Mutation predicts 40 million years of fly wing evolution

Abstract

Mutation enables evolution, but the idea that adaptation is also shaped by mutational variation is controversial. Simple evolutionary hypotheses predict such a relationship if the supply of mutations constrains evolution, but it is not clear that constraints exist, and, even if they do, they may be overcome by long-term natural selection. Quantification of the relationship between mutation and phenotypic divergence among species will help to resolve these issues. Here we use precise data on over 50,000 Drosophilid fly wings to demonstrate unexpectedly strong positive relationships between variation produced by mutation, standing genetic variation, and the rate of evolution over the last 40 million years. Our results are inconsistent with simple constraint hypotheses because the rate of evolution is very low relative to what both mutational and standing variation could allow. In principle, the constraint hypothesis could be rescued if the vast majority of mutations are so deleterious that they cannot contribute to evolution, but this also requires the implausible assumption that deleterious mutations have the same pattern of effects as potentially advantageous ones. Our evidence for a strong relationship between mutation and divergence in a slowly evolving structure challenges the existing models of mutation in evolution.

Exaggerated male forelegs are not more differentiated than wing morphology in two widespread sister species of black scavenger flies

“…Since males of the species studied here show no obvious courtship display with their wings, and wings are sexually monochromatic, we can assume that morphological divergence in wing shape is primarily shaped by natural selection. Comparative analysis of 40 million years of wing shape evolution in Drosophila revealed that interspecific shape variation is remarkably similar to that resulting from standing genetic variation or mutation, but that rates of evolution are 10,000 times slower than expected under mutation‐drift equilibrium, implying strong and consistent stabilizing selection on wing shape (Houle, Bolstad, van der Linde, & Hansen, ; see also corresponding communication by Cheverud, ). Similar conclusions have been derived from intraspecific studies of drosophilids.…”

Cross‐sex genetic covariances limit the evolvability of wing‐shape within and among species of Drosophila

“…Despite the correlative and manipulative evidence for high evolvability of wing‐shape in contemporary populations, wing‐shape is remarkably stable across the Drosophila phylogeny (Houle et al. ). Compared to many other insect species Drosophila wings are not very sexually dimorphic (Gidaszewski et al.…”

How does mutation affect the distribution of phenotypes?

“…Although the contribution of the distribution of standing genetic variance has been considered from a variety of perspectives (e.g., Blows and Hoffmann 2005;Estes and Arnold 2007;Futuyma 2010;Losos 2011), the potential role of mutation in determining patterns of phenotypic evolution, particularly on complex, multipeak adaptive landscapes, warrants further attention (Tanaka 1998;Arthur 2004;Losos 2011;Xue et al 2016;Houle et al 2017). Although the contribution of the distribution of standing genetic variance has been considered from a variety of perspectives (e.g., Blows and Hoffmann 2005;Estes and Arnold 2007;Futuyma 2010;Losos 2011), the potential role of mutation in determining patterns of phenotypic evolution, particularly on complex, multipeak adaptive landscapes, warrants further attention (Tanaka 1998;Arthur 2004;Losos 2011;Xue et al 2016;Houle et al 2017).…”

“…Further, analyses of more than two traits suggest that mutationally generated covariance results in unequal distribution of variance across multivariate phenotypic space (Camara and Pigliucci 1999;Estes and Phillips 2006;Houle and Fierst 2013;Latimer et al 2014;McGuigan et al 2014). Whether anisotropy is typical of the multivariate distribution of mutational variance of all trait types, and whether patterns of multivariate phenotypic evolution reflect this mutational characteristic remains to be determined (but see Houle et al 2017).…”