Regulation of Life-Span by Germ-Line Stem Cells in Caenorhabditis elegans

See allHide authors and affiliations

Science  18 Jan 2002:
Vol. 295, Issue 5554, pp. 502-505
DOI: 10.1126/science.1065768


The germ line of the nematode Caenorhabditis elegansinfluences life-span; when the germ-line precursor cells are removed, life-span is increased dramatically. We find that neither sperm, nor oocytes, nor meiotic precursor cells are responsible for this effect. Rather life-span is influenced by the proliferating germ-line stem cells. These cells, as well as a downstream transcriptional regulator, act in the adult to influence aging, indicating that the aging process remains plastic during adulthood. We propose that the germ-line stem cells affect life-span by influencing the production of, or the response to, a steroid hormone that promotes longevity.

Killing the germ-line precursor cells, Z2 and Z3, extends the life-span of C. elegans by ∼60% (1). This longevity is not a result of sterility, because removing the entire reproductive system (germ line and somatic gonad) has no effect on life-span. In order for germ line–ablated animals to live longer than normal, they require DAF-12, a nuclear hormone receptor, and DAF-16, a forkhead-family transcription factor. We found that this effect could be reproduced genetically:mes-1(bn7) mutants, which lack germ cells, were long lived (Fig. 1A), as were glp-1(q158)mutants (Fig. 1B) (2, 3). glp-1 encodes the receptor for a germ-line proliferation signal that is produced by the distal tip cells of the somatic gonad (4–7). In glp-1(q158)mutants, Z2 and Z3 generate only a few germ cells, which then enter meiosis and differentiate as sperm (7). In both mutants, life-span extension was suppressed by a daf-16 null mutation and by ablation of the somatic gonad precursor cells (Fig. 1) (8, 9). Many other mutants with defective germ-line proliferation were also long lived [Web table 1, experiments A and B (10); (9)].

Figure 1

Life-spans of germ-line mutants. In each graph, experimental and control animals were grown in parallel.n, total number of animals observed in each experiment/total number of uncensored animals. m, mean adult life-span (days). (–), ablated. P values refer to experimental and control animals in a single experiment. (A) Life-spans of germ-line(−) mutants. At 20°C, mes-1(bn7) is 50% penetrant; half of the animals have normal germ lines (3). mes-1(bn7) fertile,n = 48/38, m = 15.8; sterile,n = 48/31, m = 29.8; Z1/Z4(−),n = 52/51, m = 17.2, P≤ 0.0001. daf-16(mu86), n = 54/26, m = 14.1; daf-16(mu86);mes-1(bn7) sterile, n = 62/22,m = 14.2, P = 0.90. Note thatdaf-16 mutants are short-lived (25). (B) Life-spans of glp-1(q158) mutants. All of these strains contained the dpy-19(e1259) mutation. dpy-19(e1259) (control),n = 48/26, m = 21.5;dpy-19(e1259) glp-1(q158), intact, n = 95/52, m = 27.4,P ≤ 0.0001; dpy-19(e1259)glp-1(q158), Z2/Z3(–), n = 30/17, m = 25.3, P = 0.04;dpy-19(e1259) glp-1(q158), Z1/Z4(−), n = 35/34, m = 20.7,P = 0.72.

The germ-line precursors are stem cells that divide continuously during development. As development progresses, germ cells located farthest from the distal tip cells enter meiosis and then differentiate into sperm (during the L4 stage) or oocytes (during adulthood) (6), but a pool of proliferating stem cells is maintained well into adulthood (Fig. 2).

Figure 2

Germ-line stem cells regulate life-span in the adult. The panels show life-spans ofglp-1(e2141ts) animals shifted from the permissive temperature (20°C) to nonpermissive temperature (25°C) at different stages. Diagrams represent the germ lines ofglp-1 animals soon after the shift. Yellow circles, germ-line stem cells; violet circles, meiotic cells; green rectangles, oocytes; brown rectangles, sperm; black dots, distal tip cells [adapted from (30)]. (A) Animals cultured continuously at 20°C. Wild type (N2), n = 80/71,m = 19.4; glp-1(e2141ts),n = 60/46, m = 19.9, P= 0.24. (B) Animals cultured at 20°C until day 1 of adulthood, and then shifted to 25°C. N2, n = 80/57,m = 11.9; glp-1(e2141ts),n = 80/79, m = 15.0, P≤ 0.0001. (C) Animals cultured at 20°C until L4 and then shifted to 25°C. N2, n = 80/64, m = 11.6; glp-1(e2141ts), n = 65/64,m = 13.9, P = 0.008. (D) Animals shifted from 20°C to 25°C at L2. N2, n = 78/59, m = 13.3; glp-1(e2141ts),n = 80/80, m = 18.6, P≤ 0.0001. (E) Animals cultured continuously at 25°C. N2,n = 80/70, m = 11.6;glp-1(e2141ts), n = 80/77,m = 17.7, P ≤ 0.0001. This experiment was repeated twice, yielding similar results (9).

We found that neither sperm nor oocytes are required for the germ line to shorten life-span. Previously, fem-3(e1996) mutants, which do not produce sperm and develop as females, were found to have normal life-spans (11). We found that three additional female mutants, fog-1(q180), fog-2(q71), andfog-3(q470) (12–14), also had normal life-spans [Web table 1, experiments C and D (10)]. In daz-1(tj3) mutants, oocyte precursor cells arrest development at meiotic prophase and subsequently undergo apoptosis (15). We found that the life-spans of daz-1mutants were similar to those of the wild type (Table 1, experiment A). In addition, germ-line ablation extended the life-span of males, which do not produce oocytes, to the same extent as for control hermaphrodites (Table 1, experiment B).

Table 1

Life-spans of daz-1(tj3) mutants and germ line–ablated males. The animals described in each experiment (A and B) were grown in parallel (28, 29). P values are relative to the control animals grown in parallel. Life-spans were measured at 20°C.

View this table:

To determine whether germ-line stem cells might influence life-span, we forced these cells to exit mitosis and enter meiosis at different times using a temperature-sensitive allele of glp-1, e2141(16). Surprisingly, we found that inducing this switch by shifting glp-1(ts) mutants to the nonpermissive temperature either during development or in early adulthood (when the animals were already producing progeny) extended life-span (Fig. 2). This demonstrates that germ-line stem cells do influence life-span, and that they can exert their influence in the adult. The life-span increase we observed after shifting L4 larvae or young adults was smaller than we observed after shifting younger animals. Thus, germ-line proliferation in the larva may also influence life-span.

We also asked whether excessive germ-line proliferation might shorten life-span. We tested two mutants in which germ-line stem cells fail to differentiate and instead overproliferate,glp-1(oz112gf) and gld-1(q485) (17,18). Both had short life-spans [Web table 1, experiments E and F (10)], although, in principle, this could also be caused by an independent, deleterious effect of germ-line overproliferation.

Because the germ line acts in the adult to influence life-span, we also investigated the timing of DAF-16 transcription factor function. To do this, we reduced daf-16 activity at specific times using RNA interference (RNAi) (19, 20). We cultured mes-1mutants on bacteria expressing daf-16 double-stranded RNA either throughout life, or only during adulthood, and found that both treatments completely suppressed their longevity (Fig. 3, B and C). Thus, DAF-16 appears to be required specifically in the adult to effect germ- line modulation of life-span.

Figure 3

Time of action of DAF-16 in germ-line regulation of life-span. Panels show the response ofmes-1(bn7) fertile and sterile animals to a shift from control bacteria (vector only) onto bacteria expressingdaf-16 dsRNA. daf-16 RNAi bacteria were grown on plates containing ampicillin, and carried an ampicillin-resistance gene. Bacteria carrying the plasmid vector alone were not ampicillin resistant. The sterile Mes-1 phenotype was assayed using a dissecting microscope. (A) mes-1(bn7) mutants grown on bacteria carrying the plasmid vector. Fertile animals,n = 70/41, m = 12.5; sterile animals,n = 66/51, m = 19.5, P≤ 0.0001. (B) mes-1(bn7) grown ondaf-16 RNAi bacteria. Fertile animals, n = 67/40, m = 12.9; sterile animals, n = 66/42, m = 14.9, P = 0.022. (C) mes-1(bn7) shifted to daf-16 RNAi bacteria at the onset of adulthood. Fertile animals, n= 69/47, m = 12.2; sterile animals, n = 79/51, m = 14.1, P = 0.036.

Many long-lived mutants are resistant to heat and oxidative stress (21–24). We found that this was also true of germ line–ablated animals [Web table 2 (10)]. It is possible that resistance to oxidative damage causes longevity; alternatively, the germ line could influence both stress resistance and longevity independently of one another.

In summary, we have found that the aging process of C. elegans is modulated, during adulthood, by the activity of germ-line stem cells. How might these cells affect aging? One possibility is that germ-line proliferation shortens life-span by increasing energy expenditure, channeling resources that could otherwise be used for maintaining cellular integrity toward growth and reproduction. However, there does not appear to be a simple trade-off between reproduction (or energy expenditure) and aging in this system; for example, animals that lack the entire reproductive system are not long-lived. Another possibility is that the role of germ-line stem cells is simply to produce more germ-line tissue, which then influences life-span regardless of its state of differentiation. However, the gonads of daz-1 mutants have a much smaller mass than those of glp-1 mutants shifted to high temperature as adults, because oocyte precursors die instead of becoming large, mature oocytes (15). Yet daz-1 animals are not long-lived. Therefore the germ-line stem cells may be uniquely capable of influencing life-span. We propose that stem-cell proliferation influences life-span by affecting either the production of, or the response to, a steroid hormone ligand for DAF-12, which, in turn, promotes longevity. In addition, a signal dependent on stem-cell proliferation must regulate the nuclear localization of DAF-16 in somatic nongonadal tissues (25).

Surprisingly, we found that the aging process is subject to modulation by the germ line in the adult. It was particularly striking that DAF-16 acts exclusively in the adult to mediate germ-line modulation of aging. Consistent with this, ablating the germ line at hatching causes DAF-16 to accumulate in nuclei only during adulthood (25). In contrast, when life-span is extended by mutations in thedaf-16–dependent insulin/IGF-1 system, DAF-16 accumulates in nuclei throughout development and into adulthood (25).

In conclusion, our findings demonstrate that germ-line stem cells preside over two fundamental processes in the life cycle of C. elegans: reproduction and aging. These cells initiate the cascade of germ line development, thereby generating the pool of mature gametes, and they also regulate a steroid-dependent system that accelerates aging. By governing both processes, germ-line stem cells may help to coordinate the rate of aging with reproduction. Because killing germ-line precursor cells in Drosophila melanogasteralso extends life-span (26), it is possible that this system is evolutionarily conserved.

Note added in proof: While this paper was in press, Gerischet al. (31) reported that the germ line may regulate the activity of daf-9, which encodes a cytochrome p450 and may produce a steroid ligand for daf-12.

  • * Present address: Exelixis, Inc., South San Francisco, CA 94083, USA.

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


Stay Connected to Science

Navigate This Article