The Evolutionary Advantage of Aging: Ensuring an Optimal Rate of Evolution

It is commonly believed that aging is an accident of evolution, something that natural selection allowed because old animals, it is said, have no effect on fitness (Budovskaya, Y.V., Wu, K., Southworth, L.K., Jiang, M., Tedesco, P., Johnson, T.E., and Kim, S.K. (2008). An elt-3/elt-5/elt-6 GATA transcription circuit guides aging in C. elegans. Cell 134, 291-303.).

Yet there is significant and growing evidence in species as varied as yeast, nematodes and fruit flies that aging is no accident, but a carefully orchestrated genetic program (Kenyon, C., Chang, J., Gensch, E., Rudner, A., and Tabtiang, R. (1993)). Humans with drastically reduced aging have been reported, and the trait runs in families, suggesting a genetic origin. A C. elegans mutant that lives twice as long as wild type. Nature 366, 461-464; Apfeld, J., and Kenyon, C. (1999). Regulation of lifespan by sensory perception in Caenorhabditis elegans. Nature 402, 804-809; Guarente, L., and Kenyon, C. (2000). Genetic pathways that regulate ageing in model organisms. Nature 408, 255-262). Moreover, single point mutations can double lifespan, suggesting that even the smallest advantage of a long lifespan would have long ago been found by natural selection. Given that living longer generally allows an individual to produce more offspring and to thus for its genes to expand across generations, such aging genes should have long disappeared as individuals without the gene live longer and produce more offspring --unless the aging genes have an advantage that exceeds the handicap they entail.

Furthermore, aging involves no random sequence of events, as one might expect from both of the leading theories of aging, genetic drifting or environmental damage. On the contrary, aging follows a well-defined sequence of phenotypic events, which in humans includes hair loss, change of hair color, changes in the skin, and many other stereotyped changes.

Thus, the evidence suggests that animals genomes have evolved a genetic program to age. It appears that natural selection actively favors organisms that age, rather that simply tolerate them.

Here, I put forth a theory for why natural selection favors genomes that age, and thus an explanation for why animals exhibit complex genetic programs to age.

Two of the fundamental requirements of natural selection are heredity and variation. Heredity is required so that an organism selected for its competitive advantage will pass on this advantage to its offspring --otherwise the progress made by natural selection could not accumulate, and would be lost from one generation to the next. Variation is required to provide a pool from which the most fit can be chosen. From variation springs change --the essence of evolution itself.

Yet heredity, the tenet that offspring must be similar to their parents, and variation, that which dictates that they must be different from them, hang in a delicate balance. Make the variation too much, and heredity is lost, with the progress of each generation lost, preventing the accumulation of adaptations that is the hallmark of evolution. Make the variation too little, and evolution slows down to a halt. Thus, although a certain number of mutations between generations is necessary for natural selection, high mutation rates are counterproductive, because they uncouple the phenotype preferred by natural selection, operating on the parent, from the genetic composition of the offspring. Thus, evolution has itself evolved to an optimal balance of the two, carefully selected for an optimal rate of change (Nilsson and Snow, Bulletin of Mathematical Biology (2002) 64, 1033–1043. Optimal Mutation Rates in Dynamic Environments, and references therein).

And yet this genetic generational gap is age-dependent, for the germline suffers mutations throughout the life of the parent which increase the generational gap as the parent ages. Allow parents to age too much, and the generational gap, or evolutionary rate, is no longer optimal.

This is the role of aging and menopause: to preserve the optimality of the rate of evolution.

In species where these late offspring, whose genomes are too different from their parents' to make for an optimal representative of the same in their generation, would compete with their older brothers and sisters and thus decrease their brothers and sisters' survival probability, remaining fertile into old age becomes a handicap, one that decreases the probability of survival of the genome. Natural selection is the struggle between similar but different genomes for the survival of the fittest. The late offspring of old parents would be different enough from their parents so as to count against the survival of the parents' genome in that struggle, rather than for it. And so the fittest genomes are those whose carriers become infertile, unattractive or dead with the passage of time --something we call aging. Once an animal becomes unable to reproduce further, selective pressure to keep it alive falls off, and degenerative disease piles on the genetic program of aging.

In contrast, in species where siblings born over different years do not compete with each other, or where they compete with their siblings' competitors as much as with their siblings themselves, there should be been no selection for an aging program.

It is worth noting that there may be two factors at work in the selective advantage of aging: one is that the fitness of the offspring of old parents might be lower and yet not be apparent until late in life, taking precious resources from their healthy siblings until natural selection extracts its toll. This first factor will only be present in species where the life cycle is complex enough that survival as a newborn is not a good predictor of survival into adulthood; in others, natural selection among the progeny should be enough to weed out unfit offspring of the old even without aging unless pregnancy carries an unusually high cost for the whole family. The second is that the offspring of old parents may be perfectly fit but yet their genetic distance from their parents increases with the parents' age, making for negative selection for genes that cause its bearers to live to old age, as their offspring will be more different than average from their parents and thus less likely to carry non-mutated versions of those genes.

An interesting test of this theory resides in the difference in menopause and aging and menopause rates between males and females. It is well known that human males can continue to produce offspring well beyond the age of menopause in females. At first blush this may seem to run contrary to the theory, for mammalian mutation rates at any given locus in males can be more than four-fold their female counterpart (McConkey: Human Genetics), something which is assumed to derive from the fact that male gametes are the product of many more cell divisions than their female counterparts: Vogel and Motulskey (1986) estimated a 25-year-old man's sperm is the product of 300 cell divisions, compared to the roughly 24 between zygote and primary oocyte. But the fact that the average mammal, born from young parents, has more mutations in males than females says nothing about the dependence of this mutation rate on age in each gender. In fact, it is well known that the risk of mutations in a fetus increases greatly with its mother's age. Furthermore, the fact that menopause occurs in females and not males is to be expected from this theory based on the fact that females remain with their progeny more than males typically do, and thus the competition faced by early offspring from late siblings too genetically dissimilar to their parents and siblings is far greater for a female's offspring than for a male's.

I cannot speak for taxes, but as for aging and death, it seems they are not only inevitable, but good. For your genes, that is. Indeed, it seems that these two inevitable facts of life are intertwined: we age and die because children of old parents see their younger siblings as taxing their inheritance.

Alex Backer, Altadena, October 28, 2008, based on ideas I originally published in Sandia National Labs internal reports in 2004.

P.S. One prediction of the theory is that species where progeny get dispersed and do not compete with their siblings, such as some species of fish, should show less or no aging. This is indeed the case.

See also this TEDx presentation on the subject: