Healthy Animals with Extreme Longevity

See allHide authors and affiliations

Science  24 Oct 2003:
Vol. 302, Issue 5645, pp. 611
DOI: 10.1126/science.1089169

This article has a correction. Please see:

In the nematode Caenorhabditis elegans, mutations that inhibit insulin/IGF-1 (insulin-like growth factor 1) signaling, such as daf-2 insulin/IGF-1 receptor mutations, can double the life-span of the animal (1). Removing the germ-line precursor cells also extends life-span by approximately 60% (2). This life-span extension is not a result of sterility; it appears to be due to altered endocrine signaling (1, 2). Removal of the germ line or the entire reproductive system of daf-2 mutants can further extend life-span: these animals can live four times as long as normal (2).

Strong reduction of daf-2 activity causes juvenile animals to enter a quiescent state of diapause called dauer. In contrast, partial loss-of-function daf-2 mutants grow to adulthood and have long life-spans. We found that reducing daf-2 activity further by subjecting weak daf-2 mutants, such as daf-2(e1368), as young larvae to dal-2 RNA interference (RNAi) produced larger increases in life-span without triggering dauer formation (Fig. 1). Moreover, when we removed the reproductive systems of these RNAi-treated animals, they lived six times as long as normal. Whereas the mean life-span of wild type was 20 days, these animals had mean life-spans of 124 days (Fig. 1). In fact, only 15% of the animals died in the first 3 months.

Fig. 1.

Life-spans of experimental animals. N2 (wild type) on control bacteria (black), m = 20.7 ± 0.9 days, n = 56/49 [observed/uncensored events (10)]; daf-2(e1368) on control bacteria (blue), m = 32.4 ± 1.2, n = 80/48, P < 0.0001; daf-2(e1368) on bacteria expressing daf-2 dsRNA (green), m = 51.0 ± 1.9, n = 80/68, P < 0.0001; gonad-ablated daf-2(e1368) on bacteria expressing daf-2 dsRNA (red), m = 124.1 ± 5.9, n = 46/39, P < 0.0001. Like gonad-ablated daf-2(e1368,RNAi) animals, N2, e1368, e1368(RNAi) and gonad-ablated e1368 animals were healthy and active. We repeated this experiment once, initiating RNAi at the L2 stage. Again, the gonad-ablated daf-2(e1368,RNAi) mutants were very healthy and long-lived (m = 113 days). As observed previously (2), gonad ablation did not further extend the life-span of daf-2(e1368) mutants. [In wild type and daf-2 mutants, ablation of the germline extends life-span. In wild type but not daf-2(e1370) animals, ablation of the somatic gonad as well prevents this life-span extension. Thus, somatic gonad ablation may shorten life-span by increasing daf-2 activity. It seems likely that e1370 and e1368 mutants differ in their response to somatic gonad ablation because of having different levels of residual daf-2 activity (only e1368 mutants have sufficient daf-2 activity to shorten life-span after somatic gonad ablation). This would explain why whole gonad (germline plus somatic gonad) ablation can extend the life-span of daf-2(e1368,RNAi) but not daf-2(e1368) animals.]

We were particularly interested in these animals' “quality of life,” which can be addressed by examining their level of activity. Long-lived daf-2 mutants fall into two classes. Class 1 mutants resemble normal animals, whereas class 2 mutants are uncoordinated and adopt a dauerlike posture (3). Some class 2 mutations, either singly or in combination with other life-span mutations, cause very long mean life-span extensions (up to 5.5-fold); however, all of these strains have been extremely lethargic (48). This has suggested that extreme longevity may require a metabolic tradeoff.

Consistent with this, when daf-2(e1370) mutants [which behave as class 2 mutants at 25°C (3)] were subjected to gonad ablation and daf-2 RNAi, they had very long life-spans (m = 100 days) but were lethargic and dauer-like. However, when daf-2(e1368) animals [class 1 mutants (3)] were treated in the same way, not only did they have longer mean life-spans, they were quite active for most of their lives (9). For example at 120 days, when approximately half of the animals had died, two-thirds of those remaining were still moving about on their plates without any prodding by the observer. Two of these animals are shown (Movie S1) at day 144. In human terms, these animals would correspond to healthy, active 500 year olds. This indicates that extreme longevity can be uncoupled from quiescence.

The life-spans of animals in nature range from a few weeks to more than a century. Assuming that the ancestor of modern-day metazoans had a short life-span, changes in genes during evolution have extended life-span more than 1000-fold. Though far short of this achievement, these findings in C. elegans show that remarkable life-span extensions can be produced with no apparent loss of health or vitality by perturbing a small number of genes and tissues in an animal. These life-span extensions, which are the longest mean life-span extensions ever produced in any organism, are particularly intriguing because the insulin/IGF-1 pathway controls longevity in many species, including mammals (1).

Supporting Online Material

Materials and Methods

Movie S1


References and Notes

Stay Connected to Science

Navigate This Article