A dicer-related helicase opposes the age-related pathology from SKN-1 activation in ASI neurons

Abstract

Coordination of cellular responses to stress is essential for health across the lifespan. The transcription factor SKN-1 is an essential homeostat that mediates survival in stress-inducing environments and cellular dysfunction, but constitutive activation of SKN-1 drives premature aging thus revealing the importance of turning off cytoprotective pathways. Here, we identify how SKN-1 activation in two ciliated ASI neurons in Caenorhabditis elegans results in an increase in organismal transcriptional capacity that drives pleiotropic outcomes in peripheral tissues. An increase in the expression of established SKN-1 stress response and lipid metabolism gene classes of RNA in the ASI neurons, in addition to the increased expression of several classes of noncoding RNA, define a molecular signature of animals with constitutive SKN-1 activation and diminished healthspan. We reveal neddylation as a unique regulator of the SKN-1 homeostat that mediates SKN-1 abundance within intestinal cells. Moreover, RNAi-independent activity of the dicer-related DExD/H-box helicase, drh-1, in the intestine, can oppose the effects of aberrant SKN-1 transcriptional activation and delays age-dependent decline in health. Taken together, our results uncover a cell nonautonomous circuit to maintain organism-level homeostasis in response to excessive SKN-1 transcriptional activity in the sensory nervous system.

Keywords: ASI neurons; C. elegans; aging; cell nonautonomous signaling; transcriptional capacity.

Fig. 1.

Cell autonomous activity of SKN-1gf in ASI neurons. (A) Differentially expressed genes between WT and skn-1gf (Blue) were overlapped with SKN-1gf occupied loci (Yellow). ChIP-seq was normalized to a no antibody control IP, while the DE analysis was performed by comparing skn-1gf to WT transcript levels. Genes with altered expression in skn-1gf mutants and identified with occupancy of SKN-1gf on the promoter region were analyzed in WormCat 2.0 (list of genetic loci can be found in Dataset S1) to reveal enriched classes (B). Expression of SKN-1wt-GFP (C) and SKN-1gf-GFP (D) are indistinguishable and restricted to ASI neurons; costained with DiI (red) that marks ciliated neurons; arrows designate GFP in ASI cell bodies (green). (E) WormCat 2.0 analysis of genes up-regulated in FACS-enriched ASI neuron populations from skn-1gf animals.

Fig. 2.

Neddylation regulates nuclear SKN-1 stabilization in the intestine. Unlike mock treatment (A and B), acute exposure to hydrogen peroxide (H₂O₂) drives nuclear accumulation outside of the ASI neurons (white arrows) for SKN-1wt-GFP (C) and SKN-1gf-GFP (D) within intestinal nuclei (yellow arrows). As compared to control RNAi treated animals (E and F), RNAi of uba-1 (G and H) and ned-8 (I and J) stabilizes SKN-1gf-GFP but not SKN-1wt-GFP within intestinal nuclei. (K) Cartoon of the role of ubiquitinylation and neddylation on nuclear SKN-1 stability. All RNAi experiments were conducted with n = 50 N = 3, representative worms are shown.

Fig. 3.

SKN-1 activity in ASI neurons mediates peripheral stress responses. (A) Cartoon representation of strains for tissue-specific regulation of skn-1gf expression. (B) Expression of skn-1gf from gpa-4p (ASI neurons), but not vha-6p (intestine), or rgef-1p (pan-neuronal) can establish resistance to acute exposure to H₂O₂ n = 150 N = 3 per condition analyzed by one-way ANOVA *(P < 0.05) ****(P < 0.0001). (C) Tissue-specific expression of skn-1gf results in the cell nonautonomous activation of the gst-4p::gfp reporter, Green (bright reporter activation), Yellow (dim reporter activation), Red (No detectable reporter activation). Tissue-specific expression of skn-1gf is not sufficient to drive Asdf (D). Pan-neuronal, intestinal, and ASI neuron-specific degradation of SKN-1gf can partially suppress somatic lipid depletion (Asdf); n = 300; N = 3 per condition analyzed by one-way ANOVA **(P < 0.01) ***(p < 0.001) ****(p < 0.0001) (E).

Fig. 4.

DRH-1 activation delays SKN-1gf-dependent healthspan decline. (A) Cartoon schematic of EMS genetic screen for suppressors of skn-1gf activation of gst-4p::gfp. (B) Genetic linkage maps the lax257 suppressor to LGIV. As compared to control RNAi (C) drh-1 RNAi abolishes the suppression of drh-1gf (D). Ectopic expression of drh-1gf (E) suppresses skn-1gf activation of gst-4p::gfp as compared to nontransgenic siblings (F). (G) Predicted structure and amino acid substitution (wt-cyan; gf-magenta) in DRH-1gf. drh-1gf suppresses the somatic lipid depletion phenotype of skn-1gf mutant at day 3 (H) but not day 5 (I) of adulthood; n = 300; N = 3 per condition comparisons were made by one-way ANOVA ****(P < 0.0001). The resistance to acute H₂O₂ by skn-1gf exposure at day 1 of adulthood (J) and the sensitivity at day 3 of adulthood (K) is reversed by drh-1gf. (L–N) The suppression of the impaired movement phenotype of skn-1gf by drh-1gf at the day 1 stage (L) is progressively abrogated at day 3 (M) and day 5 (N) of adulthood; n = 50; N = 3 per condition. Oxidative stress assay was analyzed by one-way ANOVA **(P < 0.001) ****(P < 0.0001), n = 100; N = 3 per condition.

Fig. 5.

Intestinal DRH-1 activation reduces transcriptional load of activated SKN-1. (A–G) drh-1gf (magenta) suppresses the activation of established SKN-1 targets in skn-1gf mutant animals (blue), Blue lines represent DE between WT and skn-1gf, and magenta lines represent DE between skn-1gf drh-1gf and skn-1gf. (H) drh-1gf mutations abolishes the increased sensitivity of skn-1gf mutant animals to RNAi inhibition of transcriptional regulators n = 100 N = 3. (I) WormCat 2.0 analysis of genes activated by skn-1gf and suppressed by drh-1gf. (J–Q) drh-1gf suppresses the increased expression of ncRNA in skn-1gf mutants. The activation of the gst-4p::gfp in skn-1gf animals is suppressed by intestinal expression of drh-1gf (R and S). (T–V) drh-1gf suppresses the increased expression of signaling molecules. (W) Cartoon schematic of cell nonautonomous signaling by skn-1gf in ASI neurons and drh-1gf in the intestine. A simple linear regression model was used to compare each of the lines with P < 0.05 considered significant.