Signalling through RHEB-1 mediates intermittent fasting-induced longevity in C. elegans


Dietary restriction is the most effective and reproducible intervention to extend lifespan in divergent species1. In mammals, two regimens of dietary restriction, intermittent fasting (IF) and chronic caloric restriction, have proven to extend lifespan and reduce the incidence of age-related disorders2. An important characteristic of IF is that it can increase lifespan even when there is little or no overall decrease in calorie intake2. The molecular mechanisms underlying IF-induced longevity, however, remain largely unknown. Here we establish an IF regimen that effectively extends the lifespan of Caenorhabditis elegans, and show that the low molecular weight GTPase RHEB-1 has a dual role in lifespan regulation; RHEB-1 is required for the IF-induced longevity, whereas inhibition of RHEB-1 mimics the caloric-restriction effects. RHEB-1 exerts its effects in part by the insulin/insulin growth factor (IGF)-like signalling effector DAF-16 in IF. Our analyses demonstrate that most fasting-induced upregulated genes require RHEB-1 function for their induction, and that RHEB-1 and TOR signalling are required for the fasting-induced downregulation of an insulin-like peptide, INS-7. These findings identify the essential role of signalling by RHEB-1 in IF-induced longevity and gene expression changes, and suggest a molecular link between the IF-induced longevity and the insulin/IGF-like signalling pathway.


In an IF regimen, which has not been established in invertebrate model organisms, food is provided ad libitum to both control and experimental groups, but the experimental group is subjected to periods of fasting. We tested two IF regimens, an alternate-day fasting and an every 2 days fasting in C. elegans—an organism that has been shown to be an excellent model system for ageing research—and found that they increased lifespan by 40.4% and 56.6%, respectively (Fig. 1a, b and Supplementary Table 1). Therefore, we used fasting every 2 days as the IF regimen. This IF regimen increased resistance to heat and oxidative stress (Fig. 1c), and markedly delayed the age-related physiological decline. As animals age, the locomotion activity and muscle integrity decrease, showing impairment of cellular functions. IF markedly suppressed the age-dependent decline in these activities (Fig. 1d and Supplementary Fig. 1), suggesting that ageing is delayed in C. elegans by the IF regimen established here.

Figure 1: IF extends C. elegans lifespan.

a, Schematic representation of an IF regimen. AL, ad libitum; UV, ultraviolet. b, IF increases C. elegans lifespan. Survival curves of IF and ad libitum worms (wild type N2). c, IF increases resistance to heat (left) and oxidative stress (right). Similar results were obtained in two independent experiments. d, IF delays the age-dependent decline in locomotion activity. *P < 0.05, **P < 0.01, t-test. e, The effects of caloric restriction and IF on lifespan (measured at bacteria concentrations of 5 × 1010, 5 × 109 and 5 × 108 bacteria ml-1) are additive. **P < 1.5 × 10-7, ***P < 1.0 × 10-11, t-test. Error bars, s.e.m.

PowerPoint slide

To examine the relationship between IF and caloric restriction, we chose the solid dietary restriction method (dilution of E. coli on agar plates) as a caloric-restriction assay to perform IF and caloric restriction in the isogenic backgrounds. Consistent with previous reports, chronic restriction of food intake extended lifespan significantly; the mean lifespan in caloric restriction (5.0 × 108 bacteria ml-1) was increased by 13.2% (P < 1.1 × 10-5, t-test) compared to that in ad libitum (5.0 × 1010 bacteria ml-1; Fig. 1e). We introduced the IF regimen to these food-restricted animals. Our results showed that the effects of IF and caloric restriction are overlapping. The extent of lifespan extension by IF in ad libitum was significantly larger than that in caloric restriction (66.5% versus 51.3%). Furthermore, the mean lifespans of worms subjected to IF under both conditions are not statistically different (41.9 and 43.1 days, respectively; P = 0.38, t-test). Similarly, IF extended the lifespan of the clk-1 mutant clk-1(e2519), which is also reported to be long-lived owing to a common caloric-restriction pathway with eat-2 mutations3, to a lesser extent than that of wild type N2 (Supplementary Fig. 2). As the effects of IF and caloric restriction are overlapping, we examined the potential roles of two genes, skn-1 and pha-4, which have been shown to have essential roles in other caloric-restriction regimens, such as dilution of E. coli in liquid cultures4 and eat-2 mutants5. Notably, our IF experiments in a skn-1-null mutant, skn-1(zu135), and in pha-4 RNA interference (RNAi)-subjected animals showed that these genes are dispensable for IF-induced longevity (Supplementary Fig. 2).

In diverse species, TOR activity was downregulated by starvation and inhibition of TOR extended lifespan in a manner similar to caloric restriction6,7,8. In contrast, RHEB-1, an upstream activator of TOR, was reported to be induced in response to low nutrient levels; the expression of Drosophila Rheb was induced by protein starvation and downregulated by subsequent refeeding9. These may suggest the complexity of RHEB-1 and TOR signalling responses. We first identified C. elegans RHEB-1 as F54C8.5/rheb-1 (Supplementary Fig. 3a). rheb-1 RNAi resulted in phenotypes similar to those of TOR-deficient worms (C. elegans TOR, B0261.2/let-363), such as suppression of endo-reduplication of intestinal nuclei (Supplementary Fig. 3b–e)10. These results indicate that F54C8.5 is a C. elegans orthologue of Rheb. We then found that from early stages to adulthood, RHEB-1::GFP (a translational fusion of green fluorescent protein to the carboxy terminus of rheb-1 under the rheb-1 promoter) is expressed ubiquitously (Supplementary Fig. 3f).

We examined the role of C. elegans RHEB-1 and TOR (encoded by the let-363 gene) in dietary restriction-induced longevity. To downregulate genes of interest, worms were fed on double-stranded RNA (dsRNA)-producing E. coli after hatching up until day 2 of adulthood, and adult worms were then subjected to either caloric restriction (Fig. 2a) or IF (Fig. 2b). rheb-1 RNAi successfully suppressed endogenous RHEB-1 and RHEB-1::GFP expression (Supplementary Fig. 4). Caloric restriction extended the lifespan of control RNAi-treated animals by 18.2% (Fig. 2a, left). Consistent with previous reports, caloric restriction failed to extend the lifespan of TOR (let-363) RNAi-treated animals (data not shown). We also found that rheb-1 RNAi extended lifespan by mimicking the caloric-restriction effects (Fig. 2a, middle). Under the ad libitum condition, rheb-1 RNAi extended lifespan by 19.1%, and the longevity-promoting effect was not seen in the caloric-restriction condition (Fig. 2a, right). Next, we examined the role of the RHEB-1 and TOR pathway in IF (Fig. 2b). IF successfully extended the lifespan of control RNAi-treated animals by 49.3% (Fig. 2b, control RNAi). Surprisingly, we found that inactivation of RHEB-1 and TOR signalling did not mimic the IF effect but suppressed the IF-induced longevity. The mean lifespans of rheb-1 RNAi-treated ad libitum and IF worms were 27.7 and 28.7 days, respectively (Fig. 2b, rheb-1 RNAi). This clearly indicates the requirement of RHEB-1 for IF-induced longevity. Notably, similarly to rheb-1 RNAi, TOR (let-363) RNAi also suppressed the IF-induced longevity, but the effect of TOR (let-363) RNAi was smaller than that of rheb-1 RNAi; the mean lifespans of TOR (let-363) RNAi-treated ad libitum and IF worms were 21.6 and 27.5 days, respectively (Fig. 2b, TOR (let-363) RNAi). To exclude the possibility that this weak effect of TOR (let-363) RNAi is due to insufficient knockdown of TOR, we performed RNAi throughout the whole lifespan using both the feeding and soaking methods (Supplementary Fig. 5). This method of TOR (let-363) RNAi also partially suppressed the IF-induced longevity (37.4% lifespan extension by IF), whereas rheb-1 RNAi almost completely suppressed it (6.0% lifespan extension by IF; Supplementary Fig. 5 and Supplementary Table 1). Therefore, it is unlikely that the partial suppression of IF induced-longevity by TOR (let-363) RNAi is due to insufficient knockdown of TOR. To confirm the function of rheb-1 in adulthood, we performed the soaking RNAi after day 2 of adulthood in IF experiments. Under these conditions, IF increased the lifespan of control RNAi-, rheb-1 RNAi- and TOR (let-363) RNAi-treated worms by 48.4%, 25.3% and 41.3%, respectively, as compared to that under ad libitum conditions (Supplementary Fig. 6 and Supplementary Table 1), supporting the notion that RHEB-1 is required for IF-induced longevity. These results indicate that RHEB-1 mediates the IF effects in both TOR-dependent and -independent manners. These results, together with the recent study showing that the absence of food can act as signal independent of calorie intake11, also suggest that IF is not just an extreme caloric restriction although the caloric restriction and IF effects are overlapping, and that signalling molecules, which mediate caloric restriction and IF stimuli, are different.

Figure 2: RHEB-1 has a dual role in dietary restriction.

a, rheb-1 RNAi mimics longevity effects by caloric restriction (CR). Schematic representation of caloric-restriction experiments (top). Survival curves at different concentrations of bacteria (in parentheses) (bottom left and middle). Mean lifespans ± s.e.m. of at least two independent experiments are shown (bottom right, bacterial concentrations shown on the x-axis). AL, ad libitum; UV, ultraviolet. *P < 0.05, ***P < 5 × 10-5, t-test. b, rheb-1 RNAi and TOR (let-363) RNAi suppress longevity effects by IF. Schematic representation of IF experiments (top). Survival curves of control RNAi- (bottom left), rheb-1 RNAi- (bottom middle) and TOR (let-363) RNAi- (bottom right) treated worms in IF. Mean lifespans ± s.e.m. of three independent experiments are shown (bottom far right). *P < 0.05, t-test.

PowerPoint slide

Because RHEB-1 and TOR signalling functions to promote protein synthesis—the decrease of which increases the lifespan of worms7,12—we explored the potential roles of translation-related genes in IF-induced longevity. We performed RNAi for rps-6 (ribosomal protein subunit 6) in adulthood, and used null mutants for rsks-1 (p70 S6K) and ife-2 (translation initiation factor 4E). Both rsks-1(ok1255) and ife-2(ok306) mutants responded to IF normally (Supplementary Fig. 7). Inactivation of rps-6 extended lifespan under the ad libitum condition, whereas IF further extended the lifespan of rps-6 RNAi-treated worms to the same extent as that of control RNAi-treated animals by IF (Supplementary Fig. 7 and Supplementary Table 1). These results indicate that decreased translation efficiency cannot account for IF-induced longevity.

We tested several known longevity-regulating and nutrient-sensing genes for their roles in IF. We used their null mutants—daf-16(mgDf50), daf-16(mu86), aak-2(ok524), sir-2.1(ok434), cep-1(gk138) and Y81G3A.3(OK886) (C. elegans GCN2)13,14,15,16—and found that the IF-induced increase in lifespan was significantly diminished only in two null mutants of daf-16 (P = 0.0037 in daf-16(mgDf50), and P = 0.0003 in daf-16(mu86); Fig. 3a and Supplementary Table 1). To examine the possibility that the IF regimen used is not optimized for daf-16 null mutants, we subjected daf-16(mu86) to three kinds of IF regimens: 24 h, 48 h or 72 h of fasting per 4 days. The results showed that all the regimens extended the lifespan of daf-16(mu86) to a lesser extent than wild type N2 (Supplementary Fig. 8), showing that daf-16 partially mediates IF-induced longevity.

Figure 3: DAF-16 partially mediates IF-induced longevity.

a, Two loss-of-function mutations of daf-16 suppress IF-induced longevity. Mean lifespans ± s.e.m. of at least three experiments are shown. *P < 0.001, **P < 0.0005, t-test using lifespan extension by IF. AL, ad libitum. b, Fasting-induced DAF-16::GFP nuclear localization is suppressed by rheb-1 RNAi in intestine. Representative images are shown (left). Arrows indicate intestine, and white arrowheads indicate other tissues. Scale bar, 100 μm. Percentage of worms with nuclear localization of DAF-16::GFP in intestine after 15 h of fasting were scored in two independent experiments (right). Data are mean ± s.e.m. *P < 0.05, t-test.

PowerPoint slide

DAF-16, the forkhead transcription factor, mediates the effect of the insulin-like signalling pathway on ageing14. Environmental stresses, such as starvation, trigger DAF-16 nuclear translocation17. We examined whether rheb-1 RNAi affects fasting-induced nuclear translocation of DAF-16. In control worms, DAF-16::GFP modestly translocated to the nucleus in response to fasting (Fig. 3b). In contrast, this fasting-triggered nuclear localization of DAF-16::GFP in the intestine was suppressed by rheb-1 RNAi (Fig. 3b). Downstream targets of DAF-16—sod-3, mtl-1, hil-1 and dod-6 (ref. 18)—were induced by fasting, and their induction was suppressed by rheb-1 RNAi and TOR (let-363) RNAi (Supplementary Fig. 9, left and middle). Together these findings suggest that DAF-16 mediates, at least in part, the functions of signalling through RHEB-1 in IF-induced longevity. We found that the expression levels of two genes—rab-10 and pha-4, the expression levels of which are reported to be downregulated19 or upregulated5, respectively, by caloric restriction—were not affected by fasting (Supplementary Fig. 9, right). Interestingly, the expression level of pha-4 is upregulated by rheb-1 RNAi and TOR (let-363) RNAi (Supplementary Fig. 9, right), suggesting the possibility that inactivation of RHEB-1 and TOR signalling mimics the caloric-restriction effects through induction of pha-4 and its downstream targets. It is also reported that the longevity-promoting effect of pha-4 overexpression is independent of daf-16, and pha-4 is not required for the long lifespan of daf-2(e1368)5. Therefore, there may be two independent signalling pathways downstream of RHEB-1 and TOR signalling, one of which is the daf-16 pathway that could mediate fasting-induced longevity.

We next examined the gene expression changes during fasting in rheb-1 RNAi-treated, TOR (let-363) RNAi-treated and control RNAi-treated worms. We performed a genome-wide analysis using Affymetrix GeneChip oligonucleotide microarrays, and compared the gene expression profiles at day 4 of adulthood (see Fig. 4a). We first focused on the effect of fasting, and identified 112 genes which were upregulated by fasting more than threefold with statistical significance in control RNAi-treated worms (see Methods). Surprisingly, the fasting-induced upregulation of most of the 112 genes was dependent on RHEB-1 or TOR (100 genes in RHEB-1 and 94 genes in TOR, Fig. 4b). These 100 RHEB-1-dependent genes can be divided into two classes: genes in which the expression level in rheb-1 RNAi-fasting is either similar to (59 genes) or less than half (41 genes) that in control-fasting (Fig. 4b). These results show that rheb-1 RNAi and TOR (let-363) RNAi both suppress the induction of these genes by fasting, either by inhibiting fasting-induced upregulation or by inducing upregulation under the fed condition. We then analysed the 112 fasting-induced upregulated genes by scatter plotting (Fig. 4c). The plots of the expression levels in control-fasting versus those in TOR (let-363) RNAi-fasting (Fig. 4c, upper left) or those in rheb-1 RNAi-fasting (Fig. 4c, upper right) show that 24 (yellow) and 42 (red plus yellow) genes are downregulated more than twofold by TOR (let-363) RNAi and rheb-1 RNAi, respectively, and that the 24 genes are included completely in the 42 genes (see also Fig. 4c, lower). This clearly demonstrates that the upregulation of several genes (18 out of 42 genes, 43%) by fasting is dependent on RHEB-1, but not on TOR (Fig. 4c, lower). These results indicate the existence of a TOR-independent pathway, downstream of RHEB-1, which would also mediate IF stimuli to extend lifespan.

Figure 4: Microarray analyses identify fasting-regulated genes involved in IF-induced longevity.

a, The sampling scheme is shown. b, d, Expression profiles of fasting-induced up-regulated (b, left) and down-regulated (d) genes. Average expression profiles are shown (b, right, see text). c, Scatter plots of the expression levels for TOR (let-363) RNAi-fasting (top left) or rheb-1 RNAi-fasting (top right) worms. The black and blue lines indicate the diagonal and twofold changes between two samples. Venn diagram of fasting-induced 112 genes (bottom). e, Quantitative PCR with reverse transcription (qRT–PCR) of ins-7 expression. f, HSP-12.6, a downstream target of DAF-16, functions to mediate IF-induced longevity.

PowerPoint slide

We focused on the fasting-induced genes (41 genes, Supplementary Table 2), the induction of which is abolished by rheb-1 RNAi. We tested several of these for their role in IF-induced longevity by using null mutants (Supplementary Table 1). The loss-of-function mutation of hsp-12.6, which encodes a C. elegans orthologue of a small heat shock protein αB-crystallin, suppressed the IF-induced increase in lifespan to a similar extent to that in daf-16(mu86) (Fig. 4f), suggesting that hsp-12.6 is one of the downstream targets of DAF-16 in IF-induced longevity. In daf-16(mu86);hsp-12.6(gk156), the extent of IF-induced longevity was similar to that in single mutants hsp-12.6(gk156) or daf-16(mu86) (Figs 3a, 4f and Supplementary Table 1), confirming that daf-16 and hsp-12.6 function in the same signalling pathway. Next, we used long-lived daf-2 mutants. Low insulin/IGF-like signalling in daf-2(e1370) is known to result in constitutive activation of DAF-16 and the higher expression of hsp-12.6 than in wild type18,20. IF did not markedly extend the lifespan of daf-2(e1370) mutants (Fig. 4f), suggesting that IF acts by decreasing daf-2 activity. Furthermore, a reduction-of-function mutation in hsf-1, which has been shown to act downstream of daf-2 to promote longevity by upregulating hsp-12.6 (ref. 20), also suppressed IF-induced longevity (Supplementary Table 1). Collectively, our data support our idea that reduced daf-2 signalling, which leads to activation of daf-16, hsf-1 and hsp-12.6, mediates the IF effects. Upregulated hsp-12.6 has been reported to render daf-2(e1370) resistant to proteotoxicity, and mutations in αB-crystallin in mammals were reported to cause protein aggregation myopathies21, suggesting that there is an evolutionarily conserved role for the protein in resistance against proteotoxicity. We found that HSP-12.6::GFP expression is induced by fasting in various tissues including body wall muscles and neuronal systems (Supplementary Fig. 10).

There were 298 genes that were significantly downregulated by fasting more than threefold in control worms. These genes showed decreased expression levels after fasting even in rheb-1 RNAi- or TOR (let-363) RNAi-treated worms (Fig. 4d). Out of the 298 genes, only one, ins-7, showed the higher expression level after fasting in rheb-1 RNAi and TOR (let-363) RNAi-treated than in control-fed worms (Fig. 4d, e). It is probable that fasting-induced downregulation of ins-7—the finding shown previously22 and confirmed here (Fig. 4e)—mediates IF-induced longevity, because ins-7 is shown to negatively regulate longevity by inhibiting daf-16 activity in a daf-2-dependent manner18. In addition, in ZK1251.1&ins-7(ok1573)—the strain which carries a null mutation of ins-7——the lifespan-extending effect of IF was significantly suppressed (Supplementary Table 1, P = 0.018, t-test). However, in spite of the marked decrease in the messenger RNA level of ins-7, a null deletion of ins-7 affected IF-induced longevity only modestly. This weak phenotype of ins-7 deletion may be due to compensation by other insulin-like peptides (ins genes). We found that GFP expression under the promoter of one of the ins genes, daf-28, is markedly suppressed by fasting (Supplementary Fig. 11). As daf-28 is shown to negatively regulate lifespan23, these results indicate that only when we could suppress several such agonistic ins genes simultaneously, we could appreciably mimic the longevity promoting effect of fasting and markedly affect IF-induced longevity. Interestingly, in D. melanogaster, suppressing Tor function increases the level of Dilp2 (also known as Ilp2), an insulin-like peptide24. It is thus possible that the regulation of insulin/IGF expression by RHEB-1 and TOR signalling is evolutionarily conserved.

Recent studies have shown that diverse protocols of dietary restriction extend C. elegans lifespan through different signalling molecules4,5. Dietary food deprivation, in which animals are maintained on plates without food throughout their lives, extends lifespan independently of daf-2 and daf-16 signalling25,26, although this regimen is related to IF. DAF-16 translocates to the nucleus in response to fasting, but relocates to the cytoplasm after prolonged fasting (more than 48 h)27. Therefore, worms may have an adaptation system to fasting concerning insulin/IGF-like signalling. Thus, prolonged fasting should be different from IF. A recent study reported that daf-16 mediates the effect of solid dietary restriction, a kind of caloric restriction13. In the study, aak-2, a C. elegans orthologue of AMPK, is reported to be an upstream regulator of daf-16, whereas aak-2 is dispensable for our IF-induced longevity. Instead, daf-2 is acting as an upstream regulator of daf-16 in our study. Thus, signalling pathways mediating IF effects should not be the same as those mediating the other caloric restriction regimen-induced effects. The existence of diverse signalling pathways for so-called ‘dietary restriction’ may be reasonable, as organisms are exposed to unlimited kinds of stimuli during their lives. Accumulating evidence suggests that these different signalling pathways may converge on the limited number of signalling molecules or physiological reactions. One of these is protein turnover. Autophagy is shown to be essential for longevity in eat-2 and daf-2 mutants28, and reduced translation is shown to extend lifespan by mimicking caloric-restriction effects7,12. Bacterial food deprivation as well as lowering insulin/IGF-like signalling confers longevity and resistance to proteotoxicity through hsf-1 (ref. 29). These findings emphasize the importance of protein quality control. It should be noted that the same pathways or molecules could have different effects on lifespan depending on the ways of dietary restriction or the signalling contexts, as is the case with this study and the previous study in yeast30. Further studies are required to clarify the whole picture of signalling networks in dietary restriction. Our results may help the development of dietary restriction mimetics that improve our health without toxic side effects.

Methods Summary

Lifespan analysis

Approximately 25 young adults per each plate were moved to nematode growth medium (NGM) or RNAi plates containing 200 μg ml-1 of 5′fluoro-2′deoxyuridine (FUDR) 3 days after hatching. An adult was scored as dead when it did not respond to a mechanical stimulus. Animals that crawled off the plate, displayed extruded internal organs, or died from internally hatched progeny were censored and excluded from the statistical analysis. All experiments were performed at 20 °C and at least duplicated. P values were calculated using a log-rank test or t-test with the mean lifespan. See Supplementary Table 1 for mean lifespan, standard error of mean and P values of all IF experiments.

Online Methods

C. elegans strains

All nematodes were cultured using standard C. elegans methods31. The strains we analysed were: wild type N2, sir-2.1(ok434)(3), daf-16(mgDf50), daf-16(mu86)(3), rsks-1(ok1255)(2), rap-2(gk11)(2), akk-2(ok524)(3), ikb-1(n2027), kin-1(ok338)/mIs13, +/szT1[lon-2(e678)] I; kin-2(ok248)/szT1 X, ife-2(ok306)(2), ZK1251.1&ins-7(ok1573)(2), Y81G3A.3(ok886)(4), hsp-12.6(gk156)(4), cep-1(gk138)(2), clk-1(e2519)(2), skn-1(zu135)(3), hsf-1(sy441)(2), ccls4251(1), hil-1(gk229)(2), GR1455 mgls40[Pdaf-28::gfp] and TJ356 zls356[daf-16::gfp; rol-6](2). The numbers of outcrossing are shown in parentheses.

Intermittent fasting

Approximately 100 synchronized young adult worms raised on NGM plates with live OP50 were picked to FUDR-containing plates with live OP50. At day 2 of adulthood, worms were divided into ad libitum and IF. Worms in ad libitum were fed ultraviolet-killed OP50 ad libitum throughout their lifespan. Worms in IF were on plates with ultraviolet-killed OP50 or without food alternatively every other day. All worms were transferred to new plates every other day.

Solid dietary restriction

Solid dietary restriction was performed as described13 with modifications. The OP50 concentration was calculated by counting colony performing units. OP50 was resuspended and diluted in water. One-hundred-and-fifty microlitres of these solutions were plated and bacteria were immediately killed by ultraviolet irradiation.

RNA interference

RNAi was performed by the feeding method32 and/or the soaking method33 as described. The first 500 nucleotides of the coding region of F54C8.5/rheb-1, TOR/let-363a, rps-6 and pha-4c complementary DNA were used for RNAi. The primers used were: rheb-1 forward, 5′-ATGAGCAGTTCGCTGCAA-3′; rheb-1 reverse, 5′-AACACCTCATGCACTCGA-3′; TOR (let-363) forward, 5′-ATGCTCCAACAACACGGA-3′; TOR (let-363) reverse, 5′-GCTTTTGAAGCCATCTTG-3′; rps-6 forward, 5′-ATGAGACTTAACTTCGCC-3′; rps-6 reverse, 5′-CCATCTGGGAAGGTCTTG-3′; pha-4 forward, 5′-ATGAACGCTCAGGACTATCT-3′; pha-4 reverse, 5′-CAATTGACATTGTCTGAAAT-3′. Primers were fused with restriction enzyme sites (for feeding RNAi) or T3 and T7 promoter sequences (for soaking RNAi). In brief, in feeding RNAi, each cDNA segment was cloned into the feeding vector pPD129.36 with a Sac2 and a Kpn1 site and transformed to HT115 bacterial cells. Control animals were fed bacteria carrying an empty pPD129.36 vector. In soaking RNAi, cDNA segments were fused with T3 and T7 promoter sequences and RNA was synthesized by T3 and T7 RNA polymerase. Worms were soaked in soaking buffer containing 1–1.5 mg ml-1 dsRNA. Control animals were soaked with soaking media without dsRNA.

Generation of transgenic animals

To generate GFP reporter constructs, fragments containing the 5′ upstream sequence and coding region were amplified using PCR. The sizes of fragments were: F54C8.5 2.2 kilobase (kb), hsp-12.6-1kbp 1.6 kb and hsp-12.6-5kbp 5.5 kb. These fragments were cloned into the GFP vector pPD95.75. Transgenic worms were generated by microinjecting these plasmids into wild type N2 worms at 20 ng μl-1, 20 ng μl-1 and 50 ng μl-1, with 50 ng μl-1 pRF4 rol-6 transformation marker.


Total RNA was extracted with Sepasol(R)-RNA ISuper (Nacalai tesque), purified with RNeasy Mini Kit (Qiagen), and reverse transcribed into cDNA using M-MLV reverse transcriptase (Invitrogen) with dT primer, according to manufacturers’ instructions. cDNA was subjected to qPCR analysis using the ABI 7300Real Time PCR System (Applied Biosystems) with SYBR GreenPCR Kit (Roche). Each value was normalized to act-1, and the value in control RNAi-fed was set to 1. Primer sequences are available on request.

Heat and oxidative stress resistance

Synchronous worms at 10 day of adulthood were washed with M9 buffer and resuspended in M9 buffer with 20 animals per well. At time 0, hydrogen peroxide was added to the well in a final concentration 2.5 mM. The surviving worms were scored every hour. Heat stress was given by incubating worms on NGM plates at 35 °C. The surviving worms were scored every 5 h.

Locomotion activity

The length of worm tracks in 1 h was photographed and measured using the software application Axioplan2.

Fluorescent microscopy

For intestinal lysosomes, animals were anaesthetized with 10 mM sodium azide in M9 buffer at 120 h after synchronization and observed with Axioplan2. For intestinal nuclei, 4-day-old adult worms were fixed with 75% methanol in PBS for 1 h, washed twice with PBS and incubated with 0.1% Hoechst. For GFP::NLS in body wall muscles, the worms carrying the transgene ccls4251 were fixed with 3% formaldehyde in PBS for 5 min at room temperature. For HSP-12.6::GFP expressing strains, worms were anaesthetized with 10 mM sodium azide in M9 buffer at 4 day of adulthood. In DAF-16 localization assay, worms expressing DAF-16::GFP were synchronized and grown in the following conditions: control fed, control fasting, rheb-1 RNAi-fed and rheb-1 RNAi-fasting. After 15 h fasting, worms were fixed with 3% formaldehyde in PBS for 5 min at room temperature.

Body size

Animals were synchronized by breaching gravid hermaphrodites. One-hundred-and-twenty hours after synchronization, worms were anaesthetized with 10 mM sodium azide in M9 buffer and photographed using Axioplan2. To measure body length and width, segmented lines were drawn and the length of lines was calculated using the software application Axioplan2.

Microarray experiments

Five, four and three independent experiments were performed in control RNAi, rheb-1 RNAi and TOR (let-363) RNAi, respectively (see Fig. 4a). Total RNA was extracted as described above. Other procedures were performed according to Affymetrix protocols. Hybridized arrays were scanned using an Affymetrix GeneChip Scanner. Scanned chip images were analysed with GeneChip Operating Software v.1.4 (GCOS), and processed using default settings. The Affymetrix output (CEL files) was imported into GeneSpring 7.3 (Agilent Technologies) microarray analysis software for both statistical analysis and presentation of expression profiles (average expression profiles and scatter plots). Expression signals of probe sets were calculated using GCRMA (GC robust multi-array analysis, as implemented in GeneSpring). The log of ratio mode was used for all analyses (GeneSpring). The data have been submitted to the GEO at NCBI under the accession number GSE9682. Statistical analysis was performed by one-way analysis of variance (ANOVA) with a Benjamini and Hochberg false discovery rate (BH-FDR = 0.1) multiple testing correction followed by Tukey post-hoc tests using log-transformed data (GeneSpring). Three-thousand-and-thirty-one probe sets were identified to be regulated by fasting in control RNAi-treated worms. We defined RHEB-1-dependent or TOR-dependent genes as those in which the expression level induced by fasting was reduced to less than half under rheb-1 RNAi or TOR (let-363) RNAi conditions.

Accession codes

Primary accessions

Gene Expression Omnibus

Data deposits

Microarray data have been deposited with Gene Expression Omnibus at NCBI under the accession number GSE9682.


  1. 1

    Bordone, L. & Guarente, L. Calorie restriction, SIRT1 and metabolism: understanding longevity. Nature Rev. Mol. Cell Biol. 6, 298–305 (2005)

    CAS  Article  Google Scholar 

  2. 2

    Anson, R. M., Jones, B. & de Cabod, R. The diet restriction paradigm: a brief review of the effects of every-other-day feeding. Age (Omaha) 27, 17–25 (2005)

    CAS  Article  Google Scholar 

  3. 3

    Lakowski, B. & Hekimi, S. The genetics of caloric restriction in Caenorhabditis elegans . Proc. Natl Acad. Sci. USA 95, 13091–13096 (1998)

    ADS  CAS  Article  Google Scholar 

  4. 4

    Bishop, N. A. & Guarente, L. Two neurons mediate diet-restriction-induced longevity in C. elegans . Nature 447, 545–549 (2007)

    ADS  CAS  Article  Google Scholar 

  5. 5

    Panowski, S. H., Wolff, S., Aguilaniu, H., Durieux, J. & Dillin, A. PHA-4/Foxa mediates diet-restriction-induced longevity of C. elegans . Nature 447, 550–555 (2007)

    ADS  CAS  Article  Google Scholar 

  6. 6

    Kaeberlein, M. et al. Regulation of yeast replicative life span by TOR and Sch9 in response to nutrients. Science 310, 1193–1196 (2005)

    ADS  CAS  Article  Google Scholar 

  7. 7

    Hansen, M. et al. Lifespan extension by conditions that inhibit translation in Caenorhabditis elegans . Aging Cell 6, 95–110 (2007)

    CAS  Article  Google Scholar 

  8. 8

    Kapahi, P. et al. Regulation of lifespan in Drosophila by modulation of genes in the TOR signaling pathway. Curr. Biol. 14, 885–890 (2004)

    CAS  Article  Google Scholar 

  9. 9

    Saucedo, L. J. et al. Rheb promotes cell growth as a component of the insulin/TOR signalling network. Nature Cell Biol. 5, 566–571 (2003)

    CAS  Article  Google Scholar 

  10. 10

    Long, X. et al. TOR deficiency in C. elegans causes developmental arrest and intestinal atrophy by inhibition of mRNA translation. Curr. Biol. 12, 1448–1461 (2002)

    CAS  Article  Google Scholar 

  11. 11

    Smith, E. D. et al. Age- and calorie-independent life span extension from dietary restriction by bacterial deprivation in Caenorhabditis elegans . BMC Dev. Biol. 8, 49 (2008)

    Article  Google Scholar 

  12. 12

    Pan, K. Z. et al. Inhibition of mRNA translation extends lifespan in Caenorhabditis elegans . Aging Cell 6, 111–119 (2007)

    CAS  Article  Google Scholar 

  13. 13

    Greer, E. L. et al. An AMPK-FOXO pathway mediates longevity induced by a novel method of dietary restriction in C. elegans . Curr. Biol. 17, 1646–1656 (2007)

    CAS  Article  Google Scholar 

  14. 14

    Guarente, L. & Kenyon, C. Genetic pathways that regulate ageing in model organisms. Nature 408, 255–262 (2000)

    ADS  CAS  Article  Google Scholar 

  15. 15

    Derry, W. B., Putzke, A. P. & Rothman, J. H. Caenorhabditis elegans p53: role in apoptosis, meiosis, and stress resistance. Science 294, 591–595 (2001)

    ADS  CAS  Article  Google Scholar 

  16. 16

    Hinnebusch, A. G. Translational regulation of GCN4 and the general amino acid control of yeast. Annu. Rev. Microbiol. 59, 407–450 (2005)

    CAS  Article  Google Scholar 

  17. 17

    Henderson, S. T. & Johnson, T. E. daf-16 integrates developmental and environmental inputs to mediate aging in the nematode Caenorhabditis elegans . Curr. Biol. 11, 1975–1980 (2001)

    CAS  Article  Google Scholar 

  18. 18

    Murphy, C. T. et al. Genes that act downstream of DAF-16 to influence the lifespan of Caenorhabditis elegans . Nature 424, 277–283 (2003)

    ADS  CAS  Article  Google Scholar 

  19. 19

    Hansen, M., Hsu, A. L., Dillin, A. & Kenyon, C. New genes tied to endocrine, metabolic, and dietary regulation of lifespan from a Caenorhabditis elegans genomic RNAi screen. PLoS Genet. 1, e17 (2005)

    Article  Google Scholar 

  20. 20

    Hsu, A. L., Murphy, C. T. & Kenyon, C. Regulation of aging and age-related disease by DAF-16 and heat-shock factor. Science 300, 1142–1145 (2003)

    ADS  CAS  Article  Google Scholar 

  21. 21

    Rajasekaran, N. S. et al. Human αB-crystallin mutation causes oxido-reductive stress and protein aggregation cardiomyopathy in mice. Cell 130, 427–439 (2007)

    CAS  Article  Google Scholar 

  22. 22

    Murphy, C. T., Lee, S. J. & Kenyon, C. Tissue entrainment by feedback regulation of insulin gene expression in the endoderm of Caenorhabditis elegans . Proc. Natl Acad. Sci. USA 104, 19046–19050 (2007)

    ADS  CAS  Article  Google Scholar 

  23. 23

    Malone, E. A., Inoue, T. & Thomas, J. H. Genetic analysis of the roles of daf-28 and age-1 in regulating Caenorhabditis elegans dauer formation. Genetics 143, 1193–1205 (1996)

    CAS  PubMed  PubMed Central  Google Scholar 

  24. 24

    Luong, N. et al. Activated FOXO-mediated insulin resistance is blocked by reduction of TOR activity. Cell Metab. 4, 133–142 (2006)

    CAS  Article  Google Scholar 

  25. 25

    Lee, G. D. et al. Dietary deprivation extends lifespan in Caenorhabditis elegans . Aging Cell 5, 515–524 (2006)

    CAS  Article  Google Scholar 

  26. 26

    Kaeberlein, T. L. et al. Lifespan extension in Caenorhabditis elegans by complete removal of food. Aging Cell 5, 487–494 (2006)

    CAS  Article  Google Scholar 

  27. 27

    Weinkove, D., Halstead, J. R., Gems, D. & Divecha, N. Long-term starvation and ageing induce AGE-1/PI 3-kinase-dependent translocation of DAF-16/FOXO to the cytoplasm. BMC Biol. 4, 1 (2006)

    Article  Google Scholar 

  28. 28

    Hansen, M. et al. A role for autophagy in the extension of lifespan by dietary restriction in C. elegans . PLoS Genet. 4, e24 (2008)

    Article  Google Scholar 

  29. 29

    Steinkraus, K. A. et al. Dietary restriction suppresses proteotoxicity and enhances longevity by an hsf-1-dependent mechanism in Caenorhabditis elegans . Aging Cell 7, 394–404 (2008)

    CAS  Article  Google Scholar 

  30. 30

    Fabrizio, P. et al. Sir2 blocks extreme life-span extension. Cell 123, 655–667 (2005)

    CAS  Article  Google Scholar 

  31. 31

    Brenner, S. The genetics of Caenorhabditis elegans . Genetics 77, 71–94 (1974)

    CAS  PubMed  PubMed Central  Google Scholar 

  32. 32

    Kamath, R. S., Martinez-Campos, M., Zipperlen, P., Fraser, A. G. & Ahringer, J. Effectiveness of specific RNA-mediated interference through ingested double-stranded RNA in Caenorhabditis elegans . Genome Biol. 2, research0002.1–research0002.10 (2001)

    Google Scholar 

  33. 33

    Maeda, I., Kohara, Y., Yamamoto, M. & Sugimoto, A. Large-scale analysis of gene function in Caenorhabditis elegans by high-throughput RNAi. Curr. Biol. 11, 171–176 (2001)

    CAS  Article  Google Scholar 

Download references


We thank members of our laboratory for technical comments and helpful discussion. This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to E.N.). Some nematode strains used in this work were provided by the Caenorhabditis Genetics Center, which is funded by the NIH National Center for Research Resources (NCRR).

Author Contributions S.H. conceived the study, designed and performed the experiments, and wrote the manuscript with the help of E.N.; S.H. and T.Y. analysed the microarray data; M.U. conducted DAF-16::GFP localization experiments; E.N. supervised the project. All authors discussed the results and commented on the manuscript.

Author information



Corresponding author

Correspondence to Eisuke Nishida.

Supplementary information

Supplementary Information

This file contains Supplementary Figures1-11 with Legends and Supplementary Tables 1-2. (PDF 2910 kb)

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Honjoh, S., Yamamoto, T., Uno, M. et al. Signalling through RHEB-1 mediates intermittent fasting-induced longevity in C. elegans. Nature 457, 726–730 (2009).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing