What evolutionary developmental biology (evo devo) brings to evolutionary biology
9 January 2017
Evolutionary biology is a very vibrant and highly successful discipline. Since its reformulation during the modern synthesis it has successfully tackled many major questions and developed into a sophisticated, powerful framework. But it is important to emphasize that there are foundational questions in evolutionary biology, questions that have motivated evolutionary biology from its inception, that remain to be adequately addressed. ‘How do novel complex traits originate?’ is such a question. How does a major invention in evolution come into being in the first place? What are the baby steps of innovation?
Mainstream evolutionary biology focuses on the origin, accumulation, and differential fixation of small variants of preexisting traits within populations or among closely related groups as the critical substrate for phenotypic diversification. It lives in a world of descent with modification where everything new must come from something old, more or less gradually. There is nothing wrong with this framework, and it is applicable to and useful in a wide range of evolutionary contexts. But it encounters significant limits when applied to the issues of novelty and innovation. A novel trait, by most definitions, lacks obvious homology to preexisting structures. In other words, novelty typically begins where homology ends. The insect wing is a classic example of a morphological innovation with drastic ecological and evolutionary consequences: insects were able to take to the skies, and evolve life styles and life-cycles unreachable by non-winged arthropods, as well as diversify dramatically in the process. But the insect wing lacks obvious homology to other insect appendages and cannot be understood simply as a modified leg, or antenna, or mouthpart. At the same time we encounter another common problem: all insects (in the strict sense) have wings except those that secondarily lost them. Thus, no comparisons are possible between ancestrally wingless and slighty winged and partially winged and almost fully winged insects. Instead that variation has been lost in deep time and is unavailable for present day studies. This situation is similar to what we observe for other famous examples of evolutionary novelty: the turtles’ shell, the elephant’s trunk, or the light producing organ of fireflies. Lacking both obvious correspondence to other traits and phenotypic variation within populations or among closely related species, conventional evolutionary biology has difficulty framing an adequate research program through which to address the issue of novelty. And so for the most part, it doesn’t. As a consequence, we have learned a great deal about the quantitative and population genetic architecture of insect wing size and shape, but very little about the very origin of the first wing.
It took the discipline of evo devo to provide a more productive way of thinking and framing a research program on evolutionary novelty to make headway. While traditional approaches focus on how variation is transmitted across generations, and how this process might explain long-term evolutionary transitions, they run into trouble when suitable variation is not experimentally accessible, as in the case of insect wings. In contrast, evo devo expands the focus to include how traits are made during development, and how the process of building a trait (trait construction) compares to that of other traits, regardless of whether they do or do not share obvious homology. This seemingly simple extension makes a difference for many reasons, three of which in particular are worth highlighting here.
Understanding the biological basis of form
Evo devo provides a much deeper understanding of the biological basis of organismal form and function by filling an abstract genotype-phenotype map with the realities of developmental pathways, cells, signaling molecules, morphogenetic movements, and the processes of tissue differentiation. This can profoundly impact the research approach used to study the evolution of a given trait. For example, the tails of swallowtail butterflies (see images) are perhaps not exactly a novelty, but certainly an interesting trait which has diversified significantly within this group. It looks like an outgrowth, an extension of the wing, as if the wing’s base somehow acquired the ability to keep growing while the remainder of the wing stopped. And some species, as well as genotypes within populations, seem to possess heritable variation that enhances or mutes the further extension of these outgrowths. But once researchers examined how wing tails form during development all this had to be revised. It turns out that the wings of swallowtails at the pupal stage, which at this point have completed all of their growth, lack tails. Tails only emerge later during the pupal stage after more or less substantial programmed cell death removes existing cells and – like a cookie cutter – leaves behind what we after the final molt recognize as the adult wing’s tail. The tails of swallowtail butterflies are therefore not outgrowths – instead they are leftovers. Knowing how traits develop thus informs how one might frame a research program into their origin. In the case of wing tails, initial focus on genes and pathways that instruct outgrowth formation near the posterior end of the hindwing went nowhere. Meanwhile developmentally informed hypotheses were quickly able to zero in on the region-specific activation and inhibition of programmed cell death as the developmental means by which populations and species diversify with respect to wing tail size and shape.
Evo devo revolutionized our understanding of homology. Before evo devo, traits either were, or were not, homologous – end of story. With evo devo, traits became like onions, layered, from genes to pathways to cell types to tissues to organs. We now know that clearly non-homologous traits (such as butterfly wing patterns, beetle horns, and insect legs) can rely on the same, homologous, developmental pathways to help instruct aspects of their development. Conversely, we also learned that clearly homologous structures (such as the leg of a fly and that of a butterfly or beetle) may arrive at their final form by utilizing clearly divergent modes of development. More generally, evo devo has taught us that homology comes in shades of gray, that it can be partial, and that it can be hidden deep in the making of a trait, visible only if we carefully unpack how organisms build themselves and their parts during development.
Beginning to understand the nature of innovation
All this leads to the most important contribution of evo devo, the framing of a research program to investigate the origins of novel complex traits and evolutionary transitions. Evo devo now permits investigators to explore the origins of novel structures through comparative developmental approaches. For example, recent evo devo work on insect wings illustrates how the fusion of two body regions whose existence well predates that of wings – the lateral notum and the most proximal leg region – appears to have provided sufficient biological opportunities (e.g. in terms of cells that were already in the appropriate place to contribute to growth; or gene networks that were already wired in a way to help mediate important developmental events such as where growth should occur) – to enable the emergence of insect wings, gradually, from within the confines of ancestral variation. Many such examples have since been contributed by evo devo studies over the past decades. Collectively, they illustrate the power of the discipline, not to overthrow or discredit conventional approaches, but to extend our combined abilities to deepen our understanding of why and how evolution has unfolded the way it has.
Much of evo devo continues to reside at this junction – identifying the developmental means of evolutionary transitions. But part of the field has begun to push beyond. All evolution takes place within populations, which themselves function within, interact with, and in part actively modify specific ecological conditions. All this of course also applies to developmental evolution, and many evo devo-ists have thus become increasingly interested in the ecological, environmental, and population biological contexts that facilitate, limit, or bias the emergence and spread of developmental variants in natural populations. But that is a whole other conversation, perhaps to be had at a later point.