A test of evolutionary theories of agingKimberly A. Hughes*†, Julie A. Alipaz*‡, Jenny M. Drnevich*, and Rose M. Reynolds**School of Integrative Biology and Program in Ecology and Evolutionary Biology, University of Illinois, Urbana, IL 61801; and‡Department of Organismaland Evolutionary Biology, Harvard University, Cambridge, MA 02138Edited by Margaret G. Kidwell, University of Arizona, Tucson, AZ, and approved September 10, 2002 (received for review May 30, 2002)Senescence is a nearly universal feature of multicellular organisms,and understanding why it occurs is a long-standing problem inbiology. The two leading theories posit that aging is due to(i) pleiotropic genes with beneficial early-life effects but deleteri-ous late-life effects (‘‘antagonistic pleiotropy’’) or (ii) mutationswith purely deleterious late-life effects (‘‘mutation accumula-tion’’). Previous attempts to distinguish these theories have beeninconclusive because of a lack of unambiguous, contrasting pre-dictions. We conducted experiments with Drosophila based onrecent population-genetic models that yield contrasting predic-tions. Genetic variation and inbreeding effects increased dramat-ically with age, as predicted by the mutation theory. This increaseoccurs because genes with deleterious effects with a late age ofonset are unopposed by natural selection. Our findings provide thestrongest support yet for the mutation theory.Senescence is the decline in organismal fitness and perfor-mance with age, and it is a nearly universal feature ofmulticellular organisms (1–5). Two evolutionary models predictthat senescence will evolve because, with few exceptions, theforce of natural selection declines with adult age (6). Boththeories require the existence of genes with age-specific effects,but the kind of age-specific gene action that is required differs(6). According to the antagonistic pleiotropy (AP) theory,pleiotropic alleles that increase survival or reproduction early inlife but decrease survival or reproduction late in life canaccumulate in populations, because the selective advantage ofthe early benefits outweighs the late-life disadvantage. Underthe mutation accumulation (MA) theory, alleles with purelydetrimental effects can also accumulate if those effects areconfined to late life when selection against them is weak. Thusunder both theories, populations harbor alleles that are delete-rious in old but not in young individuals.Ramifications of the two theories are quite different. UnderAP, senescence is due to a ‘‘tradeoff’’ between early- and late-lifefitness, and any genetic or evolutionary change in senescence willbe accompanied by changes in early-life fitness components. Incontrast, the MA theory suggests that senescence is caused, atleast in part, by alleles that are neutral early in life, and thusgenetic or evolutionary changes in senescence need not beaccompanied by any change in early-life fitness. In principle,senescence could be slowed or delayed by artificially selecting onlate-life fitness or by genetic manipulation of late-acting dele-terious alleles, and there would be no cost incurred at earlierages.These two theories have been subjected to several experimen-tal tests, but previous attempts to distinguish between them havebeen inconclusive because of a lack of clear and contrastingpredictions (4, 7–9). Of the two, AP has received the strongestsupport, with studies consistently showing negative genetic cor-relations between early- and late-life fitness, which is a predictiononly of the AP theory (reviewed in refs. 9 and 10). Thus far, theevidence in support of MA is more ambiguous. Most tests haveconcentrated on measuring age-specific additive genetic varia-tion in fitness traits, because an age-related increase was pre-dicted under MA (11). Some studies have reported such anincrease (12, 13), whereas others show an early increase followedby a late-life decline in variance (7, 14, 15). These studies leavethe issue unresolved, because they differed in both experimentaland statistical methodology (8) and subsequent theoretical workshowed that increases in additive variance with age could resultunder the AP as well as the MA theories (7).Recently, new quantitative genetic models have providedseveral predictions that can be used to distinguish the twoprocesses (7). For example, if MA contributes to senescence,fitness components such as reproductive success and survival willexhibit age-related increases in dominance (VD) and homozy-gous genetic variance (VH) in addition to additive variance (VA).Inbreeding depression (ID) will also increase with age. Theincrease occurs because the genetic variances and inbreedingload are proportional to the equilibrium frequencies of delete-rious alleles, and these frequencies will increase with age underMA. In contrast, AP will not lead to age-related increases in VD,VH, or ID, although it can contribute to increases in VAundersome conditions (7).We tested these predictions by measuring age-specific repro-ductive success for 100 different genotypes of Drosophila mela-nogaster, produced from all possible crosses among 10 isogeniclines derived from a single, randomly breeding population. Wechose to measure reproductive success rather than survivalbecause of the complications involved in estimating variancecomponents for survival data (8, 16). The age-specific repro-ductive output of 6,000 flies contributed to these measures, and66,183 offspring of these flies were counted to get age-specificmeasures of genetic variance and ID.MethodsWe first created 25 different isogenic lines from wild-type fliesfrom the Ives (IV) laboratory population of D. melanogaster.This population is inversion-free and laboratory-adapted (13, 17,18), which minimizes biases in genetic variance estimates due tonovel-environment effects (19). Each line was genetically iden-tical for both the second and third chromosome, which accountsfor ⬇80% of the Drosophila genome. We first placed the secondand third chromosome balancer T (2, 3) A1-W (20) on an IVgenetic background to create a balancer stock in which the X andY chromosomes were derived from the IV population. These andall subsequent crosses were conducted so as to avoid hybriddysgenesis (e.g., ref. 13). Crossing females from the balancerstock to a single wild-type male from the IV population and thenbackcrossing a single F1male to balancer females produced F2flies that were identically heterozygous for a