Transcriptional induction of immune genes needs to be carefully controlled to promote defense against pathogen infection, without causing collateral damage to the host. In the nematode C. elegans , infection with natural pathogens of the intestine, including species of microsporidia (intracellular fungi) and a single-stranded RNA virus known as the Orsay virus , activates a transcriptional innate immune program called the intracellular pathogen response (IPR)  (Bakowski et al., 2014; Castiglioni et al., 2024; Chen et al., 2017; Reddy et al., 2019; Sarkies et al., 2013). The IPR promotes defense against infection, but IPR overactivation can slow organismal development and shorten lifespan (Reddy et al., 2019). ZIP-1 is a bZIP transcription factor that controls induction of about 1/3 of IPR genes, and promotes defense against viral infection (Lazetic et., 2022). How ZIP-1 protein expression is regulated and where ZIP-1 acts to control immunity have not been carefully explored.

 

Our previous data indicate that ZIP-1 can be expressed in the intestine or the epidermis, depending on the trigger. Specifically, an endogenously tagged ZIP-1 ::GFP protein (hereafter referred to as zip-1 p:: ZIP-1 ::GFP) is not visible under baseline conditions, but upon infection with microsporidia or the Orsay virus becomes visible in intestinal nuclei (Lazetic et al., 2022). zip-1 p:: ZIP-1 ::GFP is also visible in intestinal nuclei after 4 hours (h) of treatment with the proteasome inhibitor bortezomib, which induces mRNA expression of IPR genes, as well as zip-1 mRNA expression. In contrast, zip-1 p:: ZIP-1 ::GFP is expressed prominently in epidermal nuclei when the IPR is genetically activated by loss of the IPR inhibitor PALS-22 (Gang et al., 2022). To better understand the roles ZIP-1 has in the intestine and the epidermis, we developed tissue-specific ZIP-1 expression strains. We generated single-copy transgenic strains with ZIP-1 ::GFP expression under the control of the vha-6p intestinal-specific promoter or the dpy-7p epidermal-specific promoter. ZIP-1 expression in these strains was compared against zip-1 p:: ZIP-1 ::GFP ( Figure 1A ). Given that zip-1 mRNA is also upregulated by IPR triggers, these tissue-specific ZIP-1 strains also allowed us to investigate whether ZIP-1 protein levels might be induced by IPR triggers independently of the zip-1 promoter.

 

Similar to zip-1 p :: ZIP-1 ::GFP, we observed no expression of vha-6p :: ZIP-1 ::GFP or dpy-7p :: ZIP-1 ::GFP under baseline conditions ( Figure 1B ). However, when we treated animals with bortezomib for 4 h, we saw vha-6p :: ZIP-1 ::GFP expression in intestinal nuclei, and dpy-7p :: ZIP-1 ::GFP expression in epidermal nuclei after bortezomib treatment ( Figure 1B ). The bortezomib-induced expression of nuclear ZIP-1 ::GFP in the tissue-specific strains is consistent with our observations using endogenously tagged protein ( Figure 1B ), which was previously reported (Lazetic et al., 2022). Thus, the intestinal-specific and epidermal-specific promoters drive expression in the expected tissues, and these results indicate that ZIP-1 ::GFP protein expression can be induced by bortezomib independently of the zip-1 promoter. Under the control of constitutively active tissue-specific promoters, absence of ZIP-1 protein expression at baseline suggests that ZIP-1 protein is basally degraded. This degradation is mediated either directly or indirectly by the proteasome, as proteasomal blockade results in stabilization and nuclear expression of ZIP-1 .

 

We next examined ZIP-1 ::GFP expression with these transgenes in a zip-1 (jy13) null mutant background. Similar to the wild-type background, vha-6p :: ZIP-1 ::GFP was induced in intestinal nuclei upon 4 h bortezomib treatment in a zip-1 mutant background ( Figure 1C ). In contrast, dpy-7p :: ZIP-1 ::GFP was not induced upon 4 h bortezomib treatment in a zip-1 (jy13) mutant background ( Figure 1C ), unlike in the wild-type background. Given that endogenous zip-1 p :: ZIP-1 ::GFP is induced in the intestine by bortezomib, we hypothesized that bortezomib first affects the intestine, where wild-type intestinal ZIP-1 might be required to signal to the epidermis to induce epidermal ZIP-1 ::GFP expression. To investigate this possibility, we performed a time-course analysis to assess where and when ZIP-1 ::GFP is expressed after bortezomib treatment ( Figure 1D ). Here we found that zip-1 p :: ZIP-1 ::GFP is induced first in intestinal nuclei, and then later in epidermal nuclei. Intestinal nuclear expression is observed in roughly 50% of animals after 2 h of treatment. Similar levels of epidermal nuclear expression are only observed after 6 h of bortezomib treatment. Furthermore, all cases of epidermal nuclear expression were observed in animals that also had intestinal ZIP-1 ::GFP nuclear localization. Epidermal nuclear expression alone was not observed. This result is consistent with the model that intestinal ZIP-1 is required to send a signal to the epidermis to induce ZIP-1 ::GFP, although the specific tissue in which ZIP-1 is required for this signaling has not yet been investigated.

 

We next tested whether these tissue-specific ZIP-1 ::GFP expression constructs could rescue mRNA expression of the IPR gene pals-5 in zip-1 (jy13) mutants. zip-1 (jy13) mutants are defective in inducing pals-5 mRNA after 30 minutes of bortezomib treatment. Previous results using qRT-PCR with tissue-specific RNAi suggested that ZIP-1 was required in the intestine, but not the epidermis, for this effect (Lazetic et al., 2022).  Here we found that intestinal-specific vha-6p :: ZIP-1 ::GFP expression was sufficient to rescue the pals-5 gene induction defect of zip-1 (jy13) mutants ( Figure 1E ). Furthermore, intestinal-specific rescue of zip-1 resulted in significantly increased expression of another IPR gene, skr-5 , over wild-type and over zip-1 (jy13) mutant levels. Surprisingly, epidermal-specific dpy-7p :: ZIP-1 ::GFP expression appeared sufficient to rescue pals-5 induction, although the effect was not significant ( Figure 1E ). Thus, epidermal expression of ZIP-1 may be sufficient, but not required for inducing pals-5 mRNA expression. Of note, previous work showed a partial, non-significant reduction in pals-5 induction with epidermal-specific zip-1 RNAi (Lazetic et al., 2022).

 

Finally, we examined where ZIP-1 expression is sufficient to rescue the viral susceptibility of zip-1 (jy13) mutants. We infected animals with the Orsay virus , and then measured viral load with qRT-PCR for the Orsay virus genomic RNA1 segment. We infected zip-1 p:: zip-1 ::gfp animals, zip-1 (jy13) mutants, and zip-1 (jy13) mutants rescued with vha-6p:: zip-1 ::gfp or dpy-7p:: zip-1 ::gfp . Upon normalizing all data to zip-1 p:: zip-1 ::gfp animals, we found that intestinal-specific vha-6p :: ZIP-1 ::GFP expression was sufficient to rescue zip-1 (jy13) mutant susceptibility to viral infection. In contrast, epidermal-specific expression of ZIP-1 ::GFP was not sufficient to affect resistance to viral infection in a zip-1 (jy13) mutant background ( Figure 1F ).

 

In this work, we generated tissue-specific zip-1 expression and rescue strains to examine the regulation of protein expression and the tissue-specific effects of ZIP-1 . We demonstrated that the regulation of ZIP-1 occurs in a promoter-independent manner, suggesting that ZIP-1 is being degraded under baseline conditions. We also observed that epidermal nuclear expression of ZIP-1 is dependent on the presence of ZIP-1 in other tissues. In assessing the tissue-specific effects of ZIP-1 , we demonstrated that both intestinal and epidermal zip-1 appear sufficient to induce mRNA expression of IPR gene pals-5 , whereas only intestinal zip-1 rescued viral susceptibility in zip-1 (jy13) mutants. However, visible expression and subsequent upregulation of ZIP-1 ::GFP in a particular tissue is not necessary for IPR activation, as dpy-7p:: ZIP-1 ::GFP in a zip-1 (jy13) background displays a similar IPR activation profile following 30 minutes of bortezomib treatment to wild-type strains, despite not being visible later at 4 h. These findings support earlier findings that ZIP-1 has a functional role even when ZIP-1 ::GFP is not visible, and highlight the distinct roles of ZIP-1 at early and late stages of the IPR (Lazetic et al., 2022). Future studies could investigate where ZIP-1 is degraded basally – is it in the nucleus or the cytoplasm? What is the relationship between nuclear accumulation of ZIP-1 protein and IPR activation across tissues? Furthermore, we demonstrated that intestinal expression of zip-1 is sufficient for defense against the Orsay virus , an intestinal pathogen. Future studies could examine defense against other intestinal pathogens like microsporidia, as well as defense against other types of pathogens, including those that infect the epidermis. 

C. elegans maintenance

C. elegans were maintained at 20°C on Escherichia coli OP50-1 plated onto Nematode Growth Media (NGM) agar plates. Strains in Table 1.

 

C. elegans strain generation

Plasmids containing a tissue-specific promoter driving the expression of optimized zip-1 a with syntrons (Table 1) were used to carry out Mos1 -mediated single copy insertion by InVivo Biosystems. Strains were backcrossed to N2 wild-type worms in the Troemel lab three times before use. Genotyping primers in Table 2.

 

Bortezomib treatment

Bortezomib (Selleck Chemicals), catalog number S1013 was dissolved in DMSO to generate a 10 mM stock solution, which was diluted in M9 buffer, and added to NGM plates to achieve a final concentration of 20 μM on the plates. Synchronized L1s were grown on OP50-1 until L4, then treated with bortezomib or DMSO control. Plates were dried for 20 minutes, then incubated at 20°C for 30 minutes to 6 hours. Animals were either imaged immediately, or washed off plates with M9, then stored in Tri-reagent for RNA extraction and qRT-PCR.

 

RNA extraction and qRT-PCR

Total RNA was isolated as previously described (Lazetic et al., 2022), then used for cDNA synthesis via the iScript cDNA kit (Bio-Rad). At least 3 independent biological replicates per group were performed in each qRT-PCR analysis using Bio-Rad CFX Connect machine with Bio-Rad CFX manager 3.1. Sequences for qRT-PCR primers are provided in Table 2. All qRT-PCR data was normalized against housekeeping gene snb-1 , using the Pfaffl method (Pfaffl, 2001). For IPR activation assays, groups were normalized against a DMSO treated, N2 wild-type control. For Orsay virus infection, groups were normalized against an infected, zip-1 p:: zip-1 ::gfp control.

 

Orsay virus infection

Gravid adults were bleached to obtain synchronized L1 animals, which were grown on OP50-1 until L4 and then infected with a mixture of Orsay Virus, M9 buffer, and OP50-1 culture. After 24 hours at 20°C, animals were washed off plates, and RNA was extracted for cDNA synthesis and qRT-PCR.

 

Imaging

Imaging in Figure 1B was performed on a Zeiss AxioImager M1 compound microscope using ZEN Version 3.9.3. Imaging in Figure 1C was performed on a Zeiss LSM700 confocal microscope using ZEN 2010 Version 6.0.0.309. Image processing was done in Paint.NET Version 5.1.12.

 

Statistical analysis

Prism 10 Version 10.5.0 (774) was used for statistical analysis.