xia@cc.helsinki.fi (10/20/90)
Evolution of Sex: Two-fold Cost? In discussing evolution of sex, one should start from the evolution of diploids from haploids. The selective advantage of diploids over haploids is usually attributed to the covering-up of deleterious mutations. Thus, the formation of diploid AB from haploid A and B is favoured. Diploids are more vulnerable to Muller's ratchet than haploids. Fitness of diploids will decrease with accumulation of mutation load. One potential way for diploids to reduce the mutation load is to break the diploid AB back into haploid A and haploid B and expose the deleterious mutations to natural selection instead of covering them up. But by now haploid A and haploid B from diploid AB have both accumulated heavy mutation load, and they are far worse than the haploid A and haploid B BEFORE the formation of AB. If AB is broken into A and B, then both A and B are likely to perish with no fitness return. The only way to solve the problem is recombination. Recombination results in some A haploids and some B haploids with extremely heavy mutation load, but also some A and B haploids with little mutation load. By exposing these haploids to natural selection, those with heavy mutation load will be eliminated and the remaining haploids will produce many haploid gametes to form diploids. As a consequence, these resulting diploids have little mutation load and therefore high fitness. This is the selective advantage of the so-called "alternation of generation" observed in lower organisms. Alternation of generation cannot evolve without recombination. A huge number of haploids are required for the purpose of producing a few haploids with little mutation load from a diploid with heavy mutation load by recombination. Therefore, sex (recombination) at the beginning must be associated with production of large number of haploids (gametes), many of which are produced to perish with no fitness return. (In this sense, the cost of sex is far more than two-fold.) At first sight, this alternation of generation with recombination (A.G.R.) seems to be a good solution for early organisms, which now can enjoy the benefit of a diploid in covering up deleterious mutations while remain free from the working of Muller's ratchet. But individuals with this A.G.R. is also very susceptible to exploitation by other individuals. Suppose a population of diploid individuals that produce haploids at time T. Mortality is 90% per unit time. At time T+1, only 10% of haploids still survive. At time T+2, only 1% of haploids still survive. These 1% haploids then produce haploid gametes to form diploids. Now one mutation, say D, makes a diploid to delay its production of haploids by one time unit. Instead of producing haploids at time T, it produces haploids at time T+1. By time T+2, it still has 10% of its haploids survive. These haploids, not yet thoroughly purged by natural selection, then produce haploid gametes to form diploids NOT among themselves, but between them and those gametes produced by the "wild type". This selfish mutation D will then spread with the consequence that haploid life stage becomes shorter and shorter. (Thus, those organisms with prolonged haploid stage must guard themselves against this kind of exploitation by selfing. I hope that it should be clear by now that sex is by no means a masterpiece of nature. It is in fact something quite clumsy because, to take the full advantage of being a diploid, i.e., the advantage of covering-up deleterious mutations, the sexual organism should avoid selfing) (To be continued) Xuhua