N-acylethanolamine signalling mediates the effect of diet on lifespan in Caenorhabditis elegans

Abstract

Dietary restriction is a robust means of extending adult lifespan and postponing age-related disease in many species, including yeast, nematode worms, flies and rodents. Studies of the genetic requirements for lifespan extension by dietary restriction in the nematode Caenorhabditis elegans have implicated a number of key molecules in this process, including the nutrient-sensing target of rapamycin (TOR) pathway and the Foxa transcription factor PHA-4 (ref. 7). However, little is known about the metabolic signals that coordinate the organismal response to dietary restriction and maintain homeostasis when nutrients are limited. The endocannabinoid system is an excellent candidate for such a role given its involvement in regulating nutrient intake and energy balance. Despite this, a direct role for endocannabinoid signalling in dietary restriction or lifespan determination has yet to be demonstrated, in part due to the apparent absence of endocannabinoid signalling pathways in model organisms that are amenable to lifespan analysis. N-acylethanolamines (NAEs) are lipid-derived signalling molecules, which include the mammalian endocannabinoid arachidonoyl ethanolamide. Here we identify NAEs in C. elegans, show that NAE abundance is reduced under dietary restriction and that NAE deficiency is sufficient to extend lifespan through a dietary restriction mechanism requiring PHA-4. Conversely, dietary supplementation with the nematode NAE eicosapentaenoyl ethanolamide not only inhibits dietary-restriction-induced lifespan extension in wild-type worms, but also suppresses lifespan extension in a TOR pathway mutant. This demonstrates a role for NAE signalling in ageing and indicates that NAEs represent a signal that coordinates nutrient status with metabolic changes that ultimately determine lifespan.

Figure 1. NAE levels in C. elegans are modulated by FAAH activity

a, Levels of NAEs in first day adult wild type N2 worms measured by SID-GC-MS (mean+s.d., n=5). POEA – palmitoleoyl ethanolamide; PEA– palmitoyl ethanolamide; LOEA – linoleoyl ethanolamide; OEA – oleoyl ethanolamide; EPEA eicosapentaenoyl ethanolamide; AEA – arachidonoyl ethanolamide. b, EPEA levels are elevated in first day eri-1(mg366) IV; lin-15B(n744) adults after exposure to faah-1 dsRNA by soaking (mean+s.d., n=2). c, EPEA levels are elevated in first day wild type N2 adults after 24 h exposure to 10 μM URB597, a chemical inhibitor of mammalian FAAH (mean+s.d., n=5, p<0.05, Wilcoxon Signed Rank Test). d, Over-expression of faah-1 results in reduced EPEA levels in first day wild type N2 adults (mean+s.d., N2 n=9, rfIs22 n=7 and rfIs23 n=8, p<0.05 for both rfIs22 and rfIs23, Wilcoxon Signed Rank Test).

Figure 2. NAEs affect reproductive growth and dauer formation

a, faah-1 over-expression results in developmental delay (mean+s.d., N2 n=54, rfIs23 n=76). b, faah-1 RNAi rescues the growth delay of faah-1 over-expressors (mean+s.d., N2 n=59, rfIs23 n=53). c, Levels of EPEA during development in N2 and daf-2(e1368) animals grown at 25°C (mean+s.d., n=2). Abbreviations: L1 – 1^st larval stage; L2 – 2^nd larval stage; L3 – 3^rd larval stage; L4 – 4^th larval stage; YA – young adults; GA – gravid adults; L2d – alternate L2 stage preceding the dauer molt; L2d* – later time point in L2d; D – dauer. d, Effect of treatment with exogenous NAEs on reproductive growth in daf-2(e1368) mutants at 24°C (mean+s.d., n=2). e, Scheme illustrating genes and pathways involved in dauer formation in C. elegans. f, EPEA rescues dauer formation in multiple dauer constitutive mutants (all p<0.0001, Chi-squared test, additional data in Supplementary Table 1).

Figure 3. Reduced NAE levels are associated with DR and are sufficient to confer lifespan expression

a, EPEA levels are reduced in starved L1 larvae and increase after 6 h of exposure to food (mean+s.d., n=3). b, EPEA levels are altered in response to food availability in adult wild type N2 animals (mean+s.d., Mann-Whitney U test: 12 h Fed (n=6) vs DR (n=12) p<0.05; 24 h Fed (n=7) vs DR (n=7) p<0.001; 24 h DR vs Re-fed (n=6) p<0.005; 24 h Fed vs Re-fed p=n.s.). c, faah-1 over-expression extends lifespan in N2 wild type animals under fed conditions (1x10¹⁰ cfu/ml E. coli, p<0.0001, Log-Rank test). d, Lifespan is not different between N2 and a faah-1 over-expressing line under conditions of optimal DR (1×10⁹ cfu/ml E. coli). e, faah-1 over-expression extends lifespan in N2 wild type animals under conditions of sub-optimal DR conditions (1×10⁸ cfu/ml E. coli, p<0.0001, Log-Rank test). f, faah-1 over-expression affects lifespan in a nutrient dependent manner (mean lifespan + s.d., n=3). g, faah-1 over-expression extends lifespan in a daf-16 mutant (p<0.0001, Log-Rank test). h, Lifespan extension resulting from faah-1 over-expression requires the Foxa transcription factor PHA- 4 (N2 Control vs N2 + pha-4 RNAi, p<0.0001; rfIs22 Control vs rfIs22 + pha-4 RNAi p<0.0001, rfIs22 Control vs N2 Control p=0.0014, Log-Rank test).

Figure 4. EPEA suppresses the effects of dietary restriction on lifespan

a, EPEA treatment reduces lifespan in wild type N2 animals on control RNAi bacteria (p<0.0001, Log-Rank test). b, EPEA treatment reduces lifespan in daf-2(e1368) mutants on control RNAi bacteria (p=0.0005, Log-Rank test). c, EPEA has a minimal effect on N2 lifespan in the presence of high food concentrations (1×10¹¹ cfu/ml E. coli, p<0.0001, Log-Rank test). d, EPEA treatment completely suppresses the effect of optimal DR on wild type N2 lifespan (1×10⁹ cfu/ml E. coli, p<0.0001 Log-Rank test). e, EPEA levels are reduced in rsks-1(ok1255) mutants, a genetic model of DR (mean+s.d., n=4, p<0.05, Mann-Whitney U test). f, EPEA treatment suppresses lifespan extension in rsks-1(ok1255) mutants (p<0.0001, Log-Rank test).