Continuing my springtime frolicsome mood, a paper in this week's Nature shows that sex is good for us. Well, not necessarily us as individuals, but as a population. This has actually been a longstanding argument in evolutionary biology—sex is risky, it's hard work, and it is prone to failure. Why not just have women reproduce asexually, and bloom into pregnancy automatically as soon as they hit puberty? That would be much more efficient. Sex also has the problem of breaking up good gene combinations; as you may or may not know, my wife is perfect, but in order to reproduce, she has to water down her flawless genes by combining them with those of a lesser member of the species, me. And then of course, there's the problem of us males. We could instantly double the reproductive capacity of our population if all males were equipped with uteruses and could also bear children. It's a weird, weird system.

So why do we bother with sex? Why aren't we being displaced right now by more fecund asexual populations?

The answer is here in the abstract to the paper:

Why sex evolved and persists is a problem for evolutionary biology, because sex disrupts favourable gene combinations and requires an expenditure of time and energy. Further, in organisms with unequal-sized gametes, the female transmits her genes at only half the rate of an asexual equivalent (the twofold cost of sex). Many modern theories that provide an explanation for the advantage of sex incorporate an idea originally proposed by Weismann more than 100 years ago: sex allows natural selection to proceed more effectively because it increases genetic variation. Here we test this hypothesis, which still lacks robust empirical support, with the use of experiments on yeast populations. Capitalizing on recent advances in the molecular biology of recombination in yeast, we produced by genetic manipulation strains that differed only in their capacity for sexual reproduction. We show that, as predicted by the theory, sex increases the rate of adaptation to a new harsh environment but has no measurable effect on fitness in a new benign environment where there is little selection.

Goddard et al. are working with yeast, and have some powerful tools for cleanly testing the advantages of sex. Yeast will reproduce asexually under good conditions, but when starved, will undergo meiosis and instead reproduce sexually, so they've got the capacity to swing both ways, reproducing either sexually or asexually. The authors genetically engineered a strain of yeast, knocking out two critical genes for meiosis and replacing one with a molecular marker so they can recognize them. That means they now have two identical strains of yeast that differ genetically only at two loci, and that differ physiologically in that one can reproduce sexually, and the other cannot.

Another advantage to yeast is that the parent stock can be frozen and kept indefinitely, so you can put some individuals under selection pressure and allow them to evolve for hundreds or thousands of generations, and still be able to go to the freezer, pull out a sample of their many-times-great-grandparents, and do a comparison of their performance.

So here's the experiment. Take your two strains, one sexual and one asexual, and raise them under different selection pressures. One regime is to throw them into paradise, where there's lots of sugar around and the temperature is pleasant, and there's very little pressure to change. The second regime is to give them a little taste of hell: enough sugar to live but not much surplus, the temperature is raised to something uncomfortable, and a bit of salt is added to make everything unpleasantly briny. Then they're allowed to grow for many generations.

Periodically, samples are taken from each culture, and they are placed together with samples of the ancestral population taken from the freezer, and they are allowed to compete on a dish. What's measured is the ability of each population to outbreed the other; you can assess this by counting the number of colonies each produces relative to the ancestral form.

And here are the results.

The change in natural logarithm of fitness of asexual and sexual populations of yeast in benign and harsh environments. Points show fitness measurements for individual populations with twice log-likelihood error bars (these approximate 95% confidence limits); the error bars for the benign treatment are plotted but are mostly too small to be discriminated. The fitted model for the harsh environment is plotted for asexual (blue) and sexual (red) treatments (parameters: a1 = 0.761, a2(asexual) = -5.287, a2(sexual) = -4.901). Yellow symbols, asexual strains in the benign environment; green, sexual in the benign environment; blue, asexual in the harsh environment; red, sexual in the harsh environment.

This graph is plotting relative fitness (that is, how much better the yeast are at outbreeding their ancestors) against generation. First, look at the green and yellow dots on the straight line at the bottom of the graph; these are the sexual and asexual strains raised in a yeasty Eden. They are on an even footing with their ancestors and produce equal numbers of colonies in the competition assay; they haven't improved over time at all.

Now look at the blue and red dots. These are the populations that were raised in a harsh environment, and clearly demonstrate that that which does not kill you makes you stronger. Both the asexual and sexual forms have become stronger, faster breeders than their distant ancestors.

What you can also see is that the sexual forms, the red dots, were better adapted than their asexual blue peers. Sex led to better adaptation faster. It looks like a small difference, but as the authors explain, small differences over many generations add up.

By contrast, in the harsh environment, relative fitness increased markedly for both the asexual and sexual populations; after 100 generations the intrinsic rate of increase in the asexual populations exceeded the ancestor by 0.3, but the equivalent figure for the sexual populations was 0.4. Applying Fisher's Fundamental Theorem of Natural Selection this indicates that in the first 100 generations of the experiment there must have been genetic variance in fitness, upon which selection could act, of about 0.003 and 0.004 per generation in the asexual and sexual populations, respectively. This is consistent with the expectation from Weismann's hypothesis that the maintenance of sex is associated with increased variance in fitness. The difference in fitness between the sexual and asexual populations after 100 generations, and indeed throughout most of the experiment, is about 0.1. Although this may not seem a very great advantage, the geometric growth process underlying it quickly leads to large differences in cell numbers. Thus in the 25 mitotic generations between episodes of sex in our experiments, a sexual individual could expect to leave about 12-fold (e^0.1x25) as many descendents as an asexual individual. We can also ask whether the rates of evolution in our experiments are unrealistically high. This does not seem to be so, because the 0.3–0.4% variation in fitness we observed is in the lower range of estimates for natural populations of various plants and animals (0.1–30%, with typical values likely to be between 1% and 10%)

Goddard MR, Godfray CJ, Burt A (2005) Sex increases the efficacy of natural selection in experimental yeast populations. Nature 434:636-640.