Why have sex?

On the origins of meiosis.

Of the many innovations that occurred during the evolution of eukaryotes, one of the most mysterious and powerful was the development of meiosis and sexual reproduction. A paper from a few years back delved into how this process might first have started, given that it depends on several complex innovations. (And another paper more recently).

But first, why is sex so durable, evolutionarily speaking, resulting in an Earth where all large, complex life forms engage in it? The costs are quite severe, after all: partners must be found, which can be a particular trial for sessile plants, corals, etc.; selection for fitness can be waylaid by mate selection and other sexual games; Parents give up half their genes in creating each child, compared to creating complete copies as they would by traditional, clonal reproduction; likewise a sexual population has to have males, which double the resource needs compared to a clonal population where every member is functionally female.

The answer lies in population genetics. A clonal population, such as most bacteria, can generate mutations and adapt to external conditions. The development of antibiotic resistance is notorious. But evolution is a parallel process, where the whole population is tested, and many variants are more or less successful. Everyone has deleterious mutations mixed in amongst the beneficial ones, especially since they are always more numerous. So in a clonal population, any good mutation that occurs will be trapped in its current genome and have to compete against all the other clones with all their good and bad mutations. Another good mutation will have to fight the same battle, without the chance to team up with the first one, unless the first one has already taken over the population. This way many beneficial mutations are lost, especially those with weak effects, which are the majority, naturally.

Even worse, bad mutations tend to pile up in each clone lineage, since there is no way to get rid of them. Each mother gives the full complement of her mutations, both inherited and those that happened during her life, to her offspring. In humans, we accumulate about 175 mutations per individual before reproduction, of which roughly 1 or 2 are detectably deleterious. While astonishing repair processes have evolved to keep such errors to a minimum, there are always some, and in a clonal lineage, they are always building up, despite the ongoing selection against those which add up to worse than average problems, a process termed "Muller's ratchet".

What happens in a sexual population? Well, it is critical to realize that it does behave much more like a population- an evolutionary village, so to speak- than as competing clonal lineages. Individual alleles are recombined around and mixed up among offspring, so that there is far more diversity within the population, which allows, stochastically, for good mutations/alleles to come together in some offspring, and deleterious mutations/alleles to come together in others. Given selection where the latter die and the former flourish, the system enables far more effective use of the opportunities provided by mutations throughout the population then clonality does.

The autobiography of Julius Erving, "Dr. J", provides a graphic example. He is extraordinarily gifted in all respects, including writing. Yet his brother Marky was sickly and died very young. The difference is tragic on a human level, but routine on a genetic level. Embryos with especially deleterious alleles will frequently die before birth, hiding the true rate of this genetic "sweeping" mechanism.

Sex is so powerful that bacteria have developed several mechanisms for doing it on a small scale, such as extending pili to partners so that they can exchange limited amounts of DNA. This how antibiotic resistance spreads around so quickly, and helps bacteria exchange out some of their accumulated mutations by homologous repair, (which is common among all organisms), though at the cost of bringing in parasitic DNA elements like transposons and viruses. But bacteria have never developed the full monte: fusion of whole genomes with total remixture and sharing out to subsequent progeny, let alone obligatory sex before reproduction. Eukaryotic sex involves the fusion of complete, haploid genomes, which recombine pervasively, both by way of the independent assortment of whole chromosomes and by smaller sub-chromosomal recombintation events, to create unique, new haploid genomes, which are then sent out as new gametes.

Bacteria share small amounts of DNA though conjugation pili.

The rise of this process is a bit hard to understand because there are several complex events needed, none of which make immediate sense by themselves. First is the meiotic division, which uses most of the tools of normal mitotic cell division, plus some more (suppression of sister separation in the first division and then suppression of DNA replication in the second division) to turn a diploid genome into a carefully reduced haploid gamete genome. Second is the part of the process called synapsis where all chromosomes from the two parents align along their full length. Third is the recombination that is obligatory between these homologs in order to keep them attached at initial stages, and to interchange segments of the respective genomes. This is not to mention the fusion of gamete cells and other perphernalia of sex, which are less innovative from a molecular, and probably evolutionary, standpoint.

Basic model of meiosis, in comparison to mitosis.

The proposal made in this paper, by a luminary in the field, after whom the Holliday junction is named, is that synapsis of homologous chromosomes may have been the leading event in the development of meiosis and sex, and is understandable as a solution to a completely different problem. Quite apart from the recombination that happens when bacteria encounter DNA coming in from other cells, homologous recombination also happens after replication and before division to repair damage, of which replication is a frequent cause. But one of the hallmarks of eukaryotes is size- big cells and big genomes- they are the SUVs of cellular biology. As genomes grew, the chances of making an error in this internal recombination, with all the repetitive DNA and duplicated genes lying about, grew rapidly as well, and posed a serious danger of creating new damage. This led to the preferability of setting up a whole-genome alignment process, i.e. true synapsis, and confining it to the inverval between replication and division.

"To sum up, we propose that the selection pressures for homolog synapsis and the origins of meiosis were to improve recombinational accuracy and to restrict it to a safe interval, while retaining its short-term (repair) benefits. A cell lineage that had evolved this capability for diploid cells would be less error-prone in transmitting its genetic material."

In the original setting of haploid, single cells, as the first proto-eukaryotes undoubtedly were, this would have revolutionized post-replication homologous DNA repair, making the process far more systematic and reliable. Indeed, eukaryotes still have their DNA aligned during most of the cell cycle, though not in the elaborate synaptonemal complex now found in the first phase of meiosis. Since these organisms were not dipoid, but habitually haploid, the subsequent division would have already resembled the second division of meiosis, back to the haploid state.

"In principle, the molecular evolution of a new cohesin molecule that specifically promoted homolog pairing might have provided the crucial trigger for meiosis."

The next step to true sexuality by this model were to adopt the practice of mating between separate cells (which could have been related to the partial genetic exchange common among bacteria) to generate a diploid where truly different alleles over entire genomes are combined in one cell. While this could have worked on its own, (and exists today as parasexual cycles in fungi, where reduction back to a haploid set of chromosomes is more or less random), the addition of DNA replication, as well as synapsis and recombination as above, at this step would have necessitated a special variant of mitosis to become the first (reductional) division of meiosis, whereby the replicated homologs from the two parents align in a novel four-chromosome bundle and each segregate precisely in half, though with random polarity, prior to a second division.

"In many unicellular eukaryotes, haploid sex-cell fusion leads promptly to nuclear fusion, which immediately triggers meiosis, thus regenerating the haploid state. In contrast, in more complex, multicel- lular eukaryotes, meiosis is greatly delayed following the initial fusion of sex cells, taking place much later in the life cycle, during gametogenesis."

Subsequently, organisms would adopt the diploid state as the organismal default (not so difficult in microorganisms), which has powerful effects on genetic diversity, since it allows recessive alleles some breathing space to survive and even to fill strategic niches in the population. With the advent of multicellularity, the meiotic divisions could be re-scheduled to now-vestigial haploid cells as part of a special gamete generating process. Clearly there is a lot to chew on with this model, and a need to flesh out and gather evidence to support the many inferred steps. But it is a highly interesting idea for the stepwise development of a process that is now notoriously complex on the molecular as well as all other levels.

A problem, however, is that modern eukaryotic cells, though they use aligned sisters in a highly regulated fashion for post-replication DNA repair, do not use anything like true synapsis. That makes it difficult to suppose that synapsis ever played a role in this process, or that it was the leading element of the other steps of meiosis. One might counter that haploid cell fusion with para-sexual reduction was perhaps the first step in the sequence, (possibly even originating in a predatory setting), after which the development of replication, synapsis, diploidy, and the special homolog-separating division of meiosis I were developed to better clean up the mess. Anyhow, next week, we will delve into another theory about the origins of eukaryotes.

Overall, sex is a machine for speeding up evolution, generating a quantum leap of accessible genetic diversity within a species / population that allows bad genes to be left behind without discarding everything else in those genomes, and good genes to be concentrated in the winners. Do individuals benefit, or their genes, or the species as a whole? There is an element of group selection intrinsic to this rationale for sex, since the benefits accrue down the line to the genes of the species and the species as a whole, not to the individuals involved now, especially not those stuck with the short end of the genetic stick.

• An unusual meiotic short-circuit & genome mixing pathway in yeast cells.
• Women, and women's work, are not highly valued. Another lesson in the social construction of pay decisions.
• Seniors (and the rich) are doing relatively well, at least vs young workers afflicted by a terrible job market. "I suspect that most Fed policymakers receive relatively little input on the economy from people who are younger than 40."
• Does the US have any place for lower class workers?
• Or does it have the education and institutions to take them to a higher class?
• Mainstreaming MMT.
• Repression is hard work, and Trump, a relief.