A Functional Genomic Screen for Evolutionarily Conserved Genes Required for Lifespan and Immunity in Germline-Deficient C. elegans

The reproductive system regulates lifespan in insects, nematodes and vertebrates. In Caenorhabditis elegans removal of germline increases lifespan by 60% which is dependent upon insulin signaling, nuclear hormone signaling, autophagy and fat metabolism and their microRNA-regulators. Germline-deficient C. elegans are also more resistant to various bacterial pathogens but the underlying molecular mechanisms are largely unknown. Firstly, we demonstrate that previously identified genes that regulate the extended lifespan of germline-deficient C. elegans (daf-2, daf-16, daf-12, tcer-1, mir-7.1 and nhr-80) are also essential for resistance to the pathogenic bacterium Xenorhabdus nematophila. We then use a novel unbiased approach combining laser cell ablation, whole genome microarrays, RNAi screening and exposure to X. nematophila to generate a comprehensive genome-wide catalog of genes potentially required for increased lifespan and innate immunity in germline-deficient C. elegans. We find 3,440 genes to be upregulated in C. elegans germline-deficient animals in a gonad dependent manner, which are significantly enriched for genes involved in insulin signaling, fatty acid desaturation, translation elongation and proteasome complex function. Using RNAi against a subset of 150 candidate genes selected from the microarray results, we show that the upregulated genes such as transcription factor DAF-16/FOXO, the PTEN homolog lipid phosphatase DAF-18 and several components of the proteasome complex (rpn-6.1, rpn-7, rpn-9, rpn-10, rpt-6, pbs-3 and pbs-6) are essential for both lifespan and immunity of germline deficient animals. We also identify a novel role for genes including par-5 and T12G3.6 in both lifespan-extension and increased survival on X. nematophila. From an evolutionary perspective, most of the genes differentially expressed in germline deficient C. elegans also show a conserved expression pattern in germline deficient Pristionchus pacificus, a nematode species that diverged from C. elegans 250-400 MYA.


Introduction
The reproductive system in many animals including vertebrates, insects and nematodes regulates lifespan [1][2][3][4][5][6][7][8][9][10][11][12][13] and the interactions between the reproductive system and lifespan have been extensively studied in Caenorhabditis elegans. When the C. elegans germline is removed by either genetic means or cell ablation, its lifespan can be extended by 60% [1]. The enhanced lifespan is not a simple life-history trade-off between reproduction and somatic tissue maintenance because C. elegans rendered sterile by complete gonad ablation or other manipulations live only as long as C. elegans with intact germline [1].
The role of gonadal tissue in regulating pro-longevity or pathogen resistance genes has not been studied at a whole-genome level. Therefore, to understand the contribution of gonadal tissue to lifespan and immunity regulating genes, and the role of the innate immune system towards lifespan extension in germlinedeficient C. elegans at a genome-wide scale, we combined laser cell ablation to kill the germline and gonad precursor cells, with whole genome microarrays, followed by an RNAi knockdown screen to identify mechanistic processes involved in both longevity and immunity. We also analyzed the evolutionary conservation of germline-regulated genes required for enhanced lifespan and immunity between C. elegans and the related model nematode P. pacificus [37].

Results
Effect of germline and gonad ablation on survival of C. elegans exposed to X. nematophila The formation of the C. elegans gonad begins with four precursor cells (Z1,Z2,Z3 and Z4), with Z2 and Z3 giving rise to the germline and Z1 and Z4 making the somatic gonad [38], all of which can be removed using laser microsurgery at the L1 stage ( Figure 1). When germline cells Z(2,3) are ablated in C. elegans and P. pacificus lifespan can be extended by 60% [1,2]. We wanted to test if the pathogen resistance of C. elegans would also be enhanced, as reported in similar studies [2]. We used the bacterium Xenorhabdus nematophila as a pathogen, as it is known to be lethal to insects [39] and is also pathogenic to the nematodes C. elegans and P. pacificus [2,[40][41][42]. Further, transcriptomic profiles of nematodes infected with X. nematophila were shown to be highly similar to gene expression profiles induced in C. elegans in response to other bacterial pathogens such as S. marcescens, P. aeruginosa and Staphylococcus aureus [42][43][44]. When we fed C. elegans with X. nematophila, we observed rapid lethality, with all animals dying in one to two days ( Figure 2A, see Table S1 for summary statistics), hence we consider X. nematophila to be a bona fide nematode pathogen.
When we ablated the germline cells Z(2,3) and fed young-adult C. elegans animals X. nematophila, we found a significant increase in survival from a mean survival time of 1.8 days in control, unablated animals, to 8.5 days in Z(2,3)-ablated C. elegans ( Figure 2A). However, the ablation of Z(1,4) cells, which also leads to absence of Z(2,3) cells, reduced the mean survival time on X. nematophila to 3.7 days (Figure 2A). These results indicate that similar to its effects on longevity [1], the removal of germline contributes to resistance towards bacterial pathogens, potentially via a signal that requires the presence of a functional gonad, a finding that is in accordance with previous observations [25][26][27][28].
Some of the key genes required for the increase in C. elegans lifespan upon the removal of germline precursor cells include daf-16, daf-12, tcer-1, mir-71 and nhr-80 [1,17,19,20]. We tested whether these genes would also be required for the increase in the survival of C. elegans germline deficient animals on pathogenic X. nematophilum, and could show that germline ablations in animals carrying mutations in daf-16 (mu26, mu27 and mu86), daf-12(m20), tcer-1(tm1452), mir-71(n4115) and nhr-80(tm1011) resulted in a significant decrease in survival when fed X. nematophila as compared to germline-ablated wild-type C. elegans ( Figure 2B). It was previously reported that mutations in insulinlike receptor daf-2(e1370) can further increase the lifespan and stress tolerance of germline-ablated animals in daf-16 dependent manner [1,45]. We observed a similar synergistic effect of daf-2(e1370) mutation and germline ablation on increased pathogen resistance, resulting in further increase in their survival upon X. nematophila exposure relative to germline-ablated wild-type C. elegans exposed to X. nematophila ( Figure 2C). These results indicate that genes initially identified as being essential in regulating germline-deficient C. elegans lifespan can be essential for combating bacterial pathogens. Hence we can use survival towards pathogens as a rapid tool to identify new genes potentially involved in lifespan regulation.

Identifying gonad regulated genes in long-lived germline-ablated animals
In order to carry out an unbiased investigation of genes involved in the regulation of lifespan and pathogen resistance, we performed a genome-wide expression analysis of germline-ablated, gonad-ablated and intact C. elegans ( Figure 1). Since the presence of somatic gonad is essential for the observed increase in lifespan and survival on pathogens (Figures 2A,B,C) [1,2,25,26], we aimed to compare long-lived germline-deficient animals (Z(2,3)-ablated) against the normal-lived germline and gonad-deficient animals (Z(1,4)-ablated), which differ only in the presence of gonadal tissue. We expect this comparison to be more informative and specific for lifespan genes than the comparison between germline-ablated and un-ablated animals because the latter expression profile would be comprised of genes involved not only in lifespan regulation but also include those involved in germline development and maintenance [2,24,46]. We employed a common reference experimental design for microarray comparisons ( Figure 1) [2,42,47]. In the first set of experiments we compared the Z(2,3)-ablated animals against age-matched intact animals as reference, while the second set of experiments compared the Z(1,4)-ablated animals also against intact animals. Since the intact animals used as the reference were common to both experiments, we could compute a contrast between these two experiments to get a list of gonad-regulated genes in long-lived germline-deficient animals ( Figure 1).
The first comparison of Z(2,3)-ablated versus intact animals indicated differential expression of 2,794 genes of which 1,587 i.e. about 60% genes were down-regulated (Table S2). This finding is consistent with high transcriptional activity in the germline of intact animals. Further, this set of genes was highly enriched for genes implicated in germline development and function [46,48] ( Table S3). The comparison of Z(1,4)-ablated versus intact animals also showed a trend of predominantly down-regulated genes (more than 95% are down-regulated, Table S4), because gonad ablation takes away the genes expressed not only in the germline but in the gonad as well.
Here we found about 3,440 genes up-regulated while only 150 genes were down-regulated (Table S5). Consistent with the potential role of these genes in lifespan extension, we observed the set of up-regulated genes to be highly enriched for genes with the known RNAi phenotype ''shortened lifespan'' (p-Value of enrichment = 7.8E-16, Figure 3A, see Methods). Gene ontology enrichment analysis (Table S6) further identified broad biological processes regulating ''determination of adult lifespan'', and ''germ cell development'', as well as specific functions such as ''translation initiation factor activity''. We also observe enrichment for molecular processes and cellular components involved in ''cytochrome-c oxidase activity'', at the ''mitochondrial inner membrane'' and ''mitochondrial envelope'', as well as components of ''proteasomal core complex'' and ''structural constituent of ribosomes''. Searches for significantly enriched gene families revealed various components of proteasomal complex (e.g. pbs-, rpt-, rpngene families, Table S7), chaperon activity (cct-gene family, Table S7), fatty-acid desaturases (fat-gene family, Table  S7) and eukaryotic translation elongation factors (eef-gene family, Table S7). Together these observations suggest potential roles for proteasome function, translation initiation and elongation factors and lipid metabolism in longevity assurance, consistent with previous reports [17,20,22,49].
RNAi based screen of candidate genes identifies additional insulin pathway components regulating lifespan and innate immunity Since many of the genes identified in our microarray experiments could be involved simply in germline and gonadal tissue maintenance rather than in increased longevity and pathogen resistance, we further tested the specificity of these genes for their role in innate immune response and longevity via RNAi. We selected a set of top 100 most upregulated genes plus a random set of 50 genes upregulated at smaller fold-change values, for RNAi based assays (schematic of experiment in Figure S1, genes tested and the corresponding phenotypes observed are annotated in Tables S2-S7). In order to avoid the need for laborious cell ablation of the germline in these RNAi screens, we used the glp-1(e2141) temperature-sensitive mutant, which does not produce a germline when grown at the restrictive temperature of 25uC [50], and is reported to be long lived and resistant to bacterial pathogens [25][26][27][28]. We assume here that there are only subtle differences between rendering C. elegans germline-deficient via either cell ablation or glp-1 mutation. Any RNAi candidates that reduced survival of the glp-1(e1241) when fed X. nematophila were also tested for their effects on normal lifespan of C. elegans glp-1(e1241) mutant, and wild type C. elegans (see Fig S2, S3 Table S8-S10).
From our microarray data we observed a significant overlap of our up-regulated gene set with known DAF-16 targets [51] (onethird of known DAF-16 targets upregulated in germline-ablated animals, p-value of overlap = 5.23E-07, Figure 3B). DAF-16 is a key transcriptional regulator of insulin signaling output which is an important regulator of lifespan and innate immunity under various contexts including germline-ablation [1,2,[25][26][27][28]. We also observed transcriptional up-regulation of daf-16 itself (Table S2), and consistent with its reported role in pathogen resistance and longevity assurance [1,2,[25][26][27][28], its knock-down by RNAi reduced the survival on pathogen in our assays ( Figure 4A, see Table S8 for summary statistics). The PTEN homolog lipid phosphatase DAF-18 also acts downstream to the Insulin receptor DAF-2 and the PI3-Kinase AGE-1 to regulate the activity of DAF-16 [52]. Consistent with its lifespan-promoting role [53], we found it to be highly up-regulated in our microarray results. We further found that the RNAi based knock-down of daf-18 resulted in reduced survival of glp-1 mutants on X. nematophila ( Figure 4A, see Table  S8 for summary statistics). Thus, we demonstrate for the first time the requirement of daf-18 for the extended lifespan and pathogen resistance in a germline-deficient background.
Discovery of new genes with role in pathogen resistance and lifespan extension in germline deficient C. elegans In addition to the genes discussed above, we also found a significant reduction in pathogen resistance and lifespan upon RNAi of other genes not previously known to be involved with lifespan or innate immunity. Specifically, we could show that when the 14-3-3 protein par-5 is knocked down via RNAi, both survival on X. nematophila and lifespan are significantly impaired in glp-1(e2141) animals ( Figure 4C, Figure S2, summary statistics in Table S8, S9) but not in un-ablated C. elegans WT animals ( Figure  S3, Table S10). PAR-5 is known to regulate DAF-16 activity and lifespan [24,54], but it has so far never been implicated in affecting survival against pathogens. We also identified an uncharacterized gene T12G3.6 ( Figure 4C, Figure S2, Table S8,S9) which has been predicted to be involved in germline development in C. elegans [55] and is known to be induced in response to the fungal pathogen Drechmaria coniospora [56]. Our results show a gonaddependent up-regulation of this gene in germline-ablated animals, and a requirement for this gene in mediating the increase in lifespan and pathogen survival of germline-ablated C. elegans ( Figure 4C, Figure S2, Table S8,S9).
It is important to note that while the RNAi based knock-down of some of these candidate genes in germline-ablated animals leads to a decrease in survival on X. nematophila as well as lifespan, the lifespan of un-ablated C. elegans remains unaffected upon the RNAi of these genes ( Figure S3, Table S10). This suggests a specific role for these genes in germline-ablation mediated lifespan extension, and serves as a functional validation of our system-wide approach. However, some of the candidate genes that reduced survival on X. nematophila and lifespan in germline-ablated animals, also reduced survival of unablated C. elegans e.g. gna-2 and C14B1.2 (data not shown). Therefore, these genes seem to affect the general viability of the worm and were hence not further investigated. From these results we conclude that some of the genes observed to be upregulated in our microarray profile are specifically required for mediating the effects of germline-ablation on lifespan and immunity, while other genes might be required for general viability and normal development. Similar tests on the rest of the upregulated genes will therefore be needed to identify genes with specific roles in germline-mediated lifespan regulation.

Evolutionary conservation between C. elegans and P. pacificus
Comparison of similar biological processes in different species and model systems is helpful in identifying evolutionary trends and conserved components. To the best of our knowledge, the only study to utilize whole genome microarrays to understand the patterns and processes involved with germline deficient lifespan extension and bacterial resistance has been carried out in P. pacificus [2]. In this study whole genome microarrays comparisons between germline-ablated long-lived P. pacificus cultured either on Escherichia coli OP50 or on the nematode pathogen S. marcescens showed that the expression profiles of long-lived animals was remarkably similar to pathogen resistant animals, implying that in both scenarios similar stress response pathways are activated that contribute to the increased survival [2].
Interestingly, a significant proportion of these germline-regulated genes also show a significant overlap and similar expression patterns in both C. elegans and P. pacificus ( Figure 5). These findings highlight a great extent of evolutionary conservation in this biological process across the two species, which shared a common ancestor 250-400 million years ago [57]. This overlap is even more remarkable given the fact that our previous gene expression studies comparing C. elegans and P. pacificus in other contexts revealed relatively more diverged expression profiles [42,47]. Hence, the significant overlap between germline-regulated genes in the two species suggests a deeper conservation of underlying biological processes such as insulin signaling and proteasome function.

Discussion
The presence of gonadal tissue is essential for the increased lifespan observed in germline deficient C. elegans [1]. Although many genes have been identified as being essential in regulating C. elegans lifespan in germline deficient animals e.g. daf-16 and daf-12 [1,[14][15][16][17][18][19][20][21][22][23][24], the gamut of genes and processes regulated by the gonad has not been studied systematically at a whole genome level in this context before [58]. Here we first show that the gonad tissue is also needed for enhanced pathogen resistance phenotype of germline-ablated C. elegans. Next, using a microarray based approach we have identified a set of candidate genes that are regulated in a gonad dependent manner in germline-ablated animals, and via RNAi based knock-downs demonstrated the requirement for some of these genes in both longer lifespan as well as enhanced pathogen resistance. These data will therefore be useful in further dissection of genetic components of gonad dependent longevity signals. Furthermore, by comparing our data to germline-ablated P. pacificus [2] we demonstrate that many of these genes and processes that increase lifespan and bacterial resistance remain highly conserved during evolution. Specifically, we could show that upon germline-ablation, both P. pacificus and C. elegans upregulate genes involved in proteasomal complexes (pbs-, rpt-, rpngene families), chaperon activity (cct-gene family), fatty-acid desaturases (fat-gene family) and eukaryotic translation elongation factors, and insulin signaling (daf-18, age-1 and daf- 16), which together potentially contribute to their increased lifespan and survival on pathogens [2].

Survival assays and analysis
X. nematophila was grown in LB at 30uC overnight in a shaking incubator. The following day 100 ml were spread evenly onto predried 5 cm NGM plates and incubated overnight at 30uC. Batches of 10 ablated or un-ablated C. elegans were picked onto X. nematophila plates and monitored daily for survival. Worms that failed to respond to a touch of the worm-pick were considered dead. Survival of ablated or un-ablated C. elegans fed on X. nematophila or E. coli OP50 was compared using Kaplan meier curves and the Log Rank test using the OASIS web-tool [61] (Online Application for the Survival Analysis of Lifespan Assays Performed in Aging Research, http://sbi.postech.ac.kr/oasis).

Cell ablation and RNA collection
Wild type or mutant C. elegans L1 stage larvae were picked into 2.8 ml PBS on an agar pad containing 1 mM NaN 3 . Ablations would take place within 1 hour of hatching at 20uC. After ablation nematodes were stored at 20uC and successful ablation was verified 48 hours later. Un-ablated C. elegans were grown in parallel and served as controls. For microarrays 20 young-adult C. elegans (either Z(2,3)-ablated, or Z(1,4)-ablated, or un-ablated) were picked onto 5 separate NGM plates spotted with E. coli OP50 and incubated at 25uC for 4 hours. Nematodes were then picked into 500 ml of Trizol and stored at 280uC until further analysis. The treatments therefore included (i) Z(2,3)-ablated C. elegans fed E. coli OP50 (ii) Z(1,4)-ablated C. elegans fed E. coli OP50, and (iii) un-ablated C. elegans fed E. coli OP50. Developmental time course of ablated and un-ablated C. elegans fed E. coli OP50 was the same. Each treatment consists of approximately 100 animals and was repeated three times.

Microarray experiments and data analysis
C. elegans microarrays from Agilent Technologies were used in a two-color format, and the experiments and data analysis was performed essentially as described before [42,47], with three biological replicates per treatment condition. The first set of hybridizations compared Z(2,3)-ablated worms against un-ablated wild-type C. elegans, while the second set of hybridizations compared gonad Z(1,4)-ablated worms against un-ablated wildtype C. elegans. The differential expression between germlineablated versus gonad-ablated animals was computed via the functions ''makeContrasts'' and ''contrasts.fit'' in the ''limma'' package [62] in R/Bioconductor [63]. The raw and normalized data for all the experiments from this publication have been deposited in a MIAME compliant format at NCBI's Gene Expression Omnibus database (http://www.ncbi.nlm.nih.gov/ geo/) under accession numbers GSE44702 and GSE44703.
All overlap and enrichment analysis was statistically tested for significance using Fisher's exact test.

RNAi screening of longevity/immunity genes in C. elegans
RNAi clones from the Ahringer library (Source BioScience LifeSciences, UK), which were annotated as ''Correct'', were grown at 37uC in LB with 50 mg/ml ampicillin overnight. One hundred microlitres of overnight culture were then spread on NGM RNAi agar plates containing 25 mg/ml carbenicillin and 1 mM IPTG and allowed to induce overnight at 20uC. The next day 2-3 glp-1(e2141) young-adults were picked on to 3 plates per genotype and stored at 15uC. When numerous eggs were laid, the plates were shifted to 25uC to induce germline depletion in these progeny. After about 2.5 days, 60 to 100 young-adult nematodes were picked onto 3 to 5 plates containing X. nematophila and their survival was monitored daily (immunity assays). For longevity assays the same procedure was repeated except that the young adult worms continued to be reared on the respective RNAi clone bearing bacteria on NGM RNAi plates. We used two positive RNAi controls against unc-22 and dpy-9, to ensure that RNAi was efficient.

Supporting Information
Figure S1 Schematic of RNAi screen for genes involved with C. elegans glp-1(e2141) survival against X. nematophila.