The Changing Face of Evolutionary Thinking

In evolutionary biology, controversies, for whatever reason, never seem to die or be resolved; they fade away and can reemerge at any time. One of the oldest controversies in our field is that between selectionism and mutationism. It goes back at least to the beginning of the 20th century with Hugo de Vries being the best known proponent of the later deposed idea that mutations are the driving force of evolution. And now we have the book by Masatoshi Nei (2013) with the title Mutation-Driven Evolution. But what is remarkable is that the author is not someone from the fringes or even outside of evolutionary biology. The author is not even distant from population genetics and the molecular study of evolution but is one of the founding fathers and pioneers of what is now called the field of molecular evolution. What did we miss in the meantime? Isn’t selection well studied and the well-established driver of adaptive evolutionary change? What if anything can be gained from reopening the gates to a century-old controversy we have thought to have faded away for good? Although old controversies fade away and reemerge, they also reemerge profoundly transformed, and it is likely this transformation to which one can attach one’s hope for progress in science in spite of recurring themes appearing. While the old mutationism had many faces, described in Chapter 1 in Nei’s book, like the idea that single mutations are sufficient to create a new species. This idea was mostly due to a misunderstanding of what the nature of species is, in particular, in sexually reproducing species. The new mutationalism we encounter in Masatoshi Nei’s book has a different face, and I want to dedicate the few lines of this review to explain how I understand the core ideas of “new mutationalism.” Uncontroversial, if there is anything in evolutionary biology, is the insight that evolutionary change is a two-step process: mutation and some form of fixation of mutants in the population, be it by selection, drift, or meiotic drive. What is debatable, however, is the question what the explanatory power of each of these processes/mechanisms is. In the classic synthetic theory of evolution, which is the core of the mainstream version of evolutionary thinking, selection is the only truly causal factor, which gives direction to evolutionary change. That is to say, anything that is not a random outcome of evolution is to be explained by natural selection. An important argument was and is that mutation is a linear process with low rate, mutation rates tend to be small, while natural selection is based on an exponential growth process and thus is more effective in directing allele frequency change than mutation. This argument is one about the dynamics of allele frequency changes, where mutation seems to be a weaker force. Of course, if one defines evolutionary change as allele frequency change, then this is the appropriate resolution of controversy. However, is allele frequency change the only perspective one legitimately can take on evolution? The new mutationalism debate is not a conversation about the rate of allele frequency change but one about the outcomes of evolution. Are the products of evolution better explained by selection or by mutation? What one means by “explained by mutation” needs to be fleshed out, because not even the most dedicated proponent of a selectionist model of evolution will deny that mutations are necessary for evolutionary change. Also, it is not always easy to understand the arguments in Nei’s book, because in some places Nei asserts that mutation is necessary to explain some kind of phenomenon. But it is not clear how this assertion is different from the generally accepted idea that mutations have to occur for selection to effect change. This reviewer only came to fully appreciate the force of Nei’s argument until Chapter 9, which is the synthesis of his thinking based on the historical review and the reviews of empirical evidence from molecular evolution to which the previous eight chapters are dedicated. As far as I can see, the core idea in Nei’s book, to which evolutionary biologists need to pay attention, is constraint-breaking mutations. In the classical selection-based model of evolution, genetic variation is assumed to be abundant and supplied by mutations with little directional bias. This idea is supported by genetic data on the heritability of quantitative characters like wing length and body size. For these characters, it is plausible that natural selection explains the kind and direction of evolutionary change, such as the latitudinal variation of body size in mammals. This is plausible because of the abundance of genetic variation available for these traits in outbred populations. The alternative view, applicable to characters other than body size or wing length, is that patterns of evolutionary change can either be enabled or directed by the mutational process itself and thus properly be explained by mutation rather than natural selection. What it means to say is “explained by mutation” can be specified as meaning “explained by the specific mutational mechanisms involved in specific cases,” rather than by the abstract concept of mutation. To me three bodies of knowledge support the notion of mutation-driven evolution; most convincingly, namely, evolution of large-scale genomic features as explained by Michale Lynch’s model of genome evolution, experimental studies of protein function evolution, and studies on the evolution of development. Briefly, Michael Lynch shows that large-scale patterns of genome evolution can, to a surprising extent, be explained by the interplay of random genetic drift and the molecular biases of mutational change. If this model turns out to be correct, at least in a substantial set of cases, we would have a broad swat of evolutionary phenomenology that is driven by mutation rather than selection. Thornton and collaborators have shown in detailed evolutionary, structural, and biochemical studies of the evolution of ligand specificity in glucocorticoid receptors that evolutionary transitions in ligand specificity depend on coincidental presence of certain amino acid residues, enabling the functionally adaptive amino acid substitutions. Hence, the outcome of evolution depends on mutational events not “seen” by natural selection because they seem to be neutral and only enabling rather than functionally important themselves. Finally, the whole undertaking of developmental evolution is a rich illustration of how developmental, genetic, and genomic constraints are limiting and directing phenotypic evolutionary change. For instance, Shapiro and collaborators have shown that repeated evolutionary pelvic spine loss in sticklebacks is driven by a genomic instability in the pelvic enhancer of the Pitx1 gene. To clarify his position, Nei redefines evolution as changes (increases and decreases) in organismal complexity (Nei 2013, p. 10). One may argue that such a definition is as limited as that based on allele frequency change. (Is then an evolutionary change in body size in response to temperature change not evolution?) However, the important point here is not the comprehensiveness of this or any other “definition” of evolution. What counts is the perspective on the evolutionary process that a given “definition” opens up. The research programs and the answers we get critically depends on whether a researcher thinks she is studying allele frequency change or increase in organismal complexity. The difference matters profoundly, because it determines the range of knowledge we gain from these research activities. Of course, it is still important to understand the causes of allele frequency change, although we do have a substantial amount of mathematical theory and data about that, as it is important to understand the mechanisms that lead to complex organisms. The latter we know very little about, even though it is one of the most striking outcomes of evolution and touches on the question of human nature in a most profound way. Nei’s perspective is broadly consistent with those of developmental evolution, the functional synthesis and the mutationist view of genome evolution, and is one that is greatly expanding the scope and nature of evolutionary thinking.