Genetic analysis of Caenorhabditis elegans pry-1/Axin suppressors identifies genes involved in reproductive structure development, stress responses, and aging

Abstract The Axin family of scaffolding proteins regulates a wide array of developmental and post-developmental processes in eukaryotes. Studies in the nematode Caenorhabditis elegans have shown that the Axin homolog PRY-1 plays essential roles in multiple tissues. To understand the genetic network of pry-1, we focused on a set of genes that are differentially expressed in the pry-1-mutant transcriptome and are linked to reproductive structure development. Knocking down eight of the genes (spp-1, clsp-1, ard-1, rpn-7, cpz-1, his-7, cdk-1, and rnr-1) via RNA interference efficiently suppressed the multivulva phenotype of pry-1 mutants. In all cases, the ectopic induction of P3.p vulval precursor cell was also inhibited. The suppressor genes are members of known gene families in eukaryotes and perform essential functions. Our genetic interaction experiments revealed that in addition to their role in vulval development, these genes participate in one or more pry-1-mediated biological events. Whereas four of them (cpz-1, his-7, cdk-1, and rnr-1) function in both stress response and aging, two (spp-1 and ard-1) are specific to stress response. Altogether, these findings demonstrate the important role of pry-1 suppressors in regulating developmental and post-developmental processes in C. elegans. Given that the genes described in this study are conserved, future investigations of their interactions with Axin and their functional specificity promises to uncover the genetic network of Axin in metazoans.


Introduction
Most genetic research is aimed at linking genes to phenotypes and understanding how changes in gene function affect biological processes. Studies in animal models have demonstrated that genes exert their effects through interactions with other genes that form functional networks. Disruptions of the activity of network components can lead to various defects and in some cases premature death. Therefore, a comprehensive understanding of gene-gene interactions is crucial for the discovery of effective treatments. Our group is currently investigating the genetic network of an Axin family member in the nematode Caenorhabditis elegans. Axins are scaffolding proteins that play crucial roles in regulating conserved processes in metazoans; they integrate inputs from multiple interactors to coordinate downstream cellular signaling events. Moreover, Axin mutations have been implicated in multiple abnormalities and diseases (Mallick et al. 2019b). Therefore, elucidating the Axin signaling cascade can enhance our understanding of disease progression, and the pathway could be an attractive therapeutic target.
Work from our lab and others have shown that the C. elegans Axin homolog PRY-1 is involved in multiple developmental and post-developmental processes, including embryogenesis, neuronal differentiation, vulva formation, seam cell development, lipid metabolism, and lifespan maintenance (Maloof et al. 1999;Korswagen et al. 2002;Mallick et al. 2019b. Initial studies on pry-1 showed that the gene product acts as a negative regulator of canonical WNT signaling . PRY-1 forms a destruction complex in the absence of a WNT ligand, leading to inhibition of a WNT effector, the b-catenin homolog BAR-1. Mutations in pry-1 mimic activated WNT signaling and cause the translocation of BAR-1 to the nucleus, thereby promoting the expression of target genes (Gleason et al. 2002). During vulval development, pry-1 restricts the number of induced vulval precursor cells (VPCs). In a normal worm, three (P5.p, P6.p, and P7.p) of the six Pn.p cells (P3.p through P8.p), termed the VPCs, participate in the formation of the vulva (Sulston and Horvitz 1977;Sternberg 2005). The remaining uninduced VPCs adopt nonvulval fates and fuse with the surrounding hypodermal syncytium (Sternberg and Horvitz 1989). In the absence of pry-1, nonvulval cells are inappropriately induced to adopt specific fates, resulting in multiple ectopic ventral protrusions, a phenotype termed multivulva (Muv) (Gleason et al. 2002;Seetharaman et al. 2010). The mechanism of pry-1 action during VPC induction and cell fate specification is not well understood.
In addition to its essential function in canonical WNT signaling, pry-1 participates in the WNT asymmetric pathway to regulate the expression of heterochronic microRNAs and their targets during seam cell development (Mallick et al. 2019a). More recently, we discovered additional roles for pry-1 in the regulation of lipid metabolism, stress response, and aging, and identified its interacting genes in these processes (Mallick et al. 2019b. Whereas these findings demonstrate the essential function of pry-1 in C. elegans, how it controls diverse events and mediates specific cellular interactions remain to be explored. Given that the gene encodes a scaffolding protein, it might recruit many other factors to regulate their activities. In this study, we set out to investigate the genetic network of pry-1 by investigating a set of genes that are involved in a wide variety of cellular and molecular processes. The genes are differentially expressed in the pry-1-mutant transcriptome and while they are all associated with the GO term "reproductive structure development," many have essential roles in other tissues as well. Our RNAi knockdown experiments revealed that eight of the 26 genes tested (spp-1, clsp-1, ard-1, rpn-7, cpz-1, his-7, cdk-1, and rnr-1) strongly suppressed the pry-1 Muv phenotype and ectopic P3.p induction. We examined genetic interactions between the genes and bar-1 in vulval cells, which revealed that rpn-7 acts downstream of pry-1-bar-1-mediated signaling. All of the suppressor genes are conserved in eukaryotes and perform diverse functions such as oxidation-reduction reactions (dehydrogenase family: ard-1 and reductase family: rnr-1), protein degradation (proteasomal complex component: rpn-7 and peptidase: cpz-1), protein phosphorylation (adapter protein: clsp-1 and kinase: cdk-1), transmembrane transportation (channel protein: spp-1), and the regulation of gene expression (histone: his-7) (Davy et al. 2001;Hashmi et al. 2004;Boxem 2006;Alper et al. 2007;Mori et al. 2008;Srinivasan et al. 2008;Ossareh-Nazari et al. 2016). Four of the genes (clsp-1, his-7, cdk-1, and rnr-1) have roles in the cell cycle, DNA damage checkpoint, DNA repair, and DNA replication.
We investigated whether suppressors also participate in other pry-1-mediated processes, such as the stress response and lifespan maintenance. Whereas RNAi knockdown of four of the eight genes (cpz-1, his-7, cdk-1, and rnr-1) significantly rescued the short lifespan and stress sensitivity of pry-1 mutants, two of them (spp-1 and ard-1) affected only the stress response but not the lifespan. These results show that pry-1 utilizes overlapping subsets of genes in distinct processes. Overall, these findings demonstrate that PRY-1 interacts with a diverse set of conserved factors involved in processes such as protein modification, protein homeostasis, DNA replication, DNA repair, gene expression, and the cell cycle to control essential biological events in C. elegans.

Strains
Animals were maintained at 20 C on standard nematode growth media (NGM) plates seeded with OP50 Escherichia coli bacterial strains as described by Brenner (1974). Worm strain information can be found in Supplementary Table S1.

RNAi
RNAi mediated gene silencing was performed using a protocol previously published by our laboratory (Ranawade et al. 2018). Plates were seeded with E. coli HT115 expressing either dsRNA specific to candidate genes or empty vector (L4440). Synchronized gravid adults were bleached, and eggs were plated. Vulva or seam cell phenotypes were analyzed in young adults.
Ahringer RNAi library was the source of bacterial clones . Initially, GO term searches (in 2017; geneontology.org file "gene_ontology.obo" pulled from the version control on May 3 08:44:20 2017 UTC) had identified 36 upregulated genes associated with reproductive structure development. Of these, 26 clones were present in the RNAi library and tested. The remaining clones were either absent or did not grow. Subsequent GO analysis (performed in 2020) revealed a larger set of upregulated genes (52). Since the work on the initial 26 genes had advanced significantly by this time, additional genes were not tested.

Vulva phenotype and VPC induction analysis
Muv and protruding vulva (Pvl) phenotypes were scored in adults at plate level. Animals with multiple ventral protrusions (pseudovulvae) were termed as Muv and those with a single prominent protrusion as Pvl. VPC induction was determined during the L4 stage under a Nomarski microscope (Seetharaman et al. 2010) L4 larvae were mounted on glass slides containing 2% agar pad and the anesthetic sodium azide (1 mM). In a wild-type worm, each of the three VPCs, P5.p, P6.p, and P7.p, are induced to form the vulva (hence the induction score three, one for each precursor). The vulval progeny invaginate and fuse selectively giving rise to a "Christmas tree" appearance during the mid-to-late L4 stage. Animals with more than three induced VPCs (Muv class) show multiple distinct invaginations and are assigned induction scores of greater than three.

Lifespan analysis
All lifespan analysis was done following adult-specific RNAi treatment using a protocol described previously . Animals were grown on NGM OP50 seeded plates till late L4 stage after which they were transferred to RNAi plates. Plates were then screened daily for dead animals and surviving worms were transferred every other day till the progeny production ceased. Censoring was done for animals that either escaped, burrowed into the medium, showed a bursting of intestine from the vulva or formed a bag of worms (larvae hatches inside the worm and the mother dies). Data from the lifespan experiments are combined and represented by the Kaplan-Meier survival plot coupled with log-rank (Mantel-Cox) test (to get the statistics of average survival time) (Amrit et al. 2014).

Stress assay
Oxidative stress experiments were performed by exposing animals to 100 mM paraquat (PQ) for 1 and 2 h using a published protocol (Li et al. 2008). Final working concentrations were made in M9 buffer. At least 30 animals were tested for each strain in each replicate. Mean and standard deviation were determined from experiments performed in duplicate. Animals were considered dead if they had no response following a touch using the platinum wire pick and showed no thrashing or swimming movement in M9. Moreover, dead animals usually had an uncurled body posture compared to the normal sinusoidal shape of worms.

Body bending and pharyngeal pumping analysis
Rate of body bending per one min and the rate of pharyngeal pumping per 30 s for adults were analyzed over the period of 4 days (Collins et al. 2008). Individual hermaphrodites were analyzed for these phenotypes under the dissecting microscope by placing them on OP50 culture plates. Pharyngeal pumping was assessed by observing the number of pharyngeal contractions for 30 s. For body bending assessment, animals were stimulated by tapping once on the tail of the worm using the platinum wire pick where one body bend corresponded to one complete sinusoidal wave of the worm. Only animals that moved throughout the duration of one min were included in the analysis.

Molecular biology
RNA was extracted from synchronized L3 and day-1 adult animals. Protocols for RNA extraction, cDNA synthesis and qPCR were described earlier (Ranawade et al. 2018). Briefly, total RNA was extracted using Trizol (Thermo Fisher, USA), cDNA was synthesized using the SensiFast cDNA synthesis kit (Bioline, USA), and qPCR was done using the SYBR green mix (Bio-Rad, Canada). Primers used for qPCR experiments are listed in Supplementary  Table S1.

Statistical analyses
Statistics analyses were performed using Sigma Plot software 11, CFX Maestro 3.1, and Microsoft Office Excel 2016. For lifespan data, survival curves were estimated using the log-rank (Mantel-Cox) test and differences among groups were assessed using the log-rank test. qPCR data were analyzed using Bio-Rad CFX Maestro 3.1 software. For all other assays, data from repeat experiments were pooled and analyzed together and statistical analyses were done using GraphPad Prism 8. P-values less than 0.05 were considered statistically significant.

Results
Reproductive structure development genes are misregulated in the pry-1 mutant Given that pry-1 is involved in diverse processes, we aimed to identify its interacting genes that might act in a tissue or processspecific manner. The initial work focused on vulval development, a tissue where pry-1 plays an essential role but very little is known about the downstream target genes. We used GO enrichment analysis (http://geneontology.org/ and Wormbase release WS258) to filter differentially expressed genes associated with "reproductive structure development" (GO:0048608) in the pry-1 mutants and identified 149 genes (Supplementary Table S2). Among them, 52 and 97 genes were upregulated and downregulated, respectively, in the pry-1 mutant transcriptome ( Table S3). When examined for molecular functions, we observed enrichment in categories such as protein binding activity (64), DNA binding activity (27), RNA binding activity (20), hydrolase activity (26), signaling receptor binding activity (8), and protein kinase binding activity (7) (Supplementary Table S3). Many genes were found to be associated with cellular components, including the nucleus (74), protein-containing complexes (65), cytoplasm (52), integral components of the membrane (20), cytoskeleton (17), and nuclear chromosomes (12) (Supplementary Table S3). This suggests that pry-1 plays an essential role in regulating the expression of diverse sets of genes. Another type of analysis involved enrichment of tissues and phenotypes linked to misregulated genes (https://wormbase.org/ tools/enrichment/tea/tea.cgi, Angeles-Albores et al. 2016). The results showed significant enrichment of tissues, such as neurons and P-cell lineages (Supplementary Table S4). The genes were also significantly associated with phenotypes such as Pvl, vulval cell induction increased or decreased, and hermaphrodite reproductive system morphology variants (Supplementary Table S4).
RNAi knockdown of a subset of reproductive structure development genes suppresses the pry-1 Muv phenotype We evaluated the role of the upregulated genes in pry-1-mediated vulval development. To this end, a subset (26 of 52 genes, 50%) was experimentally tested by RNAi to determine the effect of these genes on the Muv phenotype of pry-1(mu38) animals (see Materials and Methods, Figure 2). As mentioned previously, adult pry-1 mutant hermaphrodites exhibit a Muv phenotype owing to the constitutive activation of canonical WNT signaling (Gleason et al. 2002). Knockdown of 15 of the genes by RNAi significantly suppressed ectopic pseudovulvae in pry-1 mutants (p < 0.05) ( Figure 2). For eight of these, the penetrance of the Muv phenotype was lower than the mean 6 2 standard deviations (95% confidence interval) of control RNAi-treated animals ( Figure 2); therefore, we designated them as pry-1 suppressors.
All eight genes, except one (spp-1), have homologs in higher eukaryotes including humans, suggesting their important roles in conserved biological processes. Four of the suppressors encode proteins that possess or regulate enzymatic activities, specifically acting as oxidoreductases (ard-1 and rnr-1), a protease (cpz-1), and a regulatory subunit of a proteasome complex (rpn-7). Two of the suppressors, clsp-1 and cdk-1, are involved in protein phosphorylation. cdk-1 is possibly necessary for all cell divisions in C. elegans (Boxem 2006). his-7 is a member of the human H2A family of histones that function in DNA repair and gene expression, and spp-1 is a homolog of the human gene encoding Saposin-like protein that plays a role in immunity (Table 1).
To investigate the expression of the suppressor genes in the pry-1(mu38) strain, we performed qPCR experiments at the L3 stage when VPCs undergo division to produce vulval progeny. Earlier, the transcriptome profiling revealed that all genes were upregulated in the L1 larval stage (Ranawade et al. 2018). The pattern was the same in L3 animals, except for cpz-1, which was unchanged ( Figure 3A). We also carried out qPCR analysis in day-1 adults and found that five of the suppressor genes continued to be expressed at significantly high levels. Of the remaining three, ard-1 and rpn-7 were unchanged and his-7 was downregulated ( Figure 3B). Thus, most of the suppressor genes are negatively regulated by pry-1 in L3 larvae and young adults. The results also suggest that pry-1 regulates some of the genes in a stage and tissue-specific manner. Thus, it could be that cpz-1 levels are upregulated in vulval cells but not reflected by whole animal qPCR analysis or that its temporal requirement in vulval cells is different. More work is needed to examine these possibilities. Overall, these results support the key roles of suppressor genes downstream of pry-1 in vulva formation and their potential involvement in mediating pry-1 function in adults.
To understand the cellular basis of Muv suppression, the VPC induction pattern was investigated. In control RNAi treatments, pry-1(mu38) animals showed an average VPC induction of 3.4 6 0.6 (n ¼ 29), which is higher than the N2 control ( Table 2).
The increase was mainly due to P3.p, and to a lower extent P4.p and P8.p, being ectopically induced. The RNAi knockdown of all eight genes suppressed the P3.p defect in pry-1 mutant animals ( Figure 4C, Table 2) (see Materials and Methods for details). For six of the genes (ard-1, rpn-7, cpz-1, his-7, cdk-1, and rnr-1. cdk-1 and rnr-1) the average VPC induction was also reduced (Figure 4, A and B and Table 2).
We also examined the effects of suppressors on wild-type vulval development. RNAi experiments indicated no significant reduction in VPC induction for any of the genes except cdk-1 and rnr-1 (Figure 4, A and D and Table 2). cdk-1 and rnr-1 RNAi resulted in 39% and 96.3% reductions in VPC induction, respectively, when compared to that in controls, suggesting that both genes play essential roles in vulva formation. These data, together with expression studies, support our conclusion that all  Table S2). The dotted line on the x-axis corresponds to log 2 fold change of 1 and the one on the y-axis shows p-value of 0.05. Orange and green dots represent significantly upregulated and downregulated genes, respectively. (B) Phenotype-enrichment analysis of genes shown in (A). Not all categories are listed. See Supplementary Table S4 for a complete list.

Figure 2
Quantification of the Muv phenotype following RNAi knockdown of 26 upregulated genes in pry-1(mu38) animals. Data represent the mean of two replicates (n > 40 animals in each replicate) and error bars represent the standard deviation. For eight of the genes, located on the right of the dotted vertical line, Muv penetrance was lower than the mean 6 2x standard deviation of the control (L4440). Statistical analyses were done using one-way ANOVA with Dunnett's post hoc test and significant differences are indicated by stars (*): * (p < 0.05), ** (p < 0.01), *** (p < 0.001), # (p < 0.0001). DNA replication; metabolic process; deoxyribonucleotide biosynthetic process; and oxidation-reduction process Ribonucleoside-diphosphate reductase activity Figure 3 Expression levels of suppressor genes determined by qPCR in pry-1 and bar-1 mutants. Normalized expression of suppressor genes in pry-1(mu38) L3 larvae and adults (A, B) and bar-1(ga80) L3 larvae (C). Each data represents the mean of two replicates and error bars the standard error of means. Significance was calculated using Bio-Rad software (one-way ANOVA) and significant differences are indicated by stars (*): * (p < 0.05), ** (p < 0.01), *** (p < 0.001).
of the suppressor genes act genetically downstream of pry-1 to regulate vulval development. In addition, it is possible that cdk-1 and rnr-1 act in a pathway parallel to pry-1.

Genetic interactions between suppressor genes and bar-1/b-catenin
Since PRY-1 is a component of the canonical WNT signaling, we investigated whether any of the suppressor genes function as downstream effectors of the pathway during vulval development.
To this end, we used the b-catenin homolog BAR-1, which is negatively regulated by PRY-1 (Eisenmann et al. 1998;Gleason et al. 2002). Mutations in bar-1 cause some of the VPCs to remain uninduced. In a bar-1 null mutant, bar-1(ga80), P3.p and P4.p usually adopt an F (fused) fate. The frequency of the F fate is much lower for the remaining VPCs (12-36% in each case), as they are mostly induced to form the vulva (Eisenmann et al. 1998;Eisenmann and Kim 2000). The VPC induction analysis following the RNAi knockdown experiments revealed that for six of the candidate genesspp-1, ard-1, cpz-1, his-7, cdk-1, and rnr-1-the VPC induction defect of bar-1(ga80) was significantly enhanced. cdk-1 and rnr-1 RNAi had the most severe effects, and the induction was reduced by 52% and 100%, respectively. In contrast, clsp-1 and rpn-7 did not affect VPC induction ( Figure 4E and Table 2). We also performed qPCR to examine the levels of suppressor genes in bar-1(ga80) animals, which showed that whereas four exhibited increased expression during the L3 stage (cpz-1, his-7, cdk-1, and rnr-1), one was reduced (rnr-7), and the remaining three were unchanged (spp-1, clsp-1, and ard-1) ( Figure 3C). These results, together with interaction experiments involving pry-1, are most consistent with the possibility of rpn-7 acting genetically downstream of the pry-1-bar-1 pathway to regulate VPC induction. The remaining genes might mediate pry-1 function in a bar-1-independent manner.
Suppressor genes influence pry-1-mediated nonvulval processes PRY-1 plays crucial roles in multiple developmental and postdevelopmental processes; therefore, we examined the involvement of Muv suppressors in all or subsets of PRY-1 nonvulval functions. The phenotypes examined included an increased seam cell number, molting defect, low brood size, developmental delay, stress sensitivity, increased expression of chaperones (hsp-4/BiP/GRP78, hsp-6/HSP70, and hsp-60/HSP60), and short lifespan (Ranawade et al. 2018;Mallick et al. 2019a. The partial or complete loss of function of many of the suppressor genes is known to cause defects similar to that of pry-1 mutants. These include ard-1 RNAi leading to sterile progeny and developmental delay ( To test whether the suppressor genes participate in pry-1-mediated developmental processes outside the vulva system, we first examined the seam cells. During seam cell development, pry-1 promotes asymmetric cell division, and pry-1 mutants show increased cell numbers. RNAi of the eight genes did not result in a change in the seam cells in both pry-1 mutant and wild-type animals (Supplementary Figure S1), suggesting that none of these genes play a role in pry-1-mediated signaling in generating seam cells.
rnr-1 RNAi, respectively) were much higher compared to those observed in wild-type (19.5% and 18.8% for cdk-1 RNAi and rnr-1 RNAi, respectively) (Figure 5, C and D; Table 2). The effects of all four genes on body bending and pharyngeal pumping were also examined. We observed that knockdown resulted in significant improvements in both physiological markers of aging in pry-1 mutants but not in wild-type animals ( Figure 5, A-D). Overall, these results allow us to suggest that cpz-1, his-7, cdk-1 and rnr-1 act downstream of pry-1 in restricting the lifespan of animals.
In addition to its role in aging, pry-1 is necessary for maintaining the expression of stress response signaling genes . This led us to examine whether the knockdown of suppressors would decrease the lethality of pry-1 mutants caused by acute exposure to the nonspecific oxidative stress inducer PQ.  Table 3 for lifespan data and statistical analyses. (E, F) Bar graphs showing the rates of body bending and pharyngeal pumping of pry-1 mutants over a period of 4 days following RNAi of cpz-1, his-7, cdk-1, and rnr-1. (G, H) Bar graphs showing the rates of body bending and pharyngeal pumping of N2 animals. (E-H) Data represent the mean of two replicates (n > 10 animals per replicate) and error bars represent the standard deviation. Statistical analyses for panels E-H were done using one-way ANOVA with Dunnett's post hoc test for each day and significant differences are indicated by stars (*): * (p < 0.05), ** (p < 0.01).
Except for clsp-1 and rpn-7, RNAi targeting the suppressor genes reduced the stress sensitivity in pry-1 mutants ( Figure 6A). Interestingly, ard-1 and cpz-1 RNAi conferred PQ resistance in wild-type animals as well (Supplementary Figure S3), suggesting crucial roles for both of these genes in maintaining PQ sensitivity in animals. These data suggest that pry-1 inhibits spp-1, his-7, cdk-1, and rnr-1 to regulate oxidative stress responses in animals. More work is needed to determine whether ard-1 and cpz-1 also interact with pry-1 or act in a parallel pathway.
We investigated whether downregulating the expression of six PQ-responsive genes in pry-1 mutants could lower cellular stress (Supplementary Figure S4, A and B). This was done using a set of stress response reporters. pry-1 mutants exhibit increased expression of hsp-4::GFP (endoplasmic reticulum unfolded protein response chaperone) and hsp-60::GFP (mitochondrial unfolded protein response chaperone) (Supplementary Figure S4B)  . Another stress response reporter that is sensitive to oxidative stress, sod-3::GFP (sodium dismutase), was also analyzed. Although pry-1 mutants do not affect sod-3::GFP fluorescence, muscle-specific overexpression of pry-1 causes an increase in sod-3 levels . RNAi targeting all but spp-1 caused a significant reduction in GFP fluorescence in hsp-4::GFP animals ( Figure 6B). A similar knockdown in the sod-3::GFP strain revealed that the fluorescence was strongly suppressed in the case of spp-1, cpz-1, cdk-1, and rnr-1 ( Figure 6D). There was no significant change in hsp-60::GFP expression levels ( Figure 6C). Overall, these results show that while there are differences in individual chaperon responses, all six suppressors are involved in pry-1-mediated processes and function to maintain the expression of multiple stress-responsive genes.

Discussion
In this study, we analyzed a set of genes that are upregulated in pry-1 mutants and associated with "reproductive structure development." The genes belong to GO categories that include metabolic processes, transcriptional regulation, and mitotic cell cycle. Among the 26 genes tested, RNAi for eight (spp-1, clsp-1, ard-1, rpn-7, cpz-1, his-7, cdk-1, and rnr-1) suppressed the Muv phenotype of pry-1 mutant animals with a threshold of the mean 6 2 standard deviations (95% confidence interval). The ectopic P3.p induction defect was also suppressed in all cases. Further gene expression studies and genetic interactions with a null allele of bar-1 revealed that whereas all of the suppressors function downstream of pry-1 to regulate vulva formation, only rpn-7 is involved in pry-1-bar-1 signaling. In addition, we found that cdk-1 and rnr-1 have essential roles because their downregulation caused defects in VPC induction in wild-type animals.
Whereas our work provides the first evidence for the genetic interactions between pry-1 and these eight genes (see schematic in Figure 7), some of the genes were previously reported to play roles in vulva formation. Specifically, the disruption of clsp-1 and ard-1 causes a Pvl phenotype (Simmer et al. 2003;Ceron et al. 2007). cpz-1 localizes to the developing vulva, and cpz-1 RNAi results in defective vulval morphology (Hashmi et al. 2004). cdk-1 regulates lin-12/Notch in a cell cycle-dependent manner (Nusser-Stein et al. 2012;Weinstein et al. 2015). The remaining four genes, spp-1, rpn-7, his-7, and rnr-1 had no reported function in the vulva system.
In addition to studying vulval development, we investigated the role of the suppressor genes in pry-1-mediated postdevelopmental events. spp-1, rpn-7, cdk-1, and rnr-1 are known to be involved in aging. Specifically, rpn-7 and cdk-1 are both required for glp-1-mediated lifespan extension (Ghazi et al. 2007;Seidel and Kimble 2015), and spp-1 and rpn-7 are required for the longevity of daf-2 mutants (Murphy et al. 2003;Ghazi et al. 2007; Table 3 Mean, median, and maximum lifespan of N2 and pry-1(mu38) animals following control (empty vector L4440) and gene specific RNAi  Anyanful et al. 2009). rnr-1 is downregulated in long-lived daf-2 and eat-2 mutants, consistent with the notion that high levels of rnr-1 decrease the lifespan (Gao et al. 2018). These genes have potential roles in the conserved aging pathway; therefore, we analyzed their requirement in pry-1 signaling. Our results showed that cpz-1, his-7, cdk-1, and rnr-1 act downstream of pry-1 to affect the lifespan of animals. These four genes, along with spp-1 and ard-1, also play roles in pry-1-mediated stress response maintenance, as observed in the acute PQ assay and heat shock chaperone analysis. Thus, whereas pry-1 negatively regulates many genes, it appears to utilize overlapping subsets in different events (Figure 7). We also found that similar to vulval development cdk-1 and rnr-1 extend the lifespan of wild-type animals, suggesting their essential roles in multiple tissues. Both genes may also act in a pry-1-independent manner. The involvement of spp-1 and ard-1 in the stress response but not aging is consistent with the results of other studies describing genes that have unique roles in these two processes (Chen et al. 2009;Bennett et al. 2014;Richman et al. 2018). Another mechanism by which some of the suppressor genes may mediate pry-1 signaling is by regulating proteostasis. Whereas the loss of proteostasis contributes to aging and ageassociated abnormalities, its enhancement promotes lifespan extension and results in suppression of age-related diseases (Labbadia and Morimoto 2014;Uno and Nishida 2016). Consistent with pry-1's role in proteostasis, we have found that the gene is necessary for the maintenance of stress response. In addition, preliminary work in our lab has revealed that pry-1 mutants have defects in protein folding and protein degradation.
Our data also suggest that the roles of spp-1 in pry-1 and daf-2 signaling pathways are different, although the extent to which these differences are reflected in its molecular function is unknown. Considering that spp-1 encodes a channel protein, it is possible that blocking its activity inhibits unwanted signaling in daf-2 and pry-1 mutant adults leading to improved responses. However, animals lacking spp-1 function are expected to be more sensitive to infections as this gene is involved in maintaining the innate immune response (Anyanful et al. 2009). Future experiments are needed to make any firm conclusions regarding the mechanism through which spp-1 functions in the stress response and lifespan maintenance.
In conclusion, this study identified a new set of interactors of pry-1 in C. elegans that are involved in a range of cellular events, such as protein homeostasis, signaling, gene expression, and cell proliferation, by regulating the activities and stability of proteins and changes in DNA, such as replication, the Figure 6 cpz-1, his-7, cdk-1, and rnr-1 regulate stress sensitivity in pry-1 mutants. (A) Survivability of pry-1(mu38) animals following RNAi knockdown of suppressor genes. The animals were treated with 100 mM PQ solution for 1 h. Data represent mean of two replicates (n > 30 animals). (B) Quantification of fluorescence intensity using hsp-4::GFP marker in pry-1 mutants following RNAi knockdown of spp-1, ard-1, cpz-1, his-7, cdk-1, and rnr-1. (C) Same as (B), except that fluorescence reporter is hsp-60::GFP (D) Same as B, except that fluorescence reporter is sod-3::GFP. Data represent the mean of two replicates (n > 15 animals per replicate). Error bars represent the standard deviation. Statistical analyses were done using one-way ANOVA with Dunnett's post hoc test for each day and significant differences are indicated by stars (*): * (p < 0.05), ** (p < 0.01).
damage checkpoint, and repair. Some of the suppressors affect multiple pry-1-mediated processes, whereas the others appear to have more restricted roles. All of the genes have mammalian homologs or family members, raising the possibility that their interactions with Axin might be conserved. Future studies hold promise to elucidate the mechanism by which these genes mediate tissue-specific functions of Axin in normal and disease conditions.

Data availability
Strains are available upon request. The authors affirm that all data necessary for confirming the conclusions of the article are present within the article, figures, and tables.
Supplementary material is available at G3 online.

Figure 7
A schematic diagram showing biological processes mediated by pry-1 and eight suppressor genes. The genetic relationship is based on the expression and functional data described in this study. While all genes are involved in pry-1-mediated vulval development, only six affect stress response and four lifespan. Question mark (?) indicates that cdk-1 and rnr-1 may also be regulated in a pry-1-independent manner.