The bZIP Transcription Factor ZIP-11 Is Required for the Innate Immune Regulation in Caenorhabditis elegans

Innate immunity is the first line of host defense against pathogen infection in metazoans. However, the molecular mechanisms of the complex immune regulatory network are not fully understood. Based on a transcriptome profiling of the nematode Caenorhabditis elegans, we found that a bZIP transcription factor ZIP-11 was up-regulated upon Pseudomonas aeruginosa PA14 infection. The tissue specific RNAi knock-down and rescue data revealed that ZIP-11 acts in intestine to promote host resistance against P. aeruginosa PA14 infection. We further showed that intestinal ZIP-11 regulates innate immune response through constituting a feedback loop with the conserved PMK-1/p38 mitogen-activated protein signaling pathway. Intriguingly, ZIP-11 interacts with a CCAAT/enhancer-binding protein, CEBP-2, to mediate the transcriptional response to P. aeruginosa PA14 infection independently of PMK-1/p38 pathway. In addition, human homolog ATF4 can functionally substitute for ZIP-11 in innate immune regulation of C. elegans. Our findings indicate that the ZIP-11/ATF4 genetic program activates local innate immune response through conserved PMK-1/p38 and CEBP-2/C/EBPγ immune signals in C. elegans, raising the possibility that a similar process may occur in other organisms.


INTRODUCTION
The innate immune system is an evolutionarily conserved defense mechanism against microbial infection, which elicits immediate defense and generates long-term adaptive immunity (1). Innate immunity can be divided into two distinct but overlapping components: constitutive and inducible immunity. Constitutive immunity includes basally expressed immune effectors that allow host to response to the invading pathogens instantaneously (2,3). The inducible immune response includes three stages in general (4). First, the invading pathogen is recognized by the host pattern recognition receptors (PRRs). Next, the infection signals are processed and delivered via the activation or suppression of concerned signaling pathways. Finally, the up or down-regulated immune effectors accelerate to eliminate the pathogen or alleviate its effect on the host.
The basic region leucine zippers (bZIPs) are evolutionarily conserved transcription factors in eukaryotic organisms (15). bZIPs were identified from eukaryotic genomes by a basic DNA binding region followed by a leucine-zipper coiled-coil motif, via which the bZIP proteins can form homodimers and heterodimers, further influences the DNA sites that can be bound (16). The bZIP proteins are involved in diverse biological processes such as development, reproduction, cell cycle, stress response, and immune regulation (17,18). Up to date, about 33 bZIP transcription factors have been identified in C. elegans (19), which are predominantly expressed in intestinal tissue and widely participate in the regulatory network of innate immunity. Among these immune-related bZIP transcription factors, some are directly activated or suppressed by the upstream kinases or phosphatase, such as SKN-1, CEBP-1 and ATF-7 (20)(21)(22), while some members were involved in innate immune regulation through the changes in expression levels, such as ZIP-2 and ATFS-1 (23,24).
In this study, we identified an evolutionarily conserved bZIP transcription factor ZIP-11/ATF4 as a novel innate immune modulator. Intestinal ZIP-11 promotes host resistance against pathogens through constituting a feedback loop with PMK-1/ p38 pathway. In parallel, ZIP-11 acts with a CCAAT/enhancerbinding protein, CEBP-2, to enhance resistance pathogen infection independently of PMK-1/p38 pathway. These ZIP-11-involved innate immune response pathways promote infection tolerance upon P. aeruginosa PA14 infection, which might be conserved across species and expands our knowledge of infectious disease tolerance.

P. aeruginosa PA14 Infection Induces the Expression of ZIP-11
To better understand the mechanisms of the host responses to pathogenic infection, we performed transcriptome RNAsequencing (RNA-seq) of wild-type (N2) C. elegans fed on E. coli OP50 or exposed to P. aeruginosa PA14. Differential expression analyses of triplicate samples demonstrated that 643 genes were up-regulated and 582 genes were down-regulated upon P. aeruginosa PA14 exposure as compared to the gene expression levels in C. elegans fed on E. coli OP50 ( Figure 1A; Tables S1, S2 in Supplementary Material). Subsequently, we compared our RNA-seq data with two previous studies (25,26).
The results show that about half of the differentially expressed genes in either of the previous studies were commonly regulated by P. aeruginosa infection as compared to that in this study ( Figure S1A in Supplementary Material), which give us a notion that despite the known immune molecules and pathways, further efforts are needed to explore the regulatory network involved in innate immunity. Considering the mounting evidence shows that bZIP transcription factors play crucial roles in innate immune regulation, here, we focused on the immune function of zip class genes. Out of 12 members of the zip family in C. elegans, zip-2, zip-10 and zip-11 were up-regulated upon P. aeruginosa PA14 infection ( Figure 1B; Table S3 in Supplementary Material) and those results were verified by quantitative real-time PCR (qRT-PCR) ( Figure 1C). ZIP-2 has been identified as a key mediator of infection-induced gene expression (27). Translational block caused by PA14-delivered ToxA can trigger an increase in protein level of ZIP-2 (23). Meanwhile, ZIP-10 acts downstream of the TGF-b signal pathway ligand DBL-1 and the SMAD transcription factor SMA-9 to regulate the body size and the male tail morphogenesis of C. elegans (28). Moreover, the exposure to pathogenic bacteria induces ZIP-10 expression (25), indicating ZIP-10 may also play an important role in innate immunity.
Since ZIP-11 has not been characterized yet, we focused on this bZIP transcription factor to investigate if and how it mediates the innate immunity of C. elegans. Using a tagged zip-11p::zip-11::GFP::FLAG allele in an integrated transgenic strain, we observed that the number of intestinal nuclei with visible ZIP-11::EGFP was significantly increased in P. aeruginosa PA14 exposed worms than that in E. coli OP50 fed worms ( Figure 1D). Likely because of the low baseline expression level of ZIP-11, the EGFP fluorescence signal of the normal maintained C. elegans was barely visible in intestinal tissue. Moreover, the western blot results also showed that P. aeruginosa PA14 infection dramatically upregulated the expression level of ZIP-11 protein ( Figures 1E, F). These results demonstrate that pathogen infection stimulates ZIP-11 expression which assist our presumption of possible role of ZIP-11 in innate immune response.

ZIP-11 Functions in Intestine to Promote Innate Immunity Against P. aeruginosa PA14 Infection
To test if ZIP-11 acts in innate immune response, we first examined the survival of ZIP-11 deficient worms upon P. aeruginosa PA14 infection. We found that zip-11(tm4554) null mutant worms exhibited reduced resistance to killing by P. aeruginosa PA14 as compared to wild-type ( Figure 2A). Furthermore, the sensitive phenotype to P. aeruginosa PA14 infection was also observed in worms treated with zip-11 RNA interference (RNAi) ( Figure 2B). These data suggest that the loss function or the suppression of ZIP-11 signaling impairs the innate immunity of C. elegans. Certain mutants susceptible to pathogen infection exhibit sensitivity to pathogen accumulation, while others exhibit reduced endurance to pathogen (11). Here, the numbers of bacterial cells in the intestine of zip-11 mutants were comparable to those in WT worms ( Figure S1B in Supplementary Material). Moreover, compared to WT worms, zip-11(tm4554) worms exhibited a similar accumulation pattern of P. aeruginosa/green fluorescent protein (GFP) ( Figure S1C in Supplementary Material). To ask whether the feeding or defecation behavior in zip-11(tm4554) animals may account for the immune phenotype, the pumping rate and enteric muscle contraction (EMC) cycle in the intestine were detected. We observed that there was no difference in these behaviors between zip-11(tm4554) and WT worms (Figures S1D, S1E in Supplementary Material). In addition, zip-11(tm4554) worms showed the equivalent brood size as that of WT worms ( Figure  S1F in Supplementary Material), suggesting that the observed sensitivity to P. aeruginosa infection in zip-11(tm4554) worms is not caused by the confounding effects varying on worm fecundity. There was no difference in survival rate between zip-11(tm4554) and WT worms that were fed with heat-killed P. aeruginosa ( Figure S1G in Supplementary Material). Thus, ZIP-11 activates immune response to promote the endurance to living pathogen bacteria without altering the basic health of C. elegans.
Although ZIP-11 expression was only visible in the intestinal nuclei of worms exposed to P. aeruginosa using the translational reporter zip-11p::zip-11::GFP, we generated the zip-11 transcriptional reporter zip-11p::zip-11::SL2::GFP to monitor the expression pattern of zip-11 in C. elegans. Intriguingly, the GFP signal was detectable not only in intestine but also in pharynx and hypodermis tissue ( Figure 2C; Figure S2 in Supplementary Material), indicating that ZIP-11 may play a regulatory role in multiple biological processes. In order to determine the site of action of ZIP-11 and clarify the tissue specificity of ZIP-11 in terms of immune regulation to P. aeruginosa infection, the killing assay of qRT-PCR analysis of zip-2, zip-10 and zip-11 gene expression in WT worms exposed to E. coli OP50 or P. aeruginosa PA14. (D) Representative images of the zip-11p::zip-11::GFP worms exposed to E. coli OP50 or P. aeruginosa PA14. Right panels are higher magnification. Mean percentages of ZIP-11::GFP nuclear accumulation in the anterior portion of intestinal tissue (white arrows) are indicated. n ≥ 14. Scale bar: 20 mm. (E) Western blots of the zip-11p::zip-11::GFP worms exposed to E. coli OP50 or P. aeruginosa PA14 by using anti-GFP antibody. (F) Quantification of the protein expression levels using the software program ImageJ. Data are presented as mean ± SEM. Statistical significance was determined by Student's t-test (C, F). **p < 0.01, ***p < 0.001 compared with respective controls unless specifically indicated.
worms treated with zip-11 RNAi restricted to specific tissue was performed. The specificity of tissue-specific RNAi worm strains was confirmed by tissue-specific RNAi clones, which acted only in the corresponding tissues and demonstrated specific phenotypes (29). With tissue-specific RNAi worms, only the knockdown of zip-11 in intestine reduced the resistance to P. aeruginosa infection ( Figure 2H), of which the phenotype was in accordance with that observed in zip-11 loss-function and whole-body knockdown worms. However, suppressing zip-11 expression in the neuron ( Figure 2D), or hypodermis ( Figure 2E), or muscle ( Figure 2F), or germline ( Figure 2G) did not alter the survival rate of worms upon P. aeruginosa infection. Additionally, we found that overexpression of ZIP-11 under the control of endogenous zip-11 or intestinal specific ges-1 promoter in zip-11 mutant background converted the sensitive phenotype to the resistant state ( Figure 2I). Taken together, these results support a role of ZIP-11 in activating a protective transcriptional response in the intestinal tissue of C. elegans to P. aeruginosa infection.
We further used survival rate upon P. aeruginosa infection to investigate the genetic relationship between ZIP-11 and the major innate immune pathways. As expected, the attenuated resistance to P. aeruginosa-mediated killing of zip-11(tm4554) worms was suppressed by MAPK pmk-1 RNAi ( Figure 3E :GFP worms exposed to E. coli OP50 or P. aeruginosa PA14 after being fed with vector control or pmk-1 RNAi bacteria. Mean percentages of ZIP-11::GFP nuclear accumulation in the anterior portion of intestinal tissue (white arrows) are indicated. n ≥ 15. Scale bar: 20 mm. (C) Western blots of zip-11p::zip-11::GFP worms exposed to E. coli OP50 or P. aeruginosa PA14 after being fed with vector control or pmk-1 RNAi bacteria. (D) Quantification of the protein expression levels using the software program ImageJ. (E) WT worms and zip-11(tm4554) worms fed with vector control, pmk-1 RNAi bacteria were exposed to P. aeruginosa PA14 and scored for survival. (F) Western blot analysis of p38 phosphorylation levels in WT and zip-11(tm4554) worms. (G) Quantification of the protein expression levels using the software program ImageJ.
(H) qRT-PCR analysis of the expression of PMK-1-dependent genes in WT and zip-11(tm4554) worms. (I) Representative images and quantification of T24B8.5p:: GFP in WT and zip-11(tm4554) mutant background exposed to E. coli OP50 or P. aeruginosa PA14. The number of animals analysed is indicated. Scale bar: 200 mm. (J) Representative images and quantification of F08G5.6p::GFP in WT and zip-11(tm4554) mutant background exposed to E. coli OP50 or P. aeruginosa PA14. The number of animals analysed is indicated. Scale bar: 200 mm. Data are presented as mean ± SEM. Statistical significance was determined by log-rank test (E) or Student's t-test (A, D, G-J). *p < 0.05, **p < 0.01, ***p < 0.001 compared with respective controls unless specifically indicated; n.s., not significant. Please see Table S4 in Supplementary Material for detailed statistical analysis of killing assay data.  Figures 3F, G). Furthermore, seven PMK-1 pathway dependent genes, lys-1, F35E12.5, T24B8.5, clec-85, K08D8.5, F08G5.6 and M02F4.7, were down-regulated in zip-11(tm4554) worms relative to that in WT worms ( Figure 3H). To confirm the in vivo role of ZIP-11 in the regulation of PMK-1 pathway, a transgenic transcriptional reporter strain of PMK-1 activity that expresses GFP under the control of T24B8.5 promoter was introduced (22). We observed that the T24B8.5p::GFP signal in zip-11 mutation or RNAi worms was lower than that in control worms when fed on E. coli or exposed to P. aeruginosa ( Figure 3I; Figures  Here, we reveal that PMK-1 signal can positively control ZIP-11 expression, indicating ZIP-11 lies downstream of PMK-1 pathway. On the other hand, ZIP-11 acts upstream of PMK-1 pathway in innate immune activation. These findings give us a notion that there may be a potential feedback regulatory loop between ZIP-11 and PMK-1 pathway.
ZIP-11 Acts Together With CEBP-2 to Promote Immune Response Against P. aeruginosa PA14 Infection As bZIP proteins commonly act as dimers, we attempted to identify the interaction partner that interacts with ZIP-11 in C. elegans immune response. A previous study about bZIP transcription factor protein-protein interaction network in vitro identified C. elegans CEBP-2 (CCAAT/enhancer-binding protein 2) as the highest-affinity binding partner for ZIP-11 (31). Here, the coimmunoprecipitation (Co-IP) experiments was performed to investigate the binding relationship between ZIP-11 and CEBP-2. When we co-expressed tagged ZIP-11 and CEBP-2 in 293T cells, we were able to coimmunoprecipitate the two proteins as a complex ( Figure 4A), suggesting that ZIP-11 and CEBP-2 can physically interact indeed. This interactive relationship was further revealed by GST pull-down assay ( Figure S4A in Supplementary Material). In addition, the decrease in survival upon P. aeruginosa infection of zip-11 (tm4554) worms was abolished by cebp-2 RNAi ( Figure 4B). Meanwhile, RNAi-mediated zip-11 knockdown can't further weaken the survival of cebp-2(tm5421) worms upon P. aeruginosa infection ( Figure S4B in Supplementary Material), indicating that ZIP-11 and CEBP-2 may act in the same genetic pathway in immune regulation. However, cebp-2-defect worms exhibited decreased survival compared to zip-11 mutants ( Figure 4B), which may attribute to CEBP-2 can interact with various immune-related transcription factors, such as ZIP-2 (32), and plays a pivotal role in innate immune response. Although ZIP-11 and CEBP-2 are expressed in diverse tissues and P. aeruginosa infection can't make change to CEBP-2, these two bZIP transcription factors exist noticeable co-localization in intestinal nucleus of worms upon P. aeruginosa infection ( Figure 4C; Figure S4C in Supplementary Material).
We next examined whether ZIP-11 regulates the expression of target genes known to be induced by CEBP-2 upon P. aeruginosa infection. Considering the previous study have shown that an infection response gene-1 (irg-1) was robustly induced upon P. aeruginosa infection which process is controlled by ZIP-2 and CEBP-2 (27,32), the reporter strain irg-1p::GFP was introduced to further investigate the relationship between ZIP-11 and CEBP-2 in vivo. Interestingly, upon P. aeruginosa infection, zip-11 mutation significantly reduced the induction of irg-1p::GFP in vector control worms, but not in cebp-2 RNAitreated worms ( Figures 4D, E), suggesting irg-1 induction was controlled by ZIP-11 and CEBP-2. To a certain extent, this result also supports the equivalent survival between cebp-2-and zip-11defect worms upon P. aeruginosa infection ( Figure 4B; Figure  S4B in Supplementary Material).
Our results showed that there is a connection between ZIP-11 and PMK-1/p38 pathway in innate immune response ( Figure 3). However, the induction of irg-1 upon P. aeruginosa infection was independent of the known innate immune pathways, including PMK-1/p38 pathway (27). To further investigate the relationship between CEBP-2 and PMK-1, cebp-2 and irg-1 mRNA levels were examined in WT and pmk-1(km25) worms by using qRT-PCR. We found that not only irg-1 expression, but also cebp-2 expression didn't rely on PMK-1 signal ( Figure S5A  Taken together, these results indicate that ZIP-11 and CEBP-2 function together to regulate immune genes in the context of pathogenic bacteria infection, which process is independent PMK-1/p38 pathway.

Human bZIP Transcription
Factor ATF4 Functionally Substitutes for C. elegans ZIP-11 Human activating transcription factor ATF4 (NP_001666.2), which acts as a master regulator and plays crucial role in the adaptation to stresses (33), is the top BLAST hit in the human protein database for C. elegans ZIP-11. ATF4 has a similar domain structure with ZIP-11 and most of the protein is composed of a bZIP domain. The BRLZ (basic region leucine zipper) domain of ZIP-11 shows a strong sequence similarity to that of the evolutionarily conserved basic leucine zipper transcriptional factor in eukaryotes ( Figure 5A). We next expressed the human ATF4 as a transgene in WT worms using the intestine-specific promoter, and found that the transgene enhanced worm's survival upon P. aeruginosa infection ( Figure 5B). These results suggest that human ATF4 may play conserved function in innate immune regulation across species.
Given that ZIP-11 activates innate immune response through conserved PMK-1/p38 and CEBP-2 immune signals. We next investigated whether ATF4 expression induces the key components of these two pathways in C. elegans. As expected, western blot analysis showed that the ATF4-expressed worms had higher levels of active PMK-1 compared with control worms (Figures 5C, D). Furthermore, ATF4 expression significantly enhanced the GFP levels of PMK-1-dependent transgene worms, T24B8.5p::GFP ( Figure 5E). In addition, human ATF4 also induced the expression of PMK-1-independent irg-1 upon P. aeruginosa infection ( Figure 5F). These results indicate that human ATF4 can activate PMK-1-and CEBP-2-dependent immune signals in C. elegans. WT worms and zip-11(tm4554) worms fed with vector control or cebp-2 RNAi bacteria were exposed to P. aeruginosa PA14 and scored for survival. (C) Representative images of the zip-11p::zip-11::GFP;cebp-2p::cebp-2::RFP worms exposed to E. coli OP50 or P. aeruginosa PA14. ZIP-11::GFP or CEBP-2::RFP nuclear accumulation in the anterior portion of intestinal tissue (white arrows) are indicated. Scale bar: 20 mm. (D) Representative images of irg-1p::GFP and zip-11(tm4554);irg-1p::GFP worms fed with vector control and cebp-2 RNAi bacteria were exposed to P. aeruginosa PA14. myo-2p::mcherry shows the red fluorescence of pharyngeal as a marker for the transgene. Scale bar: 200 mm. (E) Quantification of the GFP fluorescence intensity using the software program ImageJ. The number of animals analysed is indicated. Data are presented as mean ± SEM. Statistical significance was determined by log-rank test (B) or Student's t-test (E). ***p < 0.001 compared with respective controls unless specifically indicated; n.s., not significant. Please see Table S4 in Supplementary Material for detailed statistical analysis of killing assay data.   Table S4 in Supplementary Material for detailed statistical analysis of killing assay data. Combined, these data present convincing evidences supporting that human bZIP transcription factor ATF4 can functionally substitute for C. elegans ZIP-11 in innate immune regulation, raising the possibility that it may play a similar role in human.

DISCUSSION
When animals are exposed to microbial pathogens, the immune system will immediately activate a defense response through immune cells. Despite lacking professional immune cells, C. elegans can effectively protect itself from diverse invading pathogens, which largely relies on the innate immune response in epithelial cells. As a crucial immune organ, the epithelial cells can quickly recognize the pathogenic bacteria, inducing the antimicrobial peptides and secreted lysozymes to fight off the infection, which process is directly mediated by transcription factors. In the current study, we identified bZIP transcription factor ZIP-11 as a novel innate immune regulator ( Figure 5G). ZIP-11 can constitute a feedback loop with PMK-1/p38 pathway. On the other hand, ZIP-11 acts together with another bZIP transcription factor, CEBP-2, to induce immune genes expression upon P. aeruginosa infection, which process is independent of PMK-1/p38 pathway. These findings reveal that ZIP-11 plays multiple functions in the complex immune regulatory network. Moreover, the overexpression of human ATF4 in worm's intestine enhances the immune capacity of WT worms, promotes PMK-1 phosphorylation level, and induces the expression of PMK-1-or CEBP-2-dependent immune genes. Based on these findings, we proposed that the main function we identified for ZIP-11 in C. elegans may be evolutionarily conserved.
As one of the 12 members of the zip family in C. elegans, ZIP-11 has not been characterized yet. Despite the expression pattern of ZIP-11 was clarified by using the transcriptional worm zip-11p::zip-11::SL2::GFP that the GFP signal was visible in pharynx, hypodermis, and intestine ( Figure S2 in Supplementary Material), zip-11 mRNA was detected in the PVD neuron microarray profile (34). However, the tissue-specific RNAi and rescue assay indicate that only intestinal ZIP-11 is involved in the innate immune response (Figure 2). Since several transcriptome analyses showed ZIP-11 expression changed in response to multiple stress environments such as resveratrol treatment (35), rotenone treatment (36), or S. marcescens infection (37), ZIP-11 may modulate the global transcriptional regulation of stress response in C. elegans.
The p38 MAPK pathway is evolutionarily conserved in the control of cellular responses to stress and inflammation, and keeping it in balance seems critical to immune homeostasis (38). Previous studies have demonstrated that PMK-1/p38 pathway can receive various upstream signal to drive the "fight-or-flight" response and local immune response in C. elegans intestinal cells.
In current study, we show that ZIP-11 acts as an immune activator by constituting a feedback loop with PMK-1/p38 in cell-autonomous manner. Though most immune responserelated transcription factors are considered to directly modulate the expression levels of very downstream antimicrobial peptides and secreted lysozymes, which process is controlled by kinase signal. However, we didn't clarify whether the PMK-1-dependent immune genes were directly controlled by ZIP-11. In addition, while SKN-1 and ATF-7 are two important transcription factors in the intestinal immune system of C. elegans that directly activated by NSY-1-SEK-1-PMK-1 MAPK signaling (20,22), they may directly modulate ZIP-11 expression. Overall, the exact mechanism through which ZIP-11 activates PMK-1/p38 pathway and how PMK-1/p38 pathway promotes ZIP-11 expression remain unclear and require further study.
Although a series of molecules have been identified to be involved in regulating innate immune response in C. elegans, it is rarely investigated whether their functions are conserved in other organism. In human immune response, ATF4 activity is induced by the exposure of immune cells in multiple stressed environments including pathogen invasion, infection, inflammation, or tumor microenvironment (33). The regulation of immunity via ATF4 can be indirect by initiating inflammation that will secrete a series of cytokine directed immune responses. In contrast, ATF4 can also directly interfere with maturation, development, or polarization states of different immune cells such as macrophages, T cell, B cells, NK cells and dendritic cells contributing to progression of disease (39). Interestingly, ATF4 can act together with ATF6 to transactivate the transcription factor C/EBP homologous protein (CHOP), which was found to regulate mediators of apoptosis such as B-cell lymphoma 2 (BCL2) and BCL2 interacting mediator of cell death (40,41). Moreover, a previous study has shown that C/EBPg, the human homolog of C. elegans CEBP-2, can dimerize with ATF4 to control redox homeostasis in normal and cancerous cells (42), suggesting the binding relationship between ZIP-11 and CEBP-2 is conserved. Overall, these excellent work further supports our notion that the immune regulation mechanisms of ZIP-11 we revealed in C. elegans might be applied in human. Future efforts are needed to explore the role of ATF4 in immune response of human, which may provide us a potential therapeutic target for the infectious diseases.
To construct plasmid pXT07 for zip-11 RNAi, the 765 bp genomic sequence of zip-11 (from the start codon to the end of the second exon) was subcloned into the control vector L4440 between NotI and HindIII using isothermal assembly.

C. elegans Slow-Killing Assay
The slow-killing assay was performed as described previously (5) with minor modifications. The overnight culture of P. aeruginosa PA14 was seeded on modified NGM (0.35% instead of 0.25% peptone) in 35 mm-diameter plates. Plates were allowed to dry at room temperature and incubated at 37°C for 24 h and at 25°C for 8-24 h. Synchronized L4 worms were transferred to the P. aeruginosa PA14 slow killing plates and cultivated at 25°C. Animals were scored at the indicated times for survival and transferred to fresh slow killing plates daily. Animals were considered dead if they failed to respond when touched. 5-Fluoro-2'-deoxyuridine (FUdR, 50 mg/ml) (Sigma, F0503) was included in the slow killing assay to prevent the growth of progeny and "bagging". The worms died due to vulva burst or crawling off the plates were censored. Full-lawn P. aeruginosa PA14 plates were prepared for infection. All survival data were performed at least two replicates.

P. aeruginosa PA14 CFU Assay
The worms were exposed to P. aeruginosa PA14 as described above. After infection for 24 h, worms were transferred to M9 solution containing 25 mM sodium azide (Amresco, 0639) to paralyze the worms and stop the pharyngeal pumping. Worms were washed three times with an antibiotic M9 solution containing 1 mg/ml ampicillin and 1 mg/ml gentamicin (Solarbio, G8170, China) followed by 30 min of incubation in the antibiotic solution to kill the bacteria present on the exterior of the worms. After transfer onto the new NGM plates containing 1 mg/ml ampicillin and 1 mg/ml gentamicin for another 30 min to further eliminate the external P. aeruginosa PA14, the worms were lysed with a motorized pestle. Lysates were serially diluted with M9 buffer and plated on Luria-Bertani plates containing 100 mg/ml rifampicin (Solarbio, R8011, China) to select for P. aeruginosa PA14. After overnight incubation at 37°C, colonies of P. aeruginosa PA14 were counted to determine CFU per worm. Ten worms were examined per treatment, and at least three replicate assays were performed.

Behavioral Analysis
Day 1 adult animals were used for the behavioral assays at room temperature. For the brood size assay, a single L4 worm was transferred to an individual plate. The worms were transferred to fresh plates every other day until laying ceased, and the offspring were subsequently scored for the brood size. For pumping rate assay, the pharyngeal pumping was determined directly under a digital automated microscope (Nikon, AZ100, Japan). For the defecation assay, the enteric muscle contractions (EMC) were scored for each defecation cycle using a dissecting microscope.

RNA Sequencing
Total RNA was obtained as described above. 1 mg total RNA from each sample was used as input material for the RNA sample preparations. Sequencing libraries were generated using NEBNext ® Ultra ™ RNA Library Prep Kit for Illumina ® (NEB, USA) following manufacturer's recommendations and index codes were added to attribute sequences to each sample. After cluster generation on a cBot Cluster Generation System using TruSeq PE Cluster Kit v3-cBot-HS (Illumina), the library preparations were sequenced on an Illumina Hiseq platform and 125 bp/150 bp paired-end reads were generated.
The clean reads were mapped to the C. elegans genome sequence using Hisat2 v2.0.4. HTSeq v0.9.1 was used to count the reads numbers mapped to each gene. And then FPKM of each gene was calculated based on the length of the gene and reads count mapped to this gene. Differential expression analysis of two groups (three biological replicates per condition) was performed using the DESeq R package (1.18.0). The resulting pvalues were adjusted using the Benjamini and Hochberg's approach for controlling the false discovery rate. Genes with an adjusted p-value < 0.05 found by DESeq were assigned as differentially expressed.

RNA Interference
RNAi was performed by feeding worms with E. coli strain HT115 (DE3) expressing double-stranded RNA homologous of target genes as described (45). HT115 (DE3) was grown overnight in LB broth containing 100 mg/ml ampicillin at 37°C and plated over the NGM plates containing 100 mg/ml ampicillin and 2mM isopropyl 1-thio-b-D-galactopyranoside (IPTG) (Sangon Biotech, A600168). RNAi-expressing plates were allowed to grow for approximately 24 h at room temperature. The L2-L3 larval worms of the corresponding strain were placed on the RNAi plates for 2 days at 20°C. The gravid adults were then transferred onto the fresh RNAi-expressing bacterial lawns and allowed to lay eggs for 2 h to obtain the second generation of RNAi population. The eggs were allowed to develop at 20°C to reach the L4 stage for subsequent experimental use. In all experiments, uaf-1 RNAi was included as a positive control to account for RNAi efficiency.
Cell transfection and coimmunoprecipitation (Co-IP) were performed as previously described (46). Briefly, 293T cells (CRL-11268, ATCC) were maintained in DMEM medium (Gibco, C11995500BT, USA) with 10% inactive bovine fetal serum (Gibco, 10270106, USA). The cells were transiently transfected with the plasmids expressing CEBP-2-Flag, ZIP-11-GFP, or GFP by using Lipofectamine 3000 transfection reagent (Invitrogen, L3000015, USA). The cells were harvested at around 48 hours post-transfection and washed two times to remove the cell debris (800 g, 3 min). The cells were lysed with ice pre-cooled immunoprecipitation buffer (IPB, 50 mM HEPES, pH 7.7, 100 mM NaCl, 50 mM KCl, 2 mM MgCl 2 , 2 mM CaCl 2 , 1 mM EDTA, 1 mg/ml pepstatin, 1 mg/ml leupeptin, 2 mg ml-1 aprotinin, 1 mM PMSF) containing 1% Triton X-100 on ice for 60 min, then centrifuged at 20000 g for 10 min at 4°C. Total protein content was estimated by PierceTM BCA Protein Assay Kit (Thermo-Scientific, 23227, USA) following the manufacturer's protocol. The protein was diluted with IPB (with 0.05% Triton X-100) to a final concentration of 1 ug/ul. The prewashed 20 ml of anti-GFP-Trap-A beads (Chromotek, gta-20, Germany) was added and incubated overnight at 4°C with gentle rotation. The anti-GFP-Trap-A beads were collected by centrifugation at 1000 g for 3 min at 4°C. The beads were washed six times with IPB containing 0.05% (w/v) Triton X-100. After last time wash, the beads were resuspended with 60 mL IPB buffer, the immunoprecipitated proteins were boiled in 3 × Sample buffer with beta-mercaptoethanol for 10 min at 98°C, and analyzed by western blotting. For the western blotting, the primary anti-GFP Rabbit polyclonal antibody (Invitrogen, A-11122, USA) at a 1:3000 dilution or anti-Flag mouse monoclonal antibody (MBL, M185-3L, USA) at a 1:12000 dilution was used, and horseradish peroxidase (HRP)-conjugated goat anti-rabbit (Transgen, HS101-01, China) or anti-mouse (Transgen, HS201-01,China) was used as secondary antibody at a 1:2000 or 1:5000 dilution. The images of blotting membranes were captured by an imaging system (MicroChemi 4.2, DNR Bio Imaging Systems). The coimmunoprecipitation experiment was repeated at least three times independently.
To confirm the interaction of CEBP-2 and ZIP-11, CEBP-2 was expressed as GST-fusion proteins in pGEX-4T-1, and the HEK293T cells were transfected with ZIP-11-GFP plasmid. The cells and bacteria were collected and extracted with working buffer (WB, 50 mM HEPES, pH7.4, 100 mM NaCl, 50 mM KCl, 2 mM CaCl 2 , 2 mM MgCl 2 , 1 mM EDTA, 1 mg/mL pepstatin, 1 mg/mL leupeptin, 2 mg/mL aprotinin, 1mM PMSF, 1 tablet of protease inhibitor cocktail per 10 mL buffer) containing 0.5% Triton X-100. Equal amounts of GST-fused CEBP-2 or GST protein was added to the cell lysate. Then the pre-washed 25 ml of GST-beads was added to the system, and incubated overnight at 4°C with gentle rotation. The GST-beads were collected by centrifugation at 700 g for 3 min at 4°C. The beads were washed six times with ice cold WB buffer containing 0.1% Triton X-100. After last time wash, the beads were resuspended by 70 mL WB buffer, the pull-downed proteins were boiled in 3 × Sample buffer with beta-mercaptoethanol for 10 min at 98°C, and analyzed by western blotting. For the western blotting, the primary anti-GFP Rabbit polyclonal antibody (Invitrogen, A-11122, USA) at a 1:3,000 dilution or anti-GST mouse monoclonal antibody (Transgen, HT601-01, China) at a 1:1,000 dilution was used, and horseradish peroxidase (HRP)-conjugated goat anti-rabbit (Transgen, HS101-01, China) or anti-mouse (Transgen, HS201-01,China) was used as secondary antibody at a 1:2000 or 1:5000 dilution. The images of blotting membranes were captured by the imaging system. The pull down experiment was repeated at least three times independently.

Statistical Analysis
Data were presented as mean ± SEM. Survival curves were created using GraphPad Prism 8 (GraphPad, CA). The logrank (Kaplan-Meier) test was used to analyze the survival data. For fluorescence quantification, the worms in the images were outlined and the signal intensity was calculated by the ImageJ software (NIH, USA). Other data were analyzed by using an unpaired Student's t-test. The difference was considered significant from control for p-value < 0.05. All experiments were performed at least three times independently.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material.