Immune regulators PALS-25 and PALS-22 localize to mitochondria and regulate mitochondrial fragmentation in C. elegans

TL;DR

This study shows that PALS-22 and PALS-25 proteins in C. elegans localize to mitochondria and regulate mitochondrial fragmentation to enhance immunity against intracellular pathogens.

Contribution

The study reveals the mitochondrial localization and mechanism of action of PALS-22 and PALS-25 in regulating immunity.

Findings

PALS-22 and PALS-25 proteins localize to mitochondria in C. elegans.

The C-terminus of PALS-25 is responsible for mitochondrial localization.

Loss of PALS-22 causes mitochondrial fragmentation and increases resistance to microsporidia.

Abstract

The nematode C. elegans controls immunity against intracellular pathogens such as microsporidia, using the pals gene family, which has expanded in C. elegans compared to mammals. pals-22 is a negative regulator that restrains pals-25, which serves as a positive regulator of immunity. pals-22 and pals-25 encode proteins that bind each other and can act in the intestine and epidermis, but their subcellular localization and mechanism of action have not been described. Here, we show that PALS-22 and PALS-25 proteins localize to mitochondria, with PALS-25 being required for PALS-22 localization to mitochondria. The C-terminus of PALS-25 directs mitochondrial localization, and the N-terminus is required for signaling. The loss of PALS-22 causes mitochondrial fragmentation, which occurs after activating the intracellular pathogen response (IPR), a transcriptional program induced by intracellular infection. Mitochondrial fragmentation induced by knockdown of fission/fusion factors increases resistance against microsporidia infection. Thus, the PALS-22/25-mediated fragmentation of mitochondria enhances resistance against intracellular infection.

•PALS-22 and PALS-25 localize to the outer membrane of mitochondria•PALS-25 C-terminal half directs mitochondrial localization•Loss of PALS-22 triggers mitochondrial fragmentation through IPR induction

PALS-22 and PALS-25 localize to the outer membrane of mitochondria

PALS-25 C-terminal half directs mitochondrial localization

Loss of PALS-22 triggers mitochondrial fragmentation through IPR induction

Cellular physiology; Immunity; Cell biology

Figures

PALS-22 and PALS-25 are mitochondrially associated proteins(A and B) PALS-22::GFP (A) and PALS-25::GFP (B) colocalize with the mitochondria marker MitoTracker Red with robust expression in seam cells. The epidermis of adult animals is shown. White triangle denotes seam cell mitochondria, white arrow denotes non-seam cell epidermal mitochondria.(C) PALS-25::GFP and TOMM-20::mCherry colocalize in epidermal cells in young adult animals. White triangle denotes seam cell mitochondria, white arrow denotes muscle.(D) Representative images of the epidermis of PALS-22::GFP (left) or PALS-25::GFP (right

PALS-22 mitochondrial localization depends on PALS-25(A) PALS-22::GFP expression and localization in the epidermis relative to mitochondria labeled with MitoTracker Red following treatment with control or *pals-25* RNAi. *pals-25* RNAi causes PALS-22::GFP to lose association with mitochondria in seam cells of the epidermis.(B) PALS-25::GFP expression and localization in the epidermis relative to mitochondria labeled with MitoTracker Red following treatment with control or *pals-22* RNAi. *pals-22* RNAi, which induces the IPR, causes PALS-25 to form puncta localized to mitochondria in epidermal

The C-terminal half of PALS-25 is required for mitochondrial localization, and N-terminal residues are required for signaling(A) Predicted structure of PALS-25 (AlphaFold DB [Q3V5H4](Q3V5H4)). Gray corresponds to the PALS-25 C-terminal 13 amino acids, which are deleted in PALS-25(GOF), causing PALS-22 dissociation and IPR induction (Gang et al.[15](#bib15)).(B) PALS-25(GOF)::mScarlet localization with the mitochondrially localized protein VWA-8::GFP.(C) PALS-25 (Δ40 GOF)::mScarlet localization with the mitochondrially localized protein VWA-8::GFP.(D) Quantification of *pals-5p::*GFP expression

IPR reporter induction precedes mitochondrial fragmentation upon the depletion of PALS-22(A) Time course images of transgenic animals expressing ubiquitous TIR1, endogenous *pals-22* tagged with AID, and the *pals-5p::*GFP, *myo-2p::*mCherry IPR reporter treated with either ethanol vehicle control (top) or auxin (bottom) starting at 40 h post-L1. Auxin-mediated depletion of PALS-22 induces *pals-5p*::GFP IPR reporter expression as early as ∼2 h after auxin treatment (white triangles) and in most tissues by 3 h. *myo-2p::*mCherry is a marker for the transgene expressed in the pharynx. Images ar

Animals with mitochondrial fragmentation independent of IPR induction have decreased *N. parisii* pathogen load(A) Representative TEM images of *N. parisii* sporonts (red double asterisk) associated with *C. elegans* mitochondria (purple asterisks) during infection. Scale bars, 500 nm, ROI scale bars, 200 nm.(B) Wild-type animals treated for two generations on *drp-1* or *fzo-1* RNAi display increased resistance to *N. parisii* infection. ∗∗*p* < 0.01, ∗∗∗∗*p* < 0.0001, one-way ANOVA with Dunnett’s multiple comparisons test. *n* = 209–279 animals per treatment across three experimental replica

Full text

Introduction

Microsporidia are ubiquitous fungal pathogens that infect almost all animal phyla and complete their entire replicative life cycle inside host cells, similar to viruses.1^,^2 In C. elegans, the transcriptional intracellular pathogen response (IPR) provides defense against the natural microsporidian species Nematocida parisii and the Orsay virus, a single-stranded RNA virus.3 Despite their molecularly distinct nature, both N. parisii and the Orsay virus trigger IPR gene expression. How C. elegans senses N. parisii is unknown, while C. elegans senses the Orsay virus via the RIG-I homolog DRH-1 to activate the IPR.4^,^5 Although C. elegans lacks classical interferons, this RIG-I-mediated activation of the IPR resembles the type-I interferon response (IFN-I response), a key anti-viral defense program in humans that also helps promote resistance against microsporidia infections.6^,^7

Many IPR genes belong to the pals gene family, characterized by a divergent “pals” sequence signature found in C. elegans, mice, and humans.8 Humans and mice each have one pals gene ortholog, while C. elegans has at least 39 pals genes, divided into two classes: a) 26 infection-induced, and b) 13 non-induced. Among the 26 induced genes is pals-14, which promotes resistance to microsporidia,9 and pals-5, which is commonly used as a reporter for IPR induction and can reduce levels of protein aggregates.10 Some non-induced pals genes are negative regulators of the IPR (pals-17 and pals-22), while some others are positive regulators (pals-25, pals-16, and pals-20).11^,^12 These IPR regulators act in “modules,” with pals-17 repressing pals-20/pals-16, and pals-22 repressing pals-25.

The best characterized IPR module is composed of pals-22 and pals-25, which act as an OFF/ON switch of the IPR. pals-22 mutants have constitutive IPR gene expression, increased resistance to intestinal intracellular pathogens such as microsporidia and viruses, as well as increased silencing of repetitive genomic elements. All of these phenotypes are reversed in pals-22 pals-25 double mutants.11 Notably, pals-22 and pals-25 have also been shown to regulate immunity against oomycetes, which are natural eukaryotic pathogens of the epidermis.13 pals-22 mutants have slowed development and shortened lifespan, which is reversed by mutations in pals-25, indicating that PALS-22 and PALS-25 control a physiological switch from a growth state to a defense state.11 PALS-22 and PALS-25 proteins bind to each other, and when this association is lost, PALS-25 can activate IPR genes and resistance to infection.14^,^15

PALS-22 and PALS-25 are expressed in the same tissues throughout C. elegans, with demonstrated activity in the intestine and epidermis for thermotolerance and pathogen resistance, respectively.11^,^15^,^16 While phenotypes, tissue expression, and protein-protein interactions of PALS-22 and PALS-25 have been described,11^,^14^,^15 their mechanism of action is still poorly understood. Here, we show that both proteins localize to the mitochondria. PALS-22 depends on PALS-25 for its mitochondrial association, while PALS-25 localizes to mitochondria independently of PALS-22. Upon loss of PALS-22, which triggers an IPR-activated state, PALS-25 protein forms puncta along the mitochondria. We find that the C-terminus of PALS-25 is necessary and sufficient for mitochondrial localization, and the N-terminal 40 amino acids of PALS-25 are required for its induction of the IPR reporter pals-5p::GFP. We also show that loss of PALS-22 leads to the fragmentation of mitochondria, with a loss of branch number and shorter branch length. Kinetic studies of PALS-22 depletion, together with the analysis of PALS-17 knockdown and the role of ZIP-1, suggest that IPR induction may trigger mitochondrial fragmentation. We also show that N. parisii associates with C. elegans mitochondria. Furthermore, when mitochondria are fragmented independent of IPR induction, there is increased resistance to N. parisii infection. Altogether, these studies indicate that PALS-22 and PALS-25 localize to mitochondria, and the induction of mitochondrial fragmentation helps promote resistance to microsporidia infection.

Results

PALS-22 and PALS-25 proteins localize to mitochondria

To investigate the subcellular localization of PALS-22 and PALS-25 proteins, we used TransgeneOme fosmids, which are a resource collection of specific C. elegans genes tagged at their C termini with GFP, surrounded by approximately 20 kb of endogenous genomic region (Figures S1A and S1B).17 Of note, pals-22 and pals-25 are found in an operon together (Figure S1C). As previously described for strains carrying these TransgeneOme constructs,15 PALS-22::GFP and PALS-25::GFP are expressed in many tissues in C. elegans, including the intestine and epidermis. Using these strains, we focused on their subcellular expression pattern in the epidermis, due to strong expression in that tissue, particularly in seam cells. Furthermore, they can act in the epidermis to regulate the IPR.15 Here, we observed that both PALS-22::GFP and PALS-25::GFP exhibit long, branched-like expression patterns characteristic of branched mitochondrial networks. Indeed, when animals expressing PALS-22::GFP or PALS-25::GFP were stained with MitoTracker Red, we saw nearly perfect co-localization of both PALS-22::GFP and PALS-25::GFP with MitoTracker Red staining in the epidermis (Figures 1A and 1B).Figure 1PALS-22 and PALS-25 are mitochondrially associated proteins(A and B) PALS-22::GFP (A) and PALS-25::GFP (B) colocalize with the mitochondria marker MitoTracker Red with robust expression in seam cells. The epidermis of adult animals is shown. White triangle denotes seam cell mitochondria, white arrow denotes non-seam cell epidermal mitochondria.(C) PALS-25::GFP and TOMM-20::mCherry colocalize in epidermal cells in young adult animals. White triangle denotes seam cell mitochondria, white arrow denotes muscle.(D) Representative images of the epidermis of PALS-22::GFP (left) or PALS-25::GFP (right) adult animals following treatment with control, fzo-1, eat-3, and drp-1 RNAi. For A–D, scale bars, 10 μm, region of interest (ROI) scale bars, 2.5 μm.

Imaging in the intestine can be confounded by autofluorescence, and PALS-25 expression was too low to assess its subcellular localization in this tissue. However, we did observe colocalization of PALS-22::GFP with MitoTracker Red in the intestine, similar to its localization in the epidermis (Figure S2A). Next, to analyze PALS-22 and PALS-25 co-localization in the same cells, we generated a transgene that has PALS-22 tagged with GFP and PALS-25 tagged with the red fluorescent protein mScarlet. Here we made a “mini-operon” of pals-22 and pals-25 cDNAs driven by 2 kb of pals-22 upstream region, with fluorescent tags fused to the N-termini of each gene (Figure S1D). Again, we saw the localization of both GFP::PALS-22 and mScarlet::PALS-25 to mitochondria-like structures in the epidermis, with the strong co-localization of these two proteins in seam cells (Figure S2B).

The data above indicate that PALS-22 and PALS-25 localize to mitochondria. Supporting this conclusion, there were several mitochondrial proteins that appeared physically associated with PALS-22 and PALS-25 in a previously published co-IP/MS study (Table S1). Furthermore, the imaging suggests PALS-22 and PALS-25 are at the outer mitochondrial membrane instead of directly localizing with MitoTracker staining in the mitochondrial matrix (Figures 1A, 1B, and S2A). Consistent with this model, we saw close co-localization of PALS-25::GFP with TOMM-20::mCherry, which marks the outer mitochondrial membrane (Figure 1C). We also performed RNAi against known fission and fusion factors that regulate mitochondria morphology.18 Consistent with PALS-22 and PALS-25 localizing to mitochondria, we found that RNAi against the fusion factors fzo-1 and eat-3 led to more condensed PALS-22::GFP and PALS-25::GFP localization patterns resembling fragmented mitochondria (Figure 1D). We also found that RNAi against drp-1, whose wild-type function promotes the fission of the outer mitochondrial membrane, led to longer network-like PALS-22::GFP and PALS-25::GFP localization patterns, typical of elongated mitochondria (Figure 1D). Further supporting the conclusion that PALS-22 and PALS-25 localize to the outer mitochondrial membrane is the observation that drp-1 RNAi caused these two proteins to adopt a more elongated localization pattern (Figure 1D). Notably, drp-1 is required for the fission of the outer mitochondrial membrane, but not the inner mitochondrial membrane, and loss of drp-1 will cause more elongated outer mitochondrial membranes, but a fragmented inner mitochondrial membrane and matrices (see later in discussion).19^,^20

To confirm that these tagged proteins are functional, we crossed each of the TransgeneOme constructs into a pals-22 pals-25(jy80) strain background, which is a clean deletion of the pals-22/pals-25 operon, and then analyzed IPR induction (Figure S1C). Here we found that strains carrying either PALS-22::GFP or PALS-25::GFP were capable of inducing IPR gene expression upon RNAi knock-down of pals-22 (Figures S3A and S3B). Because the disruption of the interaction between PALS-22 and PALS-25 can activate the IPR, we also analyzed basal IPR gene expression in these strains as a readout for whether the tags might disrupt their interaction. Here, we did not see a significant induction of IPR gene expression (Figure S3C). Therefore, the GFP tags do not appear to interfere with PALS-22 and PALS-25 normal interactions or with their ability to induce the IPR, and thus, we conclude that the PALS-22::GFP and PALS-25::GFP transgenes are functional.

PALS-22 mitochondrial localization depends on PALS-25, which uses its C-terminus for mitochondrial localization

To examine how the localization of PALS-22 and PALS-25 is regulated, we performed RNAi knockdown of pals-25 and examined PALS-22::GFP localization. Here, we found that PALS-22::GFP lost its mitochondrial association, instead showing a diffuse cytosolic expression pattern (Figures 2A and S4A). We assessed protein levels with Western blot analysis and found that there was no significant decrease in PALS-22::GFP overall protein levels upon pals-25 RNAi (Figure S4B). In contrast, we found that there was a significant decrease in PALS-25::GFP protein levels upon RNAi knock-down of pals-22, which activates the IPR (Figure S4C). pals-25 mRNA levels do not decrease (Figure S3B), thus this decrease in PALS-25 protein caused by pals-22 RNAi cannot be explained by a decrease in pals-25 mRNA. Furthermore, we found that PALS-25::GFP formed discrete puncta after pals-22 RNAi, with these puncta still apparently localizing to mitochondria under IPR ON conditions (Figures 2B and S4D). Therefore, PALS-22 depends on PALS-25 for its localization to mitochondria (Figure 2A). In contrast, PALS-25 does not require PALS-22 for its localization to mitochondria, but does require PALS-22 to maintain wild-type protein levels, and to maintain an even distribution pattern along these mitochondria (Figure 2B).Figure 2PALS-22 mitochondrial localization depends on PALS-25(A) PALS-22::GFP expression and localization in the epidermis relative to mitochondria labeled with MitoTracker Red following treatment with control or pals-25 RNAi. pals-25 RNAi causes PALS-22::GFP to lose association with mitochondria in seam cells of the epidermis.(B) PALS-25::GFP expression and localization in the epidermis relative to mitochondria labeled with MitoTracker Red following treatment with control or pals-22 RNAi. pals-22 RNAi, which induces the IPR, causes PALS-25 to form puncta localized to mitochondria in epidermal seam cells. For A and B, scale bars, 10 μm, ROI scale bars, 2.5 μm.

Given that PALS-25 appears to be the driver of PALS-22 localization to mitochondria, we next examined which domain of PALS-25 is responsible for mitochondrial localization. Our prior work determined that the last 13 amino acids at the C-terminus of PALS-25 are required for its association with PALS-22. This conclusion is based on the analysis of the PALS-25 Q293∗ mutant, hereafter referred to as GOF (Gain Of Function), which no longer associates with PALS-22, and thus has constitutive IPR gene expression (Figures 3A and S1C). Here, we examined whether PALS-25(GOF)::mScarlet still localizes to epidermal mitochondria through the co-localization of PALS-25(GOF) and VWA-8::GFP, which localizes to mitochondria.21 Because of mScarlet aggregation in this strain, it was difficult to clearly co-localize PALS-25(GOF)::mScarlet to mitochondria. However, this protein did have a discontinuous localization pattern, similar to and near mitochondria in some areas. We obtained similar results with strains expressing PALS-25(GOF)::GFP, using TOMM-20::mCherry, to mark the outer mitochondrial membrane (Figure S5A). Overall, these results suggest that the PALS-25 C-terminal 13 amino acids may not be required for mitochondrial localization (Figure 3B).Figure 3. The C-terminal half of PALS-25 is required for mitochondrial localization, and N-terminal residues are required for signaling(A) Predicted structure of PALS-25 (AlphaFold DB Q3V5H4). Gray corresponds to the PALS-25 C-terminal 13 amino acids, which are deleted in PALS-25(GOF), causing PALS-22 dissociation and IPR induction (Gang et al.15).(B) PALS-25(GOF)::mScarlet localization with the mitochondrially localized protein VWA-8::GFP.(C) PALS-25 (Δ40 GOF)::mScarlet localization with the mitochondrially localized protein VWA-8::GFP.(D) Quantification of *pals-5p::*GFP expression normalized to body area in adults (72 h post-L1). pals-22 pals-25(jy80); jyIs8 animals (control) show low fluorescence, and two positive control lines expressing mScarlet::PALS-25(GOF) show significant induction of IPR reporter. n = 3 independent experimental replicates with 15 animals per genotype per experiment. ∗∗∗∗p < 0.0001, one-way ANOVA with Dunnett’s multiple comparisons test. The graph shows mean values, with error bars representing standard deviations. Each symbol represents an individual animal. Different symbol shapes represent animals imaged on different days.(E) Western blot analysis of different transgenically expressed PALS-25 mutant proteins in total protein lysates using commercially available HA antibody and tubulin antibody as a loading control.(F) PALS-25 N-terminal half::mScarlet has diffuse cytosolic expression.(G) mScarlet::PALS-25 C-terminal half has mitochondrial localization. For B, C, F, and G, Scale bars 10 μm. White triangles denote seam cells. For D and E, all strains tested are in a pals-22 pals-25(jy80) mutant background.

Because many proteins are directed to mitochondria via a mitochondrial leader sequence (MLS) in their N-termini, we examined the N-terminus of PALS-25 for potential MLS sequences. There was not a significant hit for an MLS using prediction programs MitoFates22 and DeepMito.23 To experimentally examine the possibility of an MLS, we deleted the N-terminal 40 amino acids of PALS-25, and then investigated subcellular localization. Here we saw that PALS-25(Δ40 GOF)::mScarlet had a localization pattern similar to the PALS-25(GOF) (Figures 3B and 3C). Furthermore, we observed that PALS-25(Δ40GOF)::GFP had a localization pattern similar to PALS-25(GOF)::GFP, when labeling mitochondria with TOMM-20::mCherry (Figure S5B). Because of the difficulties with analyzing mitochondrial localization in PALS-25(GOF) strains, we removed the N-terminal 40 amino acids from the PALS-25 wild-type sequence. Here we saw clear mitochondrial localization of a PALS-25(Δ40)::GFP fusion with TOMM-20::mCherry (Figure S5C). Thus, the N-terminal 40 amino acids of PALS-25 do not appear to be responsible for mitochondrial localization.

We found that the N-terminal 40 amino acids are required for IPR induction. In contrast to the strong induction of the IPR reporter *pals-5p::*GFP in PALS-25(GOF)::mScarlet animals, PALS-25(Δ40 GOF)::mScarlet had no *pals-5p::*GFP induction (Figure 3D). We confirmed that PALS-25(GOF)::mScarlet and PALS-25(Δ40 GOF)::mScarlet proteins are expressed at similar levels, and thus lower PALS-25 expression levels are not responsible for the lack of IPR induction upon loss of the N-terminal 40 amino acids (Figure 3E). We did notice more aggregation of PALS-25(GOF) protein, whether tagged with mScarlet or with GFP (Figures 3B and S5A), when compared to wild-type PALS-25 (Figures 1B–1D and 2B). Given that this aggregation was also seen in IPR signaling-deficient PALS-25(Δ40 GOF) strains (Figures 3C and S5B), it seems more likely that the aggregation of PALS-25(GOF) is due to a loss of association with PALS-22, rather than a result of IPR induction

To more broadly examine which region of PALS-25 directs localization to the mitochondria, we took advantage of AlphaFold, which predicts that the PALS-25 protein structure is composed of an N-terminal bundle of 4 alpha-helices and a C-terminal bundle of 4 alpha-helices (Figure 3A). We thus generated tagged versions of the N-terminal half and C-terminal half of PALS-25. Specifically, we fused mScarlet to the C-terminus of the N-terminal half of PALS-25 (1–139 aa), and we fused mScarlet to the N-terminus of the C-terminal half of PALS-25 (140–305 aa). Here, we found that the N-terminal half of PALS-25 had a diffuse cytosolic localization pattern that did not colocalize with the mitochondrial marker VWA-8::GFP21 (Figure 3F), whereas the C-terminal half of PALS-25 colocalized well with VWA-8::GFP (Figure 3G). We saw similar results with GFP-tagged versions of PALS-25; the N-terminus was diffuse and did not colocalize with the mitochondrial marker TOMM-20::mCherry, while the C-terminus co-localized well with TOMM-20::mCherry (Figures S5D and S5E). Of note, neither the N-terminal nor the C-terminal half of PALS-25 could activate the IPR independently (Figure 3D). Altogether, these results indicate that the C-terminus of PALS-25 is both necessary and sufficient for localization to mitochondria. Furthermore, the N-terminal 40 amino acids are required for the induction of IPR reporter pals-5p::GFP.

Loss of pals-22 causes mitochondrial fragmentation

Upon the analysis of pals-22 RNAi-treated animals, we noticed a change in the morphology of mitochondria (Figures 2B and S6A). We quantified these effects in two pals-22 loss-of-function mutants, pals-22(jy1) and pals-22(jy3) (Figure S1C), which have similar levels of constitutive IPR induction, pathogen resistance, increased thermotolerance, as well as other phenotypes.11^,^16 We first analyzed epidermal mitochondria using col-19p::mito-GFP (Figures 4A–4C and S7A–S7C). Here, we found that pals-22 mutants had a significantly lower mitochondrial form factor (Figure 4D), which is a metric for the complexity of mitochondria (see STAR Methods). A higher form factor indicates more elongated mitochondria, and a lower form factor indicates more spherical mitochondria (form factor = 1 is a perfect circle). In particular, epidermal mitochondria in pals-22 mutants had significantly fewer branches and shorter branches, as well as slightly wider branch diameters (Figures 4E–4G).Figure 4pals-22 mutants have mitochondrial fragmentation(A–C) Representative images of wild-type (A), pals-22(jy1) (B), and pals-22(jy3) (C) young adult animals expressing jyEx4796[col-19p::mito-GFP] in the epidermis.(D–G) pals-22 mutants display altered mitochondria morphology in the epidermis, including decreased form factor (less elongated and more spherical morphology) (D), fewer branches per mitochondria (E), decreased average length (F), and increased average diameter (G).(H–J) Representative images of wild-type (H), pals-22(jy1) (I), and pals-22(jy3) (J) young adult animals expressing mgIs48[ges-1p::mito-GFP] in the intestine.(K–N) pals-22 mutants display altered mitochondria morphology in the anterior intestine, including decreased form factor (K), fewer branches per mitochondria (L), decreased average length (M), and increased average diameter (N). For A–C and H–J, scale bars, 10 μm. The white boxes indicate a region of interest (ROI) of the epidermal seam cells, and increased magnification of mitochondria morphology in the ROI based on mito-GFP expression is shown to the right. ROI scale bars, 2.5 μm. The exposure time was set to optimize mito-GFP expression in wild-type animals. However, we noted many pals-22 mutants that displayed increased mito-GFP fluorescence intensity (white triangles). For D–G and K–N, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, Kruskal-Wallis test with Dunn’s multiple comparisons test. n = 15 ROIs analyzed per genotype, one ROI per animal, across three independent experimental replicates. Graphs show mean values, and error bars represent standard deviations. Different symbol shapes represent animals imaged on different days.

We next performed analysis of intestinal mitochondria as imaged by ges-1p::mito-GFP, and saw similar effects in this tissue; i.e., mitochondria in pals-22 mutants had a lower form factor, fewer branches, shorter branches, and slightly wider branch diameter (Figures 4H–4N and S7D–S7F). As expected, these morphology changes were reversed to a wild-type phenotype in pals-22/25(jy80) double mutants (Figure S6B), indicating that the effects on mitochondrial morphology of pals-22 loss were dependent on pals-25. Therefore, loss of pals-22 leads to broad changes in mitochondrial morphology in both the intestine and the epidermis, with a switch to mitochondria becoming more spherical, less branched, and less elongated, collectively described as being more fragmented.

To investigate how pals-22-mediated mitochondrial fragmentation might connect with known mitochondrial regulatory factors, we compared genes induced as part of the IPR to genes induced by the transcription factor ATFS-1, which mediates the induction of the mitochondrial unfolded protein response (mitoUPR).24^,^25^,^26 Here we found some overlap between these gene sets, but the majority of mitoUPR genes are not part of the IPR, including canonical mitoUPR genes such as hsp-6 and hsp-60 (Figure S6C and Table S3). We next investigated whether knockdown of the mitochondrial fission factor drp-1 might restore normal morphology in pals-22 mutants. Of note, while loss of drp-1 causes the outer mitochondrial membrane to be more elongated (Figure 1D), it causes the fragmentation of the mitochondrial matrices (Figure S6D).19 In *pals-*22 mutants, we found not only a lack of rescue by drp-1 RNAi, but instead an exacerbation of the mitochondrial morphology phenotype, with extremely compacted mitochondria when drp-1 RNAi knock-down was performed (Figure S6E). In addition, we did not see rescue of the pals-22 phenotype with RNAi against pink-1, which is required for mitophagy (Figure S6E).18 Thus, the fragmentation of mitochondria in pals-22 mutants appears to be distinct from changes caused by several canonical mitochondrial regulatory factors.

Intracellular pathogen response activation precedes mitochondrial fragmentation, which promotes defense against microsporidia infection

We next investigated whether PALS-22 exerts its effects on mitochondrial morphology directly due to its localization to that organelle, or indirectly through the transcriptional IPR program. To distinguish between these models, we performed kinetic analysis to determine whether loss of PALS-22 leads to mitochondrial morphology changes before or after IPR induction. We used a PALS-22::AID strain, which has a degron tag that enables controlled degradation of this protein upon exposure to auxin, when co-expressed with the TIR1 ubiquitin ligase adaptor.15^,^27 With a time-course of auxin treatment, we found that after 3 h of auxin treatment, there was robust induction of the IPR reporter *pals-5p::*GFP (Figures 5A and S8), whereas significant mitochondrial fragmentation in the intestine did not occur until 4 h and later (Figures 5B and 5C). Similarly, *pals-5p::*GFP induction preceded mitochondrial fragmentation in the epidermis as well (Figures 5D–5G). Specifically, there was no significant effect on mitochondrial form factor at 3 h, but there was at 48 h.Figure 5IPR reporter induction precedes mitochondrial fragmentation upon the depletion of PALS-22(A) Time course images of transgenic animals expressing ubiquitous TIR1, endogenous pals-22 tagged with AID, and the *pals-5p::*GFP, *myo-2p::*mCherry IPR reporter treated with either ethanol vehicle control (top) or auxin (bottom) starting at 40 h post-L1. Auxin-mediated depletion of PALS-22 induces pals-5p::GFP IPR reporter expression as early as ∼2 h after auxin treatment (white triangles) and in most tissues by 3 h. *myo-2p::*mCherry is a marker for the transgene expressed in the pharynx. Images are an overlay of DIC, GFP, and mCherry fluorescence channels. Scale bars, 100 μm.(B) Time course images of transgenic animals expressing ubiquitous TIR1, endogenous pals-22 tagged with AID, and mgIs48[ges-1p::mito-GFP] treated with either vehicle control (top) or auxin (bottom) starting at 40 h post-L1. The white boxes indicate an ROI in the first ring of intestinal cells, and increased magnification showing mitochondria morphology in the ROI based on mito-GFP expression is inset. The exposure time was set to optimize mito-GFP expression at the 0 h treatment time point. However, we noted that increased time exposed to auxin caused increased mito-GFP fluorescence intensity (white triangles).(C) Quantification of the treatment time course in B. Auxin-mediated depletion of PALS-22 induces significant changes in intestinal mitochondria morphology, as measured by form factor, starting 4 h after auxin treatment.(D–G) Mitochondrial morphology analysis of jyEx4796[col-19p::mito-GFP] in the epidermis following the auxin-mediated depletion of PALS-22.(D and F) Representative images of 0 h and 3 h (D) or 48 h (F) treatment timepoints of mitochondria in the epidermis positioned relative to anterior intestinal cells (located above or below the first two rings of intestinal cells). The white boxes indicate an ROI, and increased magnification showing mitochondria morphology in the ROI based on mito-GFP expression is inset.(E) Quantification of mitochondria form factor in the epidermis at 0 h vs. 3 h treatment in D. Auxin-mediated depletion of PALS-22 does not induce changes in epidermal mitochondria morphology after 3 h.(G) Quantification of mitochondria form factor in the epidermis at 48 h treatment in F. Auxin-mediated depletion of PALS-22 induces significant changes in epidermal mitochondria morphology 48 h after auxin treatment. For B, D, and F, scale bars, 10 μm, and ROI scale bars, 2.5 μm. For C, E, and G, ns = not significant, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, Welch’s t test for vehicle control vs. auxin at each time point. n = 12 ROIs analyzed per time point and condition, two ROIs per animal, across three experimental replicates. Graphs show mean values, and error bars represent standard deviations. Different symbol shapes represent ROIs from animals imaged on different days.

To test the model that loss of pals-22 causes mitochondrial fragmentation as a downstream consequence of IPR induction, we activated the IPR in a distinct manner (Figure 6A). Specifically, we performed RNAi against pals-17, which encodes a negative IPR regulator that localizes to the plasma membrane in the intestine, without obvious localization to mitochondria.12 Here, we found that RNAi against pals-17 had similar effects to the loss of pals-22, with intestinal mitochondria becoming fragmented (Figures 6B and 6C). To further test the model that the transcriptional induction of the IPR causes mitochondrial fragmentation, we analyzed the role of the bZIP transcription factor ZIP-1, which controls the induction of about one-third of IPR genes.28 Here we saw pals-22 RNAi caused less mitochondrial fragmentation in the intestines of zip-1(jy13) mutants compared to wild-type controls (Figure 6D). Overall, these findings, together with the PALS-22 depletion studies mentioned above, support the model that the pals-22/25 module controls mitochondrial fragmentation through IPR induction.Figure 6pals-17 RNAi induces mitochondrial fragmentation, and zip-1 is partially required for pals-22 RNAi-induced fragmentation(A) Representative images of *pals-5p::*GFP, *myo-2p::*mCherry IPR reporter expressing animals treated with either control or pals-17 RNAi for 24 h starting at 20 h post-L1. Control RNAi does not induce IPR reporter, and pals-17 RNAi induces pals-5p::GFP IPR reporter expression in the intestine at the 24 h timepoint shown (white arrows). *myo-2p::*mCherry is part of the same transgene and is constitutively expressed in the pharynx of transgenic animals at all life stages. Images are an overlay of DIC, GFP, and mCherry fluorescence channels. Scale bars, 100 μm.(B) Representative images of mgIs48[ges-1p::mito-GFP] animals 20 h post-L1 exposed to control or pals-17 RNAi for 24 h. The white boxes indicate an ROI in the first ring of intestinal cells, and increased magnification showing mitochondria morphology in the ROI based on mito-GFP expression is inset. Scale bars, 15 μm, ROI scale bars, 2.5 μm.(C) pals-17 RNAi induces changes in intestinal mitochondria morphology, as measured by form factor. ∗∗∗p < 0.001, unpaired t test. n = 30 ROIs analyzed per condition, two ROIs per animal, across three experimental replicates.(D) Loss of zip-1 partially rescues changes in mitochondria morphology induced by pals-22 RNAi. ∗∗∗p < 0.001, ∗p < 0.05, unpaired t-tests between strains and conditions. n = 58–70 ROIs analyzed per strain and condition across three experimental replicates. For C and D, the graph indicates mean values and error bars represent standard deviations. Different symbol shapes represent ROIs from animals imaged on different days.

We next investigated N. parisii localization to C. elegans mitochondria via transmission electron microscopy (TEM) and found several examples of close association (Figures 7A and S9). This type of association has been seen with microsporidia species infecting other hosts,29 and is thought to aid in ATP and other nucleotide transport into microsporidia, which lack their own mitochondria and nucleotide biosynthesis pathways. To determine whether host mitochondrial fragmentation may regulate resistance against microsporidia, we induced mitochondrial fragmentation through either drp-1 RNAi or fzo-1 RNAi and then measured N. parisii pathogen load. In both cases, we found that C. elegans had increased resistance to infection (Figure 7B). Notably, neither of these RNAi treatments caused the induction of the pals-5p::GFP reporter (0% of animals had induction, n > 200 animals examined for each condition). Thus, the fragmentation of the mitochondrial matrix independent of pals-5p::GFP induction appears to protect C. elegans against microsporidia infection.Figure 7. Animals with mitochondrial fragmentation independent of IPR induction have decreased N. parisii pathogen load(A) Representative TEM images of N. parisii sporonts (red double asterisk) associated with C. elegans mitochondria (purple asterisks) during infection. Scale bars, 500 nm, ROI scale bars, 200 nm.(B) Wild-type animals treated for two generations on drp-1 or fzo-1 RNAi display increased resistance to N. parisii infection. ∗∗p < 0.01, ∗∗∗∗p < 0.0001, one-way ANOVA with Dunnett’s multiple comparisons test. n = 209–279 animals per treatment across three experimental replicates. The graph shows mean values, and error bars represent standard deviations. Different symbol shapes represent animals analyzed in distinct infection experiments.(C) Model for PALS-22 and PALS-25 as regulators of mitochondrial fragmentation through IPR activation. Mitochondrial fragmentation, either through IPR activation or other morphology regulators, promotes resistance to microsporidia infection.

Discussion

The immune switch proteins PALS-22 and PALS-25 cause a dramatic rewiring of C. elegans physiology, controlling the balance between an immune state and a growth state.11 However, PALS proteins lack clear functional domains and have a poorly understood mechanism of action. Here, we show that both PALS-22 and PALS-25 proteins localize to mitochondria, apparently at the outer membrane, with PALS-22 being dependent on PALS-25 for localization to mitochondria. We define the C-terminal half of PALS-25 to be necessary and sufficient for mitochondrial localization, and the N-terminal 40 amino acids as being required to activate the pals-5p::GFP IPR reporter (Figure 7C). We also show that loss of PALS-22 causes a reduction in PALS-25 protein levels. Future studies can investigate whether the reduction in PALS-25 protein levels is controlled by targeted protein degradation, such as ubiquitylation-mediated targeting to the proteasome or lysosome. We also found that loss of PALS-22 causes the formation of PALS-25 puncta, the functional importance and nature of which can be explored in the future.

Notably, we found that loss of PALS-22 triggers mitochondrial fragmentation, which occurs after activating the IPR transcriptional program, and can be induced by loss of another IPR inhibitor, PALS-17. Furthermore, loss of the transcription factor ZIP-1, which controls the expression of one-third of IPR genes,28 partially suppressed the fragmentation caused by pals-22 knockdown. Overall, these results suggest that IPR activation may trigger fragmentation via downstream factors, rather than direct regulation by PALS-22 and PALS-25 (Figure 7C). Two recent studies in mammalian tissue culture cells have shown that microsporidia infection promotes the fragmentation of host mitochondria.30^,^31 During infection, it can be difficult to disentangle the role of host vs. pathogen-derived factors in driving fragmentation. One of these studies suggested that microsporidia infection activated DRP-1, which caused mitochondria fragmentation to ultimately aid microsporidia growth, based on analysis with the mdivi-1 inhibitor, which has multiple impacts in the cell.30^,^32 In contrast, here we report that mitochondrial fragmentation appears to impair microsporidia growth (Figures 7B and 7C). Further studies will be needed to determine when and where mitochondrial fragmentation is beneficial vs. inhibitory for microsporidia growth.

Several external stressors cause the fragmentation of mitochondria, such as wounding of the C. elegans epidermis, where fragmentation appears to promote wound healing.33 Similarly, heat shock can induce the fragmentation of mitochondria, which in some paradigms appears to then promote resistance to heat shock.34^,^35 Fragmentation is required for subsequent removal of damaged mitochondria through mitophagy. Similar themes have been observed after C. elegans infection with the extracellular bacterial pathogen Pseudomonas aeruginosa, where mitochondrial mass is reduced.36 In contrast to stressors such as heat shock and infection that trigger the fragmentation of mitochondria, upregulation of many stress/immune response pathways triggers hyperfusion or expansion of mitochondrial networks. For example in C. elegans, ATFS-1, the transcription factor that regulates the mitoUPR, appears to promote the expansion of mitochondrial networks,37 and overexpression of heat shock factor 1 (HSF-1) leads to the hyperfusion of mitochondria and increased lifespan.38 Wild-type function of the bZIP transcription factor ZIP-2, which promotes defense against P. aeruginosa, also promotes mitochondrial fusion.39 Similarly, in mammals, the activation of the endoplasmic reticulum stress pathway causes mitochondrial hyperfusion,40 and activation of the RIG-I-like receptor pathway can also activate hyperfusion, through separate kinase driven mechanisms.41

Our observation that the IPR triggers fragmentation in the absence of a stressor is perhaps unusual for a stress response pathway, as the pathways mentioned above promote mitochondrial fusion. One exception is the transcription factor HLH-30/TFEB, which promotes mitochondrial fusion in a cell-autonomous manner, but mitochondrial fragmentation in a cell non-autonomous manner.42 Similarly, mitochondrial fragmentation in neurons was found to cell non-autonomously cause fragmentation in the intestine.43 Future experiments will explore whether PALS-22/25-mediated mitochondrial fragmentation is controlled cell-intrinsically and/or systemically. Of note, fragmented mitochondria may switch to fatty acid oxidation to provide fuel, which perhaps is beneficial in the context of intracellular pathogens such as microsporidia.44 Indeed, there are indications that microsporidia cause a decrease in host fat levels and rely on host fatty acids, and thus depleting fatty acids might aid in resistance to infection.45^,^46

We found that loss of either PALS-22 or PALS-17 would trigger mitochondrial fragmentation. In contrast to PALS-22 and PALS-25, the protein PALS-17 and its positive regulator PALS-20, which is another OFF/ON switch of the IPR, localize to the apical membrane of intestinal cells.12 It is possible that membrane localization is critical for the signaling mechanism of action for these PALS proteins. Many other immune regulators are localized to membranes, such as the viral RNA sensor RIG-I, which localizes to mitochondria upon activation and binding to its downstream signaling partner MAVS. More detailed analysis of the activity of PALS-22/25 and other PALS proteins at mitochondrial and other membranes will help decipher how they trigger downstream IPR signaling and immunity.

Limitations of the study

We show that PALS-22 and PALS-25 localize to mitochondria, and define the C-terminus of PALS-25 as necessary and sufficient for mitochondrial localization, but we do not show how the C-terminus of PALS-25 localizes to mitochondria. We also show that PALS-22/25 controls mitochondrial fragmentation and that fragmentation occurs after IPR activation, but we do not show the mechanism of mitochondrial fragmentation. We also do not show the mechanism by which PALS-22/25 activate the IPR.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Emily Troemel (etroemel@ucsd.edu).

Materials availability

C. elegans strains generated in this study are available upon request.

Data and code availability

• •All data reported in this article will be shared by the lead contact upon request.
• •This article does not report original code.
• •Any additional information required to reanalyze the data reported in this article is available from the lead contact upon request.

Acknowledgments

We are grateful to N. Antao, M. Barkoulas, and V. Lažetić for comments on the article. We thank M. Wood at Scripps Research Institute for assistance in TEM experiments. We thank G. Bhabha and D. Ekiert for helpful conversations about PALS protein structure. We thank A. Chisholm and P. Murugesan for assistance with LSM800 confocal imaging. Some strains were provided by the CGC, which is funded by the 10.13039/100016958NIH Office of Research Infrastructure Programs (P40 OD010440).

This work was supported by 10.13039/100000052NIH under R01 AG052622, GM114139, AI176639, and by 10.13039/100006200NSF IOS-2301657 to E.R.T; S.S.G. received funding from 10.13039/100000052NIH award K12GM068524. M.W.S. received funding from 10.13039/100000052NIH award GM007240.

Author contributions

Conceptualization: S.S.G., M.W.S., and E.R.T.; funding acquisition: E.R.T.; investigation: S.S.G., M.W.S., and D.R.-B.; supervision: S.S.G. and E.R.T.; writing – original draft: E.R.T.; writing – review and editing: S.S.G., M.W.S., D.R.-B., and E.R.T.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesPALS-22 polyclonal antibody raised in rabbitProSCI IncorporatedGang et al.15–PALS-25 polyclonal antibody raised in rabbitProSCI IncorporatedGang et al.15–⍺-Tubulin monoclonal antibody raised in mouseMilliporeSigmaCloneDM1A; Cat#T9026; RRID: AB_477593HA monoclonal antibody raised in rabbitCell SignalingCat#C29F4Goat anti-mouse secondary antibodyMilliporeSigmaCat#401215; RRID: AB_10682749Goat anti-rabbit secondary antibodyMilliporeSigmaCat#401315; RRID: AB_10682917MicroB Red N. parisii FISH probeBiosearch TechnologiesTroemel et al.47–Bacterial and virus strainsE. coli: OP50-1Gary Ruvkun lab–E. coli: OP50 RNAiMeng Wang lab–E. coli: HT115Gary Ruvkun lab–N. parisiiTroemel et al.47; Cuomo et al.48ERTm1Chemicals, peptides, and recombinant proteinsMitoTracker Red CMXRosInvitrogenCat#M7512Indole-3-Acetic AcidAlfa AesarCat#I6125-5GTRI ReagentMolecular Research Center IncorporatedCat#TR1181-Bromo-3-ChloropropaneMolecular Research Center IncorporatedCat#BP151Amersham Enhanced Chemiluminescence ReagentCytiviaCat#RPN2209Commercial assaysiScript cDNA Synthesis KitBio-RadCat#1708890iQ SYBR Green SupermixBio-RadCat#1708880QIAprep Spin Miniprep KitQiagenCat#27104Deposited dataCo-IP/MS of PALS-22 and PALS-25Panek et al.14MassIVE: MSV000084936IPR Upregulated GenesReddy et al.11GEO: [GSE118400](GSE118400)atfs-1(et15) Upregulated GenesWu et al.25GEO: [GSE110984](GSE110984)Experimental models: Organisms/strainsC. elegans: Strain N2 wild-typeCaenorhabditis Genetics CenterN2C. elegans: Strain ERT054 jyIs8[pals-5p::GFP, myo-2p::mCherry] XBakowksi et al.49ERT054C. elegans: Strain ERT365 unc-119(ed3) III; jyEx193[pals-22::eGFP::3xFLAG, unc-119(+)]Reddy et al.16ERT365C. elegans: Strain ERT415 pals-22(jy3) IIIReddy et al.16ERT415C. elegans: Strain ERT465 unc-119(ed3) III; jyEx237[pals-25::eGFP::3xFLAG, unc-119(+)]Reddy et al.11ERT465C. elegans: Strain ERT714 pals-22 pals-25(jy80) IIIReddy et al.11ERT714C. elegans: Strain ERT751 pals-25(GOF) IIIGang et al.15ERT751C. elegans: Strain ERT930 pals-22 pals-25(jy80) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] XGang et al.15ERT930C. elegans: Strain ERT964 pals-22 pals-25(jy80) unc-119(ed3) III; jyEx193[pals-22::eGFP::3xFLAG, unc-119(+)]This studyERT964C. elegans: Strain ERT965 pals-22 pals-25(jy80) unc-119(ed3) III; jyEx237[pals-25::eGFP::3xFLAG, unc-119(+)]This studyERT965C. elegans: Strain ERT1008 pals-22 pals-25(jy80) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] X; jyEx296[col-19p::pals-25(GOF)::wrmScarlet::3xHA::unc-54 3′ UTR]Gang et al.15ERT1008C. elegans: Strain ERT1010 pals-22 pals-25(jy80) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] X; jyEx298[col-19p::pals-25(GOF)::wrmScarlet::3xHA::unc-54 3′ UTR]Gang et al.15ERT1010C. elegans: Strain ERT1017 keaSi10[rpl-28p::TIR1::mRuby::unc-54 3′UTR + Cbr-unc-119(+)] II; pals22(kea8[pals-22::GFP::degron]) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] XGang et al.15ERT1017C. elegans: Strain ERT1067 juEx4796[col-19p::mito-GFP; ttx-3p::RFP]Fu et al.33ERT1067; CZ16518C. elegans: Strain ERT1068 pals-22(jy3) III; juEx4796[col-19p::mito-GFP; ttx-3p::RFP]This studyERT1068C. elegans: Strain ERT1145 mgIs48[ges-1p::mito-GFP]Wu et al.50ERT1145; MGH50C. elegans: Strain ERT1146 pals-22(jy3) III; mgIs48[ges-1p::mito-GFP]This studyERT1146C. elegans: Strain ERT1149 pals-22(jy1) III; mgIs48[ges-1p::mito-GFP]This studyERT1149C. elegans: Strain ERT1165 pals-22(jy1) III; juEx4796[col-19p::mito-GFP; ttx-3p::RFP]This studyERT1165C. elegans: Strain ERT1187 keaSi10[rpl-28p::TIR1::mRuby::unc-54 3′UTR + Cbr-unc-119(+)] II; pals22(kea8[pals-22::GFP::degron]) III; mgIs48[ges-1p::mito-GFP]This studyERT1187C. elegans: Strain ERT1224 keaSi10[rpl-28p::TIR1::mRuby::unc-54 3′UTR + Cbr-unc-119(+)] II; pals22(kea8[pals-22::GFP::degron]) III; juEx4796[col-19p::mito-GFP; ttx-3p::RFP]This studyERT1224C. elegans: Strain ERT1255 jyEx334[pals-22p::GFP::pals-22::3xFLAG::sl2::3xHA::wrmScarlet::pals-25::unc-54 3′ UTR]This studyERT1255C. elegans: Strain ERT1301 pals-22 pals-25(jy80) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] X; jyEx354[pals-22p::pals-22::3xFLAG::sl2::3xHA::wrmScarlet::pals-25(140–305, C-term half)::unc-54 3′ UTR; gcy-8p::GFP]This studyERT1301C. elegans: Strain ERT1303 pals-22 pals-25(jy80) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] X; jyEx356[pals-22p::pals-22::3xFLAG::sl2::3xHA::wrmScarlet::pals-25(140–305, C-term half)::unc-54 3′ UTR; gcy-8p::GFP]This studyERT1303C. elegans: Strain ERT1331 pals-22 pals-25(jy80) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] X; jyEx369[col-19p::pals-25(Δ40 GOF)::wrmScarlet::3xHA::unc-54 3′ UTR; gcy-8p::GFP]This studyERT1331C. elegans: Strain ERT1332 pals-22 pals-25(jy80) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] X; jyEx370[col-19p::pals-25(Δ40 GOF)::wrmScarlet::3xHA::unc-54 3′ UTR; gcy-8p::GFP]This studyERT1332C. elegans: Strain ERT1334 pals-22 pals-25(jy80) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] X; jyEx372[pals-22p::pals-22::sl2::pals-25(1–139, N-term half)::wrmScarlet::3xHA::unc-54 3′ UTR; gcy-8p::GFP]This studyERT1334C. elegans: Strain ERT1335 pals-22 pals-25(jy80) III; jyIs8[pals-5p::GFP, myo-2p::mCherry] X; jyEx373[pals-22p::pals-22::sl2::pals-25(1–139, N-term half)::wrmScarlet::3xHA::unc-54 3′ UTR; gcy-8p::GFP]This studyERT1335C. elegans: Strain ERT1368 jyEx379[pals-22p::pals-22::sl2::pals-25(1–139, N-term half)::wrmScarlet::3xHA::unc-54 3′ UTR; gcy-8p::GFP]; ju1799[vwa-8::GFP::3xFLAG] XThis studyERT1368C. elegans: Strain ERT1369 jyEx380[pals-22p::pals-22::3xFLAG::sl2::3xHA::wrmScarlet::pals-25(140–305, C-term half)::unc-54 3′ UTR; gcy-8p::GFP]; ju1799[vwa-8::GFP::3xFLAG] XThis studyERT1369C. elegans: Strain ERT1372 jyEx383[col-19p::pals-25(Δ40 GOF)::wrmScarlet::3xHA::unc-54 3′ UTR; gcy-8p::GFP]; ju1799[vwa-8::GFP::3xFLAG] XThis studyERT1372C. elegans: Strain ERT1373 jyEx384[col-19p::pals-25(GOF)::wrmScarlet::3xHA::unc-54 3′ UTR; gcy-8p::GFP]; ju1799[vwa-8::GFP::3xFLAG] XThis studyERT1373C. elegans: Strain ERT1495 jyEx431[pals-22p::pals-22::sl2::pals-25::eGFP::unc-54 3′ UTR; gcy-8p::GFP); wbmIs98[eft-3p::tomm-20(aa1-49)::mCherry::unc-54 3′UTR]; wbmIs65[eft-3p::3XFLAG::dpy-10 crRNA::unc-54 3′UTR]This studyERT1495C. elegans: Strain ERT1496 jyEx432[col-19p::pals-25(GOF)::eGFP::3xHA::unc-54 3′ UTR; gcy-8p::GFP]; wbmIs98[eft-3p::tomm-20(aa1-49)::mCherry::unc-54 3′ UTR]; wbmIs65[eft-3p::3XFLAG::dpy-10 crRNA::unc-54 3′UTR]This studyERT1496C. elegans: Strain ERT1497 jyEx433[col-19p::pals-25(Δ40 GOF)::eGFP::3xHA::unc-54 3′ UTR; gcy-8p::GFP]; wbmIs98[eft-3p::tomm-20(aa1-49)::mCherry::unc-54 3′ UTR]; wbmIs65[eft-3p::3XFLAG::dpy-10 crRNA::unc-54 3′UTR]This studyERT1497C. elegans: Strain ERT1498 jyEx434[pals-22p::pals-22::sl2::pals-25(1–139, N-term half)::eGFP::3xFLAG::unc-54 3′ UTR; gcy-8p::GFP]; wbmIs98[eft-3p::tomm-20(aa1-49)::mCherry::unc-54 3′ UTR]; wbmIs65[eft-3p::3XFLAG::dpy-10 crRNA::unc-54 3′UTR]This studyERT1498C. elegans: Strain ERT1499 jyEx435[pals-22p::pals-22::sl2::pals-25(140–305, C-term half)::eGFP::3xFLAG::unc-54 3′ UTR; gcy-8p::GFP); wbmIs98[eft-3p::tomm-20(aa1-49)::mCherry::unc-54 3′ UTR]; wbmIs65[eft-3p::3XFLAG::dpy-10 crRNA::unc-54 3′UTR]This studyERT1499C. elegans: Strain ERT1500 jyEx436[pals-22p::pals-22::sl2::pals-25(Δ40)::eGFP::3xFLAG::unc-54 3′ UTR; gcy-8p::GFP]; wbmIs98[eft-3p::tomm-20(aa1-49)::mCherry::unc-54 3′ UTR]; wbmIs65[eft-3p::3XFLAG::dpy-10 crRNA::unc-54 3′UTR]This studyERT1500C. elegans: Strain ERT1501 *zip-1(jy13) III; mgIs48[ges-1p::mito-GFP]*This studyERT1501OligonucleotidesqPCR Primer: snb-1 F 5′-CCGGATAAGACCATCTTGACG-3′Troemel et al.51–qPCR Primer: snb-1 R 5′-GACGACTTCATCAACCTGAGC-3′Troemel et al.51–qPCR Primer: pals-5 F 5′-CATTGGAAAGCGATATTGGA-3′Bakowski et al.49–qPCR Primer: pals-5 R 5′-TCTCCAGGCACCTATCTTGTAG-3′Bakowski et al.49–qPCR Primer: F26F2.1 F 5′-TGGAACCAGGTCAGAGACAC-3′Bakowski et al.49–qPCR Primer: F26F2.1 R 5′-TTGTGAGAATTTCCGCGATA-3′Bakowski et al.49–qPCR Primer: skr-5 F 5′-CGAAGAGCAAGATGTCAAAATTG-3′Bakowski et al.49–qPCR Primer: skr-5 R 5′-AGAAGCTTGGATTGATTGGCA-3′Bakowski et al.49–qPCR Primer: pals-25 F 5′-ACTCATCTTCCAATCGCCATTG-3′This study–qPCR Primer: pals-25 R 5′-CATTTTATTGCACAACTCGGCC-3′This study–qPCR Primer: nhr-23 F 5′-CGGATATTCTATAGCTGTTGC-3′Troemel et al.51–qPCR Primer: nhr-23 R 5′-ACTTGTGGCGATGGGAAGC-3′Troemel et al.51–Recombinant DNAPlasmid: pET729 col-19p::pals-25(GOF)::wrmScarlet::3xHA::unc-54 3′ UTRGang et al.15pET729Plasmid: pET787 pals-22p::GFP::pals-22::3xFLAG::sl2::3xHA::wrmScarlet::pals-25::unc-54 3′ UTRThis studypET787Plasmid: pET789 col-19p::pals-25(Δ40 GOF)::wrmScarlet::3xHA::unc-54 3′ UTRThis studypET789Plasmid: pET792 pals-22p::pals-22::sl2::pals-25(1–139, N-term half)::wrmScarlet::3xHA::unc-54 3′ UTRThis studypET792Plasmid: pET793 pals-22p::pals-22::3xFLAG::sl2::3xHA::wrmScarlet::pals-25(140–305, C-term half)::unc-54 3′ UTRThis studypET793Plasmid: pET803: pals-22p::pals-22::sl2::pals-25::eGFP::3xFLAG::unc-54 3′ UTRThis studypET803Plasmid: pET845 pals-22p::pals-22::sl2::pals-25(1–139, N-term half)::eGFP::3xFLAG::unc-54 3′ UTRThis studypET845Plasmid: pET846 pals-22p::pals-22::sl2::pals-25(140–305, C-term half)::eGFP::3xFLAG::unc-54 3′ UTRThis studypET846Plasmid: pET847 col-19p::pals-25(GOF)::eGFP::3xHA::unc-54 3′ UTRThis studypET847Plasmid: pET848 col-19p::pals-25(Δ40 GOF)::eGFP::3xHA::unc-54 3′ UTRThis studypET848Plasmid: pET849 pals-22p::pals-22::sl2::pals-25(Δ40)::eGFP::3xFLAG::unc-54 3′ UTRThis studypET849RNAi clone: pL4440-RNAi control (HT115)Ahringer RNAi Library–RNAi clone: pL4440-RNAi control (OP50)Ahringer RNAi Library–RNAi clone: fzo-1 (HT115)Ahringer RNAi Library–RNAi clone: eat-3 (HT115)Ahringer RNAi Library–RNAi clone: drp-1 (HT115)Ahringer RNAi Library–RNAi clone: pals-22 (HT115)Ahringer RNAi Library–RNAi clone: pals-25 (HT115)Reddy et al.11–RNAi clone: pals-17 (HT115)Lažetić et al.28–RNAi clone: pals-22 (OP50)Panek et al.14–Software and algorithmsFIJI ImageJ 2.16.0NIHRRID: SCR_003070Mitochondria Analyzer FIJI PluginChaudhry et al.52https://github.com/AhsenChaudhry/Mitochondria-Analyzer/tree/masterMitochondria image pre-processingThis studyhttps://github.com/TroemelLabUCSD/Mitochondria-and-PathogenLoad-AnalysisGraphPad Prism 10GraphPad Software, Inc.RRID: SCR_002798Zeiss ZEN Microscopy SoftwareCarl Zeiss AGRRID: SCR_013672Venn Diagram MakerUniversity of Gent Bioinformatics & Evolutionary Genomicshttps://bioinformatics.psb.ugent.be/webtools/Venn/AlphaFoldJumper et al.53RRID: SCR_025454ChimeraXUCSFRRID: SCR_015872Image Lab SoftwareBio-RadRRID: SCR_014210MetaXpress 2018, version 6.5.2.351Molecular DevicesRRID: SCR_016654

Experimental model and study participant details

C. elegans worms were maintained following standard methods54 at 20 °C on Nematode Growth Media (NGM) plates seeded with streptomycin-resistant Escherichia coli OP50-1 unless stated otherwise for specific experiments. C. elegans exist as either hermaphrodites or males, and hermaphrodites were used for all assays. Results with males are unknown. See key resources table and Table S2 for more information on C. elegans strains.

Method details

Synchronization of C. elegans growth

To obtain populations of worms synchronized at the same life stage, gravid adults were washed from NGM + OP50-1 plates with M9 media into a 15 mL conical tube. The contents of the tube were then centrifuged at 3,000 rpm for 30 s to pellet the worms, and the supernatant was removed, leaving ∼2 mL of M9 with the worm pellet. 800 μL of 5.65–6% sodium hypochlorite solution and 200 μL of 2 M NaOH were added, and the tube was vortexed intermittently for ∼2 min or until the majority of adult animals had partially dissolved and released embryos into solution. The embryos were washed by filling the tube to 15 mL with M9, the tube was centrifuged to pellet the embryos, and the supernatant was removed. The 15 mL M9 wash-and-centrifuge process was repeated 4 additional times (5 washes total), and the embryos were resuspended to a final volume of ∼5 mL in M9. The tube was placed in a 20 °C incubator with continual rotation for 16-24 h to hatch synchronized L1s.

RNA interference

RNAi was performed using the E. coli HT115 feeding method. L4440 control vector, pals-22, pals-25, pals-17, fzo-1, eat-3, pink-1, and drp-1 RNAi clones in the L4440 vector background were used, except for Figure 6D, where pals-22 and L4440 control vector were used in the OP50(RNAi) background.14 Colonies for each clone were grown overnight in LB broth supplemented with 50 μg/mL carbenicillin at 37 °C with shaking at 250 rpm. 6-cm or 10-cm RNAi plates (NGM plates supplemented with 5 mM IPTG and 1 mM carbenicillin) were seeded with 400 μL or 1 mL of overnight culture, respectively, except for Figure 6D, where 10-cm RNAi plates were seeded with 750 μL of 4-fold concentrated overnight culture. Seeded plates were dried in a biosafety cabinet. The plates were then stored at room temperature in the dark for 48-72 h to allow the RNAi bacteria lawn to grow. The RNAi knockdown procedure for specific experiments is described below.

Confocal imaging

For confocal imaging, animals were mounted on a 5% agarose pad in 100 mM levamisole. Animals were imaged on an LSM700 confocal microscope with Zen 2010 software except for Figure 6D, which were imaged on an LSM800 confocal microscope using Zen 2.6 (Blue) software.

Imaging PALS-22/25::GFP, mitochondria, and co-localization of PALS-22/25

For Figures 1A, 1B, and 2: Synchronized L1 animals expressing either jyEx193[pals-22p::pals-22::GFP::3xFLAG] or jyEx237[pals-22p::pals-25::GFP::3xFLAG] were plated to 6-cm RNAi plates seeded with either L4440 control vector, pals-22(RNAi), or pals-25(RNAi) and were incubated at 20 °C for 72 h. MitoTracker Red CMXRos (Invitrogen M7512) staining of the C. elegans epidermis was performed using a 1 mM MitoTracker stock solution in DMSO diluted to 5 μM in M9, and 200 μL of 5 μM MitoTracker was then aliquoted into a 1.5 mL microcentrifuge tube for each strain and/or RNAi condition. ∼50 adult animals per strain/condition were picked into 5 μM MitoTracker and were incubated in the dark at room temperature for 10 min with mixing by intermittent inversion. The worms were then pelleted by gravity settling, the supernatant was removed, 1 mL of M9 was added to the tube, and the worms were gently vortexed. The gravity settling and M9 wash procedure was repeated twice (3 washes total), and the worms were transferred to 10-cm NGM + OP50-1 plates. The worms were left to recover from MitoTracker treatment for 2 h in the dark at 20 °C. Localization of PALS-22::GFP or PALS-25::GFP with MitoTracker Red in the epidermis following the different RNAi treatments was assessed with confocal imaging. For Figure 1D: Synchronized L1 animals expressing either jyEx193[pals-22p::pals-22::GFP::3xFLAG] or jyEx237[pals-22p::pals-25::GFP::3xFLAG] were plated to 6-cm RNAi plates seeded with either L4440 control vector, fzo-1(RNAi), eat-3(RNAi), or drp-1(RNAi) and were incubated at 20 °C for 72 h. Localization of PALS-22::GFP or PALS-25::GFP in the epidermis following the different RNAi treatments was assessed by confocal imaging. For Figure S2B: L4 animals were picked from a mixed stage population of ERT1255 grown on 6-cm NGM + OP50-1 plates and co-localization was assessed by confocal imaging. For Figure S5, animals were imaged on a Zeiss AxioImager M2 with Apotome 3 sectioning.

Molecular cloning and transgenesis

Plasmid pET787[pals-22p::gfp::pals-22::3xflag::sl2::3xHA::wrmScarlet::pals-25::unc-54-3’utr] was generated using Gibson assembly of gBlocks containing gfp::pals-22::3xflag::sl2 and sl2::3xHA::wrmScarlet::pals-25 with a vector backbone containing pals-22p and unc-54 3′UTR. pET789 was generated through a Q5 deletion protocol using pET729 as a template,15 deleting residues 2–41. pET792 was generated by Q5 deletion protocol using pET790 [pals-22p::pals-22cDNA::sl2::pals-25cDNA::wrmScarlet::3xHA::unc-54 3′UTR] deleting residues 140–305. pET793 was generated by deletion of pET791[pals-22p::pals-22::3xflag::sl2::3xHA::wrmScarlet::pals-25::unc-54-3’utr] deleting residues 2–139. wrmScarlet is a C. elegans codon-optimized mScarlet. For pET803, eGFP from a TransgeneOme fosmid was cloned and inserted into a construct containing pals-22p::pals-22::sl2::pals-25::unc-54-3’utr using Gibson assembly. For pET847, and pET848, vectors containing the desired construct without mScarlet were generated with Q5 deletion from pET729 and pET789 respectively, and eGFP was cloned from pET803 and using Gibson assembly. For pET845, 846, and 849, the template used was pET803 and a Q5 deletion protocol was used. All plasmids were sequence confirmed using Primordium and.gb files are available (see Supp Data). Gibson assembly protocol volume calculator was obtained at (https://barricklab.org/twiki/bin/view/Lab/ProtocolsGibsonCloning).

Microinjection of C. elegans

DNA for injection was prepared by overnight culture of 4 mL LB + 100 ng/μL carbenicillin and cultures were grown for ∼17 h. This overnight culture was miniprepped using the Qiagen QIAprep Spin Miniprep Kit following manufacturer protocols with the exception that no RNAse was added to buffer P1. Injection mixes were prepared to a final total [DNA] of 100 ng/μL in 10 μL. For pET787, pET792 and pET793, the construct was injected at a concentration of 20 ng/μL and co-injected with a green neuronal marker (gcy-8p::GFP) at 20 ng/μL and pBluescript at 60 ng/μL pET789 was injected at 60 ng/μL with gcy-8p::GFP at 20 ng/μL and pBluescript at 20 ng/μL. For Figures 3C–3G: At least two independent transgenic lines per construct were assessed. For Figure 3B: pET729 was 60 ng/μL and gcy-8p::GFP at 20 ng/μL, and pBluescript at 20 ng/μL. For S5A-E. For S5A,B,D and E GFP constructs were injected at 60 ng/μL and S5C was injected at 40 ng/μL. All lines in S5 were co-injected with gcy-8p::GFP at 20 ng/μL and pBluescript added to generate 100 ng/μL total DNA concentration. For Figure S5, at least three independent transgenic lines per construct were assessed.

Imaging colocalization of PALS-25 variants with mitochondria

Approximately 1-day-old adult animals were selected from a mixed stage population grown on 6-cm NGM + OP50-1 plates. For ERT1368, 1369, 1372, 1373 co-localization was assessed by confocal imaging on a Zeiss LSM700. For ERT1495, 1496,1497,1498,1499,1500 co-localization was assessed on a Zeiss AxioImager M2 with Apotome 3 sectioning.

RNA extraction and RT-qPCR

For Figures S3A and S3B: ∼3000-3,500 synchronized L1s for ERT964 pals-22 pals-25(jy80); jyEx193[pals-22p::pals-22::GFP::3xFLAG] or ERT965 pals-22 pals-25(jy80); jyEx237[pals-22p::pals-25::GFP::3xFLAG] were plated to 10-cm RNAi plates seeded with 1 mL overnight culture of L4440 control vector or pals-22 RNAi clones and were incubated at 20 °C for 48 h. For Figure S3C: ∼1000 synchronized L1s of ERT714 pals-22 pals-25(jy80), ERT751 pals-25(jy111), ERT964 pals-22 pals-25(jy80) unc-119(ed3) III; jyEx193[pals-22::**GFP::3xFLAG, unc-119(+)], and ERT965 pals-22 pals-25(jy80) unc-119(ed3) III; jyEx237[pals-25::EGFP::3xFLAG, unc-119(+)] were plated onto two separate 10-cm plates and incubated at 20 °C for 50 h. For all RT-qPCR experiments, RNA was extracted with TRI reagent (Molecular Research Center TR118), isolated with BCP reagent (Molecular Research Center BP151), washed with 100% isopropanol followed by 75% ethanol, and resuspended in HyClone nuclease-free H_2_O (Cytiva SH30538.03). cDNA was synthesized from total RNA using iScript (Bio-Rad 1708890). qPCR was performed with iQ SYBR Green Supermix (Bio-Rad 1708880) with a CFX Connect Real-Time PCR Detection System (Bio-Rad). Expression values for genes of interest were normalized to the expression of a control gene, snb-1, which does not show altered expression following IPR activation.49 The Pffafl method was used for quantification.55 Primers used for RT-qPCR analysis are provided in Table S2.

Protein collection

For Figures S4B and S4C: ∼1500–2000 synchronized L1s were plated onto RNAi plates seeded with either L4440 control vector, pals-22, or pals-25 RNAi and were incubated at 20 °C for 72 h. Worms were washed twice with ice-cold PBS +0.1% Tween 20 (PBS-T), and following the final wash were gravity settled on ice. A 15 μL aliquot of the dense worm pellet was then transferred to a 1.5 mL microcentrifuge tube and was combined with 20 μL of 4× Laemmli sample buffer (Bio-Rad #1610747) supplemented with 200 mM dithiothreitol (DTT), 25 μL ice-cold PBS-T, and mixed by vortexing. The tubes were incubated at 100 °C for 15 min to digest the worms with gentle vortexing at each 5-min interval. The protein lysates were frozen at −30 °C prior to SDS-PAGE and Western blot. For Figure 3E: PALS-25 fragment expressing strains, two separate 10-cm plates containing 15 transgenic positive L4s were plated. 72 h after plating P0, 1 mL of 10× concentrated OP50 was added to each plate. 24 h later, worms were washed off the plates with M9 + 0.1% Tween 20 (M9T) into a 15 mL conical tube, centrifuged at 3,000 rpm for 30 s to pellet the worms, and the supernatant was removed. Animals were washed twice with 15 mL of M9T, centrifuging at 3,000 RPM for 30 s to pellet worms and remove supernatant each time. After the final wash, supernatant was aspirated down to 1 mL and worms + M9T was transferred to a microcentrifuge tube and worms were then spun in a minifuge for 20 s. Supernatant was aspirated down to 100 μL. 33 μL of 4× Laemmli sample buffer (Bio-Rad #1610747) supplemented with 400 mM dithiothreitol (DTT) was mixed by vortexing. The tubes were incubated at 100 °C for 15 min to digest the worms with gentle vortexing at each 5-min interval. The protein lysates were frozen at −30 °C prior to SDS-PAGE and Western blot.

Western blot analysis

For Figures S4B and S4C: Protein lysates were thawed from −30 °C on ice, centrifuged at 13,000 rpm for 10 min, and 10 μL per sample were separated on a 4–15% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) precast gel (Bio-Rad #4561086) at room temperature. Proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane at 4 °C, 100 V, for 1.5 h. Blocking was performed in 5% nonfat dry milk dissolved in PBS-T for 2 h at room temperature. The PVDF membrane was then incubated with primary antibodies diluted in 5 mL blocking buffer overnight at 4 °C with rocking. Custom polyclonal anti-PALS-22 and anti-PALS-25 were developed and used at 1:1,000. Commercial monoclonal anti-α-tubulin produced in mice (Millipore Sigma, Cat#T9026) was used at 1:4,000. Following incubation in primary antibody, the PVDF membrane was washed three times in PBS-T, and then incubated with secondary antibodies conjugated with horseradish peroxidase in 5 mL blocking buffer for 1.25 h at room temperature with rocking (for anti-PALS-22 and anti-PALS-25 primary: goat anti-rabbit 1:10,000 (Millipore Sigma, Cat#401315-2 ML) For anti-tubulin primary: goat anti-mouse 1:10,000 (Millipore Sigma, Cat#401215-2 mL). The PVDF membrane was washed three times in PBS-T, then was treated with enhanced chemiluminescence (ECL) reagent (Cytiva RPN2209) for 5 min and imaged using a Chemidoc XRS+ with Image Lab software (Bio-Rad). Quantification of PALS-22::GFP or PALS-25::GFP band intensities was determined using Image Lab software (Bio-Rad), and each sample was normalized to its tubulin expression levels. For Figure 3E: Protein lysates were thawed from −30 °C on ice, centrifuged at 15,000 rpm for 15 min, and 20 μL per sample were separated on an Any kD Mini-PROTEAN TGX Precast Protein Gel 10-wells with 7 μL of Precision Plus Protein Dual Color Standards reference band ladder. Protein standards and each sample was loaded twice. The gel was run at 120 V for 1 h and 30 min. Proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane at 4 °C, 30 V, for 18 h. The PVDF membrane was cut in half such that each half had a reference ladder and one lane for each sample. Blocking was performed as mentioned above. Each half of PVDF membrane was then incubated with primary antibodies diluted in 5 mL blocking buffer overnight at 4 °C with rocking (1:1000 for Anti-HA (Cell Signaling, Cat #C29F4), and 1:5000 for anti-tubulin). the PVDF membrane was washed three times in PBS-T and then incubated with secondary antibodies conjugated with horseradish peroxidase in 5 mL blocking buffer for 2 h at room temperature with rocking (for anti-HA: goat anti-rabbit 1:10,000; for anti-tubulin primary: goat anti-mouse 1:1,000). The PVDF membrane was washed and imaged as mentioned above.

AlphaFold and ChimeraX visualization

AlphaFold DB version 2022-11-01, created with the AlphaFold Monomer v2.0 pipeline. ChimerX-1.4 was used to visualize.pdb files.

Measuring IPR induction in PALS-25 fragment-expressing strains

For Figure 3D: ∼500 synchronized L1s were plated onto two separate plates for strains ERT930, ERT1008, ERT1010, ERT1331, ERT1332, ERT1334, ERT1335, ERT1301, ERT1303 and incubated at 20 °C for 72 h. Plates were then washed off and pooled per genotype with 3 mL of M9T and collected into a 1.5 mL conical tube. Each tube was then washed 3 times with M9T, and allowed to gravity settle before aspirating down to 100 μL final volume. After final aspiration, 100 μL of M9T+ 10 μM levamisole was added. 96 clear bottom black well plates were used, and 4× wells per geneotype were filled with 200 μL of 10 μM levamisole, and animals were added to each well, optimizing volumes to obtain a high # of worms with little to no overlap of adult animals. The final volume per well was 300 μL. Samples were analyzed for *pals-5p::*GFP expression on an ImageXpress Nano plate reader using the 4× objective (Molecular Devices, LLC, San Jose, CA) and MetaXpress 2018 software version 6.5.2.351. The worm area was traced using FIJI software (excluding the head regions due to gcy-8p::GFP co-injection marker expression), and the average fluorescence intensity of each worm was quantified with the background fluorescence of the well subtracted. 15 animals per genotype were quantified for each experimental replicate, and three independent experiments were performed.

Imaging the mitochondria morphology of pals-22 mutants

For Figure 4: Wild-type animals expressing mgIs48[ges-1p::mito-GFP] or juEx4796[col-19p::mito-GFP] were grown for 48 h at 20 °C to the young adult life stage. pals-22(jy1) and pals-22(jy3) mutants, which are developmentally delayed,11 expressing mgIs48 or juEx4796 were grown for 54 h at 20 °C to the young adult life stage. GFP-labeled mitochondria were assessed by confocal imaging. In the intestine, imaging was restricted to the first ring (four anterior-most cells) in the same focal plane as the intestinal nuclei determined by the absence of GFP signal within the nuclei. In the epidermis, images were collected by first identifying alae in the cuticle using transmitted light in the anterior half of the worm. Next, a focal plane underneath the alae was identified using GFP fluorescence such that seam cells and surrounding hyp7 cells had GFP-labeled mitochondria in focus. To minimize changes in mitochondrial network morphology as a result of prolonged immobilization on the agar pad and/or exposure to levamisole, images were only collected within 10 min after mounting before a new slide was prepared. Confocal imaging settings were independently optimized for each transgene based on the expression in wild-type controls. See below for morphology analysis. For Figure S6B: Wild-type, pals-22(jy3), and pals-22 pals-25(jy80) adults were picked from 6-cm NGM + OP50-1 plates into a 1.5 mL microcentrifuge tube containing 200 μL of 2.5 μM MitoTracker Red diluted in M9. The worms were incubated in the dark for 10 min with mixing by intermittent inversion. The worms were then pelleted by gravity settling, the supernatant was removed, 1 mL of M9 was added to the tube, and the worms were gently vortexed. The gravity settling and M9 wash procedure was repeated twice (3 washes total), and the worms were transferred to 6-cm NGM + OP50-1 plates. The worms were left to recover from MitoTracker treatment for 2 h in the dark at 20 °C prior to confocal imaging.

Imaging IPR activation and mitochondria morphology following auxin-mediated depletion of PALS-22

For Figures 5 and S8: 6-cm NGM plates supplemented with either 200 μM auxin or vehicle control (0.15% ethanol) were seeded with 200 μL of 10× concentrated OP50-1 spread across the entire surface of the plate. For intestinal analysis, synchronized L1s of ERT1017 and ERT1187 were plated on NGM + OP50-1 and incubated at 20 °C for 40 h. A population of worms for each strain was then split and plated to auxin or vehicle treatments. Worms were then collected from each condition to collect a total of 8 time points (40–47 h post L1, 0–7 h on treatment). For epidermal analysis, synchronized L1s of ERT1224 and ERT1017 were plated to NGM + OP50-1 and incubated at 20 °C for 48 h or plated directly on auxin or vehicle plates. A population of worms for each strain plated on NGM + OP50-1 was split and plated on auxin or vehicle plates. All conditions were removed from their plates at specified time points: 0, 3, and 48 h on auxin. For all experiments, animals were removed immediately at specified time point and were mounted and imaged for confocal imaging as described above. See below for mitochondrial morphology analysis.

Imaging mitochondrial morphology in pals-17 and pals-22 RNAi-treated animals

For Figure S6: ∼250 synchronized L1s of ERT1145 mgIs48[ges-1p::mito-GFP] and ERT054 jyIs8[pals-5p::gfp, myo-2p::mCherry]X were each plated onto 4 × 6-cm OP50 plates and incubated at 20°C for 20 h. For each strain, two separate plates were washed off and pooled with 2 mL of M9 and transferred into 1.5 mL conical tubes. The worms were then washed 3 times with M9, and pelleted down to 100 μL final volume. Each 1.5 mL conical was then split onto two separate plates of either pals-17 RNAi or L4440 EV RNAi. Worms were then grown for 24 h. Worms were then imaged using the confocal as described above. At least five animals treated with pals-17 RNAi and five animals treated with L4440 RNAi were imaged. Five animal samples were randomly selected per treatment, and two regions of interest were chosen from each animal. Downstream morphology analysis and quantification were performed using the protocol below. For Figure 6D: ∼500 synchronized L1’s of ERT1145 mgIs48[ges-1p::mito::GFP] and ERT1501 zip-1(jy13) III; mgIs48[ges-1p::mito::GFP] were plated onto a 10 cm RNAi plate seeded with 750 μL of 4× concentrated RNAi overnight culture grown for 48–72 h of either pals-22 RNAi or L4440 EV RNAi. Animals were grown for 72 h at 20°C, then mounted on agar pads and imaged on the LSM800 for mitochondria morphology imaging.

Imaging mitochondrial morphology in wild-type and pals-22 mutants following RNAi mediated knockdown of mitochondrial morphology regulating factors

For Figure S6A: ∼100 synchronized L1s of juEx4796[col-19p::mito-GFP] were plated on EV or pals-22 RNAi plates seeded with 300 μL of RNAi culture. Mitochondria morphology was assessed with confocal imaging at 72 h; images were only collected within 10 min after mounting before a new slide was prepared. For Figures S6D and S6E: ∼300 synchronized L1s of ERT1145 and ERT1146 were plated on EV, drp-1, or pink-1 6-cm RNAi plates seeded with 300 μL of RNAi culture, and mitochondrial morphology was assessed with confocal imaging at 48 and 72 h.

Mitochondria morphology analysis

Confocal images of intestinal and epidermal mitochondria labeled with GFP were analyzed with FIJI software.52 For each animal, 10 μM × 10 μM square regions of interest (ROIs) were selected for analysis. Several pre-processing steps in FIJI were performed to enhance mito-GFP signal to background noise, including, in order, (a) subtract background, (b) sigma filter plus, (c) enhance local contrast, and (d) gamma correction. The processed ROI was masked using the Mitochondria Analyzer plugin (https://github.com/AhsenChaudhry/Mitochondria-Analyzer/tree/master). Slightly different masking conditions were used for pals-22 mutants compared to pals-22 and pals-17 RNAi treatments. For more information see a step-by-step protocol for image pre-processing and analysis with Mitochondria Analyzer at GitHub (https://github.com/TroemelLabUCSD/Mitochondria-and-PathogenLoad-Analysis), Fiji Macros linked here used for Figures 5C–5E, 5G, 6C, and 6D. Briefly, Mitochondria Analyzer 2D analysis was performed on a per-ROI basis (i.e., morphology characteristics for each mitochondrion in the ROI were quantified and then averaged for the number of mitochondria detected in the ROI). For analysis of intestinal mitochondria, ROIs selected in the first ring were not cell-specific and, in some cases, spanned adjacent cells. Likewise, for analysis of epidermal mitochondria, ROIs spanned both seam cells and surrounding hyp7 in the anterior half of the animal. The ‘Form Factor’ is a shape measurement where 1 equals a perfectly round 2D object and increases with object elongation.

Analysis of mitoUPR and IPR upregulated genes

Genes significantly upregulated (logFC >0, p < 0.05) by the atfs-1(et15) gain-of-function allele, which constitutively activates the mitoUPR,25 were compared to the 80 canonical IPR genes11 using WormBase IDs with the tool: https://bioinformatics.psb.ugent.be/webtools/Venn/.

N. parisii transmission electron microscopy and pathogen load assays

C. elegans were infected with N. parisii spores at 25 °C and then fixed at the late meront/early sporont stage, approximately 40 h post-inoculation for transmission electron microscopy as described.47 To perform RNAi for Figure 7B, overnight cultures were prepared at 30 °C for 19.5 h in 4 mL LB broth supplemented with μg/ml carbenicillin. RNAi plates were prepared for each experimental replicate by plating 750 μL of 4× concentrated RNAi overnight culture for L4440, pals-22, drp-1, and fzo-1. 15 L4 P0 N2 animals were plated onto two separate RNAi plates. 1 mL of 6× concentrated RNAi overnight culture that was grown at 30 °C for 19.5 h and induced with a final concentration of 2 mM IPTG for 2 h at 30 °C was added to plates ∼72 h after plating P0s. F1 worms were then bleached/synchronized ∼96 h after plating P0s to obtain F2s. N. parisii spores were prepared as described previously.56 ∼1200 F2 progeny were then plated on 1 × 10-cm RNAi plates. 45 h after plating, animals were washed with M9 into a 15 mL centrifuge tube and spun at 3,000 rpm for 30 s. Supernatant was aspirated, and worms were then transferred to a 1.5 mL microcentrifuge tube. All samples were washed with M9 and supernatant aspirated down to 100 μL and 1 mL M9 was added then spun down on minifuge for 20 s. This wash step was performed three times. Then, 100 μL of room temperature 10× OP50, 146 μL of M9 and 3.9 μL of N. parisii spores (∼20 × 10ˆ6 spores) were mixed with each sample and plated onto pre-dried 6-cm unseeded NGM plates and spread evenly across the plate. Plates were left to dry in laminar flow hood for 30 min, then moved to 25 °C for 3 h. All samples were then washed off 6-cm plates, washed with M9 three times, and plated onto fresh 10-cm RNAi plates, which were then incubated at 25 °C for an additional 27 h. Worms were then fixed in acetone and stained with FISH probes conjugated to the red Cal Fluor 610 fluorophore that hybridize to N. parisii ribosomal RNA47 (Biosearch Technologies, Hoddeson, United Kingdom) and incubated at 47 °C overnight (16–18 h). Animals were then washed and imaged using the ImageXpress Nano plate reader using the 4× objective (Molecular Devices, LLC, San Jose, CA). Worms were automatically segmented using the protocol/code located at GitHub (https://github.com/TroemelLabUCSD/Mitochondria-and-PathogenLoad-Analysis) using MetaXpress 2018 software version 6.5.2.351.

Quantification and statistical analysis

Statistical analysis was performed using Prism 9 & 10 software (GraphPad). Before statistical analysis, the D’Agostino & Pearson and Shapiro-Wilk tests were performed for all experiments to assess the distribution of the data. If data were normally distributed, standard parametric statistical tests were performed. Nonparametric tests were performed for non-normally distributed data, as determined by either the D’Agostino & Pearson or the Shapiro-Wilk tests. The corresponding figure legends describe the statistical test for each experiment, the number of data points analyzed, and the number of experimental replicates performed. Asterisks are defined in each relevant figure legend, together with the name of the statistical test.