The origin of sex and the nature of species

Abstract

By sex in eukaryotes, we understand a more-or-less regular succession of meiosis and syngamy. A natural consequence of this is the alternation of haploid and diploid phases in the life cycle. Eukaryotic sex significantly differs from prokaryotic sex in two crucial respects: the cellular mechanisms are quite different, and the transfer of genetic material in prokaryotes is less frequent and more localized (Maynard Smith et al., 1991). However, there seems to be significant continuity in the molecular mechanisms: sex in either case requires recombination enzymes, many of which are active in repair of damaged DNA as well. Thus, it seems plausible that recombinational repair was a preadaptation for sexual recombination. We mention in passing that there is a theory that selection for the recombinational repair of doublestrand DNA damage is responsible for the current maintenance of eukaryotic sex (Bernstein et al., 1981, 1988), but there are severe theoretical as well as factual problems with this theory; we will mention some factual difficulties later. Although an alternation of haploid and diploid phases follows from sex, a clue to the origin problem may lie in the idea that this alternation existed before the evolution of sexual recombination proper. The first hint that this may have been so comes from the classic paper by Cleveland (1947), where he proposed that the haploid-diploid cycle may have started with a spontaneous diploidization by endomitosis: that is, without syngamy. His suggestions were based on original observations on primitive flagellates (hypermastigotes and polymastigotes). Among them, Barbulanympha has a regular endomitosis-meiosis cycle. Margulis & Sagan (1986) called renewed attention to Cleveland’s ideas. In particular, they argued that the alternation of ploidy phases could have a primarily ecological explanation: if the environment alternates in some important factors, this may drive a haploid-diploid cycle, provided the phases are adaptations to different environments. For example, diploids have a smaller relative surface area than haploids, which may confer higher metabolic efficiency. We shall come back to such ideas soon. We focus first on the possible cellular mechanisms connecting the two phases. It is important that some protists have a one-step rather than a two-step meiosis: after syngamy, the two homologous chromosomes become disjunct without premeiotic doubling.