Longevity vs. evolutionary fitness

Over the past decade, biogerontologists have demonstrated that several types of mutations can dramatically extend the lifespan of experimental model organisms. Not only do the mutants live longer, they appear to age more slowly, maintaining vigor and a youthful appearance long after their wildtype counterparts have started to die.

Why, then, are these artificially created mutant alleles not the wildtype? In other words, if lacking a particular gene makes an animal longer-lived and more vigorous, why do all wildtype members of the species have the gene in the first place?

Chen et al. address this question in the grandmother (“grand-hermaphrodite”?) of aging models, the nematode worm C. elegans. They quantitatively measured fitness in wildtype worms and two mutants that extend lifespan in different ways — daf-2, an IGF-1 receptor, and clk-1, an enzyme involved in ubiquinone biosynthesis — and found that both longer-lived mutants are significantly less fit, in an evolutionary sense, than the wildtype.

A Demographic Analysis of the Fitness Cost of Extended Longevity in Caenorhabditis elegans

We monitored survival and reproduction of 1000 individuals of Caenorhabditis elegans wild type (N2) and 800 individuals of clk-1 and daf-2, and used biodemographic analysis to address fitness as the integrative consequence of the entire age-specific schedules of survival and reproduction. Relative to N2, the mutants clk-1 and daf-2 extended average life span by 27% and 111%, respectively, but reduced net reproductive rate by 44% and 18%. The net result of differences in survival and fertility was a significant differential in fitness, with both clk-1 (lambda = 2.74) and daf-2 (lambda = 3.78) at a disadvantage relative to N2 (lambda = 3.85). Demographic life table response experiment (LTRE) analysis revealed that the fitness differentials were due to negative effects in mutants on reproduction in the first 6-7 days of life. Fitness costs in clk-1 and daf-2 of C. elegans are consistent with the theory of antagonistic pleiotropy for the evolution of senescence.

Note that the fitness decrease isn’t caused by the overall decrease in lifetime fertility, but rather the delay in early-life reproduction. This makes sense, in the relentless logic of exponential growth: The earlier one reproduces, the earlier one’s progeny are available to do reproduction of their own. As a thought experiment, consider what would happen if one variant of a species reproduces once on day 5 of life but dies on day 6, whereas another variant reproduces on days 6 and 12 and dies on day 13. It’s pretty clear that eventually the short-lived early-reproducer would out-compete the long-lived late-reproducer, even though (in this imaginary example) the total lifetime fertility of the longer-lived variant is actually higher.

This is the idea of “antagonistic pleiotropy” referred to in the last sentence of the abstract: Traits that shorten the overall lifespan can be positively selected for if they increase the reproductive success (fitness) of the organism. Examples of this phenomenon in humans are the prostate and breast, in which high proliferative capacity is good for fitness (reproduction and child-rearing, respectively), but pose a cancer risk in the post-reproductive stage of life.


  1. […] life spans of lower animals; why didn’t evolution lead to those mutations in the first place? From Ouroboros: “Why, then, are these artificially created mutant alleles not the wildtype? In other words, if […]

  2. The reasoning is not complete. There must also be advantages that are linked to later reproduction and a longer lifespan, because there are many species that reproduce late.

    Of course there are many such advantages thinkable. There might be an optimal reproduction period in the day-night cycle or in the cycle of the seasons. It might be an advantage to start reproducing late, but reproduce often or big (invest in body mass first to avoid being predated, then have a long life with much opportunity for reproduction).

    In other words, experiments in lab conditions as described here, do not answer the real questions, but just confirm that a simple mathematic model works in a test tube.

  3. Good point. The argument of the paper as I see it, however, is that for a given body plan and ecological niche, early reproduction can be positively selected at the expense of lifespan — and that the worm provides some evidence of this.

    The authors are not arguing that it’s impossible for any organism to evolve with a longer lifespan — indeed, there are many reasons why humans live longer than worms. Rather, they’re arguing that the slowed aging in these mutants comes at a cost, and that’s why these alleles aren’t wildtype. So there’s not necessarily a fitness benefit to living longer, if it comes at the expense of reproduction.

    It’s also worth mentioning that this is not the first paper to address the fitness cost of longevity in C. elegans: see the work of Lithgow and co-workers, here. I feel like I also remember a study in which the worms were exposed to more natural-conditions (i.e., soil) but I can’t find the reference. (If anyone knows the paper I’m talking about, please comment.)

  4. All:

    Thou hast no right but to do thy Will.

    First, the study showing that daf-2 mutants “that live twice as long as wild-type worms in laboratory conditions typically die sooner than wild-type worms in a natural soil” is (1) below.

    Second, to make explicit a point that’s already implicit in Chris’ in a way that I think will better takaita’s concern: it’s absolutely true that long life and late reproduction can be a good way to achieve fitness — but only if something else doesn’t get you first. If you’re a mouse carrying genes that invest a lot of your limited resources into maintaining a body (including the reproductive system) in good shape for 3 years, that may very well pay off in fitness terms, even if it comes at the cost of a slightly lower reproductive rate because of the diversion of those resources away from quick and easy fertility — but only if you are actually able to stay alive significantly longer than your competition. But if you’re actually very unlikely to get that extra year, because (like the average mouse) you die before your first birthday due to exposure or predation, you lose. Those resources would be better invested in quick fertility — or even in warmer fur, faster metabolism, or even growing nasty spines.

    Studies like (1) and the Lithgow study Chris cited show that, once you take these slow-aging animals out of the sheltered conditions of most lab experiments — where there are no predators, your food supply is steady and reliable, and the culture medium is nearly sterile — they wind up either dying faster from starvation or other causes, or falling prey to the same causes as their wild-type cohorts at about the same time, but without having left as many young’uns behind.

    In other words, genes that provide a longer intrinsic lifespan are only selected in organisms that first either evolve traits, or are introduced to conditions, that allow for a longer extrinsic lifespan.

    Love is the law, love under Will.

    1. Van Voorhies WA, Fuchs J, Thomas S. The
    longevity of Caenorhabditis elegans in soil
    Biol Lett. 2005 Jun 22;1(2):247-9.
    PMID: 17148178 [PubMed – indexed for MEDLINE]

  5. If aging is considered a disease, wouldn’t age resistance have similar advantages to disease resistance especially when a species is under stress for its survival?

Comments are closed.