Timing Requirements for Insulin/IGF-1 Signaling in C. elegans

See allHide authors and affiliations

Science  25 Oct 2002:
Vol. 298, Issue 5594, pp. 830-834
DOI: 10.1126/science.1074240


The insulin/IGF-1 (where IGF-1 is insulin-like growth factor–1) signaling pathway influences longevity, reproduction, and diapause in many organisms. Because of the fundamental importance of this system in animal physiology, we asked when during the animal's life it is required to regulate these different processes. We find that in Caenorhabditis elegans, the pathway acts during adulthood, to relatively advanced ages, to influence aging. In contrast, it regulates diapause during development. In addition, the pathway controls longevity and reproduction independently of one another. Together our findings show that life-span regulation can be dissociated temporally from phenotypes that might seem to decrease the quality of life.

In C. elegans, mutations that decrease the activity of DAF-2, an insulin/IGF-1-like receptor, or downstream phosphatidylinositol 3-kinase, phosphoinositide-dependent kinase, or AKT (also known as protein kinase B) signaling components prolong youthfulness and double the life-span of the animal. The DAF-2 pathway influences other processes as well. All daf-2 examined mutations increase resistance to oxidative stress and delay reproduction (some alleles also reduce fertility). Strong daf-2 mutations cause juvenile animals to enter a state of diapause, called dauer, instead of growing to adulthood. All of these mutant phenotypes require the activity of DAF-16, a forkhead family transcription factor (1–3).

Mutations in components of the insulin/IGF-1 pathway also extend life-span in flies (4, 5), and mutations that inhibit pituitary development or growth hormone receptor signaling, which in turn decreases IGF-1 signaling, extend the life-span of mice (1, 6). Components of insulin/IGF-1 signaling pathways also influence reproduction, stress resistance, and entry into diapause-like states in a wide range of organisms (1–3).

To investigate when the C. elegans insulin/IGF-1 pathway acts to regulate longevity, diapause, reproduction, and stress resistance, we used RNAi (RNA interference), which decreases mRNA levels (7), to decrease daf-2 anddaf-16 activity at different times during the life cycle. Reverse transcriptase polymerase chain reaction (RT-PCR) analysis confirmed that feeding animals bacteria that are expressing daf-2or daf-16 double-stranded RNA (dsRNA) decreased mRNA levels (Fig. 1) (8). We found that culturing wild-type animals on bacteria expressing daf-2dsRNA from the time of hatching produced Daf-2(–) phenotypes similar to those produced by partial loss-of-function daf-2mutations (9–12). At 20°C the animals grew to adulthood and became long-lived, stress-resistant adults with normal brood sizes but protracted reproductive schedules (Fig. 2A, Table 1, and table S1) (8). High temperature is known to induce some dauer formation in the wild type (13). Our daf-2 RNAi treatment increased the frequency of dauer formation at high temperatures [27°C; 45% dauers (n = 72, wheren is the number of animals) for treated animals versus 10% dauers (n = 81) for control animals] (8). Strong daf-2 mutations delay growth to adulthood and cause uncoordinated movement and some embryonic lethality (9, 14, 15). We did not observe these phenotypes (not shown), suggesting that they result from a reduction ofdaf-2 activity that is greater than the reduction from RNAi treatment. Conversely, feeding daf-2(1370) mutants bacteria expressing daf-16 dsRNA completely suppressed their delayed development (16), longevity, and reproductive phenotypes (Fig. 3A). This daf-2mutation is temperature sensitive. At 25.5°C, the animals become dauers; this phenotype was partially suppressed by daf-16RNAi (17).

Figure 1

daf-2 and daf-16bacterial RNAi lower mRNA levels. (A) RT-PCR analysis ofdaf-2 mRNA after RNAi treatment. Shown are RT-PCR products from serial dilutions of total RNA isolated from control animals grown on bacteria containing vector only (lanes 1 to 5) or on bacteria expressing daf-2 dsRNA (lanes 6 to 10). (a) RNAi was initiated at hatching, and RNA was harvested at L4. (b) RNAi was initiated on day 8 of adulthood and harvested on day 10 of adulthood. (B) RT-PCR analysis of daf-16 mRNA after RNAi treatment. Conditions were the same as in (A), except animals were treated with bacteria expressing daf-16 dsRNA.

Figure 2

(A to M)daf-2 RNAi affects life-span and reproduction at different times. The time at which animals were transferred onto bacteria expressing daf-2 dsRNA is shown in the upper right corner of each panel. Blue lines, the life-spans of animals grown on control bacteria carrying the RNAi vector alone; red lines, life-spans of animals grown on bacteria expressing daf-2 dsRNA. Reproductive profiles of animals in the concurrent life-span assays are depicted to the right of each life-span profile. The percent of total progeny produced at each 12-hour time interval is shown. Scales are identical for all graphs. Reproductive profiles of animals treated with bacteria expressing daf-2 dsRNA after day 4 of adulthood are not shown because these animals were postreproductive. For statistics, see table S1.

Figure 3

(A to S) daf-16RNAi affects the life-span and reproduction of daf-2(e1370)mutants at different times during the life cycle. The time thatdaf-2(e1370) animals were transferred onto daf-16RNAi bacteria is shown in the upper right corner of each panel. Blue, life-spans of daf-2(e1370) animals grown on bacteria carrying the RNAi vector alone; red, life-spans ofdaf-2(e1370) animals grown on daf-16 RNAi bacteria. Scales of graphs (A) to (N) are identical and (O) to (S) are identical. Insets: T 0 is set to the day at which RNAi treatment was initiated in the experimental population. For statistics, see table S1. Reproductive profiles of animals in the concurrent life-span assays are depicted to the right of each life-span profile. The percent of total progeny produced at each 12-hour time interval is shown. Blue, daf-2(e1370) mutant animals grown on control bacteria; red, daf-2(e1370) mutant animals grown on bacteria expressing daf-16 dsRNA. Reproductive profiles of animals cultured on daf-16 RNAi bacteria after day 5 of adulthood are not shown because the animals were postreproductive.

Table 1

. Effect of daf-2 function on stress resistance. The results of two independent trials are shown. Trial 1 is the first row of numbers (n ≥ 25); trial 2 is the second row of numbers (n ≥ 40).

View this table:

To study when daf-2 influences adult longevity, we shifted wild-type animals to bacteria expressing daf-2 dsRNA at different ages and determined their life-spans. We found that initiating daf-2 RNAi treatment in young adults extended life-span to the same extent as did initiating RNAi at hatching (Fig. 2, A versus G; P = 0.38; table S1). Thus, daf-2is required during adulthood to regulate adult life-span.

To determine whether daf-2 could also function during development to influence adult life-span, we lowered insulin/IGF-1 signaling specifically during development. First, we initiateddaf-2 RNAi during development and then attempted to turn off the RNAi process during adulthood (8). We reasoned that if animals were exposed to the dsRNA of a gene required for RNAi to function, such as dcr-1 (dicer) (18), then RNAi activity should remain low. To this end, we shifted animals exposed todaf-2 RNAi at hatching onto bacteria expressingdcr-1 dsRNA when they reached adulthood. These animals did not have long life-spans (Fig. 4 and table S1).

Figure 4

Loss of daf-2 function during development does not increase life-span. Wild-type animals were grown on bacteria expressing daf-2 dsRNA from hatching until the first day of adulthood and then transferred to bacteria expressing dsRNA of dcr-1. Red line, life-spans of wild-type animals grown on daf-2 RNAi bacteria during development and then shifted during day 1 of adulthood to bacteria expressingdcr-1 dsRNA. Blue line, life-span of wild-type animals grown on the control RNAi bacteria during development and then shifted during day 1 of adulthood to dcr-1 RNAi bacteria. Black line, life-span of wild-type animals grown on daf-2 RNAi bacteria during development and adulthood. Life-span studies were conducted at 25°C. For statistics, see table S1.

We also turned off the effects of a daf-2 mutation at different stages by shifting daf-2(e1370) mutants to bacteria expressing daf-16 dsRNA. We found that initiatingdaf-16 RNAi treatment during young adulthood completely suppressed the long life-spans of daf-2(e1370) mutants (Fig. 3, A versus G; P = 0.19; and table S1). Thus, the presence of DAF-16 activity during development was irrelevant. Therefore, the insulin/IGF-1 pathway appears to operate exclusively during adulthood to influence adult life-span.

We found that initiating daf-2 RNAi treatment later in adulthood, throughout the reproductive period, also extended life-span (Fig. 2, G to J; table S1). The magnitude of this extension declined steadily, and became insignificant after about 6 days of adulthood (Fig. 2M and table S1), though there was a sharp decline indaf-2 mRNA at this time (Fig. 1A). Treatingdaf-2(e1370) mutants with daf-16 RNAi during the reproductive period also affected life-span (Fig. 3, H to L; table S1). Again, the magnitude of the effect declined with age, although we continued to observe effects on life-span until about day 15 of adulthood (Fig. 3, M to P; table S1). In these experiments, daf-2and daf-16 appeared to function for different durations during adulthood; however, wild-type animals (used in thedaf-2 RNAi experiments) age more quickly than daf-2mutants (used in the daf-16 RNAi experiments). In fact, as assayed by tissue morphology, day 6 wild-type animals are about the same age as day 15 daf-2(e1370) mutants (19). Thus, both genes may function at the same time, throughout the reproductive period, to influence aging. We note that the pathway could conceivably function later as well (if, for example, the rate of protein turnover falls in old animals).

We next investigated when daf-2 and daf-16RNAi could affect the timing of reproduction. We found that treating adults with daf-2 RNAi had no effect on reproduction (Fig. 2, F to J; table S1). Likewise, treating daf-2mutants with daf-16 RNAi during adulthood failed to suppress the daf-2 reproductive phenotype (Fig. 3, G to L; table S1). In contrast, initiating either RNAi treatment at hatching did affect the timing of reproduction (Figs. 2A and 3A). Thus,daf-2 appears to control reproduction and longevity independently of one another.

To determine when daf-2 and daf-16 might function in reproduction, we subjected the animals to RNAi during development. We found that initiating daf-2 RNAi treatment of wild-type animals at any time before the last larval stage, L4, delayed reproduction (Fig. 2, A to D) and that initiating daf-16RNAi treatment of daf-2(e1370) mutants at any time before L4 restored normal reproductive timing (Fig. 3, A to E). Treatment during or after L4 did not affect reproduction (Fig. 2, E to J, and Fig. 3, F to L). These findings show that reproductive timing can be specified independently of the dauer decision (which occurs before L3) (13), and they suggest that the daf-2pathway may function late in development to affect the timing of reproduction. However, we cannot rule out the possibility that the DAF-2 pathway controls reproduction during adulthood, but that initiating RNAi during or after L4 does not lower signaling activity below threshold until after DAF-2 and DAF-16 have completed their roles in reproduction. This seems less likely, because message levels fall sharply within 48 hours of RNAi treatment (Fig. 1), whereas reproduction continues for many days.

The DAF-2 pathway also regulates stress resistance (9,10, 20). Because daf-2 larvae (21) and dauers (13) are stress resistant,daf-2 must regulate stress resistance in the larvae. We found that animals treated with daf-2 dsRNA as adults were resistant to the oxidative-damaging agent paraquat (Table 1). Thus, daf-2 also acts in the adult to regulate stress resistance. Because the DAF-2 pathway regulates life-span during adulthood as well, this finding supports the hypothesis that increased resistance to oxidative stress contributes to longevity (9, 10, 20). Consistent with this, overexpression of the antioxidant superoxide dismutase gene during adulthood can extend the life-span of Drosophila(22, 23), as can administering the antioxidant Euk134 to adult worms (24).

In this study, we have investigated the temporal requirements for DAF-2 and DAF-16 action by using RNAi to reduce gene activity at specific times during the life cycle. As with any conditional expression system, we cannot be certain when any residual daf-2 or daf-16activity not removed by RNAi might function. However, the fact that daf-16 RNAi completely suppressed the strong longevity and reproductive phenotypes of the daf-2(e1370) mutant suggests that the level of residual pathway activity is likely to be minimal, as does the fact that inferences from reciprocal daf-2and daf-16 RNAi experiments were always in agreement.

Our findings suggest that the DAF-2 pathway functions exclusively during adulthood, throughout the reproductive period, to influence adult life-span. This conclusion suggests that the pathway controls downstream gene expression in an ongoing fashion through much of adulthood. Because the dauer is a juvenile form (25), the DAF-2 pathway must act in separate regulatory events to control dauer formation and adult life-span. Previously we proposed thatdaf-2 regulates a longevity process that determines the life-span of adults and that can also be expressed in conjunction with dauer-specific traits to give dauers their long life-spans (12). It will be interesting to learn whether the pathway regulates the same genes at two different life stages to influence the life-spans of adults and dauers.

Our findings indicate that the DAF-2 pathway participates in multiple, independent regulatory events to influence aging, reproduction, and diapause. In this regard, the pathway is similar to many growth factor signaling pathways (such as the epidermal growth factor or transforming growth factor-β pathways), which regulate different aspects of cell growth and differentiation independently of one another. The pleiotropy of insulin/IGF-1 pathway mutations in many organisms, particularly the linkage with diapause-like states, has raised the possibility that longevity achieved through this pathway would invariably be associated with impaired growth or reproduction. Instead, our findings suggest that, in other organisms as well, it may be possible to manipulate insulin/IGF-1 signaling during adulthood so as to extend youthfulness and life-span without affecting growth or reproduction.

Supporting Online Material

Materials and Methods

Table S1

  • * Present address: Molecular and Cell Biology Laboratory, The Salk Institute for Biological Studies, La Jolla, CA 92037, USA.

  • To whom correspondence should be addressed. E-mail: ckenyon{at}


Stay Connected to Science

Navigate This Article