**1. Introduction**

The adaptive function of sex remains, today, one of the major unsolved problems in biology. Fundamental to achieving a resolution of this problem is gaining an understanding of the function of meiosis. The sexual cycle in eukaryotes has two key stages, meiosis and syngamy. In meiosis, typically a diploid cell gives rise to haploid cells. In syngamy (fertilization), typically two haploid gametes from different individuals fuse to generate a new diploid individual. A unique feature of meiosis, compared to mitosis, is recombination between non-sister homologous chromosomes. Usually these homologous chromosomes are derived from different individuals. In mitosis, recombination can occur, but it is ordinarily between sister homologs, the two products of a round of chromosome replication. Birdsell & Wills (2003) have reviewed the various hypotheses for the origin and maintenance of sex and meiotic recombination, including the hypothesis that sex is an adaptation for the repair of DNA damage and the masking of deleterious recessive alleles. Recently, we presented evidence that among microbial pathogens, sexual processes promote repair of DNA damage, especially when challenged by the oxidative defenses of their biologic hosts (Michod et al., 2008). Here, we present evidence that meiosis is primarily an evolutionary adaptation for DNA repair. Since our previous review of this topic (Bernstein et al., 1988), there has been a considerable increase in relevant information at the molecular level on the DNA repair functions of meiotic recombination, and this new information is emphasized in the present chapter.

#### **2. Meiosis in protists and simple multicellular eukaryotes is induced in response to stressful conditions that likely cause DNA damage**

Eukaryotes appeared in evolution more than 1.5 billion years ago (Javaux et al., 2001). Among extant eukaryotes, meiosis and sexual reproduction are ubiquitous and appear to have been present early in eukaryote evolution. Malik et al. (2008) found that 27 of 29 tested meiotic genes were present in *Trichomonas vaginalis*, and 21 of these 29 genes were also present in *Giardia intestinalis*, indicating that most meiotic genes were present in a common ancestor of these species. Since these lineages are highly divergent among eukaryotes, these authors concluded that each of these meiotic genes were likely present in the common ancestor of all eukaryotes. Dacks and Roger (1999) also proposed that sex has a single

Meiosis as an Evolutionary Adaptation for DNA Repair 359

These observations suggest that meiosis is an adaptation for dealing with stress, particularly oxidative stress. It is well established that oxidative stress induces a variety of DNA damages including DNA DSBs, single-strand breaks and modified bases (Slupphaug et al., 2003). Thus we hypothesize that, in facultative sexual protists and simple multicellular eukaryotes, sex, with the central feature of meiosis, is an adaptive response to DNA damage,

**3. DNA damages induced by exogenous agents cause increased meiotic** 

If recombination during meiosis is an adaptation for repairing DNA damages, then it would be expected that exposure to DNA damaging treatments would increase the frequency of recombination, as measured by crossovers between allelic markers. Stimulation of allelic recombination was reported in the fruitfly *Drosophila melanogaster* in response to exposure to the DNA damaging agents UV light (Prudhommeau & Proust, 1973), X-rays (Suzuki & Parry, 1964), and mitomycin C (Schewe et al., 1971). X-rays induce recombination in meiotic cells not only of *D. melanogaster* females, but also of males, which normally display no

Increased meiotic recombination in response to X-irradiation has also been reported in

**4. During mitosis and meiosis, DNA damages caused by diverse exogenous** 

Molecular recombination (that is homologous physical exchange or informational exchange) during mitosis and meiosis functions as a DNA repair process designated homologous recombinational repair (HRR). Many of the gene products employed in mitotic HRR are also employed in recombination during meiosis. It is this consistent function of recombination across meiosis and mitosis in eukaryotes and transformation in prokaryotes that we seek to understand through the repair hypothesis. Mutants defective in HRR genes in *D. melanogaster* and yeast have reduced ability to repair DNA damages arising from a variety of exogenous sources. These mutants are also defective in recombination during meiosis. In general, loss of HRR capability causes increased sensitivity to killing by agents that harm cells primarily through induction of DNA damage. These agents are listed in Table 1. There have been no reports, that we know of, that HRR defective cells are sensitive to agents that

In *D. melanogaster*, mutants defective in genes *mei-41, mei-9, hdm, spnA* and *brca2* have reduced spontaneous allelic recombination (crossing over) during meiosis and increased sensitivity to killing by exposure to numerous DNA damaging agents (Table 1). The Mei-41 protein is a structural and functional homolog of the human Atm (ataxia telangiectasia) protein (Hari et al., 1995), which plays a central role in HRR. The Mei-9 and Hdm proteins are components of a multiprotein complex that resolves meiotic recombination intermediates (Joyce et al., 2009). The SpnA protein is a homolog of yeast Rad51 (Staeva-Vieira et al., 2003), and Rad51 plays a central role in strand-exchange during HRR. The *D. melanogaster* Brca2 protein, a homolog of the human Brca2 protein that protects against breast cancer, regulates the activity of Rad51 protein in HRR. The Brca2 protein is required

*Caenorhabditis elegans* (Kim & Rose, 1987), and in *S. cerevisiae* (Kelly et al., 1983).

**agents can be repaired by homologous recombination** 

harm cells by mechanisms other than primarily causing DNA damage.

for HRR of DSBs during meiosis (Klovstad et al., 2008).

particularly oxidative DNA damage.

recombination during meiosis (Hannah-Alava, 1964).

**recombination** 

evolutionary origin and was present in the last common ancestor of eukaryotes. Recently, this view received further support from a study of amoebae. Although amoebae generally have been assumed to be asexual, Lahr et al. (2011) showed that the majority of amoeboid lineages were likely anciently sexual, and that most asexual groups have probably arisen recently and independently.

Eukaryotes arose in evolution from prokaryotes, and eukaryotic meiosis may have arisen from bacterial transformation, a naturally occurring sexual process in prokaryotes. The fundamental similarities between transformation and meiosis have been explored (H. Bernstein & C. Bernstein, 2010). Bacterial transformation, like meiosis, involves alignment and recombination between non-sister homologous chromosomes (or parts of chromosomes) originating from different parents. Both during transformation and meiosis, homologs of the bacterial *recA* gene play a central role in the strand transfer reactions of recombination, indicating a mechanistic similarity. Also, bacterial transformation is induced by environmental stresses that are similar to those that induce meiosis in protists and simple multicellular eukaryotes, suggesting that there was continuity in the evolutionary transition from prokaryotic sex to eukaryotic sex. Evidence indicates that bacterial transformation is an adaptation for repairing DNA (Michod et al., 1988; Hoelzer & Michod, 1991; Michod & Wojciechowski, 1994; reviewed by Michod et al., 2008). Thus meiosis may have emerged from transformation as an adaptation for repairing DNA.

Among extant protists and simple multicellular eukaryotes sexual reproduction is ordinarily facultative. Meiosis and sex in these organisms is usually induced by stressful conditions. The paramecium tetrahymena can be induced to undergo conjugation leading to meiosis by washing, which causes rapid starvation (Elliott & Hayes, 1953). Depletion of the nitrogen source in the growth medium of the unicellular green alga *Chlamydomonas reinhardi* leads to differentiation of vegetative cells into gametes (Sager & Granick 1954). These gametes can then mate, form zygotes and undergo meiosis. Upon nitrogen starvation or desiccation, the human fungal pathogen *Cryptococcus neoformans* undergoes mating or fruiting, both processes involving meiosis (Lin et al., 2005).

In addition to starvation, oxidative stress is another condition that induces meiosis and sex. The haploid fission yeast *Schizosaccharomyces pombe* is induced to undergo sexual development and mating when the supply of nutrients becomes limiting (Davey et al., 1998). Moreover, treatment of late-exponential-phase *S. pombe* vegetative cells with hydrogen peroxide, which causes oxidative stress, increases the frequency of mating and production of meiotic spores by 4- to 18-fold (C. Bernstein & Johns, 1989). The oomycete *Phytophthora cinnamomi* is induced to undergo sexual reproduction by exposure to the oxidizing agent hydrogen peroxide or mechanical damage to hyphae (Reeves & Jackson, 1974). In the simple multicellular green algae *Volvox carteri,* sex is induced by heat shock (Kirk & Kirk, 1986). This effect can be inhibited by antioxidants, indicating that the induction of sex by heat shock is mediated by oxidative stress (Nedelcu & Michod, 2003). Furthermore, induction of oxidative stress by an inhibitor of the mitochondrial electron transport chain also induced sex in *V. carteri* (Nedelcu et al., 2004). The budding yeast *Saccharomyces cerevisiae* reproduces as mitotically dividing diploid cells when nutrients are plentiful, but undergoes meiosis to form haploid spores when starved (Herskowitz, 1988). When *S. cerevisiae* are starved, oxidative stress is increased and DNA double-strand breaks (DSBs) and apurinic/apyrimidinic sites accumulate (Steinboeck et al., 2010). Perhaps, in *S. cerevisiae,* the induction of sex by starvation is mediated by oxidative stress, analogous to the way induction of sex by heat is mediated by oxidative stress in *V. carteri*.

evolutionary origin and was present in the last common ancestor of eukaryotes. Recently, this view received further support from a study of amoebae. Although amoebae generally have been assumed to be asexual, Lahr et al. (2011) showed that the majority of amoeboid lineages were likely anciently sexual, and that most asexual groups have probably arisen

Eukaryotes arose in evolution from prokaryotes, and eukaryotic meiosis may have arisen from bacterial transformation, a naturally occurring sexual process in prokaryotes. The fundamental similarities between transformation and meiosis have been explored (H. Bernstein & C. Bernstein, 2010). Bacterial transformation, like meiosis, involves alignment and recombination between non-sister homologous chromosomes (or parts of chromosomes) originating from different parents. Both during transformation and meiosis, homologs of the bacterial *recA* gene play a central role in the strand transfer reactions of recombination, indicating a mechanistic similarity. Also, bacterial transformation is induced by environmental stresses that are similar to those that induce meiosis in protists and simple multicellular eukaryotes, suggesting that there was continuity in the evolutionary transition from prokaryotic sex to eukaryotic sex. Evidence indicates that bacterial transformation is an adaptation for repairing DNA (Michod et al., 1988; Hoelzer & Michod, 1991; Michod & Wojciechowski, 1994; reviewed by Michod et al., 2008). Thus meiosis may have emerged

Among extant protists and simple multicellular eukaryotes sexual reproduction is ordinarily facultative. Meiosis and sex in these organisms is usually induced by stressful conditions. The paramecium tetrahymena can be induced to undergo conjugation leading to meiosis by washing, which causes rapid starvation (Elliott & Hayes, 1953). Depletion of the nitrogen source in the growth medium of the unicellular green alga *Chlamydomonas reinhardi* leads to differentiation of vegetative cells into gametes (Sager & Granick 1954). These gametes can then mate, form zygotes and undergo meiosis. Upon nitrogen starvation or desiccation, the human fungal pathogen *Cryptococcus neoformans* undergoes mating or fruiting, both

In addition to starvation, oxidative stress is another condition that induces meiosis and sex. The haploid fission yeast *Schizosaccharomyces pombe* is induced to undergo sexual development and mating when the supply of nutrients becomes limiting (Davey et al., 1998). Moreover, treatment of late-exponential-phase *S. pombe* vegetative cells with hydrogen peroxide, which causes oxidative stress, increases the frequency of mating and production of meiotic spores by 4- to 18-fold (C. Bernstein & Johns, 1989). The oomycete *Phytophthora cinnamomi* is induced to undergo sexual reproduction by exposure to the oxidizing agent hydrogen peroxide or mechanical damage to hyphae (Reeves & Jackson, 1974). In the simple multicellular green algae *Volvox carteri,* sex is induced by heat shock (Kirk & Kirk, 1986). This effect can be inhibited by antioxidants, indicating that the induction of sex by heat shock is mediated by oxidative stress (Nedelcu & Michod, 2003). Furthermore, induction of oxidative stress by an inhibitor of the mitochondrial electron transport chain also induced sex in *V. carteri* (Nedelcu et al., 2004). The budding yeast *Saccharomyces cerevisiae* reproduces as mitotically dividing diploid cells when nutrients are plentiful, but undergoes meiosis to form haploid spores when starved (Herskowitz, 1988). When *S. cerevisiae* are starved, oxidative stress is increased and DNA double-strand breaks (DSBs) and apurinic/apyrimidinic sites accumulate (Steinboeck et al., 2010). Perhaps, in *S. cerevisiae,* the induction of sex by starvation is mediated by oxidative stress, analogous to the

way induction of sex by heat is mediated by oxidative stress in *V. carteri*.

recently and independently.

from transformation as an adaptation for repairing DNA.

processes involving meiosis (Lin et al., 2005).

These observations suggest that meiosis is an adaptation for dealing with stress, particularly oxidative stress. It is well established that oxidative stress induces a variety of DNA damages including DNA DSBs, single-strand breaks and modified bases (Slupphaug et al., 2003). Thus we hypothesize that, in facultative sexual protists and simple multicellular eukaryotes, sex, with the central feature of meiosis, is an adaptive response to DNA damage, particularly oxidative DNA damage.
