YAP or TEAD4 overexpression . Luc activity was even higher upon the expression of a constitutively activated YAP mutant ( YAP-S127A ) and was further enhanced by coexpression of YAP and TEAD4 ( Fig 3L ) . The high levels of Luc activity were significantly abolished when we mutated the predicted TEAD binding site ( Fig 3M ) . By contrast , ChIP assay demonstrated that TAZ did not bind to the FOXD1 promoter ( S4F Fig ) , and the Luc activity was insensitive to cellular TAZ levels ( Figs 3K and S4G ) . Therefore , the YAP–TEAD pathway , but not TAZ , transcriptionally activates FOXD1 expression . FOXD1 was initially implicated in renal development [47] , but there was a lack of evidence for a link between FOXD1 and cellular senescence . To investigate whether FOXD1 participated in YAP deficiency-induced accelerated senescence of hMSCs , we knocked out FOXD1 in hMSCs using a lentiviral vector-dependent CRISPR/Cas9 system [42 , 43] ( Figs 4A and S5A ) . FOXD1 depletion in hMSCs increased the percentage of SA-β-gal–positive cells ( Fig 4B ) , inhibited clonal expansion ( S5B Fig ) , and up-regulated P16 and P21 expression ( S5C Fig ) , recapitulating the major phenotypes implicated in premature senescence caused by the YAP deficiency . The overexpression of FOXD1 in YAP−/− hMSCs effectively alleviated the accelerated senescence ( Fig 4C ) . In addition , we also examined the gene expression profile of FOXD1 KO hMSCs using RNA-seq ( S5 Data ) . FOXD1 KO decreased the expression of genes that were mainly associated with cell division and DNA replication , which ultimately contributed to the senescence phenotypes ( Fig 4D and S6 Data ) . Combined analyses with WT and YAP−/− hMSCs showed that FOXD1-deficient and YAP−/− hMSCs were similar to each other at the transcriptomic level ( S5D and S5E Fig ) . Many differentially expressed genes were overlapped between YAP−/− ( compared to WT ) and FOXD1 KO ( compared to NTC-transduced ) hMSCs , including 116 up-regulated genes accounting for 20% of the total up-regulated genes in YAP−/− hMSCs and 276 down-regulated genes accounting for 32% of the total down-regulated genes in YAP−/− hMSCs ( Fig 4E and 4F ) , implying an important role for FOXD1 in mediating YAP deficiency-induced premature cellular aging . Of note , many of those commonly down-regulated genes were elevated upon ectopic expression of FOXD1 in YAP−/− hMSCs ( Fig 4G ) . Conversely , ectopic expression of YAP in FOXD1 KO or TEADs KD/KO hMSCs did not exert obvious rescue effect on the senescence phenotypes ( S5F and S5G Fig ) . Taken together , these data indicate that down-regulation of FOXD1 , an effector of YAP–TEAD signaling , contributes to the premature senescence induced by YAP deficiency . To further elucidate the relationship between the YAP–FOXD1 axis and human stem cell aging , we examined the expression levels of YAP , pan-TEAD , and FOXD1 in both replicative-senescent ( RS ) hMSCs and Werner syndrome ( WS ) hMSCs , a human stem cell model for premature aging disorder WS [37 , 48] . Western blotting revealed decreased levels of YAP , pan-TEAD , and FOXD1 in both types of senescent hMSCs ( Fig 5A and 5D ) . Moreover , the activity of 8 × GTIIC-Luc , a YAP/TAZ-responsive reporter , decreased in both RS hMSCs and WS hMSCs ( Fig 5B and 5E ) . Lentiviral overexpression of YAP or FOXD1 effectively attenuated the senescent features of RS hMSCs ( Figs 5C and S6A–S6C ) and WS hMSCs ( Figs 5F and S6D–S6G ) . We also observed diminished protein levels of YAP and FOXD1 in RS-primary hMSCs isolated from human bone marrow ( BM-hMSCs ) ( Fig 5G ) . In BM-hMSCs , KO of YAP or FOXD1 with a CRISPR/Cas9 system promoted cellular senescence ( Fig 5H–5K ) , whereas the overexpression of YAP or FOXD1 delayed BM-hMSC senescence ( Fig 5L and 5M ) . Collectively , these observations establish a geroprotective role for the YAP–FOXD1 axis in alleviating hMSC aging . Mesodermal cellular aging has emerged as a fundamental hallmark of aging-related disorders , including osteoarthritis , one of the most common degenerative diseases , the incidence of which increases significantly with age . Dysfunction of MSCs residing in the superficial zone of cartilage precedes osteoarthritis [15–19 , 21 , 22] that is characterized by articular cartilage degradation [49–51] . To validate a role of hMSC senescence in driving osteoarthritis , we injected young hMSCs , RS hMSCs , and RS hMSCs overexpressing YAP or FOXD1 , respectively , into the joints of immunodeficient mice and performed histological assessment of the joints 1 month later ( S7A Fig ) . In line with a previous report [52] , Safranin O staining revealed the delamination of the articular surface and erosion of articular cartilage in the RS hMSC–administrated joints ( S7B and S7C Fig ) . However , no osteoarthritis-related features manifested in the joints transplanted with young hMSCs or RS hMSCs overexpressing YAP or FOXD1 . RT-qPCR further demonstrated that RS hMSCs , rather than young hMSCs , and RS hMSCs overexpressing YAP or FOXD1 induced aging markers in mouse joints ( S7D Fig ) . These results suggest that accumulation of senescent MSCs in joints contributes to the development of osteoarthritis , which can be eliminated by YAP or FOXD1 overexpression . The elimination of local senescent cells using pharmacological or genetic approaches have been proven effective in attenuating age-associated bone loss and development of post-traumatic osteoarthritis in rodents [51 , 53] . Given the ability of YAP or FOXD1 to rejuvenate senescent MSCs , we hypothesized that intra-articular injection of lentiviral vectors expressing YAP or FOXD1 might exert a therapeutic effect on osteoarthritis . To test this , we performed an anterior cruciate ligament transection ( ACLT ) surgery widely used to trigger osteoarthritis in mice and then administrated the lentiviruses expressing flag-tagged Luc , YAP , or FOXD1 intra-articularly ( Fig 6A ) . The lentiviral vectors steadily expressed exogenous proteins in and around the joints receiving virus injection for at least 7 weeks ( S8A Fig ) . High expression levels of YAP and FOXD1 were detectable by RT-qPCR ( Fig 6B and 6C ) ; immunohistochemical analysis of the flag-tagged Luc , YAP , and FOXD1 further verified the persistent infection of the lentiviruses and expression of indicated proteins primarily in the superficial zone of articular cartilage ( S8B Fig ) . As expected , ACLT induced the accumulation of P16-positive senescent cells in the articular cartilage , particularly in the superficial zone of cartilage , of the osteoarthritis mice ( Fig 6D ) , which was accompanied by decreased levels of YAP and FOXD1 ( S8C and S8D Fig ) . YAP or FOXD1 gene therapy reduced the number of senescent cells and alleviated ACLT-induced articular cartilage erosion and clefts ( Fig 6D and 6E ) . Consistently , a substantial proportion of gene expression changes in the joints induced by ACLT were reversed by YAP or FOXD1 gene therapy ( S8C–S8F Fig and S7 Data ) . For instance , increased expression of genes associated with inflammation ( Mmp13 , Il6 , etc . ) , cellular senescence ( P21 , Serpine1 , etc . ) , and cell apoptosis ( Dapk1 , Casp4 , etc . ) were observed in ACLT-induced osteoarthritis joints , and the expression levels of most of these genes were diminished upon YAP or FOXD1 treatment . Moreover , the YAP or FOXD1 treatment enhanced the expression of proliferation markers ( Ki67 , Aspm , etc . ) and chondrocyte differentiation-related genes ( Col2a1 , Acan , etc . ) ( Fig 6F ) . Taken together , these data suggest that the YAP- or FOXD1-mediated alleviation of cellular senescence in local bone joints helps create a prochondrogenic environment and alleviates disease symptoms .
Cellular senescence and stem cell exhaustion are hallmarks of aging [54] . Accelerated attrition of the MSC pool has been observed in human stem cell and mouse models of premature aging disorders , including WS and Hutchinson Gilford progeria syndrome ( HGPS ) [37 , 55] . Transplantation of mesoderm-derived stem cells from young animals increases the lifespan of progeroid mice [56] . Quercetin has been shown to alleviate MSC senescence [57] , improve physical function , and increase lifespan in aged mice [58] . From this perspective , senescent MSCs could be good therapeutic targets for aging-associated degenerative disorders . In this study , we presented several lines of evidence supporting a geroprotective role of YAP and FOXD1 in rejuvenating hMSCs: ( 1 ) YAP is required for preventing premature senescence of hMSCs; ( 2 ) YAP transcriptionally activates FOXD1 expression whereas YAP deficiency results in down-regulation of FOXD1 , which contributes to the early-onset of cellular aging; and ( 3 ) lentiviral gene transfer of YAP or FOXD1 alleviates cellular senescence and osteoarthritis . Our findings define a critical role of the YAP–FOXD1 axis in regulating hMSC aging , which highlights new avenues for translation into geriatric and regenerative medicine . The Hippo-YAP/TAZ signaling pathway is an evolutionarily conserved pathway that regulates cell proliferation and apoptosis . Here , we focused on the function of YAP and/or TAZ in regulating hMSC senescence . We generated isogenic YAP- or TAZ-deficient hMSCs . Compared with WT cells , TAZ−/− hMSCs showed minimal effect in cell growth , whereas YAP−/− hMSCs exhibited accelerated senescence . The cytoplasmic localization of TAZ underlays its inactivation in hMSCs , whereas nuclear YAP was essential for counteracting hMSC senescence . Consistent with our observations , emerging studies have revealed the differences between YAP and TAZ . For example , TAZ promotes the myogenic differentiation of myoblasts at late stages of myogenesis , whereas YAP inhibits this process in mice [59] . However , in-depth insights into the molecular mechanisms underlying these functional differences require further investigations . FOXD1 is a new downstream target of YAP , loss of which mediates the senescent phenotype of YAP-deficient hMSCs . As a member of the forkhead box family of transcription factors , FOXD1 is known to regulate kidney development during organogenesis [60 , 61] . Recently , FOXD1 has been shown to promote cell proliferation by targeting the sonic hedgehog pathway and cyclin-dependent kinase inhibitors [62 , 63] . FOXD1 also facilitates the reprogramming of mouse embryonic fibroblasts ( MEFs ) into induced pluripotent stem cells ( iPSCs ) [64] . Here , we identified a geroprotective role for FOXD1 as a transcriptional target of YAP in rejuvenating hMSCs . Overexpression of YAP or FOXD1 delayed replicative and pathological senescence , implying a therapeutic potential of targeting the YAP–FOXD1 axis to relieve aging-associated degenerative diseases . In a therapeutic context , we provided a proof-of-concept evidence that intra-articular lentiviral transduction of a single protein exerted therapeutic effects on ACLT-induced osteoarthritis , an age-related disorder . Because ACLT-induced osteoarthritis is accompanied by the accumulation of senescent cells [65] , efforts have been made on chemical-induced elimination of senescent cells to alleviate osteoarthritis in mouse models [50 , 51] . With gene therapy offering novel therapeutic options for osteoarthritis [66] , intra-articular injection presents a minimally invasive procedure that avoids conventional barriers to joint entry , increases bioavailability , and lowers systemic toxicity [67] . For the first time , our study shows that the intra-articular injection of lentiviruses expressing YAP or FOXD1 reduces the number of senescent cells , inhibits articular inflammation and cartilage erosion , and ameliorates the pathological symptoms . Therefore , gene therapy via the introduction of geroprotective factors aiming at rejuvenating senescent cells may represent a new avenue to treating osteoarthritis in the future .
All animal experiments were conducted in compliance with animal protocols approved by the Chinese Academy of Science Institutional Animal Care and Use Committee , licensed by the Science and Technology Commission of Beijing Municipality ( SYXK-2016-0026 ) . All mice were housed under a 12-hour light/dark cycle at constant temperature ( 22°C ) . Food and water were available ad libitum . Mice were anaesthetized using isoflurane and euthanized with CO2 followed by cervical dislocation . Human H9 ESCs as well as derived YAP−/− and TAZ−/− hESCs were maintained on feeder layers of mitomycin C–inactivated MEFs in hESC medium [68] ( DMEM/F12 [Thermo Fisher Scientific , Waltham , MA] , 20% Knockout Serum Replacement [Thermo Fisher Scientific] , 0 . 1 mM nonessential amino acids [NEAAs; Thermo Fisher Scientific] , 2 mM GlutaMAX [Thermo Fisher Scientific] , 1% penicillin/streptomycin [Thermo Fisher Scientific] , 55 μM β-mercaptoethanol [Thermo Fisher Scientific] , and 10 ng/ml bFGF [Joint Protein Central , Incheon , Korea] ) or on Matrigel ( BD Biosciences , San Jose , CA , USA ) in mTeSR medium ( STEMCELL Technologies , Vancouver , Canada ) . hESCs derived hMSCs and BM-hMSCs ( purchased from Lonza , Basel , Switzerland ) were cultured in hMSC medium ( αMEM + GlutaMAX [Thermo Fisher Scientific] , 10% fetal bovine serum [Gibco , Cat: 10099–141 , Lot: 1616964] , 1% penicillin/streptomycin [Thermo Fisher Scientific] , and 1 ng/ml bFGF [Joint Protein Central] ) . No mycoplasma contamination was observed during cell culture . Antibodies used for western blotting , immunostaining and flow cytometry included anti-YAP ( 15407; Santa Cruz Biotechnology , Santa Cruz , CA ) , anti-YAP ( 52771; Abcam , Cambridge , MA ) , anti-TAZ ( 4883; Cell Signaling Technology , Danvers , MA ) , anti-GAPDH ( 25778; Santa Cruz Biotechnology ) , anti-P16 ( 550834; BD ) , anti-P21 ( 2947s; Cell Signaling Technology ) , anti-P53 ( 1101; Abcam ) , anti-β-tubulin ( 5274; Santa Cruz Biotechnology ) , anti-β-actin ( 69879; Santa Cruz Biotechnology ) , anti-Pan TEAD ( 13295; Cell Signaling Technology ) , anti-FOXD1 ( PA5-27142; Thermo Fisher Scientific ) , anti-NANOG ( 21624; Abcam ) , anti-OCT3/4 ( 5279; Santa Cruz Biotechnology ) , anti-SOX2 ( 17320; Santa Cruz Biotechnology ) , anti-TUJ1 ( T2200; Sigma , St . Louis , MO ) , anti-FOXA2 ( 8186; Cell Signaling Technology ) , anti-SMA ( A5228; Sigma ) , anti-Ki67 ( VP-RM04; Vector Labs , Burlingame , CA ) , anti-LAP2 ( 611000; BD ) , anti-HP1α ( 2616S; Cell Signaling Technology ) , anti-HP1γ ( 2619; Cell Signaling Technology ) , anti-Aggrecan ( AF1220; R&D , Minneapolis , MN ) , anti-Osteocalcin ( MAB1419; R&D ) , anti-FABP4 ( AF3150; R&D ) , anti-CD73 ( 550741; BD Bioscience ) , anti-CD90 ( 555595; BD Bioscience ) , anti-CD105 ( 17–1057; eBioscience , San Diego , CA ) , anti-CD29 ( 303004; Biolegend , San Diego , CA ) , anti-CD44 ( 550989; BD Bioscience ) , anti-CD13 ( 301705; Biolegend ) , anti-CD166 ( 343903; Biolegend ) , anti-HLA-ABC ( 560168; BD Bioscience ) , anti-CD34 ( 555822; BD Biosciences ) , anti-CD43 ( 580198; BD Biosciences ) , anti-CD45 ( 555482; BD Biosciences ) , anti-CD14 ( 555398; BD Biosciences ) , anti-CD19 ( 555415; BD Biosciences ) , anti-PDPN ( 17-9381-41; eBioscience ) , and anti-CD164 ( 324805; Biolegend ) . CRISPR/Cas9-mediated gene targeting was performed using previously described methods , with some modifications [69] . The YAP or TAZ gRNA was cloned into the gRNA vector ( Addgene #41824 ) . The donor plasmid for homologous recombination containing homology arms and a neo cassette was described previously [70] . Briefly , 5 × 106 H9 ESCs were mixed with the plasmid cocktail and electroporated . After electroporation , cells were plated on a G418-resistant MEF feeder layer . Two days later , cells were treated with 100 μg/ml G418 ( Gibco , 10131027 ) for screening . After 2 weeks of selection , G418-resistant clones were manually picked , transferred to 96-well plates , and expanded for genotyping . Gene-targeted clones were identified using genomic PCR . gRNA sequences and primers are listed in S1 Data . hMSCs were differentiated from hESCs as previously described [70–72] . Briefly , hESCs were dissociated into embryoid bodies and then plated on Matrigel-coated plates in differentiation medium ( α-MEM + GlutaMAX [Thermo Fisher Scientific] , 10% FBS [Gibco , Cat: 10099–141 , Lot: 1616964] , 1% penicillin/streptomycin [Thermo Fisher Scientific] , 10 ng/ml bFGF , and 5 ng/ml TGFβ [HumanZyme , Chicago , IL] ) . After 10 days , the confluent MSC-like cells were passaged to gelatin-coated plates and sorted by FACS to purify CD73/CD90/CD105 triple-positive hMSCs , which were further characterized by flow cytometry analysis of the surface antigens , including CD166 , CD29 , CD44 , CD13 , HLA-ABC , CD34 , CD43 , CD45 , CD14 , CD19 , PDPN , and CD164 . The functionality of hMSCs was verified by differentiation to osteoblasts , chondrocytes , and adipocytes . Lentiviral CRISPR/Cas9-mediated gene editing was performed as previously described [42] . Briefly , the sgRNA targeting YAP or FOXD1 was cloned into lentiCRISPRv2 vector ( Addgene #52961 ) , which contains 2 expression cassettes , hSpCas9 and the chimeric sgRNA . Then , the plasmids were packaged into lentiviruses and transduced into hMSCs; 72 hours later , transduced cells were treated with 1 μg/ml puromycin ( Gibco , A1113803 ) for enriching transduced cells . The sgRNA sequences are listed in S1 Data . We generated 2 lentiviral constructs to silence the expression of TEAD1 , 2 , 3 , and 4: one containing an shRNA targeting TEAD1 , TEAD3 , and TEAD4 and the other containing an sgRNA targeting TEAD2 [73] . The shRNA was cloned into the PLVTHM vector ( Addgene #12247 ) , and the sgRNA was cloned into lentiCRISPRv2 . We then cotransduced these lentiviral constructs into hMSCs . Seventy-two hours later , transduced cells were enriched by treatment with 1 μg/ml puromycin ( Gibco , A1113803 ) . The targeting sequences are listed in S1 Data . For packaging of the lentivirus , HEK293T cells were cotransfected with lentiviral vectors , psPAX2 ( Addgene #12260 ) and pMD2 . G ( Addgene #12259 ) . Viral particles were collected by ultracentrifugation at 19 , 400 g for 2 . 5 hours . For the cell cycle analysis , the Click-iT EdU Alexa Fluor 647 Flow Cytometry Assay Kit ( C-10419; Molecular Probes , Eugene , OR ) was used according to the manufacturer’s instructions . For ROS measurements , living cells were incubated with ROS indicators ( 1 μM CM-H2DCFDA , C6827; Molecular Probes ) . All experiments were measured with an LSRFortessa cell analyzer ( BD ) , and data were analyzed using FlowJo software ( TreeStar , Ashland , OR ) . Cells were fixed with 4% paraformaldehyde for 30 minutes , washed with PBS , permeabilized with 0 . 4% Trion X-100 in PBS , and then blocked with 10% donkey serum ( Jackson ImmunoResearch Labs , West Grove , PA ) . Afterwards , cells were incubated with primary antibodies in blocking solution at 4°C overnight , followed by an incubation with the corresponding secondary antibodies and Hoechst 33342 for 1 hour at room temperature . The SA-β-gal staining of hMSCs was conducted using a previously described method [74] . For western blotting , cells were lysed in RIPA buffer containing a protease inhibitor cocktail ( Roche ) and quantified with a BCA kit . Generally , 20 μg of cell lysate was subjected to SDS-PAGE and electrotransferred to a PVDF membrane ( Millipore , Billerica , MA ) . Then , the membrane was incubated with primary and HRP-conjugated secondary antibodies . Western blot data were quantified with Image Lab software for the ChemiDoc XRS system ( Bio-Rad , Hercules , CA ) . For RT-qPCR , cellular total RNA was extracted using TRIzol ( Thermo Fisher Scientific ) , and genomic DNA was removed with a DNA-free Kit ( Ambion , Austin , TX ) , followed by cDNA synthesis with the GoScript Reverse Transcription System ( Promega , Madison , WI ) . RT-qPCR was performed with qPCR Mix ( TOYOBO , Tokyo , Japan ) in a CFX384 Real-Time system ( Bio-Rad ) . For genomic PCR , genomic DNA was extracted with a DNA extraction kit ( TIANGEN , Beijing , China ) , and PCR was conducted using PrimeSTAR ( TAKARA , Tokyo , Japan ) . Two thousand cells were seeded in each well of a 12-well plate and then cultured until clear cell colonies formed to determine the clonal expansion abilities of hMSCs . The relative colony area was then determined by performing crystal violet staining and measured using ImageJ software . The indicated fragments of the FOXD1 promoter were amplified by PCR and cloned into the pGL3-Basic vector ( Promega ) . The mutant of pGL3-FOXD1 promoter 2-Luc was constructed with the Fast Mutagenesis System kit ( FM111; Transgen Biotech , Beijing , China ) . PGL3-FOXD1 promoter 2-Luc or PGL3-FOXD1 promoter 2 ( mut ) -Luc was transfected into hMSCs together with vectors expressing the proteins of interest and Renilla-Luc , which was used to normalize the transfection efficiency . For detection of the 8 × GTIIC-Luc activity , the 8 × GTIIC reporter ( Addgene #34615 ) and Renilla-Luc plasmids were cotransfected into hMSCs . Cells were harvested 72 hours later using the Dual-Luciferase Reporter Assay System ( Vigorous Biotechnology , Beijing , China ) and assayed according to the manufacturer’s instructions . ChIP was performed using a previously reported protocol with minor modifications [75] . Briefly , cells were cross-linked with 1% ( v/v ) formaldehyde for 15 minutes at room temperature , and the reaction was terminated by the addition of 125 mM glycine and an incubation for 5 minutes at room temperature . Then , cells were scraped and lysed in lysis buffer . After sonication , protein-DNA complexes were incubated with antibody-coupled Protein A beads at 4°C overnight . After elution and reverse cross-linking at 68°C , DNA was purified by phenol/chloroform extraction and ethanol precipitation and then subjected to qPCR analysis . Antibodies for ChIP included anti-YAP ( 14074; Cell Signaling Technology ) , anti-TAZ ( 4883; Cell Signaling Technology ) , anti-TEAD4 ( 101184; Santa Cruz Biotechnology ) , and normal rabbit IgG ( 2027; Santa Cruz Biotechnology ) as a negative control . For the teratoma analysis , 5 × 106 hESCs were subcutaneously injected into NOD-SCID mice ( 6 to 8 weeks , male ) . After 8 to 12 weeks , the tumors were excised , fixed , dehydrated , embedded in O . C . T . compound , sectioned while frozen , and analyzed by immunostaining . For hMSC transplantation assays , 1 × 106 hMSCs transduced with a lentivirus expressing Luc were injected into the midportion of the TA muscle of nude mice ( 6 to 8 weeks , male ) . Then , 0 , 1 , 3 , 5 , and 7 days after transplantation , mice were treated with D-luciferin ( GoldBio , St . Louis , MO ) and imaged with an IVIS spectrum imaging system ( XENOGEN , Caliper , Waltham , MA ) . Bioluminescence images were acquired in “auto” mode . For RS-hMSC–induced osteoarthritis , we transplanted PBS , young hMSCs , RS hMSCs , and RS hMSCs overexpressing YAP or FOXD1 intra-articularly into the joints of NOD-SCID mice ( 6 to 8 weeks , male ) . Firstly , the mice were anaesthetized using isoflurane , and skin around the joints were shaved . For each injection , the needle was inserted beneath the middle patellar ligament , and a volume of 10 μl containing either PBS or 3 × 106 cells was injected intra-articularly . One month later , the mice were euthanized , and the joints were collected for mRNA quantification and histological assessments . For surgically induced osteoarthritis , we performed ACLT surgery on 8-week-old male C57BL/6 mice . Animals were anaesthetized , and their hindlimbs were shaved . After the opening of the joint capsule , the anterior cruciate ligament was transected with microscissors under a surgical microscope . After irrigation with saline to remove tissue debris , the skin incision was closed . Then , 7 days later , a total volume of 10 μl of the indicated lentivirus was injected intra-articularly . At week 8 , the mice were euthanized , and the joints were collected for mRNA quantification and histological assessments . Mouse joints were fixed with 4% paraformaldehyde overnight , decalcified with 5% methanoic acid for 7 days , and embedded in paraffin . Sections ( 5 μm ) were cut from the paraffin blocks and stained with Fast Green FCF ( 0 . 02% ) and Safranin O ( 0 . 1% ) . Joint pathology was quantified using the OARSI scoring system [13] . For immunohistochemical staining , paraffin-embedded tissue sections were subjected to a heat-mediated antigen retrieval procedure , and then endogenous peroxidases were blocked with hydrogen peroxide . Next , tissue sections were incubated with a primary antibody overnight . Finally , the appropriate secondary antibody ( ZSGB-BIO , Beijing , China ) was added to the sections , which were then incubated for 30 minutes . Antigen-positive cells were visualized using the DAB Substrate kit ( ZSGB-BIO ) . Anti-P16 antibody ( 54210; Abcam ) and anti-flag ( 166355; Santa Cruz Biotechnology ) were used as the primary antibodies . First , genomic DNA was extracted using the DNeasy Blood and Tissue Kit ( Qiagen , Duesseldorf , Germany ) according to the manufacturer’s instructions . DNA was sheared into fragments of approximately 300 bp using Covaris , and then the library of the fragmented DNA was constructed using the NEBNext ultra DNA Library Prep Kit for Illumina ( NEB , Beverly , MA ) , according to the manufacturer’s protocol . The libraries were sequenced on an Illumina HiSeq 4000 platform . For CNV identification , we used the published R package HMMcopy [76] . Briefly , the genome was binned into consecutive 1 Mb windows with read Counter , and then we calculated the absolute number of reads detected in each window . We estimated the copy number with GC and mappability corrections with HMMcopy . Total RNA was extracted from cultured human cells or mouse joints using the RNeasy Mini Kit ( Qiagen ) according to the manufacturer’s protocol . For cells , 1 × 106 cells were analyzed in biological triplicate . For mouse joints , we mixed the RNA extracted from the sample group , and then divided the sample into 3 technical replicates . One to two micrograms of total RNA was used to construct sequencing libraries using the NEBNext Ultra RNA Library Prep Kit for Illumina ( NEB ) . The libraries were sequenced on an Illumina HiSeq 4000 platform . RNA-seq reads were aligned to the hg19 or mm10 reference genome using TopHat2 software [77] . The analysis of differentially expressed genes was performed using DESeq2 [78] based on read counts . Promoter for TEAD-binding sites analysis was defined as 3 kb upstream and 500 bp downstream of TSS . TEAD-binding sites with P < 1 . 0 × 10−4 among the promoter regions were found by FIMO ( http://meme-suite . org/doc/fimo . html ) using the TEAD motif downloaded from JASPAR database ( http://jaspardev . genereg . net/ ) . Results are presented as the mean ± SD . Two-tailed Student t tests were used to compare differences between treatments . P < 0 . 05 , P < 0 . 01 , and P < 0 . 001 were considered statistically significant ( "*" , "**" , and "***" , respectively ) . | Stem cell aging contributes to aging-associated degenerative diseases . Studies aiming to characterize the mechanisms of stem cell aging are critical for obtaining a comprehensive understanding of the aging process and developing novel strategies to treat aging-related diseases . As a prevalent aging-associated chronic joint disorder , osteoarthritis is a leading cause of disability . Senescent mesenchymal stem cells ( MSCs ) residing in the joint may be a critical target for the prevention of osteoarthritis; however , the key regulators of MSC senescence are little known , and targeting aging regulatory genes for the treatment of osteoarthritis has not yet been reported . Here , we show that Yes-associated protein ( YAP ) , a major effector of Hippo signaling , represses human mesenchymal stem cell ( hMSC ) senescence through transcriptional up-regulation of forkhead box D1 ( FOXD1 ) . Lentiviral gene transfer of YAP or FOXD1 can rejuvenate aged hMSCs and ameliorate osteoarthritis symptoms in mouse models . We propose that the YAP–FOXD1 axis is a novel target for combating aging-associated diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"sequencing",
"techniques",
"rheumatology",
"medicine",
"and",
"health",
"sciences",
"senescence",
"forkhead",
"box",
"green",
"fluorescent",
"protein",
"physiological",
"processes",
"developmental",
"biology",
"luminescent",
"proteins",
"organism",
"development",
"stem",
"cells",
"molecular",
"biology",
"techniques",
"rna",
"sequencing",
"mesenchymal",
"stem",
"cells",
"research",
"and",
"analysis",
"methods",
"osteoarthritis",
"musculoskeletal",
"system",
"skeletal",
"joints",
"animal",
"cells",
"proteins",
"arthritis",
"gene",
"expression",
"molecular",
"biology",
"aging",
"biochemistry",
"cell",
"biology",
"anatomy",
"physiology",
"protein",
"domains",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types"
]
| 2019 | Up-regulation of FOXD1 by YAP alleviates senescence and osteoarthritis |
The MYC oncogene has been implicated in the regulation of up to thousands of genes involved in many cellular programs including proliferation , growth , differentiation , self-renewal , and apoptosis . MYC is thought to induce cancer through an exaggerated effect on these physiologic programs . Which of these genes are responsible for the ability of MYC to initiate and/or maintain tumorigenesis is not clear . Previously , we have shown that upon brief MYC inactivation , some tumors undergo sustained regression . Here we demonstrate that upon MYC inactivation there are global permanent changes in gene expression detected by microarray analysis . By applying StepMiner analysis , we identified genes whose expression most strongly correlated with the ability of MYC to induce a neoplastic state . Notably , genes were identified that exhibited permanent changes in mRNA expression upon MYC inactivation . Importantly , permanent changes in gene expression could be shown by chromatin immunoprecipitation ( ChIP ) to be associated with permanent changes in the ability of MYC to bind to the promoter regions . Our list of candidate genes associated with tumor maintenance was further refined by comparing our analysis with other published results to generate a gene signature associated with MYC-induced tumorigenesis in mice . To validate the role of gene signatures associated with MYC in human tumorigenesis , we examined the expression of human homologs in 273 published human lymphoma microarray datasets in Affymetrix U133A format . One large functional group of these genes included the ribosomal structural proteins . In addition , we identified a group of genes involved in a diverse array of cellular functions including: BZW2 , H2AFY , SFRS3 , NAP1L1 , NOLA2 , UBE2D2 , CCNG1 , LIFR , FABP3 , and EDG1 . Hence , through our analysis of gene expression in murine tumor models and human lymphomas , we have identified a novel gene signature correlated with the ability of MYC to maintain tumorigenesis .
Overexpression of MYC is one of the most frequent events in human tumorigenesis [1] . MYC overexpression is thought to induce tumorigenesis by causing inappropriate gene expression resulting in autonomous cellular growth , proliferation , and the inhibition of cellular differentiation [2] , [3] . Many laboratories have conditionally overexpressed c-MYC ( MYC ) utilizing conditional transgenic model systems [4]–[9] . In these models , the suppression of MYC led to permanent loss of tumorigenesis through proliferative arrest , differentiation and/or apoptosis [4]–[6] , [10] . In some circumstances , even the brief suppression of MYC overexpression permanently prevents its ability to sustain tumorigenesis [6] . These and other observations have suggested the possibility that oncogenes such as MYC exhibit the phenomena of oncogene addiction [11] . However , the molecular basis of oncogene addiction is not clear . Recently , we have suggested that cellular senescence , which involves chromatin modifications and heterochromatin formation [12] , [13] , may be an important mechanism for sustained tumor regression upon MYC inactivation [14] . MYC is thought to play a role in the regulation of up to 15% of genes in the fly , mouse or human [3] , [15] , [16] . Thus , it seems likely that changes in gene expression programs , rather than individual genes , account for the phenotypic consequences of MYC inactivation . Consistent with this notion , MYC has recently been shown to globally influence chromatin structure through histone modifications [17]–[19] . Similarly , N-MYC was shown to globally regulate acetylation and methylation of histone molecules [20] . We have reported that MYC inactivation in tumors induces specific global changes in histone modification [14] . Although many MYC target genes have been identified in various cells or tissue contexts ( summarized in http://www . myc-cancer-gene . org ) , it is hard to discern which of the many of MYC targets are associated with the ability of MYC to initiate and/or sustain tumorigenesis . Many previous studies have examined changes in gene expression associated with the induction of MYC expression in cells [3] , [15] , [21]–[26] . Other groups have performed comparative analysis of gene expression profiles between murine constitutive MYC-induced tumors and human tumors in liver and prostate cancers [27] , [28] . Both of these analyses identified similarities in gene expression between MYC-induced tumor models and human tumors . Although revealing , these studies would not necessarily identify gene products that are responsible for the ability of MYC to induce tumorigenesis . We speculated that by analyzing gene expression profiles in tumors generated from conditional transgenic models would allow us to identify gene expression signature specifically associated with the ability of MYC to initiate and maintain tumorigenesis . We performed microarrays on mRNA samples from a time-course experiment with MYC inactivated and then reactivated in osteosarcoma . The expression data was then examined using the StepMiner algorithm [29] to generate a list of genes associated with MYC-induced tumorigenesis in osteosarcomas ( Figure 1 ) . The StepMiner algorithm analyzes microarray time courses by identifying genes that undergo abrupt transitions in expression level , and the time at which the transitions occur . Importantly , by ChIP we were able to demonstrate that permanent changes in gene expression were frequently associated with measurable alterations in the ability of MYC to bind to the promoter regions of these genes in osteosarcomas . Furthermore , gene expression profiles were compared between osteosarcomas and the previously published MYC conditional pancreatic tumor [25] to generate a common gene signature associated with MYC-induced tumorigenesis in mice ( Figure 1 ) . Finally , Boolean analysis was used to further examine the correlation between levels of expression of this identified subset of genes among the published dataset of 7 , 171 human microarrays in U133A format . From this analysis , we were able to deduce a list of genes strongly correlated with the ability of MYC to maintain tumorigenesis .
The MYC induced osteosarcoma derived cell line , 1325 , was grown in vitro [6] and treated with 20ng/ml of doxycycline in complete DMEM medium for various length of time to inactivate MYC expression . To reactivate MYC expression , doxycyline was removed by rinsing the bone tumor cells with an excess amount of PBS . mRNA was collected from bone tumor cells treated with doxycycline for 0 , 4 , 8 , 12 , 18 , 24 , 36 , and 48 hours , and after removal of doxycycline for 4 , 8 , 12 , 18 , 24 , 36 , and 48 hours . MYC levels were greatly reduced as early as 4 hours after doxycycline treatment ( Figure 2A and Figure S1 ) . We confirmed that the expression of MYC could be reactivated to a level similar to that of MYC-on tumors by thoroughly washing the cells with PBS ( Figure 2A and Figure S1 ) . cDNA microarray analysis was performed on the RNA samples prepared from tumors in which MYC was inactivated and reactivated for different lengths of time . StepMiner analysis ( Figure 2B , 2C ) [29] was applied to this time-course microarray experiment to identify changes in gene expression at discrete time points before and after MYC inactivation and reactivation . StepMiner fits step functions to the data points using an adaptive regression scheme and identifies time points at which a gene is significantly induced or repressed . Examples of one-step expression pattern are illustrated in Figure 2C . Recently , we have shown that MYC inactivation generally induces cellular senescence in several tumor models [14] . Therefore , we specifically examined if the expression of senescence associated genes changed upon MYC inactivation in osteosarcomas . Indeed , we did find that senescence associated genes such as p15INK4b , p21CIP , PCNA , MCM3 , CYCLIN A [12] , [30] , [31] were up-regulated or down-regulated upon MYC inactivation ( Figure S2 and Table S1 ) . Thus , our results support the notion that MYC inactivation is inducing changes in gene expression that is associated with cellular senescence . Generally , analysis of gene expression changes after StepMiner analysis revealed four discrete patterns of changes in gene expression upon MYC inactivation and reactivation: Permanently Repressed ( PR ) , Permanently Induced ( PI ) , Reversibly Repressed ( RR ) and Reversibly Induced ( RI ) . For this analysis , we set p<0 . 01 as a cutoff for statistically significant changes in gene expression . We identified 1016 unique probes in the PR group , 1777 unique probes in the PI group , 1148 unique probes in the RI group , and 1167 unique probes in the RR group ( Figure 3 and Tables S2 for lists of genes ) . Based upon our previously published observation that even brief inactivation of MYC can result in the sustained loss of the neoplastic properties of MYC-induced osteosarcomas [6] , we speculated that genes which are potentially important for sustained tumorigenesis would be permanently repressed or induced ( e . g . the PR group or the PI group ) upon MYC inactivation . To identify associated functional activities associated with the PR and PI groups of genes , we applied Gene Ontology analysis ( GO Term analysis ) to the list of genes generated above . Biological functions that were identified for each step upon MYC inactivation are listed ( Table S3 , S4 , and S5 ) . Associated functions identified include gene products known to regulate metabolism , biosynthesis of nucleotides and proteins and genes involved in the regulation or function of ribonucleoprotein complexes . Notably , MYC has been shown to regulate expression of ribosomal structure proteins and ribosomal RNAs [18] , [32] . Hence , it is striking that the mRNA expression of 61 ribosomal structural proteins out of 82 ribosomal structural protein genes was decreased upon MYC inactivation and further decreased upon MYC reactivation in bone tumor ( see Figure 4A and Table S3 for results of GO term analysis ) . To validate that these genes expression did change , we performed quantitative real-time PCR of 11 ribosomal structural proteins in osteosarcomas ( Figure 4B ) . Moreover , we found that the same ribosomal structural proteins also changed upon MYC inactivation in our conditional model of lymphomas [4] ( Figure 4B ) . We then examined if the decreased expression of ribosomal structural proteins associated with changes in rate of protein synthesis . We found that the protein synthesis rates were decreased in both bone tumor and lymphomas upon MYC inactivation ( Figure 4C ) . Furthermore , the protein synthesis rate remained lowed upon MYC reactivation in bone tumor ( Figure 4C ) . MYC has been shown to regulate the gene expression of a multitude of genes [3] , [15] , [21]–[26] , [33]–[35] . To examine if these genes changed in gene expression upon MYC inactivation and reactivation , we used two approaches . First , we retrieved the mouse homologs of MYC target genes listed in www . myc-cancer-gene . org , a collection of most of the published MYC target genes in different organisms and tissues ( total of 1697 MYC targets ) [16] . In osteosarcomas , 71 of the published MYC targets are permanently induced and 52 of the published MYC targets are permanently repressed upon MYC inactivation and reactivation ( p<0 . 01 ) ( Figure 5 , see PR and PI ) . Second , we examined direct MYC target genes identified as defined by several recent publications [33]–[35] . Interestingly , only 7–11% of these identified direct MYC target genes exhibited sustained changes upon MYC inactivation in osteosarcoma ( Figure 6 and Table S6 ) . An important recent report suggests that MYC binding to promoters is regulated by the chromatin structure at these gene loci [36] . Recently , we have shown that MYC inactivation is associated with global changes in chromatin structures [14] . Thus , it seemed that a possible explanation for the permanent changes in gene expression that we observed ( Figure 3 ) is that the ability of MYC to bind to specific gene products is perturbed by changes in chromatin structure . To address this possibility directly , we used ChIP to examine MYC binding to E-box sequences of target genes in MYC activated and MYC reactivated conditions for osteosarcomas . We specifically examine three groups of genes: the ribosomal structural proteins ( Figure 4 ) , the PR group ( Figure 6 , [35] ) and the RR group genes that were identified previously as direct MYC targets before ( Figure 6 , [35] ) . A total of 168 E-box regions were examined by ChIP . As a control , we performed ChIP for osteosarcoma in the MYC OFF condition ( Table S7 ) . Binding of MYC to E-box regions is shown as the percentage of DNA brought down by ChIP for the MYC ON versus the MYC reactivated conditions ( Figure 7 ) . Note , that upon MYC reactivation the majority of ribosomal structural genes exhibited decreased MYC binding to E-boxes relative to the MYC ON condition ( 31 out of 41 data points fall below the line of X = Y , p-value = 4 . 34×10−4 ) . Similarly , the majority of the genes with the PR pattern of gene expression exhibited a significant decrease of MYC binding to E-boxes relative to the MYC ON condition ( 42 out of 60 data points fall below the line of X = Y , p-value = 0 . 0016 ) when MYC was reactivated ( Figure 7 and Table S7 ) . In contrast , the group of genes that exhibited the RR pattern of gene expression exhibited no particular increase or decrease in MYC binding to E-boxes compared with the MYC ON condition ( 33 out of 67 data points fall below the line of X = Y , p-value = 0 . 4 ) . Our results support the possibility that the permanent changes in gene expression upon MYC inactivation can be explained in many cases because of a change in the ability of MYC to bind to specific promoter loci . To determine if the gene signature we identified would also be seen in another tumor model system , we compared our microarray data from MYC-induced osteosarcoma with a previously reported microarray data set from a MYC-induced pancreatic tumor model to identify a common expression signature for MYC-induced tumorigenesis [25] . In the published report , MYC-ERTAM was expressed specifically in β-cell pancreatic tissues with MYC-on for 2 , 4 , 8 , 24 hours , and 21 days ( referred as tumorigenesis arrays in the published paper ) , and MYC off in pancreatic tumors for 2 , 4 , and 6 days ( referred as tumor regression arrays in the published paper ) . MYC activation induced pancreatic tumors and MYC inactivation resulted in tumor regression through apoptosis [7] . cDNA from these samples was applied to oligo arrays from Affymetrix [25] . As previously suggested in the paper , we assumed that genes were induced ( repressed ) upon MYC activation and repressed ( induced ) upon MYC inactivation were potentially important for MYC induced tumorigenesis . We first used the StepMiner algorithm was applied to the raw data generated from these published experiments to obtain lists of genes that increase ( or decrease ) in expression upon tumorigenesis and decrease ( or increase ) in expression upon tumor regression ( Figure 8 and Table S8 ) . After StepMiner analysis , 196 and 65 unique probes were identified as induced and repressed genes respectively , which are associated with MYC-induced tumorigenesis . The osteosarcoma data set was filtered via the induced gene list or the repressed gene list generated from the pancreatic tumors . Then , we applied StepMiner analysis to identify genes that are permanently repressed or permanently induced with a p-value<0 . 01 . By comparing microarray data from two independent MYC conditional tumor models , we found a common gene signature with 42 genes associated with MYC-induced tumorigenesis ( Figure 9 ) . Among the list of genes , there are 34 unique genes positively correlating with MYC-induced tumorigenesis and 8 unique genes negatively correlating with MYC-induced tumorigenesis in mice ( Figure 9 ) . MYC overexpression has been implicated in the pathogenesis of many types of human cancer , in particular , hematopoietic tumors [1] . To see if the gene signature we defined in murine tumor models was predictive of genes whose expression was strongly correlated with MYC between MYC and human homologs in human lymphomas , we retrieved all publicly available human microarrays ( n = 7 , 171 ) in Affymatrix U133A platform . Then , we classified the expression level of each gene on each array as “low” or “high” relative to a threshold using Boolean analysis ( [29] and Sahoo et al . RECOMB 2007 in press , see Figure 10A ) . We found that MYC expression is “high” in human lymphomas ( 204 out of 221 lymphoma cases ignoring the “intermediate” values , see Figure 10B ) . Figure 10B shows the gene expression scatter plot of MYC and RPS2 , which are both highly expressed in lymphoma arrays ( total of 273 lymphoma microarrays are highlighted with red color ) . We then examined to see if the expression of MYC-associated genes identified above ( Figure 4 and 9 ) are “high” or “low” in more than 95% of the lymphoma microarrays . The Boolean analysis identified that the expression of both small and large ribosomal structural proteins is high in human lymphomas ( Figures S3 and S4 ) as was observed in murine osteosarcomas and lymphomas ( Figure 4 ) . We further investigated if the expression of human homologs of the common gene signature from the murine microarray data is “high” or “low” in human lymphomas . 63 unique probes from the induced list ( Figure 9 ) and 9 probes from the repressed list ( Figure 9 ) were found in the U133A format ( see Table S9 ) . We found 14 out of 63 probes correlated with the human arrays . Genes whose expression was “high” in more than 95% of human lymphomas , whose gene names include: BZW2 , H2AFY , SFRS3 , NAP1L1 , NOLA2 , UBE2D2 and CCNG1 ( p = 4 . 07×10−5 , Figure 11 ) . From the repressed list of genes 4 out of 9 probes had low expression in more than 95% of the human lymphomas , whose gene names include LIFR , FABP3 and EDG1/HEXIM1 ( p = 0 . 03 , Figure 11 ) . We have listed al the genes identified and their associated functions ( listed in the Swiss-Prot data base ) ( Figure 11 ) . Many of these genes have functions that could account for MYC activity . Notably , CCNG1 , LIFR and EDG1/HEXIM1 are involved in cell cycle or signaling pathways . H2AFY and NAP1l1 are involved in modulating chromatin structures . SFRS3 and NOLA2 are involved in mRNA and rRNA processing . BZW2 , UBE2D2 and FABP3 are involved in metabolism such as protein or fatty acid synthesis . Finally , we validated our results obtained by microarray analysis through quantitative real-time PCR ( Figure S5 ) . Moreover , we found that these identified genes exhibited similar patterns of changes in gene expression upon MYC inactivation in our model of MYC-induced lymphoma ( Figure S6 ) . Therefore , we have identified a subset of MYC regulated gene products that are highly correlated with the ability of MYC to maintain tumorigenesis .
MYC target genes have been implicated in a multitude of biological functions [16] . Many additional potential MYC targets have been identified through microarray analysis [3] , [15] , [21]–[26] , [37] . However , it has not been easy to discern which if any of these genes are involved in the ability of MYC to initiate or maintain tumorigenesis . We have combined microarray analysis of two conditional transgenic model systems and a human comparative Boolean analysis to determine which of these identified genes most strongly correlated with MYC expression from total of 273 datasets of human lymphoma microarrays in U133A format . We also utilized ChIP to demonstrate that a large number of the genes that were permanently suppressed upon MYC inactivation exhibited changes in the ability of MYC to bind to their promoter loci . Thus , we identified a gene signature strongly correlated with the ability of MYC to maintain tumorigenesis . Our results have possible implications for why MYC induces tumorigenesis in specific cellular contexts . To identify this gene signature , we utilized our conditional transgenic model system of MYC-induced osteosarcoma in which we have previously shown that upon MYC inactivation tumors permanently lost the ability of MYC to induce tumorigenesis [6] . Thereby , we defined an initial gene signature consisting of 2 , 793 unique probe sets of genes that included genes whose expression was permanently changed ( Figure 3 ) . This gene signature includes gene products that have been already implicated as MYC targets ( Figure 5 ) . Most notably , ribosomal structural proteins were strongly correlated with MYC-induced tumorigenesis in murine osteosarcomas , lymphomas ( Shachaf CM et . al . submitted ) and in human lymphomas . These results suggest that the ability of MYC to induce ribosomal gene products is important to its ability to initiate and maintain tumorigenesis . Our results are consistent with a multitude of evidence suggesting that MYC can regulate ribosomal gene expression [15] . In Drosophila , the biological connection of MYC and ribosomal structural proteins can also be seen in the small cell-size phenotypes of both MYC mutants and ribosomal structural protein genes mutants [38]–[40] . MYC globally regulates protein synthesis through regulating expression of ribosomal RNAs , tRNAs , RNA helicases , and translation elongation factors [18] , [41] . Notably , it had been shown that rate of protein synthesis was increased 3-fold in MYC-overexpressing fibroblasts compared to MYC knockout fibroblasts [42] . We confirmed that the inactivation of MYC in tumor cells resulted in a reduction of both ribosomal protein gene expression and rate of protein synthesis in murine tumor models ( Figure 4 ) . Ribosomal genes could play important function in influencing protein translation and thus in this manner influence the ability of MYC to function as an oncogene . In this regard , it is notable that a recent study in Zebra fish identified some ribosomal protein genes as tumor-suppressors [43] . Nevertheless , it is not clear how ribosomal structural protein genes function as tumor-suppressors during tumorigenesis . Interestingly , changes in the gene expression of ribosomal structural proteins , although observed in both our model of MYC induced osteosarcoma and lymphoma , were not seen in a model of pancreatic islet cell tumors ( Figure 4 , 8 , and [25] ) . Thus , it is possible that ribosomal protein genes expression play a role MYC-induced tumorigenesis only in specific types of cancer . We are reassured of the likely importance of ribosomal gene products in MYC associated tumorigenesis for we were able to confirm that MYC and ribosomal structural proteins are highly correlated in human lymphomas ( Figure S4 and S5 ) . It remains to be directly determined if these ribosomal genes are playing a role in MYC induced tumorigenesis . Genes that we identified as most strongly correlated with MYC-induced tumorigenesis ( Figure 9 ) in mice are involved in diverse biological processes such as transcription regulation , RNA processing , proliferation , fatty acid transport and cell signaling ( Figure 11 ) . Furthermore , some of the genes identified have been previously implicated in tumors or oncogenic signaling pathways . BLMH has been previously shown to be a MYC target [44] . UBE2d2 has been implicated as a target of the WNT signaling pathway in a microarray experiment [45] . NAP1l1 has been shown to be a tumor marker for colon cancer [46] . TRIP13 expression was highly elevated in tumor tissues [47] . Altered regulation of CCNG1 has been observed in breast cancer [48] . High expression of NOLA2 has been seen in squamous cell lung cancer [49] . Interestingly , anti-tumor effects have been observed for genes with expression reversely correlated with MYC . FABP3 has been proposed as tumor suppressor in breast cancer [50] . EDG1 has been shown to be an inhibitor for breast cancer growth [51] . Our data now suggest that BZW2 , H2AFY and SFRS3 , which function in translation initiation [52] , chromatin structure [53] , and mRNA splicing [54] , respectively , may also be involved in tumorigenesis . We were able to utilize our MYC conditional tumor models as tools to uncover genes that are strongly correlated with tumor maintenance . However , we recognize that it is very unlikely that any of the individual genes we identified are sufficient alone to explain the ability of MYC to initiate or maintain tumorigenesis . Rather it is highly likely that it is a constellation of gene expression changes that are responsible for the ability of MYC to maintain tumorigenesis . We can now offer a possible explanation for why the brief inactivation of MYC can result in the permanent loss of the ability of MYC to sustain tumorigenesis [6] . MYC inactivation appears to result in permanent changes in the ability of MYC to function as a transcription factor ( Figure 12 ) . Recently , we have shown that MYC inactivation induced chromatin modifications associated with cellular senescence [14] . The particular structural state of chromatin has been shown to influence the ability of MYC to bind to specific promoter loci [36] . Indeed , our results illustrate that upon MYC inactivation there were permanent changes in the ability of MYC to bind to the promoters of specific gene loci ( Figure 7 ) . It remains to be determined the mechanism of these changes in chromatin structure . One possibility is that MYC itself is contributing to changes in chromatin structure through global changes in chromatin modifications , which seems an attractive possibility based upon the work from many laboratories [14] , [20] , [55] . Regardless of the mechanism , our results point to the fact that the genes that MYC can regulate are different in different cellular contexts and that this appears to have a direct bearing on when MYC overexpression results in a neoplastic phenotype . We note that we could not explain all of the permanent changes in gene expression based upon differences in MYC binding to promoter loci . Thus , it is likely there are additional mechanisms by which MYC's ability to regulate gene expression has been altered . One of the biggest challenges in understanding how MYC contributes to tumorigenesis has been to address the conundrum that MYC has both direct and indirect influence on the expression of so many different genes and these genes are involved in a multitude of biologic functions . Many of these genes may not be relevant to how MYC overexpression contributes to tumorigenesis . Here we have illustrated by using a defined transgenic mouse model that exhibits conditional tumorigenesis such that upon MYC inactivation tumor cells permanently loses a neoplastic phenotype that we can define a specific gene list that is specifically correlated with MYC's ability to maintain tumorigenesis . To perform this analysis we combined two novel methods of gene expression analysis , the StepMiner and the Boolean analysis , as a powerful strategy to perform an unbiased comparative analysis of microarray data from conditional MYC-induced tumor models and all the available published human data with Affymetrix U133A format . Our strategy may be generally useful for the identification of gene signatures associated with the ability of specific oncogenes to initiate and sustain tumorigenesis and the identification of potential new therapeutic targets for the treatment of cancer .
Osteosarcoma-derived cell line 1325 [6] were cultured with DMEM medium supplemented with 10% FBS , 1% Pen/Strep , L-Glutamine , and non-essential amino acids ( Invitrogen ) . Lymphomas were cultured with RPMI medium supplemented with 10% FBS , 1% Pen/Strep , L-Glutamine and 3 . 96×10−4% of 2-mercaptoethanol ( Sigma ) . 20ng/ml of doxycycline was added to the medium for inactivating MYC expression . Seven times , each time with 20 mls of PBS , was applied to cells to completely remove doxycycline in the medium . For rate of protein synthesis , lymphoma-derived cell line 6780 [14] or bone tumor cell line 1325 grown in complete medium with or without doxycycline were rinsed with PBS and then replenished with DMEM ( with or without doxycyline ) without methionine and cysteine ( Invitrogen ) , containing 10% dialyzed fetal calf serum ( Invitrogen ) , 1% Pen/Strep , L-Glutamine . One hour later , cells were labeled with 30 μCi of EXPRE35S35S ( PerkinElmer ) per plate for 60 minutes and then washed with PBS . Cells were lysed and TCA precipitation was applied to determine the incorporation of radiolabeled amino acids . Aliquots of cell lysate were used for protein determination by DC Protein Assay ( Bio-Rad ) . The protein synthesis rate was calculated as TCA-precipitable counts per minute divided by micrograms of protein in the same sample . cDNA were synthesized by Superscript II ( invitrogen ) followed by manufacture's protocol . Real-time PCR for human c-MYC ( probes and primers from Applied Biosystems ) and mouse GAPDH [56] were performed in ABI PRIZM analyzer . Sequences for primers for quantitative real-time are listed in Table S8 . Mouse cDNA microarrays were produced at Stanford Functional Genomic Facility . cDNA labeling and hybridization were followed as previously described [57] . Briefly , mRNA from bone tumor cells were extracted by Trizol ( Invitrogen ) based on the protocol provided by the manufacturer . 30 μg of total RNA from bone tumor and reference RNA generated by pooling RNA from various mouse tissues were used for each microarray experiment . cDNA from bone tumor cells was labeled with Cy5-dUTP and reference cDNA was labeled with by Cy3-dUTP ( Amershan ) after reverse-transcription . Labeled cDNAs were concentrated by Microcon YM-30 ( Millipore ) before hybridizing with microarrays for 16 hours at 65°C . After hybridization , microarrays were washed and spin dry before scanned on the GenePix 40000B Array Scanner ( Axon ) . Raw array images were analyzed using the GenePix 5 software ( Axon ) . Microarray data was then submitted to the Stanford Microarray Database ( SMD ) for normalization . Data after normalization was then applied with the StepMiner algorithm to identify changes in gene expression . The StepMiner fits step functions to time-course microarray data and provides a statistical measure of the goodness of fit [29] . The steps are placed between time points at the sharpest change between low expression and high expression levels , which gives insight into the timing of the gene expression-switching event . Mathematically , steps are placed at a position that minimizes the sum of square error and an F-statistic with appropriate degrees of freedom is used to produce a p-value for the goodness of fit . The StepMiner automatically characterizes the genes in to five different groups: Up , Down , Up-Down , Down-Up and Other [29] . The genes are primarily sorted in ascending order according to the timing of their change and secondarily sorted in ascending order according to their p-values . ChIP was performed based on the protocol provided in the kit with some modifications ( ChIP assay kit by Upstate Biotech ) . Briefly , bone tumor cells were grown on the condition described above with ( MYC OFF and MYC reactivated conditions ) or without ( MYC ON condition ) doxycycline ( 20ng/ml ) . 48 hours treated with doxycycline , cells were either harvested ( as MYC OFF condition ) or extensively washed with PBS ( see above ) to remove doxycycline in the medium . 48 hours after washing , cells were harvested ( as MYC reactivated ) . Formaldehyde ( Fisher ) was added to the medium to a final concentration of 1% for cross-linking at 37°C for 10 minutes . Cross-linking was stopped by adding glycine to a final concentration of 0 . 125M . Cells were washed with cold PBS containing protease inhibitors ( 1mM PMSF , 1 μg/ml aprotinin and 1 μg/ml pepstatin A ) and pelleted by centrifugation . Cell pellets were then lysed in SDS lysis buffer ( 1% SDS , 10mM EDTA , 50mM Tris , pH 8 . 1 , with proteases inhibitors mentioned above ) . Cells were sonicated with a Branson 250 sonicator at a power setting of 3 for 3 times with 10 sec for each sonication and the cells were cooled down with ice for 1 min between each sonication . This condition of sonication yielded genomic DNA fragments with a size about 100–600 base pairs . Samples were then immunoprecipitated with c-MYC antibody ( 2 μg of N262 from Santa Cruz Biotech ) followed the protocol provided by the kit ( Upstate Biotech ) . DNA samples from the ChIP experiments were applied for quantification by Real-time PCR ( ABI PRISM 7900 HT ) with SYBR green . Promoter sequences ( −2000 to +2000 relative to the transcription start sites ) of murine MYC targets were retrieved from UCSC genome browser and primers flanking the E-box were designed by Primer3 ( http://frodo . wi . mit . edu/ ) ( Table S10 ) . Data from 7 , 171 publicly available raw Affymetrix U133A human microarrays were collected from the Gene Expression Omnibus ( GEO ) [58] and normalized together using the RMA algorithm [59] , [60] . Thresholds were assigned for each probe set by first sorting the expression values for that probe set on all arrays in ascending order , and then fitting a step function to the data using the StepMiner . This approach places the threshold cutoff at the largest jump from low values to high values . In the case where the gene expression levels are evenly distributed from low to high , the threshold cutoff tends to be near the mean expression level . If the assigned cutoff for a gene is t , expression levels above t + 0 . 5 are classified as “high , ” expression levels below t−0 . 5 are classified as “low , ” and values between t −0 . 5 and t+0 . 5 are classified as “intermediate” ( Sahoo et al . RECOMB 2007 in press ) . Two hundred and seventy three different human Lymphoma microarray experiments were identified using a simple string search “Lymphoma” in the GEO description of the experiment . Genes that are “high” or “low” in more than 95% of the Lymphoma experiments were automatically discovered . Human homologs of genes which were associated with MYC-induced tumorigenesis in mice were selected for this manuscript . | The targeted inactivation of oncogenes may be a specific and effective treatment of cancer . However , how oncogene inactivation leads to tumor regression is not clear . Previously , we have shown that even the brief inactivation of the MYC oncogene can result in the sustained regression of at least some tumors . To understand the mechanism , we have utilized several novel genomic analyses to define a set of genes that strongly correlate with the ability of the MYC oncogene to maintain tumorigenesis . First , we generated a novel data set from microarray analyses of murine tumors that we analyzed by StepMiner to identify discrete step changes in gene expression after the inactivation or the reactivation of the MYC oncogene . Second , we utilized Boolean Network Analysis to further define the subset of genes highly correlated with MYC in human tumorigenesis . Third , we utilized ChIP analysis to demonstrate that in many cases the permanent changes of gene expression we uncovered were associated with changes in the ability of MYC to occupy the promoter locus . Our general strategy could be similarly utilized in other experimental model systems to understand how specific oncogenes contribute to the maintenance of tumorigenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/cancer",
"genetics"
]
| 2008 | Combined Analysis of Murine and Human Microarrays and ChIP Analysis Reveals Genes Associated with the Ability of MYC To Maintain Tumorigenesis |
Floral organs display tremendous variation in their exterior that is essential for organogenesis and the interaction with the environment . This diversity in surface characteristics is largely dependent on the composition and structure of their coating cuticular layer . To date , mechanisms of flower organ initiation and identity have been studied extensively , while little is known regarding the regulation of flower organs surface formation , cuticle composition , and its developmental significance . Using a synthetic microRNA approach to simultaneously silence the three SHINE ( SHN ) clade members , we revealed that these transcription factors act redundantly to shape the surface and morphology of Arabidopsis flowers . It appears that SHNs regulate floral organs' epidermal cell elongation and decoration with nanoridges , particularly in petals . Reduced activity of SHN transcription factors results in floral organs' fusion and earlier abscission that is accompanied by a decrease in cutin load and modified cell wall properties . SHN transcription factors possess target genes within four cutin- and suberin-associated protein families including , CYP86A cytochrome P450s , fatty acyl-CoA reductases , GSDL-motif lipases , and BODYGUARD1-like proteins . The results suggest that alongside controlling cuticular lipids metabolism , SHNs act to modify the epidermis cell wall through altering pectin metabolism and structural proteins . We also provide evidence that surface formation in petals and other floral organs during their growth and elongation or in abscission and dehiscence through SHNs is partially mediated by gibberellin and the DELLA signaling cascade . This study therefore demonstrates the need for a defined composition and structure of the cuticle and cell wall in order to form the archetypal features of floral organs surfaces and control their cell-to-cell separation processes . Furthermore , it will promote future investigation into the relation between the regulation of organ surface patterning and the broader control of flower development and biological functions .
In contrast to other plant cell layers , the epidermis develops a unique cell wall that not merely constitutes of cellulose , hemicelluloses , pectins , and proteins but also of a cuticular matrix , which is largely composed of cutin embedded and overlaid with waxes [1] . Cutin , an insoluble cuticular polymer , is largely composed of interesterified hydroxy and hydroxy epoxy fatty acids and is attached to the outer epidermal layer of cells by a pectinaceous layer [2] . As the epidermal cell grows , the cuticle merges gradually with the cell wall components [3] . Although the role of the epidermis layer in regulating organ growth has remained controversial [4]–[5] , it is clear that it is vital for plant survival , development and the interaction with the environment [6]–[7] . Cutin and wax are synthesized exclusively in the epidermis [8] and a massive flux of lipids occurs from the sites of lipid synthesis in the plastid and the endoplasmic reticulum ( ER ) to the plant surface during cuticle deposition [9] . Significant progress has been made over the past decade in identifying genes involved in the biosynthesis and secretion of cuticular lipids [10]–[11] and in the metabolism and assembly of primary cell wall components [12]–[14] . Despite the close connection between the cell wall and the cuticular matrix , mutants and phenotypes in one of these processes were rarely examined for alteration in the other . Furthermore , to our knowledge , co-regulation of these two processes at the molecular genetic level was overlooked up to now . Biosynthesis of plant cuticle components and their secretion to the extracellular matrix involve the coordinated induction of several metabolic pathways , in which transcription factors may play a key role [9] , [15] . The Arabidopsis SHINE1/WAX INDUCER1 ( SHN1/WIN1 ) AP2-domain protein was the first transcription factor reported to control metabolic pathways generating cuticular waxes [16]–[17] . A subsequent study [18] indicated that SHN1/WIN1 controls cuticle permeability by regulating the expression of cutin biosynthesis genes , particularly LACS2 ( LONG CHAIN ACYL-COA SYNTHETASE 2 ) . The induction of wax formation in leaves by over expression of individual SHINE clade genes was suggested to be a second step , possibly an indirect process following cutin biosynthesis [18] . Nevertheless , our current knowledge is limited with respect to the SHN1/WIN1 protein's mode of action and the involvement in particular developmental processes . Arabidopsis SHN1/WIN1 transcription factor belongs to a small distinct clade of three proteins [16] . They all share two unique conserved motifs outside the AP2 domain , and all three proteins display the same shiny phenotype upon overexpression , suggesting their functional redundancy in cuticular lipid biosynthesis . Additional evidence for functional redundancy among the SHN clade members in cuticular lipid biosynthesis was provided by silencing SHN1/WIN1 [18] . In these plants , floral morphology was not altered and the subtle reduction in the levels of cutin detected in entire flower extracts was enhanced in isolated petals . Besides , their notable expression patterns in reproductive organs suggested that they are probably redundant in function . The expression of SHN1/WIN1 and SHN3 overlapped in various flower organs including in the abscission zones while SHN2 and SHN3 were both expressed in the silique dehiscence zones . Interestingly , expression of SHN2 was very specific to cell separation regions in the anthers and siliques . These expression profiles indicated that SHN transcription factors may also act in a combinatorial manner to secure reproductive organ development , protecting the exterior layers of the plants from environmental stresses . On the other hand , these three clade members differ in their spatial and temporal expression patterns , which suggests that each of them may play specific roles in various organs or under different conditions , and that the actual redundancy between the SHN factors is most probably in their target genes [16] . Further elucidation of the mode of SHN action , their target genes , and their precise connection to plant cuticle formation and plant development requires in-depth characterization of the SHN clade factors , which can be achieved by using double , possibly triple mutants to eliminate redundant activities [16]–[18] . In contrast to Arabidopsis , mutation in the barley SHN1/WIN1 ortholog ( Nud ) was sufficient to generate a severe morphological change in which the typically hulled caryopses developed into naked ones [19] . Nud was suggested to direct the deposition of a lipidic matter on the pericarp epidermis that adheres the hull to the caryopsis in a way similar to postgenital fusions displayed by numerous cuticular mutants [20]–[21] . In this study we have co-silenced the three SHN clade members in order to decipher their modes of action and resolve their biological roles . We revealed that SHN clade genes regulate the elongation and decoration ( i . e . nanoridges formation ) of reproductive organ epidermal cells , particularly in the petal surface . They also emerge as mediators of cell adhesion and separation during abscission and dehiscence . Additionally , the results suggest that beside their function in the cutin pathway , these transcription factors possess putative downstream target genes that are involved in cell wall configuration through pectin modifying enzymes and structural proteins . Thus , the study of SHN transcription factors provides novel insight to the transcriptional control that mediates the patterning of reproductive organs surfaces and their associated separation processes in between cell layers .
To circumvent the likely functional redundancy between the Arabidopsis SHN clade members we generated plants in which they were simultaneously silenced through an artificial microRNA approach ( Figure S1A and Text S1 ) . The presence of cleaved products and transcriptional downregulation of all three SHN genes was confirmed in the 35S:miR-SHN1/2/3 plants ( Figure 1A and Figure S1B–S1C ) . No visual change was observed in these plants during vegetative growth and cuticle permeability of their rosette leaves was normal ( Figure S1D–S1G ) . However , reproductive organs , particularly petals , were severely affected ( Figure 1C–1D ) . This was evident already in buds that displayed postgenital fusions between petals and other floral organs at their tops ( Figure 1H–1I ) . The expansion of petals and elongation of the carpels were restrained and they were curved and/or twisted ( Figure 1I and Figure S1L–S1M ) . The changes in flower organ morphology also impinged on self-pollination and semi-sterility was occasionally detected ( Figure 1B ) . Interestingly , mutant flower organs abscised earlier ( Figure 1E and Figure S1J–S1K ) , and in some cases the abscised flower parts stayed attached to the top of the silique due to the postgenital organ fusion between them ( Figure 1F–1G ) . Microscopic observation of floral organs surfaces in the 35S:miR-SHN1/2/3 plants revealed extensive alterations to their archetypal epidermal cells ( Figure 2 and Figure S2 ) . Both abaxial and adaxial conical cells of petals appeared less elongated , more spherical and compact in addition to being separated with wider spaces as compared to the wild-type ( WT ) cells ( Figure 2 ) . Remarkably , nanoridges , typically displayed on WT petal epidermis [22]–[23] , were either absent ( adaxial ) or significantly reduced ( abaxial ) in the 35S:miR-SHN1/2/3 petal cells ( Figure 2A–2F ) . Altered epidermis cell size , shape and nanoridge decoration was also observed in surfaces of additional floral organs such as sepals , styles , filaments , nectaries , and pedicles ( Figure S1N–S1Q and Figure S2 ) . The observed phenotypes provided evidence that the SHN clade genes function redundantly in cell elongation , separation and nanoridge formation of reproductive organs . In contrast to the 35S:miR-SHN1/2/3 floral organs , silencing SHN1/WIN1 alone did not cause any visible morphological changes in floral organs , particularly in petal surfaces ( Figure S3 ) . In order to unravel the molecular mechanism by which the SHN factors regulate the patterning of reproductive organ surfaces we compared the transcriptome of 35S:miR-SHN1/2/3 flower buds to the one of WT . A modest set of 38 differentially expressed genes was detected; 30 transcripts including SHN1 and SHN3 ( SHN2 was not represented in the array ) were downregulated while 8 others were upregulated in 35S:miR-SHN1/2/3 buds ( Table 1 ) . Interestingly , one of the two main functional categories that dominated the differential genes represented six cell wall related genes ( Table 1 ) . Four of them corresponded to enzymes associated with pectin degradation or modification , including two pectate lyases ( PLL14 and PLL23 ) , a polygalacturonase ( ADPG1 ) and a pectin methylesterase inhibitor ( PMEI ) . Two additional genes putatively encode cell wall structural proteins: a hydroxyproline-rich glycoprotein ( HRGP ) and a glycine-rich protein ( GRP ) . The second major category consisted of seven genes that putatively encode cuticular lipids ( mainly cutin ) related proteins , including 2 cytochrome P450s ( CYP86A4 and CYP86A7 ) implicated in flower cutin biosynthesis [18] , [23] , three GDSL-motif lipase/hydrolases ( RXF26 , At2g42990 , and At5g33370 ) that are highly similar to the reported cutin related lipase At2g04570 [18] , and one hydrolase ( BODYGUARD 3 , BDG3 ) , the closest homolog of BDG1 , an epidermis-specific extracellular protein associated with cuticle formation [24] . Fatty Acyl-CoA Reductase 1 ( FAR1 ) , the seventh gene was associated with primary fatty alcohol production [25]; its additional and/or alternative function with relation to surface lipids will be discussed below . Two downregulated genes encoded a potassium transporter ( KUP5 ) and an ABC transporter ( PGP13/MDR15 ) ; both are involved in cell growth [26]–[28] . Additional three downregulated genes encoded kinase and/or kinase like proteins , that are potentially involved in reporting sensing aspects of cell wall structure and function [29] . Differential expression of 24 genes including the three SHN genes was subsequently validated using realtime RT-PCR assays ( Figure S4 and Text S1 ) . Altogether , gene expression analysis results indicated that the phenotype observed in 35S:miR-SHN1/2/3 reproductive organs probably result from the altered expression of their target genes , particularly those related to cutin and cell wall remodeling and function . Because plant organ fusion and separation have been reported to be associated with cuticle [19]–[20] , [22] , we subsequently examined the changes in cuticular lipids in leaf and flower tissues of the 35S:miR-SHN1/2/3 plants . While the amount of leaf cutin was not significantly changed ( Figure S5A ) , the amount of flower cutin in the 35S:miR-SHN1/2/3 plants was reduced to 48 . 4% of the wild-type ( Figure 3A ) . The changes in flower cutin loads reflected the changes in the cuticle permeability in flower tissues ( Figure S1F–S1I ) . The substantial decrease of dioic acids ( DFA , particularly C16 , C18:2 and C18:1 ) , ω-hydroxy fatty acids ( ω-HFA , particularly C16 and C18:3 ) , 9/10 , 16-dihydroxy hexadecanoic acid ( C16-9/10 , 16-DHFA ) and 9 ( 10 ) -hydroxy-hexadecanedioic acid ( C16-9/10-HDFA ) largely contributed to the reduced flower cutin in the 35S:miR-SHN1/2/3 plants . Levels of cuticular waxes in either leaves or flowers were not significantly altered in the 35S:miR-SHN1/2/3 lines ( Figure S5B–S5C ) . The finding that co-silencing the three SHN genes affected the expression of pectin modifying genes prompted us to analyze the cell wall pectin composition in the seed mucilage and buds . GC-MS analysis did not reveal any significant compositional changes in seed mucilage and the bud cell wall pectic monosaccharides ( Figure S5D–S5E and Text S1 ) . We next used Fourier transform infrared ( FTIR ) spectroscopy to examine if petals of the 35S:miR-SHN1/2/3 plants exhibited structural changes in their cell walls . Principal component analysis ( PCA ) showed a clear separation of the petal FTIR spectra between 35S:miR-SHN1/2/3 petals and WT ones ( Figure 3C ) . The difference spectrum ( Figure 3B ) generated by digitally subtracting the average 35S:miR-SHN1/2/3 spectrum from the average WT petals spectrum showed that WT petal cell wall had more acyl esters ( 1740 cm−1 ) [30]–[31] , amide III proteins ( 1230 cm−1 ) [32] , and non-cellulosic carbohydrates ( 1100 to 900 cm−1 ) [33] . In contrast , 35S:miR-SHN1/2/3 petal cell walls contained more salt-form of pectin ( 1430 and 1600 cm−1 , respectively ) [32] , amide I and amide II proteins ( 1650 and 1550 cm−1 , respectively ) [32]–[33] , and phenolic esters or aromatic lignins ( 1635 and 1510 cm−1 ) [32]–[33] . To localize the pectic polysaccharides in the cell walls , two novel rat monoclonal antibodies LM19 and LM20 , which recognize pectic homogalacturonan ( HG ) epitopes [34] , were used to hybridize transverse sections of inflorescence stems ( pith parenchyma ) and flowers . Similar to an earlier observation in tobacco plants [34] , LM19 localized pectin to junctures ( middle lamella ) while LM20 localized pectin to the intercellular spaces ( air spaces ) in both WT and 35S:miR-SHN1/2/3 inflorescence stems ( both antibodies appeared as green fluorescence ) ( Figure 3D ) . However , the florescence of LM19 in transverse sections of the 35S:miR-SHN1/2/3 samples became weaker and they were aggregated along the middle lamella line . Moreover , the florescence of LM20 in 35S:miR-SHN1/2/3 was enhanced not only in the air spaces but also in the middle lamella . In addition , the florescence of LM20 binding to air spaces become stronger in microtome sections of 35S:miR-SHN1/2/3 petals and developing seed coats , as compared to WT ones ( Figure 3E ) . Because the binding of both LM19 and LM20 to pectin is sensitive to pectate lyase treatment and they bind preferably to HG [35] , these results indicated alteration to HG distribution in the mutants . Therefore , silencing the SHN clade genes not only affected the cutin matrix of the cuticle but also the cell wall matrix of the cell . Remarkably , in silico analysis ( Table 1 ) showed that as SHN1/WIN1 , 13 of the differentially expressed genes ( 12 downregulated and one up regulated in the 35S:miR-SHN1/2/3 plants ) display a petal-specific expression pattern [35] . Moreover , all those 12 petal-specific downregulated genes , together with SHN1/WIN1 , SHN3 , and 3 more genes display decreased expression in senescing petals [36] . Furthermore , 9 of the differential genes in addition to SHN1/WIN1 are expressed in the stamen abscission zone ( AZ ) [37] while 2 genes and SHN1/WIN1 are enriched in the nectary [38] , and 13 genes and SHN3 are differentially expressed in senescing siliques [36] . These results provided evidence that both the SHN factors and their putative targets are associated with reproductive organ development ( i . e . petals and siliques ) and possibly cell separation as well . The series of genes altered in the 35S:miR-SHN1/2/3 plants were also strongly co-expressed with the SHN factors ( Figure S6 and Table S2 ) , further indicating the functional link between the groups of genes we have identified in the array analysis . In order to examine whether loss of function of the putative SHN clade proteins target genes results in alteration to petal surface we screened for T-DNA insertions in the entire set of 28 downregulated genes . Homozygous knockout lines could be identified for thirteen of them and their petals surface was examined using scanning electron microscopy ( Figure S7 ) . Petals of the At5g23970 ( a putative acyltransferase ) and At5g33370 ( a putative GDSL-lipase ) knockout plants exhibited collapsed conical cells , while those of At4g24140 ( bodyguard3/bdg3 ) , At5g03350 ( a receptor like protein ) and At1g01600 ( cyp86a4 ) displayed abnormal abaxial nanoridges ( Figure 4A–4D ) . Some differential genes identified in microarray analysis belong to large multi-gene families as for example lipases and cytochrome P450s . This suggested that they might be functionally redundant with other family members . We therefore co-silenced the CYP86A4 with CYP86A7 , and the GDSL-lipase At5g33370 with its closest homolog At3g04290 , LTL1 [39] , via the artificial microRNA method . Plants co-silenced for either one of these pairs of genes displayed severe floral organ fusion and alteration in the conical cell shape and/or epidermis cell decoration ( Figure 4E–4H ) . These results from single knockouts and the co-silenced lines provided additional evidence for the functional link between the putative SHN proteins target genes and the patterning of the petal surface . We subsequently examined the activation of promoters of genes that were differentially expressed in the 35S:miR-SHN1/2/3 plants by the SHN transcription factors using a dual luciferase assay system [40] . Promoter regions of 23 putative targets and the 3 SHN clade genes were examined . Thirteen out of 23 were significantly activated by at least one of the three SHN transcription factors ( Figure 5 ) . Promoter regions of seven genes were activated by all three factors including the ones of RXF26 , CYP86A4 , CYP86A7 , BDG3 , FAR1 , GRP , and GRXC11 . The promoters of PRX02 ( a peroxidase ) , ARD3 ( an acireductone dioxygenase ) , and At2g43620 ( a chitinase ) were only activated by SHN1/WIN1 , SHN2 , and SHN3 , respectively . Interestingly , SHN1/WIN1 and SHN2 were able to activate each other's promoter , while SHN3 was able to activate all three SHN genes promoters . We included LACS2 promoter as a positive control [18] , however , activation of this gene promoter by the SHN transcription factors was not detected in our assay . These results further confirmed the functional redundancy of SHN transcription factors in cuticle and cell wall metabolism by acting directly on common targets and by regulating each other and possibly their own transcription . Gibberellins ( GAs ) are a class of plant hormones involved in the regulation of flower development in Arabidopsis . GA promotes the expression of floral homeotic genes APETALA3 ( AP3 ) , PISTILLATA ( PI ) , and AGAMOUS ( AG ) by antagonizing the effects of DELLA proteins , thereby allowing continued flower development [41] . Publically available array data suggested that GA promotes the expression of SHN1/WIN1 while DELLA suppresses SHN1/WIN1 expression , which was examined in the ga1-3 and the ga1-3 gai-t6 rga-t2 rgl1-1 rgl2-1 ( i . e . penta ) [35] . Remarkably , in young flower buds , GA promotes the expression of thirteen of the putative SHN target genes identified in this study while it down regulates the expression of another four putative target genes , all of them in a DELLA dependent manner ( [42] , Figure S8A–S8B ) . In addition , GA regulates another two putative SHN target genes , AT4G27450 and AT1G27940 , in a DELLA-independent way [42] . The results described above led us to suggest that GA might be involved in cuticle assembly during flower organ development via modulating the expression ( directly or indirectly ) of the SHN transcription factors and their downstream target genes . To test this assumption , we examined the expression of SHN genes in different GA biosynthesis or signaling mutants ( Figure 6A ) . Quantitative RT-PCR analysis showed that expression of SHN1/WIN1 is downregulated in the ga1-3 mutant that is defective in GA biosynthesis . It also showed that DELLA significantly suppressed SHN1/WIN1 expression , since the expression of SHN1/WIN1 in the double ( rga-t2 rgl2-1; partial loss of DELLA signaling ) and quadruple DELLA ( gai-t6 rga-t2 rgl1-1 rgl2-1 ) mutants in the ga1-3 background was recovered to equal and even much higher levels than that of the wild type , respectively . Knockout of SPY4 , another repressor of GA signaling , also enhanced SHN1/WIN1 expression as compared to the wild type . As compared to SHN1/WIN1 , SHN2 showed the opposite expression pattern in the background of the various GA biosynthesis and signaling mutants . Expression of SHN2 was upregulated in the ga1-3 background while it was significantly downregulated in the penta and spy4 mutant backgrounds . Interestingly , neither GA biosynthesis nor the signaling mutants significantly altered SHN3 expression . We also examined the expression of SHN clade genes in both the WT and ga1-3 flower buds in response to exogenous GA application ( Figure 6B ) . Quantitative RT-PCR analysis showed that GA application to the ga1-3 mutant increased the levels of SHN1/WIN1 and decreased the levels of SHN2 expression as compared to ga1-3 alone , as does the endogenous GA ( Figure 6A ) . The response of SHN3 might be different between endogenous and externally applied GA as its expression did not change significantly in the ga1-3 background alone while it was altered upon GA supplementation in either the WT or ga1-3 ( Figure 6B ) . Finally , we also carried out GC-MS analysis of the flower cuticular lipids of the GA biosynthesis and signaling mutants . While flower waxes were not significantly altered in the ga1-3 and penta mutant flowers , the total cutin load , particularly of the 9/10 , 16-dihydroxy hexadecanoic acid ( C16-9/10 , 16-DHFA ) , the predominant monomer of the Arabidopsis flower cutin , was significantly different between WT and ga1-3 and between ga1-3 and the penta mutant ( Figure S9 ) . Nevertheless , SEM observation did not reveal any significant changes in the petal surface of the open flowers in the mutant plants ( Figure S9 ) . Since we applied exogenous GA to ga1-3 plants to induce flowering [43] prior to the SEM observation , this might explain the absence of a surface phenotype in mutant petal surface . All together , these results suggest that SHN transcription factors might play a key role in the GA-mediated flower organ development regulatory network .
The lack of any visual phenotype in floral organs of SHN1/WIN1 silenced plants ( [18]; Figure S3 ) , pointed to functional redundancy among the 3 SHN clade members . Even though expression of either one of the three SHN genes was not entirely reduced , the use of an artificial microRNA targeting the entire clade was sufficient to obtain several , striking , visual phenotypes that matched the previously described SHN genes expression patterns [16] . Floral organs were affected , likely as a result of altered cuticle composition , structure and consequently permeability . However , cuticle alteration might not be the only explanation to the defects observed in organ formation since they might also be a result of SHN genes effect on the process of epidermal cell differentiation and development . This was evidenced in the altered epidermal cells size and shape in petals and sepals of the 35S:miR-SHN1/2/3 plants . These strong epidermis phenotypes ( in pavement cells , trichomes and stomata ) observed previously in plants overexpressing either one of the three SHN genes support this proposal [16] . Down regulation of the SHN clade genes had an additional effect on floral organs as SEM and transmission electron microscopy ( TEM ) revealed changes in nanoridges that typically decorate surfaces of flower organs [44] . Formation of nanoridges in Arabidopsis flowers was recently associated with cutin , particularly with C16-9/10 , 16-DHFA , the major monomer of Arabidopsis petal cutin [22]–[23] , that was also dramatically reduced in the 35S:miR-SHN1/2/3 plants . However , the absence of nanoridges on the surface of tomato fruit that also contains C16-9/10 , 16-DHFA as a major monomer , suggests additional factors including polymer structure and distribution that mediate nanoridge formation [23] , [47] . Earlier work using promoter-reporter assays suggested that SHN transcription factors act not only in the interface between the plant and its environment but also at the interface between cells and cell layers [16] . Of particular interest was SHN2 that showed strict expression in the anther and silique dehiscence zones upon organ maturation . The proposed role of SHN transcription factors in the adhesion of cell layers was strongly corroborated by the recent finding that an SHN-like gene in barley ( Nud ) mediates the contact of the caryopsis surface to the inner side of the hull by forming a specialized lipid layer [19] . In this study we detected earlier abscission of floral organs in the silenced lines which corresponded well with SHN genes expression in the base of sepals , petals , stamens and siliques in the abscission region . Organ separation events including pod shatter , seed detachment from the maternal plant , pollen separation after meiosis , anther dehiscence and floral organ abscission , are thought to be associated with alterations to properties of the cell wall matrix , mainly pectins and wall proteins [1] , [48]–[49] . The pectin degradation activity of polygalacturonases ( PGs ) has been linked with all separation events described above . Recently , three Arabidopsis PGs have been associated with cell separation during reproductive development [50] . One of these , ADPG1 , displayed altered expression in the 35S:miR-SHN1/2/3 plants and its promoter was shown here to be activated by SHN1/WIN1 and SHN2 . Thus , SHN action on organ adhesion/separation possibly combines modification to cuticular lipids ( i . e . cutin ) as well as pectins of the cell wall . Array analysis revealed a concise set of genes that are putative downstream targets of the SHN transcription factors in flower buds , only two out of them ( CYP86A4 and CYP86A7 ) overlapped with the previously reported group of 11 SHN1/WIN1 putative targets [18] . This could be explained by the fact that while Kannagara et al . ( 2007 ) detected genes that were upregulated following induction of SHN1/WIN1 in fully expanded leaves [18] , we examined flower buds in which the SHN clade genes were co-silenced . Thus , genes from these two experiments most likely represent downstream targets in either leaves or flowers or both tissues . Together , these studies also demonstrated that wax load changes in the SHN overexpression lines were probably an indirect effect . SHN transcription factors emerge as regulators of genes derived from four prominent families associated with the cuticle including two cytochrome P450s of the CYP86A clade ( CYP86A4 and CYP86A7 ) , BDG3 , encoding one of the five BDG1-like proteins [24] , three genes of the large family of GDSL-motif lipase/hydrolases [39] and one of the eight-member clade of fatty acyl-CoA reductases [25] . Apart from the latter , these genes or their family members have been reported to be involved in either cutin biosynthesis or polymer assembly in the extracellular matrix in plant reproductive organs [10] , [23] , [51]–[53] . FAR1 has been recently associated with formation of suberin , a polymer that is structurally related to cutin and is often deposited following cell to cell separation in aerial organs to form a protection layer that will shield against penetration of pathogens and dehiscence [25] , [54] . Below ground , endodermal suberin is thought to regulate the apoplastic movement of water and solutes into the stele [55]–[56] . The SHN3 expression in roots ( [16] , Figure S10 ) and the endodermal expression of FAR1 , BDG3 , CYP86A4 and At1g16760 ( Figure S10 ) suggested that the latter 4 genes are targets of SHN transcription factors both above and below ground . Hence , SHN transcription factors and their targets are not only involved in cutin assembly in reproductive organs but are likely to play a role in root suberin deposition . CYP86A4 was suggested to provide ω-hydroxylation activity that is complementary to CYP86A1 in the biosynthesis of suberin [57] and FAR1 was recently reported to be associated with generating primary fatty alcohols for suberin deposition [25] . However , the role of BDG3 and At1g16760 in root suberin remains to be determined . Previous reports regarding the SHN clade members highlighted their role in regulating the biosynthesis of cuticular lipids for surface formation [16]–[18] . However , the results of the present study imply that activity of these factors goes beyond regulating a single metabolic pathway ( i . e . cutin ) for cuticle formation and they take part in the genetic program that mediates floral organ morphogenesis , more specifically in determining organ size and shape as well as the formation of specialized epidermis cell types ( e . g . the petal conical cells ) . Related to this , gene expression changes detected in the 35S:miR:SHN1/2/3 flower buds strikingly resemble the ones implicated in the formation of the single epidermis cotton fiber cell during its elongation . These include altered expression of genes associated with cell wall loosening through modification of pectin [58] , genes associated with the build-up of a higher turgor by increased accumulation of the major osmoticum such as soluble sugars , K+ , and malate [27] , redox-related genes [59]–[60] , genes related to phytohormone biosynthesis and signaling cascades [61] . Flowering in Arabidopsis consists of three distinct phases: floral initiation , floral organ initiation and floral organ growth . Earlier studies on GA signaling revealed that GA promotes Arabidopsis petal , stamen , and anther development by opposing the function of the DELLA proteins [62] and that GA signaling is not required for floral organ specification but essential for the normal growth and development of these organs [63] . Different combinations of DELLA proteins are key to floral organ development ( RGA , RGL1 , RGL2 ) , because individual DELLA proteins have different temporal and spatial expression patterns [62] . The unique temporal and spatial expression patterns of SHN clade genes in the flower tissues [16] and their distinct expression patterns in response to the alteration of the GA signaling reported here suggest that SHNs might be part of GA floral regulatory networks . In this context , GA might act as a positive regulator of SHN1/WIN1 in the regulation of floral organs development ( i . e . elongation of petal , stamen , and anther ) [37] , [62] in the early stages of flower development . In addition , GA emerges as a negative regulator of SHN2 in modulating the cell separation processes related to silique and anther dehiscence , floral organ abscission in the later stages of flower development . Hence , GA might be involved in cuticle assembly during the expansion of petals and other floral organs . The growth and elongation of organs requires the interaction between the outer and inner cell layers , which is coordinated by hormonal signals [4]–[5] . GA has been shown to promote cutin synthesis during other growth related processes including the rapidly growing internodes of deep-water rice [64] , in extending stems of peas [65] , and in developing tomato fruit [66] . Similarly , in this study , GA application resulted in a significant increase in the cutin load of ga1-3 mutant flowers . Future studies positioning the SHN proteins in the wide genetic network that controls flower development will shed light on how cuticle and cell wall metabolism is coordinated with the processes of flowering and fertility .
All Arabidopsis plants used in miR-SHN1/2/3 experiment were in the Col-0 genetic background , while those used for DELLA or GA experiment were in Ler genetic background . Plants were grown on a soil mixture in a growth room at 20°C , 70% relative humidity , a 16/8-h light/dark cycle at a fluorescent light intensity of 100 µmol m−2s−1 . All knock out lines were bought from either ABRC or NASC , while GA biosynthesis and signaling mutant were kind gifts from Hao Yu ( National University of Singapore , Singapore ) and David Weiss ( The Hebrew University , Israel ) . Exogenous GA application was carried out as described [67] with minor modifications . 100 mM GA3 or ethanol containing water was fine sprayed daily for 6 days on 6-week-old plants , and the buds were collected for analysis . For the 35S:miR-SHN1/2/3 construct , the designed artificial miR-SHN1/2/3 sequence was directly synthesized from BIO S&T ( Bio S&T Inc . , Montreal , Canada ) . After being sequenced , it was put into pART7 vector , and finally subcloned to pART27 . Transformation to Agrobacterium tumefaciens strain GV3101 was done via electroporation and planta transformation was done via floral dipping as described [68] . Promoter sequences of the putative SHN target genes ( approximately 2 kb upstream of the start codon ) were cloned from WT genomic DNA , and coding sequences of the three members of SHN clade were cloned from WT flower cDNA , using yellow Taq DNA polymerase ( Roboklon Gmbh , Berlin , Germany ) with corresponding gene specific primer pairs ( Table S1 ) . Those promoters and TFs were cloned into pGreen II 0800-LUC vector and pBIN plus vector , respectively , and then transformed to Agrobacterium tumefaciens strain GV3101 . All DNA sequence cloned were examined by direct sequencing . Toluidine blue examination of cuticle permeability was performed as previously described [69] . For Rethinium red staining , the inflorescences of 7-week-old plants were fixed and embedded in LR White resin ( London Resin Co . , Basingstoke , UK ) as described previously [70] . Sections were cut to a thickness of 0 . 5–1 mm using a diamond knife on an Ultracut microtome ( Leica ) and sections were collected on glass slides . The slides were stained with 0 . 1% Rethinium red for 5 min and washed with double distilled water , and then observed with Nikon ECLIPSE E800 microscope . All electron microscopy works were done as previously described [22] . For scanning electron microscopy ( SEM ) , flowers from 7-week-old plants were collected , fixed with glutaraldehyde using standard SEM protocol [71] , dried using critical point drying ( CPD ) , mounted on aluminum stubs and sputter-coated with gold . SEM was performed using an XL30 ESEM FEG microscope ( FEI ) at 5–10 kV . For TEM , flowers from 7-week-old plants were collected and processed using a standard protocol [72] . The Epon-embedded samples were sectioned ( 70 nm ) using an ultramicrotome ( Leica ) and observed with a Technai T12 transmission electron microscope ( FEI ) . Total RNA was extracted from closed buds from 7-weeks-old WT and homozygous 35S:miRSHN1/2/3 T3 plants using RNeasy Plant Mini Kit ( Qiagen ) with an on column DNAse treatment . The subsequent microarray analysis and qRT-PCR analysis were performed as described previously [21] . For microarray analysis , the double-stranded cDNA was purified and served as a template in the subsequent in-vitro transcription reaction for complementary RNA ( cRNA ) amplification and biotin labeling . The biotinylated cRNA was cleaned , fragmented and hybridized to Affymetrix ATH1 Genome Array chips . Statistical analysis of microarray data was performed using the Partek® Genomics Suite ( Partek Inc . , St . Louis , Missouri ) software . CEL files ( containing raw expression measurements ) were imported to Partek GS . The data was preprocessed and normalized using the RMA ( Robust Multichip Average ) algorithm [73] . The normalized data was processed by PCA ( Principal Component Analysis ) and hierarchical clustering to detect batch or other random effects . To identify differentially expressed genes one-way ANOVA analysis of variance was applied . Gene lists were created by filtering the genes based on: fold change , p<0 . 01 , and signal above background in at least one microarray . Up-regulated genes were defined as those having a greater than or at least 1 . 5-fold linear intensity ratio while down-regulated genes were defined as those having a less than or at most −1 . 5-fold linear intensity ratio . The experiment was performed in duplicate , preparing two independent biological replicates from 5–6 plants each . Waxes were extracted and analyzed as described [22] . For cutin analysis , soluble lipids were extracted from leaf and closed buds by dipping them in 10 ml of a methanol/chloroform ( 1∶1 , v/v ) mixture for 14 days ( solvent changed daily ) . The tissues were dried , weighed ( about 10–20 mg ) and kept in N2 till analysis . The cutin was depolymerized and analyzed as described previously [22] , [54] . Petals from 7-week-old flowers were collected ( 60 petals each sample , n = 8 ) , cleared with chloroform and methanol ( 1∶1 ) , and then air-dried overnight [74] . Samples were ground with solid crystalline KBr to fine powder and pressed to 1-mm tablelets . FTIR spectra were acquired in the absorbance mode at a resolution of 4 cm−1 with 32 co-added scans at wave number range 4000 to 250 cm−1 using a NICOLE1 380 FITR Spectrometer ( Thermo Electron Corporation ) . Each spectrum was baseline corrected and spectral area normalized prior to generating average spectra and digital subtraction spectra . Primary component analysis was performed using Multiple Experiment Viewer . Inflorescence stems transverse sections were prepared according to Willats et al [75] . Regions ( 0 . 5 cm long ) of 7-week-old Arabidopsis stem ( 3th internodes from the bottom ) were excised and sectioned by hand to a thickness of ∼100–300 µm . Sections were immediately placed in fixative consisting of 4% paraformaldehyde in 50 mM PIPES , 5 mM MgSO4 , and 5 mM EGTA . Following 30 min of fixation , sections were washed in the PIPES buffer , and then in 1× PBS buffer . Petals and gynoecium transverse section were prepared as described [65] and In vitro immunocytochemistry was carried out as described by Verhertbruggen et al [34] . Sections were incubated for 1 . 5 h in 5-fold dilution of two new rat monoclonal antibody hybridoma supernatant ( LM19 and LM20 ) diluted in 5% Milk/PBS , respectively . After being washed by gently rocking in PBS at least three times , sections were incubated with a 100-fold dilution of anti-rat IgG ( whole molecule ) linked to fluorescein isothiocyanate ( FITC ) in 5% Milk/PBS for 1 . 5 h in darkness . After washing in PBS for at least 3 times , sections were mounted in a glycerol∶PBS ( vol∶vol , 1∶1 ) solution . Immunofluorescence was observed with Nikon ECLIPSE E800 microscope equipped with epifluorescence irradiation and DIC optics . Images were captured with a camera and NIS-Elements BR30 software . Transient assay was carried out as described [40] with the exception that 150 µg/ml instead of acetosyringone was included in the infiltration media [76] . Luminescence was measured using Modulus Microplate Luminometer ( Turner Biosystems , Sunnyvale , CA ) by mixing 20 µl sample extract with 80 µl Luciferase assay reagent or Renillase assay reagent , respectively , and the data was collected as ratio . Background controls were run with only the transcription factor , promoter-LUC , and pBIN Plus empty vector , and pBIN Plus empty vector with promoter-LUC in the preliminary assay , and pBIN Plus empty vector with promoter-LUC was chosen later for background control in all experiments due to its relatively higher induction of Luciferase activity than other plasmid tested . | The cuticular layer that covers all aerial parts of plants plays a vital role not only in the interaction with environment but also in plant development and growth . Despite the recent significant achievements in the identification of structural genes involved in cuticle biosynthesis and secretion , little is known regarding the regulation of metabolic pathways generating cuticular constituents , more specifically wax and cutin . The Arabidopsis AP2-type transcription factor SHINE1/WAX INDUCER1 ( SHN1/WIN1 ) was the first assigned regulator of a cuticle-related metabolic pathway; nevertheless , its mode of action and biological function remain uncertain due to redundancy with two additional clade members . Here , by co-silencing all three SHN clade members using an artificial microRNAs approach , we demonstrated that SHN transcription factors act redundantly in patterning reproductive organ surface , modulating processes associated with cell elongation , adhesion , and separation , which secure the proper function of these organs . It appears that SHN transcription factors act directly on downstream cutin and cell wall–modifying genes . These factors are likely part of the genetic network controlling floral organ development . Thus , SHN transcription factors link together cuticle assembly , cell wall remodeling , and flower development to form the archetypal surface of floral organs mediating plant reproduction through pollination and seed dispersal . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"plant",
"biology/plant",
"growth",
"and",
"development",
"plant",
"biology/plant",
"genetics",
"and",
"gene",
"expression",
"plant",
"biology/plant",
"biochemistry",
"and",
"physiology"
]
| 2011 | SHINE Transcription Factors Act Redundantly to Pattern the Archetypal Surface of Arabidopsis Flower Organs |
Vaccines that activate strong specific Th1-predominant immune responses are critically needed for many intracellular pathogens , including Leishmania . The requirement for sustained and efficient vaccination against leishmaniasis is to formulate the best combination of immunopotentiating adjuvant with the stable antigen ( Ag ) delivery system . The aim of the present study is to evaluate the effectiveness of an immunomodulator on liposomal Ag through subcutaneous ( s . c . ) route of immunization , and its usefulness during prime/boost against visceral leishmaniasis ( VL ) in BALB/c mice . Towards this goal , we formulated recombinant GP63 ( rGP63 ) -based vaccines either with monophosphoryl lipid A-trehalose dicorynomycolate ( MPL-TDM ) or entrapped within cationic liposomes or both . Combinatorial administration of liposomes with MPL-TDM during prime confers activation of dendritic cells , and induces an early robust T cell response . To investigate whether the combined formulation is required for optimum immune response during boost as well , we chose to evaluate the vaccine efficacy in mice primed with combined adjuvant system followed by boosting with either rGP63 alone , in association with MPL-TDM , liposomes or both . We provide evidences that the presence of either liposomal rGP63 or combined formulations during boost is necessary for effective Th1 immune responses ( IFN-γ , IL-12 , NO ) before challenge infection . However , boosting with MPL-TDM in conjugation with liposomal rGP63 resulted in a greater number of IFN-γ producing effector T cells , significantly higher levels of splenocyte proliferation , and Th1 responses compared to mice boosted with liposomal rGP63 , after virulent Leishmania donovani ( L . donovani ) challenge . Moreover , combined formulations offered superior protection against intracellular amastigote replication in macrophages in vitro , and hepatic and splenic parasite load in vivo . Our results define the immunopotentiating effect of MPL-TDM on protein Ag encapsulated in a controlled release system against experimental VL .
Leishmaniasis represents infections caused by obligate heterogenous kinetoplastid protozoan parasites , Leishmania , transmitted by the bite of infected female sand fly Phlebotomus . It afflicts hundreds of thousands of people in tropical and subtropical areas every year , is currently endemic in 88 countries , and poses a threat to 350 million people worldwide [1] , [2] . The spectrum of leishmaniasis ranges from self-limiting cutaneous lesions to severe chronic mucocutaneous infections to fatal visceral disease . The latter is responsible for 59 , 000 death tolls every year , a parasitic disease statistics surpassed by malarial infections [3] . Conventional chemotherapies are often inadequate , toxic and expensive or are becoming less effective because of the emergence of resistance , a clear risk for human health . Extensive studies reveal that leishmaniasis is one of the parasitic diseases that can be controlled by vaccination [4] . Until recently , live virulent L . major ( leishmanization ) was the best vaccine against human cutaneous leishmaniasis ( CL ) . However , this process was discontinued due to uncontrolled long-lasting skin lesions , the spread of HIV co-infection and the use of immunosuppressive drugs . Despite recent advances in immunology and determination of the factors that control the development of protective immune response in leishmaniasis , there are no available human vaccines against any form of leishmaniasis [5] . It is believed that persistence of a small number of parasites at the site of infection can trigger infection-induced immunity against CL [6] . This immunity requires the presence of leishmanial antigen ( Ag ) only rather than live replicating parasites [7] . An alternate way is to promote Ag exposure at the site of inoculation or to prevent Ag clearance using appropriate adjuvants like liposomes . Our group has recently demonstrated that stable cationic liposomes acted as a potent adjuvant to induce long-lasting protection against experimental visceral leishmaniasis ( VL ) [8] , [9] . However , these results were obtained through intraperitoneal ( i . p . ) immunization and without the use of any immunomodulator . The major obstacle to the development of this vaccine for human use is the route of immunization . Since the route of vaccination influences the development of immune responses for protection , or failure of protection [10] , the results obtained with i . p . immunization cannot be extrapolated to the clinically relevant subcutaneous ( s . c . ) route . Therefore to increase the prophylactic efficacy of liposomal protein vaccination through s . c . route against experimental VL , strategies are being attempted by choosing the best combination of adequate adjuvant with the vaccine delivery vehicle . The traditional paradigm of s . c . immunization proposes involvement of skin derived dendritic cells ( DCs ) , as biosensors , in Ag presentation that modulate the immune responses to the environmental stimuli . Despite the fact that delivery of liposomal Ag through s . c . route of immunization hindered the Ag uptake by draining lymph nodes ( DLN ) due to breakdown of liposomes in dermis [11] , a cationic liposomal formulation with the synthetic mycobacterial immunomodulator ( CAF01 ) exhibited substantial immune responses through activation of DCs against Mycobacterium , Chlamydia , and malaria [12] . Since the function of the liposomal formulation relied on the immunomodulator , successful liposomal vaccination would probably require an appropriate adjuvant along with the Ag delivery system . We chose cationic liposomal formulation with monophosphoryl lipid A ( MPL ) as the immunomodulator , a target for the intracellular Toll like receptor 4 ( TLR4 ) with an extensive history of use in humans . MPL has been used as an adjuvant in several human clinical trials , including vaccines for tuberculosis , malaria , hepatitis B , and leishmaniasis [13]–[16] . However , the combinatorial use of immune potentiator with suitable delivery vehicle against VL has received minimum attention . In the current study , we developed vaccination strategies using liposomal protein in association with MPL-TDM ( monophosphoryl lipid A- trehalose dicorynomycolate ) with proven vaccination efficacy through s . c . immunization in susceptible BALB/c mice against experimental VL . As the vaccine Ag we chose recombinant GP63 ( rGP63 ) from Leishmania donovani whose native form has been shown to be highly protective against VL in BALB/c mice [8] . Here in this study , we analyzed the potentiating effects of distearoylphosphatidyl choline ( DSPC ) -bearing cationic liposomes in presence of MPL-TDM for the first time . To this end , we monitored the involvement of DCs in the antigen presentation for activation of effector T cells , leishmanicidal activity of macrophages and role of T cells in eliciting protective immunity . Additionally we examined the impact of MPL-TDM and liposomes on prime-boost .
Studies were performed with 4–6 weeks old female BALB/c mice reared in the animal care facility of the Indian Institute of Chemical Biology under pathogen free conditions . All animal studies were done according to the Committee for the Purpose of Control and Supervision on Experimental Animals ( CPCSEA ) , Ministry of Environment and Forest , Govt . of India , and approved by the animal ethics committee ( 147/1999/CPSCEA ) of Indian Institute of Chemical Biology . An Indian strain of L . donovani ( MHOM/IN/83/AG83 ) was originally isolated from an Indian Kala-azar patient and maintained by serial passage in Syrian hamsters as described earlier [17] . The parasites were cultured as promastigotes at 22°C in Medium 199 ( Sigma-aldrich , St . Louis , MO ) supplemented with 2 mM glutamine , 25 mM HEPES , penicillin G sodium ( 100 U/ml ) , streptomycin sulphate ( 100 µg/ml ) and 10% heat inactivated fetal bovine serum ( FBS ) ( Gibco/BRL Life Technologies , Grand Island , USA ) . Parasites from stationary-phase culture were sub-cultured to maintain an average density of 2×106 cells/ml . A plasmid containing full-length gp63 from L . donovani ( pET16bLdgp63 ) was generated , expressed and purified as described previously [18] . Liposomal rGP63 was prepared by incorporation of rGP63 into the lipid bilayer of DSPC , cholesterol ( Sigma-aldrich ) and stearylamine ( Fluka , Buchs , Switzerland ) at a molar ratio of 7∶2∶2 and dissolved in chloroform followed by evaporating the organic solvents to form a thin film as described earlier [8] . Empty and Ag entrapped liposomes were prepared by dispersion of lipid film in either 1 ml of PBS alone or containing 250 µg/ml of Ag ( rGP63 ) . The mixture was then vortexed and sonicated in an ultrasonicator ( Misonix , New York , USA ) for 30 s , followed by incubation at 4°C for 2 h . The excess free rGP63 was removed by centrifugation at 100 , 000× g for 1 h at 4°C . The level of incorporation ranged between 65–70% . BALB/c mice were immunized s . c . two times at an interval of 2 weeks with 2 . 5 µg of rGP63 either free or entrapped within DSPC-bearing cationic liposomes , and or mixed with 25 µg of Sigma adjuvant system or MPL-TDM in a total volume of 100 µl . PBS , only liposomes , only MPL-TDM and liposomes mixed with MPL-TDM serve as controls . At 1 and 3 wks post-immunization , groups of mice were sacrificed for the analysis of immune response . In other experiments , mice were primed with liposomal rGP63 in association with MPL-TDM , followed by boosted with either rGP63 alone , or in association with MPL-TDM , liposomes or both . For experimental infections , mice were challenged 3 weeks post-immunization with 2 . 5×107 freshly transformed stationary-phase promastigotes in 200 µl PBS injected via the tail vain as reported previously [17] . Two days after immunization , DCs obtained from draining inguinal nodes were collected from the mice immunized with PBS , MPL-TDM plus liposome , and liposomal rGP63 plus MPL-TDM , and the Ag- presenting capacity was studied . In brief , lymph nodes ( LN ) were passed through 100 µm cell strainer and digested for 30 min at 37°C by 1 mg/ml of collagenase D ( Roche Diagnostics , Mannheim , Germany ) . DCs were positively isolated by using magnetized anti CD11c+ magnetic-activated cell sorting columns ( Miltenyi Biotec , Auburn , CA ) according to manufacturers' protocol and the purity was simultaneously checked by flow cytometry . To detect the Ag presenting capability of DCs , CD4+ T cells were isolated from mice vaccinated with rGP63 through magnetized anti CD4+ antibody by positive selection . 2×105 CD4+ T cells were co-cultured with 2×105 CD11c+ and CD11c− cells isolated from mice vaccinated with either liposomal rGP63 mixed with MPL-TDM , liposome plus MPL-TDM or PBS . Supernatants were collected 24 h later and assayed for IL-2 by sandwich ELISA using OptEIA kit ( BD Pharmingen , San Diego , CA ) . BMDCs were generated by culturing cells isolated aseptically from tibias and femurs of BALB/c mice in the presence of GM-CSF and IL-4 as reported previously [19] . Briefly , bone ends were cut and marrow flashed out with complete RPMI-1640 , passed through a nylon mesh to remove small pieces of muscles and debri . Cells were pelleted by centrifugation at 390 g for 10 min and resuspended in complete RPMI medium containing rGM-CSF ( 40 ng/ml ) and rIL-4 ( 30 ng/ml ) supplemented with 10% FCS at a density of 1 . 5×106/ml cells in 24 well plates . Two-thirds of the medium was replaced on day 4 of culture and maintained for additional 3 days inside 5% CO2 at 37°C . On day 8 of culture , most non-adherent and loosely adherent cells had acquired typical DC morphology and the phenotype was evaluated by flow-cytometry . On day 8 , cells were collected , washed twice with ice-cold PBS , and adjusted to 1×106 cells/ml in complete RPMI-1640 with 10% FCS and incubated in 24 well-plates with the adjuvants being tested . Supernatants were collected on days 1 , 2 , 3 and analyzed for IL-12 ( p40 ) and Nitric oxide ( NO ) . DTH was determined as an index of cell mediated immune response as described earlier [20] . The response was evaluated by measuring the difference of the footpad swelling at 24 h following intradermal immunization of the test footpad with 25 µl of rGP63 ( 200 µg/ml ) from that of control ( PBS-injected ) footpad with a constant pressure calliper . The levels of Ag-specific serum IgG and its different isotypes IgG1 and IgG2a was determined in serum samples from experimental mice before and after infection by ELISA . In brief , 96-well microtiter plates ( Maxisorp , Nunc ) were coated with rGP63 ( 5 µg/ml ) diluted in 0 . 02 M phosphate buffer ( pH 7 . 5 ) overnight at 4°C . The plates were blocked with 1% BSA in PBS at room temperature for 3 h to prevent non-specific binding . After washing with PBS containing 0 . 05% Tween-20 ( Sigma ) , the plates were incubated with 1∶1000 dilutions of mice sera at 4°C . The next day , the plates were incubated for 3 h at room temperature with horseradish peroxidase-conjugated goat anti-mouse IgG1 or IgG2a diluted 1∶500 in blocking buffer . Substrate solution ( 0 . 8 mg/ml o-phenylene diamine dihydrochloride 0 . 05 M phosphate-citrate buffer pH 5 . 0 containing 0 . 04% H2O2 ) ( 100 µl ) was added for 30 min . The absorbance was determined using an enzyme-linked immunosorbent assay ( ELISA ) plate reader ( Thermo , Waltham , MA ) at 450 nm . The spleen cells were aseptically removed from the immunized and infected BALB/c mice and single cell suspensions were prepared in RPMI 1640 supplemented with 10 mM NaHCO3 , 10 mM HEPES , 100 U/ml penicillin , 100 µg/ml streptomycin sulphate , 50 µM 2-ME and 10% heat inactivated fetal bovine serum . Erythrocytes were removed by lysis with 0 . 14 M Tris buffered NH4Cl . The splenocytes were then washed twice , resuspended in the culture medium and viable mononuclear cell number was determined by Trypan blue exclusion [21] . Then the cells were cultured in a triplicate in a 96-well flat bottom plate at a density of 2×105 cells/well in a final volume of 200 µl and stimulated with rGP63 ( 2 . 5 µg/ml ) . The supernatants collected were stored at −70°C for cytokine analysis . Measurement of IFN-γ , IL-2 , IL-12 ( p40 ) , IL-4 , IL-10 levels was carried out as detailed in the instructions supplied with a cytokine ELISA kit ( BD Biosciences ) . The accumulation of nitrite in the culture medium was measured as described previously [21] . Briefly , 100 µl of splenocyte culture supernatants were mixed with an equal volume of Griess reagent ( 1% sulfanilamide and 0 . 1% N-1-naphthylethylene diamine hydrochloride in 50% H3PO4 ) and incubated at room temperature for 10 min . Absorbance was then measured at 540 nm . At 3 weeks after the last booster immunization , macrophages collected from peritoneal exudates of vaccinated mice were allowed to adhere to cover slips in 0 . 5 ml RPMI 1640 containing 10% FCS at 37°C in 5% CO2 . Nonadherent cells were removed by washing with warm PBS after 2 h . Macrophages were infected with promastigotes on glass cover slips ( 18 mm2; 106 macrophages per cover slip ) in 0 . 5 ml of RPMI/10% FCS at a ratio of approximately 10 parasites/macrophage for 4 h . The unphagocytosed parasites were removed by warm PBS washing , and the infected macrophages were further incubated in complete medium for 72 h at 37°C in 5% CO2 . The cells were then fixed in methanol followed by staining with Giemsa for determination of intracellular parasite numbers . Prior to fixation , culture supernatants were removed at 72 h and frozen at −70°C for cytokine analysis . Parasite load was evaluated by limiting dilution assay ( LDA ) with slight modifications [22] . Briefly , a weighted piece of liver and spleen were isolated from mice and homogenized in complete Schneider's Drosophila Medium ( Invitrogen , Grand Island , USA ) containing 10% heat inactivated FCS and diluted with a same medium to a final concentration of 1 mg/ml . Five-fold serial dilutions of homogenized tissues were cultured in a 96 well tissue culture plate ( Nunc , Rosklide , Denmark ) for one month at 22°C . The reciprocal of the highest dilution that was positive for parasite growth was considered to be the concentration of parasites per mg of tissue . The total organ burden was calculated using the weight of the respective organs . For intracellular analysis of IFN-γ produced by CD4+ and CD8+ T lymphocytes of vaccinated and infected mice , flow-cytometry ( FACS Canto , BD Biosciences ) using FACS Diva Software was carried out . All the antibodies were purchased from BD pharmingen . Single-cell splenocyte suspensions were stimulated overnight with 5 µg/ml rGP63 or left unstimulated . Brefeldin A ( 10 µg/ml ) was added to the cultures 2 h before harvest . The cells were then washed in PBS containing 0 . 1% NaN3 and 1% FCS at 4°C and stained with PE-conjugated anti-CD3 , PerCP Cy5 . 5 conjugated anti-CD4 and FITC-conjugated anti-CD8 mAb at 4°C for 30 min . The cells were then permeabilized with Cytofix/Cytoperm ( BD Biosciences ) solution for 20 min at 4°C , and then stained with APC conjugated anti-IFN-γ mAb . After incubation at 4°C for 30 min in the dark , cells were washed with wash buffer and re-suspended in staining buffer prior to analysis . One-way ANOVA statistical test was performed to assess the differences among various groups . Multiple comparisons Tukey-Kramer test was used to compare the means of different treatment groups using the GraphPad InStat software . In some experiments two-tailed Student's t test was performed . A value of p<0 . 05 was considered to be significant .
We have previously shown that the i . p . administration of native GP63 in association with DSPC bearing cationic liposomes was sufficient to confer durable immunity against BALB/c mice challenged with virulent L . donovani [8] . To design effective strategies that are crucial in s . c . immunization , we developed and optimized suitable vaccine preparations using recombinant form of L . donovani GP63 adjuvanted with either MPL-TDM or cationic DSPC liposomes or both . Mice were immunized with rGP63 alone or in association with either MPL-TDM , liposomes or both and the immune response was investigated by rGP63-specific immunity ( Figure 1 ) . Since DTH response is the measure of cell mediated immunity in vivo [20] , we found mice receiving either rGP63 alone , in association with either MPL-TDM , liposomes or combined with both adjuvants showed significantly higher response compared to controls just 1 wk after final vaccination ( data not shown ) and maintained thereafter up to 3 weeks ( Figure 1A ) . Mice vaccinated with liposomal rGP63 mixed with MPL-TDM , however , exhibited significantly higher level of DTH compared to mice receiving either rGP63 alone or mixed with MPL-TDM ( p<0 . 001 ) ( Figure 1A ) . We next analyzed the cytokine responses in splenocytes isolated from vaccinated mice , following stimulation with Ag . The assays for IFN-γ , and IL-12 p40 production from immunized mice showed that vaccination with liposomal rGP63 mixed with MPL-TDM exhibited substantially higher immune response 1 week after final immunization ( data not shown ) and was further enhanced at 3 weeks post vaccination ( Figures 1B and 1C ) . Moreover , mice vaccinated with liposomal rGP63 in association with MPL-TDM showed significantly higher IFN-γ and IL-12 p40 than the mice receiving rGP63 plus MPL-TDM or liposomal rGP63 ( p<0 . 001 ) . Interestingly , the IL-4 response was also clearly evident in the mice receiving rGP63 combined with two adjuvants and was significantly higher than the mice receiving rGP63 with MPL-TDM or liposomes ( p<0 . 001 ) ( Figure 1D ) . In vitro analysis of mice sera obtained after the final immunization demonstrated that animals vaccinated with rGP63 adjuvanted with either MPL-TDM , or cationic liposomes or both generated enhanced levels of IgG2a ( associated with Th1-biased response ) . Moreover , combining liposomal rGP63 with MPL-TDM resulted in a significantly high IgG2a compared to mice receiving MPL plus rGP63 or liposomal rGP63 ( p<0 . 001 ) ( Figure S1A ) . Although low level of serum IgG1 was observed in all vaccinated mice , maximum generation was observed in mice vaccinated with liposomal rGP63 alone , and combined with MPL-TDM ( Figure S1B ) . These data suggest that s . c . immunization with rGP63 adjuvanted with combined adjuvants elicited significant cellular and humoral immunity at 3 weeks with early mixed Th1/Th2 immune response . It has been previously reported that TLR4 agonists like MPL can induce DCs to produce high levels of proinflammatory cytokines [23] , [24] , usually to a lower extent than its related compound LPS . To optimize formulations for prophylactic use , we investigated innate signals on DCs induced by MPL in presence or absence of cationic DSPC liposomes . Thus , we assessed the adjuvanticity of cationic DSPC liposomes , MPL-TDM and combined formulation ( MPL-TDM plus liposomes ) on DCs derived from the bone marrow of BALB/c mice . As shown in Figure 2A , culture with MPL-TDM , or cationic DSPC liposomes stimulated relatively high levels of IL-12 ( p40 ) . Interestingly , combining MPL-TDM with DSPC bearing cationic liposomes resulted in release of significantly higher levels ( 1100±43 . 3 pg/ml ) of IL-12 ( p40 ) compared to MPL-TDM ( 873 . 3±30 . 32 pg/ml ) ( p<0 . 05 ) or cationic DSPC liposomes ( 553 . 3±40 . 55 pg/ml ) ( p<0 . 001 ) and the response with the combinations was comparable to LPS ( 1013±7 . 26 pg/ml ) . Similarly , upon combined stimulation with MPL-TDM and cationic DSPC liposomes , the NO production from BMDCs was greatly increased ( 37 . 33±5 . 2 µM ) and were significantly higher than MPL-TDM ( 25±2 . 8 µM ) ( p<0 . 01 ) and liposomes ( 22±2 . 64 µM ) ( p<0 . 01 ) ( Figure 2B ) . These data indicated that combining MPL-TDM with cationic DSPC liposomes resulted in increased adjuvant activity on DCs in vitro . Taken together , these data illustrated the rational for using combined vaccine formulations to obtain the sustained immune responses in BALB/c mice . DCs , the most potent APC of the immune system , play a crucial role in priming T cell immunity during Leishmania infection [25] . However , the activation of T cells by DCs depends on their state of processing , maturation and activation . Since the finding that s . c . immunization with cationic liposomal rGP63 in association with MPL-TDM demonstrated highest immune response after vaccination , we explored whether the early immune responses were due to activation of T-cells through draining node DCs . Naïve BALB/c mice were immunized s . c . with liposomal rGP63 mixed with MPL-TDM . Injections with PBS and liposomes mixed with MPL-TDM served as controls . We isolated both CD11c+ and CD11c− cells from the LN draining the immunization site after 2d of immunization and co-cultured with CD4+ T cells isolated from mice immunized with liposomal rGP63 through i . p . route . Figure 2C illustrates that CD11c+ cells isolated from mice receiving liposomal rGP63 in combination with MPL-TDM , induced substantial amount of IL-2 from CD4+ T cells suggesting the ability of DCs to activate T cells . These data thus exemplified the role of DCs in the Ag presentation and concomitantly highlight the stimulatory effects of the vaccine comprising combined adjuvants on the immune system . Based on the result obtained above , it appeared that above MPL-TDM in presence of liposomes contributed to the adjuvant effect of the liposomal rGP63 based protein vaccine . We further examined whether the combined adjuvant formulation was required for boosting to obtain maximal responses . To address this point , we developed several vaccine regimens in which priming was carried out using liposomal rGP63 in association with MPL-TDM , while boosting was either with rGP63 alone , in association with MPL-TDM , with DSPC liposomes , or both . Boosting with rGP63 either with MPL-TDM , cationic liposomes or both showed comparable level of DTH , Ag-specific splenocyte proliferation , and IL-2 production ( data not shown ) 3 weeks after final vaccination . Significant enhancement of both the IgG2a and IgG1 isotypes with a dominance of IgG2a were observed in mice boosted with either rGP63 in association with MPL-TDM , or liposomes or both ( Figure S2 ) . However , mice boosted with liposomal rGP63 with ( 492 . 6±11 . 4 pg/ml IFN-γ , 183 . 2±6 . 8 pg/ml IL-12 p40 ) or without MPL-TDM ( 480 . 8±12 . 8 pg/ml IFN-γ , 174 . 2±6 . 46 pg/ml IL-12 p40 ) elicited comparable level of Th1 cytokine responses ( IFN-γ , IL-12 p40 ) and the magnitude was significantly higher ( p<0 . 001 ) than the group of mice boosted with rGP63 plus MPL-TDM ( 345±21 pg/ml IFN-γ , 93 . 2±6 . 1 pg/ml IL-12 p40 ) ( Figures 3A and 3B ) 3 weeks after final vaccination . Furthermore , IL-4 response was significantly higher in mice boosted with liposomal rGP63 with or without MPL-TDM ( Figure 3D ) compared to PBS control ( p<0 . 001 ) . In addition , although highest level of NO was produced by splenocytes of mice boosted with liposomal rGP63 in association with MPL-TDM , the response was not significantly higher than mice boosted with liposomal rGP63 ( Figure 3C ) . We next further compared and analyzed the IFN-γ producing CD4+ and CD8+ T cells in mice boosted with various vaccine potentials by intracellular cytokine staining through flow cytometry . Mice boosted with rGP63 alone showed significantly higher IFN-γ producing CD4+ ( 3 . 0±0 . 4% ) as well as IFN-γ producing CD8+ T cells ( 1 . 4±0 . 2% ) over PBS ( p<0 . 01 ) immunized group ( Figure 4 ) . Moreover , mice boosted with rGP63 plus MPL-TDM showed significantly higher expression of CD4+ IFN-γ+ ( 4 . 1±0 . 3%; p<0 . 01 ) and CD8+ IFN-γ+ T cells ( 2 . 6±0 . 1%; p<0 . 01 ) compared to controls after 3 weeks post-vaccination . Interestingly , highest expression of IFN-γ producing CD4+ ( 6 . 9±0 . 1% ) and CD8+ T cells ( 4 . 1±0 . 7% ) was observed in mice boosted with MPL-TDM in combination with liposomal rGP63 , which was comparable with mice boosted with liposomal rGP63 ( CD4+ IFN-γ+ T cells 5 . 7±1%; CD8+ IFN-γ+ T cells 3 . 9±0 . 4% ) . Taken together , these results indicated that liposomal rGP63 either alone or in combination with MPL-TDM during boost seemed to modulate the immune system more efficiently than mice receiving rGP63 plus MPL-TDM during boost . Since boosting with liposomal Ag or liposomal Ag with MPL-TDM showed almost comparable post vaccination immune responses , the question we asked that whether MPL-TDM is required in boosting or not . Because we observed that boosting with liposomal rGP63 either alone or in combination with MPL-TDM have accounted for comparable vaccine efficacy before L . donovani infection , we investigated whether this effect was also extended to experimental systems involving challenge infection with virulent L . donovani , an etiological agent of VL . It is well established that activated macrophages from vaccinated mice inhibit Leishmania multiplication and arrest the parasite growth efficiently [26] . To assess the activation of antimicrobial responses elicited by macrophages to limit parasite multiplication , murine peritoneal macrophages were isolated from different prime-boost regimens and incubated with L . donovani promastigotes for 4 h . We observed that the numbers of infected macrophages from mice boosted with either rGP63 alone , entrapped within liposomes , in association with MPL-TDM or adjuvanted with both were significantly lower than the controls after 72 h infection ( Figure 5A ) . Mean number of amastigotes per macrophage was significantly controlled in groups of mice boosted with liposomal rGP63 with MPL-TDM than mice receiving liposomal rGP63 alone . Thus , highest anti leishmanial activities were observed in mice receiving liposomal rGP63 and MPL-TDM during both prime and boost . Since stimulated macrophages produce IL-12 ( p40 ) in response to intracellular pathogens , we investigated the production of this cytokine in supernatants of macrophages infected with parasite in vitro . Although detectable IL-12 ( p40 ) was released from all the vaccinated mice , highest release of IL-12 was observed in mice receiving liposomal rGP63 in association with MPL-TDM ( Figure 5B ) . The production of IL-12 ( p40 ) in these macrophages correlated with the production of NO ( Figure 5C ) . Collectively , these results indicated that boosting with liposomal rGP63 in presence of MPL-TDM showed highest action toward parasite elimination in L . donovani infected macrophages compared to mice boosted with liposomal rGP63 only . We next analyzed the hepatic and splenic parasite load in all the different vaccine regimens at 3 months post challenge measured by limiting dilution . The results showed that mice boosted with rGP63 adjuvanted with both cationic DSPC liposomes and MPL-TDM had ∼2 and ∼1 . 5-log-fold reduced parasite burden in liver and spleen respectively compared to mice boosted with liposomal rGP63 only ( Figure 6 ) . Furthermore , the data shown in Figure 6 indicated that both the formulations reduced parasite burden ∼5–7-log-fold in liver , and ∼7–9 . 5-log-fold in spleen , compared to unvaccinated mice . This extent of protection obtained so far has not been achieved in subunit protein based vaccination through s . c . route in susceptible BALB/c mice against VL . To understand the mechanism underlying substantial and significant protection showed by the groups of mice receiving MPL-TDM in combination with liposomal rGP63 during prime-boost , we investigated the cellular and humoral responses after challenge infection with L . donovani , identifying cell types that can produce IFN-γ and cells responsible for sustained protection . Since previous reports described that T cell proliferation is impaired during active VL [27] , we explored Ag-induced T cell proliferation after infection ( Figure S3 ) . Although comparable levels of DTH and IgG isotype responses were observed in groups of mice boosted either with liposomal rGP63 alone or in association with MPL-TDM ( data not shown ) , the proliferative responses were significantly higher in mice boosted with rGP63 and MPL-TDM compared to mice boosted with liposomal rGP63 ( Figure S3 ) . Strategies that favour Th1 response during infection have been shown to reduce Leishmania infection [28] . Moreover , induction of IFN-γ has been found to be involved in resistance to Leishmania infection in murine model [29] , and is an essential upregulator of NO production by the macrophages . In contrast , IL-10 correlates with disease susceptibility [30] . Based on the important roles contributed to these cytokines , we investigated the levels of IFN-γ , IL-12 ( p40 ) , ( Th1 cytokines ) and IL-4 , IL-10 ( Th2 cytokines ) after 3 months of infection . Our data suggested that boosting comprising combined adjuvants showed significantly higher IFN-γ responses ( 583 . 6±16 . 6 pg/ml ) compared to boosting with liposomal rGP63 ( 531±11 pg/ml ) alone after 3 months post infection ( p<0 . 05 ) ( Figure 7A ) . Since IL-12 is a potent inducer of Th1 cells and the control of Leishmania infection requires generation of strong Th1 response [31] , we analyzed the IL-12 response in splenocytes of immunized mice after L . donovani infection . Figure 7B showed that mice boosted with liposomal rGP63 along with MPL-TDM exhibited significantly ( p<0 . 05 ) highest IL-12p40 ( 214 . 6±7 . 7 pg/ml ) compared to mice receiving liposomal rGP63 during boost ( 185 . 4±6 . 5 pg/ml ) . This response was also reflected in the generation of NO . There are significant differences in NO production between the groups of mice receiving either liposomal rGP63 alone or in combination with MPL-TDM at 3 months post-infection ( Figure 7C ) . Moreover , combined adjuvant administration with rGP63 or mice boosted with liposomal delivery and Ag showed comparable levels of IL-4 production . These responses were significantly lower in comparison to unvaccinated mice ( p<0 . 001 ) ( Figure 7D ) . In contrast , the Th1 suppressive cytokine , IL-10 , in the mice boosted with combined adjuvant formulations was significantly down-regulated than mice boosted with MPL-TDM plus rGP63 ( p<0 . 01 ) ( Figure 7E ) . Moreover , flow cytometric analysis showed 7 . 9±0 . 3% and 4 . 9±0 . 8% IFN-γ producing CD4+ and CD8+ , respectively , in the L . donovani challenged liposomal rGP63 and MPL-TDM immunized mice during prime-boost , compared with 5±0 . 9% and 3 . 3±0 . 4% of IFN-γ producing CD4+ and CD8+ , respectively , in the liposomal rGP63 boosted L . donovani challenged BALB/c mice ( Figure 8 ) . Therefore , mice boosted with liposomal rGP63 in association with MPL-TDM showed significantly higher IFN-γ producing CD4+ T cells compared to mice boosted with liposomal rGP63 ( p<0 . 05 ) after 3 months L . donovani infection . Collectively , these results suggest that combined administration of cationic DSPC liposomes and MPL-TDM with rGP63 in a prime-boost regimen resulted in a significantly sustained Th1 biased immune response , which is extremely effective against L . donovani multiplication in susceptible BALB/c mice .
Despite the fact that the safety of protein-based subunit vaccination makes them highly attractive for human administration , a major limitation in the development of such vaccine remains the poor immunogenicity when used alone [32] . We believe that the development of protein-based vaccination could be greatly influenced by using suitable and effective adjuvant systems that de facto promote the slow release of Ag at the site of immunization with their ability to trigger the immune system . With the wide assortment of vaccine strategies that are immunologically effective , several studies have demonstrated that combining immunomodulators with controlled-release technologies led to potent and impressive immune responses [12] , [33]–[36] . Toward this goal , our study shed new light on the approaches by using TLR4 agonist with liposomal delivery system , through s . c . route , in the susceptible BALB/c mouse model for translation of experimental results into humans . In the present study , we investigated the potentiating effects of MPL-TDM on cationic liposomal formulation against experimental VL . In choosing effective immunomodulators , MPL is of interest based on its protective efficacy in experimental models of leishmaniasis [16] , [37] , [38] , and safety and immunogenicity in humans [28] . MPL can activate the APCs through maturation of co-stimulatory molecules , and to secrete cytokines such as IL-6 , IFN-γ , IL-12 which are crucial for activation and maturation of T and B cells [39]–[41] . Moreover , MPL-TDM was found to induce Th1 biased responses [42]–[44] . In our recent study , we demonstrated that vaccination with leishmanial Ag in combination with MPL-TDM elicited protective immune response against L . donovani challenge infection in BALB/c mice [45] . Therefore it is speculated that the effect of MPL-TDM might be enhanced in the presence of vaccine delivery system , such as liposomes , typically for the protein-based vaccination . Although immunogenicity of purified Ag is enhanced by cationic liposomal delivery [8] , [9] , [17] , several studies showed that combination of a delivery vehicle and a Th1 inducing immunomodulator are prerequisite against Leishmania and tuberculosis [46]–[48] . Therefore , new generation vaccines are likely to be comprised of recombinant Ags used in conjunction with immunomodulators and delivery systems [49] . Developing effective vaccines against leishmaniasis is found to be difficult due to lack of appropriate vaccine adjuvants . Many attempts have been made to overcome this problem . Our earlier reports with leishmanial antigen ( LAg ) entrapped within DSPC liposomes showed protective efficacy against VL in BALB/c mice [21] . Moreover , cationic liposomes are commonly combined with immunomodulators to enhance the desired immune response towards Th1 . Interestingly , soluble leishmanial antigen ( SLA ) in association with cationic liposomes and noncoding pDNA bearing immunostimulatory sequences elicited impressive protective responses against VL [36] . However , in all these studies , animals were immunized through i . p . route . In addition , recent studies have shown that s . c . vaccination with liposomal rGP63 partially protected susceptible BALB/c mice when challenged with L . major promastigotes [50] . Therefore , the immune response elicited by liposomal formulation could be modified by using MPL , a potent immunostimulator [51] . MPL signals via TLR4 , which in turn activates the NF-κB and subsequent expression of pro-inflammatory cytokines . Since s . c . administration of MPL has been found to be successful against leishmaniasis [38] , we used MPL as a potentiator of liposomal rGP63 . Here , in this study , we evaluate the adjuvant role of MPL-TDM in liposomal rGP63 vaccine and compared the usefulness of these adjuvants in different prime boost regimens for the first time against VL . Although strong adjuvanticity is prerequisite during priming , the impact of boosting is essential to study [52] . However , the use of heterologous prime-boost strategy in which immune response is primed with DNA followed by boosting with recombinant protein Ag has been successfully used in various diseases [53] . Since the prime-boost approach can improve the effectiveness of existing vaccines and quality of immune responses , several leishmanial Ags have been examined against experimental VL [54] . However , optimization of immune response of a vaccine can also be tested by the use of different adjuvants during prime/boost with same Ag . Interestingly , optimizing immune response by modulating vaccine components during boost is an essential approach to design effective vaccines [55] . In this study , we extended these findings by designing several vaccine formulations particularly during boost . Following priming with liposomal rGP63 in association with MPL-TDM , a strong immune response was observed . Although the adjuvant effect of MPL-TDM during boost was less than the cationic liposomes , the response was stimulated in combination with liposomal Ag delivery . While both the prime and boost influenced the immune response after immunization comparably , the ultimate difference was observed after challenge infection . Therefore , combination of delivery systems along with MPL is a novel approach for designing effective vaccines [51] . This adjuvant combination efficiently inhibited L . donovani multiplication in macrophages in vitro when administered during both prime and boost . Resistance to leishmaniasis is associated with a predominant IFN-γ and IL-12 production from the Ag-specific T cells [56] , [57] . The results of the current study showed significantly higher levels of IFN-γ , and IL-12 ( p40 ) in groups of mice receiving rGP63 in presence of liposomes and MPL-TDM during prime and boost than mice receiving liposomal rGP63 during boost , after L . donovani challenge infection . While MPL is reported to promote IFN- γ production by Ag-specific CD4+ T cells and skewing the immune response towards Th1 type [40] , liposomal Ag showed protective response against leishmaniasis [21] . Because , NO is generated after macrophage activation by IFN-γ and plays an important role in controlling leishmaniasis [58] , we measured the NO content in splenocytes of all vaccinated mice after L . donovani infection . Significantly higher NO was obtained in mice receiving combined adjuvant formulations than mice boosted with liposomal rGP63 . To this end , the protective response showed that vaccination with liposomal rGP63 along with MPL-TDM showed almost 2-log-fold , and 7–10-log-fold reduction in parasite burden compared to mice boosted with liposomal rGP63 and unvaccinated mice , respectively . The ultimate impact of boosting was seen after challenge infection might be due to high Th1-biased response . Therefore , a clear requirement of all the vaccine components ( rGP63 , MPL-TDM , and liposomes ) was evident for both priming and boosting . Until now , this extent of protection achieved using recombinant protein-based subunit vaccination formulated with MPL-TDM and a delivery system has not been achieved with any other subunit protein-based vaccination against experimental VL . In conclusion , the advantage of simultaneous administration of liposomal Ag and TLR ligand through s . c . route is efficacious against experimental VL . This is the first report of using different prime/boost approach with TLR4 ligand along with liposomal delivery of subunit vaccination against experimental VL in susceptible BALB/c mice . This formulation resulted in activation of DCs , increased T cell response , higher IgG2a/IgG1 ratio resulting in superior protection against L . donovani infection . | Visceral leishmaniasis ( VL ) , a vector-transmitted disease caused by Leishmania donovani , is potentially fatal if left untreated . Vaccination against VL has received limited attention compared with cutaneous leishmaniasis , although the need for an effective vaccine is pressing for the control of the disease . Earlier , we observed protective efficacy using leishmanial antigen ( Ag ) in the presence of either cationic liposomes or monophosphoryl lipid A-trehalose dicorynomycolate ( MPL-TDM ) against experimental VL through the intraperitoneal ( i . p . ) route of administration in the mouse model . However , this route of immunization is not adequate for human use . For this work , we developed vaccine formulations combining cationic liposomes with MPL-TDM using recombinant GP63 ( rGP63 ) as protein Ag through the clinically relevant subcutaneous ( s . c . ) route . Two s . c . injections with rGP63 in association with cationic liposomes and MPL-TDM showed enhanced immune responses that further resulted in high protective levels against VL in the mouse model . This validates the combined use of MPL-TDM as an immunopotentiator and liposomes as a suitable vaccine delivery system . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"immunity",
"immunology",
"biology",
"immune",
"response"
]
| 2011 | Potentiating Effects of MPL on DSPC Bearing Cationic Liposomes Promote Recombinant GP63 Vaccine Efficacy: High Immunogenicity and Protection |
Bacteria exploit an arsenal of antimicrobial peptides and proteins to compete with each other . Three main competition systems have been described: type six secretion systems ( T6SS ) ; contact dependent inhibition ( CDI ) ; and bacteriocins . Unlike T6SS and CDI systems , bacteriocins do not require contact between bacteria but are diffusible toxins released into the environment . Identified almost a century ago , our understanding of bacteriocin distribution and prevalence in bacterial populations remains poor . In the case of protein bacteriocins , this is because of high levels of sequence diversity and difficulties in distinguishing their killing domains from those of other competition systems . Here , we develop a robust bioinformatics pipeline exploiting Hidden Markov Models for the identification of nuclease bacteriocins ( NBs ) in bacteria of which , to-date , only a handful are known . NBs are large ( >60 kDa ) toxins that target nucleic acids ( DNA , tRNA or rRNA ) in the cytoplasm of susceptible bacteria , usually closely related to the producing organism . We identified >3000 NB genes located on plasmids or on the chromosome from 53 bacterial species distributed across different ecological niches , including human , animals , plants , and the environment . A newly identified NB predicted to be specific for Pseudomonas aeruginosa ( pyocin Sn ) was produced and shown to kill P . aeruginosa thereby validating our pipeline . Intriguingly , while the genes encoding the machinery needed for NB translocation across the cell envelope are widespread in Gram-negative bacteria , NBs are found exclusively in γ-proteobacteria . Similarity network analysis demonstrated that NBs fall into eight groups each with a distinct arrangement of protein domains involved in import . The only structural feature conserved across all groups was a sequence motif critical for cell-killing that is generally not found in bacteriocins targeting the periplasm , implying a specific role in translocating the nuclease to the cytoplasm . Finally , we demonstrate a significant association between nuclease colicins , NBs specific for Escherichia coli , and virulence factors , suggesting NBs play a role in infection processes , most likely by enabling pathogens to outcompete commensal bacteria .
Bacteriocins were the first of the interbacterial competition systems to be discovered [1] . They take the form of antimicrobial peptides or proteins released to the environment that have activity against closely related organisms . Experimental and theoretical work has suggested that bacteriocins are important agents of competition between microbial communities [2–6] . The potent antimicrobial activity of bacteriocins has generated interest in their application in agriculture [7] , the food industry [8] , and as potential clinical therapies for bacterial infections [9 , 10]; however , without understanding the distribution of bacteriocins , the significance of their role in bacterial competition remains uncertain and their exploitation consequently limited . Here , we circumvent one of the problems in identifying protein bacteriocins in bacterial genomes , their sequence and domain diversity , by developing a bioinformatics pipeline that identifies enzymatic nuclease bacteriocins ( NBs ) . NBs are one of the major classes of protein bacteriocins . The new tools we have developed allowed us to determine NB species distribution and evaluate current ideas as to how such molecules are deployed by Gram-negative bacteria . The best studied of the Gram-negative protein bacteriocins are the colicins produced by E . coli . Colicin gene expression is typically coupled to the SOS stress response and nutrient status of the cell [11 , 12] . Despite varying in size , colicins have a common tripartite structure , composed of an N-terminal translocation ( T- ) domain , a central receptor binding ( R- ) domain and a C-terminal cytotoxic domain . Colicins deliver one of five cytotoxic activities into cells; a pore-forming ionophore or lipid II hydrolase are delivered to the periplasm whereas DNases , tRNases or rRNases are delivered to the cytoplasm [13] . Colicin-producing bacteria protect themselves through the action of a small immunity protein that in the case of nuclease colicins is co-expressed and released from cells bound tightly to its colicin [14–16] . Colicins enter a target cell by binding an outer membrane receptor , typically an iron or vitamin transporter , and then parasitizing another outer membrane protein as a translocation portal for its T-domain to contact one of two trans-periplasmic proton motive force ( pmf ) -coupled systems: Tol for group A ( e . g . ColE9 ) or Ton for group B ( e . g . ColD ) colicins [13] . The immunity proteins of nuclease colicins are dissociated at the cell surface in a pmf-dependent step during translocation [17] . Colicin-like NBs have been identified in pathogens such as Klebsiella pneumoniae ( klebicins ) and Pseudomonas aeruginosa ( pyocins ) , all having a distinct multi-domain architecture and associated immunity protein [18–20] . Colicins ( and colicin-like proteins ) are thought to play a role in the co-existence of mixed microbial communities , and possibly as virulence determinants during proliferation within such communities [2 , 21] . ExPEC bacteria , for example , show a significant association between virulence factors and colicins in E . coli isolates that cause bacteraemia [22 , 23] . In a mouse gut infection model , colicin production by Salmonella enterica serovar Typhimurium is exploited by the organism to outcompete commensal E . coli during inflammation [24] . A recent survey of Shigella sonnei isolates in Vietnam spanning three decades identified a colicin-producing plasmid as one of the factors that overcame a dispersal bottleneck , helping the organism become established in the population as the major cause of dysentery [25] . Hence identifying the ecological niches in which colicins and colicin-like proteins are found and ascertaining their prevalence amongst pathogens are missing pieces of the microbial ecology puzzle that has ramifications for clinical epidemiology , biotechnology and biomedicine . Multiple factors make NB genes difficult targets for in silico identification . First , NBs are often ( but not always ) plasmid-encoded and only present in a fraction of isolates [26] . Second , extensive sequence diversification and domain rearrangements hamper identification of NBs . Finally , the nuclease domains of NBs are similar to the cytotoxic domains of other polymorphic toxin families , such as contact-independent inhibition ( CDI ) systems , further complicating their identification . In this study , we establish methods for the identification of NBs from large bacterial genome datasets . To overcome the low sequence identity between NBs , we use profile hidden markov models to identify conserved motifs in NB cytotoxic domains and their associated immunity proteins . We find that NBs are present at differing levels of abundance in bacterial species and that organisms exploit them in many varied ecological niches . Our analysis , validated for a novel NB shown to be active against P . aeruginosa , reveals the extent of domain rearrangements and sequence diversification that has occurred in NB genes , identifies a critical motif implicated in NB translocation and associates NBs in bacterial pathogenesis .
Protein bacteriocins are difficult to identify in bacterial genomes using conventional pairwise alignment strategies . We therefore developed a bioinformatics pipeline to identify bacteriocins using profile Hidden Markov models ( HMMer 3 . 1 ) [27] centred on the catalytic regions of the nuclease cytotoxic domain of the bacteriocin . Pore-forming bacteriocins were excluded from the study due to the absence of similarly conserved motifs . Five different NB families have been described in the literature , two DNases and three RNases ( Fig 1 and S1 Fig ) and the profile pairs were based on conserved motifs identified in their catalytic centres . For example , HNH-type endonuclease bacteriocins have a 30-residue motif ( also referred to as the ββα-Me motif ) as their active site ( HHX14NX8HX3H ) ( Fig 1 ) . Conserved motifs are similarly identifiable in the cytotoxic domains of non-HNH DNase , tRNase and rRNase NBs [28 , 29] ( S1 Fig ) . These motifs are however frequently found in other bacterial contexts; for example , the HNH motif is also found in Type II restriction and homing endonucleases , mismatch repair enzymes , and is the catalytic core of the CRISPR-Cas9 complex [30–32] . Hence three additional criteria were used to distinguish NBs from other systems and to validate them as bona fide NBs . First , the presence of conserved motifs within the adjoining immunity protein . Immunity proteins are specific , tight-binding inhibitors of bacteriocins that protect the producing cell from the action of its own toxin [16] . In the case of NBs , immunity genes are co-expressed with the bacteriocin . The resulting heterodimeric complex is often released from the cell through the action of a bacteriocin-release protein or lysis protein , which is often encoded within bacteriocin operons [13] . In cases where a bacteriocin-release protein has not been identified latent prophage genes have been implicated in release [33] . NB-specific immunity proteins are only jettisoned during translocation of the toxin into a susceptible cell [17] . To increase the specificity of our searches we exploited the highly conserved genetic link between NB genes and their cognate immunity genes , creating ‘profile pairs’ where the immunity profile is within 60bps ( or 200bps in the case of ColE5/pyocinS4 tRNase ) downstream of the cytotoxic profile ( S1 Table , S1 and S2 Figs ) . Second , the length of the NB sequence . NBs need multiple domains to translocate across the two membranes of the cell envelope and are typically >60 kDa in size . To be conservative , we filtered by sequence length , accepting sequences between 350–950 residues . Third , putative bacteriocins and the genes at the N-terminus are analysed for the presence of domains readily identifiable through the PFAM database as linked to other polymorphic toxin systems are excluded; for example , TSS6 effector proteins , which are often NB-related nucleases , or hemagglutinin repeats that are associated with contact dependent inhibitors . We subjected 100 , 154 PubMLST Multispecies isolate database , covering >3 , 000 species , to our pipeline and identified 3094 bacteriocin genes in 2479 isolates across 53 species ( S3 Fig , S2 Table ) . The largest non-bacteriocin contaminants of the database were uropathogenic-specific protein ( usp ) from E . coli and various short DNase containing proteins predicted to be effectors of the T6SS . E . coli usp has 40–50% sequence identity to the HNH DNase cytotoxic domain of NBs and was identified in over 2000 E . coli and Salmonella enterica genomes [35] . They were not included in our analysis as Usp has been shown to have genotoxic effects towards mammalian not bacterial cells and also contain an N-terminal domain of unknown function ( DUF769 ) , predicted to be a member of the Hcp-1 proteins of T6SS [35 , 36] . A group of HNH containing nucleases were also located in the Gram-positive Bacillus genera which are linked to the SUKH ( Syd , US22 , Knr4 homology ) domain immunity protein superfamily that bind to a variety of bacterial toxins of diverse nuclease and nucleic acid deaminase families [37] . The HNH motif was dissimilar to the NB DNase motif , having an extended linker region between the final histidine residues , HHX14N8HX8H . A non-HNH DNase T6SS effector was present in over half of Klebsiella pneumoniae strains . Other polymorphic toxin systems that share similar cytotoxic domains with NBs include the MAF proteins of Neisseria and a family of large T6SS effectors in the Vibrio genus that were identified by a LysM domain ( PFAM 01476 ) and a recently described motif associated with type VI secretion , the MIX domain [38 , 39] . Interestingly , we identified several bacterial species that utilize NBs as well as other competition systems such as T6SS , including Pseudomonas aeruginosa , Escherichia coli and Yersinia pseudotuberculosis . The dynamics of bacteria could utilise both diffusible and contact dependant competition systems is still unknown . We next looked for an association between NBs across different environmental niches . The PATRIC database contains 6256 genomes with associated environmental data [40] . The >1500 types of listed environment were grouped by an automated keyword search using custom python scripts . 120 unique environments were identified across 239 NB containing bacteria . Whilst it is not surprising that the majority of NB containing E . coli were identified in the GI tract and the majority of Pseudomonas species associated with either soil or plant based environments , we also observed these bacteria in the invasive environments of blood and urine including patients suffering septicemia ( P . aeruginosa VRFPA01 and P . aeruginosa VRFPA02 ) or urinary tract infection ( E . coli UMEA-3342-1 ) ( S4 Fig ) . NBs are found in significant proportions in the Enterobacteriaceae and Pseudomonadaceae families , however the proportion of a species that contained NB genes varied greatly; <5% E . coli strains encoded NBs compared to 85% for Pseudomonas aeruginosa and 31% for Klebsiella pneumoniae , which agrees with previous experimental work [41–43] . A prevalent feature of Pseudomonas and Klebsiella isolates is their possession of multiple NBs , which is likely to be a factor in NB evolution through increased rates of recombination , as has been proposed for pore-forming bacteriocins [41] . Klebsiella pneumoniae genomes commonly encode a DNase bacteriocin as well as cloacin , a rRNase bacteriocin . This trend has also been observed in Pseudomonas species [44] . A similar distribution is observed with Yersinia mollaretii genomes where 4 isolates contained both DNase and rRNase NBs . An important aspect of bacteriocin biology is how bacterial genomes encode these toxic molecules since this will have an impact on their distribution in bacterial populations . NBs of E . coli ( colicins ) are known to be plasmid encoded whereas NBs of P . aeruginosa ( pyocins ) are located on the chromosome [13] . As part of our analysis we instigated a bioinformatics protocol for establishing whether NBs were chromosomal or plasmid-encoded as this has not been shown for many species containing NB-encoding genes . Distinguishing these gene locations is , however , challenging due to the differing levels of assembly of the genomes interrogated by our pipeline . Using the Plasmid finder database and the Carattoli typing scheme [45] we could identify plasmid replicons within NB containing contigs . Addition of a conserved region identified for pColE1 replicons [46] allowed the typing of additional plasmids including large tRNase containing plasmids in E . coli ( S5a Fig ) . To demonstrate plasmid or chromosomal association of contigs containing NB genes for species not covered within this typing scheme , that could not be typed using the Plasmid Finder database , a python script was implemented to measure the percentage alignment of a contig to a database of known plasmids . As anticipated , NBs from Escherichia and Pseudomonas exhibited a strong association with plasmid and chromosomes , respectively ( S5b Fig ) . NBs from Yersinia species , however , demonstrated little or no alignment to plasmid sequences , suggesting these are chromosomally-located . Interestingly , species from Enterobacter displayed a bimodal distribution suggesting their NBs could be plasmid encoded or chromosomal . For a small number of NBs , these predictions were confirmed using NCBI fully assembled genomes; both plasmid and chromosomal NB genes were identified for Enterobacter isolates using this approach . In total , chromosomal NBs were identified in four genera: Pseudomonas , Serratia , Yersinia and Enterobacter . In the fully assembled genome of Klebsiella pneumoniae KPNIH12 , plasmid transfer genes and an NB gene ( klebicin B ) were incorporated into the genome and associated with a putative IS903 transposase . NBs with high sequence identity to that of Enterobacter cloacae ( cloacin DF13; CloDF13 ) were identified in species across multiple genera . CloDF13 is a ~60 kDa NB that utilizes an rRNase cytotoxic domain . MAUVE 2 . 0 was used to identify a plasmid in seven species that included transfer and mobilisation genes with over 90% identity to pCloDF13 , a broad range transmissible bacteriocinogenic plasmid [47] . pCloDF13 was identified in Enterobacter cloacae , Enterobacter aerogenes , Escherichia coli , Klebsiella pneumoniae and Salmonella enterica . A surprising conclusion of our analysis is that NBs are found exclusively in γ-proteobacteria ( Fig 2 , S6 Fig ) . Indeed , only one enzymatic bacteriocin has been identified outside of the γ-proteobacteria , a colicin M like bacteriocin in the β-proteobacterium Burkholderia [48] . We note , however , that unlike NBs which translocate to the bacterial cytoplasm to cleave nucleic acids , colicin M kills bacteria by cleaving the lipid II precursor of peptidoglycan within the periplasm [49] . In trying to rationalise why NBs appear only to be part of the antimicrobial armoury of γ-proteobacteria we considered two possibilities , the environment in which these organisms typically reside and the machinery involved in their uptake . Environment can be dismissed as a cause of this exclusivity since α- , β- and ε-proteobacteria live in many of the same environments as γ-proteobacteria that contain NB genes; for example , NBs are abundant in soil dwelling γ-proteobacteria yet α- , β- and δ-proteobacteria are all more prevalent in soil [50] . The translocation machinery hijacked by NBs , the Ton and Tol systems , are found beyond the γ-proteobacteria ( defined using a profile HMM strategy; S3 Table , S7 Fig ) , which suggests that the mechanism of NB import is also not a limiting factor in their species distribution . It remains unclear why NBs are exclusively found in γ-proteobacteria . No systematic in silico analysis has been conducted to ascertain the diversity of bacteriocin protein sequences in bacteria . Domain organization in NBs has conventionally been based on that found in colicins and pyocins . Colicins are composed of an amino terminal T-domain , a central R-domain and a C-terminal cytotoxic domain ( T-R-C ) , while pyocins are thought to have their T- and R- domains switched ( R-T-C ) . Rather than assign functionality to domains , which is largely unknown in our dataset , we analyzed the NB protein sequences identified through our bioinformatics pipeline for structural elements/motifs commonly associated with NBs , including intrinsically disordered and coiled-coil regions and structured domains described in the PFAM database ( Fig 3 ) . For this analysis , the cytotoxic nucleases of the sequences , which are always at the C-termini of NBs , were excluded as they provide little information on domains involved in import . Our approach captured a large proportion of previously identified NBs reported in the literature thereby validating the pipeline . Our analysis revealed five principles that underpin the diversity of NB sequences involved in their import . First , with the exception of some Pseudomonas species , ( which appear too diverse for meaningful analysis by CLANS therefore we present a phylogenetic analysis in S8 Fig ) ( see below ) , the majority of NB sequences ( 2866: ~93% ) can be classified into eight groups ( I-VIII ) based on sequence identity and predicted domain organization . Second , six of the eight groups come from the Enterobacteriaceae ( groups I , II , IV , V , VI and VII ) and two from P . aeruginosa ( groups III and VIII ) . Third , groups are generally not comprised of sequences from a single genus ( e . g . group V includes E . coli , Serratia marescens , Salmonella enterica and Klebsiella pneumoniae ) but can be populated entirely by a single genus ( e . g . group VI contains only Salmonella enterica and group IV contains only Yersinia spp NBs ) . Conversely , some species appear in multiple groups ( e . g . klebicins of K . pneumoniae are found in groups V and VII ) . Fourth , the degree of sequence identity within groups is high ( 40–90% ) but is very low between groups ( <20% ) . Fifth , only one sequence motif was common to all NBs across the eight groups , identified as the DPY motif in Fig 3 ( see below ) . Four of the eight groups have duplicated DPY motifs ( groups II , IV , V and VI ) although the biological relevance of this is currently unknown . Below , we highlight some of the main structural features of NBs for a select few species that encode them . E . coli NBs of the rRNase and DNase type were identified in the 1970s [54 , 55] and are commonly referred to as E-type colicins [16] . The present analysis places these NBs in group I , which also includes NBs from Klebsiella pneumoniae , Serratia marcescens , Shigella sonnei , Enterobacter cloacae , Xenorhabdus , and Citrobacter ( Fig 3 ) . Structures for both the rRNase NB ColE3 and DNase NB ColE9 from this group have been reported [56] , plus there have been numerous biophysical and structural studies of other NBs from this group [57] . The main structural features of group I NBs are a T-domain composed of a disordered N-terminal region adjoining a folded domain , the latter identified in the PFAM database as PFAM 03515 , and a central coiled-coil region which constitutes the R-domain . In E-type colicins the R-domain binds the vitamin B12 transporter BtuB , while disordered regions ( known as the Intrinsically Unstructured Translocation Domain or IUTD ) are involved in binding OmpF in the outer membrane and TolB in the periplasm as part of their import mechanism [58 , 59] . Group I NBs ( apart from those within Morganella ) have disordered regions at their N-terminus , which is also seen in five of the eight NB groups ( Fig 3 ) . It is likely these disordered regions are involved in the translocation mechanism of all NBs that contain them . Once in the periplasm the nuclease domains of NBs are translocated across the cytoplasmic membrane by the AAA+ ATPase/protease FtsH , which also proteolytically releases the nuclease to the cytosol [60 , 61] . The processing site identified by de Zamaroczy and co-workers [52] can be found in the NBs of groups I and V . K . pneumoniae is an important human pathogen and a leading cause of drug resistant infections . Multiple NBs , including four well characterized klebicins ( A-D ) , have been identified in K . pneumoniae isolates [62] . We identified klebicins A ( equivalent to cloacin DF13 , group I ) , klebicin B ( group VII ) and klebicin C ( group II ) , as well as two novel DNases which share less than 50% similarity to klebicin B or C at the N-terminus ( group VII ) . Sequences within group VII contained a large coiled-coil region at the N-terminus whereas group II klebicins have much shorter coiled-coil regions similar to group V NBs , which includes the well-studied colicin D [29 , 63] . Group II and group V NBs only differ in the presence of N-terminal disorder and overlap in terms of their sequence similarity clustering . The klebicins of group II contain two copies of the DPY motif whereas group VII NBs only have one . Serratia marcescens is a ubiquitous environmental organism that has become a major cause of healthcare associated infections , with increasing reports of antimicrobial resistance [64 , 65] . Only one Serratia bacteriocin ( the pore forming marcescin 28b ) has been identified to-date [66] . Using our pipeline , 11 NB families with no significant similarity to 28b were identified in 44 Serratia isolates . Bacteriocins containing either DNase and tRNase cytotoxic domains appeared to be interchangeable suggesting recombination of cytotoxic domains , a trait that is common in polymorphic toxins [67] . 8 of the 11 clusters were predicted to be chromosomally encoded and are present in group V . Three of the NBs are more closely related to cloacin DF13 , containing an identifiable N-terminal T-domain and 42% similarity to the T- and R- domains of cloacin DF13 ( group I ) , but have a long linker region and a repeat of the DPY motif and therefore form a separate group ( group II ) . The genus Yersinia has previously been shown to contain a number of bactericidal toxins only two of which have been characterized as bacteriocins . Pesticins from Yersinia pestis degrade peptidoglycan using murimidase activity and colicin FY is a pore forming bacteriocin [68 , 69] . We identified 18 NB sequence clusters at 90% sequence identity in 8 Yersinia species and identified both DNase and rRNase cytotoxic motifs . None of these sequences had significant similarity to pesticin or colicin FY . Yersinia NBs formed a separate group within the Enterobacteriaceae , the group IV NBs , although somewhat unexpectedly this did not include Y . pestis . These sequences have a ~90 residue unstructured region at the N-terminus similar to the IUTD of colicins and a long coiled-coil region , likely to be the receptor binding domain . 15 Yersinia NBs had a conserved region within the disordered N-terminal domain suggesting they may target the same translocation machinery . Pseudomonas aeruginosa is an opportunistic pathogen and a leading cause of nosocomial infections [70 , 71] . The S-type pyocins of Pseudomonas aeruginosa share a similar domain structure to colicins but with a rearrangement of their T- and R- domains and a small domain of unknown function between them [42] . An in silico investigation into the diversity of Pseudomonas toxins has previously identified several novel S-type pyocins [44] . Similar to this earlier study , we identified pyocins S1-9 ( excluding the pore former pyocin S5 ) and S11-12 , as well as two additional recombinations; pyocin S13 ( or SD1 ) which contains the colicin D like cytotoxic domain of S11 and S12 associated with the N-terminus of S1 [72] , and a pyocin we term S3C which contains the rRNase domain of colicin E3 associated with the pyocin S3 N-terminal domains and has not been reported previously in literature ( Fig 3 ) . Two novel HNH DNase motifs were also identified with 40–60% similarity to the S1 , S2 and AP41 cytotoxic domains . Overall , 85% ( 874/1024 genomes ) P . aeruginosa strains contained NB genes , which largely segregate to a single group in our analysis ( group VIII ) . A second , much smaller , group ( group III ) is populated only by a single pyocin ( S9 ) , also identified by Ghequire et al [44] . These proteins are predicted to have a disordered N-terminal domain , a single DPY motif and no identifiable coiled-coil region . NBs were identified in 20 Pseudomonas species other than P . aeruginosa . NBs containing an rRNase domain were identified in P . fluorescens , P . synxantha , P . poae , P . brassicacearum and 10 Pseudomonas sp . isolates . In total , 175 Pseudomonas spp . ( excluding P . aeruginosa ) genomes contain a DNase or E3-type rRNase domain ( 51 . 5% and 21% , respectively ) . NBs with a tRNase cytotoxic domain were not identified outside of P . aeruginosa species which is in line with previous analysis [44] ( S8 Fig ) . Many of the species observed were represented in the database by only a single genome suggesting the true extent of the sequence diversity within Pseudomonas spp . has yet to be established . Ghequire et al [44] also reported the presence of a group of NBs that contain a dual tandem repeat of the so-called pyocin_s domain ( see below ) , followed by a DNase domain ( either HNH or non-HNH ) in 10 isolates . We observed a similar domain organisation but note that the additional DNase domain does not contain a functional HNH motif . We analyzed the 3094 NB sequences captured by our bioinformatics pipeline using MEME [73] and identified a ~15-16-residue motif ( D-X4-FP-X8-Y ) located within the T-domain of nuclease colicins , which we refer to as the DPY motif ( Fig 4a ) . T-domains of NBs such as colicins are typically ~300 amino acids in length and share very low sequence identity ( <15% ) , with many loop insertions/deletions between the β-sheet secondary structure elements . The DPY motif is found in all NB sequences we examined and as such is a clear identifier of T-domains , the first of its kind . Sano et al . [20] previously highlighted a similar region in a few nuclease pyocins and colicins the deletion of which abolished cytotoxic activity [74] . The structures of three NB T-domains are known ( annotated in the PFAM database as PFAM 03515 ) ; colicin E3 ( 2B5U ) , colicin E9 ( 5EW5 [56] ) and the S-type pyocin domain from Erwinia carotovora ( 3MFB ) as well as the T-domain of the pore forming colicin B ( 1RH1 ) which shares high sequence identity to the T-domain of the NB colicin D . We performed structure-based sequence alignments and found that the DPY motif is integral to a much larger segment spanning the C-terminal half ( ~140 amino acids ) of the T-domain ( Fig 4b ) . This segment ( coloured blue in Fig 4b ) is identified in the PFAM database as the pyocin_s domain ( PFAM 06958 ) [42] . Nine amino acids identified within PFAM 06958 are conserved across the T-domains of all NBs , which are coincident with five β-strands of the structure ( Fig 4a ) . Finally , it is interesting to note that the location and indeed number of DPY motifs within NBs varies; it is generally placed at the C-terminal end of T-domains , which can be proximal or distal to the cytotoxic domain of NBs ( Fig 3 ) . The DPY motif forms a buried hydrogen bond network within PFAM 06958 that in colicin E9 involves Tyr285 and Asp270 , which is salt-bridged to Arg185 at the N-terminal end of PFAM 06958 ( Fig 4c ) . We mutated these three residues to alanine separately and in combination and found that mutation of Tyr285 ( by itself or in combination ) abolished cell killing whereas Asp270 and Arg185 mutations had no observable effect ( Fig 4d ) . With the exception of colicin B , a pore-forming colicin that shares a common ancestry with the nuclease colicin D , the DPY motif is found exclusively in NBs . The motif is not present in bacteriocins that are active in the periplasm such as the pore-forming colicins E1 , A , Ia , Ib and pyocin S5 and lipid II hydrolases such as colicin M . Hence , the DPY motif is specific to bacteriocins active in the cytoplasm . The involvement of the motif in translocation to the cytoplasm could be either at the outer membrane step , defining a path through the cell envelope , or the inner membrane , in a step prior to FtsH-mediated translocation of the nuclease to the cytoplasm . The ability of the pipeline to predict novel bacteriocins could improve our repertoire of potential species-specific antimicrobials for a number of clinically relevant species . As a test of the pipeline , we cloned and expressed a novel bacteriocin from Pseudomonas aeruginosa , which we call pyocin Sn . Pyocin Sn ( 832 residues ) resides within group VIII , has no detectable disorder domain , is characterized by a large coiled-coil region and a C-terminal , non-HNH type DNase domain . Purified pyocin Sn had cytotoxic activity against three different strains of P . aeruginosa ( S9 Fig ) . Hence our pipeline accurately identifies novel NBs in bacterial genomes . Recent work has highlighted the potential role of colicins as virulence factors in pathogenic bacteria , which can be used to displace commensal bacteria [24 , 25] . As yet however there has been no assessment of whether specific bacteriocins such as colicins are associated with pathogenicity . To look for an association between colicinogenicity and virulence we calculated the pangenome of 357 E . coli strains ( 166 colicinogenic strains and 191 non-colicinogenic strains ) using the roary pangenome . 41 , 723 gene groups were identified ( 1905 core genes ( 100–99% ) , 739 soft core genes ( 99–95% ) , 3 , 500 shell genes ( 95–15% ) and 35 , 579 cloud genes ( 15–0% ) ) . Association was determined by a fisher’s test followed by a Cochran-Mantel-Haenszel on BAPS clustered populations to correct for population stratification . 28 genes ( excluding genes that code for a colicin , immunity protein or plasmid mobility genes ) were found to be significantly associated with the trait of colicinogenicity ( Table 1 ) . These genes included virulence factors , toxin/anti-toxin modules , phages and genes involved in LPS and O-antigen biogenesis ( Fig 5 ) . Specific virulence factors are associated with pathotypes of colicinogenic E . coli such as the shigella toxin producing E . coli STEC and the enterohemorrhagic E . coli factor for adherence . Other genes associated with the carriage of colicin genes are general virulence factors such as hemolysins , which target the eukaryotic plasma membrane , and ureases that increase cytoplasmic pH to allow E . coli to survive in acidic environments . We also identify an association with phage genes suggesting either horizontal transfer of NB genes via bacteriophages or that associated phage lysis genes are being recruited for release of NBs , as has been proposed for other NBs [33] . The association of NBs with virulence factors suggests a role in the pathogenicity mechanisms of E . coli . This in silico evidence supports the hypothesis that bacteria use bacteriocins to displace native microflora in order to colonize and cause infection . In summary , we have developed and validated a pipeline for the identification of NB genes in bacterial genomes . Using this pipeline , we are able to show that the five NB families are widespread in γ-proteobacteria and that bacteria containing these genes occupy diverse ecological niches . Most have yet to be characterized . We also show that while NBs have high sequence diversity and multiple domain arrangements , they retain a conserved motif that is likely required for translocation of the nuclease to the cytoplasm . Finally , using a pangenome analysis of E . coli isolates we show that NB genes are associated with virulence factors supporting the hypothesis that NBs are exploited by invasive bacteria to displace host microflora .
Nucleotide fasta files for genomes from PATRIC [40] and NCBI assembled bacteria are freely available from ftp . patricbrc . org/patric2 ( accessed 11/07/2014 ) and ftp . ncbi . nlm . nih . gov/genomes/ASSEMBLY_BACTERIA ( accessed 14/05/2015 ) , respectively . We also made use of the PubMLST Multispecies isolate database website ( https://pubmlst . org/rmlst/ accessed 10/12/2014 ) [77] . All databases were subjected to the nuclease bacteriocin pipeline however due to the size and coverage of the pubMLST database ( 10-fold larger then PATRIC ) results are only shown for pubMLST . PATRIC was used to calculate environmental association as it has greater metadata and NCBI used to identify Ton/Tol operons as genomes are assembled . PATRIC and NCBI databases did not produce any novel NBs that were not identified in pubMLST . The PubMLST Multispecies isolate database contains NCBI Nucleotide database ( finished genomes ) and community contributed NCBI assembled genomes ( NCBI Assembly database ) as well as genomes assembled in-house from the publically available raw read data at the European Nucleotide Archive ( ENA-SRA ) . The PubMLST Multispecies isolate database aims to collect a wide range of bacterial species from all bacterial Classes that have been NGS sequenced in order to reflect the current knowledge of bacterial genomes and does not explicitly bias itself towards any particular subset of bacteria; however , it is subject to any bias of sequences deposited in the ENA . PubMLST Multispecies isolate database genomes are assembled in-house from the ENA-SRA database . The short read sequences were assembled using the Velvet genome assembly program ( v1 . 2 . 08 ) [78] . All odd-numbered kmer lengths between 21 and the read length were sampled using the VelvetOptimiser software ( v2 . 2 . 4 , bioinformatics . net . au/software . velvetoptimiser . shtml ) to automatically calculate the optimal assembly parameters for Velvet . Default parameters were used throughout with the exception that no scaffolding was performed and only contigs with 200bp or more were included in the final assembly . Data from pubMLST used in this study including source ( NCBI or ENA-SRA ) and accession within source database is provided ( S1 Data ) . Profile pairs shown in S1 Table were used to search for homologues of a NB and immunity protein in each of the translations using HMMscan ( HMMer3 . 1 ) with default search parameters ( http://hmmer . org/ ) . PFAM profiles were used for all five cytotoxic domains , HNH DNase , the non-HNH DNase pyocin S3 cytotoxic domain , tRNase ( both Colicin D-like and Colicin E5-like ) and rRNase [79] . To test for sensitivity , NBs identified in a first iteration of the pipeline were used to make secondary profile pairs ( iterated HMM ) . Regions that matched the conserved motifs were extracted using the easel toolkit associated with HMMer3 . 1 . Sequence alignments were performed using either MUSCLE [80] or CLUSTAL Omega [81] . Genomes from all databases were submitted through the pipeline as shown in S3 Fig . Briefly , genomes were translated into six frames . Each frame was aligned to the profile pairs in S1 Table using HMMer3 . 1 HMMscan with default parameters E-value cut off 5x10-2 . Custom python scripts were used to find the intergenic distance , here defined as the bases between the end of the cytotoxic profile alignment and the start of the immunity profile alignment . We do not measure the actual intergenic region between two open reading frames , as ORFs for immunity genes are often not identified by annotation software . For all cytotoxic domains apart from the ColE5 tRNase profile pairs with intergenic distances of greater than 60 base pairs were discarded . For the ColE5 like tRNase cytotoxic domain the distance was extended to 200bp as the profiles did not cover the end of the cytotoxic domain and the beginning of the immunity leading to an extended region between the two ( S2 Fig ) . Open reading frames containing the cytotoxic and immunity pairs were extracted using Prodigal v2 . 6 . 1 . ORFs greater than 950 residues and less than 350 residues were discarded as non-NB ORFs [13] . Sequences were aligned to the PFAM-A version 27 profile database . ORFs with significant alignments ( E-value <5x10-5 ) to active secretion systems or profiles not biologically relevant to NBs were rejected . To correct for incorrectly predicted ORFs we cluster sequences using cd-hit to find the longest ORF prediction for each sequence . Two additional profiles were used to discriminate T6SS from Vibrio and Klebsiella . A profile built containing the recently described MIX domain of Vibrio T6SS . A second profile was generated to remove T6SS effectors from Klebsiella pneumoniae which were located downstream of a T6SS Rhs protein . Conserved replicon sequences from the Carattoli typing scheme were downloaded from Plasmid Finder [45] . In addition , a conserved region from the ColE1 replicon were added [45] . An E-value cut off of 5x10-5 was used . Sequenced plasmids from NCBI and EBI were downloaded to generate a custom blast database of 3603 plasmids . Contigs containing a suspected NB were aligned to the database using blastn with default settings . Regions of the contig that aligned to a plasmid ( E value < 5x10-7 ) were summed and overlaps removed to determine the percentage of plasmid association . A small number of contigs could be verified using the NCBI fully assembled genomes that contain fully assembled chromosomes and plasmids . To detect horizontal gene transfer by colicinogenic plasmids found across multiple genomes , sequences were clustered using cd-hit with sequence identity of 90% and genome comparisons were performed using MAUVE2 [82] . 2785 fully assembled bacterial genomes with open reading frames predicted using Glimmer3 . 0 were downloaded from NCBI . S3 Table describes the PFAM profiles ( Pfam 27 ) used to identify conserved regions within proteins of the Ton/Tol operons . HMMscan ( HMMer3 . 1 ) was again used to identify profile matches in the open reading frames ( E-value cutoff 5x10-5 ) . Within each genome , regions of interest were identified by clustering co-ordinates of profile matches along the chromosome with a threshold of 2000 bases forming a new cluster . To differentiate between Ton , Tol and Mot operons , simple presence/absence rules were applied . For a region of interest to be classified as a Tol operon it needed to contain ExbD , MotA_ExbB and PD40 . Ton operons were defined as containing ExbD , MotA_ExbB , TonB_C but not the PD40 repeat which is a structural motif within TolB that does not have a homologue in the Ton operon . Cytotoxic domains were removed from bacteriocin sequences by removing the section of sequence that overlapped with the Pfam cytotoxic domains using HMMsearch and the easl toolkit . Sequences with cytotoxic domains removed were clustered using CLANS . Colored clusters were calculated using network approach where each sequence emits the clusters it is linked to weighted by the negative log of the p-value . This is iterated until clusters experience no further change . Only HSP hits p<1x10-10 were used in the analysis [51] . Coiled-coil regions were predicted using COILS [83] . Disorder was predicted using IUPRED [84] . Disordered regions were defined as predictions of long disorder greater than 0 . 5 stretching for ≥ 30 residues . 357 E . coli genomes were accessed from the ENA-SRA ( S2 Data ) and assembled using a Velvet based in—house assembly and improvement pipeline before annotation using Prokka [85] . Pangenome analysis was performed using the Roary pangenome software [86] and with similarity cut-off set at 95% . Phylogenetic trees were calculated from the core gene alignment produced by Roary using RAxML ( WAG model with Gamma correction ) [87] . The R stats package was used to perform the analysis . Gene clusters were tested for association with colicinogenic genomes by a fisher’s exact test . Population stratification was evaluated using BAPS [88] clustering and a Cochran-Mantel-haenzsel test . PFAM ( Pfam 27 ) domains from were predicted using HMMscan ( HMMer3 . 1 ) . Killing assays were performed by inoculating soft LB agar with either colicin sensitive E . coli JM83 for DPY mutations or one of three P . aeruginosa strains ( PAO1 , UCBPP-PA14 or PA14 ) for pyocin Sn sensitivity , and overlaid on LB agar . Serial dilutions were prepared using 20mM Tris-HCl pH 7 and 2 μl spotted onto the agar . Plates were incubated overnight at 37°C and scored on presence of a zone of clearance in the soft LB agar . Codon optimised synthetic genes encoding pyocin Sn ( GenBank: AHA26272 . 1 ) and its immunity protein in series ( Eurofins Genomics ) were ligated into the NdeI / XhoI sites of pET24a ( Novagen ) to give pNGH252 such that the immunity protein contained a C-terminal hexa-histidine tag . A second copy of the immunity protein , again with a C-terminal hexa-histidine tag , was ligated into the NcoI / HindIII sites of pACYCDuet-1 ( Novagen ) , to give pNGH260 . Both plasmids were transformed into BL21 ( DE3 ) cells to ensure the immunity protein expressed in excess of the bacteriocin . Cultures of pNGH252 pNGH260 BL21 ( DE3 ) were grown at 28°C in LB 50 μg/ml kanamycin , 34 μg/ml chloramphenicol to an OD600 nm of 0 . 7 upon which expression from both plasmids was induced through the addition of IPTG to a final concentration of 1 mM . Cells were grown for a further 16 hours , before being harvested by centrifugation , and lysed through sonication in 25 mM Tris-HCl , pH 7 . 5 , 500 mM NaCl , and 1 mM PMSF . Cell debris was removed through centrifugation at 17 , 500 xg with the supernatant being passed through a 0 . 45 μm filter . The pyocin Sn-ImHis6 complex was purified on a 5 ml HisTrap FF column ( GE Healthcare ) equilibrated in 25 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl eluting bound protein with a 0 to 250 mM imidazole gradient over 10 column volumes . The pyocin Sn-ImHis6 complex was further purified on a HiLoad 26/60 Superdex 200 column ( GE Healthcare ) equilibrated in 25 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl . The purified complex was quantified through A280 nm using a sequence based extinction coefficient of 70 , 250 M-1 . cm-1 . | Bacteria deploy a variety of antimicrobials to kill competing bacteria . Nuclease bacteriocins are a miscellaneous group of protein toxins that target closely related species , cleaving nucleic acids in the cytoplasm . It has proved difficult to establish how widespread bacteriocins are in bacterial populations due to the high diversity of bacteriocin-encoding genes . Here , we describe an in silico approach to identify nuclease bacteriocin genes in bacterial genomes and to distinguish them from other competition toxins . Bacteria that contain nuclease bacteriocin genes are found in many different types of environment but are prevalent in niches where interbacterial competition is likely to be high . Nuclease bacteriocins are found exclusively in γ-proteobacteria and are particularly abundant in the Enterobacteriaceae and Pseudomonadaceae families . Although the sequences we identify are indeed diverse ( <20% sequence identity between protein families ) we show that all nuclease bacteriocins contain an invariant motif , usually within a common structural scaffold , that is implicated in translocating the cytotoxic nuclease to the cytoplasm . Finally , we show that nuclease bacteriocins in pathogenic E . coli are strongly associated with virulence factors suggesting they play a role in pathogenicity mechanisms . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"nucleases",
"deoxyribonucleases",
"enzymes",
"pathogens",
"dna-binding",
"proteins",
"enzymology",
"microbiology",
"toxic",
"agents",
"toxicology",
"pseudomonas",
"aeruginosa",
"genome",
"analysis",
"sequence",
"motif",
"analysis",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"pseudomonas",
"sequence",
"analysis",
"bacterial",
"toxins",
"bacteriocins",
"genomics",
"bioinformatics",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"biological",
"databases",
"biochemistry",
"hydrolases",
"sequence",
"databases",
"database",
"and",
"informatics",
"methods",
"genetics",
"protein",
"domains",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"genomic",
"databases",
"organisms"
]
| 2017 | Diversity and distribution of nuclease bacteriocins in bacterial genomes revealed using Hidden Markov Models |
Many plant cells can be reprogrammed into a pluripotent state that allows ectopic organ development . Inducing totipotent states to stimulate somatic embryo ( SE ) development is , however , challenging due to insufficient understanding of molecular barriers that prevent somatic cell dedifferentiation . Here we show that Polycomb repressive complex 2 ( PRC2 ) -activity imposes a barrier to hormone-mediated transcriptional reprogramming towards somatic embryogenesis in vegetative tissue of Arabidopsis thaliana . We identify factors that enable SE development in PRC2-depleted shoot and root tissue and demonstrate that the establishment of embryogenic potential is marked by ectopic co-activation of crucial developmental regulators that specify shoot , root and embryo identity . Using inducible activation of PRC2 in PRC2-depleted cells , we demonstrate that transient reduction of PRC2 activity is sufficient for SE formation . We suggest that modulation of PRC2 activity in plant vegetative tissue combined with targeted activation of developmental pathways will open possibilities for novel approaches to cell reprogramming .
Plant cells have long been recognized for their capacity to become pluripotent and to enter various differentiation pathways in response to chemical or mechanical stimuli . This is exemplified by ectopic organ formation and entire plant regeneration from vegetative tissue in response to plant hormone and/or stress signaling , which is widely used for clonal propagation of horticultural species and as a developmental model system [1 , 2] . The most extensive reprogramming establishes high level of pluripotency or even totipotency that allows somatic embryogenesis—the development of ectopic embryos from somatic cells in a process that is independent of gamete formation , fertilization or seed development [3] . Somatic embryo ( SE ) formation is usually induced by exposing somatic tissue to plant hormones or to abiotic stress [1] . In Arabidopsis thaliana ( Arabidopsis ) , SEs can be efficiently produced by applying the synthetic auxin 2 , 4-dichlorophenoxyacetic acid ( 2 , 4-D ) to immature zygotic embryos [4–7] . In addition to hormone-mediated SE-induction , SE in Arabidopsis can be also promoted by ectopic overexpression of specific transcription factor ( TF ) genes , such as the homeodomain TF WUSCHEL ( WUS ) ) [8] , the AP2 TFs PLETHORA 4/BABY BOOM ( PLT4/BBM ) ) [9] or PLT5/EMBRYO MAKER ( PLT5/EMK ) ) [10] , the MADS box TF AGAMOUS-LIKE 15 ( AGL15 ) ) [11 , 12] or the LEAFY COTYLEDON genes LEC1[13] or LEC2[14] , or by the overexpression of the SOMATIC EMBRYOGENESIS RECEPTOR-LIKE KINASE 1 ( SERK1 ) ) [15] . In many plant species , hormone-mediated SE induction requires sexual reproduction , as a high potential for SE formation is limited to microspores or zygotic embryos ( ZEs ) [16–19] . This also applies to Arabidopsis where ZEs present the most efficient source of auxin-induced SEs [16 , 17] . Molecular barriers that prevent somatic embryogenesis from vegetative tissue remain poorly understood . Polycomb repressive complexes ( PRCs ) are well-studied epigenetic executors of developmental phase transitions and cell fate specification in animals and plants [20–22] . By establishing repressive chromatin structures at developmental regulators , they ensure stable down-regulation of developmental programs that are not required at a particular time [21 , 23] . PRC2 has histone methyltransferase activity for histone H3 lysine 27 trimethylation ( H3K27me3 ) . Distinct PRC2 complexes exist in Arabidopsis , which differ in composition and in function during development . The EMBRYONIC FLOWER ( EMF ) and VERNALIZATION ( VRN ) complexes are the PRC2 complexes acting in sporophytic tissues . The EMF and VRN complexes contain the Suppressor of Zeste 12 homologs EMF2 and VRN2 , respectively . In addition , the complexes share the Extra sex comb homolog FERTILIZATION-INDEPENDENT ENDOSPERM ( FIE ) , the p55 homolog MULTICOPY SUPRESSOR OF IRA 1 ( MSI1 ) and the partly redundant Enhancer of Zeste homologs CURLY LEAF ( CLF ) or SWINGER ( SWN ) [24 , 25] . Among other functions , PRC2 represses embryo maturation programs during the establishment of vegetative development in Arabidopsis [26 , 27] . Absence of PRC2 is accompanied by ectopic activation of late embryogenesis developmental programs in vegetative tissue , marked by expression of embryo maturation genes and ectopic accumulation of embryonic storage molecules [26 , 28–30] . Neutral lipids and expression of embryo maturation genes are also found in somatic tissue of mutants such as pickle [28 , 31] , suggesting that these features may also mark activation of biochemical pathways in the absence of altered development . It is , therefore , unclear what level of dedifferentiation is reached in PRC2-depleted plant cells . Here we show that absence of PRC2 by itself is not sufficient to achieve full cell dedifferentiation required for somatic embryogenesis but that PRC2-depleted somatic cells respond to external hormone and stress treatments by becoming competent for somatic embryogenesis . We define a set of morphological , anatomical and molecular features of SEs that distinguish them from neutral lipid-containing structures that spontaneously develop in PRC2 mutant plants . By combining stable and transient PRC2-depletion with hormone treatments , we demonstrate that cell fate can be reset and somatic embryogenesis induced in PRC2-depleted shoot and root tissue . We propose that lowering the PRC2-imposed epigenetic barrier combined with hormonal stimuli allows ectopic co-activation of key developmental regulators and establishment of embryogenic potential in plant vegetative tissue .
In Arabidopsis , external application of the synthetic auxin 2 , 4-D to immature ZEs induces reprogramming and development of SEs [4–7 , 16] . Somatic embryogenesis is induced following the removal of 2 , 4-D and requires the establishment of a gradient of the natural auxin indole-3-acetic acid ( IAA ) and expression of root- ( RAM ) and shoot-apical meristem ( SAM ) specifying genes [2 , 32–34] . We used a modification of an established protocol for SE induction from immature ZEs [16] to determine the shortest duration of the 2 , 4-D treatment required for efficient SE development . Wild-type ZEs ( early bent cotyledon stage , Fig 1A ) were exposed to 5 μM 2 , 4-D for 3 , 5 or 7 days ( Fig 1B and 1E ) after which they were cultured on hormone-free medium for additional 7 days . These treatments stimulated the development of mature primary somatic embryos ( Fig 1D and 1E ) . While exposure to 2 , 4-D for 3 or 5 days resulted in the emergence of trichome-bearing true leaves from the SAM region of the ZE ( Fig 1C ) , an exposure for 7 days resulted in the formation of callus-like tissue in the SAM region ( Fig 1B ) from which primary SEs emerged in 60–70% of ZEs ( Fig 1D–1F ) . Using the 7-day 2 , 4D-treatment ( Fig 1E ) , we tested the efficiency of SE formation using ZEs at different stages of development and seed germination . We found that the potential to form SEs decreases to below 10% in embryos from dry seeds and is lost in germinating embryos ( Fig 1G ) . PRC2 activity is dispensable for zygotic embryo development [26] but it is required for the repression of seed dormancy and late embryogenesis genes in germinating seedlings [26 , 27] . Mutants in the genes encoding the catalytic sporophytic PRC2 subunits CLF and SWN ( clf swn ) spontaneously develop poorly differentiated shoot organs and ectopic finger-like protrusions that ectopically accumulate neutral lipids [28 , 29] ( S1A Fig ) . Occasional SE formation was reported in clf swn [29] , indicating that under specific conditions , the PRC2-depleted tissue may become embryogenic . Interestingly , the loss of embryogenic potential in germinating seeds ( Fig 1G ) coincides with transcriptional upregulation of genes encoding PRC2 subunits , especially of CLF ( [27] S1B Fig ) . We hypothesized that the onset of PRC2 function during embryo germination may be one of the factors restricting the time window available for an effective auxin response and limit the SE potential in Arabidopsis . Although the SE-formation potential in ZEs is not significantly affected by the absence of either CLF or SWN ( S1C Fig ) , its loss in germinating clf mutant seeds is slightly retarded ( S1C Fig ) . As CLF and SWN act at least partly redundantly [29] , we tested the effect of combined CLF and SWN depletion on the SE potential in seedlings . Since clf-29 swn-3 ( further clf swn ) plants do not produce seeds and mutant phenotypes become visible only after germination [26] , clf swn seedlings were selected from the progeny of clf/+ swn/- plants 7 days after induction of germination . We exposed separated shoot and root explants of 7 , 14 or 21 day-old wild-type , clf , swn and clf swn plants to 5 μM 2 , 4-D for different periods of time followed by 7-day culture on hormone-free medium ( Fig 2 , S2A Fig ) . Callus tissue formed in wild type , swn and the majority of clf shoot and root explants after 7-day 2 , 4-D-treatment and 7-day cultivation on hormone-free medium ( Fig 2A and 2C ) . Structures containing neutral lipids that resembled wild-type ZE-derived SEs formed in 95% and 2 . 5% of clf swn shoot and root explants , respectively ( Fig 2A , 2D and 2K ) . Even a 60-hour 2 , 4-D-pulse was sufficient to induce SE formation in 87% of clf swn shoot explants ( Fig 2B and 2D ) , suggesting accelerated response to the treatment compared to wild-type ZEs ( Fig 1F ) . Staining by Sudan Red 7B showed that all three observed types of structures–namely the finger-like protrusions that spontaneously develop in clf swn [28 , 29] ( S1A Fig , Fig 2F–2I ) , the 2 , 4-D-induced clf swn structures ( Fig 2J–2L ) and the wild-type ZE-derived SEs ( Fig 2M–2O ) –contain neutral lipids ( Fig 2G , 2K and 2O , respectively ) . The 2 , 4-D-induced clf swn structures ( Fig 2J–2L ) nevertheless differed from mock-treated clf swn ( S1A Fig , Fig 2F–2I ) and resembled wild-type SEs ( Fig 2M–2O ) in several aspects . First , they were larger in size , being 2–3 mm long in 2 , 4-D-derived clf swn and wild-type SEs compared to ca 1 mm in the spontaneously-developing clf swn structures ( Fig 2J vs . 2G ) . Second , they formed morphologically clear root and shoot apical poles . Third , they were loosely attached to the parental explant at the root pole or at the root-hypocotyl transition zone , generally lacking vasculature and vascular connection to the parental explant ( S2B Fig ) . Due to their morphological similarity to wild-type SEs , we designated the 2 , 4-D-induced clf swn structures "SEs" in contrast to the "SE-like" structures spontaneously formed in clf swn that also contain neutral lipids typically present in the embryo but lack other resemblance to wild-type SEs . Efficiency of SE formation in clf swn shoot explants and SE morphology were independent of plant age ( S2A Fig ) but the SE morphology depended on the duration of the 2 , 4-D treatment ( Fig 2D vs . 2E ) . While short ( 60 h ) 2 , 4-D-exposure led to the formation of SEs resembling wild-type ZE-derived SEs ( Fig 2D ) , extended treatment induced 95% of explants to develop stunted malformed SEs with detectable root pole but lacking clearly distinguished cotyledons ( Fig 2B–2E ) . This observation is consistent with a reported inhibitory effect of extended auxin treatment [2] . Under the tested conditions , we have not observed SEs developing from clf swn hypocotyls , true leaves or the majority of root explants ( Fig 2P–2R ) . Most SEs originated form the shoot apex region ( Fig 2J ) and , occasionally , from the cotyledon margins ( Fig 2S ) . Similarly to wild-type ZEs ( Fig 1B ) , where no SEs were morphologically distinguishable after 7-day 2 , 4-D treatment , no SEs were visible on the shoot apex at the end of a 7-day induction in clf swn ( Fig 2T ) , suggesting that embryogenesis mainly occurs after the removal of 2 , 4-D . Since the perception , transport and metabolism of 2 , 4-D differs from other synthetic and natural auxins [35] , we tested the ability of 1-naphthaleneacetic acid ( NAA ) and indole-3-butyric acid ( IBA ) , two auxins routinely used in horticulture , to induce SEs in clf swn . 50 μM IBA induced 12% and 50 μM NAA 48 . 5% of the explants to form SEs , demonstrating that the SE-inducing effect is not limited to 2 , 4-D but can also be observed for other auxins ( S2C Fig ) . Together , these results showed that PRC2 activity in the shoot apex of germinated seedlings prevents auxin-induced somatic embryogenesis . The high induction efficiency ( approximately 85% of shoots forming SEs ) and relatively short induction time ( 50–70 hours ) offered the opportunity to establish an experimental model system to address processes associated with somatic embryogenesis . Size , formation of apical and basal poles and loose attachment to parental explants distinguished clf swn-derived SEs from SE-like structures ( Fig 2J–2L vs . 2F–2I , respectively ) . We therefore tested whether the morphological differences between SEs and SE-like structures reflected different activation of genes required for apical meristem and organ identity establishment . We compared the expression of genes marking root identity ( WOX5 , SCR , PLT1 , PLT2 ) and shoot apical meristem ( SAM ) identity ( WUS , STM ) in wild-type seedlings , ZEs and SEs to clf swn seedlings , SEs and SE-like structures ( Fig 3A , S3 Fig ) . No relevant differences that could distinguish SEs from SE-like structures were observed in the expression level of WUS , STM , WOX5 and SCR ( S3 Fig ) . In contrast , the expression of the root stem cell niche patterning genes PLT1 and PLT2 [36] ( Fig 3A ) , but not of PLT3 and PLT4/BBM [37] ( S3 Fig ) , distinguished wild-type ZEs , wild-type SEs and clf swn SEs from the SE-like structures . Although both PLT1 and PLT2 are PRC2 targets [38–40] , transcriptional activation in clf swn was limited to SEs suggesting that the 2 , 4-D treatment is required for the induction of PLT1 and PLT2 expression in the mutant background . We next tested whether SEs or SE-like structures have functional apical meristems . First , the respective SAM- and RAM-specific reporter lines CLV3pro::GUS and WOX5pro::NLS-GUS [34 , 41–43] were introgressed into clf swn . CLV3pro::GUS was expressed in a single focus in the apical part of wild-type ZEs as expected ( Fig 3B ) . Wild-type SEs had one or more GUS-expressing foci , indicating multiple SAM regions . In general , clf swn SEs had a single CLV3pro::GUS-expressing region . In contrast , SE-like structures had multiple regions of ectopic GUS expression ( Fig 3B ) , suggesting the presence of multiple ectopic SAMs in untreated clf swn . In the wild type , WOX5pro::NLS-GUS was expressed in a single domain marking the RAM in ZEs but its expression was mostly dispersed throughout the basal part of ZE-derived SEs similar to the previously reported PIN4 and DR5::GUS expression in SEs [32] . We could only detect well localized WOX5pro::NLS-GUS expression in the clf swn-derived SEs but not in SE-like structures ( Fig 3B ) , suggesting that an organized RAM region distinguishes SEs from SE-like structures . Next , we tested whether the apical meristems in SEs are functional by germinating SEs or SE-like structures on hormone-free medium for 7 days ( Fig 3C ) . The majority of SEs and some SE-like structures produced more shoot tissue , probably reflecting the number of established ectopic SAMs in the detached explants . Whereas about 73% of wild-type and 92% of clf swn SEs developed roots , only 6% of the SE-like structures did so ( Fig 3C ) , demonstrating that presence of a functional RAM is the rule in clf swn SEs but an exception in SE-like structures . In summary , PRC2 depletion alone results in the formation of neutral lipid-containing SE-like structures that can develop a functional SAM but usually lack a RAM . A RAM almost exclusively develops in response to external auxin treatment and differentiates the developmentally autonomous SEs from the SE-like structures . SE development in Arabidopsis ZE-based embryogenic cultures ( Fig 1A–1F ) , as well as in clf swn shoots ( Fig 2B , 2D , 2E and 2T ) requires sufficient duration of 2 , 4-D-mediated induction phase followed by 2 , 4-D removal which facilitates embryogenesis . We first asked which of these two phases was repressed by PRC2 utilizing a genomic CLF-GR construct ( CLFpro::CLF-GR [44] ) that allowed dexamethasone ( dex ) -inducible translocation of the CLF-GR fusion protein into the nucleus , complementing CLF activity in the clf swn mutant plants ( S4 Fig ) . We observed partial CLF activity even in the absence of dex in all transgenic clf swn CLF-GR lines , which was manifested by a higher amount of H3K27me3 at selected PRC2 target genes , a lower extent of their up-regulation and a less severe developmental phenotype of clf swn CLF-GR than clf swn plants ( S4A–S4E Fig ) . Partial CLF activity in the absence of dex was also reported for other clf swn CLF-GR lines [45] . Consequently , efficient SE formation in clf swn CLF-GR shoot explants required 7 days of 2 , 4-D inductive treatment ( Fig 4 ) , similarly to what was required for SE induction from wild-type ZEs ( Fig 1B ) . In the absence of dex , SE efficiency in independent transgenic lines varied between 35% and 50% ( Fig 4A–1 , S4F Fig ) . Addition of dex to the 2 , 4-D-containing medium completely prevented SE formation ( Fig 4A–2 ) , demonstrating an immediate effect of CLF in repressing the embryogenic potential . Importantly , dex-induced translocation of CLF to the nucleus at the time of auxin withdrawal did not significantly affect the SE development ( Fig 4A–3 ) . Together , these results established that reduced activity of CLF-containing PRC2 during the 2 , 4-D inductive treatment is necessary and sufficient for SE formation but that reduced CLF-PRC2 activity after 2 , 4-D withdrawal does not considerably affect the efficiency or morphology of the developing SEs . We next tested whether SEs generated by 2 , 4-D in the transient absence of PRC2 are capable of regenerating entire plants . To this end , we compared plants germinated from clf swn-derived SEs in the presence or absence of dex with plants regenerated from wild-type ZE-derived SEs ( Fig 4B ) . The germination efficiency of the clf swn CLF-GR SEs was comparable to wild-type SEs and was independent of the presence of dex . The germinated SEs from mock-treated clf swn CLF-GR failed to stably establish a vegetative phase and differentiate true rosette leaves , resembling the clf swn SE-derived plants . On the contrary , the clf swn CLF-GR SEs germinated and grown in the presence of dex produced plants with differentiated rosette leaves that phenotypically resembled the wild-type SE-derived plants . Altogether , these results show that transient absence of PRC2 activity during initial phases of 2 , 4-D induction is sufficient for producing SEs that are capable of regenerating fully differentiated plants if PRC2 activity is restored during germination and plant growth . Wounding can contribute to cell fate reprogramming in plants [46] . Generating explants used for the 2 , 4-D-mediated reprogramming involves seedling dissection , and a wound response could potentially serve as an additional factor contributing to SE development in our experimental system . We therefore addressed the relative contribution of the individual treatments ( Fig 5 , S5 Fig ) and found that applying 2 , 4-D for 60-hours to dissected ( wounded ) or intact clf swn seedlings induced different morphological outcomes ( Fig 5A ) . Only the combination of wounding and 2 , 4-D triggered efficient SE formation in 87% of explants . 2 , 4-D treatment without wounding induced SE formation in 19 . 6% , wounding only in 2 . 8% and absence of treatment in 0 . 7% of shoot explants ( Fig 5A ) . Notably , extended ( 7-day ) treatment with 2 , 4-D alone resulted in 52% SE efficiency compared to the 19 . 6% after 60 hours ( S5A Fig ) , suggesting that wounding combined with 2 , 4-D treatment enhances the induction rate but that 2 , 4-D is the main trigger . As wounding triggers a rapid increase of active forms of jasmonic acid [47] , we further tested whether methyl jasmonate ( MeJA ) can substitute wounding in SE induction ( S5B Fig ) . Concentrations of 5–20 μM MeJA induced SE formation with half the efficiency of wounding ( approximately 40% compared to 81% ) and using higher concentrations resulted in decrease of SE formation , indicating that jasmonate signaling contributes to the 2 , 4-D-mediated SE induction but does not fully explain or substitute the wounding effect . We next asked what gene expression changes are associated with the induction of SE development . We performed RNA-sequencing using dissected shoot apexes from wild-type and clf swn plants after single or combined wounding and 2 , 4-D treatments , which represented tissue enriched or depleted for embryogenic cells ( S5C Fig , Fig 5A ) . Principle component analysis ( Fig 5B ) identified two main principal components ( PCs ) which together explained 79% of the variance , separating samples on the basis of the applied treatment and the genetic background . The first PC , which explains 62% of the variance , represents mainly the effect of 2 , 4-D-treatments on wild type . The second PC , which explains 17% of the variance , represents the effect of the combined wounding and 2 , 4-D-treatment on clf swn and the background difference between wild type and clf swn . Thus , PC2 reflects embryogenic potential of the probed tissue . The wounded and 2 , 4-treated clf swn samples were separated from all other samples including the clf swn single treatments , suggesting a unique effect of the combined treatment in clf swn ( Fig 5B ) . The embryogenic potential therefore seemed to a large extent determined by different effects of the combined 2 , 4-D and wounding treatments in the wild-type and in the clf swn genetic backgrounds . We searched for gene expression signatures associated with the embryogenic competence . 2664 genes were down-regulated and 2890 genes up-regulated in wounding- and 2 , 4-D-treated clf swn shoot explants that were competent to develop SEs compared to identically treated wild-type plants that never developed SEs ( Fig 5C , S1 Table ) . The upregulated genes were enriched for gene ontology ( GO ) categories related to carpel , ovule and embryo development , lipid storage , photosynthesis and redox processes ( S5D Fig ) and included 318 transcription factor ( TF ) genes ( Fig 5C , S1 Table ) . To identify similarities between TFs upregulated in the wounding- and 2 , 4-D-treated clf swn tissue and organ-specific TF expression , we performed biclustering analyses against anatomy categories in the Genevestigator data warehouse [48] . This established that many of the upregulated TFs are expressed in the shoot apex ( 70 TFs– 22% ) , embryo ( 69 TFs– 22% ) , ovule ( 61 TFs– 19% ) , suspensor ( 38 TFs– 12% ) and endosperm ( 43 TFs– 14% ) of wild-type plants ( S6A Fig ) . Known shoot and embryo-specific TFs such as WUS , CUP-SHAPED COTYLEDON ( CUC1 , CUC2 and CUC3 ) , LEC1 , LEC2 , FUSCA 3 ( FUS3 ) or ABSCISIC ACID INSENSITIVE ( ABI3 , ABI4 , ABI5 ) were among the upregulated genes . Among the genes down-regulated in treated clf swn compared to wild type ( Fig 5C , S1 Table ) , overrepresented GO categories included defense response to biotic and abiotic stresses or response to hormones , including abscisic and salicylic acid or auxin ( S5D Fig , S1 Table ) , suggesting lower amplitude of response to the treatment in the mutant . We next addressed the contribution of the combined wounding and 2 , 4-D treatment in clf swn ( Fig 5C , S2 Table ) . The treatment in clf swn resulted in the down-regulation of 240 genes related to metabolic processes and oligonucleotide transport ( S5E Fig , S2 Table ) and the up-regulation of 1451 genes enriched for GO categories related to cell wall organization , auxin and oxidative stress response , gravitropism and root development–S5E Fig , S2 Table ) . The upregulated genes included 140 TF genes expressed in wild-type root ( 34 TFs– 24% ) , root stele ( 33 TFs– 24% ) and phloem ( 31 TFs– 22% ) ( S6B Fig ) . Among them were known developmental regulators , such as MONOPTEROS ( MP ) , TARGET OF MONOPTEROS 6 ( TMO6 ) [49] , PLT2 , WOX5 , or root-specific LATERAL ORGAN BOUNDARY DOMAIN genes ( LBD16 , 18 and 29 ) [50] . Finally , we identified 35 and 139 genes respectively down- and up-regulated specifically in the wounding- and 2 , 4-D-treated clf swn samples . The GO category response to oxidative stress was enriched among the up-regulated genes ( Fig 5C , S5F Fig , S3 Table ) . The upregulated genes included 16 TF genes ( Table 1 ) , for example the shoot apical meristem and embryo patterning genes DORNRöSCHEN / ENHANCER OF SHOOT REGENERATION 1 ( DRN/ESR1 ) and DRN-LIKE ( DRNL/ESR2 ) [51 , 52] . Together , we identified a non-redundant set of 485 TF genes whose expression marks the clf swn shoot apex sample upon induction of SEs ( Fig 5D ) . These TFs specify shoot- , root- and embryo-identity and development . 16 of these TFs are up-regulated specifically in the samples that can efficiently develop SEs and may serve as conservative markers of embryogenic potential ( Table 1 , S4 Table ) . The main biological processes activated in response to the SE-inductive treatment were mainly related to oxidative stress response and cell wall remodeling . TF genes upregulated in clf swn in response to the inductive treatment are significantly enriched for PRC2-target ( H3K27me3-marked ) genes compared to the genome-wide average [39] ( Fig 5C , S5 Table ) . In particular , 110 of the 140 TFs induced by the treatment in the clf swn background are PRC2 targets . This finding is consistent with the notion that loss of PRC2 by itself does not cause activation of these target genes unless additional signals such as hormones are present . 2 , 4-D-treatment efficiently reprogrammed PRC2-depleted shoots to form SEs . In contrast , SE formation in root explants was rare and stochastic ( Fig 2A and 2P ) . We therefore hypothesized that PRC2-depleted shoots constitutively express genes needed for SE formation that remain inactive in 2 , 4-D treated PRC2-depleted roots and asked what genes are constitutively upregulated in the untreated clf swn shoot apexes compared to wild type . 1794 genes were up-regulated , including 214 TF genes ( Fig 5C , S6 Table ) . GO categories enriched among the up-regulated genes were related to lipid metabolism , carpel and ovule development , redox processes and abscisic acid ( ABA ) response ( S6 Table ) . Biclustering of the 400 most highly up-regulated genes against perturbation samples in Genevestigator revealed similarity to ABA-treated plants ( S7 Fig ) , indicating that constitutive ABA responses mark the clf swn shoot tissue regardless of an embryogenesis-inducing treatment . 160 of the 214 TF genes were common to the set of TFs up-regulated in the wounding and 2 , 4-D-treated clf swn sample , including the ABA-responsive TFs ABI3/4/5 , indicating that active ABA response persists after the induction treatment . Because ABA can promote embryogenic competence in different species and contribute to the establishment of a polarized auxin response [53–55] , we hypothesized that absence of ABA signaling may limit the establishment of embryogenic potential in clf swn root tissue . To assess the level of constitutive ABA signaling in the clf swn root , we measured the transcript levels of ABA-responsive TFs up-regulated in the clf swn shoot—ABI3 , ABI4 and its downstream target PLT5 [56] ( Fig 6A ) . The transcript level was low in the untreated clf swn root , supporting absence of constitutive ABA signaling in this tissue , but it was increased by external ABA treatment in the clf swn ( but not the wild-type ) root ( Fig 6A ) . Induction of ABI3pro::GUS marker expression by combined 2 , 4-D and ABA treatment in clf swn but not in wild-type root confirmed the RT-PCR results , further localizing the transcriptional activation to the root stele ( S8 Fig ) . Within the ABI3pro::GUS positive region in clf swn roots , we observed local DR5::GUS expression maxima 3 days after the hormone removal , suggesting that the combined treatment contributes to the formation of an auxin response gradient within the tissue ( S9 Fig ) . Next , we tested whether activation of the same TFs that marked the embryogenic potential in the clf swn shoot ( Fig 5D ) is conditioned by external ABA in the root . 2 , 4-D was sufficient to activate the root identity TFs PLT1 , PLT2 and WOX5 as well as PLT4/BBM ( Fig 6B and 6C ) in both clf swn and wild-type shoot and root , but the external ABA treatment promoted the transcription of the shoot and/or embryonic TFs WUS , CUC1 , CUC2 and DRNL in the clf swn but not in the wild-type root ( Fig 6C and 6D ) . These results showed that a constitutive ABA response is present in PRC2-depleted shoot but absent in PRC2-depleted roots , and that external ABA treatment can specifically and locally induce the expression of ABA-responsive and shoot/embryo-identity TFs in PRC2-depleted roots and contribute to the establishment of localized auxin response maxima . The results further confirmed the expression patterns of TF genes identified as markers of embryogenic competence by the genome-wide approach ( Fig 5D ) . Finally , to determine whether the ABA-induced transcriptional changes in the root correlate with embryogenic potential , we tested the effect of a 60 h combined treatment of ABA , 2 , 4-D and wounding in wild-type and clf swn root explants ( Fig 7A ) . This protocol efficiently induced the formation of ectopic embryo-like structures and also induced the development of bipolar SEs in clf swn ( Fig 7B ) but not in wild-type root explants ( Fig 7A ) . The efficiency of bipolar SE formation from roots was relatively low in comparison to the shoot explants ( Fig 7B ) . When the SE-like structures were however used as source explants for 2 , 4-D mediated SE induction , the shoot-like , but not the root-like , structures developed SEs with an efficiency similar to the clf swn shoot apex explants ( Fig 7C ) . Together , these results established that PRC2 absence by itself is not sufficient for ectopic activation of root , shoot and embryo identity genes that mark tissue with embryogenic potential and that externally added 2 , 4-D and ABA each contribute to ectopic activation of a subset of TFs involved in organ specification . The combined 2 , 4-D and ABA treatment in root explants is efficient in inducing the development of ectopic SE-like structures but less efficient in inducing complete bipolar SEs . The shoot-like subset of the SE-like structures can nevertheless respond to external 2 , 4-D by efficient complete SE-formation .
Reprogramming of Arabidopsis vegetative tissue to somatic embryogenesis is problematic due to molecular barriers that remain largely unknown . Here , we identify the activity of the histone-methyltransferase complex PRC2 during Arabidopsis vegetative development as a barrier to hormone-induced reprogramming to somatic embryogenesis . PRC2 is known to be required for a stable embryo-to-seedling developmental transition [26 , 28 , 29] . More recently , PRC2 was shown to ensure the maintenance of differentiated state of Arabidopsis root hair cells [30] . Vegetative PRC2-depleted tissue can adopt embryonic identity , which has been mainly demonstrated by the accumulation of neutral lipids and activation of embryo maturation genes [26 , 28–30] . Occasional SE development was also reported [26 , 29 , 30] but the frequency of SE development or factors that influence it have never been followed . It was therefore difficult to conclude on the extent of cell dedifferentiation or trans-differentiation in PRC2 depleted tissue . Here we show that spontaneous development of complete SEs in clf swn vegetative tissue is rare unless additional hormonal treatments are applied to explants . We support our conclusion by identifying features of complete SEs [19 , 32] that differentiate structures developing in response to the external treatment from SE-like shoot-derived structures that develop spontaneously in clf swn . Apart from the accumulation of neutral lipids common to both types of structures , complete SEs are distinguished from the SE-like structures ( i ) by the general absence of vascular connection to the parental explant , and ( ii ) by the development of correctly localized and functional root and shoot apical meristems ( Figs 2 and 3 ) . We therefore suggest that spontaneous cell dedifferentiation that would allow somatic embryogenesis in PRC2-depleted cells is rare but can be efficiently induced by additional treatments or environmental conditions . Treatments with the synthetic auxin 2 , 4-D , wounding or other abiotic stress and ABA are known to contribute to somatic embryogenesis ( reviewed in [1] ) . Here , we identify PRC2-repression as a factor that modulates the transcriptional response to these treatments in a way that limits their efficiency for SE induction in wild-type vegetative tissue . By showing that external addition of ABA is necessary for the reprogramming in the PRC2-depleted root but not shoot ( Fig 7 ) , we demonstrate that the requirement for external treatments varies in different cell types depending on the existing level of hormonal signaling and/or cell-type specific transcriptional background . Even using the identified treatments , only some regions of shoot and root explants responded by formation of SEs–the responsive tissue was mainly restricted to the shoot apical meristem region and to the root stele , which both contain stem cell niches [57] . It is therefore likely that under the tested conditions , the capacity for developmental reprogramming is limited to specific PRC2-depleted cell types . Several observations support this notion . First , the ABI3pro::GUS reporter is activated by combined 2 , 4-D and ABA treatment locally in the root stele ( S8 Fig ) . Second , the low efficiency of complete SE development in clf swn root explants contrasts with efficient SE induction using shoot-like root-derived structures as explants ( Fig 7C ) . It is possible that establishing ectopic shoot apical cell-identity promotes the capacity for 2 , 4-D-mediated reprogramming . Third , although almost all clf swn shoot apical explants developed SEs following 2 , 4-D treatment ( Fig 2B ) , the number of SEs per explant usually varied between 1–10 , arguing against a homogeneous developmental response in all cells within the explant . Whether the responsive cell types localize within the stem cell niches remains to be determined . It has nevertheless been previously shown that fully differentiated PRC2-depleted root hair cells can undergo spontaneous dedifferentiation [30] . It is therefore possible that other cells than stem cells may undergo dedifferentiation required for SE development upon external treatments in PRC2-depleted tissue . Although we have not observed hormone-induced SE-formation from root hairs , it is possible that additional triggers are required for inducing complete SE development in these cells . We found here that the establishment of embryogenic competence in the PRC2-depleted tissue coincides with simultaneous ectopic up-regulation of TF genes that are commonly expressed in the wild-type root , shoot apex and embryonic tissues . Among them are TF genes whose overexpression alone can trigger ectopic developmental reprogramming of wild-type cells in the absence of external hormone treatments . These include for example LEC1 , LEC2 , PLT4/BBM , AGL15 or WUS , which induce the vegetative-to-embryonic transition or retention of embryogenic potential [8 , 9 , 11 , 13 , 14] , DRN/ESR1 , DRNL/ESR2 or WUS , which induce ectopic shoot development [58–60] , PLT2 , which induces ectopic root development [37] , or LBD16 , LBD18 and LBD29 , which induce callus formation [61] . Samples containing multiple cell types were studied here and it needs to be established whether the identified TF genes are co-expressed in one specific cell-type and how their expression contributes to the establishment of embryogenic competence . Among the identified TF genes are 16 genes that were expressed specifically in the wounding- and 2 , 4-D-treated clf swn samples ( Table 1 ) and provide a conservative set of potential markers of embryogenic competence to be studied . The combined 2 , 4-D and wounding treatment of wild type or clf swn tissue triggered very different transcriptional and developmental effects . While the induction of root identity genes is a common response to 2 , 4-D treatment both in wild-type [59 , 62] and in clf swn tissue [59 , 62] , the activation of shoot and embryo-identity TF genes is limited to PRC2-depleted tissue . This suggests that PRC2 repression prevents the transcriptional response of a subset of genes to the applied treatment , which may be crucial for establishing embryogenic competence . In support of this , PRC2-target genes are significantly enriched among the TF genes upregulated in the samples competent to produce SEs . Similarly , 2 , 4-D-mediated induction of somatic embryogenesis in cultured ZEs coincides with extensive upregulation of TF genes [63] , among which PRC2-targets are also significantly enriched ( S4 Table ) . Acquisition of totipotency during the development of the megaspore mother cell ( MMC ) in Arabidopsis is marked by depletion of PRC2-deposited H3K27me3 [64] and extensive reshaping of the H3K27me3 landscape follows reprogramming of Arabidopsis leaf cells to callus [65] , supporting the need for global transcriptional reprogramming of the PRC2-target genes . Changes in chromatin structure have recently been associated with the initiation and progression of somatic embryogenesis in different plant species . The histone deacetylase inhibitor trichostatin A ( TSA ) increases the efficiency of somatic embryo induction during microspore embryogenesis in Brassica napus [66] and retention of embryogenic potential in Norway spruce [67] . Decrease of H3K9me2 and increase in histone acetylation are associated with the progression of microspore embryogenesis in B . napus [68] and reduction of DNA methylation by 5-azacytidine promotes the induction of microspore embryogenesis in B . napus and Hordeum vulgare but hinders embryo differentiation [69] . It is therefore possible that temporal reduction of repressive chromatin structure is one of the requirements of efficient reprogramming to totipotency in plant cells . It will be interesting to test the effect of PRC2 depletion or its combination with other chromatin modifiers in other experimental model systems and plant species . Despite the fact that the expression pattern of most of the 485 TF genes expressed in the wounding- and 2 , 4-D-treated clf swn tissue correlate with the general role of PRC2 in gene repression , some of these TF genes displayed lower responsiveness to the inductive treatment in clf swn than in wild type ( cluster A2 , Fig 5D ) . In addition , combined treatment of 2 , 4-D and wounding in wild-type shoot explants led to the down- and up-regulation of a similar number of genes as in clf swn ( S5 Table , S8 Table ) , but resulted in different transcriptional ( Fig 5B ) and developmental outcomes ( Fig 2C ) . Among the TF genes upregulated in the treated wild type , PRC2-target genes were also enriched ( S5 Table ) . These observations suggest a complex reprogramming of PRC2-target gene expression in response to the combined treatment in general , arguing against a simple prevention of the treatment-responsive gene activation in clf swn . The GO categories enriched among the genes down-regulated in treated clf swn compared to treated wild type involved mainly response to stress and hormone stimuli . In support of this , we noticed that while the 2 , 4-D treatment in wild type resulted in massive pericycle cell multiplication as expected , it had a weaker effect on cell multiplication in the clf swn root ( S8 Fig , S9 Fig ) . The dosage and type of stress is crucial for inducing somatic embryos in Arabidopsis vegetative tissue [70] , which could also explain reduced efficiency of SE formation in clf swn treated with high concentration of MeJA ( S5 Fig ) . Both a lower amplitude and varied response to the undergone stress and hormone treatment may therefore be an important contributing factor to the SE induction in PRC2-depleted tissue . In summary , we demonstrate here that the histone methyltransferase activity of PRC2 constitutes a barrier to hormone-mediated cell dedifferentiation and somatic embryogenesis in Arabidopsis vegetative tissue . We show that absence of PRC2 activity is required but by itself is not sufficient for full dedifferentiation of somatic cells , and that additional triggers are needed to induce reprogramming leading to SE development . Transient absence of PRC2 activity is sufficient and residual PRC2 activity does not inhibit the reprogramming . This opens possibilities for targeted modulation of PRC2 activity to enhance the efficiency of somatic embryogenesis approaches in recalcitrant species .
The strong PRC2 mutants alleles clf-29 ( SALK_021003 ) [71] and swn-3 ( SALK_050195 ) [29] were used . Wild-type control plants were Col-0 . WOX5pro::NLS-GUS ( 42 ) and CLV3pro::GUS ( NASC N9610 ) [41] reporters were introduced into clf-29 swn-3 by crossing and F3 seedlings were used . The ABI3pro::GUS and DR5::GUS marker lines were generated by crossing ABI3pro::GUS [72] and DR5::GUS [73] with clf-29/+ swn3/- and F2 seedlings were used . clf-29/+ swn3/- CLF-GR plants were created by transformation of a pCLF::CLF-GR construct [44] into clf-29/+ swn/- plants by floral dip . 18 independent transgenic lines were obtained , from which two lines with least CLF activity and most pronounced clf swn-like phenotypes in the absence of dex were selected for further experiments [72 , 73] . As a standard , seeds were surface sterilized using 70% and 90% ethanol , placed on ½ strength MS medium ( Duchefa , M0222 ) containing 1% ( w/v ) sucrose and 0 . 8% ( w/v ) agar ( standard MS ) and were stratified for 48 h . Plants were grown under long-day growth conditions ( 16 hrs light 110 μmoles m-2 s-1 , 22°C and 8 hrs dark , 20°C ) . The following plant growth regulators were used at concentrations specified in each experiment: 2 , 4-Dichlorophenoxyacetic acid ( 2 , 4-D , Sigma-Aldrich #D7299 ) , abscisic acid ( ABA , Sigma-Aldrich #A1049 ) , 1-naphthylacetic acid ( NAA , Sigma-Aldrich #N1641 ) , indole-3-butyric acid ( IBA , Sigma-Aldrich #I5386 ) , dexamethasone ( Sigma-Aldrich #D1756 ) was used at 10 μM . All chemicals were dissolved in DMSO which was used in mock treatments . SE from immature ZEs or from seedlings were induced as described previously [16] with modifications: For ZE-derived SEs , siliques of 6–7 week plants were surface-sterilized for 20 minutes using 2% sodium hypochlorite solution containing 3 drops of Tween 20 per 100 ml and ZEs were excised from seeds as previously described [16] . In contrast with the published protocol , all steps of primary SE induction , development , germination and plant growth were carried out using standard MS medium and long-day growth conditions as specified above . For SE induction , the standard MS medium was supplemented with 5 μM 2 , 4-D as described [16] after which the ZEs were transferred to the standard ( hormone-free ) MS medium for 7 days after which the percentage of SE-forming ZEs was determined . For SE induction from seedlings , seeds were surface sterilized using ethanol , placed on standard MS medium , stratified for 48 h and grown under long-day conditions as specified above . SE induction was carried out under the same conditions as described for ZEs using 14 DAG clf swn and 10 DAG clf swn CLF-GR unless specified otherwise . Seedlings were transferred to induction 2 , 4-D-containing plate and wounding was realized by separating the shoot from the root using a scalpel blade to cut the seedling in the top half of the hypocotyl . Whole seedlings represented the non-wounded controls . Matching mock ( DMSO ) -treated controls were included in all experiments . After the specified induction times , the explants were kept on hormone-free plates for another 7 days after which percentage of SE-forming explants was calculated . Each experiment was repeated 2 to 4 times using at least 25–30 explants as indicated in the figure legends . Embryonic lipids were visualized by 20-minute staining with Sudan Red 7B ( Sigma-Aldrich #46290 ) as described [28] . GUS staining was performed as described [74] . For more details refer to S1 Text ( Supporting Material and Methods ) . RNA from seeds was extracted using the hot borate buffer method modified after [75] ( for details see S1 Text ) . RNA from other samples was extracted using TRIzol according to the manufacturer’s instruction . RNA was treated with DNaseI ( Fermentas ) and 1 μg of RNA was reverse-transcribed using the RevertAid First Strand cDNA Synthesis Kit ( Thermo Scientific ) with oligo ( dT ) primers . Quantitative PCR was performed using the MyiQ Single Color Real Time PCR detection system ( BIO-RAD ) with gene-specific primers ( S1 Text ) and 5X HOT FIREPol Eva Green qPCR Mix Plus ( ROX ) ( Solis Biodyne ) . PP2A ( AT1G13320 ) was used as reference gene . All experiments were performed at least twice using biological replicates and technical triplicates . Wild-type and clf swn shoot apex explants were exposed for 55 hours to different combinations of treatments as shown in Fig 5A , S6A Fig . The experiment was performed in independent biological triplicates . RNA was isolated using a MagJET Plant RNA Purification Kit ( Thermo Scientific ) . Sequencing libraries were prepared using the TruSeq RNA Library Preparation Kit v2 ( Illumina ) and 12 samples were pooled for sequencing in one lane of the Illumina HiSeq2000 platform ( S7 Table ) . Differential gene expression analyses were performed using the R package DESeq v1 . 18 . 0 [76] , applying threshold values of p = 0 . 05 after multiple testing correction according to [77] and a minimal log2 fold change = 0 . 6 between any pair of replicates for calling of differentially expressed genes . PCA analysis was performed using the R statistical software environment ( www . r-project . org ) . For additional details on sequencing and data analysis , see S1 Text . RNA-seq data has been submitted to GEO data repository under the accession no . GSE81166 . Unless stated otherwise , all experimental data was analyzed using unpaired two-tailed Student’s t-test . Letters above bars in bar charts represent statistical significance level where at least three biological replicates were analyzed ( same letter–same significance level ) . | Somatic embryogenesis provides the strongest support for plant cell totipotency but reprogramming of non-reproductive tissue is problematic or even impossible in many plant species . Here we show that the activity of Polycomb Repressive Complex 2 ( PRC2 ) constitutes a major barrier to hormone-mediated establishment of embryogenic competence in plant vegetative tissue . We identify a conservative set of transcription factors whose expression coincides with the establishment of embryogenic competence in vegetative tissue , among which are key developmental regulators of root , shoot and embryo development . We show that lowering the PRC2-imposed barrier combined with activating hormone treatments establishes embryogenic competence in different tissue types , which opens possibilities for novel strategies to plant cell identity reprogramming . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"plant",
"growth",
"and",
"development",
"gene",
"regulation",
"regulatory",
"proteins",
"brassica",
"dna-binding",
"proteins",
"hormones",
"developmental",
"biology",
"plant",
"science",
"model",
"organisms",
"plant",
"hormones",
"experimental",
"organism",
"systems",
"transcription",
"factors",
"seedlings",
"embryos",
"plants",
"arabidopsis",
"thaliana",
"research",
"and",
"analysis",
"methods",
"embryology",
"plant",
"embryogenesis",
"plant",
"development",
"proteins",
"gene",
"expression",
"biochemistry",
"plant",
"biochemistry",
"plant",
"and",
"algal",
"models",
"embryogenesis",
"genetics",
"biology",
"and",
"life",
"sciences",
"auxins",
"organisms"
]
| 2017 | PRC2 Represses Hormone-Induced Somatic Embryogenesis in Vegetative Tissue of Arabidopsis thaliana |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.