Cellular stress induces a protective sleep-like state in C. elegans

Abstract

Sleep is recognized to be ancient in origin, with vertebrates and invertebrates experiencing behaviorally quiescent states that are regulated by conserved genetic mechanisms. Despite its conservation throughout phylogeny, the function of sleep remains debated. Hypotheses for the purpose of sleep include nervous-system-specific functions such as modulation of synaptic strength and clearance of metabolites from the brain, as well as more generalized cellular functions such as energy conservation and macromolecule biosynthesis. These models are supported by the identification of synaptic and metabolic processes that are perturbed during prolonged wakefulness. It remains to be seen whether perturbations of cellular homeostasis in turn drive sleep. Here we show that under conditions of cellular stress, including noxious heat, cold, hypertonicity, and tissue damage, the nematode Caenorhabditis elegans engages a behavioral quiescence program. The stress-induced quiescent state displays properties of sleep and is dependent on the ALA neuron, which mediates the conserved soporific effect of epidermal growth factor (EGF) ligand overexpression. We characterize heat-induced quiescence in detail and show that it is indeed dependent on components of EGF signaling, providing physiological relevance to the behavioral effects of EGF family ligands. We find that after noxious heat exposure, quiescence-defective animals show elevated expression of cellular stress reporter genes and are impaired for survival, demonstrating the benefit of stress-induced behavioral quiescence. These data provide evidence that cellular stress can induce a protective sleep-like state in C. elegans and suggest that a deeply conserved function of sleep is to mitigate disruptions of cellular homeostasis.

Figure 1

C. elegans experiences ALA-dependent recovery quiescence following heat stress. A, Time course of feeding quiescence during and after a 30 min 35°C heat shock in wild-type and ALA neuron-defective ceh-17(np1) animals compared to untreated control animals. B, Automated (multi-worm tracker) analysis of locomotor quiescence in wild-type and ceh-17(np1) animals during and after a 30 min 35°C heat shock compared to untreated control animals. Shaded regions represent time during heat shock. Mean and SEM are shown. *p < 0.001; Fisher’s exact test (A) and Student’s t test (B).

Figure 2

Recovery quiescence displays sleep-like properties and requires components of EGF signaling. A, Reversibility of locomotor quiescence after perturbation by harsh touch at 10, 20, and 30 minutes after a 30 min 35°C heat shock. B and C, Fraction of heat shocked wild-type young adults that do not show locomotor response within 5 seconds after a 3 second exposure to blue light (B) or during 5 second exposure to 1-octanol (C). D and E, Time course of feeding quiescence (D) and locomotor quiescence (E) following a 30 min 35°C heat shock in wild-type, lin-3(n1058) itr-1(sy290), let-23(sy10) unc-4(e120) and plc-3(tm1340) animals. let-23 and plc-3 mutants show impaired feeding quiescence at all time points, and lin-3 mutants at time points indicated by asterisks. The failure of lin-3 mutant animals to recover activity may reflect altered kinetics of ligand release caused by the n1058 intracellular domain mutation. Locomotion data was not collected for let-23 mutant animals due to a movement defect conferred by the unc-4(e120) mutation. Mean and SEM are shown. *p < 0.05, **p < 0.01, ***p < 0.001; Fisher’s exact test (A–D) and Student’s t test (E).

Figure 3

Cellular stressors induce ALA-dependent recovery quiescence. A and B, Time course of feeding quiescence (A) and locomotor quiescence (B) in wild-type and ALA-defective ceh-17(np1) animals after 15 min exposure to a 500 mM NaCl solution. Grey dotted line indicates behavior of untreated wild-type animals. C and D, Time course of feeding quiescence (C) and locomotor quiescence (D) in wild-type and ceh-17(np1) animals after a 30 min exposure to 5% ethanol. E and F, Time course of feeding quiescence (E) and locomotor quiescence (F) in wild-type and ceh-17(np1) animals after a 15 min exposure to −15 °C. G and H, Time course of feeding (G) and locomotor (H) quiescence in wild-type and ceh-17(np1) animals during (shaded region) and after a 15 min exposure to Cry5B toxin. Mean and SEM are shown. *p < 0.001; Fisher’s exact test. Note that each stressor also has ALA-independent effects of varying duration on feeding and/or locomotion, observable at the earliest time points.

Figure 4

Recovery quiescence is associated with restoration of proteostasis and survival following noxious heat exposure. A and B, Time course of feeding quiescence in wild-type and ceh-17(np1) animals following a 30 min 37°C heat shock (A) and a 20 min 40°C heat shock (B). *p < 0.001; Fisher’s exact test. C, Time course of feeding quiescence in wild-type, hsf-1(sy441), daf-16(mu68), and hsp-4(gk514) animals following a 30 min 37°C heat shock. Peak quiescence is reduced in hsf-1 and daf-16 mutant animals compared to wild type (asterisks not shown, p = 0.0019 in both cases; Fisher’s exact test), while the duration of quiescence is significantly increased (*p < 0.05; Fisher’s exact test.) in each of the mutant strains. For panels A–C, feeding quiescence is shown but animals were similarly quiescent for locomotion. D, Fold expression of hsp-16.2:GFP and hsp-4:GFP transcriptional reporters 24 hours after a 37°C 30 min heat shock compared to untreated controls. #p < 0.01, ##p < 0.001; Student’s t test. E and F, Survival (E) and locomotor quiescence (F) among wild-type, ceh-17(np1), egl-4(ad450gf), and ceh-17(np1);egl-4(ad450gf) animals following a 20 min 40°C heat shock. ceh-17 animals are significantly impaired for survival compared to wild type (p < 0.001; log rank test). In ceh-17;egl-4(gf) mutants, locomotor quiescence is partially restored (*p < 0.05 vs. ceh-17(np1); Fisher’s exact test) and the survival defect is partially rescued (p < 0.001 vs. ceh-17(np1); log rank test). egl-4(gf) animals do not show significantly greater survival than wild-type animals (p = 0.647; log rank test) but display increased quiescence at certain time points (p < 0.05 vs. wild type at 6, 8, 22, and 24 hrs after heat shock; Fisher’s exact test). Mean and SEM are shown for all panels except C and F, which show mean only. On the X axes, “B” indicates baseline (untreated).