Hormetic heat stress and HSF-1 induce autophagy to improve survival and proteostasis in C. elegans

Abstract

Stress-response pathways have evolved to maintain cellular homeostasis and to ensure the survival of organisms under changing environmental conditions. Whereas severe stress is detrimental, mild stress can be beneficial for health and survival, known as hormesis. Although the universally conserved heat-shock response regulated by transcription factor HSF-1 has been implicated as an effector mechanism, the role and possible interplay with other cellular processes, such as autophagy, remains poorly understood. Here we show that autophagy is induced in multiple tissues of Caenorhabditis elegans following hormetic heat stress or HSF-1 overexpression. Autophagy-related genes are required for the thermoresistance and longevity of animals exposed to hormetic heat shock or HSF-1 overexpression. Hormetic heat shock also reduces the progressive accumulation of PolyQ aggregates in an autophagy-dependent manner. These findings demonstrate that autophagy contributes to stress resistance and hormesis, and reveal a requirement for autophagy in HSF-1-regulated functions in the heat-shock response, proteostasis and ageing.

Figure 1. Heat shock induces autophagy.

(a–d) GFP::LGG-1/Atg8 punctae were counted in wild-type C. elegans maintained under control conditions (CTRL) or subjected to heat shock for 1 h at 36 °C (HS) followed by the indicated recovery period (Rec). Punctae were examined in (a) hypodermal seam cells (N=63-98 cells), (b) body wall muscle (N=10–12 animals), (c) nerve ring neurons (N=12 animals) and (d) proximal intestinal cells (N=14–16 animals). See also Supplementary Table 3 for a summary of repeat experiments. (e,f) Autophagy-flux measurements were performed on day 1 of adulthood in animals maintained at 15 °C (CTRL) or subjected to heat shock for 1 h at 36 °C (HS) followed by injection with vehicle (DMSO) or bafilomycin A (BafA) to block autophagy at the lysosomal acidification step. The number of GFP::LGG-1/Atg8 punctae was counted in (e) hypodermal seam cells (N=28-39, n=2) and (f) the proximal intestine (N=14-17, n=2). (g) Transcript levels of genes involved in various steps of the autophagy process in wild-type (WT) animals maintained under control conditions (CTRL) or subjected to heat shock for 1 h at 36 °C (HS). Data are the mean±s.e.m. of four biological replicates, each with three technical replicates, and are normalized to the mean expression levels of four housekeeping genes. All error bars are s.e.m. Scale bars, 20 μm. TB, terminal pharyngeal bulb. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 by Student's t-test (a–d), two-way ANOVA (e,f) and multiple t-tests (g).

Figure 2. Autophagy is induced in HSF-1-overexpressing animals.

(a–d) GFP::LGG-1/Atg8 punctae were counted in (a) hypodermal seam cells (N=131–163 cells), (b) body-wall muscle (N=10 animals), (c) nerve-ring neurons (N=11–12 animals) and (d) proximal intestinal cells (N=13 animals) of wild-type (WT) and HSF-1-overexpressing (HSF-1 OE) animals. See also Supplementary Table 7 for a summary of repeat experiments. (e,f) Autophagy-flux measurements were performed on day 1 of adulthood in animals maintained at 20 °C. WT and HSF-1 OE animals were injected with vehicle (DMSO) or bafilomycin A (BafA) to block autophagy at the lysosomal acidification step. The number of GFP::LGG-1/Atg8 punctae was counted in (e) hypodermal seam cells (N=129–162, n=3) and (f) the proximal intestine (N=21–26, n=3). (g) Transcript levels of genes involved in various steps of the autophagy process in WT and HSF-1 OE animals. Data are the mean±s.e.m. of four biological replicates, each with three technical replicates, and are normalized to the mean expression levels of four housekeeping genes. All error bars are s.e.m. Scale bars, 20 μm. TB, terminal pharyngeal bulb. ns: P>0.05, *P<0.05, **P<0.01, ***P<0.001 and ****P<0.0001 by Student's t-test (a–d), two-way ANOVA (e,f) and multiple t-tests (g).

Figure 3. Autophagy genes are required for heat shock- and HSF-1-mediated survival.

(a) Survival of wild-type (WT) animals subjected to hormetic heat shock on day 1 of adulthood and then incubated for 8 h at 36 °C on day 4 of adulthood. Animals were fed from day 1 of adulthood with control bacteria (empty vector, CTRL) or bacteria expressing dsRNA targeting the autophagy genes unc-51/ATG1, bec-1/ATG6 and lgg-1/ATG8 (N=65-90 animals, n=4 plates). (b–d) Lifespan analysis of animals subjected to hormetic heat shock with RNAi-mediated autophagy gene reduction from day 1 of adulthood. WT-CTRL animals (19.2 days, N=104) compared with WT-HS animals (23.7 days, N=94): P<0.0001, (b) unc-51/ATG1 RNAi-CTRL (18.5 days, N=110) compared with unc-51/ATG1 RNAi-HS (17.5 days, N=107): P=0.04, (c) bec-1/ATG6 RNAi-CTRL (19.2 days, N=116) compared with bec-1/ATG6 RNAi-HS (18.4 days, N=112): P=0.3, (d) lgg-1/ATG8 RNAi-CTRL (18.1 days, N=108) compared with lgg-1/ATG8 RNAi-HS (17.9 days, N=79): P=0.7. (e) Survival of WT or HSF-1-overexpressing (HSF-1 OE) animals incubated for 8 h at 36 °C on day 3 of adulthood. Animals were fed from day 1 of adulthood with control bacteria (empty vector, CTRL) or bacteria expressing dsRNA targeting the indicated autophagy genes (N=113–220 animals, n=4 plates). Error bars indicate s.e.m. ns: P>0.05, *P<0.05 and ***P<0.001 by one-way ANOVA. (f–h) Lifespan analysis of WT and HSF-1 OE animals subjected to RNAi-mediated autophagy gene reduction from day 1 of adulthood. WT animals (18.1 days, N=113) compared with HSF-1 OE animals (23.0 days, N=121): P<0.0001, (f) WT: CTRL compared with unc-51/ATG1 RNAi (18.3 days, N=128): P=0.9, HSF-1 OE: CTRL compared with unc-51/ATG1 RNAi (15.4 days, N=133): P<0.0001, (g) WT: CTRL compared with bec-1/ATG6 RNAi (16.7 days, N=123): P=0.02, HSF-1 OE: CTRL compared with bec-1/ATG6 RNAi (16.3 days, N=140): P<0.0001, (h) WT: CTRL compared with lgg-1/ATG8 RNAi (16.7 days, N=109): P=0.02, HSF-1 OE: CTRL compared with lgg-1/ATG8 RNAi (14.9 days, N=147): P<0.0001, by log-rank test. See Supplementary Tables 1, 2, 11 and 12 for details on thermorecovery and lifespan analyses and replicate experiments.

Figure 4. Hormetic heat shock reduces PolyQ protein aggregation.

(a) Intestinal PolyQ aggregates detected on day 5 of adulthood in wild-type (WT) animals maintained under control conditions (WT-CTRL) or subjected to hormetic heat shock (1 h at 36 °C) on day 1 of adulthood (WT-HS) and in HSF-1-overexpressing animals maintained under control conditions (HSF-1 OE-CTRL). Animals expressing PolyQ44::YFP under the control of the intestine-specific promoter vha-6 were fed from hatching with control bacteria (empty vector; upper row) or bacteria expressing dsRNA targeting bec-1/ATG6 RNAi (lower row). Arrowheads indicate prominent aggregates. Scale bar, 500 μm. (b–d) Quantification of PolyQ aggregates on day 5 of adulthood in animals fed from hatching (whole life, WL) with control bacteria (empty vector, CTRL) or bacteria expressing dsRNA targeting unc-51/ATG1, bec-1/ATG6 or hsf-1 in (b) WT animals, (c) WT animals subjected to hormetic heat shock on day 1 of adulthood and (d) HSF-1 OE animals (N=14–23). Dotted line represents number of aggregates of WT-CTRL on empty vector control and grey asterisk represents P-value compared to WT-CTRL on empty vector. The experiments were repeated at least three times with similar results. (e) Lifespan analysis of animals expressing intestinal PolyQ44::YFP and subjected to hormetic heat shock on day 1 of adulthood. Intestinal Q44-CTRL animals (15.2 days) compared with Intestinal Q44-HS animals (19.3 days): P<0.0001, see Supplementary Table 13 for details on lifespan analyses and replicate experiments. Error bars indicate s.e.m. ns: P>0.05, *P<0.05, **P<0.01 and ***P<0.001 by one-way ANOVA (b–d) and log-rank test (e).