Neuronal activity protects from chronic oxidative stress

(A) Top, experimental timeline indicating timing and duration of histamine and PQ treatment for survival assays. Left: Bar graph comparing IP50 measurements for control (His-) and His-treated (His+) kyEx4571 (tag-168p::HisCl1::SL2::GFP) transgenic animals animals in the presence of PQ. The day at which His-treated animals reach half of their initial population is significantly advanced compared to control, indicating reduced survival. Each point represents one independent trial. Bars indicate mean ±SEM. *p=0.0135, unpaired two-tailed t-test with Welch’s correction. Number of independent trials: n=6. Right: Kaplan-Meier survival curves comparing His+ and His-groups. Pan-neuronal silencing (24 hr) during oxidative stress reduces survival in the continued presence of PQ. Kaplan-Meier curves are cumulative of 6 trials. ****p= 3.4E-08, Fisher’s exact test at 50% of survival. Total number of animals: n= 59 (His-), 64 (His+).

(B) Top, experimental timeline indicating timing and duration of histamine and PQ treatment for survival assays. Note 8 hr recovery period between the end of histamine treatment and the onset of PQ exposure. Left: Bar graph comparing (IP50 measurements for control (His-) and His-treated (His+) kyEx4571 (tag-168p::HisCl1::SL2::GFP) transgenic animals animals in the presence of PQ. The day at which His-treated animals reach half of their initial population is significantly advanced compared to control, indicating reduced survival. Each point represents one independent trial. Bars indicate mean ± SEM. *p=0.0156, unpaired two-tailed t-test with Welch’s correction. Number of independent trials: n=6. Right: Kaplan-Meier survival curves comparing His+ and His-groups. Pan-neuronal silencing (24 hr) prior to the onset of oxidative stress reduces organismal survival in the continued presence of PQ. Kaplan-Meier curves are cumulative of 6 trials. ****p= 1.20E-07, Fisher’s exact test at 50% of survival. Total number of animals, n= 82 (His-), 89 (His+).

Disruption of either glutamate or acetylcholine neurotransmission increases vulnerability to PQ

(A) Left: Bar graph comparing IP50 measurements for the indicated genotypes. The day at which eat-4 (glutamate), unc-17 (acetylcholine) or unc-25 (GABA) mutants reach half of their initial population is significantly advanced compared to control, indicating reduced survival. The reduction in IP50 is greatest for eat-4 and unc-17 mutants. Each point represents one independent trial. Bars indicate mean ± SEM. ****p <0.0001, eat-4(ky5); ****p<0.0001, unc-17(e245); ***p=0.0006, unc-25(e156); p=0.2718 (ns), tdc-1(n3420); p=0.9128 (ns), tph-1(mg280); p=0.2104 (ns), cat-2(n4547). One-way ANOVA with Dunnett’s multiple comparisons test. Number of trials: n≥5. Right: Upper, table shows the six conserved neurotransmitters and corresponding mutant genes involved with either vesicular loading or neurotransmitter biosynthesis that were surveyed in the survival assays. Kaplan -Meier survival curves for wild type and indicated neurotransmitter-deficient mutants with chronic PQ exposure. Kaplan-Meier curves are cumulative of at least 5 trials. ****p=5.00E-06, eat-4(ky5); ****p=1.50E-12, unc-17(e245); ****p=2.40E-12, unc-25(e156); ****p=1.10E-08, tdc-1(n3420); p=0.2492 (ns), tph-1(mg280); ****p=4.30E-12, cat-2(n4547). Fisher’s exact test at 75% of survival. Total number of animals: n= 527(N2), 251 (eat-4), 155 (unc-17), 148 (unc-25), 155 (tdc-1), 205 (tph-1), 145 (cat-2).

(B) Left: IP50 comparisons showing glutamate and acetylcholinedeficient mutants have no significant difference in their normal longevity. Each data point is one independent trial. Bars indicate mean ±SEM. p =0.9712(ns), eat-4(ky5); p=0.6226 (ns), unc-17(e245). One-way ANOVA with Dunnett’s multiple comparisons test. Number of trial ≥3. Right: Kaplan-Meier curve shows there is a modest reduction of longevity in unc-17(e245) and no difference in eat-4(ky5) compared to WT(N2). The difference in longevity for unc-17(e245) compared to WT(N2) was observed after 9 days. Kaplan-Meier curves are cumulative of at least 3 trials. p =0.2425(ns), eat-4(ky5), **p=0.0012, (unc-17(e245). Fisher’s exact test at 50% of survival. Total number of animals: n=65, N2; 67, eat-4(ky5) and 80, unc-17(e245).

Disruption of cholinergic transmission from motor neurons reduces survival during chronic oxidative stress

(A) Bar graph comparing IP50 measurements on PQ for wild type, unc-17(e245), unc-17(e113), or motor neuron-specific knockdown of unc-17 by RNAi. unc-17(e113) mutants with reduced unc-17 expression in motor neurons or RNAi downregulation of unc-17 in motor neurons each show significant reductions in IP50, comparable to that of severely hypomorphic unc-17(e245) mutants. Each data point is one independent trial. Bars indicate mean ±SEM. ****p <0.0001, unc-17(e245); ****p<0.0001, unc-17(e113); *p =0.0224, unc-17β RNAi. One-way ANOVA with Dunnett’s multiple comparisons test. Number of trials, n≥4. Total number of animals: n= 303, N2; 200, unc-17(e245); 200, unc-17(e113); and 91, unc-17β RNAi.

(B) IP50 comparison of of wild type (WT), unc-17(e245), and unc-17(e113) show there is no significant reduction in longevity under control (PQ-) conditions though unc-17(e113) is more variably affected. Each data point is one independent trial. Bars indicate mean ±SEM. p= 0.7316 (ns), unc-17(e245); p=0.1826 (ns), unc-17(e113). One-way ANOVA with Dunnett’s multiple comparisons test. Number of trials: n≥3. Total number of animals: n= 65 WT(N2); 80, unc-17(e245); 63, unc-17(e113).

Extended oxidative stress mobilizes a broad transcriptional response

(A) Heatmap shows clustering of treatment (PQ+) and control (PQ-) replicates for WT(N2). Number of replicates: n= 4 for both conditions.

(B) Volcano plot (log2FoldChange, -log10padj) of differentially expressed genes following 48 hrs of PQ (4 mM) treatment compared to age-matched controls. Red: upregulated, Blue: Downregulated, Grey: not significantly different. Statistical cutoff for differential expression: FoldChange>2, Padj<0.01 and False Discovery Rate (FDR) <0.01.

(C) Functional categorization of upregulated and downregulated genes that showed differential expression following 48 hr PQ exposure in wild type. The transcription factor, stress response and proteolysis proteasome categories were most strongly enriched amongst the differentially expressed genes. Grey squares indicate no significant enrichment in that category. Gene-count and P-value scales as indicated. A larger radius indicates a higher number of genes in that category. A darker color indicates a more significant P value.

Comparison of transcriptomic responses between acute and extended oxidative stresses

(A) Venn diagram shows the extent of overlap between genes upregulated in the presence of arsenite (Oliveira et al., 2009) compared with genes we identified as upregulated in response to extended PQ exposure. Approximately 36% (43 out of 118) of genes upregulated by arsenite treatment were also upregulated by PQ. Representative factor 3.5, p < 2.816e-14.

(B) Venn diagram shows the extent of overlap between genes upregulated in the presence of tBOOH (Oliveira et al., 2009) compared with genes we identified as upregulated in response to extended PQ exposure. Approximately 29% (81 out of 281) of genes upregulated by tBOOH treatment are also upregulated by PQ. Representative factor 2.8, p < 2.205e-18.

(C) Venn diagram shows the extent of overlap between genes upregulated in the presence of juglone (Wu et al., 2016) compared with genes we identified as upregulated in response to extended PQ exposure. Approximately 13% (103 out of 811) of genes upregulated by juglone treatment are also upregulated by PQ. Representative factor 1.2, p < 0.014.

Acetylcholine deficiency blunts transcriptomic responses to oxidative stress

Whole worm RNA sequencing data shows 48h continuous exposure to PQ (4 mM) does not change the transcriptomic landscape in unc-17(e113) mutants as severely as in wild type.

(A) Heatmap shows clustering of treatment (PQ+) and control (PQ-) replicates for unc-17(e113). Number of replicates: n=3 for both conditions.

(B) Volcano plot (log2FoldChange, -log10padj) of genes differentially expressed in unc-17(e113) mutants in response to 48 hrs of PQ (4 mM) treatment compared to age-matched controls. Red: upregulated, Blue: Downregulated, Grey: not significantly different. Statistical cutoff for differential expression FoldChange>2, padj<0.01, and False Discovery Rate (FDR) <0.01.

(C) Functional categorization of differentially expressed genes (upregulated and downregulated) in unc-17(e113) mutants following 48 hr of PQ exposure. Grey squares indicate no significant enrichment in that category. Gene count and P-value scale as indicated. A larger radius corresponds to a higher number of genes in that category. A darker color corresponds to a more significant P value.

(D) Venn diagram shows the intersection of upregulated genes in response to PQ for wild type (N2) (total 1811) and unc-17(e113) (total 788). 1,313 genes that are a part of the wild type transcriptional response to PQ fail to be upregulated in ACh-deficient e113 mutants.

(E) Venn diagram shows the intersection of PQ-responsive and ACh-dependent wild type genes compared with genes differentially expressed in unc-17(e113) versus wild type under control conditions without PQ treatment. 104 genes (14%) are shared between the two groups. Wild-type PQ-responsive genes that are no longer PQ-responsive in unc-17 mutants are largely distinct from those differentially expressed between unc-17 and wild type in the absence of PQ.

Cholinergic activation of GPCR signaling via mAChR GAR-3 promotes survival in the presence of oxidative stress

(A) Bar graph showing ACh signaling genes upregulated in wild type in response to PQ. acr-12 and lev-8 iAChR subunit, chaperone ric-3 and mAChR gar-3 transcripts are significantly upregulated. The bar graph shows log2foldchange value for all detected ACh signaling gene transcripts (y axis).

(B) Bar graph comparing IP50 measurements for wild type, gar-3(gk305) mAChR mutants and ric-3(hm9) chaperone mutants in the presence of PQ. Mutation of gar-3 significantly reduces IP50 values compared to wild type, whereas the effect of ric-3(hm9) is variable. Each data point represents one independent trial. Bars indicate mean ±SEM. ****p<0.0001, gar-3(gk305); p= 0.0531(ns), ric-3(hm9). One-way ANOVA with Dunnett’s multiple comparisons test.

(C) Bar graph comparing IP50 measurements for wild type, gar-3(gk305) mutants and gar-3;ric-3 double mutants in the presence of PQ. The IP50 of gar-3;ric-3 double mutants is similar to that of gar-3(gk305) single mutants. Each data point represents one independent trial. Bars indicate mean ± SEM. **p= 0.0044, WT (N2) vs.gar-3(gk305);ric-3(hm9); p=0.0961(ns) for gar-3(gk305) vs. gar-3(gk305);ric-3(hm9). One-way ANOVA with Tukey’s multiple comparisons test. Number of trials: n=4, N2; n=3, gar-3(gk305); n=4, gar-3(gk305);ric-3(hm9).

(D) Bar graph comparing IP50 measurements for wild type and gar-3(gk305) mutants under control condition in the absence of PQ. gar-3 mutants do not show an appreciable reduction in longevity compared to wild type under control (PQ-) conditions. Each data point represents one independent trial. Bars indicate mean ± SEM. p= 0.7377 (ns), unpaired two-tailed t-test with Welch’s correction. Number of trials: n= 3. Total number of animals: n= 65 WT(N2); 85, gar-3(gk305).

(E) Bar graph comparing IP50 measurements for wild type, unc-17(e245) and gar-3(gk305) mutants in the presence of PQ. Mutation of unc-17 or gar-3 reduces survival to a similar extent in the presence of PQ. Each data point represents one independent trial. Bars indicate mean ± SEM. ****p<0.0001, one-way ANOVA with Dunnett’s multiple comparisons test. Number of trials: n= 9, N2; 8, unc-17(e245); 11, gar-3(gk305).

(F) Bar graph comparing IP50 measurements for wild type, unc-17(e245), unc-17(e113), gar-3(gk305) and each of the double mutant combinations in the presence of PQ. Combined mutation of gar-3 and unc-17 elicits a similar reduction in IP50 to that of either single mutant, providing evidence that unc-17 and gar-3 may act in the same pathway. Each data point represents one independent trial. Bars indicate mean ± SEM. ***p=0.0006, WT vs. unc-17(e245); **p=0.0011, WT vs. unc-17(e113); ***p=0.0005, WT vs. gar-3(gk305); p= 0.8811(ns) gar-3(gk305) vs. gar-3(gk305);unc-17(e245); p= 0.9983 (ns) gar-3(gk305) vs. gar-3(gk305);unc-17(e113); p=0.9312 (ns), unc-17(e245) vs. gar-3(gk305); p =0.9827(ns), unc-17(e113) vs. gar-3(gk305); p= 0.9997(ns), unc-17(e113) vs. gar-3(gk305);unc-17(e113) and p>0.9999(ns), unc-17(e245) vs. gar-3(gk305);unc-17(e245). One-way ANOVA with Tukey’s multiple comparisons test. Number of trials: n≥3.

(G) Bar graph comparing IP50 measurements for wild type, gar-3(gk305) mutants and transgenic gar-3 mutants expressing gar-3(+). Expression of wild type gar-3 using the native gar-3 promoter region normalizes the survival of gar-3 mutants in the presence of PQ. Each data point is one independent trial. Bars indicate mean ± SEM. p =0.9585(ns), WT vs. gar-3p::gar-3(+);gk305 and ***p =0.0002, gar-3(gk305) vs. gar-3p::gar-3(+);gk305. One-way ANOVA with Tukey’s multiple comparisons test. Number of trials: n≥4.

Deficiency of cholinergic signaling via GAR-3 blunts transcriptomic responses to oxidative stress

(A) Heatmap shows clustering of treatment (PQ+) and control (PQ-) replicates for gar-3(gk305). Number of replicates n= 2 for both conditions.

(B) Volcano plot (log2FoldChange, -log10padj) of genes differentially expressed in gar-3(gk305) mutants in response to 48 hr of PQ (4 mM) treatment compared to age-matched controls. Red: upregulated, Blue: Downregulated, Grey: not significantly different. Statistical cutoff for differential expression FoldChange>2, Padj<0.01 and False Discovery Rate (FDR) <0.01.

(C) Functional categorization of differentially expressed genes (upregulated and downregulated) in gar-3(gk305) mutants in the presence of extended PQ exposure. Grey squares indicate no significant enrichment in that category. Gene count and P-value scales as indicated. A larger radius corresponds to a higher number of genes in that category. A darker color corresponds to a more significant P value.

(D) Venn diagram showing the intersection of upregulated genes in wild type (total 1811), unc-17(e113) (total 788), and gar-3(gk305) (total 868) in response to PQ. 1055 genes that are upregulated in the wild transcriptional respoinse to PQ fail to upregulate in both ACh-deficient unc-17 and gar-3 mAChR mutants.

Cholinergic motor neuron-specific overexpression of gar-3 is sufficient to protect against oxidative damage

(A) Confocal maximum intensity projection of the head region of an adult animal expressing genome-inserted gar-3::SL2::GFP. Note expression in nerve ring processes (white box) and in the pharynx. Anterior is to the left in all panels. Scale, 10 µm.

(B) Confocal maximum intensity projection showing zoomed-in view of gar-3 expression in nerve ring neuronal cell bodies. Scale, 10 µm.

(C) Confocal maximum intensity projection of the midbody region showing expression in ventral nerve cord motor neurons and body wall muscles. Scale, 10 µm.

(D) Confocal maximum intensity projections of a segment of the ventral nerve cord showing expression of gar-3::SL2::GFP in cholinergic motor neurons labeled by acr-2p::mCherry. White arrows indicate gar-3 expressing cholinergic neurons. Scale bar, 10 μm.

(E) Confocal maximum intensity projections of a segment of the ventral nerve cord showing expression of gar-3::SL2::GFP in GABAergic motor neurons labeled by unc-47p::mCherry. White arrows indicate gar-3 expressing GABAergic neurons. Scale bar, 10 μm.

(F) Bar graph comparing IP50 measurements for wild type wild type, unc-17(e113), transgenic wild type animals overexpressing gar-3 in cholinergic motor neurons (unc-17βp::gar-3 OE) and transgenic unc-17 mutants overexpressing gar-3 in cholinergic motor neurons (unc-17βp::gar-3(+);unc-17) in the presence of PQ. Cholinergic overexpression of gar-3 significantly increases survival compared to wild type. Mutation of unc-17 reverts survival to the level of unc-17(e113) mutants. Each data point is one independent trial. Bars indicate mean ±SEM. ****p<0.0001, WT(N2) vs. unc-17(e113), and WT(N2) vs. unc-17βp::gar-3OE, and unc-17(e113) vs. unc-17βp::gar-3OE, and unc-17βp::gar-3OE vs. unc-17βp::gar-3OE;unc-17. ***p=0.0005, WT(N2) vs. unc-17βp::gar-3(+);unc-17. p=0.8673 (ns), unc-17(e113) vs. unc-17βp::gar-3OE;unc-17. One-way ANOVA with Tukey’s multiple comparisons test. Number of trials: n=19, WT(N2); 7, unc-17(e113); 10, unc-17βp::gar-3OE; 3, unc-17βp::gar-3(+);unc-17.

Chronic exposure to 4mM paraquat (PQ) results in oxidative stress

(A) Experimental timeline for PQ chronic exposure assay. Animals are continuously exposed to 4 mM paraquat (PQ). At day 0 of the experiment, animals are day 1 (D1) adults.

(B) Chronic PQ exposure reduces C. elegans longevity. Kaplan-Meier curves comparing the longevity of wild type (WT) N2 animals under control conditions (PQ–) to survival on 4 mM PQ. Percentage survival (y-axis) is plotted as a function of time (x-axis). ****p= 1.4E-12, Fisher’s exact test at 50% of survival. Total number of independent trials: n=6. The Kaplan-Meier curve is cumulative of all trials. Total number of animals over all trials: n= 112 (PQ-) and 136 (PQ+).

(C) Left: IP50comparison between wild type animals exposed to either PQ (PQ+) or PQ+NAC. For all, IP50 indicates the time to reach half of the original population. Each data point is one independent trial. Bars indicate mean ± SEM. ****p=0.0007, unpaired two-tailed t-test with Welch’s correction. Total number of independent trials n=7. Right: Kaplan-Meier survival curves comparing the profiles of animals exposed to either PQ (PQ+) or PQ+NAC. The antioxidant N-acetyl cysteine (NAC) enhances survival during chronic PQ treatment. Kaplan-Meier curve is cumulative of 7 trials. ****p= 3.60E-12, Fisher’s exact test at 50% of survival. Total animals: n= 172 (PQ+), n=175 (PQ+NAC).

(D) Left: IP50 comparison of wild type, sod-2(ok1030) and sod-2(gk257) treated with PQ. sod-2 mutants mutants are hypersensitive to PQ. Each data point is one independent trial. Bars indicate mean ± SEM. ***p= 0.0004 sod-2(ok1030), ****p<0.0001 sod-2(gk257), one-way ANOVA with Dunnett’s multiple comparisons test. Total number of trials: n=5 (WT), n=4 for sod-2(ok1030) and sod-2(gk257). Right: Kaplan-Meier survival curves comparing wild type, sod-2(ok1030) and sod-2(gk257). SOD-2 deficiency reduces survival during chronic PQ exposure. Kaplan-Meier curve is cumulative of at least 4 trials. ****p= 1.1E-12, sod-2(ok1030), ****p=4.5E-13, sod-2(gk257). Fisher’s exact test at 50% of survival. Total number of animals: n= 123, WT; 91, sod-2(ok1030); 90, sod-2(gk257).

(E) Immunoblotting data of protein carbonylation in wild type animals after 48 hr of PQ exposure (starting from D1 adult) compared to untreated controls. Left, Chemiluminescence image of the blot with primary antibody against dinitrophenylhydrazone (DNP) (top) and α-tubulin control (bottom). Right, The ratio of the area under the curve for DNP chemiluminescence in a particular lane relative to α-tubulin. Each data point indicates one trial. Bars indicate mean ± SEM.

*p=0.0454, unpaired two-tailed t-test with Welch’s correction.

Pharmacological silencing of pan-neuronal activity without exposure to oxidative stressor PQ has a nominal effect in the reduction of organismal longevity

(A) Top, experimental timeline indicating the period of histamine-mediated neuronal silencing (24 hrs) at day 2 of the longevity assay. Left: Bar graph comparing IP50 measurements for control (His-) and His-treated (His+) kyEx4571 (tag-168p::HisCl1::SL2::GFP) transgenic animals animals under control conditions (PQ-). The mean IP50 value is not significantly altered by histamine exposure, though variability increases in the His-treated group. Each data point represents one independent trial. Bars indicate mean ± SEM. p=0.1843 (ns), unpaired two-tailed t-test with Welch’s correction. Number of trials: n=4. Right: Kaplan-Meier survival curves for control (His-) and His-treated (His+) groups. Modest differences in longevity are apparent after day 9 of the lifespan assay. Kaplan-Meier curves are cumulative of 4 trials. *p = 0.0241, Fisher’s exact test at 50% of survival. Total number of animals: n= 46 (His-), 49 (His+).

(B) Top, experimental timeline indicating the periods of histamine-mediated neuronal silencing (24 hr) and recovery (8 hr). Left: Bar graph comparing IP50 measurements for control (His-) and His-treated (His+) kyEx4571 (tag-168p::HisCl1::SL2::GFP) transgenic animals animals under control conditions (PQ-). Each data point represents one independent trial. Bars indicate mean ± SEM. p=0.3629 (ns), unpaired two-tailed t-test with Welch’s correction. Number of trials, n=6. Right: Kaplan-Meier survival curves comparing profiles of control (His-) and His-treated (His+). Kaplan-Meier curves are cumulative of 6 trials. *p = 0.0390, Fisher’s exact test at 50% of survival. Total number of animals: n=68 (His-),67 (His+).