C. elegans fmo-4 shares significant similarities to its family member and longevity gene, fmo-2, including 88% conservation in catalytic residues (Figure 1—figure supplement 1), a transmembrane domain (Figure 1A) and predicted subcellular localization in the ER (Figure 1—figure supplement 2; Kishore et al., 2020). Additionally, fmo-4 is induced by the longevity intervention DR (Rossner et al., 2017). These structural similarities and DR-mediated induction led us to hypothesize that fmo-4 may play a role in aging. To test this, we first asked whether fmo-4 is required for well-studied longevity pathways, including DR. We utilized fmo-4(ok294) knockout (KO) animals (C. elegans Deletion Mutant Consortium, 2012) on five conditions reported to extend lifespan in C. elegans. Since fmo-4 is induced by dietary restriction, we started with fed and DR (sDR Greer and Brunet, 2009) conditions. Our results show that loss of fmo-4 has no significant effect on control-fed worms but prevents nearly all the lifespan extension seen under DR (Figure 1B). Wild-type (WT) worms on DR experience a ~35% lifespan extension compared to fed WT worms, but when fmo-4 is knocked out this extension is reduced to ~10% and this interaction is significant by Cox regression (p-value <4.50e^–6). These data support that fmo-4 is required for the DR longevity pathway (Figure 1B). Having established this role, we continued lifespan analyses of fmo-4 KO worms exposed to RNAi knockdown of the S6-kinase gene rsks-1 (mTOR signaling; Pan et al., 2007), the Von Hippel-Lindau gene vhl-1 (hypoxic signaling; Leiser et al., 2013), the insulin receptor daf-2 (insulin-like signaling; Kimura et al., 1997), and the cytochrome c reductase gene cyc-1 (mitochondrial electron transport chain, cytochrome c reductase; Lee et al., 2010; Figure 1C–F). The resulting data show that fmo-4 is fully required for rsks-1-mediated longevity as marked by a complete abrogation of the rsks-1 RNAi lifespan extension when fmo-4 is knocked out (Figure 1C), placing fmo-4 downstream of mTOR. fmo-4 was not required for the hypoxic response (Figure 1D), insulin-like signaling (Figure 1E), or cytochrome c reductase inhibition pathways (Figure 1F) to increase lifespan. It is notable that, unlike fmo-4, fmo-2 is required for vhl-1-mediated longevity (Leiser et al., 2015). This result, coupled with fmo-4 and fmo-2 each being required for DR, suggests that these two genes in the same family are overlapping but distinct in their requirement. Together, these data suggest that fmo-4 plays a necessary role in C. elegans longevity regulation downstream of at least two pathways: DR and mTOR signaling.

Having established that, like fmo-2, fmo-4 is required for DR-mediated longevity, we next hypothesized that fmo-2 and fmo-4 may be acting in the same pathway. To test this, we crossed fmo-2 overexpressing (OE) worms with fmo-4 KO worms to create an fmo-2 OE; fmo-4 KO strain (fmo-2OE;4KO). We validated this strain via qPCR analysis (Supplementary file 1). Upon measuring their lifespan compared to fmo-2 OE animals, we find that knocking out fmo-4 completely abrogates the lifespan extension from fmo-2 OE (Figure 2A). This result suggests that fmo-4 is required for fmo-2-mediated longevity and thus may act downstream of fmo-2 to promote longevity in C. elegans.

We previously showed that fmo-2 OE is not just sufficient for lifespan extension but is also sufficient to promote resistance to multiple forms of stress, including oxidative stress (paraquat), heat stress, and ER stress (tunicamycin; Leiser et al., 2015). Since fmo-4 is required for fmo-2-mediated longevity, we hypothesized that fmo-4 would also be required for fmo-2-mediated stress resistance. Following exposure to 5 mM paraquat, 37 °C heat stress, or 5 µg/mL tunicamycin, we find that the fmo-2OE;4KO worms survive similarly to WT worms and are no longer resistant to these stresses (Figure 2B–D). Thus, knocking out fmo-4 blocks fmo-2-mediated broad stress resistance.

Healthspan is a crucial component to longevity, and we previously found that fmo-2 OE is sufficient to promote healthspan benefits with age in C. elegans (Leiser et al., 2015). As fmo-4 is required for both the lifespan extension and stress resistance observed with fmo-2 OE, we hypothesized that fmo-4 would also be required for fmo-2-mediated healthspan benefits. To test this, we measured the thrashing rate of fmo-2 OE;4KO worms at day 2 and day 10 of adulthood. We find that knocking out fmo-4 abrogates the healthspan benefits of the fmo-2 OE worms with age (Figure 2E, F). Thus, fmo-4 is required for fmo-2-mediated health and resilience benefits, further supporting its acting downstream of fmo-2 to promote health and longevity.

While being necessary for increased healthspan, lifespan, and stress resistance shows that fmo-4 is required for these benefits, it does not test whether it can modulate aging directly. Therefore, to better understand fmo-4’s role in longevity and health, we next asked whether fmo-4 is sufficient for lifespan extension, stress resistance, and healthspan benefits. We created a ubiquitous fmo-4 OE worm strain with fmo-4 expressed under the eft-3 promoter via multicopy extrachromosomal array followed by random integration. qPCR analysis shows that fmo-4 is overexpressed ~150 fold over WT in the ubiquitous fmo-4 OE worms (Supplementary file 1), and their development time is slightly (~1.5 hr) but statistically significantly delayed compared to WT (Figure 3—figure supplement 1). Interestingly, we find that ubiquitous fmo-4 OE is sufficient to reproducibly extend lifespan by ~20% (Figure 3A). Epistatic analysis with fmo-2 shows that fmo-4 OE lifespan extension is not additive with fmo-2 OE (Figure 3—figure supplement 2A), as determined by comparing the individual OE strains to the fmo-2 OE;fmo-4 OE (fmo-2OE;4 OE) worm strain (Supplementary file 1). This further supports that these two genes in the same family act in the same pathway. To test whether fmo-4 requires fmo-2, we overexpressed fmo-4 in the context of fmo-2 knockout (fmo-4 OE;2KO) (Supplementary file 1). The results show that fmo-4 OE does not require fmo-2 for its longevity benefits (Figure 3—figure supplement 3A). This further validates that fmo-4 likely acts downstream of fmo-2.

Increased longevity is frequently observed in tandem with increased healthspan and resistance to toxic stress (Soo et al., 2023). We assessed the healthspan of the ubiquitous fmo-4 OE worms at days 2 and 10 of adulthood by measuring thrashing rates (movement). We find that ubiquitous fmo-4 OE is sufficient for an improvement (p=0.011) in healthspan benefits with age (Figure 3B, C). We next measured the stress resistance of the ubiquitous fmo-4 OE worms to 5 mM paraquat (oxidative stress), 37 °C heat stress, 5 µg/mL of tunicamycin (ER glycosylation stress), and 1 mg/mL thapsigargin (ER calcium stress). We included two ER stresses because FMO-4 is predicted to be an ER transmembrane protein. We find that ubiquitous fmo-4 OE is sufficient to convey paraquat stress resistance (Figure 3D) but does not offer any protection from heat (Figure 3E) or tunicamycin stress (Figure 3F). Interestingly, ubiquitous fmo-4 overexpressing animals are sensitive to thapsigargin, as shown by the worms’ significantly diminished size and inability to develop compared to WT control worms (Figure 3G–J). These results suggest a narrow stress resistance and even some sensitivity for fmo-4 in comparison to other longevity genes. To test whether fmo-4 OE would interact with fmo-2 OE, we combined the strains and measured their stress resistance. We find that the fmo-2 OE strain is still resistant to paraquat, heat, and tunicamycin stress in the presence of increased fmo-4, suggesting that overexpression of fmo-4 does not affect resistance to these stresses (Figure 3—figure supplement 2B–D). Interestingly, the fmo-2 OE strain is sensitive to thapsigargin stress in the presence of increased fmo-4, suggesting that fmo-4 likely also acts downstream of fmo-2 in this stress pathway (Figure 3—figure supplement 2E–I). Overall, these results suggest that fmo-4 (1) is sufficient to improve healthspan, (2) is both necessary and sufficient to promote paraquat resistance, (3) is necessary but not sufficient to improve heat and tunicamycin resistance, and (4) promotes sensitivity to calcium-mediated ER stress from thapsigargin. As expected under the hypothesis that fmo-4 acts downstream of fmo-2, the fmo-4OE;2KO strain phenocopies the fmo-4 OE strain in all stress assays (Figure 3—figure supplement 3B–I). Together, these data suggest that fmo-4 OE is sufficient to promote lifespan extension and paraquat stress resistance as well as a healthspan benefit, while exhibiting sensitivity to ER calcium stress.

While using the ubiquitous overexpressing worm strain is helpful to probe into fmo-4’s involvement in longevity, stress resistance, and healthspan, it is also important to note that fmo-4 is normally localized and expressed primarily in the hypodermis (Figure 4A–C; Petalcorin et al., 2005). Thus, we asked whether hypodermal-specific fmo-4 overexpression is sufficient for these health benefits. We created a hypodermal-specific fmo-4 OE worm strain expressing fmo-4 under the dpy-7 promoter via multicopy extrachromosomal arrays followed by random integration. qPCR data show that fmo-4 is expressed ~45 fold over WT in the hypodermal-specific fmo-4 OE worms (Supplementary file 1), and their developmental time does not differ from WT (Figure 3—figure supplement 1). We measured lifespan, resistance to paraquat, heat, tunicamycin, and thapsigargin stress, as well as the healthspan of the hypodermal-specific fmo-4 OE worms (fmo-4 OE^Hyp). We find that the fmo-4 OE^Hyp strain exhibits a lifespan extension (Figure 4D), resistance to paraquat (Figure 4E) but not heat or tunicamycin stress (Figure 4F, G), and sensitivity to thapsigargin stress (Figure 4H–J). However, the fmo-4 OE^Hyp strain did not have a statistically significant effect on healthspan unlike the ubiquitous fmo-4 OE strain (Figure 4K, L). Overall, these results agree with the C. elegans single-cell atlas as well as previously published work (Petalcorin et al., 2005) showing that fmo-4 is mostly localized to the hypodermis (Figure 4A) and is consistent with the ubiquitous strain primarily benefiting from expressing additional fmo-4 in the hypodermis. Thus, ubiquitous or hypodermal-specific overexpression of fmo-4 is sufficient for longevity and paraquat stress resistance (Figures 3A, D, 4D, E), does not affect heat or tunicamycin resistance (Figure 3E, F, Figure 4F, G), and results in thapsigargin sensitivity (Figures 3G–J–4H–J).

To identify the downstream effects of fmo-4 expression, we analyzed the transcriptome of fmo-4 OE animals and compared them with control animals (Figure 5A). Based on a p-value of <0.05, we find ~800 transcripts are upregulated and ~500 transcripts are downregulated when fmo-4 is overexpressed (Source data 1). Of the upregulated and downregulated transcripts unique to the fmo-4 OE profile, we noticed that some of the significant pathways include transcription, Wnt signaling, TGFβ signaling, protein processing in the ER, and other subsets of signaling, as determined by the Database for Annotation, Visualization and Integrated Discovery (DAVID) Functional Annotation Bioinformatics Microarray Analysis (Figure 5A). Based on its known role as a xenobiotic enzyme, we were intrigued to find that fmo-4 OE is modulating signaling pathways. We were further intrigued to find that one of the signaling pathways affected by fmo-4 is calcium signaling (Source data 2). This was interesting considering that fmo-4 OE worms are sensitive to thapsigargin (Figure 3G–J), implicating an interaction between calcium signaling and fmo-4. To further verify this possibility, we used another tool called PANTHER Classification System Protein Class, which identified FMO-4 as a putative transmembrane signal receptor (G-protein coupled receptor), ion transporter, and/or calcium-binding protein (Source data 2). Together, these results support the possibility that FMO-4 is involved in calcium-related processes.

To begin exploring this interaction, we determined how changes in intracellular calcium levels impact fmo-4 expression and lifespan. We first manipulated calcium levels by supplementing the acetylcholine agonist, carbachol. Carbachol activates acetylcholine receptors, increasing overall intracellular calcium levels (Masoumi et al., 2023). We first validated that carbachol increases calcium levels by utilizing worms that neuronally express the calcium indicator GCaMP7f (SWF702; Dag et al., 2023). When these worms were exposed to 300 µM carbachol, a significant increase in GFP fluorescence was reported, indicating an increase in calcium levels (Figure 5—figure supplement 1A–D). We then measured fmo-4 gene expression upon exposure to carbachol using an fmo-4p::mCherry transcriptional reporter strain. Interestingly, we find that supplementation of 300 µM carbachol induces fmo-4 promoter expression fluorescence nearly twofold over a water control (Figure 5B, C). This was similar to the level of induction of fmo-4 in fasted (DR-like) conditions (Figure 5B, C). We postulate that fmo-4 is induced by carbachol because increased intracellular calcium activates fmo-4 gene expression. Since carbachol supplementation induces fmo-4, we hypothesized that carbachol may affect lifespan and would interact with fmo-4. We measured the lifespan of worms supplemented with 50 µM carbachol and find that while WT lifespan is significantly extended by carbachol, this extension is not additive with fmo-4 OE (Figure 5D), and requires fmo-4, as the fmo-4 KO is shorter-lived when exposed to carbachol (Figure 5E).

To test whether reducing calcium has an opposing effect to increasing it, we utilized EthyleneDiamineTetraAcetic acid (EDTA) to deplete calcium levels (Mellau and Jørgensen, 2003). We validated that EDTA depletes calcium levels by exposing GCaMP7f expressing worms (Dag et al., 2023) to 10 mM EDTA and we observed a significant decrease in GFP fluorescence, indicating a decrease in calcium levels (Figure 5—figure supplement 1A, B, E, F). Then, we placed fmo-4p::mCherry reporter worms on plates containing 10 mM EDTA and measured fluorescence. Surprisingly, we find that EDTA also significantly induces fmo-4 expression (Figure 5B, C). We then assessed the survival of WT, fmo-4 OE and KO worms on plates supplemented with 50 µM EDTA. We find that EDTA significantly extends WT lifespan without further extending fmo-4 OE lifespan (Figure 5F), while shortening the lifespan of fmo-4 KO animals (Figure 5G). Overall, these data suggest that fmo-4 is highly sensitive to changing calcium levels in either direction, and that both increasing and depleting calcium (1) induces fmo-4, (2) extends WT lifespan, (3) is not additive with fmo-4 OE lifespan, and (4) is deleterious to fmo-4 KO animals. Thus, these results support an interaction between fmo-4 and calcium signaling.

Having established an interaction between fmo-4 and calcium perturbations, we next asked if fmo-4 interacts with genes involved in calcium signaling. These genes include calreticulin (crt-1), inositol triphosphate receptor (IP₃R, itr-1), and the mitochondrial calcium uniporter (mcu-1; Burkewitz et al., 2020). Calreticulin is responsible for binding and sequestering calcium in the ER lumen (Groenendyk et al., 2021; Park et al., 2001). It was previously shown that FMO protein extracted from rabbit lung can form a complex with calreticulin, suggesting a physical interaction (Guan et al., 1991). When testing for genetic interactions between crt-1 and fmo-4, we find that crt-1 RNAi extends WT lifespan, as previously reported (Burkewitz et al., 2020), and that this lifespan extension is not additive with fmo-4 OE (Figure 6A). Treating fmo-4 KO worms with crt-1 RNAi can significantly extend lifespan (Figure 6—figure supplement 1A), but this result was inconsistent (Source data 3). Overall, this suggests that fmo-4 and crt-1 are acting in the same genetic pathway. IP₃R, or itr-1 in C. elegans, is located in the ER membrane and is critical for exporting calcium from the ER lumen into the cytosol (Groenendyk et al., 2021; Berridge, 1993). Previous results suggest that loss of itr-1 decreases WT lifespan (Burkewitz et al., 2020), which our results confirm (Figure 6B). Importantly, we also find that itr-1 RNAi decreases lifespan of fmo-4 KO worms similarly (Figure 6—figure supplement 1B) and fmo-4 OE (Figure 6B) worms to a greater extent than WT, negating the relative longevity of fmo-4 OE worms compared to WT. Thus, itr-1 is required to increase lifespan downstream of fmo-4. MCU (mcu-1) is located in the inner mitochondrial membrane and allows cytosolic calcium into the inner mitochondrial matrix (Groenendyk et al., 2021; Marchi and Pinton, 2014). We also find that mcu-1 is required downstream of fmo-4 OE, as mcu-1 RNAi abrogates the fmo-4 OE lifespan (Figure 6C). The fmo-4 KO lifespan is slightly decreased on mcu-1 RNAi (Figure 6—figure supplement 1C). These data support that fmo-4 extends lifespan through its interactions with the calcium signaling genes, crt-1, itr-1, and mcu-1.

In addition to lifespan assessment, we were curious whether these genes also interact with fmo-4 expression to modify resistance to paraquat. We exposed WT, fmo-4 OE, and fmo-4 KO worms to 5 mM paraquat plates with RNAi for crt-1, itr-1 or mcu-1 and assessed survival over time. Interestingly, we find that crt-1 RNAi promotes resistance in WT worms and this effect is not additive in fmo-4 OE worms (Figure 6D). Furthermore, crt-1 RNAi extends the lifespan of fmo-4 KO worms (Figure 6—figure supplement 1D), suggesting that crt-1 and fmo-4 act in the same pathway and that crt-1 may act downstream of fmo-4 to promote paraquat stress resistance. Neither itr-1 RNAi nor mcu-1 RNAi confers resistance to paraquat, and fmo-4 OE resistance is lost when exposed to these RNAi (Figure 6E, F). The fmo-4 KO worms show a decrease in paraquat survival when treated with itr-1 and mcu-1 RNAi (Figure 6—figure supplement 1E, F). We also measured paraquat stress resistance of fmo-4 OE and KO worms exposed to RNAi of the voltage-dependent anion channel 1 (vdac-1), which is located in the outer mitochondrial membrane and also regulates calcium flow into and out of the mitochondria (Shoshan-Barmatz et al., 2018). Interestingly, fmo-4 OE resistance is lost when exposed to vdac-1 RNAi and the fmo-4 KO worms show no difference in paraquat survival compared to the EV RNAi control (Figure 6—figure supplement 1G, H). These data support a model where fmo-4 OE converges onto a shared pathway mediating paraquat stress resistance and longevity regulation. Together, we conclude that fmo-4 interacts with ER to mitochondrial calcium signaling through crt-1, itr-1, and mcu-1, to promote longevity and paraquat stress resistance.

Our data establish that fmo-4 interacts with ER and mitochondrial calcium signaling to promote lifespan extension and resistance to paraquat. A previous study linked many of these components to a major regulator of UPR^ER, activating transcription factor-6 (atf-6) (Burkewitz et al., 2020). When atf-6 is lost, lifespan is extended through changes in calcium signaling between the ER and mitochondria requiring crt-1, itr-1, and mcu-1 (Burkewitz et al., 2020). Based on our transcriptomics data and fmo-4’s interactions with calcium signaling genes, we hypothesized that fmo-4 could interact with atf-6 to promote longevity and paraquat resistance. To test whether atf-6 is regulating fmo-4, we utilized our fmo-4p::mCherry transcriptional reporter strain. We measured fmo-4 gene expression after RNAi knockdown of atf-6 and find that fmo-4p::mCherry worms on atf-6 RNAi show a consistent ~two fold increase in fluorescence (Figure 7A, B), similar to what we observe from calcium perturbations (Figure 5B, C). Since atf-6 is one of three branches of UPR^ER, we were curious to see if fmo-4 interacts specifically with atf-6 or also with the other two branches, ire-1/xbp-1 and pek-1/atf-4 (Read and Schröder, 2021). We treated the fmo-4p::mCherry reporter worms with ire-1, xbp-1, pek-1, or atf-4 RNAi and measured fluorescence. We find that knocking down the components of the IRE-1 or PEK-1 branches does not induce fmo-4 gene expression (Figure 7—figure supplement 1A, B). Thus, only knockdown of the ATF-6 branch of UPR^ER induces fmo-4 expression.

We hypothesized that atf-6 limits fmo-4 expression and thus fmo-4 acts downstream of atf-6 knockdown to modulate lifespan extension. To test this, we assessed survival and find that while atf-6 RNAi extends the lifespan of WT worms, as reported (Burkewitz et al., 2020), this effect is abrogated when fmo-4 is knocked out (Figure 7C). This result suggests that fmo-4 is indeed acting downstream of atf-6 to promote longevity. To determine if atf-6 is also involved in fmo-4-mediated paraquat resistance, we exposed WT and fmo-4 KO worms to 5 mM paraquat plates with atf-6 RNAi. We find that atf-6 RNAi promotes stress resistance to paraquat, but this effect is lost when fmo-4 is knocked out, supporting the hypothesis that atf-6 is involved in fmo-4-mediated paraquat resistance (Figure 7D). Together, these data suggest that fmo-4 modulates lifespan and paraquat stress resistance downstream of the reduction in atf-6, ultimately regulating ER to mitochondria calcium signaling (Figure 7E).

Standard C. elegans cultivation procedures were used as previously described (Leiser et al., 2015). Briefly, all worm strains were maintained on solid nematode growth media (NGM) using E. coli OP50 throughout life except where double stranded (ds) RNAi (E. coli HT115) were used. Worms were transferred using a platinum wire. All worm strains were kept at 20 °C. RNAi used is listed in Supplementary file 3. Worm strains are listed in Supplementary file 4. Genotyping primers are listed in Supplementary file 5.

Animals were synchronized by placing 10 gravid adult worms on NGM plates seeded with E. coli OP50 to lay eggs for 1 hr at 20 °C. The gravid adult worms were then removed, and the eggs were allowed to hatch and develop at 20 °C until larval stage 2 (L2). At this point the L2 worms were moved to individual 35 mm NGM plates seeded with E. coli OP50, one worm per plate. This was done for each worm strain tested and ten total 35 mm plates per worm strain were prepared. The worms were then followed through development and watched hourly during young adulthood to score the time of the first egg lay, as previously described (Leiser et al., 2015). Development time was reported in hours since egg lay. Three replicate experiments were performed. For development on DMSO control or thapsigargin, the same protocol was followed except that the L2 worms were moved to individual 35 mm NGM plates seeded with E. coli OP50 that were spotted with either DMSO or 1 mg/mL thapsigargin, one worm per plate.

Animals were synchronized by placing 10 gravid adult worms on NGM plates seeded with E. coli OP50 and allowing them to lay eggs for 2 hr at 20 °C. The gravid adult worms were then removed, and the eggs were allowed to hatch and develop at 20 °C until day 1 adulthood. Worms were placed in a drop of M9 solution, as previously described (Leiser et al., 2015). The individual body bends were counted at maximum rate for 30 s. Thrashing was assayed on day 2 and day 10 of adulthood. The animals that were not used for the day 2 assay were transferred to fresh fed FUdR plates four times until they were ready to be assayed. Three replicates were performed. Data were analyzed in GraphPad Prism using unpaired two-tailed t tests with Welch’s correction as well as one-way ANOVA.

Gravid adults were placed on NGM plates containing 1 mM β-D-isothiogalactopyranoside (IPTG), 25 μg/ml carbenicillin, and the corresponding RNA interference (RNAi) clone from the Vidal or Ahringer RNAi library. 200 µL of HT115 bacteria expressing double stranded (ds) RNA of either the control empty vector (EV) or the RNAi of interest at optical density (OD) of 3.0 and concentration of 3 X was added to each plate. After 3 hr, the adults were removed, and the eggs were allowed to develop at 20 °C until they reached late L4/young adult stage. From here, 70 worms were placed on each RNAi plate and transferred to fresh RNAi +FUDR plates on day 1, day 2, day 4, and day 6 of adulthood. A minimum of two plates per strain per condition were used per replicate experiment. Experimental animals were scored every 2–3 d and considered dead when they did not move in response to prodding under a dissection microscope. Worms that crawled off the plate were not considered, but ruptured worms were considered as previously described (Leiser et al., 2015). A similar method was used for non-RNAi lifespan experiments, except the plates did not contain IPTG and worms were fed E. coli OP50. 200 µL of E. coli OP50 at OD of 3.0 and concentration of 3 X was added to each plate. A similar method was also used for carbachol (Thermo Fisher, L06674.14) and EDTA (Thermo Fisher, AM9260G) supplementation lifespan experiments, except 50 µM carbachol or 50 µM EDTA was added to the NGM plates without IPTG, and worms were fed E. coli OP50. 200 µL of E. coli OP50 at OD of 3.0 and concentration of 3 X was added to each plate. Optimal concentrations of carbachol and EDTA (50 µM) were determined by assessing survivability of worms exposed to a range of concentrations. Log-rank test was used to derive p-value for lifespan assays using p<0.05 cut-off threshold compared to EV or wild-type controls. Cox regression was also used to assess interactions between genotype and condition for lifespan assays sing p<0.01 cut-off threshold compared to controls.

Gravid adults were placed on NGM plates seeded with 200 µL E. coli OP50. After 3 hr, the adults were removed, and the eggs were allowed to develop at 20 °C until they reached late L4/young adult stage. From here, 70 worms were transferred to fed FUdR plates seeded with 200 µL of E. coli OP50 at OD of 3.0 and concentrated 3 X on days 1 and 2 of adulthood. On day 3 of adulthood, DR conditions were transferred to plates with 10⁹ seeded lawns and transferred every other day four times while the corresponding controls were transferred equally to fed ad-lib plates. This form of DR is termed solid DR (sDR; Greer and Brunet, 2009). 80 µL of 10 mM palmitic acid (Sigma-Aldrich) dissolved in 100% EtOH was added to the outer rim of the plate to prevent fleeing. A minimum of two plates per strain per condition were used per replicate experiment. Experimental animals were scored every 2–3 d and considered dead when they did not move in response to prodding under a dissection microscope. Worms that crawled off the plate were not considered, but ruptured worms were considered as previously described (Leiser et al., 2015).

Approximately 600 day 1 adult worms per biological replicate were washed with M9 buffer three times, frozen in liquid nitrogen, and then stored at –80 °C. RNA was extracted by adding 500 mL of Trizol reagent to the frozen worm pellets, followed by three freeze-thaw cycles with liquid nitrogen and water bath at 42 °C. Then, 500 µL of ethanol was added to the samples, and RNA was isolated using the Direct-Zol Miniprep Plus Kit (Cat#R2072). Purified RNA was sent to Novogene (Novogene Corporation Inc) for sequencing on the Illumina HWI-ST1276 instrument. Messenger RNA was purified from total RNA by using poly-T oligo-attached magnetic beads. After fragmentation, first strand cDNA was synthesized using random hexamer primers, followed by second strand cDNA. The library was constructed, which entailed end repair, A-tailing, adapter ligation, size selection, amplification, and purification, and then was sequenced on an Illumina device using paired-end sequencing. Gene ontology analysis was done using the National Institutes of Health DAVID Bioinformatics tool. The dataset has been submitted to GEO NCBI (GEO accession #GSE288007).

RNA was isolated from day 1 adult worms (approximately 600 worms per strain) following three rounds of freeze-thaw in liquid nitrogen using Invitrogen’s Trizol extraction method, similar to the method described above. 1 µg of isolated and purified RNA was reverse transcribed to cDNA using SuperScript II Reverse Transcriptase (18064071, Invitrogen,). Gene expression levels were measured using 600 ng of cDNA and SYBR Green PCR Mastermix (A25742, Applied Biosystems) with primers at 10 μM concentration. mRNA levels were normalized using Y45F10D.4 as a reference gene (Taki and Zhang, 2013). List of qPCR primers used are in Supplementary file 6.

Gravid fmo-4p::mCherry transcriptional reporter adult animals were placed on NGM plates containing 1 mM β-D-isothiogalactopyranoside (IPTG), 25 μg/ml carbenicillin, and the corresponding RNAi clone from the Vidal or Ahringer RNAi library. After 3 hr, the adults were removed, and the eggs were allowed to develop at 20 °C until they reached late L4/young adult stage. Then 40–50 worms per plate were transferred to similar plates that contain FUdR for overnight. The following day, ~20 worms per condition were picked off these plates, added to a 2% agarose pad on a glass slide, anesthetized in 0.5 M sodium azide (Sigma), and imaged at 6.3 x magnification with the LASx software and Leica scope using the mCherry fluorescence channel (Miller et al., 2022). Three replicates were performed. Fluorescent intensity in the mCherry channel was measured in ImageJ. Brightness of images within a dataset was increased to the same level. Data were analyzed in GraphPad Prism using unpaired two-tailed t tests with Welch’s correction.

Gravid fmo-4p::mCherry transcriptional reporter adult animals were placed on NGM plates seeded with E. coli OP50. After 3 hr, the adults were removed, and the eggs were allowed to develop at 20 °C until they reached late L4/young adult stage. Then 30 worms were transferred to NGM plates seeded with E. coli OP50 also containing either 300 µM carbachol (Thermo Fisher, L06674.14) or water. An optimal concentration of carbachol was determined by first assessing survivability of worms exposed to a range of concentrations and then by assessing fluorescence of the fmo-4p::mCherry reporter worms exposed to a range of concentrations. The worms were incubated for 24 hr at 20 °C. After 24 hr, ~20 worms per condition were then picked off these plates and added to unseeded NGM plates, anesthetized in 0.5 M sodium azide (Sigma), and imaged at 6.3 x magnification with the LASx software and Leica scope using the mCherry fluorescence channel (Miller et al., 2022). Three replicates were performed. Each worm was measured for fluorescence in ImageJ. Data were analyzed in GraphPad Prism using t tests.

Gravid fmo-4p::mCherry transcriptional reporter adult animals were placed on NGM plates seeded with E. coli OP50. After 3 hr, the adults were removed, and the eggs were allowed to develop at 20 °C until they reached late L4/young adult stage. Then 30 worms were transferred to NGM plates seeded with E. coli OP50 topically spotted with 150 µL of 0.5 M ethylenediaminetetraacetic acid (EDTA) (ThermoFisher, AM9260G) or 150 µL of water. An optimal concentration of EDTA was determined by first assessing survivability of worms exposed to a range of concentrations and then by assessing the fluorescence of the fmo-4p::mCherry reporter worms exposed to a range of concentrations. The worms were incubated for 24 hr at 20 °C. After 24 hr, ~20 worms per condition were then picked off these plates and added to unseeded NGM plates, anesthetized in 0.5 M sodium azide (Sigma), and imaged at 6.3 x magnification with the LASx software and Leica scope using the mCherry fluorescence channel (Miller et al., 2022). Three replicates were performed. Each worm was recorded in ImageJ. Data were analyzed in GraphPad Prism using t tests.

Gravid GCaMP7f (SWF702; Dag et al., 2023) adult worms were placed on NGM plates seeded with E. coli OP50. After 3 hr, the adults were removed, and the eggs were allowed to develop at 20 °C until they reached late L4/young adult stage. Then 30 worms were transferred to NGM plates seeded with E. coli OP50 topically spotted with 150 µL of 0.5 M EDTA(Thermo Fisher, AM9260G), 300 µM carbachol (Thermo Fisher, L06674.14), or water. The worms were incubated for 24 hr at 20 °C. After 24 hr, ~20 worms per condition were then picked off these plates and added to unseeded NGM plates, anesthetized in 0.5 M sodium azide (Sigma), and imaged at 6.3 x magnification with the LASx software and Leice scope using the GFP fluorescence channel. Three replicates were performed. Each worm was recorded in ImageJ, with focus on the head region to assess neuronal expression. Data were analyzed in GraphPad Prism using t tests. For time course assays, gravid GCaMP7f adult worms were placed on unseeded NGM plates, anesthetized in 0.5 M sodium azide (Sigma), and imaged at 6.3 x magnification with the LASx software and Leica scope using the GFP fluorescence channel. Then 2 µL of either 300 µM carbachol or 10 mM EDTA were added to the worms and they were imaged using the GFP fluorescence channel at 0.5 min, 5 min, 10 min, 15 min, and 20 min. Three replicates were performed. Each worm was recorded in ImageJ, with focus on the head region to assess neuronal expression. Data were analyzed in GraphPad Prism using t tests.

Log-rank test was used to derive p-value for lifespan and stress resistance survival assays using p<0.05 cut-off threshold compared to EV or wild-type controls. Cox regression was also used to assess interactions between genotype and condition for lifespans and stress resistance survival assays using p<0.01 cut-off threshold compared to controls. Source data 3 provides the results of the Log-rank test and Cox regression analysis, which were run in RStudio.