Professor at the Institute of Science and Technology, Austria.
Breaking the Wall that Limits Evolution. How Sexual Recombination Accelerates Adaptation.
Thank you for the invitation to speak here, and thank you for coming to listen. This meeting is about breaking walls, breaking the walls that limit our understanding of the world, breaking the walls that separate different disciplines. Many of the talks, like the talk we just heard today, will be about immediate human problems and how we can try to solve them. But what I want to do is do something rather different, and I think the talks in the first session following me will be rather like this. I will be taking a much longer view, going all the way back to the formation of the planet, 4,500 million years ago, to the origin of life just after, and then following the long span of evolution through to the present extraordinary diversity of life that we see around us. I want to explain how it is possible that sexual reproduction works with natural selection to build up the extraordinarily complex and sophisticated organisms that we see around us; how it works by breaking the walls between individuals.
When we look at the living world, we are impressed by all kinds of extraordinary adaptations, for example the eyes of the owl, which could detect a few photons of light, or the antennae of a female moth, which can detect a few molecules of pheromones sent out by a male many kilometres away. We are impressed by the sophisticated social structure of a beehive, by the elaborate behaviours of, for example, a New Caledonian crow, which can dig out insects with a very simple tool. Perhaps we are most impressed by our own brain, which has laid the basis for a new form of evolution – the evolution of ideas rather than the evolution of genes.
But most of the living world is actually invisible to us. It consists of single-celled organisms, which in their own way are just as extraordinary. I just showed a few random pictures here, but these organisms can display an extraordinary biochemical versatility and an extraordinarily precise genetic system. We all share, indeed all organisms share, this basic genetics and biochemistry, a system that allows us to replicate our DNA with extreme accuracy, with less than one mistake per thousand million copying events, to carry out precise reactions that catalyse specific reactions at ambient pressure and temperature, something completely beyond the reach of human chemists. We all share this genetic system, which descends from our common ancestor – the common ancestor of all living organisms, which lived about 3,600 million years ago.
Here is a schematic picture of our current understanding of the “tree of life”, the relationships between all living organisms. At the centre, labelled “the root”, is this common ancestor. Shortly after that ancestor lived, life diversified into three domains: the Bacteria, the Archaea and the Eukaryotes. All I want to emphasise from this picture is that almost all of these organisms are single-celled, invisible microbes. The organisms that we are familiar with: ourselves, the animals, fungi, the plants and so on, are tucked away within the vast and ancient diversity of the eukaryotes.
Arguably, one of the most important transitions in evolution was the emergence of the eukaryotes about 1,800 million years ago. These organisms are characterised by a complex cellular structure: contrasting on the left, bacteria, and on the right a much larger eukaryote cell, in which the nucleus encloses the genetic material, the DNA; and there are many other cellular compartments, which carry out specific functions: photosynthesis, respiration and so on. This complex structure arose through an ancient symbiosis, through the coming together of different organisms, when an ancient microbe engulfed two or more species of bacteria. The genes from those bacteria were gradually incorporated into the eukaryote genome, so that our own genomes are a mosaic of bacterial and archaeal genes. The remnants of those symbiotic bacteria still carry out today the key functions of respiration, of photosynthesis, and perhaps many other functions.
The eukaryotes diversified into a whole range of mostly single-celled organisms, but also green plants and animals, like jellyfish. But these multi-cellular organisms, these large organisms, the ones that we can really see in the fossil record, only emerged relatively recently – meaning, 565 million years ago, which is recent set against the long span of 4.5 thousand million years. The first soft-bodied forms were quickly followed shortly by a radiation at the base of the Cambrian into the whole range of organisms that we see in the conventional fossil record.
So, we see this very, very long span of evolutionary time. The things we think about when we think of evolution are really much more recent than that: the extinction of the dinosaurs 65 million years ago, the diversification of the primates, our divergence from the chimpanzee and the gorilla lineages – all very recent on this time scale – and, of course, human history, recent history, our own culture, absolutely occurring at the twinkling of an eye, set against this long span of time.
So, what I want to do is to ask two very closely related questions. First, why is sexual reproduction so widespread amongst the eukaryotes? I said that the eukaryotes are characterised by a complex internal organisation, which arose originally by symbiosis, by the coming together of different organisms. But, they are also characterised, almost always, by an organised process of sexual reproduction, by the coming together of different cells to produce a diploid – that is, a cell with two genomes – and then by a complex process of meiosis, in which these genomes are aligned, broken and rejoined to produce new combinations of genes. This process of sexual reproduction, the coming together, reshuffling and separation of genomes seems to be essentially obligatory across the eukaryotes. Second, and closely connected with this: how is it that such complicated organisms can evolve and diversify, especially given that the eukaryotes have larger cells, typically smaller population sizes, slower reproduction – all disadvantages in evolution – and yet they have diversified to produce extraordinarily complex and sophisticated structures?
What I will be trying to do is really to sketch out the consensus view that has emerged over the last century or so in evolutionary biology, but also to hint at the breakthroughs that are in progress with the current deluge of data from DNA sequencing.
The existence of sexual reproduction is really completely baffling, because it is a lot of trouble; why do we bother finding a mate? How do single cells find each other? It is hard to accurately recombine genomes, to line them up, to cut, to rejoin, without making mistakes, without generating errors – a slow, expensive, difficult process. Once we have gone to all this trouble, we produce new combinations of genes, randomising combinations that have been built up by selection to work well together. In the short term, that brings a substantial and measurable cost.
Immediately after sexual production evolves, there is a specialisation into males and females: males produce large numbers of small gametes, females do all the work, provide the resources, such as large eggs and parental care etc. in higher organisms. That leads to an immediate so-called two-fold cost: a female who produces entirely daughters will increase in the population at a rate twice that of a sexual, which must waste its efforts producing males. There is the possibility of the spread of parasites, of venereal disease and so on; and there is the evolution of sexual selection in which males waste their energy making large displays or fighting each other: stag beetles fighting other males for mates, flowers displaying to attract pollinators, and so on.
Why? Why does this happen? It is a real puzzle, but basically, the answer is simply that in the long term sexual recombination is essential for natural selection to work effectively. To explain this, we have to understand why selection works so well.
The problem is to understand how selection can pick out, from the vast space of possible DNA sequences, that tiny fraction that can actually work to actually produce organisms that can survive and reproduce. Only a tiny fraction of DNA sequences will work, will be fit. A process of simple selection from random variation will be hopeless: there has to be a stepwise process of building up slight changes that increase fitness, by gradual accumulation of individually favourable changes, improving sequences in a stepwise manner.
Favourable combinations of mutations have to be brought together, and this is very difficult in the absence of sex. Here we have a simple diagram, showing time on the horizontal axis. We begin with “a”, “b”, two genes, which don’t work so well. A mutation from “a” to “A” is favoured, and similarly, a mutation from “b” to “B” is favoured, and so they each increase. Crucially, however, without recombination, separate mutations cannot come together. You cannot get big “A” and big “B” until, lets say, “A” is fixed in the population and then “B”. Evolution has to work in series, rather than in parallel. Thus, almost all the mutations that occur at any one time are wasted, because only one of them can win – they compete with each other. Indeed, asexual species do not persist – they tend to go extinct. They may do well for a short time, but over hundreds of thousands of years or millions of years, they do not persist.
Thus selection is only effective in the presence of recombination. In this example, selection and recombination have to work together to bring together the best combination “AB” here. In this diagram, you see a twist to the story: any gene that favours recombination, or favours sexual union, therefore favours the production of these fit combinations “AB”. That gene, lets say “M”, which increases recombination, will itself gain an advantage by being associated with the good gene combinations that it produces.
My contribution to all this has really been to develop the mathematical theory that quantifies the advantage to sex and recombination and to show that, in principle, it can actually give a sufficient advantage to spread, to maintain sex, despite its obvious costs. However, though the theoretical development – contributed to by many people – has been quite mature, we had very little data with which to check the theory. Suddenly, though – over just the past few years – we have abundant data from the genome sequencing projects, and this is actually quite disconcerting: to go from a field in which there was virtually no data, to a field in which we suddenly have samples of thousands of complete genomes, from many different species. Soon we will, I think, have a good quantitative understanding of the extent of selection, and we will know whether selection is in general sufficiently strong to actually account for the extent of sex and recombination. We know already that most functional regions are constrained, and thus evolve more slowly, but a few change much more rapidly. Indeed, it may be that most of the differences in protein sequence between species are driven by selection. However, we do not yet know whether selection is sufficiently strong and widespread to maintain sex and recombination. Still, I hope I have given you at least a hint of the way in which combination of theory and data can explain this basic feature of the living world. Thank you.