"What Use is Sex?" is the title of an article by John Maynard Smith. In it, the biologist refers to sexual reproduction as the big unsolved mystery in the history of evolution. Although the article in question was published as far back as 1971, and Smith unfortunately died this year, the problem he gave a name to is still seeking a solution. This does not mean that there are no hypotheses on the evolution of sexuality. However, owing to a lack of concrete experiments, they have been neither verified nor falsified for the main part.

The interplay of genes as the key player?

One of the theories on the evolution of recombination most frequently referred to is based on the phenomenon of negative epistasis. This has to do with the fact that two different locations in the genome act upon eachother, whereby the interaction cannot be simply deduced from the single effects. In the case of negative epistasis, for example, the loss of fitness that results when two detrimental mutations occur is greater than what one would expect to find in the two single mutations. By contrast, positive epistasis means that the effects of two changes is less drastic than one would expect from single effects.

Using mathematical models to discover the evolution of recombination: Sebastian Bonhoeffer, ETH Professor of Theoretical Biology. large

Mathematical models now show that sexual reproduction is advantageous when two faulty mutations combine and cause unexpected levels of damage. The same applies if, together, two advantageous, fitness-promoting mutations produce a proportionally lower gain in fitness. In both cases, one is looking at negative epistasis, because the corresponding mathematical formula carries a minus sign. If one can now show that negative epistasis occurs in living organisms, one has concrete indications as to why sex evolved.

Researching sexual reproduction using HI-viruses

This was precisely the object of Sebastian Bonhoeffer's (1) investigations. Even though the scientist is professor of theoretical biology, he analysed the occurrence of epistatic effects using data from concrete HI-viruses, chosen because this organism possesses a double genome.

With the help of HI-viruses, ETH scientist Sebastian Bonhoeffer was able to refute a theory on the evolution of sexual reproduction.

If a host cell is attacked, as happens regularly, by multiple organisms, the different HIV genomes mix. The pathogen thus represents a simple model of recombination. Another reason for the choice was that enough data exists on HI-viruses. Bonhoeffer obtained the sequence and fitness data of around 10,000 viruses from a Californian company.

Out of this data chaos, it was a question of deducing which forms of epistasis occur in HI-viruses. In his work, published last week in "Science" (2), Bonhoeffer is able to show that on average, in HI-viruses a positive epistasis can be found. Although fitness values vary widely, according to Bonhoeffer these results are strong evidence that epistasis can hardly be considered as the motivating force behind sexual reproduction, because otherwise a negative epistasis would have emerged.

Further refutation needed

So, what else could have given rise to sexuality as a means of reproduction in the history of evolution? Bonhoeffer admits that his HIV results cannot necessarily be directly translated to the sexuality of higher organisms. This is because the duplication of the genome in HI-viruses might simply be a repair mechanism, something that can practically be excluded as the sole reason for the recombination in higher organisms. However, the high rate of combination in HI-viruses does not support the repair function argument.

There is a further reason, however, not to totally rule out the epistasis as the reason for sexual reproduction, yet. It might be that various forms of epistasis occur depending on the environment–something that Bonhoeffer has not yet investigated. Corresponding analyses are already being planned, though, for HI-viruses. Moreover, the ETH scientist is thinking of examining yeast to find out whether he can find negative epistasis in organisms with two inactivated genes, so-called double knockouts. Here, too, talks are underway.

But what if no negative epistasis is found in all these investigations? Then the hypothesis of epistasis as the reason for recombination can be definitely put to rest, thinks Bonhoeffer, not hiding the fact that he would not be surprised by such a result. Which theory he will thus topple is something the scientist leaves open. John Maynard Smith's question on the purpose of sexuality would thus be narrowed down but would still remain open for a while longer.