Sci. Aging Knowl. Environ., 22 May 2002
Vol. 2002, Issue 20, p. pe6
[DOI: 10.1126/sageke.2002.20.pe6]


Genes, Culture, and Aging Flies--What the Lab Can and Cannot Tell Us About Natural Genetic Variation for Senescence

Christine C. Spencer , and Daniel E. L. Promislow

The authors are in the Department of Genetics at the University of Georgia, Athens, GA 30602, USA. E-mail: (D.E.L.P.);2002/20/pe6

Key Words: laboratory culture • Drosophila • evolution • longevity genes • senescence • trade-offs

Studies of lab-reared model organisms have contributed enormously to our understanding of the genetics of aging. But recent work with the fruit fly Drosophila melanogaster suggests that such studies might be biased by the use of these conveniently cultured organisms. Here we discuss how laboratory culture might influence results from genetic studies of aging in flies, as well as in other model organisms such as mice and worms.

In laboratory fly culture, adult flies are placed in bottles or vials containing a standard medium, usually made of yeast, agar, sugar, and cornmeal, and allowed to lay eggs for 1 or 2 days. Adults are then discarded, and 2 weeks later, newly emerged adults are transferred to fresh containers to begin the process anew. Under this regimen, selection favors flies with high fecundity immediately after sexual maturity, because no fly older than 5 or 6 days has the opportunity to lay eggs. Because of the cost of producing and laying eggs, high early fecundity can lead to shorter life-spans, as has been shown by Partridge and colleagues (1, 2). As a further consequence of this standard culturing technique, mutations that deleteriously affect only flies older than 5 or 6 days can accumulate without selection ever acting on them. Clark (3) and Promislow and Tatar (4) have suggested that this mutation accumulation could also lead to a decrease in life-span over many generations.

Several recent papers show in detail how genetic changes that affect demographic traits occur in the lab (5, 6) and how these changes might bias our interpretation of studies on aging (5, 7). In these analyses, researchers collected strains of flies from the wild and documented how life history characters changed over many years in the laboratory. Matos et al. (6) compared the age at maturity of a newly wild-caught fly strain over the course of 47 generations with that of flies previously collected from the same population and established in the lab (Fig. 1). Initially, the newly caught flies were significantly older when they reached reproductive maturity than were flies from the established strain. Within 10 generations, however, the age at maturity in the more recently caught flies was reduced to that of the established lab strain. In a separate study, Sgr� and Partridge (5) tested whether lab cultures do indeed select for high early fecundity and shorter life-span. To this end, they examined several fitness traits in wild-caught Drosophila maintained under the following three conditions: (i) The first group of flies was maintained in bottles and cycled to fresh bottles every 2 weeks (as described above), so that only young flies were able to lay eggs, and parents never coexisted with their adult offspring. (ii) The second group of flies was maintained as large populations in cages, where old as well as young flies were able to reproduce. (iii) The third group consisted of long-lived flies that had been recently obtained from the wild. In line with Clark's (3) and Promislow and Tatar's (4) predictions, the proportion of flies exhibiting early fecundity increased dramatically in bottle cultures over time, whereas in cage cultures, the proportion of flies exhibiting early fecundity showed no change over the same time period and remained significantly lower than in the bottle cultures (Fig. 2). In addition, flies recently obtained from the field survived longer than laboratory flies cultured in cages, and flies in cages survived longer than flies cultured in bottles [see figure 1 in (8)]. This result suggests that keeping populations of flies in cages with overlapping generations might be one way to reduce, if not eliminate, the confounding effects of laboratory culture.

View larger version (15K):
[in this window]
[in a new window]
Fig. 1. Lab selection on age at maturity. Data shown here were derived from data in table 1 of Matos et al. (6). The figure illustrates the convergence over time of age at maturity in recently obtained wild flies with that of the same strain previously established in the lab.


View larger version (6K):
[in this window]
[in a new window]
Fig. 2. Lab selection on early fecundity. This figure is based on data in table 11 of Sgr� and Partridge (5). Early fecundity is greater in flies maintained in bottles for 2 years versus 1 year. Flies maintained in cages with overlapping generations do not show a similar increase in early fecundity. For each culturing regimen, early fecundity was measured as the average number of eggs laid per day for adults aged 6 to 8 days.

So what does all this have to do with genetic studies of senescence? Three general approaches have been used to study the genetics of aging: (i) artificial selection, in which genotypes that yield long-lived phenotypes are created by selectively breeding older flies (9), (ii) quantitative genetic analyses that estimate genetic variance components or identify broad chromosomal regions associated with life-span (10-12), and (iii) molecular genetic screens designed to identify single genes that affect aging (13-17). Results from all of these approaches can be confounded by laboratory culturing techniques. This is because in each case, the genes that are identified might have little to do with genes that cause variation in life-span in natural populations.

Whether or not this cultural effect confounds studies of aging is an important question. For example, artificial selection studies have been carried out to determine whether aging has a genetic basis and, in particular, whether the genes that influence aging have antagonistic pleiotropic effects (that is, effects that increase fitness at early ages, but decrease fitness late in life). If lab culture leads to shorter life-span, then artificial selection to increase life-span might simply be restoring genotypes to their original state, which is associated with a life-span that is longer than those observed in lab-cultured organisms.

To test the hypothesis that previous artificial selection experiments have simply restored normal life-span to lab lines, Linnen et al. (7) compared age-specific mortality in a number of different lines. These lines included flies that had been selected for long life-span by Rose et al. (9) ("O" lines), the control lines from which these long-lived strains were derived ("B" lines), a standard lab strain (Canton S), and a recently wild-caught strain from a peach orchard in Georgia.

In line with many previous studies, these researchers found that flies from the O lines lived significantly longer than those from the B lines. This comparison illustrates the power of artificial selection to extend life-span dramatically. But are the O lines unnaturally long-lived? Linnen et al. (7) found that the B lines and Canton S had similar life-spans, as expected. More surprisingly, after 20 years of artificial selection for long life-span, the O lines had the same life-span as did the Georgia line that had been brought in from the wild a few months earlier.

There is no way to determine whether the original stock from which the B lines were derived was long-lived and then evolved a shorter life-span in the lab. Similarly, if this process did occur, we cannot determine now whether it was a result of mutation accumulation in late-acting genes or selection for high fecundity at an early age. In any case, these studies suggest that artificial selection for long life-span could simply reverse the genetic changes that have already occurred in lab culture.

Although the focus of this critical attention has been on artificial selection studies of aging, these results have important implications for findings from molecular genetic studies as well. In the past few years, researchers have identified numerous genes that lead to substantial life extension in Drosophila, including methuselah (MTH) (13), Chico (17), Insulin Receptor (InR) (15), I'm-not-dead-yet (Indy) (16), and human superoxide dismutase (SOD1) (14, 18). We know that mutations at each of these loci (or overexpression, in the case of SOD1 ) lead to a significant increase in longevity in lab strains. However, the studies discussed above suggest that the effects of these modified genes might be limited to, or most dramatic in, the short-lived lab strains in which they are studied. This idea is supported by one study in particular (19), which suggests that Cu/ZnSOD overexpression extends life-span to a greater degree in short-lived organisms.

In the future, researchers need to examine the effects of aging-related genes in strains that have not been subject to long periods of lab adaptation. But why should scientists care whether aging-related genes increase life-span in lab strains but not in wild strains?

It is not sufficient to identify genes that influence life-span, whatever the genetic background. First, in the search for genes that extend life-span, researchers must verify that they have not simply identified a gene that somehow ameliorates the deleterious effects of a genotype that produces an abnormally short life-span. For example, consider a hypothetical isolated human population in which all members carry the allele that gives rise to Huntington's disease (HD). Even with the best medical care, mean life-span in this population would not exceed 45 or 50 years. If a geneticist were somehow to introduce the wild-type HD allele into members of the population, and this gene therapy resulted in an increase in life-span, it would appear to the na�ve observer as though the geneticist had discovered an aging-related gene. Biogerontologists must be sure that they are not mirroring this hypothetical example in their efforts to uncover aging-related genes.

Second, although researchers have successfully identified single-gene mutations that have major effects on survival at all ages in worms, flies, and mice, the alleles that are associated with existing differences in life-span in natural populations (including human populations) are likely to be those with more subtle effects. Genetic variation in human populations is unlikely to be due to alleles of major effect, especially if they influence survival rates at all ages. Natural selection would drive such alleles to fixation if they increased survival or would remove them from the population if they decreased survival. Aging-related genes found in lab studies might identify pathways involved in the aging process (such as insulin signaling). But only by studying natural populations are scientists likely to identify specific allelic variants associated with longevity.

However, wild-caught strains are not without their problems. In a population at genetic equilibrium in the wild, all organisms exhibit about the same overall level of fitness. The antagonistic pleiotropy theory of aging predicts that such fitness levels are obtained through trade-offs, such that individual traits--fertility, fecundity, larval viability, adult survival and so forth--will be negatively correlated. For example, animals with high early fecundity might also have short life-spans. Researchers who study such phenomena using wild-caught strains recently brought into the lab, however, are likely to find that individual fitness traits are positively correlated. As discussed by Service and Rose (20), this positive correlation might occur, not in violation of theory, but because genotypes that evolved in one environment (the wild) are being studied in another (the lab). Among recently wild-caught flies, some genotypes will be better adapted than others to the novel laboratory environment. Until the population has adapted to this environment (that is, reached a genetic equilibrium where gene frequencies are relatively stable), some genotypes will result in high values for all fitness traits, whereas others will result in low values for all of these traits. Thus, among organisms with different genotypes in the lab, fitness traits measured early in life will appear to be positively correlated with fitness traits measured late in life. A second problem is that lab-born offspring of wild-caught females are likely to be affected not only by the genes they inherit but also by the environments (for example, food availability and temperature) that their mothers experienced in the wild. Third, wild-caught flies are limited in the degree to which they can be used for genetic manipulations. Recently derived wild-caught Drosophila all contain P elements. These now common transposable elements were not found in most strains collected before the 1960s. A substantial component of fly genetics is built on our ability to manipulate genes using P-element vectors within strains that are normally free of P elements. Thus, given that P-element-free stocks no longer exist in the wild, scientists are limited in the extent to which they can carry out genetic manipulations in recently derived stocks.

Ultimately, we need to identify genes of major effect as well as genes with more subtle effects that extend longevity in "natural" long-lived strains of flies. No gene is an island. Each gene functions within a complex network of other genes. Thus, one can expect that genes that increase life expectancy in one genetic background might have no effect--or even the opposite effect--in other backgrounds. Scientists are still at a stage where their efforts are primarily directed toward the search for genes with major effects on life-span. In time, it might turn out that modifier genes--those that have no direct effect on life-span but that influence the effect of aging-related genes through epistatic interactions--are responsible for a substantial amount of the existing genetic variation for aging in natural populations.

The problems with the use of fruit fly lab cultures that we have noted here might also pertain to studies on nematodes and mice. Lab-maintained mouse stocks are much shorter-lived and more fecund than are wild-caught strains (21, 22). And although worms can now be frozen, thereby limiting any genetic change in the lab, the most commonly studied strains (for example, Bergerac and Bristol) were collected a half-century ago (23). These strains were probably kept in rapidly cycling cultures for hundreds of generations before they were moved to Sydney Brenner's lab 5 or 6 years later, where they were eventually preserved. Unless the research community chooses to modify decades of lab practice, the problems created by cultural effects in laboratory strains will persist. The time has come to consider broadening our approach to the genetics of aging. In light of the high early fecundity and increased mortality rates caused by laboratory culture, we must turn our attention to genetic variation for longevity in natural populations and modify how we handle lab strains in order to minimize the impact of lab culture on life-span. Such a shift in focus might be crucial if we are to fully understand the genetic basis of aging in longer-lived organisms, such as humans.

May 22, 2002

  1. L. Partridge, A. Green, K. Fowler, Effects of egg-production and of exposure to males on female survival in Drosophila melanogaster. J. Insect Physiol. 33, 745-749 (1987).[CrossRef]
  2. L. Partridge, R. Andrews, The effect of reproductive activity on the longevity of male Drosophila melanogaster is not caused by an acceleration of ageing. J. Insect Physiol. 31, 393-395 (1985).[CrossRef]
  3. A. G. Clark, Senescence and the genetic correlation hang-up. Am. Nat. 129, 932-940 (1987).[CrossRef]
  4. D. E. Promislow, M. Tatar, Mutation and senescence: Where genetics and demography meet. Genetica 102-103, 299-314 (1998).
  5. C. M. Sgr�, L. Partridge, Evolutionary responses of the life history of wild-caught Drosophila melanogaster to two standard methods of laboratory culture. Am. Nat. 156, 341-353 (2000).[CrossRef]
  6. M. Matos, M. R. Rose, M. T. R. Pit�, C. Rego, T. Avelar, Adaptation to the laboratory environment in Drosophila subobscura. J. Evol. Biol. 13, 9-19 (2000).
  7. C. Linnen, M. Tatar, D. E. L. Promislow, Cultural artifacts: A comparison of senescence in natural, lab-adapted and artificially selected lines of Drosophila melanogaster. Evol. Ecol. Res. 3, 877-888 (2001).
  8. C. M. Sgr�, L. Partridge, A delayed wave of death from reproduction in Drosophila. Science 286, 2521-2524 (1999).[Abstract/Free Full Text]
  9. M. Rose, Laboratory evolution of postponed senescence in Drosophila melanogaster. Evolution 38, 1004-1010 (1984).[CrossRef]
  10. M. R. Rose, B. Charlesworth, Genetics of life-history evolution in Drosophila melanogaster. I. Sib analysis of adult females. Genetics 97, 173-186 (1981).[Abstract/Free Full Text]
  11. D. E. L. Promislow, M. Tatar, A. Khazaeli, J. W. Curtsinger, Age-specific patterns of genetic variance in Drosophila melanogaster. I. Mortality. Genetics 143, 839-848 (1996).[Abstract/Free Full Text]
  12. S. V. Nuzhdin, E. G. Pasyukova, C. L. Dilda, Z.-B. Zeng, T. F. C. Mackay, Sex-specific quantitative trait loci affecting longevity in Drosophila melanogaster. Proc. Natl. Acad. Sci. U.S.A. 94, 9734-9739 (1997).[Abstract/Free Full Text]
  13. Y. J. Lin, L. Seroude, S. Benzer, Extended life-span and stress resistance in the Drosophila mutant methuselah. Science 282, 943-946 (1998).[Abstract/Free Full Text]
  14. T. L. Parkes, A. J. Elia, D. Dickinson, A. J. Hilliker, J. P. Phillips, G. L. Boulianne, Extension of Drosophila lifespan by overexpression of human SOD1 in motorneurons. Nature Genet. 19, 171-174 (1998).[CrossRef][Medline]
  15. M. Tatar, A. Kopelman, D. Epstein, M. P. Tu, C. M. Yin, R. S. Garofalo, A mutant Drosophila insulin receptor homolog that extends life-span and impairs neuroendocrine function. Science 292, 107-110 (2001).[Abstract/Free Full Text]
  16. B. Rogina, R. A. Reenan, S. P. Nilsen, S. L. Helfand, Extended life-span conferred by cotransporter gene mutations in Drosophila. Science 290, 2137-2140 (2000).[Abstract/Free Full Text]
  17. D. J. Clancy, D. Gems, L. G. Harshman, S. Oldham, H. Stocker, E. Hafen, S. J. Leevers, L. Partridge, Extension of life-span by loss of CHICO, a Drosophila insulin receptor substrate protein. Science 292, 104-106 (2001).[Abstract/Free Full Text]
  18. W. C. Orr, R. S. Sohal, Extension of life-span by overexpression of superoxide dismutase and catalase in Drosophila melanogaster. Science 263, 1128-1130 (1994).[Abstract/Free Full Text]
  19. J. Sun, J. Tower, FLP recombinase-mediated induction of Cu/Zn-superoxide dismutase transgene expression can extend the life span of adult Drosophila melanogaster flies. Mol. Cell Biol. 19, 216-228 (1999).[Abstract/Free Full Text]
  20. P. M. Service, M. R. Rose, Genetic covariation among life-history components: the effects of novel environments. Evolution 39, 943-945 (1985).[CrossRef]
  21. R. A. Miller, Kleemeier award lecture: Are there genes for aging? J. Gerontol. A Biol. Sci. Med. Sci. 54, B297-B307 (1999).[Medline]
  22. S. N. Austad, Does caloric restriction in the laboratory simply prevent overfeeding and return house mice to their natural level of food intake? Science's SAGE KE (2001),;2001/6/pe3.
  23. E. L. Hansen, E. A. Yarwood, W. L. Nicholas, F. W. Sayre, Differential nutritional requirements for reproduction of two strains of Caenorhabditis elegans in axenic culture. Nematologica 5, 27-31 (1959).

Science of Aging Knowledge Environment. ISSN 1539-6150