Endoplasmic Reticulum Stress Caused by Lipoprotein Accumulation Suppresses Immunity against Bacterial Pathogens and Contributes to Immunosenescence

ABSTRACT The unfolded protein response (UPR) is a stress response pathway that is activated upon increased unfolded and/or misfolded proteins in the endoplasmic reticulum (ER), and enhanced ER stress response prolongs life span and improves immunity. However, the mechanism by which ER stress affects immunity remains poorly understood. Using the nematode Caenorhabditis elegans, we show that mutations in the lipoproteins vitellogenins, which are homologs of human apolipoprotein B-100, resulted in upregulation of the UPR. Lipoprotein accumulation in the intestine adversely affects the immune response and the life span of the organism, suggesting that it could be a contributing factor to immunosenescence. We show that lipoprotein accumulation inhibited the expression of several immune genes encoding proteins secreted by the intestinal cells in an IRE-1-independent manner. Our studies provide a mechanistic explanation for adverse effects caused by protein aggregation and ER stress on immunity and highlight the role of an IRE-1-independent pathway in the suppression of the expression of genes encoding secreted proteins.

atf-6, and perk-1-help to maintain ER homeostasis via a number of sophisticated mechanisms that control translation, protein degradation, mRNA degradation, and chaperone levels (6,7). However, the UPR becomes compromised with age (8,9), leading to enhanced protein aggregation in the ER.
Recent studies have shown that the UPR plays crucial roles in the regulation of the life span and immunity of organisms (9)(10)(11). Constitutive expression of the active form of XBP-1, that is part of the most conserved branch of the UPR, has been found to prolong life span (9). It has also been shown that bacterial infection causes damage to proteins at the site of infection, which leads to their aggregation (12), and that bacterial infection upregulates the UPR (13,14). In addition, an increase in the UPR enhances the immunity of animals against bacterial infection (15)(16)(17).
We conducted a forward genetic screen in the nematode model organism, Caenorhabditis elegans, to identify dominant mutants exhibiting high UPR levels. We obtained several mutant animals with an upregulated UPR. Whole-genome sequencing (WGS) revealed that the UPR upregulation in these mutants resulted from mutations in vitellogenin proteins (VIT), which are lipoproteins involved in the transport of lipids from the intestine to oocytes and are homologs of human apolipoprotein B-100 (apoB-100). We show that mutations in lipoproteins result in their accumulation in the intestine, which causes ER stress and adversely affects the life span of the organisms and their resistance to pathogen infection. The upregulated UPR counteracts the toxic effects of ER stress and improves immunity and prolongs the life span of the mutants. Our studies also suggest that accumulation of lipoproteins may be a factor in immunosenescence. Quantification of the levels of expression of genes using RNA sequencing showed that ER stress caused by lipoprotein accumulation significantly reduced the expression levels of genes encoding secreted immune effectors. Our studies uncovered a novel mechanism by which ER stress suppresses innate immunity.

RESULTS
Forward genetic screen for mutants exhibiting UPR upregulation. We conducted a forward genetic screen using the UPR reporter strain SJ4005 Phsp-4::gfp to isolate mutants exhibiting an upregulated UPR (Fig. 1A). The transcription of hsp-4, the C. elegans ortholog of the mammalian ER-localized Hsp70 chaperone BiP, is upregulated in response to ER stress. An F 1 screen of approximately 80,000 ethyl methanesulfonate (EMS)-mutagenized haploid genomes resulted in the isolation of seven dominant mutants with enhanced green fluorescent protein (GFP) expression compared to the parental strain SJ4005 (see Fig. S1A in the supplemental material). All of the mutants showed very high levels of GFP in the intestine compared to the SJ4005 animals. We randomly selected five mutants for in-depth analysis (Fig. 1B). The enhanced GFP expression was completely blocked by knockdown of the UPR gene xbp-1 and partially blocked by knockdown of the UPR gene ire-1 by RNA interference (RNAi) ( Fig. 1C and Fig. S1A). We confirmed activation of the UPR by monitoring the transcript levels of the ER chaperone genes hsp-4 and hsp-3 (Fig. 1D). The mutants had significantly higher levels of the ER chaperone transcripts than the SJ4005 animals did. However, the mutants did not show any significant changes in the transcript levels of hsp-6 ( Fig. S1B), a mitochondrial chaperone that is activated in the mitochondrial UPR. The mutants showed a slight but nonsignificant increase in the transcript levels of the cytosolic chaperone hsp-16.2 (Fig. S1B), but these changes are marginal compared with those in hsp-16.2 transcript levels upon heat stress (18), which induces hsp- 16.2. Interestingly, the mutants did not show enhanced GFP levels in comparison to SJ4005 animals until 62 h of growth at 20°C (Fig. 1E). The levels of GFP started to increase in the mutant animals only at 66 h, which coincides with the developmental stage at which L4 larvae molt into young adults. This finding indicated that the UPR was activated in the mutants only in the adult stage and not in the larval stages. Furthermore, the male mutant animals did not display any upregulation of the UPR compared to the SJ4005 animals (Fig. 1F). Taken together, these observations indicated that activation of the UPR in the mutant animals was specific to the adult hermaphrodites. genome ( Fig. 2A). VIT proteins are lipoproteins that are expressed in the intestine of the adult hermaphrodite and are involved in the transport of lipids from the intestine to oocytes (19). Moreover, they have a high propensity to misfold and cause ER stress, and they are one of the most abundant proteins in C. elegans (20). This high abundance and propensity to cause ER stress may explain why all the mutations corresponded to vit genes.
Consistent with our results indicating that activation of the UPR was observed only in mutant hermaphrodites, vit genes are not expressed in males. Moreover, the UPR was upregulated in the mutants only in the intestine ( Fig. 1B and Fig. S1A), the site of VIT protein synthesis. We reasoned that if upregulation of the UPR is caused by mutations in vit genes, inhibition of the mutated genes should rescue the phenotype. Among the genes encoding vitellogenins, a high level of sequence similarity is observed, resulting in their classification into three groups. The vit-1 and vit-2 genes correspond to a group closely related to the group comprising vit-3, vit-4, and vit-5, while vit-6 has a relatively low similarity to other vit genes (Fig. S3). The level of sequence similarity indicates that vit-1 and vit-2, vit-3, vit-4, and vit-5 would undergo complete cross RNAi (21). RNAi clones for vit-2 and vit-6 are unavailable in the two commercial C. elegans RNAi libraries, but given the aforementioned sequence similarity, RNAi against vit-1 is expected to inhibit vit-2. Consistently, RNAi against vit genes caused a complete suppression of the enhanced UPR of the tested mutants (Fig. 2B).
Mutations in VIT proteins affect their transport and lead to accumulation in the intestine. Next, we sought to identify the mechanism by which the dominant mutations in vit genes activate the UPR. A previous study showed that reduction of unsaturated fatty acids in membranes upregulates the UPR in C. elegans (22). Because VIT proteins are involved in the transport of lipids, we decided to study whether the identified mutations may disturb lipid homeostasis in the intestine. To test whether the UPR upregulation was due to a reduction of unsaturated fatty acids, we supplemented the animal diets with different unsaturated fatty acids. Supplementation of the diet with fatty acids did not suppress the upregulated UPR (Fig. S4), indicating that the UPR was not upregulated due to a reduction of unsaturated fatty acids.
All of the mutations identified in the VIT proteins are nonconservative in nature. For example, in three isolated mutants, the smallest and neutral amino acid, glycine, is substituted by large and charged amino acids ( Fig. 2A). These kinds of amino acid changes may lead to perturbations in the conformation of the protein, potentially resulting in aggregation. Two of the mutants also exhibited the same residue change, glycine for arginine, in a highly conserved region that is also present in human apoB-100 (23) ( Fig. 2A and C). This conserved region of apoB-100 is known to have several disease-linked mutations (Fig. 2C), which promote misfolding and hampered secretion (24)(25)(26). VIT proteins are expressed in the intestine, from where they are secreted and transported to oocytes so that they can be incorporated into the developing embryos to provide nutrients (19). Thus, we tested whether the upregulated UPR observed in the isolated mutants might be caused by hampered secretion and subsequent accumulation of VIT proteins in the intestine. To examine this possibility, we generated transgenic animals expressing VIT-2::GFP and VIT-2(G839R)::GFP. VIT-2(G839R) carries a mutation analogous to the G841R and G836R mutations found in the vit-1(ac2) and vit-5(ac5) mutants.
Consistent with earlier studies (19), we observed VIT-2::GFP in both the intestine and embryo ( Fig. 3A and Fig. S5). In contrast to VIT-2::GFP, VIT-2(G839R)::GFP remained in the intestine and was not transported to embryos (Fig. 3A). This finding indicated that the G839R mutation could lead to the misfolding and aggregation of VIT proteins, affecting secretion and transport to oocytes. Consistently, VIT-2(G839R)::GFP aggregates were observed in the intestine as well as in the pseudocoelomic region (Fig. 3B). Taken together, these data showed that activation of the UPR in the isolated mutants was caused by the accumulation of VIT proteins in the intestine.
Accumulation of VIT proteins leads to enhanced susceptibility to pathogens and a shortened life span. On the basis of the previous results, we hypothesized that the accumulation of VIT proteins might affect the intestinal immune response of all isolated mutants. Concomitant blockage of the ATF-6 and XBP-1 branches of the UPR and exposure to bacterial toxins are known to cause intestinal degeneration that results in smaller animals and paler intestines that are visible under the light microscope (13,27). As shown in Fig. 4A, vit-2(ac3) animals exhibited smaller sizes and paler appearances compared to the parental strain SJ4005, and these morphological changes correlated with a strong susceptibility to infection by the pathogenic bacterium Pseudomonas aeruginosa PA14 ( Fig. 4B and Table S1).
When the UPR was downregulated by RNAi against ire-1, vit-2(ac3) and vit-6(ac6) animals showed strong morphological changes associated with degeneration ( Fig. 4C), while RNAi against xbp-1 imposed those changes in all vit mutants (Fig. 4E). In contrast with the results shown in Fig. 4B, all of the mutants showed a strong susceptibility to  Table S1). The susceptibility to pathogen infection was even stronger in mutants where xbp-1 was inhibited by RNAi (comparison of Fig. 4D and F). Consistent with these results, xbp-1 RNAi inhibited the observed UPR more efficiently than ire-1 RNAi did (Fig. S1A).
There were no significant differences between the brood size of the mutants and that of SJ4005 animals (Fig. S6), suggesting that the accumulation in the intestine of a single VIT protein did not affect the overall transport of nutrients to oocytes. To  coli HT115 containing empty vector at 20°C until they were 1-day-old adults before images were acquired. (B) Survival of SJ4005 and mutant animals on P. aeruginosa PA14 at 25°C after treatment with control RNAi. The animals were grown on E. coli HT115 containing empty vector at 20°C until they were 1-day-old adults before transferring to P. aeruginosa PA14 at 25°C. P values comparing the values for mutant animals to the values for SJ4005 animals are as follows: P Ͻ 0.001 for vit-1(ac2), P Ͻ 0.001 for vit-2(ac3), P Ͻ 0.001 for vit-4(ac4), P Ͻ 0.001 for vit-5(ac5), and P Ͻ 0.001 for vit-6(ac6). (C) Photomicrographs of SJ4005 and mutant animals grown on E. coli HT115 containing vector for RNAi against ire-1 at 20°C until they were 1-day-old adults before images were acquired. (D) Survival of SJ4005 and mutant animals on P. aeruginosa PA14 at 25°C after treatment with ire-1 RNAi. The animals were grown on E. coli HT115 containing vector for RNAi against ire-1 at 20°C until they were 1-day-old adults before being transferred to P. aeruginosa PA14 at 25°C. The P values for all mutant animals relative to the values for SJ4005 animals are P Ͻ 0.001. (E) Photomicrographs of SJ4005 and mutant animals grown on E. coli HT115 containing vector for RNAi against xbp-1 at 20°C until they were 1-day-old adults before images were acquired. (F) Survival of SJ4005 and mutant animals on P. aeruginosa PA14 at 25°C after xbp-1 RNAi. The animals were grown on E. coli HT115 containing vector for RNAi against xbp-1 at 20°C until they were 1-day-old adults before being transferred to P. aeruginosa PA14 at 25°C. P values for all mutant animals relative to SJ4005 animals are P Ͻ 0.001. Survival curves are representative assays for three independent experiments. n ϭ 60 to 100. ER Stress in the Intestine Suppresses Immune Genes ® examine the effects of the mutations on the life spans of the animals, we compared the survival of the mutants with that of SJ4005 animals grown on nonpathogenic Escherichia coli OP50. While vit-1(ac2), vit-4(ac4), and vit-6(ac6) did not show any significant differences in life span, vit-2(ac3) and vit-5(ac5) had a significantly reduced life span at 20°C (Fig. 5A and Table S1). In contrast, all of the mutants showed significantly enhanced susceptibility to the pathogenic bacteria P. aeruginosa PA14 at 20°C (Fig. 5B and Table S1). These results indicated that while some mutants might have been able to survive as long as SJ4005 animals under nonstressed conditions, under the stress caused by infection, all of the mutants exhibited reduced survival compared to the SJ4005 animals. Consistent with the idea that an increase in temperature enhances proteotoxic stress (28), all of the mutants displayed reduced survival when grown on E. coli OP50 at 25°C (Fig. 5C and Table S1). The mutant animals also showed enhanced susceptibility to another pathogen, Salmonella enterica (Fig. S7).
The mutant animals exhibited a reduced life span upon xbp-1 RNAi, and more than 50% of the mutant animals died within 5 days of adulthood (Fig. 5D). Taken together, these results show that the increased ER stress caused by accumulation of VIT proteins reduces the life span and enhances the pathogen susceptibility of animals, indicating that the upregulated UPR helped to alleviate the toxic effects of the ER stress on the animals. The results also indicate that the enhanced susceptibility of vit-1(ac2), vit-4(ac4), and vit-6(ac6) animals to pathogen infection at 20°C is not due to general sickness, suggesting that mutations in VIT proteins reduce immunity against bacterial infections. Singh and Aballay ® High levels of VIT proteins enhance susceptibility to pathogen infection and contributes to immunosenescence. It is known that the UPR decreases with aging (8,9), with a concomitant increase in the levels of VIT proteins (29,30). It is also known that the immune response decreases with aging (31,32). To examine whether protein accumulation with aging may affect immunity, we decided to mimic some aspects of aging by blocking the UPR and by overexpressing VIT-2. While control RNAi animals displayed VIT-2::GFP mainly in the embryos, xbp-1 RNAi animals primarily showed VIT-2::GFP in the intestine (Fig. 6A and Fig. S8). Consistent with our previous results indicating that VIT protein accumulation affects intestinal physiology, animals expressing VIT-2::GFP exhibited pale intestines upon xbp-1 RNAi but not control RNAi (Fig. 6A and Fig. S8). As shown in Fig. 6B, xbp-1 RNAi of animals expressing VIT-2::GFP had an enhanced susceptibility to P. aeruginosa infection compared to control RNAi. These results suggest that the accumulation of VIT protein in the intestine may negatively affect the response to infection. To address whether accumulation of VIT proteins with age affects the response to infection, we studied the pathogen susceptibility of aging animals upon knockdown of vit genes. We found that RNAi against vit-1, which is known to inhibit expression of other vit genes (30), enhanced the resistance to P. aeruginosa infection of 1-day-old adult animals (Fig. 6C). The resistance to P. aeruginosa infection upon vit-1 RNAi was further enhanced in 3-day-old adult animals :GFP on P. aeruginosa PA14 at 25°C after treatment with control RNAi or xbp-1 RNAi. The animals were grown on RNAi plates at 20°C until they were 1-day-old adults before being transferred to P. aeruginosa PA14 at 25°C. Comparing the value for control RNAi versus xbp-1 RNAi for animals expressing VIT-2::GFP, P Ͻ 0.001. (C to E) Survival of N2 animals on P. aeruginosa PA14 at 25°C after treatment with control RNAi and vit-1 RNAi. The animals were grown on RNAi plates at 20°C until they were 1-day-old, 3-day-old, and 5-day-old adults before being transferred to P. aeruginosa PA14 at 25°C. ER Stress in the Intestine Suppresses Immune Genes ® (Fig. 6D). Aging C. elegans nematodes are known to have diminished UPR (9) and enhanced aggregation of the proteome (33,34), and this overall proteotoxic stress correlates with enhanced susceptibility to pathogen infection (comparison of Fig. 6C to E). As shown in Fig. 6C and D, vit-1 RNAi, which reduced the expression of one of the most abundant intestinal proteins, improved the resistance to P. aeruginosa infection of mildly aged C. elegans. However, vit-1 RNAi had no effect on the resistance to P. aeruginosa infection in 5-day-old adult animals (Fig. 6E), suggesting that the deleterious effects of the overall misfolded proteome of 5-day-old animals cannot be reversed by only inhibiting VIT protein expression. Taken together, these results suggest that the accumulation of VIT proteins in the intestine may affect not only longevity but also immunity and that they may contribute to immunosenescence.
VIT accumulation leads to reduced gene expression of genes encoding secreted immune effectors. To understand the mechanism of enhanced susceptibility to pathogen infection caused by VIT accumulation, we used RNA sequencing to focus on transcriptional changes induced by VIT protein accumulation in vit-2(ac3) animals compared to SJ4005 animals. Of the 542 differentially regulated genes, 186 genes were upregulated, while 356 genes were downregulated (Table S2). Consistent with the idea that VIT protein accumulation induces ER stress, gene ontology analysis for biological function of the upregulated genes showed enrichment in genes linked to ER stress and the UPR (Table S2). Interestingly, we found that a subset of downregulated genes was enriched in ontology clusters associated with defense response, innate immune response, and defense responses to Gram-negative and Gram-positive bacteria (Table S2). These downregulated genes included lysozymes, saposin-like proteins (spp genes), and C-type lectins (Fig. 7A), which are conserved markers of immune activation that encode immune effectors secreted by the intestinal cells. For further analysis, we selected genes that have been experimentally linked to host defense against infections and confirmed their downregulation by quantitative reverse transcription-PCR (qRT-PCR) (Fig. 7B).
Because enhanced ER stress is known to activate regulated IRE-1-dependent degradation (RIDD) of mRNAs (35), we studied whether genes encoding secreted immune effectors were repressed by IRE-1. Thus, we compared the mRNA levels of immune genes of vit-2(ac3) ire-1 RNAi animals to those of SJ4005 ire-1 RNAi animals. We found that the expression level of 7 out of 8 immune genes was significantly inhibited in vit-2(ac3) animals in an IRE-1-independent manner (Fig. 7B). Thus, only a single studied gene, thn-2, showed partial IRE-1-dependent inhibition. These studies indicate that the expression of immune genes encoding secreted proteins is downregulated by the ER stress caused by accumulation of VIT proteins in an IRE-1-independent manner.

DISCUSSION
We demonstrated that mutations in lipoproteins in C. elegans led to their accumulation in the intestine and resulted in upregulation of the UPR. Although protein accumulation enhanced the UPR, which mitigated the adverse effects of accumulated VIT proteins on the isolated mutant animals, the animals exhibited enhanced susceptibility to pathogen infection and a shortened life span. Our results also indicate that the enhanced susceptibility to pathogen infection of animals exhibiting high ER stress due to accumulation of VIT proteins may be due to a mechanism activated by the resultant ER stress that inhibits the expression of a small subset of genes in the C. elegans genome that is highly enriched in immune effectors (Fig. 7C).
Surprisingly, even after cessation of the self-fertile reproductive period, the levels of VIT proteins continue to increase (29,30,36,37). Moreover, not only does the amount of VIT proteins increase with age but also the rate of VIT protein synthesis increases (37). It is not clear why the VIT proteins are expressed even after cessation of the self-fertile reproductive period. One possibility is that VIT proteins remain necessary for reproduction by cross-fertilization, which occurs up to day 13 of adulthood (38) and is more than two times longer than the self-fertilization period. Another possibility is that the animals are not under selection pressure after the reproductive period, and hence, no active mechanism has evolved to inhibit VIT protein expression during this period.
Gene expression and proteomic studies of several long-lived mutants have shown that these animals have reduced levels of vit genes and VIT proteins (30,(39)(40)(41)(42). In addition, RNAi against different vit genes prolongs life span in multiple studies (30,40,43). Interestingly, long-lived mutants, which have reduced levels of vit genes and VIT proteins, carry mutations affecting different processes that are capable of controlling longevity, such as the daf-2/daf-16 pathway, sterility, dietary restriction, and AMPactivated protein kinase pathway, among others. Even chemical substances such as garlic extract (44) and tyrosol (45), which enhance life span, lead to reduced levels of VIT proteins in animals. VIT proteins were also found to be highly abundant in insoluble fractions of the proteomes from aged animals (33,34), likely because of their aggregation. Because all of the different longevity pathways seem to merge at reduced levels of VIT proteins, these studies highlight that one of the most downstream strategies utilized to enhance life span could be a reduction of the expression of aggregationprone proteins such as lipoproteins. It has been shown that animals with a longer life span would also possess improved pathogen resistance (46). This could be due, at least in part, to reduced levels of aggregation-prone proteins, such as lipoproteins. Indeed, we observed that vit-1 RNAi enhanced the resistance of animals to pathogens.
Several bacterial pathogens are known to induce ER stress in host cells. In C. elegans, both Gram-positive and Gram-negative pathogenic bacteria have been shown to enhance protein aggregation (12). These findings are further supported by the observation that pathogenic P. aeruginosa bacteria upregulate the UPR (14). Moreover, ER Stress in the Intestine Suppresses Immune Genes ® P. aeruginosa infection led to reduced expression levels of genes lys-7, spp-1, and thn-2, which are required for optimal survival on P. aeruginosa (47). We found that these genes, which are also markers of immune activation by exposure to other pathogenic bacteria such as Salmonella enterica serovar Typhimurium and Enterococcus faecalis, were downregulated by the ER stress caused by VIT accumulation. Taken together, these results suggest that the ER stress caused by P. aeruginosa infection leads to downregulation of genes encoding secreted immune effectors.
In summary, our results provide evidence that the ER stress caused by protein accumulation results in the downregulation of genes encoding secreted proteins. Enhanced ER stress is known to lead to repression of specific mRNAs, which encode components of the plasma membrane and secreted proteins, by their degradation by IRE-1 (35). Here, we show that suppression of the immune genes encoding secreted proteins is independent of IRE-1. Our results also provide a mechanistic explanation for the adverse effects that high ER stress has on immunity. In future studies, it will be important to characterize the IRE-1-independent pathway(s) that may sense increased ER stress and elicit the suppression of genes encoding secreted proteins.

MATERIALS AND METHODS
Bacterial strains. The following bacterial strains were used: Escherichia coli OP50, E. coli HT115(DE3), Pseudomonas aeruginosa PA14, and Salmonella enterica serovar Typhimurium 1344. The cultures for these bacteria were grown in Luria-Bertani (LB) broth at 37°C.
C. elegans strains and growth conditions. The C. elegans parental strain SJ4005 (Phsp-4::gfp) and wild-type Bristol N2 strains were obtained from the Caenorhabditis Genetics Center (University of Minnesota, Minneapolis, MN). The C. elegans strains were cultured under standard conditions and fed E. coli OP50.
Forward genetic screen for UPR upregulation. Ethyl methanesulfonate (EMS) mutagenesis was performed using the SJ4005 strain, which expresses green fluorescent protein (GFP) under the promoter of the hsp-4 gene. Approximately 3,000 synchronized late L4 larvae of SJ4005 were treated with 50 mM EMS for 4 h and then washed three times with M9 medium. The washed animals (P0 generation) were then transferred to four large petri dishes (15 cm) containing E. coli OP50 and allowed to lay eggs of F 1 progeny overnight. The P0 animals were then washed away with M9 medium, while the F 1 eggs remained attached to the bacterial lawn. The F 1 eggs were allowed to grow to adulthood. The animals that showed enhanced GFP expression in the F 1 progeny were selected on individual plates. Approximately 80,000 haploid genomes were mutated in the screen. All of the mutants were backcrossed six times with the parental SJ4005 strain before analysis.
Fluorescence imaging. Animals were anesthetized using an M9 salt solution containing 30 mM sodium azide and mounted onto 2% agar pads. The animals were then visualized using a Leica M165 FC fluorescence stereomicroscope.
RNA interference. RNA interference (RNAi) was used to generate loss-of-function RNAi phenotypes by feeding nematodes E. coli strain HT115(DE3) expressing double-stranded RNA (dsRNA) homologous to a target gene (48,49). Briefly, E. coli with the appropriate vectors were grown in LB broth containing ampicillin (100 g/ml) and tetracycline (12.5 g/ml) at 37°C overnight and plated onto NGM plates containing 100 g/ml ampicillin and 3 mM isopropyl-␤-D-thiogalactoside (IPTG) (RNAi plates). RNAiexpressing bacteria were allowed to grow overnight at 37°C. Gravid adults were transferred to RNAiexpressing bacterial lawns and allowed to lay eggs for 2 h. The gravid adults were removed, and the eggs were allowed to develop at 20°C to 1-day-old adults (96 h at 20°C from eggs) for subsequent assays. For age-dependent assays of N2 on vit-1 RNAi, the eggs were allowed to reach 1-day-old adult stage on vit-1 RNAi plates at 20°C (96 h at 20°C from eggs), 3-day-old adult stage (144 h at 20°C from eggs), and 5-day-old adult stage (192 h at 20°C from eggs) before transferring them to P. aeruginosa PA14 plates. unc-22 RNAi was included as a positive control in all experiments to account for the RNAi efficiency. RNAi clones for vit-2 and vit-6 are unavailable in the libraries of J. Ahringer and M. Vidal. The rescue of the enhanced UPR in vit-2(ac3) was tested by vit-1 RNAi thanks to cross-reactivity. The rescue of the enhanced UPR in vit-6(ac6) was not tested by RNAi due to the lack of cross-reactivity.
Whole-genome sequencing and data analysis. The mutant animals were grown at 20°C on NGM plates seeded with E. coli OP50 until starvation. The animals were rinsed off the plates with M9 medium, washed three times, incubated in M9 medium with rotation for 2 h to eliminate food from the intestine, and washed again three times with M9 medium. Genomic DNA extraction was performed using the Gentra Puregene kit (Qiagen, Netherlands). DNA libraries were prepared according to a standard Illumina (San Diego, CA) protocol. The DNA was subjected to whole-genome sequencing (WGS) on an Illumina HiSeq 4000 sequencing platform using 50 single-end nucleotide reads. Library preparation and WGS were performed at the Duke Center for Genomic and Computational Biology.
The whole-genome sequence data were analyzed using the EMS density mapping workflow from the CloudMap program of the Galaxy web platform (50). The list of single nucleotide polymorphisms (SNPs) was generated by comparison to the reference genome WS220. Common SNPs were subtracted, and linkage maps were generated.