HSF-1 regulators DDL-1/2 link insulin-like signaling to heat-shock responses and modulation of longevity

Abstract

Extended longevity is often correlated with increased resistance against various stressors. Insulin/IGF-1-like signaling (IIS) is known to have a conserved role in aging and cellular mechanisms against stress. In C. elegans, genetic studies suggest that heat-shock transcription factor HSF-1 is required for IIS to modulate longevity. Here, we report that the activity of HSF-1 is regulated by IIS. This regulation occurs at an early step of HSF-1 activation via two HSF-1 regulators, DDL-1 and DDL-2. Inhibition of DDL-1/2 increases longevity and thermotolerance in an hsf-1-dependent manner. Furthermore, biochemical analyses suggest that DDL-1/2 negatively regulate HSF-1 activity by forming a protein complex with HSF-1. The formation of this complex (DHIC) is affected by the phosphorylation status of DDL-1. Both the formation of DHIC and the phosphorylation of DDL-1 are controlled by IIS. Our findings point to DDL-1/2 as a link between IIS and the HSF-1 pathway.

Figure 1. Inactivation of daf-2 positively regulates HSF-1 activity and heat-shock response

(A) DNA-binding activity of C. elegans HSF-1 in response to 90 min of heat-shock at 37°C (HS), measured by electrophoretic mobility shift assay. Nuclear extracts were incubated with biotin-labeled HSE probes (Biotin-HSE). The specificity of the DNA binding was determined by competition with 200x unlabeled HSE probes (HSE), randomly synthesized DNA (rDNA), anti-HSF-1 antibodies (Ab1), or anti-GFP antibodies (Ab2) before being applied to gel eletrophoresis. (B) Nuclear accumulation of HSF-1 in response to IIS inactivation. EQ73 animals (hsf-1::gfp) grown on vector control (VC) or different RNAi bacteria were unstressed or heat-shocked for 30 min (HS) before being classified into three groups according to the nuclear/cytosolic (n/c) ratio of GFP intensity in the intestinal cells (right panels). “c”, “wn”, and “sn” are animals with n/c ratio < 1.2, 1.2~2.0, and >2.0, respectively. The mean of three independent experiments were pooled and shown (left panel). *, p < 0.0001 vs VC under same conditions (chi²-test). n ≥ 300. (C–D) The DNA-binding activity of HSF-1 in daf-2(e1370) mutants in response to 90 min of heat-shock (HS). The result of a representative experiment is shown in (C). The mean ± SD of three independent experiments (mean ± SD), normalized to the control (N2 with unlabeled HSE), is presented in (D). (E) N2 or daf-2(RNAi) animals were unstressed or heat-shocked for 90 min (HS). Worm whole cell extracts (WCE) of these animals were subjected to immunoblotting analysis using anti-HSF-1 (top) or anti-β-actin (bottom) antibodies. Detail quantification in Fig. S1D. (F–G) Relative abundance of (F) hsp-16.2 and (G) hsp-70 (F44E5.5) mRNA in wild types (N2) or daf-2(e1370) mutants fed with control or hsf-1 RNAi bacteria. The inset shows the mRNA level under unstressed conditions (without 90 min HS). Data were combined from at least three experiments, and the mean ± SD of each treatment are shown.

Figure 2. A common hsf-1-dependent mechanism mediates the longevity effects of ddl-1, ddl-2 and hsb-1

(A) Lifespan analysis of wild-type (N2) animals or hsf-1(sy441) mutants grown on empty vector control or ddl-1 RNAi bacteria at 20°C. (B) Lifespan analysis of N2 animals or hsf-1(sy441) mutants grown on control or ddl-2 RNAi bacteria. (C) Lifespan analysis of N2 animals or hsb-1(cg116) mutants grown on control or hsf-1 RNAi bacteria. (D) Lifespan analysis of N2 animals grown on control, ddl-1 RNAi, ddl-2 RNAi, or 1:1 mixture of ddl-1 and ddl-2 RNAi bacteria at 20°C. (E) Lifespan analysis of N2, ddl-1(ok2916), hsb-1(cg116), or ddl-1(ok2916);hsb-1(cg116) mutants. (F) Lifespan analysis of N2, ddl-2(ok3235), hsb-1(cg116), or ddl-2(ok3235);hsb-1(cg116) mutants. Statistical details are summarized in Table 1 and Table S1.

Figure 3. DDL-1 and DDL-2 negatively regulate HSF-1 activity

(A-B) Lowering ddl-1 or ddl-2 expression increases HSF-1 DNA-binding activity. The DNA-binding activity of HSF-1 in N2, ddl-1(RNAi), or ddl-2(RNAi) animals with or without 90 min of heat-shock (HS). A representative experiment is shown in (A). Quantification of three independent experiments (mean ± SD) is presented in (B). (C–D) Relative abundance of (C) hsp-16.2, and (D) hsp-70 (F44E5.5) mRNA in N2, hsf-1(RNAi), ddl-1(RNAi), or ddl-2(RNAI) animals with or without heat-shock (90 min). The inset shows the mRNA level under unstressed conditions. The mean ± SD of three independent experiments were pooled and shown here. (E) Nuclear accumulation of HSF-1 in response to DDL-1/2 inactivation. EQ73 animals (hsf-1::gfp) grown on control, ddl-1, or ddl-2 RNAi bacteria were unstressed or heat-shocked for 30 min (HS). The result of three experiments were pooled and shown here. Data are mean, n ≥ 300 per RNAi treatment. *, p < 0.0001; ^#, p = 0.097 (chi²-test). (F) Worm extracts (WCE) prepared from N2, ddl-1(RNAi), or ddl-2(RNAi) animals with or without 90 min of HS were subjected to immunoblotting analysis using anti-HSF-1 (top) or anti-β-actin (bottom) antibodies. Detail quantification in Fig. S1D.

Figure 4. DDL-1 forms a protein complex with HSF-1, HSB-1, and DDL-2 in a cell culture model and in C. elegans

(A) DDL-1 interacts with DDL-2 in 293T cells. Here and in (B-C), 293T cells were transfected with indicated combinations of pCMV-HA-DDL-1, pFLAG-CMV2-DDL-2, pFLAG-CMV2-DDL-1, pCMV-Myc-HSB-1, or pCMV-Myc-HSF-1 plasmids. Whole cell extracts were then immunoprecipitated (IP) and subsequently western-blotted (WB) using indicated antibodies. (B) DDL-1 interacts with HSB-1 in 293T cells. (C) DDL-1 may form a protein complex with HSF-1 in 293T cells. (D) DDL-1 may form a protein complex with HSF-1 in C. elegans. Worm whole cell extracts (WCE) were prepared from N2, EQ136 [HA-ddl-1 oe], or EQ193 [HA-ddl-1 oe; hsb-1(cg116)] animals. Samples were immunoprecipitated and subsequently western-blotted using indicated antibodies. (E) DDL-1 interacts with DDL-2 in C. elegans. WCE were prepared from N2 or EQ155 animals expressing both HA-DDL-1 and FLAG-DDL-2 proteins. Samples were immunoprecipitated and western-blotted using indicated antibodies.

Figure 5. Both the formation of DHIC and the threonine phosphorylation of DDL-1 are regulated by IIS

(A) The formation of DHIC is disrupted by IIS inactivation. Worm whole cell extracts (WCE) prepared from N2 or EQ136 [HA-ddl-1 oe] adult animals grown on control, daf-2, akt-1, or a 1:1 mixture of daf-2 and daf-16 RNAi bacteria were subjected to immunoprecipitation (IP) and western blotting analysis (WB) with indicated antibodies. The total HSF-1 input of each IP experiment was measured by blotting each WCE sample with anti-HSF1 antibodies. (B) Whole cell extracts prepared from 293T cells over-expressing HA-DDL-1 were treated with buffer or 1U/μg protein CIP (alkaline phosphatase) for 1 hr. Samples were then subjected to western blotting analysis (WB) with anti-HA antibodies. (C) Whole cell extracts prepared from 293T cells over-expressing HA-tagged wild-type or mutated (T182A) DDL-1 were subjected to western blotting analysis (WB) using anit-HA or anti-phosphothreonine antibodies. (D) 293T cells were transfected with indicated combinations of pCMV-driven HA-DDL-1(WT), HA-DDL-1(T182A) or Myc-HSF-1 plasmids. Whole cell extracts prepared from these cells were immunoprecipitated (IP) and subsequently western-blotted (WB) using indicated antibodies. (E) The level of threonine phosphorylated DDL-1 is elevated in daf-2 mutants. WCE prepared from N2 or EQ136 worms grown on control or daf-2 RNAi bacteria were immunoprecipitated and western-blotted using indicated antibodies. Shown here (A, D, and E) are the immunoblots of a representative experiment. Quantification of at least three experiments is shown in Fig. S4.

Figure 6. Model of HSF-1 activation regulated by IIS in C. elegans

Upon heat stress stimulation, HSF-1 undergoes oligomerization, post-translational modification, and nuclear translocation in an undefined order before acquiring DNA-binding and transcriptional activity. The formation of a DDL-1 containing HSF-1 inhibitory complex (DHIC), consisting of HSF-1, HSB-1, DDL-1 and DDL-2, reduces the pool of HSF-1 susceptible to heat stress stimulation. Increased insulin/IGF-1-like signaling (IIS) promotes the formation of DHIC, while reducing DAF-2 activity promotes DDL-1 phosphorylation and disrupts DHIC formation and consequently increases HSF-1 activity under both stressed and unstressed conditions.