HIF-1-dependent regulation of lifespan in Caenorhabditis elegans by the acyl-CoA-binding protein MAA-1

Abstract

In yeast, the broadly conserved acyl-CoA-binding protein (ACBP) is a negative regulator of stress resistance and longevity. Here, we have turned to the nematode C. elegans as a model organism in which to determine whether ACBPs play similar roles in multicellular organisms. We systematically inactivated each of the seven C. elegans ACBP paralogs and found that one of them, maa-1 (which encodes membrane-associated ACBP 1), is indeed involved in the regulation of longevity. In fact, loss of maa-1 promotes lifespan extension and resistance to different types of stress. Through genetic and gene expression studies we have demonstrated that HIF-1, a master transcriptional regulator of adaptation to hypoxia, plays a central role in orchestrating the anti-aging response induced by MAA-1 deficiency. This response relies on the activation of molecular chaperones known to contribute to maintenance of the proteome. Our work extends to C. elegans the role of ACBP in aging, implicates HIF-1 in the increase of lifespan of maa-1-deficient worms, and sheds light on the anti-aging function of HIF-1. Given that both ACBP and HIF-1 are highly conserved, our results suggest the possible involvement of these proteins in the age-associated decline in proteostasis in mammals.

Keywords: ACBP; C. elegans; HIF-1; aging; proteostasis.

Figure 1. ACBPs regulate C. elegans lifespan

(A) Downregulation of maa-1 promotes longevity. Lifespans of wildtype worms (N2) subjected to control RNAi vs maa-1 dsRNA, P<0.0001. (B) Downregulation of acbp-1 or acbp-3 modestly increases longevity. Lifespans of wild-type worms subjected to control RNAi vs acbp-1 RNAi, P<0.01; and control RNAi vs acbp-3 RNAi, P=0.056. (C-D) Loss-of-function mutations maa-1(ok2033) (C) and maa-1(sv38) (D) prolong lifespan (both P<0.0001). P values were calculated using the log-rank (Mantel-Cox) method. Replicate experiments and details of statistical analyses are shown in Table S1 and S2.

Figure 2. Loss of maa‐1 promotes heat, oxidative, and proteotoxic stress resistance

A‐B) maa‐1(ok2033) mutants showed enhanced resistance to incubation at 35°C (A) and exposure to 150 mM paraquat (B). Error bars represent SEM from three independent experiments (*P<0.05, **P<0.01 compared to wildtype worms by Student's t‐test). (C‐D) maa‐1 RNAi increases resistance to paralysis induced by aggregation of a 35‐residue polyglutamine repeat (C; P<0.0001) or human β‐amyloid (D; P<0.0001). P values were calculated using the log‐rank (Mantel‐Cox) method. Replicate experiment is shown in Figure S2.

Figure 3. MAA-1 functions predominantly in the intestine to regulate longevity

Intestinal-specific RNAi is sufficient to extend longevity. Lifespans of rde-1(ne219) mutants in which rde-1 expression is restored in the intestine (A) or the hypodermis (B); animals were subjected to control or maa-1 RNAi (P<0.0001 and P<0.05 for A and B, respectively). (C) Lifespans of the control strain rde-1(ne219) subjected to control or maa-1 RNAi (P=0.8513). P values were calculated using the log-rank (Mantel-Cox) method. Replicate experiments are shown and additional statistical analysis are shown in Table S1 and S2.

Figure 4. HIF‐1 mediates lifespan extension of maa‐1–deficient animals

(A) Downregulation of hif‐1 reduces the longevity of maa‐ 1(ok2033) mutants. Lifespans of wild‐type and maa‐1(ok2033) mutants subjected to hif‐1 or control RNAi (P=0.1091 for control RNAi vs hif‐1 RNAi of maa‐1(ok2033) mutants). (B) Deletion of hif‐1 reverses the lifespan extension conferred by the maa‐1 mutation. (P=0.2661 for wild‐type vs maa‐1(ok2033); hif‐1(ia4) mutants; P<0.0001 for maa‐1(ok2033) vs maa‐1(ok2033); hif‐1(ia4) mutants). (C) qPCR analysis of the HIF‐1 targets nhr‐57 and F22B5.4 in maa‐1(ok2033) and maa‐1(ok2033); hif‐1(ia4) mutants. Results are relative to levels in wildtype animals. Error bars represent SEM (t‐test: *P<0.05, **P<0.001 for maa‐1(ok2033) vs wildtype animals). (D) HIF‐1 stability is not affected by downregulation of maa‐1. Western blot of protein extracts from wildtype, transgenic iaIs34 animals carrying HIF‐1 (P621G)::myc, and transgenic iaIs28 animals carrying HIF‐1::myc, subjected to control or maa‐1 RNAi. Blots were probed with anti‐myc and anti‐α‐tubulin antibodies (upper panel). Quantification of band intensity is shown in lower panel (N=3, **P<0.001). (E) A maa‐1 loss‐of‐function mutation does not further increase the lifespan of long‐lived vhl‐1(ok161) mutants (P=0.0692 for vhl‐1(ok161) vs maa‐1(ok2033); vhl‐1(ok161), P=0.8449 for maa‐1(ok2033) vs maa‐1(ok2033); vhl‐1(ok161)). (F) Downregulation of maa‐1 does not affect the lifespan of long‐lived transgenic animals overexpressing HIF‐1(P621G)::myc (hif‐1 OE). (P=0.0678 for hif‐1 OE on control vs maa‐1 RNAi). P values were calculated using the log‐rank (Mantel‐Cox) method. Replicate experiments are shown in Table S1. Additional statistical analysis is shown in Table S2.

Figure 5. DAF-16 is required for lifespan extension in maa-1–deficient animals

Lifespan of maa-1(ok2033); daf-16(mu86) double mutants is significantly shorter than that of maa-1(ok2033) single mutants (P<0.0001). P values were calculated using the log-rank (Mantel-Cox) method. Replicate experiments and additional statistical analysis are shown in Table S1 and S2.

Figure 6. Molecular chaperones promote longevity downstream of HIF‐1 in maa‐1–deficient mutants

(A) The expression of two shsp genes is induced by maa‐1 deficiency. qPCR analysis of hsp‐16.1 and hsp‐16.49 in wildtype, maa‐1(ok2033), maa‐1(ok2033); hif‐1(ia4), and maa‐1(ok2033); daf‐16 (mu86) animals. mRNA levels are expressed relative to those in wildtype animals (t‐test: *P<0.05 and **P<0.001 for maa‐1(ok2033) and maa‐1(ok2033); daf‐16 (mu86) vs wildtype) respectively. (B) Downregulation of shsp expression has little effect on the lifespan of wildtype animals (P=0.1024 and P=0.2988 for control vs hsp‐16.1 and hsp‐16.49, respectively). (C) Downregulation of shsp expression markedly shortens the lifespan of maa‐1(ok2033) mutants (P<0.0001 for control vs each RNAi). P values for lifespan analyses were calculated using the log‐rank (Mantel‐Cox) method. Replicate experiments and additional statistical analysis are shown in Table S1 and S2.